Ferrocenes Ligands Materials and Biomolecules

Ferrocenes Ferrocenes: Ligands, Materials and Biomolecules Editor ˇ EPNI ˇ ˇ PETR ST CKA Charles University, Prague ...

Author: Petr Stepnicka

76 downloads 1430 Views 11MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

Ferrocenes

Ferrocenes: Ligands, Materials and Biomolecules Editor ˇ EPNI ˇ ˇ PETR ST CKA Charles University, Prague

Copyright  2008

John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, England Telephone (+44) 1243 779777

Email (for orders and customer service enquiries): [email protected] Visit our Home Page on www.wileyeurope.com or www.wiley.com All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except under the terms of the Copyright, Designs and Patents Act 1988 or under the terms of a licence issued by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London W1T 4LP, UK, without the permission in writing of the Publisher. Requests to the Publisher should be addressed to the Permissions Department, John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, England, or emailed to [email protected], or faxed to (+44) 1243 770620. Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The Publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the Publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought. The Publisher and the Author make no representations or warranties with respect to the accuracy or completeness of the contents of this work and speciﬁcally disclaim all warranties, including without limitation any implied warranties of ﬁtness for a particular purpose. The advice and strategies contained herein may not be suitable for every situation. In view of ongoing research, equipment modiﬁcations, changes in governmental regulations, and the constant ﬂow of information relating to the use of experimental reagents, equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert or instructions for each chemical, piece of equipment, reagent, or device for, among other things, any changes in the instructions or indication of usage and for added warnings and precautions. The fact that an organization or Website is referred to in this work as a citation and/or a potential source of further information does not mean that the author or the publisher endorses the information the organization or Website may provide or recommendations it may make. Further, readers should be aware that Internet Websites listed in this work may have changed or disappeared between when this work was written and when it is read. No warranty may be created or extended by any promotional statements for this work. Neither the Publisher nor the Author shall be liable for any damages arising herefrom. Other Wiley Editorial Ofﬁces John Wiley & Sons Inc., 111 River Street, Hoboken, NJ 07030, USA Jossey-Bass, 989 Market Street, San Francisco, CA 94103-1741, USA Wiley-VCH Verlag GmbH, Boschstr. 12, D-69469 Weinheim, Germany John Wiley & Sons Australia Ltd, 42 McDougall Street, Milton, Queensland 4064, Australia John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop #02-01, Jin Xing Distripark, Singapore 129809 John Wiley & Sons Canada Ltd, 6045 Freemont Blvd, Mississauga, Ontario, L5R 4J3, Canada Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books. Library of Congress Cataloging-in-Publication Data Stepnicka, Petr. Ferrocenes : ligands, materials and biomolecules / Petr Stepnicka. p. cm. Includes bibliographical references and index. ISBN 978-0-470-03585-6 (cloth) 1. Ferrocene. 2. Ligands. 3. Biomolecules. I. Title. QD412.F4S84 2008 547 .05621 – dc22 2007044547 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN 978-0-470-03585-6 Typeset in 10/12pt Times by Laserwords Private Limited, Chennai, India Printed and bound in Great Britain by Antony Rowe Ltd, Chippenham, Wiltshire This book is printed on acid-free paper responsibly manufactured from sustainable forestry in which at least two trees are planted for each one used for paper production.

Contents Preface

vii

Contributors

xi

PART I

1

FERROCENE LIGANDS

1 Monodentate Ferrocene Donor Ligands Robert C.J. Atkinson and Nicholas J. Long 2 The Coordination and Homogeneous Catalytic Chemistry of 1,1 -Bis(diphenylphosphino)ferrocene and its Chalcogenide Derivatives Sheau W. Chien and T.S. Andy Hor

3

33

3 Synthesis, Coordination Chemistry and Catalytic Use of dppf Analogs Thomas J. Colacot and S´ebastien Parisel

117

4 Other Symmetric 1,1 -Bidentate Ferrocene Ligands Ulrich Siemeling

141

5 1 -Functionalised Ferrocene Phosphines: Synthesis, Coordination Chemistry and Catalytic Applications ˇ epniˇcka Petr Stˇ

177

6 Chiral 1,2-Disubstituted Ferrocene Diphosphines for Asymmetric Catalysis Hans-Ulrich Blaser, Weiping Chen, Francesco Camponovo and Antonio Togni 7 Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors ˇ epniˇcka and Martin Lamaˇc Petr Stˇ

PART II MATERIALS, MOLECULAR DEVICES AND BIOMOLECULES 8 Ferrocene Sensors Simon R. Bayly, Paul D. Beer and George Z. Chen

205

237

279 281

vi

Contents

9 Ferrocene-Based Electro-Optical Materials J¨urgen Heck and Markus Dede

319

10 Ferrocene-Containing Polymers and Dendrimers Nicholas J. Long and Konrad Kowalski

393

11 Ferrocene-Containing Thermotropic Liquid Crystals Robert Deschenaux

447

12 Crystal Engineering with Ferrocene Compounds Dario Braga, Marco Curzi, Stefano Luca Giaffreda, Fabrizia Grepioni, Lucia Maini, Anna Pettersen and Marco Polito

465

13 The Bioorganometallic Chemistry of Ferrocene Nils Metzler-Nolte and Mich`ele Salmain

499

Index

641

Preface

Ferrocene, the ﬁrst known and archetypal metallocene, stirred up a great deal of attention immediately after its discovery in the early 1950s.1,2 However, after the vigorous era of pioneering research aimed predominantly at understanding its basic properties and reactivity had subsided, research activity did not cease in the slightest. Instead, it spread further into many ﬁelds of chemistry and also to related neighbouring disciplines, where it continues with still increasing publication activity (Figure 1). Nowadays, ferrocene is no longer considered a chemical curiosity, but serves as a widely applicable organometallic scaffold for the preparation of functional derivatives that are ﬁnding use in very many areas. These range from mostly academic research aimed at exploring its use in the preparation of various organometallic compounds, to practically directed applications in catalysis, material science and, more recently, also in biomedicinal chemistry. About twelve years have already passed since the legendary book Ferrocenes edited by A. Togni and T. Hayashi was published.3 This, together with the recent developments, certainly justiﬁes publication of a new guide to ferrocene chemistry. Although the areas originally highlighted in the subtitle of the book (Homogeneous Catalysis, Organic Synthesis, Materials Science) still represent the major research ﬁelds in ferrocene chemistry, there have emerged some additional, rapidly developing areas with a strong practical relevance, such as organometallic crystal engineering and, above all, bioorganometallic chemistry, which are also included in this book. Consequently, the book is formally divided into parts, dealing with ferrocene ligands, material aspects of ferrocene chemistry and, ﬁnally, with bioorganometallic chemistry involving ferrocene compounds. However, in view of the amount of literature concerning ferrocene and its derivatives, particularly in the popular areas, this book can no longer provide any kind of an exhaustive literature survey. Therefore, it is conceived for the most part as a combination of the necessary introductory information and a summary of important recent results published up to about the end of the year 2006. For readers seeking

viii

Preface

Figure 1 The number of research articles published per year on ‘ferrocene’ according to CAS

more exhaustive literatures sources, references to literature reviews focusing on particular research areas have been included where appropriate. The book should not only provide a timely summary of important ﬁndings in each particular ﬁeld covered but should also be of help for newcomers (or even random visitors) to ferrocene chemistry who simply make use of some property of a ferrocene derivative. I believe that this book may also serve as a starting point for those who want to become really ‘involved’ with ferrocene. Finally, I wish to thank all the authors for the preparation of their individual contributions that really make this book. Their work, which often nearly turned to a Herculean task because of the enormous amount of recent literature, was always reliable and accurate. My sincere thanks go also to Paul Deards, a commissioning editor at Wiley in Chichester, and his whole editorial team for their encouragement in the initial stages, continuing support during the manuscript preparation and smooth guidance through the entire production process. I hope that the readers will enjoy reading this book and ﬁnd it useful in their own work. ˇ epniˇcka Petr Stˇ Prague, summer 2007

References 1 (a) T.J. Kealy, P.L. Pauson, Nature, 1951, 168, 1039–1040; (b) S.A. Miller, J.A. Tebboth, J.F. Tremaine, J. Chem. Soc., 1952, 632–635; (c) G. Wilkinson, M. Rosenblum, M.C. Whiting,

Preface

ix

R.B. Woodward, J. Am. Chem. Soc., 1952, 74, 2125–2126; (d) E.O. Fischer, W. Pfab, Z. Naturforsch., 1952, 7b, 377–379. 2 For historical essays, see: G. Wilkinson, J. Organomet. Chem., 1975, 100, 273–278; and historical notes published in the special issue of J. Organomet. Chem. dedicated to the 50th anniversary of the discovery of ferrocene, J. Organomet. Chem., 2001, 637–639. 3 A. Togni, T. Hayashi (Eds), Ferrocenes: Homogeneous Catalysis, Organic Synthesis, Materials Science, VCH, Weinheim, Germany (1995).

Note Since the use of abbreviated symbols for ferrocene and ferrocenyl groups is somewhat confusing in the literature, in this book the ferrocenyl group [Fe(η5 -C5 H4 )(η5 -C5 H5 )] is denoted as Fc and the ferrocene-1,1-diyl group [Fe(η5 -C5 H4 )2 ] as fc. By using these abbreviations, ferrocene [Fe(η5 -C5 H5 )2 ] can be formulated either as FcH or fcH2 . Deﬁnitions of the frequently or chapter-speciﬁc abbreviations are provided in the individual parts.

Contributors

Robert C. J. Atkinson and Nicholas J. Long Department of Chemistry, Imperial College London, South Kensington, London, United Kingdom Sheau W. Chien and T. S. Andy Hor Department of Chemistry, National University of Singapore, Singapore Thomas J. Colacota and S´ebastien Pariselb a Johnson Matthey Catalysts, Catalysis and Chiral Technologies, West Deptford (NJ), United States of America b Johnson Matthey Catalysts, Catalysis and Chiral Technologies, Royston, United Kingdom. Ulrich Siemeling Institute of Chemistry, University of Kassel, Kassel, Germany ˇ epniˇcka Petr Stˇ Department of Inorganic Chemistry, Faculty of Science, Charles University in Prague, Czech Republic Hans-Ulrich Blaser,a Weiping Chen,a Francesco Camponovob and Antonio Tognib a Solvias AG, Basel, Switzerland b Department of Chemistry and Applied Biosciences, Swiss Federal Institute of Technology, ETH Zurich, Zurich, Switzerland ˇ epniˇcka and Martin Lamaˇc Petr Stˇ Department of Inorganic Chemistry, Faculty of Science, Charles University in Prague, Czech Republic

xii

Contributors

Simon R. Bayly,a Paul D. Beera and George Z. Chenb a Department of Chemistry, Inorganic Chemistry Laboratory, University of Oxford, Oxford, United Kingdom b School of Chemical and Environmental Engineering, University of Nottingham Nottingham, United Kingdom Jurgen Heck and Markus Dede ¨ Institute of Inorganic and Applied Chemistry, University of Hamburg, Hamburg, Germany Nicholas J. Longa and Konrad Kowalskib Department of Chemistry, Imperial College London, South Kensington, London, United Kingdom b Department of Organic Chemistry, Institute of Chemistry, University of Ł´ od´z, Ł´od´z, Poland a

Robert Deschenaux Institut de Chimie, Universit´e de Neuchˆatel, Neuchˆatel, Switzerland Dario Braga, Marco Curzi, Stefano Luca Giaffreda, Fabrizia Grepioni, Lucia Maini, Anna Pettersen and Marco Polito Dipartimento di Chimica G. Ciamician, Universit`a degli Studi di Bologna, Bologna, Italy Nils Metzler-Noltea and Mich`ele Salmainb a Department of Chemistry and Biochemistry, Ruhr-Universit¨at Bochum, Bochum, Germany b Ecole Nationale Sup´erieure de Chimie de Paris, Laboratoire de chimie et biochimie des complexes mol´eculaires, UMR CNRS 7576, Paris, France

PART I Ferrocene Ligands

1 Monodentate Ferrocene Donor Ligands Robert C.J. Atkinson and Nicholas J. Long

1.1

Introduction and Scope

Even though it was ﬁrst discovered over 50 years ago, research into ferrocene-containing compounds continues apace, largely due to applications within catalysis and materials science.1, 2 In coordination chemistry, the ferrocene moiety has played a signiﬁcant role as a backbone or a substituent in ancillary ligands due to: the speciﬁc and unique geometries that the ferrocene provides; and its electronic (redox) properties, whereby the possibility of switching the redox state of the ferrocene backbone gives potential access to control of reactivity at a metal centre. The gamut of ligands formed via substitution of ferrocenes by various donor heteroatoms have found wide application.3, 4 This chapter focuses on the synthesis and coordination chemistry of monofunctional or monodentate ferrocene ligands, along with a survey of the applications of these ligands, particularly in homogeneous catalysis. Our scope has been those monofunctional ferrocene ligands in which the donor groups are bonded to the ferrocene unit either directly or via a simple methylene spacer, the classiﬁcation being (i) nitrogen donors, (ii) oxygen donors, (iii) phosphorus donors and (iv) chalcogen donors. We have detailed those examples where there has been an application of the ligand or at least extensive coordination chemistry, rather than solely a ligand synthesis. There is also a short section on general synthetic routes to monosubstituted ferrocenes. The chapter concentrates on monofunctional ligands synthesised up to December 2006. It should be noted that, in some cases, the coordination mode of the ligand to Ferrocenes: Ligands, Materials and Biomolecules  2008 John Wiley & Sons, Ltd

Edited by Petr Stepnicka

4

Ferrocenes: Ligands, Materials and Biomolecules

the metal centre in a catalytic system has been established through the synthesis and characterisation of model metal complexes which may provide mechanistic information about the catalytic process under study. However, in other cases, the catalytic efﬁcacy of the ligand has been determined with little evidence for the manner in which the ligand actually interacts with the catalytic metal centre – these ligands are also considered herein, however. There is an excellent summary by Max Herberhold of ‘ferrocene compounds containing heteroelements’ that appeared in the original ‘Ferrocenes’ book of 1995.5 Our aim is not to duplicate the information contained there as much of it is still relevant, especially the more historical synthetic methods and heteroatom derivatives of ferrocene – but rather to bring the ﬁeld up to date, with more recent synthetic methodology and applications of monosubstituted ferrocenes being discussed. Herberhold’s article also contains material on disubstituted ferrocenes, which is not our remit here. In fact, the various di- or tri-substituted heteroatom ferrocenes are covered in Chapters 2–6 of this book. The monofunctional ligands have tended to be overshadowed by the wealth of information and catalytic application of the disubstituted ferrocenes. With more facile and reproducible synthetic routes now available however, the monosubstituted species are undergoing something of a renaissance, especially for applications in catalysis. The ligands are of interest as the substituents may be designed to electronically and/or sterically alter the environment around the catalytic metal centre in such a way as to increase the turnover or, in some cases, to allow the catalysis to happen at all. Enantiomerically pure versions of chiral ligands may favour the formation of a product with a particular conﬁguration so allowing asymmetric catalysis.

1.2

General Synthetic Routes to Monosubstituted Ferrocenes

The isolation of monosubstituted ferrocene derivatives is often not a trivial exercise due to the lack of suitable synthetic routes and difﬁculties in their separation from disubstituted analogues. Although ferrocene undergoes facile electrophilic substitution and mercuration, it is sensitive to oxidation and thus reactions such as halogenation and nitration cannot be used for the synthesis of substituted ferrocenes. In fact, only radical substitution and electrophilic substitution under nonoxidising conditions, i.e. Friedel–Crafts acylation, Mannich reactions, borylation,6 lithiation and mercuration, can be used in the formation of substituted ferrocenes. To incorporate just one heteroatom directly onto a cyclopentadienyl ring can really only be carried out via lithiation and mercuration. Metallation of ferrocene has long been the best method to obtain halogenated derivatives and these species are vital intermediates in the synthesis of heteroatom-substituted ferrocenes. The useful intermediates for the synthesis of monosubstituted ferrocenes are brieﬂy summarised in Scheme 1.1. Lithiation is possibly the most convenient entry into the preparation of ferrocene derivatives. Lithiation with n-butyllithium (nBuLi) generally leads to a mixture of mono- and 1,1 -dilithiated species (Scheme 1.2), though dilithioferrocene can be formed exclusively when n-BuLi is used along with N ,N ,N ,N -tetramethylethylenediamine (tmeda) in hexane.7, 8 To obtain exclusively monolithioferrocene, t-BuLi has to be used in Et2 O solution. Kagan and co-workers

Monodentate Ferrocene Donor Ligands

5

B(OH)2 (i) B(OnBu)3

Li Fe

Fe

(ii) hydrolysis

n-BuLi

tosyl bromide

or t-BuLi

Br

or (BrCX2) 2

Fe

Fe

(i) Hg(OAC)2 NBS HgCl

(ii) LiCl

I2

Fe

I Fe

Scheme 1.1 Ferrocene intermediates used for the incorporation of heteroatoms

Li

Li +

Fe

Fe

Li

n-BuLi hexane Fe

t-BuLi Li

Et2O Fe

Scheme 1.2 Formation of lithioferrocenes using n- or t -BuLi

have made exhaustive investigations into the best conditions for the lithiation of ferrocene and subsequent monosubstituted ferrocene derivatives. The lithiation agent, reaction time, temperature and solvent each play an important role and they used trin-butylstannyl derivatives for puriﬁcation purposes and as precursors for reaction with electrophiles giving monosubstituted ferrocenes in nearly quantitative yields.9, 10 The mercuration of ferrocene to give chloromercuri-ferrocenes is normally facilitated via a one-pot reaction of ﬁrstly Hg(OAc)2 followed by addition of a chloride salt such as potassium chloride or lithium chloride.11 The mixture of mono- and disubstituted ferrocene–HgCl species formed can be puriﬁed by Soxhlet extraction and sublimation.12 The HgCl substituent can be easily exchanged via halogenating agents to give mono-halogenated ferrocenes – along with the lithio-species, the most versatile precursors towards ferrocenes bearing heteroelements. For instance, as shown in

6

Ferrocenes: Ligands, Materials and Biomolecules OPh

OC(O)CH3

Fe

Fe

(ii)

(iii)

SPh

Br (i)

Fe

Cl

(iv)

Fe

Fe

(vi)

(v)

CN

NPh2 Fe

Fe

Scheme 1.3 Some examples of copper-assisted substitution reactions of FcBr; (i) CuSPh, pyridine; (ii) KOPh/Cu, xylene, 160 ◦ C; (iii) Cu(OAc)2 , 135–140 ◦ C; (iv) CuCl, pyridine; (v) CuCN, pyridine, 135–140 ◦ C; (vi) NaNPh2 , CuBr, 120 ◦ C

Scheme 1.3, the halogen group in FcX (generally bromine and iodine, and to a lesser extent chlorine) can be replaced with other anionic groups via nucleophilic substitution in the presence of copper(I) salts and polar solvents such as pyridine.13, 14 Using this methodology, chloro-15 and cyano-16 substituents can be incorporated, as well as derivatives featuring nitrogen, oxygen and sulfur, and these are discussed in the following sections.

1.3

Nitrogen-Substituted Ferrocenes

Aminoferrocene (FcNH2 , 1) has long been a ‘holy grail’ for ferrocene chemists and coordination chemists in general and although substituted ferrocene amines are known, the formation of N-substituted ferrocene species has been hampered by the lack of good synthetic routes. However, in recent years more efﬁcient routes to the synthesis of the primary amine (FcNH2 ), and indeed the diamine fc(NH2 )2 , have opened up the area. It remains the most reliable and versatile route into appending a nitrogen substituent directly onto the ferrocene cyclopentadienyl rings. The ﬁrst report on 1 came from Nesmeyanov et al. in 195517 who reacted FcLi with the O-benzyl ether of hydroxylamine (H2 NOCH2 Ph). Yields though were disappointing (25 %) so the same group devised routes to aminoferrocene from (i) the reaction of N -ferrocenylphthalimide and hydrazine hydrate (N2 H4 •H2 O) in boiling ethanol18, 19 and (ii) FcN3 , and its reduction by lithium aluminium hydride (LiAlH4 )20 (Scheme 1.4). Both routes produced aminoferrocene in yields of ca. 70–80 %, though

Monodentate Ferrocene Donor Ligands

Fc

Br

NaN3 HCONMe2/H2O

+

Fc

−

N=N=N

LiAlH4

Fc

7

NH2

Et2O 1

Scheme 1.4 The azide route to amine 1

the preparations were not straightforward due to difﬁculties in handling the air- and moisture-sensitive material. The azide of ferrocenyl carboxylic acid (FcC(O)N3 ) could also be gainfully used either to form a urethane21, 22 or an acetamide.23 The amine is formed via hydrolysis (Scheme 1.5) and Herberhold et al. were able to isolate the product in high yield after puriﬁcation as the crystalline hydrochloride salt. H N COCH2Ph

Fe

O

PhCH2OH

10% KOH

O NH2

C Fe

N3

Fe

H Ac2O, H3

O+

10% KOH

N

1

CMe

Fe

O

Scheme 1.5 Synthesis of 1 via urethane and acetamide intermediates

In recent times, a number of these routes have been revisited and improvements made. Bildstein et al. have formed 1 in large scale and reasonable yields via the sequence FcH – solid FcLi – FcI – N -ferrocenylphthalimide – 1.24 The main advantage was the selective monometallation of ferrocene and the direct bromination and iodination of lithioferrocene on a large scale (>30 g), along with the avoidance of intermediates such as ferrocene boronic acid and (chloromercurio)ferrocene. Butler and Richards have used a modiﬁed Curtius rearrangement to form 1 and its pentaphenylferrocene derivative in improved yields.25 In the formation of isocyano derivatives, van Leusen and Hessen have perhaps detailed the most convenient and widely used route to 1 – via α-azidostyrene.26 The method was based on a procedure developed by Hassner et al. for the synthesis of anilines and heteroaromatic amines27 and involves the reaction of aryllithium reagents with α-azidostyrene – a reagent that is readily available from styrene in three simple steps.28 Ferrocene is lithiated in tetrahydrofuran (THF) with 0.9 equiv. of t-BuLi, and then reacted with α-azidostyrene at −70 ◦ C. Acidiﬁcation with hydrochloric acid followed by extraction with water and precipitation with base gave crude 1 in ca. 50 % yield; vacuum sublimation facilitated further

8

Ferrocenes: Ligands, Materials and Biomolecules

puriﬁcation. This attractive route enables the preparation of 1 of good purity in multigramme quantities. The electrochemistry of 1 and other ferrocene amines indicates that the amine substituent acts as an unusually potent activating group for ferrocene oxidation, with 1 oxidising at a potential 0.37 V more negative than ferrocene itself.29 With the development of better synthetic routes, the chemistry of 1 has been studied in detail, with facile alkylation and acylation and this has led to a wide range of derivatives, some of which are now discussed. Amine 1 reacts with chlorosilanes in the presence of Et3 N to give N -silylated derivatives 2 such as FcNH(SiMe3 ) (2a) and FcNH(SiMe2 H) (2b), whilst N-lithiation of 1 followed by reaction with Me3 SnCl forms N -stannyl derivatives, FcNH(SnMe3 ) or FcN(SnMe3 )2 (2c). These N-functionally substituted derivatives can also be extended to the N -boryl analogues such as FcN(SiMe3 )BEt2 or FcNHBEt2 .30 Carre et al. have described the synthesis of 1,1 -bis(N -t-butyl-N -hydroxyamino) ferrocene, where the two hydroxylamino substituents are in eclipsed positions31 whilst Knochel detailed the synthesis of FcNHAr via the reaction of arylazotosylates with functionalised organomagnesium compounds.32 This general and elegant amination method features a one-pot reaction sequence consisting of a Grignard reaction, allylation and reduction to give the functionalised diarylamines, e.g. 3, in good yields (Scheme 1.6). In a ‘one-off’ but useful reaction, (di-p-tolylamino)ferrocene was synthesized using palladium-catalysed C−N bond formation.33 This route to (diarylamino) ferrocenes was developed as an alternative to the rather unpredictable Ullmann-type coupling reactions. Amine 1 has also featured in the synthesis of a range of ‘donoracceptor’ complexes incorporating ferrocene species. The ferrocene unit was linked to a metal–nitrosyl acceptor via a variety of conjugated bridges and the compounds exhibited reasonable second order nonlinear optical (NLO) behaviour that could be redox-switchable.34, 35 N2Ts

H N

MgBr Fe

Br

Fe

+ Br

3

Scheme 1.6 Preparation of ferrocenyl aryl amines via arylazo tosylates32

Amine 1 is also the parent compound for the important derivative isocyanoferrocene (4), formed via the dehydration of formamidoferrocene (Scheme 1.7).36 The same group also published the synthesis of the analogous isothiocyanatoferrocene (FcNCS). The isocyanides in general, have been extensively employed as ligands in organometallic chemistry since they are analogous to, but more basic than, carbon monoxide. Isocyanide ligands are more versatile than carbon monoxide in the sense that the substituent on nitrogen can be varied to inﬂuence the donor/acceptor properties of the ligand and to manipulate the architectures of metal complexes comprising the ligand. Aryl isocyanides are better π-acceptors than alkyl isocyanides and the

Monodentate Ferrocene Donor Ligands

9

H N

N NH2 Fe

HC(O)OEt

CH

Fe

O

POCl3

C

Fe

(i-Pr2NH) 4

Scheme 1.7 Synthesis of isocyanoferrocene (4)

ferrocene derivative is a useful addition – it being a stronger σ -donor but a slightly weaker π-acceptor than isocyanobenzene. The solid state structure of 4 shows an almost undistorted ferrocene-like geometry37 and it has recently been used in the stabilisation of ansa-chromocene derivatives38 and to form the unusual [Cr(CNFc)6 ] compound, where there is the incorporation of seven transition metal atoms within the relatively compact ML6 motif.39 The ligands are said to represent a new class of aromatic isocyanides incorporating nonbenzenoid π-systems. Schiff-base ligands are ubiquitous within coordination chemistry and, in recent years, salicylaldiminato complexes of the early transition metals have played important roles in homogeneous catalysis, in particular as active pre-catalysts for ethylene polymerisation. Within this area, ferrocenyl-substituted Schiff-base ligands and their complexes have been widely explored, largely due to the easily accomplished condensation reactions of amines with acyl- or formyl-ferrocenes. The ﬁrst such ferrocene ligand, FcN=CH(C6 H4 OH-2) (5; Scheme 1.8), was reported in 197740 and further investigated 10 years later, it being synthesised by the condensation reaction of 1 with salicylaldehyde. A range of late transition metal complexes have been formed with 5 to study the electrochemical and magnetic properties,41 but it is the applications within ethylene polymerisation that have brought the ligand motif to the fore. Long and Gibson et al. have formed a range of sterically-hindered ligands 6 by the condensation of 1 with a range of salicylaldehydes, which have then been bound to nickel or chromium metal centres (Scheme 1.8).42 Although very sensitive to air, the chromium complexes were found to act as pre-catalysts for the polymerisation of ethylene. Similar nickel-based complexes featuring pyridyl- and quinoidyl-N -substituted ferrocene ligands have proven to be very efﬁcient pre-catalysts for the formation of short chain ethylene oligomers43 and the same collaborative team is focusing on using the redox-active nature of the ferrocene unit to effect redox-switching within homogeneous catalysis.44, 45 A series of related magnesium, titanium and zirconium complexes of ferrocenyl-substituted salicylaldiminato species have been recently reported, with the titanium complex exhibiting moderate activity for ethylene polymerisation and the zirconium species being highly active for ethylene oligomerisation.46 An interesting set of ligands also derived from 1 have been N-heterocyclic carbenes with N -ferrocenyl substitution.24, 47 N-Heterocyclic carbenes are of great current interest due to their potential as easily modiﬁed ligands for metal complexes with catalytic applications. The ferrocene substituent, with its unique spatial requirements and powerful electron-donating capacity, may offer an additional stabilisation of the electron-deﬁcient carbene moiety. Bildstein and co-workers have reported the synthesis of benzimidazoline2-ylidenes with one and two N -ferrocenyl groups appended (Scheme 1.9).

10

Ferrocenes: Ligands, Materials and Biomolecules Fc N CrCl2 R

O

R p-tolylCrCl2.3thf

R′

HC Fc

N

R R′

HO

(TMEDA)NiMe2 Fc (5) R = R′ = H (6a) R = H, R′ = 9-Anthracenyl (6b) R = H, R′ = Tryptocenyl (6c) R = R′ = t-Butyl

R′

N Ni

O

N

O

R′

Fc

R

Scheme 1.8 Sterically-hindered ferrocene Schiff base ligands and their nickel and chromium complexes

Fc N

Fc N

+ −

+ −

BF4 N

B(C6H5)4 N

CH3

Fc

Scheme 1.9 Examples of precursors to N -ferrocenyl N-heterocyclic carbenes24, 47

Electrochemical studies indicate a signiﬁcant electronic communication between the carbene moiety and the N -ferrocenyl substituent. Synthetic routes to these directly attached N -ferrocenyl species are not trivial but they do offer some intereresting catalytic potential. For example, when bound to palladium(0), they have been used in the efﬁcient telomerisation of 1,3-butadiene with alcohols,48 showing remarkable catalyst productivities and regioselectivities. The authors hope that the efﬁciency of the catalyst system, formed in situ, as well as the simplicity of the reaction will yield industrial application. Finally, the FcN=motif has featured in a number of other studies. In 1993, phenylazoferrocene was formed and shown to undergo cyclometallation.49 Starting from 1, Imhof has formed a series of heterocyclic imine ligands with a ferrocenyl group

Monodentate Ferrocene Donor Ligands

11

N N

C

Fe

N 7

Scheme 1.10 The fully cyclopentadienyl-conjugated ligand (7)

as the substituent at the imino nitrogen atom.50 In 2001, Hall described the synthesis of the ﬁrst ferrocene-functionalised ligand 7 in which all the donor atoms are cyclopentadienyl-conjugated (Scheme 1.10).51 This ligand was designed to be a responsive metal-binding species, and indeed exhibited enhanced electrochemical response (of the ferrocene moiety) to copper (Cu+ ) ion binding relative to similar ligands in which the donor atoms are not conjugated with the cyclopentadienyl (Cp) ring.

1.4

Oxygen-Substituted Ferrocenes

Nesmeyanov and coworkers ﬁrst reported hydroxyferrocene (FcOH, 8) in 1959, generating it from either ferrocenylboronic acid FcB(OH)2 (via reaction with Cu(OAc)2 and then potassium hydroxide) or more conveniently from alkaline hydrolysis (with potassium hydroxide) of the acetate FcOAc (the acetate being accessible from FcBr and Cu(OAc)2 as given above).52, 53 Alcohol 8 is a yellow, very air-sensitive solid and a slightly weaker acid than phenol. The difﬁculties in handling 8 mean that its chemistry has not been fully developed though a range of simple derivatives are now known (Scheme 1.11). For example, methoxyferrocene can be formed via methylation

OMe

OR

Fe

Fe

(ii) OEMe3 Fe

(iii) OH

(i) Fe

OC(O)R (iv)

Fe

8

Scheme 1.11 Some reactions of FcOH(8); (i) Me3 ECl (E = silicon, tin); (ii) Me2 SO4 ; (iii) RX (R = CH2 COOH, CH2 CH2 =CH2 ); (iv) R−C(O)Cl (R = Ph, Fc)

12

Ferrocenes: Ligands, Materials and Biomolecules

of 8 with dimethyl sulfate (Me2 SO4 ) and alkylation with activated halides RX gives ferrocenyl ethers FcOR.54 The routes to ferrocenyl ethers have been improved upon over the years. Ferrocenyloxy2-tetrahydrofuran was formed from FcB(OH)2 55 and in 1981, Akabori and co-workers produced a convenient preparation for ferrocenyl esters and ethers (Scheme 1.12). Acylation of a ferrocenylhalide, followed by reduction with sodium hydride (NaH) gives the desired products in reasonable yield.56, 57 More recently, Plenio et al. showed that ferrocenyl aryl ethers could be formed via copper(I)-catalysed routes.58 For example, the coupling reaction of iodoferrocene with various phenols, a base (such as caesium carbonate or potassium phosphate) and CuI/2,2,6,6-tetramethylheptane-3,5-dione as the catalyst gives the products in excellent yield. X

RCO2H

Fe

Cu2O, MeCN X = bromine, iodine

O CR

Fe O

R′Br NaH/15-crown-5

R = alkyl,aryl

OR′

Fe

R′ = alkyl

Scheme 1.12 Synthesis of ferrocenyl ethers

Further derivatives where the oxygen atom is connected directly to the ferrocene unit can be obtained from the reaction of 8 with organoelement chlorides such as Me3 SnCl, t-BuPCl2 , t-Bu2 PCl and t-Bu2 AsCl.59 Analogous trimethylsiloxy derivative (FcOSiMe3 ) can be obtained via the reaction of lithiated ferrocene with bis (trimethylsilyl)peroxide – a route that has also been used in the formation of novel 1,1 -P/O ferrocenediyl ligands (see Chapter 5).60 Due the sensitivity of the FcO− species and the difﬁculty in producing large quantities of ligands, the coordination chemistry is rather limited, though there are a few interesting examples in the literature. In situ generation of potassium salts has been used to form crown ether-type polyoxaferrocenophanes.61 These in turn can bind metal cations so acting as ‘chemical sensors’. Reaction of 8 with various chlorides of both three and ﬁve valent phosphorus leads to a series of mono- to trinuclear ferrocenolato derivatives Ph3−n P(OFc)n (n = 1–3)62, 63 and organogold compounds of methoxyferrocene are known.64, 65 Cyclopalladation is one of the most studied organometallic reactions and usually involves exceptionally high regioselectivity. As an analogue of phenol, hydroxyferrocene 8 has been converted into a phosphite ester with chiral (racemic) butane-1,3diol, and undergoes cyclopalladation similarly to hydroxyarene phosphites.66 Planar chirality is present but no diastereoselectivity is observed. An interesting example of a redox-switchable hemilabile ligand (RHL) 9 has been reported, starting from ferrocenylacetate and reacting with TsOCH2 CH2 Cl followed by KPPh2 (Scheme 1.13; see Chapter 4 for 1,1 -analogues to 9). On complexation to rhodium(I), the authors have demonstrated electrochemical control of the coordination environment around the metal centre.67 When the ligand chelates, the Rh−O (ether) bond is weak and oxidation of the adjacent ferrocenyl group further weakens this bond to dissociation point

Monodentate Ferrocene Donor Ligands

O

13

PPh2

Fe

9

Scheme 1.13 A redox-switchable ferrocenyl ligand 9

and a η6 -arene-bridged dimer is formed. The oxidation state dependent behaviour of the rhodium chelate complex is proof of the RHL concept: the electrochemical interconversion of the square planar chelate complex and the arene-bridged dimer illustrates the use of RHLs for controlling the electronic and steric environment of transition metal centres, and has potential application in catalysis.

1.5

Phosphorus-Substituted Ferrocenes

Phosphorus-substituted ferrocenes are the most well-studied class of heteroatomsubstituted ferrocenes and whilst much of their interest lies in the disubstituted species, i.e. the applications of 1,1 -bis(diphenylphosphino)ferrocene (dppf) and analogues in catalysis (see Chapter 2), there have also been many interesting investigations into monosubstituted ferrocenylphosphines and their applications within homogeneous catalysis. The possibility of almost limitless variation of the substituent groups has made phosphines extremely popular ligands in organometallic chemistry, in particular chiral phosphines and their complexes for use in asymmetric catalysis. Phosphorus derivatives of ferrocenes were ﬁrst investigated in 1962 by Sollott and co-workers.68 Air-stable ferrocenylphenylphosphines were formed by the interaction of ferrocene with phenylphosphonous- and phosphinous chlorides in the presence of anhydrous aluminium(III) chloride (AlCl3 ), i.e. under Friedel–Crafts conditions. However, the method has not proven wide-ranging due to derivatives being poorly characterised, obtained as mixtures or relatively inaccessible. Knox and Pauson published an improved synthesis of ferrocenyldimethylphosphine (FcPMe2 ) that involved methylation of FcPCl2 , sometimes a difﬁcult precursor to reliably obtain.69 Methanolysis of FcPCl2 yields unstable dimethyl ferrocenylphosphonite (FcP(OMe)2 ) that can be converted to methyl ferrocenylphosphinite (FcPH(O)(OMe)) on heating or chromatography. An improved synthesis for FcPCl2 has been reported70 and the availability of this species gives access to the other previously unknown members of the halo series – the diﬂuorides, dibromides and diiodides. The general class of ferrocenyldihalophosphines is a valuable synthon for the generation of a large variety of ferrocenylphosphines. As expected the major application of phosphorus-donor ferrocenes has been in catalysis. For example, Carretero and co-workers71 have used readily available and air-stable ferrocenylphosphines as new catalysts for Baylis–Hillman reactions between aldehydes and acrylates. In the search for highly nucleophilic yet air-stable phosphines the ferrocenyldialkylphosphines have come to the fore (being impressively more reactive than PPh3 and PCy3 ), and catalysed the reaction affording adducts in high yields

14

Ferrocenes: Ligands, Materials and Biomolecules

and short reaction times, with the least hindered diethylphosphine (FcPEt2 , 10) giving complete conversion within one hour with 98 % adduct yield. The authors have also tested a range of planar chiral ferrocenyldialkylphosphines in asymmetric Baylis–Hillman reactions. Indeed, planar chiral ferrocenylphosphines have provided countless examples of excellent enantiocontrol in catalytic asymmetric metalcatalysed reactions – see the excellent review articles by Colacot72 and Richards.73 Monophosphines containing a ferrocenyl moiety have been particularly effective ligands for catalytic asymmetric metal-catalysed reactions such as dialkyl-zinc additions to aldehydes, allylic alkylations, cross-coupling reactions and aldol reactions. To form enantiomerically pure P-chiral phosphines, PCl3 is generally used as the starting material with three sequential nucleophilic displacements to introduce alkyl or aryl groups. In 1997, Brown and Laing examined the methods of asymmetric synthesis of P-chiral monophosphines featuring a bulky group.74 These routes included the arylation of P-chlorooxazaphospholidine, and formation of diarylphosphine boranes, pioneered by Jug´e and Genet.75 The latter route appeared most successful with tertiary phosphines being formed in greater than 92 % ee. Jamison and co-workers have formed a series of ferrocenylphosphines with high ee also by ephedrine-based oxazaphospholidine borane complexes, with primary alkyl, secondary alkyl and substituted aromatic substituents introduced at the P centre.76 The synthetic route (Scheme 1.14) provides facile access to this underdeveloped class of chiral monophosphines. Examples of their use in catalysis include: the nickel-catalysed reductive coupling of aldehydes,77 regioselective, asymmetric reductive coupling of 1,3-enynes and ketones,78 formation of enantiomerically pure primary allylic amines79 and asymmetric conjugate addition of diethylzinc to enones.80 Recently, a method involving reaction of a dichlorophosphine with a chiral lithiated ferrocene, followed by a second organometallic reagent has been communicated. The relatively straightforward method gives access to a range of highly stereoselective ferrocene-based P-chiral phosphine ligands.81 The coordination chemistry of ferrocenylphosphines is extensive, especially ferrocenyldiphenylphosphine, which is analogous to PPh3 . Examples of its unidentate Ph R O

RLi

Me Ph

H3B

N

H 3B

OH

N

Ph

P

Ph

retention

P

Me

Me

inversion MeOH, H2SO4

Me OMe R Et2NH, heat

P

Ph

Fc R

retention

Ph H3B

FcLi inversion

P Fc

Ph H 3B

P R

Scheme 1.14 Formation of chiral ferrocenylmonophosphines via ring-opening of an oxazaphospholidine borane

Monodentate Ferrocene Donor Ligands

15

coordination include: binding to group 10 metal centres to form square planar complexes,82, 83 with the ferrocenyl ligands taking up a transoid orientation, as stabilising ligands in Rh(I) Vaska-type complexes,84, 85 in formation of transition metal [60]fullerene complexes,86 within Group 8 metal clusters87–91 and as a redox-active substituent.85, 92–96 Wrighton and co-workers showed that the electron density around a metal centre such as rhenium could be achieved via control of the redox state of a ferrocenylphosphine ligand.97 For example, in 11, the ferrocenyl units can be reversibly oxidised by one electron each, the oxidation being ferrocenyl-centred (Scheme 1.15). The authors showed that the electron density at the metal centre can be predictably adjusted and tuned by oxidation of a pendant redox centre. Thus examination of the effect of the oxidation state of a pendant redox ligand on the rate of reaction at an affected metal centre could be made. Cl CO

FcPh2P Re

CO

FcPh2P OC 11

Scheme 1.15 A ferrocenyl–rhenium complex (11)

Although the coordination chemistry of tertiary phosphine ligands is well-known, that of primary and secondary ferrocenylphosphines has been largely neglected as most of these phosphines are highly air-sensitive and therefore difﬁcult to handle. Nevertheless, Hey-Hawkins and co-workers have successfully bound species such as FcPH2 to molybdenum(II) and tungsten(II) complexes and used them as single-component catalysts for the metathesis polymerisation of norbornene and norbornadiene.98, 99 For several years, Hartwig has been interested in the development of ligand structures for palladium-catalysed cross-coupling reactions, whereby the ligands can activate aryl chlorides under mild conditions, and effect high conversions with very low catalyst loadings.100–105 Electron rich, sterically hindered monodentate ligands have been investigated, with di(t-butyl)phosphinoferrocene (12) and its 1 ,2 ,3 ,4 ,5 pentaphenyl analogue 13 a particular focus.106 The ligands can be synthesised in relatively facile fashion and in reasonable yields (Scheme 1.16). In the coupling of phenoxides with unactivated aryl halides, complexes of 12 showed excellent activities. However, it was discovered that perarylation occurred in the catalytic process and the true catalyst was based on ligand 13 which indeed exhibited higher activities when isolated in its pure form. This remarkable ligand, known as Q-phos has been shown to be a very general ligand for cross-coupling processes, and examples include Suzuki reactions of aryl and primary alkylboronate esters, aryl halide etheriﬁcations at room temperature, aryl halide aminations, arylation of malonates and the Heck arylation of oleﬁns at room temperature.107 Ligand 13 is indeﬁnitely stable in air as a solid and also in solution. This stability was assumed to be a kinetic phenomenon and probably results from a steric hindrance of the ligand that is increased by a preferential conformation of 13 that

16

Ferrocenes: Ligands, Materials and Biomolecules

Fe

P(t-Bu)2

(i) t-BuLi

Fe

(ii) ClP(t-Bu)2

Pd(OAc)2 NaO-t-Bu, PhCl solvent, 95–110 °C

12

P(t-Bu)2 Ph

Fe

Ph

Ph Ph

Ph 13

Scheme 1.16 Formation of di(t -butyl)phosphinoferrocene ligands 12 and 13

pushed the lone pair towards the aryl groups on the ferrocene. Although not trivial due to mixtures of products being formed, the authors have formed a series of arylated di(t-butyl)phosphinoferrocenes to investigate the effects of sterics and electronics on catalytic activity. Dicyclohexylphosphinoferrocene has recently been used in the annulation of aromatic imines via directed C−H bond activation107 and the general class of ferrocenyl monophosphines can catalyse the Suzuki–Miyaura coupling of aryl chlorides.107 The efﬁcient activation of the latter remains an important goal due to their inexpensive costs and convenient availability, and electron-rich ferrocenylphosphines have a role to play in this area. Xiao and co-workers108 have formed a series of ortho-arylated ferrocenyl phosphines (Scheme 1.17) based on Buchwald’s biphenyl-based ligands – now ubiquitous in palladium-catalysed cross-coupling reactions.109 The ligands can be synthesised in three steps from ferrocenyl phosphine oxides. The ﬁrst step is ortho-lithiation of the oxides to give the iodo-substituted product, then the aryl groups are introduced by reaction with arylboronic acids and ﬁnally, the free phosphines can be easily obtained from the oxides by reduction with trichlorosilane. This neat method allows for the facile synthesis of (arylferrocenyl)phosphines and these electron-rich species were very effective ligands for the palladium-catalyzed Suzuki–Miyaura process, coupling aryl chlorides with efﬁciently low catalyst loadings. MeO

Fe

PPh2

Fe

PPh2

Me

Fe

PCy2

Fe

PCy2

Scheme 1.17 Examples of (monoarylferrocenyl)phosphines108

Examples of other phosphorus-containing substituents on monofunctional ferrocenes include phosphinate, phosphonate and thiophosphonate derivatives. Though not widespread, these species present a potential useful alternative to the straightforward phosphine ligands, via the incorporation of harder oxygen or softer sulfur donor atoms. Phosphinates are usually resistant to oxidation and hydrolysis, the phosphinate groups acting as bridging ligands and have found application in hybrid organic–inorganic materials and molecular level devices. The synthesis of the ligands is relatively straightforward

Monodentate Ferrocene Donor Ligands

17

depending on the availability of starting materials. For example, ethyl ferrocenylphenylphosphinate (FcP(O)(Ph)(OEt), 14) is formed via the reaction of monolithioferrocene with chloroethyl phenylphosphonate, itself obtained by chlorination of ethyl phenylphosphinate, prepared by ethanolysis of PhPCl2 .110 Zinc, cadmium and manganese complexes are known with the structures involving tetracoordinated metals doubly-bridged by the phosphinate groups and exhibiting high thermal stability. Other alkyl derivatives of these ligands have been reported111 as have some interesting ferrocenyl hydroxymethylphosphines (FcP(CH2 OH)2 ) and their oxide, sulﬁde and selenide derivatives.112 These hydroxymethylphosphines are attractive ligands as the hydroxyl groups confer water-solubility and their reactivity with amines and alkenes for example provide access to a wide range of derivatives. Their formation is via reaction of the primary phosphine FcPH2 with formaldehyde and puriﬁcation by dynamic vacuum gives the hydroxymethylphosphine as a brown crystalline solid, soluble in polar organic solvents and indeﬁnitely stable in air. The chalcogenide derivatives, FcP(E)(CH2 OH) (E = oxygen, sulfur, selenium), can be formed by reaction of the parent phosphine with hydrogen peroxide, and powdered sulfur or selenium and ultrasound. All derivatives are crystalline and soluble in polar organic solvents and their coordination chemistry and applications in catalysis is certainly worthy of investigation. Reaction of the ferrocenyl Lawesson’s Reagent Fc(S)PS2 P(S)Fc with NaOR (R = Me, i-Pr) gives the nonsymmetric phosphonodithiolato anions [Fc(RO)PS2 ]− , which can be complexed to a range of metals.113 The versatile coordination behaviour and stability may ﬁnd application as phosphodithiolates in general have been used in many commercial applications. The search for redox-active ligands has seen the formation of a ferrocenylfunctionalised phosphaneiminato ligand114 and a phosphido ligand which can be generated by ring-opening of the P−C bond of a phosphorus-bridged [1]ferrocenophane, and subsequent insertion of a Cp(CO)Fe fragment (see Chapter 5).115

1.6

Chalcogen-Substituted Ferrocenes

There are two main routes into sulfur-substituted ferrocenes: (i) the electrophilic sulfonation of ferrocene to form FcSO3 H and (ii) the insertion of sulfur into the carbon– lithium bond of lithioferrocene. With the improvements in mono-lithiation techniques (see earlier in this Chapter) the latter method is perhaps most useful, especially in the preparation of mercaptoferrocene (FcSH) and the related thioethers, FcSR. The original method of forming FcSH featured the hydrogenation of the sulfonyl chloride derivative FcSO2 Cl,116 and this can be improved upon by forming an intermediate and easily puriﬁed ammonium salt FcSO3 NH4 , which is then treated with PCl3 to give FcSO2 Cl. Following reduction with lithium aluminium hydride, FcSH can be isolated as an orange–brown solid. The dimeric FcS−SFc can also be treated with LiAlH4 to give the desired FcSH, and although this route requires extra steps the dimer can be puriﬁed and handled easily thus ensuring more efﬁcient conversion to the thiol.117 Insertion of sulfur into the Fc−Li bond, leads to FcSLi(THF) when the reaction is carried out in THF.118 This is a light yellow, very air-sensitive solid that can be easily hydrolysed to give FcSH, though mixtures of products are often obtained.

18

Ferrocenes: Ligands, Materials and Biomolecules

FcSH is a very reactive compound, easily combining with activated oleﬁns118 and acyl chlorides,119 where once again the crucial step in the synthesis involves insertion of sulfur into the Li−C bond followed by treatment with acyl chloride (RC(O)Cl). Ferrocenylthioethers can be conveniently prepared via the lithiation route and quenching with a suitable electrophile, such as alkyl disulﬁdes or phenyl disulﬁde, where the weak S−S bond is broken.120 Another route involves reaction of iodoferrocene with a copper bronze and an organic thiol.121 Though not versatile the route does give FcSR ligands in reasonable yields. In 2002, Bonini et al. produced a series of enantiomerically pure hydroxyalkyl- and aminoalkyl- ferrocenyl sulﬁdes from the reaction of FcSLi and substituted epoxides.122 Meanwhile, Brown and co-workers have reported that ferrocenyl sulﬁdes afford meta-lithiation products with up to 94 % regioselectivity on reaction with s-BuLi.123 The FcSR species were prepared by either reaction of ferrocenyllithium and the corresponding disulﬁde124 or via the corresponding sulfoxide.125 This route thus enabled the formation of a range of 1,3-disubstituted ferrocenes – an unusual but desirable substitution pattern. In recent years, chiral ferrocenyl sulfoxides have proven to be of great interest due to their involvement in the preparation of enantiopure 1,2-disubstituted ferrocenes with a predictable absolute conﬁguration.126–128 The three step process is shown in Scheme 1.18 and involves a highly diastereoselective ortho-functionalisation step. The routes should lead to a diverse range of 1,2-disubstituted ferrocenes for use in asymmetric catalytic reactions. Other S=O substituted ferrocene systems include substituents featuring (i) chiral sulﬁnyl groups129 – a useful chiral controller in asymmetric aziridination and allylation of hydrazones – and (ii) chiral sulfoximido groups.130

p-Tol

R1

S Fe

Fe

p-Tol

R1 R2

S O

Fe

O Fe

Scheme 1.18 Formation of chiral ferrocenyl sulphoxide and subsequently enantiopure 1,2-disubstituted ferrocenes

As could be expected, the coordination chemistry of FcSH and FcSR is wellestablished (though not nearly as widespread as with the ferrocenediyl analogues), especially when softer donating atoms are called for; some representative examples of transition metal complexes are now detailed. In similar studies regarding redoxactive hemilabile ligand, as mentioned in Section 1.4, Mirkin et al. have formed a square planar rhodium(I) complex featuring bis-bidentate coordination of RHL ligand FcSCH2 CH2 PPh2 (15).131 The complex undergoes small molecule-induced intramolecular electron ‘pinch and catch’, as the authors remark, i.e. the uptake and release of small molecules can be engineered. The complex reacts with the π-acid CO but is inert towards the σ -donor ligand CH3 CN. However, oxidation of the complex, via the ferrocene groups effects an uptake of CH3 CN and in fact the paramagnetic, square planar doubly oxidised

Monodentate Ferrocene Donor Ligands

19

rhodium(I) complex becomes a diamagnetic, octahedral rhodium(III) species with two additional CH3 CN ligands. Thus, this is an initial example of the way ligands can be designed to allow for the controlled uptake and release of small molecules at transition metal centres. Transition metal complexes with chalcogen ligands have been seen as synthetic models for active sites of metalloenzymes and heterogeneous sulﬁde catalysts. In this area, several novel ferrocenylchalcogenolate-bridged diruthenium complexes 16 have been formed via the oxidative addition of diferrocenyl dichalcogenides (Scheme 1.19).132 Ferrocenylchalcogenato bridges have also featured in binuclear cyclopentadienyl vanadium and tantalum complexes, where the chalcogen can be sulfur, selenium or tellurium.133 Finally, ferrocenylthioethers have been used as supporting ligands for multi-metallic clusters, adding redox-active behaviour and imparting stability via the unique characteristics of the fc unit, e.g. ruthenium carbonyl clusters,134 iron carbonyl clusters135 and molybdenum–iron–sulfur clusters.136 *Cp

Cl Ru EFc

FcE Ru *Cp

Cl 16

Scheme 1.19 An example of ferrocenylchalcogenolate-bridged diruthenium complexes (16) (E = sulfur, selenium)

The insertion of sulfur into the carbon–lithium bond of lithioferrocenes to give FcSLi is a synthetic method that is conveniently extended to the heavier chalcogens selenium137 and tellurium.138, 139 The lithium chalcogenates FcELi (E = sulfur, selenium, tellurium) can then be used in situ for reactions with halogen-containing compounds. In fact, a method has been reported for gaining pure FcLi from FcTeBu,140 and for the synthesis of a range of selenium and tellurium substituted ferrocenes.140, 141 Similar methods for the formation of ferrocenyl thioether ligands have been employed for the seleno analogues,120, 142 i.e. reaction of lithioferrocene with RSe−SeR. The ligands and their palladium(II) complexes have been used as catalysts for selective hydrogenation and Grignard cross-coupling reactions. Examples include the selective hydrogenation under both homogeneous and heterogeneous conditions for the reduction of dienes to monoenes, and the Grignard cross-coupling for haloalkanes and allylmagnesium halides. Ferrocenyl selenoether ligands have also been used in the synthesis of a Cd4 Se6 adamantoid cluster complex, functionalising the surface with redox active centres143 and as part of polysiloxane-supported metal complexes 144 17 (Scheme 1.20). The ligand was immobilised on fumed silica and then reacted with potassium chloroplatinate. The platinum complexes were efﬁcient catalysts for hydrosilylation of oleﬁns with triethoxysilane.

20

Ferrocenes: Ligands, Materials and Biomolecules O Se(CH2)11Si-SiO2 Fe

Se(CH2)11Si-SiO2 O 17

Scheme 1.20 Polysiloxane-supported ferrocenyl selenoether ligands (17)

The ﬁrst examples of ferrocenyltellurium compounds were reported in 1987.139 Insertion of tellurium into FcLi gives diferrocenylditelluride (Te2 Fc2 ) in 50 % yield. The reaction of Te2 Fc2 with organolithium compounds (RLi) was then used to prepare ferrocenyltellurides FcTeR (R = Fc, n-Bu). Other ferrocene-containing telluroether ligands (formed via FcLi and FcTe−TeFc respectively) and complexes have been reported by Singh et al.145 and Nishibayashi et al.,146 though overall this is a very underdeveloped area.

1.7

Monosubstituted Ferrocene Donor Ligands Featuring a Carbon Spacer

There is extensive literature on the class of compounds where the donor atom is not directly substituted onto one of the ferrocene cyclopentadienyl rings but attached via a carbon spacer unit. This is largely due to more facile synthetic routes due to the ready availability of stable and inexpensive starting materials (e.g. acetylferrocene, ferrocenylmethanol) and that the ﬁnal compounds are generally more stable. Compounds with the various donor atoms detailed previously, e.g. nitrogen, oxygen, phosphorus and chalcogen will be reviewed, but it is the nitrogen-containing derivatives that dominate the ﬁeld. 1.7.1

Nitrogen-Donor Compounds

Ferrocene was ﬁrst aminomethylated in 1955 by Lindsay and Hauser via the reaction of ferrocene with paraformaldehyde and dimethylamine in glacial acetic acid.147–149 The product (N ,N -dimethylaminomethyl)ferrocene (FcCH2 NMe2 , 18) represented a signiﬁcant breakthrough as it provided a new route to hitherto unavailable ferrocene derivatives, such as alcohols, oximes, imines and aldehydes. The ligand can be derivatised by reaction with 6-amino-2-picoline to introduce a pyridyl unit alongside the amino-nitrogen donor atom, and then coordinated to metal centres such as gold, silver and copper.150, 151 Via lithiation, the ligand 18 itself can be bound to silver and platinum.152 An unusual ligand featuring two ethylpyridine linkages bound to the amino nitrogen centre has been formed by Halcrow and co-workers.153 A bidentate binding mode is observed when coordinated to zinc (Scheme 1.21), but the compound soon decomposes via C−N bond cleavage. However, the expected tridentate

Monodentate Ferrocene Donor Ligands

21

Fc N N

Zn

N Br

Br

Scheme 1.21 The zinc complex of a ferrocenyl bis[2-(pyrid-2-yl)ethyl]amine derivative

coordination mode is prevalent on coordination to cobalt, nickel and copper and no decomposition is seen. The aminomethyl ferrocene fragment has also been used as an integral part of redox-responsive systems. As (N -propylaminomethyl)ferrocene, the fragment operates as a proton-sensitive redox-responsive unit, illustrated by dramatic changes in the ferrocene/ferrocenium redox potential.154 The fragment has been bound to naphthalimide, itself an electroluminescent ﬂuorophore used in molecular switches and bioprobes, to investigate whether oxidation state dependence may be a factor in switching on or off the ﬂuorescence. It was found that the N -ferrocenyl substituents do not perturb the energy levels of the ﬂuorophore although emission is quenched.155 An interesting application of the derivatised aminomethyl ferrocene fragment has been in pseudorotaxane formation.156 The potential to form molecular shuttles and machines continues to excite and so it was of note when the redox active unit was reported to form a pseudorotaxane with a crown ether via an electrochemical stimulus and in the presence of a suitable hydrogen source. Ugi has shown that ferrocenyl derivatives such as a C-chiral amine FcCH(Me)NMe2 can be lithiated with high diastereoselectivity, and the resulting organolithium may be trapped with various electrophiles (E+ ) to provide 1,2-disubstituted ferrocenes [Fe{η5 C5 H3 (E)(CH(Me)NMe2 )-1,2}(η5 -C5 H5 )] (19; see Chapter 6).157 Rather than a methylene spacer unit, a C=N linkage has also commonly been used in the formation of functional ligands. For example, some new ferrocenyl Schiff bases and their aluminium and zinc compounds have been used in catalysis158 and in second-order nonlinear optics.159 A series of substituted ferrocenyl compounds have been formed (Scheme 1.22) and analysed for their second-order optical nonlinearity. All show a reasonable response compared to the urea standard and there is correlation with the electron withdrawing nature of the substituted benzene ring. The C=N linkage from the ferrocenyl unit has also been used in the formation of nickel dithiocarbamate complexes bearing ferrocenyl units (Scheme 1.22).160 The authors were attempting to form molecular systems capable of exchanging electrons with an electrode, via the coupling of multiple, identical metal-centered fragments. Thus, a new ferrocene ligand 20 (Scheme 1.22) was prepared from ferrocenecarboxaldehyde and then Schiff-base chemistry undertaken. Nickel(II) dithiocarbamate bearing two ferrocenyl groups and its oxidised product nickel(IV) dithiocarbamate featuring three ferrocenyl groups were synthesised. Electrochemical investigations revealed that, in spite of the chemical equivalence of the ferrocene groups, a mixedvalence state persists in solution, presumably due to electrostatic effects.

22

Ferrocenes: Ligands, Materials and Biomolecules SK H

N

H N

R

Fe

S

N Fe

R = OCH3, H, Cl, NO2

20

Scheme 1.22 Some NLO-active Schiff-base linked ferrocenyl compounds

R NMe2 Pd Cl

Fe

NR′2 Pd Cl

Fe

L

L

Scheme 1.23 Cyclopalladated (aminomethyl)ferrocene derivatives

The most widely-studied class of compounds of these C−N substituted ferrocene derivatives has been the cyclometallated compounds of dimethylaminomethylferrocene and its imine analogue, and in particular the cyclopalladated species (Scheme 1.23). The reasons for their ubiquitous nature stem from their use in asymmetric catalysis via their planar chirality (palladium complexes) and their anti-tumour activity (platinum complexes). Shaw and Gaunt produced the ﬁrst cyclopalladated derivative in 1975161 reacting 18 with sodium chloropalladate(II) in the presence of sodium acetate. For many years Lopez and co-workers have been one of the leaders in this ﬁeld, investigating the effects of the alkyl or other donor substituents,162 the nature of the nitrogen donor atom (sp2 versus sp3 ),163 formation of diastereomerically pure metallacycles164 and the inclusion of additional donor atoms to effect tridentate coordination as opposed to bidentate (Scheme 1.24).165–168 In 1994, they found that in cyclopalladated compounds containing the imine as the chelate ligand, the Pd−N bond H N Fe

Pd Cl L

Scheme 1.24 coordination

NMe2

N Fe

Pd

SMe

L

Cyclopalladated (aminomethyl)ferrocenes with possible bi- or tri-dentate

Monodentate Ferrocene Donor Ligands

23

is clearly less reactive than those containing the amine. In addition, the palladium(II) acts an electron-withdrawing group in the cyclometallated derivatives and that most of the electron density is withdrawn from the CH=N unit. An interesting point is that in the cycloplatinated compounds there are two possible centres for antitumour activity – the ferrocene and the platinum. In a series of papers in 1994,169 – 171 Robinson and Simpson and co-workers synthesised and characterised a range of cyclometallated ferrocenylamine complexes of platinum(II) with a view to examining their cytotoxic activity. Due to the similarities with cis-platin it was thought that the compounds had the potential to produce a spectrum of toxicity and tumour activity. Toxicity, histological and antitumour studies in mice showed that the cyclometallated ferrocenylamines cause kidney rather than liver dysfunction, that they have reasonable toxicity and are mildly cytotoxic against standard tumours. Although only poorly soluble in water or saline solution, they were active against cis-platin resistant cell lines.172 The same group has also produced switchable cycloplatinated ferrocenylamine derivatives of acridone, naphthalimide and anthraquinone.173 Catalytic studies with cyclopalladated ferrocenylamines have centered on asymmetric catalysis and the advantages of such catalysts include their ease of synthesis, facile modiﬁcation and convenience of handling (insensitivity to air and moisture). One of the early examples came from Overman and co-workers174 who used a series of enantiopure cyclopalladated ferrocenyl amines and imines as catalysts for the [3,3]-rearrangement of allylic benzimidates to allylic benzamides. Reasonable enantioselection was observed and this was also found to be highly dependent upon the nature of the counterion. Mak and co-workers175 have formed a series of enantiopure bis(µ-acetato)-bridged planar chiral cyclopalladated species, and air and moisture stable tricyclohexylphosphine adducts of cyclopalladated ferrocenylamines (Scheme 1.25) have been easily synthesised and used in the palladium-catalysed Suzuki cross-coupling of aryl chlorides.176 The catalysts gave the coupled products in excellent yields in the reaction of nonactivated and deactivated aryl chlorides with phenylboronic acid. The catalyst loadings could also be lowered to 0.01 % mol % without loss of activity. Other palladium-catalysed cross-coupling reactions, such as Heck, Sonogashira177 and Mizoroki–Heck178 have also been catalysed by these classes of compounds. H3C

CH3

H3C N

N

R Pd

Fe L

Fe

Pd

Cl

Cl

PCy 3

Scheme 1.25 Cyclopalladated ferrocene Schiff bases and tricyclohexylphosphine adducts

Finally the ligand class has also stabilised main group and transition metal centres via cyclometallation and formed for example, some unusual heterotrimetallic metalloplumbylene compounds (FcN)2 PbM(CO)5 (M = chromium, molybdenum, tungsten).179–181

24

Ferrocenes: Ligands, Materials and Biomolecules

1.7.2

Oxygen-Donor Compounds

Ferrocenecarboxaldehyde (FcCHO, 21) was ﬁrst prepared in the late 1950s by two different methods, though both used (N ,N -dimethylaminomethyl)ferrocene methiodide ([FcCH2 NMe3 ]I) as the starting material. Hauser and Lindsay182 showed that 21 would undergo typical addition and condensation reactions to form a range of other ferrocene derivatives. Pauson and co-workers183 produced 21 via the Sommelet reaction and also illustrated its versatility as a starting material, forming the analogous oxime, nitrile, alcohol and carboxylic acid derivatives. Ferrocenylmethanol is probably the most studied and reacted species in this class of compounds, largely due to its ready availability. Displacement of the hydroxyl group by amines is relatively easy when carried out in dilute acid184 and FcCH2 OH reacts smoothly with mercaptosuccinic acid to give ferrocenylmethylthiosuccinic acids.185 The alcohol can also be deprotonated with NaN(SiMe3 )2 to form an unsolvated sodium alkoxide which is a useful intermediate in the preparation of early transition metal and lanthanide derivatives containing the ferrocenylmethoxide ligand.186 Swarts and co-workers187 have formed a series of primary ferrocenylalcohols Fc(CH2 )m OH (m = 1–4) via the reduction of the appropriate ferrocenecarboxylic acids (Scheme 1.26). In-depth electrochemical measurements were carried out and the ferrocene group showed reversible electrochemistry with the reduction potential of the ferrocene group being inversely proportional to the side chain length. The inﬂuence of the side chain length on reduction potential was more pronounced for the acids because the electron-withdrawing properties of the carbonyl group are stronger than that of the alcohol group. Ion pairing was also found to play a major role in the electrochemical behaviour of ferrocenylmethanol. Finally, germatranes bearing a ferrocenylalkoxyl moiety have been obtained by the reaction of HOGe(OCH2 CH2 )3 N with various ferrocenyl alcohols188 and used for antitumour and antibacterial activity.

O

O OC

COOH

CO

Fe

Zn/HgCl2

Fe

HCl, CH3OH

AlCl3, CH2Cl2

Zn/HgCl2 HCl, CH3COOH

LiAlH4 AlCl3, CH2Cl2

COOCH3 Fe

LiAlH4, ether O

(CH2)nCOOH Fe

n = 0–3

LiAlH4, ether

(CH2)mOH Fe

NaBH4 alcohol

H Fe

m = 1–4

Scheme 1.26 Synthesis of some ferrocene-containing alcohols187

21

Monodentate Ferrocene Donor Ligands

1.7.3

25

Phosphorus-Donor Compounds

In contrast to the large number of ferrocene-derived phosphines, ligands where there is a carbon spacer between the two functionalities are rare.189, 190 However in recent years the chemistry of ferrocene alkylphosphines (22) and ferrocene hydroxymethylphosphines (23) has developed (Scheme 1.27). Henderson and co-workers have been very active in this area and have investigated the properties and coordination abilities of various primary phosphines and an arsine bearing ferrocene substituents.191–194 Although these ligand types are usually very air-sensitive, they found that primary phosphines with an alkyl linkage between the cyclopentadienyl ligand of ferrocene and the phosphorus or arsenic atom were unexpectedly stable in air over two years. In comparison, ferrocenylphosphine oxidises in air after a few days. The same group has also described the synthesis of the ﬁrst example of a ferrocenederived hydroxymethylphosphine ligand, together with some derivatives.195, 196 Reaction of [FcCH2 NMe3 ]I with an excess of P(CH2 OH)3 gives the air-stable ferrocenylphosphine FcCH2 P(CH2 OH)2 . Further reaction of the CH2 OH groups on phosphorus is facile and a range of derivatives are now known, such as the phosphine oxide, phosphine sulﬁde and various cyano- and amino-derivatives. Removal of formaldehyde from FcCH2 P(CH2 OH)2 with one mole equivalent of Na2 S2 O5 gives the crystalline and completely air-stable primary phosphine FcCH2 PH2 and exhibits all the typical coordinative properties of a primary phosphine.193, 194 Electrochemical studies of 22 and 23 and other derivatives show that the free ligands exhibit complex voltammetric responses due to participation of the P lone pair in the redox reactions. Uncomplicated ferrocene-based redox chemistry is observed for phosphorus(V) derivatives and when the ligands are coordinated to metal centres.192 In 2001, Henderson and co-workers reported the synthesis of ferrocenyl-phosphonic and -arsonic acids197 and showed that platinum(II) complexes of these ligands show moderate activity against P388 leukaemia cells, whereas the parent ligands are inactive.198 More recently, 23 has been reacted with a range of amino acids to form novel phosphino amino acids, being notably water soluble.199 The same group has bound the related primary phosphine ligands 22 (where R = H) to molybdenum and tungsten centres and examined the dynamic behaviour of the complexes in solution,200, 201 whilst Laguna et al. have formed gold and silver complexes with the same ligand.202 Tertiary ferrocenylmethylphosphines are known and have been used as ligands for Suzuki–Miyaura palladium catalysts.203 These phosphines possess an aryl substituent on the methyl bridge in order to maximise steric bulk. Catalytic activity of the

CH2PR2 Fe

22 (R = hydrogen, alkyl, aryl)

CH2P(CH2OH)2 Fe

23

Scheme 1.27 Ferrocene alkylphosphines (22) and ferrocene hydroxymethylphosphines (23)

26

Ferrocenes: Ligands, Materials and Biomolecules

complexes is high but the t-butyl substituted ligand is not stable in air. In contrast, di(t-butyl)(ferrocenylmethyl)phosphine, which lacks the phenyl group on the methylene bridge, is reasonably air-stable as a solid and possesses an electron donating ability similar to that of tri-i-propylphosphine.204 Palladium complexes of this ligand can catalyse room temperature Suzuki–Miyaura coupling reactions with aryl bromides, and exhibit modest yields in Heck couplings. The modest activity seems to stem from the fact that whilst oxidative addition occurs efﬁciently, other steps in the catalytic cycle, such as transmetallation or migratory insertion, are inefﬁcient. Butler and coworkers have shown that ferrocenylmethylphosphine ligands can have an effect in the palladium-catalysed reaction of carbon monoxide, methanol and ethane to obtain methyl propionate, a key intermediate in the preparation of methyl methacrylate. 1.7.4

Chalcogen-Donor Compounds

There are only a very few reports in the literature on this type of compound. In 1999, Laguna and co-workers formed the ferrocene derivative FcCH2 N(CH2 )2 SH from the condensation reaction of 21 with β-mercaptoethylamine. It can be easily oxidised to its disulﬁde and the parent ligand reacts smoothly with gold (I) phosphine cations.205 Some ferrocene-containing penicillins and cephalosporins (featuring thioglycolic acids) are known206 and Bonini et al. have synthesised a range of new thioferrocenoylsilanes (Scheme 1.28). The derivatives can lead to planar enantiomerically pure chiral thioferrocenylsilanes, and have been investigated for applications within asymmetric catalysis.207 S SiR3 Fe

(R = Me, Ph)

Scheme 1.28 Structure of thioferrocenoylsilanes

1.8

Conclusions

As stated previously, the ﬁeld of monofunctional ferrocene ligands has been overshadowed by that of its nearest neighbour, the disubstituted ferrocenes. Difﬁculties in preparing the monofunctional species cleanly (free from the difunctional analogues) and in high yields has held back research in this area. However, with more facile and reproducible synthetic routes now available, the monosubstituted species are once again being focused on and the ﬁeld has become topical. For instance, Hartwig’s work on the donor-substituted pentamethylferrocenyl species has brought catalysis by this type of ligand class into a wider arena. Although many of the ligands and substitution patterns are now well-established, there does seem enormous scope for further growth. Admittedly, having the donor heteroatom directly attached to the Cp ring often

Monodentate Ferrocene Donor Ligands

27

leads to stability problems and careful handling is required. Nevertheless, with more sophisticated glassware and laboratory equipment, this is not such a handicap anymore. Indeed, derivatives of ferrocenylamine are now widespread and even those of ferrocenol are increasing. As with the difunctional analogues, the P-substituted species dominate, especially in catalysis where the ligand properties (i.e. electronics, sterics and chirality) can easily be harnessed and tuned. Importantly, a carbon spacer can be used to aid stability with little loss in donating behaviour or power and air- and water-stable and water-soluble ligands such as ferrocene alkylphosphines and ferrocene hydroxymethylphosphines respectively will surely be further exploited.

References 1. A. Togni, R.L. Halterman (Eds), Metallocenes, Wiley-VCH Verlag GmbH, Weinheim, Germany, 1998. 2. N.J. Long, Metallocenes: An Introduction to Sandwich Complexes, Blackwell Science, Oxford, UK, 1998. 3. A. Togni, T. Hayashi (Eds), Ferrocenes: Homogeneous Catalysis, Organic Synthesis, Materials Science, VCH, Weinheim, 1995. 4. For some excellent reviews on ferrocene ligands see: (a) R.C.J. Atkinson, V.C. Gibson, N.J. Long, Chem. Soc. Rev., 2004, 33, 313–328; (b) P. Barbaro, C. Bianchini, G. Gianbastini, S.L. Parisel, Coord. Chem. Rev., 2004, 248, 2131–2150; (c) U. Siemeling, T-C. Auch, Chem. Soc. Rev., 2005, 34, 584–594; (d) T.J. Colacot, Platinum Met. Rev., 2001, 45, 22–30. 5. M. Herberhold, Ferrocene Compounds Containing Heteroelements, in Ferrocenes: Homogeneous Catalysis–Organic Synthesis–Materials Science, VCH, Weinheim (1995), pp. 219–278. 6. (a) T. Renk, W. Ruf, W. Siebert, J. Organomet. Chem., 1974, 64, C45–C47; (b) W. Ruf, T. Renk, W. Siebert, Z. Naturforsch., 1977, 31b, 1028–1034. 7. J.J. Bishop, A. Davison, M.L. Katcher et al. J. Organomet. Chem., 1971, 27, 241–249. 8. B. McCulloch, D.L. Ward, J.D. Woollins, C.H. Brubaker, Jr., Organometallics, 1985, 4, 1425–1532. 9. F. Rebiere, O. Samuel, H.B. Kagan, Tetrahedron Lett., 1990, 31, 3121–3124. 10. D. Guillaneux, H.B. Kagan, J. Org. Chem., 1995, 60, 2502–2505. 11. R.W. Fish, M. Rosenblum, J. Org. Chem., 1965, 30, 1253–1254. 12. M.D. Rausch, L.P. Klemann, A. Siegel et al. Synth. React. Inorg. Metal-Org. Chem., 1973, 3, 193–199. 13. M. Sato, I. Motoyama, K. Hata, Bull. Chem. Soc. Jpn., 1971, 44, 812–815. 14. M.D. Rausch, J. Org. Chem., 1961, 26, 1802–1805. 15. M. Sato, I. Motoyama, K. Hata, Bull. Chem. Soc. Jpn., 1970, 43, 1860–1863. 16. M. Sato, T. Ito, I. Motoyama et al. Bull. Chem. Soc. Jpn., 1969, 42, 1976–1981. 17. A.N. Nesmeyanov, E.G. Perevalova, R.V. Golovnya, L.S. Shilovtseva, Dokl. Akad. Nauk SSSR, 1955, 102, 535–538. 18. A.N. Nesmeyanov, V.A. Sazonova, V.I. Romanenko, Chem. Ber., 1960, 9, 2717–2729. 19. A.N. Nesmeyanov, V.A. Sazonova, V.I. Romanenko, Dokl. Akad. Nauk SSSR, 1964, 157, 922–925. 20. A.N. Nesmeyanov, V.N. Drozd, V.A. Sazonova, Dokl. Akad. Nauk SSSR, 1963, 150, 321–324. 21. F.S. Arimoto, A.C. Haven, Jr., J. Am. Chem. Soc., 1955, 77, 6295–6297.

28 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63.

Ferrocenes: Ligands, Materials and Biomolecules E.M. Acton, R.M. Silverstein, J. Org. Chem., 1959, 24, 1487–1490. M. Herberhold, M. Ellinger, W. Kremnitz, J. Organomet. Chem., 1983, 241, 227–240. B. Bildstein, M. Malaun, H. Kopacka et al. Organometallics, 1999, 18, 4325–4336. D.C. Butler, C.J. Richards, Organometallics, 2002, 21, 5433–5436. D. van Leusen, B. Hessen, Organometallics, 2001, 20, 224–226. A. Hassner, P. Munger, B.A. Belika, Tetrahedron Lett., 1982, 23, 699–702. G. Smolinsky, J. Org. Chem., 1962, 27, 3557–3559. W.E. Britton, R. Kashyap, M. El-Hashash et al. Organometallics, 1986, 5, 1029–1031. B. Wrackmeyer, H.E. Maisel, M. Herberhold, J. Organomet. Chem., 2001, 637–639, 727–732. F. Carre, C. Guerin, B.J.L. Henner, C. Uerpmann, J. Organomet. Chem., 2002, 654, 210–215. I. Sapountzis, P. Knochel, Angew. Chem., Int. Ed., 2004, 43, 897–900. A. Mendiratta, S. Barlow, M.W. Day, S.R. Marder, Organometallics, 1999, 18, 454–456. M. Malaun, R. Kowallick, A.M. McDonagh et al. Dalton Trans., 2001, 3025–3038. S-M. Lee, R. Kowallick, M. Marcaccio et al. Dalton Trans., 1998, 3443–3450. G.R. Knox, P.L. Pauson, D. Willison et al. Organometallics, 1990, 9, 301–306. B. Wrackmaeyer, H.E. Maisel, O.L. Tok et al. Z. Anorg. Allg. Chem., 2004, 630, 2106–2109. P.J. Shapiro, R. Zehnder, D.M. Foo et al. Organometallics, 2006, 25, 719–732. T.C. Holovics, S.F. Deplazes, M. Toriyama et al. Organometallics, 2004, 23, 2927–2938. I.A. Alekseeva, G.N. Yashchenko, T.A. Sinitsyna, L.A. Petrov, Zh. Obshch. Khim, 1977, 47, 1874–1878. M. Bracci, C. Ercolani, B. Floris et al. Guastini Dalton Trans., 1990, 1357–1363. V.C. Gibson, C.K.A. Gregson, C.M. Halliwell et al. J. Organomet. Chem., 2005, 690, 6271–6283. V.C. Gibson, C.M. Halliwell, N.J. Long et al. Dalton Trans., 2003, 918–926. C.K.A. Gregson, V.C. Gibson, N.J. Long et al. J. Am. Chem. Soc., 2006, 128, 7410–7411. R.C.J. Atkinson, V.C. Gibson, N.J. Long et al. Organometallics, 2007, 26, 316–320. R.K.J. Bott, M. Schormann, D.L. Hughes et al. Polyhedron, 2006, 25, 387–396. B. Bildstein, M. Malaun, H. Kopacka et al. J. Organomet. Chem., 1999, 572, 177–187. R. Jackstell, A. Frisch, M. Beller et al. J. Mol. Cat. A: Chem., 2002, 185, 105–112. G.R. Knox, P.L. Pauson, D. Willison, J. Organomet. Chem., 1993, 450, 177–184. W. Imhof, J. Organomet. Chem., 1997, 541, 109–116. T.G. Sprigings, C.D. Hall, Organometallics, 2001, 20, 2560–2564. A.N. Nesmeyanov, V.A. Sazonova, V.N. Drozd, L.A.M.V. Nikonova, Dokl. Akad. Nauk SSSR, 1960, 133, 126–129. A.N. Nesmeyanov, V.A. Sazonova, V.N. Drozd, Tetrahedron Lett., 1959, 1, 13–17. A.N. Nesmeyanov, W.A. Ssazonowa, V.N. Drozd, Chem. Ber., 1960, 93, 2717–2729. R. Epton, G. Marr, G.K. Rogers, J. Organomet. Chem., 1978, 150, 93–100. S. Akabori, M. Sato, S. Ebine, Synthesis, 1981, 4, 278–279. K. Shibata, Y. Saito, M. Matsui, Y. Takase, Bull. Chem. Soc. Jpn., 1984, 57, 3349–3350. M.R. Van der Heiden, G.D. Frey, H. Plenio, Organometallics, 2004, 23, 3548–3551. M. Herberhold, H-D. Brendel, A. Hofmann et al. J. Organomet. Chem., 1998, 556, 173–187. (a) R.C.J. Atkinson, V.C. Gibson, N.J. Long, et al. Organometallics, 2004, 23, 2744–2751; (b) R.C.J. Atkinson, V.C. Gibson, N.J. Long, et al. Dalton Trans., 2004, 1823–1826. S. Akabori, Y. Habata, M. Sato, S. Ebine, Bull. Chem. Soc. Jpn., 1983, 56, 537–541 and 1459–1461. M. Herberhold, A. Hofmann, W. Milius, J. Organomet. Chem., 1998, 555, 187–200. M. Herberhold, A. Hofmann, W. Milius, Z. Anorg. Allg. Chem., 1997, 623, 1599–1608.

Monodentate Ferrocene Donor Ligands

29

64. A.N. Nesmeyanov, N.N. Sedova, S.K. Moiseev, V.A. Sazonova, Dokl. Akad. Nauk SSSR, 1981, 256, 98–101. 65. A.N. Nesmeyanov, E.G. Perevalova, K.I. Grandberg, et al. J. Organomet. Chem., 1974, 65, 131–144. 66. L.L. Troitskaya, S.T. Ovseenko, Y.L. Slovokhotov et al. J. Organomet. Chem., 2002, 642, 191–194. 67. E.T. Singewald, C.A. Mirkin, C.L. Stern, Angew. Chem., Int. Ed., 1995, 34, 1624–1627. 68. G.P. Sollott, H.E. Mertwoy, S. Portnoy, J.L. Snead, J. Org. Chem., 1963, 28, 1090–1092. 69. G.R. Knox, P.L. Pauson, D. Willison, Organometallics, 1992, 11, 2930–2933. 70. R. Pietschnig, E. Niecke, Bull. Soc. Chim. Fr., 1997, 134, 605–608. 71. S.I. Pereira, J. Adrio, A.M.S. Silva, J.C. Carretero, J. Org. Chem., 2005, 70, 10175–10177. 72. T.J. Colacot, Chem. Rev., 2003, 103, 3101–3118. 73. C.J. Richards, A.J. Locke, Tetrahedron: Asymmetry, 1998, 9, 2377–2407. 74. J.M. Brown, J.C.P. Laing, J. Organomet. Chem., 1997, 529, 435–444. 75. (a) S. Jug´e, J-P. Genet, Tetrahedron Lett., 1989, 30, 2783–2786; (b) S. Jug´e, M. Stephan, R. Merdes et al. J. Chem. Soc., Chem. Commun., 1993, 531–532. 76. E.A. Colby, T.F. Jamison, J. Org. Chem., 2003, 68, 156–166. 77. R.M. Moslin, T.F. Jamison, Org. Lett., 2006, 8, 455–458. 78. K.M. Miller, T.F. Jamison, Org. Lett., 2005, 7, 3077–3080. 79. S.J. Patel, T.F. Jamison, Angew. Chem., Int. Ed., 2005, 97, 3941–3944. 80. A. Alexakis, J. Burton, J. Vastra et al. Eur. J. Org. Chem., 2000, 24, 4011–4027. 81. W. Chen, W. Mbafor, S.M. Roberts, J. Whittal, J. Am. Chem. Soc., 2006, 128, 3922–3923. 82. N.J. Long, A.J.P. White, D.J. Williams, M. Younus, J. Organomet. Chem., 2002, 649, 94–99. 83. S. Otto, A. Roodt, J. Smith, Inorg. Chim. Acta, 2000, 303, 295–299. 84. S. Otto, A. Roodt, Inorg. Chim. Acta, 2004, 357, 1–10. 85. F. Estevan, J. Latorre, E. Peris, Polyhedron, 1993, 12, 2153–2156. 86. L-C. Song, G-A. Yu, H-T. Wang et al. Eur. J. Inorg. Chem., 2004, 4, 866–871. 87. T.C. Zheng, W.R. Cullen, S.J. Rettig, Organometallics, 1994, 13, 3594–3604. 88. W.R. Cullen, S.J. Rettig, T.C. Zheng, J. Organomet. Chem., 1993, 452, 97–103. 89. W.R. Cullen, S.J. Rettig, T.C. Zheng, Organometallics, 1992, 11, 853–858. 90. W.R. Cullen, S.T. Chacon, M.I. Bruce et al. Organometallics, 1988, 7, 2273–2278. 91. W.R. Cullen, S.J. Rettig, T.C. Zheng, Organometallics, 1992, 11, 277–283. 92. J.C. Kotz, C.L. Nivert, J.M. Lieber, R.C. Reed, J. Organomet. Chem., 1975, 91, 87–95. 93. F. Estevan, P. Lahuerta, J. Latorre et al. Dalton Trans., 1993, 1681–1688. 94. B. Therrien, L. Vielle-Petit, J. Leanneret-Gris et al. J. Organomet. Chem., 2004, 689, 2456–2463. ˇ epniˇcka, R. Gyepes, O. Lavastre, P.H. Dixneuf, Organometallics, 1997, 16, 95. P. Stˇ 5089–5095. 96. S. Perruchas, N. Avarvari, D. Rondeau et al. Inorg. Chem., 2005, 44, 3459–3465. 97. T.M. Miller, K.J. Ahmed, M.S. Wrighton, Inorg. Chem., 1989, 28, 2347–2355. 98. R. Sommer, P. Loennecke, J. Reinhold et al. Organometallics, 2005, 24, 5256–5266. 99. S.I.M. Paris, F.R. Lernke, R. Sommer et al. J. Organomet. Chem., 2005, 690, 363–370. 100. Q. Shelby, N. Kataoka, G. Mann, J.F. Hartwig, J. Am. Chem. Soc., 2000, 122, 10718–10719. 101. G. Mann, C. Incarvito, A.L. Rheingold, J.F. Hartwig, J. Am. Chem. Soc., 1999, 121, 3224–3225. 102. J.P. Stambuli, S.R. Stauffer, K.H. Shaughnessy, J.F. Hartwig, J. Am. Chem. Soc., 2001, 123, 2677–2678. 103. J.P. Wolkowski, J.F. Hartwig, Angew. Chem., Int. Ed., 2002, 41, 4289–4291. 104. S.R. Stauffer, J.F. Hartwig, J. Am. Chem. Soc., 2003, 125, 6977–6985.

30

Ferrocenes: Ligands, Materials and Biomolecules

105. F. Barrios-Landeros, J.F. Hartwig, J. Am. Chem. Soc., 2005, 127, 6944–6945. 106. N. Kataoka, Q. Shelby, J.P. Stambuli, J.F. Hartwig, J. Org. Chem., 2002, 67, 5553–5566. 107. R.K. Thalji, K.A. Ahrendt, R.G. Bergman, J.A. Ellman, J. Org. Chem., 2005, 70, 6775–6781. 108. C. Baille, L. Zhang, J. Xiao, J. Org. Chem., 2004, 69, 7779–7782. 109. J. Yin, M.P. Rainka, X.X. Zhang, S.L. Buchwald, J. Am. Chem. Soc., 2002, 124, 1162–1163. 110. O. Orms, J. Le Bideau, A. Vioux, D. Leclerq, J. Organomet. Chem., 2005, 690, 363–370. 111. O. Orms, F. Maurel, F. Carre et al. J. Organomet. Chem., 2005, 689, 2654–2661. 112. W. Henderson, S.R. Alley, J. Organomet. Chem., 2002, 658, 181–190. 113. I.P. Gray, H.L. Milton, A.M.Z. Slawin, J.D. Woollins, Dalton Trans., 2003, 3540-3457–. 114. U. Siemeling, A Stammler, H-G. Stammler, O. Kuhnert, Z. Anorg. Allg. Chem., 1999, 625, 845–847. 115. T. Mizuta, M. Onishi, T. Nakazono et al. Organometallics, 2002, 21, 717–726. 116. G.R. Knox, P.L. Pauson, J. Chem. Soc., 1958, 692–696. 117. M. Herberhold, O. Nuyken, T. Pohlmann, J. Organomet. Chem., 1991, 405, 217–227. 118. M. Herberhold, O. Nuyken, T. Poehlmann, J. Organomet. Chem., 1995, 501, 13–22. 119. M. Herberhold, P. Leitner, C. Doernhoefer, J. Ott-Lastic, J. Organomet. Chem., 1989, 377, 281–289. 120. R.V. Honeychuck, M.O. Okoroafor, L.H. Shen, C.H. Brubaker, Jr, Organometallics, 1986, 85, 482–490. 121. M.D. Rausch, J. Org. Chem., 1961, 26, 3579–3580. 122. L. Bernardi, B.F. Bonini, M. Comes-Franchini et al. Eur. J. Org. Chem., 2002, 16, 2776–2784. 123. C. Pichon, B. Odell, J.M. Brown, Chem. Commun., 2004, 5, 598–599. 124. P. Diter, S. Taudien, O. Samuel, H.B. Kagan, J. Org. Chem., 1994, 59, 370–373. 125. K. Nagasawa, A. Yoneta, T. Umezawa, K. Ito, Heterocycles, 1987, 26, 2607–2609. 126. O. Riant, G. Argouarch, D. Guillaneux et al. J. Org. Chem., 1998, 63, 3511–3514. 127. D.H. Hua, N.M. Nadege, Y. Chen et al. J. Org. Chem., 1996, 61, 4508–4509. 128. F. Rebiere, O. Riant, L. Ricard, H.B. Kagan, Angew. Chem., Int. Ed. Engl., 1993, 32, 568–570. 129. I. Fernandez, V. Valdivia, B. Gori et al. Org. Lett., 2005, 7, 1307–1310. 130. C. Bolm, K. Muniz, N. Aguilar, M. Kesselgruber, G. Raabe, Synthesis, 1999, 7, 1251–1260. 131. I.V. Kourkine, C.S. Slone, C.A. Mirkin et al. Inorg. Chem., 1999, 38, 2758–2759. 132. D. Marquarding, H. Klusacek, G. Gokel et al. J. Am. Chem. Soc., 1980, 92, 5389–5393. 133. M. Herberhold, J. Peukert, M. Kruger et al. Z. Anorg. Allg. Chem., 2000, 626, 1289–1295. 134. W.R. Cullen, S.J. Rettig, T.C. Zheng, Polyhedron, 1995, 14, 2653–2661. 135. J. Adeleke, M. Adebanjo, Y.W. Chen, L.K. Liu, Organometallics, 1992, 11, 2543–2550. 136. K. Tanaka, M. Nakamoto, Y. Tashiro, T. Tanaka, Bull. Chem. Soc. Jpn., 1985, 58, 316–321. 137. R. Broussier, A. Abdulla, B. Gautheron, J. Organomet. Chem., 1987, 332, 165–173. 138. M. Herberhold, P. Leitner, C. Dornhofer, J. Ott-Lastic, J. Organomet. Chem., 1989, 377, 281–289. 139. M. Herberhold, P. Leitner, J. Organomet. Chem., 1987, 336, 153–161. 140. A. Chiefﬁ, J.V. Cornasseto, V. Snieckus, Synlett , 2000, 2, 269–271. 141. M.R. Burgess, C.P. Morley, M. Di Vaira, J. Organomet. Chem., 2005, 690, 3099–3104. 142. A.A. Naiini, C-K. Lai, D.L. Ward, C.H. Brubaker, Jr., J. Organomet. Chem., 1990, 390, 73–90. 143. T.P. Lebold, D.L.B. Stringle, M.S. Workentin, J. Corrigan, Chem. Commun., 2003, 12, 1398–1399. 144. J-Z. Yao, Y-Y. Chen, B-S. Tian, J. Organomet. Chem., 1997, 534, 51–56.

Monodentate Ferrocene Donor Ligands 145. 146. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 166. 167. 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 178. 179. 180. 181. 182. 183. 184. 185. 186. 187. 188.

31

H.B. Singh, A.V. Regini, J.P. Jasinski et al. J. Organomet. Chem., 1994, 464, 87–94. Y. Nishibayashi, T. Chiba, J.D. Singh et al. J. Organomet. Chem., 1994, 473, 205–213. C.R. Hauser, J.K. Lindsay, D. Lednicer, C.E. Cain, J. Org. Chem., 1957, 22, 717–718. J.K. Lindsay, C.R. Duke, J. Org. Chem., 1957, 22, 355–358. C.R. Hauser, J.K. Lindsay, J. Org. Chem., 1956, 21, 382–383. E.M. Barranco, O. Crespo, M.C. Gimeno et al. Eur. J. Inorg. Chem., 2004, 24, 4820–4827. E.M. Barranco, M.C. Gimeno, A. Laguna, M.D. Villacampa, Inorg. Chim. Acta, 2005, 358, 4177–4182. K. Jacob, F. Voihgt, K. Merzweiler, C. Pietzsch, J. Organomet. Chem., 1997, 545–546, 421–433. Q.F. Mokuolu, C.A. Kilner, S.A. Barrett et al. Inorg. Chem., 2005, 44, 4136–4138. G. De Santis, L. Fabbrizzi, L. Manotti et al. Inorg. Chim. Acta, 1998, 267, 177–182. J.C. McAdam, B.H. Robinson, J. Simpson, Organometallics, 2000, 19, 3644–3653. M. Horie, Y. Suzaki, K. Osakada, J Am. Chem. Soc., 2004, 126, 3684–3685. D. Marquarding, H. Klusacek, G. Gokel et al. J. Am. Chem. Soc., 1980, 92, 5389–5393. E. Hecht, Z. Anorg. Allg. Chem., 2001, 627, 2351–2358. S.K. Pal, A. Krishnan, P.K. Das, A.G. Samuelson, J. Organomet. Chem., 2000, 604, 248–259. K. Oyaizu, K. Yamamoto, Y. Ishii, E. Tsuchida, Chem. Eur. J., 1999, 5, 3193–3201. J.C. Gaunt, B.L. Shaw, J. Organomet. Chem., 1975, 102, 511–516. C. Lopez, R. Bosque, X. Solans, M. Fnt-Bardia, J. Organomet. Chem., 1997, 539, 99–107. C. Lopez, R. Bosque, X. Solans et al. Dalton Trans., 1994, 3039–3046. C. Lopez, A. Caubet, S. Perez et al. Chem. Commun., 2004, 5, 540–541. C. Lopez, S. Perez, X. Solans, M. Font-Bardia, J. Organomet. Chem., 2005, 690, 228–243. S. Perez, C. Lopez, A. Caubet et al. Organometallics, 2006, 25, 596–601. C. Lopez, A. Caubet, S. Perez et al. J. Organomet. Chem., 2002, 651, 105–113. S. Perez, C. Lopez, A. Caubet et al. New. J. Chem., 2003, 27, 975–982. N.W. Duffy, J.C. McAdam, B.H. Robinson, J. Simpson, Inorg. Chem., 1994, 33, 5343–5350. P. Ranatunge-Bandarage, R. Ramani, J. Simpson et al. Organometallics, 1994, 13, 511–521. P. Ranatunge-Bandarage, R. Ramani, J. Simpson, B.H. Robinson, Organometallics, 1994, 13, 500–510. K. McGrouther, D.K. Weston, D. Fenby et al. Dalton Trans., 1999, 1957–1966. E.M. McGale, E.R. Murray, J.C. McAdam et al. Inorg. Chim. Acta, 2003, 352, 129–135. F. Cohen, L.E. Overman, Tetrahedron: Asymmetry, 1998, 9, 3213–3222. G. Zhao, Q-C. Yang, T.C.W. Mak, Organometallics, 1999, 18, 3623–3636. J. Gong, G. Liu, C. Du et al. J. Organomet. Chem., 2005, 690, 3963–3969. Z. Tibor, A. Csampai, A. Kotschy, Tetrahedron, 2005, 61, 9767–9774. X.M. Zhao, X.Q. Hao, B. Liu et al. J. Organomet. Chem., 2006, 691, 255–260. N. Seidel, K. Jacob, A.K. Fischer, Organometallics, 2001, 20, 578–581. N. Seidel, K. Jacob, A.K. Fischer et al. Eur. J. Inorg. Chem., 2001, 1, 145–151. K. Jacob, F. Voigt, K. Merzweiler, et al. J. Organomet. Chem., 1998, 552, 265–276. C.R. Hauser, J.K. Lindsay, J. Org. Chem., 1957, 22, 906–908. G.D. Broadhead, J.M. Osgerby, P.L. Pauson, J. Chem. Soc., 1958, 650–656. A.L.J. Beckwith, G.G. Vickery, Perkin Trans. 1 , 1975, 18, 1818–1821. R. DAbard, B. Misteriewicz, H. Platin, J. Wasielewski, J. Organomet. Chem., 1987, 328, 185–192. H. Gornitzka, F.T. Edelmann, K. Jacob, J. Organomet. Chem., 1992, 436, 325–332. W.L. Davis, R.F. Shago, E.H.G. Langner, J.C. Swarts, Polyhedron, 2005, 24, 1611–1616. L. Chen, J-X. Chen, L. Sun, Q. Xie, Appl. Organomet. Chem., 2005, 19, 1038–1042.

32

Ferrocenes: Ligands, Materials and Biomolecules

189. 190. 191. 192. 193. 194. 195. 196. 197. 198. 199. 200. 201.

S. Hoppe, H. Welchmann, K. Jurkschat et al. J. Organomet. Chem., 1995, 505, 63–72. Y. Yamamoto, T. Tanase, I. Mori, Y. Nakamura, Dalton Trans., 1994, 3191–3192. A.J. Downard, N.J. Goodwin, W. Henderson, J. Organomet. Chem., 2003, 676, 62–72. W. Henderson, S.R. Alley, J. Organomet. Chem., 2002, 656, 120–128. N.J. Goodwin, B.K. Nicholson, Inorg. Chim. Acta, 1999, 295, 18–24. N.J. Goodwin, W. Henderson, B.K. Nicholson et al. Dalton Trans., 1999, 1785–1794. N.J. Goodwin, W. Henderson, J.K. Sarfo, Chem. Commun., 1996, 13, 1551–1552. N.J. Goodwin, W. Henderson, B.K. Nicholson et al. Dalton Trans., 1997, 4377–4384. S.R. Alley, W. Henderson, J. Organomet. Chem., 2001, 637–639, 216–229. W. Henderson, S.R. Alley, Inorg. Chim. Acta, 2001, 322, 106–112. A.A. Karasik, R.N. Naumov, R. Sommer et al. Polyhedron, 2002, 21, 2251–2256. R. Sommer, P. Loennecke, J. Reinhold et al. Organometallics, 2005, 24, 5256–5266. R. Sommer, P. Loennecke, P.K. Baker, E. Hey-Hawkins, Inorg. Chem. Commun., 2002, 5, 115–118. E.M. Barranco, O. Crespo, M.C. Gimeno et al. Inorg. Chem., 2000, 39, 680–687. Z-Y. Tang, Y. Lu, Q-S. Hu, Org. Lett., 2003, 5, 297–300. M.D. Sliger, G.A. Broker, S.T. Grifﬁn et al. J. Organomet. Chem., 2005, 690, 1478–1486. E.M. Barranco, M.C. Gimeno, P.G. Jones et al. Inorg. Chem., 1999, 38, 702–706. D. Scutaru, L. Tatani, I. Mazilu et al. J. Organomet. Chem., 1991, 401, 81–85. B.F. Bonini, M. Comes-Franchini, M. Fochi et al. J. Organomet. Chem., 2001, 637–639, 407–417.

202. 203. 204. 205. 206. 207.

2 The Coordination and Homogeneous Catalytic Chemistry of 1,1-Bis(diphenylphosphino)ferrocene and its Chalcogenide Derivatives∗ Sheau W. Chien and T. S. Andy Hor

2.1

Introduction

Since our last reviews on the coordination and catalytic chemistry of 1,1 -bis(diphenylphosphino)ferrocene (dppf),1 research work with this ligand has further intensiﬁed. Today, not only has it become one of the ‘choice ligands’ among the diphosphines, its wide acceptance has also inspired the development of other derivatives such as R2 PfcPPh2 (R = Ph)2 and R2 PfcPR2 (R = R = Ph). There have also been numerous attempts to develop other relatives, e.g. through replacing the metal by its congenors, namely [M(η5 -C5 H4 PPh2 )2 ] (M = ruthenium3 and osmium)4 , oxidation to chalcogenides dppfE and dppfE2 (E = oxygen5 , sulfur6 , selenium7 ), alkylation ([Ph2 PfcPR Ph2 ]X), adduct formation (e.g. dppf•2BH3 8 ), and also via other modiﬁcations at the phosphorus atom(s) such in the related phosphine-phosphonites R2 PfcP(OR)2 9 and phosphonates R2 PfcP(O)(OR)2.10 As this chapter is dedicated solely to the parent ligand, namely dppf, and its oxides and chalcogenides, other derivatives and modiﬁcations are excluded from the coverage. The unsymmetric derivatives bearing different ∗ Speciﬁc abbreviations used throughout this chapter are given at the end of the chapter before the Reference List.

Ferrocenes: Ligands, Materials and Biomolecules  2008 John Wiley & Sons, Ltd

Edited by Petr Stepnicka

34

Ferrocenes: Ligands, Materials and Biomolecules

substituents at the phosphorus atoms and the phosphine–phosphonites and phosphonates are covered in Chapters 3 and 5. There are many advantages of dppf over other alkyl or aryl based diphosphines. A typical example of this is found in the etheration of aryl halides.11 The skeletal ﬂexibility remains a key feature that is almost unmatched by any other diphosphines. This topic was extensively covered in our original article.1 The electro-,12 photo-,13 and material14 activities of the metallocenyl core also bring along a host of practical uses to dppf and its relatives. Without question, among these applications, the catalytic potential has attracted the most attention. Hence, in this review, two of the most significant and vigorous advances in dppf research are focused on – coordination chemistry and homogeneous catalysis. Since our last review, Bandoli and Dolmella15 have updated the ligand chemistry of dppf, and Colacot16 has summarised some developments of dppf in its catalytic coupling chemistry. These articles provided adequate coverage of the background and essential developments of dppf up to early 2000s. The main objective of this review is to summarise some representative works since the turn of the new millennium and to highlight some new or unusual chemistry that has been discovered recently. Dppf is a typical metalloligand. It carries an organometallic moiety but the metal rarely participates in active interaction with other functional substrates. The chemistry is largely dictated by the two phosphine donors at the peripheral ends of the metallocenyl backbone. The metal (iron) henceforth does not usually interfere with the coordination and activity of the donors.17 Yet, the electronic buffer and stereogeometrical tuning provided by the metallocenyl core are inherent features of dppf. Not only do they help the phosphines to coordinate to a range of metals in different redox and geometric environments through different coordination modes but they also enable the ligand to react to changes in the metal coordination sphere. This adaptation helps the metal to adjust to the needs of the incoming substrates, as well as to stabilise and solubilise the resultant complex. The collective outcome is the improvement of catalytic performance. In this chapter, the recent reactions that are promoted by dppf complexes or use dppf as a supporting ligand are also summarised. A word of caution is that in many of the reactions described, other diphosphines could also be active. The differences are not always clear. A few cases in which comparative study between dppf and other mono- or diphosphines was reported, are highlighted accordingly.

2.2 2.2.1

Coordination Chemistry Preparation of dppf Complexes

The most common method used in the preparation of dppf complexes is ligand replacement from precursors with labile donors. A typical example is the formation of [Pt(η2 −PhC≡CC≡CPh)(dppf)] from [Pt(cod)2 ], dppf and 1,4-diphenylbutadiyne. When there are no labile departing ligands, it is customary to use thermal, photolytic, or chemo-oxidative techniques to activate the leaving group (for examples, see Table 2.1). Addition reaction of unsaturated complexes with the dppf ligand is also easy to carry out. A good example would be the formation of [CuI(dppf)]2 directly

The Coordination and Catalytic Chemistry of dppf

35

from copper(I) iodide with dppf. Use of dinuclear bridging halide complexes as precursors is a variation of this approach. For example, addition of dppf to [RhCl2 Cp*]2 would yield [RhCl2 Cp*]2 (µ-dppf). Many unsaturated species can be generated in situ in solution from stable compounds. This method allows many halide or pseudohalide complexes to be used as precursors for dppf complexes. This is best represented by the use of a halide abstractor such as Tl+ (as in Tl[PF6 ]), Ag+ (as in AgOTf), or even H+ (such as H[BF4 ]) or a metathesis agent such as Na+ in sodium salts to activate and remove an anionic ligand. The cationic species with a coordination site that is either vacant or taken up by a weak donor (e.g. solvent) would then readily capture the basic dppf. Good examples include the preparations of [Rh(cod)(dppf)](OTf) from [RhCl(1,5-cod)]2 , AgOTf and dppf, and [RhClCp*(dppf)][PF6] from [RhCl2 Cp*]2 , dppf and Na[PF6 ]. It is also possible to combine metathesis with addition and displacement reactions in one step. Formation of [Ru(dppf)2 ] from [RhCl(C8 H12 )]2 and Ag[BF4 ] is representative of this approach. An unusual source of dppf is its selenide derivative, i.e. dppfSe2 . This is fairly stable, but towards zero-valent metal carbonyls, such as [Ru3 (CO)12 ], it undergoes reductive and dechalcogenative cleavage to give [Ru3 (µ3 -Se)2 (CO)7 (µ-dppf)]. One single reagent thus serves three purposes – releasing dppf as a donor, oxidising the metal and providing a source for single atomic selenium. Its parent compound, dppf, usually serves as a neutral ligand with no synthetically signiﬁcant redox properties. However, towards high-valent metals that are strongly oxidising, it is possible to use dppf as a reducing source as well as its usual donor character. Synthesis of [Tc(N)Cl2 (dppf)] from [Ph4 As][Tc(N)Cl4 ] and dppf is an example of such approach. A list of dppf complexes prepared and characterised in the period 1999 to 2007 is given in Tables 2.1 and 2.2. Their preparative conditions are also summarised. 2.2.2

Reactivity

The metallocene moiety of dppf is fairly stable and generally considered as nonfunctional. The reactivity at the phosphine sites is typical of a diphosphine. The most distinctive and signiﬁcant activity of dppf is the change of coordination mode, especially with respect to changes in the ligand environment. Such changes are closely associated with the catalytic state of the complex, or the metal, to be precise. It is, therefore, essential to understand the dynamic behavior of coordinated dppf. A list of the known changes in coordination modes of the recently established dppf complexes is given in Table 2.3. 2.2.2.1

Coordination Mode Changes

A coordinated dppf ligand is generally chemically stable. Skeletal disintegration rarely occurs except under strongly thermolytic or photolytic conditions. Such decomposition usually takes place in a metal cluster core, giving rise to common fragments such as phenyl, cyclometalated C5 and C6 rings, hydride, cyclopentadienyl, and halfsandwiched Fe−Cp moieties and even Fe−M species etc. As these functionalities are easily trapped by metals in vicinity, the thermal or photolytic products are usually generated in form of a complex mixture of clusters. This mode of skeletal decomposition is unique in dppf as compared to other alkyl- or aryl-based diphosphines.18

7

6

Group

Re

Tc

Mn

W

Mo

Metal

60 )]

No No No No No No Yes Yes No No No

mer -[W(CO)3 (dppf)(η2 -C70 )]

mer -[W(CO)3 (dppf)]2 (η2 ,η2 C70 )

[W(CO)3 (EtCN)(dppf)]

fac -[W(CO)3 (SO2 )(dppf)]

mer -[W(CO)3 (CH2 SO2 )(dppf)]

[Mn(CO)3 (FBF3 ) (dppf)]•3/2CH2 Cl2

[Mn(CO)4 (dppf)][BF4 ]•CH2 Cl2

[Tc(N)Cl2 (dppf)]

[Tc(NPh)Cl3 (dppf)]

[Re(O)Cl2 (OEt)(dppf)]

Yes

mer-[W(CO)3 (dppf)]2 (η2 ,η2 -C60 )

mer-[W(CO)3

No

[Mo(CO)3 (EtCN)(dppf)]

(dppf)(η2 -C

Yes

X-ray structure

fac-[Mo(CO)3 (CH3 CN) (dppf)]•1/2H2 O

Compound

C

C

C

C

C

C

C

C

C

C

C

C

C

C

Coordination mode of dppfa

1 2 2 2 3 3 4 4 5

Ditto [W(CO)3 (NCEt)3 ] + dppf in EtCN for 1 h [W(CO)3 (EtCN)(dppf)] + SO2 in CH2 Cl2 at 0 ◦ C for 2 h, then 20◦ C for 24 h [W(CO)3 (SO2 )(dppf)] + CH2 N2 in ether/CH2 Cl2 at 0 ◦ C [MnH(CO)3 (dppf)] + H[BF4 ] in CH2 Cl2 for ∼10 min [Mn(CO)3 (FBF3 )(dppf)]•3/2CH2 Cl2 + CO in CH2 Cl2 for ∼19 h [Ph4 As][Tc(N)Cl4 ] + dppf, in C6 H6 for ∼2 h [Tc(NPh)Cl3 (PPh3 )2 ] + dppf, in C6 H6 /CH2 Cl2 for 2 h [Re(O)Cl3 (OPPh3 )(Me2 S)] + dppf, in EtOH

1

[W(CO)4 (dppf)] + C70 , hν in chlorobenzene

1

[W(CO)4 (dppf)] + C60 , hν in chlorobenzene

1

2

[Mo(CO)3 (EtCN)3 ] + dppf, in EtCN for 1 h Ditto

1

Ref.∗

[Mo(CO)3 (CH3 CN)3 ] + dppf, in MeCN for 30 h

Substrates and preparative conditions

Table 2.1 A selected list of recent mono- and polynuclear homometallic dppf complexes

36 Ferrocenes: Ligands, Materials and Biomolecules

8

C B

No No Yes No No

Yes No Yes Yes

No

No No

[ReCl3 (NC6 H4 Fc-4)(dppf)]

[ReCl3 {(NC6 H4 -4) CO2 C6 H4 Fc-4}(dppf)]

fac -[Re(O)Cl3 (dppf)]

fac -[Re(NPh)Cl3 (dppf)]

mer -[ReCl3 (NC6 H3 -2,6-i -Pr2 ) (dppf)]

fac -[Re{OC(O)OMe}(CO)3 (dppf)]

[Re(O)Cl2 (OH)(dppf)]

[Re2 Cl6 (dppf)]

[Fe2 (µ-SCH2 OCH2 S)(CO)5 ]2 (µ-dppf)

[Ru(µ-O2 CC2 F4 CO2 )(CO) (H2 O)(dppf)]2 •H2 O

[Ru{η5 :σ -Me2 C(C5 H4 )(C2 B10 H10 )}(dppf)]

[RuCl2 (η6 -C6 H5 (CH2 )2 OC(O)Fc)]2 (µ-dppf)

Fe

Ru

B

C

C

C

C

C

C

C

C

C

C

Yes

[Re(N)Cl2 (dppf)]

C

Yes

fac -[ReH(CO)3 (dppf)] 7

fac -[Re(O)Cl3 (dppf)] + 4-FcC6 H4 NH2 , in C6 H6 for 24 h

15

[RuCl2 {η6 -C6 H5 (CH2 )2 OC(O)Fc}]2 + dppf in CH2 Cl2 for 24 h

(continued overleaf )

14

[Ru(cod)][η5 :σ -Me2 C(C5 H4 )(C2 B10 H10 )]b + dppf, in THF for 24 h

12

[Fe2 (µ-SCH2 OCH2 S)(CO)6 ] + Me3 NO•2H2 O + dppf in MeCN at r.t. for 4h

13

11

(Bu4 N)2 [Re2 Cl8 ] + dppf, in MeOH for 3 h

[Ru(µO2 CC2 F4 CO2 )(CO)(H2 O)(PPh3 )2 ]2 + dppf, in toluene for 1 h

10

9

[Re2 (CO)10 ] + Me3 NO•2H2 O + dppf in MeOH/THF under vacuum at r.t. for 4 h [Re(O)Cl3 (dppf)] in aq. Me2 CO or THF or repetitive recrystallization from wet ether

8

8

mer,trans-[ReCl3 (NPh)(PPh3 )2 ] + dppf, in C6 H6 for 1 h mer,cis-[ReCl3 (NC6 H3 -2,6-i Pr2 )2 (C5 H5 N)] + dppf in CH2 Cl2 for 48 h at r.t.

8, 10

[Re(O)Cl3 (PPh3 )2 ] + dppf in C6 H6 at r.t. for ∼2 h or in THF for 30 min

7

4

[Ph4 As][Re(N)Cl4 ] + dppf, in C6 H6 for ∼2 h

fac -[Re(O)Cl3 (dppf)] + 4-FcC6 H4 CO2 (4-C6 H4 NH2 ), in C6 H6 for 72 h

6

[Re2 (CO)10 ] + dppf, in 1-pentanol for 1 d

The Coordination and Catalytic Chemistry of dppf 37

Group

Metal

X-ray structure No No

Yes No Yes Yes Yes Yes Yes Yes No Yes No

Compound

[RuCl2 (η6 -C6 H5 (CH2 )3 OH)]2 (µ-dppf)

[RuCl2 (η6 -C6 H5 (CH2 )3 OC(O)Fc)]2 (µ-dppf)

[Ru(NCS)Cp(dppf)]

[Ru(NCS)(HMB)(dppf)][PF6 ]c

[Ru(NCS)2 (HMB)]2 (µ-dppf)c

[Ru(CH3 CN)Cp(dppf)][BPh4 ]

[Ru(CH3 CN)(HMB) (dppf)] [PF6 ]2 c

[Ru(PMe3 )Cp(dppf)][PF6 ]

[Ru(PMe2 Ph)Cp(dppf)][PF6 ]

[Ru(S2 CNEt2 )Cp]2 (µ-dppf)

[Ru(S2 CNEt2 )2 (dppf)]

[{Ru(µ-S2 )Cp(dppf)}2 ][BPh4 ]Cl

[RuCl2 (1,2,4-C6 H3 Me3 )]2 (µdppf)

B

C

C

B

C

C

C

C

B

C

C

B

B

Coordination mode of dppfa

Table 2.1 (continued )

15

16 16 16 16 16 16 16 16 16 16 17

[RuClCp(dppf)] + NaNCS in MeOH for 6 h [RuCl(HMB)(dppf)][PF6 ] + NaNCS + in MeOH for 23 h [RuCl(HMB)(dppf)][PF6 ] + NaNCS in MeCN for 2–3 days [RuClCp(dppf)] + Na[BPh4 ] in MeCN for 1 h [RuCl(HMB)(dppf)][PF6 ] + NH4 [PF6 ], in MeCN for 24 h [RuClCp(dppf)] + PMe3 + NH4 [PF6 ] in MeOH [RuClCp(dppf)] + PMe2 Ph + NH4 [PF6 ] in MeOH [RuClCp(dppf)] + NaS2 CNEt2 , in MeOH for 10 h [RuCl(HMB)(dppf)][PF6 ]c + NaS2 CNEt2 •3H2 O, in MeOH for 24 h [RuClCp(dppf)] + S8 + Na[BPh4 ] in CH2 Cl2 for 9 h [RuCl2 (1,2,4-C6 H3 Me3 )]2 + dppf, in MeOH for 4 h

15

Ref.∗

[RuCl2 (η6 -C6 H5 (CH2 )3 OH)]2 (µdppf) + FcCO2 H + DCC/DMAP/PPy in CH2 Cl2 at r.t. for 3 d

[RuCl2 {η6 -C6 H5 (CH2 )3 OH}]2 + dppf in CH2 Cl2 for 12 h

Substrates and preparative conditions

38 Ferrocenes: Ligands, Materials and Biomolecules

Yes Yes Yes Yes No No Yes No No No Yes No Yes

[RuH(Cl)(PCy3 )(dppf)]d

[Ru(RCOO)2 (dppf)] (R = Me, Et, Ph)

[Ru(PhCOO)2 (CH3 CN)(H2 O) (dppf)]

[RuCl(η3 -C3 H5 )(CO)(dppf)]

[RuBr(η3 -C3 H5 )(CO)(dppf)]

[RuX(η3 -2-MeC3 H4 )(CO) (dppf)] (X = Cl and Br)

[{RuCl(µ-Cl)(CO)(dppf)}2 ]

[{RuBr(µ-Br)(CO)(dppf)}2 ]

cis,cis,cis-[RuCl2 (CO)2 (dppf)]

cis,cis-[RuCl2 (CO)(NCCH2 Ph) (dppf)]

cis,cis-[RuCl2 (CO)(Py)(dppf)]e

cis,cis-[RuCl2 (CO)(PhNH2 ) (dppf)]

[RuCl2 (=CH2 )(C6 H5 )(dppf)]

C

C

C

C

C

C

C

C

C

C

C

C

C

20 20 20 20 20 20 20 20 20

[RuCl(η3 -C3 H5 )(CO)3 ] + dppf, in toluene for 3 h [RuBr(η3 -C3 H5 )(CO)3 ] + dppf, in toluene for 3 h [RuX(η3 -2-MeC3 H4 )(CO)3 ] + dppf, in THF for 7 h [RuCl(η3 -C3 H5 )(CO)(dppf)] + HCl in CH2 Cl2 at r.t. [RuBr(η3 -C3 H5 )(CO)(dppf)] + HBr in CH2 Cl2 at r.t. [{RuCl(µ-Cl)(CO)(dppf)}2 ] + CO, in THF for 5h [{RuCl(µ-Cl)(CO)(dppf)}2 ] + PhCH2 CN in CH2 Cl2 at r.t. for 2 h [{RuCl(µ-Cl)(CO)(dppf)}2 ] + Py in CH2 Cl2 at r.t. for 2 h [{RuCl(µ-Cl)(CO)(dppf)}2 ] + PhNH2 in CH2 Cl2 at r.t. for 2 h

(continued overleaf )

21

19

[Ru(PhCOO)2 (dppf)] + MeCN for 3 h at r.t.

[RuCl2 (=CH2 )(PCy3 )2 ] + dppf, in toluene

19

18

[Ru(RCOO)2 (PPh3 )2 ] + dppf in CH2 Cl2 for ∼1 h at r.t.

[RuH(Cl)(H2 )(PCy3 )2 ] + dppf in CH2 Cl2 at −78 ◦ C

The Coordination and Catalytic Chemistry of dppf 39

Group

Metal

C C

No No No No No No No Yes No Yes

[RuCl{HB(pz)}3 (dppf)]f

[Ru(CN-t -Bu){HB(pz)3 }(dppf)] [PF6 ]f

[RuH{HB(pz)3 }(dppf)]f

[Ru(=C=C=CPh2 ){HB(pz)3 } (dppf)][PF6 ]f

[Ru(=C=CHC6 H4 Me-4) {HB(pz)3 }(dppf)][PF6 ]f

[Ru(C≡CC6 H4 Me-4){HB(pz)3 } (dppf)]f

[RuCl2 (en)(dppf)]g

[RuCp(dppf)(SPh)]

[RuCp(dppf)(S(C5 H4 NH))] [BPh4 ]

[RuCp(dppf)(SC(NH2 )2 )][PF6 ]

C

C

C

C

C

C

C

C

C

Yes

[RuCl2 (=CHPh)(dppf)]

B

Coordination mode of dppfa

Yes

X-ray structure

[RuCl2 (η3 :η3 -C10 H16 )]2 (µ-dppf)

Compound

Table 2.1 (continued )

24 24 25 26 26 26

[Ru(=C=CHC6 H4 Me-4){HB(pz)3 }(dppf)][PF6 ] + NaOMe for 1 hr [RuCl2 (η6 -C6 H6 )]2 + dppf, in DMF for 10 min, then + en for 1 h [RuClCp(dppf)] + HSPh in EtOH at r.t. for 1 h [RuClCp(dppf)] + 2-mercatopyridine + NaBPh4 in MeOH at r.t. for 15 min [RuClCp(dppf)] + NH4 [PF6 ] in MeOH at r.t. for 30 min, then + NH2 C(S)NH2 for 3 h

24

24

[RuCl{HB(pz)3 }(dppf)] + Ag[PF6 ] + 4-MeC6 H4 C≡CH, in THF for 1 h

[RuCl{HB(pz)3 }(dppf)] + HC≡CCPh2 (OH) + Ag[PF6 ] in THF for 40 min

[RuCl{HB(pz)3 }(dppf)], in NaOMe for 4 h

24

24

[RuCl(PPh3 )2 {HB(pz)3 }] + dppf, in C6 H6 for 40 min [RuCl{HB(pz)3 }(dppf)] + Ag[PF6 ] + t -BuNC in CH2 Cl2 for 1 h

23

22

Ref.∗

[RuCl2 (=CHPh)(PPh3 )2 ] + dppf in CH2 Cl2 at −78◦ C for 20 min

[{RuCl(µ-Cl)(η3 :η3 -C10 H16 )}2 ] + dppf in CH2 Cl2 at r.t. for 5 min

Substrates and preparative conditions

40 Ferrocenes: Ligands, Materials and Biomolecules

Os

B B

C M

Yes No Yes Yes Yes Yes Yes Yes No

No

[RuCp(dppf)(SMe2 )][PF6 ]

[Ru2 (CO)4 (µ-PFu2 )(µ-η1 ,η2 Fu)(dppf)]n h

[Ru2 (CO)5 (µ-PFu2 )(µ-η1 ,η2 Fu)]2 (dppf)h

[Ru3 (µ3 -Se)2 (CO)7 (µ-dppf)]

[Ru3 Se(µ-dppf)(µ-OCPh) (CO)6 ]

[{Ru3 O(O2 CMe)6 (Py)2 }2 (dppf)] [PF6 ]2 e

[Ru3 (CO)11 ]2 (µ-dppf)

[Ru3 (CO)7 (µ3 -S)2 (µ-dppf)]

OsB5 H9 (PPh3 )2 (CO):[2,2,2(PPh3 )2 (CO)-nido-2OsB4 H7 -3-(BH2 •dppf)]

[2,2,2-(PPh3 )2 (CO)-nido-2OsB4 H7 -3-BH2 )2 (dppf)]

B

B

B

B

B

C

C

Yes

[RuCp(dppf)(SCS(CH2 )2 S)]Cl

C

Yes

[RuCp(dppf)(SCS(CH)2 S)]Cl

28

[Ru3 (CO)12 ] + dppfSe2 + Me3 NO, in toluene for 90 min

Ditto

32

32

(continued overleaf )

[OsB5 H9 (PPh3 )2 (CO)] + dppf in CH2 Cl2 , 5◦ C for 24 h

31

28

[Ru3 (CO)12 ] + dppfSe2 + Me3 NO in toluene, for 3 h

[Ru3 (CO)9 (µ3 -S)2 ] + dppf in toluene at r.t. for 2h

27

[Ru2 (CO)6 (µ-PFu2 )(µ-η1 ,η2 -Fu)] + dppf (2:1), in toluene for ∼1 h

30

27

[Ru2 (CO)6 (µ-PFu2 )(µ-η1 ,η2 -Fu)] + dppf (1:1), in toluene for ∼1 h

[Ru3 (CO)12 ] + dppf + Na/benzophenone ketyl in THF for ∼10 min

26

[RuClCp(dppf)] + [AuCl(SMe2 )] + NH4 [PF6 ] in Me2 CO at r.t. for 10 h

29

26

[RuClCp(dppf)] + ethylene trithiocarbonate in CH2 Cl2 /MeOH at r.t. for 30 min

[Ru3 O(O2 CMe)6 (Py)2 (CH3 OH)][PF6 ] + dppf in CH2 Cl2 for 2 d

26

[RuClCp(dppf)] + vinylene trithiocarbonate in CH2 Cl2 /MeOH at r.t. for 30 min

The Coordination and Catalytic Chemistry of dppf 41

Co

9

Rh

Metal

Group

No

[{Co2 (CO)4 }2 (µ-η1 :η1 -dppf)2 (η1 -µ-η1 -dppf)(µ-η2 :η2 FcC≡CH)2 ]

Yes

No

[CoH(dppf)2 ]

[Rh(dppf)2 ]

No

[Co(dppf)2 ]

No

No

[CoI(η5 -C5 H5 )(dppf)]I

[Rh(dppf)2 ][BF4 ]

No

[Os7 (CO)17 (µ4 -η2 -CO) (MeCN)(µ-dppf)]

No

Yes

[Os3 (µ-H)(CO)8 (µ-η2 -NO2 ) (µ-dppf)]

[Rh(O2 )(Qs )(dppf)]j

Yes

[OsCl(Cym)(dppf)][PF6 ]i

Yes

No

[2,2,2-(PPh3 )2 (CO)-nido-2OsB4 H7 -3-(BH2 •dppf•BH3 )]

[Rh(cod)(dppf)](OTf) •2CH2 Cl2 b,v

X-ray structure

Compound

C

C

C

C

B

C

C

C

C

C

C

M

Coordination mode of dppfa

Table 2.1 (continued )

41 41

[Na(THF)2 ][Rh(dppf)2 ] + [Rh(dppf) 2 ][BF4 ] in THF/toluene for 10 min

b

[RhCl(cod)]2 + Ag[BF4 ] in Me2 CO at r.t., then dppf and in CH2 Cl2 for 1.5 h

37

[CoCl2 (dppf)] + dppf + NaHgx in THF at r.t. for 30 min

40

37

[CoCl2 (dppf)] + dppf + sodium naphthalenide in THF at r.t. for 30 min

[Rh(cod)(Qs )] + dppf in Et2 O for ∼2 h

36

[CoI2 (η5 -C5 H5 )(CO)] + dppf in CH2 Cl2 at r.t. for 30 min

39

35

[Os7 (CO)19 (MeCN)2 ] + dppf in CH2 Cl2 at r.t. for 12 h

[RhCl(1,5-cod)]2 + AgOTf + dppf in THF for 2h

34

[Os3 (µ-H)(CO)10 (µ-η2 -NO2 )] + dppf + Me3 NO in CH2 Cl2 at r.t. for 12 h

38

33

[(Cym)OsCl2 ]2 + dppf, in CH3 OH/THF for 11 h

[Co2 (CO)6 (µ-η2 :η2 -FcC≡CH)] + dppf in CH2 Cl2 at r.t. for 12 h

32

Ref.∗

Ditto

Substrates and preparative conditions

42 Ferrocenes: Ligands, Materials and Biomolecules

Ir

C C

Yes No No Yes No

No No No Yes No Yes

[RhClCp*(dppf)][PF6 ]

[Rh(CH3 CN)Cp*(dppf)][PF6 ]2

[Rh(CNR)Cp*(dppf)][PF6 ]2 (R = Xylk , Mesl )

[Rh(4-MeC6 H4 CH2 NC)Cp* (dppf)][PF6 ]2

[Rh {3-((l )-PhCH(Me)NHCO) C6 H4 NC}Cp*(dppf)][PF6 ]2

[Rh(CO)Cp*(dppf)][PF6 ]2

[Ir(dppf)2 ][BF4 ]

[IrH(dppf(-H))(dppf)][BF4 ]m

[Ir(dppf)2 ]

[Na(THF)2 ][Ir(dppf)2 ]

[Na(THF)5 ][Ir(dppf)2 ]•THF

C

C

C

C

C

C

C

C

C

B

Yes

[RhCl2 Cp*]2 (µ-dppf)

C

Yes

[Na(THF)5 ][Rh(dppf)2 ]

42

42 43

[Rh(CH3 CN)Cp*(dppf)][PF6 ] + 3-((l )-PhCHMeNHCO)C6 H4 NC in CH2 Cl2 at r.t. for 2 h [Rh(CH3 CN)Cp*(dppf)][PF6 ] + CO (10 min) in CH2 Cl2 for 1 h [IrCl(C8 H14 )2 ]2 + Ag[BF4 ] in Me2 CO at r.t., then dppf in CH2 Cl2

43 (continued overleaf )

obtained from a saturated solution of the complex in toluene upon addition of THF

43

42

[Rh(CH3 CN)Cp*(dppf)][PF6 ] + 4-MeC6 H4 CH2 NC in CH2 Cl2 at r.t. for 2 h

[Ir(dppf)2 ][BF4 ] + sodium naphthalenide in THF

42

[Rh(MeCN)Cp*(dppf)][PF6 ] + RNC in CH2 Cl2 at r.t. for 2 h

43

42

[RhClCp*(dppf)][PF6 ] + Na[PF6 ] + AgNO3 in MeCN at r.t. for 3 h

[Ir(dppf)2 ][BF4 ] + [Na(THF)2 ][Ir(dppf)2 ] in THF/toluene

42

[RhCl2 Cp*]2 + dppf + Na[PF6 ] in CH2 C12 /Me2 CO at r.t. for 3 h

43

42

[RhCl2 Cp*]2 + dppf in CH2 Cl2 at r.t. for 2 h

[Ir(dppf)2 ][BF4 ], in CH2 Cl2 for 2 days

41

[Rh(dppf)2 ][BF4 ] + sodium naphthalenide (1:3) + Na in THF at r.t. for 1 h

The Coordination and Catalytic Chemistry of dppf 43

10

Group

Ni

Metal No No No No No No Yes No No Yes Yes No

[Ni(bzi prdtc)(dppf)]Xn (X = I, ClO4 , NCS)

[Ni(but2 dtc)(dppf)]ClO4 •H2 Oo

[Ni(but2 dtc)(dppf)]Io

[Ni(plddtc)(dppf)]Xp (X = I, ClO4 )

[Ni(plddtc)(dppf)]Br•H2 Op

[Ni(tzdtc)(dppf)]Xq (X = I, ClO4 )

[Ni(hmidtc)(dppf)]ClO4 r

[Ni(hmidtc)(dppf)]Ir

[Ni(pipdtc)(dppf)]Xs (X = I, NCS)

[Ni(pipdtc)(dppf)]ClO4s

[Ni(pe2 dtc)(dppf)]ClO4 t

[Ni(bz2 dtc)(dppf)]ClO4 u

C

C

C

C

C

C

C

C

C

C

C

C

Coordination mode of dppfa

Table 2.1 (continued ) X-ray structure

Compound

44

[Ni(hmidtc)2 ] + Ni(ClO4 )2 •nH2 O + dppf, in EtOH for 5–6 h [Ni(hmidtc)2 ] + NiI2 •nH2 O + dppf, in EtOH for 5–6 h

45 45 46 46

[Ni(pipdtc)2 ] + NiX2 •2H2 O + dppf, in EtOH for 5–6 h [Ni(pipdtc)2 ] + Ni(ClO4 )2 •nH2 O + dppf, in EtOH for 5–6 h [Ni(pe2 dtc)2 ] + dppf + Ni(ClO4 )2 •nH2 O, in MeOH for 10 h [Ni(bz2 dtc)2 ] + dppf +Ni(ClO4 )2 •nH2 O, in MeOH for 10 h

44

44

[Ni(tzdtc)2 ] + NiX2 •nH2 O + dppf, in in EtOH for 5–6 h

44

[Ni(plddtc)2 ] + NiBr2 •nH2 O + dppf, in EtOH for 5–6 h

44

[Ni(but2 dtc)2 ] + NiI2 •nH2 O + dppf, in EtOH for 5–6 h

44

44

[Ni(but2 dtc)2 ] + Ni(ClO4 )2 •nH2 O + dppf, in EtOH for 5–6 h

[Ni(plddtc)2 ] + NiX2 •nH2 O + dppf, in EtOH for 5–6 h

44

Ref.∗

[Ni(bzi prdtc)2 ] + NiX2 •nH2 O + dppf, in EtOH for 5–6 h

Substrates and preparative conditions

44 Ferrocenes: Ligands, Materials and Biomolecules

Pd

C C C

No No No

No No No

No

[Pd(CH2 CHCH(OCH3 ))(dppf)] (FSO3 )

[Pd(CH2 CHCH(OCH3 ))(dppf)] (CH3 SO3 )

[Pd(CH2 CHCH(OCH3 ))(dppf)] (4-NO2 C6 H4 SO3 )

[Pd(CH2 CHCH(OCH3 ))(dppf)] [BF4 ]

[Pd(η2 -C70 )(dppf)]

[(Pd(dppf))6 (µ-L1 )4 (OTf)12 ]v

[(Pd(dppf))6 (µ-L2 )4 (OTf)12 ]v

C

C

C

C

C

Yes

[Pd(CH2 CHCH(OCH3 ))(dppf)] (OTf)v

C

C

No

Yes

[Ni(dppf)2 ][PF6 ]

C

[PdMe(P(O)(OPh)2 )(dppf)]

No

[Ni(dppf)2 ]

48 48 48

48 49 50

50

[Pd(CH2 =CHCHO)(dppf)] + CH3 O3 SF in CD2 Cl2 at r.t. [Pd(CH2 =CHCHO)(dppf)] + CH3 O3 SCH3 in C2 D4 Cl2 at r.t. for 3 days [Pd(CH2 =CHCHO)(dppf)] + CH3 O3 SC6 H4 NO2 -4 in CD2 Cl2 at r.t. for 12 h [Pd(CH2 =CHCHO)(dppf)] + [(CH3 )3 O][BF4 ] in CD2 Cl2 at r.t. for 3 h [Pd(PPh3 )4 ] + C70 , then + dppf; in C6 H5 Cl at r.t. for 1.5 h [Pd(H2 O)2 (OTf)2 (dppf)] + 1,3,5-tris[(pyrid-4-yl)ethynyl]benzene (L1 ) in CH2 Cl2 for 8 days [Pd(H2 O)2 (OTf)2 (dppf)] + 1,3,5-tris[(pyrid-4-yl)ethenyl]benzene (L2 ) in CH2 Cl2 for 24 h

(continued overleaf )

48

[Pd(CH2 =CHCHO)(dppf)] + TfOMe in toluene at r.t.

47

38

[Ni(dppf)2 ] + [Fe(C5 H5 )2 ][PF6 ] in THF at r.t. for 12 h [PdMe{4,4 -(t -Bu)2 bipy}(P(O)(OPh)2 )] + dppf in CH2 Cl2 at −43◦ C for 24 h

38

[NiCl2 (dppf)] + dppf + sodium napthalenide in toluene at r.t. for 30 min

The Coordination and Catalytic Chemistry of dppf 45

Group

Metal

No

No

No No Yes Yes

No

No No No

[Pd(C2 O4 )(dppf)]

[Pd{(OOC)2 CH2 }(dppf)]

[{Pd(µ-OH)(dppf)}2 ](NO3 )2

[PdCl(NO3 )(dppf)]•CHCl3

[PdCl{5-(CHO)C6 H3 C(H)=NCy-κ 2 C ,N }]2 (µ-dppf)d

[Pd{5-(CHO)C6 H3 C(H)=NCyκ 2 C ,N }(dppf)][PF6 ]d

[Pd{η3 -CH(CO2 Me)COCH (CO2 Me)}(dppf)]

[PdCl{4-(CHO)C6 H3 C(H)=N(Cy)κ 2 C ,N }]2 (µ-dppf)d

[Pd{4-(CHO)C6 H3 C(H)=N(Cy)κ 2 C ,N }(dppf)][PF6 ]d

C

B

C

C

B

C

C

C

C

C

Coordination mode of dppfa

Table 2.1 (continued ) X-ray structure

[(Pd(dppf))6 (µ-L3 )4 (OTf)12 ]v

Compound

[Pd{4-(CHO)C6 H3 C(H)=N(Cy)-κ 2 C ,N }(µCl)]2 + dppf in Me2 CO at r.t. for 2 h, then NH4 [PF6 ]

54

54

53

[Pd2 (dba)3 ]•CHCl3 bb + dppf + 3-oxopentanedioic acid dimethyl ester for 48 h [Pd{4-(CHO)C6 H3 C(H)=N(Cy)-κ 2 C ,N } (µ-Cl)]2 + dppf in Me2 CO for 12 h at r.t.

52

52

[Pd{5-(CHO)C6 H3 C(H)=NCy-κ 2 C ,N }(µ-Cl)]2 + dppf in Me2 CO at r.t. for 2 h, then NH4 [PF6 ]

[Pd{5-(CHO)C6 H3 C(H)=NCy-κ 2 C , N }(µ-Cl)]2 + dppf in Me2 CO for 12 h at r.t.

51

51

[PdCl2 (dppf)] + AgNO3 in H2 O at 80◦ C for 2 h [PdCl2 (dppf)] + AgNO3 (1:1) in H2 O at 80◦ C for 2 h

51

51, 55

50

Ref.∗

[PdCl2 (dppf)] + CH2 (CO2 K)2 , in H2 O for ∼3 h

[PdCl2 (dppf)] + (CO2 K)2 , in H2 O for ∼3 h; or [PdBr2 (dppf)] + (CO2 Ag)2 in CH2 Cl2 at r.t. for 20 h

[Pd(H2 O)2 (OTf)2 (dppf)] + 2,4,6-tris(4-pyridyl)-1,3,5-triazine (L3 ) in CH2 Cl2 for 24 h

Substrates and preparative conditions

46 Ferrocenes: Ligands, Materials and Biomolecules

C B

Yes No No Yes No Yes Yes No No

No

[Pd(O2 C(CF2 )2 CF3 )2 (dppf)]

[Pd(O2 CC6 H4 Cl-4)2 (dppf)]

[Pd(O2 CPh)2 (dppf)]

[Pd(O2 CCHCl2 )2 (dppf)]

[Pd(O2 CCH2 CO2 H)2 (dppf)]

[Pd(O2 CCH=CHCO2 H)2 (dppf)]

[PdCl(O2 CPh)(dppf)]

[PdCl(O2 CCF3 )(dppf)]

[PdCl{2,3,4-(MeO)3 C6 HC(H)=NCH2 CH2 OH}]2 (µ-dppf)

[Pd(H2 O)2 (dppf)](OTs)2 w

C

C

C

C

C

C

C

C

C

Yes

[Pd(O2 CCF2 CF3 )2 (dppf)]

C

Yes

[Pd(O2 CCF3 )2 (dppf)]

58

[PdCl2 (dppf)] + AgOTs in CH2 Cl2 for ∼12 h at r.t.

(continued overleaf )

57

[Pd{2,3,4-(MeO)3 C6 HC(H)=NCH2 CH2 OH} (µ-Cl)]2 + dppf in Me2 CO for 12 h

55

Pd(OAc)2 + dppf + CH2 (CO2 H)2 in MeOH/Et2 O for ∼5 h

55

55

[PdCl2 (dppf)] + NaO2 CCHCl2 + AgNO3 in CH2 Cl2 at r.t. for ∼16 h

[Pd(µ-Cl)(dppf)]2[BF4 ]2 + NaO2 CCF3 in CH2 Cl2 for ∼20 h

55

[PdCl2 (dppf)] + AgO2 CPh in CH2 Cl at r.t. for 20 h

55

55

[PdBr2 (dppf)] + AgO2 CC6 H4 Cl-4 in CH2 Cl2 at r.t. for 20 h

[PdCl2 (dppf)] + AgO2 CPh in CH2 Cl2 for ∼20 h

55

[PdCl2 (dppf)] + AgO2 C(CF2 )2 CF3 in CH2 C12 at r.t. for 6 h

55

56

[PdCl2 (dppf)] + AgO2 CCF2 CF3 in CH2 Cl2 at r.t. for 20 h

Pd(OAc)2 + dppf + maleic acid in MeOH/Et2 O for ∼12 h

55

[PdBr2 (dppf)] + AgO2 CCF3 in CH2 Cl2 at r.t. for ∼20 h

The Coordination and Catalytic Chemistry of dppf 47

Group

Metal

No No

No

No

Yes No No

No

[Pd(OAc)2 (dppf)]

[Pd{2,3,4-(MeO)3 C6 HC(H)=N(2-(O)C6 H4 )}]2 (µ-dppf)

[Pd{2,3,4-(MeO)3 C6 HC(H)=N(2-(O)-4MeC6 H3 )]2 (µ-dppf)

[PdCl{2-ClC6 H3 C(H)=NCH2 CH2 SMe}]2 (µ-dppf)

[Pd(9S3)(dppf)][PF6 ]2 •CH3 NO2 x

[PdCl{(η5 -C5 H5 )Fe(η5 -C5 H3 ) C(H)=NMes}]2 (µ-dppf)l

[Pd{(η5 -C5 H5 )Fe(η5 -C5 H3 ) C(H)=NMes}(dppf)][PF6 ]l

[Pd(LNC )(NCO)]2 (µ-dppf)y No Yes

[Pd(L ) (NCO)(dppf)]•CH2 Cl2 y

[Pd(C5 H9 )(dppf)][BF4 ]z

NC

No

C

C

B

C

B

C

B

B

B

C

C

Coordination mode of dppfa

Table 2.1 (continued ) X-ray structure

[Pd(NCMe)2 (dppf)][BF4 ]2

Compound

59

60

61 62 62

63

[Pd{2,3,4-(MeO)3 C6 HC(H)=N(2-(O)-4MeC6 H3 )}] + dppf in CH2 Cl2 for 12 h [PdCl{2-ClC6 H3 C(H)=NCH2 CH2 SMe}] + dppf in Me2 CO for 12 h [PdCl2 (dppf)]•CH2 Cl2 + 9S3 + in MeOH for 1 h, then NH4 [PF6 ] and for 30 min [Pd{(η5 -C5 H5 )Fe(η5 -C5 H3 )C(H)=NMes} (µ-Cl)]2 + dppf in Me2 CO for 24 h at r.t. [Pd{(η5 -C5 H5 )Fe(η5 -C5 H3 )C(H)=NMes} (µ-O2 CMe)]2 + dppf in Me2 CO for 2 h at r.t., then NH4 [PF6 ] for 2 h [Pd(LNC )(µ-NCO)]2 + dppf in CH2 Cl2 for 2h

[(C5 H9 )Pd(µ-Cl)]2 + dppf + Ag[BF4 ] in CH2 Cl2

64

63

59

[Pd{2,3,4-(MeO)3 C6 HC(H)=N(2-(O)C6 H4 )}]n + dppf in Me2 CO for 12 h

)(µ-NCO)]2 + dppf (2:1) in CH2 Cl2 for

58

Pd(OAc)2 + dppf in C6 H6 for ∼12 h

[Pd(L 2h

58

[PdCl2 (dppf)] + Ag[BF4 ] in CH2 Cl2 /MeCN at −78◦ C, then at r.t. for 1 h

NC

Ref.∗

Substrates and preparative conditions

48 Ferrocenes: Ligands, Materials and Biomolecules

No No No Yes No No

No No Yes Yes No

Yes

[PdI(PhCH2 OH)(dppf)]

[Pd(OCH2 Ph)(dppf)]

[ClPdN(Cy)=C(H)C6 H2 C(H)=N(Cy)PdCl]2 (µ-dppf)2 d

[BrPdN(Cy)=C(H)C6 H2 C(H)=N(Cy)Pd(Br)]2 (µ-dppf)2 d

[PdCl{o-C6 H4 C=NC(Me)=C(Me) NMe}]2 (µ-dppf)]

[(Pd{o-C6 H4 C=NC(Me)=C(Me) NMe})(dppf)]ClO4

[PdBr(H2 DPP)(dppf)]aa

[Pd(η2 -C60 )(dppf)]

[Pd(η2 -C60 )(dppf)]

[PdCl(LNC )]2 (µ-dppf)y

[Pd(LNC )(dppf)][PF6 ]y

[PdCl(C6 H4 CH2 NHMe2 -κ C ) (dppf)][PF6 ]

C

C

B

C

C

C

C

B

B

B

C

C

66

[ClPdN(Cy)=C(H)C6 H2 C(H)=N(Cy)}PdCl]n + dppf in CH2 Cl2 at r.t. for 24 h

68 69 49 70 70

70

[Pd2 (dba)3 ]bb + dppf, in toluene for 10 min; then (H2 DPP)Br and for 2.5 h [Pd2 (dba)3 ]•C6 H6 bb + C60 + dppf in at r.t. in o-xylene [Pd(PPh3 )4 ] + C60 , then dppf in toluene at r.t. for 1 h [Pd(LNC )(µ-Cl)]2 + dppf in CH2 Cl2 at r.t. for 1 h [PdCl(LNC )]2 (µ-dppf)•CH2 Cl2 + dppf in CH2 Cl2 /Me2 CO, then Na[PF6 ] at r.t. for 2h [Pd(LNC )(dppf)][PF6 ]y + aq. HCl in Me2 CO at r.t. for 1 h

(continued overleaf )

67

67

[Pd{o-C6 H4 C=NC(Me)=C(Me)NMe}(µCl)]2 + dppf in Me2 CO for 2 h at r.t., then NaClO4 for 1 h

[Pd{o-C6 H4 C=NC(Me)=C(Me)NMe} (µ-Cl)]2 + dppf in Me2 CO for 12 h at r.t.

66

65

[Pd(OCH2 Ph)PPh3 ]2 + dppf in CH2 Cl2 at r.t. for 2 h

[BrPdN(Cy)=C(H)C6 H2 C(H)=N(Cy)PdBr]n + dppf in CH2 Cl2 at r.t. for 24 h

65

[PdI(PhCH2 OH)(PPh3 )2 ] + dppf in CH2 Cl2 at r.t. for 1 h

The Coordination and Catalytic Chemistry of dppf 49

Group

Metal

X-ray structure No Yes Yes No

Yes No

Yes No Yes Yes Yes

Compound

[Pd(H)(PCy3 )(dppf)](CF3 CO2 )d

[Pd(dppf)(SC6 F4 Y-4)2 ] (Y = F, H)

[Pd(dppf)(SC6 H4 X)2 ] (X = 2-CF3 , 3-F, 4-F)

[PdBr(dppf)(NS)]

[Pd(MeC(O)S-κS)2 (dppf)]

[Pd2 {1,3[C(H)=NCH2 C4 H7 O]2 C6 H2 }(µ-dppf)]

[Pd2 (dppf)2 (XylNC)2 ][PF6 ]2 k

[Pd2 (dppf)2 (MesNC)2 ][PF6 ]2 l

[Pd3 Cl2 (η2 -dppf)(µ-dppf) (µ3 -S)2 ]

[Pd3 Cl(η2 -dppf)2 (PPh3 ) (µ3 -S)2 ]Cl

[Pd3 Cl(η2 -dppf)2 (PPh3 ) (µ3 -S)2 ]NO3

C

C

C/B

C

C

B

C

C

C

C

C

Coordination mode of dppfa

Table 2.1 (continued ) Ref.∗ 71 72 72 73

74 75

76 76 77 77 77

Substrates and preparative conditions [Pd(PCy3 )2 ] + dppf in toluene-d8 at r.t. for 1 h, then CF3 CO2 H for 1 h [PdCl2 (dppf)] in CH2 Cl2 + Pb(SC6 F4 Y-4)2 in Me2 CO for 12 h [PdCl2 (dppf)] in CH2 Cl2 + Pb(SC6 H4 X)2 in Me2 CO for 12 h [Pd2 (dba)3 ]•CHCl3 bb + dppf in CH2 Cl2 for 0.2 h, then N -bromosuccinimide (NSBr) in CH2 Cl2 [PdBr2 (dppf)] + MeC(O)SK in CH2 Cl2 at r.t. for 20 h [Pd2 {1,3-[C(H)=NCH2 C4 H7 O]2 C6 H2 } (µ-Cl)2 ]2 + dppf in Me2 CO for 1 h [Pd2 (XylNC)6 ][PF6 ]2 + dppf in CH2 Cl2 at r.t. for 12 h [Pd2 (MesNC)6 ][PF6 ]2 + dppf in CH2 Cl2 at r.t. for 12 h [Pd2 (dppf)2 (µ-S)2 ] + [PdCl2 (PPh3 )2 ] in THF at r.t. for 1 day [Pd2 (dppf)2 (µ-S)2 ] + [PdCl2 (PPh3 )2 ] in MeOH at r.t. for a day [Pd2 (dppf)2 (µ-S)2 ] + [Cu(NO3 )(PPh3 )2 ] in THF at r.t. for a day

50 Ferrocenes: Ligands, Materials and Biomolecules

Pt

C C

No

Yes No Yes No No

No

[Pt(C6 H4 NMe2 -4)(C6 H4 X4)(dppf)] (X = H, Me, OMe, Cl, and CF3 )

[Pt(CH2 CHCH(OCH3 )) (dppf)](OTf)

[Pt(CH2 CHCH(OCH3 ))(dppf)] (p-NO2 C6 H4 SO3 )

[Pt(η2 -C60 )(dppf)]

[Pt(η2 -C70 )(dppf)]

[(Pt(dppf))6 (µ-L1 )4 (OTf)12 ]v

[(Pt(dppf))6 (µ-L2 )4 (OTf)12 ]v

C

C

C

C

C

C

No

[Pt(C6 H4 X-4)(C6 H4 Y-4)(dppf)] (X/Y = Me/MeO, Me/H, Me/Cl, Me/F, Me/CF3 ; OMe/H, OMe/Cl, OMe/F, OMe/CF3 , CF3 /H, and CF3 /Cl)

C

No

[Pt(C6 H4 X-4)2 (dppf)] (X = H, Me, OMe, Cl, CF3 , and NMe2 ) 78

78

48 48 49 49 50

50

[Pt(cod)(C6 H4 X-4)(C6 H4 Y-4)]b + dppf in C6 H6 at r.t. for 1 h

[PtCl(cod)(C6 H4 X-4)]b + dppf, in C6 H6 for 30 min; then 4-NMe2 C6 H4 MgBr in toluene for 2 h at r.t. [Pt(CH2 =CH2 )(dppf)] + CH2 =CHCHO + TfOMe in C6 H6 at r.t. [Pt(CH2 =CHCHO)(dppf)] + 4-O2 NC6 H4 SO3 CH3 in CD2 Cl2 at r.t. [Pt(PPh3 )4 ] + C60 in toluene at r.t. for 0.5 h, then dppf for 0.5 h [Pt(PPh3 )4 ] + C70 in PhCl at r.t. for 1 h, then dppf for 0.5 h

cis-[Pt(H2 O)2 (OTf)2 (dppf)] + 1,3,5-tris[(pyrid-4-yl)ethynyl]benzene (L1 ) in CH2 Cl2 for 8 d [Pt(H2 O)2 (OTf)2 (dppf)] + 1,3,5-tris[(pyrid-4-yl)ethenyl]benzene (L2 ), in MeNO2 for 6 d

(continued overleaf )

78

[Pt(cod)(C6 H4 NX-4)2 ]b + dppf in C6 H6 at r.t. for 1 h

The Coordination and Catalytic Chemistry of dppf 51

Group

Metal

No

No No

No No

No Yes Yes Yes

[Pt{(O2 C)2 CCH2 CH2 CH2 } (dppf)]

[Pt{(O2 C)2 fc}(dppf)]

[Pt(O2 CR)2 (dppf)] (R = t -Bu, Cyd )

[{Pt(µ-OH)(dppf)}2 ](NO3 )2

[Pt{OC(=NO)C6 H4 OH} (dppf)]

[Pt{OC6 H4 C(O)NOH}(dppf)]

[Pt(C≡CPh)2 (dppf)]

[Pt(9S3)(dppf)][PF6 ]2 •CH3 NO2 x

[Pt(dmit)(dppf)]•1.5CHCl3 cc

(dppf)]

C

C

C

C

C

C

C

C

C

No

[PT{(O2C)2CCH2CH2CH2}

C C

Yes

[Pt(C2 O4 )(dppf)]

C

Coordination mode of dppfa

No

X-ray structure

[(Pt(dppf))6 (µ-L3 )4 (OTf)12 ]v

Compound

Table 2.1 (continued )

79

79

[PtCl2 + dppf in MeOH + salicylhydroxamic acid + Me3 N, in MeOH for 2.5 h [Pt{OC6 H4 C(O)NOH}(cod)]b + dppf in CH2 Cl2

61 81

[PtCl2 (9S3)] + dppf, in MeNO2 for 3 h; then NH4 [PF6 ] and for 30 min [PtCl2 (dppf)] + Na2 (dmit) in CHCl3 at r.t. for 1h

[PtCl2 (dppf)] + HC≡CPh + CuI in (i −Pr)2 NH/CH2 Cl2 at r.t. overnight

80,82

51

[PtCl2 (dppf)] + AgNO3 + in H2 O for ∼2 h (cod)]b

51

[PtCl2 (DMSO)2 ] + AgNO3 , in H2 O for ∼2 h; then (i) RCO2 K in H2 O, and (ii) dppf in CHCl3

51

[PtCl2 (dppf)] + fc(CO2 Ag)2 , in acetone for ∼2 h

51

51

51

[PtCl2 (dppf)] + (CO2 K)2 , in H2 O for ∼3 h [PtCl2 (dppf)] + CH2 (CO2 K)2 , in H2 O for ∼3 h [PtCl2 (dppf)] + dipotassium cyclobutane-1,1-dicarboxylate; in H2 O for ∼3 h

50

Ref.∗

[Pt(H2 O)2 (OTf)2 (dppf)] + 2,4,6-tris(4-pyridyl)-1,3,5-triazine (L3 ), in MeNO2 for 8 d

Substrates and preparative conditions

52 Ferrocenes: Ligands, Materials and Biomolecules

C

B

C C

No No Yes Yes

No

Yes No No No Yes Yes

[Pt(mtdt)(dppf)]gg

[Pt(i-mnt)(dppf)]hh

[Pt(η2 -PhC≡CC≡CPh)(dppf)]

[Pt{3-CH3 (CH2 )5 OC6 H3 C(Me)=NN=C(S)NH2 }]2 (µ-dppf)

[PtCl(Me)(dppf)]

[PtMe(Me2 SO)(dppf)](OTf)v

[{PtMe(dppf)}4 (TpyP)](OTf)4 ii,v

[Pt2 (µ-C8 H4 S2 )(MeCN)2 (dppf)2 ](TfO)2 jj,v

[Pt2 (dppf)2 (µ2 -η1 (C), η1 (S)-C4 H3 S)2 ](TfO)2 kk,v [Pt4 (µ2 -isonic)4 (dppf)4 ](OTf)4 jj,v

[Pt(dphdt)(dppf)]

ff

C

C

C

C

C

C

C

C

Yes

[Pt(phdt)(dppf)]ee

C

No

[Pt(dddt)(dppf)]dd

81 81 81

[PtCl2 (dppf)] + K2 (dphdt) in CHCl3 at r.t. for 3h [PtCl2 (dppf)] + K2 (mtdt) in CHCl3 at r.t. for 3 h [PtCl2 (dppf)] + K2 (i-mnt) in CHCl3 at r.t. for 3h

84 84 84 85 85

trans-[PtCl(Me)(DMSO)2 ] + dppf in CH2 Cl2 for 1 h [PtCl(Me)(dppf)] + TfOAg in CH2 Cl2 /DMSO for 1 h [PtMe(DMSO)(dppf)](OTf) + TpyP in CH2 Cl2 at r.t. for 12 h [Pt2 Br2 (µ2 -C8 H4 S2 )(dppf)2 ] + TfOAg in MeCN/Me2 CO [PtBr(dppf)(C4 H3 S)] + TfOAg in MeCN/CHCl3 at r.t. for 30 min [Pt2 Br2 (µ-C8 H4 S2 )(dppf)2 ] + TfOAg in MeCN/ CHCl3 at r.t. for 45 min, then + Hisonic+ for 5 h

(continued overleaf )

85

83

[Pt{3-CH3 (CH2 )5 OC6 H3 C(Me) =NN=C(S)NH2 }]4 + dppf in acetone for 4 h

82

81

[PtCl2 (dppf)] + K2 (phdt) in CHCl3 at r.t. for 3 h

[Pt(cod)2 ] in petroleum ether at 0◦ C, dppf in Et2 O at r.t. for 1 h, then 1,4-diphenylbutadiyne for 1 h

81

[PtCl2 (dppf)] + K2 (dddt) + in CHCl3 at r.t. for 3h

The Coordination and Catalytic Chemistry of dppf 53

11

Ag

Cu

Group Metal

C C

Yes Yes Yes

Yes Yes Yes Yes Yes No No Yes Yes

[CuI(dppf)]2

[Cu(dppf)(dppfO)][PF6 ]

[Cu(dppf) (µ-dppf)Cu(dppf)][PF6]2 • 0.75H2 O

[Cu{S2 C(t −Bu-Hfy)}(dppf)]nn

[Cu(µ-Cl)](dppf)]2

[Cu2 {[SC=(t -Bufy)]2 S}(dppf)]nn

[Cu2 (NCS)2 (dppf)2 ]•2CH2 Cl2

[Cu2 (dppa)(dppf)2 ][BF4 ]2 oo

[Cu2 (dpbp)(dppf)2 ][BF4 ]2 pp

[Cu4 {S2 C=(t -Bufy)}2 (dppf)2 ]nn

[AgX(dppf)]2 •3pipqq (X = Cl, Br)

[AgX(dppf)]2 (X = I, SCN, NCO, CN)

C

C

C

C

C

C

C

B

C

C

C

Yes

[Cu(PCHO) (dppf)][BF4 ]mm

C

Yes

X-ray Coordination structure mode of dppfa

[{Cu(dppf)}3 {hat(CN)6 }] [PF6 ]2 ll

Compound

AgX + dppf in MeCN at r.t. for 24 h

88

88

93

[Cu2 (dpbp)2 (NCMe)4 ][BF4 ]2 + dppf in CH2 Cl2 for 6h

AgX + dppf in 1:1 MeCN/pip at r.t. for 24 h

92

[Cu2 (dppa)3 (CH3 CN)2 ][BF4 ]2 + dppf in CH2 Cl2 at r.t. for 8 h

90

91

CuSCN/NaNCS + dppf in CH2 Cl2 at r.t. for several days

[Cu(NCMe)4 ][PF6 ] + (pipH)[(t -Bu-Hfy)CS2]qq + dppf for 1 h

90

[Cu{S2 C(t -Bu-Hfy)}(dppf)] in MeCN + NEt3 for 24 h

90

89

[Cu(NCMe)4 ][PF6 ] in EtOH + dppf, in toluene for 48 h

89

89

[Cu(NCMe)4 ][PF6 ] in EtOH + dppf, in toluene for 2 h

[Cu{S2 C(t -Bu-Hfy)}]n + dppf in CH2 Cl2 for 1 h

88

CuI + dppf in CH2 Cl2 at r.t. for 2 h

[Cu(NCMe)4 ][PF6 ] + dppf in CH2 Cl2 at r.t.for 48 h

87

[Cu(PCHO)2 (NCMe)][BF4 ] + dppf, in THF for 40 min

Ref.∗ 86

Substrates and preparative conditions [Cu(CH3 CN)4 ][PF6 ] + dppf in CH2 Cl2 + hat(CN)6 in Me2 CO/CH2 Cl2

Table 2.1 (continued ) 54 Ferrocenes: Ligands, Materials and Biomolecules

Yes Yes Yes Yes No Yes No Yes No No No

[Ag2 {H2 B(btz)2 }2 (dppf)]rr

[Ag{H2 B(tz)2 }(dppf)]ss

[Ag(O2 CCF3 )(dppf)]n

[Ag(dppf)(PMe2 Ph)2 ][PF6 ]

[Ag(dppf)](CF3 SO3 )

[Ag(dppf)2 ](CF3 SO3 )

[Ag(NO2 )(dppf)]

[Ag{H2 B(Pz)2 }(dppf)]f

[Ag{H2 B(m2 pz)2 }(dppf)]tt

[Ag{B(Pz)4 }(dppf)]f

}(dppf)]f No No

[Ag{B(mpz)4 }(dppf)]uu

[Ag(Tm)(dppf)]vv

[Ag{B(Pz)4

B

Yes

[Ag(O2 CCF3 )(dppf)]∞ , H2 O/MeCN solvate

C

C

C

C

C

C

C

C

C

C

B

C

C

C

Yes

[Ag(O2 CCH3 )(dppf)]2

C

Yes

[Ag(NO3 )(dppf)]2 •3H2 O•5 MeCN 88

Ag(O2 CCF3 ) + dppf in MeCN at r.t for 24 h

96 97 97

[Ag2 (O2 CCF3 )2 (dppf)] + PMe2 Ph in MeOH at r.t. for 1 h, then NH4 PF6 at r.t. for 30 min CF3 SO3 Ag in CH2 Cl2 + dppf in toluene at r.t. for 8h CF3 SO3 Ag in MeOH + dppf in THF at r.t. for 8 h

100

AgNO3 + dppf + K(Tm) in MeOH at r.t. for 12 h

(continued overleaf )

99

99

99

99

99

AgNO3 + dppf + K[B(mpz)4 ] in MeOH at r.t. for 1h

AgNO3 + dppf + K[B(pz)4 ] (2:1:2) in MeOH at r.t. for 1 h

AgNO3 + dppf + K[B(pz)4 ]] in MeOH at r.t. for 1 h

AgNO3 + dppf + K[H2 B(m2 pz)2 ] in MeOH at r.t. for 1 h

AgNO3 + dppf + K[H2 B(Pz)2 ] in MeOH at r.t. for 1h

98

96

Ag(O2 CCF3 ) + dppf in CH2 Cl2 at r.t. for 3.5 h

AgNO2 + dppf in MeOH for 48 h at 40◦ C, then for 2 h at r.t.

95

AgNO3 + K[H2 B(tz)2 ] + dppf in MeOH at r.t. for 2h

94

88

AgO2 CCH3 + dppf in MeCN at r.t for 24 h

AgNO3 + K[H2 B(btz)2 ] + dppf in MeOH at r.t. for 3h

88

AgNO3 + dppf in MeCN at r.t for 24 h

The Coordination and Catalytic Chemistry of dppf 55

Au

Metal

B C

Yes Yes Yes No No No

[Ag4 (O2 CCF3 )4 (dppf)2 ]

[Au2 (µ-dppf)(C≡CCH2 OC6 H4 )2 (SO2 )]

[4-BrC6 H4 CH(4-C6 H4 OCH2 C≡CAu)2 (µ-dppf)]

[Au2 {(C≡CCH2 OC6 H4 )2 CMe2 }(µ-dppf)]

[Au2 (SeC2 B10 H11 )2 (µ-dppf)] B

B

B

B

B

[Ag2 (O2 CCF3 )2 (dppf)]xx 2n

B

Yes Yes

[Ag2 (O2 CCF3 )2 (dppf)2 ]

B

Yes

Coordination mode of dppfa

Table 2.1 (continued ) X-ray structure

[Ag2 (O2 CCF3 )2 (dppf)]ww n

[Ag2 (O2 CCF3 )2 (dppf)]

Compound

97 101 102

AgO2 CCF3 + dppf in CHCl3 at r.t. for 20 h [Au2 (C≡CCH2 O)2 (SO2 )(C6 H4 )]n + dppf in CH2 Cl2 at r.t. for 3 h [(4-C6 H4 OCH2 C≡CAu)2 (CHC6 H4 Br-4)] + dppf in CH2 Cl2 at r.t. for 2 h

104

97

AgO2 CCF3 + dppf in CHCl3 at r.t. for 1.5 h

[(B10 H11 C2 )SeH] + Na2 CO3 + [Au2 Cl2 (µ-dppf)] in CH2 Cl2 for 30 min

97

AgO2 CCF3 + dppf in CHCl3 at r.t. for 1.5 h

103

96

AgO2 CCF3 + dppf in CH2 Cl2 at r.t. for 3 h

[CMe2 (C6 H4 OCH2 C≡CAu)2 ]n + dppf in CH2 Cl2 at r.t. for 3 h

96

Ref.∗

AgO2 CCF3 in MeOH and dppf in THF at r.t. for 8 h

Substrates and preparative conditions

Fe(C5 H4 PPh2 )(C5 H4 PPhC6 H4 ); n bzi prdtc = (iso-propyl)(benzyl)carbamodithioato-κ 2 S ,S ; o but2 dtc = dibutylcarbamodithioato-κ 2 S ,S ; p pldtdtc = 1-pyrrolidinecarbodithioatoκ 2 S 1,S 1 ; q tzdtc = 3-thiazolidinecarbodithioato-κ 2 S 3,S 3 ; r hmidtc = azepane-1-carbodithioato-κ 2 S 1,S 1 ; s pipdtc = 1-piperidinecarbodithioato-κ 2 S 1,S 1 ; t pedtc = dipentylcarbamodithioato-κ 2 S ,S ; u bz2 dtc = dibenzylcarbamodithioato-κ 2 S ,S ; v TfO− = triﬂuoromethansulfonate(1-); w TsO− = 4-toluenesulfonate(1−); x 9S3 = 1,4,7-trithiacyclononane; y LNC = 2-(N ,N -dimethylaminomethyl)phenyl-κ 2 C ,N ; z C H = (1,2,3-η)-3-methyl-2-butenyl; aa H DPP = 5,15-diphenylporphyrin; bb dba = dibenzoylacetone; cc dmit = 1,35 9 2 dithiole-2-thione-4,5-dithiolate; dd dddt = 5,6-dihydro-1,4-dithiin-2,3-dithiolate; ee phdt = 6-hydro-5-phenyl-1,4-dithiin-2,3-dithiolate; ff dphdt = 5,6-diphenyl-1,4-dithiin-2,3dithiolate; gg mtdt = 1,2-bis(methylthio)ethylene-1,2-dithiolate; hh i-mnt = 2,2-dicyano-1,1-ethylenedithiolate; ii TpyP = 5,10,15,20-tetrakis(4-pyridyl)-21H , 23H -porphyrin; jj C H S = 2,2 -bithiophene; kk C H S = 2-thienyl; ll hat(CN) = hexaazatriphenylene hexacarbonitrile; mm PCHO = 2-(diphenylphosphino)benzaldehyde; nn (t -Bu-Hfy)CS = 4 3 8 4 2 6 2 2,7-di-tert-butyl-9H -ﬂuorene-9-carbodithioate; oo dppa = bis(diphenylphosphino)acetylene; pp dpbp = 4,4-bis(diphenylphosphino)biphenylene; qq pip = piperidine; rr btz = 1,2,3ss tt uu vv benzotriazol-1-yl; tz = 1,2,4-triazol-1-yl; m2 pz = 3,5-dimethylpyrazol-1-yl; mpz = 3-methylpyrazol-1-yl; Tm = hydrotris(3-methyl-1-imidazolyl-2-thioxo)borate(1−); ww linear polymer based on disilver repeating units; xx 2D polymer based on tetrasilver repeating units.

∗ References for Table 2.1 are given after the main Reference List. Notes: a C = chelating, B = bridging, M = monodentate; b cod = cycloocta-1,5-diene; c HMB = η6 -C6 Me6 ; d Cy = cyclohexyl; e Py = pyridine; f Pz = pyrazol-1-yl; g en = ethane-1,2-diamine; h Fu = 2-furyl; i Cym = η6 -p-cymene; j HQs = 1-phenyl-3-methyl-4-(2-thenoyl)pyrazol-5-one; k Xyl = 2,6-xylyl; l Mes = mesityl; m dppf-H =

Group

56 Ferrocenes: Ligands, Materials and Biomolecules

C

C C

C C

Yes

Yes Yes Yes No No

[MnPd(µ-PPh2 )(CO)4 (η2 -dppf)]•(CH3 )2 CO

[PdAg(µ-O2 CCF2 CF3 )2 (O2 CCF2 CF3 -O)(dppf)]

[PdAg2 (µ-O2 CCF2 CF3 )2 (O2 CCF2 CF3 -O)2 (dppf)]

[Ag2 Pd2 (NO3 )2 (dppf)2 -(µ3 S)2 ]•2CH2 Cl2

[Au2 Pd2 Cl2 (dppf)2 (µ3 -S)2 ]

[GaPd2 Cl2 (dppf)2 (µ3 -S)2 ] [GaCl4 ]

Mn

Ag

Au

Ga

Pd

C

No

[RuCp(dppf)(SnCl3 )]

Sn

Ru

C

C

No

[Re(CO)3 {η5 -5, 10-(µ-H)2 -exo{Rh(dppf)2 }-7, 8-C2 B9 H9 }]

Rh

Re

B

Yes

[W2 Ir2 (µ-dppf)(CO)8 (η5 -C5 H4 Me)2 ]

Ir

W

C

Yes

B

Coordination mode of dppfa

[MoS4 {Cu(dppf)}2 ] •2DMF•CH3 CN

Cu

Mo

X-ray Structure No

W

Cr

Compound

5 6

7 7 7 8 8

[RuClCp(dppf)] + SnCl2 in MeOH/toluene at r.t. for 6 h [PdCl2 (dppf)] + AgNO3 in CH3 CN for 1h, then PPN[Mn2 (CO)8 (µ-PPh2 )]b at 60◦ C for 24 h AgO2 CCF2 CF3 + [Pd(O2 CCF2 CF3 )2 (dppf)] in CH2 Cl2 for 30 min [PdAg(O2 CCF2 CF3 )3 (dppf)] + AgO2 CCF2 CF3 in CH2 Cl2 for 30 min [Pd2 (dppf)2 (µ-S)2 ] + AgNO3 in THF at r.t. for 4h [Pd2 (dppf)2 (µ-S)2 ] + [AuCl(SMe2 )] in THF at r.t. for 3 h [Pd2 (dppf)2 (µ-S)2 ] + GaCl3 in THF at r.t. for 4h

(continued overleaf )

4

3

[W2 Ir2 (CO)10 (η5 -C5 H4 Me)2 ] + dppf in CH2 Cl2 at r.t. for 16 h [RhCl(PPh3 )3 ] + dppf in CH2 Cl2 at r.t. for 1 h, then Cs[Re(CO)3 (η5 -7,8-C2 B9 H11 )] in CH2 Cl2 at r.t. for 2 h

2

1

Ref.∗

(Bu4 N)4 [MoS4 Cu6 Br8 ] in MeCN + dppf in DMF at 60◦ C for 30 min

[W(CO)6 ] in THF/hν, then (i) [Cr(CO)6 ]/hν, (ii) dppf in THF

Substrates + Preparative Conditions

Table 2.2 A selected list of recently characterised di- and polynuclear heterometallic complexes

[Cr(CO)5 (µ-dppf)W(CO)5 ]

M2

M1

The Coordination and Catalytic Chemistry of dppf 57

Pt

M1

C

No No No No No Yes No

[Pd2 SnF4 (dppf)2 (µ3 -S)2 ]•CH2 Cl2

[PbPd2 (NO3 )(dppf)2 (µ3 -S)2 ]NO3

[PbPd2 (NO3 )(dppf)2 (µ3 -S)2 ][PF6 ]

[BiPd2 Cl2 (dppf)2 (µ3 -S)2 ]Cl

[BiPd2 Cl2 (dppf)2 (µ3 -S)2 ][PF6 ]

[Ag2 Pt2 Cl2 (dppf)2 (µ3 -S)2 ]

[Pt{η3 -C(C≡CPh)=C(Ph)AuPPh3 }(dppf)](O3 SCF3 )

Sn

Pb

Ag

Au

Bi

C

No

[TlPd2 (dppf)2 (µ3 -S)2 ][PF6 ]3

C

C

C

C

C

C

C

No

[TlPd2 (dppf)2 (µ3 -S)2 ]NO3

Tl

C

Coordination mode of dppfa

Yes

X-ray Structure

[InPd2 Cl3 (dppf)2 (µ3 -S)2 ]

Compound

8 8 8 8 8 8 8 9

[Pd2 (dppf)2 (µ-S)2 ] + TlNO3 in THF at r.t. for 4h [Pd2 (dppf)2 (µ-S)2 ] + TlNO3 + NH4 [PF6 ] in MeOH + in THF at r.t. for 4 h [Pd2 (dppf)2 (µ-S)2 ] + SnF4 in THF at r.t. for 6 h [Pd2 (dppf)2 (µ-S)2 ] + Pb(NO3 )2 in THF at r.t. for 4 h [Pd2 (dppf)2 (µ-S)2 ] + Pb(NO3 )2 + NH4 [PF6 ] in MeOH + in THF at r.t. for 4 h [Pd2 (dppf)2 (µ-S)2 ] + BiCl3 in THF at r.t. for 4 h [Pd2 (dppf)2 (µ-S)2 ] + BiCl3 + NH4 [PF6 ] in MeOH, in THF at r.t. for 4 h [Pt2 (dppf)2 (µ-S)2 ] + [AgCl(PPh3 )] in THF for 2h

10

8

[Pd2 (dppf)2 (µ-S)2 ] + InCl3 in THF at r.t. for 6 h

[Pt(η2 -PhC≡CC≡CPh)(dppf)] + [Au (PPh3 )](O3 SCF3 ) in THF for 5 min/5◦ C, then 10 min/r.t.

Ref.∗

Substrates + Preparative Conditions

Table 2.2 (continued )

In

M2

58 Ferrocenes: Ligands, Materials and Biomolecules

No No

(Et4 N)2 [(MoS4 )2 Au2 (dppf)]

(Et4 N)2 [(MoOS3 )2 Au2 (dppf)]

B

B

B

B

C

C

C

12 13 14 14 14 14

Ag2 SO4 + (Et4 N)2 [Ni(mnt)2 ] + dppf in MeCl for 12 h [AuCl(SC4 H8 )]e + (Et4 N)2 [MoOS4 ] + dppf in CH2 Cl2 [AuCl(SC4 H8 )]e + (Et4 N)2 [MoOS4 ] + dppf in CH2 Cl2 [AuCl(SC4 H8 )]e + (Et4 N)2 [WS4 ] + dppf in CH2 Cl2 [AuCl(SC4 H8 )]e + (Et4 N)2 [WOS3 ] + dppf in CH2 Cl2

12

[Pt(C≡CPh)2 (dppf)] in CHCl3 + Ag[BF4 ] in MeCN at r.t. for 1 h crystallization of [Pt(C≡CPh)2 Ag(dppf)][BF4 ] from CH2 Cl2 /hexane

12

12

[Pt(C≡CPh)2 (dppf)] in CHCl3 + [Cu (NCMe)4 ][BF4 ] in MeCN at r.t. for 1 h [Pt(C≡CPh)2 Cu(NCMe)(dppf)][BF4 ] in CH2 Cl2 /hexane at r.t for 1 d

11

[Pt(dppf)(η2 -dba)]c + [Ru3 (CO)9 (µ3 -S)2 ] in toluene at −40◦ C/1 h, then 0◦ C/1 h

∗ References for Table 2.2 are given after the main Reference List. Notes: a C = chelating, B = bridging; b PPN = [N(PPh3 )2 ](1+); c dba = dibenzoylacetone; d mnt = maleonitriledithiolate(2−); e SC4 H8 = tetrahydrothiophene.

W

Mo

Au

No

Yes

[Ag(dppf)]2 [Ni(mnt)2 ]d

Ni

Ag

(Et4 N)2 [(WOS3 )2 Au2 (dppf)]

Yes

[{Pt(C≡CPh)2 (dppf)}2 Ag][BF4 ]

Yes

No

[Pt(C≡CPh)2 Ag(dppf)][BF4 ]

(Et4 N)2 [(WS4 )2 Au2 (dppf)]

Yes

[{Pt(dppf)(C≡CPh)2 }2 Cu][BF4 ]

Ag

C

No

[Pt(C≡CPh)2 Cu(NCMe)(dppf)] [BF4 ]

Cu C

C

Yes

[PtRu3 (CO)6 (µ-CO)2 (dppf) (µ4 -S)2 ]

Ru

The Coordination and Catalytic Chemistry of dppf 59

Pd

Ag

10

11

M

(PPh3 )2 (CO)OsB5 H9 :[2,2,2-(PPh3 )2 (CO)-nido-2-OsB4 H7 -3(BH2 •dppf)] + nido-Os(PPh3 )2 (CO)B5 H9

Bd

B

[Ag(PMe2 Ph)2 (dppf)][PF6 ] + dppf

+

PF− 6 B

2

(µ-dppf)c

[Ag2 (O2 CCF3 )2 (dppf)] + PMe2 Ph/PF− 6

[PdCl(LNC )]

B

C

[RuX(HMB)(dppf)][PF6 ] (X = Cl, NCS)b + NaSCN

[RhCl2 Cp*]2 (µ-dppf) + dppf/PF− 6

C

Coordination mode of dppfa

[RuClCp(dppf)] + NaS2 CNEt2

Starting complex + Reagent(s)

6]

[Ag(O2 CCF3 )(dppf)]n

[Ag(PMe2 Ph)2 (dppf)][PF6 ]

[Pd(LNC )(dppf)][PF

Be

C

C

C

B

[2,2,2-(PPh3 )2 (CO)-nido-2OsB4 H7 -3-BH2 ]2 (dppf)

[RhClCp*(dppf)][PF6 ]

B

B

[RuX2 (HMB)]2 (µ-dppf)

[RuCp(S2 CNEt2 )]2 (µ-dppf)

Coordination mode of dppfa

5

5

4

3

2

1

1

Ref.∗

∗ References for Table 2.3 are given after the main Reference list. Notes: a C = chelating, B = bridging, M = monodentate; b HMB = η6 -hexamethylbenzene; c LNC = 2-(N ,N -dimethylaminomethyl)phenyl-κ 2 C ,N ; d intramolecular bridge; e dppf-bridged polymer.

Rh

Os

Ru

Metal

9

8

Group

Table 2.3 A selected list of dppf complexes that show changes of coordination mode of dppf during reaction

60 Ferrocenes: Ligands, Materials and Biomolecules

The Coordination and Catalytic Chemistry of dppf

61

As given in our earlier review,1 dppf generally shows a greater tendency to adopt three major coordination modes: unidentate, chelating and bridging. This is attributed to its skeletal ﬂexibility, which allows for a greater degree of torsional freedom and lowers the energy barrier between different modes of bonding. The unique sandwiched skeleton, as compared to other common diphosphines, presents an array of metallocene-centered fragmentations that are not found in other diphosphine ligands such as Ph2 P(CH2 )n PPh2 , binap etc. Under oxidative or oxygenated conditions, it is much more likely for a coordinated phosphine in a M1 (µ-dppf)M2 fragment to be oxidised to its phosphine oxide derivative, namely M1 (µ-dppfO)M2 or M1 (µ-dppfO2 )M2 . Similarly, in a sulfur or selenium containing system, it is possible to witness unwelcome chalcogenisation to give M1 (µ-dppfE)M2 and M1 (µ-dppfE2 )M2 (E = sulfur, selenium). The source of oxygen or chalcogen is not necessarily molecular oxygen and elemental Sn or Sen . Co-ligands that contain sulfur, selenium or oxygen are possible sources. The inter-ligand migratory pathway can be complex and generally not well understood. The insertion of oxygen or chalcogen to the M−P bond may also be catalysed by the metal. In mono and dinuclear complexes, a much more common reactivity shown is the change of coordination mode. A list of some recent representative examples is given in Table 2–3. There are four common types of mode changes: monodentate to chelating (Type 1); monodentate to bridging (Type 2); bridging to chelating (Type 3); and chelating to bridging (Type 4). Type 1 is initiated by removal of a co-ligand, therefore creating a neighbouring vacant site for chelate formation (Equation 2.1). X X

M

P

P

−L

X

L

X

M

P

P

X X

M

P

(2.1)

P

Type 2 is driven by the basic character of the dangling phosphine. It enables the phosphine to capture a second Lewis acidic metal. The latter can be generated in situ from the same system that carries the pendant phosphine. This would lead to inter-molecular interaction, giving homo-dinuclear complexes (Equation 2.2). A recent example of such is given by the formation of [2,2,2-(PPh3 )2 (CO)-nido-2-OsB4 H7 -3-BH2 ]2 (dppf) from (PPh3 )2 (CO)OsB5 H9 :[2,2,2-(PPh3 )2 (CO)-nido-2-OsB4 H7 -3-(BH2 .dppf)], although in this case dppf strictly bridges two B5 H9 units in the coordination sphere, not the metal (osmium) itself. If another metal is introduced to the system, it would bind to the pendant phosphine, thus giving hetero-dinuclear complexes. X X

M

P

X

L

M

X X

X

L

−L

P

L

P

P

L

L

M′

M′

M

P

X

P dimerisation

X

M

P

P

P

P

M

X X

L

(2.2)

X X

62

Ferrocenes: Ligands, Materials and Biomolecules

Changes of Types 3 and 4 may not have any thermodynamic advantage. The kinetric barrier may also be high. They could proceed only with the help of a chemical drive. Type 3 is activated by selective de-ligation on either metal centre across the bridge, thus providing a strongly acidic site that provides a driving force for cyclisation and chelation (Equation 2.3). A good example of this is illustrated in the formation of [Pd(C6 H4 CH2 NMe2 -κ 2 C,N )(dppf-κ 2 P , P )][PF6 ] from [PdCl(C6 H4 CH2 NMe2 -κ 2 C, N )]2 (µ-dppf). Use of Na[PF6 ] as a metathesis agent removes the stable choride, thereby creating an acidic center for cyclisation. L L

M

L P

P

L

M

L

L

−L

L

L

M

P

P

L

M

L

−[ML3]

L

L

L

M

P

(2.3)

P

Type 4 requires the use of excess nucleophile, whose attack on the metal prompts the release of one end of the chelate. The resultant dangling phosphine, which is strongly basic, rapidly captures a neighbouring molecule, thus resulting in a bridge (Equation 2.4). This bridge formation would only be feasible if the metal under attack is unsaturated or acidic. This is achieved through the liberation of free dppf into the solution. One useful approach, which is often neglected, is therefore is to use a strong Lewis base to attack the metal and compete with dppf. It forces the chelate to open and create the site for an incoming group. This is best exempliﬁed by the formation of [Ru(NCS)2 (η6 -C6 Me6 )]2 (µ-dppf) from [Ru(NCS)(η6 -C6 Me6 )(dppf)][PF6 ]. The use of excess NaNCS promotes NCS− attack on the ruthenium(II), forcing the chelate to open on one metal, and liberation of dppf in the other. This results in bridge formation and ﬁlling up of the vacated sites by the nucleophilic thiocyanate. X X X

Nu M

X

P

X

P

M

P

P

X

X

X M X

2.3

M

P Nu

M

P Nu

P

P

Nu

X

M

Nu

P P

(2.4)

Nu

P P

X X

X

M

P P

Nu

Nu

X

M

X M X

Coordination Chemistry of the Oxide, Sulﬁde and Selenide Derivatives of dppf

The chemistry of the oxides and chalcogenides of dppf, namely dppfE and dppfE2 (E = oxygen, sulfur, selenium) is emerging along the same line as dppf, except that, in the case of the mono-oxide, sulﬁde or selenide forms, the ligands are inherently

The Coordination and Catalytic Chemistry of dppf

63

difunctional and hybrid in characters, whereas for the dioxide, sulﬁde or selenide forms, the donor characters have completely changed. These modiﬁcations allow the formation of bigger macrocyclic chelate rings or longer spacers in coordination polymer but keep the properties inherited from the metallocene skeletal backbone. Like the parent dppf, they are electrochemically active. The oxide and sulﬁde show good oxidation reversibility whereas the selenide is irreversible.19 Preparation of dppfE2 from the oxidation (with H2 O2 , Sn , Sen etc) is generally easy. Selective preparation of the mono-form dppfE is much trickier. Grushin developed a novel and valuable method to make dppfO (up to 65 % isolated yield) under palladium(II)-catalysed biphasic conditions promoted by iodide. It is interesting to note the dual reducing [palladium(II) to palladium(0)] and oxygenation [M(dppf) to M(dppfO)] functions of hydroxide.20 Reoxidation of palladium(0) to palladium(II) is facilitated by dibromoethane (Scheme 2.1). P

+ PdX2

P

Biphasic Organic Aqueous

CH2

X

CH2

P

PdII

P

X

P

P

Pd0

X 2OH−

X

P

P

P

P

P P

P

O

P P

Pd0

P P

2X− + H2O

O

P

Scheme 2.1 Proposed biphasic catalytic preparation of dppf monoxide (The scheme is adapted from Ref 20)

Almost all of the dppfE and dppfE2 complexes are prepared by standard use of dppfE or dppfE2 ligands based on known coordination chemistry. A notable exception is the use of dppfSe2 , which serves as a source of both dppf and selenium in a redox process involving cleavage of the weaker P−Se bond. This is best illustrated in the work done by Tiripicchio et al. in 1997 who synthesised the disubstituted clusters [Fe3 (µ3 -Se)2 (CO)7 (µ-dppf)] (1) (Scheme 2.2) from the simultaneous oxidation and ligand substitution of [Fe3 (CO)12 ] and dppfSe2 .21 Predieri et al. reported similar reactions of [M3 (CO)12 ] (M = iron or ruthenium) with dppfSe2 in the presence of Me3 NO. They generally give the clusters 1 and [Ru3 (µ3 -Se)2 (CO)7 (η-dppf)] (2)22

64

Ferrocenes: Ligands, Materials and Biomolecules (CO)3 M Se (CO)2M P

Se (CO)3 Ru

M(CO)2 Se P

Se (CO)2Ru

Se

P Ru(CO)2

fc

P 1 (M = Fe), 3 (M = Ru)

2

PPh2 (CO)2 Ru Ru(CO)2 PPh Ru CPh (CO)2 O 4

Scheme 2.2 The molecular structures of [Fe3 Se2 (CO)7 (µ-dppf)] (1), Ru3 Se2 (CO)7 (dppf)] (2), [Ru3 Se2 (CO)7 (µ-dppf)] (3), and [Ru3 Se{µ-P(Ph)fcPPh2 }(µ-OCPh)(CO)6 ] (4)

(Scheme 2.2). The dppf ligand in cluster 1 traverses two non-bonding iron atoms, whereas in 2, it functions as a chelate on ruthenium at the basal plane of the square pyramid. The latter is among the ﬁrst carbonyl clusters with a chelating dppf. Both compounds are ﬂuxional, showing a rocking motion of the bidentate bridging ligand below the square basal plane (in 1) and an exchange of the axial and equatorial positions between the two chelating P atoms in 2. Thermal reactions of [Ru3 (CO)12 ] with dppfSe2 at 60 and 110◦ C, respectively, give the two isomeric nido-clusters 2 and [Ru3 Se2 (CO)7 (µ-dppf)] (3) (Scheme 2.2).23 The latter, which is a congeneric analogue of 2, contains bridging dppf. The former 2, which is obtained under kinetic control, can be thermally converted to the more stable bridged cluster [Ru3 Se{µ-P(Ph)fcPPh2}(µ-OCPh)(CO)6] (4) (Scheme 2.2) in toluene. The LC/MS technique has been applied successfully to separate and identify solvent mixtures containing many of these clusters without subject to any sample pre-treatment.24 It is generally known that dppf-coordinated complexes can undergo oxygenation or chalcogenisation to dppfE- or dppE2 -coordinated complexes. A good example is [CoCpI(dppf)]I which, upon exposure to air in a tetrahydrofuran (THF) or dichloromethane (CH2 Cl2 ) solution, affords the paramagnetic polymer [CoI2 (µ-dppfO2 )]n .25 Intriguingly, the obvious reaction between cobalt(II) iodide and dppfO2 yields, not the polymer, but the monomeric [CoI2 (dppfO2 )] instead. Both compounds are electrochemically and magnetically active. However, there are examples when the dppf chelate can resist sulfur attack. This is exempliﬁed in the reaction of [RuClCp(dppf)]/ Na[BPh4 ] with elemental sulfur in CH2 Cl2 , giving [{RuCp(dppf)}2 (µ-S2 )][BPh4 ]Cl.26 It is not uncommon to see dynamic changes of coordination mode in dppf complexes. Observation of such is relatively rare in the dppfE2 complexes. A good example is found in the complex salt [Au(dppfSe2 )][AuCl2 ] in which there is a dynamic exchange between a chelating complex in the salt form, and a bridging complex in the covalent and neutral dinuclear form (Equation 2.5). Related to this, it is also notable that chlorination of [AuCl]2 (µ-dppfS2 ) by chlorine in carbon tetrachloride (CCl4 ) could revert from a bridge to chelate in [Au(dppfS2 )][AuCl4 ]. Ph2 P Se AuCl

Ph2 P Se Fe

Au

Ph2 P Se

[AuCl2]

Fe Ph ClAu Se P2

(2.5)

Co

Pd

Cu

9

10

11

No No

[{Ag(PPh3 )}2 (dppfO2 )](ClO4 )2

No

[Cu(dppfO2 )(PPh3 )2 ][NO3 ]

[Ag(dppfO2 )(PPh3 )]ClO4

No

[Cu(dppfO)2 ][BF4 ]

No

Yes

[PdCl2 (dppfO2 )]

[Ag(dppfO2 )2 ]ClO4

Yes

trans-[PdCl2 (dppfO)2 ]

No

Yes

[CoI2 (dppfO2 )]

[Ag(OClO3 )(dppfO2 )]

Yes

No

[RuCl2 (η3 :η3 -C10 H16 )(dppfO)]

[CoI2 (µ-dppfO2 )]n

X-ray structure

Compound

∗ References for table 2.4 are given after the main reference List. Notes: a C = chelating, B = bridging, M = monodentate.

Ag

Ru

Metal

8

Group

B

C

C

C

C

C

C

C

C

B

M

Coordination mode of dppfOan

4 5 6 6

[PdCl2 (MeCN)2 ] + dppfO2 in CH2 Cl2 for 12 h [Cu(MeCN)4 ][BF4 ] + dppfO in CH2 Cl2 at r.t. for 1 h [Cu(NO3 )(PPh3 )2 ] + dppfO2 in CH2 Cl2 for 45 min AgClO4 + dppfO2 in CH2 Cl2 for 1 h

Ditto

6

6

3

Na2 [PdCl4 ] + dppfO in CH2 Cl2

[Ag(OClO3 )(PPh3 )] + dppfO2 in CH2 Cl2 for 90 min

2

CoI2 + dppfO2 in CH2 Cl2 at r.t. for 1 h

6

2

[CoICp(dppf)]I + air in CH2 Cl2 or THF

Ditto

1

Ref.∗

[Ru(µ-Cl)Cl(η3 :η3 -C10 H16 )]2 + dppfO in CH2 Cl2 at r.t.

Substrates and preparative conditions

Table 2.4 A selected list of recently characterised dppfO and dppfO2 complexes

The Coordination and Catalytic Chemistry of dppf 65

Pd

Cu

10

11

No Yes No Yes No No No No No Yes

[{Cu(dppf)}2 (µ-dppfS2 )][BF4 ]2

[Cu(dppfS)2 ][BF4 ]

[Cu(dppfS2 )][BF4 ]

[Cu2 (dppfS2 )3 X2 ]n (X = BF4 , PF6 , ClO4 )

[(AuX)2 (µ-dppfS2 )] (X = Cl and Ph)

[(AuPPh3 )2 (µ-dppfS2 )](ClO4 )2

[(AuPPh3 )(dppfS2 )]ClO4

[Au(C6 F5 )3 ]2 (µ-dppfS2 )

[AuCl(C6 F5 )2 ]2 (µ-dppfS2 )

B

B

C

B

B

B

C

B

B

C

C

No

[Rh(CO)(dppfS) (µ-Cl)RhCl(CO)2 ]

[PdCl2 (dppfS)]

C

No

Coordination mode of dppfSan

[RhCl(CO)(dppfS)]

X-ray structure

1 1 2 2 3

[RhCl(CO)2 ]2 + dppfS in toluene at r.t. for 1 h PdCl2 + dppfS in toluene at r.t. for 72 h [Cu(MeCN)4 ][BF4 ] + dppf in CH2 Cl2 at r.t., then with dppfS2 for 15 min [Cu(MeCN)4 ][BF4 ] + dppfS (2 equiv.) in CH2 Cl2 for 1 h [Cu(MeCN)4 ][BF4 ] + dppfS2 in CH2 Cl2 at r.t. for 1 h

4 4 4

[Au(C6 F5 )3 (OEt2 )] + dppfS2 in CH2 Cl2 for 1 h [Au(C6 F5 )3 (OEt2 )] + dppfS2 in CH2 Cl2 for 1 h

4

[Au(OClO3 )(PPh3 )] + dppfS2 in CH2 Cl2 for 90 min Ditto

4

[AuX(C4 H8 S)]b + dppfS2 in CH2 Cl2 for 90 min

3

1

[RhCl(CO)2 ]2 + dppfS in CH2 Cl2 at r.t. for 2 h

[Cu(MeCN)4 ]X + dppfS2 in CH2 Cl2 at r.t. for 1h

Ref.∗

Substrates and preparative conditions

Table 2.5 A selected list of recently characterised dppfS and dppfS2 complexes

Compound

∗ References for Table 2.5 are given after the main Reference List. Notes: a C = chelating, B = bridging, M = monodentate; b C4 H8 S = tetrahydrothiophene.

Au

Rh

Metal

9

Group

66 Ferrocenes: Ligands, Materials and Biomolecules

Au

Ag

Cu

Metal

No

[Ag(dppfSe2 )](CF3 SO3 )

No

No

[Ag(dppfSe2 )][PF6 ]

[Au(C6 F5 )3 ]2 (µ-dppfSe2 )

Yes

[Ag(dppfSe2 )]ClO4

Yes

No

[{Cu2 (dppfSe2 )3 (X)2 }n ] (X = BF4 , PF6 , ClO4 )

[Au(dppfSe2 )][AuCl2 ]

No

[Cu(dppfSe2 )](CF3 SO3 )

No

No

[Cu(dppfSe2 )]X (X = BF4 , PF6 )

[Au(dppfSe2 )]ClO4

Yes

X-ray structure

[Cu(dppfSe)2 ][BF4 ]

Compound

B

C

C

C

C

C

B

C

C

C

Coordination mode of dppfSan

∗ References for Table 2.6 are given after the main Reference List. Notes: a C = chelating, B = bridging, M = monodentate; b C4 H8 S = tetrahydrothiophene.

11

Group

5 2b 3 3

[Ag(MeCN)4 ]ClO4 + dppfSe2 in CH2 Cl2 at r.t. for 1 h [Ag(NCCH3 )4 ][PF6 ] + dppfSe2 in CH2 Cl2 at r.t. for 1 h CF3 SO3 Ag + dppfSe2 in CH2 Cl2 for 30 min + dppfSe2 in CH2 Cl2 for

3 3

[AuCl(C4 H8 S)]b (2 equiv.) + dppfSe2 in CH2 Cl2 for 30 min [Au(C6 F5 )3 (OEt2 )] + dppfSe2 in CH2 Cl2 for 30 min

30 min

[Au(C4 H8 S)2 ]ClOb4

4

3

[Cu(NCMe)4 ](CF3 SO3 ) + dppfSe2 in THF for 30 min [Cu(MeCN)4 ]X + dppfSe2 in CH2 Cl2 at r.t. for 1h

2a, b

1

[Cu(MeCN)4 ][BF4 ] + dppf, then dppfSe2 and in ClCH2 CH2 Cl [Cu(MeCN)4 ]X + dppfSe2 in CH2 Cl2 at r.t. for 1h

Ref.∗

Substrates and preparative conditions

Table 2.6 A selected list of recently characterised dppfSe and dppfSe2 complexes

The Coordination and Catalytic Chemistry of dppf 67

1 1

[Ag(dppfO2 )(bipy)]ClO4 [Ag(dppfO2 ){(Ph2 P(S))2 CH2 }]ClO4 [Ag(dppfO2 )(S2 CNEt2 )]

bipy (Ph2 P(S))2 CH2 Na(S2 CNEt2 )

3 3

[Ag(dppfSe2 )(bipy)](CF3 SO3 ) [Ag(dppfSe2 )2 ](CF3 SO3 )

bipy

∗ References for Table 2.7 are given after the main Reference List. Notes: a The dppfE2 coordinate as E , E -chelating donors in all cases.

dppfSe2

3

[Ag(dppfSe2 ){(Ph2 P(Y))2 CH2 }](CF3 SO3 )

(Ph2 P(Y))2 CH2 (Y = S and Se)

[Ag(dppfSe2 )](CF3 SO3 )

2

[Au(dppfS2 )][AuCl4 ]

Cl2 (in CCl4 )

[AuCl]2 (µ-dppfS2 )

2

PPh3

[Au(dppfS2 )(PPh3 )](CF3 SO3 )

1

[Ag(dppfO2 )(PPh3 )2 ]ClO4

PPh3

1

1

Ref.∗

[Ag(dppfO2 )(P(S)Ph3 )]ClO4

Product

P(S)Ph3

Substrate

[Au(dppfS2 )](CF3 SO3 )

[Ag(OClO3 )(dppfO2 )]

Starting material

Table 2.7 Representative reactions of some dppfE2 complexes (E = O, S, Se)a

68 Ferrocenes: Ligands, Materials and Biomolecules

The Coordination and Catalytic Chemistry of dppf

69

A list of dppfE and dppfE2 (E = oxygen, sulfur, selenium) complexes prepared and characterised in the period 1999 to 2007 is given in Tables 2.4, 2.5 and 2.6, respectively. Their preparative conditions are also summarised. In addition, representative reactions of some dppfE2 complexes (E = oxygen, sulfur, selenium) are given in Table 2.7.

2.4

Catalytic Reactions Involving dppf Complexes

Dppf-supported metal complexes are widely used in promoting coupling reactions. The inherent redox ability of dppf enables its complex to be electronically sensitive to the changing demands of the substrates and co-ligands along its catalytic pathway. This raises the catalyst stability without compromising on its catalytic activity.27 Some recent and representative examples are illustrated below, while Chapter 3 reports on many related systems based on dppf analogues. 2.4.1 2.4.1.1

C−C Coupling Reactions Suzuki Reaction

The palladium-catalysed coupling reaction between organoboronic acid and halides,28 commonly termed as Suzuki–Miyaura (or Suzuki) coupling,29 is effectively supported by phosphine ligands. In the presence of dppf, [Pd2 (dba)3 ] shows impressive activity and selectivity in the coupling between bromobenzene and the activated 4bromoacetophenone with phenylboronic acid (Equation 2.6).30 The high performance is attributed to the large bite angle of the ligand (99◦ in [MLn (dppf)]30 ), which promotes interaction between the organic substituents thus driving the reductive elimination step. B(OH)2

+

R

Br

Pd2(dba)3/dppf base

R

R = H, C(O)Me

(2.6) Many dppf complexes have been used to support catalysis. Among them [PdX2 (dppf)] is probably the most common. Colacot et al.31 prepared a complete series of [MX2 (dppf)] (M = Pd, Pt; X = Cl, Br, and I) and compare their catalytic activities as a function of P–M–P and X–Pd–X angles. The platinum complexes are not active while [PdCl2 (dppf)] and [PdBr2 (dppf)] show the highest activities under similar conditions. Dppf generally performs better than other mono- and diphosphines. Notable successes through the use of dppf have been experienced even for couplings involving the challenging aryl chloride substrates. For example, high yields of 3substituted propanals (8) can be obtained from acrolein diethyl acetal (5) and aryl chloride (7) when catalysed by Pd(OAc)2 and dppf in reﬂuxing THF (Equation 2.7).32 Such coupling using the borane derivatives 6 can be extended to a variety of aryl and vinyl halides. One-pot borylation with Suzuki coupling has been achieved in the synthesis of functionalised 4,4 -bisquinolones (10) promoted by microwaves with the aid of palladiumcatalyst (Table 2.8).33 These methods are generally suitable for efﬁcient symmetrical

70

Ferrocenes: Ligands, Materials and Biomolecules O

Boc N N

H

OEt OEt 8 (94% from 7)

5

B

OEt

(2.7)

HClaq, iPrOAc

9-BBN, THF Boc N N

+

OEt 6

7

Cl

Pd(OAc)2/dppf

OEt

Boc N N

OEt

K2CO3, THF 65 °C, 20 h

biaryl coupling. In general, the palladium-catalysed cross-coupling method (Method A) is more reliable and better yielding (68–85 %) when compared to the nickel-mediated homo-coupling method (Method B, 39–90 %). An additional disadvantage of the latter is the extensive puriﬁcation needed to remove large quantities of free PPh3 by ﬂash chromatography. Hilt et al. have shown that dihydroaromatic alkenyl-substituted boronic esters (11), which can be generated by neutral cobalt(I)-catalysed Diels–Alder reactions, are suitable precursors for the synthesis of different classes of polycyclic compounds.34 Table 2.8 Synthesis of biaryls via microwave-assisted cross- and homo-coupling of (hetero)aryl chlorides R1

R4 R3

N R2

O

Method A: PdCl2(dppf), [B(pin)]2, KOH, BuCl, MW, 130 °C, 35 min

Cl

Method B NiCl2, Zn, PPh3, KI, DMF, MW, 205 °C, 25 min

O

R1

R2 R3

N

R4 R4 R3

N R2

9 Substrate

R1

O 10

Yield (%) Method A

Method B

85

90

68

68

83

39

9d: R1 = Me, R2 = R3 = H, R4 = OMe

83

70

9e: R1 = Me, R2 = R4 = H, R3 = OMe

70

74

9f: R1 = Me, R2 = H, R3 = R4 = OMe

82

41

9a: R1 = Me, R2 = R3 = R4 = H 1-R2 =

9b: R

1=

9c: R

Ph,

(CH2)3, R2 =

R3 =

R3 =

R4 =

R4 =

H

H

The Coordination and Catalytic Chemistry of dppf

71

They react under palladium catalytic conditions with diiodobenzene, bromoiodobenzene and iodoaniline derivatives to give regioselectively substituted phenanthrene and phenanthridine derivatives (Equation 2.8). This Suzuki coupling reaction proceeds easily to deliver biphenylamine products 12.

R1

R1

I B

+

O

R2

1. PdCl2(dppf) 2. DDQ

H 2N

R2

O

NH2

11

12

(2.8) Colobert et al.35 reported the ﬁrst examples of asymmetric biaryl Suzuki coupling reactions using enantiopure β-hydroxy- and methoxy-sulfoxides as chiral auxiliaries with an effective control of the axial chirality (up to 98 % de). Highly diastereoselective biaryl Suzuki coupling reactions of (1R)-1-(2-iodo or bromophenyl)-2-((R)-4tolylsulﬁnyl)-1-ethanol derivatives (13) with various aryl- or naphthyl-boronic acids (or esters) (14) lead to high quantitative yields of 15 (Equation 2.9). The diastereoselectivity is essentially controlled by the stereogenic carbon atom close to the new C−C bond. Interestingly, it is also ligand dependent. R3

+ R2

S OR1

X

Pd(OAc)2, dppf

14

CsF/dioxane 100 °C, 1 h

pTol

O

13

X = I, Br

B(OR)2

B(OR)2

R4

R5

(2.9)

or

14 =

R3

R3

15 =

15

R2

S

* OR1 R4

O

pTol

or

R2

S

* OR1 R5

O

pTol

72

Ferrocenes: Ligands, Materials and Biomolecules

2.4.1.2

Heck Reaction

In a typical Heck reaction an unsaturated halide or triﬂate cross-couples with an alkene in the presence of palladium catalyst and a strong base to form a substituted alkene. The regioselectivity generally depends on the substrates and experimental conditions. Highly regioselective Heck couplings of α,β-unsaturated tosylate and mesylate derivatives (16) with vinyl ethers and N -acyl-N -vinylamines (17) have been reported.36 Several 2-alkoxy-1,3-dienes and 2-acylamino-1,3-butadienes (18) are prepared in good yields using 1.5 mol% of [Pd2 (dba)3 ], 3 mol% of dppf, and NEt(i-Pr)2 (DIPEA) in dioxane (Equation 2.10). Use of α,β-unsaturated ketones and esters would provide a cheaper alternative to similar couplings using a triﬂate electrophile. R

OTs/OMs +

Y X

1.5 mol% Pd2(dba)3, 3 mol% dppf, DIPEA

R

(2.10)

Y

dioxane, 85 °C

O

O

X

16 17 X = C,O; R = OR′, NHC(O)R′′

18

Coupling between methyl acrylate and 4-nitrophenyl triﬂate, using complex 19 as the catalyst gives 3-(4-nitrophenyl)acrylic acid methyl ester (20) as the major product and nitrobenzene as the side product (Equation 2.11).37 Reaction with phenyl triﬂate gives trace of methyl cinnamate as the Heck addition product. Addition of halides or acetate salts has a modest effect on the rate of oxidative addition and results in neutral aryl-palladium(II) complexes. Addition of Li+ or Eu3+ salts [notably Eu(OTf)3 ] raises the reaction rate, possibly by promoting a competing dissociative pathway. P

OTf

+

CO2Me

P

CO2Me Pd

Eu(OTf)3, NEt3, THF, 65 °C

O2N

P P = dppf

CO2Me

19

+ O2N

O2N

20

(2.11) The Heck coupling effect of the complexes [Pd(dppf)(SRF )2 ], (SRF = SC6 F5 , SC6 F4 -4-H, SC6 H4 -2-CF3 , SC6 H4 -4-F, SC6 H4 -3-F) have been studied in the reaction of bromobenzene with styrene to give (E)-stilbene.38 The effects of the thiolates and the P–Pd–P bite angles have been examined. Electron-withdrawing substituents tend to favor higher yields in the Heck reaction catalysed by [Pd(SRF )2 (dppf)]. Hallberg et al.39 reported a highly regioselective, one-pot, palladium-catalysed synthesis of the 2-aryl-3-(N, N -dialkylamino)-1-propenes (21) promoted by microwaves (Equation 2.12). The very high regioselectivity observed in the arylation of allylamines or allyl alcohol is attributed to Pd–N and Pd–O coordination, respectively. Unexpectedly, phenyl triﬂate and 22 couple under carbon monoxide (1 atm) to give N, N -dimethylbenzamide (23) (Equation 2.13).

The Coordination and Catalytic Chemistry of dppf

Pd(OAc)2/dppf

OTf +

NR2 R = Alkyl

73

NR2

(2.12)

Base, CH3CN or DMF 80 °C, 20 h or microwave 3–5 min

21 O

Pd(OAc)2/dppf, CO

OTf +

NMe2

NMe2

80 °C, K2CO3, <2 h

22

(2.13)

23

Allylsilanes are important intermediates in organic synthesis as they show high regioselectivity towards electrophiles.40 When the Heck reaction is extended to allylsilanes such as H2 C=CHCH2 SiMe3 (24), the resultant arylation is ligandcontrolled and highly regioselective, delivering branched β-products of the type ArC(CH2 SiMe3 )=CH2 (25).41 The high preference for internal over terminal doublebond arylation suggests a strong contribution from the β-cation-stabilising effect of silicon. These reactions can be sped up further (to minutes) by microwaves, without sacriﬁcing the regioselectivity in most cases. 2.4.1.3

Other Palladium-Catalysed C−C Cross-Coupling Reactions

One of the simplest forms of C−C coupling is arylation or alkylation. Many techniques have been used. One is the combinative use of copper(I) and palladium(II)/palladium(0) in the arylation of an activated pyrrolidine. Treatment of (α –aminoalkyl)lithium reagents with aryl or vinyl iodides in the presence of catalytic amounts of copper(I) cyanide and [PdCl2 (PPh3 )2 ] or [Pd{P(C6 H4 OMe-4)3 }4 ] affords 2-aryl substituted amines (26 in Equation 2.14).42 The yields can be improved by the use of dppf taking advantage of its large bite. This method applies well to electron-rich aryl iodides (XArI, X = Me, OMe) but is ineffective towards their electron-poor counterparts (XArI, X = NO2 , CO2 Li).

N Boc

5 mol% Pd cat., PhI P, As, Sb ligand, 10 mol% CuCN BuLi, TMEDA, −78 °C, 2 h, THF or Et2O

N

Ph

(2.14)

Boc 26

2-Bromo-, 2-iodo- or 3-bromothiophenes, 2-bromothiazole and 2-bromofuran can be catalytically reduced to their corresponding bithiophene, bithiazole and bifuran derivatives.43 A basic alcohol medium favors the reductive coupling pathway over the hydrodehalogenation pathway, the latter of which is generally more facile with other reducing agents. The complex [PdCl2 (dppf)] is the most efﬁcient catalyst which gives the highest yield of the coupling products and bithiophene/thiophene ratio in the shortest time.

74

Ferrocenes: Ligands, Materials and Biomolecules

An unprecedented direct reaction between an aryl halide and a ketone catalysed palladium(0) under basic conditions has been reported by Hartwig et al. (Equation 2.15).44 Arylation of ketones gives secondary, tertiary, and quaternary carbon centres, along with independent generation of the palladium enolate intermediate. The likely catalytic pathway (Scheme 2.3) is initiated by oxidative addition to palladium(0) giving [PdAr(X)(dppf)], which captures the alkali enolate to give an unusual enolate aryl complex. Reductive elimination would then yield the unusual C−C bond-forming product. O

O

H

R” + ArBr

R R’ R” = Aryl, t-Bu

Pd(dba)2/dppf base

Ar

(2.15)

R’’ R

R’

O Ar R

X

[Pd(dppf)n]

Ar’

Y

R’

Ar

O Ar’

Pd(dppf) R’

(dppf)Pd

Ar X

R

O R Ar’ + base R’

Scheme 2.3 Proposed catalytic pathway of ketone arylation (The scheme is adapted from Ref. 44)

A similar coupling approach can be applied in conjunction with other methodologies to produce important compounds through domino reactions. For example, preparation of (Z)-6-arylidene/vinylidene-6a-methyl-1,5,6a-tetrahydro-2H -pyrrolo[3,2-c] pyrazol-3-ones (28) can be achieved through a remarkable series of base-promoted annulation, domino 1,5-enyne arylation and vinylation reactions (Equation 2.16).45 This sequential reaction is the ﬁrst example of a domino, palladium-catalysed intermolecular arylation/vinylation combined with an intramolecular carbopalladation process of a 1,5-enyne derivative (27). The preparation gives rise to fused heterocycles through a regioselective 5-endo-dig annulation reaction with a diagonal ring closure. It signiﬁes a positive outcome of β-hydride elimination with cleavage of the N−H bond.

The Coordination and Catalytic Chemistry of dppf

75

H H N

N

OR

H N

NHR1 + R2 X

O

O

Pd(OAc)2, dppf, piperidine

R2

N

r.t., 24 h

NH O

N H

27

28

(2.16) The enantio- and diastereomerically pure metal complexes of a chirally dynamic dppf ligand can be prepared. The chirality control is achieved through the coordination of enantiopure diaminobinaphthyl (DABN) to control the axial chirality of the group-10 d8 complexes (Equation 2.17).46 The nickel–dppf catalyst [Ni(dppf)(DABN)][SbF6]2 affords higher enantioselectivity and catalytic efﬁciency than its palladium or platinum counterparts or [Ni((R)-DABN)2 ][SbF6 ]2 without the dppf ligand. R2

O

H +

R1

OEt

H

[M(dppf)((R)-DABN)]2+ M = Pt, Pd, Ni

O

R2

OH OEt

R1

(2.17)

O

Decarboxylative coupling is an unusually creative way to create C−C bonds. Tunge et al.47 demonstrated a bio-inspired method for the synthesis of protected homoallylic amines (29) from allylic esters of amino acids in which the amine is protected as a diphenyl ketimine. This is effectively carried out at room temperature with the aid of [Pd2 (dba)3 ] and dppf (Equation 2.18). The key step involves decarboxylative metalation of α-amino acid derivatives, giving nucleophilic α-imino anion equivalents, and addition to electrophilic π-allyl palladium intermediates. Ph

Ph

5 mol% Pd2(dba)3, 10 mol% dppf

O

Ph

Ph +

N

N

Ph

Ph

N

(2.18)

O Ph

CO2

Ph

Ph

29

Another variation to the above is represented by the extrusion of not only carbon dioxide, but also a base such as methoxide that can capture a proton from [CH] and create a site for coupling. This is illustrated in the alkylation of active methine compounds (31) with benzylic carbonates (30). This reaction proceeds without the aid of a base when promoted by [PdCp(η3 -C3 H5 )]/dppf. Addition of 1,5-cod extends the lifetime of the palladium catalyst, thus leading to high yields of the desired benzylation products 32 (Equation 2.19).48 A plausible mechanism for catalytic benzylation is shown in Scheme 2.4. A palladium(0)-dppf species 34 could be formed from the reduction of [PdII (η3 -allyl)(dppf)] (33) via nucleophilic attack of malonate carbanion. It then captures and activates benzyl methyl carbonate to give a (η3 -benzyl)palladium dppf species (35), releasing methoxide which serves as the deprotonation source of the malonate.

76

Ferrocenes: Ligands, Materials and Biomolecules R

R1 OCOMe

+

H C R2

O

1–2 % [PdCp(h3-C3H5)]/dppf

R1

R C R3

10 % cod

R3

30 31 R = H; p-MeO; p-Me; p-MeO2C; p-CF3; o-Me

R2

32

(2.19) [Pd(h3-C3H5)(cod)]+ + dppf

(dppf)Pd+

OCOMe

33

−OCOMe

O

CO2 + MeO−

O

CO2R

−

CO2R (dppf)Pd0

Pd+(dppf)

34

35 −

CO2R

CO2R CO2R

CO2R

Scheme 2.4 Possible reaction mechanism of catalytic benzylation (The scheme is adapted from Ref. 48)

Similar C−H activation of dimethyl malonate (36) can be promoted by palladium/dppf system, which catalyses benzylic substitutions of benzyl esters with malonates and amines.49 Palladium complexes with a diphosphine ligand show the best performance (Equation 2.20). A Pd(η3 -benzyl) intermediate, similar to the well documented Pd(η3 -allyl), could also be invoked. Ar

OCO2Me

+

CO2Me CO2Me 36

1 mol%[Pd]-dppf BSA-cat. KOAc THF, 80 °C, 3 h

Ar

CO2Me CO2Me

Ar +

Ar

CO2Me CO2Me

Ar = Ph; 2-MeC6H4

(2.20) Imidates and thioimidates are important precursors of heterocyclic compounds.50 Whitby and Furber et al.51 reported the high yield synthesis of imidates/thioimidates ArC(Y)=NR (37; Y = OEt, OPh, S(i-Pr)) from a three-component coupling among aryl- or heteroaryl-bromides (ArBr), alkoxides, aryloxides, thioalkoxides and isocyanides, promoted by palladium(II)/dppf.

The Coordination and Catalytic Chemistry of dppf

2.4.1.4

77

Non-Palladium Catalysed C−C Coupling Reactions

Many d-block metals apart from palladium have found their roles in different types of cross-couplings. Among them, rhodium seems to be very popular. For example, cationic rhodium(I)/dppf-catalysed regio- and diastereoselective intermolecular [4 + 2] carbocyclisation of 5-trimethylsilyl-4-pentynals (38) with electron-deﬁcient alkenes leads to cyclohexanones (40 and 41) (Table 2.9).52 Every carbocyclisation produces a single oleﬁn isomer. The regioselectivity of alkene insertion depends on the nature of alkene. A key intermediate in this intermolecular [4 + 2] carbocyclisation is a ﬁvemembered acylrhodium intermediate. Table 2.9 Rhodium-catalysed regio- and diastereoselective [4 + 2] carbocyclization of 5-trimethylsilyl-4-pentynals with electron-deﬁcient alkenes O

OH

O

E H

E +

R

10 %[Rh(dppf)]BF4 CH2Cl2,25 °C, 23–70 h

SiMe3 38

+

Me3Si

H

E

Me3Si

40

39

Entry

R

R

38R

41 de (%) (41)

Yield (%) (40:41)

39E

H

1

Me

CO2nBu

76(81:19)

>99

2

Me

CO2Cy

79(73:27)

>99

3

Me

CO2tBu

86(28:72)

>99

4

nBu

CO2nBu

77(84:16)

>99

5

nBu

CO2tBu

77(32:68)

>99

A rhodium-based catalyst system of [RhCl(cod)]2 /dppf/Na2 CO3 promotes coupling between 2-hydroxybenzaldehydes with various internal and terminal alkynes. This is accompanied by cleavage of the C−H bond of the aldehyde to give the corresponding 2-alkenoylphenols (42) in good yields (Equation 2.21).53 OH

OH X

H O

+

R1

R2

[RhCl(cod)]2/dppf

R1 (or R2)

X

Na2CO3

R2 (or R1) O 42

(2.21) Cooperative use of rhodium(I) with a second organometallic reagent is common. An addition reaction of 1-alkenylboronic acids or their esters to aldehydes or ketones can be promoted by catalytic amounts of [RhCl(dppf)] or [Rh(OH)(dppf)] in aqueous

78

Ferrocenes: Ligands, Materials and Biomolecules

MeOH or DME in the presence of potassium hydroxide.54 The utility of this protocol has been demonstrated in the corresponding intramolecular reaction that leads to cyclic homoallylic alcohols. Other boron reagents such as 2-(3-methoxypropenyl) benzo[1,3,2]dioxaborole (43) have been applied to oleﬁn addition, such as vinylcyclopropanation.55 Thus, 3-exo-vinyltricyclo[3.2.1.02,4 ]octane (45) can be prepared from 43 and norbornene (44) when catalysed by [RhCl(C2 H4 )2 ]2 and dppf (Equation 2.22). The pathway involves an norbornyl-rhodium(I) intermediate and multiple carborhodation steps, including an intramolecular 3-exo-trig cyclisation, and a termination step with β-oxygen elimination, producing [RhI (OMe)]. MeO

[RhCl(C2H4)2]2/dppf

+ R

BCat

R

43

NEt3, H2O, dioxane, 100 °C, 3 h

R R

44

45

(2.22) Another methodology on cyclisation catalysed by copper(II) and supported by dppf is found in the construction of carbocyclic and heterocylic rings.56 Intramolecular reductive aldol reaction of α,β-unsaturated esters with ketones gives ﬁve and sixmembered β-hydroxylactones (46) in high stereoselectivity (Equation 2.23).57 Use of chiral non-racemic diphosphines renders the cyclisation enantioselective. R2

R2

O

R1

Cu(OAc)2 • H2O, dppf

O R3

n

TMDS, THF, r.t.

O

O

R1

O R3 HO 46

(2.23)

n

The same catalytic combination using copper(II) and dppf has been applied to reductive aldol cyclisation of α,β-unsaturated amides with ketones to yield 4hydroxypiperidin-2-ones (47) in high diastereoselectivity (Equation 2.24).58 When used in combination with proline-catalysed asymmetric Mannich reactions, this methodology enables the enantioselective synthesis of more highly functionalised piperidin-2-ones and hydroxylated piperidines. O O CO2Et O N PMP

Cu(OAc)2, dppf Me H2O, TMDS, THF, r.t.

Me Me HO

N

PMP CO2Et

BH3•THF THF, reflux

Me

N

PMP OH

Me HO 47

(2.24) Organozinc compounds have also been used. They can be prepared from transmetalation of commercial organolithium with zinc(II) chloride. The resultant organozinc attacks the 2-position of 2,4-dibromothiazole to give 2-alkyl-4-bromothiazoles (48) at room temperature in the presence of palladium(0). The use of dppf helps to suppress undesired β-hydride elimination and isomerisation.59

The Coordination and Catalytic Chemistry of dppf

79

Zinc(II) could also cooperate with rhodium(I) to facilitate arylation. Rhodiumcatalysed addition of organometallic reagents to aryl alkynyl ketones for the synthesis of highly substituted indanones has been developed.60 The proposed intermediate of an aryl alkynyl ketone species undergoes hydrorhodation followed by a 1,4-rhodium migration. The key to success requires an optimised choice of ligand (dppf) and solvent (ClCH2 CH2 Cl) (Equation 2.25). O

O +

ArZnCl

R1

[RhCl(C2H4)2]2, dppf ClC2H4Cl, 25 °C, H2O

(2.25) R

Ar

R1

R

Another rhodium/zinc cooperation is found in the cross-coupling reaction between primary alkyl halides bearing β-hydrogens and arylzinc compounds with carbonyl groups such as ester, amide, or ketone at the ortho position.61 It can tolerate a range of functional groups such as ester, nitrile, or acyloxylate moieties on the halides. Arylzinc compounds free of ortho-carbonyl groups react well with ethyl 3-iodopropanoate, which suggests that the essential intramolecular interaction between carbonyl groups and rhodium promotes the reductive elimination (Equation 2.26). R-I or R-Br/Rh-dppf cat.; X = OMe, NMe2, Ph; R = Et, C7H15, Cl(CH2)3, CH2=CH(CH2)3, EtO2C(CH2)3, BzO(CH2)4, NC(CH2)4, etc.

ZnI O

R

(2.26) O

in N,N,N’,N’-tetramethylurea, r.t.−80 °C, 1–72 h

X

X

The catalytic activity of a rhodium complex in cross-coupling between ArZnI and Me3 SiCH2 I has been examined. The rhodium complex, generated in situ from [RhCl(cod)]2 and dppf, shows excellent catalytic activity in the formation of various functionalised benzylsilanes ArCH2 SiMe3 .62 31 P NMR analysis points to a rapid and quantitative transfer of aryl groups from ArZnI to the rhodium-complex to form a Rh–Ar species. Shintani and Hayashi et al.63 have developed a rhodium-catalysed multicomponentcoupling reaction starting with a transmetallation of aryl from zinc(II) to rhodium(I) (Equation 2.27). This reaction possibly proceeds via a carborhodation-oxidative addition-reductive elimination pathway (Scheme 2.5). It is initiated by a RhI −Ar complex formed from transmetalation of an aryl group from zinc to rhodium. Alkyne insertion in 49 to Rh−C would give an intermediate rhodium(I) alkenyl 49a. Intramolecular oxidative addition of the iodoarene would give rhodium(III) species 49b, which regenerates the rhodium(I) through reductive elimination giving product 50. R2

R1

[RhCl(C2H4)2]2

I + N Me

O

R2ZnCl

R1

10 mol% Rh, 11 mol% dppf dioxane 40 °C, 20 h

O N Me

(2.27)

80

Ferrocenes: Ligands, Materials and Biomolecules R I RhI-Ar N

O ArZnCl

Me

I I Rh N 49a

Ar RhI-I

R O

Me reduction elimination I oxidation addition

Rh

Ar

Ar

R

III

N

R

O N

O

Me

Me 49b

50

Scheme 2.5 Proposed catalytic cycle for the rhodium-catalysed multicomponent coupling involving intramolecular alkyne insertion into Rh−C bond (oxidative addition/reductive elimination) (The scheme is adapted from Ref. 63)

2.4.1.5

Main-Group-Metal-Assisted Cross-Coupling Reactions

Numerous cross-coupling reactions could also be assisted by the s- and p-block metals in conjunction with dppf. Co-presence of catalytic amount of palladium compounds provides the needed drive. For example, Mongin et al.64 introduced the ﬁrst cross-coupling using lithium triarylmagnesates. 2-, 3- and 4-Bromoquinolines (51) are converted to the corresponding lithium tri(quinolyl)magnesates (52) on treatment with LiMgBu3 in THF in a ‘one-pot’ procedure. In the presence of [Pd(dba)2 ]/dppf, the resulting organomagnesium intermediates cross-couple with heteroaryl bromides and chlorides to afford functionalised quinolines (53) (Table 2.10). 2-Bromo substrates generally give better yields. Germylation of aryl or vinyl halides can also be catalysed by Pd(OAc)2 /dppf.65 Reaction of aryl halides with tri(2-furyl)germane (54) gives aryltri(2-furyl)germanes (55) in good yields (Table 2.11). Electron-rich aryl iodides give the corresponding arylgermanes in relatively high yields (Table 2.11, entries 1–3). Vinyl iodides and bromides as well as 2-bromopyridine are also active. Whereas 2-iodotoluene gives low yield (entry 4), aryl bromide does not undergo the coupling at all (entry 7). Further cross-coupling of the aryltri(2-furyl)germanes with aryl halides provides a

The Coordination and Catalytic Chemistry of dppf

81

Table 2.10 Cross-coupling reactions of organomagnesium 52 with heteroaryl bromides Br

N

0.35 eq. Bu3MgLi THF, −10 °C, 2.5 h

N

1) Pd(dba)2/dppf RBr, THF, r.t., 18 h

Li Mg *

R

2) H2O

3

N

52

51 R

Entry

53 Product

Yield (%)

56

1

N Br

N N Br

Br

53

2 Br

N

N N

51

N

3 Br

N

N

Table 2.11 Palladium-catalyzed germylation of aryl or vinyl halides

GeH O 54 Entry

+

RX

Pd2(OAc)2, dppf Cs2CO3

3

DMF, r.t.

GeR O

3

55 RX

Yield %

1

4-MeOC6H4I

80

2

3-MeOC6H4I

88

3

4-MeC6H4I

85

4

2-MeC6H4I

43

5

6-Iodo-dodec-6-ene

83

6

2-Iodo-prop-2-en-1-ol

59

7

4-MeOC6H4Br

0

8

2-Bromo-propene

70

9

1-Bromo-propene

54

10

a-Bromostyrene

49

11

(E)- b-Bromostyrene

74

12

2-Bromopyridine

85

82

Ferrocenes: Ligands, Materials and Biomolecules

facile synthetic route for unsymmetrical biaryls from two different aryl halides. This is achieved through the use of catalytic [Pd2 (dba)3 ] supported by (Bu4 N)F. Hypervalent organobismuth compounds such as 6-tert-butyl-5,6,7,12-tetrahydrodibenz[c, f ][1,5]azabismocines (56) also facilitate coupling with aryl and alkenyl chlorides when catalysed by Pd(OAc)2 /dppf.66 2.4.1.6 2.4.1.6.1

C–X Coupling Reactions C–B coupling reactions

A series of electron-rich aryl boronates have been prepared under microwave conditions from the palladium-catalysed reactions of aryl bromides with bis(pinacolato) diboronate, [B(pin)]2 .67 This method could provide a good entry to a range of organoboron reagents suitable for Suzuki-type couplings. The use of microwaves signiﬁcantly enhances the reaction rates (reducing reaction duration from days/hours to minutes) when compared to the conventional thermal methods. 2.4.1.6.2

C–N coupling reactions

Among the coupling reactions studied, C–N coupling, especially amination and hydroamination, probably attracts the most attention outside the C–C coupling domain. Dppf is also a very popular ligand choice. For example, aryl chlorides can be converted to aniline derivatives (57) under catalytic amounts of [Ni(cod)2 ] and dppf or 1,10phenanthroline in the presence of NaO(t-Bu) (Equation 2.28).68 Both electron rich and electron poor aryl chlorides, as well as chloropyridine derivatives are suitable. Primary and secondary amines can also be used to give the desired aryl amine products. The potential of nickel-catalysed Ar−Cl activation towards C−C and C−N bond formation is promising.

Ar-Cl + HN(R)(R’)

cat. Ni(cod)2, dppf, NaO(t-Bu) toluene, 70–100 °C

Ar-N(R)R’

(2.28)

57

A convenient one-step synthesis of 2-[(nitroaryl)amino]-3-chloro-1,4-naphthoquinones (60) has been reported.69 This is based on a direct amination of 2,3dichloro-1,4-naphthoquinone (58) with nitro-substituted aryl amines (59) catalysed by [PdCl2 (dppf)]/dppf/NaO(t-Bu) (Equation 2.29). Traces of the 2,3-di[(nitroaryl)amino]1,4-naphthoquinones are also formed. One-step synthesis of N -substituted 6-amino-2,2 :6 ,2 -terpyridine (61) and 6,6 diamino-2,2 :6 ,2 -terpyridine (62) can be realised in palladium-catalysed amination of bromo-substituted terpyridines with various amines.70 To achieve reasonable yields, it is important to use an appropriate chelating phosphine ligand (such as dppf) in this type of amination since terpyridines are strongly coordinative. Many functional amination products can be made using similar approaches since the catalytic system is amenable to a range of substrates. Another example is the preparation of 1,3-bis (5-diarylaminothiophen-2-yl)isothianaphthenes (65) from 1,3-bis

The Coordination and Catalytic Chemistry of dppf R4 O

NH2 R4

Cl

+ R2

O

R3

58

59

R3

O R1

Cl

83

PdCl2(dppf)/dppf

R2

NH R1

NaO(t-Bu), toluene, 80 °C

Cl O

60 Major + R3 4 O R

O

R2 NH

R1

NH

R1

R4

Trace

R2 R3

(2.29) (5-bromothiophen-2-yl)isothianaphthene (64) and diarylamines (63) (Scheme 2.6).71 The presence of [PdCl2 (dppf)]/dppf enables an isothianaphthene group to be captured by the bis(arylamino)oligothiophenes, forming novel hole transport dyes that exert high molar extinction coefﬁcients. Triarylamines with different substituents can also be similarly prepared. A good illustration is a one-pot C−N bond-forming reaction, whereby two aryl bromides are sequentially added to an arylamine in the presence of [Pd2 (dba)3 ]/dppf/NaO(t-Bu).72 This methodology has been applied to the synthesis of 4,4 -bis(m-tolylphenylamino) biphenyl (66) derivatives that are useful as the hole transport component of vapourdeposited organic light-emitting diodes. Application of similar methods can help to achieve a variety of cyclisation reactions. A good illustration is the development of the synthesis of systems, in which dihydroazaphenanthrene is fused to macrocycles (67), medium-ring heterocycles (69) and 1,4-benzodiazepine-2,5-diones (70).73 Catalytic palladium and copper are used differently to develop a divergent synthesis of two different heterocyclic scaffolds from the same starting materials. Use of an acetate source with palladium(II) drives linear amide 68 to 69 in a domino intramolecular N-arylation/C−H activation/aryl–aryl bond forming process whereas copper(I) iodide leads to 70 in an intramolecular N-arylation reaction (Scheme 2.7). There are many variations to this approach. Instead of the standard aryl or alkyl primary amines, the use of a range of functional secondary amines could also be envisaged. A good example of such is represented by the coupling of lactams (71)

84

Ferrocenes: Ligands, Materials and Biomolecules

Br

Th

Br

63a,b

+

Ar1 NH

Ar2 64a,b

Thiophene

Ar1

dppf, t-BuONa, o-xylene, 125 °C

S

S

Br

S

Ar1 N Th

N

Ar2

Ar2

65a,b,c Product

Amine OMe

Br

PdCl2(dppf)

OMe

MeO

NH

S

N

S

N

S

63a 64a

S

Br

S

Br

S

OMe

MeO

OMe

65a

NH

S

N

S

N

S

63a 64b

Br

S S

Br

65b

NH

S

N

S

63b 64b

N

65c

Scheme 2.6 Synthesis of 1,3-bis(5-diarylaminothiophen-2-yl)isothianaphthenes

to bromobenzenes (Equation 2.30).74 The reaction proceeds well with a variety of lactams as well as electron rich and electron poor bromobenzenes. The overall reaction efﬁciency also depends on the ring size of the lactam. Formation of the ﬁve-membered N -aryl lactams is favourable, whereas the four, six and seven-membered ring systems are less reactive. O

O

R NH n

71 n = 1–4

+ Br

Pd(OAc)2, dppf NaO(t-Bu), toluene 120 °C

R N n

(2.30)

The Coordination and Catalytic Chemistry of dppf

I (a)

I H N

O

PdCl2(dppf)

H N

X n

DMSO, KOAc, 120 °C

N

n

m

O

85

H N

O

X m

O

I

R1

H N

(b)

O PdCl2(dppf)

N O

67

Me

O

N

DMSO, KOAc, 120 °C

R1

I

N

68 1

Me

69 O

R = H, Me, Bn, iPr, Me2CHCH2

I I (c)

O

H N

CuI DMSO, K2CO3, 110 °C

N O

I O

N

Me N O

Me

70

Scheme 2.7 (a) From linear diamide to macrocycles 67; (b) Palladium-catalysed synthesis of polyheterocycle 69; (c) Copper-catalysed intramolecular N -arylation

Substituted hydrazines can also be suitable amination sources. For example, coupling of N -Boc arylhydrazine, (4-O2 NC6 H4 )N(Boc)NH2 (72) with 4-nitrophenyl bromide (73) has been used as a model for amination, amidation and carbamation.75 However, only aryl halides with electron-withdrawing groups are effective under the conditions used. When the use of (substituted) hydrazine is combined with other aromatic systems, including the non-halides, many new materials can be prepared conveniently. For example, cyclic and acyclic β-bromovinyl aldehydes (74) are cyclised with an array of arylhydrazines (75) in the presence of palladium(II)/dppf/NaO(t-Bu) to give 1-aryl1H -pyrazoles (76) in reasonable yields (Equation 2.31).76 CHO +

H2N

NHAr

Br 74

Pd(OAc)2, dppf NaO(t-Bu), toluene, ∆ , 24 h

N N

75

76

Ar

Ar = Ph; 2-MeC6H4; 3-MeC6H4; 4-MeC6H4; 4-MeOC6H4; 2-CF 3C6H4; 4-NO2C6H4

(2.31)

86

Ferrocenes: Ligands, Materials and Biomolecules

Use of isonitriles with amines presents another interesting approach. Insertion of the former to aryl bromides that carry pendant amine or alcohol groups on the ortho position leads to cyclic amidines or imidates.77 This highly efﬁcient one-step and concurrent intramolecular reaction with intermolecular as well as C–N with C–C coupling is promoted by PdCl2 /dppf catalyst (Equation 2.32). HN(t-Bu) Br

n

NH2

+ (t-Bu)NC

PdCl2/dppf

(2.32)

N

Cs2CO3/toluene, 1 h, ∆

n

n = 1,2

Organometallic complexes immobilised on a thin ﬁlm of supported ionic liquids can be used to prepare novel bi-functional catalysts combining soft Lewis acid and strong Brønsted acid functions. These materials effectively catalyse aniline addition to styrene, yielding the Markovnikov product (77) under kinetic conditions and mainly the thermodynamic anti-Markovnikov product (78) (Equation 2.33).78 +

Pd(CF3CO2)2(dppf)

H2N

TfOH

CH3 HN 77 +

(2.33)

NH 78

Catalytic hydroamination of oleﬁns provides a powerful means to synthesise amines with regio- and enantio-control. The ﬁrst reported transition metal-catalysed oleﬁn hydroamination with alkylamines gives products with Markovnikov regiochemistry (e.g. 79 in Equation 2.34).79 Reaction of the (η3 -phenethyl)palladium intermediate with the more basic alkylamine leads to oleﬁn elimination that competes with amine addition to form the hydroamination product. The palladium-catalysed hydroamination of styrene with aniline exhibits a substantial 13 C isotope effect at the benzylic carbon.80 It points to a rate-determining nucleophilic attack of amine on a (η3 -phenethyl)palladium complex. The ease of palladium displacement after reversible hydropalladation of the alkene determines the selectivity. Dppf serves a functional role in the stabilisation of such intermediate.

+

HN

O

5 mol% Pd(O2CCF 3)2; 10 mol% dppf; 20 mol% CF 3SO3H

N

dioxane, 120 °C, 24 h

(2.34) O

79

The Coordination and Catalytic Chemistry of dppf

87

The scope of the palladium-catalysed hydroamination of vinylarenes has been signiﬁcantly expanded. The formation of the C−N bond is evaluated from the reaction between the dppf-ligated palladium(II) intermediate (80) and two equivalents of morpholine (81) in the presence of added dppf in an attempt to trap the palladium(0)-dppf product (Equation 2.35).

p-Tol

N

+ Pd(dppf)2 O

(dppf)Pd −

13%

+

+ HN

OTf

Me 80

dppf

O

dioxane, 110 °C, 2 equiv. 5 min 81 + O

p-Tol 84%

NH2 (OTf) + Pd(dppf)2

(2.35) Palladium-catalysed N -arylation of 2-oxazolidinones with aryl bromides gives 3aryl-2-oxazolidinones (82) in good yields (Equation 2.36).81 The reaction outcome is affected by the nature of aryl bromides, phosphine ligands, bases and solvents. Use of a catalytic mixture comprising Pd(OAc)2 , dppf and NaO(t-Bu) favours aryl bromides with a para electron-withdrawing substituent whereas [Pd2 (dba)3 ], Xantphos and NaO(t-Bu) favour neutral and mild electron-rich/poor bromides. R1

R2

R1

R2

Pd(OAc)2/dppf, NaO(t-Bu)

ArBr + HN

O

toluene, 120 °C

N Ar

O

O

(2.36)

O 82

Reaction of polyamines with primary amino groups (e.g. ethane-1,2-diamine and propane-1,3-diamine) with ArBr or ArI proceeds selectively to yield N -monoarylsubstituted derivatives of polyamine in the presence of [PdCl2 (dppf)] and NaO(t-Bu).82 This methodology provides a convenient method for arylation of di-, tri- and tetra-amines. The palladium-catalysed reactions of propane-1,3-diamine and 3,3 diaminodipropylamine with the more reactive 1-bromonaphthalene can be used for the preparation of sym-dinaphthyl derivatives of these amines. This palladium-catalysed arylation reaction has been successively used in the preparation of a variety of aryl substituted polyamine compounds (Equation 2.37). Diarylation of di- and triamines can also be achieved (Equation 2.38). RNH Br

(NR)x

NH2

PdCl2(dppf)/dppf, NaO(t-Bu),dioxane

RNH

(NR)x

HN

(2.37)

88

Ferrocenes: Ligands, Materials and Biomolecules

H 2N

NH2

NH NH

Br PdCl2(dppf)/dppf, 2

(2.38)

NaO(t-Bu), dioxane NH HN

N H

HN

NH2

NH2

A catalytic mixture containing [Pd2 (dba)3 ]•CHCl3 , dppf and acetic acid can also promote the addition of protected and/or functionalised amines to allenes with high regio- and stereoselectivity.83 With the use of different protected amines, which can be easily deprotected, this method could be useful for allylamine synthesis (e.g. 83 in Equation 2.39). The mild conditions and easy workup are among the advantages in these hydroamination reactions. R C

+ H

NXY

H

Pd2(dba)3/dppf AcOH, THF, 80 °C

a: X = Y = CH2CO2Et b: X = Y = CH2Ph c: X = Y = Ph d: X = b-Naphthyl, Y = Ph

R

NXY 83

(2.39)

Hydroamination can also be carried out intramolecularly.84 For example, allenes with amine or sulfonyl amide groups at the terminus of the carbon chain afford the corresponding 2-vinylpyrrolidines and 2-vinylpiperidines in the presence of (η3 C3 H5 )PdCl]2 /dppf under weakly acidic conditions (Equation 2.40). This new type of hydroamination is expected to proceed through the insertion of a Pd−H bond to an allenic double bond. C

NH R

C

[(h3−C3H5)PdCl]2/dppf AcOH, THF 70 °C, 5 h

Pd-H insertion HNPd R

H

PdLn N R

–Pd

N

R

(2.40)

The Coordination and Catalytic Chemistry of dppf

89

The cyclisation of an allenylamine having a shorter carbon chain has also been examined. Use of 84 leads to 1,3-diene 85, instead of the expected endo-cyclised product (Equation 2.41). The product formed suggests a (π-allyl)palladium intermediate is generated from ‘hydropalladation’ of 84. The stabilising effect of dppf towards such an intermediate seems obvious. [(h3-C3H5)PdCl]2/dppf NHTs 84

NHTs

AcOH, THF 70 °C, 5 h

(2.41)

85

Hartwig et al. reported the use of [Pd(dba)2 ] with dppf to promote the amination of aryl triﬂates.85 The ability of dppf complexes of palladium to catalyse the amination of a variety of aryl triﬂates with anilines and alkylamines has been investigated (Equation 2.42). OTf + HNRR’ R

Pd(dba)2/dppf NaO(t-Bu)

NRR’

(2.42)

R

Protected pyridylhydrazine derivatives (86) can be prepared in a one-step palladiumcatalysed amination reaction supported by chelating phosphines such as dppf (Equation 2.43).86 2-Pyridyl chlorides, bromides and triﬂates are effective electrophiles whereas di-tert-butyl hydrazodiformate is an excellent hydrazine substrate. Deprotection can be accomplished easily under mild conditions. This catalytic amination provides a direct route to protected bifunctional hydrazinopyridine linkers, which are useful for metal-bioconjugate syntheses. O

R

O

R + N

X

HN

O(t-Bu)

Pd2(dba)3/dppf

HN

O(t-Bu)

Cs 2CO3, toluene

N

N

O(t-Bu)

HN

O(t-Bu)

O 86

O 20 % HCl.EtOH

R

N

NH NH2

2HCl

(2.43) Reaction of dimethyl (Z)-2-butenylene dicarbonate (87) with primary amines in the presence of [Pd(η3 -C3 H5 )Cl]2 and dppf gives vinyl-oxazolidone compounds (88) (Equation 2.44).87 Cyclisation of diamide 89 under modiﬁed Miyaura–Suzuki conditions catalysed by [PdCl2 (dppf)] gives polyheterocycle 90 instead of macrocycle

90

Ferrocenes: Ligands, Materials and Biomolecules

91 (Equation 2.45).88 This domino process produces 90 from linear diamide 89. It involves a sequence of intramolecular Buchwald–Hartwig amination, C−H activation and aryl–aryl bond formation. Accordingly, the proline derivative 92 can be cyclised to give the desired pentacyclic compound 93 (Equation 2.46).

MeO2CO

OCO2Me

+

R NH2

[Pd(h3-C3H5)Cl]2/dppf

O

CH2Cl2

87 I

N R

(2.44)

O 88

O

H N

N H

O

I 89

DMSO, KOAc, 120 °C PdCl2(dppf)

(2.45)

O

H N N O H N

H N

O

O 90

91

I PdCl2(dppf)

O O

DMSO, KOAc, 120 °C

HN

(2.46)

N

N I

92

O

N

O 93

The Coordination and Catalytic Chemistry of dppf

91

Similarly, an intramolecular cyclisation of a β-lactam with a propargyl gives carbapenam in the presence of catalytic [Pd2 (dba)3 ] and a bidentate ligand (Equation 2.47).89 For instance, phosphate 94 affords the desirable β-methylcarbapenams 95 and 96. The mechanism involves the formation of σ -allenylpalladium complex from palladium(0) and propargyl phosphate and subsequent attack from the lactam nitrogen on the central sp carbon of the ligand. OSi H

H

OSi H

H

Pd2(dba)3/dppf NH

H

MeCO2Na, THF, 40 °C, 22 h

O

N O

OPO(OEt)2

Si = TBDMS

OR

95

94

+ OSi H

H

N O OR 96

(2.47) Pharmaceutical research in carbapenam antibiotics has been gathering pace since the discovery by Merck’s group of thienamycin, the ﬁrst naturally occurring carbapenam antibiotic.90 β-Lactam reacts with propargyl phosphate and sodium acetate when catalysed by [Pd2 (dba)3 ] and a bidentate ligand such as dppf giving carbapenam in high yield. The lactam nitrogen attacks the central carbon of a (η3 -propargyl)palladium complex, which is generated from palladium(0) and propargyl phosphate. The ligand plays an important role in determining the ring size of the cyclised compound. Carbapenam (97) which has a β-methyl group at the 1-position has been constructed (Equation 2.48). 5 mol% Pd2(dba)3, 20 mol% dppf, 1.5 equiv. MeCOONa

OTBS H H

O

THF, 40 °C, 22 h

NH R

OTBS H H OCOMe

N O

(2.48)

97

R = OP(O)(OEt)2

Although palladium is commonly used in this type of hydroamination, other metal complexes such as rhodium(I) could also be used in conjunction with dppf. For

92

Ferrocenes: Ligands, Materials and Biomolecules

example, the hydroaminomethylation91 of arylethylenes with anilines proceeds under mild conditions in the presence of [Rh(cod)2 ][BF4 ]/dppf to give the corresponding branched amphetamine derivatives in good selectivity and yield (Equation 2.49). The 4-position of substituents on the styrene imposes only minimum electronic inﬂuence on the yield and selectivity.

+

Ar

R2R3NH

[Rh(cod)2]BF4, dppf, 10 mol% HBF4 60 °C, 30bar (CO/H2=1:5)

Ar

NR2R3

+

NR2R3

Ar

(2.49) A simple colorimetric assay of various metal catalysts showed that a mixture comprising [Ni(cod)2 ], dppf and acid is a highly active catalyst system for the hydroamination of dienes by alkylamines to form allylic amines (Equation 2.50).92 The reaction scope is broad enough to support reactions between various primary/secondary alkylamines and 1,3-dienes. The thermodynamics favors the reaction of a nickel(0) complex with allylic amine in the presence of acid to give a nickel(II)-allyl intermediate. These results led to the discovery that nickel and some palladium complexes can catalyse amine exchanges in allylic amines. Ni(cod)2/dppf, CF3CO2H

NR1R2 +

HNR3R4

NR3R4 +

THF, r.t., 24–72 h

HNR1R2

(2.50)

Catalytic amination of 1,3-dibenzyl-5-iodouracil leads to heterocyclic 5-aminouracil derivatives under the catalytic system [Cu(OTf)]2 •PhH/phen/dba/Cs2 CO3 .93 The copper-based system is, however, ineffective for the coupling of 1,3-dibenzyl-5iodouracil (98) with imidazole and 2-aminothiazoline. Instead, such a coupling can be facilitated by [Ni(cod)2 ]/dppf/NaO(t-Bu) (Equation 2.51). These are the ﬁrst examples of catalytic amination with a uracil substrate. O I

N S

NH2

Ni(cod)2/dppf

NBz

+ N Bz

O

NH3, NaO(t-Bu), toluene, 100 °C, 24 h

H N

N S

O NBz N Bz

(2.51)

O

98

2.4.1.6.3

C−P coupling reactions

The synthesis of a range of 5 -deoxy-5 -methylidene phosphonate-containing thymidine dimers (101) through C−P coupling has been achieved.94 The optimised catalytic mixture comprising Pd(OAc)2 , dppf and propylene oxide is effective towards condensation from a range of H-phosphonates (99) and vinyl bromide (100) (Equation 2.52). This method allows rapid access to a range of nuclei acids with modiﬁed backbone.

The Coordination and Catalytic Chemistry of dppf

93

O HN O N TBSO

N

O

O

P

MeO

N

HN

N

N TBSO

O

N

O

N N O

H O

99 Pd(OAc)2, dppf, THF

+

MeO

P

O

NH O

reflux, propylene oxide

N O

O Br

TBDPSO

NH

101

N

O

O TBDPSO 100

(2.52) 2.4.1.6.4

C−O coupling reactions

An intramolecular palladium-catalysed coupling reaction of an aryl halide with an alcohol is used to prepare the oxazepine ring systems containing pyridazinone moiety (102).95 The best conditions for this reaction requires the use of catalytic Pd(OAc)2 with dppf and potassium carbonate (K2 CO3 ) at 80◦ C in toluene (Equation 2.53). The product has biological and medicinal potential. The proposed mechanism involves oxidative addition of the aryl bromide, followed by the generation of the palladium oxametallacycle and C−O bond forming reductive elimination. The [PdII ArBr] intermediate can be generated independently but the oxametallacycles could not be isolated or observed. Br

R

Pd(OAc)2, dppf

N

N

OH

N

O

Base, solvent, ∆ THP

R O

N

N N

O

THP

O

O 102

(2.53) Dppf has found its role in promoting the synthesis of natural products containing furan rings. Substituted 2,3-dihydrofurans and benzofurans can be prepared from substituted propargylic carbonates (103) and nucleophiles (Equation 2.54).96 An intramolecular

94

Ferrocenes: Ligands, Materials and Biomolecules

substitution using a nucleophilic phenoxy group could also yield the corresponding dihydrofuran. OH

OCO2Me

5 mol% Pd2(dba)3 • CHCl3, 20 mol% dppf

103 +

R

O O

(2.54)

dioxane, 60 °C R

HO

R = 2-OMe; 4-Me; 2,4,6-trimethyl; H; 1-Naphthyl; 4-Cl; 4-F; 4-acetyl

2.4.1.6.5

C−S coupling reactions

A rather uncommon palladium(0)/dppf catalysed C−S coupling is witnessed in the reaction between thiols and aryl or vinyl halides.97 Acid condensation between a cysteine derivative 104 and the protected (4-iodophenyl)alanine (105) under optimised conditions gives the pseudodipeptide 106 in 67 % yield (Equation 2.55). Similar coupling with various vinyl and alkynyl halides 107 also takes place (Equation 2.56). SH Boc

Boc N H

COOEt

HN

104

Pd2(dba)3, dppf, NEt3, acetone, reflux

+

EtOOC

S CO2Me HN

I

Boc 106 Boc

N CO2Me H 105

(2.55) R1

104

R1

R3

R2

X

+ 107

R3 Pd2(dba)3, dppf,

R2 S

‘base’ Boc

N H

(2.56)

COOEt

The C−S bond forming mechanism of the palladium-catalysed coupling reaction of aryl or vinyl halides with a thiol derived from a protected chiral cysteine 108 (RSH) has

The Coordination and Catalytic Chemistry of dppf

95

been proposed (Equation 2.57).98 Oxidative addition would generate [PdI(Ph)(dppf)] (109), which then exchanges with RSH to receive the thiolate. Reductive elimination of [PdPh(SR)(dppf)] would deliver the coupling product PhSR. An interesting intermediate prior to the exchange is the thiol complex with a dangling dppf, [PdI(Ph)(SHR)(η1dppf)] (110) (31 P NMR evidence). Dppf in such a monodentate mode is not common but was established about two decades ago.99 SH + Ar-X or Vinyl-X BocHN

COOEt

S-Ar

[Pd] Base

S-Vinyl +

BocHN

COOEt

BocHN

COOEt

108

(2.57) Diphenyl disulﬁde and selenide are suitable phenylchalcogenolate sources. An efﬁcient cross-coupling using these reagents on substituted aryl bromides has been developed (Equation 2.58).100 The process, which is catalysed by [PdCl2 (dppf)]/Zn, gives unsymmetrical aryl sulﬁdes and selenides in good to excellent yields under neutral conditions. Br + PhXXPh

PdCl2(dppf)

R

Zn, THF

XPh R

(2.58)

X = S, Se

Another use of disulﬁde is represented in the alkylthiolation reaction of 1-alkynes to give (alkynyl)thioethers R1 C≡CSR2 .101 The catalyst used is [RhH(PPh3 )4 ] but the ligand of choice (dppf) remains the same. In contrast, this method, without the need for any base, would transform triethylsilylacetylene (111) and bis{[2-(tbutoxycarbonylamino)]ethyl} disulﬁde (112) to 1-{2-(t-butoxycarbonylamino)ethyl}2-triethylsilylethyne (113), 2-(t-butoxycarbonylamino)ethanethiol (69 %) (114) and trace amount of 1-{2-(t-butoxycarbonylamino)ethyl}-1-triethylsilylethene (115) (Equation 2.59). Et3SiC CH 111

+

(SCH2CH2NHBoc)2 112

RhH(PPh3)4/dppf acetone, ∆ ,1 h

Et3SiC CSCH2CH2NHBoc 113 +

(2.59)

Et3Si C CH2 BocNHCH2CH2S 115

+

HSCH2CH2NHBoc 114

[RhH(PPh3 )4 ] and dppf also catalyse the regio- and stereoselective additions of diaryl disulﬁdes and diselenides to 1-alkynes giving predominantly (Z)-1-arylseleno2-arylthio-1-alkenes, RC(STol)=CH(SePh) (116), along with all other possible products: RC(SePh)=CH(STol) (117), RC(STol)=CH(STol) (118), RC(SePh)=CH(SePh) (119).102 Addition reaction of dibutyl disulﬁde and dibutyl diselenide with 1-octyne takes place with similar selectivity, albeit lower activity, giving (Z)-1-butylseleno2-butylthio-1-octene. Finally, a mixture comprising Cu(II)/AgClO4 /dppf catalyses the

96

Ferrocenes: Ligands, Materials and Biomolecules

dehydrative glycosylation of tri-O-benzylated 1-hydroxyribofuranose (120) to give the ribofuranoside (121) with high stereoselectivity (Equation 2.60).103 O

O

OH +

BnO BnO 120

ROH (R’SH)

CuCl2/AgClO4/dppf

OH (SR’)

BnO

CaSO4, CHCl3 BnO

OBn

OBn 121

OH O

R = cyclo-C6H11, i-C3H7, n-C8H17, cyclo-C6H11CH2, BnO BnO

OBn OCH3

R’ = Ph

(2.60) 2.4.2 2.4.2.1

Non-Coupling Catalysed Reactions Carbonylation and Carboxylation

Alterman et al.104 have reported a fast method of preparing phthalides (122) using CO generated in situ from DMF and [Mo(CO)6 ] in the microwave-promoted carbonylation–lactone formation reactions catalysed by Pd(OAc)2 /dppf. [Mo(CO)6 ] is particularly useful as it readily generates phthalides as well as dihydroisocoumarin, dihydroisoindone and phthalimide from the corresponding aryl bromide via an efﬁcient CO insertion (Equation 2.61). The electronic effect of the aromatic ring of bromobenzyl alcohols on the product yield was examined. O Br

Pd(OAc)2, dppf, CO-source

O

DMF, 150–160 °C

OH

(2.61)

122 CO-source: Mo(CO)6, DMAP, DIEA, dioxane

In the aminocarbonylation of aryl halides, formamide is used not only as a solvent, but also a source of NH3 and CO.105 Aryl bromides in formamide under microwave irradiation give primary benzamides in good yields when catalysed by Pd(OAc)2 /dppf, imidazole and KO(t-Bu) (Equation 2.62). This in situ CO generation and carbonylation methodology works well with all the aryl halides tested. Imidazole is used as a nucleophilic catalyst. O Br R

+ H

O

Pd(OAc)2/dppf NH2

Imidazole, KO(t-Bu), 180 °C

R

NH2

(2.62)

For the ﬁrst time, carbonylations of unprotected bromoindoles with different Nand O-nucleophiles give excellent yields of indole carboxylic acid derivatives.106 This is achieved by a catalyst combining a palladium source with dppf and Et3 N

The Coordination and Catalytic Chemistry of dppf

97

(Equation 2.63). For example, aminocarbonylation of 4-, 5-, 6- or 7-bromoindole with arylethylpiperazines provides a direct one-step synthesis for central nervous systemactive amphetamine derivatives. Br

N H

O PdCl2(PhCN)2/dppf

+

N

NEt3, CO, 130 °C

N

N H

Ar

NH

R

N Ar R

(2.63) Hydroesteriﬁcation of acenaphthylene with MeOH/CO yields a mixture of acenaphthene-1-carboxylic acid methyl ester (123), 1-methoxyacenaphthene (124) and polyacenaphthylene (125) (Equation 2.64).107 This is catalysed by various palladiumII / phosphine systems, including the one based on dppf. For the palladium/ monophosphine/p-TsOH systems, only ligands with intermediate electronic and steric properties could yield active catalysts that result in satisfactory chemoselectivities. The yields are also inﬂuenced by the reaction concentration in MeOH and the Pd/p-TsOH ratio. Up to 85 % conversion with 93 % chemoselectivity of ester 123 can be achieved. Diphosphines which are used as auxiliary ligands give less efﬁcient catalysts, yielding conversions up to 60 % and chemoselectivity in ester 123 of 85 %. In toluene, both dppf and BINAP [with Pd(OAc)2 ] are more soluble and give ester 123 with higher conversions and chemoselectivities. COOMe Pd(OAc)2/dppf +

CO

p-TsOH, ClC2H4Cl, MeOH, 30 bar CO, 24 h 123 +

(2.64) OMe

+

n 125

124

98

Ferrocenes: Ligands, Materials and Biomolecules

The preparation of pyrimidine-2-carboxylates and 2-pyrimidineacetates from alkoxycarbonylation of 2-chloropyrimidine and 2-(chloromethyl)pyrimidine respectively with CO in the presence of Pd(OAc)2 /dppf has been introduced by Bessardand & Crettaz (Equation 2.65).108 The advantages are the use of readily available starting materials and good yields of the esters. O

O N

O Pd(OAc)2/dppf, CO, ROH

N

O N

CH3COONa

N

(2.65)

Cl

COOR

X = Cl; R = Me, Et, i-Pr, Cy, Bz X = CH2Cl; R = CH3, C3H5, CH(CH3)2

Complexes [Pd(H2 O)2 (P-P)](OTs)2 , where P−P = dppf (126), and 1,1 -bis (diphenylphosphino)octamethylferrocene (dppomf) (127), are effective catalysts for the methoxycarbonylation of ethene.109 The dppf-modiﬁed catalyst produces several low molecular weight oxygenates, ranging from methyl propanoate to alternating oligomers of CO and ethene, whereas the dppomf catalyst yields exclusively methyl propanoate (Equation 2.66). This study suggested that the behavior of the dppf precursor is similar to that of the other palladium(II) diphosphine catalysts. The formation of β-chelate intermediates from either Pd−H or Pd−OMe and their role in controlling the perfect alternation of monomers have been experimentally demonstrated. The selective production of methyl propanoate with the use of the dppomf-modiﬁed catalyst has been attributed to the greater propensity of dppomf versus dppf to form palladium(II) complexes with a dative Fe → Pd bond, which forces the phosphorus atoms to be trans to each other, yielding Pd-acyl species that do not react with ethene in MeOH. β-Chelate species are not formed by the dppomf-modiﬁed catalyst. O

O

O

O OMe

O

126

O O

MeO

OMe

OMe

n

alt-E-CO

O

MeO

+

CO +

OMe

MeOH

O 127 O OMe

(2.66) Palladium(II) complexes with dppf, dppomf, 1,1 -bis(diphenylphosphino)ruthenocene (dppr) and osmocene (dppo) have been used to catalyse the methoxycarbonylation of styrene.110 All the reactions give methyl phenylpropanoates with the linear isomer,

The Coordination and Catalytic Chemistry of dppf

99

methyl 3-phenylpropanoate (128) dominating (up to 85 % regioselectivity). The highest turnover frequency is obtained with the dppr precursor in the presence of ptoluenesulphonic acid co-catalyst (334 mol of styrene converted with 1 mol of catalyst in 1 h). Cyclocarbonylation of 2-iodoanilines with heterocumulenes takes place under catalytic conditions provided by Pd(OAc)2 /dppf to give the corresponding 4(3H )-quinazolinone derivatives (129) in good yields (Equation 2.67).111 Use of isocyanates, carbodiimides and ketenimines as substrates would give 2,4-(1H, 3H )-quinazolinediones, 2-amino4(3H )-quinazolinones and 2-alkyl-4(3H )-quinazolinones, respectively. The reaction probably proceeds via urea-type intermediates, followed by catalytic carbonylation and cyclisation. The addition of dppf or dppb in some cases gives similar isolated product yields. O R

Pd(OAc)2/dppf

I +

R1N=C=CR2R3

NH2

R

N

THF, K2CO3 100 °C, 300 psi CO, 24 h

R1 CHR2R3

N 129

R = H, Cl, CH3, OH, CN

(2.67) Carbonylative ring-forming reactions of 2-iodothiophenol and its derivatives with allenes and CO are catalysed by Pd(OAc)2 /dppf, giving thiochroman-4-ones (130) in good to excellent isolated yields with high regioselectivity (Equation 2.68).112 Regioselective addition of the sulfur function to the more positive end of the allene could account for the catalytic heteroannulation. The proposed pathway involves arylpalladium formation, CO insertion, intramolecular cyclisation and then reductive elimination. The reducing action of CO helps to reduce Pd(OAc)2 to the catalytically active [Pd0 (dppf)]. The one-pot procedure and the mild conditions used make this a useful methodology in the synthesis of thiochromanones.

R

R2

R1

I +

C

SH R3

R4

R1

O Pd(OAc)2/dppf i-Pr2NEt, 400 psi CO, C6H6, 100 °C

R = H, Me, Cl R1, R2, R3 and R4 = H, alkyl, -CO2Et or cycloalkyl

R2

R

S

R4

(2.68)

R3

130

A new type of palladium(0)-catalysed carbon dioxide recycling reaction using allylic carbonates has been developed by Yoshida and Ihara et al.113 A CO2 elimination– ﬁxation process takes place with trans-4-methoxycarbonyloxy-2-buten-1-ols (131) under catalytic palladium(0)/dppf, giving cyclic carbonates functionalised by a vinyl group (Equation 2.69). A variety of allylic carbonates is expected to give cyclic carbonates with high efﬁciencies (Equation 2.70).

100

Ferrocenes: Ligands, Materials and Biomolecules O Pd2(dba)3/dppf

HO

OBz

Pen

O

BSA, dioxane, 50 °C, CO2, 12 h

Pen 131

Pen

O

(2.69)

Pen O

HO

Ph

Pd2(dba)3/dppf

OCO2Me

Ph

O

O

(2.70)

H

dioxane, 50 °C, 4 h

Ni(cod)2 /phosphine (PPh3 or dppf) catalysts promote decarboxylative ring-opening reaction of a wide variety of cyclic carbonates 132 to give ω-dienyl aldehydes 133 in good yields (Scheme 2.8).114

+

Ni0

O 133

Ni(cod)2/dppf O

O

Et3B or Et2Zn O

CO2 O 132

Ni

OH

H

L

Scheme 2.8 Nickel catalyzed fragmentation of 132 to 133 (The scheme is adapted from Ref. 114)

2.4.2.2

Cyanation Reactions

K4 [Fe(CN)6 ] has been used as a relatively safe cyanating agent for aryl halides, under catalytic conditions of Pd(OAc)2 /dppf, to give benzonitriles in good yields (Equation 2.71).115 X + K4[Fe(CN)6] R X = Br, Cl

CN

Pd(OAc)2/dppf

(2.71)

Na2CO3 R

Cyanation of aryl bromides to aromatic nitriles using DMF can be achieved with the use [Pd2 (dba)3 ]/dppf. Zn/Zn(OAc)2 is introduced to protect the active palladium(0)

The Coordination and Catalytic Chemistry of dppf

101

in oxygenated conditions.116 Another use of zinc(II) in cyanation of ArCl is reported by Jin et al. through the use of Zn(CN)2 in conjunction with Pd(0)/dppf/Zn (Equation 2.72).117 Remarkably, both electron deﬁcient and electron rich aryl chlorides can be effectively cyanated. It offers a practical improvement to the Rosenmund–von Braun reaction.118 CF3

CF3 N

O

N

N

O N

Zn(CN)2, Pd2(dba)3/dppf, Zn

(2.72)

DMA, 120 °C, 4 h Cl

CN

F

F

Similar use of Zn(CN)2 towards aryl bromides has been reported by Maligres et al. (Equation 2.73).119 The optimised cyanation features extremely low levels of palladium(0). This is achieved through the use of dppf as ligand which also helps to maintain a clean and robust reaction, giving near-quantitative yields. The more sterically hindered and electron rich l-bromonaphthalene and 2-bromoanisole can also be converted to their corresponding nitriles. Br + N

2.4.2.3

Zn(CN)2

Pd2(dba)3/dppf

NH2

CN

(2.73) NH

N

2

Oxidation Reactions

Singlet oxygen can be prepared in a large scale from ozone when catalysed by dppf (Equation 2.74).120 This is an illustration of the potential of substituted ferrocenes (134) as oxidation catalysts. The process can be run as batch procedure, or in a semicontinuous process. The resultant dppfO2 can be captured and reduced to dppf through hydrogenation, thus allowing a continuous process to be developed. PO3(Ph)2 dppf

2O3

Fe

dppfO2

+

1O 2

(2.74)

PO3(Ph)2 134

Dehydrogenation of 1-phenylethanol can be promoted by the bis(tetraﬂuorosuccinate)bis{Ru(dppf)(H2 O)(CO)} complexes (135) under mild conditions in the absence of acid or base.121 2.4.2.4

Hydrogenation Reactions

Hydrogenation of heteromeric oleﬁnic glycine dimers catalysed by rhodium(I)/dppf represents an efﬁcient route to diastereomerically pure, orthogonally protected

102

Ferrocenes: Ligands, Materials and Biomolecules

diaminosuccinic acid derivatives.122 A cis-selective hydrogenation of oleﬁnic glycine dimers affords either the syn- or anti -isomers of the diamino dicarboxylic acids, depending on the double bond geometry. The products are obtained as racemates. Hydrogenation of each isomer gives the racemic diaminosuccinic acid derivatives (anti )-136 and (syn)-136, respectively, in good yields (Scheme 2.9). tBuO

O HN

HN Ph

O

O

tBuO

OtBu O

8 eq HNEt2

O

MeOH, 70 °C

OCH3

O

OtBu

HN

O

O

NH

H3CO O

O

Ph

[RhCl(cod)]2/dppf 90 bar H2, toluene, 80 °C

tBuO

O HN

O

tBuO

OtBu

O

O

OtBu

HN

O NH

H3CO O

O

NH

H3CO O

Ph

rac-(anti)-136

O

O

Ph

rac-(syn)-136

Scheme 2.9 Stereospeciﬁc hydrogenation of orthogonally protected dimers (The scheme is adapted from Ref. 122)

Instead of using stoichiometric sodium tetrahydridoborate to reduce ketones chemioselectively to racemic secondary alcohols, a homogeneous catalytic method using ruthenium(II) diamine-diphosphine complexes has been developed.123 For instance, isophorone (137) and 3-dimethylaminopropiophenone are effectively hydrogenated using [RuCl2 (dppf)(en)] (en = ethane-1,2-diamine; Equation 2.75). OH

O RuCl2(en)(dppf) KO(t-Bu), i-PrOH, H2, 150 psi

(2.75)

137

The effect of diphosphines, including those that are conﬁgurationally ﬂexible, on the ruthenium(II) catalysed enantioselective hydrogenation of 1-acetonaphthone has been examined in the presence of a chiral diamine.124 They exert signiﬁcant effects on both the activity and enantioselectivity of ruthenium(II)–diamine catalysts. The ligand

The Coordination and Catalytic Chemistry of dppf

103

with the smallest bite angle yields the lowest conversion whereas the largest gives the lowest enantioselection. When dppf is used, an additional feature is evident. Most dppf chelating complexes contain staggered Cp rings, which can lead to two possible enantiomers. Introduction of a chiral diamine such as (R, R)-1,2-diphenylethane-1,2diamine (dpen) would give two diastereomers, one of which could dominate and give rise to an active and enantioselective catalyst (Equation 2.76). Since both dppf and biphen have dynamic and potential axial chirality, they could exert similar chemical control of chirality. Ph2 P

Ph2 P M

M P Ph2

P Ph2 H2N

Ph

H2N

Ph

(2.76) H2 N

Ph2 P

Ph

Ph

H2 N

Ph

N H2

M

M P Ph2

2.4.2.5

Ph2 P

N H2

Ph

P Ph2

Substitution Reactions

Nucleophilic substitution of aryl bromides with sodium aryl oxides to give diaryl ethers (138) can be promoted by [Pd(dba)2 ] and dppf (Equation 2.77).125 Isolated yields of over 90 % can be achieved with electron deﬁcient aryl bromides and electron rich sodium aryl oxides. Electron poor dppf derivatives (CF3 -dppf and MeO-dppf) give higher yields. Similar alkyl and silyl ethers can also be prepared from aryl halides under the catalytic use of [Ni(cod)2 ] with a suitable ligand.126 Use of dppf mediates the formation of (tert-butyl)aryl, methylaryl and (tert-butyldimethylsilyl)aryl ethers from aryl halides and sodium alkoxides or siloxides under mild conditions. The formation of silyl-aryl ethers occurs in higher yields and under milder conditions when catalysed by a combination [Ni(cod)2 ] and dppf when compared to palladium. This catalytic mixture also performs better than the three common palladium catalysts, namely [Pd(dba)2 ], Pd(OAc)2 and [Pd(PPh3 )n ], towards the formation of methyl-aryl ethers. Similar success is also experienced in the preparation of tert-butyldimethylsilyl aryl ether from sodium tert-butyldimethylsiloxide (NaOTBDMS) and electron deﬁcient aryl halides. Reactions of many of these ethers with Brønsted or Lewis acids or ﬂuoride would lead to phenols. Y

Br + NaO

R

Pd(dba)2/dppf or CF3-dppf 100–120 °C, 6−30 h

Y

O

R

138 + NaBr

(2.77)

104

Ferrocenes: Ligands, Materials and Biomolecules

2.4.2.6

Isomerisation Reactions

The catalytic isomerisation of 2-methyl-3-butenenitrile (139) to 3-pentenenitrile (140) is promoted by a catalyst from [Ni(cod)2 ] and dppf (Equation 2.78).127 Solution studies revealed nickel(II)-allyl as the intermediate and suggested C−CN bond cleavage and re-formation. CN

[Ni(cod)2], dppf 100 °C

(2.78)

CN

139

140

The reactions between 1,6-enynes with alkynes to produce cyclohexadiene derivatives are catalysed by [IrCl(cod)]2 /diphosphine (Scheme 2.10).128 Dppe among the diphosphines is most condusive in supporting the cycloaddition. In the absence of alkynes, 1,6-enynes cycloisomerise to (Z)-1-alkylidene-2-methylenecyclopentane derivatives (141). In these ﬁrst examples of highly (Z)-selective cycloisomerisation, dppf gives the best support. [IrCl(cod)]2/dppf benzene, ∆

E E 141

E E

toluene, ∆

Et

E

[IrCl(cod)]2/dppe

E = CO2Et

Et

Et

E

Et

Scheme 2.10 Iridium complex-catalyzed reaction of 1,6-enynes

Facile regioselective isomerisation of terminal propargylic alcohols HC≡CCR1 R2 (OH) to α,β-unsaturated aldehydes R1 R2 C=CHCHO (R2 = CHR3 R4 ) or ketones R3 R4 C=C(R1 )COMe can be promoted by [Ru(η3 -2-MeC3 H4 )(CO)(dppf)][SbF6 ] (142) (Scheme 2.11).129 The same catalyst can be used in the preparation of conjugated 1,3-enynes from dehydration of propargylic alcohols. The same 16 valence electron complex also catalyses the propargylic substitution reaction of 1,1-diphenyl-2-propyn-1-ol with alcohols to produce propargylic ethers (143), and the formal isomerisation of 1,1-diphenyl-2-propyn-1-ol to 3,3-diphenyl-2propenal (144) (Equation 2.79).130 OH H

Ph Ph

H

OR 142 ROH, 75 °C

H

Ph Ph 143

+

Ph

O

(2.79) H

Ph 144

The Coordination and Catalytic Chemistry of dppf OH (a) H

C

C

C

H

R2

C

C

C

C

THF, ∆

R2 C C R1

H

OH H

O

142, CF3CO2H

R1

(b)

105

CHR2R3 1

Me O

142, CF3CO2H

C

THF, ∆

R3 C C R2

R1

R

Scheme 2.11 (a) Isomerization of propargylic alcohols into α ,β-unsaturated aldehydes (Meyer–Schuster rearrangement); (b) Isomerization of propargylic alcohols into α ,βunsaturated ketones (Rupe rearrangement)

2.4.2.7

Oligomerisations and Polymerisations

The ﬁrst selective and catalytic synthesis of fulvenes 146a–e through a [2 + 2 + 1] cyclotrimerisation of alkynes 145a–e has been reported by Yamamoto et al. (Equation 2.80).131 These products are not easily accessible via conventional methods.

R

[PdCl(h3-C3H5)]2/dppf

R

R

toluene, 70 °C, 2–3 days R 146

145

(2.80)

a: R = (CH2)3CH3; d: R = (CH2)9CH3 b: R = (CH2)2CH3; c: R = (CH2)5CH3 e: R = CH2

Kim et al. reported the synthesis of some new rhodium and iridium complexes of ferrocene-based ligands and their catalytic activities towards polymerisation of phenylacetylene.132 A series of cationic palladium(II)-diphosphine complexes generated in situ from the chloride abstraction reactions of [PdCl(CH3 )(dppf)], [PdCl(CH3 ) (dippf)] (dippf = 1,1 -bis(diisopropylphosphino)ferrocene), [PdCl2 (dppf)] and [PdCl2 (dippf)] by AgOTf was developed by Darkwa and Pollack et al.133 The polymerisation is catalysed by dppf and dippf palladium(II)-complexes (Equation 2.81).

Ph

H PA

[Pd(CH3)(P-P)(NCMe)](OTf) or [Pd(P-P)(NCMe)2]2(OTf)2

H

Ph H

Ph

(2.81)

CH2Cl2-CH3CN H

Ph H

Ph

PPA

The synthesis of substituted poly(phenylene)s, in particular poly(1,4-phenylene)s, by palladium-catalysed Suzuki coupling of 2,5-dialkyl-l,4-phenylenediboronic acid 1,3propanediol diester with various aryl dibromides has been described.134 The mild

106

Ferrocenes: Ligands, Materials and Biomolecules

method can be applied to polymerisation of nitro-containing monomers. For example, poly(4,6-dinitro-2 ,5 -dihexyl-3,4 -biphenylylene) (147) is obtained from Suzuki-type reaction catalysed by [PdCl2 (dppf)] in quantitative yield at 37◦ C in THF and aqueous sodium hydrogencarbonate (Equation 2.82). The very mild conditions used allow monomers with a variety of substituents to be used, such as ﬂuoro, nitrile, N, N dimethylamino and triﬂuoromethyl. 2,5-Dibromopyridine can also be used as monomer. C6H13 O B

O B O

O C6H13 + C6H13

Br [PdCl2(dppf)] Br

NO2

n

O

THF, NaHCO3, 37 °C

O2N

B

NO2

O C6H13

O2N

147

(2.82) Copolymerisation of aryl dichlorides with aryl primary diamines can also be promoted by [Ni(cod)2 ] with dppf (Equation 2.83).135 The system can also be adapted to prepare m-polyaniline from m-dichlorobenzene and m-phenylenediamine. Related examples can be found in Chapter 10. n Cl

2.4.2.8

Ar Cl + H2N

Ar’ NH2

Ni(cod)2/dppf Base

H Ar N Ar’

H N

(2.83) n

Asymmetric Ring Opening

The mechanism of the palladium-catalysed ring opening of oxabicyclic alkenes with ZnR2 has been studied (Scheme 2.12).136 The reaction proceeds through an enantioselective carbopalladation of the alkene with a cationic palladium complex instead of a π-allyl mechanism. Carbometalated products have been trapped. The combination of palladium and dialkylzinc is unique in that the latter functions both in the transmetalation and as a Lewis acid in the formation of the reactive cationic palladium species.

2.5

Conclusion

Although a plethora of information is available on the catalytic advantages of dppf, it is still not possible to predict with conﬁdence speciﬁc organic transformations or their reaction conditions that can be promoted best by dppf over other diphosphines. This would only be possible if there was a thorough understanding of a speciﬁc mechanistic pathway, the associated kinetic information and the chemical behaviour

The Coordination and Catalytic Chemistry of dppf

107

OTIPS reflux, 48 h OTIPS (a)

O

OH Me2Zn, PdCl2(dppf), Zn(OTf)2 ClCH2CH2Cl OTIPS

r.t., 20 h

OTIPS

O

(b)

O

OTIPS

OTIPS Me2Zn, PdCl2(dppf), Zn(OTf)2 ClCH2CH2Cl, r.t., 20 h

I2 or D2O O ZnR

O X X = I or D

Scheme 2.12 (a) Ring opening of oxabicyclic alkenes; (b) Involvement of an organozinc Intermediate (The scheme is adapted from Ref. 136)

of discrete catalytic species both in and off (but feeding into) the catalytic cycle. A good example of such is presented in the recent work by Hartwig, Blackmond, Buchwald et al.137 Even commonly used species such as Pd(dppf)2 can be deceptively simple. For example, its catalytic behaviour, e.g. towards oxidative addition, is generally assumed to be similar to other Pd(diphosphine)2 , but there are sufﬁcient evidence to suggest that such assumption is not always correct. The different behavior of Pd(dppf)2 and Pd(binap)2 towards oxidative addition in the amination of aryl halides is one of such examples.138 In fact, since the dissociation of dppf from Pd(dppf)2 to give the catalytically active Pd(dppf) tends to compete with at least two other processes, namely oxidation addition to give [PdX(R)(dppf)] and ligand replacement or attachment by substrates (such as amines, thiols etc.), there could be many inter-dependent pathways operating in a seemingly simple singular conversion. Since the dissociation and re-association of dppf in palladium(0) could be the turnover limiting step, and the oxidative addition could also be the rate-limiting step, a good understanding of the coordination and redox chemistry of dppf complexes would be essential before the value of dppf in catalysis can be fully exploited. Isolation of key reaction intermediates, or at least models to such intermediates, remains an attractive avenue. This often requires clever ligand design and a good understanding of metal-ligand compatibility. Recent use of iminophosphines and other related hybrid ligands is an example of such approach.139

108

Ferrocenes: Ligands, Materials and Biomolecules

The rate-determining step in the advances of dppf in catalysis is, however, not on the number of dppf catalysts that can be designed, but on the ability to translate the coordination chemistry of dppf complexes to its catalytic chemistry. To carry out such translational research of dppf, it is necessary to obtain and couple three sets of highquality data sets – structural, kinetic and mechanistic, as well as computational output. It was over 40 years ago that dppf was ﬁrst discovered. Its coordination, organometallic and catalytic chemistry have advanced almost beyond recognition. However, relatively little has been achieved in connecting the coordination property to the catalytic efﬁciency and then to the mechanistic intricacies. This ‘silo effect’ must be addressed before a quantum leap can be taken in the chemistry of this fascinating metalloligand.

Abbreviations 9-BBN Boc BSA Bz cod Cy DABN dba DCC DDQ DIEA or DIPEA DMA DME DMAP Pen PPy DMF DMSO Dppe Phen TBS or TBDMS TBDPS TfO THF THP TIPS TMEDA TMDS TsO Tol

9-Borabicyclo[3.3.1]nonane tert-Butoxycarbonyl N ,O-Bis(trimethylsilyl) acetamide benzyl Cycloocta-1,5-diene Cyclohexyl Diaminobinaphthyl Dibenzoylacetone N ,N -dicyclohexylcarbodiimed 2,3-Dichloro-5,6-dicyano-1,4-benzoquinone Diisopropylethyl amine Dimethylacetamide 1,2-Dimethoxyethane 4-Dimethylaminopyridine Pentyl 4-Pyrrolidinopyridine N ,N -Dimethylformamide Dimethyl sulfoxide 1,2-Bis(diphenylphosphino)ethane 1,10-Phenanthroline tert-Butyldimethylsilyl tert-Butyldiphenylsilyl Triﬂuoromethansulfonate Tetrahydrofuran Tetrahydro-2H-pyran-2-yl Triisopropylsilyl N ,N ,N ,N -Tetramethylethylenediamine 1,1,3,3-Tetramethyldisiloxane 4-Toluenesulfonate Tolyl

The Coordination and Catalytic Chemistry of dppf

109

References 1. (a) K.S. Gan, T.S.A. Hor, 1,1 -Bis(diphenylphosphinolferrocene – Coordination Chemistry, Organic synthesis and Catalysis, In A. Togni, T. Hayashi, (Eds), Ferrocenes: Homogeneous Catalysis, Organic Synthesis, Materials Science, VCH, Weinheim, Germany (1995), pp. 3–104; (b) S.W.A. Fong, T.S.A. Hor, J. Cluster Sci., 1998, 9, 351–392. 2. (a) L.E. Hagopian, A.N. Campbell, J.A. Golen et al. J. Organomet. Chem., 2006, 691, 4890–4900; (b) F.N. Blanco, L.E. Hagopian, W.R. McNamara et al. Organometallics, 2006, 25, 4292–4300; (c) J.H.L. Ong, C. Nataro, J.A. Golen, A.L. Rheingold, Organometallics, 2003, 22, 5027–5032; (d) C. Bianchini, A. Meli, W. Oberhauser et al. Organometallics, 2005, 24, 1018–1030. 3. (a) C. Nataro, A.N. Campbell, M.A. Ferguson et al. J. Organomet. Chem., 2003, 673, 47–55; (b) S. Li, B. Wei, P.M.N. Low et al. Dalton Trans., 1997, 1289–1293; (c) B. Wei, S. Li, H.K. Lee, T.S.A. Hor, J. Organomet. Chem., 1997, 527, 133–136. 4. (a) S.L. Martinak, L.A. Sites, S.J. Kolb et al. J. Organometal. Chem., 2006, 691, 3627–3632. (b) O.V. Gusev, A.M. Kalsin, P.V. Petrovskii et al. Organometallics, 2003, 22, 913–915. 5. (a) L-T. Phang, K.S. Gan, H.K. Lee, T.S.A. Hor, Dalton Trans., 1993, 2697–2702; (b) V.V. Grushin, Organometallics, 2001, 20, 3950–3961. 6. Z-G. Fang, T.S.A. Hor, Y-S. Wen et al. Polyhedron, 1995, 14, 2403–2409. 7. R. Broussier, E. Bentabet, M. Laly et al. J. Organomet. Chem., 2000, 613, 77–85. 8. (a) K.J. Donaghy, P.J. Carroll, L.G. Sneddon, Inorg. Chem., 1997, 36, 547–553; (b) B.D. Swartz, C. Nataro, Organometallics, 2005, 24, 2447–2451. 9. J-C. Hierso, F. Lacassin, R. Broussier et al. J. Organomet. Chem., 2004, 689, 766–769. ˇ epniˇcka, I. C´ısaˇrov´a, R. Gyepes, Eur. J. Inorg. Chem., 2006, 926–938. 10. P. Stˇ 11. G. Mann, J.F. Hartwig, J. Am. Chem. Soc., 1996, 118, 13109–13110. 12. (a) L.M. Scolaro, M.R. Plutino, A. Romeo et al. Dalton Trans., 2006, 21, 2551–2559; (b) F.N. Blanco, L.E. Hagopian, W.R. McNamara et al. Organometallics, 2006, 25, 4292–4300; (c) A.C. Ohs, A.L. Rheingold, M.J. Shaw, C. Nataro, Organometallics, 2004, 23, 4655–4660; (d) A.R. O’Connor, C. Nataro, A.L. Rheingold, J. Organomet. Chem., 2003, 679, 72–78; (e) G.J. Grant, S.M. Carter, A.L. Russell et al. J. Organomet. Chem., 2001, 637–639, 683–690. 13. S.Y. Ho, E.C-C. Cheng, E.R.T. Tiekink, V.W-W. Yam, Inorg. Chem., 2006, 45, 8165–8174. 14. V.D. de Castro, G.M. de Lima, A.O. Porto et al. Polyhedron, 2004, 23, 63–69. 15. G. Bandoli, A. Dolmella, Coord. Chem. Rev., 2000, 209, 161–196. 16. T.J. Colacot, Platinum Met. Rev., 2001, 45, 22–30. 17. M. Sato, M. Sekino, S. Akabori, J. Organomet. Chem., 1988, 344, C31–C34. 18. M.I. Bruce, P.A. Humphrey, O. Shawkataly et al. Organometallics, 1990, 9, 2910–2919. 19. B.D. Swartz, C. Nataro, Organometallics, 2005, 24, 2447–2451. 20. V.V. Grushin, Organometallics, 2001, 20, 3950–3961. 21. D. Cauzzi, C. Graiff, et al. J. Organomet. Chem., 1997, 536–537, 497–507. 22. D. Cauzzi, C. Graiff, C Massera et al. Dalton Trans., 1999, 3515–3521. 23. F.F. de Biani, C. Graiff, G. Opromolla et al. J. Organomet. Chem., 2001, 637–639, 586–594. 24. M. Careri, C. Graiff, A. Mangia et al. Rapid Commun. Mass Spectrom., 1998, 12, 225–230. 25. T. Avil´es, A. Dinis, J.O. Gonc¸alves et al. Dalton Trans., 2002, 4595–4602. 26. X.L. Lu, J.J. Vittal, E.R.T. Tiekink et al. J. Organomet. Chem., 2004, 689, 1978–1990. 27. R.T. Hembre, J.S. McQueen, V.W. Day, J. Am. Chem. Soc., 1996, 118, 798–803. 28. N. Miyaura, K. Yamada, A. Suzuki, Tetrahedron Lett., 1979, 20, 3437–3440.

110

Ferrocenes: Ligands, Materials and Biomolecules

29. N. Miyaura, A. Suzuki, Chem. Commun., 1979, 866–867. 30. S. Doherty, E.G. Robins, M. Nieuwenhuyzen et al. Organometallics, 2002, 21, 1383–1399. 31. T.J. Colacot, H. Qian, R. Cea-Olivares, S. Hernandez-Ortega, J. Organomet. Chem., 2001, 637–639, 691–697. 32. (a) C.J. Cowden, D.C. Hammond, B.C. Bishop et al. Tetrahedron Lett., 2004, 45, 6125–6128. (b) S.P. Keen, C.J. Cowden, B.C. Bishop et al. J. Org. Chem., 2005, 70, 1771–1779. 33. J. Hashim, T.N. Glasnov, J.M. Kremsner, C.O. Kappe, J. Org. Chem., 2006, 71, 1707–1710. 34. G. Hilt, W. Hess, F. Schmidt, Eur. J. Org. Chem., 2005, 2526–2533. 35. P-E. Broutin, F. Colobert, Org. Lett., 2003, 5, 3281–3284. 36. A.L. Hansen, T. Skrydstrup, Org. Lett., 2005, 7, 5585–5587. 37. A. Jutand, K.K. (Mimi) Hii, M. Thornton-Pett, J.M. Brown, Organometallics, 1999, 18, 5367–5374. ´ 38. C. Herrera-Alvarez, V. G´omez-Ben´ıtez, R. Red´on et al. J. Organomet. Chem., 2004, 689, 2464–2472. 39. K. Olofsson, M. Larhed, A. Hallberg, J. Org. Chem., 2000, 65, 7235–7239. 40. K.A. Horn, Chem. Rev., 1995, 95, 1317–1350. 41. K. Olofsson, M. Larhed, A. Hallberg, J. Org. Chem., 1998, 63, 5076–5079. 42. R.K. Dieter, S. Li, J. Org. Chem., 1997, 62, 7726–7735. 43. Y. Xie, G.K. Tan, Y.K. Yan et al. Dalton Trans., 1999, 773–779. 44. B.C. Hamann, J.F. Hartwig, J. Am. Chem. Soc., 1997, 119, 12382–12383. 45. E. Rossi, A. Arcadi, G. Abbiati et al. Angew. Chem. Int. Ed., 2002, 41, 1400–1402. 46. K. Mikami, K. Aikawa, Org. Lett., 2002, 4, 99–101. 47. E.C. Burger, J.A. Tunge, J. Am. Chem. Soc., 2006, 128, 10002–10003. 48. R. Kuwano, Y. Kondo, Org. Lett., 2004, 6, 3545–3547. 49. R. Kuwano, Y. Kondo, Y. Matsuyama, J. Am. Chem. Soc., 2003, 125, 12104–12105. 50. D.G. Neilson, The Chemistry of Amidines and Imidates, S. Patai, Eds., John Wiley & Sons, Chichester (1975), 385–489. 51. C.G. Saluste, R.J. Whitby, M. Furber, Tetrahedron Lett., 2001, 42, 6191–6194. 52. K. Tanaka, Y. Hagiwara, M. Hirano, Eur. J. Org. Chem., 2006, 3582–3595. 53. K. Kokubo, K. Matsumasa, Y. Nishinaka et al. Bull. Chem. Soc. Jpn., 1999, 72, 303–311. 54. A. Takezawa, K. Yamaguchi, T. Ohmura et al. Synlett , 2002, 1733–1735. 55. T. Miura, T. Sasaki, T. Harumashi, M. Murakami, J. Am. Chem. Soc., 2006, 128, 2516–2517. 56. (a) J. Montgomery, Angew. Chem., Int. Ed. 2004, 43, 3890–3908; (b) A.B. Dounay, L.E. Overman, Chem. Rev , 2003, 103, 2945–2964; (c) E. Negishi, C. Cop´eret, S. Ma et al. Chem. Rev., 1996, 96, 365–393; (d) I. Ojima, M. Tzamarioudaki, Z. Li, R.J. Donovan, Chem. Rev., 1996, 96, 635–662. 57. H.W. Lam, P.M. Joensuu, Org. Lett., 2005, 7, 4225–4228. 58. H.W. Lam, G.J. Murray, J.D. Firth, Org. Lett., 2005, 7, 5743–5746. 59. T. Bach, S. Heuser, J. Org. Chem., 2002, 67, 5789–5795. 60. R. Shintani, T. Hayashi, Org. Lett., 2005, 7, 2071–2073. 61. H. Takahashi, S. Inagaki, Y. Nishihara et al. Org. Lett., 2006, 8, 3037–3040. 62. H. Takahashi, K.M. Hossain, Y. Nishihara et al. J. Org. Chem., 2006, 71, 671–675. 63. R. Shintani, T. Yamagami, T. Hayashi, Org. Lett., 2006, 8, 4799–4801. 64. S. Dumouchel, F. Mongin, F. Tr´ecourt, G. Qu´eguiner, Tetrahedron Lett., 2003,44, 3877–3880. 65. T. Nakamura, H. Kinoshita, H. Shinokubo, K. Oshima, Org. Lett., 2002,4, 3165–3167. 66. O. Yamazaki, T. Tanaka, S. Shimada et al. Synlett , 2004, 1921–1924. 67. P. Appukkuttan, E. Van der Eycken, W. Dehaen, Synlett , 2003, 1204–1206.

The Coordination and Catalytic Chemistry of dppf 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109.

110. 111. 112.

111

J.P. Wolfe, S.L. Buchwald, J. Am. Chem. Soc., 1997, 119, 6054–6058. X-L. Wang, X-F. Zheng, J. Reiner, Synlett , 2006, 942–944. J-D. Cheon, T. Mutai, K. Araki, Tetrahedron Lett., 2006, 47, 5079–5082. R. Kisselev, M. Thelakkat, Chem. Commun., 2002, 1530–1531. S. Thayumanavan, S. Barlow, S.R. Marder, Chem. Mater, 1997, 9, 3231–3235. G. Cuny, M. Bois-Choussy, J. Zhu, J. Am. Chem. Soc., 2004, 126, 14475–14484. W.C. Shakespeare, Tetrahedron Lett., 1999, 40, 2035–2038. Y-K. Lim, K-S. Lee, C-G. Cho, Org. Lett., 2003, 5, 979–982. C.S. Cho, D.B. Patel, Tetrahedron, 2006, 62, 6388–6391. C.G. Saluste, S. Crumpler, M. Furber, R.J. Whitby, Tetrahedron Lett., 2004, 45, 6995–6996. O. Jimenez, T.E. M¨uller, C. Sievers et al. Chem. Commun., 2006, 2974–2976. M. Utsunomiya, J.F. Hartwig, J. Am. Chem Soc., 2003, 125, 14286–14287. L.K. Vo, D.A. Singleton, Org. Lett., 2004, 6, 2469–2472. S. Cacchi, G. Fabrizi, A. Goggiamani, G. Zappia, Org. Lett., 2001, 3, 2539–2541. I.P. Beletskaya, A.G. Bessmertnykh, R. Guilard, Tetrahedron Lett., 1997, 38, 2287–2290. M. AI-Masum, M. Meguro, Y. Yamamoto, Tetrahedron Lett., 1997, 38, 6071–6074. M. Meguro, Y. Yamamoto, Tetrahedron Lett., 1998, 39, 5421–5424. J. Louie, M.S. Driver, B.C. Hamann, J.F. Hartwig, J. Org. Chem., 1997, 62, 1268–1273. J.B. Arterburn, K.V. Rao, R. Ramdas, B.R. Dible, Org. Lett., 2001, 3, 1351–1354. S. Tanimori, M. Kirihata, Tetrahedron Lett., 2000, 41, 6785–6788. G. Cuny, M. Bois-Choussy, J. Zhu, Angew. Chem. Int. Ed., 2003, 42, 4774–4777. Y. Kozawa, M. Mori, Tetrahedron Lett., 2001, 42, 4869–4873. Y. Kozawa, M. Mori, J. Org. Chem., 2003, 68, 8068–8074. L. Routaboul, C. Buch, H. Klein et al. Tetrahedron Lett., 2005, 46, 7401–7405. J. Pawlas, Y. Nakao, M. Kawatsura, J.F. Hartwig, J. Am. Chem. Soc., 2002, 124, 3669–3679. J.B. Arterburn, M. Pannala, A.M. Gonzalez, Tetrahedron Lett., 2001, 42, 1475–1477. S. Abbas, C.J. Hayes, Tetrahedron Lett., 2000, 41, 4513–4517. C. Ma, S-J. Liu, L. Xin et al. Tetrahedron, 2006, 62, 9002–9009. M. Yoshida, Y. Morishita, M. Fujita, M. Ihara, Tetrahedron Lett., 2004, 45, 1861–1864. X. Moreau, J.M. Campagne, J. Organomet. Chem., 2003, 687, 322–326. X. Moreau, J.M. Campagne, G. Meyer, A. Jutand, Eur. J. Org. Chem., 2005, 3749–3760. (a) T.S.A. Hor, L-T. Phang, J. Organomet. Chem., 1990, 381, 121–125, and refs. therein; (b) T.S.A. Hor, L-T. Phang, J. Organomet. Chem., 1989, 373, 319–324, and refs. therein. S. Fukuzawa, D. Tanihara, S. Kikuchi, Synlett , 2006, 2145–2147. M. Arisawa, K. Fujimoto, S. Morinaka, M. Yamaguchi, J. Am. Chem. Soc., 2005, 127, 12226–12227. M. Arisawa, Y. Kozuki, M. Yamaguchi, J. Org. Chem., 2003,68, 8964–8967. T. Suzuki, S. Watanabe, T. Yamada, K. Hiroi, Tetrahedron Lett., 2003, 44, 2561–2563. X. Wu, A.K. Mahalingam, Y. Wan, M. Alterman, Tetrahedron Lett., 2004, 45, 4635–4638. Y. Wan, M. Alterman, M. Larhed, A. Hallberg, J. Comb. Chem., 2003, 5, 82–84. K. Kumar, A. Zapf, D. Michalik et al. Org. Lett., 2004, 6, 7–10. J. Giron`es, J. Duran, A. Polo, J. Real, J. Mol. Catal. A: Chem., 2003, 198, 77–88. Y. Bessard, R. Crettaz, Tetrahedron, 1999, 55, 405–412. (a) C. Bianchini, A. Meli, W. Oberhauser et al. Organometallics, 2003, 22, 2409–2421; (b) O.V. Gusev, A.M. Kalsin, M.G. Peterleitner et al. Organometallics, 2002, 21, 3637–3649. C. Bianchini, A Meli, W. Oberhauser et al. J. Mol. Catal. A: Chem., 2004, 224, 35–49. C. Larksarp, H. Alper, J. Org. Chem., 2000, 65, 2773–2777. W.J. Xiao, H. Alper, J. Org. Chem., 1999, 64, 9646–9652.

112

Ferrocenes: Ligands, Materials and Biomolecules

113. 114. 115. 116. 117. 118. 119. 120.

M. Yoshida, Y. Ohsawa, M. Ihara, J. Org. Chem., 2004, 69, 1590–1597. M. Mori, M. Kimura, Y. Takahashi, Y. Tamaru, Chem. Commun., 2006, 4303–4305. T. Schareina, A. Zapf, M. Beller, Chem. Commun., 2004, 1388–1389. R. Chidambaram, Tetrahedron Lett., 2004, 45, 1441–1444. F. Jin, P.N. Confalone, Tetrahedron Lett., 2000, 41, 3271–3273. J. Zanon, A. Klapars, S.L. Buchwald, J. Am. Chem. Soc., 2003, 125, 2890–2891. P.E. Maligres, M.S. Waters, F. Fleitz, D. Askin, Tetrahedron Lett., 1999, 40, 8193–8195. W.G. Jary, T. Ganglberger, P. P¨ochlauer, H. Falk, Monatshefte f¨ur Chemie, 2005, 136, 537–541. J. van Buijtenen, J. Meuldijk, J.A.J.M. Vekemans et al. Organometallics, 2006, 25, 873–881. K. Zeitler, W. Steglich, J. Org. Chem., 2004, 69, 6134–6136. P.D. de Koning, M. Jackson, I.C. Lennon, Org. Process Res. Dev., 2006, 10, 1054–1058. S. Subongkoj, S. Lange, W. Chen, J. Xiao, J. Mol. Catal. A: Chem., 2003, 196, 125–129. G. Mann, J.F. Hartwig, Tetrahedron Lett., 1997, 38, 8005–8008. G. Mann, J.F. Hartwig, J. Org. Chem., 1997, 62, 5413–5418. A. Acosta-Ram´ırez, M. Mu˜noz-Hern´andez, W.D. Jones, J.J. Garc´ıa, J. Organomet. Chem., 2006, 691, 3895–3901. S. Kezuka, T. Okado, E. Niou, R. Takeuchi, Org. Lett., 2005, 7, 1711–1714. V. Cadierno, S.E. Garc´ıa-Garrido, J. Gimeno, Adv. Synth. Catal., 2006, 348, 101–110. V. Cadierno, J. D´ıez, S.E. Garc´ıa-Garrido, J. Gimeno, Chem. Commun., 2004, 2716–2717. U. Radhakrishnan, V. Gevorgyan, Y. Yamamoto, Tetrahedron Lett., 2000, 41, 1971–1974. S-I. Lee, S-C. Shim, T-J. Kim, J. Polym. Sci., Part A: Polym. Chem., 1996, 34, 2377–2386. K. Li, G. Wei, J. Darkwa, S.K. Pollack, Macromolecules, 2002, 35, 4573–4576. C. Kowitz, G. Wegner, Tetrahedron, 1997, 53, 15553–15574. T. Kanbara, Y. Miyazaki, K. Hasegawa, T. Yamamoto, J. Polym. Sci., Part A: Polym. Chem., 2000, 38, 4194–4199. M. Lautens, S. Hiebert, J-L. Renaud, J. Am. Chem. Soc., 2001, 123, 6834–6839. S. Shekhar, P. Ryberg, J.F. Hartwig et al. J. Am. Chem. Soc., 2006, 128, 3584–3591. L.M. Alcazar-Roman, J.F. Hartwig, A.L. Rheingold et al. J. Am. Chem. Soc., 2000, 122, 4618–4630. (a) Z. Weng, S.H. Teo, T.S.A. Hor, Organometallics, 2006, 25, 4878–4882; (b) S. Teo, Z. Weng, T.S.A. Hor, Organometallics, 2006, 25, 1199–1205; (c) Z. Weng, S. Teo, L.L. Koh, T.S.A. Hor, Chem. Commun., 2006, 1319–1321; (d) Z. Weng, S. Teo, L.L. Koh, T.S.A. Hor, Angew. Chem. Int. Ed., 2005, 44, 7560–7564; (e) Z. Weng, S. Teo, L.L. Koh, T.S.A. Hor, Organometallics, 2004, 23, 4342–4345; (e) Z. Weng, L.L. Koh, T.S.A. Hor, J. Organomet. Chem., 2004, 689, 18–24; (f) Z. Weng, S.H. Teo, T.S.A. Hor, Acc. Chem. Rec., 2007, 40, 676–684.

121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139.

References for Tables Table 2.1 1. 2. 3. 4. 5. 6.

L-C. Song, J-T. Liu, Q-M. Hu et al. Organometallics, 2000, 19, 5342–5351. W.A. Schenk, K. Nielsen, N.I. Burzlaff, M. Hagel, Inorg. Chem., 2002, 41, 1079–1085. J.A.K. Bauer, T.M. Becker, M. Orchin, J. Chem. Crystallogr., 2005, 35, 141–146. F. Tisato, F. Refosco, M. Porchia et al. J. Organomet. Chem., 2001, 637–639, 772–776. O. Sigouin, A.L. Beauchamp, Inorg. Chimica Acta, 2005, 358, 4489–4496. D.M. Kimari, A.M. Duzs-Moore, J. Cook et al. Inorg. Chem. Commun., 2005, 8, 14–17.

The Coordination and Catalytic Chemistry of dppf 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51.

113

C. Imrie, J.R. Dilworth, Y. Zheng et al. Dalton Trans., 2001, 2624–2633. N-S. Choi, S.W. Lee, J. Organomet. Chem., 2001, 627, 226–232. Y-K. Yan, W. Koh, C. Jiang et al. Polyhedron, 2000, 19, 641–647. O. Knoesen, H. G¨orls, S. Lotz, J. Organomet. Chem., 2000, 598, 108–115. N.D. Reddy, P.E. Fanwick, R.A. Walton, Inorg. Chem., 2001, 40, 1732–1733. (a) L-C. Song, Z-Y. Yang, H-Z. Bian et al. Organometallics, 2005, 24, 6126–6135. (b) LC. Song, Z-Y. Yang, H-Z. Bian, Q-M. Hu, Organometallics, 2004, 23, 3082–3084. J. van Buijtenen, J. Meuldijk, J.A.J.M. Vekemans et al. Organometallics, 2006, 25, 873–881. Y. Sun, H-S. Chan, P.H. Dixneuf, Z. Xie, Organometallics, 2004, 23, 5864–5872. B. Therrien, L. Vieille-Petit, J. Jeanneret-Gris et al. J. Organomet. Chem., 2004, 689, 2456–2463. X.L. Lu, J.J. Vittal, E.R.T. Tiekink et al. J. Organomet. Chem., 2004, 689, 1978–1990. B. Therrien, G. S¨uss-Fink, Inorg. Chim. Acta, 2004, 357, 219–224. S. Jung, C.D. Brandt, J. Wolf, H. Werner, Dalton Trans., 2004, 375–383. X.L. Lu, S.Y. Ng, J.J. Vittal et al. J. Organomet. Chem., 2003, 688, 100–111. V. Cadierno, P. Crochet, J. D´ıez et al. Organometallics, 2003, 22, 5226–5234. I.R. Butler, S.J. Coles, M.B. Hursthouse et al. Inorg. Chem. Commun., 2003, 6, 760–762. V. Cadierno, S.E. Garc´ıa-Garrido, J. Gimeno, J. Organomet. Chem., 2001, 637–639, 767–771. P. Nieczypor, P.W.N.M. van Leeuwen, J.C. Mol et al. J. Organomet. Chem., 2001, 625, 58–66. B. Buriez, I.D. Burns, A.F. Hill et al. Organometallics, 1999, 18, 1504–1516. J. Zhao, J.F. Hartwig, Organometallics, 2005, 24, 2441–2446. X.L. Lu, J.J. Vittal, E.R.T. Tiekink et al. J. Organomet. Chem., 2004, 689, 1444–1451. W-Y. Wong, F-L. Ting, W-L. Lam, Eur. J. Inorg. Chem., 2002, 2103–2111. F.F. de Biani, C. Graiff, G. Opromolla et al. J. Organomet. Chem., 2001, 637–639, 586–594. J-L. Chen, L-Y. Zhang, L-X. Shi et al. Inorg. Chim. Acta, 2005, 358, 859–864. A.R. O’Connor, C. Nataro, A.L. Rheingold, J. Organomet. Chem., 2003, 679, 72–78. S.W.A. Fong, J.J. Vittal, T.S.A. Hor, Organometallics, 2000, 19, 918–924. P. McQuade, R.E.K. Winter, L. Barton, J. Organomet. Chem., 2003, 688, 82–91. W. Kaim, T. Sixt, M. Weber, J. Fiedler, J. Organomet. Chem., 2001, 637–639, 167–171. E.N-M. Ho, B.K-M. Hui, W-T. Wong, J. Organomet. Chem., 2001, 637–639, 276–283. K.S-Y. Leung, Inorg. Chem. Commun., 1999, 2, 498–502. T. Avil´es, A. Dinis, J.O. Gonc¸alves et al. Dalton Trans., 2002, 4595–4602. G. Pilloni, A. Toffoletti, G. Bandoli, B. Longato, Inorg. Chem., 2006, 45, 10321–10328. E. Champeil, S.M. Draper, Dalton Trans., 2001, 1440–1447. M.D. Jones, F.A.A. Paz, J.E. Davies, B.F.G. Johnson, Acta Cryst., 2003, E59, m111–m113. C. Pettinari, F. Marchetti, A. Cingolani et al. J. Organomet. Chem., 2002, 651, 5–14. B. Longato, R. Coppo, G. Pilloni et al. J. Organomet. Chem., 2001, 637–639, 710–718. J-F. Ma, Y. Yamamoto, J. Organomet. Chem., 1999, 574, 148–154. B. Longato, L. Riello, G. Bandoli, G. Pilloni, Inorg. Chem., 1999, 38, 2818–2823. R. Pastorek, J. Kamen´ıcˇ ek, M. Pavl´ıcˇ ek et al. J. Coord. Chem., 2002, 55, 1301–1308. R. Pastorek, Z. Tr´avn´ıcˇ ek, J. Marek et al. Polyhedron, 2000, 19, 1713–1717. R. Pastorek, J. Kamen´ıcˇ ek, B. Cvek et al. J. Coord. Chem., 2006, 59, 911–919. R.A. Stockland, Jr., A.M. Levine, M.T. Giovine et al. Organometallics, 2004, 23, 647–656. M. Morita, K. Inoue, S. Ogoshi, H. Kurosawa, Organometallics, 2003, 22, 5468–5472. L-C. Song, G-F. Wang, P-C. Liu, Q-M. Hu, Organometallics, 2003, 22, 4593–4598. S-S. Sun, J.A. Anspach, A.J. Lees, Inorg. Chim. Acta, 2003, 351, 363–368. T.A.K. Al-Allaf, H. Schmidt, K. Merzweiler et al. J. Organomet. Chem., 2003, 678, 48–55.

114 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93.

Ferrocenes: Ligands, Materials and Biomolecules R. Ares, M. L´opez-Torres, A. Fern´andez et al., J. Organomet. Chem., 2003, 665, 76–86. A. Mentes, R.D.W. Kemmitt, Polyhedron, 2002, 21, 2653–2657. R. Ares, M. L´opez-Torres, A. Fern´andez et al. Polyhedron, 2002, 21, 2309–2315. Y.C. Neo, J.S.L. Yeo, P.M.N. Low et al. J. Organomet. Chem., 2002, 658, 159–168. Y.C. Neo, J.J. Vittal, T.S.A. Hor Dalton Trans., 2002, 337–342. A. Fern´andez, D. V´azquez-Garc´ıa, J.J. Fern´andez et al. Eur. J. Inorg. Chem., 2002, 2389–2401. O.V. Gusev, A.M. Kalsin, M.G. Peterleitner et al. Organometallics, 2002, 21, 3637–3649. A. Fern´andez, D. V´azquez-Garc´ıa, J.J. Fern´andez et al. New J. Chem., 2002, 26, 398–404. A. Fern´andez, D. V´azquez-Garc´ıa, J.J. Fern´andez et al. New J. Chem., 2002, 26, 105–112. G.J. Grant, S.M. Carter, A.L. Russell et al. J. Organomet. Chem., 2001, 637–639, 683–690. J.M. Vila, E. Gayoso, T. Pereira et al. J. Organomet. Chem., 2001, 637–639, 577–585. S.R. Ananias, A.E. Mauro, V.A. de L. Neto, Transition Met. Chem., 2001, 26, 570–573. R.J. van Haaren, K. Goubitz, J. Fraanje et al. Inorg. Chem., 2001, 40, 3363–3372. C. Fern´andez-Rivas, D.J. C´ardenas, B. Mart´ın-Matute et al. Organometallics, 2001, 20, 2998–3006. M. L´opez-Torres, A. Fern´andez, J.J. Fern´andez et al. Organometallics, 2001, 20, 1350–1353. M. Lousame, A. Fern´andez, M. L´opez-Torres et al. Eur. J. Inorg. Chem., 2000, 2055–2062. D.P. Arnold, P.C. Healy, M.J. Hodgson, M.L. Williams, J. Organomet. Chem., 2000, 607, 41–50. V.V. Bashilov, T.V. Magdesieva, D.N. Kravchuk et al. J. Organomet. Chem., 2000, 599, 37–41. J-F. Ma, Y. Yamamoto, Inorg. Chim. Acta, 2000, 299, 164–171. D.H. Nguyen, G. Laurenczy, M. Urrutigo¨ıty, P. Kalck, Eur. J. Inorg. Chem., 2005, 4215–4225. ´ C. Herrera-Alvarez, V. G´omez-Ben´ıtez, R. Red´on et al. J. Organomet. Chem., 2004, 689, 2464–2472. C.M. Crawforth, S. Burling, I.J.S. Fairlamb et al. Tetrahedron, 2005, 61, 9736–9751. Y.C. Neo, J.J. Vittal, T.S.A. Hor, J. Organomet. Chem., 2001, 637–639, 757–761. A. Fern´andez, E. Pereira, J.J. Fern´andez et al. New J. Chem., 2002, 26, 895–901. T. Tanase, J. Matsuo, T. Onaka et al. J. Organomet. Chem., 1999, 592, 103–108. J.S.L. Yeo, G. Li, W-H. Yip et al. Dalton Trans., 1999, 435–441. S. Shekhar, J.F. Hartwig, J. Am. Chem. Soc., 2004, 126, 13016–13027. W. Henderson, C. Evans, B.K. Nicholson, J. Fawcett, Dalton Trans., 2003, 2691–2697. W-Y. Wong, G-L. Lu, K-H. Choi, J. Organomet. Chem., 2002, 659, 107–116. D-Y. Noh, E-M. Seo, H-J. Lee et al. Polyhedron, 2001, 20, 1939–1945. C.V. Ursini, G.H.M. Dias, M. H¨orner et al. Polyhedron, 2000, 19, 2261–2268. D. V´azquez-Garc´ıa, A. Fern´andez, J.J. Fern´andez et al. J. Organomet. Chem., 2000, 595, 199–207. L.M. Scolaro, M.R. Plutino, A. Romeo et al. Dalton Trans., 2006, 2551–2559. P. Teo, L.L. Koh, T.S.A. Hor, Inorg. Chem., 2003, 42, 7290–7296. S. Furukawa, T. Okubo, S. Masaoka et al. Angew. Chem. Int. Ed., 2005, 44, 2700–2704. W-Y. Yeh, Y-C. Liu, S-M. Peng, G-H. Lee, Inorg. Chim. Acta, 2005, 358, 1987–1992. C.D. Nicola, Effendy, C. Pettinari et al. Inorg. Chim. Acta, 2005, 358, 695–706. P. Pinto, M.J. Calhorda, V. F´elix et al. Monatshefte f¨ur Chemie, 2000, 131, 1253–1265. J. Vicente, P. Gonz´alez-Herrero, Y. Garc´ıa-S´anchez et al. Eur. J. Inorg. Chem., 2006, 115–126. D. Li, Y-F. Luo, T. Wu, S.W. Ng, Acta Cryst., 2004, E60, m927–m929. Y-C. Liu, C-I Li, W-Y. Yeh et al. Inorg. Chim. Acta, 2006, 359, 2361–2368. C-H. Chou, W-Y. Yeh, G-H. Lee, S-M. Peng, Inorg. Chim. Acta, 2006, 359, 4139–4143.

The Coordination and Catalytic Chemistry of dppf

115

94. 95. 96. 97. 98. 99. 100. 101. 102.

G.G. Lobbia, M. Pellei, C. Pettinari et al. Inorg. Chim. Acta, 2005, 358, 3633–3641. G.G. Lobbia, M. Pellei, C. Pettinari et al. Polyhedron, 2005, 24, 181–187. X.L. Lu, W.K. Leong, L.Y. Goh, A.T.S. Hor, Eur. J. Inorg. Chem., 2004, 2504–2513. X.L. Lu, W.K. Leong, T.S.A. Hor, L.Y. Goh, J. Organomet. Chem., 2004, 689, 1746–1756. Effendy, J.V. Hanna, F. Marchetti et al. Inorg. Chim. Acta, 2004, 357, 1523–1537. Effendy, G.G. Lobbia, M. Pellei et al. Inorg. Chim. Acta, 2001, 315, 153–162. C. Santini, C. Pettinari, G.G. Lobbia et al. Inorg. Chim. Acta, 1999, 285, 81–88. F. Mohr, M.C. Jennings et al. Eur. J. Inorg. Chem., 2003, 217–223. C.P. McArdle, S. Van, M.C. Jennings, R.J. Puddephat, J. Am. Chem. Soc., 2002, 124, 3959–3965. 103. C.P. McArdle, M.J. Irwin, M.C. Jennings, J.J. Vittal, R.J. Puddephatt, Chem. Eur. J., 2002, 8, 723–734. 104. S. Canales, O. Crespo, M.C. Gimeno et al. Dalton Trans., 2003, 4525–4528.

Table 2.2 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

A.C. Ohs, A.L. Rheingold, M.J. Shaw, C. Nataro, Organometallics, 2004, 23, 4655–4660. Y-Y. Niu, T-N. Chen, S-X. Liu et al. Dalton Trans., 2002, 1980–1984. J.P. Blitz, N.T. Lucas, M.G. Humphrey, J. Organomet. Chem., 2002, 650, 133–140. D.D. Ellis, P.A. Jelliss, F.G.A. Stone, Organometallics, 1999, 18, 4982–4994. X.L. Lu, J.J. Vittal, E.R.T. Tiekink et al. J. Organomet. Chem., 2004, 689, 1444–1451. Y. Liu, K.H. Lee, J.J. Vittal, T.S.A. Hor, Dalton Trans., 2002, 2747–2751. Y.C. Neo, J.J. Vittal, T.S.A. Hor, Dalton Trans., 2002, 337–342. G. Li, C-K. Lam, S.W. Chien et al. J. Organomet. Chem., 2004, 690, 990–997. Z. Li, K.F. Mok, T.S.A. Hor, J. Organomet. Chem., 2003, 682 73–78. C.V. Ursini, G.H.M. Dias, M. H¨orner et al. Polyhedron, 2000, 19, 2261–2268. S.W.A. Fong, J.J. Vittal, T.S.A. Hor, Organometallics, 2000, 19, 918–924. W-Y. Wong, G-L. Lu, K-H. Choi, J. Organomet. Chem., 2002, 659, 107–116. K-T. Youm, Y. Kim, Y. Do, M-J. Jun, Inorg. Chim. Acta, 2000, 310, 203–209. L. Song, S-Q. Xia, S-M. Hu et al. Polyhedron, 2005, 24, 831–836.

Table 2.3 1. 2. 3. 4. 5.

X.L. Lu, J.J. Vittal, E.R.T. Tiekink et al. J. Organomet. Chem., 2004, 689, 1978–1990. P. McQuade, R.E.K. Winter, L. Barton, J. Organomet. Chem., 2003, 688, 82–91. J-F. Ma, Y. Yamamoto, J. Organomet. Chem., 1999, 574, 148–154. J-F. Ma, Y. Yamamoto, Inorg. Chim. Acta, 2000, 299, 164–171. X.L. Lu, W.K. Leong, L.Y. Goh, A.T.S. Hor, Eur. J. Inorg. Chem., 2004, 2504–2513.

Table 2.4 1.

V. Cadierno, S.E. Garc´ıa-Garrido, J. Gimeno, J. Organomet. Chem., 2001, 637–639, 767–771. 2. T. Avil´es, A. Dinis, J.O. Gonc¸alves et al. J. Chem. Soc., Dalton Trans., 2002, 4595–4602. 3. R.J. Coyle, Y.L. Slovokhotov, M.Y. Antipin, V.V. Grushin, Polyhedron, 1998, 17, 3059–3070. 4. J.S.L. Yeo, J.J. Vittal, T.S.A. Hor, Chem. Commun., 1999, 1477–1478. 5. G. Pilloni, B. Longato, G. Bandoli, Inorg. Chim. Acta, 1998, 277, 163–170. 6. M.C. Gimeno, P.G. Jones, A. Laguna et al. Inorg. Chim. Acta, 2001, 316, 89–93.

116

Ferrocenes: Ligands, Materials and Biomolecules

Table 2.5 1. R. Broussier, E. Bentabet, M. Laly et al. J. Organomet. Chem., 2000, 613, 77–85. 2. G. Pilloni, B. Longato, G. Bandoli, Inorg. Chim. Acta, 1998, 277, 163–170. 3. G. Pilloni, B. Longato, G. Bandoli, B. Corain, Dalton Trans., 1997, 819–825. 4. M.C. Gimeno, P.G. Jones, A. Laguna, C. Sarroca, J. Organomet. Chem., 2000, 596, 10–15.

Table 2.6 1. G. Pilloni, B. Longato, G. Bandoli, Inorg. Chim. Acta, 1998, 277, 163–170. 2. (a) M.C. Gimeno, P.G. Jones, A. Laguna et al. Inorg. Chim. Acta, 2001, 316, 89–93; (b) H. Liu, N.A.G. Bandeira, M.J. Calhorda et al. J. Organomet. Chem., 2004, 689, 2808–2819. 3. S. Canales, O. Crespo, M.C. Gimeno et al. J. Organomet. Chem., 2000, 613, 50–55. 4. G. Pilloni, B. Longato, G. Bandoli, B. Corain, Dalton Trans., 1997, 819–825. 5. G. Pilloni, B. Longato, G. Bandoli, Inorg. Chim. Acta, 2000, 298, 251–255.

Table 2.7 1. 2. 3.

M.C. Gimeno, P.G. Jones, A. Laguna et al. Inorg. Chim. Acta, 2001, 316, 89–93. M.C. Gimeno, P.G. Jones, A. Laguna, C. Sarroca, J. Organomet. Chem., 2000, 596, 10–15. S. Canales, O. Crespo, M.C. Gimeno et al. J. Organomet. Chem., 2000, 613, 50–55.

3 Synthesis, Coordination Chemistry and Catalytic Use of dppf Analogs Thomas J. Colacot and S´ebastien Parisel

3.1

Introduction

Ferrocene-based phosphines have emerged as one of the most powerful classes of ligands in chiral and achiral catalysis.1 Hayashi’s original work,2 demonstrating the importance of bite angle on the applications of a C2 symmetric metallocene based ligand, dppf and its palladium complexes for carbon–carbon coupling reaction, was a milestone in accelerating the developments of phosphinoferrocene ligands for homogeneous catalysis applications. Since then, numerous publications have appeared on dppf and [PdCl2 (dppf)]: notable publications include a book chapter by Hor3 and reviews by Hartwig4 on carbon–heteroatom coupling. From an industrial perspective, Johnson Matthey was the ﬁrst company to commercialise dppf and [PdCl2 (dppf)] in multikilogram quantities for applications in homogeneous catalysis in the ﬁne chemical and pharmaceutical industries.5 Since the chemistry of dppf is covered by Hor in this book (Chapter 2), this chapter focuses only on dppf analogs: their general synthesis, coordination chemistry and catalysis.6 Compared to the signiﬁcant amount of work on dppf itself, articles on its analogs have been limited until recently. A few interesting review articles published in the past ﬁve years highlight the catalytic applications of some of these derivatives.1d, 7 The purpose of this chapter is not to duplicate any of the earlier efforts but to consolidate the syntheses, coordination chemistry and catalytic applications of dppf analogs with Ferrocenes: Ligands, Materials and Biomolecules Edited by Petr Stepnicka  2008 John Wiley & Sons, Ltd

118

Ferrocenes: Ligands, Materials and Biomolecules

particular emphasis given to 1,1 -disubstituted symmetrical and unsymmetrical, chiral and achiral derivatives.

3.2

Achiral Ferrocenylphosphines

3.2.1

1,1 -Disubstituted Achiral Symmetrical Ferrocenylphosphines

In addition to the bite angle (ligand–metal–ligand angle formed when a bidentate ligand coordinates as a chelate to a metal center) of the 1,1 -disubstituted ferrocene ligands, the electron donating abilities and steric properties play prominent roles in catalysis. Although the synthesis of dppf was reported in detail in 1971,1b it took about two decades to make its analogs with varying electronic and steric properties. The basic synthetic route is very similar to the dppf synthesis, where 1,1 -dilithioferrocene is coupled with two equivalents of R2 PCl.8 The corresponding arsenic derivative can also be prepared by quenching the dilithioferrocene with R2 AsCl. Alternatively, some of these ligands have been prepared by reacting 1,1 -bis(dichlorophosphino)ferrocene with alkyl/aryl lithium or magnesium reagents. In general, sterically congested chlorophosphines such as Cl(o-i PrC6 H4 )2 P prefer the latter route, as the former one produces predominantly P–P coupled diphosphine monoxide impurities.9 Phosphonites are also obtained by treatment of fc(PCl2 )2 with alcohol in the presence of a base such as pyridine.9 These routes are summarised in Scheme 3.1. 1) 2 n-BuLi, TMEDA, hexane, reflux FcH

PR2

2 RM

PR2

THF, −80 °C

Fe 2) 2R2PCl, THF, −40 °C R Me Et i-Pr Cy t-Bu o-tolyl o-MeOC6H4 p-MeOC6H4 p-CF3C6H4 3,5-CF3C6H3 2-furyl 5-Me-2-furyl o-i PrC6H4 C6F5

Abbreviation dmpf (1) depf (2) dippf (3) dcypf (4) dtbpf (5) dtpf (6) o-MeOdppf (7) p-MeOdppf (8) p-CF3dppf (9) 3,5-CF3dppf (10) dfpf (11) 5-Medfpf (12) o-i Prdppf (13) (14)

PCl2 Fe PCl2

Pyridine, 4 R'OH

P(OR')2 Fe P(OR')2 R' = Et (15)

Scheme 3.1 General synthetic routes for 1,1 -bis(dialkyl/aryl/alcoxy phosphino)ferrocene11

Unlike dppf, puriﬁcation and handling of fc(PR2 )2 , where R is an alkyl group, is very tedious due to its air-sensitivity and required chromatographic separations, which

Synthesis, Coordination Chemistry and Catalytic Use of dppf Analogs

119

limits the development of this area. However, commercial quantities of palladium complexes of 1,1 -bis(di-tert-butylphosphino)ferrocene (dtbpf, 5) and 1,1 -bis(di-isopropylphosphino)ferrocene (dippf, 3) have been available from Johnson Matthey since 2004.5, 10 3.2.2

1,1 -Disubstituted Non-Chiral Unsymmetrical Ferrocenylphosphines

Compared to the development of symmetrical ferrocenylphosphines, the area of unsymmetrical achiral ones is very limited, but has the potential for growth. To the best of our knowledge, there are fewer than a dozen 1,1 -unsymmetrical ferrocene ligands known in the literature (Figure 3.1).

PR32 Fe PR1R2

16 : R3 = Ph, R1 = R2 = t-Bu 17 : R3 = Ph, R1 = R2 = i-Pr 18 : R3 = t-Bu, R1 = t-Bu, R2 = Ph 19 : R3 = Ph, R1 = R2 = p-MeO-Ph 20 : R3 = Ph, R1 = R2 = 2-furyl

21 : R3 = Ph, R1 = R2 = Cl 22 : R3 = Ph, R1 = R2 = OPh 23 : R3 = Ph, R1 = R2 = Men 24 : R3 = Ph, R1 = R2 = p-CF3-Ph 25 : R3 = p-MeO-Ph, R1 = R2 = p-CF3-Ph

Figure 3.1 1,1 -unsymmetrical ferrocenylphosphines

Cullen’s work12 was instrumental in developing the ﬁrst examples of unsymmetrical 1,1 -P/P ligands via P-[1]-ferrocenophane ring opening methodology, developed by the pioneering work of Seyferth and Osborne.13 The other prominent route involves stepwise lithiation followed by phosphination of ferrocene or dibromoferrocene. These methods are described in Chapter 5.14 3.2.3

Coordination Chemistry

Both symmetrical and unsymmetrical ferrocenylphosphines have been used to prepare metal complexes, mostly by making use of the late transition metals. Palladium and platinum complexes10, 15, 16 are typically synthesised by reacting the ligand with [MCl2 (MeCN)2 ], [MCl2 (PhCN)2 ], [MCl4 ]2− , [Pd(Me)Cl(COD)], or [MCl2 (COD)] (M = platinum, palladium; COD = cycloocta-1,5-diene) in a suitable solvent. The metal precursor and reaction conditions, especially the reaction solvent, are critical in isolating the complexes in good purity and yield when the ligand contains alkyl groups on the phosphorus. Interestingly the nickel complexes have been prepared by reacting the ligands with hydrated nickel(II) chloride or nickel(II) bromide. In general, these complexes of ferrocenylphosphines, especially the bulkier ones, are more difﬁcult to synthesise in pure form than other bidentate phosphines, such as dppe, dppp, etc. Complexes of cobalt, zinc, cadmium, mercury and gold are also known.15, 17, 18, 19, 20 The X-ray crystal structures of several examples of these compounds are reported. A comparison of the P-M-P bite angle of several of these complexes is given in Table 3.1. The bite angle seems to be inﬂuenced by several factors: the central metal of the metallocene; groups on the phosphorus atoms; and the complexing metal. The inﬂuence of the X groups on the bite angle seems to be trivial for less bulky ligands,

120

Ferrocenes: Ligands, Materials and Biomolecules Table 3.1 Bite angle of metal complexes of ferrocenyldiphosphines Complex

Bite Angle (◦ )

Reference

[PdCl2 (dppf)] (26) [PdCl2 (dppr)]a (27) [PdCl2 (dppo)]b (28) [PdCl2 (dmpf)] (29) [PdCl2 (depf)] (30) [PdCl2 (dippf)] (31) [PtCl2 (dippf)] (32) [ZnCl2 (dippf)] (33) [PdCl2 (dcypf)] (34) [PdCl2 (dtbpf)] (35) [PdCl2 (5-Medfpf)] (36) [PdCl2 (o-MeOdppf)] (37) [PdCl2 (o-i Prdppf)] (38)

97.98(4) 100.02(2) 101.29(4) 98.95(9)/99.64(9) 97.74(3) 103.59(4) 103.78(5) 109.98(2) 102.45(3) 104.22 (5) 96.98(3) 100.27(5) 101.54(7)

12 6e 9 17 9 18 19 19 20 21 22 9 9

a dppr = 1,1 -bis(diphenylphosphino)ruthenocene. b dppo = 1,1 -bis(diphenylphosphino)osmocene.

although in the case of [PtX2 (dppf)] (where X = chlorine, bromine, iodine or Ph) there was a slight increase in the bite angle when X = I or Ph in comparison to X = Cl.16 Moving down in the periodic table from iron to ruthenium and osmium (e.g., 26 → 27 → 28) there is a bite angle increase of 2–3◦ , which could be due to the larger size of the metal on the metallocenes. The effect of bulkier substituents on the phosphorus atoms also has an inﬂuence on the bite angle, as seen with palladium complexes 26, 30, 31, 34 and 35. The nature of the coordinating metal also inﬂuences the bite angle. For example, the bite angles of dippf complexes of various MCl2 fragments (M = palladium, platinum, zinc) illustrate the inﬂuence of the metal coordination to the phosphorus atoms. The bite angle of [PdCl2 (dippf)] (31) is about 103.6◦ , 6.0◦ smaller than that of its zinc analogue (33). Interestingly, the effect is very subtle between platinum and palladium. That difference could be due to the pseudo–square planar geometry of Group 10 metal complexes vs the pseudo–tetrahedral geometry of Group 12 metal complexes (Figure 3.2).19 The X-ray structure of 34 was published recently by Nataro et al.20 The bite angle is about 1◦ lower than that of 31. Very recently we have been successful in obtaining the X-ray structure of 35, which once again establishes the relationship between the bite angle (104.22(5)◦ ) and the steric bulkiness of the substituents on the phosphorus.21 This value is very close to what Hartwig observed for the X-ray structure of the oxidative addition product, [PdBr(4-CN-C6 H4 )(dtbpf)] (104.28(5)◦ ).23 The bite angle increases dramatically (159.75[4]◦ ) when the halide from the same complex is removed to make it cationic (Figure 3.3). Similar observation was made by van Leeuwen for [PdMe(dippf)]+ (158.21(2)◦ ).24 Unlike in the case of the three-coordinated Tshaped d8 complexes, these cationic complexes adopt a square–planar structure with Pd−Fe bonding interactions. Hartwig observed an interesting solvent assisted equilibrium between [PdBr(4-CN-C6 H4 )(dtbpf)] and [Pd(4-CN-C6 H4 )(dtbpf)]+ , where polar solvents favour the formation of cationic species.23

Synthesis, Coordination Chemistry and Catalytic Use of dppf Analogs

Cl(1)

C(3) Fe(1)

121

C(11)

C(4) Zn(1)

C(2) C(5)

C(14)

Cl(2)

P(1)

C(1) P(2) C(6)

C(10)

C(6) C(9)

P(1A)

C(1)

C(17)

C(8)

P(1)

C(20)

C(7)

Pt(1)

Fe(2) C(11) Cl(1)

Cl(1A)

Figure 3.2 X-ray structures of 32 (left, bite angle 103.78[5]◦ ) and 33 (right, bite angle 109.98[2]◦ ). Reprinted with permission from Ref. 19. Copyright 2003, American Chemical Society. C(18)

C24 C26 C25 C30

C31

C29

C28

C21

Br1

C33

C(4)

Fe1

Pd1

C20

C19

C15

C(14) P(1) C(32) C(31)

C(5)

Pd(1)

C13 C8

N(1)

C(30) C(27) C(28) C(29)

C(7)

C(9) C(10) C(6) C(23)

C4

C(25)

C6

C12

C(33)

P(2)

C1 C9

C17

C(24)

C5

C(20)

C(19)

C(22)

C7

C14 C11

C(8)

C3 P1

C(11) C(1)

Fe(1)

C2

C16

C18

C(2) C(3)

C22

P2 C32

C23

C(17)

C(15)

C(12)

C27

C10

N1

C(21)

Figure 3.3 X-ray structures of cationic [Pd(Br)(4-CN-C6 H4 )(dtbpf)] (left) and the cation in [Pd(4-CN-C6 H4 )(dtbpf)][BF4 ] (right). Reprinted with permission from Ref. 23. Copyright 2003, American Chemical Society.

Palladium complexes of the general formula, [Pd(H2 O)(OTs)(P–P)]OTs (OTs = 4toluenesulfonate), where P-P is dmpf, depf or dippf were recently characterised by Bianchini and Gusev.17 About two decades ago, Cullen8a, 12a was able to synthesise the cationic rhodium complex of dtbpf by reacting [{Rh(NBD)Cl}2 ] with the diphosphine ligand in the presence of sodium perchlorate. The X-ray structure (Figure 3.4) shows that the P−Rh−P bond angle (bite angle) is 103.75(5)◦ . The corresponding bite angle is much larger than that of [Rh(NBD)(PPFA)]+ (95(1)◦ ; NBD = norbornadiene) and that of [Rh((S,S)-chiraphos)(COD)]+ (83.82(6)◦ ). The angles across the diagonals are 148.9 and 150.9◦ , respectively, showing a distortion from square-planar environment, in comparison to the other examples of bidentate phosphine complexes of rhodium.

122

Ferrocenes: Ligands, Materials and Biomolecules C(54)

C(53)

C(53) C(55) C(22)

C(52)

C(62)

C(51) C(2)

C(13)

C(64)

Fe

C(23)

C(42)

C(41)

P(1)

C(1)

C(65) C(61) P(2) C(4)

C(21) C(12)

Rh C(31)

C(11) C(71)

C(72)

C(77) C(73)

C(76)

C(3)

C(33) C(43)

C(32)

C(75) C(74)

Figure 3.4 Crystal Structure of the cation of [Rh(NBD)(dtbpf)][ClO4 ]. Reprinted with permission from Ref. 8(a). Copyright 1983, American Chemical Society.

The extent of this extreme distortion could be due to the repulsion between NBD and ˚ are signiﬁcantly t-butyl substituents. The Rh−P distances of 2.466(1) and 2.458(1) A longer than those found in the corresponding PPFA- and (S,S)-chiraphos-chelated, phosphine complexes.12 The 31 P-NMR chemical shifts of some of the common ferrocenylphosphine ligands as well as their metal complexes are summarised in Table 3.2. Typically, most of the ligands have negative chemical shifts (shielded vs. 85 % phosphoric acid), except in the case of dtbpf. Proceeding from the less bulky 29 to the more bulky electronrich 35 the phosphorus chemical shifts move to lower ﬁeld (deshielding). For the corresponding platinum compounds the shifts are almost half that of palladium, while the shifts for zinc complexes are negative (33 and 40). No systematic work in this area correlating the effect of electronic and steric properties of the complexes with their respective NMR chemical shifts has been reported. Such a study might be useful in understanding the chemical and catalytic properties of these complexes. 3.2.4

Homogeneous Catalysis Applications

Unlike the case of dppf, the homogeneous catalysis of dppf analogs is not well explored. However, applications of some of the palladium-based compounds in carbon–carbon and carbon–nitrogen coupling have been reported recently with some excellent results. Rhodium-catalysed hydrogenation is yet another area of interest. 3.2.4.1

Carbon–Carbon Coupling

The work of Koie/Fu29 on the use of bulky electron-rich monodentate ligand t-Bu3 P in conjunction with palladium sparked a renewed interest in the area of coupling, mainly

Synthesis, Coordination Chemistry and Catalytic Use of dppf Analogs Table 3.2

31 P-NMR

123

data of 1,1 -diphosphinoferrocene ligands and their metal complexes

Ligand

δP (ppm)

Complex

δP (ppm)

dppf

−16.6b[16]

dppr dppo dmpf (1) depf (2) dippf (3)

−16.7b[6a] −13.4b[26] – −26.1b[9] −0.2a[8d]

[PdCl2 (dppf)] (26) [PtCl2 (dppf)] [PdCl2 (dppr)] (27) [PdCl2 (dppo)] (28) [PdCl2 (dmpf)] (29) [PdCl2 (depf)] (30) [PdCl2 (dippf)] (31) [PtCl2 (dippf)] (32) [ZnCl2 (dippf)] (33) [PdCl2 (dcypf)] (34) [PtCl2 (dcypf)] (39) [ZnCl2 (dcypf)] (40) [PdCl2 (dtbpf)] (35) [PtCl2 (dtbpf)] (41) [PdCl2 (dtpf)] (42) [PdCl2 (o-MeOdppf)] (37) [PdCl2 (5-Medfpf)] (36) [PdCl2 (o-i Prdppf)] (38) [PdCl2 (14)] (43) [PdCl2 (15)] (44)

34.5b[5a] 13.3b[5a] 32.3b[25] 37.9b[26] 15.4b[17] 42.8b[17] 65.8a[8d, 10] 30.7b[19] −6.8b[19] 56.6b[20] 23.0b[20] −15.1b[20] 66.7b[5b, 10] 35.7b[28] 33.5d , 38.5d and 48.8e[9] 40.1b[9] −5.25b[22] 38.1 and 41.0b,c[9] 11.5b[9] 119.9b[9]

dcypf (4)

−8.0b[5b, 27]

dtbpf (5)

28.2a[5b]

dtpf (6) 7 12 13 14 15

−36.8b[11e] −44.4b[9] −63.8b[22] −40.1b[9] −58.7b,f[9] 157.8b[9]

a in C D . b in CDCl . c both doublets, J = 47.7 Hz. d both doublets, J = 33.2 Hz. e in CDCl at −50 ◦ C. f quintuplet, 6 6 3 3

J = 30.5 Hz.

Figure 3.5 X-ray molecular structure of 35 showing its relatively larger bite angle (104. 2◦ ).21 Reprinted with permission from Ref. 21(a). Copyright 2007, American Chemical Society.

124

Ferrocenes: Ligands, Materials and Biomolecules

because of their applications in more challenging coupling reactions with aryl chlorides, electron rich substrates, sulfur containing heterocyles, etc. The earlier work of Hayashi demonstrated the importance of bite angle of the bidentate ligand, [PdCl2 (dppf)] in coupling.2 Studies by van Leeuwen further substantiated the importance of the bite angle.30 1,1 -Bis(di-tert-butylphosphino)ferrocene (5) is unique in the sense that it is an electron-rich and bulky bidentate ligand with a relatively larger bite angle. Its palladium complex is stable in air, unlike in the case of the bis(t-Bu3 P) analogue, making it very attractive for industrial processes. We found that [PdCl2 (dtbpf)] (35) (Figure 3.5) is one of the most active catalysts (Tables 3.3 and 3.4) for the Suzuki coupling of challenging substrates.10 The screening results obtained, with 2-halo-ﬂuoroanisoles as model substrates are summarised in Table 3.3. The superior activity of 35 in comparison with some of the commercially successful catalysts for coupling of aryl bromides and even heterocyclic chlorides is clearly demonsted.31 The general applicability of 35 in Suzuki coupling of electron rich Ar−Br and Ar−Cl substrates is shown in Table 3.4. Although 35 has the largest bite angle (104.2◦ , Figure 3.5) among the known dppf analogs,21 it is not signiﬁcantly larger than that of 31 (103.6◦ ).19 However, based on the results described in Table 3.3 (entries 9 and 10), 35 is more active in catalysis, especially for Suzuki coupling reactions. The uniqueness of this catalyst seems to be a combination of several factors such as electron richness, the steric bulkiness of the t-butyl group and the larger bite angle. In several instances, air-stable 35 was as active or even superior to the relatively air-sensitive palladium(I) and palladium(0) based catalysts: [{Pd(µ-Br)(t-Bu3 P)}2 ] and [Pd(t-Bu3 P)2 ].10 Table 3.3 Comparison of screening results of various palladiumcomplexes for Suzuki coupling reactions10 EtOH-Water [Pd] 1.0 mol%

MeO Ph-B(OH)2 +

Entrya 1 2 3 4 5 5 6 7 8 9 10 11

X

F

K2CO3, 80 °C 12 h

Catalyst

X

[PdCl2 (PPh3 )2 ] [PdCl2 (PCy3 )2 ] [PdCl2 (dppe)] 26 31 [PdCl2 (PPh3 )2 ] [PdCl2 (PCy3 )2 ] [PdCl2 (dppe)] 26 31 35 35

Br Br Br Br Br Cl Cl Cl Cl Cl Cl Cl

MeO Ph

F

Yield (%) 93 100 100 100 100 2 2 5 4 9 65 100

a All the reactions were conducted in EtOH-Water (1:1) at 80 ◦ C, except for entry 11, where DMF at 120 ◦ C was used.

Synthesis, Coordination Chemistry and Catalytic Use of dppf Analogs

125

Table 3.4 Generality of 35 in challenging coupling reactions10 B(OH)2

+

35 1.0 mol% ArX

Ar DMF, K2CO3, 120 °C, 15 h

Entry 1 2 3 4 5 6 7 8

Substrate 4-Bromoanisole 4-Bromo-3-methylanisole Bromomesitylene 4-Chlorotoluene 4-Cloroanisole 2-Chloro-4,6-dimethoxytriazine 2-Chloro-3-methylpyridine 2-Chlorothiophene

Yield (%) 100 96 85 98 98 100 89 84

The same ligand, 5 (dtbpf) in conjunction with Pd(OAc)2 was identiﬁed to be the best system for the practical synthesis of C-2-arylpurines,32a a class of biologically important compounds. Itoh also successfully explored the use of 5 in the direct synthesis of hetero-biaryl compounds containing an unprotected NH2 group via Suzuki coupling.32b In a palladium-catalysed Suzuki reaction of 4-bromotoluene with 4-methoxyphenylboronic acid, Beltskaya observed that ligands 7 and 13 are effective in giving quantitative conversion of the coupled product.9 However, these ligands are not expected to be as active as 5 or its palladium complex 35, but are more active than [PdCl2 (dppf)]. 3.2.4.2

Carbon–Heteroatom (Hartwig–Buchwald) Coupling

Hartwig has performed a systematic evaluation of the role of dppf analogs and other ligands in the amination of aryl halides.11e In this investigation, dppf was used as a ‘standard’. This interesting study demonstrated that ligands with smaller bite angles (∼90◦ ) gave best selectivities for monoarylation versus diarylation of amines, as did ligands with increased steric bulkiness. No ﬁrm explanation has been provided for this observation. The arylation of n-butylamine with (4-bromobutyl)benzene was conducted in the presence of various ligands in conjunction with [Pd(dba)2 ] (dba = dibenzylideneacetone). Reactions conducted with less electron rich ligands with respect to dppf: 10 and 11 showed an increase in the formation of the dehalogenated product (arene), while electron rich ligands such as 8 gave decreased amounts of the arene product. This is contrary to the expectation that electron poor ligands accelerate reductive elimination, thereby increasing the amine/arene ratio. Amination reactions of iso-butylamine catalysed by a Pd–dppf complex gave a two-fold increase in the amount of dehalogenation (arene) product relative to the analogous reactions with n-butylamine. A larger increase in arene was observed when dtbpf was used. Increasing the size of the amine

126

Ferrocenes: Ligands, Materials and Biomolecules

led to an increased formation of mono arylation products for both of the ligands, dppf and dtbpf. In fact, only a trace amount of diarylation product was observed for dtbpf. Very interestingly, subsequent work from Hartwig’s group identiﬁed 5 as an exceptionally good ligand for palladium-catalysed amination of unactivated aryl chlorides with aniline in excellent yield of the mono arylation product (Scheme 3.2).33 Our internal study indicates that the fully formed catalyst (35) is also effective for similar transformations. Palladium complexes of 7 and 13 were also reported recently to show high activity in amination reactions.9

Pd(dba)2/5/Base NH2 +

Cl

NH

110 °C/24 h 95 %

Scheme 3.2 Palladium-catalysed Ar-Cl coupling of aniline using dtbpf (5)

Hartwig also carried out some kinetic studies to determine the trans effect on the rates of reductive elimination from arylpalladium amino complexes bearing symmetrical and unsymmetrical dppf analogs.14b Solutions of symmetrical and unsymmetrical [L2 Pd(Ar) (NMeAr )], where L2 = dppf, 8, 9, 19, 24, 25; Ar = p-CF3 C6 H4 ; and Ar = p-CH3 C6 H4 , Ph and p-MeOC6 H4 , underwent C−N bond forming reductive elimination at – 15 ◦ C to form the corresponding N -methyldiarylamine in high yield. Complexes of symmetrical dppf analogs with electron withdrawing groups underwent eliminations faster than those with electron donating groups. The orientation of the unsymmetrical ligand affected the rate of reductive elimination. Complexes with the weaker donor trans to nitrogen and stronger donor trans to the palladium-bond aryl group underwent reductive elimination faster than the regioisomeric complex with stronger donor trans to nitrogen and weaker donor trans to palladium-bound aryl group. This could be due to the fact that Pd−P distance trans to the amino group is shorter than Pd−P distance trans to the aryl group. The magnitude of the electronic effect of the aryl group on amino nitrogen was larger than the electronic effect of the aryl group on the phosphorus atom. At the same time, Hartwig also studied steric and electronic effects on the reductive elimination of diaryl ethers from palladium(II) intermediates. Reductive elimination of diaryl ether from bulky dtbpf was 100 times faster than with the corresponding dppf complex. This study also suggested that a bulky monodentate ferrocene-based ligand, 1-(di-tert-butylphosphino)-1 ,2 ,3 ,4 ,5 -pentaphenylferrocene (Q-Phos), favoured the formation of diaryl ethers in excellent yield.23 3.2.4.3

Palladium-Catalysed Arylation of Ketones and Malonates

The works of Hartwig,34 Buchwald35 and Satoh36 were instrumental in developing the ﬁeld of palladium-catalysed arylation. In 1999, it was reported that dtbpf was an excellent ligand for the palladium-catalysed arylation of a series of ketones and malonates (Scheme 3.3).37

Synthesis, Coordination Chemistry and Catalytic Use of dppf Analogs

+

X

R

X=Br, Cl, OTs

O

Pd(dba)2 / 5

O

R

R'

127

R

NaOt Bu, 25–100 °C

R' Ar

R, R'=alkyl, aryl or R ester and R' alcohol

Scheme 3.3 General α-ketone arylation reaction

Once again the activity of the dtbpf ligand was attributed to: i) steric effects, increasing the energy of the more stable high coordinate species; and ii) alkylphosphines being more resistant toward P−C cleavage processes than arylphosphines. However, subsequent works by Hartwig on α –amide arylation,38 α-cyanoester arylation39 and monoarylation of nitriles40 suggested that bulky monophosphines of the type P(t-Bu)3 or Q-Phos were adequate ligands for such C−C bond-forming reactions. Recent work from our laboratory indicated that per-formed catalyst [PdCl2 (dtbpf)] (35) is superior to the in situ systems. With this catalyst, we have been able to accomplish α-arylation of various ketones with a wide range of aryl bromides and chlorides in excellent conversion (80–100 %) and selectivity at catalyst loading as low as 0.1 %.21 Part of the results are summarized in Table 3.5. 3.2.4.4

Palladium-Catalysed Indolisation

Lu and Senanayake developed an elegant, practical and economical process41 for the synthesis of 2,3-disubstituted indole compounds via palladium-catalysed indolisation of 2-bromo- or chloroanilines and their derivatives with internal alkynes (Scheme 3.4). In comparison to the use of various bulky electron rich monodentate ligands such as Cy3 P, t-Bu3 P and the (biaryl)PR2 , dtbpf seems to be the best ligand for giving good yield and regioselectivity of the ﬁnal products.

R3

X R1

+ NH2

R2

Pd(OAc)2 5 mol% 5 10 mol% K2CO3 2.5 eq. NMP, 110–130 °C

R2 R1

R2 N H

Scheme 3.4 Palladium-catalysed indolization of 2-bromo- or chloroanilines

3.2.4.5 Cullen12

Rhodium-Catalysed Hydrogenation

explored the use of [Rh(NBD)(dtbpf)]ClO4 in achiral hydrogenation during the mid 1980s. The study was not detailed nor did it explore other systems, especially in the context of achiral catalysis. However, it implied that a match between the electronic and steric effects of the ligands and the substrates was a key factor in deciding the kinetics. The electron-rich ligand dtbpf increases the reactivity of the complex. However, the bulkiness of dtbpf seems to have a negative rate effect in the

128

Ferrocenes: Ligands, Materials and Biomolecules Table 3.5 Effect of the Ketone Substrate in the 35-catalyzed αArylation21 O Ar-Cl

entry

+

R'

R

O

2mol% (DtBPF)PdCl2 1.1 equiv. NaOtBu

R'

product

substrate O

Ar

R

100 °C, dioxane, 1M

conv. (%)

O 99 (95)

1 O O

99 (95)

2 F F

O

O 3

98 (91) OMe OMe

O

O 99

4 MeO

MeO O

5

F3C

O F3C 98 (60)

CF3 O

CF3 O

6

99 (92)

O

OMe

O

7

53

O

OMe

O

8

92

OMe (Reproduced from Grasa and Colacot Org. Lett., 9, 5489–5492. Copyright (2007), with permission from American Chemical Society.)

Synthesis, Coordination Chemistry and Catalytic Use of dppf Analogs

129

case of certain bulky substrates. The other ferrocene-based phosphine ligands, such as PPFA and even its t-Bu2 P substituted analogue, are less effective in comparison to [Rh(NBD)(dtbpf)]ClO4 . 3.2.4.6

Palladium-Catalysed Cooligomerisation and Copolymerisation

Dinjus was the ﬁrst to report the activity of 1,1 -bis(dialkylphosphino)ferrocene ligands for synthesising δ-lactone by a palladium-catalysed reaction of 1,3-butadiene with CO2 (Scheme 3.5).8d Ligands 3 and 23 proved to be much more effective in these palladiumcatalysed transformations, in comparison with the classic dppf, in terms of selectivity to δ –lactone and Turn Over Number (TON).

2

+

CO2

[Pd] / L / CH3CN O

70 °C, 15 h

O

L = dppf, 3, 23

Scheme 3.5 Palladium-catalysed telomerisation, formation of 2-ethyliden-6-hepten-5-olid

Collaborative work between the groups of Bianchini and Gusev led to a study on ethylene carbonylation by palladium-catalysts containing 1,1 -bis(dialkylphosphino) ferrocenes, such as ligands 1, 2 and 3.17 It was demonstrated that the nature of the substituents on the phosphorus atoms has a major role on the distribution of products: copolymers (high molecular weight) or oligomers (low molecular weight). The inﬂuence of the ligand in the reaction between ethylene and CO is summarised in Scheme 3.6. O OMe

n + CO

O

[Pd(OTs)(H2O)(L-L)]OTs

OR

MeOH, additive

When L-L = 3 (dippf) O

O

O

When L-L = 1 (dmpf) or 2 (depf) n>1

OMe OMe O O

O OMe

MeO O

MeO

OMe O

O

Scheme 3.6 Inﬂuence of ligand on the selectivity of CO/ethylene co-polymerisation

Steric effects apparently play a major role in determining the catalytic activity of 3, hence the formation of only low molecular weight ketone, ester, keto-ester and di-ester. The formation of a dative Fe → Pd bond (when L–L = 3) during the catalytic cycle

130

Ferrocenes: Ligands, Materials and Biomolecules

is forced by the presence of the bulky and electron-donating i-Pr groups, explaining the selective formation of low molecular weight oxygenates.17 This phenomenon had been previously observed with Pd-catalysts containing bulky ligands such as 5.42, 43

3.3

Chiral Ligands

Recently there has been a renewed interest in the area of chiral ferrocenylphosphines in asymmetric organic syntheses, as evidenced by the recent review articles by Colacot,1d Zhang,44 Bianchini7c and Long7d highlighting the applications of these ligands in organic synthesis via homogeneous catalysis. However, in this chapter only 1,1 -disubstituted ferrocene ligands will be focused on. These bis-substituted ligands can be further classiﬁed into symmetrical, unsymmetrical and P-chiral. 3.3.1

1,1 -Disubstituted Chiral Symmetrical Ferrocenylphosphines

The number of ligands known in this area is still limited. However, catalytically active systems have already been reported in the literature. In 1994 Burk45 prepared a DuPhos (ﬁve membered phospholane) analog of a ferrocene based ligand by reacting 1,1 -bisphosphinoferrocene with a lithiating agent, followed by treatment with two equivalents of chiral, hexanediol cyclic sulfates and another two equivalents of n-BuLi to facilitate the ring closure (50–51). Subsequently, Marinetti/Genˆet46 and Berens/Burk47 modiﬁed the ligand by using the same methodology to construct conformationally constrained 4-membered cyclic phosphines, known today as FerroTANEs (45–49). Recently, Zhang extended this methodology to construct a ‘DuPhos’ ligand with a mannitol skeleton (52).48 Zhang also synthesised the ferrocene-based binaphane 53 from 1,1 -bis(phosphino)ferrocene.49 The various synthetic routes are summarised in Scheme 3.7. In 1998, Reetz50 very cleverly synthesised and reported the X-ray structure of 1,1 bis(binaphthylphosphonito)ferrocene (54) from binaphthol in 90 % yield. The synthesis details are summarised in Scheme 3.8. FerroPHOS (55) and FERRIPHOS (56–60) are new types of C2 -symmetric ligands developed by Kang51 and Knochel,52 respectively. The basic schemes are summarised in reviews by Knochel52b and Santelli.53 The bis oxazoline based ligands, N,N,P,P type (61–62) have been prepared from chiral amino-alcohol.1d, 54 In addition, Brunner reported the preparation of 63, a symmetric chiral bis(dimenthyl)phosphine.14a The basic structures of these complexes are represented in Figure 3.6. 3.3.2

1,1 -Disubstituted Chiral Unsymmetrical Ferrocenylphosphines

To our knowledge there are only three 1,1 -disubstituted chiral unsymmetrical phosphines known today. With limited examples, these ligands do not seem to have any special advantages in catalysis. The synthetic routes are similar to the methods described in the earlier sections (Scheme 3.9). The menthyl14a and phospholane55 derivatives (23 and 64) were synthesised from 1-(diphenylphosphino)-1 -bromoferrocene, while 65, the menthoxy derivative,56 was prepared from 1-(diphenylphosphino)-1 (dichlorophosphino)ferrocene.

Synthesis, Coordination Chemistry and Catalytic Use of dppf Analogs R

R O O

S

R (2 eq)

R O

R

O O

(2 eq)

R = Me (50) R = Et (51)

P

R Fe

P

R

R

PH2 H 2P

R

R

LDA/THF

Fe

Fe

P

R S

P

R

n-BuLi/THF

O

R = Me (45) R = Et (46) R = Pr (47) R = i Pr (48) R = tBu (49)

O O

131

O

O

O O S O O

O (2 eq)

P

O

O

Fe

n-BuLi/THF

P

52

O

Cl (2 eq)

P

Cl

Fe NaH/THF

P 53

Scheme 3.7 General synthetic routes to 1,1 -bis(phosphaheterocyclic)ferrocenes P(NEt2)2 n-BuLi/TMEDA Fe

PCl2 HCl/Et2O

Fe

Fe

−78 °C

2 (NEt2)2PCl P(NEt2)2

PCl2

R-(+)-BINOL Toluene/Heat 36 h

O P O

Fe

O P O

54

Scheme 3.8 Synthesis of Reetz’s ferrocene based diphosphonite

132

Ferrocenes: Ligands, Materials and Biomolecules CHEt2

O

R

PPh2 Fe

Fe PPh2

CHEt2 (S,S)- FerroPHOS, 55

N

Ar PPh2 PPh2 Ar

R FERRIPHOS

Fe

R PMen2

PPh2 PPh2 N

Fe PMen2 R

O Bis-oxazoline based fc ligand

R= Me, Ar = Ph, 56 R= Me, Ar = o-Tol, 57 R= Me, Ar = 2-Np, 58 R= i-Pr, Ar = Ph, 59 R= NMe2, Ar = Ph, 60

63

R= i-Pr, 61 R= t-Bu, 62

Figure 3.6 Structures of the main C2 symmetric bisphosphine based ligands

X = PH2

O O S O O

PPh2 Fe P

n-BuLi 64

PPh2 X

Fe

X = Br/n-BuLi

PPh2 Fe

(Men)2PCl

P(Men)2

23

X = PCl2

Menthol

Fe

PPh2 P(OMen)2

Et3N 65

Scheme 3.9 Synthetic routes for 1,1 -disubstituted chiral unsymmetrical ferrocenylphosphines

3.3.3

1,1 -Disubstituted P-Chiral Ferrocenylphosphines

The ﬁrst P-chiral bisphosphine, DIPAMP, was discovered and its catalytic applications developed by Knowles about two and a half decades ago, for which he shared the 2001 Nobel Prize in Chemistry.57 However, the subsequent developments of new P-chiral

Synthesis, Coordination Chemistry and Catalytic Use of dppf Analogs

133

ligands have been very slow due to the difﬁculties in synthesising them as well as their inherent propensity to undergo racemisation. Zhang’s review summarises the various ligands developed in this area.44 A few efﬁcient P-chiral 1,1 -bis(phosphino)ferrocenes have been reported independently by Mezzetti58 and van Leeuwen.59 The general synthetic route is similar to the Jug´e –Genˆet60 approach, which avoids optical resolution, although there are subtle differences between the methods employed by Imamoto,61 Mezzetti58 and van Leeuwen.59 Using this methodology, both symmetrical and unsymmetrical ligands can be synthesised.62 The general pathway for the synthetic methodology is summarised in Scheme 3.10. The X-ray structures of a few examples of these ligands are reported in the literature.59, 61 An X-ray structure of the (R,R)-66, is represented in Figure 3.7.

Ar P Ar

P

Fe

R1

R1

OMe Ar P R1 BH3 OMe Ar P R1 BH3

Br Fe

Br

sec-BuLi (1 eq.) inversion

tBu

P Me

P

Fe

Fe Br

OMe P R Ar 2 BH3

1. sec-BuLi (1 eq.) inversion 2. CF3SO3H/KOH retention

1. sec-BuLi (1 eq.) inversion 2. CF3SO3H/KOH retention

Ar

tBu

P Ar

P

Ar = 1-naphthyl, R1 = 3-anisyl, 74 R1 = 3-CF3Ph, 75

Ar P BH3 R1

Me

77, via HPLC resolution of the borane adduct.

Ar = Ph, R1 = 1-naphthyl, 66 R1 = 2-naphthyl, 67 R1 = 1-anisyl, 68 R1 = 2-anisyl, 69 R1 = 3-anisyl, 70, R1 = 3-CF3Ph, 71, R1 = 2-biphenylyl, 72 R1 = 9-phenanthryl, 73

Fe

Ar = 1-naphthyl, R1 = 3-anisyl and R2 = 3-CF3Ph, 76

R1

R2

Scheme 3.10 General synthetic methodology for P-chiral analogs of dppf

3.3.4

Coordination Chemistry

Since these chiral ligands are used mainly in asymmetric hydrogenation reactions, their coordination chemistry is dominated by rhodium, iridium and ruthenium, although limited examples of platinum and palladium compounds are known.

134

Ferrocenes: Ligands, Materials and Biomolecules C8′ C15′

C8′

C10′ C14′

C7′ C11′

C13′

C6′

C12′

C17′

C5′

C1′

C4′ c19

C20

P1′ C16′ C21′

C18′ C2′

C3′

C18′

C18

Fe C5

C20′

C21 C16

C17 P1 C1

C4 C3

C12

C2

C13

C11

C6 C7

C14 C10 C8

C15

C9

Figure 3.7 X-ray structure of the P-chiral ligand (R,R)-66.59 Reprinted with permission from Ref. 59. Copyright 1999, American Chemical Society.

Typically, cationic [Rh(COD)2 ]X or [Rh(NBD)2 ]X (X = ClO4 , BF4 , PF6 , CF3 SO3 etc) are used as the rhodium precursors. Since the catalysts can be prepared in situ, fully formed complexes have not been isolated in all cases, although isolated pure complexes should offer advantages in increasing the selectivity and activity. In 2003, Heller synthesised and solved the X-ray structure of [Rh{(R,R)-Et-FerroTANE}(NBD)][BF4] based on ligand 46.63 The P−Rh−P bite angle in this case is 98.27(5)◦ . In the same year, Marinetti reported the X-ray structure of [Rh{(S,S)-i-Pr-FerroTANE}(COD)]OTf, based on ligand 48, with a bite angle of 96.90(3)◦ .64 Ruthenium compounds are more difﬁcult to synthesise in pure form, however Marinetti synthesised and reported the X-ray structure of a Ru–FerroTANE complex based on 48.64 The ruthenium atom is approximately octahedral. The P–Ru–P bite angle, 98.57(3)◦ is deviated from the ideal 90◦ ; however, such deviations are expected for these types of ferrocenylphosphine ligands.64 3.3.5

Catalysis Applications in Asymmetric Synthesis

The ligands mentioned above are utilised mostly for chiral hydrogenation reactions of C=C, C=O and C=N double bonds. There are some reports on allylic alkylation and chiral Kumada coupling as well.

Synthesis, Coordination Chemistry and Catalytic Use of dppf Analogs

3.3.5.1

135

Hydrogenation of C=X (where X = C, O and N)

One of the major applications for chiral 1,1 -disubstituted ferrocenylphosphines is the rhodium-catalysed hydrogenation of oleﬁns. The three model substrates that are typically employed to test the activities of different ligands are shown in Scheme 3.11. CO2R1 R3

CO2R1 ∗ NHCOR2

[Rh], H2

NHCOR2

R3

Methyl-2-AcetamidoAcrylate, MAA : R1=R2=Me, R3=H Methyl AcetamidoCinnamate, MAC : R1=R2=Me, R3=Ph

CO2R1 R3

CO2R2

[Rh], H2 R3

CO2R1 ∗ CO2R2

DiMethyl Itaconate, DMI : R1=R2=Me, R3=H

Scheme 3.11 Model reactions for asymmetric hydrogenation

Burk and Gross45 used disubstituted ferrocenylphosphines successfully in rhodiumcatalysed asymmetric hydrogenation. In most of the cases, the active catalyst was formed in situ by mixing the metal precursor [Rh(L-L)]2 X (where L−L = NBD or COD and X = BF4 or ClO4 ) with the chiral bidentate phosphine. However, studies show that the use of a fully formed catalyst increases the ee of the hydrogenation.58 Table 3.6 is a compilation of the successful results obtained with a number of chiral 1,1 -disubstituted ferrocenylphosphines for the hydrogenation of the model substrates MAA, MAC and DMI. In the space available, it is practically impossible to summarise all the work that has been done in the area of asymmetric hydrogenation of oleﬁns using 1,1 -disubstituted ferrocenylphosphines. However, it has been attempted to reference all the relevant work of Burk,45, 47 Marinetti/Genˆet/Jug´e,46, 60, 64 Zhang,48, 49 Reetz,50 Kang,51 Knochel,52 Mezzetti,58 van Leeuwen,59 Brunner,14a Hsiao/Rivera/Rosner65 and Heller.63, 66 Iridium complexes of Zhang’s f-binaphane 53 were found to be efﬁcient for the enantioselective hydrogenation of acyclic imines49 and even more active for the reductive amination of aryl ketones.67 The use of iodine (I2 ) as an additive was also explored to obtain higher ee for less hindered aryl substrates. Interestingly, when 2,6-dimethylphenyl-substituted substrate was used in conjunction with iodine, it led to a dramatic decrease in the conversion and ee.49 It is therefore thought that the catalytic process must differ from those reactions that use no additives. The use of iridium complexes of f-binaphane allows the formation of chiral primary amines from aryl ketones, via Cerium Ammonium Nitrate (CAN) oxidation.68 However Ir-53 catalysed reductive amination does not work with alkyl ketones.

136

Ferrocenes: Ligands, Materials and Biomolecules Table 3.6 Rh-catalysed hydrogenation of the three model substrates with various 1,1 -ferrocene diphosphines ee (%)

Ref.

1 1

98.0 97.0

47

4.0

18

91.0 90.0

46

1.0 5.0 4.0

24 24 18

96.0 83.0 94.0

5.5 1.0 3.1

12 0.5 1

89.9 99.9 99.5

Ligand

Substrate

Solvent

H2 (bar)

46 47

DMIa DMIa

MeOH MeOH

5.5 5.5

45

DMIb MACb

MeOH

MACc MACd MAAa

MeOH

DMIe MAAe MACe

MeOH

45 48 52

MeOH

THF

t (h)

64 46

48

54

DMIf MAAg

DCM

1.3

20

99.5

50

55

MACh MAAh

EtOH MeOH

2.0 2.0

3–12 3–12

97.6 97.5

51

1.0 1.0 1.0 1.0

see see see see

i i j j

98.6 97.9 98.6 97.7

52d

58b

59

56 57

MACi MAAi MACj MAAj

MeOH MeOH

MeOH

1.0

17

66

MACk MACl N -Me-MACm N -Me-MACm

91.0 96.0 97.0 90.0

66 73

MACn MACn

MeOH

2.0

6

97.3 98.7

68

a 0.5 mol % of fully formed complex [Rh(L-L)(COD)][BF ] at 20 ◦ C. b 0.5 mol % of in situ 4 formed catalyst using [Rh(COD)2 ](CF3 SO3 ) at r.t. c 0.5 mol % of fully formed complex, [Rh(LL)(COD)](CF3 SO3 ) at 50 ◦ C. d Same conditions as in (c), except at 23 ◦ C. e 1.0 mol % of in situ complex formed with [Rh(COD)2 ][PF6 ] at r.t. f 0.5 mol % of fully formed complex, [Rh(LL)(COD)][BF4 ] at r.t. g 0.1 mol % of in situ complex formed using [Rh(COD)2 ][BF4 ] at r.t. h 1.0 mol % of in situ complex formed using [Rh(COD) ][BF ] at 20–23 ◦ C. i 1.0 mol % of in 4 2 situ formed complex using [Rh(COD)2 ][BF4 ] at r.t., half-time is less than 2 min. j 1.0 mol % of in situ formed complex using [Rh(NBD)2 ][BF4 ] at r.t., half-time is less than 2 min. k 0.5 mol % of in situ formed complex with [Rh(NBD)2 ][BF4 ] at 20 ◦ C. l Same as (k) but fully formed complex [Rh(L-L)(COD)][BF4 ] was used. m Same as (l) except at 25 ◦ C. n 1.0 mol % of in situ formed complex using [Rh(NBD)2 ][ClO4 ] at 25 ◦ C.

Another way of obtaining simple chiral compounds, in this case alcohols, has been developed by Imamoto.61 Indeed rhodium complexes of 77 were found to be very efﬁcient catalysts for the asymmetric hydrosilylation of a series of aryl-alkyl ketones to chiral alcohols. The silylation agent of choice is (1-naphthyl)phenylsilane in combination with pre-formed catalyst, [Rh(53)(NBD)][BF4]. The P-stereogenic ligands 66

Synthesis, Coordination Chemistry and Catalytic Use of dppf Analogs

137

and 68 were also tested for the Ru-catalysed hydrogenation of α- and β-keto esters.58a The ee’s of the hydrogenated products were only moderate. Therefore, these ligands cannot compete with the ligands of the BINAP family. 3.3.5.2

Chiral C−X Bond Formation, (where X = C or P)

Glueck reported the formation of chiral palladium(0) trans-stilbene complexes of 46 and 50.69 These new compounds can then be used as catalyst precursors for asymmetric phosphination. It must be noted that palladium complexes of Et-FerroTANE (46) have also been applied for the design of chiral triarylphosphines.68 This enantioselective palladium-catalysed C−P cross-coupling reaction between aryl iodides and diarylphosphines is a powerful tool for generating new chiral phosphines. The hydroformylation of alkenes is one of the largest industrial reactions in homogeneous catalysis, which has been recently extended to the asymmetric synthesis of aldehydes.70 van Leeuwen has performed a fairly detailed investigation on the enantioselective hydroformylation, using a series of 1,1 -bis[(1-naphthyl)arylphosphino]ferrocene (66, 74–76, Scheme 3.10).62 In this study, only low Turn Over Frequency (TOF) and poor selectivity (low branched/linear ratio; superior formation of achiral terminal aldehyde over the formation of chiral branched aldehyde and low ee’s) have been observed. However, ligands with electron-withdrawing group on the phosphorus (75 > 76 > 74) seem to positively inﬂuence the TOF and ee’s.62, 71 The palladium-catalysed allylic substitution reaction is yet another way to generate new C–C bonds.72 The C2 -symmetrical P,P-chelating ferrocene moieties of type, 55 (FerroPHOS derivatives) have been used by Ikeda successfully.73 A combination of the planar-chiral ligands with (π-allyl)palladium chloride is a very active in situ system for the allylation of dimethylmalonate with sterically hindered allylic substrates. The FerriPhos family of C2 -symmetrical ligands, and especially palladium complexes of 60, showed good activity for the asymmetric Kumada–Hayashi coupling reaction.52a The Sanoﬁ–Aventis ligand, JAFAPhos is also a good ligand system for asymmetric Kumada coupling reaction, a challenging reaction in asymmetric catalysis.5b, 74

3.4

Conclusion

The chemistry of dppf analogs has shown a very positive impact in the area of metal-catalysed organic reactions such as C–C and C–N coupling, asymmetric hydrogenation, etc. This area has tremendous growth potential due to the commercialisation of some of the ligand/catalysts (e.g. 3, 34, 35). It is already evident in the manufacture of APIs (Active Pharmaceutical Ingredients), that the catalytic processes described above play a key part in the production of new drugs.

3.5

Acknowledgements

Fred Hancock and Bill Tamblyn, the respective Technical and Commercial Directors of Johnson Matthey Catalysis and Chiral Technologies, are acknowledged for their useful discussions and editorial assistance.

138

Ferrocenes: Ligands, Materials and Biomolecules

References 1. (a) A. Togni, T. Hayashi (Eds), Ferrocenes: Homogeneous Catalysis, Organic Synthesis, Materials Science, Wiley-VCH Verlag GmbH, Weinheim, Germany (1995); (b) J.J. Bishop, A. Davison, M.L. Katcher et al. J. Organomet. Chem., 1971, 27, 241–249; (c) G. Bandoli, A. Dolmella, Coord. Chem. Rev., 2000, 209, 161–196; (d) T.J. Colacot, Chem. Rev., ˇ epniˇcka, Eur. J. Inorg. Chem., 2005, 19, 3787–3803; 2003, 103, 3101–3118; (d) P. Stˇ (e) R.G. Array´as, J. Adrio, J.C. Carretero, Angew. Chem. Int. Ed., 2006, 45, 7674–7715. 2. T. Hayashi, M. Konishi, Y. Kobori et al. J. Am. Chem. Soc., 1984, 106, 158–163. 3. K-S. Gan, T.S.A. Hor, 1,1 -Bis(diphenylphosphino)ferrocene – Coordination Chemistry, Organic Syntheses, and Catalysis in A. Togni, T. Hayashi (Eds), Ferrocenes: Homogeneous Catalysis, Organic Synthesis, Materials Science, Wiley-VCH Verlag GmbH, Weinheim Germany (1995), pp. 1–104. 4. (a) J.F. Hartwig, Angew. Chem. Int. Ed., 1998, 37, 2046–2067; (b) J.F. Hartwig, Acc. Chem. Res., 1998, 31, 852–860. 5. (a) T.J. Colacot, H. Qian, R. Cea-Olivares, S. Hernandez-Ortega, J. Organomet. Chem., 1998, 637–639, 691–697; (b) www.jmcatalysts.com/pharma 6. Although not discussed in detail in this chapter, examples of osmium and ruthenium based dppf analogs are also known in the literature. (a) S. Li, B. Wei, P.M.N. Low et al. Dalton Trans., 1997, 1289–1293; (b) S.L. Martinak, L.A. Sites, S.J. Kolb et al. J. Organomet. Chem., 2006, 691, 3627–3632; (c) O.V. Gusev, T.A. Peganova, A.M. Kalsin et al. J. Organomet. Chem., 2005, 690, 1710–1717; (d) B. Wei, S. Li, H. Kee, T.S.A. Hor, J. Organomet. Chem., 1997, 527, 133–136; (e) C. Nataro, A.N. Campbell, M.A. Ferguson et al. J. Organomet. Chem., 2003, 673, 47–55. 7. (a) T.J. Colacot, Platinum Met. Rev., 2001, 45, 22–30; (b) L-X. Dai, T. Tu, S-L. You et al. Acc. Chem. Res., 2003, 36, 659–667; (c) P. Barbaro, C. Bianchini, G. Giambastiani, S.L. Parisel, Coord. Chem. Rev., 2004, 248, 2131–2150; (d) R.C.J. Atkinson, V.C. Gibson, N.J. Long, Chem. Soc. Rev., 2004, 33, 313–328. 8. (a) W.R. Cullen, T-J. Kim, F.W.B. Einstein, T. Jones, Organometallics, 1983, 2, 714–719; (b) M. Kumada, Y. Kiso, M. Umeno., J. Chem. Soc. Chem Commun., 1970, 611–614; (c) I.R. Butler, W.R. Cullen, T-J. Kim, Synth. React. Inorg. Met-Org. Chem. 1985, 15, 109; (d) A.R. Elsagir, F. Gassner, H. G¨orls, E. Dinjus, J. Organomet. Chem., 2000, 597, 139–145. 9. (a) O.V. Gusev, T.A. Peganova, A.M. Kalsin et al. Organometallics, 2006, 25, 2750–2760; (b) A.M. Kalsin, P.V. Petrovsky, K.A. Lyssenko, O.V. Gusev, Organometallics, 2006, 25, 1854–1856. 10. T.J. Colacot, H.A. Shea, Org. Lett., 2004, 6, 3731–3734 and references therein. 11. (a) J.D. Unruth, J.R. Christenson, J. Mol. Catal., 1982, 14, 19–34; (b) I.E. Nifant’ev, A.A. Boricenko, L.F. Manzhukova, E.E. Nifant’ev, Phosphorus, Sulfur, Silicon, 1992, 68, 99–106; (c) S. Hillebrand, J. Bruckmann, C. Kr¨uger, M. Haenel, Tetrahedron Lett., 1995, 36, 75–78; (d) R-J. de Lang, J. van Soolingen, H.D. Verkruijsse, L. Brandsma, Synth. Commun., 1995, 25, 2989–2991; (e) B.C. Hamann, J.F. Hartwig, J. Am. Chem. Soc., 1998, 120, 3694–3703. 12. (a) W.R. Cullen, T-J. Kim, F.W.B. Einstein, T. Jones, Organometallics, 1985, 4, 346–351; (b) I.R. Butler, W.R. Cullen, T-J. Kim et al. Organometallics, 1985, 4, 972–980. 13. (a) A.G. Osborne, R.H. Whiteley, J. Organomet. Chem., 1975, 101, C27–C28; (b) D. Seyferth, H.P. Withers, Organometallics, 1982, 1, 1275–1282. 14. (a) H. Brunner, M. Janura, Synthesis, 1998, 45–55 see references therein; (b) M. Yamashita, J.V.C. Vicario, J.F. Hartwig, J. Am. Chem. Soc., 2003, 125, 16347–16360. 15. (a) Y. Kiso, M. Kumada, K. Tamao, M. Umeno, J. Organomet. Chem., 1973, 50, 297; (b) K. Li, I.A. Guzei, J. Darkawa, Polyhedron, 2003, 22, 805–810.

Synthesis, Coordination Chemistry and Catalytic Use of dppf Analogs

139

16. T.J. Colacot, R.A. Teichman, R. Cea-Olivares et al. J. Organomet. Chem., 1998, 557, 169–179. 17. C. Bianchini, A. Meli, W. Oberhauser et al. Organometallics, 2005, 24, 1018–1030. 18. A.L. Boyes, I.R. Butler, S.C. Quayle, Tetrahedron Lett., 1998, 39, 7763–7766. 19. J.H.L. Ong, C. Nataro, J.A. Golen, A.L. Rheingold, Organometallics, 2003, 22, 5027–5032. 20. L.E. Hagopian, A.N. Campbell, J.A. Golden et al. J. Organomet. Chem., 2006, 691, 4890–4900. 21. (a) G. Grasa, T.J. Colacot, Org. Lett., 2007, 9, 5489–5492. (b) G. Grasa, T.J. Colacot, Org. Process Res. Dev., 2008 (in press). 22. A. Fihri, J-C. Hierso, A. Vion et al. Adv. Synth. Catal., 2005, 347, 1198–1202. 23. G. Mann, Q. Shelby, A.H. Roy, J.F. Hartwig, Organometallics, 2003, 22, 2775–2789. 24. M.A. Zuideveld, B.H.G. Swennenhuis, M.D.K. Boele et al. Dalton Trans., 2002, 2308– 2317. 25. J.M. Brown, P.J. Guiry, Inorg. Chim. Acta, 1994, 220, 249–259. 26. O.V. Gusev, A.M. Kalsin, P.V. Petrovskii et al. Organometallics, 2003, 22, 913–915. 27. T-J. Kim, Y-H. Kim, H-S. Kim et al. Bull. Korean Chem. Soc., 1992, 13, 588–592. 28. F.N. Blanco, L.E. Hagopian, W.R. McNamara et al. Organometallics, 2006, 25, 4292–4300. 29. (a) M. Nishiyama, T. Yamamoto, Y. Koie, Tetrahedron Lett., 1998, 39, 617–620; (b) T. Yamamoto, M. Nishiyama, Y. Koie, Tetrahedron Lett., 1998, 39, 2367–2370; (c) A.F. Littke, G.C. Fu, Angew. Chem., Int. Ed., 2002, 41, 4176–4211 and references therein; (d) G.C. Fu, J. Org. Chem., 2004, 69, 3245–3249. 30. P.W.N.M. van Leeuwen, P.C.J. Kamer, J.N.H. Reek, P. Dierkes, Chem. Rev., 2000, 100, 2741–2769. 31. (a) N. Kudo, M. Perseghini, G.C. Fu, Angew. Chem., Int. Ed., 2006, 45, 1282–1284; (b) C. Yi, R. Hua., J. Org. Chem., 2006, 71, 2535–2537; (c) A.S. Guram, A.O. King, J.G. Allen et al. Org. Lett., 2006, 8, 1787–1789. 32. (a) T. Itoh, K. Saito, T. Mase, Adv. Synth. Catal., 2004, 346, 1859–1867; (b) T. Ito, T. Mase, Tetrahedron Lett. 2005, 46, 3573–3577. 33. B.C. Hamann, J.F. Hartwig, J. Am. Chem. Soc., 1998, 120, 7369–7370. 34. B.C. Hamman, J.F. Hartwig, J. Am. Chem. Soc., 1997, 119, 12382–12383. 35. M. Palucki, S.L. Buchwald, J. Am. Chem. Soc., 1997, 119, 11108–11109. 36. T. Satoh, Y. Kawamura, M. Miura, M. Nomura, Angew. Chem., Int. Ed., 1997, 36, 1740–1742. 37. M. Kawatsura, J.F. Hartwig, J. Am. Chem. Soc., 1999, 121, 1473–1478. 38. S. Lee, J.F. Hartwig, J. Org. Chem., 2001, 66, 3402–3415. 39. N.A. Beare, J.F. Hartwig, J. Org. Chem., 2002, 67, 541–555. 40. L. Wu, J.F. Hartwig, J. Am. Chem. Soc., 2005, 127, 15824–15832. 41. M. Shen, G. Li, B.Z. Lu et al. Org. Lett., 2004, 6, 4129–4132. 42. M.A. Zuideveld, B.H.G. Swennenhuis, P.C.J. Kamer, P.W.N.M. van Leeuwen, J. Organomet. Chem., 2001, 637–639, 805–808. 43. P.W.N.M. van Leeuwen, M.A. Zuideveld, B.H.G. Swenhuis et al. J. Am. Chem. Soc., 2003, 125, 5523–5539. 44. W. Tang, X. Zhang, Chem. Rev., 2003, 103, 3029–3070. 45. M.J. Burk, M.F. Gross, Tetrahedron Lett., 1994, 35, 9363–9366. 46. A. Marinetti, F. Labrue, J-P. Genˆet, Synlett. 1999, 1975–1977. 47. U. Berens, M.J. Burk, A. Gerlach, W. Hems, Angew. Chem., Int. Ed., 2000, 39, 1981–1984. 48. D. Liu, W. Li, X. Zhang, Org. Lett., 2002, 4, 4471–4474. 49. D. Xiao, X. Zhang, Angew. Chem., Int. Ed., 2001, 40, 3425–3428. 50. M.T. Reetz, A. Gosberg, R. Goddard, S.H. Kyung, Chem. Commun., 1998, 2077–2078. 51. (a) J. Kang, J.H. Lee, A.H. Ahn, J. Choi, Tetrahedron Lett., 1998, 39, 5523–5526; (b) J. Kang, J.H. Lee, J.B. Kim, G.J. Kim, Chirality, 2000, 12, 378–382.

140

Ferrocenes: Ligands, Materials and Biomolecules

52. (a) L. Schwink, P. Knochel, Chem. Eur. J. 1998, 4, 950–968; (b) A. Boudier, L.O. Bromm, M. Lotz, P. Knochel, Angew. Chem., Int. Ed., 2000, 39, 4414–4435; (c) M. Lotz, T. Ireland, J.J.A. Perea, P. Knochel, Tetrahedron: Asymmetry, 1999, 10, 1839–1842; (d) J.J. Almena Perea, A. B¨orner, P. Knochel, Tetrahedron Lett., 1998, 39, 8073–8076; (e) J.J.A. Perea, M. Lotz, P. Knochel, Tetrahedron: Asymmetry, 1999, 10, 375–384. 53. D. Laurenti, M. Santelli. Organic Prep. Proced. Int., 1999, 31, 245–294. 54. J. Park, K.H. Lee, C.W. Cho, Tetrahedron Lett., 1995, 36, 7263–7266. 55. S. Basra, J.G. de Vries, D.J. Hyett et al. Dalton Trans., 2004, 1901–1905. 56. J-C. Hierso, F. Lacassin, R. Broussier, R. Amardeil, P. Meunier, J. Organomet. Chem., 2004, 689, 766–769. 57. T.J. Colacot, Platinum Met. Rev., 2002, 46, 82–83. 58. (a) F. Maienza, M. Santoro, F. Splinder et al. Tetrahedron: Asymmetry, 2002, 13, 1817–1824. (b) F. Maienza, M. W¨orle, P. Steffanut et al. Organometallics, 1999, 16, 1041–1049. 59. U. Nettekoven, P.C.J. Kamer, P.W.N.M. van Leeuwen et al. J. Org. Chem., 1999, 64, 3996–4004. 60. S. Jug´e, M. Stephan, J.A. Lafﬁtte, J.P. Genˆet, Tetrahedron Lett., 1990, 31, 6357–6360. 61. H. Tsuruta, T. Imamoto, Tetrahedron: Asymmetry, 1999, 10, 877–882. 62. U. Nettekoven, P.C.J. Kamer, M. Widhalm, P.W.N.M. van Leeuwen, Organometallics, 2000, 19, 4596–4607. 63. J. You, H-J. Drexler, S. Zhang et al. Angew. Chem., Int. Ed., 2003, 42, 913–915. 64. A. Marinetti, F. Labrue, B. Pons et al. Eur. J. Inorg. Chem., 2003, 2583–2590. 65. Y. Hsiao, N.R. Rivera, T. Rosner et al. J. Am. Chem. Soc., 2004, 126, 9918–9919. 66. D. Heller, H.J. Drexler, J. You et al. Chem. Eur. J., 2002, 8, 5196–5202. 67. Y. Chi, Y.G. Zhou, X. Zhang, J. Org. Chem., 2003, 68, 4120–4122. 68. C. Korff, G. Helmchen, Chem. Commun., 2004, 530–531. 69. T.J. Brunker, N.F. Blank, J.R. Moncarz et al. Organometallics, 2005, 24, 2730–2746. 70. F. Ungv´ary, Coord. Chem. Rev., 2002, 228, 61–82 and references therein. 71. A.T. Axtell, J. Klosin, K.A. Abboud, Organometallics, 2006, 25, 5003–5009. 72. J. Tsuji, H. Takashi, M. Morikawa, Tetrahedron Lett., 1965, 6, 4387–4388. 73. W. Zhang, T. Shimanuki, T. Kida et al. J. Org. Chem., 1999, 64, 6247–6251. 74. H. Jendralla, E. Paulus, Synlett 1997, 471–472.

4 Other Symmetric 1,1-Bidentate Ferrocene Ligands∗ Ulrich Siemeling

4.1

Introduction

Dppf is one of the most useful and popular chelate ligands in coordination chemistry (see Chapter 2). It was ﬁrst reported in 1965, and chiral variants were described in 1974. The overwhelming success of dppf and its many offspring has overshadowed, and arguably even delayed, the development and use of ferrocene-based chelate ligands with two ligating heteroatoms other than phosphorus. The closest analogues are bidentate arsane ligands, which, although used as early as 1971,1 have not found widespread application. In contrast, 1,1 -bidentate ferrocene ligands with nitrogen, oxygen, sulfur, selenium or tellurium donor atoms have received considerably more attention, and dynamic progress concerning such homo-donor ligands can currently be noted. This chapter focuses on symmetric 1,1 -bidentate ligands of this type. It provides an update for and extends the scope of a tutorial review published in 2005.2 The literature is covered up to autumn 2006. Only those ferrocene derivatives that have actually been used as chelate ligands are included. These ligands are collected in a Chart for reference purposes. The structure of this chapter is simple and straightforward. The material is organised in sections according to the nature of the chelating heteroatoms. These sections are divided into subsections, each dealing with a distinct individual ligand class, the ordering principle being that electroneutral ligand classes are treated ﬁrst in each section. ∗ Speciﬁc abbreviations used throughout this chapter are given at the end of the chapter before the Reference List.

Ferrocenes: Ligands, Materials and Biomolecules  2008 John Wiley & Sons, Ltd

Edited by Petr Stepnicka

Fe

Fe

Ph

Ph

Ph

Ph

5

1

Fe

O

O

N

N

10

PR3

PR3

N

N

O

O

a b

O

R Cy Ph

Py

Py Py

Py

Fe

Fe

Fe

6

2

O

O

N

N

O

O

11

R

a b c d e f g

CHPh

CHPh

R

N

N

3

O

O

R Ph CH2Ph SiMe3 Mes TIPP CHMePh CHMeTol

Fe

Fe

N

N O O

O

O

7

12

Fe

R

2

1

2

R 1 R

R

N

O

O

1

8

Fe

Fe

R H t-Bu 9-anthracenyl t-Bu

O

O

N

a b c d

O

O

2

R H H H t-Bu

13

O

O

4

PR2

PR2

Nn-Pr

Fe

R a Cy b Ph

Fe

N

N

9

O O

O

O

14

Fe

O O

O

O

142 Ferrocenes: Ligands, Materials and Biomolecules

Fe

26

Fe

Fe

15

S

R

R

S

S

S

S

X

S

S

S

S

S

S

S

31

S

Et2N

20

Fe

S

Et2N

Fe

X a S b Se

Fe

27

S

R a Me b t-Bu

16

Fe

S

S

Fe

S

R2

R1

a b c d e f

O

O

Fe

21

S

S

32

S

R

R1 Ph Mes Me Me Me t-Bu

Fe

g h i j k l

R a Me b Mes c t-Bu

X a S b Se

R2 Me Et i-Pr n-Bu i-Bu CH2Ph

S

O

X

R1 Me Et i-Pr n-Bu i-Bu CH2Ph

S

S

Fe

28

R2 Ph Mes t-Bu Ph Mes Ph

Fe

S

S

22

S

S N

33

Y

Y

S

S

Y a S b Se c Te

Chart

Fe

17

Fe

Fe

Fe S

S

R

R

Se

Se

S

S

Fe

Fe

24

PPh2

PPh2

R Me Ph CH2Ph (CH2)9CH=CH2

18

a b c d

S

S

Fe

t- Bu 29

34

t-Bu

Fe

S

S

23

S

S

(CH2)8

Fe

S

S

S

S

t-Bu

S

S

Fe

30

N

19

S

S

t- Bu

S

S

Fe

CH2

CH2

25

Fe

S

S

S

S

Other Symmetric 1,1 -Bidentate Ferrocene Ligands 143

144

Ferrocenes: Ligands, Materials and Biomolecules

4.2

[N,N] Ligands

4.2.1

The Diamino Framework fc(NR2 )2

Transition metal chelates of this ligand family have been investigated by Plenio and coworkers, who have used compound 1 in their studies.3, 4 This species contains six potentially ligating nitrogen atoms. A detailed electrochemical study of the behaviour of 1 towards zinc triﬂate in acetonitrile revealed that this system is rather complicated (Scheme 4.1). Py

Ph Ph Ph

Py Py

Fe

2 Zn

Ph Ph

+

+

PY [1Zn2]4+ ∆E = 0.79 V

2 +

Py

Ph

Ph

Zn 2

N

1

Ph

Py Py

Fe

Ph

Py

Zn

Zn2

N

2 +

N

Ph

Py

Ph

N

N Py Py

Fe N

Ph

Py

Py Ph

+ Zn 2

+

N

Ph Ph

Fe

Ph

A ∆E = 0.41 V

Py Zn 2 +

N

Py Py

B ∆E = 0.85 V

Py = 2-pyridyl

Scheme 4.1

1 can accommodate two zinc centres, each being coordinated by two pyridyl nitrogen atoms and the corresponding cyclopentadienyl-connected nitrogen atom. This leads to an anodic shift of the redox potential of 1 of 0.79 V. The situation is less obvious for a 1:1 stoichiometry of 1 and zinc triﬂate. Two different isomers of [1Zn]2+ , A and B, are formed, which differ dramatically in their redox potentials. An anodic shift of the redox potential of 1 of 0.41 V is approximately half as large as the effect observed for the 2:1 complex and is therefore compatible with isomer A. The second isomer seems to be chelate B, where the zinc centre is coordinated by both cyclopentadienylconnected nitrogen atoms. The short distance between the redox-active unit and the coordinated metal centre causes a particularly large anodic shift of the redox potential of 1. Essentially the same behaviour was observed towards Co2+ . 4.2.2

The Diimino Framework fc(N=CR1 R2 )2

Hor and coworkers have prepared the palladium chelate [2PdCl2 ] by reaction of [PdCl2 (MeCN)2 ] with 2 (Scheme 4.2).5 This air-stable compound effectively catalyses

Other Symmetric 1,1 -Bidentate Ferrocene Ligands +

Ph N [Pd]

Fe

145

[BAF]

Ph [Pd] = PdClMe Na[BAF] CH2Cl2, MeCN

N

N

NCMe Pd

Fe N

Ph

Me Ph

[Pd] = PdCl2 PdClMe

[BAF] = [B{3,5-(CF3)2C6H3}4]

Scheme 4.2

Suzuki cross-coupling reactions of aryl iodides and bromides with aryl boronic acids in aqueous media under non-homogeneous conditions in which the products can be easily isolated and the catalyst retrieved. Reusability of the catalyst proved to be good, with yields decreasing only marginally in ﬁve consecutive runs. With a view to potential application in oleﬁn polymerisation, Gibson and Long have used the closely related methyl derivative [2PdClMe] for the formation of [2PdMe(MeCN)][B{3,5(CF3 )2 C6 H3 }4 ] (Scheme 4.2), which, however, failed to polymerise ethylene.6 Several salen type analogues of 2 have been investigated by Arnold as well as by Gibson and Long (Scheme 4.3). Metathesis of 3aH2 –3cH2 with Ti(Oi-Pr)4 afforded hexacoordinate titanium chelates with cis oriented Oi-Pr groups,6 whereas hexacoordinate zirconium chelates with trans oriented benzyl groups and a square-planar conformation of the chelate ligand was obtained from the reaction of 3bH2 and 3dH2 , respectively, with Zr(CH2 Ph)4 .7 The closely related [4Zr(CH2 Ph)2 ] was obtained from 4H2 and Zr(CH2 Ph)4 (Scheme 4.4).7 Reaction of 3bH2 with [TiCl4 (THF)2 ] afforded the trans-dichloro complex [3b TiCl2 ].6 No base was needed to mop up the hydrogen chloride produced in this reaction. The analogous species [3dMCl2 ] (M = titanium, zirconium) were obtained from the respective metal tetrachloride and the magnesium complex [3dMg(THF)2 ] in THF.7 The oxo-bridged dimers [3aTi(µ-O)]2 and [{3aTiCl}2 (µ-O)] were obtained by serendipity.6 Their formation seems to be due to the presence of adventitious moisture during the attempted recrystallisation of the corresponding dichloro derivative. 4.2.3

The Diphosphoraneimino Framework fc(N=PR3 )2

Metallinos and coworkers have reported the formation of [5aPdCl2 ] from 5a and [PdCl2 (MeCN)2 ] (Scheme 4.5).8 With 5b, however, the ionic product [5bPdCl]Cl was ˚ obtained; its cation exhibits an exceptionally short Fe−Pd bond length of 2.6716(7) A. This is a unique result, since the question of whether the iron atom may act as an additional electron donor to chelated metal centres had been addressed before exclusively with [P,P], [O,O] and [S,S] ligands (vide infra). The fact that Fe−Pd bond formation occurs with 5b, but not with 5a, is tentatively ascribed to electronic differences between the PCy3 and PPh3 substituents, which render 5b less electrondonating for palladium(II) than 5a.

146

Ferrocenes: Ligands, Materials and Biomolecules

R2

R2

N

N R'

Fe

O

R1 Zr(CH2Ph)4 R1 toluene

HO HO

Fe

Zr

N

N R' [3Zr(CH2Ph2)] R1 R2 b t-Bu H

3H2 R2 b: [TiCl4(THF)2] toluene d: 1) n-Bu2Mg THF 2) MCl4

d Ti(Oi-Pr)4 toluene

R2

M

O

N

N

[3MCl2]

b

O R1 Ti (Oi-Pr)2 R1 O

Fe

N Cl

d d

t-Bu t-Bu R2

R1 R1

O

[3Ti(Oi-Pr)2]

R2 M

R1

Ti Ti

t-Bu

H

t-Bu

Zr

t-Bu

t-Bu t-Bu

R2 a

R2

R1

R2

H

H

H b t-Bu c 9-anthracenyl H

Scheme 4.3

N HO HO

Fe

R2

R' = CH2Ph

N Cl Fe

O

R1 R1

N R'

Zr(CH2Ph)4 toluene

N

O Fe

Zr

O

N R'

4H2

[4Zr(CH2Ph)2] R' = CH2Ph

Scheme 4.4

Other Symmetric 1,1 -Bidentate Ferrocene Ligands

147

PCy3 N a

PR3

Pd

Fe N

N [PdCl2(MeCN)2]

Fe PR3

5

Cl PCy3

[5aPdCl2]

toluene

N

Cl

PPh3 N

b Fe

R

+ Cl

Pd

Cl

N

a Cy b Ph

[5bPdCl]Cl

PPh3

Scheme 4.5

4.2.4

The Diamido Framework fc(NR)2 2−

Diamido chelate ligands have been exploited extensively in the search for new αoleﬁn polymerisation catalysts,9, 10 and this has provided an important stimulus for the great current interest in such species. The ﬁrst examples of transition metal chelates containing the ferrocene-based diamido ligand framework fc(NR)2 2− (6) were published in 2001 by three different research teams (Arnold,11 R = SiMe3 ; Bildstein and Siemeling,12 R = Ph; • Gibson and Long,13 R = CH2 Ph). The repertoire of the parent ferrocenes 6H2 has been extended to compounds containing bulky aryl groups14, 15 instead of simple phenyl, and also includes the C2 -symmetric species 6fH2 and 6gH2 .16 To date, mainly titanium and zirconium chelates have been reported with this ligand platform, using primarily metathesis reactions of 6H2 with M(CH2 Ph)4 and M(NMe2 )4 (M = titanium, zirconium), respectively. The only examples from main group element chemistry are the magnesium complex [6cMg(THF)2 ] published by the group of Arnold11 and the aluminium complexes [6cAlR(py)] [R = t-Bu, CH(SiMe3 )2 , Si(SiMe3 )3 ] reported by Wrackmeyer and coworkers.17 Finally, the vanadium imido complex [6cVCl(NTol)]18 and the uranium tetraamide [(6c)2 U]19 were described very recently by Westmoreland and Arnold. The reaction of 6aH2 with M(NMe2 )4 (M = titanium, zirconium) afforded [6aTi (NMe2 )2 ] and [6aZr(NMe2 )2 (HNMe2 )], respectively (Scheme 4.6).12 The zirconium atom is pentacoordinate, containing HNMe2 as an additional ligand, which is liberated in the metathesis reaction as the second product. The phenyl rings in this compound are in conjugation with the lone pairs of the chelating nitrogen atoms, competing with the zirconium atom for lone pair electron density. With the bulkier 6eH2 , the tetracoordinate [6eZr(NMe2 )2 ] was obtained.20 The absence of HNMe2 in this compound appears to be due to a higher π-loading (and hence lower Lewis acidity) of the chelating nitrogen atoms. Their lone pairs are not in conjugation with the bulky aryl rings, since these are oriented perpendicular to

148

Ferrocenes: Ligands, Materials and Biomolecules Ph N [Zr]

Fe

a R = Ph

HNMe2

N NHR

Ph [6aZr(NMe2)2(HNMe2)]

Zr(NMe2)4

Fe

toluene

R

NHR N

b R = CH2Ph 6H2

e R = TIPP [Zr] = Zr(NMe2)2 TIPP = 2,4,6-i-Pr3C6H2

Zr(CH2Ph)4 toluene b R = CH2Ph

[Zr]

Fe N

R [6bZr(NMe2)2] [6eZr(NMe2)2]

e R = TIPP

TIPP N

NR Fe

Zr NR 2

[(6b)2Zr]

CH2Ph Zr

Fe N

CH2Ph

TIPP [6eZr(CH2Ph)2]

Scheme 4.6

the chelate ring plane (Figure 4.1). This stereoelectronic reasoning is in line with the ﬁnding that with 6bH2 , which contains electron-donating alkyl substituents, the tetracoordinate [6bZr(NMe2 )2 ] was formed.13 Interestingly, metathesis of 6bH2 with one equivalent of Zr(CH2 Ph)4 afforded the tetraamido compound [(6b)2 Zr],13 whereas the dibenzyl complex [6eZr(CH2 Ph)2 ] was obtained in the analogous reaction of 6eH2 (Scheme 4.6). [6aZr(NMe2 )2 (HNMe2 )] could be transformed to the, equally pentacoordinate, dichloro derivative [6aZrCl2 (HNMe2 )] by reaction with [H2 NMe2 ]Cl. Furthermore, oxidation to [6aZr(NMe2 )2 (HNMe2 )][PF6 ] proceeded cleanly and swiftly with the ferrocenium salt [FeCp2 ] [PF6 ].12 The group of Arnold has addressed the question whether the ferrocene unit can act as an additional donor for a chelated metal centre.21 A crucial starting material for this investigation was the dimethyl complex [6cTiMe2 ], which was obtained from 6cH2 in a three-step reaction, involving deprotonation with n-Bu2 Mg, reaction of the resulting magnesium complex [6cMg(THF)2 ] with [TiCl4 (THF)2 ] and subsequent methylation of the dichloro product [6cTiCl2 ] with MeLi.11 The benzyl analogue [6cTi(CH2 Ph)2 ] proved to be accessible in a single step by metathesis of 6cH2 with Ti(CH2 Ph)4 .11 Treatment of [6cTiMe2 ] with B(C6 F5 )3 afforded the contact ion pair ˚ is 0.25 A ˚ shorter than that of [6cTiMe][MeB(C6 F5 )3 ], whose Fe−Ti distance of 3.07 A

Other Symmetric 1,1 -Bidentate Ferrocene Ligands

149

Figure 4.1 Molecular structures of [6aZr(NMe2 )2 (HNMe2 )] (left) and [6eZr(NMe2 )2 ] (right)

its dimethyl precursor and is compatible with a weak Fe → Ti donor interaction present in the electron-poor cationic alkyl complex (Scheme 4.7).21 With one equivalent of [Ph3 C][B(C6 F5 )4 ] as methyl-abstracting reagent [6cTiMe][B(C6 F5 )4 ] was obtained, whereas with 0.5 equivalents formation of the methyl-bridged [(6cTiMe)2 (µ-Me)][B (C6 F5 )4 ] was observed. [6cTiMe][B(C6 F5 )4 ] afforded [{6cTi(µ-Cl)}2 ][B(C6 F5 )4 ]2 in dichloromethane solution by reaction with the solvent. This chloro-bridged dimer ˚ which is equal to the sum of the exhibits an Fe−Ti distance of only 2.491(2) A, covalent radii for iron and titanium and compares well to Fe−Ti single bond lengths reported in the literature. Reaction with dichloromethane was also observed, when [6cZr(CH2 Ph)2 ] was treated with B(C6 F5 )3 in this solvent, affording the chloro-bridged dimer [{6cZrCl(µ-Cl)}2 ].22 This compound forms a monomeric solvent adduct in THF solution. [6cTiMe][B(C6 F5 )4 ] has been shown to oligomerise 1-hexene, producing shortchain oligomers of 5–6 monomer units with an activity of 102 g (oligomer)/(mmol catalyst)•h.21 The related zirconium complex [6cZr(CH2 Ph)][B(C6 F5 )4 ] was found to polymerise ethylene with an activity of 102 g of PE/(mmol catalyst)•h•bar.22 It also inserted diphenylacetylene, affording [6cZr{CPh=CPh(CH2 Ph)}][B(C6 F5 )4 ]. When [B(CH2 Ph)(C6 F5 )3 ]− was used as the counter anion, close ion pairing occurred in the solid state and also in solution, leading to a greatly reduced activity towards ethylene. The reaction was sufﬁciently slow to detect the formation of the mono-insertion

150

Ferrocenes: Ligands, Materials and Biomolecules SiMe3

SiMe3

N

N B(C6F5)3

TiMe2

Fe

TiMe

Fe

pentane

N

MeB(C6F5)3

N

SiMe3

SiMe3

3.07 Å [6cTiMe][BMe(C6F5)3]

[6cTiMe2] [Ph3C][B(C6F5)4 ] chlorobenzene

+

SiMe3

[B(C6F5)4 ]

N

N TiMe

Fe

CH2Cl2

Cl Cl

Ti

N

SiMe3

+

2[B(C6F5)4]

N Ti

Fe

N

2

SiMe3 Me3Si Fe N SiMe3 Me3Si

[6cTiMe][B(C6F5)4]

2.49 Å [{6cTi(m-Cl)}2][B(C6F5)4]2

Scheme 4.7

+

SiMe3 N

CH2Ph Zr

Fe

SiMe3 N R RC CR benzene

B(C6F5)3

N SiMe3

SiMe3 [6cZr{CR=CR(CH2Ph)}][B(CH2Ph)(C6F5)3] R = Me, Ph, Tol

[6c(CH2Ph)][B(CH2Ph)(C6F5)3 ] C 2H 4

benzene

SiMe3 N

CH2CH2CH2Ph Zr

Fe

R

Zr

Fe

N

X

B(C6F5)3

N SiMe3 [6c(CH2CH2CH2Ph)][B(CH2Ph)(C6F5)3]

Scheme 4.8

Other Symmetric 1,1 -Bidentate Ferrocene Ligands

151

product [6cZr(CH2 CH2 CH2 Ph)][B(CH2 Ph)(C6 F5 )3 ] (Scheme 4.8) as well as the diand tri-insertion products by1 H NMR spectroscopy after 20 minutes. Insertion of acetylenes RC≡CR still occurred easily, and the corresponding products [6cZr{CR=CR(CH2 Ph)}][B(CH2 Ph)(C6 F5 )3 ] (R = Me, Ph, Tol) were formed in clean and swift reactions (Scheme 4.8). The insertion reaction with ethylene was slowed down, when the SiMe3 substituents were replaced by mesityl groups. [6dZr(CH2 Ph)] [B(CH2 Ph)(C6 F5 )3 ], which was formed from [6dZr(CH2 Ph)2 ] and B(C6 F5 )3 , afforded the mono-insertion product [6dZr(CH2 CH2 CH2 Ph)][B(CH2 Ph)(C6 F5 )3 ] in the presence of an excess of ethylene on a time scale of ca. 30 minutes at room temperature.14 No oligomerisation of 1-hexene could be observed in this case. In contrast, the catalyst obtained by activation of [6dZr(CH2 Ph)2 ] with [Ph3 C][B(C6 F5 )4 ] proved to be able to polymerise up to 400 equivalents of 1-hexene, yielding poly(1-hexene) with molecular weights up to 20 000 with relatively low polydispersities (PDI = 1.3–1.4). In this context, Arnold and coworkers have also reported the synthesis of [6fZr(CH2 Ph)2 ] and [6gZr(CH2 Ph)2 ] (Scheme 4.9), which both contain a C2 -symmetric chelate ligand framework, potentially relevant to the stereospeciﬁc polymerisation of α-oleﬁns.16 Me Ar N CHAr Fe N CHAr

N H

1) 2 MeLi toluene, − 60 °C 2) H2O

Fe

H N Ar

Ar = Ph, Tol

Ar

6H2

f Ph g Tol

Me Zr(CH2Ph)4 Me Ar

[6Zr(CH2Ph)2]

N Zr(CH2Ph)2

Fe N

Ar Me

Scheme 4.9

4.3 4.3.1

[O,O] Ligands The Diether Framework fc(OR)2

Sato and coworkers have described the only example so far of a chelate based on a simple bidentate diether ligand.23 Reaction of 7 with [Pd(MeCN)4 ][BF4 ]2 in the presence of PPh3 afforded [7Pd(PPh3 )][BF4 ]2 (Scheme 4.10), which proved to be rather unstable, so that only limited spectral data (IR, UV–Vis) could be obtained. These data

152

Ferrocenes: Ligands, Materials and Biomolecules 2

+

2[BF4]

Me OMe Fe

O

[Pd(MeCN)4][BF4]2, PPh3 Fe

acetone OMe

7

Pd

PPh3

O Me [7Pd(PPh3)][BF4]2

Scheme 4.10

are in accord with a tetracoordinate palladium complex which exhibits an Fe → Pd interaction. Convincing evidence for the presence of dative iron–metal bonds has been provided for closely related thioether analogues (vide infra). Most of the work in this area has been performed with crown ether derivatives and related oligodentate species (lariat ethers, cryptands; see Chapter 8). Not surprisingly, only s-block metal chelates have been reported with these comparatively hard donors. Akabori and coworkers have obtained a series of crystalline complexes comprising [8Li]ClO4 ,24 [9M]ClO4 (M = Li, Na),24 [10M]SCN (M = Li, Na)25, 26 and [11M]SCN (M = Li, Na, K)25, 26 by reaction of the respective ligand with the corresponding metal salt in acetonitrile. [10Na]SCN was structurally characterised and turned out to be a contact ion pair in the crystal. The sodium atom interacts with the thiocyanate nitrogen atom and is therefore hexacoordinate (Figure 4.2). In their investigation of the cryptand 12, Plenio and coworkers were able to isolate the chelates [12Na]ClO4 and [12Ca(H2 O)](ClO4 )2 , which was structurally characterised.3 In addition to the four oxygen atoms and two nitrogen atoms present in 12, the calcium atom interacts with one water molecule and one perchlorate oxygen atom and is therefore octacoordinate (Figure 4.2). Cyclic voltammetry revealed that metal coordination leads to particularly large anodic shifts in the case of 12 (0.40 V for Na+ , 0.47 V for Ca2+ ) and closely related cryptands.

Figure 4.2 Molecular structures of [10Na]SCN (left) and [12Ca(H2 O)](ClO4 )2 (right, only the perchlorate ion which interacts with the calcium ion is shown)

Other Symmetric 1,1 -Bidentate Ferrocene Ligands

153

An interesting modiﬁcation of the diether motif was introduced by Mirkin and coworkers, who employed redox-switchable hemilabile phosphane–ether ligands to investigate fundamental aspects of hemilability.27 A series of rhodium(I) and palladium(II) complexes was synthesised by the respective reaction of 13 with [RhCl(cyclooctene)2 ]n in THF in the presence of Ag[BF4 ] and with [Pd(MeCN)4 ][BF4 ]2 in acetone (Scheme 4.11). Electrochemical investigations revealed a rather dramatic inﬂuence of the coordinated metal centre on the redox potential of the chelate ligand, which was anodically shifted by ca. 0.4 and 0.6 V for rhodium(I) and palladium(II), respectively. +

O

[RhCl(cyclooctene)2], Ag[BF4] THF O

PR2

O

PR2

a b

13 R Cy Ph

Rh

Fe O

Fe

PR2

[13Rh][BF4]

acetone

2

+

2[BF4]

PR2

O [Pd(MeCN)4][BF4]2

[BF4]

PR2

Pd

Fe O

PR2

[13Pd][BF4 ]2

Scheme 4.11

Oxidation of the chelate ligand caused a decrease of the complex formation constant by factors of more than 1010 for palladium(II) and up to ca. 107 for rhodium(I). The different anodic shifts of the half-wave potential reﬂect differences in inductive withdrawal of electron density from the ferrocene unit as well as different electrostatic interactions between the oxidised ferrocene backbone and the coordinated metal ˚ for all compounds investigated. A cations. The iron–metal distances were ca. 4.0 A detailed analysis revealed that electrostatic effects contribute substantially to the thermodynamic destabilisation of the chelates upon oxidation. Not surprisingly, the large changes in complex stability coincide with substantial changes in reactivity. Electrochemical oxidation of the rhodium complex [13aRh]+ in acetonitrile solution caused a large increase in afﬁnity towards acetonitrile (Scheme 4.12). The Rh−O bond weakens upon oxidation of the ferrocene unit, which leads to an increase of the complex formation constant of the acetonitrile complex by a factor of more than 107 . 4.3.2

The Dialkoxo Framework fc(O)2 2−

The dialkoxo ligand fc(O)2 2− (14) can be classiﬁed as ‘hard’ according to Pearson’s HSAB principle. In contrast to the related ‘soft’ analogues fc(Y)2 2− (33, Y = sulfur, selenium, tellurium; vide infra), the use of this ligand has been very limited. Akabori and coworkers have published the palladium chelate [14Pd(PPh3 )] (Scheme 4.13,

154

Ferrocenes: Ligands, Materials and Biomolecules +

PCy2

O

KRed + 2 MeCN

Rh

Fe O

+

MeCN Rh

Fe

− 2 MeCN

PCy2

PCy2

O

O

NCMe

PCy2

+

[13aRh] + e−

+

∆E1/2 = −435 mV

− e−

KOx / KRed = 2•107

2 O

PCy2 Rh

Fe O

PCy2

+ e−

− e−

+

2

KOx + 2 MeCN

MeCN

Fe

− 2 MeCN

+

PCy2

O

O

Rh

NCMe

PCy2

Scheme 4.12

Et2N S

O Fe

Pd O

[14Pd(PPh3)]

3 Ag

Fe

Ag

3ClO4

PPh3

S PPh3

+

PPh3

S S

Ag

PPh3

Et2N [15(AgPPh3)3](ClO4)3

Scheme 4.13

left), which was obtained from the reaction of Na2 14 with [PdCl2 (PPh3 )2 ].28 The ruthenocene analogue was structurally characterised by X-ray diffraction and was found ˚ compatible with a weak dative bond. Thereto have a Ru−Pd distance of 2.692(1) A, fore, [14Pd(PPh3 )] is also supposed to exhibit an Fe → Pd interaction. No further reports of related chelates have been published. This paucity is rather surprising in view of the great importance of transition metal dialkoxo chelates.29

4.4 4.4.1

[S,S] Ligands The Dithiocarbamate Ligand Framework fc(S2 CNR2 )2

Laguna and coworkers have reported the synthesis of [15(AgPPh3 )3 ](ClO4 )3 by reaction of 15 with three equivalents of [Ag(OClO3 )PPh3 ].30 This is the only example

Other Symmetric 1,1 -Bidentate Ferrocene Ligands

155

so far in this category. Only one of the three AgPPh3 units is actually chelated by 15 (Scheme 4.13, right) and therefore exhibits a three-coordinate silver atom, as opposed to the silver atoms of the other two AgPPh3 units, which are two-coordinate. Exclusively two-coordinate silver atoms of the latter type are present in the product of the reaction of 15 with two equivalents of [Ag(OClO3 )PPh3 ], which afforded [15(AgPPh3 )2 ](ClO4 )2 . With one equivalent of silver (I) perchlorate (AgClO4 ), a linear coordination polymer was obtained, where 15 acts as a bridging ligand.31 4.4.2

The Dithioether Framework fc(SR)2

A vast number of simple chelates of the type [16{M}] have been reported (Table 4.1), including also complexes of the unsymmetrical dithioether ligands 16i–16l investigated by Long and coworkers. A few examples of homoleptic species [(16)2 M]X [M = copper(I), silver(I)] have also been described as well as the dinuclear platinum complex [(µ2 -16g){16gPt(C6 F5 )}2 ](ClO4 )2 , where 16g acts both as a bridging and a chelating ligands towards platinium(II). Not surprisingly, essentially all of these compounds have been prepared by standard ligand (CO, RCN, oleﬁn, solvent) substitution reactions. Many of these chelates have been shown to be ﬂuxional in solution by variable temperature NMR studies. The ﬂuxionality can be ascribed to two processes, that is sulfur inversion and bridge reversal involving the chelated metal atom. Bridge reversal turned out to be rapid on the NMR time scale even at low temperatures, whereas sulfur inversion was sufﬁciently slow below ca. −30 ◦ C to allow observation of static isomers by NMR spectroscopy. Table 4.1 Chelates containing a dithioether ligand 16 Compound type

Speciﬁc members

[16M(CO)4 ]

M: a Cr, Mo, W;32 i, l W33

[16aReX(CO)3 ]

X: Cl, Br, I34

[16MX2 ]

M/X: a Pd/Cl,35, 36 Pd/Br,35, 36 Pt/Cl,35–37 Pt/Br;35–37 b Pt/Cl;38 c Pd/Cl, Pd/Br, Pt/Cl, Pt/Br;35, 36 d Pd/Cl;23 e Pd/Cl, Pd/Br, Pt/Cl, Pt/Br;35, 36 f Pd/Cl,36 Pd/Br,36 Pt/Cl,36 Pt/Br;35, 36 g Pd/Cl,36 Pd/Br,35, 36 Pt/Cl,36 Pt/Br;35, 36 h Pd/Cl, PtCl;39, 40 i, j Pd/Cl, PtCl;33 k Pd/Cl, PtCl;40 l Pd/Cl, Pt/Cl;33

[16M(PPh3 )][BF4 ]2

M: a Pd,23, 41 Pt;23, 42 c Pd;23, 41 d Pd;23 e Pd,23, 41 Pt;23 f Pd23, 41

[16aPtXMe3 ]

X: Cl, Br, I43

[(16)2 M]X

M/X: a, c Cu/PF6 , Ag/BF4 ;44 g Ag/OTf45

[16M(L)]X

M/L/X: a Cu/PPh3 /PF6 ;44 g Ag/PPh3 /OTf, Ag/phen/OTf, Au/PPh3 /OTf45

[16gMX(C6 F5 )]

M/X: Pd/C6 F5 , Pt/C6 F5 , Pt/Cl, Pt/Br46

miscellaneous

[16bRhCl3 ],38 [(µ2 -16g){16gPt(C6 F5 )}2 ](ClO4 )2 ,46 [16gAg(OTf)]45

156

Ferrocenes: Ligands, Materials and Biomolecules

Activation parameters for sulfur inversion have been determined for quite a number of chelates [16{M}] containing hexacoordinate tungsten,32 rhenium,34 platinum35 and tetracoordinate palladium36 and platinum.36, 37 Sato and coworkers have reported the structurally characterised platinum chelate [16aPt(PPh3 )][BF4 ]2 (Scheme 4.14, left),23, 42 which was obtained from [16aPtCl2 ] by chloride abstraction with two equivalents of Ag[BF4 ] in acetone in the presence of PPh3 .

2

+

+

2[BF4]

S Fe

Pt

PPh3

S

R a Me e i-Bu

PPh3

S Rh

Fe

PPh3

S

[16Pt(PPh3)][BF4]2 [18Rh]

+

Scheme 4.14

˚ compatThis compound was found to exhibit an Fe–Pt distance of 2.851(2) A, ible with a dative Fe → Pt bond. The coordination of the platinum atom can best be described as distorted square-planar. In platinum chemistry, [16ePt(PPh3 )][BF4 ]2 (Scheme 4.14, left) is the only other analogue known,23 whereas palladium congeners have been reported with 16a and 16c–g23, 41 as well as for the ferrocenophane 17, whose thioether sulfur atoms are connected by a (CH2 )8 chain. In analogy to [16aPt(PPh3 )][BF4 ]2 , all of these compounds are supposed to exhibit an iron → metal interaction, which is expected to reduce the ﬂexibility of the S−M−S unit. This is supported by the fact that these complexes are not ﬂuxional on the NMR time scale in solution at room temperature. This is in sharp contrast to related tetracoordinate species like [16MCl2 ] (M = palladium, platinum), where an iron → metal interaction can be ruled out. Sulfur inversion was found to be rapid at room temperature for such species (vide supra). In addition to the phosphane–ether ligands 13, Mirkin and coworkers have also utilised the phosphane–thioether 18 in their investigations concerning redoxswitchable hemilabile systems.27 Oxidation of the ferrocene unit of [18Rh]+ (Scheme 4.14, right) was found to destabilise the complex thermodynamically by a factor of more than 1011 . This effect is much larger than that observed for the related ether analogue [13bRh]+ , where a destabilisation by a factor of ca. 107 was observed (vide ˚ in both monocationic complexes, so that elecsupra). The Fe–Rh distance is ca. 4.0 A trostatic effects are expected to be very similar. Ligand-based oxidation weakens the comparatively strong Rh−S bond more than the weaker Rh−O bond. Long and coworkers have been able to obtain, more or less by serendipity, a chelate containing the tetradentate ligand 19, whose thioether groups each carry an additional carbanionic ligating unit. This was achieved by a cyclometalation reaction, which occurred upon prolonged heating of the platinum(II) complex [16hPtCl2 ]39, 40

Other Symmetric 1,1 -Bidentate Ferrocene Ligands

157

Mes SMes

S [MCl2(PhCN)2]

Fe SMes

toluene

S M = Pt toluene reflux

16h

MCl2

Fe

Mes [16hMCl2] M = Pd, Pt

S Cl Fe

Pt

[19PtCl2]

S Cl

Scheme 4.15

in reﬂuxing toluene solution, affording the hexacoordinate platinum(IV) species [19PtCl2 ] (Scheme 4.15).39 No such reaction was observed for the palladium analogue [16hPdCl2 ]. Sato and coworkers have reported a large number of chelates (Table 4.2) of ferrocenophane-type ligands, which contain one (20–22) and two (23–26) donor atoms in addition to the two thioether sulfur atoms attached to the cyclopentadienyl rings, the additional donor atoms being sulfur (20a, 21a, 23–26), selenium (20b, 21b) and nitrogen (22). On top of that, Ebine has described a silver complex of a ligand with three additional ether oxygen atoms (27), and Sato et al. have used ligand 28, which contains two ferrocene units. Soft metal centres [palladium(II),42, 47–50, 53, 58 platinum(II),42, 49, 54, 58 copper (I),47, 51, 52, 55, 58 silver(I),47, 55–58 mercury(II)47 ] have been used almost exclusively with these ligands, the only exception being the copper(II) chelates [25Cu][BF4 ]2 and [26Cu][BF4 ]2 ,51, 52 which were obtained from the reaction of one equivalent of Cu[BF4 ]2 with the tetradentate thiacrown-type ligands 25 and 26, respectively, in nitromethane solvent. The analogous reaction with the tridentate 21a in ethanol afforded [(21a)2 Cu][BF4 ]2 ,51, 52 whose M¨ossbauer spectrum turned out to be in accord with a copper(I) complex containing a tridentate 21a and a monodentate 21a+ ligand (Scheme 4.16). The only structurally characterised example of a chelate containing a monovalent metal centre is [23Ag]ClO4 ,55, 56 which, like all related MI chelates described by Sato and Ebine (Table 4.2), was obtained by reacting the chelate ligand with the corresponding metal salt in acetonitrile solvent. The [23Ag]+ units are aggregated as dimers in the crystal through one bridging sulfur atom per ligand, the coordination of the silver atoms being distorted square-pyramidal (Figure 4.3).56

158

Ferrocenes: Ligands, Materials and Biomolecules Table 4.2 Chelates containing a cyclic dithioether ligand with additional donor atom(s) (L = 20–28) Compound type

Speciﬁc members

[LMCl2 ]

L/M: 20a/Pd, 20a/Hg, 21a/Hg47

[LPd(MeCN)][BF4 ]2

L: 20a,42, 48, 49 20b,49 21a,42, 48, 49 21b,49 2250

[(L)2 M][BF4 ]2

L/M: 20a/Pd,48, 49 20b/Pd,49 21a/Cu51, 52

[LM]Xn

M/Xn : 20a Cu/ClO4 ;47 21a Pt/[BF4 ]2 ,42, 48, 49 Ag/BF4 ;47 21b Pt/[BF4 ]2 ;49 23 Pd/[BF4 ]2 ,48, 53 Pt/[BF4 ]2 ,54 Cu/BF4 ,55 Ag/ClO4 ,55, 56 Ag/NO3 ,55 Ag/OTf;55 24 Pd/[BF4 ]2 ,48, 53 Pt/[BF4 ]2 ,54 Cu/BF4 ,55 Ag/ClO4 ;55 25 Pd/[BF4 ]2 ,48, 53 Pt/[BF4 ]2 ,54 Cu/ClO4 ,55 Cu/[BF4 ]2 ;51, 52 26 Pd/[BF4 ]2 ,48, 53 Pt/[BF4 ]2 ,54 Cu/ClO4 ,55 Cu/[BF4 ]2 ;51, 52 27 Ag/NO3 ;57 28 Pd/[BF4 ]2 , Pt/[BF4 ]2 , Cu/ClO4 , Ag/BF4 58

miscellaneous

[20aCu(MeCN)][BF4 ]47

2 S Cu[BF4]2 S

Fe S 21a

ethanol

S Fe II

S

+

2[BF4]

S Cu I

Fe III

S

S

S

[(21a)2Cu][BF4]2

Scheme 4.16

Figure 4.3 View of a dimeric aggregate of the cationic unit of [23Ag]ClO4

Other Symmetric 1,1 -Bidentate Ferrocene Ligands

159

Among the chelates with divalent metal centres, the palladium complexes [20aPd (MeCN)][BF4 ]2 ,42 [21aPd(MeCN)][BF4 ]2 42 and [22Pd(MeCN)][BF4]2 50 have been structurally characterised. They exhibit notable differences, which are due to their individual chelate ligands. 21a is a [9]ferrocenophane, whereas 20a and 22 are [7]ferrocenophanes; the donor set of the former two comprises three sulfur atoms, whereas two sulfur atoms and a nitrogen atom are present in the latter. The pentacoordinate palladium atom of [20aPd(MeCN)][BF4 ]2 is in a distorted square pyramidal ligand ˚ environment. It exhibits a strongly coordinated acetonitrile ligand [Pd–N 2.181(3) A] ˚ In the less strained and a fairly weak Fe → Pd interaction [Fe–Pd 3.0962(8) A]. ˚ [21aPd(MeCN)][BF4 ]2 , the Fe → Pd interaction is much stronger [Fe–Pd 2.827(3) A], ˚ The palladium while the acetonitrile is hardly bonded to the metal [Pd–N 2.84(1) A]. atom in this complex therefore is essentially tetracoordinate and in a distorted squareplanar ligand environment. Similar structures can be assumed for the selenium analogues [20bPd(MeCN)][BF4]2 and [21bPd(MeCN)][BF4]2 .49 [22Pd(MeCN)][BF4 ]2 , on the other hand, contains a pyridine-type nitrogen donor atom in addition to the two thioether sulfur atoms in the ansa chain of the chelate ligand. The acetonitrile nitrogen atom and the pyridine-type nitrogen atom are both strongly bonded to the ˚ respectively], while the Fe → Pd interaction palladium [Pd–N 2.00(1) and 1.91(1) A, ˚ Interestingly, the dicationic [22Pd(MeCN)] units is very weak [Fe–Pd 3.228(2) A]. ˚ which aggregate as dimers in the crystal, exhibiting a Pd–Pd distance of 3.278(2) A, renders each palladium atom effectively hexacoordinate (Figure 4.4). In contrast to their palladium congeners, no acetonitrile solvent is present in the platinum chelates [21Pt][BF4 ]2 ,49 whose platinum atoms therefore are most likely tetracoordinate with an Fe → Pt bond. All the complexes just described were obtained straightforwardly from [M(MeCN)4 ][BF4 ]2 (M = palladium, platinum) and the respective ligand in acetone (M = palladium) or acetonitrile (M = platinum) solution with subsequent recrystallisation from solvent mixtures containing acetonitrile. With the [7]ferrocenophane-type ligands (20a and 20b), the homoleptic [(20)2 Pd][BF4 ]2 were obtained in low yield as side products in these reactions.

Figure 4.4 View of a dimeric aggregate of the cationic unit of [22Pd(MeCN)][BF4 ]2

160

Ferrocenes: Ligands, Materials and Biomolecules

As an extension of the ferrocenophane-type ligands just described, Sato and coworkers have also reported the tetrathia[5.5]ferrocenophane 28 and its chelates with palladium(II), platinum(II), copper(I) and silver(I).58 [28Cu]ClO4 and [28Ag][BF4 ] (Scheme 4.17) were characterised crystallographically. Both exhibit tetracoordinate metal centres in a distorted tetrahedral ligand environment, the distortion being much more pronounced for silver than for copper. +

S

S M

Fe S

Fe S

X M X Cu ClO4 Ag [BF4]

[28M]X

Scheme 4.17

Copper(I) and silver(I) coordination was found to lead to an anodic shift of the redox potential of 28 of ca. 0.2 and 0.35 V, respectively. No wave splitting was detected for the two-electron redox process in the cyclic voltammogram, indicating non-interacting ferrocene units. This is compatible with ﬁndings reported by Long and coworkers for [(16c)2 Cu][PF6 ] and [(16c)2 Ag][BF4 ],44 which also contain two ferrocene units per MI . A metal-induced anodic shift of ca. 0.42 V was observed for these species together with a single-stepped two-electron redox process associated with non-interacting ferrocene units. In contrast to this, the palladium chelate [28Pd][BF4 ]2 was found to exhibit two oxidation waves separated by ca. 0.13 V, which indicates a weak interaction between the ferrocene units. In their work with unsymmetrical dithioether ligands, Long and coworkers have also used the tetradentate ligands 29 and 30, which, just like 28, contain two ferrocene moieties.59 Reaction of 29 with one equivalent of [PdCl2 (cod)] and [Cu(MeCN)4 ][PF6 ], respectively, in dichloromethane afforded the corresponding dinuclear chelates [(µ29)(PdCl2 )2 ] and [(µ-29){Cu(µ-PO2 F2 )}2 ] (Figure 4.5, left), which were structurally characterised by X-ray diffraction. The unusual anion bridging the two copper atoms in the latter complex was inadvertently formed in situ by hydrolysis of hexaﬂuorophosphate. Corresponding analogues with 30 have also been obtained and are assumed to exhibit very similar structures. Obviously, these two ligands are not suitable for tetracoordination of a single metal centre, which can be ascribed to the unfavourable distance between the two sulfur atoms connected by a ﬁve atom bridge. 4.4.3

The Mixed Thioether/Disulﬁde Framework [{Fe[C5 H4 (SR)]}2 (µ-C5 H4 SSC5 H4 )]

The previous subsection ended with a description of copper and palladium complexes obtained by Long and coworkers with the unsymmetrical dithioether ligands 29 and 30.

Other Symmetric 1,1 -Bidentate Ferrocene Ligands

161

Figure 4.5 Molecular structures of [(µ-29){Cu(µ-PO2 F2 )}2 ] (left) and [(µ-31b){Cu(µPO2 F2 )}2 ] (right)

In the same vein, this group has also prepared four analogous complexes with the mixed thioether/disulﬁde ligands 31a and 31b.59 The crystal structure of [(µ-31b){Cu(µPO2 F2 )}2 ] was determined by X-ray diffraction (Figure 4.5, right) and revealed close similarities in copper coordination with [(µ-29){Cu(µ-PO2 F2 )}2 ] (vide supra). Just like 29 and 30, 31a and 31b appear to be unsuitable for tetracoordination of a single metal centre, since the distance between two of the four sulfur atoms is either too long (29, 30) or too short (31). 4.4.4

The Mixed Thioether/Thiolate Framework fc(SR)(S)−

First examples of this still rather small class of compounds were described in 2000.60, 61 Hidai and coworkers obtained [(µ-32a){Ru(p-cymene)}2](OTf) from the reaction of the dinuclear [(µ-33a){Ru(p-cymene)}2 ] with one equivalent of MeOTf in toluene (vide infra).60 Related compounds were described by Long and coworkers,33, 40, 61 who have devised an elegant synthetic route to unsymmetrical 1,1 -disubstitued ferrocenes containing a thiol as well as a thioether substitutent, starting from 1,2,3-trithia[3]ferrocenophane. In late transition metal chemistry, the sulfur-bridged dimers [{µ32b(MCl)}2 ] (M = palladium, platinum) were prepared from trans-[MCl2 (PhCN)2 ] and Li 32b (Scheme 4.18, left).40, 61 The t-Bu analogue [{µ-32c(PdCl)}2 ] was formed, more or less by serendipity, in an unusual insertion into the S−S bond of 31b upon reaction with trans-[PdCl2 (PhCN)2 ] (Scheme 4.19).33

162

Ferrocenes: Ligands, Materials and Biomolecules Mes

Mes

Cl

S

S

M

Fe S

S

S

M Cl

MCl2(THF)2

Fe

Fe S

[32bMCl2(THF)2] M = V, Cr

Mes [{m-32b(MCl)}2] M = Pd, Pt

Scheme 4.18 t-Bu S

S

Fe

Pd

Fe

Fe

Cl

S

[PdCl2(PhCN)2]

S t-BuS

31b

St-Bu

S Pd

Cl

Fe S

t-Bu [{(m-32c)(PdCl)}2]

Scheme 4.19

In Group 5 and Group 6 metal chemistry, chelates of the type [32bMCl2 (THF)2 ] (M = vanadium, chromium) (Scheme 4.18, right) were obtained from the metathesis reaction of [MCl3 (THF)3 ] with Li32b.40 These compounds were used as precatalysts in the polymerisation of ethylene. By far the best result was obtained with the vanadium complex, which, after activation with dichloromethylaluminium (100 equivalents), exhibited an activity of 28 g/(mmol h bar). 4.4.5

The Dithiolato Framework fc(S)2 2−

The focus has clearly been on late transition metal chemistry with the dithiolato ligand 33a, which is in line with expectations based on Pearson’s HSAB principle. Comparatively few early transition metal chelates have been published to date. Steudel et al. synthesised [33aTiCp2 ] from 33aH2 and Cp2 TiCl2 in the presence of NEt3 as a base (Scheme 4.20).62 Interestingly, the use of sodium amide (NaNH2 ) instead of NEt3 furnished [33aTiClCp].63 The zirconium analogue was obtained similarly with Cp2 ZrCl2 . Gibson et al. reported the chelates [33aM(NMe2 )2 ], which were prepared from 33aH2 and M(NMe2 )4 (Scheme 4.20).64 In the presence of AlMe3 and MAO, both amido complexes polymerised ethylene, albeit with low activities [M = titanium: 3 g/(mmol h bar), M = zirconium: 15 g/(mmol h bar)].64

Other Symmetric 1,1 -Bidentate Ferrocene Ligands

163

S [Cp2TiCl2] NEt3

TiCp2

Fe S

SH

[33aTiCp2]

Fe SH

M(NMe2)4 S

33aH2

M(NMe2)2

Fe S

[33aM(NMe2)2] M = Ti, Zr

Scheme 4.20

In Group 5 metal chemistry, [33aV(O)Cp ] (Cp = Cp, Cp*) are the only examples known to date. They were prepared by Herberhold and coworkers from Li2 33a and [V(O)Cp Cl2 ].65 The selenium analogues containing 33b instead of 33a as chelate ligand were obtained similarly. The groups of Herberhold and Jin have published several series of chelates containing fc(Y)2 2− (33a: Y = sulfur; 33b: Y = selenium, 33c: Y = tellurium). For the sake of brevity, the selenium and tellurium containing analogues will be mentioned in this section. Jin et al. have described a range of Group 6 nitrosyl complexes [33aMCp (NO)] (Cp = Cp: M = molybdenum, tungsten; Cp = Cp*: M = chromium, molybdenum).66 The selenium analogue [33bMoCp*(NO)] was also prepared. While [33aCrCp*(NO)] was synthesised from 1,2,3-trithia[3]ferrocenophane and the carbonyl complex [CrCp* (NO)(CO)2 ], the molybdenum and tungsten complexes were formed more straightforwardly by reacting Li2 33a with [MCp I2 (NO)]. An essentially identical synthetic approach was used for the homoscorpionate analogue [33aMoTp*(NO)], which was obtained from 33aH2 and [MoTp*I2 (NO)] by Hamor and coworkers.67 Oxidation of [33aMoCp*(NO)] proved possible with Ag[BF4 ], affording [33aMoCp*(NO)][BF4 ] in high yield.66 In addition to these 16 VE complexes, the authors have also reported the synthesis of the 18 VE species [33aWCp2 ] from Li2 33a and [WCp2 Cl2 ].66 In Group 7 metal chemistry, the ﬁrst compound to be described was [PPh4 ][(33a)2 ReO], which was obtained by Dilworth and Ibrahim from the reaction of 33aH2 with [Re(O)Cl3 (PPh3 )] in the presence of diethyl amine as a base and subsequent precipitation with [PPh4 ]Br.68 The complex showed three irreversible oxidation processes in the cyclic voltammogram, two of which have been attributed to the oxidation of the two ferrocene units. In view of results obtained by Long and Zanello in rhodium chemistry (vide infra), sulfur-centred oxidation processes have also to be taken into account here. The last three members in this category are [33aRe(Nt-Bu)Cp*], [33aRe(O)Cp*] and [33aRe(O)Tp], which were prepared by Herberhold and coworkers by reacting Li2 33a with [Re(Nt-Bu)Cp*Cl2 ], [Re(O)Cp*Cl2 ] and [Re(O)TpCl2 ], respectively.69 The diselenolato analogues, which contain 33b instead of 33a, were also obtained. The homoscorpionate complex [33aRe(O)Tp] was structurally characterised by X-ray diffraction (Figure 4.6).

164

Ferrocenes: Ligands, Materials and Biomolecules

Figure 4.6 Molecular structure of [33aRe(O)Tp]

In view of the great importance of iron–sulfur redox systems it is quite amazing that in Group 8 metal chemistry only two iron chelates containing 33a have been reported to date. The ﬁrst is [(µ-33a){Fe(CO)3 }2 ], which was prepared as early as 1983 by Seyferth and Hames by thermal reaction of 1,2,3-trithia[3]ferrocenophane with [Fe3 (CO)12 ] (Scheme 4.21).70

M = Fe, Ru

S

M(CO)3

S

M(CO)3

Fe

S Fe

S

[(m-33a){M(CO)3}2]

[M3(CO)12]

S

S

X

Fe

M = Os

Os(CO)3

S

Os(CO)3

+ higher nuclearity clusters

X = S: [(m-33a)(m-S){Os(CO)3}2] X = Os(CO)4: [(m2-33a){Os(CO)3}2{Os(CO)4}]

Scheme 4.21

The ruthenium analogue was synthesised similarly from [Ru3 (CO)12 ] by Cullen and coworkers, while with the less reactive [Os3 (CO)12 ] the sulfur-bridged analogue [(µ33a)(µ-S){Os(CO)3 }2 ] was obtained, together with the higher nuclearity clusters [(µ2 33a){Os(CO)3 }2 {Os(CO)4 }], [(µ2 -33a){Os(CO)3 }{Os3 (µ3 -S)(CO)8 }] and [(µ2 -33a) {Os(CO)3 }{Os3 (µ3 -S)2 (CO)6 }] (Scheme 4.21).71 As the second example from iron chemistry, the redox-active iron–sulfur cluster [{(µ-33a)Fe(µ-S)}2][Nn-Pr4 ]2 was prepared by Lorkovi´c et al. from 1,2,3-trithia[3]ferrocenophane and iron(III) chloride in

Other Symmetric 1,1 -Bidentate Ferrocene Ligands 2 S

S Fe

Fe S

S

Fe

S

2[Nn-Pr4]

165

+

Fe

S

[{(33a)Fe(m-S)}2][Nn-Pr4]2

Scheme 4.22

the presence of NaSt-Bu as reducing agent (Scheme 4.22).72 The sulfur-coordinated iron(III) centres proved to be antiferromagnetically coupled. The cluster core of this compound is akin to structural models of [2Fe–2S] redox proteins, the presence of two additional iron(II) centres being a new aspect. A reversible one-electron reduction of the FeIII 2 S2 cluster core was easily achieved. However, in contrast to related Fe2 S2 type clusters, full reduction to a FeII2 S2 system was not possible down to −2.5 V (vs. ferrocenium/ferrocene), probably owing to the presence of the electron-rich ferrocene moieties. In ruthenium chemistry, Herberhold and coworkers contributed the two isoelectronic compounds [33aRu(CO)(η6 -C6 Me6 )] and [33aRuCp*(NO)],73 which were obtained straightforwardly from the reaction of Li2 33a with [RuCl2 (CO)(η6 -C6 Me6 )] and [RuCl2 Cp*(NO)], respectively. The corresponding analogues containing 33b were prepared, too, as well as the closely related osmium complexes [33OsCp*(NO)] (Scheme 4.23). Y

NO M

Fe Y

Cp∗

a b c

Y S Se Te

M Ru, Os Ru, Os Os

[33MCp∗(NO)]

Scheme 4.23

In addition to Cullen’s [(µ-33a){Ru(CO)3 }2 ] (vide supra), higher nuclearity ruthenium chelates were also described by Hidai and coworkers (Scheme 4.24),60 who synthesised the ruthenium(II) complex [(µ-33a)(µ-Cl){Ru(p-cymene)}2][PF6 ] from 33aH2 and [(p-cymene)RuCl2 ]2 in the presence of [NH4 ][PF6 ]. Reduction of this compound with sodium amalgam afforded the ruthenium(I) chelate [(µ-33a){Ru(p-cymene)}2 ]. Protonation with triﬂic acid occured at the Ru−Ru bond, affording the hydrido-brigded ruthenium(II) species [(µ-33a)(µ-H){Ru(p-cymene)}2] (OTf). In contrast, alkylation with methyl triﬂate occured at one of the sulfur atoms, yielding [(µ-32a){Ru(p-cymene)}2](OTf), which contains a chelate ligand with a mixed IV thioether/thiolate donor set (vide supra). On top of that, the 50 VE RuIII clus2 Ru ter [(µ3 -33a)(µ3 -S)(µ2 -Cl)(RuCp*)3 ][FeCl4 ] (Figure 4.7) was obtained from 1,2,3trithia[3]ferrocenophane and [RuClCp*]4 (1.7 equivalents) in THF.60

166

Ferrocenes: Ligands, Materials and Biomolecules p-cymene SH

Fe

[(p-cymene)RuCl2]2 [NH4][PF6]

S

+

p-cymene

[PF6]

Ru

SH

S

p-cymene

+

OTf

H S

MeOTf

p-cymene

Me

Ru

Fe

Ru

p-cymene [(m-33a){Ru(p-cymene)}2]

HOTf

S

S

Fe

Ru

p-cymene [(m-33a)(m-Cl){Ru(p-cymene)}2][PF6]

33aH2

Ru

Na / Hg Cl

Fe

S

S

Ru

S

Ru

+

OTf

Fe

Ru

p-cymene [(m-33a)(m-H){Ru(p-cymene)}2](OTf)

p-cymene [(m-32a){Ru(p-cymene)}2](OTf)

Scheme 4.24

The groups of Herberhold and Jin have studied a range of Group 9 metal chelates [33MCp*L] (M = rhodium, iridium) utilising 16 VE metal ligand fragments of the type Cp*ML (M = rhodium: L = PMe3 , CNt-Bu; M = iridium: L = PMe3 , PPh3 , CNtBu) (Scheme 4.25).75–78 33 acts a neutral 2 VE donor in these complexes. Cyclic voltammetry revealed two electrochemically reversible one-electron oxidation steps for the iridium complexes, while the second oxidation of [33RhCp*(PMe3 )] proved to be irreversible. Y

L M

Fe Y

Cp∗

a b c

Y S Se Te

M L Rh PMe3, CNt-Bu Ir PMe3, PPh3, CNt-Bu

[33MCp∗L]

Scheme 4.25

Chemical oxidation of [33aMCp*L] (M = rhodium, L = CNt-Bu;77 M = iridium, L = PMe3 76 ) was performed with Ag[BF4 ] and afforded the respective salt [33a MCp*L][BF4 ]. [33bRhCp*(CNt-Bu)][BF4] was prepared similarly.77 EPR spectroscopic investigations of the electrochemically generated cations [33aMCp*L]+ (M = rhodium, L = PMe3 ; M = iridium, L = PPh3 ) revealed the absence of ferrocenium species. This is in accord with the results of theoretical calculations which demonstrate that the ﬁrst oxidation step is essentially sulfur-centred and does not involve the ferrocene unit.76 Treatment of [33aIrCp*(PPh3 )] with elemental sulfur removed the phosphane ligand.74 This procedure did not lead to the formation of a dative Fe → M bond. Instead, the sulfur-bridged dimeric complex [{(µ-33a)IrCp*}2 ] was

Other Symmetric 1,1 -Bidentate Ferrocene Ligands

167

Figure 4.7 Molecular structure of the cation in [(µ3 -33a)(µ3 -S)(µ2 -Cl)(RuCp*)3 ][FeCl4 ]

isolated in high yield (Scheme 4.26). The same species was obtained in low yield from the metathesis reaction of Li2 33a with the chloro complex [IrCp*(µ-Cl)Cl]2 .78, 79 Each dithiolato ligand functions as a neutral four electron donor in this iridium(III) complex. The analogous reaction of [RhCp*(µ-Cl)Cl]2 afforded the rhodium(III) complex [{(µ33a)RhCp*}2 ] as a side product,79 the main product being [(µ-33a)(RhCp*)2 ].78, 79 For analogues containing 33b or 33c, the reader is referred to Scheme 4.26. The rhodium(II) compound [(µ-33a)(RhCp*)2 ] is a close relative of Hidai’s ruthenium(I) chelate [(µ-33a){Ru(p-cymene)}2 ] (vide supra). Both contain an M−M bond, and the single dithiolato ligand acts a neutral six electron donor, bridging the two metal centres. Redox processes also occurred in the case of cobalt, where the paramagnetic species [{(µ2 -33a)CoCp*}2 Co] was obtained from Li2 33a and [CoCp*(µ-Cl)Cl]2 .78 It contains a bent Co3 chain. As in [(µ-33)(RhCp*)2 ] and [(µ-33a){Ru(p-cymene)}2 ], the bridging ligand acts as a neutral six electron ligand. A lot of work has been done with the dithiolato ligand 33a in Group 10 metal chemistry, addressing, inter alia, the question whether the iron atom can act as an electron donor to metal centres chelated by this ligand. The ﬁrst example, [33aPd(PPh3 )], was prepared in 1983 by Seyferth et al.80 by reacting [Pd(PPh3 )4 ] with 1,2,3-trithia[3] ferrocenophane and later on by Akabori and coworkers81 from 33aH2 and [Pd(PPh3 )4 ] (Scheme 4.27, left).

168

Ferrocenes: Ligands, Materials and Biomolecules S Y=S M = Co

Fe Cp

∗ Co

S Co CoCp∗ Fe

S

S

[{(m2-33a)CoCp∗}2Co] Cp∗ [Cp∗M(m-Cl)Cl]

Li2[Fe(C5H4Y)2] Y = S, Se, Te 2

Y

Rh

Y

Rh

Fe

M = Rh

[(m-33)(RhCp∗)

2]

Cp∗

Y Y=S M = Rh, Ir

M

Fe

Y

Y Ir(PPh3

Fe

)Cp∗

Y

Cp∗

1/8 S8

Y M

− SPPh3

Cp∗

Fe Y

[{(m-33)MCp∗}2]

[33IrCp∗(PPh3)] Y a S b Se

Y M a S Rh, Ir b Se Ir

Scheme 4.26

S Fe

M

S PPh3

Fe

S [33aM(PPh3)] M = Pd, Pt

Ni

PMe2Ph

S [33aNi(PMe2Ph)]

Scheme 4.27

The analogous platinum complex was also prepared by Akabori’s group, using 1,2,3-trithia[3]ferrocenophane,81 while the reaction of 33aH2 with [Pt(PPh3 )4 ] afforded [33aPt(PPh3 )2 ].81 Both compounds [33aM(PPh3 )] contain a dative Fe → M bond ˚ for palladium80, 82 and platinum,81 respectively], rendering [2.878(1) and 2.935(2) A the M atom tetracoordinate. The nickel analogue could not be prepared with PPh3 as a ˚ ligand. However, the chelate [33aNi(PMe2 Ph)] with an Ni−Fe distance of 2.886(1) A was obtained by Hidai and coworkers from the reaction of [NiCl2 (PMe2 Ph)2 ] with 33aH2 in the presence of potassium hydroxide as a base (Scheme 4.27, right).83 Obviously, the comparatively weak Fe → Ni donation is compensated for by the more electron-donating nature of the PMe2 Ph ligand. Use of the bis(phosphane) ligand dppe enforced formation of chelates [33aM(dppe)] without a dative Fe → M bond for

Other Symmetric 1,1 -Bidentate Ferrocene Ligands SH Fe

Ph2 P

S [MCl2(dppe)] base

Fe

169

M

SH

S

P Ph2

[33aM(dppe)] M = Ni, Pd, Pt

33aH2 [RuCl2(p-cymene)]2 [NH4][PF6] acetonitrile

p-cymene S

Ru

S

M

Fe

[Cp2Fe][PF6] CH2Cl2

+

[PF6]

Cl Ph2 P

2

Ph2 P S

M

S

M

Fe

+

2[PF6]

PPh2 Ph2 P P Ph2

P Ph2 [(m-33a{RuCl(p-cymene)}{M(dppe)}][PF6]

[(m-33a){M(dppe)}2][PF6]2

Scheme 4.28

˚ for M = nickel, palladium, platinum, as evidenced by Fe–M distances well above 4 A the structurally characterised chelates of nickel and palladium (Scheme 4.28).83 Oxidation of these compounds with one equivalent of [FeCp2 ][PF6 ] in dichloromethane afforded the ligand-bridged complexes [(µ-33a){M(dppe)}2 ][PF6 ]2 , together with polymeric 1,1 -ferrocenylene disulﬁde (Scheme 4.28). The nickel compound was found to ˚ between its nickel centres.83 The mixed metal anaexhibit a distance of 2.972(3) A logues [(µ-33a){RuCl(p-cymene)}{M(dppe)}][PF6] were obtained from the reaction of [33aM(dppe)] (M = nickel, palladium, platinum) with [RuCl2 (p-cymene)]2 in acetonitrile in the presence of [NH4 ][PF6 ] (Scheme 4.28). The ruthenium atom is in a distorted pseudotetrahedral ligand environment, when the aromatic ligand is viewed as occupying a single coordination site, whereas the respective M atom exhibits the expected square-planar coordination that seems to be slightly inﬂuenced by an additional weak interaction with the chloro ligand of the neighbouring ruthenium atom (Figure 4.8).

4.5 4.5.1

[Se,Se] Ligands The Diselenoether Framework fc(SeR)2

The number of selenoether ligands used for chelate formation is small in comparison to their thioether relatives. In close analogy to their work with the thioether ligand 16a, Abel and coworkers have systematically investigated a range

170

Ferrocenes: Ligands, Materials and Biomolecules

Figure 4.8 Molecular structure of the cation in [(µ-33a){RuCl(p-cymene)}{Ni(dppe)}][PF6]

of chelates containing 34a, namely [34aM(CO)4 ] (M = chromium, molybdenum, tungsten),32 [34aReX(CO)3 ] (X = chlorine, bromine, iodine),34 [34aPtXMe3 ] (X = chlorine, bromine, iodine)43 and [34aPdCl2 ].37 The last compound was also reported by Sato and coworkers.23 Activation parameters for selenium inversion were obtained for all of these chelates. Selenium inversion turned out to be appreciably higher in energy than sulfur inversion. In their studies concerning chelates which exhibit an ironmetal interaction, Sato and coworkers have prepared [34aPd(PPh3 )][BF4 ]2 .23 As for the sulfur analogue [16aPd(PPh3 )][BF4 ]2 (vide supra) and similar closely related thioether derivatives, the synthesis involved chloride abstraction from the dichloro derivative [34aPdCl2 ] with Ag[BF4 ] in the presence of PPh3 . [34bPd(PPh3 )][BF4 ]2 was obtained similarly from [34bPdCl2 ]. The group of Laguna studied a number of silver chelates containing the aromatic selenoether 34b.45 Reaction of this ligand with silver triﬂate in dichloromethane afforded [34bAg(OTf)]. Conductivity measurements indicated that this species hardly dissociates in solution. The compound forms triﬂate-bridged dimeric units in the solid state (Figure 4.9). The reaction of [34bAg(OTf)] with 1,10-phenanthroline (phen) afforded the ionic [34bAg(phen)](OTf), which turned out to be a typical 1:1 electrolyte. The closely related [34bAg(PPh3 )](OTf) was prepared from 34b and [Ag(OTf)(PPh3 )] and behaved similarly. The same holds true for the gold analogue [34bAu(PPh3 )](OTf). When

Other Symmetric 1,1 -Bidentate Ferrocene Ligands

171

Figure 4.9 View of a dimeric aggregate of [34bAg(OTf)]

silver triﬂate was allowed to react with two equivalents of 34b, [(34b)2 Ag](OTf) was obtained, whose conductivity data again were indicative of a 1:1 electrolyte. In each case, Laguna and coworkers have also prepared the corresponding thioether analogue containing ligand 16g (vide supra). Finally, the platinum compounds [34cPtCl2 ]84 and [34dPtCl2 ],85 which were prepared by Chen and coworkers, turned out to be suitable for catalytic hydrosilylation reactions with (EtO)3 SiH. The latter was thoroughly investigated by multinuclear NMR spectroscopy to extract rate constants for conformational exchange.86–88 4.5.2

The Diselenolato Framework fc(Se)2 2−

In close analogy to their work with the dithiolato ligand 33a, the groups of Herberhold and Jin have reported a number of diselenolato relatives of vanadium, molybdenum, tungsten, rhenium, ruthenium, osmium, rhodium and iridium (vide supra). The only compounds from these two research groups not already indicated in Section 4.4.5 of this chapter are [33bMoCp*(NO)] and [33bMoCp2 ] published by Jin et al.66 In platinum chemistry, Akabori and coworkers have obtained [33bPt(PPh3 )2 ] from the reaction of 1,2,3-triselena[3]ferrocenophane with [Pt(PPh3 )4 ].81 The sulfur analogue [33aPt(PPh3 )2 ] was prepared by a different route involving 33aH2 and [Pd(PPh3 )4 ], since with 1,2,3-trithia[3]ferrocenophane the metal–metal bonded species [33aPt(PPh3 )] was formed (vide supra). Brown and Corrigan have prepared the Pn-Bu3 analogue [33bPt(Pn-Bu3 )2 ] from 33b(SiMe3 )2 and cis-[PtCl2 (Pn-Bu3 )2 ] in THF, while [{(µ-33b)Pt(Pn-Bu3)}2 ] was obtained with trans-[PtCl2 (Pn-Bu3 )2 ] (Scheme 4.29).89 The palladium analogue was prepared similarly. These ligandbridged complexes [{(µ-33b)M(Pn-Bu3 )}2 ] (M = palladium, platinum) contain two

172

Ferrocenes: Ligands, Materials and Biomolecules SeSiMe3

Se

cis-[PtCl2(Pn-Bu3)2]

Fe

THF SeSiMe3

Pt

Fe Se

Pn-Bu3

[33bPt(Pn-Bu3)2]

THF trans-[MCl2(Pn-Bu3)2]

Se

Pn-Bu3

Pn-Bu3 M

Fe

Se

Se M n-Bu3P

Fe Se

[{(m-33b)M(Pn-Bu3)}2] M = Pd, Pt

Scheme 4.29

identical ferrocene units, which gave rise to two one-electron redox waves in the respective cyclic voltammogram with a wave splitting of 162 and 135 mV for M = palladium and platinum, respectively. These rather low E 1/2 values indicate that the corresponding mixed-valence species most likely belong to Class I according to the Robin and Day classiﬁcation.90 [33bPt(Pn-Bu3 )2 ] also showed two one-electron oxidation waves, although only one ferrocene unit is present here. In line with results obtained by Herberhold and Zanello with [33bMCp*(PMe3 )] (M = rhodium, iridium),76 the additional oxidation was assigned to a selenium-based process. Finally, Corrigan and coworkers have used 33b(SiMe3 )2 for the synthesis of a range of copper complexes, which formed under low temperature conditions with this reagent and CuOAc in the presence of suitable phosphanes. Most of the products turned out to contain copper–selenium cluster units. Cluster formation was mainly controlled by the stoichiometric ratio of the starting materials and the steric properties of the phosphanes used. Reaction of one equivalent of 33b(SiMe3 )2 with two equivalents of CuOAc in the presence of four equivalents of Pi-Pr3 afforded [(µ-33b){Cu(Pi-Pr3)}2 ], while with Pn-Pr3 [(µ3 33b)2 Cu4 (Pn-Pr3 )4 ] was formed (Figure 4.10).91 With six equivalents of PEtPh2 , the cluster [(33b)4 Cu8 (PEtPh2 )4 ] was obtained.92 This method also afforded the considerably larger clusters [(33b)4 Cu20 Se6 (Pn-Pr3 )10 ],91 [(33b)4 Cu20 Se6 (Pn-Bu3 )10 ],93 [(33b)4 Cu20 Se6 (PEtPh2 )10 ],93 [(33b)6 Cu36 Se12 (Pn-Pr2 Ph)12 ],91 [(33b)6 Cu36 Se12 (PnPr3 )10 {P[(CH2 )3 SH]Ph2 }2 ]91 and [(33b)8 Cu40 Se12 (PPh3 )9 ]91 . In the chemistry of silver(I), a single example has been described to date, namely [(µ2 -33b)3 Ag4 ][Nn-Bu4 ]2 , which was obtained in a stoichiometric reaction from silver(I) chloride (four equivalents) and 33b(SiMe3 )2 (three equivalents) in the presence of Pn-Bu3 (eight equivalents) and [Nn-Bu4 ]Br (two equivalents) in THF.94

Other Symmetric 1,1 -Bidentate Ferrocene Ligands

173

Figure 4.10 Molecular structures of [(µ-33b){Cu(Pi -Pr3 )}2 ] (left) and [(µ3 -33b)2 Cu4 (PnPr3 )4 ] (right)

4.6 4.6.1

[Te,Te] Ligands The Ditellurolato Framework fc(Te)2 2−

The groups of Herberhold and Jin have described the only examples to date of ferrocene-based chelates with two ligating tellurium atoms. All of these complexes contain the ditellurolato ligand 33c and have already been mentioned in Section 4.4.5. Electrochemical investigations were hampered by very fast decomposition of the compounds upon oxidation.76 Nevertheless, two oxidation processes have been observed for [33cMCp*(PMe3 )] (M = rhodium, iridium), which is in accord with the results obtained with the corresponding thiolato and selenolato analogues (vide supra).

4.7

Conclusion and Outlook

Much progress has been made in the subject area of this review over the past decades. The focus has clearly been on synthetic methods and structural aspects so far, where a state of maturity has now been reached. Considering the number of recent papers, the two most active domains currently are those concerning ferrocene-based diamido ligands and sulfur-containing ligand systems, whereas the chemistry of ferrocene-based

174

Ferrocenes: Ligands, Materials and Biomolecules

[O,O] and [Te,Te] chelate ligands appears to be in a dormant state. There is a huge lack of knowledge especially in the area of transition metal chelates containing the long known fc(O)2 2− ligand (14), where just a single example has been reported to date. Examples for successful applications have been reported with quite a number of compounds already, and it is clear that this is only the beginning, since ‘studies on the way’ have been announced in several very recent papers, most of them addressing catalytic applications. In view of the rather well-developed synthetic methodology available, this augurs well for new and exciting results, especially in the area of catalysis. The well-behaved redox-chemical properties of ferrocene derivatives will be beneﬁcial for future investigations concerning the redox-tunability of molecular structure and reactivity with these compounds. This fascinating aspect has been addressed in only a few papers so far and is still in its infancy. Another hitherto almost neglected aspect which will most likely attract more attention in the future concerns stereochemistry, since it is easy to introduce chirality into ferrocene derivatives, and the ferrocene-based planar chirality never undergoes racemisation.95

Abbreviations Cy Mes Py TIPP Tol Tp Tp*

cyclohexyl mesityl 2-pyridyl 2,4,6-triisopropylphenyl p-tolyl κ 3 -hydrotri(pyrazol-1-yl)borate κ 3 -hydrotris(3,5-dimethylpyrazol-1-yl)borate

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.

J.J. Bishop, A. Davison, Inorg. Chem., 1971, 10, 826–831. U. Siemeling, T-C. Auch, Chem. Soc. Rev., 2005, 34, 584–594. H. Plenio, C. Aberle, Chem. Eur. J., 2001, 7, 4438–4446. H. Plenio, C. Aberle, Angew. Chem. Int. Ed., 1998, 37, 1397–1399. Z. Weng, L.L. Koh, T.S.A. Hor, J. Organomet. Chem., 2004, 689, 18–24. V.C. Gibson, C.K.A. Gregson, C.M. Halliwell et al. J. Organomet. Chem., 2005, 690, 6271–6283. A. Shaﬁr, D. Fiedler, J. Arnold, Dalton Trans., 2002, 555–560. C. Metallinos, D. Tremblay, F.B. Barrett, N.J. Taylor, J. Organomet. Chem., 2006, 691, 2044–2047. V.C. Gibson, S.K. Spitzmesser, Chem. Rev., 2003, 103, 283–316. G.J.P. Britovsek, V.C. Gibson, D.F. Wass, Angew. Chem. Int. Ed., 1999, 38, 428–447. A. Shaﬁr, M.P. Power, G.D. Whitener, J. Arnold, Organometallics, 2001, 20, 1365–1369. U. Siemeling, O. Kuhnert, B. Neumann et al. Eur. J. Inorg. Chem., 2001, 913–916. V.C. Gibson, N.J. Long, E.L. Marshall et al. Dalton Trans., 2001, 1162–1164. A. Shaﬁr, J. Arnold, Inorg. Chim. Acta, 2003, 345, 216–220. U. Siemeling, T-C. Auch, O. Kuhnert et al. Z. Anorg. Allg. Chem., 2003, 629, 1334–1336. A. Shaﬁr, D. Fiedler, J. Arnold, Chem. Commun., 2003, 2598–2599.

Other Symmetric 1,1 -Bidentate Ferrocene Ligands 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29.

30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59.

175

B. Wrackmeyer, E.V. Klimkina, W. Milius, Inorg. Chem. Commun., 2006, 9, 716–721. I. Westmoreland, J. Arnold, Acta Crystallogr. Sect. E , 2006, m2272–m2273. I. Westmoreland, J. Arnold, Acta Crystallogr. Sect. E , 2006, m2303–m2304. U. Siemeling, T-C. Auch, S. Tomm et al. Organometallics, 2007, 26, 1112–1115. A. Shaﬁr, J. Arnold, J. Am. Chem. Soc., 2001, 123, 9212–9213. A. Shaﬁr, J. Arnold, Organometallics, 2003, 22, 567–575. M. Sato, M. Sekino, M. Katada, S. Akabori, J. Organomet. Chem., 1989, 377, 327–337. S. Akabori, Y. Habata, M. Sato, Bull. Chem. Soc. Jpn., 1985, 58, 783–784. S. Akabori, Y. Habata, M. Sato, Bull. Chem. Soc. Jpn., 1984, 57, 68–72. S. Akabori, Y. Habata, Y. Sakamoto et al. Bull. Chem. Soc. Jpn., 1983, 56, 537–541. A.M. Allgeier, C.S. Slogane, C.A. Mirkin et al. J. Am. Chem. Soc., 1997, 119, 550–559. S. Akabori, T. Kumagai, T. Shirahige et al. Organometallics, 1987, 6, 2105–2109. For example, Ti dialkoxo complexes are known to be useful not only for α-oleﬁn polymerisation,9,10 but also for the addition, both stoichiometric and catalytic, of nucleophiles to carbonyl groups, and enantioselective reactions of this type are easily feasible with chiral dialkoxo ligands; for a review, see R.O. Duthaler, A. Hafner, Chem. Rev., 1992, 92, 807–832. M.C. Gimeno, P.G. Jones A. Laguna, C. Sarroca, Z. Naturforsch. B, 2004, 59, 1365–1371. O. Crespo, C. Gimeno, P.G. Jones, et al. Chem. Commun., 1998, 1481–1482. E.W. Abel, N.J. Long, K.G. Orrell et al. J. Organomet. Chem., 1989, 367, 275–289. V.C. Gibson, N.J. Long et al. Organometallics, 2004, 23, 957–967. E.W. Abel, N.J. Long, K.G. Orrell et al. J. Organomet. Chem., 1990, 383, 253–269. ˇ K.G. Orrell, V. Sik, C.H. Brubaker, Jr, B. McCulloch, J. Organomet. Chem., 1984, 276, 267–279. B. McCulloch, D.L. Ward, J.D. Woolins, C.H. Brubaker, Jr, Organometallics, 1985, 4, 1425–1432. E.W. Abel, N.J. Long, K.G. Orrell et al. J. Organomet. Chem., 1991, 405, 375–382. Y.-y. Chen, J.-j. Ni, J.-z. Yao, Cuihua Xuebao, 1991, 12, 466–471. V.C. Gibson, N.J. Long, A.J.P. White et al. Organometallics, 2000, 19, 4425–4428. V.C. Gibson, N.J. Long, C.K. Williams et al. Dalton Trans., 2003, 3599–3605. M. Sato, M. Sekino, S. Akabori, J. Organomet. Chem., 1988, 344, C31–C34. M. Sato, K. Suzuki, H. Asano et al. J. Organomet. Chem., 1994, 470, 263–269. E.W. Abel, N.J. Long, K.G. Orrell et al. J. Organomet. Chem., 1989, 378, 473–483. N.J. Long, J. Martin, G. Opromolla et al. Dalton Trans., 1999, 1981–1986. S. Canales, O. Crespo, A. Fortea et al. Dalton Trans., 2002, 2250–2255. S. Coco, P. Espinet, J. Organomet. Chem., 1994, 484, 113–118. M. Sato, K. Suzuki, S. Akabori, Bull. Chem. Soc. Jpn., 1986, 59, 3611–3615. M. Sato, K. Suzuki, S. Akabori, Chem. Lett., 1987, 2239–2242. M. Sato, H. Asano, K. Suzuki et al. Bull. Chem. Soc. Jpn., 1989, 62, 3828–3834. M. Sato, H. Asano, S. Akabori, J. Organomet. Chem., 1993, 452, 105–109. M. Sato, S. Akabori, M. Katada et al. Chem. Lett., 1987, 1847–1850. M. Sato, M. Katada, S. Nakashima et al. Dalton Trans., 1990, 1979–1984. M. Sato, S. Akabori, Bull. Chem. Soc. Jpn., 1989, 62, 3492–3497. M. Sato, H. Asano, S. Akabori, J. Organomet. Chem., 1991, 401, 363–370. M. Sato, S. Tanaka, S. Akabori, Y. Habata, Bull. Chem. Soc. Jpn., 1986, 59, 1515–1519. S. Akabori, S. Sato, M. Sato, Y. Takanohashi, Bull, Chem. Soc. Jpn., 1989, 62, 1582–1586. M. Sato, M. Kubo, S. Ebine, S. Akabori, Bull. Chem. Soc. Jpn., 1984, 57, 421–425. M. Sato, H. Anano, J. Organomet. Chem., 1998, 555, 167–175. C.K.A. Gregson, N.J. Long, A.J.P. White, D.J. Williams, Organometallics, 2004, 23, 3674–3682.

176

Ferrocenes: Ligands, Materials and Biomolecules

60. S. Takemoto, S. Kuwata, Y. Nishibayashi, M. Hidai, Organometallics, 2000, 19, 3249– 3252. 61. V.C. Gibson, N.J. Long, A.J.P. White et al. Chem. Commun., 2000, 2359–2360. 62. R. Steudel, K. Hassenberg, J. Pickardt et al. Organometallics, 2002, 21, 2604–2608. 63. C-S. Liu, Y-K. Zhou, Youji Huaxue, 1996, 16, 270–273. 64. V.C. Gibson, N.J. Long, J. Martin et al. J. Organomet. Chem., 1999, 590, 115–117. 65. M. Herberhold, M. Schrepfermann, A.L. Rheingold, J. Organomet. Chem., 1990, 394, 113–120. 66. G-X. Jin, Y. Liu, X-Y. Yu, Chin. J. Org. Chem., 2000, 20, 352–356. 67. R.P. Sidebotham, P.D. Beer, T.A. Hamor et al. J. Organomet. Chem., 1989, 371, C31–C34. 68. J.R. Dilworth, S.K. Ibrahim, Transition Met. Chem., 1991, 16, 239–240. 69. M. Herberhold, G.X. Jin, W. Milius, J. Organomet. Chem., 1996, 512, 111–116. 70. D. Seyferth, B.W. Hames, Inorg. Chim. Acta, 1983, 77, L1–L2. 71. W.R. Cullen, A. Talaba, S.J. Rettig, Organometallics, 1992, 11, 3152–3156. 72. I.M. Lorkovi´c, X. Bu, P.C. Ford, Inorg. Chim. Acta, 2000, 307, 77–80. 73. M. Herberhold, G-X. Jin, I. Trukenbrod, W. Milius, Z. Anorg. Allg. Chem., 1996, 622, 724–728. 74. M. Herberhold, G-X. Jin, A.L. Rheingold, G.F. Sheats, Z. Naturforsch. B, 1992, 47, 1091–1098. 75. M. Herberhold, C. D¨ornh¨ofer, A. Scholz, G.X. Jin, Phosphorus, Sulfur, Silicon, 1992, 64, 161–168. 76. P. Zanello, M. Casarin, L. Pardi et al. J. Organomet. Chem., 1995, 503, 243–250. 77. G-X. Jin, Y. Liu, X-Y. Yu, Chin. J. Org. Chem., 2000, 20, 202–205. 78. M. Herberhold, G-X. Jin, A.L. Rheingold, Angew. Chem. Int. Ed. Engl., 1995, 34, 656–657. 79. M. Herberhold, G-X. Jin, A.L. Rheingold, Z. Anorg. Allg. Chem., 2002, 628, 1985–1990. 80. D. Seyferth, B.W. Hames, T.G. Tucker et al. Organometallics, 1983, 2, 472–474. 81. S. Akabori, T. Kumagai, T. Shirahige et al. Organometallics, 1987, 6, 526–531. 82. M. Cowie, R.S. Dickson, J. Organomet. Chem., 1987, 326, 269–280. 83. S. Takemoto, S. Kuwata, Y. Nishibayashi, M. Hidai, Inorg. Chem., 1998, 37, 6428–6434. 84. J.Z. Yao, B.S. Tian, Y.Y. Chen, Chin. Chem. Lett., 1993, 4, 601–602. 85. J-Z. Yao, Y-Y. Chen, B-S. Tian, J. Organomet. Chem., 1997, 534, 51–56. 86. X-A. Mao, J-Z. Yao, B-S. Tian, Y-Y. Chen, Magn. Res. Chem., 1996, 34, 109–115. 87. J-L. Yan, X-A. Mao, L-F. Shen, Chem. Phys. Lett., 1997, 272, 278–283. 88. J. Yan, X. Mao, L. Shen, Wuli Huaxue Xuebao, 1997, 13, 853–856. 89. M.J. Brown, J.F. Corrigan, J. Organomet. Chem., 2004, 689, 2872–2879. 90. M.B. Robin, P. Day, Adv. Inorg. Chem. Radiochem., 1967, 10, 247–422. 91. C. Nitschke, D. Fenske, J.F. Corrigan, Inorg. Chem., 2006, 45, 9394–9401. 92. A.I. Wallbank, J.F. Corrigan, Chem. Commun., 2001, 377–378. 93. A.I. Wallbank, A. Borecki, N.J. Taylor, J.F. Corrigan, Organometallics, 2005, 24, 788–790. 94. A.I. Wallbank, J.F. Corrigan, J. Cluster Sci., 2004, 15, 225–232. 95. R.G. Array´as, J. Adrio, J.C. Carretero, Angew. Chem. Int. Ed., 2006, 45, 7674–7715.

5 1-Functionalised Ferrocene Phosphines: Synthesis, Coordination Chemistry and Catalytic Applications ˇ epniˇcka Petr Stˇ

5.1

Introduction

This chapter focuses on the preparation, coordination behaviour and catalytic applications of heterodifunctional ferrocene donors that combine a phosphino moiety with another non-phosphine, potentially donating group. It also gives examples of the use of these compounds as organometallic synthons. The scope is restricted to the cases where the donor moieties occupy 1,1 -positions of the ferrocene framework and are directly attached to it.1 The related compounds with the donor asymmetry arising only from the presence of different substituents at trivalent phosphorus atoms in 1,1 -positions (e.g., R1 R2 PfcPR3 R4 and R12 PfcP(OR2 )2 ) or from partial derivatisation of symmetrical 1,1 -diphosphines such in 1,1 -bis(diphenylphosphino)ferrocene monochalcogenides (dppfE, E = oxygen, sulfur, and selenium) have been addressed in Chapters 2 and 3. Due to space constraints, the chapter concentrates only on the most relevant compound and reaction types reported up to autumn 2006. At the very beginning it should be noted that functionalised ferrocene monophosphines have been well established since the early days of ferrocene chemistry. As a class of ligands, however, they have been put into the shade by the exceedingly rich chemistry of their symmetric congeners, particularly dppf (Chapter 2), and by the numerous donors with combined planar and central chirality (Chapter 6). Only Ferrocenes: Ligands, Materials and Biomolecules  2008 John Wiley & Sons, Ltd

Edited by Petr Stepnicka

178

Ferrocenes: Ligands, Materials and Biomolecules

recently, the interest in the chemistry of 1 -functionalised ferrocene phosphines has renewed.2 This revival can be mainly accounted for by the great versatility of these donors. The easy modiﬁcation at the phosphine part together with a wide choice of the synthetically accessible functional groups and their subsequent chemical transformations allow donors to be synthesised that have electronic and steric properties which can vary over a wide range. The combination of the unlike donor centres often results in interesting coordination chemistry and, for a speciﬁc metal–ligand combination, may lead to hemilabile systems.3, 4 When uncoordinated, the modifying groups can form bridges to other metal centres or can enter into noncovalent intermolecular interactions, thus enabling the formation of supramolecular assemblies in the solid state. This often occurs with soft metal ions5 and phosphines bearing polar donor groups that bind preferably as P-monodentate donors. It should also be pointed out that ferrocene donors represent unique redox-active metalloligands, the incorporation of which increases the nuclearity of the whole coordination assembly. In some cases, coordinated ferrocene ligands may serve as probes at the molecular level, since coordination-induced changes in the overall ‘chemical structure’ can be conveniently followed via monitoring the properties of the ferrocene unit by physicochemical (typically spectral) methods and, particularly, electrochemical techniques.

5.2

Synthetic Methods for the Preparation of 1,1 -Unsymmetrically Disubstituted Ferrocenes

Owing to difﬁculties associated with selective monometalation of ferrocene and its substitution derivatives, the preparation of 1,1 -unsymmetrically disubstituted ferrocenes typically relies on step-wise transmetalation/functionalisation reactions of suitable 1,1 symmetric precursors. In 1990, Wright reported that 1,1 -bis(tributylstannyl)ferrocene (2) undergoes selective lithium–tin exchange upon addition of one molar equivalent of butyllithium at low temperature to give Bu3 SnfcLi intermediate, which reacts smoothly with electrophilic reagents to yield monostannyl derivatives. Repeating this reaction sequence (without or after modiﬁcation of the previously introduced functional group) allows the preparation of asymmetric products (Scheme 5.1, route A).6 The starting bis(stannyl)ferrocene 2 is obtained as an air-stable, viscous liquid in reasonable yields and purity by reacting 1,1 -dilithioferrocene–N ,N ,N ,N -tetramethyl1,2-diaminoethane (TMEDA) adduct (1)7 with chloro(tributyl)stannane.6, 8 Later, this approach has been successfully extended to 1,1 -dibromoferrocene (3, Scheme 5.1, route A).9 Syntheses starting with 3 are particularly attractive as they avoid toxic, environmentally troublesome organotin compounds, producing only watersoluble side-products (lithium salts). Moreover, the starting dibromide is an easy-tohandle, air-stable crystalline solid, which can be synthesised in multigram quantities from 1 and 1,2-dibromotetraﬂuoroethane9b, 10, 11 or 1,1,2,2-tetrabromoethane.12 Selective transmetalation reactions involving other 1,1 -symmetric ferrocene derivatives have not yet found practical use.13 An older approach that is complementary to the above methods makes use of the high reactivity of phosphorus-bridged [1]ferrocenophanes,14 apparently resulting from

The Chemistry of 1 -Functionalised Ferrocene Phosphines

179

route A Y

E

1

1. LiBu Fe 2.

E1X

1. LiBu Fe

Y

SnClBu3 (2) or 1,2-C2X4Br2 (3)

E1 Fe

2. E2X

E2

Y

2, Y = SnBu3 3, Y = Br Li (TMEDA)x

Fe Li 1

route B

PhPCl2

P(R)Ph

P(R)Ph Fe

PPh

LiR

EX

Fe Li

5

Fe E

6 (R = Ph)

Scheme 5.1

steric strain imposed by the single-atom bridge.15 In the early 1980s, (ferrocene-1,1diyl)phenylphosphine (5, fcPPh) was shown to undergo facile ring-opening under the action of organolithium reagents (LiR) at low temperature, yielding the reactive intermediates Ph(R)PfcLi. These organolithiums readily react further with electrophiles to afford functionalised phosphines Ph(R)PfcE in moderate to good yields (Scheme 5.1, route B; R = Ph: E (reagent) = H (H2 O), SiMe3 (ClSiMe3 ), SiPh3 (ClSiPh3 ), SnMe3 (ClSnMe3 ), PPh2 (ClPPh2 ); R = Me: E = H; R = t-Bu: FcP(S)Ph(t-Bu) (H2 O, then S8 )).16 Phosphaferrocenophanes 5,16, 17 fcPMe,16 fcP(t-Bu) as well as some Cp-ring substituted derivatives18 all react similarly. The ferrocenophanes are obtained as air sensitive solids in low to moderate yields from simple metathesis of 1 with the appropriate dichlorophosphine. It is noteworthy that the use of fcPR1 and LiR2 (R1 = R2 ) leads to racemic mixtures of a P-chiral phosphines; however, this route to P-chiral ferrocene donors has not yet been pursued. The functional ferrocene phosphines studied to date are mainly those derived (formally) from (diphenylphosphino)ferrocene, very likely due to the satisfactory properties of the Ph2 PfcX-type compounds and also for practical reasons such as their relative stability and availability of necessary educts. This rendered 1-(diphenylphosphino)-1 bromoferrocene (Ph2 PfcBr, 4) as the frequently used precursor for the synthesis of 1 -functionalised ferrocene phosphines. Bromide 4 is smoothly and cleanly converted to 1 -(diphenylphosphino)-1-lithioferrocene (6) and can be prepared by all methods outlined above; most easily, however, from dibromide 3.9c Yet another synthetic approach towards 1 -functionalised ferrocene phosphines is represented by speciﬁcally directed metalation of monosubstituted ferrocenes, though this is limited in scope. For instance, ferrocenecarbaldehyde (7) reacts with lithium

180

Ferrocenes: Ligands, Materials and Biomolecules

salt of N -methypiperazine to give an intermediate aminal, which can be functionalised at the unsubstituted cyclopentadienyl ring in good selectivity but with only modest yields by treating sequentially with Li(t-Bu) and an electrophile. The aldehyde group is restored during the subsequent hydrolytic work-up (Scheme 5.2; 17–69 % yields and (1,1 ):(1,2) selectivity of 90:10–96:4 were reported for the series where E = Me, Et, SiMe3 , SnBu3 , PPh2 , B(OH)2 , I, and CHO).19 H CHO Fe

Li

N N

N

Me

room temp., 2 h

OLi

Fe

N Me

7

CHO Fe

CHO >>

Fe

E

1. Li(t-Bu), 0 °C 2. EX, then H2O

E

Scheme 5.2

Similarly, the reaction of Boc-protected 1-(ferrocenyl)ethylamine, FcCH(Me)NHBoc (Boc = (t-butyloxy)carbonyl), with two equivalents of butyllithium results in selective N ,1 -dimetalation. Subsequent reactions with electrophiles followed by standard aqueous work-up afford 1 -functionalised, N -protected amines in very good yields. The course of the metalation reaction changes dramatically with the nature of the substituent at the one-carbon spacer that connects the amino group to the ferrocene unit, and also with the kind of the nitrogen-protecting group. For instance, lithiation of various N -acyl amines proceeds with much lower selectivity.20

5.3

Donors of the P,C-type

Unsymmetric 1,1 -ferrocene ligands combining a σ - or π-donating hydrocarbyl moiety with a phosphino group are still rather scarce. Early attempts at preparation of complexes with σ -bonded [Ph2 Pfc]− anion, the archetypal member of this family,21 can be exempliﬁed by the reaction of 1-(diphenylphosphino)-1 -lithioferrocene (6) with [(η5 -C5 H5 )Fe(CO)2 I] (FpI), which produces chromatographically separable mixture of two ferrocenophanes 8 and 9, Ph2 PfcI and [{(η5 -C5 H5 )Fe(CO)2 }2 ] (Scheme 5.3). The formation of the insertion product 9 was rationalised by its stabilisation under the reaction conditions (low temperature). A similar reaction between 6 and [(η5 C5 H5 )Fe(CO)(PPh3 )I] gave a mixture of Ph2 PfcI and ferrocenophane 8, resulting from intramolecular, chelation-assisted replacement of the PPh3 ligand.22

The Chemistry of 1 -Functionalised Ferrocene Phosphines PPh2

PPh2 FpI 6

+

Fe

−LiI

181

+ Ph2PfcI

Fe

Fe

+ [{(C5H5)Fe(CO)2}2]

CO

C

Fe

O

CO 8 (30 %)

9 (<5 %)

Scheme 5.3

Only shortly afterwards it was shown that the reaction of 6 with an excess of Group 6 metal carbonyls followed by alkylation with [Me3 O]BF4 gives a mixture of Pchelated Fischer-type23 alkoxycarbenes 10 and zwitterionic carbonylate-phosphonium compounds 11 (Scheme 5.4; M = chromium (a) and tungsten (b)). The structures of 10b and 11a were conﬁrmed by single-crystal X-ray diffraction.24 Ph P 6

1. [M(CO)6] 2. [Me3O]BF4

Ph M(CO)4

Fe

+

PPh2Me +

Fe _

C 10

M(CO)5 OMe

11

Scheme 5.4

The related aminocarbenes25 were synthesised by reacting tertiary 1(diphenylphosphino)-1 -ferrocenecarboxamides (12, NR2 = NEt2 (a) or morpholin-4yl (b)) with the in situ generated carbonylate salts Na2 [M(CO)5 ] (M = chromium and tungsten) and chloro(trimethyl)silane.26 The starting amides were obtained in practically quantitative yields by amidation of 1-(diphenylphosphino)-1 -ferrocenecarboxylic acid27 (Hdpf; see below). The reactions with the chromium carbonylate gave directly the corresponding P-chelated carbenes 13a-b, together with the amidophosphine complexes 14a-b as by-products (Scheme 5.5). By contrast, the reaction of the tungsten carbonylate with 12a under identical conditions yielded a mixture of non-chelated carbene 15, trinuclear ferrocene-bridged complex 16 and the expected side product 17. Removal of 17 by chromatography and thermolysis of the 15–16 mixture ﬁnally led to a separable mixture of P-chelated carbene 18, unchanged 16, and [W(CO)5 (Ph2 PfcCHO-κP )] (19) as a decomposition product (Scheme 5.6, Figure 5.1). The isolation of the non-chelated carbene suggested that the reaction proceeds in a step-wise manner, the carbene moiety being formed ﬁrst. Attempts to prepare similar iron(0) carbenes failed; the reaction between 12a and [Fe(CO)4 ]2− /SiMe3 Cl28 gave only the amidophosphine complex [Fe(CO)4 (12a-κP )].25 An electrochemical study performed on amides 12 and all the chromium complexes mentioned above showed that whereas amides 12 and the phosphine complexes 14

182

Ferrocenes: Ligands, Materials and Biomolecules Ph

Ph

P

Cr(CO)4

Fe C PPh2

PPh2 HNR2

Fe

Fe

EDC/HOBt

2. Me3SiCl

O

12a-b

Hdpf

Ph

Ph

C

CO2H

NR2

13a-b +

1. Na2[Cr(CO)5]

P

Cr(CO)5

NR2 Fe O C

NR2 14a-b

Scheme 5.5 Ph

Ph PPh2

P

+

Fe C

NEt2

12a

(80 °C/4 h)

NEt2

W(CO)5

16

15

2. Me3SiCl

Fe C

W(CO)5 1. Na2[W(CO)5]

W(CO)5

Ph

Ph P

Ph

W(CO)5

P

Fe C(O)NEt2

16 +

17

W(CO)4 +

Fe

P

W(CO)5

Fe CHO

C NEt2

18

Ph

Ph

Ph

19

Scheme 5.6

behave as simple localised (though communicating) redox systems (ferrocene/ferrocenium for 12, ferrocene/ferrocenium and chromium(0)/chromium(I) for 14), the carbenes represent electronically coupled systems where redox changes encompass the whole molecule.25 1-(Diphenylphosphino)-1 -vinylferrocene (20) as the simplest potentially P:η2 coordinating ferrocene donor has been reported ﬁrstly as an undesired product resulting from acid-catalysed dehydration29 of 1-(diphenylphosphino)-1 -(1-hydroxyethyl) ferrocene (21) during attempted preparation of 1-(diphenylphosphino)-1 -[1-(N ,N dimethylamino)ethyl]ferrocene (22) (Scheme 5.7).30 The hydroxy derivative 21 was

O1

C28

C29

W1

O5

C32

C31

C11

N

C13

O4

C1

C2

C12 O10

Fe O9

C37

C7

C33

O6

C6

C36

C22

P

W2

C23

C34

C16

C35

O7

C17

O8

C2

Fe

C7

C13

C1

O3

C6

C12

C11

C29

C23

P

C22

N

W

C16

C17

C14

O2

C28

C31

C15

C30

O5

O4

Figure 5.1 Molecular structures of 16 (left) and 18 (right). Reprinted with permission from Ref. 25. Copyright 2004, American Chemical Society.

O2

C30

O3

C14

C15

The Chemistry of 1 -Functionalised Ferrocene Phosphines 183

184

Ferrocenes: Ligands, Materials and Biomolecules

obtained by metathesis of in situ formed 6 with acetyl chloride to give 1-(diphenylphosphino)-1 -acetylferrocene (23) followed by hydride reduction. Changing the educt ratio (5/LiPh/MeC(O)Cl) inﬂuenced the reaction course so that it afforded considerable amounts of 1,1 -bis[1 -(diphenylphosphino)ferrocenyl]ethene (24) (Scheme 5.7).31, 32 When reacted with Grubbs complex [(Cy3 P)2 Cl2 Ru=CHPh] (Cy = cyclohexyl), phosphinoalkene 20 gave a complicated equilibrium mixture containing free 20 and the P-chelated carbene [(Cy3 P)Cl2 Ru{Ph2 PfcCH-κ 2 C 1 ,P }].33 Ph2P

PPh2 6

MeC(O)Cl

1. 5

Fe

−LiCl

2. −H2O

PPh2 Fe

Fe

C(O)Me

C CH2

23 (39 %)

24

Li[AlH4] PPh2

PPh2 [H+]

Fe

−H2O CH(OH)Me

21 (crude ca. 95 %)

Fe CH CH2 20

Scheme 5.7

Phosphinoalkene 20 was alternatively synthesised by Wittig vinylation of 1-(diphenylphosphino)-1 -ferrocenecarbaldehyde (25; see below). With soft metal precursors, 20 gives simple phosphane complexes, forming trans-[PdCl2 (20-κP )2 ] from [PdCl2 (cod)], [PdCl(LNC )(20-κP )] from [{Pd(µ-Cl)(LNC )}2 ], and trans-[RhCl(CO)(20-κP )2 ] from [{Rh(µ-Cl)(CO)2}2 ] (cod = η2 :η2 -cycloocta-1,5-diene, LNC = 2-(dimethylaminomethyl-κN )phenyl-κC 1 ). Even so, the reactions with copper(I) iodide and [Cu(MeCN)4 ][PF6 ] at appropriate molar ratios yielded heterocubane [(µ3 -I)4 {Cu(20κP )}4 ] and the bis(phosphane) complex [Cu(MeCN)(20-κP )2 ][PF6 ], respectively,34 whilst the metal carbonyl precursors [W(CO)4 (cod)] and [Fe2 (CO)9 ] gave rise exclusively to trans-[W(CO)4 (20-κP )2 ] and [Fe(CO)4 (20-κP )].35 The behaviour of 20 contrasts with that of its planarly chiral isomer, (Sp )-1-(diphenylphosphino)-2-vinylferrocene (27), which readily forms η2 :κP -chelate complexes.36, 37

5.4

P,N-Donors

1,1 -Ferrocene ligands combining phosphorus and nitrogen donor groups are relatively common, mainly because of their obvious relation to catalytically relevant chiral P,Ndonors and their great synthetic potential, as well as the structural and synthetic variability

The Chemistry of 1 -Functionalised Ferrocene Phosphines

185

of the accessible N-functional groups. However, compounds featuring simple primary amino groups were often avoided because of their relatively low stability and undesired acid–base reactions that may complicate their synthesis and subsequent use as ligands. Signiﬁcant stabilisation of ferrocene phosphinoamines was typically achieved through introduction of non-hydrogen substitutents at the nitrogen atom. The archetypal representative of ferrocene-based 1,1 -P,N donors, 1-(diphenyphosphino)-1 -aminoferrocene (28), was ﬁrst reported by Butler and Quayle as late as in 1998. It was synthesised by metalation of 1-(diphenyphosphino)-1 -bromoferrocene (4) with LiBu and reacting the in situ formed 6 with O-benzylhydroxylamine (38 % yield after crystallisation). The phosphinoamine was further reacted with [PdCl2 (cod)] to give [PdCl2 (28)].38 The aminomethyl derivative Ph2 PfcCH2 NH2 has not yet been described. However, Widhalm et al. reported the preparation of the related amine bromide 29 and used it in the synthesis of ferrocene P,N-ligands with axially chiral substituents. Firstly, a mixture of tertiary amine 30 and ammonium salt [Ph2 PfcCH2 NMe2 Bu]Br (31a) was prepared by monometalation of 3 with LiBu and treatment with Eschenmoser’s salt, [CH2 =NMe2 ]I. Heating the salts 31a and 31b (the latter is readily obtained by alkylation of 30 with MeI) with aqueous ammonia–benzene mixture yielded the primary amine. Finally, alkylation of 29 with 2,2 -bis(bromomethyl)-1,1 -binaphtyl gave 32, which underwent smooth phosphinylation to afford axially chiral N -{[1 (diphenylphosphino)ferrocenyl]methyl}-3,5-dihydro-4H -dinaphth[2,1-c:1 ,2 -e]azepine (33) – either as a racemic mixture or enantiopure, axially chiral compound depending on the biphenyl precursor used (Scheme 5.8).39, 40

3

1. LiBu 2. [CH2

NMe2]I

Br

Br

Br +

Fe

Fe

Bu CH2NMe2 +

CH2NMe2 30 (56%)

31a (33%)

MeI

Br /NEt3 I

Br

Br

Me2NH H2O/C6H6 Fe

Fe + CH2NMe3

29

31b

PPh2

Br 1. LiBu 2. Ph2PCl

Fe

Fe

N 33 (81%)

N 32

Scheme 5.8

NH2

Br

186

Ferrocenes: Ligands, Materials and Biomolecules

The N ,N -dimethyl amine 34 was prepared by reductive amination of aldehyde 35 (obtained by mono-lithiation of 2 and reacting with N ,N -dimethyformamide) and replacement of the stannyl with phosphino group (Scheme 5.9; cf. route A in Scheme 5.1). The related C-chiral amine 36 was prepared using a different strategy: via hydride reduction of intermediate imine 37, which results from condensation of aldehyde 25 with (R)-phenylethylamine.6 SnBu3 2

SnBu3 Me2NH

Fe

Fe

Na[BH3(CN)] CHO

PPh2 1. LiBu 2. Ph2PCl

CH2NMe2

CH2NMe2

35

34

PPh2 2

PPh2 amine [H+]

Fe

PPh2 Li[AlH4]

Fe

Fe

Ph

CHO 25

Fe

H N

Ph

CH N 37

36

Scheme 5.9

Hor and coworkers recently used similar methods to synthesise a series of Schiff bases R2 PfcCH=NR (38, R/R = t-Bu/Ph (a), Ph/C6 F5 (b), Cy/CH(Me)Ph (c); Cy = cyclohexyl) and tested these compounds for their coordination and catalytic potential. It was shown that compound 38a reacts with [Pd2 (dba)3 ] (dba = dibenzylideneacetone) to give a mixture of coordinatively unsaturated palladium(0) complexes 39 (16 valence electrons) and 40 (14 valence electrons; Scheme 5.10), which were both isolated and structurally characterised. Addition of pentaﬂuorophenyl iodide to 38a/[Pd2 (dba)3 ] mixture or directly to 40 resulted in oxidative addition of the C−I bond across palladium(0), yielding square-planar palladium(II) complex [Pd(C6 F5 )(I)(38a-κ 2 N ,P )]. This complex reacted with phenylboronic acid in the presence of a base to give 2,3,4,5,6-pentaﬂuorobiphenyl as the C−C coupling product. Indeed, the 38a/[Pd2 (dba)3 ] mixture itself catalyzed efﬁciently Suzuki–Miyaura cross-coupling of arylboronic acids with aryl chlorides.41 Phosphinoimine 38a was also reacted with [NiCl2 (dme)] (dme = 1,2-dimethoxyethane) to give the nickel(II) complex 41 which, upon addition of methyllithium (1 equiv.), underwent formal one-electron reduction to give the nickel(I) complex 42 (Scheme 5.11). Treatment of 41 and 42 with two equivalents of AlMe3 or with MAO (‘methylalumoxane’) was shown to give complex 43 featuring an ‘activated’ aldimine moiety. As revealed by X-ray crystallography, the activation is manifested in a signif˚ for 42 and 1.435(9) for 43) icant elongation of the C=N bond length (cf. 1.288(3) A as well as in changed nickel–donor distances, while maintaining planar coordination environment around the nickel atom (Figure 5.2). Finally, reacting [Ni(cod)2 ], t-BuNC, and 38a or 38b produced the ligand-substitution products [Ni(L-κP )(CNt-Bu)3 ]: 44a

The Chemistry of 1 -Functionalised Ferrocene Phosphines Ph

t-Bu

O

t-Bu P Pd

Fe

Ph

CH N Ph

P(t-Bu)2 39 +

[Pd2(dba)3]

Fe

HC

PhN

CH NPh 38a

Fe

t-Bu t-Bu P Fe

P

Pd

t-Bu

t-Bu CH

NPh 40

Scheme 5.10

t-Bu

t-Bu P

38a + [NiCl2(dme)]

Cl Ni

Fe

Cl

CH N Ph

41 LiMe/−30 °C t-Bu

t-Bu

t-Bu P

t-Bu P

Cl Ni

Fe

−30 °C

N

C

AlMe3 or MAO

Cl Ni

Fe

CH N Ph

H 43

42

Scheme 5.11

Ph

187

P(1)

C(1)

C(22)

Ni(1)

N(1) C(12)

Cl(1) Fe(1)

C(12) C(16)

P(1)

C(1)

N(1)

C(1A)

Ni(1) Cl(1)

Figure 5.2 Molecular structures of 42 (left) and 43 (right). Reproduced from Z. Weng, et al. Angew Chem Int. Ed., 44, 7560–7564. Copyright (2005), with permission from Wiley-VCH Verlag GmbH.

C(18)

Fe(1)

C(2)

188 Ferrocenes: Ligands, Materials and Biomolecules

The Chemistry of 1 -Functionalised Ferrocene Phosphines

189

(L = 38a) and 44b (L = 38b), respectively.42 Other nickel(0) complexes featuring additional carbonyl, t-BuNC and diphenylethyne supporting donors have been prepared from [Ni(cod)2 ], 38a or 38c, and the appropriate ligand.43 All nickel complexes activated with MAO or EtAlCl2 cocatalysts (at Al/Ni ratio of 1000) were tested as catalysts for ethene oligomerisation under moderate conditions. The catalytic systems produced mostly dimerisation products though with different selectivity and activity (turnover frequency).42, 43 A different, catalytically active agostic Ni−Al complex was isolated from a reaction of 38b, [Ni(cod)2 ], and AlMe3 in hexane at −30 ◦ C.44 Phosphinoferrocenes bearing N-heteroaryl groups in position 1 of the ferrocene unit constitute another prominent class of unsymmetric ferrocene donors. Thus, 1(diphenylphosphino)-1 -(2-pyridyl)ferrocene (45) was obtained by simple reaction of lithoferrocene 6 with excess pyridine, though in a poor yield of about 10 %.45 Starting with 2,2 -bipyridine, 4,4 -bipyridine and 1,10-phenanthroline as the heterocyclic component, the same method gave 1-(diphenylphosphino)-1 -(hetaryl)ferrocenes, where hetaryl is 2,2 -bipyridyl-6-yl (46), 4,4 -bipyridyl-2-yl and 1,10-phenanthrolin-2-yl, respectively, in much better yields (53–83 %). A related compound possessing a stereogenic P-atom, 1-[(butyl)phenylphosphino]-1 -(2,2 -bipyridyl-6-yl)ferrocene, was synthesised in a similar manner from racemic Ph(Bu)PfcLi.46 Reactions of 45 with [{M(µ-Cl)(cod)}2 ] (M = rhodium and iridium) and Ag[PF6 ] as the halide scavenger gave a pair of isomeric but structurally different cationic complexes. Whereas the reaction with iridium(I) precursors yielded the iridium(III) hydride 47 as a product of activation (oxidative addition) of the ferrocene C−H bond proximal to the iridium centre, the rhodium complex was found to be the expected substitution product 48 in which 45 coordinates as a cis-P,N donor (Scheme 5.12). On the basis of X-ray crystallographic analysis which revealed a short Rh· · ·H−C contact ˚ the complex was formulated alternatively as pseudo squarefor 48 (Rh· · ·H 2.39(5) A), pyramidal with the weakly coordinating hydrogen occupying the ﬁfth coordination site. The solution NMR data for 48 suggested slow exchange of the α/α protons at the pyridyl-substituted cyclopentadienyl ring.47 Ph

Ph

PF6

Ph

P ....Rh Fe H N

Ph

PF6

P M = Rh 1/2 [{MCl(cod)}2] + 45 + AgPF6 −AgCl

48

M = Ir −AgCl

H Fe

Ir N

47

Scheme 5.12

The structurally related phosphinobipyridine 46 reacts with [PdCl2 (cod)] to give the stereochemically unique, square-planar palladium(II) complex [PdCl2 (46-κ 2 N ,P )], where the bidentate ligand spans the trans positions (Scheme 5.13). The structure of this complex has been established by X-ray structure determination for the adduct

190

Ferrocenes: Ligands, Materials and Biomolecules Ph N Fe

Ph P

PPh2 [PdCl2(cod)]

Fe

Cl Pd N

Cl N

N 46

Scheme 5.13

[PdCl2 (46-κ 2 N ,P )]•[PdCl2 (cod)] and its formation was followed in situ by cyclic voltammetry and NMR spectroscopy.48

5.5

Phosphines Bearing Chalcogen Donor Sites

There has been reported quite a large number of (diphenylphosphino)ferrocene derivatives bearing an oxygen functional group at the non-phosphinylated cyclopentadienyl ring. However, most of them were regarded only as (potential) synthons while their coordination and catalytic chemistry remains largely unexplored. Prominent examples of such compounds are the aforementioned acetyl derivative Ph2 PfcC(O)Me, alcohols obtained thereof (Ph2 PfcCH(Me)OH and PhPfcC(Ph)Me(OH)),31 and their unsubstituted counterpart Ph2 PfcCH2 OH.49 There is, also, only little known about the coordination properties of aldehyde 25,9c which already proved to be a versatile organometallic synthon (see above). It is readily accessible from step-wise functionalisation of 26 and 39c (route A in Scheme 5.1), or from ferrocenophane 5 (route B in Scheme 5.1).49 The derived cyclic acetals (ketals) Cy2 Pfc(CH(R)OCH2CH2 O) (R = Ph (49a), R = H (49b); Cy = cyclohexyl) and (tBu)2 Pfc(CH2 OCH2 CH2 O) (49c) were recently obtained by Hor et al. by acetalisation50 of the respective acylferrocenes BrfcC(O)R with ethylene glycol and subsequent phosphinylation (LiBu/Ph2 PCl). When combined with palladium precursors, they form highly efﬁcient catalysts for palladium-catalysed Suzuki-Miyaura cross coupling of substituted aryl chlorides with phenylboronic acid. To characterise intermediates involved in the catalysed reaction, the system comprising [Pd2 (dba)3 ], L (49a or 49b), and C6 F5 I as a reactive halide, has been studied. This led to characterisation of the following oxidative addition products: [{Pd(µ-I)(C6 F5 )(L-κP )}2 ], [PdI(C6 F5 )(L-κ 2 O, P )2 ], cis- and trans-[PdI(C6 F5 )(L-κP )2 ].51 Similarly to amine 27, the simplest hydroxy derivative 1-(diphenylphosphino)-1 hydroxyferrocene (50) is rather unstable52 and has been prepared only recently by the reaction of 6 with bis(trimethylsilyl)peroxide and hydrolysis of the silyloxy intermediate (Scheme 5.14). Compound 50 was deprotonated at the hydroxyl group with sodium hydride and the intermediate alkoxide was reacted with trans-[NiCl(Ph)(PPh3 )2 ] to give diamagnetic chelate complex [Ni(Ph2 PfcO-κ 2 O,P )(Ph)(PPh3 )] with the phosphorus groups presumably in trans positions.53 In the synthesis of the respective alkoxy compound Ph2 PfcOMe (51), 1,1 -dibromoferrocene (3) was ﬁrst converted to acetyloxy ferrocene 52 by reacting with

The Chemistry of 1 -Functionalised Ferrocene Phosphines PPh2 5

1. LiBu 2. (Me3SiO)2

191

PPh2 H2O

Fe

Fe

OSiMe3

OH 50

Scheme 5.14

stoichiometric amounts of acetic acid and copper(I) oxide in acetonitrile, and then alkylated to give 53. A better-yielding, one-pot procedure for the preparation of methoxide 53 directly from 3, relating to the synthesis of 50 was also devised. The standard lithiation and phosphinylation afforded phosphinoether 51 (Scheme 5.15).54 Br 3

AcOH/MeCN Cu2O (cat.) 13 %

Fe

Br NaH/MeI

15-crown-5 77 % OAc

52

Fe

53

PPh2 1. LiBu

2. Ph2PCl 40 % OMe

Fe OMe 51

1. LiBu, 2. (Me3SiO)2, 3. Me2SO4/K2CO3 56 %

Scheme 5.15

Coordination ability of methoxy-phosphine 51 was probed in a series of copper, rhodium and Group 10 metal complexes. The compound coordinates both in chelate ([NiBr2 (51-κ 2 O,P )], [PdCl2 (51-κ 2 O,P )]) and P-unidentate fashion ([PtCl2 (51-κP )2 ], trans-[RhCl(CO)(51-κP )2 ], and [Cu(MeCN-κN )(51-κP )2][PF6 ]) (Figure 5.3). When combined with palladium(II) acetate or [Pd2 (dba)3 ]•C6 H6 , 51 gives an efﬁcient catalytic system for Suzuki–Miyaura reaction of phenylboronic acid with 4-bromotoluene.54 Another donor belonging to the class of 1,1 -unsymmetric ferrocene P,O-donors is 1-(diphenylphosphino)-1 -ferrocenecarboxylic acid (Ph2 PfcCO2 H or Hdpf).55 Originally, it had been prepared by carboxylation of in situ formed 6 and acidiﬁcation of the formed lithium salt (cf. Scheme 5.1, route B).27 Complementary approaches relying on step-wise functionalisation of 256 and 39c were published later. As a ligand, Hdpf has been studied quite thoroughly. With soft metals (palladium(II),57, 58, 59 platinum(II),57 nickel(II),60 mercury(II),61 copper(I),62 rhodium(I),63 ruthenium(II),64 and Group 6 carbonyls65 ), it usually binds as a P-monodentate donor whilst the uncoordinated carboxyl group takes part in hydrogen bonding to carboxyl group from a proximal ligand moiety or to solvent molecules, thus forming various supramolecular assemblies. Being a redox-active ligand, Hdpf has been used as a redox probe at the molecular level.66 Of particular interest is an extensive series of (η6 -arene)Ru complexes (52) and carbenes synthesised thereof (53, Scheme 5.16), which combine

Cl

Fe′

P′

0′

C(26)

C(25)

C(32)

C(33)

C(27)

C(40)

C(31)

Fe(2)

C(30)

C(34)

O(29)

Cu

C(23)

C(29)

C(24) C(28)

P(2)

C(46)

C(5)

Fe(1)

C(52)

C(4)

C(9) C(10)

C(51)

C(2)

N(50)

C(1)

P(1)

C(17)

C(7)

C(11)

C(6)

C(6)

C(3)

C(8)

Figure 5.3 Views of the molecular structure of [PtCl2 (51-κ P )2 ] (left) and of the cation in the structure of [Cu(MeCN-κ N )(51-κ P )2 ][PF6 ] (right). Reprinted with permission from Ref. 54. Copyright 2004, American Chemical Society.

C(13)

C(12)

C(17)

P

Cl′

Pt

C(7)

C(6)

C(4)

Fe

C(8)

C(5)

C(14)

C(16)

C(23)

C(3)

C(15)

C(19) C(18)

C(20)

C(21)

0

C(22)

C(1)

C(2)

C(11) C(10)

C(9)

192 Ferrocenes: Ligands, Materials and Biomolecules

The Chemistry of 1 -Functionalised Ferrocene Phosphines

193

alkyne = FcC CH (a) PhC CH (b) Me3SiC CH (c) Ru Cl

Cl

Cl Ru

Cl

L

Ru

Cl

L

alkyne/MeOH/NaPF6 −NaCl

Ru L

Cl

Cl

52a-d

53

L = PMe3 (a), PPh3 (b) FcPPh2 (c), Hdpf (d)

L/R′ = PMe3/Fc

OMe PF 6 R’

(aa)

PPh3/Fc (ba)

FcPPh2/Fc (ca)

Hdpf/Fc (da)

FcPPh2/H (cb)

Hdpf/H (db)

FcPPh2/Ph (cc)

Hdpf/Ph (dc)

Scheme 5.16

several redox centres (Ru(II)/Ru(III) with one or two chemically different ferrocene units). The electrochemical study showed that primary ferrocene/ferrocenium oxidation occurring at the ruthenium-bonded ferrocene phosphine (e.g. in 52c,d) inﬂuences the following Ru(II) → Ru(III) oxidation (i.e. the redox properties of the arene–ruthenium moiety), indicating redox coupling of these parts. In contrast, ferrocenyl units in the carbene moiety appear isolated from the rest of molecule, which corresponds with the presence of the methylene spacer.64 Simple salts of carboxylate dpf− are mostly ill-deﬁned materials due to their reluctance to crystallise and strong tendency to hold reaction solvents. Hence, when required, they are best prepared in situ.27, 67 A well-deﬁned, paramagnetic carboxylate complex [Ti(dpf-κ 2 O,O ) (η5 -C5 HMe4 )2 ] (54) has been obtained from the reaction between Hdpf and the octamethyltitanocene precursor [Ti(η5 -C5 HMe4 )2 (η2 -Me3 SiC≡CSiMe3 )] and has been characterised by X-ray diffraction (Figure 5.4).68 Finally, a coordination mode involving both donor sites of Hdpf has been achieved in a series of square-planar rhodium(I) complexes featuring O,P-chelate coordinated phosphinocarboxylate anion dpf− . These compounds resulted from complexes 55a–c via acid–base reaction of acetylacetonate with Hdpf (Scheme 5.17, Figure 5.4). The direct reaction between [Rh(acac)(CO)2] (Hacac = acetylacetone), which is the precursor to 55, and Hdpf gave an interesting complex, where P-monodentate acid and O,P-chelating carboxylate coexist (57 in Scheme 5.17). Spectral data indicated that both forms of the phosphinocarboxylic donor exchange the acidic proton which, in turn, points to hemilabile coordination.63 Complexes 56a,b and 57 (as such or with phosphine or phosphite cocatalysts) as well as catalytic system formed in situ from various rhodium(I) precursors and Hdpf were tested in hydroformylation of 1-hexene. Compound 57 proved to be particularly attractive catalyst precursor, producing heptanals in both good yield and with favourable (n/iso) ratio. Moreover, it could be used repeatedly.69 Various Hdpf-based catalytic systems (in the form of deﬁned complexes as well as catalyst formed in situ) were successfully employed in palladium-catalysed Suzuki–Miyaura

Figure 5.4 The molecular structures of 54 (left) (Reproduced from K. Mach, et al. Acta Crystallogr., Sect. E, Cryst. Struct. Commun. 58, ˇ epniˇcka, m116–m118. Copyright (2002), with permission from International Union of Crystallography.) and 55b (right) (Reproduced from P. Stˇ J. Chem. Soc., Dalton Trans., 2807–2811. Copyright (1998), with permission from Royal Society of Chemistry.)

194 Ferrocenes: Ligands, Materials and Biomolecules

The Chemistry of 1 -Functionalised Ferrocene Phosphines Ph

Ph

195

CO

P Rh Fe

2 Hdpf −Hacac −CO

Ph P

O

Ph

C

[Rh(acac)(CO)2] −CO

PR3

Fe

O

[Rh(acac)(L)(CO)]

HO2C

55a-c L = PPh3 (a) PCy3 (b) PPh2 (c)

57

−Hacac Hdpf

Ph

Ph P

CO

Rh Fe O

L

C O 56a-c

Scheme 5.17

reaction of 4-bromotoluene and phenylboronic acid.58 Hdpf was also used as a modiﬁer to MCM-41-supported ruthenium catalysts promoting addition of benzoic acid to propargyl alcohol to give 2-oxopropyl benzoate.70 An O,P donor combining phosphine and phosphonate moieties, Ph2 PfcP(O)(OEt)2 (58), was recently prepared from 4 via lithiation and treatment with ClP(O)(OEt)2 (i.e. via route A in Scheme 5.1).71, 72 It has been further converted to asymmetric bis(phosphane) Ph2 PfcPH2 and to P-chiral diphosphine hydrogenation catalysts.71 Its coordination properties were studied in a series of palladium(II) and Group 12 metal complexes. Depending on the reaction stoichiometry, the reaction of 58 with [PdCl2 (cod)] gave the halidebridged dimer [{Pd(µ-Cl)Cl(58-κP 2)}2 ] or bis-phosphine complex [PdCl2 (58-κP 2 )2 ]. The reactions with Group 12 metal dibromides gave uniformly the ‘adducts’ [MBr2 (58)] (M = zinc, cadmium and mercury) that possess distinctly different solid state structures. Whereas the zinc complex [ZnBr2 (58-κ 2 O 1 ,P 2 )2 ] (59) is the molecular chelate, its cadmium congener 60 was found to be a coordination polymer involving bromide and 58 as bridging ligands (Scheme 5.18). Finally, the HgBr2 -58 system showed structural ﬂuxionality, allowing for isolation of the isomeric complexes [{Hg(µ-Br)Br(58-κP 2)}2 ] and [(µ-58-1κO 1 :2κP 2 )(µ-58-1κP 2 :2κO 1 ){HgBr2 }2 ].72 In comparison with their P,O counterparts, the unsymmetric ferrocene P,S and P,Se donors are still quite uncommon, being restricted to Ph2 PfcER compounds, where E = sulfur and selenium, and R = hydrogen or a hydrocarbyl group (for P/SO donors see below). The sulfur analogue of 50, phosphinothiol 61, was obtained in a modest

196

Ferrocenes: Ligands, Materials and Biomolecules Ph

Ph P

Br Zn Br

Fe O P

OEt OEt

59

Fe Fe

Br Br

Ph P O P OEt Ph EtO OEt Ph EtO O P P Ph

P

Br Cd

Cd Br Br

O EtO

Ph OEt

P

Ph Ph EtO OEt O P P Ph

Br Cd

Cd Br Br

Fe Fe

x 60

Scheme 5.18

yield from 62 via reductive cleavage of the disulﬁde bridge and reaction with potassium diphenylphosphide (Scheme 5.19). The key intermediate, disulﬁde 62, was synthesised from 3 by metalation with LiBu (1 equiv.) and subsequent treatment with S2 Cl2 (0.5 equiv.). Thiolate 61 readily underwent deprotonation at the thiolate group (with LiBu or as a result of acid–base reactions with metal precursors) to give mononuclear nickel(II) (63) and binuclear, sulfur-bridged palladium(II) and rhodium(I) complexes (64 and 65; cf. the behaviour of 50).73 The apparently more stable phenylthio derivative Ph2 PfcSPh (66) was reported in 1992. Its was obtained from step-wise functionalisation of 2 (Scheme 5.1, route A: E1 X = Ph2 PCl, ‘E2 X’ = PhSSPh) and has been structurally characterised.74 The same compound and (i-Pr)2 PfcSPh (67) were obtained similarly from 3 (via BrfcSPh).9d Later, the original synthetic route was used to prepare 6675, 76 and its methylthio analogue, Ph2 PfcSMe (68)75 from 5 (see Scheme 5.1, route B: 1. LiPh, 2. MeSSMe; yield 18 %). A better yield of 68 has been achieved by sequential addition of Ph2 PCl and MeSSMe to in situ formed 2 (i.e. in one pot).75 Direct addition of an equimolar Ph2 PCl/RSSR mixture to 2 has been shown to further increase the yield of Ph2 PfcSR to about 40 % while making the whole procedure even simpler. This has been demonstrated by the synthesis of 68 and Ph2 PfcSMes (69, Mes = mesityl).77 Donor 68 reacted with [M(MeCN)4 ]X under complete replacement of the acetonitrile ligands to give the cationic bis-chelate complexes [M(68-κ 2 P ,S)2 ]X (M/X = Cu/PF6

The Chemistry of 1 -Functionalised Ferrocene Phosphines

197

3 1. LiBu

1. LiBu 2. 1/2 S2Cl2

2. [PdCl2(PhCN)2] (for 64) or [{RhCl(CO)2}2] (for 65)

S

PPh2 1. Li[BHEt3]

Fe Br

2. KPPh2, then H+

P

Fe SH

2 61

62

Ph

Ph

L M

Fe

S

S

Ph

Ph

M Me

P

[NiMe2(TMEDA)]

Ni

Fe S

L

P Ph

NCMe

Fe

Ph

64 (M = Pd, L = Cl) 65 (M = Rh, L = CO)

63

Scheme 5.19

and Ag/BF4 ).75 P,S-Chelate complexes were also obtained with Group 10 metal precursors ([MX2 (68-κ 2 P ,S)], where M/X = Ni/Br, Pd/Cl and Pt/Cl) and with [{Rh(µCl)(CO)2 }2 ] ([RhCl(CO)(68-κ 2 P ,S)]). Increasing the ligand-to-metal ratio led to the [PdCl2 (68-κ 2 P )2 ] complex featuring P-monodentate 68, while the reaction of 68 with photo-generated [W(CO)5 (THF)] (THF = tetrahydrofuran) afforded a pair of tungsten–carbonyl complexes, differing in the coordination of the phosphinosulﬁde ligand: [W(CO)4 (68-κ 2 P ,S)] and [W(CO)5 (68-κP )]. The complexes were studied electrochemically and the solid state structures of several representatives have been determined by X-ray diffraction. In addition, 68 was tested as a ligand in palladiummediated Suzuki–Miyaura cross-coupling and showed good activity in the reaction of 4-bromotoluene with phenylboronic acid.77 Only recently, compound 66 and its heavier congener, Ph2 PfcSePh (70; prepared from 5 via route B in Scheme 5.1: 1. LiPh, 2. PhSeSePh), have been as ligands for Group 11 complexes.76 With an extensive series of isolated and structurally characterised complexes with 66 and 70,78 it has been demonstrated that gold(I), as the softest metal in the group, tends to accommodate the ligands as P-monodentate donors, while copper(I) and silver(I) display a clear tendency to form P,E-chelated complexes (66: E = sulfur; 70: E = selenium). It has been also shown that gold(I) complexes with P-bonded ligands are good precursors for the synthesis of bridged heteromultinuclear complexes by means of the uncoordinated phenylchalcogenate groups. Complexes [Au(µ-1κP :2κS-66)Ag](ClO4 )2 prepared from [Au(66-κP )2 ]ClO4 and silver(I)perchlorate, and [(ClAuPh2 PfcSePh-κSe)2PdCl2 ] obtained from [AuCl(70-κP )] and half molar equivalent of [PdCl2 (MeCN)2 ] have been studied as representative examples.76

198

Ferrocenes: Ligands, Materials and Biomolecules

5.6

Donors Bearing Composite and Potentially Multidentate Groups

Obviously the most thoroughly studied ferrocene phosphines modiﬁed with composite functional groups are represented by C-chiral 2-[1 -(diphenylphosphino)ferrocenyl]4,5-dihydrooxazoles (or oxazolines79 ) 71. Chirality of these donors originates from β-aminoalcohols, which are available from natural α-amino acids. Consequently, they are mostly (S)-conﬁgured at the stereogenic centre and typically have R = i-Pr, t-Bu, or Ph. Three approaches towards 71 were reported independently in the literature. Albeit that the synthetic methodology is rather straightforward, the synthetic routes deserve a comment as they are complementary and demonstrate the synthetic versatility of some ferrocene precursors. The method devised by Ikeda et al.56 (A in Scheme 5.20; R = (S)-i-Pr, (S)-t-Bu, and (S)-Ph) starts with 2, which is ﬁrst converted via Hdpf to Hdpf pentaﬂuorophenyl ester 72 or directly to Hdpf methyl ester 73 (cf. Scheme 5.1). Both esters are then amidated with the appropriate β-aminoalcohols to give amides 74 which undergo ring closure leading to the desired oxazolines. Alternatively, the same series of phosphinooxazolines was prepared by Park and coworkers80 (B in Scheme 5.20) from acid 75 (available from 3, cf. Scheme 5.1), which was ﬁrst amidated to give 76 and subsequently cyclised to (1-bromoferrocenyl)oxazolines 77. The synthesis of 71 is completed by the standard phosphinylation. Finally, the route used by Dvoˇra´ k et al.81 (C in Scheme 5.20) resembles that of Ikeda as it makes use of Hdpf27 (Scheme 5.20). However, Hdpf was ﬁrst protected at the phosphorus by oxidation to the corresponding phosphine oxide (HdpfO) and then directly amidated to 78. Standard closure of the oxazoline ring (→ 79) followed by reduction of the phosphinoyl moiety then gives oxazolines 71 (R = (S)-i-Pr, (R)-i-Pr, (S)-t-Bu, and (S)-Ph). When combined with (allyl)palladium(II) precursors, the phosphinoxazolines afford highly efﬁcient catalytic systems for palladium-mediated allylic alkylation of rac1,3-diphenylprop-2-en-1-yl acetate (80) with deprotonated malonate esters (yields and enantioselectivities up to 99 %).56, 80, 81, 82 Attempts have been made to clarify the reaction course and characterise intermediates involved in the catalysed reactions by means of in situ NMR studies and reactions with other palladium(II) compounds.56, 80, 81 Furthermore, oxazolines 71 were tested in asymmetric Heck reaction of 2,3-dihydrofuran with phenyl triﬂate83 and were also used in the preparation of 1,1 -bis(phosphino)-2oxazolinylferrocenes.84 Other phosphinoferrocene ligands bearing chiral functional groups in position 1 are represented by S-chiral phosphinoferrocene sulfoxides 81 (Scheme 5.21; R = 4XC6 H4 , X = Me, MeO, and Cl; 1-naphthyl and 2,4,6-triisopropylphenyl) and C-chiral phosphinohydrazones 82 (Scheme 5.21; R = hydrogen, Me, and Et). Whereas the sulfoxides were obtained in good yields from 2 (via route A in Scheme 5.1),85 the latter compounds featuring potential (PNNO) donor arrays were synthesised by acidcatalyzed condensation of 25 with the appropriate 1-amino-2-(R2 -methoxymethyl) pyrrolidines.86 Both ligand series were tested in palladium-catalysed asymmetric allylic substitution with malonate esters of rac-cyclohex-2-en-1-yl acetate (sulfoxides)85 or 80 (hydrazones),86 showing moderate to very good yields and enantioselectivities.

The Chemistry of 1 -Functionalised Ferrocene Phosphines 2

i.

iv.

PPh2

PPh2

Br

Fe

Fe

Br ix.

Fe N

CO2H

CO2Me 73

Hdpf ii.

PPh2 iii.

Fe

Fe

H N

76

x.

B

PPh2

PPh2 vi.

74

71

R

CO2H 75

O

vii.

xv.

C

5 xi. Hdpf

xii.

HdpfO

xiii.

O

O

PPh2

PPh2

Fe

78

H N

xiv. R

3

Fe N

OH

O

OH

Fe N

OH

O

O

R

Br

Fe

R

H N

viii.

CO2C6F5 72

Fe

R

O

77

v.

A

199

79

R

O

Legend: i. (a) LiBu, (b) Ph2PCl, (c) LiBu, (d) CO2, then H; ii. C6F5OH/N,N′-dicyclohexylcarbodiimide; iii. aminoalcohol/NEt3/Me2NCHO (solvent); iv. LiBu, (b) Ph2PCl, (c) LiBu, (d) ClCO2Me; v. aminoalcohol/Na; vi. MeSO2Cl/NEt3; vii. (a) LiBu, (b) CO2, then H+; viii. (a) PCl5, (b) aminoalcohol/NEt3; ix. 4-MeC6H4SO2Cl/4(dimethylamino)pyridine/NEt3; x.(a) LiBu, (b) Ph2PCl; xi. (a) LiPh, (b) CO2, then H+; xii. H2O2; xiii. (a) (COCl)2, (b) aminoalcohol/NEt3; xiv. 4-MeC6H4SO2Cl/NEt3; xv. SiHCl3/NEt3.

Scheme 5.20 A summary of synthetic routes available for the synthesis of phosphinoferrocene oxazolines 71

PPh2 Fe

Fe

Fe

O

N C N H

S(O)R 81

PPh2

PPh2

82

MeO

N

C R

R

HN

( )n

83a (n = 1) 83b (n = 2)

Scheme 5.21

200

Ferrocenes: Ligands, Materials and Biomolecules

Two homologous Hdpf amides bearing N -(2-pyridyl)alkyl substituents 83 represent further examples of potentially multidentate ferrocene donors. They were prepared by amidation of Hdpf with the appropriate 2-(ω-aminoalkyl)pyridine and were tested as ligands for palladium and in palladium-catalysed Suzuki–Miyaura reactions. Depending on the reaction stoichiometry, the phosphinoamides react with [PdCl2 (cod)] to give bis-phosphine and monophosphine complexes, [PdCl2 (L-κP )2 ] and ‘[PdCl2 (L)]’ (L = 83a and 83b). The former compounds posses the expected trans-planar coordination geometry. In the latter, the ligands coordinate as P,N donors forming a molecular chelate with trans-spanning P,N-ligand [PdCl2 (83a-κ 2 N ,P )] and a ligandbridged dimer [{PdCl2 (µ(N ,P )-83b)}2 ], respectively (cf. the behaviour of 46).87 S

S

PPh2 25

S8

PPh2 NaBH4

Fe

Fe

CHO

CH2OH

84

85 N

N R

S BF4

PPh2 R N

Fe

Raney Ni

R N

Fe N

87a-b

86a-b R = Me (a) R = Mes (b)

For 87a 1. t-BuOK, [{RhCl(LL}2] 2. NaBF4

Ph

BF4

PPh2

N

Ph P

L Rh

Fe N

BF4 L N

Me

88a-b L−L = 1,5-cyclooctadiene (a) L−L = norbornadiene (b)

Scheme 5.22

/HBH4

The Chemistry of 1 -Functionalised Ferrocene Phosphines

201

A rather speciﬁc example of mixed donor ferrocene donors has been recently offered by Manoury et al. who synthesised phosphinoferrocene imidazolium salts (87) as precursors of phosphinoferrocenyl-substituted N-heterocyclic carbenes. The synthetic approach is illustrated in Scheme 5.22. The synthesis starts with phosphinoaldehyde 25, which is ﬁrst protected at the phosphorus group (→ 84)88 and reduced to alcohol (→ 85). Subsequent acid-promoted substitution reaction affords the imidazolium phosphine sulﬁdes 86 which, after deprotection, gave the phosphine imidazolium derivatives 87. Salts 87 react with rhodium(I)-precursors in the usual manner, giving phosphinotethered carbene complexes 88.89

5.7

Summary and Outlook

The chemistry of functionalised phosphines derived for 1,1 -disubstituted ferrocene framework has developed to a wide and interesting ﬁeld. Both the absolute number and diversity of the known representatives indicate that these compounds constitute a highly variable class of organometallic ligands that hold great potential for the preparation of unique coordination compounds and for catalysis. However, there is still enough space left for future developments, the area of catalysis being particularly promising ﬁeld. The compounds have already proved to be valuable catalyst components for achiral, most often C−C bond forming reactions and, when equipped with chiral substitutents, also for enantioselective transformations. More applications can be expected in the near future. A great advantage of functionalised phosphines derived from 1,1 -ferrocene scaffold can be seen in the well-developed synthetic methodology, relative stability and great variability. They allow not only the preparation of new ligand classes but also the creation of wide ligand libraries via rather independent variations at both the phosphine and the functional moieties. The latter aspect is particularly important for tailoring the catalytic systems properties for a speciﬁc reaction-substrate combination. Besides, ferrocene ligands usually represent deﬁned redox systems than can be utilised as electrochemical probes at the molecular level or tools for tuning the complex (catalyst) properties via changing the oxidation state of the iron centre.

References 1. For clarity, the phosphine part is assumed to occupy the 1 position albeit this does not always comply with the priority rules. 2. Chemistry of 1 -functionalised ferrocene phosphines has not been reviewed separately but was a part of some previous review articles. See, for instance: (a) T.J. Colacot, Chem. Rev. 2003, 103, 3101–3118; (b) R.C.J. Atkinson, V.C. Gibson, N.J. Long, Chem. Soc. Rev. 2004, 33, 313–328; (c) P. Barbaro, C. Bianchini, G. Giambastiani, S.L. Parisel, Coord. Chem. Rev. 2004, 248, 2131–2150. 3. This term was introduced in 1979 by Jeffrey and Rauchfuss: J.C. Jeffrey, T.B. Rauchfuss, Inorg. Chem. 1979, 18, 2658–2666.

202

Ferrocenes: Ligands, Materials and Biomolecules

4. (a) A. Bader, E. Lindner, Coord. Chem. Rev. 1991, 108, 27–110; (b) C.S. Slone, D.A. Weinberger, C.A. Mirkin, Prog. Inorg. Chem. 1999, 48, 233–350; (c) P. Braunstein, F. Naud, Angew. Chem. Int. Ed. 2001, 40, 680–699. 5. Notation according to the Pearson’s HSAB concept; for reference, see: R.G. Pearson, J. Am. Chem. Soc. 1963, 85, 3533–3539. 6. M.E. Wright, Organometallics 1990, 9, 853–856. 7. (a) M.D. Rausch, D.J. Ciappenelli, J. Organomet. Chem. 1967, 10, 127–136. See also: (b) M.D. Rausch, G.A. Moser, C.F. Meade, J. Organomet. Chem. 1973, 51, 1–11; (c) J.J. Bishop, A. Davison, M.L. Katcher et al. J. Organomet. Chem. 1971, 27, 241–249. 8. T. Dodo, H. Suzuki, T. Takiguchi, Bull. Chem. Soc. Jpn. 1970, 43, 288–290. 9. (a) L-L. Lai, T-Y. Dong, Chem. Commun. 1994, 2347–2348; (b) T-Y. Dong, L-L. Lai, J. Organomet. Chem. 1996, 509, 131–134; (c) I.R. Butler, R.L. Davies, Synthesis 1996, 1350–1354; (d) T-Y. Dong, C-K. Chang, J. Chin. Chem. Soc. 1998, 45, 577–579; (e) T-Y. Dong, P-H. Ho, C-K. Chang, J. Chin. Chem. Soc. 2000, 47, 421–424. 10. R.F. Kovar, M.D. Rausch, H. Rosengberg, Organomet. Chem. Synth. 1970–1971, 1, 173–181. 11. Bromination of lithioferrocenes can be performed also with 1,2-C2 Br2 Cl4 though with lower yields. Representative examples: (a) L. Xiao, K. Mereiter, W. Weissensteiner, M. Wildham, Synthesis 1998, 1354–1362; (b) A.J. Locke, T.E. Pickett, C.J. Richards, Synlett 2001, 141–143; (c) K. Tappe, P. Knochel, Tetrahedron: Asymmetry 2004, 15, 91–102. 12. A. Shaﬁr, M.P. Power, G.D. Whitener, J. Arnold, Organometallics 2000, 19, 3978–3982. 13. It was shown that the C−Si bonds in fc(SiMe3 )2 can be cleaved with mercury(II) chloride to give Me3 SifcHgCl (51 %) and fc(HgCl)2 .13a Me3 SifcHgCl reacts with copper(II) salts CuX2 and iodine to give Me3 SifcX (X = Cl, Br, SCN, SO2 Ph; and I), and can be transmetalated with butyllithium to give Me3 SifcLi, which undergoes standard reactions with electrophiles.13b (a) G. Marr, T.M. White, J. Chem. Soc. C 1970, 1789–1792; (b) G. Marr, T.M. White, J. Organomet. Chem. 1971, 30, 97–101. 14. M. Herbehold, Angew. Chem. Int. Ed. Engl. 1995, 34, 1837–1839 (a review). 15. N.B. The tilt of the cyclopentadienyl rings in 4 as determined by X-ray crystallography is as high as 26.7◦ : H. Stoeckli-Evans, A.G. Osborne, R.H. Whiteley, J. Organomet. Chem. 1980, 194, 91–101. 16. (a) D. Seyferth, H.P. Withers, Jr., J. Organomet. Chem. 1980, 185, C1–C5; (b) D. Seyferth, H.P. Withers, Jr., Organometallics 1982, 1, 1275–1282. 17. A.G. Osborne, R.H. Whiteley, R.E. Meads, J. Organomet. Chem. 1980, 193, 345–357. 18. I.R. Butler, W.R. Cullen, F.W.B. Einstein et al. Organometallics 1983, 2, 128–135. 19. G. Iftime, C. Moreau-Bossuet, E. Manoury, G.G.A. Ballavoine, Chem. Commun. 1996, 527–528. 20. J.M. Chong, L.S. Hegedus, Organometallics 2004, 23, 1010–1014. 21. Metal-bonded [R2 Pfc] and related motifs were identiﬁed in products resulting from thermolysis of some mixed-donor carbonyl complexes with phosphinoferrocene ligands: (a) W.R. Cullen, S.J. Retting, T-C. Zheng, Organometallics 1992, 11, 277–283. See also: (b) M.I. Bruce, P.A. Humprey, O. bin Shawkataly et al. Organometallics 1990, 9, 2910–2919; (c) W.R. Cullen, S.J. Rettig, T. Zheng, Organometallics 1992, 11, 853–858. 22. I.R. Butler, W.R. Cullen, Organometallics 1984, 3, 1846–1851. 23. E.O. Fischer, A. Maasb¨ol, Angew. Chem. 1964, 76, 645; (b) Transition Metal Carbene Complexes, Verlag Chemie, Weinheim , Germany (1983). 24. I.R. Butler, W.R Cullen, F.W.B. Einsten, A.C. Willis, Organometallics 1985, 4, 603–604. 25. L. Meca, D. Dvoˇra´ k, J. Ludv´ık et al. Organometallics 2004, 23, 2541–2551. 26. (a) R. Imwinkelried, L.S. Hegedus, Organometallics 1988, 7, 702–706; (b) M.A. Schwindt, T. Lejon, L.S. Hegedus, Organometallics 1990, 9, 2814–2819.

The Chemistry of 1 -Functionalised Ferrocene Phosphines

203

ˇ epniˇcka, J. Ludv´ık, I. C´ısaˇrov´a, Organometallics 1996, 15, 543–550. 27. J. Podlaha, P. Stˇ 28. D. Dvoˇra´ k, Organometallics 1995, 14, 570–573. 29. Vinylferrocenes are readily formed by dehydration of (1-hydroxyethyl)ferrocenes and their esters. For representative examples, see: (a) K. Schl¨ogl, A. Mohar, Naturwissenchaften 1961, 48, 376–377; (b) F.S. Arimoto, A.C. Haven, Jr., J. Am. Chem. Soc. 1955, 77, 6295–6297; (c) M.D. Rausch, A. Siegel, J. Organomet. Chem. 1968, 11, 317–324; (d) J.K. Stille, H. Su, D.H. Hill et al. Organometallics 1991, 10, 1993–2000; (e) H. C. L. Abbenhuis, U. Burckhardt, V. Gramlich et al. Organometallics 1994, 13, 4481–4493; (f) M.E. Kuehne, U.K. Bandarage, J. Org. Chem. 1996, 61, 1175–1179; (g) C.R. Landis, R.A. Sawyer, E. Somsook, Organometallics 2000, 19, 994–1002. 30. 1 -(Diphenylphosphino)-1-(1-phenylethen-1-yl)ferrocene was obtained similarly from Ph2 PfcCPh(OH)Me (Ref.31 ). 31. I.R. Butler, W.R. Cullen, Can. J. Chem. 1983, 61, 147–153. 32. Reaction between in situ formed 5 (from 4 and excess LiPh) and MeC(O)Cl gave a mixture of 23 (39 % after chromatography and crystallisation), 24, Ph2 PfcC(Ph)=CH2 , and (diphenylphosphino)ferrocene. 33. I.R. Butler, S.J. Coles, M.B. Hursthouse, et al. Inorg. Chem. Commun. 2003, 6, 760–762. ˇ epniˇcka, I. C´ısaˇrov´a, Collect. Czech. Chem. Commun. 2006, 71, 215–236. 34. P. Stˇ ˇ epniˇcka, J. Organomet. Chem. in print (doi: 10.1016/j.jorganchem.2007.10.056). 35. P. Stˇ ˇ epniˇcka, I. C´isaˇrov´a, Inorg. Chem. 2006, 45, 8785–8798. (b) P. Stˇ ˇ epniˇcka, 36. (a) P. Stˇ M. Lamaˇc, I. C´isaˇrov´a, J. Organomet. Chem. in print (doi: 10.1016/j.jorganchem. 2007.11.016). 37. It can be noted that a related P-protected ferrocene phosphinoalkyne Ph2 P(S)fcC≡CH has been prepared recently by Corey–Fuchs alkynylation of Ph2 P(S)fcCHO and further reacted with [Co2 (CO)8 ] to give [(µ-η2 :η2 -Ph2 P(S)fcC≡CH){Co(CO)3 }2 ](Co−Co) and ˇ epniˇcka, with K2 [HgI4 ]/KOH to give the acetylide complex [Hg{C≡CfcP(S)Ph2 }2 ]: P. Stˇ I. C´ısaˇrov´a, J. Organomet. Chem. 2006, 691, 2863–2871. 38. I.R. Butler, S.C. Quayle, J. Organomet. Chem. 1998, 552, 63–68. 39. M. Widhalm, U. Nettekoven, K. Mereiter, Tetrahedron: Asymmetry 1999, 10, 4369–4391. 40. N.B. Compound 38 results as a minor by-product from lithiation/phosphinylation of N[(ferrocenyl)methyl]-3,5-dihydro-4H -dinaphth[2,1-c:1,2 -e]azepine. 41. Z. Weng, S. Teo, L.L. Koh, T.S.A. Hor, Organometallics 2004, 23, 4342–4345. 42. Z. Weng, S. Teo, L.L. Koh, T.S.A. Hor, Angew. Chem. Int. Ed. 2005, 44, 7560–7564. 43. Z. Weng, S. Teo, T.S. Andy Hor, Organometallics 2006, 25, 4878–4882. 44. Z. Weng, S. Teo, L.L. Koh, T.S.A. Hor, Chem. Commun. 2006, 1319–1321. 45. I.R. Butler, Organometallics 1992, 11, 74–83. 46. I.R. Butler, Can. J. Chem. 1992, 11, 3117–3121. 47. I.R. Butler, M. Kalaji, L. Nehrlich, et al. J. Chem. Soc., Chem. Commun. 1995, 459–460. 48. T. Yoshida, K. Tani, T. Yamagata, et al. J. Chem. Soc., Chem. Commun. 1990, 292–294. ˇ epniˇcka, T. Baˇse, Inorg. Chem. Commun. 2001, 4, 682–687. 49. P. Stˇ 50. R.D. Moulton, A.J. Bard, Organometallics 1988, 7, 351–357. 51. S. Teo, Z. Weng, T.S.A. Hor, Organometallics 2006, 25, 1199–1205. 52. See, for instance: M. Herbehold, H-D. Brendel, A. Hofmann et al. J. Organomet. Chem. 1988, 556, 173–187 and references cited therein. 53. R.C.J. Atkinson, V.C. Gibson, N.J. Long et al. Dalton Trans. 2004, 1823–1826. 54. R.C.J. Atkinson, V.C. Gibson, N.J. Long et al. Organometallics 2004, 23, 2744–2751. ˇ epniˇcka, Eur. J. Inorg. Chem. 2005, 3787–3803 (review). 55. P. Stˇ 56. W. Zhang, Y. Yoneda, T. Kida et al. Tetrahedron: Asymmetry 1998, 9, 3371–3380. ˇ epniˇcka, J. Podlaha, R. Gyepes, M. Pol´asˇek, J. Organomet. Chem. 1998, 552, 293–301. 57. P. Stˇ ˇ epniˇcka, M. Lamaˇc, I. C´ısaˇrov´a, Polyhedron 2004, 23, 921–928. 58. P. Stˇ

204 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78.

79. 80. 81. 82. 83. 84.

85. 86. 87. 88. 89.

Ferrocenes: Ligands, Materials and Biomolecules ˇ epniˇcka, I. C´ısaˇrov´a, Collect. Czech. Chem. Commun. 2006, 71, 279–293. P. Stˇ ˇ J. Pinkas, Z. Bastl, M. Slouf et al. New J. Chem. 2001, 25, 1215–1220. ˇ epniˇcka, I. C´ısaˇrov´a, J. Podlaha et al. J. Organomet. Chem. 1999, 582, 319–327. P. Stˇ ˇ epniˇcka, R. Gyepes, J. Podlaha, Collect. Czech. Chem. Commun. 1998, 63, 64–74. P. Stˇ ˇ epniˇcka, I. C´ısaˇrov´a, Dalton Trans. 1998, 2807–2811. P. Stˇ ˇ epniˇcka, R. Gyepes, O. Lavastre, P.H. Dixneuf, Organometallics 1997, 16, 5089–5095. P. Stˇ ˇ epniˇcka, Collect. Czech. Chem. Commuun. 2000, L. Lukeˇsov´a, J. Ludv´ık, I. C´ısaˇrov´a, P. Stˇ 65, 1897–1910. ˇ epniˇcka, ECS Trans. 2007, 2, 17–25. See also Refs. 61, 64, 65. J. Ludv´ık, P. Stˇ ˇ epniˇcka, J. Podlaha, Inorg. Chem. Commun. 1998, 1, (a) Ref. 27 (alkali metals); (b) P. Stˇ 332–334 (alkali earth metals). ˇ epniˇcka, Acta Crystallogr., Sect. C: Cryst. Struct. ComK. Mach, J. Kubiˇsta, I. C´ısaˇrov´a, P. Stˇ mun. 2002, 58, m116–m118. ˇ epniˇcka, E. Mieczy´nska, J.J. Zi´ołkowski, J. Organomet. Chem. 2005, A.M. Trzeciak, P. Stˇ 690, 3260–3267. ˇ epniˇcka, J. Demel, J. Cejka, ˇ P. Stˇ J. Mol. Catal. A: Chem. 2004, 224, 161–169. S. Basra, J.G. de Vries, D.J. Hyett et al. Dalton Trans. 2004, 1901–1905. ˇ epniˇcka, I. C´ısaˇrov´a, R. Gyepes, Eur. J. Inorg. Chem. 2006, 926–938. P. Stˇ V.C. Gibson, N.J. Long, A.J.P. White et al. Organometallics 2002, 21, 770–772. J.A. Adeleke, Y-W. Chen, L-K. Liu, Organometallics 1992, 11, 2543–2550. N.J. Long, J. Martin, G. Opromolla et al. Dalton Trans. 1999, 1981–1986. J.E. Aguado, S. Canales, M.C. Gimeno et al. Dalton Trans. 2005, 3005–3015. V.C. Gibson, N.J. Long, A.J.P. White et al. Dalton Trans. 2002, 3280–3289. The following gold(I) complexes with P-monodentate were isolated: [AuCl(L-κP )] and [Au(C6 F5 )(L-κP )] (L = 66 and 70); [Au(L-κP )2 ]ClO4 (L = 66 and 70), [Au(66κP )(PPh3 )](CF3 SO3 ), and [Au(C6 F5 )3 (L-κP )] (L = 66 and 70). In other gold(I) compounds the donors were shown to coordinate as bridging ([Au2 (66-κ 2 P , S)2 ](ClO4 )2 and [(µ70){Au(C6 F5 )3 }2 ]) or P,Se-chelating donors ([Au(C6 F5 )3 (70-κ 2 P,Se)]ClO4 ). By contrast, silver(I) and copper(I) formed exclusively chelate complexes: [Ag(O3 SCF3 )(L-κ 2 P , E)], [Ag(66-κ 2 P , S)2 ](CF3 SO3 ), [Ag(O3 SCF3 )(L-κ 2 P , E)(PPh3 )], and [Cu(L-κ 2 P , E)2 ][PF6 ] (L = 66 and 70). (a) O.B. Sutcliffe, M.R. Bryce, Tetrahedron: Asymmetry 2003, 14, 2297–2325; (b) C.J. Richards, A.J. Locke, Tetrahedron: Asymmetry 1998, 9, 2377–2407. (reviews). J. Park, Z. Quan, S. Lee et al. J. Organomet. Chem. 1999, 584, 140–146. ˇ epniˇcka et al. Collect. Czech. Chem. Commun. 2001, 66, D. Drahoˇnovsk´y, I. C´ısaˇrov´a, P. Stˇ 588–604. W-P. Deng, X-L. Hou, L-X. Dai et al. Chem. Commun. 2000, 285–286. W-P. Deng, X-L. Hou, L-X. Dai, X.W. Dong, Chem. Commun. 2000, 1483–1484. Representative examples: (a) W-P. Deng, S-L. You, X-L. Hou et al. J. Am. Chem. Soc. 2001, 123, 6508–6519; (b) T. Tu, W-P. Deng, X-L. Hou et al. Chem. Eur. J. 2003, 9, 3073–3081; (c) L. Wang, W.H. Kwok, A.S.C. Chan et al. Tetrahedron: Asymmetry 2003, 14, 2291–2295; (d) T. Tu, X-L. Hou, L-X. Dai, J. Organomet. Chem. 2004, 689, 3847–3852. S. Nakamura, T. Fukuzumi, T. Toru, Chirality 2004, 16, 10–12. T. Mino, H. Segawa, M. Yamashita, J. Organomet. Chem. 2004, 689, 2833–2836. J. K¨uhnert, M. Duˇsek, J. Demel et al. Dalton Trans. 2007, 2802–2811. 37 For preparation of compound 85, see Ref. A. Labande, J-C. Daran, R. Poli, E. Manoury, XXII International Conference on Organometallic Chemistry, Zaragoza, Spain, July 23–28, 2006. Poster 8, Book of abstracts, vol. 2, p. 196.

6 Chiral 1,2-Disubstituted Ferrocene Diphosphines for Asymmetric Catalysis∗ Hans-Ulrich Blaser, Weiping Chen, Francesco Camponovo and Antonio Togni

6.1

Introduction

A ferrocene derivative bearing two different substituents in position 1,2 or 1,3 is chiral and represents a very prototypical example of a planar-chiral compound. The 1,2-disubstitution pattern is much more often encountered than the 1,3 isomeric form which is, in fact, quite rare (see also Chapter 7). The importance of enantiomerically pure 1,2-disubstituted ferrocenes and their 1,2,1 -tri- and 1,2,1 ,2 -tetrasubstituted derivatives is undoubtedly connected to the synthesis of ligands for asymmetric catalysis, among which diphosphines have attracted a great deal of attention in both academic and industrial laboratories. The 1,2-disubstituted ferrocene scaffold as a backbone for chiral ligands was introduced by Kumada and Hayashi1 based on Ugi’s pioneering work related to the synthesis of enantiopure ferrocenes (e.g. so-called Ugi’s amine, 1; Scheme 6.1). PPFA (2) as well as BPPFA (3) and BPPFOH (4) were not only the ﬁrst ligands with planar chirality but also proved to be very effective ligands for a variety of catalytic asymmetric transformations. From this starting point, several families of ligands displaying a wealth of structural variations and containing up to all three types of elements of chirality have been developed in the last few years. ∗ Speciﬁc abbreviations used throughout this chapter are given at the end of the chapter before the Reference List.

Ferrocenes: Ligands, Materials and Biomolecules  2008 John Wiley & Sons, Ltd

Edited by Petr Stepnicka

206

Ferrocenes: Ligands, Materials and Biomolecules

NMe2 Fe

Fe

NMe2 PPh2

Fe

X PPh2 PPh2

Ugi's amine (1)

(R,Sp)-PPFA (2)

BPPFA: X = NMe2 (3) BPPFOH: X = OH (4)

Scheme 6.1 Structures of Ugi’s amine and the ﬁrst ferrocene-based chiral phosphine ligands

The literature in the ﬁeld of chiral ferrocene in general has been growing exponentially during the last decade, such that a comprehensive treatise of this chemistry would take encyclopedic dimensions. Therefore, it is our intention to limit the discussion here to diphosphines and related compounds, as the main representatives in the area of chiral ferrocenes, and to focus on structures having a combination of different elements of Model substrates for C=C hydrogenation Typical reaction conditions: R

2

preformed [Rh(diene)LX] or in situ formed complex s/c 100, 20–30 °C, 1–3 bar

enamides NHAc R1

a -acetamido acrylic acid derivatives

COOR2 NHAc R1

R1 = Ph, R2 = Me

methyl acetamido cinnamate (MAC)

R1

acetamido cinnamic acid (ACA)

R1

COOR

= Ph, = H,

R2

=H

= Me

methyl acetamido acrylate (MAA)

itaconic acid (ITA)

R=H

COOR

R2

R = Me dimethyl itaconate (DMI)

Model substrates for C=O hydrogenation O

O

O

COOR2

R1

Typical reaction conditions: preformed [Ru(arene)LX2] or in situ formed complex s/c 100, 20–80 °C, 10–40 bar

Model substrate for allylic alkylation, typical reaction conditions OAc Ph

Ph

[Pd(allyl)Cl]2 + L

CH2(COOMe)2

CH(COOMe)2

BSA, NaOAc

+ s/c 100, r.t., 48 h

Ph

Ph

Scheme 6.2 Model substrates and standard reaction conditions for ligand tests

Chiral 1,2-Disubstituted Ferrocene Diphosphines for Asymmetric Catalysis

207

chirality: planar chirality due to the ferrocene fragment, central chirality due to stereogenic carbon and/or phosphorus centres as well as axial chirality. Several recent reviews cover some of the area of ferrocene-based chiral ligands and their application to various asymmetric transformations.2 These reviews and other chapters of this book should be consulted to put the present account into a broader perspective. The discussion is structured according to the elements of chirality present in the ligand: • Planar and central carbon chirality • Planar and central phosphorus chirality • Planar, central carbon and central phosphorus chirality • Planar, axial and central carbon chirality The descriptions of the ligands will convey representative examples of their applications in catalysis. It is important to note that many of the ligands described in this chapter were tested on model substrates under standard reaction conditions. Unless otherwise noted, this means that usually a substrate to catalyst ratio (s/c) of 100 was employed and that neither Turn Over Number (TON) nor Turn Over Frequency (TOF) were optimised. A few of the most frequently used reactions and standard conditions are depicted in Scheme 6.2.

6.2

Planar and Central Carbon Chirality

With only a few exceptions, the ferrocenyl ligands depicted in Scheme 6.3 are inherently modular. Therefore, they can easily be tuned sterically and electronically by the choice of the two PR2 groups. For this reason, the ligands are among the most versatile for catalytic applications. Indeed, many of the ligand families have been commercialised and a large variety is currently available in technical quantities. The most important method for synthesising ligands with planar and central chirality uses the highly diastereoselective ortho-lithiation of 1 followed by the introduction of an appropriate electrophile (Scheme 6.4).1, 3 6.2.1

BPPFA and BPPFOH

As already pointed out, these were the ﬁrst ferrocene-based ligands and their stereoselective synthesis depends on the highly diastereoselective ortho-lithiation of amine 1 followed by the introduction of an appropriate electrophile (Scheme 6.4).1, 3 Depending upon the electrophile and the lithiation conditions, this can result in a monophosphine such as PPFA, a diphosphine such as BPPFA, or in halides 14a–b, which are themselves starting materials for a large array of ligands obtained via replacement of the halogen and the dimethylamino group. The latter can be easily replaced by an acetoxy as shown for the preparation of acetate 15, obtained by reaction of BPPFA with a large excess of acetic anhydride (Scheme 6.5).4 The acetoxy group in 15 can be hydrolysed to give BPPFOH or replaced with the appropriate nucleophile such as a secondary amine leading to ligands 16a–c or with MeOH to give BPPFOMe. Note that the nucleophilic substitution reaction at the pseudo benzylic stereogenic center of Ugi’s amine derivatives is an SN 1-type process always occurring with retention of conﬁguration.

208

Ferrocenes: Ligands, Materials and Biomolecules

Fe

X PPh2

PR'2

Fe PPh2 PR'2

Josiphos (5)

BoPhoz: X = NR (6) X=O

Fe

PR2 PR'2

X

PR2

Fe

PPh2 BPPFA: X = NMe2 (3) BPPFOH: X = OH (4)

R'

R R' PPh2

Fe PPh2

Fe

H Me

R PPh2 PPh2 R

PR2

PR2 Me Fe Fe

R' Mandyphos (9) (FERRIPHOS)

Taniaphos (8)

Walphos (7)

H

TRAP (10)

R'2P CONH HNOC Fe

PPh2

Fe

R2P

R'2P Fe

Fe

12

13

PR2

Ph2P 11

Scheme 6.3 Structure of representative ferrocene diphosphines with planar and central carbon chirality

The catalytic application of a large variety of PPF and BPPF ligands has been thoroughly reviewed by Hayashi.1 For this reason only the most important results are summarised brieﬂy. Nickel or palladium complexes of PPFA and BPPFA are excellent catalysts for the cross-coupling reaction of secondary Grignard reagents with vinyl bromides giving ee’s up to 95 % in very high yield due to dynamic kinetic resolution of the racemic reagents. Substituted 1-NpMgBr and 1-NpBr can be coupled with nickel/BPPFOMe to give atropisomeric binaphthyl derivatives with enantioselectivities of 95–99 %. Palladium/16b complexes are excellent catalysts for the allylic nucleophilic substitution of various allylic substrates (Scheme 6.6) whereas rhodium/16c complexes proved to be quite active for the rather difﬁcult hydrogenation of tetrasubstituted α,β-unsaturated acids with ee’s up to 98 %.5 For the hydrogenation of α-amino ketones, rhodium/BPPFOH is still one of the best catalysts with enantioselectivities of 92–95 % and a pilot process has been developed for an intermediate of levoprotiline (Scheme 6.6).6 A spectacular reaction is the so-called gold aldol reaction, formally a [3 + 2]-cycloaddition of an aldehyde and an α-isocyanoacetate, allowing the twostep preparation of β-hydroxy-α-amino acid derivatives (Scheme 6.6). Best ligands are prepared from BPPFA and must contain a 1,2-diamino side chain. One of the most effective ligands is 16c giving ee’s of 92–97 % and trans/cis ratios higher than 90/10. Several variants of the reaction have been described: instead of the simple aldehydes,

Chiral 1,2-Disubstituted Ferrocene Diphosphines for Asymmetric Catalysis

1) n-BuLi 2) Ph2PCl NMe2 Fe

NMe2 (R,Sp)-PPFA (2) Fe PPh2

1) n-BuLi, TMEDA 2) Ph2PCl

NMe2 BPPFA (3) Fe PPh2 PPh2

1) n-BuLi 2) I2 or CBr2Cl4

(R)-1

209

NMe2 Fe

X

14a X = I 14b X = Br

Scheme 6.4 Synthesis of ligands and ligand precursors based on the diastereoselective lithiation of Ugi’s amine

1) n-BuLi 2) H2O NMe2 Ac2O Fe PPh2 PPh2 BPPFA (3)

OAc Fe PPh2 PPh2 15

Me HN R MeOH or EtOH

OH Fe PPh2 PPh2

BPPFOH (4)

R

16 a R = (CH2)2NEt2 b R = CH(CH2OH)2

N Fe PPh2 Me c PPh2

R = (CH2)2N

Scheme 6.5 Synthesis of classical BPPFA derivatives

activated ketones as well as chiral aldehydes can be used; silver-based catalysts are also active for selected reactions and α-isocyanophosphonates are also suitable reactants. The reaction has been applied to the synthesis of MeBmt, the unusual amino acid of cyclosporine, and was actually run on a multi kilogramme scale (Scheme 6.6).7 6.2.2

Josiphos

The Josiphos ligands arguably constitute one of the most versatile and successful ligand families, second probably only to the BINAP ligands. Because the two phosphine groups are introduced in consecutive steps with very high yields (Scheme 6.7),8, 15 a variety of ligands are readily available with widely differing steric and electronic properties. Up to now, only the (R,Sp )-family (and its enantiomers) but not the (R,Rp ) diastereomers have led to high enantioselectivities. At present, about 150 different Josiphos ligands have been prepared and 40 derivatives are available in a ligand kit for screening and on a multi kilogramme scale for production.9, 10 The most successful ligands, their numbering and applications are depicted in Scheme 6.7.

210

Ferrocenes: Ligands, Materials and Biomolecules Allylic substitution OAc + Ar

CH2(COOMe)2

Pd / 16b

ee 86–96 %

BnNH2

Pd / 16b

ee >97 %

Ar

Ar = Ph, 3-MeOPh, 1-Np OCOOEt + R

R

R = Ph, i-Pr Au- aldol reaction R O R

+

[Au(CNCy)2]BF4 / 16c

CN-CH2COOMe

CH2Cl2, 0–20 °C

H

O

COOMe + cis isomer

N

Industrial applications O

H

NHMe +

COOMe

O

HO

NH2

(MeBmt)

+ diastereomers

CN-CH2COOMe

Au / 16a 40 °C, s/c 100 yield 85 %, 15 % other diastereomers bench scale process, Ciba-Geigy

Rh / BPPFOH 50 °C, 80 bar ee 97 %; TON 2000 pilot process for levoprotiline Ciba-Geigy

Scheme 6.6 Some representative reactions with BPPF ligands

NMe2 Fe

Fe (R)-1

NMe2 R'2PH PR2 AcOH, 80 °C

(R,Sp)-PPFA (2)

Fe

PR'2 PR2

(R,Sp)-Josiphos (5)

ligand

R

R'

important applications

5a

Ph

Cy

5b 5c 5d 5e 5f 5g 5h

Ph Cy Cy Ph Cy 4-CF3-Ph 4-MeO-Xyl

t-Bu Cy Ph 3,5-Xyl t-Bu t-Bu t-Bu

process for jasmonate; hydrogenation of enamides, itaconates; hydroboration; allylic alkylations; Michael additions; PMHS reduction of C=C; addition to meso anhydrides opening of oxabicycles; biotin process hydrogenation of phoshinylimines Michael additions Metolachlor process; methoxycarboxylation hydrophosphination; cross-coupling process for MK-0431 dextromethorphane process

Scheme 6.7 Josiphos ligands and some of their representative applications

Chiral 1,2-Disubstituted Ferrocene Diphosphines for Asymmetric Catalysis

211

X-ray structures of both the free ligand 5a as well of its rhodium, palladium and platinum complexes were determined and conﬁrmed both structure and sense of the planar chirality.11 To avoid interaction of the large PR2 group with the rest of the molecule, the hydrogen and Me group attached the side chain are forced below the Cp plane while the PR2 group is above. This places the two phosphine groups in an ideal position for the formation of metal complexes and, as such, dictates a similar coordination behaviour to different metal centres while still allowing the adoption of different bond lengths, oxidation states and coordination geometries. Catalytic applications of the Josiphos ligand family have been reviewed elsewhere up to 2002.9 For this reason, only a summary is provided here of the most important processes operated with Josiphos ligands and an update on new results is given. Up to now, Josiphos ligands are (or have been) applied in four production processes (Scheme 6.8) and about ﬁve or six pilot and bench scale processes involving rhodium-, iridium- and ruthenium-catalysed hydrogenation reactions of C=C and C=N bonds.16 The most important application is undoubtedly the Ir/5e-catalysed hydrogenation of a hindered N-aryl imine of methoxyacetone, the largest known enantioselective process operated for the enantioselective production of the herbicide (S)-Metolachlor.15, 43 The hydrogenation of a tetrasubstituted oleﬁn was the key step for two production processes developed for the synthesis of biotin by Lonza and of methyl dihydrojasmonate by Firmenich, respectively.43a Pilot processes were developed by Lonza for a building block of crixivan and for dextromethorphane.43a Recently, Merck chemists reported the unprecedented hydrogenation of unprotected dehydro β-amino acid derivatives catalysed by Rh/Josiphos with ee’s up to 97 %. It was shown that not only simple CF3

MeO

COOMe

F N

N

N N

NH2 O

F

O

F Ir/Josiphos 5e; ee 80 % TON 2000 000; TOF >400 000 h−1 very large scale production Ciba-Geigy/Syngenta/Solvias

Rh/Josiphos 5g; ee 94 % TON 350; TOF ~50 h−1 small scale production process Merck

O N

O N

NH OHC O

Ru/Josiphos 5a; ee 90 % TON 2000; TOF 200 h−1 medium scale production Firmenich

O

Rh/Josiphos 5b; ee 99 % TON 2000; TOF n.a. medium scale production Lonza

N

Et O

N . H3PO4

NHtBu Rh/Josiphos 5b; ee 97 % TON 1000; TOF 450 h−1 pilot process, >200 kg Lonza

OMe Ir/Josiphos 5h; ee 90 % TON 1500; TOF n.a. pilot process Lonza

Scheme 6.8 Important industrial applications of Josiphos ligands

212

Ferrocenes: Ligands, Materials and Biomolecules

derivatives but also the complex intermediate for MK-0431 depicted in Scheme 6.8 can be hydrogenated successfully. Regular production on a multi tonne per annum scale with ee’s up to 98 % started in 2006.17 Rhodium and iridium complexes with chiral Josiphos ligands are highly selective, active and productive catalysts for various enantioselective reductions. Very high enantioselectivities are obtained for the enantioselective hydrogenation of enamides, itaconic acid derivatives, acetoacetates as well as N-aryl imines (in the presence of acid and iodide) and phosphinylimines.9 Josiphos 5a is the ligand of choice for the coppercatalysed reduction of activated C=C bonds with PMHS23 (Scheme 6.9) with very high enantioselectivities for nitro alkenes,18 α,β-unsaturated ketones,19a esters19b, c and nitriles.24 The same ligand has recently been reported to give high stereoselectivity in the iridium-catalysed hydrogenation of an α-aminoacetophenone,25 while 5b was preferred for the rhodium-catalysed hydrogenation of an α-amino-β-ketoester with moderate ee’s but high diastereoselectivity26 and for the hydrogenation of a tetrasubstituted sulfonamidoacrylic acid derivative with ee’s up to 97 %.27

R1

NO2

+ PMHS

0.1–1 mol % [CuO-t-Bu]4 / 5a PhSiH3, H2O toluene

R2 R = Ar, HetAr, Alk

R2 NO2

R1

ee 80–96 % yield 60-90 %

R1

R3 R2

+ PMHS

O

R = (subst)Alk, PhMe2Si

0.1–1 mol % CuCl or [CuH(PPh3)]6 / 5a toluene / t-BuONa or t-BuOH

R' = Alk, OEt

CN

R3 R2

O

ee 95–99 % yield >90 %

3 mol % [Cu(OAc)2] / 5a

Ar

R1

+ PMHS

R

toluene

Ar

CN R

R = (subst)Ph, pyridine

ee 95–99 %

R = Alk

yield 80–96 %

Scheme 6.9 Asymmetric Copper-catalysed reductions using Josiphos

Josiphos complexes have been successfully applied to various other catalytic reactions, such as allylic alkylation or hydroformylation.9 Recently, Feringa’s group reported high ee’s for the copper-catalysed Michael addition of Grignard reagents to α,β-unsaturated ester,20a thioesters20b (Scheme 6.10), and for selected cyclic enones.20c Preferred ligands were 5a and 5d. The Rh/5a-catalysed Michael addition of organotriﬂuoroborates to cyclic enones with enantioselectivities up to 99 % was reported by Genˆet and coworkers (Scheme 6.10).28 The same catalyst type was also very effective for the nucleophilic ring opening of oxabicyclic substrates leading to tetrahydronaphthalene and cyclohexene derivatives (Scheme 6.11).22 This reaction was

Chiral 1,2-Disubstituted Ferrocene Diphosphines for Asymmetric Catalysis 0.5–2.5 mol% Cu/Josiphos COOMe

R

213

R'

+ R'MgBr

COOMe t-BuOMe −75 °C, 1–2 h

R = Alk, (subs)Ar, hetAr R' = mostly Et

ee 86–97 % yields 75–99 %

R = alkyl: preferred ligand 5a R = aryl: preferred ligand 5d for thioesters: preferred ligand 5a; ee 86–96 %, yields 80–95 % O

O

3 mol% Rh/5a + K[BPhF3]

ee 99 % toluene / H2O 100 °C, 1–2 days

Ph

Scheme 6.10 Asymmetric 1,4-addition reactions using Josiphos

O

R

Rh/5a

+ NuH

THF, 80 °C

R Nu ee 95–99 % ee 93–99 % ee 95 %

NuH = various substituted phenols NuH = various alcohols NuH = PhSO2NH2

OH

O + NuH

OR

Rh/5b

ee 93 % ee 95 % ee 94–99 %

NuH = PhOH NuH = PhNHMe NuH = ArB(OH)2 H

H

OH

O O

R

OR

Nu

OR

R

OR

+ Ph2Zn

Pd/5a THF, 80 °C

O

R R

H O Ph H

COOH

ee up to 97 %, s/c 20, yield 70–95 %

cyclic and acyclic CN

CN Pd/5a DMF, 65–90 °C

O

O X

X = I, Br, OTf

ee 60–94 %, s/c 10, yield 80–92 %

Scheme 6.11 Various rhodium- and palladium-catalysed reactions using Josiphos ligands

214

Ferrocenes: Ligands, Materials and Biomolecules

scaled up to the kilogramme scale.29 The palladium-catalysed opening of various cyclic anhydrides with Ph2 Zn was described by Bercot and Rovis to occur with very high ee’s in the presence of 5a (Scheme 6.11).21 The Hartwig group described the Pd/5bcatalysed addition of acetylacetone to dienes with ee’s up to 81 %30 and Lorman et al. reported up to 95 % ee for an intermolecular Heck reaction using a Pd/5a catalyst (Scheme 6.11).31 Josiphos ligands are effective in the palladium-catalysed methoxycarbonylation of styrene (the best ligand is 5e, low branched/linear ratio)33 and for the hydrophosphonation of C=C bonds with ee’s up to 88 % for 5f.34 Finally, aryl and vinyl bromides can be cross-coupled with allyl triﬂuoroborate salts with high branched/linear ratios and enantioselectivities up to 90 %; the best catalyst for this transformation is [Pd(OAc)2 ]/5f.32 Immobilised Josiphos and Josiphos Analogs The Josiphos backbone has been covalently attached to organic or inorganic polymeric supports, and modiﬁed by the introduction of hydrophilic groups (to make the ligands water soluble) or imidazolium tags (in order to immobilise the ligand in ionic liquids; Schemes 6.12 and 6.13).12–14 The synthesis of the functionalised ligand precursor starts from 1, which is conventionally bis-lithiated with n-BuLi/TMEDA and subsequently reacted with 1,2-C2 Br2 Cl4 to furnish the dibromide (17). Selective lithium–bromine

NMe2 Fe

1) n-BuLi TMEDA

Br

Fe

2) (CF2Br)2

NMe2 1) n-BuLi

Br

1

NMe2 R PH 2 PPh2

Fe 2) Ph2PCl

17

AcOH

Br 18 O

1) Fe

PR2 1) n-BuLi PPh2 Br

Fe

2) Cl(CH2)3Me2SiCl

PR2 PPh2

1) n-BuLi 2) DMF

Fe

PR2 PPh2 CHO

23

PR2 PPh2

Si 21

Cl

COOH 22

Fe

20

1) n-BuLi 2) CO2

Fe

O 2) H2NNH2

Si

19

NK

PR2 PPh2

NH2

1) NH2OH 2) red.

Fe

PR2 PPh2 CH2NH2

24

Scheme 6.12 Synthesis of Josiphos derivatives bearing functional groups at position 1 of the ferrocene core

Chiral 1,2-Disubstituted Ferrocene Diphosphines for Asymmetric Catalysis

215

Dendrimer bound Josiphos (starting from 21) O

X

NHR Fe X

X = COO

X

RNH =

Cy2 PPh2

Si HN

X

NHR

25

O

SiO2 bound Josiphos (starting from 21)

O

O O

Si

N H

Fe N H

PXyl2 PPh2

Si

O 26 Polymer bound Josiphos (starting from 21) O N H

polystyrene

O N H

N H

PXyl2 PPh2

Fe N H

Si

27 Ionic liquid and water soluble Josiphos (starting from 22 and 24)

Fe

PR2 PPh2

(CF3SO2)2NG N

+ N Me

Fe

PR2 PPh2

COOH O

NH

HN

O

HN

O 28

29

O

COOH

O

COOH

Scheme 6.13 Structure of functionalised Josiphos ligands

exchange followed by quenching with ClPPh2 affords 18, which is converted to the bromo-substituted Josiphos 19 as described above. A second lithium–bromine exchange followed by trapping with (3-chloropropyl)dimethylchlorosilane gives 20, which was then converted to the functionalised ligand 21 by reaction with potassium phthalimide and subsequent treatment with hydrazine. Ligands with a COOH or CH2 NH2 group can be prepared by the same strategy from 19 upon reaction of the lithiated intermediate either with CO2 or with DMF, followed by reaction with hydroxylamine and reduction, respectively. To attach Josiphos to a dendrimer backbone, 21 was ﬁrst reacted with 5-(tertbutyldimethylsiloxy)isophthaloyl dichloride and, after deprotection, with adamantanetetracarboxylic acid chloride. The resulting dendrimer 25 contains eight Josiphos

216

Ferrocenes: Ligands, Materials and Biomolecules

units.14 For immobilisation on silica gel,13 21 was reacted ﬁrst with 3-isocyanatopropyltrimethoxysilane and then with silica to afford SiO2 -Josiphos 26. To immobilise Josiphos on polystyrene (PS),13 the amino groups of the aminomethylated polystyrene were ﬁrst reacted with an excess of 2,4-diisocyanatotoluene to give a reactive polymer followed by addition of 21. The remaining isocyanate groups were reacted with an excess of ethanol to give the immobilised ligand PS–Josiphos 27. The ionic liquid soluble ligand 2812 was prepared from 22 by DCC mediated coupling with the corresponding aminopropyl methyl imidazolium salt. Finally, the water soluble Josiphos 2912 was synthesised by condensation of 24 with (EtO2 C(CH2 )2 OCH2 )3 CNH2 in the presence of an activated urea. The functionalised ligands were tested for various hydrogenation reactions. The silica-supported ligand 5e (26) as well as the water soluble ligand 29 gave TONs >100 000 and TOFs up to 20 000 h−1 for iridium-catalysed imine reductions, while the polymer bound ligand 27 was much less active; in all cases ee’s were comparable to those obtained with the homogeneous catalyst.13 Dendrimeric Rh/Josiphos complexes 25 hydrogenated DMIT with ee’s up to 98.6 % with similar activities as the mononuclear catalyst.14 Similarly, ligands with an imidazolium tag 28 had a catalytic performance comparable to that of the non-functionalised ligands for the rhodium-catalysed hydrogenation of MAA and DMI both in classical solvents and under two-phase conditions in ionic liquids. In the latter case they were easy to separate and to recycle.12 Rh/Josiphos complexes were also ion-exchanged onto MCM-4135 and adsorbed onto a heteropolyacid-modiﬁed alumina.36 Both catalysts were active for the hydrogenation of DMI with good to very good ee’s and could easily be recycled or operated continuously in supercritical carbon dioxide, respectively. Ligands 1237 and 1338 (Scheme 6.14), two analogs of Josiphos, were prepared by Weissensteiner and colleagues to test the effect of restricting the conformational ﬂexibility of the ligand on its catalytic performance. The synthesis of 12 started with the reduction of (Rp )-30 with lithium aluminium hydride to give alcohol (S,Rp )-31, which was subsequently transformed with HNMe2 /AlCl3 to amine (S,Rp )-32. Treatment with Me2N

HO

O LiAlH4

Fe

Me2NH AlCl3

Fe

R'2P

Me2N Ph2P Fe

(S,Rp)-33

Ph2PCl

(S,Rp)-32

(S,Rp)-31

(Rp)-30

n-BuLi Fe

R2PH AcOH

R'2P

R 2P Fe

Fe

12

13

PR2

Scheme 6.14 Synthesis of conformationally rigid Josiphos-type ligands 12 and the structure of ferrocenophane ligand 13

Chiral 1,2-Disubstituted Ferrocene Diphosphines for Asymmetric Catalysis

217

n-BuLi followed by quenching with R2 PCl led to the aminophosphine (S,Rp )-33. Finally, reaction of aminophosphine (S,Rp )-33 with a secondary phosphine in acetic acid gave diphosphines 12. Compound 13 was synthesised analogously to the preparation of Josiphos (see Chapter 7 for details). The X-ray crystal structure of 12 and 13 show that while the phosphino groups are well positioned to form metal complexes, the ligands cannot adapt as easily as Josiphos ligands to the requirements of a metal since the conformational space is restricted. In view of this structural information it was not unexpected that the enantioselectivity and, in many cases, also the catalytic activity observed in the hydrogenation of various C=C, C=O and C=N containing substrates,37 as well as in the allylic alkylation and amination reactions39 with metal complexes containing 12 and 13, were often signiﬁcantly lower compared to those of corresponding Josiphos catalysts. 6.2.3

BoPhoz and Analogues

Similarly to Josiphos, BoPhoz is a modular ligand class with a PR2 group on the Cp ring and an aminophosphine at the side chain. Its preparation starts from PPFA and leads via reaction of acetate 34 to the secondary amines 35 (Scheme 6.15).40 Coupling of the amino substituent with chlorophosphines affords BoPhoz. Recently, Boaz developed a new modular synthesis by reaction of 35 with phosphorus trichloride followed by reaction with a Grignard reagent.41 This new method allows the incorporation of a wide range of nitrogen and phosphorus substituents R and R . Even though a variety of ligands with different R and R groups have been prepared, the preferred ligand up to now is the N -methyl derivative with R = Ph. Bophoz ligands are stable in air but depending on the solvent the stability of the N-PR2 bond may be a critical issue. Selected BoPhoz ligands are available from Johnson Matthey.46 A phosphinite analog 37 prepared from the hydroxyl derivative 36 by reaction with R2 PCl was recently described by Jia et al.70

NMe2 Fe PAr2

OAc Fe PAr2

Ac2O 100 °C

i-PrOH, 50 °C

34

(R,Sp)-PPFA

OH Fe PAr2

NHR Fe PAr2 35

PCl3, Et3N toluene

1) Ac2O 2) LiAlH4

36

RNH2

Ar'2PCl

OPAr'2

N

R

Fe PAr2

Fe PAr2 PCl2

37

38

R'2PCl, Et3N toluene

R'MgX

Scheme 6.15 Synthesis of BoPhoz ligands

N

R

Fe PAr2 PR'2 BoPhoz (6) preferred ligand R = Me, Ar; R' = Ph

218

Ferrocenes: Ligands, Materials and Biomolecules

BoPhoz ligands are very effective for the rhodium-catalysed hydrogenation of a variety of activated C=C bonds such as enamides (ee 96–99 %, s/c up to 10 000) and itaconates (ee 80–99 %).40 As observed for several ligands forming seven-membered chelates, high activities can be reached (maximum TOFs for enamides up to 68 00 h−1 ).40b While Boaz found that R = Ph gave the best results, Chan and co-workers42 showed that the 3,5-Me2 or 3,5-(CF3 )2 substitution pattern of the phenyl groups attached to phosphorus often had a beneﬁcial effect on ee’s and ligand stability for the hydrogenation of a variety of enamides (ee up to 99.7 %) and MAC (ee up to 99.5 %). Similar results were reported by Jia et al.70 for the hydrogenation of a variety of substituted acetamido acrylates (ee 95–99.5 %) catalysed by Rh/37 (Ar = Ph, 3,5(CF3 )2 -Ph). Feasibility studies for the technical preparation of cyclopropylalanine40b and 2-naphthylalanine40b have been reported. BoPhoz is also suitable for the rhodiumcatalysed hydrogenation of α-keto esters (ee 88–92 %) and keto pantolactone (ee 97 %). Recently, it was reported that various β-ketoesters are hydrogenated with Ru/BoPhoz complexes with enantioselectivities of 94–95 % (s/c 200, room temperature, 20 bar).41c 6.2.4

Walphos

Modular Walphos ligands form eight-membered metallocycles due to the additional phenyl ring attached to the cyclopentadienyl fragment. There are noticeable electronic effects but the scope of this ligand family is still under investigation; several derivatives are available from Solvias on a technical scale.10 Walphos was developed by the group of Weissensteiner starting from 1 (Scheme 6.16).44 In a Negishi coupling reaction of (R)-1 with 2-bromoiodobenzene, the enantiomerically pure key intermediate (R,Rp )-39 was obtained. A subsequent lithiation of this bromide followed by quenching with the appropriate R2 PCl resulted in the formation of the corresponding tertiary phosphine. To prevent a ring closure reaction in the next step, the phosphines were converted to the corresponding phosphine oxide (R,Rp )-40. Nucleophilic substitution of the dimethylamino group with either various R2 PH in acetic

NMe2 Fe

1) s-BuLi 2) ZnCl2,1,2-BrC6H4I [Pd]

(R)-1

Fe

NMe2 1) s-BuLi R2(O)P Me 2) R2PCl H

3) H2O2

(R,Rp)-39 R2(O)P

PR'2 PMHS, [Ti(OPr-i)4] PR2 or Me

R'2PH AcOH

Br

Fe

(R,Rp)-41

H

SiHCl3, Et3N

PR'2 Me 7a 7b Fe H 7c 7d 7e Walphos (7) 7f 7g

NMe2 Me Fe

H

(R,Rp)-40 R

R'

Ph Ph Ph 4-MeO-Xyl Ph Cy Xyl

3,5-(CF3)2-Ph Ph Cy 3,5-(CF3)2-Ph Xyl 3,5-(CF3)2-Ph Xyl

Scheme 6.16 Synthesis of Walphos ligands

Chiral 1,2-Disubstituted Ferrocene Diphosphines for Asymmetric Catalysis

219

acid led to (R,Rp )-41. The diastereoselectivity of the nucleophilic substitution step seems to depend strongly on the nucleophilicity of the phosphine, since only the electron-rich phosphines gave products with full retention of conﬁguration. Reduction of (R,Rp )-41 with PMHS/Ti(O-i-Pr)4 or HSiCl3 /NEt3 gave the Walphos ligands. Walphos ligands show promise for various enantioselective hydrogenations. Rh/ Walphos catalysts gave good results for dehydro amino and itaconic acid derivatives (ee 92–95 %)44b, 47 and of vinylboronates (Scheme 6.17).45 Ru/Walphos complexes were highly selective for β-ketoesters (ee 91–95 %)44b, 47 and acetylacetone (ee >99.5 %, s/c 1000, 7a).44b, 47 The copper-catalysed enantioselective reduction of α,β-unsaturated ketones with PMHS was carried out with 92–95 % ee’s (s/c 100 at −78 ◦ C), ligand 7a being the preferred one.19a The ﬁrst industrial application has just been realised in collaboration with Speedel/Novartis for the hydrogenation of SPP100-SyA, a sterically demanding α,β-unsaturated acid intermediate of the renin inhibitor SPP100 (Scheme 6.17). The process has already been operated on a multi tonne scale. Recently, two novel transformations were reported to be catalysed by Rh/Walphos complexes with high enantioselectivities (Scheme 6.17): the [4 + 2]-addition of 4-alkynals with an acrylamide by Tanaka and co-workers48 and the reductive coupling of enynes with α-keto esters by Krische’s group.49 O O O

Rh / Walphos 7a; ee 95 % TON 5000; TOF ~800 h−1 medium scale production Novartis / Solvias

OH

O

O R

B

O

R' = H, boronate

Rh/7a or 7f

+ H2

O

s/c 50, 15–35 bar

R'

B

O

R

ee 85–95 % R' yield 60–90 % O

O H

Rh/7b

+

CONMe2

R

s/c 10–20 CH2Cl2, 80 °C

R H ee 97 to >99 % yield 50–90 %

R = (subst)Ar, Alk

R R = (subst)Alk

O

+ R'

CONMe2

Rh/7g + H2 COOMe

R' = Alk, (subst)Ph

s/c 50 1 bar, 80 °C

R COOMe R'

OH

ee 88–93 % yield 80–97 %

Scheme 6.17 Industrial and synthetic applications of Walphos ligands

220

Ferrocenes: Ligands, Materials and Biomolecules

6.2.5

Taniaphos

Taniaphos ligands developed by the Knochel group have an additional phenyl ring inserted at the side chain of Ugi’s amine. Besides the two phosphine moieties, the substituent at the stereogenic centre can also be varied and, up to now, three generations of Taniaphos ligands with different substituent types have been prepared (Scheme 6.18). Several Taniaphos ligands are being marketed by Solvias in collaboration with Umicore.10 R1 R2 PR32 Fe PR42

R1

R2

N(Alkyl)2 H or MeO H

H MeO or H Alkyl

Taniaphos (8)

first generation (8a) second generation (8b) third generation (8c)

Taniaphos ligands of the first generation:

8aa 8ab 8ac 8ad 8ae

R1 (R2 = H)

R3,4

NMe2 NMe2 NMe2 NBu2 morph

Ph Cy 4-MeO-Xyl Cy Xyl

Scheme 6.18 Structure of Taniaphos ligands

The ﬁrst generation Taniaphos is readily prepared in ﬁve steps starting from ferrocene in a highly stereoselective process (Scheme 6.19).50a, b Friedel–Crafts acylation of ferrocene with 2-bromobenzoyl chloride furnished ketone 42 in 80 % yield. The CBS reduction of 42 afforded alcohol 43 in 95 % yield and 96 % ee (99.5 % after recrystallisation from heptane). Acylation of the alcohol followed by treatment with a dialkylamine led to 44. Finally, the ﬁrst generation Taniaphos (8a) was obtained by the lithiation with t-BuLi followed by the reaction with ClPR2 . O

Fe

o-BrC6H4COCl, AlCl3

Br

OH CBS catalyst BH3•SMe2

Fe

Fe

42 (Alkyl)2N 1)Ac2O 2)NH(Alkyl)2

Fe 44

43 (Alkyl)2N

Br 1) t-BuLi 2) R2PCl

PR2

Fe PR2 Taniaphos (8a) (first generation)

Scheme 6.19 First generation synthesis of Taniaphos ligands

Br

Chiral 1,2-Disubstituted Ferrocene Diphosphines for Asymmetric Catalysis

221

Unlike the synthesis of ﬁrst generation of Taniaphos ligands, Knochel’s route to the second generation is not stereoselective (Scheme 6.20).50c The reaction of lithiated sulfoxide (S)-45 with 2-(diphenylphosphino)benzaldehyde furnished the two diastereomeric alcohols (Sp ,αS)-46a and (Sp ,αR)-46b (55:45) in 82 % yield which were separated by column chromatography. (Sp ,αS)-46a was converted to the corresponding methyl ether 47 by deprotonation and methylation. Subsequent lithiation followed by the addition of Ar2 PCl, provided the second generation of Taniaphos ligands (8b). OH

S

Fe

1) LDA O 2) o-(Ph P)C H CHO 2 6 4

OH

O S

Fe

OH

PPh2

O

Tol (Sp, aR)-46b 37 %

OMe PPh2

Fe

Tol

OMe PPh2 1) t-BuLi 2) Ar2PCl

O

KH,MeI

S

S

separation

S

Fe

(Sp, aS)-46a 45 %

(S)-45

PPh2

O +

Tol

Tol

Fe

PPh2

Fe PAr2

Tol

(Sp, aS)-46a

Taniaphos (8b) (second generation)

47

Scheme 6.20 Second generation synthesis of Taniaphos ligands

This synthesis has the advantage of allowing the preparation of 1,5-diphosphines with two different phosphorus substituents, but includes a tedious separation of the diastereomeric alcohols. Very recently, Chen et al. reported a highly stereoselective synthesis of second generation Taniaphos (Scheme 6.21).51 Reaction of (S)-α-2bromoferrocenecarboxaldehyde (48) with the Grignard reagent prepared from the readily available (2-bromophenyl)diphenylphosphine, gave (Sp ,αS)-49 in 98 % yield as a MgBr CHO Br

Fe

OH

PPh2

KH, MeI Br

Fe

(Sp, aS)-49

50

Fe

(S)-48

OMe PPh2

PPh2

Br

OMe PPh2 1) t-BuLi 2) Ar2PCl

Fe PAr2

Taniaphos (8b) (second generation)

Scheme 6.21 Alternative route to second generation Taniaphos ligands

222

Ferrocenes: Ligands, Materials and Biomolecules

single diastereomer. Methylation of alcohol (Sp ,αS)-49 furnished the ether (Sp ,αS)-50, whereafter exchange of the bromine atom for the diphenylphosphino moiety afforded second generation Taniaphos (8b) in 92 % overall yield. The diastereomeric ligand with (Sp ,αR) conﬁguration can be prepared from 46b. Finally, Knochel developed a stereoselective synthesis for third generation Taniaphos starting from 51 (Scheme 6.22).50d Substitution of the dimethylamino groups with o-bromophenylzinc bromide in the presence of acetyl chloride and subsequent recrystallisation from ether led to the diastereomerically pure ligand precursors 52 in high yields. Double halogen–lithium exchange followed by the reaction with ClPPh2 gave third generation Taniaphos (8c). Note that the term ‘third generation’ is here only valid because of the different reagents used as compared to the ﬁrst-generation synthesis by which some of the same derivatives were already accessible. ZnBr

R NMe2 Fe

X

51

X = halide

R

PPh2

1) t-BuLi 2) Ph2PCl

Br AcCl, −78 °C

R

Br

Fe 52

X

Fe PPh2 Taniaphos (8c) (third generation)

Scheme 6.22 Third generation synthesis of Taniaphos ligands

A variety of Taniaphos ligands have been shown to be very selective in a number of model hydrogenation reactions.50, 52 Both the nature of two phosphine moieties (R3 , R4 ) and of the substituents at the stereogenic center (R1 , R2 ) have a strong but not systematic effect on the catalytic performance. With very few exceptions, relatively electron rich all-aryl substituted derivatives (R3 , R4 = Ph, Xyl, MeO–Xyl) gave the best performance. Rather surprisingly, a change of the substituents can even lead to a different sense of induction for the rhodium-catalysed hydrogenation of MAC (ee’s 94–99.5 %) and DMI (ee’s 91–99.5 %). For MAC, methyl or methoxy substituents lead to the opposite absolute conﬁguration of the product compared to R1 = NMe2 , i-Pr or H. Similar effects are also observed for DMI and for the hydrogenation of an enol acetate where ee’s up to 98 % but low activities have been obtained. Interestingly, changing the absolute conﬁguration of the stereogenic centre for the second generation Taniaphos (R1 or R2 = MeO, 8b) only has an effect on the ee-level but not on the sense of induction. Enamides are hydrogenated with 92–97 % ee’s but low TOFs (Rh/8ad) and a β-dehydro acetamido acid with 99.5 % (Rh/8ac). Taniaphos 8aa is very efﬁcient for the rhodium-catalysed hydrogenation of β-diketones (ee 99.4 %) and β-ketoesters (ee’s up to 94 %; s/c up to 25 000). Recently, a variety of highly enantioselective transformations were described using Taniaphos complexes; selected reactions are depicted in Scheme 6.23. Ligands 8aa were found to be very selective for the rhodium-catalysed nucleophilic ring opening of an azabicycle53 and for the copper-catalysed Michael addition of Grignard reagents to cyclohexenone (ee’s 94–96 %).20c Copper-Taniaphos complexes were also very

Chiral 1,2-Disubstituted Ferrocene Diphosphines for Asymmetric Catalysis

223

NBoc + R2NH

Rh/8aa

ee 84–96 % yield 75–90 %

R 2N NHBoc

O R

+

R'

Cu/8

COOMe + PhSiH3

−40 °C

HO R' R

R = Alk, (subst)Ph R' = H: preferred ligand 8aa, 50–88 % erythro, ee 68–97 % R' = Me: preferred ligand 8ab, 86–95 % erythro, ee 83–95 % O +

Ar

•

O

COOMe +

BH O

Cu/8ae −20 °C

R N

Ar +

SO2Ar'

Cu/8aa Et3N 0 °C

R = H, Me

R

COOMe threo

erythro yield 70–99 % HO Ar

COOMe + diastereomer

ee 82–84 %; dr >8 yield ~90 %

Ar = (subst)Ph, 2-Np

MeOOC

HO R' COOMe +

Ar = (subst)Ph Ar' = Ph, 1-Np, 2-thienyl

SO2Ar' R MeOOC

Ar N H ee 79–85 % yield 70–90 %

Scheme 6.23 Synthetic applications of Taniaphos ligands

effective for the reductive addition of aldehydes55a to methyl acrylate and of methyl ketones to methyl acrylate55b and allene carboxyl ester56 (Scheme 6.23). Coppercatalysed additions of aryl methyl ketones to silylated ketene acetals in the presence of Taniaphos 8ad gave ee’s up to 92 %57 while Cu/Taniaphos 8aa was effective for the cycloaddition of azomethine ylides to vinyl sulfones (Scheme 6.23)58 and for the reaction of various allylic bromides with Grignard reagents with 92–97 % ee.59 6.2.6

Mandyphos (Ferriphos)

Ferriphos/Mandyphos (9a) was ﬁrst prepared by Hayashi65 as a bidentate analog of PPFA (Scheme 6.24). Later, Knochel54 developed a general synthesis for a highly modular ligand family called Ferriphos/Mandyphos where not only the PR32 moieties but also the R2 substituents can be used for ﬁne tuning purposes. Selected Ferriphos/ Mandyphos derivatives are commercialised by Solvias in collaboration with Umicore.10 The synthesis of Ferriphos/Mandyphos 9d (Scheme 6.25)54a, b starts from diamine 53 which was dilithiated with t-BuLi and halogenated to give dibromides 54 as a single diastereoisomer. Direct substitution of the dimethylamino groups of 54 with diorganozincs in the presence of acetyl chloride gave intermediates 55 (R1 = Me, i-Pr, allyl). The treatment of 55 with n-BuLi followed by the reaction with ClPPh2 provided Ferriphos/Mandyphos 9d. Ferriphos/Mandyphos 9a–c54c are prepared by dilithiation of the corresponding diamines 53 with t-BuLi, followed by quenching with ClPPh2 , thus affording dimethylamino derivative 56. Ferriphos/Mandyphos with different amino

224

Ferrocenes: Ligands, Materials and Biomolecules R1 R2 PR32 PR32 R2

Fe

9a 9b 9c 9d

R1 Ferriphos/ Mandyphos (9)

R1

R2

R3

NMe2 NMe2 NMe2 Alkyl

Me Ph Ph Aryl

Ph Ph 4-MeO-Xyl Ph

Scheme 6.24 Structure and numbering of Ferriphos/Mandyphos ligands R1

NMe2 1) t-BuLi 2) (CCl2Br)2

NMe2

R2 Fe

Br Br R

R2 Fe

53 R2

1) t-BuLi 2) R32PCl

NMe2

= Ph, Me, Et Fe

R 2Zn AcCl

R1 R

Fe

2

2

Br Br R

1) n-BuLi 2) Ph2PCl

NMe2

OAc

R2 PR32 PR32 R2

R2 PR32 PR32 R2

Fe

NMe2

N(Alkyl)2 NH(Alkyl)2

OAc

56

R2 PPh2 PPh2 R2

R1 Mandyphos (9d)

55

Ac2O

Fe

2

R1

NMe2

54

R2

1

Fe

R2 PR32 PR32 R2

N(Alkyl)2 Mandyphos (9a-c)

57

Scheme 6.25 Synthesis of Ferriphos/Mandyphos ligands R

COOMe O

[Rh(nbd)2]BF4/9c; 5 bar, r.t.

NHCbz

R = H, Me, Et, OSiR3, CF3 NBoc

R

R

R' R' + R2NH

R

ee 80–97 %, yield 80–97 %

Rh/9a

R' R = H, Me R' = H, Me, MeO, F

R 2N BocHN

R' R

ee 94–99 % yield 80–95 %

Scheme 6.26 Synthetic applications of Ferriphos/Mandyphos ligands

groups were prepared by transforming 56 into the corresponding acetate 57 followed by the reaction with secondary amines. Even though the scope of this family is not yet fully explored, screening results52 indicate high enantioselectivities as well as high activity for several Ferriphos/

Chiral 1,2-Disubstituted Ferrocene Diphosphines for Asymmetric Catalysis

225

Mandyphos derivatives in the rhodium-catalysed hydrogenation of dehydro aminoacid derivatives (preferred ligand 9c, ee’s 95 to >99 %, s/c up to 20 00) and the rutheniumcatalysed hydrogenation of tiglic acid (9c, ee 97 %). 9c was also the ligand of choice for the ruthenium-catalysed hydrogenation of various methyl 2-furylacrylates (Scheme 6.26)60 while Rh/Ferriphos/Mandyphos was highly selective for the rhodium-catalysed ring opening of azabicycles with amines (Scheme 6.26).61 6.2.7

TRAP

The TRAP ligands developed by Ito and coworkers62 form nine-membered chelate rings and were conceived as trans-chelating diphosphines. The X-ray structure of several metal complexes has been determined and show that the major isomer has indeed trans-conﬁguration. However, NMR experiments have shown that cis-isomers are also present. The synthesis is depicted in Scheme 6.27 and starts with a derivative of Ugi’s amine 58 which was converted to tertiary phosphine 59 and then oxidised to the corresponding phosphine oxide 60. Homocoupling of 60 with activated copper powder without solvent produced the biferrocene 61, which was ﬁnally reduced with HSiCl3 /NEt3 to give TRAP.

NMe2 Fe

I

Fe

P(O)R2

PR2 H O 2 2 Fe

59

58

H Me

I

P(O)R2 Fe Fe 61

P(O)R2 Me SiHCl3, Et3N H

I

Cu

60

H Me

PR2 Fe Fe

PR2 Me H

TRAP (10)

Scheme 6.27 Synthesis of TRAP

Up to now only a few different PR2 fragments have been tested, but it is clear that the choice of R strongly affects the level of enantioselectivity and sometimes even the sense of induction. The rhodium-catalysed hydrogenation of MAA, MAC and itaconates gives enantioselectivities of 92–96 % if carried out at pressures of 0.5–1 bar.62b, c A number of difﬁcult substrates depicted in Scheme 6.28, such as N -acetylindole derivatives,62a protected β-hydroxy-α-amino and α,β-diamino acid derivatives,63e, f and an indinavir intermediate,63d are effectively hydrogenated by Rh/TRAP complexes. A number of N -Boc-indole derivatives were hydrogenated with 87–96 % ee using Ru/Ph–TRAP catalysts64a and Rh/Alkyl-TRAP complexes were shown to be effective catalysts for the hydrosilylation of substituted acetophenones (ee 80–92 %), FcAc (ee 97 %), α-disubstituted β-ketoesters (ee 98 %), and various diketones (ee 89–99 %).65b Rh/Ph-TRAP complexes catalysed the Michael addition of α-cyanocarboxylates to vinyl ketones with ee’s up to 89 %.66

226

Ferrocenes: Ligands, Materials and Biomolecules R

R N

R'3SiO

R

NHAc

N O

X = OOCt-Bu, NHCbz

R = Me, COOEt Rh / Ph-TRAP ee 94–95 %

N

COOCH3

NHAc

O

PhOOC

X COOCH3

COOtBu

NHt-Bu

Rh / i-Bu-TRAP ee 97 %

Rh / Pr-TRAP ee 82–97 %

Scheme 6.28 Various substrates for hydrogenations with Rh/TRAP complexes

6.2.8

Other diphosphines

Compound 11 (Scheme 6.29), independently developed by Zhang’s67 and Hou’s68 groups, is easily prepared from chiral phosphino ferrocenyl oxazoline 62 (R = i-Pr or t-Bu).69 Hydrolysis of the oxazoline ring in 62 with triﬂic acid followed by acylation gave the amidoester 63, which was converted in high yield to the acid 64 upon treatment with sodium hydroxide or t-BuOK. Condensation of 64 with cyclohexane1,2-diamine in the presence of DCC afforded 11 (shown is the matched conﬁguration R,R,Sp ,Sp ). O

O N PPh2

Fe

R 1) TFA, Na2SO4, H2O

Fe

2) Ac2O, pyridine

62

O PPh2

NHAc or NaOH, THF, MeOH

63

O

Fe 64

t-BuOK, Et2O, H2O

R

OH PPh2

NH2 H2N DCC

CONH HNOC Fe

PPh2

Fe

Ph2P

(R,R,Sp,Sp)-11

Scheme 6.29 Synthesis of ligand 11

Ligand 11 has been tested for a variety of palladium catalysed allylic alkylation reactions with moderate to very high enantioselectivities. Of synthetic interest are the allylation of methyl tetralone68b and the kinetic resolution of cyclohexenyl acetate.67

6.3

Planar and Central Phosphorus Chirality

A small number of bidentate ligands combining planar chirality with stereogenic phosphorus donors have been reported (Scheme 6.30).

Chiral 1,2-Disubstituted Ferrocene Diphosphines for Asymmetric Catalysis

Ar P Ph

Fe

P

Ph Ar

Ph 1-Np

P P

Fe

(Sp,Sp,Sp,Sp)-65c

(Rp,Rp,Sp,Sp)-65 a: Ar = 1-Np b: Ar = biphenyl-2-yl

Ph P

H3 B

Structure of ferrocenyl diphosphines with planar and central phosphorus

OCH3 1) FcLi Ar

Fe

Ph P Ar H2O2 Fe

2) Et2NH

(RP)-66 (SP)-67 Ar = 1-Np, biphenyl-2-yl Ph

Ar P

O I

Cu

Ph Ar

(Sp,RP)-69

Ph P Ar 1) (i-Pr2N)MgBr O 2) I2

(Sp,RP)-69 + (Rp,RP)-69

(RP)-68

O P

Ph Fe Fe

Fe

1-Np Ph

Fe

Fe

Scheme 6.30 chirality

227

Ph P Ar O

(Rp,Rp,RP,RP)-70

1) HSiCl3, Et3N 2) BH3 3) chromatography 4) Et2NH

Ar P Ph

Fe

P

Ar

Fe (Rp,Rp,SP,SP)-65a-b

Ph 1-Np P

I

O

Ph 1-Np

P P

Fe

1-Np Ph

Fe

Fe (Rp,RP)-69

(Sp,Sp,SP,SP)-65c

Scheme 6.31 Synthesis of ligands containing stereogenic phosphorus atoms

Ligands 65 are P-chiral analogs of BIPHEP, ﬁrst described by Sawamura et al.71 The synthesis of 65 as described by Nettekoven et al.72 is illustrated in Scheme 6.31. LiFc was reacted with (R)-phosphinite–borane (66),73 yielding after deprotection conﬁgurationally inversed ferrocenyl monophosphines 67. These were stereoselectively oxidised with hydrogen peroxide, giving the optically pure phosphine oxides 68. Compounds 68 were ortho-magnesiated with (i-Pr2 N)MgBr, followed by quenching with iodine/THF solution at −30 ◦ C to give a mixture of the two ortho-iodo products 69 in good yields. After column chromatographic separation the ratio of the two diastereomers was found to be 75:25 for Ar = 1-Np and 97:3 for Ar = biphenyl-2-yl, respectively. The ortho-iodo ferrocenyl phosphine oxides 69 were subjected to Ullmann coupling at 130 ◦ C using activated copper powder to give the desired biferrocenyl–diphosphine

228

Ferrocenes: Ligands, Materials and Biomolecules

dioxides 70. Complete reduction to the diphosphines 65 required rather harsh reaction conditions (HSiCl3 /toluene/triethylamine at 130 ◦ C for 72 hours) leading to partial epimerisation of the stereogenic phosphorus centres. Isolation and puriﬁcation was achieved by column chromatographic separation of their borane adduct, followed by deprotection using diethylamine to give the P-chiral biferrocenyldiphosphines 65a and 65b. The diastereomeric biferrocenyl diphosphine (Sp ,Sp ,SP ,SP )-65c was prepared from the minor, (Rp ,RP )-conﬁgured ortho-iodo ferrocenylphosphine oxide (Rp ,RP )-69. These ligands were tested in the standard allylic alkylation and amination reactions of 1,3-diphenylallyl acetate with methyl malonate and benzylamine, respectively. Best results were obtained for Pd/65a, which achieved 88 % for the alkylation and 93 % for the amination reaction.

6.4

Planar, Central Carbon and Central Phosphorus Chirality

The main ligand classes discussed above may be extended to include an additional element of central chirality when at least one of the two phosphorus atoms bears two different substituents. Some examples of these relatively rare ligands are shown in Scheme 6.32.

Ph PCy2

P Fe PPh2 Ar

Fe P

N

R2

R2

72

71

Fe

PPh2 Me2N

Fe P

R1

NMe2

Me

P

Ph

Ar

R1

Ph P Ar 74

73 Ph

MeO H

PPh2

H

P

NMe2

Fe

Fe Fe P Ar

Ph

PingFer (75)

Fe

Me2N

P Ph

H

TriFer (76)

Scheme 6.32 Structure of ferrocene diphosphines with planar, central carbon and central phosphorus chirality

The ﬁrst example of a ferrocene-based diphosphine with planar, central carbon and central phosphorus chirality was described by Togni et al.11 Derivatives of 71 (Ar = Cy, 2-anisyl) were prepared in analogly to the Josiphos ligands by reacting PPFOAc with racemic ArPhPH to give 1:1 mixtures of diastereomeric 71 which were separated by column chromatography. Disappointingly, the P-chiral ligands only gave moderate results for the hydrogenation of various C=C bonds.29

Chiral 1,2-Disubstituted Ferrocene Diphosphines for Asymmetric Catalysis

229

Recently, several families of ferrocene-based diphosphine combining the three named elements of chirality were developed by the group of Chen.74 Chen’s strategy for introducing central phosphorus chirality is straightforward and encompasses reaction of a chiral lithiated ferrocene with a dichlorophosphine, followed by a second organometallic reagent (Scheme 6.33). As a rule, the reactions are highly stereoselective and lead to a variety of modular ligands. X* Fe

X* R1PCl2

Li

Fe

X*

P Cl

R2M

1 P R 2 R

Fe

R1

high yield high stereoselectivity

X* = chiral directing group

Scheme 6.33 Strategy for the generation of P-stereogenic ligands

Ligands 71, 72 and 73 are P-chiral analogs of Josiphos and BoPhoz, respectively (Scheme 6.34).74a Accordingly, lithiated Ugi’s amine is reacted with R1 PCl2 , followed by the appropriate organometallic reagent R2 M to afford, in most cases, a single diastereomer 77. As expected, signiﬁcant matched/mismatched effects are observed for the rhodium-catalysed hydrogenation of enamides. On average, 73 with R1 = Ph and R2 = 1-Np gives about 2–3 % better ee’s and slightly higher activities than BoPhoz.

PCy2 Fe NMe2 Fe 1

1) t-BuLi 2) R1PCl2 3)

R2M

P

R1

R2 NMe2 Fe

P

72

R1

R2 77

N Fe

P

R1

PPh2

R2 73

Scheme 6.34 Synthesis of P-stereogenic Josiphos and BoPhoz derivatives

PingFer (75 in Scheme 6.35),74b a P-chiral version of second generation Taniaphos is prepared in high yield with up to >99 % diastereoselectivity via bromine–lithium exchange starting from (Sp ,αS)-50 followed by the reaction with PhPCl2 and then an appropriate organometallic reagent. Also in this case, signiﬁcant matched/mismatched effects are observed for the rhodium-catalysed hydrogenation of enamides. On average, PingFer with Ar = 1-Np gives about 2–3 % better ee than the corresponding

230

Ferrocenes: Ligands, Materials and Biomolecules OMe PPh2

Fe

OMe PPh2 1) t-BuLi 2) PhPCl2

Br

Fe P

3) ArLi

Ar (Sp,αS)-50

Ph

PingFer (75)

Scheme 6.35 Synthesis of P-stereogenic PingFer

Taniaphos. Best results were reported for MAC (99.6 % ee) and brominated ACA (99.9 % ee). C2 -Symmetric ferrocene-based diphosphine ligands with planar, central carbon and central phosphorus chirality can be readily prepared either by reaction of a chiral 1,1 -dilithioferrocenyl species with a dichlorophosphine, followed by a second organometallic reagent or by reaction of the lithiated Ugi’s amine with a dichlorophosphine, followed by a bis-organometallic reagent (Scheme 6.36). In this way, C2 -symmetrical amine 53 was lithiated with t-BuLi, and reacted with two equivalents of PhPCl2 followed by an aryl-lithium reagent to afford a single diastereomer of 74, a P-stereogenic version of Mandyphos. For the preparation of TriFer,74c (R)-1 was lithiated with tBuLi, followed by reaction with PhPCl2 and then with fcLi2 , to afford a mixture of diastereomer (R,R)-(Sp ,Sp )-(SP ,SP )-TriFer and the meso-compound, (R,R)-(Sp ,Sp )(SP ,RP )-diastereomer, in about 95:5 ratio. NMe2 NMe2

1) t-BuLi 2) PhPCl2

Fe

3) ArLi

Fe Me2N

P

Ph

R Ph P Ar 74

NMe2 53

Ph H NMe2 Fe

1

NMe2

1) t-BuLi 2) PhPCl2 3) fcLi2

P

Fe

Fe Fe

P

Me2N Ph TriFer (76)

H

Scheme 6.36 Synthesis of Mandyphos-type ligands containing P-stereogenic donors

TriFer was tested in the commercially important hydrogenation of SPP100-SyA, a sterically demanding α,β-unsaturated carboxylic acid, which is an intermediate in the synthesis of the renin inhibitor SPP100. A bench scale process was developed affording 98–99 % ee. Rh/Trifer complexes were also highly selective (ee’s 95–98 %) for the difﬁcult rhodium-catalysed hydrogenation of a variety of substituted α-ethoxycinnamic acids.

Chiral 1,2-Disubstituted Ferrocene Diphosphines for Asymmetric Catalysis

6.5

231

Planar, Axial and Central Carbon Chirality

Compounds combining the three basic elements of chirality, planar, axial and central, are inherently rare. However, that such compounds are relatively easy to realise starting from 1,2-disubstituted ferrocenes is obvious. Compounds 78a–c, the only representatives of this class, independently developed by Zheng’s75 and Chan’s70 groups, are easily prepared in nearly quantitative yields by the reaction of 79 and 80, respectively, with chlorophosphites 81 in the presence of Et3 N (Scheme 6.37).

O P Cl O Y Fe PPh2 Y = NHR, 79 Y = OH, 80

81

X Fe PPh2

O P

X = NH X = NMe X=O

O

78a 78b 78c

Scheme 6.37 Synthesis of ligands displaying central, axial and planar chirality

These ligands were tested in the rhodium-catalysed hydrogenation of various acetamido oleﬁns. Signiﬁcant matched/mismatched effects were observed for 78a–c (best combination R,Sp ,Rax ), but were less pronounced for 78c. Compound 78a is the ligand of choice for the hydrogenation of a variety of dehydro β-amino acid derivatives (ee’s 96 to >99 %, s/c up to 5000); (E) and (Z) substrates give the opposite enantiomer with similar stereoselectivity.75a 78b is very effective for the hydrogenation of substituted aryl enamides (ee 98–99.6 %, s/c up to 5000), for DMI and MAC (ee’s 99–99.9 %, s/c up to 10 000)75b as well as ACA (ee 99.9 %).41a Finally, 78c achieved ee’s of 97–99.6 % in the hydrogenation of various substituted MAC derivatives.70

6.6

Conclusions

Ferrocenes displaying the 1,2-disubstitution pattern constitute a fundamentally important class of chiral compounds. For more than three decades they have been exploited mainly for the synthesis of enantiomerically pure ligands for asymmetric catalysis and, among such ligands, diphosphines arguably play the most prominent role. This is true not only from a fundamental point of view, but also because a number of ferrocenyl ligands have been already exploited on an industrial scale. In this chapter an attempt has been made to convey the most signiﬁcant and recent aspects of the chemistry of ferrocenyl diphosphines. While sharing our unbroken fascination for ferrocene and its chiral derivatives, it is hoped that inspiration for future work by the steadily growing community of their users has been provided.

232

Ferrocenes: Ligands, Materials and Biomolecules

Abbreviations Boc BSA CBS Cbz Cy DCC LDA morph nbd Np PMHS TFA TMEDA Xyl

t-Butyloxycarbonyl bis(trimethylsilyl)acetamide Corey-Bakshi-Shibata (as referred to the corresponding reduction) benzyl carbamate cyclohexyl dicyclohexylcarbodiimide lithium diisopropylamide N -morpholinyl norbornadiene naphthyl polymethylhydrosiloxane triﬂuroacetic acid N ,N ,N ,N -tetramethylethylenediamine xylyl

References 1. For previous accounts, see: (a) T. Hayashi, Asymmetric catalysis with chiral ferrocenyphosphine ligands in Ferrocenes, A. Togni and T. Hayashi (Eds), VCH, Weinheim, Germany (1995), pp. 105–142; (b) A. Togni, New chiral ferrocenyl ligands for asymmetric catalysis in Metallocenes, A. Togni and R.L. Halterman (Eds), Wiley-VCH Verlag GmbH, Weinheim, Germany (1998), pp. 685–721; (c) M. Perseghini, A. Togni, Organometallic complexes of iron; Product subclass 8: Ferrocenes in Science of Synthesis (Houben-Weyl), Vol. 1, M. Lautens (Ed.), Thieme, Stuttgart, Germany (2001), pp. 889–929. 2. (a) T.J. Colacot, Chem. Rev., 2003, 103, 3101–3118; (b) L-X. Dai, T. Tu, S-L. You et al. Acc. Chem. Res., 2003, 36, 659–667; (c) P. Barbaro, C. Bianchini, G. Giambastiani, S.L. Parisel, Coord. Chem. Rev., 2004, 248, 2131–2150; (d) R.C.J. Atkinson, V.C. Gibson, N.J. Long, Chem. Soc. Rev., 2004, 33, 313–328; (e) R.G. Array´as, J. Adrio, J.C. Carretero, Angew. Chem. Int. Ed., 2006, 45, 7674–7715. 3. D. Marquarding, H. Klusacek, G. Gokel et al. J. Am. Chem. Soc., 1970, 92, 5389–5393. 4. T. Hayashi, A. Yamazaki, J. Organomet. Chem., 1991, 413, 295–302. 5. (a) T. Hayashi, N. Kawamura, Y. Ito, J. Am. Chem. Soc., 1987, 109, 7876–7878; (b) T. Hayashi, N. Kawamura, Y. Ito, Tetrahedron Lett., 1988, 29, 5969–5972. 6. H-U. Blaser, R. Gamboni, G. Rihs et al. Route evaluation for the synthesis of (R)Levoprotiline in Process Chemistry in the Pharmaceutical Industry, K.G. Gadamasetti (Ed.), Marcel Dekker Inc., New York (1999), pp. 189–200. 7. A. Togni, S. Pastor, G. Rihs, Helv. Chim. Acta., 1989, 72, 1471–1478 and unpublished results. 8. A. Togni, C. Breutel, A. Schnyder et al. J. Am. Chem. Soc., 1994, 116, 4062–4066. 9. H-U. Blaser, W. Brieden, B. Pugin et al. Top. Catal., 2002, 19, 3–16. 10. For more information, see: (a) M. Thommen, H.U. Blaser, PharmaChem, 2002, 7/8, 33; (b) http://www.solvias.com/english/products-and-services/chemicals/ligands/chiralligand-kit.html 11. A. Togni, C. Breutel, M.C. Soares et al. Inorg. Chim. Acta, 1994, 222, 213–224. 12. X. Feng, B. Pugin, E. K¨usters et al. Adv. Synth. Catal., 2007, 349, 1803–1806 13. B. Pugin, H. Landert, F. Spindler, H-U. Blaser, Adv. Synth. Catal., 2002, 344, 974–979.

Chiral 1,2-Disubstituted Ferrocene Diphosphines for Asymmetric Catalysis

233

14. (a) C. K¨ollner, B. Pugin, A. Togni, J. Am. Chem. Soc., 1998, 120, 10274–10275; (b) C. K¨ollner, A. Togni, Can. J. Chem., 2001, 79, 1762–1774. 15. For a scalable synthesis of Ugi amine see H-U. Blaser, H.P. Buser, K. Coers et al. Chimia, 1999, 53, 275–280. 16. H-U. Blaser, F. Spindler, M. Thommen, Industrial applications in Handbook of Homogeneous Hydrogenation, J.G. de Vries and C.J. Elsevier (Eds), Wiley-VCH Verlag GmbH, Weinheim, Germany (2006), pp. 1279–1326. 17. (a) M. Rouhi, Chemical & Engineering News, 2004, 82(37), 28–29; (b) Y. Hsiao, N.R. Rivera, T. Rosner et al. J. Am. Chem. Soc., 2004, 126, 9918–9919. 18. (a) C. Czekelius, E.M. Carreira, Angew. Chem. Int. Ed., 2003, 42, 4793–4795; (b) C. Czekelius, E.M. Carreira, Org. Lett., 2004, 6, 4575–4577. 19. (a) B.H. Lipshutz, J.M. Servesko, Angew. Chem. Int. Ed., 2003, 42, 4789–4792; (b) B.H. Lipshutz, J.M. Servesko, B.R. Taft, J. Am. Chem. Soc., 2004, 126, 8352–8353; (c) B.H. Lipshutz, N. Tanaka, B.R. Taft, C-T. Lee, Org. Lett., 2006, 8, 1963–1966. 20. (a) F. L´opez, S.R. Harutyunyan, A. Meetsma et al. Angew. Chem. Int. Ed., 2005, 44, 2752–2756; (b) R. Des Mazery, M. Pullez, F. L´opez et al. J. Am. Chem. Soc., 2005, 127, 9966–9967; (c) B.L. Feringa, R. Badorrey, D. Pena et al. Proc. Nat. Acad. Sci., 2004, 101, 5834–5838. 21. E.A. Bercot, T. Rovis, J. Am. Chem. Soc., 2004, 126, 10248–10249. 22. (a) M. Lautens, K. Fagnou, S. Hiebert, Acc. Chem. Res., 2003, 36, 48–58; (b) M. Lautens, K. Fagnou, Proc. Nat. Acad. Sci., 2004, 101, 5455–5460. 23. J.F. Carpentier, V. Bette, Curr. Org. Chem., 2002, 6, 913–936 (review). 24. D. Lee, D. Kim, J. Yun, Angew. Chem. Int. Ed., 2006, 45, 2785–2787. 25. B.M. Andresen, S. Caron, M. Couturier et al. Chimia, 2006, 60, 554–560. 26. K. Makino, T. Fujii, Y. Hamada, Tetrahedron: Asymmetry, 2006, 17, 481–485. 27. C.S. Shultz, S.D. Dreher, N. Ikemoto et al. Org. Lett., 2005, 7, 3405–3408. 28. M. Pucheault, S. Darses, J-P. Genˆet, Eur. J. Org. Chem., 2002, 3552–3557. 29. Solvias AG, unpublished results. 30. A. Leitner, J. Larsen, C. Steffen, J.F. Hartwig, J. Org. Chem., 2004, 69, 7552–7557. 31. M.E.P. Lormann, M. Nieger, S. Br¨ase, J. Organomet. Chem., 2006, 691, 2159–2161. 32. Y. Yamamoto, S. Takada, N. Miyaura, Chem. Lett., 2006, 35, 1368–1369. 33. C. Godard, A. Ruiz, C. Claver, Helv. Chim. Acta, 2006, 89, 1610–1622. 34. G. Xu, L-B. Han, Org. Lett., 2006, 8, 2099–2101. 35. W.P. Hems, P. McMorn, S. Riddel et al. Org. Biomol. Chem., 2005, 3, 1547–1550. 36. P. Stephenson, B. Kondor, P. Licence et al. Adv. Synth. Catal., 2006, 348, 1605–1610. 37. T. Sturm, W. Weissensteiner, F. Spindler et al. Organometallics, 2002, 21, 1766–1774. 38. (a) A. Mernyi, C. Kratky, W. Weissensteiner, M. Widhalm, J. Organomet. Chem., 1996, 508, 209–218; (b) E.M. Cayuela, L. Xiao, T. Sturm et al. Tetrahedron: Asymmetry, 2000, 11, 861–869. 39. T. Sturm, B. Abad, W. Weissensteiner et al. J. Mol. Catal. A: General , 2006, 255, 209–219. 40. (a) N.W. Boaz, D.D. Debenham, E.B. Mackenzie, S.E. Large, Org. Lett., 2002, 4, 2421–242; (b) N.W. Boaz, E.B. Mackenzie, S.D. Debenham et al. J. Org. Chem., 2005, 70, 1872–1880. 41. (a) N.W. Boaz, J.A. Jr. Ponasik, S.E. Large, Tetrahedron: Asymmetry, 2005, 16, 2063–2066; (b) N.W. Boaz, S.D. Debenham, S.E. Large, M.K. Moore, Tetrahedron: Asymmetry, 2003, 14, 3575–3580; (c) N.W. Boaz, J.A. Posanik, S.E. Large, Tetrahedron Lett., 2006, 47, 4033–4035. 42. X. Li, X. Jia, L. Xu et al. Adv. Synth. Catal., 2005, 347, 1904–1908. 43. H-U. Blaser, Adv. Synth. Catal., 2002, 344, 17–31. For more information see Reference 9. 44. (a) T. Sturm, L. Xiao, W. Weissensteiner, Chimia, 2001, 55, 688–693; (b) T. Sturm, W. Weissensteiner, F. Spindler, Adv. Synth. Catal., 2003, 345, 160–164.

234

Ferrocenes: Ligands, Materials and Biomolecules

45. (a) J.B. Morgan, J.P. Morken, J. Am. Chem. Soc., 2004, 126, 15338–15339; (b) W.J. Moran, J.P. Morken, Org. Lett., 2006, 6, 2413–2415. 46. http://www.jmcatalysts.com/pct/producttype.asp?producttypeid=265 47. Solvias AG, unpublished screening results. 48. K. Tanaka, Y. Hagiwara, K. Noguchi, Angew. Chem. Int. Ed., 2005, 44, 7260–7263. 49. J-R. Kong, M-Y. Ngai, M.J. Krische, J. Am. Chem. Soc., 2006, 128, 718–719. 50. (a) T. Ireland, G. Grossheimann, C. Wieser-Jeunesse, P. Knochel, Angew. Chem, Int. Ed., 1999, 38, 3212–3215; (b) T. Ireland, K. Tappe, G. Grossheimann, P. Knochel, Chem. Eur. J., 2002, 8, 843–852; (c) M. Lotz, K. Polborn, P. Knochel, Angew. Chem, Int. Ed., 2002, 41, 4708–4711; (d) K. Tappe, P. Knochel, Tetrahedron: Asymmetry, 2004, 15, 91–102. 51. W. Chen, S.M. Roberts, J. Whittall, A. Steiner, Chem. Commun., 2006, 2916–2918. 52. F. Spindler, C. Malan, M. Lotz et al. Tetrahedron: Asymmetry, 2004, 15, 2299–2306. 53. M. Lautens, K. Fagnou, V. Zunic, Org. Lett., 2002, 4, 3465–3468. 54. (a) J. Almena Perea, A. B¨orner, P. Knochel, Tetrahedron Lett., 1998, 39, 8073–8076; (b) M. Lotz, T. Ireland, J. Almena Perea, P. Knochel, Tetrahedron: Asymmetry, 1999, 10, 1839–1842; (c) J. Almena Perea, M. Lotz, P. Knochel, Tetrahedron: Asymmetry, 1999, 10, 375–384. 55. (a) O. Chuzel, J. Deschamp, C. Chausteur, O. Riant, Org. Lett., 2006, 8, 5943–5946; (b) J. Deschamp, O. Chuzel, J. Hannedouche, O. Riant, Angew. Chem. Int. Ed., 2006, 45, 1292–1297. 56. D. Zhao, K. Oisaki, M. Kanai, M. Shibasaki, J. Am. Chem. Soc., 2006, 128, 14440–14441. 57. K. Oisaki, D. Zhao, M. Kanai, M. Shibasaki, J. Am. Chem. Soc., 2006, 128, 7164–7165. 58. T. Llamas, R. G´omez Array´as, J.C. Carretero, Org. Lett., 2006, 8, 1795–1798. 59. (a) F. L´opez, A.W. van Zijl, A.J. Minnaard, B.L. Feringa, Chem. Commun., 2006, 409–411; (b) K. Geurts, S.P. Fletcher, B.L. Feringa, J. Am. Chem. Soc., 2006, 128, 15572–15573. 60. A.S.K. Hashmi, P. Haufe, C. Schmid et al. Chem. Eur. J., 2006, 12, 5376–5382. 61. Y-h. Cho, V. Zunic, H. Senboku et al. J. Am. Chem. Soc., 2006, 128, 6837–6846. 62. (a) M. Sawamura, H. Hamashima, Y. Ito, Tetrahedron: Asymmetry, 1991, 2, 593–596; (b) M. Sawamura, H. Hamashima, M. Sugawara et al. Organometallics, 1995, 14, 4549–4558. (c) R. Kuwano, M. Sawamura, S. Okuda et al. Bull. Chem. Soc. Jpn., 1997, 70, 2807–2822. 63. (a) R. Kuwano, K. Sato, T. Kurokawa et al. J. Am. Chem. Soc., 2000, 122, 7614–7615; (b) R. Kuwano, M. Sawamura, Y. Ito, Bull. Chem. Soc. Jpn., 2000, 73, 2571–2578; (c) R. Kuwano, M. Sawamura, Y. Ito, Tetrahedron: Asymmetry, 1995, 6, 2521–2526; (d) R. Kuwano, Y. Ito, J. Org. Chem., 1999, 64, 1232–1237; (e) R. Kuwano, S. Okuda, Y. Ito, J. Org. Chem., 1998, 63, 3499–3503; (f) R. Kuwano, S. Okuda, Y. Ito, Tetrahedron: Asymmetry, 1998, 9, 2773–2775. 64. (a) R. Kuwano, M. Kashiwabara, Org. Lett., 2006, 8, 2653–2655; (b) R. Kuwano, M. Sawamura, Y. Ito, Bull. Chem. Soc. Jpn., 2000, 73, 2571–2578. 65. T. Hayashi, A. Yamamoto, M. Hojo, Y. Ito, Chem. Commun., 1989, 495–496. 66. M. Sawamura, H. Hamashima, Y. Ito, J. Am. Chem. Soc., 1992, 114, 8295–8296. 67. J.M. Longmire, B. Wang, X. Zhang, Tetrahedron Lett., 2000, 41, 5435–5439. 68. (a) S-L. You, X-L. Hou, L-X. Dai et al. Chem. Commun., 2000, 1933–1934; (b) S-L. You, X-L. Hou, L-X. Dai, X-Z. Zhu, Org. Lett., 2001, 3, 149–151; (c) S-L. You, X-L. Hou, LX. Dai, J. Organomet. Chem., 2001, 637–639, 762–766. 69. (a) C.J. Richards, T. Damalidis, D.E. Hibbs, M.B. Hursthouse, Synlett , 1995, 74–76; (b) Y. Nishibayashi, S. Uemura, Synlett , 1995, 79–81. 70. X. Jia, X. Li, W.S. Lam et al. Tetrahedron: Asymmetry, 2004, 15, 2273–2278. 71. M. Sawamura, A. Yamauchi, T. Takegawa, Y. Ito, Chem. Commun., 1991, 874–875. 72. (a) U. Nettekoven, M. Widhalm, P.C.J. Kamer et al. Organometallics, 2000, 19, 2299–2309; (b) U. Nettekoven, M. Widhalm, H. Kalchhauser et al. J. Org. Chem., 2001, 66, 759–770.

Chiral 1,2-Disubstituted Ferrocene Diphosphines for Asymmetric Catalysis

235

73. U. Nettekoven, P.C.J. Kamer, P.W.N.M. van Leeuwen et al. J. Org. Chem., 1999, 64, 3996–4004. 74. (a) W. Chen, W. Mbafor, S.M. Roberts, J. Whittall, J. Am. Chem. Soc., 2006, 128, 3922–3923; (b) W. Chen, S.M. Roberts, J. Whittall, A. Steiner, Chem. Commun., 2006, 2916–2918; (c) W. Chen, P.J. McCormack, K. Mohammed et al. Angew. Chem Int. Ed., 2007, 46, 4141–4144. 75. (a) X-P. Hu, Z. Zheng, Org. Lett., 2005, 7, 419–422; (b) X. Hu, Z. Zheng, Org. Lett., 2004, 6, 3585–3588.

7 Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors ˇ epniˇcka and Martin Lamaˇc Petr Stˇ

7.1

Introduction

Chiral ferrocene ligands are among the most frequently used catalyst components in common laboratory practice and have found applications even in industry. This chapter, the last among those dealing with ferrocene ligands, attempts to brieﬂy summarise those synthetic routes to planar chiral ferrocene donors that have not been dealt with in detail in the previous chapters. However, since many of the areas covered here have been extensively reviewed, this chapter provides only the necessary introductory information and important and recent examples from each particular ﬁeld.

7.2 7.2.1

Synthetic Routes Leading to Planar Chiral Ferrocene Compounds Ugi’s Amine and Related N-directing Groups

The preparation of planar chiral, 1,2-difunctionalised ferrocene compounds is typically achieved by diastereoselective ortho-lithiation of suitable ferrocene derivatives followed by reaction of the lithium salts formed with appropriate electrophiles. With the properly selected chiral directing group or chiral bases, such methodology is, typically, both high yielding and stereoselective, producing compounds amenable to Ferrocenes: Ligands, Materials and Biomolecules  2008 John Wiley & Sons, Ltd

Edited by Petr Stepnicka

238

Ferrocenes: Ligands, Materials and Biomolecules

further synthetic transformations. The scope of the accessible compounds can be further widened by manipulating the chiral auxiliary, which opens access to many derivatives and also to planar-only chiral ferrocenes. A real breakthrough in the synthesis of enantiopure planar chiral ferrocenes came with the discovery of the highly diastereoselective ortho-lithiation of C-chiral [1(N ,N -dimethylamino)ethyl]ferrocene (1; so-called Ugi’s amine). Amine 1, which is now readily available in optically pure form by CBS reduction of acetylferrocene and amination,1, 2 can be not only selectively lithiated/functionalised to give 2-mono or 1 ,2-dilithio derivatives3 but can also be subjected to stereoconservative replacement of the amino group with various nucleophiles.4 Typical reactions are summarised in

Me O Fe

Me OH

CBS

Fe

2. Me2NH 1. acetylation Me NMe2 Fe

1

1. LiBu (2 equiv.)/TMEDA 2. electrophile (sequential functionalisation with different electrophiles is possible)

1. LiBu (1 equiv.) 2. electrophile

Me

Me

E1

Fe

NMe2

1. LiBu 2. (E2)+

Fe

Me

NMe2

E1

Fe

E2 MeI Me

Fe

E1 E2

Nu

E2

Ac2O Me

I

NMe3 OR

E1

Fe

E1

nucleophile (Nu)

OAc

E2

Scheme 7.1 Summary of important synthetic transformations involving Ugi’s amine (1)

Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors

239

Scheme 7.1. The reaction chemistry and examples of donors synthesised from 1 can be found in numerous review articles5, 6 and also in Chapter 6. The concept of diastereoselective ortho-metalation was extended to alkoxy analogues of Ugi’s amine (i.e. FcCH(R)OMe)7 and, particularly, to ferrocenylmethyl amines bearing chiral substituents at the nitrogen atom, FcCH2 (NRR)∗ , where (NRR)∗ is (S)-NMe[CH(Me)Ph] (2),8 (S)-2-methylpiperidinyl (3),9 (2R,5R)-2,5-dimethylpyrrolidinyl (4),10 (S)-2-methoxymethylpyrrolidinyl (5),11 2-pyrrolidinyl (6, R = Me, iPr, t-Bu, allyl),12 axially chiral azepine moiety (7)13 and ephedrine-based auxiliaries such in 8 (Scheme 7.2).8, 14

N

N

Ph

N

N

OMe

Fe

Fe

Fe

Fe

2

3

4

5

Ph N

N

O N Fe

6

R

Fe

Fe

7

8

Scheme 7.2

The successful application of the directed metalation approach naturally prompted interest in the utility of other chiral auxiliaries. Enders et al. have reported that lithiation of benzoylferrocene-SAMP hydrazone (SAMP = (S)-1-amino-2-methoxymetylpyrrolidine) followed by trapping of the lithiated intermediate with electrophiles afford ortho-functionalised products 9 in good yields and with excellent diastereoselectivity. The ketone moiety was subsequently restored to give 10 (Scheme 7.3).15, 16 Acylferrocenes FcC(O)R, where R = i-Pr and Cy, reacted similarly. By contrast, SAMPhydrazones of propionyl and butyrylferrocene were ﬁrst deprotonated at activated Cα of the aliphatic substituent and then at the ferrocene unit. Both lithiation steps are highly diastereoselective (de ≥ 96 and 90 %, respectively) and, in combination with reductive removal of the chiral auxiliary (via the corresponding hydrazine), allow

240

Ferrocenes: Ligands, Materials and Biomolecules FcC(O)R SAMP AlMe3 Ph

R = Ph N

N

1. LiBu

Fc

Fe

2. EX

R

E

C(O)Ph

N N Fe

E

MeO

OMe 9

10

R = Et, Pr 1. LiBu (R1 = Me, Et) 2. E1X R1

R1

E1 N

N

Fe MeO

1. LiBu 2. E2X

Fe

E2

R1

E1 N N Fe

E1

E2

MeO 11

Scheme 7.3 Preparation of planar chiral ferrocenes via SAMP hydrazones [E-X = I2 , Me-I, Me3 Si-Cl, PPh2 P-Cl, Ph2 CO (E = Ph2 C(OH)), Me2 NCHO (E = CHO), E1 /E2 = various SR and PR2 groups]

for the preparation of planar chiral ferrocenes with chiral, donor-functionalised side chains 11 (Scheme 7.3). The catalytic potential of such donors has been demonstrated in enantioselective allylic alkylation.17, 18 The related imines (S)-FcC(R1 )=NCH(R2 )CH2 OMe (R1 = Me, Et; R2 = t-Bu, iBu, and i-Pr) resulting from the condensation of ketones FcC(O)R1 with (S)-chiral β-aminoalcohols and methylation at the terminal hydroxyl group can also be ortholithiated (de > 90 %, (Rp )-isomer favoured). Reaction with electrophiles followed by hydrolytic removal of the chiral auxiliary affords the corresponding planar chiral ferrocenyl ketones.19 7.2.2

Planar Chiral Ferrocene Oxazolines

The oxazoline (dihydrooxazole) group is another moiety that combines excellent orthodirecting ability with donor properties. Currently, ferrocenyloxazolines constitute a prominent class of ferrocene derivatives that have been successfully applied to catalysis.20 Although some reports describing the preparation of 2-ferrocenyloxazolines appeared in the 1980s,21 rapid development did not commence until 1995 when a convenient synthetic procedure was discovered independently by three research groups (Scheme 7.4). The protocols of Richards22 and Sammakia23 are based on the reaction of ferrocenoyl chloride with chiral β-aminoalcohols to yield the corresponding hydroxyamides 12. The subsequent cyclisation to oxazolines 13 is achieved either

Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors

FcCO2H

241

NEt3

FcCOCl

O

R OH

H2N FcCO2R

R

Fc

OH

N H 12

Me3Al or base

PPh3/CCl4/NEt3 or TsCl/NEt3 (base catalyst) or MsCl/NEt3 O R

FcCN

H2N

N

OH

R

Fe

ZnCl2/ 13

Scheme 7.4

by action of PPh3 /CCl4 and NEt3 or with p-toluenesulfonyl chloride/NEt3 . Another modiﬁcation of the cyclisation step involves methanesulfonylation of the hydroxyl group.24 Amides 12 are accessible also by amidation of simple ferrocenecarboxylic esters.25 An alternative approach to 13, though less efﬁcient, has been described by Uemura.26 It is based on the condensation of cyanoferrocene with β-aminoalcohols in the presence of catalytic amount of ZnCl2 , which leads directly to the desired compounds (Scheme 7.4). The ortho-lithiation of C-chiral ferrocenyloxazolines has been described simultaneously by Richards,22a Sammakia23a and Uemura26a (Scheme 7.5). Varying distribution of diastereomeric products was obtained using different conditions (temperature, solvent, lithium alkyls) for the ortho-lithiation of 4-i-propyl-2-ferrocenyloxazoline and subsequent trapping with Me3 SiCl. Finally, the selectivity was increased up to >500:1 using s-BuLi in n-hexanes with TMEDA as an additive.23b Sammakia has also shown that the ortho-substitution is directed exclusively by the nitrogen atom.23c Until now, a large array of electrophiles has been used and various optically pure planar chiral derivatives have been prepared. An access to (S,Rp )-diastereomers was provided by introducing the trimethylsilyl moiety as a temporary protecting group, followed by metalation/functionalisation and removal of the silyl group with Bu4 NF.22, 25 It should be noted that diastereoselective lithiation of analogous imidazolines has been also reported.27 The ortho-lithiation of 1,1 -bis(oxazolinyl)ferrocenes has been reported, too.28, 24b However, it proceeds with lower diastereoselectivity, leading to mixtures of products with their distribution strongly dependent on the reaction conditions. On the other hand, bis(oxazolinyl)biferrocenes have been successfully ortho-functionalised and ligands thereby obtained have been used in asymmetric catalysis.29 As already indicated, chiral ferrocene oxazolines have been frequently used as ligands in enantioselective, metal-mediated organic reactions. A brief summary is

242

Ferrocenes: Ligands, Materials and Biomolecules

R′

O

O R

N Li

Fe

N Fe

R

Li

O N Fe

major

R R'

LiR'

R Li

13

Li

N O

Fe

O N

R

Fe

minor

Scheme 7.5 Plausible mechanism of enantioselective lithiation of oxazolines 13

provided in Table 7.1, while the ligand types are shown in Scheme 7.6. In general, the stereodiscrimination and enantioselectivities achieved with chiral ferrocene oxazolines vary over a wide range. Hence, a careful tailoring of the catalytic system (the ligand in particular) is vital to achieve good results. The application of oxazoline-based catalysts in C−C bond forming reactions can be demonstrated by cross-coupling of Grignard reagents with alkenyl halides (Table 7.1, entry 1),30 palladium-catalysed conjugate addition of Grignard reagents to enones (Table 7.1, entry 2),31 nickel-mediated coupling of allylic substrates with Grignard reagents32, 33 and with arylboronic acids (Table 7.1, entry 3).34 However, the most frequently studied catalytic C−C bond forming reactions are enantioselective addition of organozinc reagents to carbonyl compounds (Table 7.1, entry 4),35 and palladium-catalysed allylic alkylation36 and amination reactions (Table 7.1, entries 5 and 6).37, 36c, 36l The use of ferrocene oxazolines in palladium-catalysed asymmetric Heck reaction is also well documented (Table 7.1, entries 7 and 8).38, 36l Yet another broad application ﬁeld comprises reductions of unsaturated substrates or reduction-like reactions. Typical examples comprise asymmetric hydrogenation of ketones (ruthenium complexes; Table 7.1, entry 9),39 asymmetric hydrogenation of (E)-α-phenylcinnamic acid over a heterogenised ruthenium catalyst,40 and iridiumcatalysed asymmetric hydrogenation of quinolines (Table 7.1, entry 10).41 Ferrocene oxazoline proved efﬁcient also in asymmetric transfer hydrogenation of ketones (Table 7.1, entry 11)42, 36f and in the related, ruthenium-catalysed oxidative kinetic resolution of racemic secondary alcohols.43 Furthermore, the oxazolines have been tested in asymmetric hydrosilylation of ketones,26a–b, 44 imines,44b, 45 and ketoximes with ruthenium, rhodium and iridium based catalysts (Table 7.1, entries 12 and 13).46 The

4

3

2

1

Entry

O

R

MgCl

Y R′

CuI

RMgCl

+ Br

O

H

ZnR′2

R

O

R

OH

R′

[Ni]

ArB(OH)2

[Ni]

R

R

R

[Ni or Pd]

RMgX

(other R′ sources also tested)

R

Y = OAc, OCO2Me, OMe, OPh, OH, Br, OP(O)(OEt)2

Ph

Me

Reaction

Ph

Ar

R

R′

R′

Me R

30

14 = i -Pr, A = PPh2 , R2 = B = H)

35b, c, d, f, g, k, m, n, p, q 35b, d, f

14 (R1 = t -Bu, A = CPh2 (OH), R2 = B = H) 14 (R1 = Ph, A = CPh2 (OH), R2 = B = H)

(continued overleaf )

35a

32, 34

14 (R1 = i -Pr, A = CHO, R2 = B = H)

14 (R1 = i -Pr, t -Bu, CH2 Ph, Ph, A = PPh2 , R2 = B = H); (S ,S )-15 (R1 = R2 = Ph, A = PPh2 , B = H)

14 (R1 = i -Pr, Me, CH2 Ph, Ph, t -Bu, A = PPh2 , R2 = B = H) 31

Ref.

(R1

Ligand

Table 7.1 Catalytic use of chiral ferrocene oxazolines Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors 243

Entry Reaction

Table 7.1 (continued )

14 (R1 = t -Bu, i -Pr, CH2 Ph, Ph, A = SiMe2 (OH), Si(i -Pr)2 OH, SiPh2 (OH), R2 = B = H)

16 (R1 = i -Pr, t -Bu, CH2 Ph, A = Me, C = CPh2 (OH), R2 = B = H); 16 (R2 = Ph, A = Me, C = CPh2 (OH), R1 = B = H); 16 (R1 = i -Pr, t -Bu, B = Me, C = CPh2 (OH), R2 = A = H)

17 (R1 = R3 = i -Pr, A = D = CPh2 (OH), R2 = R4 = B = C = H); 17 (R1 = R3 = i -Pr, t -Bu, A = C = CPh2 (OH), R2 = R4 = B = D = H)

17 (R1 = R3 = t -Bu, A = D = CPh2 (OH), R2 = R4 = B = C = H)

14 (R1 = t -Bu, B = CPh2 (OH), R2 = A = H); 14 (R1 = R2 = Me, A = CPh2 (OH), B = H)

Ligand

35o

35l

35e

35e, h

35b, d

Ref.

244 Ferrocenes: Ligands, Materials and Biomolecules

5

R

R′

OAc [Pd]/base

CH2(CO2Me)2 R

MeO2C R′

CO2Me

36f

36c, i, f

14 (R1 = R2 = H, Me, CH2 Ph, A = SPh, B = H) 14 (R1 = i -Pr, t -Bu, CH2 Ph, A = SPh, SMe, S(C6 H4 Me-4), R2 = B = H); 14 (R1 = i -Pr, B = SPh, R2 = A = H)

36j

36c

(continued overleaf )

14 (R1 = i -Pr, A = Me3 Si, B = SPh, R2 = H);14 (R1 = i -Pr, t -Bu, CH2 Ph, A = SePh, R2 = B = H);14 (R1 = t -Bu, B = SePh, R2 = A = H);

14 (R2 = Ph, B = SPh, R1 = A = H); 14 (R1 = t -Bu, B = SPh, R2 = A = H)

36i

36c

14 (R1 = R2 = H, Me, CH2 Ph, A = PPh2 , B = H)

14 (R1 = i -Pr, A = Me3 Si, B = SPh, R2 = H)

36a, c

36a, b, c

14 (R1 = t -Bu, B = PPh2 , R2 = A = H)

14 (R1 = i -Pr, t -Bu, A = PPh2 , R2 = B = H)

Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors 245

Entry

Reaction

Table 7.1 (continued ) 2

36d

36a

17 (R1 = R3 = t -Bu, A = D = PPh2 , R2 = R4 = B = C = H) 17 (R1 = R3 = i -Pr, A = B = C = D = PPh2 , R2 = R4 = H); 17 (R1 = R3 = i -Pr, B = C = PPh2 , A = Me3 Si, Et3 Si, Ph3 Si, R2 = R4 = D = H);

36a, b

36k, l

36e, f, l

Ref.

17 (R1 = R3 = i -Pr, t -Bu, B = C = PPh2 , R2 = R4 = A = D = H);17 (R1 = R3 = i -Pr, t -Bu, B = C = PPh2 , R2 = R4 = A = D = H)

16 (A = Me3 Si, C = PPh2 , R1 = R2 = H, Me, B = H); 16 (B = Me3 Si, C = PPh2 , R1 = R2 = H, Me, A = H)

16 (R1 = i -Pr, A = Me3 Si, Me, Bu3 Sn, C = PPh2 , R2 = B = H); 16 (R1 = i -Pr, B = Me3 Si, Me, C = PPh2 , R2 = A = H)

14 (R = Ph, B = SePh, R2 = A = H)

Ligand

246 Ferrocenes: Ligands, Materials and Biomolecules

6

R′

Y = OAc, OCO2Et

R

Y [Pd]

PhCH2NH2 R

HN R′

Ph

(continued overleaf )

36c, l

37a

36n

19 (E = PPh2 , P(t -Bu)2 , SPh, R = i -Pr, t -Bu, CH2 Ph) 14 (R1 = i -Pr, t -Bu, A = PPh2 , R2 = B = H) 14 (R1 = t -Bu, A = PPh2 , R2 = B = H); 14 (R1 = t -Bu, B = PPh2 , R2 = A = H); 14 (R1 = R2 = H, Me, CH2 Ph, A = PPh2 , B = H)

36h

18 (E = PPh2 , SPh)

17 (R1 = R3 = i -Pr, A = B = C = PPh2 , R2 = R4 = D = H); 17 (R1 = R3 = i -Pr, A = B = D = PPh2 , R2 = R4 = C = H); 17 (R1 = R3 = i -Pr, A = D = PPh2 , B = Me3 Si, R2 = R4 = C = H); 17 (R1 = R3 = i -Pr, B = C = PPh2 , A = D = Me3 Si, R2 = R4 = H)

Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors 247

7

Entry

R

R

R = H, Me, Et

O

Reaction

R′OTf [Pd] R′

O R

R

Table 7.1 (continued )

16 (R1 = t -Bu, A = PAr2 (Ar = Ph; C6 H3 (CF3 )2 -3,5; C6 H3 Me2 -3,5; C6 H4 CF3 -4; C6 H4 OMe-4), C = PPh2 , R2 = B = H); 16 (R1 = t -Bu, A = PPh2 , C = PAr2 , R2 = B = H)

16 (R1 = CH2 Ph, A = Me3 Si, C = PPh2 , R2 = B = H); 16 (R1 = CH2 Ph, B = Me3 Si, Me, C = PPh2 , R2 = A = H); 16 (R2 = Ph, A = Me, C = PPh2 , R1 = B = H); 16 (R2 = Ph, B = Me3 Si, C = PPh2 , R1 = A = H)

14 (R1 = i -Pr, t -Bu, A = PPh2 , R2 = B = H)

(R1

16 = i -Pr, A = Me3 Si, C = PPh2 , R2 = B = H); 16 (R1 = i -Pr, B = Me, C = PPh2 , R2 = A = H); 16 (R1 = R2 = Me, A = Me3 Si, C = PPh2 , B = H)

Ligand

38f

38e

38a, b

36l

Ref.

248 Ferrocenes: Ligands, Materials and Biomolecules

11

10

9

8

R

R

Ar

O

O

R′

R

CO2Me

N

R′

R′′OH/[Ru]

N

H2/[Ru]

ArOTf [Pd]

OH

OH

R′

R

H2/[Ir]

R

Ar

Ar

CO2Me

N

R

N H

R′

42a, c, d

14 (R1 = Me, i -Pr, t -Bu, s-Bu, CH2 Ph, Ph, A = PPh2 , R2 = B = H)

(continued overleaf )

36f

14 (R1 = i -Pr, A = PPh2 , R2 = B = H); 14 (R1 = i -Pr, B = PPh2 , R2 = A = H)

41

39

14 (R1 = i -Pr, t -Bu, Ph, A = PAr2 (Ar = Ph; C6 H3 (CF3 )2 -3,5; C6 H3 Me2 -3,5; C6 H4 CF3 -4; 3,5-Me2 -4-(MeO)-C6 H2 ), R2 = B = H) 14 (R1 = i -Pr, t -Bu, CH2 Ph, Ph, A = PPh2 , R2 = B = H); 14 (R1 = t -Bu, B = PPh2 , R2 = A = H)

38g

16 (R1 = i -Pr, t -Bu, CH2 Ph, Ph, A = C = PPh2 , R2 = B = H); 16 (R1 = i -Pr, B = C = PPh2 , R2 = A = H)

Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors 249

13

12

Entry

R

R

R′

NR′′

O

R′

Reaction

Ph2SiH2/[M]

then H+

Ph2SiH2/[M]

R

R

R′ HNR′′

OH

R′

Table 7.1 (continued )

42a, c, d

26a, 44b

26b, 44a

14 (R = Me, i -Pr, t -Bu, s-Bu, CH2 Ph, Ph, A = PPh2 , R2 = B = H) 14 (R1 = Ph, A = PPh2 , R2 = B = H) (S ,S )- and (R,R)-15 (R1 = R2 = Ph, A = PPh2 , B = H)

14 (R1 = i -Pr, A = PPh2 , R2 = B = H)

14 (R1 = t -Bu, A = PPh2 , R2 = B = H) (S ,S )-15 (R1 = R2 = Ph, A = PPh2 , B = H)

14 (R1 = Ph, A = PPh2 , R2 = B = H)

14 (R1 = i -Pr, t -Bu, A = PPh2 , R2 = B = H)

45, 46

45

44b, 45, 46

44b

Ref.

1

Ligand

250 Ferrocenes: Ligands, Materials and Biomolecules

15

14

Ar

R′′

R

RN

N

+

R′

CO2Me

CO2R′

O

[Cu or Ag]

R′′

RN

R′ O

Ar

R′O2C

CO2Me

CO2Me

N

R

14 (R1 = i -Pr, A = PAr2 (Ar = Ph; C6 H3 (CF3 )2 -3,5; C6 H3 Me2 -3,5; C6 H4 CF3 -4; C6 H4 OMe-4), R2 = B = H); 16 (R1 = i -Pr, A = C = PPh2 , R2 = B = H)

14 (R1 = i -Pr, t -Bu, CH2 Ph, Ph, H A = PPh2 , R2 = B = H); 14 (R1 = CH2 Ph, A = PAr2 (Ar = Ph; C6 H3 (CF3 )2 -3,5; C6 H3 Me2 -3,5; C6 H4 CF3 -4; C6 H4 OMe-4), R2 = B = H); 14 (R1 = CH2 Ph, B = PPh2 , R2 = A = H)

14 (R1 = i -Pr, t -Bu, CH2 Ph, Ph, A = PAr2 (Ar = Ph; C6 H3 Me2 -3,5), R2 = B = H)

dimeric palladacycles 20 (see Scheme 7.8)

54

53d

53c

50

Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors 251

252

Ferrocenes: Ligands, Materials and Biomolecules R2 R2

O

B

R1

N Fe

O

B

A

N Fe

R2

O

B R1

N

A

Fe

R1

A C

14

15

R2

O

B

R1

N Fe

E

A D

Fe

N

C

O

O

Ph N

O

17

16

N O

(t-Bu)Me2Si

Fe

R

E

R4 R3

18

19

Scheme 7.6

hydrosilylation of ketones with polymethylhydrosiloxane (PMHS) was shown to proceed also in the presence of [2-(oxazolinyl)ferrocenyl]thiolate zinc complexes.47 Of practical interest is also a sequential, one-pot alkylative reduction of aromatic ketones with primary alcohols, producing homologated secondary alcohols in good yields and with excellent ee’s (Scheme 7.7).48 O + RCH2OH Ar

OH

1. [{IrCl(cod)}2], PPh3, KOH 2. [RuCl2(L)(PPh3)], i-PrOH/i-PrONa L = chiral ferrocene oxazoline

Ar

*

R

Scheme 7.7

Other examples include asymmetric methoxyselenation of alkenes,49 rearrangement of allylic imidates to N-allylamides performed in the presence of oxazoline palladacycles 20 (Scheme 7.8) as catalyst precursors (Table 7.1, entry 14),50 and enantioselective, palladium-catalysed ring opening of aza- and oxabicyclic alkenes with organozinc reagents.51 When combined with metal Lewis acids, the oxazolines induce poor to moderate enantiodiscrimination in asymmetric Diels–Alder reaction52 copper(I)- and silver(I)–oxazoline systems form very efﬁcient catalysts for enantioselective [3 + 2] cycloadditions (Table 7.1, entry 15).53, 54 The potential of chiral ferrocene oxazolines to act as Lewis bases has been demonstrated in asymmetric aza-Baylis–Hillman reaction of N-sulfonated imines with activated alkenes, though with only moderate stereodiscrimination (Scheme 7.9).55

Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors Y

O R

N Pd

I

Fe I

Pd N Y Fe

253

R

O

20

Scheme 7.8 Palladacyles used as catalysts for rearrangement of allylic imidates [Y = H, SiR3 , aryl; R = t -Bu or CEt2 (OMe)] O R′

+ ArCH

HNTs

chiral Lewis base

NR

Ar

*

R′

Scheme 7.9

Apart from being valuable catalyst components, ferrocene oxazolines can serve as useful synthons in the preparation of other ferrocene compounds. As the oxazoline group represents a versatile carboxyl protecting group, the oxazolines are typically used for the preparation of substituted ferrocenecarboxylic acids.56 This approach leads to numerous valuable synthetic building blocks and interesting ligands including planar chiral 2-(diphenylphosphino)ferrocenecarboxylic acid,57 which was used as a chiral auxiliary58 and in the preparation of other chiral phosphinoferrocene ligands.56a, 56d, 57a, 59 Other examples include a general route to racemic, 2-substituted ferrocenecarboxaldehydes60 and the preparation of a 2-(oxazolinyl)ferrocenyl Nheterocyclic carbene.61 7.2.3

O-Donor Directing Groups

In addition to the numerous N-donor moieties, there is also a wide choice of efﬁcient O-donor ortho-directing groups, ranging from simple alkoxy groups (see above) and frequently used sulfoxides and acetals to P-chiral phosphine oxides62 and [1,3,2]oxazaphospholidine-2-oxides.63 The use of the sulfoxide moiety as an efﬁcient ortho-directing group in ferrocene chemistry has been coined by Kagan and coworkers.64 Chiral-at-sulfur sulfoxides FcS(O)R (R = alkyl or aryl; most often t-Bu) are available by enantioselective oxidation of the corresponding sulﬁdes (FcSR) with organic peroxides (e.g. PhMe2 COOH) in the presence of titanium reagent prepared from Ti(O-i-Pr)4 , (+)-diethyl tartarate (DET) and water (e.g. (SS )-21 in Scheme 7.10). Another route is represented by the reaction of LiFc with chiral sulﬁnates. The preparation of p-tolyl sulfoxide (RS )22 from (1R)-menthyl (SS )-4-toluenesulﬁnate depicted in Scheme 7.10 serves as a representative example.65

254

FcLi

Ferrocenes: Ligands, Materials and Biomolecules

R2S2

FcSR

Ti(O-i-Pr)4/DET/H2O

1. LiBu

Fe

2. E+

(SS)-21

O S

Fe

cumyl hydroperoxide

O S R

E O S R

23

O

Fe

(RS)-22

S

p-Tol O

1. LiN(i-Pr)2 2. E+

E

Fe

S

p-Tol O

24

Scheme 7.10 Preparation and lithiation of chiral ferrocene sulfoxides

Lithiation of sulfoxides with an appropriate base and under optimised conditions occurs with a high stereoselectivity in the position adjacent to the sulfoxide moiety, thus providing a route to highly optically pure, 2-functionalised derivatives (23 and 24 in Scheme 7.10).65c–d, 66 However, apart from being an ortho-directing group, the sulfoxide moiety represents a reactive site suitable for subsequent synthetic modiﬁcations. The sulfoxide group in the functionalised ferrocene sulfoxides can be: converted to the corresponding sulfone or sulﬁde; exchanged for lithium (preferably with Li(t-Bu)); or efﬁciently removed. The synthetic potential of planar chiral ferrocene sulfoxides can be demonstrated by their use in the synthesis of planar-only chiral ferrocene diphosphines 25,65c, 67 enantiopure 2-aryl/heteroaryl-1-phosphinoferrocenes and their corresponding phosphites 26 and 27,68 chiral aminopyridine 28,69 Taniaphos ligands (see Chapter 6),70 C2 -symmetric 1,1 -diphosphino-2,2 -bis(sulfenyl) or bis(sulfonyl)ferrocenes 29 (see also below),71 phosphinosulfanes 30 and 31,72 1-sulﬁnyl-2-aminoferrocenes (32),73 a planar-chiral N-heterocyclic carbene 33,74 biferrocene diphosphines 3475 and (Rp ,Rp )-bis[2-(diphenylphosphino)ferrocenyl]methane76 (Scheme 7.11). Most of these donors have been tested as chiral ligands for various enantioselective reactions.64 The use of sulfoxide directing groups in the preparation of enantiopure, asymmetric 1,3-disubstituted ferrocenes is discussed below. Another approach developed by Kagan, Riant et al. makes use of a chiral dioxolane moiety. Chiral acetal 35 as the key compound is conveniently prepared by transacetalisation of intermediate ferrocenecarboxaldehyde dimethylacetal (36) with (S)-1,2,4-butanetriol and subsequent standard methylation at the terminal hydroxyl group (Scheme 7.12). Lithiation of acetal 35 followed by reaction with electrophiles (→ 38) and hydrolytic removal of the protective group yield synthetically valuable, 2-substituted ferrocenecarboxaldehydes 39 in optically pure form.77 Starting with (S)1,2,4-butanetriol,78 this method produces only (Sp )-aldehydes (provided the E-group is senior to CHO according to Cahn–Ingold–Prelog rules). However, it is very attractive due to a wide choice of applicable electrophiles (see examples in Scheme 7.12) as

Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors

255

Y

Fe

PR2

Ar

Ar

PPh2 Fe

PY2

Fe

NMe2

PPh2

Fe N

25 (R = Me, Cy)

26 (Ar = aryl or heteroaryl; Y = R or OR)

SR′

SE2R′ Fe

PR2 PR2

27 (Y = OMe, PPh2, NHR)

Fe

PR2

28 (Ar = aryl)

S(O)R′

SR′ Fe

Y

PR2

Fe

NR2

SE2R′ 29 (E = void or O)

30

31 (Y = CH2, NH)

Fe

SiMe3 Fe N

32

Fe

PR2 PR′2

N 33

34

Scheme 7.11

well as the possibility of isolating the functionalised acetals (i.e. protected aldehydes) and converting the lithiated intermediate into other reactive compounds. For instance, the acetals bearing a halide or boronic moiety (38, where E = halide or B(OH)2 ) were used in cross-coupling reactions while the zincated acetal (38, E = ZnCl; obtained from 38, E = lithium and ZnCl2 ) or the corresponding cuprate proved to be suitable substrates for C−C bond forming reactions and for amination reactions (→ 39 with E = NH2 and NHAc).77b Manoury et al. extended this approach to fc(CHO)2 79 and have shown that lithiation of the SiMe3 -substituted acetal 38 occurs in position 5, giving rise, after quenching with an electrophile, to 2,5-disubstituted acetal.80 Both 2-mono and 2,5-difunctionalised acetals are selectively lithiated with lithium N -methylpiperazide to give 1 -lithioderivatives (cf. Chapter 5).80, 81 Chiral ferrocenecarboxaldehydes 39 have been typically used in the preparation of material precursors82 whilst the aldehyde bearing the diphenylphosphino group (39, E = PPh2 ) has been used in the preparation of numerous planar-only chiral

256

Ferrocenes: Ligands, Materials and Biomolecules OH FcCHO

CH(OMe)3 cat.

H+

OH

FcCH(OMe)2

O

OH

cat. H+ (removal of H2O)

36

Fc O 37

OH

1. NaH 2. MeI

O CHO Fe

E

H+

Fe

E

O

1. t-BuLi OMe

2. EX

O Fc O OMe

39

38

35

Scheme 7.12 Preparation of 2-substituted ferrocenecarboxaldehydes via chiral acetal 35 [E (E-X) = SiMe3 (ClSiMe3 ), SiMe2 (t -Bu) (Cl-SiMe2 (t -Bu)), SnBu3 (Cl-SnBu3 ), PPh2 (ClPPh2 ), CO2 Me (Cl-CO2 Me), I (1,2-C2 H4 I2 ), Br (p-xylylene dibromide), OH (Me3 SiOOMe3 ), SC6 H4 Me-4 [(4-MeC6 H4 )2 S2 ], B(OH)2 (B(OR)3 , then hydrolysis), CHO (HCONMe2 , then hydrolysis), Ph (the lithio intermediate was ﬁrst zincated wit ZnCl2 (i.e., E = ZnCl) and then cross-coupled with PhI under Pd-catalysis)]

ligands. Typical examples (Scheme 7.13) include phosphino-thioethers 40,83 phosphinoalkenes 41,84 phosphino-imines 42,85 diphosphines 43,86 phosphino-imidazolidines 44,87 Taniaphos-type ligands (see Chapter 6)88 and bisferrocene donors such as 45 (via pinacol coupling),77b, 89 46 (via McMurry reaction) and 47 (via condensation with chiral 1,2-diaminocyclohexanes).90, 91 Most of these compounds have been tested in enantioselective allylic alkylation (40–44 and 47) and in rhodium-catalysed hydrogenation of dehydroamino acids (43). Both isomers of the phosphinoaldehyde were recently obtained by standard resolution via diastereoisomeric derivatives and converted to a series of acetals with combined planar and central chirality, that were subsequently used as ligands in palladium-catalysed allylic alkylation.92 7.2.4

The Use of Chiral Bases

In contrast to the frequent use of chiral ortho-directing groups, the complementary approach based on chiral bases is still less common. Chiral amides or adducts formed from alkyl lithiums and chiral N-donor lithiate ferrocenes that are substituted with polar functional groups. The substituent not only directs (coordinates) the reagent but also simultaneously activates the ferrocene unit towards nucleophilic attack. For instance, FcP(O)Ph2 was deprotonated with (R,R)-[PhCH(Me)]2 NLi and then treated with Me3 SiCl to give 2-silylated product in 95 % yield and 54 % ee. Unfortunately, other substrates tested were either not appreciably deprotonated (FcCH2 OH, FcCH2 OMe, FcPPh2 and FcSPh) or gave racemic products (FcSO2 Ph and FcCON(iPr)2 ).93 Another amide prepared from LiBu and (R,R)-N ,N ,N ,N -tetramethyl-1,2diaminocyclohexane was shown to ortho-lithiate amines FcCH2 NR2 (NR2 = NMe2 ,

Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors

PPh2

Fe

Fe

40

HO

PPh2

Fe

41

PPh2

OR

PR2

CHO

PPh2 Fe 45

"materials"

42

OH

Ph2P

CpFe

N R

R

SR

PPh2

Fe

PPh2

Fe

PPh2

FeCp (R = H, Ac)

43 (R = Ph, Cy, t-Bu etc)

Ph2P

Me N

FeCp N

CpFe + (Z)-isomer

N

N Fe PPh2 Me

PPh2 PPh2

PPh2 46

257

CpFe

FeCp 47

44

Scheme 7.13

piperidinyl and morpholinyl) stereosectively with ee’s in the range 67–80 % whereas sulfones FcSO2 R (R = 4-tolyl or t-Bu) gave only disappointing results.94 Adduct of LiBu with (−)-sparteine has been used ﬁrstly with isopropylferrocene, leading predominantly to optically enriched 3,1 -dilithiated product.9 Tertiary ferrocenecarboxamides proved more successful, affording the ortho-substituted products. Thus, metalation of amide 48 followed by treatment with electrophiles leads to functionalised amides 49 in good yields and ee’s up to 99 % while diamide 50 gives rise to mono- (51) and disubstituted products (52), depending on the reaction conditions (Scheme 7.14). Selected functionalised amides have been tested as ligands for palladium-catalysed allylic alkylation and in enantioselective addition of diethylzinc to benzaldehyde.95 However, subsequent synthetic use of such amides is limited by their high hydrolytic stability.96 This can be partly circumvented by changing the N-substituents.97 Metalation of C-chiral methyl ferrocenesulfonates with the LiBu-(−)sparteine adduct has also been studied.98

7.3

Synthesis of 1,3-Disubstituted Ferrocene Donors

The number of 1,3-disubstituted (or 1,1 ,3-trisubstituted) ferrocenes still remains low compared to their 1,2 (or 1,1 ,2) counterparts. This is largely due to the lack of efﬁcient and generally applicable synthetic methods available for their preparation. Formerly,

258

Ferrocenes: Ligands, Materials and Biomolecules E FcCON(i-Pr)2 48

1. LiR/(–)-sparteine Fe

2. electrophile (E+)

CON(i-Pr)2

49 E fc{CON(i-Pr)2}2 50

1. LiR/(–)-sparteine 2. electrophile (E+)

Fe

CON(i-Pr)2 CON(i-Pr)2

51 OR E Fe

CON(i-Pr)2 E

E +

Fe

E

CON(i-Pr)2 meso-52

CON(i-Pr)2 CON(i-Pr)2

dl-53

Scheme 7.14 Directed lithiation of tertiary ferrocene carboxamides

the access to 1,3-disubstituted ferrocenes relied predominantly on the ability of alkyl substituents in monoalkylferrocenes (FcR) to direct electrophiles in electrophilic substitutions, typically acetylations, to all the possible positions (i.e. 2-, 3-, and 1 -) – but with a slightly different efﬁciency. The desired 1-acetyl-3-R derivatives were usually isolated from the isomer mixtures by tedious chromatography and converted to other, more reactive derivatives.99, 100 When appropriate, resolution of chiral 1,3disubstituted ferrocenes was achieved via separation of diastereoisomeric derivatives101 or by enzymes.102 A similar directing effect has been observed also in lithiation,9, 99f, 103 formylation99f, 104 and borylation105, 106 reactions involving alkyl- and 1,1 -dialkylferrocenes, and in lithiation of FcPPh2 .107 A better yielding and more versatile approach towards 1,3-disubstituted ferrocenes is based on lithiation of ferrocenyl sulﬁdes (54, R = 4MeC6 H4 , Ph, t-Bu) that, with a proper base, affords predominantly 3-lithiated products. Trapping the lithio intermediates with electrophiles together with a replacement of the sulﬁde group (e.g. via oxidation to sulfoxide and reaction with t-BuLi to give a lithio derivative; see above) and transformations of the introduced substituents lead to 1,3-functionalised ferrocenes 55–57 (Scheme 7.15).108 Alternative routes to 1,3-disubstituted ferrocenes require the ferrocene framework to be constructed. For instance, metathesis between 1,3-disubstituted cyclopentadienide salts with iron(II) chloride affords 1,1 ,3,3 -tetrasubstituted ferrocenes.109 Alternatively, the 1,3-disubstituted ferrocenes are accessible via photochemically-assisted replacement of arene ligand in [Fe(η5 -C5 H5 )(η6 -p-C6 H4 Me2 )][PF6 ] with 6-(dimethylamino)fulvenes. Hydrolytic work-up of the intermediate iminium salts gives rise to 3-substituted acylferrocenes 58 (Scheme 7.16).110

Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors

259

FcH SR

i. ii.

FcLi

FcSR

v.

E

S(O)R vi.

Fe

E

Fe

54 iii.

iv.

55

56

FcS(O)R vii.

E' E

Fe

57

Scheme 7.15 Preparation of 1,3-disubstituted ferrocenes by lithiation of ferrocene sulﬁdes [i . s-BuLi in THF (0 ◦ C/5 h); ii. RSSR (R = t -Bu, Ph, and 4-tolyl); iii. (4-MeC6 H4 )SO2 (i -Pr) or t -BuS(O)S(t -Bu); iv. Me2 SiCl2 /Zn in acetone (0 ◦ C/10 min); v . s-BuLi in THF (0 ◦ C/7 h), then electrophile (E+ ); vi. oxidation; vii. excess t -BuLi in THF at −78 ◦ C, then electrophile]

NMe2

NMe2 PF6 Fe

R2

R1

R2

PF6

R1 Fe

OH−/H2O

O

R2

R1 Fe

58

Scheme 7.16 Preparation of 1-R-3-acylferrocenes from 6-(dimethylamino)fulvenes. R1 /R2 = CO2 Me/t -Bu (a), H/CO2 Et (b), NMe2 /t -Bu (c), NMe2 /C(O)NMe2 (d)

260

Ferrocenes: Ligands, Materials and Biomolecules

This ‘fulvene route’ has been used in the preparation of pincer-type diphosphine ligands that, in analogy to their benzene counterparts,111 undergo C−H activation when reacted with rhodium(I) and palladium(II) precursors to afford complexes with triplyligating (PCP) 1,3-bis{(dialkylphosphino)methyl}ferrocen-2-yl donors.112 A comparison of the routes available for the preparation of 1,3-disubstituted ferrocenes has been offered during the preparation of 1,3-bis[(S)-4-(i-propyl)-4,5-dihydrooxazol-2yl]ferrocene.113 Only recently, an efﬁcient methodology has been developed that allows the preparation of enantiopure 1,3-disubstituted derivatives by means of a temporary chiral ortho-directing group, which is removed after functionalisation at both adjacent positions. This approach has so far been used with the well-established C-chiral Ugi’s amine (R)-1 (Scheme 7.17), chiral ferrocene sulfoxides114 and also with ferrocenes combining two chiral directing groups.115, 116

NMe2 Fe

i.

(R)-1

Br

Fe

NMe2

ii.

OHC

NMe2 Fe

(R,Sp)-63

Br

NMe2

(R,Rp)-60

(R,Sp)-59

57 % overall

Ph2P

Fe

iii.

v.

NMe2

HO Fe

(R,Sp)-62

iv.

HO Fe

Br

NMe2

(R,Rp)-61

Scheme 7.17 Preparation of 1,3-disubstituted ferrocenes by using a temporary orthodirecting group [i . s-BuLi (THF, −78 ◦ C/4 h), then CF2 BrCF2 Br (THF, room temperature/17 h); ii. LiTMP (HTMP = 2,2,6,6-tetramethylpiperidine; THF, −78 ◦ C/4 h), then Me2 NCHO (0 ◦ C/16 h); iii. LiAlH4 (room temperature/16 h); iv. LiBu (THF, −78 ◦ C/30 min), then water; v . HPPh2 /HBF4 (CH2 Cl2 , room temperature/16 h)]

Despite the intrinsic synthetic difﬁculties, 1,3-disubstituted ferrocenes have been used in the preparation of electrochemically sensing molecules,117 ferrocene-bridged bis(crown ethers),118 liquid–crystalline materials119 as well as precursors to organometallic polymers103b and photoresponsive molecules that change conformation as the consequence of conﬁgurational changes of their photosensitive parts or non-covalently bonded hosts.120

Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors

7.4

261

Chiral Donors with Bridged Cyclopentadienyl Rings

Ferrocenes with intramolecularly bridged cyclopentadienyl rings, the so-called ferrocenophanes, were formerly looked upon mainly as chemical (structural) curiosities. Later, however, interest in bridged ferrocenes was renewed, being stimulated by their applications as electrochemical sensors (see Chapter 8) as well as precursors to ferrocene ligands and organometallic polymers (see Chapters 5 and 10). Because the number of ring-linking groups (1–5) and their nature can be varied rather independently, ferrocenophanes constitute a highly variable class of ferrocene derivatives. However, for practical reasons,121 singly-bridged compounds have been studied most often. The nature of the bridging group(s), particularly the length and the atom sequence, are decisive for the ferrocenophane properties: Whereas short bridges typically induce steric strain to the ferrocene unit, resulting in mutual tilting of the cyclopentadienyl planes and in enhanced reactivity, the longer ones just limit mobility of the ferrocene cyclopentadienyls while leaving the geometry of the ferrocene core virtually intact. The ﬁrst ferrocenophanes prepared were those featuring hydrocarbyl bridges. To date, there have been reported many examples of such compounds differing by the number of the linking groups and carbon atoms therein. They are accessible directly by reacting iron(II) chloride with the appropriate bis(cyclopentadienide) salts or from suitably modiﬁed ferrocene derivatives via intramolecular cyclisation reactions (e.g. acylation, Dieckman, acyloin and aldol-type reactions), by alkene metathesis as well as from reactions of some α,ω-dihalides with fcLi2 .122 Indeed, there has been reported a number of compounds containing potentially donating heteroatoms within the bridge.123 Only a few representatives were, however, studied as ligands. Among the derivatives with a single heteroatom bridge (fcE),124 the ferrocene-1,1-diyl phosphines fcPR have been studied most thoroughly. These reactive compounds, which are accessible by metathesis of the fcLi2 –TMEDA adduct with the appropriate RPCl2 ,125, 126 have been studied as ligands and polymer precursors27d, 127 and were also used in the preparation of other phosphinoferrocene ligands (see Chapter 5). Yet another example of the ferrocenophane ligand with donor atoms located in the bridge is represented by phosphine 64 whose preparation is depicted in Scheme 7.18. Compound 64 was used as a P-monodentate donor in palladium(II), platinum(II) and chromium and iron carbonyl complexes of which several representatives were structurally and/or electrochemically characterised.128

PhPH2 +

1. LiBu 2. H2O

Ph

P

1. LiBu 2. FeCl2

Fe

64

Scheme 7.18

P

Ph

67

Fe

NR*

H2NR*

NaBH4

rac-67

66

68

Fe

Fc

Fe

[H]

H2/PtO2

Fe

O

(CF3CO)2O

CO2H

OH

NHR*

HCHO NaBH4

then base rac-65

Fe

69

Fe

FcCHO

AcOH Et2NH

NMeR*

resoln. via tartrate salts

NMe2

C5H5N/C5H10NH (cat.)

CH2(CO2H)2

Me2NH/AlCl3

CO2H

Me2NH

70

Fe

(R)-65

(S)-65

OAc

Me2NH

MeI

71

Fe

NMe2R*

I

Scheme 7.19 Summary of synthetic routes to enantiopure amines 65; (R∗ = (S )-1-(1-naphthyl)ethyl (a), and (S )-1-phenylethyl (b))

Fc

262 Ferrocenes: Ligands, Materials and Biomolecules

Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors

263

On the other hand, ferrocenophanes bearing a single hydrocarbyl bridge are well established as scaffolds for the preparation of chiral ferrocene donors. The most systematically studied are undoubtedly compounds with a single three-carbon tether derived from C-chiral amine 65. In its synthesis, 3-ferrocenylpropanoic acid is ﬁrst cyclised to give ketone 66,129 which is then reduced to alcohol 67129 and, ﬁnally, converted to racemic 65.130 Resolution of racemic 65 can be achieved by fractional crystallisation of tartrate salts (Scheme 7.19).131 Another route to optically pure 65 relies on chiral auxiliary approach, starting again with ketone 66. The ketone is ﬁrst condensed with (S)-1-(1-naphthyl)ethylamine (H2 NR∗ ) to give imide syn-67 as a thermodynamic product, whose subsequent reduction gives diastereomerically pure amine (R,S)-68 in 95 % yield. Amine (R)-65 is then obtained in three steps via reductive methylation, acetylation and nucleophilic replacement of the acetate group (i.e. via 69 and 70). The use of the less bulky amine PhCH(Me)NH2 results in the formation of a mixture of isomeric imines synand anti -67a which, upon reduction, gives chromatographically separable mixture of diastereoisomeric amines (R,S)- and (S,S)-68a (75 % and 16 % isolated yields, respectively). The major isomer can be converted to amine (R)-65 either by the reaction sequence mentioned above or via its ammonium salt (R,S)-71 (Scheme 7.19).132 The reactivity of amine 65 clearly parallels that of Ugi’s amine (1). For instance, monolithiation of rac-65 takes place exclusively at the cyclopentadienyl ring in position adjacent to the nitrogen atom and, in the case of enantiomerically pure 65, with a high diastereoselectivity. This was demonstrated by a series of lithiation/phosphinylation reactions of (S)-65 giving phosphinoamines (S,Sp )-72 (Scheme 7.20; N.B.: Analogous products resulting from (S)-1 and (S)-65 have the opposite conﬁgurations at the chirality plane). The following lithiation step is less selective. It was also shown that the phosphinoamines can be efﬁciently converted to their respective acetates, alcohols and, ﬁnally, to hydrocarbons133 – again in a manner similar to their 1-based counterparts. Likewise, the standard phosphinylation of 72 (Scheme 7.20) affords diphosphines 73 with retention of conﬁguration at both chirality elements. NMe2

Fe

(S)-65

1. LiBu 2. ClPR2

R2P

NMe2 Fe

(S,Sp)-72(R)

R'2PH /AcOH

R2P

PR'2 Fe

(S,Sp)-73(R/R')

Scheme 7.20

The coordination behaviour of 72 and 73 has been probed in a series of palladium134 and ruthenium complexes.135 Enantiopure diphosphines (R,Rp )-73(Ph,Ph) and (R,Rp )73(Ph,Cy) were tested in rhodium-catalysed hydrogenations of prochiral alkenes and ketones, in iridum-catalysed reduction of 2-ethyl-N -(2-methoxy-1-methylethylidene)6-methylaniline,136 and in palladium-mediated allylic alkylation and amination reactions.137 Additionally, platinum complexes of the type [PtCl2 (L)] and [PtCl(SnCl3 )(L)],

264

Ferrocenes: Ligands, Materials and Biomolecules

where L = (R,Rp )-73(Ph/Cy), (R,Rp )-73(Ph/Cy) and (R,Rp )-73(C6 H4 F-4/Ph), and some of their palladium(II) analogues were studied as catalysts in hydroformylation and methoxycarbonylation of styrene.138 In 1999, Erker and coworkers reported amines 75 possessing a further chirality element. They are obtained by hydrogenation of their bridge-unsaturated precursors 74 that are in turn accessible from Mannich-type reaction of 1,1 -diacetylferrocene, the appropriate amine (in excess) and one-molar equivalent of titanium(IV) chloride as a Lewis acid promoter (Scheme 7.21). Amines 75 with achiral N-substituents result as mixtures of racemic cis and trans isomers, the latter prevailing.139, 140 The reaction mechanism has been corroborated by in situ generation of the anticipated intermediate, 1,1 -bis{1-(dimethylamino)vinyl}ferrocene, from iron(II) chloride and the corresponding cyclopentadienide.140 Me

Me O Fe Me

R2NH TiCl4

H2

Fe

Pd/C NR2

O

74a–g

Fe NR2 cis/trans-75a,b,e,g

Scheme 7.21 The synthesis of amines 74 and 75 (NR2 = NMe2 (a), NEt2 (b), NMe(i -Pr) (c), pyrrolidin-1-yl (d), piperidin-1-yl (e), morpholin-4-yl (f), and (R)- or (S )-NMe{CH(Ph)Me} (g))

The preparation of enantiopure trans-75a was ﬁrst attempted via reduction of Cchiral enamine 74g followed by a replacement of the amine function and, alternatively, via fractional crystallisation of diastereoisomeric tartrate salts. Finally, a satisfactory resolution was achieved by using a covalently bound chiral auxiliary. This method outlined in Scheme 7.22 works well with both (S)- and (R)-PhCH(Me)NH2 as the source of the additional chirality, providing access to both forms of trans-75g [i.e., (R,R)- and (S,S)-75g]; amines resulting from the minor cis-form are removed during the chromatography. The NMe2 group is then restored by a sequence of reductive debenzylation and reductive amination.141 A similar approach has been applied to the synthesis of analogous primary amines.142 Similarly to 65, the amine trans-75a undergoes ortho-lithiation and stereoconservative replacement of the amine function. The course of the amine substitution reaction has been rationalised in terms of the stabilisation of the intermediate carbocation by the electron rich iron atom. This explanation was supported by spectral data for 79, obtained from racemic trans-75a via hydride abstraction with B(C6 F5 )3 , and by structural characterisation of a product of its partial hydrolysis and, ﬁnally, by DFT calculations.143 The cation in 79 undergoes clean hydrolysis to give ketone 80 (Scheme 7.23). Both rac- and (R)-80 were shown to be accessible by several routes via various imine intermediates from rac- and (R)-75a.144, 145 The selective metalation and substitution reactions of rac- and (R,R)-trans-75a enabled the preparation of a series of mono- and diphosphines (Scheme 7.24).146 Thus,

Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors Me

265

I i.

Fe

cis/trans-75a

NMe3

Me

Me

Me

76 ii.

iv.

Fe

Ph

Me

75g

N

Ph N Me

NH

Me

*

Fe NMe2

Me (R,R)-78

Me (R,R)-77g

iii.

Fe

v.

Fe

Me iv.

Fe

(R,R)-75a

N

v.

Fe

Ph

Me

Fe

NH Me (S,S)-78

Me (S,S)-77g

NMe2 (S,S)-75a

Scheme 7.22 A method devised for the preparation of enantiopure (R,R)- and (S ,S )-75a. (Only the use of (R)-PhCH(Me)NHMe as chiral auxiliary is shown for clarity.) i . MeI; ii. PhCH(Me)NHMe/K2 CO3 ; iii. chromatographic separation; iv. H2 /Pd-C; v . CH2 O/NaBH4 ] Me

Me

rac-75a

B(C6F5)3

[HB(C6F5)3]

Fe

NaOH − Me2NH

Fe

NMe2

O 80

79

Scheme 7.23 Me

trans-75a

1. LiBu 2. R2PCl

Me

R'2PH

Fe

AcOH

Fe

NMe2 PR2 81(R)

Scheme 7.24

PR'2 PR2 82(R/R')

266

Ferrocenes: Ligands, Materials and Biomolecules

ligands 81(Ph) and 82(Ph/Ph) synthesised from rac-trans-75a were shown to act as cis-chelating donors and their palladium(II) complexes served as precursors to efﬁcient catalytic systems for alternating CO/ethene copolymerisation.146a The enantiopure counterparts (R,R,Rp )-82(Ph/Ph) and (R,R,Rp )-82(Ph/Cy) were used in rhodiumcatalysed hydrogenations of dimethyl itaconate and methyl (Z)-acetamidocinnamate, and in palladium-catalysed alternating copolymerisation of propene with CO.146b Attempts to prepare a ruthenium(II) P ,N -chelate complex by the reaction of aminophosphine 81(Ph) and [{RuCl2 (η6 -p-cymene)}2 ] in the presence of K[PF6 ] led to a mixture of cationic complexes featuring: the ligand converted to an anionic P ,N ,Cligand due to C–H activation of one methyl group; and the N-protonated form of 81(Ph) as a P -monodentate donor.147 Synthetic versatility of ferrocenophane trans-75a has been further demonstrated by its conversion to chiral iodoamines, (iodo)acetates, chlorophosphines, and by preparation of phosphinoamines that relate to 81 but possess primary amino groups. The latter compounds were tested as ligands for hydrogenations of prochiral ketones.148

7.5

Polydentate Ferrocene Donors

Ferrocene derivatives bearing several donor groups constitute a diverse group of compounds, some of which have been described elsewhere in this book. This particular section restricts itself to the cases where at least two similar or identical donor groups are bound directly to the ferrocene moiety. (Several such polyfunctional ferrocenes have been already mentioned in Section 7.1.2 dedicated to oxazoline compounds.) Undoubtedly, the most abundant family of such donors are ferrocene polyphosphines. A possible route to these compounds via lithiation of other phosphinoferrocenes (namely dppf and FcPPh2 ) and a subsequent reaction with a chlorophosphine has been explored by Butler et al.107a This approach unfortunately proved to be rather impractical due to low regioselectivity and difﬁculties with separation of the products. For FcPPh2 , the procedure has been slightly modiﬁed which allowed, after tedious chromatographic puriﬁcation, the isolation of 1,2,1 - and 1,3,1 -tris(diphenylphosphino) ferrocenes.107b In 1999, a new impulse came to this area when Butler reported ortho-lithiation of 1,1 -dibromoferrocene (fcBr2 , 83) with lithium diisopropylamide (LDA).149 A number of derivatives were then prepared by quenching the lithiated species with various electrophiles. The primary products can be regarded as valuable organometallic synthons. Their subsequent transformations typically involve lithiation with n-BuLi and reactions with electrophilic reagents including chlorophosphines (Scheme 7.25). In addition, it was established that an excess of LDA in the metalation of 83 leads to 2,5- and 2,2 dilithiated species, thus opening a route to 1,3-bis(diphenylphosphino)ferrocene after removing the bromine atoms, as well as to 1,2,3,1 - and 1,1 ,2,2 -tetrakis(diphenylphosphino)ferrocene (Scheme 7.25).116 Meunier, Broussier, Hierso et al. coined a different approach towards ferrocene di-, tri- and tetraphosphines with or without additional alkyl substituents at the cyclopentadienyl rings, consisting of metathesis between appropriately substituted cyclopentadienides150 and iron(II) chloride. This methodology proved to be very attractive because

Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors

267

Y = PPh2

Br 1. excess LDA

Fe

2. ClPPh2 Br Y

83

Br

1. LDA (1 equiv.) 2. electrophile

Fe

Y

Br +

Fe

Y

Br Br Fe

Br +

Fe

Br

1. LiBu 2. ClPPh2

1. LiBu 2. H2O

Y Y Br

1. LiBu 2. ClPPh2

E Y Br

Y Fe

Y

Y Fe

Y

Y

Fe

Y Y Y

Scheme 7.25

of good yields and synthetic versatility (e.g. ferrocenes featuring differently substituted cyclopentadienyl rings result from sequential addition of two cyclopentadienide salts). Several phosphines thus resulting have been structurally characterised, studied as ligands in various complexes, and also converted to phosphine oxides and sulﬁdes. Analogous polythioethers were synthesised in a similar manner.151 One of the most thoroughly examined in the series, 1,1 ,2,2 -tetrakis(diphenylphosphino)-4,4 -di-t-butylferrocene (84 in Scheme 7.26) exerts interesting coordination properties152 and catalytic chemistry153 due to its ﬁxed molecular conformation. Testing of 84 in palladium-catalysed Suzuki, Heck153a and allylic amination153b reactions showed high activity of the formed catalytic systems even at very low catalyst/substrate ratios. This was attributed to a stabilisation of the active species resulting from the blocked conformation of the ferrocene cyclopentadienyls and to the presence of multiple metal-stabilising donor groups. On the other hand, a triphosphine combining two types of phosphorus groups, 1,2-bis(diphenylphosphino)-1 -(di-i-propylphosphino)-4(t-butyl)ferrocene (85 in Scheme 7.26) showed promising results in Sonogashira crosscoupling reaction, coupling terminal alkynes with various aryl halides even at catalyst loadings as low as 10−4 mol%.154 There is only a limited number of other known multidentate ferrocene donors besides the phosphines. In addition to alkylthioethers prepared by Broussier et al. (see above), Long and coworkers reported the synthesis of 1,1 ,2-tris and 1,1 ,2,2 tetrakis(methylsulfanyl)ferocene155 by lithiation of fc(SMe)2 156 and subsequent reaction with dimethyl disulﬁde. Complexes of these ligands with transition metals of Groups 6, 7 and 10 have been prepared and structurally characterised.157

268

Ferrocenes: Ligands, Materials and Biomolecules Me3C Me3C

PPh2 Fe

Me3C

PPh2

PPh2 Fe

PPh2

PPh2 P(i-Pr)2

PPh2 84

85

Scheme 7.26

Only recently, optically pure chiral ferrocene 1,1 -bis(phosphines) with sulfurcontaining groups (sulfanyl, sulﬁnyl and sulfonyl; 29) in positions 2 and 2 have been reported by Zhang et al.71 Not surprisingly, their preparation is based on orthometalation/functionalisation of chiral sulfoxide fc{S(O)(t-Bu)}2 followed by redox transformations of the sulﬁnyl group. The ligands were successfully tested in enantioselective hydrogenation over rhodium catalysts and palladium-catalysed allylic alkylation reactions.

References 1. 2. 3. 4. 5.

6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

K. Tappe, P. Knochel, Tetrahedron: Asymmetry 2004, 15, 91–102. Amine 1 is available also commercially. D. Marquarding, H. Klusacek, G. Gokel et al. J. Am. Chem. Soc. 1970, 921, 5389–5393. G.W. Gokel, D. Marquarding, I.K. Ugi, J. Org. Chem. 1972, 37, 3052–3058. (a) Ferrocenes: Homogeneous Catalysis, Organic Synthesis, Materials Science, A. Togni, T. Hayashi (Eds), VCH, Weinheim, Germany, 1995; (b) A. Togni, New Chiral Ferrocenyl Ligands for Asymmetric Catalysis in Metallocenes–Synthesis, Reactivity, Applications, A. Togni, R.L. Halterman (Eds), vol. 2, pp. 685–721, Wiley-VCH Verlag GmbH, Weinheim, Germany, 1998; (c) C.J. Richards, A.J. Locke, Tetrahedron: Asymmetry 1998, 9, 2377–2407; (d) T.J. Colacot, Chem. Rev. 2003, 103, 3101–3118; (e) R.C.J. Atkinson, V.C. Gibson, N.J. Long, Chem. Soc. Rev. 2004, 33, 313–328; (f) R.G. Array´as, J. Adrio, J.C. Carretero, Angew. Chem. Int. Ed. 2006, 45, 7674–7715. See also: T. Hayashi, T. Mise, M. Fukushima et al. Bull. Chem. Soc. Jpn. 1980, 53, 1138–1151. (a) R = Me: P.B. Valkovich, G.W. Gokel, I.K. Ugi, Tetrahedron Lett. 1973, 14, 2947–2950; (b) R = Ph: M. Lotz, T. Ireland, K. Tapper, P. Knochel, Chirality 2000, 12, 389–395. L. Xiao, K. Mereiter, W. Weissensteiner, W. Widhalm, Synthesis 1999, 1354–1362. (a) T. Aratani, T. Gonda, H. Nozaki, Tetrahedron Lett. 1969, 2265–2268; (b) T. Aratani, T. Gonda, H. Nozaki, Tetrahedron 1970, 26, 5453–5464. A. Farrell, R. Goddard, P.J. Guiry, J. Org. Chem. 2002, 67, 4209–4217. C. Ganter, T. Wagner, Chem. Ber. 1995, 128, 1157–1161. T. Ahern, H. M¨uller-Bunz, P.J. Guiry, J. Org. Chem. 2006, 71, 7596–7602. M. Widhalm, K. Mereiter, M. Bourghida, Tetrahedron: Asymmetry 1998, 9, 2983–2986. (a) R. Kitzler, L. Xiao, W. Weissensteiner, Tetrahedron: Asymmetry 2000, 11, 3459–3462; (b) L. Xiao, R. Kitzler, W. Weissensteiner, J. Org. Chem. 2001, 66, 8912–8919. (a) D. Enders, R. Peters, R. Lochtman, J. Runsink, Eur. J. Org. Chem. 2000, 2839–2850; (b) D. Enders, R. Peters, R. Lochtman, J. Runsink, Synlett 1997, 1462–1464.

Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors

269

16. FcCHO-SAMP hydrazone adds LiR across the N=N double bond to give the corresponding hydrazines, which after reductive cleavage and acetylation afford (R)-FcCH(R)NHAc (ee 91–98 %); Dialdehyde fc(CHO)2 reacts similarly: D. Enders, R. Lochtman, Eur. J. Org. Chem. 1998, 689–696. 17. (a) D. Enders, R. Peters, R. Lochtman, G. Raabe, Angew. Chem. Int. Ed. 1999, 38, 2421–2423; (b) D. Enders, R. Peters, R. Lochtman et al. Eur. J. Org. Chem. 2000, 3399–3426. 18. D. Enders, R. Peters, J. Runsink, J.W. Bats, Org. Lett. 1999, 1, 1863–1866. 19. B. Ferber, S. Top, P. Herson, G. Jaouen, Organometallics 2007, 26, 1686–1691. 20. A comprehensive review covering the chemistry and applications of chiral ferrocene oxazolines since their ﬁrst appearance in 1995 up to 2002 has been published: O.B. Sutcliffe, M.R. Bryce, Tetrahedron: Asymmetry 2003, 14, 2297–2325. An update of important catalytic applications throughout years 2003–2006 has been provided in ref. 5f. 21. G. Schmitt, P. Klein, W.J. Ebertz, J. Organomet. Chem. 1982, 234, 63–72. 22. (a) C.J. Richards, T. Damalidis, D.E. Hibbs, M.B. Hursthouse, Synlett 1995, 74–76; (b) C.J. Richards, A.W. Mulvaney, Tetrahedron: Asymmetry 1996, 7, 1419–1430. 23. (a) T. Sammakia, H.A. Latham, D.R. Schaad, J. Org. Chem. 1995, 60, 10–11; (b) T. Sammakia, H.A. Latham, J. Org. Chem. 1995, 60, 6002–6003; (c) T. Sammakia, H.A. Latham, J. Org. Chem. 1996, 61, 1629–1635. 24. (a) A. Chesney, M.R. Bryce, R.W.J. Chubb et al. Synthesis 1998, 413–416; (b) W. Zhang, Y. Adachi, T. Hirao, I. Ikeda, Tetrahedron: Asymmetry 1996, 7, 451–460. 25. K.H. Ahn, C-W. Cho, H-H. Baek et al. J. Org. Chem. 1996, 61, 4937–4943. 26. (a) Y. Nishibayashi, S. Uemura, Synlett 1995, 79–81; (b) Y. Nishibayashi, K. Segawa, K. Ohe, S. Uemura, Organometallics 1995, 14, 5486–5487; (c) Y. Nishibayashi, K. Segawa, Y. Arikawa et al. J. Organomet. Chem. 1997, 545-546, 381–398. 27. R. Peters, D.F. Fischer, Org. Lett. 2005, 7, 4137–4140. 28. (a) J. Park, S. Lee, K.H. Ahn, C-W. Cho, Tetrahedron Lett. 1995, 36, 7263–7266. (b) J. Park, S. Lee, K.H. Ahn, C-W. Cho, Tetrahedron Lett. 1996, 37, 6137–6140. 29. The resulting ligands have been used in the intermolecular copper(I)-catalysed cyclopropanation of styrene with diazoacetates and also in the intramolecular version of this reaction with ene-diazoacetates. (a) S-G. Kim, C-W. Cho, K.H. Ahn, Tetrahedron: Asymmetry 1997, 8, 1023–1026; (b) S-G. Kim, C-W. Cho, K.H. Ahn, Tetrahedron 1999, 55, 10079–10086. 30. C.J. Richards, D.E. Hibbs, M.B. Hursthouse, Tetrahedron Lett. 1995, 36, 3745–3748. 31. E.L. Strangeland, T. Sammakia, Tetrahedron 1997, 53, 16503–16510. 32. K-G. Chung, Y. Miyake, S. Uemura, Perkin Trans. 1 2000, 2725–2729. 33. Copper-based catalytic system: A.S.E. Karlstr¨om, F.F. Huerta, G.J. Meuzelaar, J-E. B¨ackvall, Synlett 2001, 923–926. 34. K-G. Chung, Y. Miyake, S. Uemura, Perkin Trans. 1 2000, 15–18. 35. (a) S-I. Fukuzawa, H. Kato, Synlett 1998, 727–728; (b) C. Bolm, K. Mu˜niz-Fern´andez, A. Seger et al. J. Org. Chem. 1998, 63, 7860–7867; (c) C. Bolm, K. Mu˜niz-Fern´andez, A. Seger, G. Raabe, Synlett 1997, 1051–1052; (d) C. Bolm, K. Mu˜niz, J.P. Hildebrand, Org. Lett. 1999, 1, 491–493; (e) W. Zhang, H. Yoshinaga, Y. Imai et al. Synlett 2000, 1512–1514; (f) C. Bolm, K. Mu˜niz, Chem. Commun. 1999, 1295–1296; (g) C. Bolm, N. Hermanns, J.P. Hildebrand, K. Mu˜niz, Angew. Chem., Int. Ed. 2000, 39, 3465–3467; (h) C. Bolm, N. Hermanns, M. Kesselgruber, J.P. Hildebrand, J. Organomet. Chem. 2001, 624, 157–161; (i) C. Bolm, N. Hermanns, A. Claßen, K. Mu˜niz, Bioorg. Med. Chem. Lett. 2002, 12, 1795–1798; (j) C. Bolm, M. Kesselgruber, A. Grenz et al. New. J. Chem. 2001, 25, 13–15; (k) C. Bolm, J. Rudolph, J. Am. Chem. Soc. 2002, 124, 14850–14851; (l) M. Li, K. Yuan, Y-Y. Li et al. Tetrahedron: Asymmetry 2003, 14, 3347–3352; (m) J. Rudolph, F. Schmidt, C. Bolm, Adv. Synth. Catal. 2004, 346, 867–872; (n) J. Rudolph, N. Hermanns,

270

36.

37. 38.

39. 40. 41. 42.

43. 44. 45. 46. 47. 48. 49. 50. 51. 52.

53.

Ferrocenes: Ligands, Materials and Biomolecules ¨ ¸ ubukc¸u, F. Schmidt, C. Bolm, C. Bolm, J. Org. Chem. 2004, 69, 3997–4000; (o) S. Ozc Org. Lett. 2005, 7, 1407–1409; (p) J. Rudolph, F. Schmidt, C. Bolm, Synthesis 2005, 840–842; (q) F. Schmidt, J. Rudolph, C. Bolm, Synthesis, 2006, 3625–3630. (a) K.H. Ahn, C-W. Cho, J. Park, S. Lee, Tetrahedron: Asymmetry 1997, 8, 1179–1185; (b) W. Zhang, T. Hirao, I. Ikeda, Tetrahedron Lett. 1996, 37, 4545–4548; (c) S-L. You, X-L. Hou, L-X. Dai et al. J. Org. Chem. 2002, 67, 4684–4695; (d) S. Lee, J.H. Koh, J. Park, J. Organomet. Chem. 2001, 637-639, 99–106; (e) W-P. Deng, X-L. Hou, L-X. Dai et al. Chem. Commun. 2000, 285–286; (f) L-X. Dai, X-L. Hou, W.P. Deng, et al. Pure Appl. Chem. 1999, 71, 1401–1405; (g) R. Shintani, M.M-C. Lo, G.C. Fu, Org. Lett. 2000, 2, 3695–3697; (h) E. Manoury, J.S. Fossey, H. A¨ıt-Haddou et al. Organometallics 2000, 19, 3736–3739; (i) S-L. You, Y-G. Zhou, X-L. Hou, L-X. Dai, Chem. Commun. 1998, 2765–2766; (j) S-L. You, X-L. Hou, L-X. Dai, Tetrahedron: Asymmetry 2000, 11, 1495–1500; (k) W-P. Deng, S-L. You et al. J. Am. Chem. Soc., 2001, 123, 6508–6519; (l) L-X. Dai, T. Tu, S-L. You et al. Acc. Chem. Res. 2003, 36, 659–667; (m) C-W. Cho, J-H. Son, K.H. Ahn, Tetrahedron: Asymmetry, 2006, 17, 2240–2246; (n) F.M. Geisler, G. Helmchen, J. Org. Chem. 2006, 71, 2486–2492. Y.M. Malone, P.J. Guiry, J. Organomet. Chem. 2000, 603, 110–115. (a) A.J. Hennessy, Y.M. Malone, P.J. Guiry, Tetrahedron Lett. 1999, 40, 9163–9166; (b) T.G. Kilroy, A.J. Hennessy, D.J. Connolly et al. J. Mol. Catal. A: Chem. 2003, 196, 65–81; (c) D. Kiely, P.J. Guiry, J. Organomet. Chem. 2003, 687, 545–561; (d) D. Kiely, P.J. Guiry, Tetrahedron Lett. 2003, 44, 7377–7380; (e) W-P. Deng, X-L. Hou, L-X. Dai, X-W. Dong, Chem. Commun. 2000, 1483–1484; (f) T. Tu, W-P. Deng, X-L. Hou et al. Chem. Eur. J. 2003, 9, 3073–3081; (g) T. Tu, X-L. Hou, L-X. Dai, Org. Lett. 2003, 5, 3651–3653. (a) F. Naud, C. Malan, F. Spindler et al. Adv. Synth. Catal. 2006, 348, 47–50; (b) D.M. Tellers, M. Bio, Z.J. Song et al. Tetrahedron: Asymmetry 2006, 17, 550–553. J. Rouzaud, M.D. Jones, R. Raja et al. Helv. Chim. Acta 2003, 86, 1753–1759. S-M. Lu, X-W. Han, Y-G. Zhou, Adv. Synth. Catal. 2004, 346, 909–912. (a) T. Sammakia, E.L. Strangeland, J. Org. Chem. 1997, 62, 6104–6105; (b) Y. Nishibayashi, I. Takei, S. Uemura, M. Hidai, Organometallics 1999, 18, 2291–2293; (c) Y. Arikawa, M. Ueoka, K. Matoba et al. J. Organomet. Chem. 1999, 572, 163–168. Y. Nishibayashi, A. Yamauchi, G. Onodera, S. Uemura, J. Org. Chem. 2003, 68, 5875–5880. (a) Y. Nishibayashi, K. Segawa, H. Takada et al. Chem. Commun. 1996, 847–848; (b) Y. Nishibayashi, I. Takei, S. Uemura, M. Hidai, Organometallics 1998, 17, 3420–3422. I. Takei, Y. Nishibayashi, Y. Arikawa et al. Organometallics 1999, 18, 2271–2274. I. Takei, Y. Nishibayashi, Y. Ishii et al. Chem. Commun. 2001, 2360–2361. S. G´erard, Y. Pressel, O. Riant, Tetrahedron: Asymmetry 2005, 16, 1889–1891. G. Onodera, Y. Nishibayashi, S. Uemura, Angew. Chem. 2006, 118, 3903–3906; Angew. Chem. Int. Ed. 2006, 45, 3819–3822. S-I. Fukuzawa, K. Takahashi, H. Kato, H. Yamazaki, J. Org. Chem. 1997, 62, 7711–7716. (a) Y. Donde, L.E. Overman, J. Am. Chem. Soc. 1999, 121, 2933–2934; (b) C.E. Anderson, Y. Donde, C.J. Douglas, L.E. Overman, J. Org. Chem. 2005, 70, 648–657. (a) M. Lautens, S. Hiebert, J-L. Renaud, Org. Lett. 2000, 2, 1971–1973; (b) M. Li, X-X. Yan, W. Hong et al. Org. Lett. 2004, 6, 2833–2835. (a) S. Fukuzawa, Y. Yahara, A. Kamiyama et al. Org. Lett. 2005, 7, 5809–5812; (b) D. Carmona, R. Medrano, I.T. Dobrinovich et al. J. Organomet. Chem. 2006, 691, 5560–5566. (a) R. Shintani, G.C. Fu, J. Am. Chem. Soc. 2003, 125, 10778–10779; (b) A. Su´arez, C.W. Downey, G.C. Fu, J. Am. Chem. Soc. 2005, 127, 11244–11245; (c) W. Gao,

Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors

54. 55. 56.

57.

58.

59.

60. 61. 62. 63. 64. 65.

66. 67. 68.

69. 70. 71. 72.

271

X. Zhang, M. Raghunath, Org. Lett. 2005, 7, 4241–4244; (d) W. Zeng, Y-G. Zhou, Org. Lett. 2005, 7, 5055–5058. X-X. Yan, Q. Peng, Y. Zhang et al. Angew. Chem. 2006, 118, 2013–2017; Angew. Chem. Int. Ed. 2006, 45, 1979–1983. M. Shi, L-H. Chen, C-Q. Li, J. Am. Chem. Soc. 2005, 127, 3790–3800. (a) S-L. You, Y-M. Luo, W-P. Deng et al. J. Organomet. Chem. 2001, 637-639, 845–849; (b) Ref. 36c. (c) T.E. Pickett, C.J. Richards, Tetrahedron Lett. 2001, 42, 3767–3769; (d) S-L. You, X-L. Hou, L-X. Dai et al. Chem. Commun. 2000, 1933–1934. (e) Ref. 35b. (f) R. Kuwano, T. Uemura, M. Saitoh, Y. Ito, Tetrahedron Lett. 1999, 40, 1327–1330; (g) W. Zhang, T. Kida, Y. Nakatsuji, I. Ikeda, Tetrahedron Lett. 1996, 37, 7995–7998; (h) W. Zhang, T. Shimanuki, T. Kida et al. J. Org. Chem. 1999, 64, 6247–6251; (i) R. Salter, T.E. Pickett, C.J. Richards, Tetrahedron: Asymmetry 1998, 9, 4239–4247; (j) A.J. Locke, T.E. Pickett, C.J. Richards, Synlett 2001, 141–143; (k) W. Zhang, Y-I. Yoneda, T. Kida et al. J. Organomet. Chem. 1999, 574, 19–23. (a) J.M. Longmire, B. Wang, X. Zhang, Tetrahedron Lett. 2000, 41, 5435–5439; ˇ epniˇcka, New J. Chem. 2002, 26, 567–575. For alternative route that affords both (b) P. Stˇ enantiomers after resolution, see: (c) B. Breit, D. Breuninger, Synthesis 2005, 2782–2786. (a) B. Breit, D. Breuninger, Synthesis 2005, 147–157; (b) B. Breit, D. Breuninger, Eur. J. Org. Chem. 2005, 3916–3929; (c) B. Breit, D. Breuninger, J. Am. Chem. Soc. 2004, 126, 10244–10245. (a) S-L. You, X-L. Hou, L-X. Dai, J. Organomet. Chem. 2001, 637-639, 762–766; (b) S-L. You, X-L. Hou, L-X. Dai, X-Z. Zhu, Org. Lett. 2001, 3, 149–151; (c) J.M. Longmire, B. Wang, X. Zhang, J. Am. Chem. Soc. 2002, 124, 13400–13401. (d) M. Lamaˇc, ˇ epniˇcka Organometallics 2007, 26, 5042–5049. J. Tauchman, I. C´ısaˇrov´a, P. Stˇ J. Garc´ıa, A. Moyano, M. Rosol Tetrahedron, 2007, 63, 1907–1912. Y. Yuan, G. Raabe, C. Bolm, J. Organomet. Chem. 2005, 690, 5747–5752. U. Nettekoven, M. Widhalm, P.C.J. Kamer et al. Organometallics 2000, 19, 2299–2309. D. Vinci, N. Mateus, X. Wu et al. Org. Lett. 2006, 8, 215–218. B. Ferber, H.B. Kagan, Adv. Synth. Catal. 2007, 349, 493–507 (review). (a) P. Ditter, O. Samuel, S. Taudine, H.B. Kagan, Tetrahedron: Asymmetry 1994, 5, 549– 552; (b) D. Guillaneux, H.B. Kagan, J. Org. Chem. 1995, 60, 2502–2505; (c) O. Riant, G. Argouarch, D. Guillaneux et al. J. Org. Chem. 1998, 63, 3511–3514; (d) D.H. Hua, N.M. Lagneau, Y. Chen et al. J. Org. Chem. 1996, 61, 4508–4509; (e) N.M. Lagneau, Y. Chen, P.M. Robben et al. Tetrahedron 1998, 54, 7301–7334; (f) H.K. Cotton, F.F. Huerta, J-E. B¨ackvall, Eur. J. Org. Chem. 2003, 2756–2763. F. Rebi`ere, O. Riant, L. Ricard, H.B. Kagan, Angew. Chem., Int. Ed. Engl. 1993, 32, 568–570. G. Argouarch, O. Samuel, O. Riant et al. Eur. J. Org. Chem. 2000, 2893–2899. (a) H.L. Pedersen, M. Johansen, Chem. Commun. 1999, 2517–2518; (b) H.L. Pedersen, M. Johansen, J. Org. Chem. 2002, 67, 7982–7994; (c) M. Lotz, G. Kramer, P. Knochel, Chem. Commun. 2002, 2546–2547; (d) J.F. Jensen, I. Søtofte, H.O. Sørensen, M. Johannsen, J. Org. Chem. 2003, 68, 1258–1265; (e) A. Herrbach, A. Marinetti, O. Baudoin et al. J. Org. Chem. 2003, 68, 4897–4905; (f) R. Kloetzing, P. Knochel, Tetrahedron: Asymmetry 2006, 17, 116–123; (g) V.E. Albrow, A.J. Blake, R. Fryatt et al. Eur. J. Org. Chem. 2006, 2549–2557. J.G. Seitzberg, C. Dissing, I. Søtofte et al. J. Org. Chem. 2005, 70, 8332–8337. M. Lotz, K. Polborn, P. Knochel, Angew. Chem. Int. Ed. 2002, 41, 4708–4711. M. Raghunath, W. Gao, X. Zhang, Tetrahedron: Asymmetry 2005, 16, 3676–3681. (a) J. Priego, O.G. Manche˜no, S. Cabrera et al. Chem. Commun. 2002, 2512–2513; (b) O.G. Manche˜no, J. Priego, S. Cabrera et al. J. Org. Chem. 2003, 68, 3679–3686;

272

73. 74. 75. 76. 77. 78. 79. 80. 81. 82.

83. 84. 85.

86. 87. 88. 89. 90. 91. 92. 93. 94. 95.

96. 97. 98.

Ferrocenes: Ligands, Materials and Biomolecules (c) O.G. Manche˜no, R.G. Array´as, J.D. Carretero, J. Am. Chem. Soc. 2004, 126, 456–457; (d) S. Cabrera, R.G. Array´as, J.C. Carretero, Angew. Chem. Int. Ed. 2004, 43, 3944–3947; (e) A.S. Gonz´alez, R.G. Array´as, J.C. Carretero, Org. Lett. 2006, 8, 2977–2980. J. Priego, O.G. Manche˜no, S. Cabrera, J.D. Carretero, J. Org. Chem. 2002, 17, 1346–1353. C. Bolm, K. Kesselgruber, G. Raabe, Organometallics 2002, 21, 707–710. L. Xiao, K. Mereiter, F. Spindler, W. Weissensteiner, Tetrahedron:Asymmetry 2001, 12, 1105–1108. Y. Wang, W. Weissensteiner, K. Mereiter, F. Spindler, Helv. Chim. Acta. 2006, 89, 1772–1782. (a) O. Riant, O. Samuel, H.B. Kagan, J. Am. Chem. Soc. 1993, 115, 5835–5836; (b) O. Riant, O. Samuel, T. Flessner et al. J. Org. Chem. 1997, 62, 6733–6745. (S)-1,2,4-Butanetriol is accessible from l-malic acid by reduction with BH3 -B(OMe)3 : S. Hanessian, A. Ugolini, D. Dub´e, A. Glamyan, Can. J. Chem. 1984, 62, 2146–2147. G. Iftime, J-C. Daran, E. Manoury, G.G.A. Balavoine, J. Organomet. Chem. 1998, 565, 115–124. J. Chiffre, Y. Coppel, G.G.A. Balavoine et al. Organometallics 2002, 21, 4552–4555. G. Iftime, J-C. Daran, E. Manoury, G.G.A. Balavoine, Organometallics 1996, 15, 4808–4815. Representative examples: (a) A.G. Neo, A. Gref, O. Riant, Chem. Commun. 1998, 2353–2354; (b) A. Skibniewski, G. Bluet, N. Druz´e, O. Riant, Synthesis 1999, 459–462; (c) G.G.A. Balavoine, J-C. Daran, G. Iftime et al. Organometallics 1999, 18, 21–29; (d) G. Iftime, G.G.A. Balavoine, J-C. Daran et al. C. R. Acad. Sci. Paris, Ser. IIc, Chimie 2000, 3, 139–146; (e) V. Mamane, O. Riant, Tetrahedron 2001, 57, 2555–2561; (f) V. Mamane, A. Gref, F. Leﬂoch, O. Riant, J. Organomet. Chem. 2001, 637–639, 84–88. L. Routaboul, S. Vincendeau, J-C. Daran, E. Manoury, Tetrahedron:Asymmetry 2005, 16, 2685–2690. ˇ epniˇcka, I. C´ısaˇrov´a, Inorg. Chem. 2006, 45, 8785–8798. (b) P. Stˇ ˇ epniˇcka, (a) P. Stˇ M. Lamaˇc, I. C´ısaˇrov´a, J. Organomet. Chem. 2008, 693, 446–456. (a) J.H. Lee, S.U. Son, Y.K. Chung, Tetrahedron: Asymmetry 2003, 14, 2109–2113. For a polymer-supported form, see: (b) H-J. Park, J.W. Han, H. Seo et al. J. Mol. Catal. A: Chem. 2001, 174, 151–157. G. Argouarch, O. Samuel, H.B. Kagan, Eur. J. Org. Chem. 2000, 2885–2891. M-J. Jin, V.B. Takale, M.S. Sarkar, Y-M. Kim, Chem. Commun. 2006, 663–664. W. Chen, S.M. Roberts, J. Whittall, A. Steiner, Chem. Commun. 2006, 2916–2918. N. Taniguchi, M. Uemura, Tetrahedron Lett. 1998, 39, 5385–5388. X-X. Yan, C-G. Liang, Y. Zhang et al. Angew. Chem. Int. Ed. 2005, 44, 6544–6546. A similar compound derived from ethane-1,2-diamine has been also reported: A. MassonSzymczak, O. Riant, A. Gref, H.B. Kagan, J. Organomet. Chem. 1996, 511, 193–197. N. Mateus, L. Routaboul, J-C. Daran, E. Manoury, J. Organomet. Chem. 2006, 691, 2297–2310. D. Price, N.S. Simpkins, Tetrahedron Lett. 1995, 36, 6135–6136. Y. Nishibayashi, Y. Arikawa, K. Ohe, S. Uemura, J. Org. Chem. 1996, 61, 1171–1174. (a) M. Tsukazaki, M. Tinkl, A. Roglans et al. J. Am. Chem. Soc. 1996, 118, 685–686; (b) H. Jendralla, E. Paulus, Synlett 1997, 471–472; (c) R. Laufer, U. Veith, N.J. Taylor, V. Snieckus, Org. Lett. 2000, 2, 629–631; (d) R. Laufer, U. Veith, N.J. Taylor, V. Snieckus, Can. J. Chem. 2006, 84, 356–369. For a review on ortho-metalation of tertiary amides, see: V. Snieckus, Chem. Rev. 1990, 90, 879–933. C. Metallinos, H. Szillat, N.J. Taylor, V. Snieckus, Adv. Synth. Catal. 2003, 345, 370–382. C. Metallinos, V. Snieckus, Org. Lett. 2002, 4, 1935–1938.

Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors

273

99. (a) M. Rosenblum, R.B. Woodward, J. Am. Chem. Soc. 1958, 80, 5443–5449; (b) A.N. Nesmeyanov, E.V. Leonova, N.S. Kochetkova et al. J. Organomet. Chem. 1975, 96, 275–278; (c) M. Hisatome, O. Tachikawa, M. Sasho, K. Yamakawa, J. Organomet. Chem. 1981, 217, C17–C20; (d) T-Y. Dong, M-C. Lin, L. Lee et al. J. Organomet. Chem. 2003, 679, 181–193; (e) Y. Nagai, J. Hooz, R.A. Benkeser, Bull. Chem. Soc. Jpn. 1964, 37, 53–60; (f) H. Falk, G. Haller, K. Schl¨ogl, Monatsh. Chem. 1967, 98, 592–602; (g) A. Kasahara, T. Izumi, Y. Yoshida, I. Shimizu, Bull. Chem. Soc. Jpn. 1982, 55, 1901–1906; (h) T-Y. Dong, C-H. Huang, C-K. Chang et al. A. Yeh, J. Am. Chem. Soc. 1993, 115, 6357–6368. 100. It is noteworthy that acylation of 1,1 -bis(trimethylsilyl)- or 1,1 -bis(tributylstannyl) ferrocene with acetyl chloride/AlCl3 affords a mixture of 1 -acetyl-, 2-acetyl, and 3-acetyl1-(trimethylsilyl or tributylstannyl)ferrocenes (sic!): (a) A.F. Cunningham, Jr, J. Am. Chem. Soc. 1991, 113, 4864–4870. Likewise, borylation of 1,1 -bis(trimethylsilyl)ferrocene with BCl3 gives mixtures of 1,1 - and 1,3-bis(dichloroboryl)ferrocenes: (b) P.A. Deck, T.S. Fisher, J. Sloan Downey, Organometallics 1997, 16, 1193–1196. 101. Representative examples: (a) L. Westman, K.L. Rinehart, Jr., Acta. Chim. Scand. 1962, 100, 1199–1205; (b) H. Falk, O. Hofer, K. Schl¨ogl, Monatsh. Chem. 1969, 100, 624–648; (c) H. Lehner, K. Schl¨ogl, Monatsh. Chem. 1970, 101, 895–911. 102. (a) T. Izumi, T. Hino, J. Chem. Tech. Biotechnol. 1992, 55, 325–331; (b) T. Izumi, T. Hino, A. Ishihara, J. Chem. Tech. Biotechnol. 1993, 56, 45–49; (c) N. D’Antona, D. Lambusta, R. Morrone et al. Tetrahedron: Asymmetry 2004, 15, 3835–3840. 103. (a) D.L. Compton, T.B. Rauchfuss, Organometallics 1994, 13, 4367–4376; (b) G. Masson, P. Beyer, P.W. Cyr et al. Macromolecules 2006, 39, 3720–3730. 104. Vilsmeier–Haack formylation of long-chain 1,1 -dialkylferrocenes also produces 3formylated product: P. Massiot, M. Imp´eror-Clerc, M. Veber, R. Deschenaux, Chem. Mater. 2005, 17, 1946–1951. 105. (a) Borylation of alkylferrocenes and 1,1 -dibromoferrocene: B. Wrackmeyer, U. D¨orﬂer, J. Rinck, M. Herberhold, Z. Naturforsch. 1994, 49b, 1403–1406; (b) Iridium-catalysed borylation: A. Datta, A. K¨ollhofer, H. Plenio, Chem. Commun. 2004, 1508–1509. 106. With boron trihalides, ferrocene itself becomes borylated in a step-wise manner to give monoboryl derivative, which is subsequently converted to a mixture of 1,1 - and 1,3diborylferrocenes. Extensive borylation gives exclusively 1,1 ,3,3 -tetraborylferrocenes. (a) B. Wrackmeyer, U. D¨orﬂer, M. Herberhold, Z. Naturforsch. 1993, 48b, 121–123; (b) A. Appel, H. N¨oth, M. Schmidt, Chem. Ber. 1995, 128, 621–626. 107. Lithiation of FcPPh2 with LiBu in the presence of TMEDA followed by quenching with chloro(diphenyl)phosphine was shown to afford mixtures of dppf, and its 1,2-, and 1,3isomers; however, with the latter dominating: (a) I.R. Butler, W.R. Cullen, Organometallics 1986, 5, 2537–2542; (b) I.R. Butler, L.J. Hobson, S.M.E. Macan, D.J. Williams, Polyhedron 1993, 12, 1901–1905. 108. C. Pichon, B. Odell, J.M. Brown, Chem. Commun. 2004, 598–599. 109. (a) S. Bradley, M.M. Corradi, K.D. Camm et al. J. Organomet. Chem. 2002, 656, 49–56; (b) J. Christoffers, T. Werner, A. Baro, P. Fischer, J. Organomet. Chem. 2004, 689, 3550–3555; (c) P.A. Deck, C.E. Kroll, W.G. Hollis Jr, F.R. Fronczek, J. Organomet. Chem. 2001, 637–639, 107–115; (d) The reaction between Li[(Ph2 P)2 C5 H3 -1,3] and FeCl2 gives the expected 1,1 ,3,3 -tetraphosphinylated ferrocene, while a simultaneous addition of Li[C5 H5 ] and Li[(Ph2 P)2 C5 H3 -1,3] to FeCl2 yields predominantly [Fe{η5 -C5 H3 (Ph2 P)2 1,3}(η5 -C5 H5 )]: R. Broussier, E. Bentabet, P. Mellet et al. J. Organomet. Chem. 2000, 598, 365–373. 110. (a) P. Bickert, B. Hildebrandt, K. Hafner, Organometallics 1984, 3, 653–657; (b) W.E. Lindsell, L. Xinxin, J. Chem. Research (S) 1998, 62–63.

274

Ferrocenes: Ligands, Materials and Biomolecules

111. (a) B. Rybtchinski, D. Milstein, Angew. Chem. Int. Ed. 1999, 38, 870–883; (b) M. Albrecht, G. van Koten, Angew. Chem. Int. Ed. 2001, 40, 3750–3781. 112. (a) A.A. Koridze, A.M. Sheloumov, S.A. Kuklin et al. Russ. Chem. Bull., Int. Ed. 2002, 51, 1077–1078 (original reference: Izv. Akad. Nauk, Ser. Khim. 2002, 51, 988–990). (NB The ruthenocene analog and the related 1-[(diphenylphosphino)methyl]3-N,N-[(dimethylamino)methyl]ferrocene were also synthesised.); (b) E.J. Farrington, E.M. Viviente, B.S. Williams et al. Chem. Commun. 2002, 308–309; (c) A.A. Koridze, S.A. Kuklin, A.M. Sheloumov et al. Russ. Chem. Bull., Int. Ed. 2003, 52, 2754–2756 (original reference: Izv. Akad. Nauk, Ser. Khim. 2003, 52, 2607–2609); (d) A.A. Koridze, S.A. Kuklin, A.M. Sheloumov et al. Russ. Chem. Bull., Int. Ed. 2003, 52, 2757–2759 (original reference: Izv. Akad. Nauk, Ser. Khim. 2003, 52, 2610–2612); (e) A.A. Koridze, S.A. Kuklin, A.M. Sheloumov et al. Organometallics 2004, 23, 4585–4593. 113. B.F. Bonini, E. Capit´o, M. Comes-Franchini et al. Arkivoc 2006, 85–96. 114. M. Steurer, K. Tiedl, Y. Wang, W. Weissensteiner, Chem. Commun. 2005, 4929–4931. 115. (a) [chiral sulfoxide and acetal] B. Ferber, S. Top, R. Welter, G. Jaouen, Chem. Eur. J. 2006, 12, 2081–2086; (b) [chiral amine and sulfoxide] see Ref. 102c. See also: (c) F. Rebi`ere, O. Riant, L. Ricard, H.B. Kagan, Angew. Chem. Int. Ed. Engl. 1993, 32, 568–570. 116. Lithiation of 1,1 -dibromoferrocene with LiN(i-Pr)2 followed by reaction with ClPPh2 gives a mixture of 2-(diphenylphosphino)-1,1-dibromoferrocene, 2,5-bis(diphenylphosphino)1,1 -dibromoferrocene, and 2,2 -bis(diphenylphosphino)-1,1-dibromoferrocene. The second compound gives rise 1,3-bis(diphenylphosphino)ferrocene after lithiation and quenching with water. I.R. Butler, M.G.B. Drew, C.H. Greenwell et al. Inorg. Chem. Commun. 1999, 2, 576–580. 117. 1,3-Bis[(6-propanoylamino)pyrid-2-yl]ferrocene: J. Westwood, S.J. Coles, S.R. Collinson et al. Organometallics 2004, 23, 946–951. 118. T. Izumi, S. Oohashi, Y. Tate, J. Heterocycl. Chem. 1993, 30, 967–975. 119. Prominent examples include compounds derived from 1,3-ferrocenedicarboxylic (a-c) and 3(1 ,3-dialkylferrocen-1-yl)propanoic acids (d): (a) T. Chuard, S.J. Cowling, M. FernandezCiurleo et al. Chem. Commun. 2000, 2109–2110. (b) R. Deschenaux, J. Santiago, Tetrahedron Lett. 1994, 35, 2169–2172; (c) See Ref. 110b; (d) See Ref. 104b (only selected examples). 120. (a) T. Muraoka, K. Kinbara, T. Aida, Nature 2006, 440, 512–515; (b) T. Muraoka, K. Kinbara, T. Aida, J. Am. Chem. Soc. 2006, 128, 11600–11605. 121. The preparation of multiply bridged ferrocenes is usually tedious (convergent) and often hampered by low yields. 122. (a) W.E. Watts, Organomet. Chem. Rev. 1967, 2, 231–254; (b) R.W Heo, T.R. Lee, J. Organomet. Chem. 1999, 578, 31–42; (c) G.B. Shul’pin, M.I. Rybinskaya, Usp. Khim. 1974, 53, 1524–1553 (English translation: G.B. Shul’pin, M.I. Rybinskaya, Russ. Chem. Rev. 1974, 43, 716–732). 123. See also Refs. 122a, 122c. Selected examples [bridge type]: [S(VI)N] (a) R.A. Abramovitch, C.I. Azogu, R.G. Sutherland, Chem. Commun. 1969, 1439–1140; (b) R.B. Abramovitch, J.L. Atwood, M.L. Good, B.A. Lampert, Inorg. Chem. 1975, 14, 3085–3089; [CP] and [CS] (c) R. Resendes, J.M. Nelson, A. Fischer et al. J. Am. Chem. Soc. 2001, 123, 2116–2126; [NCC] (d) H. Plenio, J. Yang, R. Diodone, J. Heinze, Inorg. Chem. 1994, 33, 4098–4104; [CNC] (e) H-J. Lorkowski, P. Kieselack, Chem. Ber. 1966, 99, 3619–3627; (f) H. Plenio, J. Yang, R. Diodone, J. Heinze, Inorg. Chem. 1994, 33, 4098–4104; (g) K. Osakada, T. Sakano, M. Horie, Y. Suzaki, Coord. Chem. Rev. 2006, 250, 1012–1022; (h) Y. Suzaki, M. Horie, T. Sakano, K. Osakada, J. Organomet. Chem. 2006, 621, 3403–3407; (i) T. Moriuchi, S. Bandoh, Y. Miyaji, T. Hirao, J. Organomet. Chem. 2000, 599, 135–142; (j) C.M. N’Diaye, L.A. Maciejewski, J.S. Brocard,

Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors

124.

125.

126. 127.

275

C. Biot, Tetrahedron Lett. 2001, 42, 7221–7223; (k) T. Moriuchi, T. Ikeda, T. Hirao, Organometallics 1995, 14, 3578–3580; [COC] (l) K.L. Rinehart, A.K. Frerichs, P.A. Kittle et al. J. Am. Chem. Soc. 1960, 82, 4111–4112; (m) G.R. Knox, J.D. Munro, P.L. Pauson et al. J. Chem. Soc. 1961, 4619–4624; (n) T.A. Mashburn, Jr, C.R. Hauser, J. Org. Chem. 1961, 26, 1671–1672; (o) E.C. Winslow, E.W. Brewster, J. Org. Chem. 1961, 26, 2982; (p) K. Schl¨ogl, A. Mohar, Monatsh. Chem. 1961, 92, 219–235; (q) A.N. Nesmeyanov, E.G. Perevalova, Yu.A. Ustynyuk et al. Izv. Akad. Nauk SSSR, Ser. Khim. 1966, 9, 1646–1650; (r) P.L. Pauson, M.A. Sandhu, W.E. Watts, J. Chem. Soc. C 1966, 251–255; (s) M.J.A. Hibb, W.E. Watts, J. Chem. Soc. C 1969, 1469–1472; (t) E.S. Bolton, P.L. Pauson, M.A. Sandhu, W.E. Watts, J. Chem. Soc. C 1969, 2260–2263; (u) K. Yamakawa, M. Hisatome, J. Organomet. Chem. 1973, 52, 407–424; (v) M. Hillman, J.D. Austin, Organometallics 1987, 6, 1737–1743; (w) R.C. Petter, C.I. Milberg, S.J. Rao, Tetrahedron Lett. 1990, 31, 6117–6120; (x) G. Iftime, J-C. Daran, E. Manoury, G.G.A. Balavoine, Organometallics 1996, 15, 4808–4815; (y) S. Lisac, V. Rapi´c, Z. Zori´c, N. Filipovi´c-Marini´c, Croat. Chem. Acta 1997, 70, 1021–1037; (z) G. Iftime, J-C. Daran, E. Manoury, G.G.A. Balavoine, Angew. Chem. Int. Ed. 1998, 37, 1698–1701; (aa) G. Iftime, J-C. Daran, E. Manoury, G.G.A. Balavoine, J. Organomet. Chem. 1998, 565, 115–124; (bb) A. Kundu, S. Prabhakar, M. Vairamani, S. Roy, Organometallics 1999, 18, 2782–2785; (cc) M.A. Carroll, A.J.P. White, D.A. Widdowson, D.J. Williams, Perkin Trans. 1 2000, 1551–1557; [SiOSi] and [GeOGe]: (dd) R.L. Schaaf, P.T. Kan, C.T. Lenk, J. Org. Chem. 1961, 26, 1790–1795; (ee) M. Kumada, T. Kondo, K. Mimura et al. J. Organomet. Chem. 1972, 43, 307–314; (ff) T. Kondo, K. Yamamoto, M. Kumada, J. Organomet. Chem. 1972, 43, 315–321; [CP(III)C] (gg) T. H¨ocher, A. Cinquantini, P. Zanello, E. Hey-Hawkins, Polyhedron 2005, 24, 1340–1346; [CP(V)C] (hh) E. Soulier, J-J. Yaouanc, P. Laurent et al. Eur. J. Org. Chem. 2000, 3497–3503; [EEE] and [EYE] systems, where E = chalcogenide, and Y = chalcogenide or other atom (C, Si, Ge, Sn, P, As, Sb, and B): (ii) A. Davison, J.C. Smart, J. Organomet. Chem. 1969, 19, P7–P8; (jj) J.J. Bishop, A. Davison, M.L. Katcher et al. J. Organomet. Chem. 1971, 27, 241–249; (kk) R.E. Hollands, A.G. Osborne, I. Townsend, Inorg. Chim. Acta 1979, 37, L541; (ll) A. Davison, J.S. Smart, J. Organomet. Chem. 1979, 174, 321–334; (mm) M. Herbehold, P. Leitner, U. Thewalt, Z. Naturforsch. 1990, 45b, 1503–1507; (nn) M. Herbehold, P. Leitner, J. Organomet. Chem. 1991, 411, 233–237 and references therein; (oo) M. Herbehold, C. D¨ornh¨ofer, A. Scholz, G.-X. Jin, Phosphorus, Sulfur, and Silicon 1992, 64, 161–168; (pp) M. Herbehold, H.-D. Brendel, J. Organomet. Chem. 1993, 458, 205–209. (a) M. Herbehold, Angew. Chem. Int. Ed. Engl. 1995, 34, 1837–1839 (a review); E = S and Se: (b) J.K. Pudelski, D.P. Gates, R. Rulkens et al. Angew. Chem. Int. Ed. Engl. 1995, 34, 1506–1508; (c) R. Rulkens, D.P. Gates, E. Balaishis et al. J. Am. Chem. Soc. 1997, 111, 10976–10986; E = BNR2 : (d) H. Braunschweig, R. Dirk, M. M¨uller et al. Angew. Chem. Int. Ed. Engl. 1997, 36, 2338–2340; (e) A. Berenbaum, H. Braunschweig, R. Dirk et al. J. Am. Chem. Soc. 2000, 122, 5765–5774. (a) D. Seyferth, H.P. Withers, Jr., J. Organomet. Chem. 1980, 185, C1–C5; (b) D. Seyferth, H.P. Withers, Jr., Organometallics 1982, 1, 1275–1282; (c) A.G. Osborne, R.H. Whiteley, R.E. Meads, J. Organomet. Chem. 1980, 193, 345–357; (d) I.R. Butler, W.R. Cullen, F.W.B. Einstein et al. Organometallics 1983, 2, 128–135; (e) H. Brunner, J. Klankermayer, M. Zabel, J. Organomet. Chem. 2000, 601, 211–219. The synthesis and coordination of fcPNR2 has also been reported: M. Herbehold, F. Hertel, W. Millius, B. Wrackmeyer, J. Organomet. Chem. 1999, 582, 353–357. (a) C.H. Honeyman, D.A. Foucher, F.Y. Dahmen et al. Organometallics 1995, 14, 5503–5512; (b) T. Mizuta, T. Yamasaki, H. Nakazawa, K. Miyoshi, Organometallics

276

128.

129.

130. 131. 132. 133.

134.

135.

136. 137. 138. 139. 140. 141. 142. 143. 144. 145.

146. 147. 148. 149. 150. 151.

152.

Ferrocenes: Ligands, Materials and Biomolecules 1996, 15, 1093–1100; (c) T. Mizuta, M. Onishi, K. Miyoshi, Organometallics 2000, 19, 5005–5009; (d) C.E.B. Evans, A.J. Lough, H. Grondey, I. Manners, New. J. Chem. 2000, 24, 447–453; (e) T. Mizuta, Y. Imamura, K. Miyoshi, J. Am. Chem. Soc. 2003, 125, 2068–2069; (f) Y. Imamura, K. Kubo, T. Mizuta, K. Miyoshi, Organometallics 2006, 25, 2301–2307. (a) O.J. Curnow, G. Huttner, S.J. Smail, M.M. Turnbull, J. Organomet. Chem. 1996, 524, 267–270; (b) J.J. Adams, O.J. Curnow, G. Huttner et al. J. Organomet. Chem. 1999, 577, 44–57. (a) M. Rosenblum, A.K. Banerjee, N. Danieli et al. J. Am. Chem. Soc. 1963, 85, 316–324. For a practical, one-pot synthesis of 66 by acylation of ferrocene with acryloyl chloride, see: (b) T.D. Turbitt, W.E. Watts, J. Organomet. Chem., 1972, 46, 109–117. P. Dixneuf, R. Dabard, Bull. Chim. Soc. France 1972, 2847–2854. G. Tainturier, K. Chhor y Sok, B. Gautheron, C.R. Acad. Sci., Ser. C 1973, 277, 1269–1270. E.M. Cayuela, L. Xiao, T. Sturm et al. Tetrahedron: Asymmetry 2000, 11, 861–869. (a) G. Kutschera, C. Kratky, W. Weissensteiner, M. Widhalm, J. Organomet. Chem. 1996, 508, 195–208; (b) A. Mernyi, C. Kratky, W. Weissensteiner, M. Widhalm, J. Organomet. Chem. 1996, 508, 209–218. (a) (R,Rp )-72(Ph): F. G´omez-de la Torre, F.A. Jal´on, A. L´opez-Agenjo et al. Organometallics 1998, 17, 4634–4644; (b) (S,Sp )-73(Ph/Ph) and (S,Sp )-73(Ph/Cy): M.C. Carri´on, F.A. Jal´on, A. L´opez-Agenjo et al. J. Organomet. Chem., 2006, 691, 1369–1381. (a) F.A. Jal´on, A. L´opez-Agenjo, B.R. Manzano et al. Dalton Trans. 1999, 4031–4039 (b) M.C. Carri´on, E. Garc´ıa-Vaquero, F.A. Jal´on et al. Organometallics 2006, 25, 4498–4503. T. Sturm, W. Weissensteiner, F. Spindler et al. Organometallics, 2002, 21, 1766–1774. T. Sturm, B. Abad, W. Weissensteiner et al. J. Mol. Catal. A: Chem. 2006, 255, 209–219. T. Sturm, W. Weissensteiner, K. Mereiter et al. J. Organomet. Chem. 2000, 595, 93–101. (a) S. Kn¨uppel, R. Fr¨ohlich, G. Erker, J. Organomet. Chem. 2000, 595, 308–312; (b) S. Kn¨uppel, R. Fr¨ohlich, G. Erker, J. Organomet. Chem. 1999, 586, 218–222. P. Liptau, S. Kn¨uppel, G. Kehr et al. J. Organomet. Chem. 2001, 637–639, 621–630. P. Liptau, L. Tebben, G. Kehr et al. Eur. J. Inorg. Chem. 2003, 3590–3600. L. Tebben, G. Kehr, R. Fr¨ohlich, G. Erker, Synthesis 2004, 1971–1976. P. Liptau, M. Neumann, G. Erker et al. Organometallics 2004, 23, 21–25. L. Tebben, M. Neumann, G. Kehr et al. Dalton Trans. 2006, 1715–1720. Racemic 80 has been previously obtained by intramolecular cyclisation of 3ferrocenylbutanoic acid. (a) J.W. Huffman, R.L. Asbury, J. Org. Chem. 1965, 30, 3941–3943; (b) T.E. Bitterwolf, Inorg. Chim. Acta 1986, 117, 55–64. (a) P. Liptau, T. Seki, G. Kehr et al. Organometallics 2003, 22, 2226–2232; (b) P. Liptau, L. Tebben, G. Kehr et al. Eur. J. Org. Chem. 2005, 1909–1918. P. Liptau, D. Carmona, L.A. Oro et al. Eur. J. Inorg. Chem. 2004, 4586–4590. C. Nilewski, M. Neumann, L. Tebben et al. Synthesis 2006, 2191–2200. I.R. Butler, S. M¨ussig, M. Plath, Inorg. Chem. Commun. 1999, 2, 424–427. Synthesis of phosphinylated cyclopentadienes is described in: R. Broussier, S. Ninoreille, C. Legrand, B. Gautheron, J. Organomet. Chem. 1997, 532, 55–60. (a) R. Broussier, S. Ninoreille, C. Bourdon et al. J. Organomet. Chem. 1998, 561, 85–96; (b) Ref. 109d; (c) J-C. Hierso, V.V. Ivanov, R. Amardeil et al. Chem. Lett. 2004, 33, 1296–1297; (d) V.V. Ivanov, J-C. Hierso, R. Amardeil, P. Meunier Organometallics 2006, 25, 989–995. (a) Pd, Rh: R. Broussier, E. Bentabet, R. Amardeil et al. J. Organomet. Chem. 2001, 637–639, 126–133; (b) Cr, Mo, W, Mn carbonyls: E. Andr´e-Bentabet, R. Broussier, R. Amardeil et al. Dalton Trans. 2002, 2322–2327; (c) The results have been also summarised in the review article covering the coordination chemistry of polydentate phosphines

Synthesis and Catalytic Use of Planar Chiral and Polydentate Ferrocene Donors

153.

154. 155. 156. 157.

277

in general: J-C. Hierso, R. Amardeil, E. Bentabet et al. Coord. Chem. Rev. 2003, 236, 143–206. (a) J-C. Hierso, A. Fihri, R. Amardeil et al. Organometallics 2003, 22, 4490–4499; (b) D.H. Nguyen, M. Urrotigo¨ıty, A. Fihri et al. Appl. Organomet. Chem. 2006, 20, 845–850. J-C. Hierso, A. Fihri, R. Amardeil, P. Meuiner, Org. Lett. 2004, 6, 3473–3476. N.J. Long, J. Martin, A.J.P. White, D.J. Williams, Dalton Trans. 1997, 3083–3085. B. McCulloch, D.L. Ward, J.D. Woollins, C.H. Brubaker, Jr Organometallics 1985, 4, 1425–1432. (a) W: See Ref. 155; (b) Re(I), Pt(II): K. Bushell, C. Gialou, C.H. Goh et al. J. Organomet. Chem. 2001, 637–639, 418–425.

PART II Materials, Molecular Devices and Biomolecules

8 Ferrocene Sensors Simon R. Bayly, Paul D. Beer and George Z. Chen

8.1

Introduction

The well developed and highly adaptable synthetic chemistry of ferrocene, together with its accessible ferrocene/ferrocenium redox couple has lead to its frequent use in electrochemical molecular receptors for cations, anions and neutral species. In the vast majority of examples the ferrocene moiety is incorporated as a signalling/reporter group, whose redox (or spectroscopic) response is perturbed upon proximal binding of the guest species. Alternatively, it can simply serve as a structural component, allowing control over the topology of the guest binding site. Ferrocene receptors, particularly where they combine these roles, show a range of functionality not available to purely organic structures. The generic design for a redox-active receptor utilises a spacer group to covalently link the guest binding group to the ferrocene unit. Because of the presence of the two functional groups (redox centre and receptor moiety), ferrocene molecular receptors can undergo simultaneous or successive electron and guest transfer to and from, respectively, the ferrocene centre and the receptor moiety. When a suitable spacer or linker group is used to allow interaction between the redox centre and the complexed guest, the electron and guest transfer processes can inﬂuence each other (i.e. they are coupled). Such bifunctional molecules are of great fundamental interest and can have many different applications. At the fundamental level for example, some ferrocene alkyl benzoaza crown ethers were recently used as a model for the quantitative investigation of intramolecular electrostatic interactions which play a vital role in determining the structure and functionality of macromolecules such as supramolecules Ferrocenes: Ligands, Materials and Biomolecules  2008 John Wiley & Sons, Ltd

Edited by Petr Stepnicka

282

Ferrocenes: Ligands, Materials and Biomolecules

and biomolecules.1, 2 The very interesting and important result from these investigations is that, in all the tested ferrocene crown ether molecules, the electrostatic interaction between the ferrocene centre and the guest cation bound to the crown ether ring becomes ineffective when the two are separated more than about one nanometre. For more practical applications, there have been many more investigations in the past two decades, aiming at developing new and special molecular devices such as electrochemical molecular probes or sensors with ultra-high selectivity towards cationic, anionic or neutral molecular guests.3 In this chapter, the development of electrochemical molecular sensors based on various ferrocene receptor molecules is described for cations, anions, ion-pairs and neutral guest species. This will begin with the thermodynamic basis of electrochemical molecular recognition, leading to the selective sensing, followed by representative examples on cations, anions, ion-pairs and neutral molecules, with detailed information on the relevant molecular design, synthesis strategies and electrochemical test results. Examples of sensing devices based on ferrocenylated biomolecules and enzyme containing systems are given in Chapter 13. 8.1.1 8.1.1.1

Thermodynamics Coupled Electron and Guest Transfer Processes

The combined chemical processes in which a redox responsive receptor molecule selectively recognises (complexation) and electrochemically senses (signalling) guest species have been termed as electrochemical molecular recognition.4 The process is illustrated in Figure 8.1, using the N -ferrocenylmethyl aza-18-crown-6 receptor and

O

O N Fe

O

K

N Fe +

+ Na+

O

K+

O

O

O

+ Na+

O O

Eo

+

+e

N Fe +

O

O

O Na+

N Fe

+e

O

O

O

O Eo

O Na+

O

O

O

nF(Eo − Eo+) = RT ln (K+ /K)

Figure 8.1 Scheme of one-square showing coupled electron and guest (Na+ ) transfer to and from a ferrocene crown ether molecule

Ferrocene Sensors

283

its complexation with sodium (Na+ ) as an example. Electrochemical molecular recognition can be represented by the scheme of one-square in which guest binding (vertical reactions) at the receptor site (crown ether) induces a potential shift of the electron transfer (horizontal reactions) from the adjacent redox centre (ferrocene), represented by (E o − E o + ). Alternatively, electron transfer at the redox centre also changes the afﬁnity of the neighbouring receptor moieties for the guest species, represented by ln(K/K+ ). These mutual inﬂuences can be linked thermodynamically by the following simple equation: nF (E o − E o + ) = RT ln(K+ /K) (8.1) The left hand side of Equation 8.1, i.e. the potential shift, (E o − E o + ) = E o , is measurable electrochemically, which forms the device basis of an electrochemical molecular probe or sensor. The thermodynamic derivation of Equation 8.1 can be found in the literature.4 For signalling purposes, the potential shift must be sufﬁciently large, at least larger than the experimental errors of the technique employed. This is for example about ±5 mV in many cyclic voltammetric measurements, but can be reduced to less than ±1 mV in potentiometric titration. From the thermodynamic point of view, the value of the potential shift is, according to Equation 8.1, determined by the ratio of K+ /K, instead of the absolute value of either K or K+ . As a consequence a successful ferrocene receptor need not necessarily have a very high binding strength for the guest to be sensed. So long as electron transfer leads to a sufﬁciently large change in the stability of the receptor–guest complex, a measurable change in redox potential can be observed. For voltammetric measurements, such as cyclic voltammetry and square-wave voltammetry, the sensing signal is related to the current waves. The resolution for distinguishing current waves on the voltammogram is about 60 mV at room temperature (∼300 K). It can then be calculated from Equation 8.1 that the value of K should differ from that of K+ by at least one order of magnitude. In other words, Equation 8.1 indicates that the further away the K+ /K ratio is from unity, the greater the potential shift that is observed. For ferrocene receptor molecules, cation complexation will lead to a positive (anodic) shift in potential (E o + > E o ) and make the K+ /K ratio smaller than one, whilst anion binding causes the potential to shift negatively (cathodic shift) with the K+ /K ratio larger than one. Obviously, the K+ /K ratio is a measure of how efﬁcient the mutual inﬂuence or coupling is between the electron and guest transfer reactions, and can be termed as the reaction coupling efﬁciency (RCE).4 The RCE is directly determined by the nature and magnitude of the intramolecular interactions between the guest and the redox centre before and after electron transfer. The interactions are themselves dependent on the properties of redox centre, receptor moiety and/or guest species, and on the structural relationship between them that is dictated by the linker or spacer group. Therefore, the synthetic design for guest sensing purposes requires careful selection and balanced incorporation of the redox centres and the guest receptor moieties into either rigid or ﬂexible structures that can accommodate coupled electron and guest transfer processes. Two structural incorporation strategies have been employed in the synthesis of redox responsive receptor molecules: use of a spacer group to link the redox centres with the guest receptor moieties in close proximity; and arrangement of the redox centres into the guest binding framework. These

284

Ferrocenes: Ligands, Materials and Biomolecules

are discussed below according to the nature of guest species, and the guest–redox centre interaction pathways, using ferrocene based receptor molecules as examples. 8.1.1.2

Effect of Substituents – the Linker Group

Ferrocene is a neutral molecule and is soluble in many common organic solvents. It can be readily and reversibly oxidised to the ferrocenium cation under ambient conditions by either chemical or electrochemical means. The redox potential (strictly speaking, this is the standard electrode potential, E o , but can be represented by the o mean potential, E o = (Epa + Epc )/2, as measured from cyclic voltammograms or the half-wave potential, E1/2 , as measured from the steady state voltammogram) for the ferrocene/ferrocenium couple depends on the solvent used and is E o = 0.31 V vs. saturated calomel electrode (SCE) in acetonitrile.5–7 Due to its electrochemical reversibility, the ferrocene/ferrocenium couple is also often used as a so-called internal reference in conjunction with a pseudo-reference electrode (e.g. a platinum wire immersed in the electrolyte) in many organic systems in which a suitable reference electrode is not available.8 The oxidation potential of the ferrocene/ferrocenium couple changes according to the nature and number of substituents.9 For example, alkyl substituents may each negatively shift the potential by about 50 mV, whereas phenyl and carboxyl will positively increase the potential by 23 mV and 280 mV, respectively.7 In general, the potential shift depends on the substituent’s electron-donating or -accepting power (electrophilicity) that is often described by the Hammett constant.9 The substituent effect is important in designing receptor molecules so that the value of E o is an available parameter for selection of the working potential range of the sensor. However, it should be pointed out that because the substituent effect is present in both free receptor and complex forms of the molecule, it makes no net contribution to the potential shift as predicted by Equation 8.1, unless guest binding leads to further change to the substituent’s electrophilicity. This is in theory an inevitable event when the substituent is changed from the un-complexed to the guest-complexed forms. The practical signiﬁcance of this change of course depends on the magnitude of inﬂuence of the complexed guest on the electrophilicity of the substituent. For ferrocene receptors, the substituent combines the spacer group and the receptor moiety. Because of its closer proximity to the ferrocene centre, the spacer group exerts a greater inﬂuence than the receptor moiety on the redox potential of the free receptor molecule. However, when a guest is complexed to the receptor moiety, the electrophilicity of the spacer group may change depending on the polarising nature (donating or withdrawing) and power of the guest, and also on the polarisability of the spacer group (saturated or conjugated). This is a particularly important and complicated factor to be taken into consideration when designing ferrocene receptors for sensing a speciﬁc guest, and should be dealt with by individual cases. For example, in a very early work, it was demonstrated that when a conjugated spacer group is used to link ferrocene and crown ether, signiﬁcantly larger Group I metal cation induced potential shifts were observed in comparison with their counterparts with a saturated spacer group.10 A very recent detailed voltammetric study of a number of alkyl linked ferrocene crown ethers revealed that even the saturated alkyl spacer, if it consists of

Ferrocene Sensors

285

no more than three methylene sub-units, may also be polarised upon cation complexation to enable through-bond electrostatic communication.2 The consequence is that cation complexation induced potential shift in the ferrocene centre is much larger than would be expected from a point charge at the same distance from the ferrocene centre according Coulomb’s Law. 8.1.1.3

Effect of Complexed Cations

Ferrocene receptors are capable of losing one electron to form the cationic or ferrocenium version of the molecule. This means that if the guest is a cation, a positive (anodic) potential shift is expected, whilst anion complexation should lead to a negative (cathodic) potential shift. In the case of cation sensing, the potential shift increases linearly against the effective charge of the cation, Qcation = ze/r, where z and r are, respectively, the valence and coordination radius of the cation. In the case of alkyl spacer linked ferrocene crown ethers, Equation 8.2 was derived from Coulomb’s Law (middle part) and voltammetric data (right hand part):1 nQFc Qcation Qcation d o (8.2) E = =A 1− 4πε0 e εd dmax d Cancelling the Qcation /d terms from the middle and right hand parts, Equation 8.2 can be re-arranged to Equation 8.3: 1 d 4πε0 e A 1− (8.3) = ε nQFc dmax where n is the number of bound cations of equidistance from the ferrocene centre, QFc the variation in effective charge on the ferrocene moiety upon the transfer of one electron, d the distance between the bound cation and the ferrocene centre, ε0 the vacuum permittivity, ε the relative permittivity of the local medium between the bound cation and the ferrocene centre, e the elementary charge, Qcation the effective charge on the bound cation (see above), and A and dmax are both parameters determined by the environment and framework of the molecule. Particularly, the physical meaning of dmax is the maximum interaction distance between the ferrocene centre and bound cation at which the potential shift, E o , drops to zero. In other words, beyond this length, the intramolecular electrostatic interactions are ineffective. In ferrocene crown ether receptors with alkyl spacers, the maximum effective interaction distance was observed to be around one nanometre, which deserves attention in designing and selection of other ferrocene receptors with saturated spacer groups for electrochemical molecular sensing. The parameter A, as shown by Equation 8.3, is independent of the bound cation, but affects the local dielectric property, ε. Therefore, it must be a function of the environment and framework of the receptor molecule. In other words, the value of A will differ between different types of receptor molecules and also between different electrolytes.1 8.1.1.4

Effect of Complexed Anions

For anion complexation, reports on a quantitatively or qualitatively predictable relationship are rare between the potential shift and the property of the anion. For ferrocene

286

Ferrocenes: Ligands, Materials and Biomolecules

receptors, the fact that they become cationic upon electron transfer (oxidation) means an increase in the stability constant of the anion complex, i.e. K+ /K > 1. However, the situation becomes more complex because of this direct electrostatic attraction between the ferrocenium and the guest anion. Firstly, anions, particularly polyatomic anions, are more polarisable than simple metallic cations and will be more affected by close proximity to the ferrocenium cation. This electron transfer induced polarisation may then lead to an underestimation of the effective charge on an anion. Secondly, most ferrocene based anion receptors have ﬂexible acyclic structures and often employ multiple hydrogen bond donor groups. This means that the electrostatic attractive force is likely to distort the structure of the receptor to pull the anion closer to the ferrocenium. Thirdly, precipitation of the ferrocenium-anion salt onto the electrode surface may also add more complication to the electrochemical measurement. Consequently, the magnitude of potential shift in ferrocene based anion receptors is difﬁcult to predict and in many cases is signiﬁcantly larger than that expected from simple structural parameters. This feature of ferrocene receptors upon complexation with an anion is of course beneﬁcial for sensing purposes, although it is necessary to design the receptor molecules on the basis of individual cases. 8.1.1.5

Effect of Complexed Ion-Pairs

Receptors consisting of both cation and anion receptor moieties have the potential to complex simultaneously a cation and an anion as an ion-pair. The electrostatic attraction between the complex cation and anion will further increase the stability of the ion-pair complex. In such cases, the potential shift is more likely determined by the dipole strength and direction of the ion-pair towards the ferrocene centre. For example, if it is the negatively charged end that is closer to the ferrocene centre, a cathodic potential shift is expected. The overall potential shift of a ferrocene receptor upon complexation with an ionpair is smaller than complexing the individual ions because the opposite nature of the electrostatic interactions between the ferrocenium moiety and the complexed cation and anion. However, if the complexation with the cation and anion occurs successively, each of the forward and backward potential shifts will still be signiﬁcant, and can be exploited for sensing purposes.11 8.1.1.6

Effect of Neutral Guests

It is not appropriate to apply electrostatic interactions for sensing neutral molecules. Therefore, even though a redox active receptor may be designed to have satisfactory selectivity and complexation strength for a neutral guest molecule, electrochemical signalling of guest complexation requires a different strategy from those used for an ionic guest. This can be achieved by combining a receptor moiety with two or more ferrocene groups (or other redox responsive groups) into an appropriate structure in which the binding of the neutral guest interferes with the electronic interaction between the ferrocene groups, causing the redox potentials of individual ferrocene groups to shift. In designing a ferrocene receptor following the strategy above it is important to arrange the ferrocene groups at asymmetrical positions in the molecule, so that electron transfer to the ferrocene groups occurs at different potentials. This will give rise

Ferrocene Sensors

287

to two or more redox processes. The potential difference between these processes is mainly determined by the electrostatic interactions between the ferrocene groups upon oxidation, which is also a function of the ‘local dielectric environment’ and is inﬂuenced ﬁrstly by the solvent molecules and the neutral guest before and after complexation respectively. This type of redox responsive receptor is particularly helpful for sensing nonpolar guests in a polar solvent, or polar guests in a nonpolar solvent. While quantitative prediction of the potential variation against the property of the neutral guest is not yet attainable, it can be qualitatively anticipated that the variation will be larger if the polarity difference between the guest and solvent molecules is signiﬁcant. Two types of documented parameters can be used for such qualitative estimations: dielectric constant (ε) and the dipole moment (p). Another approach, which has been demonstrated for recognition of cations but not yet for neutral molecules or any other types of guests, is speciﬁcally relevant to ferrocenes that are substituted with relatively large or bulky receptor moieties on each of the two cyclopentadienyl rings. Because of the steric repulsion, the two substituents would be arranged at the opposite sides of the ferrocene centre. Therefore, if the two substituents are drawn to the same side of the ferrocene centre upon complexation with the guest molecule, electrochemical responses are expected. For voltammetry, the change may be in the redox potential if, for example, there are interactive groups on the two substituents close to the ferrocene centre,12 or it may be in the amplitude of the current, as a result of the change in shape and volume between the complex (folded) and the free receptor (extended).1 8.1.2

Biomolecules and Other Large Molecules

Although this chapter covers ferrocene based receptor molecules as the larger host for sensing a smaller guest, it should be pointed out that there are also a large number of ferrocene based donor molecules as the guest for sensing much larger host molecules, such as the redox active glucose oxidase which is poly-anionic. Direct electron transfer between an electrode and the enzyme molecule is kinetically difﬁcult because of the deep location of the redox centre inside the large enzyme, which is about a few nanometers in size. In such cases, a mediator is needed to relay electron transfer from the electrode into the redox active centre deep inside the enzyme molecule. This mediated electron transfer process is schematically illustrated in Figure 8.2 using a ferrocene mediator as an example. Ferrocene donor molecules have been widely used as such electron transfer mediators. As schematically shown in Figure 8.2, successful mediation requires the redox potential of the ferrocene/ferrocenium couple of the ferrocene donor molecule, E o Fc , to be more positive than that of the Ez+ /Ez couple of the enzyme host, E o Ez . Because the value of E o Fc can be signiﬁcantly varied by changing the electrophilicity of substitutes, it is possible to use ferrocene donor molecules to probe or sense many enzymes and other biomolecules. The voltammetry of ferrocene mediated electron transfer is characterised by the unique feature of an electro-catalytic process: the oxidation peak current of the enzyme solution increases signiﬁcantly upon the addition of the ferrocene donor compound. If the voltammogram is ﬁrst recorded in the solution of the ferrocene donor compound,

288

Ferrocenes: Ligands, Materials and Biomolecules Ez–Fc+ complex

Reduced enzyme (Ez) Ferrocenium (Fc+)

Electrode

2

e

e

1 Electrolyte

3 Electrolyte

5

4

Ferrocene (Fc) Oxidised enzyme (Ez+) Oxidisable site

Ez+–Fc complex Diffusion into bulk solution

Anionic site

Figure 8.2 Schematic representation of the process of ferrocene mediated electron transfer between an electrode and a polyanionic enzyme

addition of the enzyme may also result in some change in current, but is more likely to negatively shift the potential of the oxidation current peak.13 8.1.3

Interaction Pathways

Clearly, as discussed above, electrochemical molecular recognition of a guest species depends on how the intramolecular interactions are realised between the complexed guest and the redox centre. In other words, it is the nature of the guest–redox centre interaction pathway that determines the effectiveness of guest sensing. Although appearing to be different from each other, all the interaction pathways that have been demonstrated in the literature can be categorised into two groups. The ﬁrst group takes advantage of the electrostatic interactions and the second involves the direct participation of the redox group(s) in the guest complexation process. The two can be further classiﬁed into ﬁve sub-groups, namely: through-space electrostatic interaction; through-bond electrostatic interaction; direct coordination; guest interference; and conformational variation. These are schematically explained in Figure 8.3 and discussed in details below. 8.1.3.1

Intramolecular Electrostatic Interactions

The ﬁrst strategy is to incorporate a particular spacer group between the redox centre and the receptor moiety so that the intramolecular electrostatic interactions are exploited either through-space or through-bond or a combination of both. Obviously,

Ferrocene Sensors Redox Centre–Guest Electrostatic Interactions

Redox Centre Participation in Guest Binding

Through-Space (Medium) signal receptor G

RC

signal source

289

Direct Coordination G

RC (d)

(a) Through-Bond

G

RC

RC

(b) RC

RC1

RC2 G Interference Between Redox Centres (c)

Conformational Change in Redox Centre

G

(e)

Figure 8.3 Schematic illustration of ﬁve different intramolecular interaction pathways for electrochemical molecular recognition of guest species using redox responsive receptor molecules (RC: Redox Centre; G: Guest)

in such receptor molecules, there is no direct interaction between the redox centre and the guest molecule or ion. Therefore, the strength of the electrostatic interaction decides the value of RCE. Examples for each of the two communication pathways are shown: Figure 8.3(a) (through-space) and Figure 8.3(b) (through-bond). Because electrostatic interactions are stronger between charged guest and the redox centre upon electron transfer, they are more often employed for electrochemical recognition or sensing ionic guests, including ion-pairs. 8.1.3.2

Redox Centre Participation in Guest Binding

Coupled electron–guest transfer can be achieved via incorporation of redox centre(s) into the cyclic or acyclic receptor structure. In such receptor molecules, upon complexation of the guest, electron transfer to or from the redox centre(s) is directly inﬂuenced by the presence of the guest. Figure 8.3(c) shows an example in which the redox centre contributes directly to binding the guest, which leads to relatively large voltammetric variation. This type of coupling mechanism is named the direct coordination pathway. Another coupling strategy, as illustrated in Figure 8.3(d), involves multiple and mutual inﬂuencing redox centres in the receptor’s binding structure. Coupling of electron–guest transfer in such receptor molecules takes advantage of the dependence of the intra-redox centre interactions on the polarity or polarisability of the complexed guest. In comparison with the solvent molecule, a more polar or polarisable guest

290

Ferrocenes: Ligands, Materials and Biomolecules

molecule can in effect increase the permittivity of the local medium between the redox centres, and hence decrease the intra-redox centre interaction which is electrostatic in nature. The opposite effect is expected for nonpolar and nonpolarisable guest molecules. This type of coupling mechanism is called the guest-interference pathway, and is particularly suitable for developing molecular sensors for neutral guests. It is also possible to incorporate a conformation sensitive redox centre into the guest binding framework that undergoes conformation changes upon guest binding, (Figure 8.3(e)).14 Such conformational variation pathways were in fact ﬁrst demonstrated using a ferrocene bis crown ether receptor molecule.15 It was observed that when the two crown ether substituents were pulled to the same side of the ferrocene to form a sandwich complex with an appropriate cation, direct interaction between the two sulfur containing spacer groups on the two substituents led to a shift in the redox potential.

8.2

Cation Receptors

Metal cations are essential in biology. The functioning of the nervous system depends on the control of sodium (Na+ ) and potassium (K+ ) concentrations, while transition metal cations are active in the catalytic sites of many enzymes. Furthermore, the selective sensing of metal ions in aqueous systems is important from an industrial and environmental viewpoint. These diverse applications have lead to the development of many molecular sensor systems for cationic guest species, the majority of which use ferrocene as a redox reporter group. The ﬁrst redox active cation receptors to be studied were the ferrocene crown ether conjugates of Saji 1 and Beer 2–4 (Scheme 8.1).16–20 These are sensors for Group I cations, where the electron-rich crown ether provides the guest binding site – binding

Scheme 8.1

Ferrocene Sensors

291

Table 8.1 Cation induced anodic shifts in the potential of the ferrocene/ferrocenium redox couple of 1–3 and 5 (Electrochemistry carried out in acetonitrile solution containing 0.2 M [Bu4 N][BF4 ] as supporting electrolyte) Receptor +

E(Na )/mV E(K+ )/mV E(Li+ )/mV

1a

2

3

5

– 180 –

40 20 70

35 20 75

<10 <10 390

a Carried out in dichloromethane (CH Cl ) containing 0.5 M 2 2 [Bu4 N][PF6 ].

of sodium, potassium and caesium (Cs+ ) to 2–4 was observed by 13 C NMR (in CD3 OD/acetone solution) and in the solid state X-ray crystal structures. The rotational freedom of ferrocene about the molecular axis also allows the formation of both 2:1 cation:receptor complexes and 1:1 intramolecular ‘sandwich’ complexes depending on the receptor and cation used. Crucially, binding of a cation at the crown ether site in receptors 1–3 causes a signiﬁcant anodic shift in the redox potential of the ferrocene/ferrocenium redox couple (Table 8.1). No cation induced shift in redox potential was observed with 4 and this is attributed to the distance and lack of conjugation between the ferrocene and ionophore. As predicted by simple electrostatics, oxidation of receptor 1 was found to decrease its binding afﬁnity for alkali metal cations due to repulsion between 1+ and the cation. This is the origin of the cation induced shift in redox potential. The reduced stability of the 1+ : cation complex means that K+ /K < 1 and thus E o + > E o (Equation 8.1). The ferrocene unit therefore not only enables redox sensing of alkali metal cations, but also allows the host–guest interaction to be switched on and off depending on its oxidation state. Saji has demonstrated that this phenomenon can be used in electrochemical ion transport across liquid membranes.16 A crown ether binding site is not a prerequisite for the sensing of Group I cations. Lithium (Li+ ) has a high afﬁnity for carbonyl oxygen donors and actually gives a greater anodic shift in redox potential in the presence of lithium with the simple ferrocene amide 5 than with 2–4.21 Further studies on ferrocene crown ethers have demonstrated that electronic conjugation can propagate a redox response to cation binding even when the binding unit is located at a distance from the ferrocene reporter group. This is seen in receptor 6 (Scheme 8.2) where an aza-crown is connected to ferrocene via a trans-stilbene like linkage.10 A maximum anodic shift in the ferrocene/ferrocenium redox couple of 110 mV was observed for this receptor in the presence of Mg2+ in acetonitrile solution. In the reference system, 7, which has a saturated linker and therefore no pathway for electronic communication between the binding site and the ferrocene, no measurable cation induced shift in redox potential was observed. Molecular models suggest that the distance between the ferrocene and the bound cation would be about one nanometre, beyond the predicted maximum interaction distance (dmax , Equation 8.2) for such systems. The related receptor 8, in which the aryl aza-crown is directly linked to the ferrocene, also shows a 110 mV maximum anodic shift in the presence of Mg2+ .22 The reduced distance between the cation binding site and the ferrocene unit would

292

Ferrocenes: Ligands, Materials and Biomolecules

Scheme 8.2

be expected to lead to an increased anodic shift, however in this case it is likely that electronic conjugation between the two is somehow reduced. As seen in Equation 8.2, the magnitude of the change in redox potential is dependent on the effective charge of the cation bound. Using this idea Beer and coworkers have devised a system that can simultaneously detect and differentiate between two different Group II cations. In the presence of Mg2+ and Ba2+ in acetonitrile solution, the ditopic aza-crown receptor 9 produces three new peaks in the voltammogram. These correspond to the 9:2Ba2+ , 9:2Mg2+ and 9:Ba2+ Mg2+ complexes with anodic shifts (compared to 9 alone) of 150, 395 and 275 mV respectively.23, 24 The smaller magnesium ion has a greater charge density than the barium ion, and hence produces a larger shift in redox potential. As well as metal cations, ferrocene aza-crown systems have also been shown to sense the ammonium (NH4 + ) cation.25 On the basis of effective charge, a small Group I cation such as potassium would be expected to induce a greater shift in redox potential than the diffuse ammonium cation. This is not observed in the bis- and trisferrocenyl macrocycles 10–13 (Scheme 8.3). Receptor 10 gives an anodic shift of 60 mV in response to two equivalents of either of these cations, whereas 11–13 are selective for NH4 + over K+ (e.g. 11 gives shifts of 210 mV for NH4 + and 40 mV for K+ ). It is clear that in these examples the value of K+ /K (and hence the cation induced shift) is not primarily dictated by the electrostatic properties of the cation. Hydrogen bonding interactions must increase the stability of the NH4 + :ferrocene complex relative to the NH4 + :ferrocenium complex. This is particularly evident in 12 and 13, which respond strongly to NH4 + binding despite the ferrocene being separated from the aza-crown by a propan-1,3-diyl spacer. It is thought that the ﬂexible hydrocarbyl chain enables close contact of the cation and metallocene redox centre through a hydrogen bonding interaction between the carbonyl oxygen of the ferrocene amide and the coordinated NH4 + . Another example where conformational changes on cation binding lead to a surprising redox response is the thioether linked receptor 14 (Scheme 8.4).15 Complexation of Na+ in acetonitrile solution results in a 70 mV anodic shift in the ferrocene/ferrocenium redox couple, whereas K+ generates a cathodic shift of 60 mV. The origin of this unexpected cathodic shift is related to the structure of the complex with the larger potassium cation. 13 C NMR and FAB MS experiments show that whereas 14 binds

Ferrocene Sensors

Scheme 8.3

Scheme 8.4

293

294

Ferrocenes: Ligands, Materials and Biomolecules

Na+ with a 1:2 host:guest stoichiometry, it forms a 1:1 sandwich complex with K+ . In this conformation it is thought that the lone pairs of electrons on the sulfur atoms are directed towards the iron atom of the ferrocene. The extra electron density at the metal centre makes the ferrocene easier to oxidise than in the free receptor and produces the observed cathodic shift in redox potential. Ferrocene cryptands have also been studied as cation receptors. Hall et al. have investigated the coordination of Group II and lanthanide metal cations by cryptand 15 using cyclic voltammetry in acetonitrile solution.26 As expected the dication induced anodic shift of the ferrocene/ferrocenium redox couple was found to broadly correlate with the charge density of the cationic guest. Interestingly the analogous receptor 16, in which the amide linkages have been reduced to alkyl amines reported by Kaifer, Gokel and coworkers, gives a larger magnitude of voltammetric response to cations than 15.27 The maximum anodic shifts due to Ca2+ in acetonitrile electrolyte solution were 155 mV and 275 mV for 15 and 16 respectively. Ferrocene 16 also gives a measurable response to Na+ and K+ whereas 15 does not. Clearly the more ﬂexible alkyl linker allows greater coupling between the ferrocene and the cryptand binding site. Sensor molecule 16 shows a remarkable selectivity for silver (Ag+ ). The stability constant of the 16:Ag+ complex in acetonitrile electrolyte solution was found to be an order of magnitude higher than even the 16:Ca2+ complex (log K = 8.35 vs. 7.16), in which the complexed cation has a much higher charge density. Silver (Ag+ ) and calcium (Ca2+ ) also produce similar anodic shifts in the ferrocene/ferrocenium redox couple of 16 (282 mV vs. 274 mV) despite their comparative charge. Spectroscopic (1 H NMR, UV/Vis) and solid state X-ray crystal structure evidence indicate that the Ag+ ion is bound not only by the nitrogen donors of the macrocycle, but also interacts ˚ compared strongly with the iron atom of the ferrocene. The Fe–Ag+ distance is 3.37 A ˚ in the relevant complexes (the ionic radius of Ag+ is to an Fe–Na+ distance of 4.39 A ˚ compared to 0.95 A ˚ for Na+ ). In this example ferrocene acts as a cation binding 1.3 A group as well as a redox signalling unit. In examples 17–19 (Scheme 8.5) the ferrocene unit is directly attached to the cryptand via oxygen donor atoms.28 In acetonitrile solution in the presence of Group I

Scheme 8.5

Ferrocene Sensors

295

monocations and Group II dications large anodic shifts in redox potential were measured, in the range of 80 mV for 19:Li+ to 380 mV for 17:Ca2+ . It was observed that the magnitude of the redox response due to dications was larger than that due to monocations, reﬂecting their effective charge (Qcation ). Larger cryptands gave smaller anodic shifts as a result of the reduced proportion of the cationic charge experienced by the O-donors attached to the ferrocene. For each cryptand the metal ion with the best size match for the binding cavity gave the greatest response, since in this situation the cation is held closest to the ferrocene (d/dmax is smaller). The excellent transition metal cation coordination properties of nitrogen donors have lead to their extensive use in ferrocene sensors for cations. Examples range from the simple amines and imines 20–22 (Scheme 8.6), to polyamine chains and macrocycles such as 23–27 (Scheme 8.7). The former allow the selective electrochemical sensing of Mg2+ and Ca2+ in acetonitrile solution,29 whilst the latter provide pH sensitive detection of transition metal cations in aqueous media.30–33 Sensor molecule 27 has been used for the quantitative sensing of Cu2+ , Zn2+ and Cd2+ in THF/water (70/30) solution by Martinez-Manez and coworkers.23

Scheme 8.6

Scheme 8.7

296

Ferrocenes: Ligands, Materials and Biomolecules

Perhaps the largest known anodic shifts in redox potential due to cation complexation have been observed in 28, where ferrocene has been directly conjugated to two di(2pyridylmethyl) groups with an amine linkage.34 Plenio and coworkers showed that in acetonitrile solution this receptor produces anodic shifts of 380 and 760 mV for 28:Co2+ and 28:2Co2+ respectively. Complexation of Zn2+ generates shifts of 330 and 720 mV for 28:Zn2+ and 28:2Zn2+ respectively. Molina and coworkers have recently developed an unusual series of azaferrocenophanes 29–37 (Scheme 8.8) which incorporate both a quinoline nitrogen and an alcohol oxygen for cation coordination.35, 36 Receptors 32 and 33 include an additional ferrocene unit, whereas 35–37 possess thienyl, 2-pyridyl and 3-pyridyl receptor groups respectively. Coordination of Mg2+ , Ca2+ Ni2+ and Zn2+ in acetonitrile/dichloromethane (3/2) solution produced anodic shifts of 300–400 mV in the redox couple of the ferrocenophane. The effect of cations on the redox couple of the distal ferrocenes of 32 and 33 was far less marked, with shifts of 30 mV or less with magnesium and calcium due to the distance from and lack of conjugation between the ferrocene and the receptor site. A colourimetric response to cations was demonstrated using receptor 37. Upon chelation of Mg2+ in dichloromethane solution the ferrocene MLCT band at λmax = 462 nm was replaced with a new band at λmax = 525 nm, with increased absorbance. The 37:Mg2+ complex was found by spectrophotometric titration to possess a stability constant of 1.5 × 1010 M−1 in dichloromethane solution. This is a rare example of the use of ferrocene as a chromophoric reporter group in cation sensing.

Scheme 8.8

Fluorescence spectroscopy is a powerful technique in molecular sensing due to its sensitivity. The relatively low oxidation potential of ferrocene that makes it a useful redox reporter group means that it generally quenches the emission from ﬂuorophores to which it is attached. A number of research groups have described systems which circumvent this problem. Delavaux-Nicot and coworkers have developed 38 and 39 (Scheme 8.9), ferrocene based ﬂuorescent sensors for the detection of Ca2+ .37, 38 These molecule do not function as reversible electrochemical sensors since they undergo irreversible oxidation before the ferrocene/ferrocenium redox couple is reached. In both cases 1 H and 13 C NMR experiments in CD3 CN indicate that Ca2+ interacts strongly with the carbonyl groups, leaving the amines uncoordinated. A spectrophotometric

Ferrocene Sensors

297

Scheme 8.9

response to Ca2+ coordination in acetonitrile solution was also observed in the UV/Vis. More striking is the change in the ﬂuorescence properties of the molecules. As Ca2+ was added to an acetonitrile solution of 38 the intensity of the emission at 560 nm increased until two equivalents of cation were reached. Further addition led to a steady decrease in emission intensity until at relatively high calcium concentration total quenching occurred. Complex ‘multistep behaviour’ was observed with derivative 39 under similar conditions. Molina and coworkers have studied the ferrocene–anthracene dyad 40, a ﬂuorescent sensor for Li+ which operates in aqueous conditions.39 The molecule is only weakly ﬂuorescent (excitation at 370 nm, φ = 0.042–0.008), but in CH3 CN/H2 O (70/30) mixtures at pH = 5 the addition of one equivalent of lithium ions results in a 7.3 fold enhancement in ﬂuorescence. At this pH the receptor is selective for Li+ over Na+ , K+ , Ca2+ , Cu2+ , and Zn2+ . Using ﬂuorescence titration the stability constant of the 40:Li+ complex in CH3 CN/H2 O (70/30) solution was found to be 11.76 (±0.02) M−1 . A colourimetric response to Li+ was also observed by UV/Vis spectroscopy, with the MLCT absorption band moving to higher energy and decreasing in intensity. Neither of these effects is observed in anhydrous conditions.1 H NMR and DFT analysis indicate that the Li+ cation interacts directly with the ferrocene iron atom and is stabilised by a network of hydrogen bonds involving two water molecules and the protonated imine nitrogen of the receptor. In this case the ferrocene moiety itself is acting as the cation binding group, albeit in concert with hydrogen bonding. Molina has also developed the fully conjugated diferrocenyl macrocyclic receptor 41.40 This functions as a redox switchable sensor which can selectively complex and decomplex Mg2+ . No perturbation of the two distinct ferrocene/ferrocenium couples was observed on addition of Ca2+ , Li+ or Na+ to a dichloromethane solution of 41. However, addition of Mg2+ caused an anodic shift in the redox couple of the ferrocene at the allylic end of the molecule, while the redox couple of the other ferrocene remained largely unperturbed. The UV/Vis spectrum of 41 was also found to be sensitive to Mg2+ , and titration results indicated that the 41:2 Mg2+ complex is formed via the 40:Mg2+ complex with stability constants of K11 = 9.8 × 105 M −1 and K12 = 6.3 × 105 M −1 . When a solution of 41: Mg2+ was oxidised electrochemically

298

Ferrocenes: Ligands, Materials and Biomolecules

it gave a UV/Vis absorption spectrum identical to 41+ (the oxidised form of 41), indicating that decomplexation of Mg2+ occurs on oxidation of the receptor. Reduction of the solution containing Mg2+ regenerated the 41:2 Mg2+ . This demonstrates again the origin of the cation induced shift in redox potential.

8.3

Anion Receptors

Anionic species play a fundamental role in numerous biological and chemical processes. Many biologically important species are anionic, such as nucleic acids and the majority of enzyme substrates. Anions also play a role in diseases such as cystic ﬁbrosis, which is caused by the misregulation of Cl− ion channels. In the environment pollutant anions such as NO3 − and PO4 3− can cause the eutrophication of waterways. In view of this the design of receptors that can selectively bind and sense anions is an important goal and has become the focus of a great deal of current research interest. The binding of anions presents a greater challenge than the binding of cations. The reduced effective nuclear charge of anions means that they are larger than isoelectronic cations and have smaller charge to radius ratios. Because of their diffuse nature, anions are more difﬁcult to complex. Anions are also more likely to be polyatomic (adding to their diffuse nature) and exist in a wide variety of geometries. Solvation effects and pH dependence are also important factors to take into account in the coordination of anions. The ﬁrst redox active anion sensors were reported by Beer et al. in 1989.41 They were based on cobaltocenium, because this reporter group possesses a positive charge for electrostatic interaction with anions as well as an accessible redox couple. Cobaltocenium systems have since been studied extensively as anion sensors and it has been found that augmentation with hydrogen-bond donor groups, such as amides, generates both stronger and more selective anion binding and enhances the electrochemical sensing of anions.42 The chemical and structural similarity between ferrocene and cobaltocenium has meant that much of the anion receptor chemistry developed with cobaltocenium has subsequently been applied to ferrocene. The most pertinent difference is that the ferrocene is neutral, and has no inherent electrostatic attraction for anions. Hence, in their unoxidised form ferrocene receptors are comparatively weaker anion receptors. This is not necessarily a disadvantage for redox sensing, since the magnitude of E o relates to the relative stabilities of the neutral and oxidised receptor:anion complexes, and the anion binding afﬁnity of ferroecen receptors is increased by oxidation of ferrocene to the positively charged ferrocenium ion. Molecules 42–46 (Scheme 8.10) are a selection of ferrocene-based receptors that include secondary amide groups for the hydrogen-bonding of anions.43, 44 Electrochemical measurements on 42–45 in acetonitrile showed cathodic shifts in the ferrocene/ferrocenium oxidation wave of up to 240 mV induced by H2 PO4 − . Competition experiments demonstrated the same shift even in the presence a ten-fold excess of Cl− or HSO4 − . Receptor 46 displays the opposite selectivity, with HSO4 − generating a cathodic shift of 220 mV which is not affected by the presence of excess H2 PO4 − . In this case the HSO4 − anion donates a proton to the basic amine function of the receptor and it is the resulting cationic complex which binds the residual SO4 2− anion.

Ferrocene Sensors

299

Scheme 8.10

A very simple cationic receptor 47 has been investigated by Moutet and co-workers (Scheme 8.11). It is able to sense H2 PO4 − and ATP2− in a range of solvents, displaying a cathodic shift of 470 mV in CH2 Cl2 with H2 PO4 − . It appears that in this case the high afﬁnity of this receptor for anions (due to a strong ion-pairing interaction) does lead to a larger value of E o . Other anion coordinating groups to which ferrocene has been attached include imidazolium, calix[4]pyrrole, pyridinium substituted tripods, and cyclotriveratrylene; e.g. 48, 49, 50, and 51 respectively.45–49 Anion induced cathodic shifts in redox potential were observed in each case. According to 1 H NMR studies in CD3 CN molecules 52–54 (Scheme 8.12), in which a ferrocene-1,1 -bisamide is bridged across the upper rim of a calix[4]arene, preferentially bind carboxylate anions (acetate and benzoate) over H2 PO4 − and Cl− .50 This is thought to be the result of topological complementarity between the amide hydrogen bond donor cavity and the bidentate carboxylate anions. In these examples the ferrocene group effectively controls the selectivity of the anion recognition unit to which it is attached. Receptor 55 incorporates a zinc(II) porphyrin backbone with four ferrocene amides.51 The Lewis acid metal atom at the centre of the porphyrin provides an additional binding site for anion recognition. With the free-base precursor of 55 in CD2 Cl2 solution no signiﬁcant anion induced shifts were seen in the 1 H NMR signals of the amide protons,

300

Ferrocenes: Ligands, Materials and Biomolecules

Scheme 8.11

Scheme 8.12

Ferrocene Sensors

301

whereas the metalloporphyrin binds Br− (K = 6200 M−1 ), NO3 − (K = 2300 M−1 ) and HSO4 − (K = 2100 M−1 ). Electrochemical studies in 3:2 CH2 Cl2 /CH3 CN revealed anion induced cathodic shifts in both the porphyrin (E = 85–115 mV) and tetraferrocene oxidation (E = 20–60 mV) waves. Atropisomers of 55 (the α,α,α,αatropisomer is shown) were also studied and displayed different selectivities but generally lower stability constants. This demonstrates the importance of the ferrocene amide groups in the anion recognition site and suggests that they operate cooperatively in the α,α,α,α-atropisomer. Ferrocene anion sensors are also capable of operating in an aqueous environment, an important consideration in the sensing of biologically relevant anions. Anion recognition in polar solvent media requires the use of strong electrostatic interactions. Beer, MartinezManez and coworkers have synthesised a series of ferrocene-based receptors appended with various open chain and cyclic amine functional groups (Scheme 8.13), e.g. 56 and 57 bind ATP2− and H2 PO4 − in water,52–54 giving cathodic shifts of 60–80 mV in the ferrocene/ferrocenium couple. The selectivity of this class of receptors is pH dependent, since the protonation state of the various amine groups varies with pH.

Scheme 8.13

Metallacyclic receptors 26 and 58 were shown to be able to quantitatively determine PO4 3− and SO4 2− in the presence of other anions. The electrochemistry of these receptors was studied in 70:30 THF/H2 O over a range of pH values. A maximum selective redox shift of 54 mV for SO4 2− over PO4 3− was observed at pH 4 for 26, whereas 58 gave a maximum selective redox shift of 35 mV for PO4 3− over SO4 2− at pH 8. Anions such as carboxylates can also be sensed by ferrocene based receptors. Neutral molecule 59 (Scheme 8.14) uses hydrogen bonding to bind mono and dicarboxylate anions with 2:1 and 1:1 guest: host stoichiometry.55 However it is not selective, giving a maximum cathodic shift of 150 mV for both acetate and phthalate. Another recent example shows that selectivity for dicarboxylate anions over monocarboxylates and other simple anions can be achieved. The tetra-ammonium macrocycle 60 binds phthalate, isopthalate and dipicolinate with a 2:1 guest:host stoichiometry, giving maximum cathodic shifts in the redox potential of 275, 193 and 168 mV respectively.56

302

Ferrocenes: Ligands, Materials and Biomolecules

Scheme 8.14

In comparison, the monoacid 4-nitrobenzoate produced a maximum cathodic shift of only 49 mV. Recognition of ﬂuoride in aqueous media is particularly challenging due to the strongly hydrated nature of the anion. The ferrocene–boronic acid, 61, of Shinkai and coworkers (Scheme 8.15) acts a selective redox sensor for aqueous F− .57 The stability constant of 700 M−1 for the ﬂuoride–ferrocenium complex (determined electrochemically) is due to the favourable interaction between boron and F− , a hard acid and hard base respectively. Stability constants for both the Br− and Cl− complexes are lower than 2 M−1 .

Scheme 8.15

Ferrocene Sensors

303

Aldridge et al. have demonstrated that a similar boryl–ferrocene 62 can be used as a selective colourimetric sensor for F− .58 When F− was added to a dichloromethane solution of 62 under aerobic conditions a colour change from orange to pale green was observed. This did not occur with any other anion tested. Spectroscopic and electrochemical measurements suggest that complexation of F− causes the spontaneous formation of a ferrocenium species, i.e. the 150 mV anodic shift in the ferrocene/ferrocenium redox potential caused by F− complexation reduces the redox potential enough for the 62:2F− complex to be oxidised by atmospheric oxygen. Selective ﬂuorescence sensing of F− has also been achieved using ferrocene receptor 63.59 This example uses amide groups for anion binding and naphthalene groups to provide the ﬂuorescence signal. Addition of F− to a DMSO solution of 63 lead to a threefold enhancement (at 5 equivalents) of the intramolecular naphthalene–naphthalene excimer emission at 492 nm. Of the other anions tested only H2 PO4 − generated a signiﬁcant response, causing a two-fold enhancement at 5 equivalents. In electrochemical studies in DMF electrolyte solution, F− generated a 120 mV cathodic shift in the redox potential. The receptor 64, based on an azaferrocenophane structure bearing two urea groups as linkers between the redox active (ferrocene) and ﬂuorescent (naphthalene) signalling subunits, also shows both ﬂuorescent and electrochemical sensing of F− .60 On addition of excess F− it displays an enhancement factor of 13 in the naphthalene emission bands at 362 and 380 nm in DMF solution and a cathodic shift of the ferrocene/ferrocenium couple of 190 mV in DMSO electrolyte solution. The unusual new [3,3]ferrocenophane 65·H+ also acts as a selective ﬂuorescent anion sensor – in this case for NO3 − .61 The protonated receptor is weakly ﬂuorescent ( = 0.043) in dichloromethane solution and on addition of NO3 − the naphthalene based emission at 354 nm is quenched (to = 0.020). Addition of AcO− , HSO4 − and H2 PO4 − merely induced deprotonation of the receptor. This receptor is also able to act as a redox sensor for other anions and as a ﬂuorescent sensor for Group II cations. Combining ferrocene with another organometallic unit can lead to augmented anion sensing. For instance, the incorporation of phosphine groups into ferrocene-amide receptor 66 (Scheme 8.16) allowed the coordination of various transition metal carbonyls to generate mixed-metal complexes 67 and 68.62 These were found by 1 H NMR titration in CD2 Cl2 solution to bind Cl− approximately an order of magnitude more strongly that the parent phosphine. The complexes of 67 and 68 with Cl− , Br− , I− and H2 PO4 − were also found to be more stable, but to a lesser degree. In these examples the addition of a more Lewis acidic metal centre enhances anion binding. Signiﬁcant anion induced cathodic shifts in both the ferrocene/ferrocenium couple and the irreversible oxidation wave of the second metal were observed in the cyclic voltammetry. An unusual bimetallic anion sensor has been developed by Jurkschat and coworkers.63 In macrocycle 69 two ferrocene reporter units are linked together by two Lewis acid organotin spacers. Electrochemical measurements in dichloromethane solution show anion induced cathodic shifts in the ferrocene/ferrocenium redox couple of 130 mV for Cl− , 210 mV for F− and 480 mV for H2 PO4 − .

304

Ferrocenes: Ligands, Materials and Biomolecules

Scheme 8.16

8.4

Receptors for Neutral Guest Species

Neutral guest species are perhaps the most challenging target in synthetic receptor chemistry, since their lack of electrostatic charge necessitates the use of weaker noncovalent interactions to bind them. The forces involved in neutral molecular recognition and complexation are principally hydrogen bonding, charge transfer interactions, and hydrophobic or solvophobic effects. Ferrocene containing receptors for neutral diamine guest species such as DABCO were ﬁrst reported in 1991 by Gokel and coworkers, although these were not demonstrated as redox sensors since guest complexation was only observed by 1 H NMR.64, 65 Tucker and coworkers have developed ferrocenyl receptors which use hydrogen bonding interactions to give an electrochemical response to neutral guest species.66, 67 The amidopyridine residues in 70 and 71 (Scheme 8.17) act as complementary two-point hydrogen bonding sites for carboxylic acids, resulting in 1:1 and 1:2 receptor:guest binding in CDCl3 solution for the monotopic receptor 70 and the ditopic receptor 71 respectively. In dichloromethane solution alkyl monocarboxylic acids cause a small cathodic shift in the ferrocene/ferrocenium redox potential, 20–25 mV for 70 and 55–60 mV for 71. The ditopic ferrocene can also act as a 1:1 receptor for diacid guests. Complexation of 71 with pentane-1,5-dioic acid complex generates a cathodic shift of 85 mV. The fact that the observed shift in redox potential is cathodic indicates that the host–guest complex becomes more stable on oxidation, i.e. the guest stabilises the oxidised form of the receptor, or the receptor has a greater afﬁnity for the guest in its oxidised form than when it is neutral. This is consistent with hydrogen bonding increasing the electron density on the receptor, or with the positive charge on the receptor increasing its capacity as a hydrogen bond donor.66 Ferrocene 71 also binds the urea and barbiturate derivatives 72–74. It was observed that the guest induced redox shift does not correlate directly to the stability constant of

Ferrocene Sensors

305

Scheme 8.17

the host:guest complex with the netral receptor – ethyleneurea (72) gives the largest redox response, yet the 71:72 complex possesses the lowest stability constant. Again this demonstrates that it is the relative stabilities of the ferrocene:guest and ferrocenium:guest complexes that dictate the magnitude of the redox shift. The 1,3-ditopic receptor 75, like 70, is selective for the trimethyleneurea (73), but overall binding strength is increased. The additional hydrogen-bonding groups in 76 and 77 modify their selectivity such that these receptors are selective for barbital (74). The 1,3-ditopic receptors however, where they show well deﬁned electrochemistry, only show small (≤20 mV) cathodic shifts in redox potential with these guests. It is thought that the 1,1 -disubstited ferrocenes give a greater electrochemical response due to the guest species being bound close to the ferrocene centre by the two amidopyridine units, whereas in the 1,3-systems the guest is held further away.67 Inouye and coworkers have made signiﬁcant progress developing organometallic receptors for sensing nucleobases.68–72 In molecules 78–87 (Scheme 8.18) ferrocene is connected via an ether or ethynyl linker to 2,6-diamidopyridine, which provides a complementary hydrogen-bonding site for the nucleobase 1-butylthymine. The other ring of the ferrocene is functionalised with various planar aromatic groups to provide additional π-stacking interactions with nucleobase guests. In this instance the function of the ferrocene is to hold the hydrogen-bonding and π-stacking sites at the correct orientation and distance for guest binding.68 The receptors bind 1-butylthymine in CDCl3 solution with 1:1 stoichiometry and Ka values, calculated from 1 H NMR studies, between 960 and 2230 M−1 (Table 8.1). The series 78–82 shows increasing afﬁnity for 1-butylthymine as the size of the aromatic π-stacking site increases. Derivatives 83–87 have a rigid spacer between the ferrocene and the 2,6-diamidopyridine unit and examples 83–85 exhibit signiﬁcantly lower stability constants with 1-butylthymine than 86 and 87. In the latter examples the pyrene and perylene groups are able to

306

Ferrocenes: Ligands, Materials and Biomolecules

Scheme 8.18

project into space beyond 2,6-diamidopyridine unit (in the eclipsed conformation) presenting a greater surface area for π-stacking with the guest. The related ditopic receptors 88 and 89 (Scheme 8.19) incorporate two ferrocenyl hydrogen bond donor units which are attached to a pyrene ring at the 2 and 7 positions. As predicted these bind 1-butylthymine in a 1:2 host:guest ratio with overall stability constants in CDCl3 of 6.7 × 106 M−2 for 88 and 2.5 × 107 for 89.69 1 H NMR studies show evidence of additional π-stacking interactions compared to the monotopic receptors. The addition of two hydrogen-bonding groups to ferrocene in a 1,1 fashion generates receptors for dinucleosides, for instance 90 binds thymidyl(3 -5 )thymidine in CDCl3 /DMSO-d6 (85/15).72 Here ferrocene is used a structural linker because the inter-ring spacing of the two cyclopentadienyl rings (0.33 nm) is near to the spacing between the stacked base pairs. Incorporation of a spiropryidopyran unit between the ferrocene and hydrogen-bonding group provides a chromophore for colourimetric sensing of dinucleoside derivatives. In the presence of guanine 91 (Scheme 8.20) undergoes isomerisation to the merocyanine form 91 and shows a dramatic increase in the absorption band at 575 nm. This colour change is observed in the presence of guanine–guanine derivative 92, but it was not observed with analagous thymine–thymine and adenine–adenine guest species.71 Selective detection of amino acids is difﬁcult to achieve due to their multiple hydrogen bonding sites and zwitterionic character. Ferrocene crown ether 93 (Scheme 8.21) is able to extract several l-amino acids into CH2 Cl2 from acidic aqueous solution. The host:guest stiochiometry was found to be 1:1. The amino acid is presumed to hydrogen

Ferrocene Sensors

307

Scheme 8.19

Scheme 8.20

bond ditopically via its −NH3 + and CO2 H groups to the crown ether and carboxylic acid sites of the receptor respectively.73 Circular dichroism (CD) spectroscopy was used to sense binding of the chiral guest species. Addition of l-amino acids by 93 resulted in the appearance of a ‘W-shaped’ CD band at the absorption maximum

308

Ferrocenes: Ligands, Materials and Biomolecules

Scheme 8.21

of the ferrocene. Complexes of d-amino acids gave the symmetrically opposite ‘Mshaped’ spectra. Another ferrocene-based receptor for amino acids has recently been described by Roy and coworkers.74 Strong 1:1 binding of 94 with amino acids in MeCN/H2 O (55/45) mixtures at pH = 7.2 was observed by UV/Vis, ﬂuorescence, electrochemistry, isothermal calorimetry (ITC) and 13 C NMR. Glutamic acid (Glu)and aspartic acid (Asp) which exist in their anionic form at this pH, were found to have the greatest afﬁnity for the receptor (Ka (94:Glu) = 98 × 103 M−1 and Ka (94:Asp) = 98 × 103 M−1 , determined by UV/Vis titration). Zwitterionic amino acids glutamine (Gln) and glycine (Gly), gave slightly lower stability constants [Ka (94:Gln) = 68 × 103 M−1 and Ka (94:Gly) = 39 × 103 M−1 ]. The highly conjugated receptor is ﬂuorescent and on addition of Glu the emission maximum at 554 nm increases in intensity and shifts to 550 nm. NMR studies suggest that both the ester carbonyl groups and terminal hydroxyls are involved in amino acid recognition. Carbohydrates are another important biological target for ferrocene sensor molecules. Receptor 95 is able to sense saccharides such as d-glucose, d-fructose, d-galactose and d-mannose electrochemically.75 In this case the ferrocene redox reporter group is teamed with a diamine diboronic acid binding unit whose saccharide binding properties have previously been studied. Saccharides were found to generate anodic shifts in the ferrocene/ferrocenium redox couple. This suggests that oxidation of the ferrocene removes electron density from the amine nitrogen atoms, reducing their ability to accept hydrogen bonds and therefore destabilising the ferrocenium:saccharide complex relative to the ferrocenium:saccharide complex. Alternatively, the saccharide may

Ferrocene Sensors

309

increase the degree of interaction between neighbouring amine and boron atoms, which in turn reduces the electron density on the amines thereby destabilising the oxidised form of the ferrocene.

8.5

Ion-Pair Receptors

Interest in ditopic receptors capable of binding ion-pairs has increased markedly in recent years. One reason for this is that recognition of charged species by charge-neutral receptors is difﬁcult to achieve if there are ions of the opposite charge competing with the receptor in solution. Indeed, coordination of the competing counter-ion to the receptor in addition to removing the ion from solution can lead to a considerable enhancement of the afﬁnity of the receptor for the ion of opposite charge. Furthermore charged species are not easily transported across biological membranes. This problem can be circumvented by a receptor that binds the cation or anion of interest as one of the partners in a charge-neutral ion pair. Most examples of ion-pair receptors comprise separate binding domains for the anion and cation,76–78 but the contact ion-pair binding approach, wherein the anion and cation are bound together as one moiety, is gaining interest as this avoids the energetically unfavourable separation of the two ions.79–81 Tucker and coworkers have recently reported a ferrocene ion-pair receptor which acts as a chromogenic molecular switch.82 The ferrocene of molecule 96 (Scheme 8.22)

Scheme 8.22

310

Ferrocenes: Ligands, Materials and Biomolecules

is substituted in a 1,1 fashion by a phenyl urea unit for anion binding and a crown ether for cation binding. On addition of F− to a solution of 96 in acetonitrile, a signiﬁcant perturbation of the UV/Vis spectrum was observed, including the appearance of a new absorption at 472 nm. The stability constant for the 96:F− complex was determined as 9340 M−1 . The addition of 10 equivalents of KPF6 to the solution of 96 containing 10 equivalents of F− caused the complete disappearance of the 475 nm absorption. The stability constant of the receptor with K+ in the presence of F− was found to be 1460 M−1 . The model sensor 97, lacking the cation binding site, exhibited similar switching properties and the Ka of the 97:F− complex was also found to be alike (9660 M−1 ). As predicted the Ka for K+ in the presence of F− was far lower (230 M−1 ). The reason that K+ binding causes the loss of the characteristic electronic absorption due to 96:F− (since F− is not displaced) is not understood, but it is thought that the ion-pairing interaction between F− and K+ plays a signiﬁcant role. Tuntunlani and coworkers have described a ferrocene cryptand based on the ionpair receptors of Smith.83, 84 In the absence of metal cations sensor 98 binds Cl− and Br− weakly in 5 % CD3 CN/CDCl3 solution, giving stability constants of <40 M−1 . In the presence of alkali metal ions two different phenomena were observed. Formation of the uncomplexed metal halide ion-pair occurred when Cl− or Br− were added to a solution containing 98:K+ and when Cl− was added to 98:Na+ . In these systems the halide anion effectively removes the cation from the macrocycle:cation complex. However, in the presence of Na+ receptor 98 was found to bind Br− with a stability constant of 16 100 M−1 . Thus in 5 % CD3 CN/CDCl3 98 is a selective ion-pair receptor for NaBr. Electrochemical experiments in 40:60 CH3 CN/CHCl3 electrolyte solution were used to demonstrate ion-pair sensing. In this solvent mixture the presence of Na+ leads to an increase in the observed cathodic shift in redox potential due to Cl− or Br− . Cl− induces a cathodic shift of 107 mV with uncomplexed 98, which is increased to 122 mV for 98:Na+ . Br− induces a cathodic shift of 28 mV with uncomplexed 98, which is increased to 46 mV for 98:Na+ . For this increase in cathodic shift to occur, the increased stability of the ion-pair complex must be so great that it offsets the reduced negative charge experienced by the ferrocene unit due to the proximity of the cation. Zwitterions can also be regarded as ion-pairs. Tucker and coworkers have recently investigated a ferrocene sensor for phenylalanine, an amino acid that is zwitterionic under certain conditions.85 Ditopic molecule 99 carries a pyridyl amide for recognition of the carboxylate portion of phenylalanine zwitterions and a crown ether for binding the ammonium function. In binding studies phenylalanine was added as the ammonium salt, but on contact with 99 proton transfer to the pyridyl ring of the receptor occurs, and it is effectively the zwitterions that are bound. In acetonitrile electrolyte solution the carboxylate dominates the redox sensing behaviour, generating a maximum overall cathodic shift in the redox potential of 129 mV. The monotopic reference compound 70 which bears only the anion recognition site shows a cathodic shift of 107 mV in same conditions, demonstrating that binding of the ammonium function to the crown ether enhances the redox response.

Ferrocene Sensors

8.6

311

Nanoscale Sensors

A recent signiﬁcant advance in the development of organometallic receptors has been their incorporation into nanoscale structures. Pre-organised groups of closely spaced receptors have the potential to give rise to advantageous effects such as inter-receptor cooperativity and guest desolvation. These methods of harnessing receptors in the solid state represents a potential route for the fabrication of robust sensing devices. Self-assembled monolayers and thin polymer ﬁlms of ferrocene receptors on electrode surfaces exhibit ampliﬁed anion binding. Similarly ferrocene functionalised dendrimers and gold nanoparticles have been found to exhibit enhanced binding of anions. 8.6.1

Dendrimers

Astruc and coworkers are pioneers in the ﬁeld of ferrocene dendrimers. As with simple ferrocene species these are able to electrochemically sense anions via a cathodic shift in the ferrocene/ferrocenium couple.86 Evidence of a dendritic effect was observed in 100–102 (Scheme 8.23), consecutive dendrimer generations which comprise up to 18 amido–ferrocene units. As the number of ferrocene units increases, so does the magnitude of the cathodic shift in the ferrocene/ferrocenium couple, the redox response due to H2 PO4 − or HSO4 − . The stability constants calculated for the HSO4 − complexes of 101 and 102 in CH2 Cl2 solution are reported to be 9390 and 216 900 M−1 respectively (assuming 1:1 binding between ferrocene groups and anions). The same group has recently investigated ﬁve generations of pentamethyl–amidoferrocene dendrimers using the DSM polyamine core (not shown).87 The pentamethyl substituted ferrocene was chosen to overcome the irreversible electrochemistry and electrode adsorption occurring with 100–102. In this series the magnitude of the dendritic effect varied according to the anion studied. Progression from lower to higher dendrimer generations leads to modest increases in the anion induced cathodic shift of the ferrocene/ferrocenium couple with H2 PO4 − in DMF solution, whereas with ATP2− anion binding progressed from weak to relatively strong. This is likely to be the result of a change in anion binding stoichiometry with the two different anions; ATP2− forms a 1:2 anion:ferrocene complex whereas H2 PO4 − forms a 1:1 complex. 8.6.2

Thin Polymer Films and Self-Assembled Monolayers

Beer and coworkers have investigated anion sensing using self-assembled monolayers (SAMs) of ferrocene amide disulﬁdes. SAMs were formed of the 1,1 -bis(alkyl-N amido)ferrocene 103 (Scheme 8.24) on gold electrodes.88 The pendant disulﬁde groups serve to covalently anchor the receptor to the gold surface. In diffusive electrochemical experiments on 103 in CH3 CN/CH2 Cl2 solution, anion induced cathodic shifts of the ferrocene/ferrocenium redox couple were observed, with values of 40 mV for Cl− , 20 mV for Br− and 210 mV for H2 PO4 − . When conﬁned to a monolayer the anion induced shifts measured were 100 mV for Cl− , 30 mV for Br− and 300 mV for H2 PO4 − in the same solvent system–consistently greater than for the solution phase receptor.

312

Ferrocenes: Ligands, Materials and Biomolecules

Scheme 8.23

This represents a signiﬁcant ‘surface sensing ampliﬁcation’. The modiﬁed electrodes were also able to selectively detect H2 PO4 − in the presence of a 100-fold excess of halide. In aqueous solution the selectivity of the system was altered, enabling the detection of the poorly hydrated anion ReO4 − in the presence of H2 PO4 − . A very recent development are SAMs comprising ferrocene appended oligonucleotide sequences. Radi and O’Sullivan used a guanine rich aptamer oligonucleotide with a thiol terminus to form SAMs on gold electrodes. The guanine rich sequence is known to convert from a loose randomly coil structure to a compact quadruplex in

Ferrocene Sensors

313

Scheme 8.24

the presence of K+ . Square wave voltammetry and electrochemical impedence spectroscopy both showed current signals for the Fe2+ to Fe3+ oxidation which increased with K+ concentration. It is thought that the change in structure from coil to quadruplex brings the ferrocene tag into close proximity with the gold surface, thus signiﬁcantly reducing the barrier to charge transfer. Fan et al. have applied a similar system to the femtomolar sequence speciﬁc detection of DNA. Here the ferrocene tagged DNA sequence is designed to form a hairpin loop structure in the original SAM, such that the ferrocene is held close to the gold electrode surface. Hybridisation of the strand with a complementary DNA strand causes the ferrocene tag to move away from the gold surface, with a concomitant reduction in the current observed for the Fe2+ to Fe3+ oxidation by cyclic voltammetry. The current was found to respond logarithmically to the concentration of target DNA over at least six orders of magnitude. Work on ferrocene containing thin-polymer ﬁlms has primarily been focussed on anion sensing (presumably due to the positively charged nature of pyrrole electropolymers, the most commonly used type), although cation sensing by this method has also been demonstrated. Monomer 104 incorporates a ferrocene amide reporter with a polymerisable pyrrole unit. Thin ﬁlms of the receptor were prepared by electropolymerisation on a platinum or carbon electrode.89 Anion induced shifts of the ferrocene/ferrocenium couple were measured in CH3 CN giving values of 30, 180, 220 mV for HSO4 − , ATP2− and H2 PO4 − respectively. Polymerisation of 104 gives rise to a surface sensing ampliﬁcation The anion sensing properties of ferrocenyl viologen 105 in thin polymer ﬁlms have also recently been explored.90 Poly-105 registered small anion induced cathodic shifts of the ferrocene/ferrocenium couple of up to 35 mV for ATP2− in aqueous solution. Cation sensing can also be accomplished using thin polymer ﬁlms.91 Moutet and coworkers have electropolymerised 106 onto platinum electrodes. The resulting ﬁlms show a 120 mV cathodic shift in redox potential in the presence of Ba2+ ions in MeCN solution.

314

8.6.3

Ferrocenes: Ligands, Materials and Biomolecules

Functionalised Nanoparticles

Astruc and coworkers have prepared the amidoferrocenylalkylthiol (AFAT)–gold nanoparticle system represented as 107 (Scheme 8.25). The proportion of AFAT to dodecanethiol obtained by ligand substitution of dodecanethiol stabilised nanoparticles ranged from 7–38 %, corresponding to an average of 8–39 AFAT units per nanoparticle. In CH2 Cl2 solution the ferrocene functionalised nanoparticles show a single reversible redox wave for the Fe2+ /Fe3+ couple at identical potential irrespective of surface loading. Addition of H2 PO4 − generated a cathodic shift in the redox potential of 220 mV. The magnitude of the shift was found to be the same irrespective of surface loading and is considerably larger than observed the comparable amidoferrocene monomer FcCONHCH2 CH2 OPh (45 mV) or even a representative ferrocene tripod PhC(CH2 CH2 CH2 NHCOFc)3 (110 mV).

Scheme 8.25

The same group has also investigated the anion sensing properties of gold nanoparticles 108 and 109 substituted with dendrons comprising three amidoferrocene or silyl ferrocene branches (Scheme 8.26).92 The surface loadings of 108 and 109 were 3 % and 4.8 % respectively, corresponding to an average three and ﬁve dendrons per nanoparticle. Nanoparticles of type 108 show very similar properties to the AFAT

Ferrocene Sensors

315

Scheme 8.26

modiﬁed nanoparticles 107, with a H2 PO4 − induced cathodic shift of 210 mV in dichloromethane solution. Despite lacking any hydrogen bonding groups the nanoparticles dendronised with silyl ferrocenes (109) gave a H2 PO4 − induced cathodic shift of 110 mV in the same solvent.

8.7

Conclusion

Ferrocene based receptors continue to be the mainstay of electrochemical molecular sensing chemistry. The useful physico-chemical properties of ferrocene have lead to its use as a multifunctional component in sensors for all types of guest species. Cation coordination is a well-established subject and the ferrocene redox-active centre features in the majority of redox-responsive examples in this ﬁeld. Anion recognition has been less studied, but accelerating progress in this area means that ferrocene anion sensors

316

Ferrocenes: Ligands, Materials and Biomolecules

are becoming equally advanced. Harnessing the weak noncovalent interactions required for recognition of neutral guest species is an even more difﬁcult challenge and ion-pair sensing is also relatively unexplored. Nevertheless, ferrocene has been at the inception and has remained the focus of much of the work on these two types of sensor, which are destined to generate a great deal of interest in the future. A signiﬁcant recent advance has been the incorporation of ferrocene receptors into nanoscale structures such as dendrimers, nanoparticles, thin polymer ﬁlms and self-assembled monolayers. The origin and development of the ‘surface sensing ampliﬁcation’ observed in these pre-organised systems is an observation that needs to be exploited further. In the light of these exciting recent developments ferrocene is certain to play a leading role in the continuing emergence of electrochemical molecular sensing nanotechnology.

References 1. S. Jin, D.H. Wang, X.B. Jin, G.Z. Chen, Chem. Phys. Chem., 2004, 5, 1623–1629. 2. S. Jin, X.B. Jin, D.H. Wang et al. J. Phys. Chem. B, 2005, 109, 10658–10667. 3. P.D. Beer, S.R. Bayly, Anion sensing by metal-based receptors in Anion Sensing, I. Stibor (Ed.), Top. Curr. Chem., Springer, Berlin (2005), vol. 255, pp 125–162. 4. P.D. Beer, P.A. Gale, G.Z. Chen, Dalton Trans., 1999, 1897–1910. 5. Z.H. Zhong, T. Matsumura-Inoue, A. Ichimura, Anal. Sci., 1992, 8, 877–9. 6. I. Noviandri, K.N. Brown, D.S. Fleming et al. J. Phys. Chem. B, 1999, 103, 6713–6722. 7. N.J. Long, Metallocenes – An Introduction to Sandwich Complexes. Blackwell Science: London, UK (1998). 8. N.G. Connelly, W.E. Geiger, Chem. Rev., 1996, 96, 877–910. 9. M. Emilia, N. Silva, A.J.L. Pombeiro et al. J. Organomet. Chem., 1994, 480, 81–90. 10. P.D. Beer, C. Blackburn, J.F. McAleer, H. Sikanyika, Inorg. Chem., 1990, 29, 378–81. 11. P.D. Beer, Z. Chen, M.I. Ogden, Faraday Trans., 1995, 91, 295–302. 12. P.D. Beer, C.G. Crane, J.P. Danks et al. J. Organomet. Chem., 1995, 490, 143–8. 13. N.J. Forrow, G.S. Sanghera, S.J. Walters, Dalton Trans., 2002, 3187–3194. 14. P.D. Beer, Z. Chen, A.J. Goulden et al. Chem. Commun., 1994, 1269–1271. 15. P.D. Beer, J.P. Danks, D. Hesek, J.F. McAleer,, Chem. Commun., 1993, 1735–7. 16. T. Saji, I. Kinoshita, Chem. Commun., 1986, 716–17. 17. T. Saji, Chemistry Letters, 1986, 275–6. 18. P.D. Beer, Chem. Commun., 1985, 1115–16. 19. P.D. Beer, C.G. Crane, A.D. Keefe, A.R. Whyman, J. Organomet. Chem., 1986, 314, C9–C11. 20. P.D. Beer, H. Sikanyika, A.M.Z. Slawin, D.J. Williams, Polyhedron, 1989, 8, 879–86. 21. P.D. Beer, H. Sikanyika, C. Blackburn et al. J. Organomet. Chem., 1988, 356, C19–C22. 22. P.D. Beer, A.D. Keefe, H. Sikanyika et al. Dalton Trans., 1990, 3289–94. 23. Z. Chen, A.J. Pilgrim, P.D. Beer, Faraday Trans., 1995, 91, 4331–3. 24. Z. Chen, A.J. Pilgrim, P.D. Beer, J. Electroanal. Chem., 1998, 444, 209–217. 25. P.D. Beer, D.B. Crowe, M.I. Ogden et al. Dalton Trans., 1993, 2107–16. 26. C.D. Hall, N.W. Sharpe, I.P. Danks, Y.P. Sang, Chem. Commun., 1989, 419–21. 27. J.C. Medina, T.T. Goodnow, M.T. Rojas et al. J. Am. Chem. Soc., 1992, 114, 10583–10595. 28. H. Plenio, C. Aberle, Organometallics, 1997, 16, 5950–5957. 29. A. Chesney, M.R. Bryce, A.S. Batsanov et al. Chem. Commun., 1998, 677–678. 30. P.D. Beer, D.K. Smith, Dalton Trans., 1998, 417–423.

Ferrocene Sensors

317

31. M.J.L. Tendero, A. Benito, R. Martinez-Manez et al. Dalton Trans., 1996, 343–51. 32. A. Benito, R. Martinez-Manez, J. Soto, M.J.L. Tendero, Faraday Trans., 1997, 93, 2175–2180. 33. M.E. Padilla-Tosta, R. Martinez-Manez, T. Pardo et al. Chem. Commun., 1997, 887–888. 34. H. Plenio, D. Burth, Organometallics, 1996, 15, 4054–4062. 35. J.L. Lopez, A. Tarraga, A. Espinosa et al. Chem. Eur. J., 2004, 10, 1815–1826. 36. A. Tarraga, P. Molina, J.L. Lopez, M.D. Velasco, Dalton Trans., 2004, 1159–1165. 37. J. Maynadie, B. Delavaux-Nicot, S. Fery-Forgues et al. Inorg. Chem., 2002, 41, 5002–5004. 38. B. Delavaux-Nicot, J. Maynadie, D. Lavabre, S. Fery-Forgues, Inorg. Chem., 2006, 45, 5691–5702. 39. A. Caballero, R. Tormos, A. Espinosa et al. Org. Lett., 2004, 6, 4599–4602. 40. A. Caballero, V. Lloveras, A. Tarraga et al. Angew. Chem., Int. Ed., 2005, 44, 1977–1981. 41. P.D. Beer, A.D. Keefe, J. Organomet. Chem., 1989, 375, C40–C42. 42. P.D. Beer, Chem. Commun., 1996, 689–96. 43. P.D. Beer, A.R. Graydon, A.O.M. Johnson, D.K. Smith, Inorg. Chem., 1997, 36, 2112–2118. 44. P.D. Beer, Z. Chen, A.J. Goulden et al. Chem. Commun., 1993, 1834–6. 45. J-L. Thomas, J. Howarth, K. Hanlon, D. McGuirk, Tetrahedron Lett., 2000, 41, 413–416. 46. P.A. Gale, M.B. Hursthouse, M.E. Light et al. Tetrahedron Lett., 2001, 42, 6759–6762. 47. L.O. Abouderbala, W.J. Belcher, M.G. Boutelle et al. Chem. Commun., 2002, 358–359. 48. O. Reynes, F. Maillard, J-C. Moutet et al. J. Organomet. Chem., 2001, 637–639, 356–363. 49. J.W. Steed, Chem. Commun., 2006, 2637–2649. 50. B. Tomapatanaget, T. Tuntulani, O. Chailapakul, Org. Lett., 2003, 5, 1539–1542. 51. P.D. Beer, M.G.B. Drew, R. Jagessar, Dalton Trans., 1997, 881–886. 52. P.D. Beer, Z. Chen, M.G.B. Drew et al. Chem. Commun., 1993, 1046–8. 53. P.D. Beer, Z. Chen, M.G.B. Drew et al. Inorg. Chim. Acta, 1996, 246, 143–150. 54. P.D. Beer, J. Cadman, J.M. Lloris et al. Dalton Trans., 1999, 127–134. 55. D.S. Kim, H. Miyaji, B.Y. Chang et al. Chem. Commun., 2006, 3314–3316. 56. X.L. Cui, R. Delgado, H.M. Carapuca et al. Dalton Trans., 2005, 3297–3306. 57. C. Dusemund, K.R.A.S. Sandanayake, S. Shinkai, Chem. Commun., 1995, 333–4. 58. S. Aldridge, C. Bresner, I.A. Fallis et al. Chem. Commun., 2002, 740–741. 59. B.G. Zhang, J. Xu, Y.G. Zhao et al. Dalton Trans., 2006, 1271–1276. 60. F. Oton, A. Tarraga, A. Espinosa et al. J. Org. Chem., 2006, 71, 4590–4598. 61. F. Oton, A. Tarraga, P. Molina, Org. Lett., 2006, 8, 2107–2110. 62. J.E. Kingston, L. Ashford, P.D. Beer, M.G.B. Drew, Dalton Trans., 1999, 251–258. 63. R. Altmann, O. Gausset, D. Horn et al. Organometallics, 2000, 19, 430–443. 64. J.C. Medina, C. Li, S.G. Bott et al. J. Am. Chem. Soc., 1991, 113, 366–7. 65. C. Li, J.C. Medina, G.E.M. Maguire et al. J. Am. Chem. Soc., 1997, 119, 1609–1618. 66. J.D. Carr, L. Lambert, D.E. Hibbs et al. Chem. Commun., 1997, 1649–1650. 67. J. Westwood, S.J. Coles, S.R. Collinson et al. Organometallics, 2004, 23, 946–951. 68. M. Inouye, Y. Hyodo, H. Nakazumi, J. Org. Chem., 1999, 64, 2704–2710. 69. M. Inouye, M-a.S. Itoh, H. Nakazumi, J. Org. Chem., 1999, 64, 9393–9398. 70. M. Inouye, M. Takase, Angew. Chem., Int. Ed., 2001, 40, 1746–1748. 71. M. Takase, M. Inouye, Chem. Commun., 2001, 2432–2433. 72. M. Takase, M. Inouye, J. Org. Chem., 2003, 68, 1134–1137. 73. H. Tsukube, H. Fukui, S. Shinoda, Tetrahedron Lett., 2001, 42, 7583–7585. 74. P. Debroy, M. Banerjee, M. Prasad et al. Organic Letters, 2005, 7, 403–406. 75. S. Arimori, S. Ushiroda, L.M. Peter et al. Chem. Commun., 2002, 2368–2369. 76. N. Pelizzi, A. Casnati, A. Friggeri, R. Ungaro, Perkin Transactions 2 , 1998, 1307–1311. 77. D.J. White, N. Laing, H. Miller et al. Chem. Commun., 1999, 2077–2078. 78. D.M. Rudkevich, J.D. Mercerchalmers, W. Verboom et al. J. Am. Chem. Soc., 1995, 117, 6124–6125.

318

Ferrocenes: Ligands, Materials and Biomolecules

79. M. Cametti, M. Nissinen, A.D. Cort et al. J. Am. Chem. Soc., 2005, 127, 3831–3837. 80. J.M. Mahoney, A.M. Beatty, B.D. Smith, Inorg. Chem., 2004, 43, 7617–7621. 81. B.D. Smith, Ion-Pair Recognition by Ditopic Macrocyclic Receptors in Macrocyclic Chemistry: Current Trends and Future Perpectives, K. Gloe (Ed.), Springer: Dordrecht, The Netherlands (2005), pp 137–151. 82. H. Miyaji, S.R. Collinson, I. Prokes, J.H.R. Tucker, Chem. Commun., 2003, 64–65. 83. C. Suksai, P. Leeladee, D. Jainuknan et al. Tetrahedron Lett., 2005, 46, 2765–2769. 84. M.J. Deetz, M. Shang, B.D. Smith, J. Am. Chem. Soc., 2000, 122, 6201–6207. 85. H. Miyaji, G. Gasser, S.J. Green et al. Chem. Commun., 2005, 5355–5357. 86. C. Valerio, J-L. Fillaut, J. Ruiz et al. J. Am. Chem. Soc., 1997, 119, 2588–2589. 87. M-C. Daniel, J. Ruiz, J-C. Blais et al. Chem. Eur. J., 2003, 9, 4371–4379. 88. P.D. Beer, J.J. Davis, D.A. Drillsma-Milgrom, F. Szemes, Chem. Commun., 2002, 1716–1717. 89. I. del Peso, B. Alonso, F. Lobete et al. Inorg. Chem. Commun., 2002, 5, 288–291. 90. O. Reynes, C. Bucher, J-C. Moutet et al. Chem. Commun., 2004, 428–9. 91. A. Ion, I. Ion, J.C. Moutet et al. Sens. Actuators, B:, 1999, B59, 118–122. 92. A. Labande, D. Astruc, Chem. Commun., 2000, 1007–1008.

9 Ferrocene-Based Electro-Optical Materials J¨urgen Heck and Markus Dede

9.1

Introduction

In recent years materials capable of achieving the modulation of light have received a lot of attention due to their potential applications in electro-optical devices.1–7 To develop electro-optical materials with well deﬁned properties it is most desirable to vary as many parameters as possible. In this respect substances based on organometallic compounds provide several general advantages. The tuning of the properties can be achieved by variation of metal atoms in complexes, their oxidation state, the ligand environment and the complex geometry. Furthermore, coordination of metal atoms can stabilise reactive organic molecules.8 Ferrocene in particular is seen as a favourable component for electro-optical materials. Its rich chemistry9 offers the variation of substituents for attaching to the electro-optical systems as well as a high diversity of both structural and electronic properties.10 Ferrocene features a reversible one electron oxidation11 and in conjugation with the stability of the corresponding ferrocenium salts, ferrocene/ferrocenium is a very stable redox pair.11, 12 The easy and reversible release of an electron establishes ferrocene moiety as a strong donor8, 11 that stabilises adjacent electron deﬁcient systems.12 The purpose of this chapter is to provide a reasonably comprehensive survey of the research on electro-optical materials containing ferrocene units between 1995 and 2006, with primary emphasis on the relationship between structure and properties. Patent literature and results based solely on calculations are excluded. The main focus is set upon experimental results of nonlinear optical materials complemented with Ferrocenes: Ligands, Materials and Biomolecules  2008 John Wiley & Sons, Ltd

Edited by Petr Stepnicka

320

Ferrocenes: Ligands, Materials and Biomolecules

selected references concerning self-assembled monolayers (SAMs), thin ﬁlms, as well as artiﬁcial photosynthesis, data storage and nanotubes. Materials based on or resulting from ferrocene polymers are dealt with in Chapter 10. Nonlinear optical (NLO) effects are caused by the interaction of very strong electromagnetic ﬁelds (e.g. laser light) with matter, i.e. polarisable electron density that inﬂuences the nature of the incident light.8 The result is an induced polarisation response (P ) that can be expressed – at the molecular level – as a function of the applied electric ﬁeld (E) according to the following power series expansion (Equation 9.1): P D αE C βE 2 C γ E 3 C Ð Ð Ð

(9.1)

Only under a very strong electric ﬁeld do the βE 2 and γ E 3 terms become signiﬁcant and give rise to the quadratic and cubic NLO effects, respectively. The coefﬁcients β and γ are called ﬁrst and second molecular hyperpolarisabilities, respectively. When macroscopic material is concerned, Equation 9.1 converts into: P D χ (1) E C χ (2) E 2 C χ (3) E 3 C Ð Ð Ð

(9.2)

The coefﬁcients χ (2) and χ (3) are the bulk NLO susceptibilities. All of the mentioned NLO coefﬁcients can have either positive or negative signs, but usually only their absolute values are quoted.10 Quadratic NLO effects arising from β and χ (2) include second harmonic generation (SHG), the electro-optic (EO, Pockels) effect and frequency mixing, while cubic NLO effects based on γ and χ (3) include third harmonic generation (THG), optical phase conjugation and the optical Kerr effect (OKE).10 So far the best understood NLO phenomena are frequency doubling (SHG) and tripling (THG).10 General requirements for large β values (SHG) at the molecular level are: a dipolar structure (usually achieved by linking a donor and an acceptor by a conjugated π-bridge, i.e. D-π-A systems);13 a large difference in dipole moments between the ground and excited state;8 the existence of low-energy excited states, which are coupled with charge-transfer transitions;10, 13 large oscillator strengths;8 and a non-centrosymmetric arrangement of the NLO chromophore (NLOphore), especially as a bulk material.10 Dipolar organometallic complexes possessing MLCT or LMCT transitions meet these criteria.8 Nevertheless, octupolar molecules are known which possess zero ground state dipole moments but also demonstrate large β values.10 In contrast, substances with large γ values (THG) rely foremost on extensive πconjugated systems.10 There are no conditions on symmetry to be met for large γ values, not even for bulk material. For the reasons mentioned above, ferrocene is one of the preferred choices as the donor component in dipolar organometallic D-π-A systems. There are several methods for determining SHG properties. Quadratic NLO properties of the solid state (χ (2) ) can be measured by Kurtz-powder tests.14, 15 This method is semiquantitative only; its results are affected by particle size and no SHG activity is found for centrosymmetric crystals. Electric-ﬁeld-induced SHG (EFISHG)15, 16–18 and hyper-Rayleigh scattering (HRS)15, 19–21 are suitable methods for the measurement of the molecular unit β in solution. HRS offers some advantages as knowledge of molecular dipole moments (µ) is not necessary and the method is applicable to charged and octupolar compounds. On the other hand, it does not discriminate between HRS and

Ferrocene-Based Electro-Optical Materials

321

ﬂuorescence due to multi-photon absorption.22 This disadvantage can be circumvented by femtosecond HRS experiments.22, 23 Dilute solutions of non ionic but dipolar compounds are also used for electro-optical absorption measurements (EAOM), which allows the static ﬁrst hyperpolarisability β0 to be calculated:15, 24 β D β0

2 2 Emax /f[1 (2Ef )2 (Emax )1 ][(Emax )2 Ef2 ]g

β0 D 3µ12 (µ12 )2 /2(Emax )2

(9.3) (9.4)

25, 26

The theoretical two-state model (Equations 9.3 and 9.4) can be used to correct the resonance enhancement of results from EFISHG or HRS experiments. Ef is the fundamental laser energy, Emax is the energy of maximal absorption, µ12 is the dipole moment change for ICT excitation, and µ12 is the transition dipole moment. The two-state model is reasonably valid for dipolar molecules in which β is primarily associated with a single intervalence charge transfer (ICT) excitation.10 Generally, dipolar ferrocene derivatives show two strong absorption bands in UV/Vis spectra which both contribute to the SHG effect. Now it is generally accepted that the low energy CT transition is attributable to a donor metal (DM )-acceptor ligand CT transition whereas the stronger, high energy CT transition is caused by a donor ligand (DL )-acceptor ligand transfer.27, 28 Cubic NLO properties are primarily determined by degenerate four-wave mixing (DFWM) experiments29–32 or by the z-scan technique.33–35 EFISHG can be used as well.10

9.2

SHG Active Compounds

In the following text the SHG active ferrocene derivatives have been sorted into different sub-classes primarily according to the acceptor groups present in the respective molecules. Even though there are different conventions in use,36, 37 in many papers it is not stated clearly which one is actually employed. Nevertheless, in this chapter all β values are given as absolute values. Since there are two sets of units commonly used for β values in the literature, esu and C m3 V2 respectively, the published data have been converted into the respective other set of units according to Wortmann et al. (β:1 ð 1050 C m3 V2 D 2.694 ð 1030 esu)37 for a better comparison. The converted numbers are written in italics. 9.2.1

Tris(pyrazolyl)borato Complexes

For D-π-A compounds with an acceptor group composed of a transition metal complex including a tris(pyrazolyl)borato (Tp) ligand (Figure 9.1) a large number of β values has been determined (Tables 9.1 and 9.2).38–42 The listed β values range between 36 ð 1030 esu and 975 ð 1030 esu and allow for deriving pertinent relationships between structure and properties. McCleverty et al. probed the inﬂuence of the halogeno ligand within the organometallic acceptor group.38 It was deduced from the results for the amino or anilino bridged complexes 1–3 and 5–7 that changing from a chloro to a bromo ligand did not affect the nonlinear optical response of the complex to a major extent whereas the introduction of a iodo ligand caused an increase of the β value (Table 9.1). This did not hold for the pyridinium bridged complexes 14 and 16 which exhibited similar β values (42 ð 1030 esu

322

Ferrocenes: Ligands, Materials and Biomolecules R1 Ferrocene

Compound

Bridge

Ferrocene donor

R2

1 N N R2 ONR M N N BH X N N R1 R2

Bridge

Tris(pyrazolyl)borato Moiety M

X

R1

R2

Mo

Cl

CH3

CH3

Mo

Br

CH3

CH3

Mo

I

CH3

CH3

438

W

Cl

CH3

CH3

538

Mo

Cl

CH3

CH3

Mo

Br

CH3

CH3

Mo

I

CH3

CH3

838

W

Cl

CH3

CH3

938,39

Mo

Cl

CH3

CH3

Mo

I

CH3

CH3

W

Cl

CH3

CH3

Mo

Cl

CH3

CH3

Mo

Cl

CH3

CH3

Mo

Cl

CH3

CH3

Mo

Cl

CH3

CH3

N

Mo

I

CH3

CH3

+

Mo

Cl

CH3

CH3

Mo

Cl

CH3

CH3

138 238

H N

Fc–

338

638

H N

Fc– 738

1039

H N

Fc–

1139 N

H N

1238

Fc–

1338

H3C CH3 H3C H3C Fe CH3 H3C CH3 H3 C

N

1438

Fc–

N

1538

H3C H3C

Fe

N

+

+

+

N

CH3 CH3

+

1638 Fc–

1738

1838

N

N

H3 C H3C

CH3

H3C H3C

CH3 CH3

H3C Fe

N

+

N

Figure 9.1 Compounds containing tris(pyrazolyl)borate (Tp) complexes as acceptor group

Ferrocene-Based Electro-Optical Materials R1 Ferrocene

Compound

Ferrocene donor

323

R2

1 N N R2 ONR M N N BH X N N R1 R2

Bridge

Bridge

Tris(pyrazolyl)borato Moiety

1938

M

X

R1

R2

Mo

Cl

CH3

CH3

Mo

Cl

pC6H4– OMe

H

+

N +

Fc–

N

2038 O H3C 2140,42

N

H3C H N

N 2241

O

2341,42

H N N

2441,42

N

Fc–

H N H N

2541,42 H3C

N

2641,42

N

H N

H3C 2741,42

N

H3C N

H N

Figure 9.1 (continued )

and 36 ð 1030 esu, respectively). The inﬂuence of the metal atom of the acceptor fragment was also tested.38 Although the metal inﬂuenced the nonlinear response, there was no obvious dependency. For the amino bridged compounds 1 and 4 the change from molybdenum to tungsten caused an increase in the β value from 119 ð 1030 esu to 240 ð 1030 esu. Replacing the amino bridge by an anilino function reversed the result. The molybdenum complex 5 exhibited a markedly higher β value than the analogous tungsten complex 8 (297 ð 1030 esu and 56 ð 1030 esu, respectively). The inﬂuence of different degrees of methylation of the ferrocenyl unit was derived from the compounds 13, 14 and 15.38 With an increasing number of methyl groups the nonlinear optical response increased, rising from β D 42 ð 1030 esu for the not methylated 14 to β D 205 ð 1030 esu for the eight-fold methylated 13. Oxidation of the ferrocenyl donor (13 ! 13C ) resulted in diminishing the nonlinear activity.38

324

Ferrocenes: Ligands, Materials and Biomolecules

Table 9.1 Linear optical and quadratic nonlinear optical response parameters of compounds containing tris(pyrazolyl)borate (Tp) complexes as acceptor group (part I) Compound

λmax /nm (ε/103 M1 cm1 )

β/1030 esu

β/1050 C m3 V2

β0 /1030 esu

β0 /1050 C m3 V2

119 š 12e

44.2 š 4.5

57 š 6f

21 š 2

50.1 š 5.2

67 š

7f

25 š 3

73.3 š 14

101 š

10f

138

737 (3.9)a,b,c,d

238

753

(4.0)a,b,c,d

135 š

14e

764

(3.9)a,b,c,d

197 š

39e

338 38

4

589 (3.4)

a,b,c,d

538

596

638

609 (7.3)a,b,c,d

738

(8.0)a,b,c,d

627

(8.3)a,b,c,d

89.1 š 8.9

38 š 4

14 š 1

297 š

30e

110 š 11

52 š

5f

19 š 2

289 š 29e

107 š 11

60 š 6f

22 š 2

431 š

160 š 16

43e e

21 š 3

975 š

98e

498 519 (–)b,h 516 (–)i,h

35 š

7j,m

1039

514 (–)g,h 536 (–)b,h 536 (–)i,h

85 š 10j,n

1139

415 (–)g,h 430 (–)b,h 431 (–)i,h

8 š 2j,o

1238

580 (sh)a,b,c,d

38

8

938 938, 39

38

510 (4.4) 530

a,b,c,d

(19.2)a,b,c,d (–)g,h

a,b,c,d

13

555 (4.5)

13C 38

589 (1.7)a,b,c,d

1438 1538 38

16

56 š 8

1 š 0.2

362 š 36

3š

0.3f

1 š 0.1

564 š 56e

209 š 21

85 š 9f

30 š 3

e

205 š 21

76.1 š 7.8

f

14 š 2

5.2 š 0.7

129 š 13e

47.9 š 4.8

20 š 2f

7.4 š 0.7

16 š 2

3š

0.45f

65.7 š 6.7

8š

1f

511

42 š

547

(8.2)a,b,c,d

177 š

18e

36 š 6

e

13 š 2

2 š 0.3

570

(5.3)a,b,c,d

k

k

1838

532 (3.7)a,b,c,d

l

l

1938

569 (1.0sh)a,b,c,d 570

(2.0)a,b,c,d

166 š 17e 74 š

11e

61.6 š 6.3 27 š 4.1

1 š 0.17 3.0 š 0.4

f

1738

2038

40.8 š 4

11f

4 š 0.6

6e

516 (6.0)

110 š

f

f

(5.7)a,b,c,d a,b,c,d

37.5 š 4

240 š 24

e

0.7 š 0.1

17 š 2f

6.3 š 0.7

8š

3.0 š 0.4

1f

a Compound is transparent at the fundamental frequency 1064 nm. b Measured in CH Cl . c Measured at 30Ž C. 2 2 d Concentration unknown. e Method: SHG, measured at 1064 nm in CH Cl , ca. 4 ð 105 mol L1 , reference: p2 2 nitroaniline: β D 21.6 ð 1030 esu. f Data corrected for resonance enhancement at 532 nm using the two-level model

with β0 D β[1 (λmax /1064)2 ][1 (λmax /532)2 ]; damping factors not included. g Measured in hexane. h Measured at ambient temperature. i Measured in DMF. j Method: Kurtz-powder, measured at 1907 nm, reference: urea. k Response too low to measure. l Fluorescence detected. m SHG activity relative to urea: 35 C/ 7. n SHG activity relative to urea: 85 C/ 10. o SHG activity relative to urea: 8 C/ 2.

Ferrocene-Based Electro-Optical Materials

325

Table 9.2 Linear optical and quadratic nonlinear optical response parameters for Tp containing compounds (part II) Compound 2140,42 2241 2341,42 2441,42 2541,42 2641,42 2741,42

λmax /nm (ε/103 M1 cm1 )

Concentration /1019 cm3

484 (–)a 780 (2.405)a 588 (2.393)a 584 (sh)a 528 (11.664)a 549 (sh)a 550 (sh)a

10.5 13 18 7.1 16 6.9 7.2

(2) χ31 9 /10 esu

(2) χ15 9 /10 esu

(2) χ33 9 /10 esu

0.4 (0.04)b,c

0.4b

0.3b

b,d

b,d

0.5b

0.4b 0.6b 0.7b 0.4b 0.7b

b,d

(0.07)b,c

1.3 1.9 (0.27)b,c 2.4 (0.15)b,c 1.4 (0.20)b,c 1.5 (0.21)b,c

0.7b 1.0b 0.4b 0.4b

a Measured in CH Cl . b Corona-poled ﬁlms of the host polymer poly(methyl methacrylate) (PMMA) and the organometal2 2 lic compound measured at 1064 nm. c Normalised values in brackets, i.e. χ (2) of the same molecular concentration. d Very low response.

Kurtz-powder experiments on similar D-π-A systems in which the acceptor group is represented by a Tp–Mo complex were carried out by Hamor et al.39 For a number of the donor groups tested, only the ferrocenyl derivatives (9–11) exhibited detectable SHG activity in the solid state. It was concluded that bulky substituents at each end are necessary to produce packing arrangements favourable for SHG. The groups of Agull´o-L´opez and Cano explored the SHG properties of Tp containing compounds on corona-poled spin-coated ﬁlms of poly(methyl methacrylate) (PMMA) including these organometallic compounds.40–42 The nonzero components of the susceptibility tensor are listed in Table 9.2. Signiﬁcant values were found for (2) most compounds, the highest for the χ31 component. Kleinman symmetry was not (2) (2) 6D 3χ31 , which means that the tested complexes can not be considered obeyed, i.e. χ33 linear (one dimensional). The extension of the π-conjugated bridge improved the SHG response, as can be seen for the anilide bridged complex 23 and the diarylazoamide bridged compound 25. The incorporation of methyl substituents into the diarylazoamide bridge (complexes 24, 26 and 27) did not signiﬁcantly modify χ (2) components. Furthermore, from a comparison of the normalised χ (2) values (i.e. of the same molecular concentration) of 24 and 2541, 42 it appeared that the NDN group within the π-bridge was more effective for SHG than the CDC containing bridge (Table 9.2). 9.2.2

Sesquifulvalenes

In Tp containing compounds there is a bond between the π-linker and the metal atom, the properties of the latter being inﬂuenced by its ligands. A different group of compounds are complexes based on sesquifulvalene and closely related systems. The terminal accepting group is an organic moiety that can be modiﬁed by coordination of metal complexes (Figures 9.2 and 9.3; Table 9.3).43–51 In the ﬁeld of sesquifulvalene complexes that are NLO active, extensive research was done by Heck et al.43–46, 48, 52 The SHG activity of a great variety of sesquifulvalene complexes was determined by HRS resulting in β values, which ranged

326

Ferrocenes: Ligands, Materials and Biomolecules +

Ferrocene

Bridge MLn

Bridge

Metal Fragment MLn

2843–45

none

none

2943,44,46

none

Cr(CO)3

3043,44

none

RuCp

none

RuCp*

(E)-CH=CH

none

3345

(E,E)-(CH=CH)2

none

3445

(E,E,E)-(CH=CH)3

none

(E)-CH=CH

Cr(CO)3

H3C Fe CH3 CH HC 3 3 CH

(E)-CH=CH

Cr(CO)3

Fc–

(E)-CH=CH

RuCp*

Compound

Ferrocene

3143,44 3243,45

Fc–

3543,46 3643

3

3743 3843,45 3943

none Fc–

Cr(CO)3

S

4043

RuCp

4145

S

2

Fc–

4245 4343,46

S – –C–C– –

none none Cr(CO)3

Figure 9.2 Complexes based on sesquifulvalene as accepting group

between zero (or detecting ﬂuorescence) and up to 570 ð 1030 esu (Table 9.3). The results allowed structure–property relationships to be deduced. Archetype sesquifulvalene complexes that possess a direct link between the C5 and the C7 moiety and are mononuclear (such as ferrocenyl derivative 28) showed no SHG activity or just ﬂuorescence due to two-photon absorption. A second transition metal atom coordinated at the C7 ring inﬂuenced the electron accepting capabilities of that acceptor group. It can be derived from the compounds 29–31 that the type of metal atom is decisive. The coordination of a Cr(CO)3 moiety was not sufﬁcient to produce a measurable SHG response, whereas ruthenium derivatives (e.g. RuCp, RuCp*) were

Ferrocene-Based Electro-Optical Materials Ferrocene Compound

4448

4548

Ferrocene

HO2C

–

O2Cfc–

4647

Bridge

327

MLn

Bridge

Metal Fragment MLn

(E)-CH=CH

H3C Ru CH3 CH3 H3C CH3

(E)-CH=CH

H3C Ru CH3 CH3 H3C CH3

(E)-CH=CH Fc–

4747

(E,E)-(CH=CH)2

4849

– –C–C– –

4949

– –C–C– –

5049

– –C–C– –

Fc–

5150

none

5251

– –C–C– –

5351

– – –C–C–C–C– – –

Cr (CO)3 B Ru

B H3C Rh CH3 CH3 H3C CH3 B H3C Ir CH3 CH3 H3C CH3 B Co

Cr (CO)3 Cr (CO)3

Figure 9.3 Complexes based on systems closely related to sesquifulvalenes

more efﬁcient. Within the ruthenium containing group a substitution of Cp by Cp* did not modify the β value to any signiﬁcant extent. Furthermore, it was discovered that, in contrast to ferrocene derivatives, compounds based on ruthenocene did not exhibit ﬂuorescence due to two-photon absorption; thus giving ruthenocene derivatives a clear advantage.45

328

Ferrocenes: Ligands, Materials and Biomolecules

Table 9.3 Linear optical and quadratic nonlinear optical response parameters of complexes based on sesquifulvalene and closely related systems Compound

λmax /nm (ε/103 M1 cm1 )

2843

698 (3.1)a

β/1050 C m3 V2

β0 /1030 esu

b,c

– –

2844

725 700 (3.335)a,e

e,f,g

2845

726 (3.570)d 700 (3.330)a

h,g

–

2946

590 (3.100)d,i 570 (–)j,i

b,k

–

2944

590 (3.100)d,e 574 (4.880)a,e

e,f,l

–

2943

590 (–)d

b,c

–

3043

(–)a

44

(3.565)d,e

β/1030 esu

615

80b

d,e

29 e,f

17m n

β0 /1050 C m3 V2

6.3

30

637 (–) 614 (5.270)a,e

125

46.4

28

10

3143

604 (–)a

90b

33

18m

6.7

3144

628 (–)d,e 605 (4.510)a,e

115e,f

42.7

23m

8.5

3243

816 (5.500)d

o

–

3245

(10.460)d

816 765 (9.510)a

g

–

3345

865 (18.090)d 780 (12.770)a

g

–

3445

903 (20.215)d 782 (23.900)a

g

–

3543

670 (4.700)d

322b

120

113m

41.9

3546

670 (4.700)d,p 590 (6.000)j,p

320b

119

113m

41.9

3643

750 (–)d

230b

114m

42.3

43

38

845 (11.200)

d

85.4

b,c

– – –

3845

845 760 (9.470)a

g

3943

595 (–)d

o

4043

(–)a

370f

(11.210)d

– –

(11.210)d

735

4145

844 700 (9820)a

g

4245

885 (8.120)d 775 (2.440)a

g

137

175m

65.0

Ferrocene-Based Electro-Optical Materials

329

Table 9.3 (continued ) Compound

λmax /nm (ε/103 M1 cm1 )

4343

600 (3.100)d

570b

4346

600 560 (2.500)j,i

570b

4448

487 (1.419)d

b,k

4548

(1.655)d

q,k

47

46

4747

(3.100)d,i

β/1030 esu

507

304 (–) 334

r

β/1050 C m3 V2

β0 /1030 esu

β0 /1050 C m3 V2

212

105m

39.0

212

105m

39.0

– – s,t

193

(–)r

300s,t

(0.970)d

71.6 111

119m,t

44.2

164m,t

60.9

4849

446 440 (0.600)u

f,k

4949

456 (1.330)d 450 (sh)u

59f

22

–

5049

442 (0.985)d 442 (0.480)u

46f

17

–

5150

650 (2.000)d,p 600 (1.650)j,p 585 (1.640)v,p

90 š 30b

33 š 11

–

5251

387 (–)r 395 (–)v

9w

3

7x

3

5351

419 (–)r 411 (6.800)v

11w

9x

3

–

4.1

a Measured in nitromethane. b Method: HRS, measured at 1064 nm in CH Cl , reference: p-nitroaniline: β D 21.6 ð 2 2 1030 esu. c Not measurable. d Measured in CH2 Cl2 . e Concentration ca. 105 mol L1 . f Method: HRS, measured at 1064 nm in nitromethane, reference: p-nitroaniline: β D 34.6 ð 1030 esu. g Fluorescence detected. h Method: HRS, measured at 1064 nm, reference: p-nitroaniline. i Concentration: ca. 103 mol L1 . j Measured in acetone. k Not observed. l HRS signal not detectable. m Calculated by taking the two-level model into account. n Calculated from β D β (HRS) 0 [(1 4λ2max /λ2 )(1 λ2max /λ2 )]. o Not determined. p Concentration: ca. 104 mol L1 . q Method: HRS, measured at r s 1064 nm in acetonitrile, reference: p-nitroaniline. Measured in CHCl3 . Method: HRS, measured at 1064 nm in CHCl3 , reference: p-nitroaniline (β D 23 ð 1030 esu). t β values š15 %. u Measured in acetonitrile. v Measured in dimethylsulfoxide (dmso). w β333 values, method: HRS, measured at 1500 nm, in CHCl3 , reference: p-dimethylamino 0 values which were calculated by applying the two-level model an cinnamic aldehyde (β333 D 35 ð 1030 esu). x β333 the π -π * absorption band.

The incorporation of π-linkers such as (conjugated) double and triple bonds, thiendiyl and ethynylthienyl between the ﬁve and seven membered rings of the sesquifulvalene molecule can considerably affect the NLO activity. For the mononuclear sesquifulvalene complexes (32–34, 38, 41, 42) no SHG activity could be determined regardless of the presence of a spacer group. In complexes containing a Cr(CO)3 moiety in the acceptor group, the introduction of an ethylene bridge (35) increased the β value to ca. 320 ð 1030 esu and the incorporation of an ethinylene bridge (43) even further to β D 570 ð 1030 esu.43,45

330

Ferrocenes: Ligands, Materials and Biomolecules

Substitution of a Cp ring (35) by a Cp* ring (36) in the donor group (i.e., at the ferrocenyl moiety) resulted in diminishing the SHG activity.43, 46 When the C7 ring of the sesquifulvalene complex 35 was replaced by a benzene ring (46) the SHG activity dropped from β D ca. 320 ð 1030 esu to β D 193 ð 1030 esu, conﬁrming the superior electron accepting capabilities of sesquifulvalene system. On extending the π-system by an additional ethylene spacer (47) the SHG response increased to β D 300 ð 1030 esu.47 Some complexes were also tested in which the C7 ring of the sesquifulvalene moiety was replaced by a borabenzene ring (48–51).49, 50 For 48, which contained a ruthenium benzene moiety in the acceptor group, no SHG activity could be observed. When this moiety was exchanged for a Cp*Rh (49) or a corresponding iridium unit (50), β values of 59 ð 1030 esu and 46 ð 1030 esu, respectively, could be determined.49 The abovementioned borabenzene derivatives possessed an ethynylene bridge. Compared with these compounds the directly coupled borabenzene complex 51 exhibited an unexpectedly high SHG activity of β D 90 š 30 ð 1030 esu.59.2.3

Pyridine-Containing Complexes

The properties of a pyridyl group, which itself acts as an electron withdrawing group, can be modiﬁed by the coordination of electron deﬁcient fragments or converting to pyridinium salts. This, in turn, has an impact on the overall NLO properties of the considered substance. Research on this subject (Figures 9.4 and 9.5; Tables 9.4 and 9.5)

Ferrocene Compound

1338

Bridge

Ferrocene H3C CH3 H3C H3C Fe H3C CH3 CH3 H3C

N Bridge

R

Metal Fragment R H C 3

(E)-CH=CH

H3C 1438

1538

Fc–

H3C H C 3

Fe

(E)-CH=CH

CH CH3 3

(E)-CH=CH

CH3

H C N N CH3 ON 3 Mo N N BH Cl N N H3C CH3 CH3

H C N N CH3 ON 3 Mo N N BH Cl N N H3C

CH3

H3C

CH3

H C N N CH3 ON 3 Mo N N BH Cl N N H3C

CH3

Figure 9.4 NLO active compounds based on pyridinium and pyridyl groups as acceptor groups (I)

Ferrocene-Based Electro-Optical Materials Ferrocene Compound

Bridge

Ferrocene

N

R

Bridge

Metal Fragment R H3C

1638

(E)-CH=CH

H3C H3C

Fc– 1738

1838

–N=CH–

CH3 H3C H3C H3C Fe H3C CH3 CH3 H3C

1938

CH3

H C N N CH3 ON 3 Mo N N BH I N N

–CH=N–

CH3 CH3

H C N N CH3 ON 3 Mo N N BH Cl N N H3C

CH3

H3C

CH3

H C N N CH3 ON 3 Mo N N BH Cl N N H3C

CH3

H3C

CH3

H C N N CH3 ON 3 Mo N N BH Cl N N CH3 H3C

Fc–

H3C

CH3

H C N N CH3 ON 3 Mo N N BH Cl N N CH3 H3C

2038

Fc–

(E)-CH=CHC(=O)

5438,47,53–55

Fc–

(E)-CH=CH

none

(E)-CH=CH

none

(E)-CH=CH

none

5538

5638

H3C H3C

Fe

CH3 CH3

CH3 H3C H3C H3C Fe H3C CH3 CH3 H3C

–CH3+ I−

5758,59

–Et+ I−

5859

–Pr+ I−

5959 6159

–Bu+ I− –C5H11+ I−

6259

–C10H21+ I−

6354,55,59

–C18H37+ I−

6059

Fc–

(E)-CH=CH

Figure 9.4 (continued )

331

332

Ferrocenes: Ligands, Materials and Biomolecules Ferrocene Compound

Ferrocene

Bridge

N Bridge

6459 6547,53 6647,53 6747,53

7053

Metal Fragment R –O–

Fc–

(E)-CH=CH

Cr(CO)5 Mo(CO)5 W(CO)5

6853 6953

R

none Cr(CO)5

Fc–

Mo(CO)5

7153

W(CO)5

7253

(E)-CH=CH

–CH3+ I−

–CH3+ I−

7353 Fc–

–CH3+ I−

7453 2

75

38

7638

7738 7838 7938

CH3 H3C H3C H3C Fe H3C CH3 CH3 H3C

(E)-CH=CH

W(CO)5

CH3 H3C H3C H3C Fe H3C CH3 CH3 H3C

–CH=N–

none

S

–CH3+ I−

CH3 H3C H3C H3C Fe H3C CH3 CH3 H3C

none

W(CO)5

Figure 9.4 (continued )

was conducted by the groups of McCleverty,38 Peris,47, 53 Qian,54, 55 Huang,56, 57 Beck,58 Silver,59 Chung,60, 61 and Lee.61 Compound 54, which can be considered as the basic system since it features a fairly small π-system and an unmodiﬁed pyridyl group, exhibited a β value (HRS measurements) between 16 ð 1030 esu38 and 21 ð 1030 esu47 (Table 9.4). Modiﬁcation of the donor part by successive methylation of the cyclopentadienyl rings led to a noticeable increase of the SHG efﬁciency (cf. 55 and 56).38 A similar effect was observed for the methylation of 67 leading to 75.38, 47 The same trend was found also for ferrocene pyridyl compounds attached to tris(pyrazolyl)borato molybdenum complexes. Starting from the non methylated ferrocene moiety (14, β D 42 ð 1030 esu) the tetramethylated complex 15 exhibited a SHG efﬁciency of β D 177 ð 1030 esu, which was further enhanced for the octamethylated compound 13 (β D 205 ð 1030 esu).38 Alkylation at the nitrogen atom of the pyridyl ring was also realised (Table 9.5).56–59 Due to different experimental set-ups, obtained β values can hardly be compared. For

Ferrocene-Based Electro-Optical Materials +

Fe

−

Ph

N

O

N C16H33

333

Ln

Cocrystallisation

Fe

O Ph

N

4

Ln = La (80), Nd (81), Dy (82), Yb (83)56

8460 −

+

Fe

N

O

N C16H33

Ln O F3C

8557

Fe

N

Fe N

4

8761

Re(CO)3Br 2

8658

Figure 9.5 NLO active compounds based on pyridinium and pyridyl groups as acceptor groups (II)

Table 9.4 Linear optical and quadratic nonlinear optical response parameters of NLO active compounds based on pyridinium and pyridyl groups as acceptor groups (part I) Compound

λmax /nm (ε/103 M1 cm1 )

1338

555 (4.5)a,b

13C 38 1438 1538 38

16

β/1030 esu

β/1050 C m3 V2

β0 /1030 esu

β0 /1050 C m3 V2

205 š 21c

76.1 š 4.8

14 š 2d

5.2 š 0.7

589

(1.6)a,b

129 š

13c

47.9 š 4.8

20 š

7.4 š 0.7

511

(5.7)a,b

42 š

6c

16 š 2

547

(8.2)a,b

177 š

18c

65.7 š 6.7

516 (6.0)

a,b

36 š 6

1738

570

(5.3)a,b

e

1838

532 (3.7)a,b

f

38

a,b

19

2038 5438 5447 5453 5454,55

495 (3.1)

13 š 2

c

2d

3š

0.45d

1 š 0.17

8š

1d

3 š 0.4

2 š 0.3

d

0.7 š 0.1

e f c

166 š 17

61.6 š 6.3 27 š 4.1

570

(2.0)a,b

74 š

11c

466

(1.5)a,b

16 š

2c

468

(–)g

21h

468

(–)j

21h

7.1

3.3k

1.2

17 š 2d

6.3 š 0.7

8š

1d

3 š 0.4

5.4 š 0.7

3š

5d

1š2

7.1

4i

1

(continued overleaf )

334

Ferrocenes: Ligands, Materials and Biomolecules Table 9.4 (continued )

Compound

λmax /nm (ε/103 M1 cm1 )

5538

489 (2.9)a,b

5638

503

(4.4)a,b

5758

521 (–)l

6354,55

β/1030 esu

β/1050 C m3 V2

27 š 4.1

9 š 1d

3 š 0.4

76 š

28 š 4.1

6š

2 š 0.4

11c

68m 62.1k 63h

23

6553

477

(–)j

63h

23

478

(–)g

95h

35

6653

487

(–)j

95h

35

6747

491 (–)g

101h

37.5

53

491 (–)

j

h

101

37.5

6853

459

(–)j

146h

54.2

462

(–)j

369h

137

j

h

6953 53

476 (–)

448

166

7153

487

(–)j

535h

199

7253

553 (–)j

40n

15

7353

(–)j

197n

73.1

j

n

70

53

74

7538 75C 38

1d

25

401 (–)g

67

β0 /1050 C m3 V2

73 š 11cc

6547 6647

β0 /1030 esu

503

462 (–) 552

(4.8)a,b

402

(10.5)a,b

8.5

12i

4.5

12i

4.5

170

458

252 š

23i

25c

93.5 š 9.3

f

14 š 2d

5.2 š 0.7

f

38

76

503 (1.8)a,b

67 š 10c

25 š 4

5 š 1d

2 š 0.4

7738

538 (6.5)a,b

410 š 41c

152 š 15

8 š 1d

3 š 0.4

7838

515 (31.1)a,b

7938

571

(6.3)a,b

79C 38

463

(36.1)a,b

8658

501 (9.77)b

f

306 š

f

31c

113 š 12

f

24o

334 š 4d

124 š 1

f

8.9

a Measured at 30 Ž C. b Measured in CH Cl . c Method: HRS, measured at 1064 nm in CH Cl , ca. 4 ð 105 mol L1 , 2 2 2 2 reference: p-nitroaniline (β D 21.6 ð 1030 esu). d Data corrected for resonance enhancement at 532 nm using the two-level model with β0 D β[1 (λmax /1064)2 ][1 (λmax /532)2 ]; damping factors not included. e Response too low to measure. f Fluorescence occurs. g Measured in CHCl3 . h Method: HRS, measured at 1064 nm in CHCl3 , reference: p-nitroaniline (β D 23 ð 1030 esu), β values š15%. i Calculated by taking the two-level model into account. j Solvent not stated. k Method: EFISHG, measured at 1064 nm. l ZINDO-derived value.183 m Method: HRS, measured at 1910 nm in CHCl3 , reference: p-dimethylamino cinnamic aldehyde (DMAZ). n Method: HRS, measured at 1064 nm in acetone, reference: p-nitroaniline (β D 23 ð 1030 esu), β values š15%. o Method: HRS, measured at 1300 nm in CHCl3 , reference: p-dimethylamino cinnamic aldehyde (DMAZ, β D 41 ð 1030 esu).

Ferrocene-Based Electro-Optical Materials

335

Table 9.5 Linear optical and quadratic nonlinear optical response parameters of NLO active compounds based on pyridinium and pyridyl groups as acceptor groups (part II) β/1030 esu

β/1050 C m3 V2

Area per molecule /nm2

χ (2) /107 esu

Compound

λmax /nm (ε/103 M1 cm1 ) or [absorbance]

SHG activity relative to urea

5759

542.8 [0.337]a 542.8 [0.208]b

9.2c,d 11.7c,e

5859

541.6 [0.293]a 543.2 [0.287]b

14.6c,f 0.4c,e

5959

542.4 [0.169]a 542.4 [0.311]b

0.5c,g 2.9c,e

6059

543.2 [0.322]a 548.0 [0.327]b

0.3c,h 0.2c,e

6159

543.2 [0.163]a 548.8 [0.433]b

0.4c,i 0.3c,e

6259

544.0 [0.186]a 548.6 [0.517]b

0.5c,j 0.4c,e

6359

550.4 [0.275]a 548.8 [0.511]b

0.6c,k 0.3c,e

6459

471.6 [0.244]a

0.3c,e

8056

ca. 344 (–)l

158m

58.6

1.11

2.3

8156

ca. 344

(–)l

137m

50.9

1.02

2.2

8256

ca. 344

(–)l

150m

55.7

1.03

2.4

8356

ca. 344 (–)l

144m

53.5

1.02

2.9

84 (š)binaphthol60

–

0.3n

84 ()binaphthol60

518 (–)o

0.4n

8557

575 (–)l

8761

–

150m

55.7

0.80

2.8 ca. 4p

a Measured in water. b Measured in acetonitrile. c Method: Kurtz-powder, measured at 1907 nm, reference: urea. d Particle size: 200 µm. e Particle size: 20 µm. f Particle size: 300 µm. g Particle size: 150 µm. h Particle size: 500 µm. i Particle size: 160 µm. j Particle size: 125 µm. k Particle size: 250 µm. l Measured in CHCl . m SHG signals derived from 3 Langmuir-Blodgett (LB) ﬁlms by measurements at 1064 nm. n Kurtz-powder measurements at 1295 nm. o Solid state p spectrum. Kurtz-powder measurements at 1.36 µm.

a series of alkylated pyridinium compounds with alkyl groups ranging from methyl to octadecyl (57–63) the SHG activity was determined by the Kurtz-powder method.59 It turned out that only for the smaller alkyl groups, such as methyl (57), ethyl (58), and propyl (59), signiﬁcant SHG activity could be detected. All compounds used in Langmuir–Blodgett ﬁlms (80–83, 85) featured the same alkyl chain (n-hexadecyl).56, 57 They differed only in the type of metal used in the counter ion complex, which in turn did not inﬂuence signiﬁcantly the SHG efﬁciency, all β values being in the range of 137 ð 1030 esu and 158 ð 1030 esu. In contrast, the coordination of an electron withdrawing metal fragment, such as M(CO)5 (M D Cr (65), Mo (66), W (67)), led

336

Ferrocenes: Ligands, Materials and Biomolecules

to an enhancement of the nonlinear optical properties (Table 9.4), the effect being more pronounced for tungsten than for molybdenum or chromium.47,53 An increase in β values was also achieved when the octamethylated 56 formed a complex with a (CO)5 W moiety (75).38 By far the largest increase in SHG activity could be achieved by enlarging the π-system, e.g. for the related tungsten containing compound 71 the β value reached 535 ð 1030 esu.53 The pyridine nitrogen atom was also coordinated to tris(pyrazoly)borato molybdenum complexes (Table 9.4).38 For compound 54 this coordination led to an increase of the nonlinear optical activity (15, 16), though not as pronounced as for the abovementioned pentacarbonyl metal fragments. For the methylated compound 76 the introduction of the tris(pyrazolyl)borato molybdenum fragment (18) caused the occurrence of ﬂuorescence. The type of halogen atom used as a ligand in the molybdenum complex (e.g. 14 vs. 16) did not exert a signiﬁcant inﬂuence on the β values of the complexes. Far more pronounced, however, was the inﬂuence of methylation of the ferrocene moiety (13–15), that has already been mentioned above. The oxidation of the ferrocene moiety of 13 (β D 205 ð 1030 esu) caused a decrease of the NLO activity (13C , β D 129 ð 1030 esu).38 9.2.4

Complexes Containing a Nitro Function

A number of complexes that contain a nitro function as the acceptor group have been synthesised (Figures 9.6 and 9.7) and their nonlinear optical properties tested in various R2

Bridge Fe

Compound

Substituent R1

NO2

R1 Substituent R2

8854,55,62–64

H

H

8964

H

H

9054

H

H

9154,55

H

H

9262,63

Si(CH3)3

H

9362,63

CH3

H

9462,63

H

CH2OH

9563

PPh2

H

9663

H

9763

CHO

Bridge

Cl

OCH3 H

Figure 9.6 Complexes with a nitro group as electron withdrawing group (I)

Ferrocene-Based Electro-Optical Materials R2

Bridge Fe

Compound

Substituent R1

337

NO2

R1 Substituent R2

9865

H

H

9966

H

H

10066

H

H

10166

H

H

10266

H

H

10366

H

H

10466

H

H

10541,42

H

H

Bridge N N

N

N

N N

N

H3C 10640,42

H

H

N

H3C N

H N

Figure 9.6 (continued )

ways (Tables 9.6 and 9.7).38, 40–42, 54, 55, 62–70 The group of Balavoine and Daran probed a series of enantiomerically pure (E)-(2-(4-nitrophenyl)ethenyl)ferrocenes with diverse substituents in the 2-position on the cyclopentadienyl ring in solution by EFISHG measurements as well as in the solid state by Kurtz-powder experiments (92–97, 107, in Table 9.6).62, 63 It has been found that while these different chromophores had closely related SHG responses in solution (β values between 24 ð 1030 esu (93) and 53 ð 1030 esu for (94)) the same chromophores displayed large differences for the bulk SHG efﬁciencies (SHG activity relative to urea between 0 for 95 and 100 for 92). This effect was mainly ascribed to crystal packing: the chirality of the constituting parts prevented centrosymmetric crystal assemblies and therefore preserved SHG efﬁciency in the solid state. Even so, 93 showed a crystal packing close to centrosymmetry and thus a rather small SHG activity (6 relative to urea). 92 having a trimethylsilyl substituent in 2-position possessed the highest SHG activity (100 relative to urea) and showed also that the particle size is a key factor, i.e. the smaller the particle size the better the SHG efﬁciency.

338

Ferrocenes: Ligands, Materials and Biomolecules NO2 Fe

SH O

Fe

H3 C CH3 H3C H3C Fe CH3 H 3C H3 C CH3

10763

S

NO2

10938,68,69 10867

O

SH ON

Fe

2

SH

O2N

Fe

11067

11167

NO2 Fe

NO2

O2N

Fe NO2 Fe 11270 O2N 11370 ON 2

11470

Figure 9.7 Complexes with a nitro group as electron withdrawing group (II)

Ma et al. investigated (by means of Kurtz-powder measurements) a series of complexes (99–104) that contained a nitro accepting group as well as an imino function within the π-bridge.66 The highest SHG activity was determined for 99 (190 ð urea), in which the nitro group was in ortho position relative to the π-bridge. The para isomer 101 exhibited a considerably smaller SHG efﬁciency (60 ð urea). Also it was found that the orientation of the imino group did inﬂuence the extent of the SHG activity. Compounds with the N -terminus of the imino group proximal to the ferrocene moiety (101, 103) exhibited SHG efﬁciencies inferior to 102 and 104. Self-assembled monolayers of ferrocene based chromophores containing a nitro function (110, 111) and 108 (as a reference molecule without a nitro function) on a gold surface were examined by Uosaki et al.67 These systems showed a jump in SHG activity when the ferrocene moiety was oxidised to the ferrocenium cation. The βzzz ratios between the reduced and oxidised states were 1.7 and 2.0 for 110 and 111, respectively. The origin of the difference in SHG intensity was attributed to changes in the orientation of the SAM and in the hyperpolarisabilities of these molecules. The former was represented by an increased angle between the surface of the gold electrode and the chromophore containing chain. It was stated that due to the reversible oxidation, which could be repeated many times, the described SAMs possessed switchable NLO

10466

10366

10266

10166

10066

9362,63 9462,63 9563 9663 9763 9865 9966

ca. 500 (–)a 510 (5.700)e 496 (–)h

8862,63 8864 8854 8855 8964 9054 9154,55 9262,63 24b 53b

51.2o

490 (1.76)n 448 (–)q 485 (4.48)r 595 (–)s 437 (–)q 481 (2.44)r 599 (–)s 422 (–)q 458 (2.72)r 529 (–)s 429 (–)q 470 (2.56)r 533 (–)s 431 (–)q 454 (2.64)r 537 (–)s 439 (–)q 467 (2.72)r 543 (–)s

31b 152f 31i 32.1i 110f 66i 60.6i 36b

β/1030 esu

ca. 500 (–)a ca. 500 (–)a 500 (–)l 504 (–)m

ca. 500 (–)a

490 (2.200)e 500 (–)h

λmax /nm (ε/103 M1 cm1 )

Compound

19.0

8.9 20

11 56.4 12 11.9 40.8 24 22.5 13

β/1050 C m3 V2

6.1p

60g

72g

β0 /1030 esu

2.3

22

27

β0 /1050 C m3 V2

110t

75t

95t

60t

95t

190t

100c,d 90c,j 80c,k 6c,d 20c,d 0c,d,m 20c,d 4c,d

0c,d

βzzz /1030 cm5 esu1

(continued overleaf )

SHG activity

Table 9.6 Linear optical and quadratic nonlinear optical response parameters of complexes with a nitro group as electron withdrawing group (part I)

Ferrocene-Based Electro-Optical Materials 339

494 (–)l 497 (–)ag 495 (0.47)ah 493 (–)l 487 (–)ag 488 (0.56)ah 481 (–)l 481 (0.30)ah

646 (4.7)z 586 (–)aa 655 (–)s 387 (14.2)z 851 (–)ad

λmax /nm (ε/103 M1 cm1 )

25 š 4ab

316 š 32ab

esu

9.3 š 1

117 š 12

β/1050 C m3 V2

10 š 2ac

95 š 10ac

β0 /1030 esu

Table 9.6 (continued )

3.7 š 0.7

35 š 3.7

β0 /1050 C m3 V2

ai,aj

ai,aj

0.28ai

0.08u,w,af 0.33u,x,af

0.11u,w,ae 0.61u,x,ae

8c,d 0.04u,v,w 0.06u,v,x

SHG activity

164y

250y 82.3y

147y

11.4y

25.4y

βzzz /1030 cm5 esu1

a Measured in dioxane. b Method: EFISHG, measured at 1907 nm in dioxane, in various concentrations (0 to 5 ð 103 mol L1 ), reference: 2-methyl-4-nitroaniline (MNA). c Method: Kurtz-powder measurements at 1907 nm, reference: urea (particle size: 50– 80 µm). d Particle size: 50– 80 µm. e Measured in CHCl3 at ambient temperature, concentration: 105 mol L1 . f Method: HRS, measured at 1.06 µm in CHCl3 . g Determined by the Oudar-Chemla equation. h Solvent not stated. i Method: EFISHG, measured at 1064 nm. j Particle size: 80– 125 µm. k Particle size: 125– 180 µm. l Measured in CHCl3 . m Decomposition. n Measured in acetonitrile. o Method: HRS, measured at 1064 nm in acetonitrile, concentration 105 –106 mol L1 , reference: p-nitroaniline (β D 23.0 š 2.2 ð 1030 esu). p Calculated using the two-state model. q Measured in Et2 O. r Measured in CH2 Cl2 . s Measured in dmso. t Method: Kurtz-powder measurements at 1907 nm, reference: urea (particle size: <74 µm). u In situ SHG activity of SAM on gold, measured in 0.1 mol L1 HClO4 solution at 1064 nm, values of SHG intensity were normalised to the SHG intensity at the bare gold electrode at 0 mV and the intensity of the incident light. v Surface coverage (SAM): 4.0 ð 1014 molecules cm2 . w SHG at 0 mV. x SHG at 850 mV. y Calculated from the optical parameters f , µgn , ωgn when the wavelength of the incident light and the second harmonic are 1.06 µm and 532 nm, respectively. z Measured in CH2 Cl2 at 30 Ž C. aa Measured in hexane. ab Method: HRS, measured at 1064 nm in CH2 Cl2 , concentration ca. 105 mol L-1 , reference: p-nitroaniline (β D 21.6 ð 1030 esu). ac Data corrected for resonance enhancement at 532 nm using the two-level model with β D β[1 (λ 2 2 max /1064) ][1 (λmax /532) ], damping factors not included. ad) 0 Measured in CH2 Cl2 , weak transition. ae Surface coverage (SAM): 1.7 ð 1014 molecules cm2 . af Surface coverage (SAM): 2.0 ð 1014 molecules cm2 . ag Measured in methanol. ah Measured in THF. ai Method: Kurtz-powder measurements at 1.3 µm, reference: urea. aj Not determined unambiguously.

11470

11370

111C 67 11270

110C 67 11167

11067

109C 38

108C 67 10938, 68, 69

10763 10867

Compound

β/1030

340 Ferrocenes: Ligands, Materials and Biomolecules

Ferrocene-Based Electro-Optical Materials

341

Table 9.7 Linear optical and quadratic nonlinear optical response parameters of complexes with a nitro group as electron withdrawing group (part II) Compound

λmax /nm (ε/103 M1 cm1 )

(2) χ31 /109 esu

(2) χ15 /109 esu

(2) χ33 /109 esu

10542

6.6 (1.10)a,b

4.2a

2.4a

10541

6.6a,c

4.2a

2.4a

10642

1.4 (0.09)a,b

0.9a

0.5a

1.4a,c,e

0.9a,c,e

0.5a,c,e

10640

500 (–)d

a Corona-poled ﬁlms of the host polymer poly(methyl methacrylate) (PMMA) and the organometallic compound measured at 1064 nm. b Normalised values in brackets, i.e. χ (2) of the same molecular concentration. c Method: Maker (2) D 1.96 ð 107 esu). d Measured in CH2 Cl2 at ambient temperature, fringes technique, reference: LiNbO3 plate (χ33 4 5 1 e concentration ca. 10 10 mol L . Molecular concentration: 15 ð 1019 cm3 , ﬁlm thickness: 1.6 µm.

properties. On the other hand, McCleverty et al. reported that the oxidation of 109 causes a sharp drop in NLO efﬁciency; cf. β of the reduced state: 316 ð 1030 esu vs. 25 ð 1030 esu for the oxidised state (109C ).38, 69 As considered in detail later (Figures 9.15 and 9.16; Tables 9.16 and 9.17), cis/trans isomers of a complex with a nitro function as the electron accepting unit (88/89) exhibited markedly different SHG efﬁciencies, clearly favouring the trans isomers (Table 9.6).62–64 Attempts were made to ﬁnd organometallic/inorganic groups that are even more electron accepting than the nitro function. However, it turned out that the nitro group behaved as a much better acceptor group than the Mo(TpAr )(NO)(Cl)(X) or carborane acceptor groups in compounds having the same ferrocenyl donor group and polarising bridging system.41, 64 9.2.5

Compounds with Barbituric Acid Derivatives

Derivatives of barbituric acid are another group of molecules that were used as electron accepting moieties (Figure 9.8, Table 9.8).71–75 In accordance with the general trend, the SHG efﬁciencies of such compounds increased upon an enlargement of the π-system. Chai et al. used ferrocene as part of the π-bridge between an anilino donor and an acceptor based on barbituric acid (120).74 Langmuir–Blodgett ﬁlms of this compound possessed β values as high as 1.8 ð 1028 esu (Bloembergen and Pershan method,184 measured at 1064 nm). The intention was that the redox active ferrocene moiety within the π-linker could be used for switching the overall SHG properties of the ﬁlm. Das et al. tested whether supramolecular assemblies that were formed by electron donors and acceptors via hydrogen bonds possessed SHG activity.75 The highest SHG response (I2ω (a.u.) D 80 mV; HRS method, measured at 1064 nm, in CHCl3 in varying concentrations) was found for 121 using a ratio of 1:5 for the barbituric acid derivative and the diaminopyridine derivative, respectively. It was assumed that because of the hydrogen bonds a noncentrosymmetric arrangement was formed, favouring large NLO properties.

342

Ferrocenes: Ligands, Materials and Biomolecules

Fe Compound 115

71

116

71

Bridge

Accepting Group

Bridge

Accepting Group O

none

NH S

11772,73

none

11872,73

(E)-CH=CH

11972,73

NH

(E)-CH=CH

O O N S N O

(E,E)-(CH=CH)2 O

(C18H37)2N

Fe Fe

12074

HN

N

NH2

O NH

O

Fe

O

HN O

O

NH HN O 12175

Figure 9.8 NLO active compounds based on barbituric acid derivatives as electron accepting groups

Table 9.8 Linear optical and quadratic nonlinear optical response parameters of complexes with a barbituric acid derivative as electron withdrawing group Compound

λmax /nm (ε/103 M1 cm1 )

β/1030 esu

11571

601 (2.600)a 586 (5.900)b 602 (22.300)a 636 (5.500)b

11671

Compound 11772,73 11872,73 11972,73

λmax /nm (ε/103 M1 cm1 ) 579 (1.0)e 618 (1.4)e 632 (2.3)e

β/1050 C m3 V2

β0 /1030 esu

β0 /1050 C m3 V2

181c

67.2

34d

13

214c

79.4

40d

15

µβ/1048 esu 100f 850f 1900f

a Measured in acetone. b Measured in CHCL . c Method: HRS, measured at 1064 nm in acetone, reference: p-nitroaniline 3 (β D 25.9 ð 1030 esu). d Calculated from the two-level model. e Measured in CH2 Cl2 . f Method: EFISHG, measured at 1907 nm in CHCl3 , reference: 4-N ,N -dimethylamino-40 -nitrostilbene (µβ ¾500 ð 1048 esu), error: š20 %, values have not been corrected for the electronic deformation contribution to the EFISHG signal.

Ferrocene-Based Electro-Optical Materials

9.2.6

343

Compounds with Nitrile Functions

The cyano function is a powerful acceptor group and was incorporated in a number of ferrocene containing NLOphores (Figures 9.9 and 9.10; Tables 9.9 and 9.10).47, 54, 55, 70–73, 76–82 Peris et al. examined a series of complexes, of which some were coordinated via their cyano group to a metal fragment to form heterobimetallic Bridge

Accepting Group

Fe Compound 12254,55 123

Bridge

Accepting Group

(E)-CH=CH

–CN

47,54

–CN

12447

–CNCr(CO)5

12547

–CNW(CO)5

12680

–RuCp(PPh3)2

12780

–FeCp(CO)2 NC

12876

Ph S

12976

NC

CN CN CN CN CN

13071 13171 13277

S S

–CH=C(CN)2 S NC

H N

CN NH2

13378 13478 13571,73,79 13671∗ 13771,73,79 13873,79

13979

none

–CH=C(CN)2

none (E)-CH=CH (E,E)-(CH=CH)2

(all E)-(CH=CH)4

Figure 9.9 Compounds with cyano units as acceptor group (I)

344

Ferrocenes: Ligands, Materials and Biomolecules Bridge

Accepting Group

Fe Compound

Bridge

Accepting Group

14072,73

none

14172,73

(E)-CH=CH

14272,73

(E,E)-(CH=CH)2

14373

CN

NC

O

(all E)-(CH=CH)3

S O CN

NC 14481

none O NC

14581

CN

none NC

CN

∗ In this compound the ferrocene moiety is methylated nine-fold, Me9Fc−

Figure 9.9 (continued ) NC N

CN

H CN

N Fe

Fe

Fe

Si Si Si Si Si Si 14782

14677 H Fe

Si Si Si Si Si Si

CN

CN CN

NC

Fe CN

14882

14970 CN

Fe

NC 15070

Figure 9.10 Compounds with cyano units as acceptor group (II)

Ferrocene-Based Electro-Optical Materials

345

Table 9.9 Linear optical and quadratic nonlinear optical response parameters of compounds with cyano units as acceptor group (part I) β/1030 esu

β/1050 C m3 V2

β0 /1030 esu

β0 /1050 C m3 V2

34e

13

101

39e

14 18 16

Compound

λmax /nm (ε/103 M1 cm1 )

12254,55

466 (–)a

5.4b

2.0

c

203d

75.4

47

123

473 (–)

12354

10b

3.7

12447

184 (–)c

271d

12547

487

(–)c

375d

139

48e

126[BF4 ]80

484 (–)c

325d

121

44f

]80

(–)c

186d

69.0

25f

12780

474 (–)c

171d

63.5

28f

10

12876

663 (–)c

180g

66.8

82h

30

12976

724

(–)c

140g

52.0

51h

19

13071

555 (6500)i 570 (7.300)c

217j

80.5

14f

13171

555 (10.800)i 554 (13.400)c

173k

64.2

5f

2

13277

508 (–)c

476l

177

40f

15

13571

525 (2.900)i 536 (2.800)c

90k

33

0.5f

13671

581 (3.700)i 582 (3.400)c

55k

20

8f

13771

558 (6.300)i 570 (10.700)c

140k

52

10f

14677

587 (–)c

576l

214

52f

126[PF6

485

9.3

5.2

0.2 3 3.7 19

a Unknown solvent. b Method: EFISHG, measured at 1064 nm. c Measured in CHCl . d Method: HRS, measured at 3 1064 nm in CHCl3 , reference: p-nitroaniline (23 ð 1030 esu), value š15 %. e Calculated from the two-level model, damping factors not included. f Calculated from the two-level model. g Method: Harmonic Light Scattering (HLS), measured at 1.91 µm in CHCl3 in various concentrations, reference: N -4-nitrophenyl-prolinol (NPP). h No calculation model given. i Measured in acetone. j Method: HRS, measured at 1064 nm, in acetone, reference: p-nitroaniline (25.9 ð 1030 esu). k Method: HRS, measured at 1064 nm in CHCl3 , reference: p-nitroaniline (23 ð 1030 esu). l Method: HRS, measured at 1064 nm in CHCl , concentration 105 mol L1 , reference: p-nitroaniline (β D 3 17.4 š 0.6 ð 1030 esu by the external method using p-nitroaniline in dioxane (β D 16.9 š 0.4 ð 1030 esu).

compounds.47 In relation to the monometallic compound 123 (β D 203 ð 1030 esu) it was found that the coordination of an electron accepting moiety such as M(CO)5 (M D chromium (124), tungsten (125)) enhanced the NLO activity of the compound, whereas the effect was more pronounced for tungsten than for chromium (Table 9.9). The groups of Peris and Persoons reported that the counter ion inﬂuenced the SHG response of a compound.80 The complexes 126[BF4 ] and 126[PF6 ] only differed in their counter ion. The former compound exhibited a β value of 325 ð 1030 esu compared to 186 ð 1030 esu of the latter complex (Figure 9.9, Table 9.9).

346

Ferrocenes: Ligands, Materials and Biomolecules

Table 9.10 Linear optical and quadratic nonlinear optical response parameters of compounds with cyano units as acceptor group (part II) Compound

λmax /nm (ε/103 M1

13378 13478 13579 13573 13779 13773 13879 13873 13979 14072 14073 14172 14173 14272 14273 14373 14481

µβ/1048 esu

µβ0 /1048 esu

cm1 )

543 (10.000)a 500 (sh, 12.000)a 526 (–)e 526 (0.3)e 556 (–)e 556 (0.6)e 568 (–)e 568 (0.9)e i

(0.7)a

670 667 (0.7)a 724 (1.0)a 721 (1.0)a 745 (3.3)a 746 (3.3)a 743 (2.3)a 613 (0.4)k 622 (0.4)l 632 (0.5)m

450 š 50b,c 570 š 30b,c 92f 92h 420f 420h 1120f 1120h 4600f 160j 160j 1400j 1400j 3000j 3000j 11200j 160j

14581

637 (0.7)k 656 (0.7)l 659 (0.6)m

14782

367 (17.300)n 363 (–)p 364 (–)a

o

14882

368 (17.500)n 364 (–)p 363 (–)a

o

14970q 15070s

492 (0.34)r 473 (–)m 473 (–)t 469 (0.45)r

200 š 20c,d 370 š 20c,d 60–80[70]g 250–340[300]g 660–875[770]g 3300g

280j

a Measured in CH Cl . b Method: EFISHG, measured at 1.54 µm in CH Cl , concentration: (1 10) ð 104 mol L1 . 2 2 2 2 c µβ and µβ values are deﬁned according to the ‘traditional’ EFISHG deﬁnition.185 d Calculated from the two-level 0 model, taking into account (where necessary) the spectral width of the absorption to the lowest excited state. e Measured

in acetone. f Method: EFISHG, measured at 1907 nm in acetone, measurements are calibrated relative to a quartz wedge, for which the experimental value of the quadratic susceptibility d11 D 1.2 ð 1019 esu determined at 1.06 µm was used; to account for dispersion, this value is extrapolated to d11 D 1.1 ð 1019 esu at 1.91 µm; accuracy ranges between 5 and 20 %. g Range of µβ0 values derived from the two-level model, using two absorption maxima; an approximated µβ0 value can then be estimated by taking the average and is indicated between square brackets. h Method: EFISHG, measured at 1907 nm, in acetone. i The two absorption bands tend to be closer together with increasing chain length; as a result, the two bands overlap for the longest derivative and the maximum of the lowest energy band cannot be determined. j Method: EFISHG, measured at 1907 nm, in CHCl3 . k Measured in hexane. l Measured in benzene. m Measured in CHCl . n Measured in cyclohexane. o Method: EFISHG, quite exceptional values 3 of the ﬁrst hyperpolarisability β have been measured; up to three times that of the open chain analogues. p Measured in acetonitrile. q Activity: 0.24 relative to urea, method: Kurtz-powder, measured at 1.3 µm, reference: urea. r Measured in THF. s Method: Kurtz-powder, measured at 1.3 µm, not determined unambiguously. t Measured in methanol.

Ferrocene-Based Electro-Optical Materials

347

When cyano compounds with π-systems of different size were considered, the general result was that the extension of the π-systems caused an increase of the SHG response (Table 9.10). The incorporation of a p-vinylenephenylene group into the πsystem of 133 raised the µβ value from 450 ð 1048 esu to 570 ð 1048 esu (134).78 The extension of the polyenic bridge exempliﬁed by the cyano complexes 135 and 137–139, was accompanied with a marked rise in µβ values79 (a more detailed description of the inﬂuence of the length of polyenic chains on the NLO response is given below). Complexes 128 and 129 (Table 9.9) combine a larger π-system with an increased number of acceptor groups, i.e. cyano groups. This led to a higher β value76 but not to such an extent than the introduction of another ferrocene donor into 132 resulting in 146.77 9.2.7

Polyene Bridged Compounds

The effect of the extension of a polyenic bridge on the SHG efﬁciency (Tables 9.11 and 9.12) of the respective complexes (Figure 9.11) has been studied by numerous groups.45, 47, 54, 55, 71–73, 79, 83–86 It has been mentioned before that mononuclear ferrocene complexes of sesquifulvalene derivatives showed strong ﬂuorescence due to two photon absorption (section 9.2.2). This was also true for similar complexes featuring a polyenic chain of different length between the C5 and C7 rings of the sesquifulvalene moiety (compounds 32–34 in Table 9.11).45 Coordination of a Cr(CO)3 acceptor moiety to a closely related system featuring a benzene ring instead of the C7 ring, resulted in a detectable SHG response. Furthermore, the introduction of a second double bond caused an increase in β values from 193 ð 1030 esu (46) to 300 ð 1030 esu (47).47 The same trend was observed for other complexes with polyenic chains. For push-pull polyenes with ferrocene as the donor group and a Fischer carbene complex as the acceptor moiety (151–157), an increase in SHG activity was achieved by increasing the number of oleﬁnic double bonds, by choosing a solvent with a higher dielectric constant (from hexane to acetonitrile), and by changing chromium for tungsten as the metal centre of the Fischer carbene complex moiety.83 No bulk susceptibility of these compounds could be determined because of crystallisation in the centrosymmetric space groups. When the Fischer carbene complex was replaced by electron accepting groups such as formyl (172–175) or dicyanovinyl (135, 137–139) a lengthening of the conjugation path resulted in a pronounced increase in µβ values (Table 9.12).73,79 In addition, overall higher µβ values and a steeper rise were found for the dicyanovinyl substituted complexes. Complexes, in which the acceptor group was represented by azulenylium (158–161) or guaiazulenylium (162–165 in Table 9.11) cations, yielded quite high β values despite relatively short donor–acceptor distances, the azulenylium function being a better acceptor than the methyl and isopropyl groups bearing guaiazulenylium function.84 The π-linker of compounds with a dinuclear iron complex as acceptor group (166–171) was systematically modiﬁed, i.e. lengthened. The addition of a thiophene group enhanced β to a larger extent (β D 679 ð 1030 esu for 170) than two additional double bonds (β D 562 ð 1030 esu for 169).85 For a series of compounds using dicyanomethylene-2,3-dihydrobenzothiophene as acceptor (140–143), the group of Marder and Blanchard-Desce found a sharp increase of the ﬁrst molecular hyperpolarisability when the π-system was enlarged.72, 73 Separating donor and acceptor by three additional CDC double bonds resulted in an increase of the µβ value from 160 ð 1048 esu (140) to 11 200 ð 1048 esu (143).

348

Ferrocenes: Ligands, Materials and Biomolecules

Table 9.11 Linear optical and quadratic nonlinear optical response parameters of compounds with different polyene chain lengths as part of the π -bridge (part I) Compound

λmax /nm (ε/103 M1 cm1 )

β/1030 esu

3245

816 (10.460)a 765 (9.510)b 865 (18.090)a 780 (12.770)b 903 (20.215)a 782 (23.900)b 304 (–)d 334 (–)d 496 (–)g

c

3345 3445 4647 4747 8854 8855 9054 11571

15183

500 (–)g 601 (2.600)i 586 (5.900)d 602 (22.300)i 636 (5.500)d 525 (2.900)i 536 (2.800)d 558 (6.300)i 570 (6.700)d 543.2 (–)k

15283

546.3 (–)k

15383

549.9 (–)k

15483

558.8 (–)k

15583

558.2 (–)k

15683

550.9 (–)k

11671 13571 13771

β/1050 C m3 V2

β0 /1030 esu

β0 /1050 C m3 V2

193e 300e 31h 32.1h 66h 181j

71.4 111 11.5 11.9 24 67.2

119f 164f

44.2 60.9

34f

13

214j

79.4

40f

15

90e

33

140j

52.0

46.8k,l,m 69a,l 89l,i 110l,n 160.6k,l,m 240a,l 306l,i 410l,n 342.8k,l,m 440a,l 760l,i 1030l,n 66.8k,l,m 82a,l 102l,i 160l,n 220.4k,l,m 402a,l 520l,i 600l,n 408.0k,l,m 580a,l 840l,i 1140l,n

17.4 26 33 41 59.61 89.1 114 152 127.2 163 282 382 24.8 30 37.9 59.4 81.81 149 193 223 151.4 215 312 423.2

c c

0.5f

0.2

10f

3.7

(continued overleaf )

Ferrocene-Based Electro-Optical Materials

349

Table 9.11 (continued ) Compound

λmax /nm (ε/103 M1 cm1 )

β/1030 esu

β/1050 C m3 V2

15783

541.4 (–)k

15884 15984 16084 16184 16284 16384 16484

716 (10.200)a 798 (17.300)a 866 (22.900)a 920 (35.900)a 724 (13.500)a 782 (21.600)a 827 (35.400)a

289.7 356 638 898.3 121

16584 16685

859 (43.300)a 655 (10.953)a 634 (9.390)n 734 (13.881)a 698 (10.029)n 775 (18.277)a 717 (16.275)n 813 (23.106)a 720 (14.954)n 757 (19.966)a 688 (17.147)n 764 (22.363)a 677 (19.199)n 474 (–)g 565 (5.100)i 574 (5.000)d 581 (10.000)i 593 (10.700)d

780.4k,l,m 960a,l 1720l,i 2420l,n 326o – 810o – 220o 1539o,p 1373o,p (360)q – 113r

41.9

36s

13

156r

57.9

74s

27

227r

84.3

120s

44.5

562r

209

313s

116

679r

252

343s

127

1195r

443.6

615s

228

5.4h 224e

2.0 83.1

26f

9.7

182j

67.6

25f

9.3

3.6h 35t 39t 42t 8.4h 22.6h

1.3 13 14 16 3.1 8.39

16785 16885 16985 17085 17185 17254,55 17671 17771 17854,55 17886 17986 18086 18155 18255

(–)n

295 320 (–)n 346 (–)n

β0 /1030 esu

145f

β0 /1050 C m3 V2

53.8 –

301

451f

167 –

81.7 571.3 509.7 (134)

101f 821f 770f (132)q

37.5 30.5 285 (49.0)

–

a Measured in CH Cl . b Measured in nitromethane. c Fluorescence occurs. d Measured in CHCl . e Method: HRS, 2 2 3 measured at 1064 nm in CHCl3 , reference: p-nitroaniline (β D 23 ð 1030 esu). f Calculated from the two-level model. g Solvent not stated. h Method: EFISHG, measured at 1064 nm. i Measured in acetone. j Method: HRS, measured at 1064 nm in acetone, reference: p-nitroaniline (β D 25.9 ð 1030 esu). k Measured in hexane. l Method: HRS, measured at 1064 nm, in CHCl3 , concentration: 105 107 mol L1 , reference: p-nitroaniline (β D 24.5 ð 1030 esu). m Value š8 %. n Measured in acetonitrile. o Method: HRS, measured at 1064 nm in CH Cl , reference: p-nitroaniline. 2 2 p Enhanced due to two-photon absorption ﬂuorescence. q Obtained from high frequency demodulation at λ D r 6 1300 nm. Method: HRS, measured at 1064 nm in dry CH2 Cl2 , concentration: 10 104 mol L1 , reference: p-nitroaniline (β D 21.6 ð 1030 esu). s Calculated from the two-level model using the lower energy charge-transfer band. t Method: HRS.

350

Ferrocenes: Ligands, Materials and Biomolecules

Table 9.12 Linear optical and quadratic nonlinear optical response parameters of compounds with different polyene chain lengths as part of the π -bridge (part II) Compound

λmax /nm (ε/103 M1 cm1 )

µβ/1048 esu

11772

579 (1.0)a

100b

11773

576 (1.0)a

100c

11872

618 (1.4)a

850b

11873

617 (1.4)a

850c

11972

632 (2.3)a

1900b

11973

632 (2.3)a

1900c

13579

(–)d

526

92e

13573

526 (0.3)d

92c

13779

556 (–)d

420e

13773

556 (0.6)d

420c

13879

568 (–)d

1120e

13873

568 (0.9)d

1120c

13979

d,g

4600e

14072

670 (0.7)a

160b

14073

667 (0.7)a

160c

14172

724 (1.0)a

1400b

14173

721 (1.0)a

1400c

14272

745 (3.3)a

3000b

14273

746

(3.3)a

3000c

14373

743 (2.3)a

11200c

17279

480 (–)d

60e

17273

480 (0.2)d

60c

17379

494 (–)d

215e

17373

µβ0 /1048 esu

60–80[70]f 250–340[300]f 660–875[770]f 3300f

42–49[45]f 147–184[165]f

494 (0.3)d

215c

79

174

d,g

560e

440f,g

17579

d,g

1150e

870f,g

a Measured in CH Cl . b Method: EFISHG, measured at 1907 nm, in CHCl , error: š20 %, µβ values have not been 2 2 3 corrected for the electronic deformation contribution to the EFISHG signal. c Method: EFISHG, measured at 1907 nm, in CHCl3 . d Measured in acetone. e Method: EFISHG, measured at 1907 nm in acetone, measurements are calibrated relative to a quartz wedge, for which the experimental value of the quadratic susceptibility d11 D 1.2 ð 1019 esu determined at 1.06 µm was used; to account for dispersion, this value is extrapolated to d11 D 1.1 ð 1019 esu at 1.91 µm; accuracy ranges between 5 and 20%. f Range of µβ0 values derived from the two-level model, using two absorption maxima; an approximated µβ0 value can then be estimated by taking the average and is indicated between square brackets. g The two absorption bands tend to be closer together with increasing chain length; as a result, the two bands overlap for the longest derivative and the maximum of the lowest energy band cannot be determined.

Ferrocene-Based Electro-Optical Materials

351

Linker Fe Compound

n Linker

Accepting Group

Chain Length n

3245

none

1

3345

none

2

3445

none

3

4647

none

1

4747

none

2

15183

none

1

15283

none

2

83

none

3

15483

none

1

15583

none

2

15683

none

3

15783

none

4

15884

none

0

15984

none

1

16084

none

2

16184

none

3

16284

none

0

16384

none

1

16484

none

2

16584

none

3

16685

none

1

16785

none

2

85

none

3

153

168

16985

Accepting Group

Cr(CO)3

MeO Cr(CO)5

MeO W(CO)5

H3C

none

CH3

O

170

17185

S S

C Fe

C

C O Fe

4 O

85

CH3

C

1 2

Figure 9.11 Compounds with different polyene chain lengths as part of the π -bridge

352

Ferrocenes: Ligands, Materials and Biomolecules Linker Fe Compound

n Linker

Accepting Group

Chain Length n

17254,55,73,79

none

1

17373,79

none

2

17479

none

4∗

17579

none

6∗

13571,73,79

none

0

13771,73,79

none

1

13873,79

none

2

13979

none

4∗

11571

none

0

71

none

1

11772,73

none

0

72,73

none

1

Accepting Group

–CHO

116

118

–CH=C(CN)2

O NH S NH O O N S N

11972,73

none

2

14072,73

none

0

14172,73

none

1

14272,73

none

2

14373

none

3

17671

none

0

177 71

none

1

O

NC

O

CN

S O O

O 8854,55

none

1

9054

none

2

17854,55,86

none

1

179

86

none

2

180

86

none

3

18155

none

1

18254,55 none 2§ ∗ The marked polyenic chain features two additional methyl groups § The polyenic chain features a chloro substituent

Figure 9.11 (continued )

p-C6H4NO2

–CO2CH3

p-C6H4Cl

Ferrocene-Based Electro-Optical Materials

9.2.8

353

Schiff-Bases as Linking Groups

The groups of Yamamoto, Samuelson, Ma and Das probed compounds in which acceptors such as functionalised benzene rings substituted with carborane, methoxy, chloro, nitro, dicyanomaleonitrile or anthrachinone moieties were linked to ferrocene through a Schiff base CDN bridge (Figures 9.12 and 9.13; Table 9.13 and 9.14).64–66, 77 Bimetallic complexes were investigated as well as charge transfer complexes. Complexes containing a carborane as the electron accepting group exhibited higher β values, when para-carborane (189) was used rather than ortho- or meta-carborane analogues (187 and 188).64 Even though carboranes acted as accepting group, their electron withdrawing ability was not as high as that of a nitro function. A series of aromatic Schiff bases were attached to ferrocene to form mononuclear (183–186) or dinuclear complexes (192–196 in Figure 9.12 and Table 9.13).65 Some of those were subjected to the formation of charge transfer complexes with acceptor molecules such as iodine (I2 ), p-chloranil (CA), 2,3-dichloro-5,6-dicyano1,4-benzoquinone (DDQ), tetracyanoethylene (TCNE) and 7,7,8,8-tetracyanoquinodimethane (TCNQ), which caused a signiﬁcant increase in the β value of the mono as well as the bisferrocenyl compounds. The best results were achieved with p-chloranil (183ž(CA)2 , 192ž(CA)2 , 193ž(CA)2 , 194ž(CA)2 , 195ž(CA)3 , 196ž(CA)2 ). Furthermore, bisferrocenyl complexes displayed higher β values than their monoferrocene counterparts, the major distinction being the different extent of the π-conjugation, which may be responsible for the higher β values. Comparing mono and bisferrocenyl complexes with anthraquinone (190/191) and dicyanomaleonitrile (132/146) accepting groups, respectively, the anthraquinone derivatives exhibited higher β values.77 Ma et al. performed Kurtz-powder experiments on a series of NLOphores containing an imino group as part of the π-bridge (Figure 9.13, Table 9.14).66 The largest SHG response, i.e. 190ð urea, was found for 99, which possessed a nitro moiety as acceptor in ortho position to the π-bridge. There were three pairs of isomers that used the same accepting group, differing only in the orientation of the imino function (101/102, 103/104 and 197/198). In all cases, the isomer in which the nitrogen atom was proximal to the acceptor exhibited a higher SHG efﬁciency. 9.2.9

µ-Carbyne Diiron Complexes

The SHG activity of organometallic NLOphores containing the [Fe2 (η-C5 H5 )2 (CO)2 (µCO)(µ-C)]C acceptor moiety were investigated by Farrell, Hudson, Manning and coworkers (Figure 9.14, Table 9.15).85, 87, 88 When the linker included an aromatic ring, the β and β0 values of that compound were higher than for the corresponding complexes with pure polyenes as linkers (166–169). Within the group of aromatic linkers a benzene ring (212) was more effective than a thiophene (170) or furan ring (211).85 When a terthiophene linker was modiﬁed to form a dithienylbenzo[c]thiophene, the hyperpolarisability dropped from β D 650 ð 1030 esu to 344 ð 1030 esu (204 and 210, respectively).88 For a variety of complexes with thiophene containing linkers β values ranging from 344 ð 1030 esu (210) to 1424 ð 1030 esu (203) could be determined.85, 87, 88 A general trend seemed to be that if there was more than one thiophene ring in a bridge, it was favourable to ﬁx them in a ‘linear’ and planar arrangement. Furthermore, an increase in the number of thiophene rings was accompanied by a decline of SHG activity.

354

Ferrocenes: Ligands, Materials and Biomolecules

N

Terminal Group H

H

Fe

Compound

Bridge

N

N H Fe

Fe

Terminal Group

Compound

Bridge O

18364,65

19177 O

18465

OCH3

14677

18565

Cl

19265

NC

CN

OCH3 18665

NO2

18764

19365

19465 Cl

= CH

18864

H3C

CH3

H3C

CH3

19565 = CH

18964

19665 = CH

O

19077 O

NC 13277

CN NH2

Figure 9.12 Compounds with imino groups as part of the π -bridge (I)

Ferrocene-Based Electro-Optical Materials Y X

Fe

355

Terminal Group

Compound

X

Y

Terminal Group

9966

N

CH

10066

N

CH

10166

N

CH

10266

CH

N

10366

N

CH

10466

CH

N

19766

N

CH

19866

CH

N

19966

CH

N

p-C6H4Br

20066

N

CH

(E)-CH=CHPh

20166

N

CH

o-C6H4OH

NO2

NO2 m-C6H4NO2

p-C6H4NO2

p-C6H4Cl

Figure 9.13 Compounds with imino groups as part of the π -bridge (II)

Table 9.13 Linear optical and quadratic nonlinear optical response parameters of compounds with imino groups as part of the π -bridge (part I) Compound

λmax /nm (ε/103 M1 cm1 )

β/1030 esu

β/1050 C m3 V2

β0 /1030 esu

β0 /1050 C m3 V2

13277 14677 18364 18365 183žI3 65 183ž(DDQ)2 65 183žTCNE65 183žTCNQ65 183ž(CA)2 65 18465 18565 18665

508 (–)a 587 (–)a 465 (1.200)a 448 (2.21)f 525 (2.62)f 565 (1.35)f 565 (3.34)f 470 (8.49)f 560 (5.04)f 457 (1.36)f 463 (0.93)f 490 (1.76)f

597b 717b 53d 13.8g 130.9g 132.2g 146.3g 53.5g 221.3g 11.9g 20.9g 51.2g

22.2 266 20 5.12 48.59 49.07 54.31 19.9 82.15 4.42 77.6 19.0

476c 576c 32e 3.3h 2.6h 12.1h 13.4h 9.4h 17.3h 2.5h 4.1h 6.1h

177 214 12 1.2 9.7 4.49 4.97 3.5 6.42 0.97 1.5 2.3 (continued overleaf )

356

Ferrocenes: Ligands, Materials and Biomolecules Table 9.13 (continued )

Compound

λmax /nm (ε/103 M1 cm1 )

β/1030 esu

β/1050 C m3 V2

β0 /1030 esu

β0 /1050 C m3 V2

18764 18864 18964 19077 19177 19265 192žI5 65 192ž(DDQ)2 65 192ž(TCNE)2 65 192ž(TCNQ)2 65 192ž(CA)2 65 19365 193žI4.5 65 193ž(DDQ)2 65 193ž(TCNE)2 65 193ž(TCNQ)2 65 193ž(CA)2 65 19465 194žI4.5 65 194ž(DDQ)4 65 194ž(TCNE)2 65 194ž(TCNQ)2 65 194ž(CA)2 65 19565 195žI3 65 195ž(DDQ)2 65 195ž(TCNE)2 65 195ž(TCNQ)2 65 195ž(CA)2 65 19665 196žI2 65 196ž(DDQ)2 65 196ž(TCNE)2 65 196ž(TCNQ)2 65 196ž(CA)2 65

469 (1.600)a 467 (1.700)a 467 (1.700)a 495 (–)a 498 (–)a 453 (1.06)f 514 (5.10)f 546 (5.16)f 498 (7.41)f 482 (2.54)f 634 (20.50)f 460 (9.30)f 510 (9.80)f 582 (3.61)f 479 (11.65)f 476 (17.96)f 570 (12.51)f 474 (2.39)f 521 (5.35)f 588 (3.63)f 490 (11.59)f 476 (3.39)f 601 (4.40)f 455 (3.75)f 502 (4.91)f 494 (5.34)f 478 (11.13)f 475 (15.56)f 438 (2.73)f 465 (4.71)f 524 (7.06)f 590 (7.71)f 505 (5.69)f 478 (13.58)f 448 (4.14)f

59d 60d 107d 939b 889b 18.3g 126.5g 173.5g 259.2g 165.9g 134.9g 22.9g 136.6g 149.9g 181.2g 143.6g 263.3g 33.1g 159.9g 175.0g 125.5g 204.0g 279.1g 22.7g 120.8g 152.6g 160.5g 170.7g 170.1g 35.9g 149.4g 131.8g 288.1g 172.5g 196.6g

22 22 39.7 34.9 330 6.79 45.96 64.40 96.21 61.58 50.07 8.50 50.71 55.64 67.26 53.30 97.74 12.3 59.35 64.96 46.59 75.72 103.6 8.43 44.84 56.64 59.58 633.6 63.14 13.3 55.46 48.92 106.9 64.03 72.98

37e 36e 64e 90c 45c 4.1h 6.5h 6.8h 25.1h 23.6h 36.6h 4.7h 8.5h 20.7h 27.3h 22.9h 27.8h 5.5h 4.9h 26.9h 33.6h 32.5h 52.5h 4.9h 10.3h 16.5h 24.7h 27.7h 45.5h 6.9h 3.4h 20.9h 22.1h 26.5h 47.0h

14 13 24 33 17 1.5 2.4 2.5 9.32 8.76 13.6 1.7 3.2 7.68 10.1 8.50 10.3 2.0 1.8 9.99 12.5 12.1 19.5 1.8 3.82 6.12 9.17 10.3 16.9 2.6 1.3 7.76 8.20 9.84 17.4

a Measured in CHCl . b Method: HRS, measured at 1064 nm in CHCl , concentration: 105 mol L1 , reference: 3 3 p-nitroaniline (β D 17.4 š 0.6 ð 1030 esu by the external reference method using p-NA in dioxane (β D 16.9 š

0.4 ð 1030 esu) as the external reference. c Calculated using the two-state model. d Method: HRS, measured at 1.06 µm in CHCl3 , reference: p-nitroaniline. e Determined by the Oudar-Chemla equation. f Measured in acetonitrile. g Method: HRS, measured at 1064 nm in acetonitrile, concentration: 105 to 106 mol L1 , reference: p-nitroaniline (β D 23 š 2.2 ð 1030 esu). h No calculation model given.

Ferrocene-Based Electro-Optical Materials

357

Table 9.14 Linear optical and quadratic nonlinear optical response parameters of compounds with imino groups as part of the π -bridge (part II) Compound

λmax /nm (ε/103 M1 cm1 ) Et2 O

λ max /nm (ε/103 M1 cm1 ) CH2 Cl2 a

λmax /nm (ε/103 M1 cm1 ) DMSO

9966

448 (–)

485 (4.48)

595 (–)

190

10066

437 (–)

481 (2.44)

599 (–)

95

10166

422 (–)

458 (2.72)

529 (–)

60

66

102

429 (–)

470 (2.56)

533 (–)

95

10366

431 (–)

454 (2.64)

537 (–)

75

10466

439 (–)

467 (2.72)

543 (–)

110

19766

423 (–)

465 (3.24)

538 (–)

25

19866

418 (–)

462 (2.34)

529 (–)

30

19966

421 (–)

465 (2.86)

532 (–)

20

20066

429 (–)

461 (2.92)

536 (–)

13

66

417(–)

458 (2.72)

495 (–)

13

201

SHG activity relative to ureab

a Concentration: 5 ð 105 mol L1 . b Method: Kurtz-powder, measured at 1907 nm, reference: urea, particle size:

<74 µm.

9.2.10

Inﬂuency of Stereochemistry and Symmetry

When SHG properties are concerned, symmetry is a key issue both on the molecular level and for bulk material. Usually, centrosymmetry renders SHG measurements impossible. Therefore, attempts have been made to circumvent this obstacle by preventing a centrosymmetric alignment (Figures 9.15 and 9.16; Tables 9.16 and 9.17; see also Chapter 12).40–42, 60, 61, 64, 66, 70, 89 In many cases symmetrically 1,10 -substituted ferrocenes adopt an orientation in anti -fashion that results in a zero molecular dipole moment. Chung et al. tried to overcome these difﬁculties by cocrystallisation of 1,10 bis(ethenyl-4-pyridyl)ferrocene (84) with resorcinol and phoroglucinol, respectively.61 Although the resulting complexes exhibited asymmetric structures on the molecular level veriﬁed by X-ray analysis, they were either packed in a centrosymmetric manner or possessed a pseudoinversion centre. Consequently, the net dipole moments of the bulk materials were close to zero. Only 1,10 -bis(ethenyl-4-quinolinyl)ferrocene (87) – even without cocrystallisation – exhibited an NLO active conformation and a SHG efﬁciency of about four times that of urea (Table 9.16). Cocrystallisation of 84 with (š)-1,10 -binaphthol in ethanol yielded crystals with a noncentrosymmetric arrangement that showed an SHG efﬁciency of 0.3 times that of urea.60 When optically pure ()-1,10 -binaphthol was used instead, the crystal structure of the resulting complex revealed that all the molecular dipoles were aligned in the same direction. Nevertheless, it exhibited only modest nonlinear optical properties (ca. 0.4 times that of urea).60 Another approach to avoid a centre of symmetry pursued by Cano et al. is

358

Ferrocenes: Ligands, Materials and Biomolecules O

C Fe

Bridge Fe

C O

Compound

Bridge

C O Fe

C

Compound

16685

(E)-CH=CH

20587

16785

(E,E)-(CH=CH)2

20687

Bridge S

S S

S 16885

(all E)-(CH=CH)3

S

20787

S S

16985

20887

S

S

20285

20987

S

S

17085

21088

(all E)-(CH=CH)4

Cl

S S

17185

21185

S 2

S

S

O

Cl 20385

21285 S

20488

S S

S

Figure 9.14 Compounds with acceptor groups based on a diiron complex

to change the substitution pattern of bridging aromatic rings from para to ortho as in compounds 21 and 106, respectively (Figure 9.15).40,42 Indeed, this led to a noncentrosymmetric space group (Pna21 ) for 21 but also to much smaller but still signiﬁcant nonzero χij(2) values when compared with the related para-ferrocenyl molecule (27).41 1,10 -Disubstituted ferrocene derivatives examined by the groups of Peris and Humphrey (112–114, 149, 150) differed in the orientation of the alkenyl group

Ferrocene-Based Electro-Optical Materials

359

Table 9.15 Linear optical and quadratic nonlinear optical response parameters of compounds with acceptor groups based on a diiron complex β/1030 esu

β/1050 C m3 V2

β0 /1030 esu

β0 /1050 C m3 V2

Compound

λmax /nm (ε/103 M1 cm1 )

16685

655 (10.953)a 634 (9.390)b

113c

41.9

36d

13

16785

734 (13.881)a 698 (10.029)b

156c

57.9

74d

27

16885

775 (18.277)a 717 (16.275)b

227c

84.3

120d

16985

813 (23.106)a 720 (14.954)b

562c

209

313d

116

17085

757 (19.966)a 688 (17.147)b

679c

252

343d

127

17185

764 (22.363)a 677 (19.199)b

1195c

443.6

615d

228

20285

746 (17.173)a 713 (13.986)b

459c

170

225d

20385

731 (24.000)a 644 (30.695)b

1424c

528.6

674d

20488

658 (28.550)a 619 (30.720)b

650e

241

20587

733 (10.500)a 670f (15600)b

429c

159

202d

20687

747 (19.400)a 680f (23.400)b

624c

232

307d

114

20787

750 (31.200)a 680f (35.350)b

867c

323

430d

160

20887

712f (10.600)a

625c

232

273d

101

d

259

87

a

c,g

209

830 (41.850) 769 (40.900)b

21088

720 (35.390)a 707 (38.350)b

344e

128

21185

747 (22.325)a 683 (21.202)b

352c

131

173d

21285

705 (12.162)a 616 (12.803)b

1106c

410.5

469d

1243

461.4

697

44.5

83.5 250

75.0

64.2 174

a Measured in CH Cl . b Measured in acetonitrile. c Method: HRS; measured at 1064 nm in CH Cl , reference: p2 2 2 2 nitroaniline (β D 21.6 ð 1030 esu). d Calculated using the two-level model. e Method: HRS, measured at 1300 nm, external reference method. f Estimated from US-vis spectra. g Fluorescence contribution.

360

Ferrocenes: Ligands, Materials and Biomolecules Compound

Substituent H3C

R N

2140,42

H3C

N N ON R NH Mo N N BH Cl N N

N

R = −C6H4-OCH3 H 3C

R R

N

2741,42

H3C

N N ON R NH Mo N N BH Cl N N

N

R = −C6H4-OCH3

R

H3C

10640,42

N

H3C N

NO2

21389

21489

21564

21664

21764 = CH 21864 = CH

21964 = CH

22064 = CH

Figure 9.15 Isomeric NLO systems of monosubstituted ferrocenes, Fc-Substituent

Ferrocene-Based Electro-Optical Materials Compound

361

Substituent

22164 = CH 22264 = CH N

O2N

9966

N

10066

NO2

10166

N

NO2

Figure 9.15 (continued )

Compound

8761

Configuration

Substituent Y

trans / trans N

8460

trans / trans

11270

trans / trans

11370

trans / cis

11470

cis / cis

14970

trans / trans

15070

trans / cis

N

p-C6H4NO2

p-C6H4CN

Figure 9.16 NLO systems of 1,10 -disubstituted ferrocenes, fc(CHDCHY)2

(Figure 9.16, Table 9.16).70 The all-trans complexes 112 and 149 exhibited only modest SHG responses (0.28 ð urea and 0.24 ð urea, respectively). The SHG efﬁciency of their mixed cis/trans isomers could not be determined unambiguously. Further evidence that ‘bent’ conjugated systems showed diminished SHG efﬁciencies compared to ‘linear’ systems was provided by Yamamoto et al.64, 89 The ferrocene

362

Ferrocenes: Ligands, Materials and Biomolecules

Table 9.16 Linear optical and quadratic nonlinear optical response parameters of isomeric NLO systems (part I) Compound

λmax /nm (ε/103 M1 cm1 )

Concentration 1019 cm3

2142 2140

484 (10.500)c

2741

550 (sh)f

10.5

(2) χ31 /109 esu

(2) χ15 /109 esu

(2) χ33 /109 esu

0.4a (0.04)b

0.4a

0.3a

0.4d,e

0.4d,e

0.3d,e

1.5d

0.4d

0.7d

SHG activity relative to urea

84 (š)-binaphthol60

0.3g

84 ()-binaphthol60

0.4g

8761

4h

9966

485 (4.48)i

190j

10066

481 (2.44)i

95j

10166

458 (2.72)i

60j

10640

348 (15.100)c

11270

494 (–)l 497 (–)m 495 (0.47)n

11370

493 (–)l 487 (–)m 488 (0.56)n

o,p

11470

481 (–)l 481 (0.30)n

o,p

14970

492 (0.34)n

15070

473 (–)l 473 (–)m 469 (0.45)n

15

1.4d,k

0.9d,k

0.5d,k 0.28o

0.24o o,p

a Spin-coated, corona-poled ﬁlms of the host polymer poly(methyl methacrylate) (PMMA) and the organometallic compound measured at 1064 nm. b Normalised values in brackets, i.e. χ (2) of the same molecular concentration. c Measured in CH Cl at ambient temperature and concentrations of ca. 104 – 105 mol L1 . d Spin-coated, corona2 2

poled ﬁlms of the host polymer poly(methyl methacrylate) (PMMA) and the organometallic compound measured at (2) D 1.96 ð 107 esu). e Film thickness: 1.8 µm. 1064 nm, Maker fringes technique, reference: χ -cut LiNbO3 plates (χ33

f Measured in CH Cl . g Method: Kurtz-powder, measured at 1295 nm, sample size: 75– 150 µm, thickness of the 2 2 sample: 270 µm. h Method: Kurtz-powder, measured at 1.36 µm, reference: urea. i Measured in CH2 Cl2 , concentration: 5 ð 105 mol L1 . j Method: Kurtz-powder, measured at 1907 nm, in CH2 Cl2 , reference: urea, particle size <74 µm. k Film thickness: 1.6 µm. l Measured in CHCl . m Measured in methanol. n Measured in THF. o Method: Kurtz-powder, 3 measured at 1.3 µm, reference: urea. p Not determined unambiguously.

Ferrocene-Based Electro-Optical Materials

363

Table 9.17 Linear optical and quadratic nonlinear optical response parameters of isomeric NLO systems (part II) Compound

λmax /nm (ε/103 M1 cm1 )

β/1030 esu

β/1050 C m3 V2

β0 /1030 esu

β0 /1050 C m3 V2

21389

374a

139

21489

508a

189

c

24

41d

15

38.2

62d

23

30

50d

19

42.3

66d

24

21

53d

20

41.2

65d

24

36

61d

23

d

29

64

215

21664

448 (0.500)

b

64

458

(1.100)b

21764

448

(0.900)b

81c

21864

463 (1.900)b

114c

21964

449

(0.800)b

84c

460

(1.600)b

111c

448

(0.800)b

98c

b

c

22064 22164 64

222

460 (1.700)

103c

131

48.6

77

a Method not stated. b Measured in CHCl at ambient temperature and concentrations of 104 mol L1 . c Method: HRS, 3 measured at 1.06 µm in CHCl3 , reference: p-nitroaniline. d Determined by the Oudar–Chemla equation.

derivatives considered (213–222 in Figure 9.15 and Table 9.17) possessed double bonds with cis or trans orientation. The determined β values for the trans compounds including acceptor groups such as carboranes or NO2 or no acceptor group at all were signiﬁcantly larger than for the complexes oriented in cis fashion.64 Much larger β values for such cis/trans systems were achieved when fullerene was chosen as the acceptor group (214, trans: β D 508 ð 1030 esu; 213, cis: β D 374 ð 1030 esu).89 Kurtz-powder experiments were carried out on complexes featuring an imino function as part of the π-bridge and a nitro group in ortho, meta, or para position relative to the π-bridge (Figure 9.15, Table 9.16).66 For all compounds, whether ‘linear’ or ‘bent’, their SHG response could be determined. The nitro group in para position (100) resulted in a higher SHG activity than in the meta position (101) while the ortho arrangement of 99 proved to be far more favourable (190 ð urea) than the para formation of 100 (95 ð urea).66 9.2.11

Fluorene and Related Compounds

The groups of Bryce, Perepichka and Khodorkovsky studied NLOphores with ﬂuorene derivatives as electron acceptors (Figure 9.17, Table 9.18).78 Using EFISHG measurements µβ values in the range of 410–5000 ð 1048 esu were determined, depending on the type of the linker and the kind of substituents on the ﬂuorene moiety. A comparison of 224 and 225 showed that the latter exhibited a comparatively low µβ value (410 ð 1048 esu) despite featuring an extended and trans-oriented π-system which usually is deemed to be favourable for a large NLO response. When the CO2 CH3 substituent of the ﬂuorene moiety of 227 was exchanged for a CO2 (CH2 CH2 O)3 CH3

364

Ferrocenes: Ligands, Materials and Biomolecules NO2 R

O2N

Bridge Fe

H

Compound 223

78

NO2

Bridge

Substituent R

– –C – – C–

CO2(CH2CH2O)CH3

22478 CO2(CH2CH2O)3CH3 22578 22678

CO2(CH2CH2O)3CH3 (E,E)-(CH=CH)2

22778

CO2CH3

Figure 9.17 Compounds with ﬂuorene-based acceptor groups

Table 9.18 Linear optical and quadratic nonlinear optical response parameters of compounds with ﬂuorene-based acceptor groups Compound

λmax /nm (ε/103 M1 cm1 )

µβ/1048 esu

µβ0 /1048 esu

22378

616 (4.570)a

700 š 100b,c

170 š 30c,d

22478

570 (sh)a

1300 š 130b,c

830 š 80c,d,e

22578

570 (sh, 4.470)a

410 š 35b,c

240 š 20c,d,f

22678

660 (15.100)a

2700 š 200b,c

470 š 50c,d,f

22778

663 (–)a

5000 š 1500b,c

900 š 300c,d,g

a Measured in 1,2-dichloroethane at 25 Ž C. b Method: EFISHG, measured at 1.54 µm in CH2 Cl2 , concentration: (1–10) ð 104 mol L1 . c µβ and µβ0 values are deﬁned according to the ‘traditional’ EFISHG deﬁnition. d Calculated using a two-level dispersion model. e µβ0 calculated using the energy of the main peak (430 nm) and not low-intensity, low-energy shoulder. f µβ0 calculated using the energy of the main peak (460 nm) and not low-intensity, low-energy shoulder. g The uncertainty in µβ and µβ0 for this molecule is relatively large due to its low solubility.

moiety (226) the result was a decrease in the µβ value, which is probably due to an aggregation in antiparallel fashion resulting from intermolecular donor acceptor interaction.78 Structurally related to the above mentioned ﬂuorene based moieties are derivatives of indane-1,3-dione (Figure 9.18, Table 9.19).71 Contrary to the overall general

Ferrocene-Based Electro-Optical Materials

365

O O

Bridge Fe

H

Compound 176

Bridge

71

none

22871∗

none

17771

(E)-CH=CH

22971

S

23071

S

S

∗

The ferrocene moiety is nona-methylated, Me9Fc–

Figure 9.18 Compounds with acceptor groups based on indane-1,3-dione Table 9.19 Linear optical and quadratic nonlinear optical response parameters of compounds with acceptor groups based on indane-1,3-dione β/1030 esu

β0 /1030 esu

β0 /1050 C m3 V2

83.1

26d

9.7

67.6

25d

9.3

84e

31

24d

8.9

563 (9.000)a 582 (6.800)b

336e 305c

125 113

29d 42d

11 16

568 (5.600)a 562 (22.000)b

672e

249

67d

25

Compound

λmax /nm (ε/103 M1 cm1 )

17671

565 (5.100)a 574 (5.000)b

224c

17771

581 (10.000)a 593 (10.700)b

182e

22871

639 (6.400)a 650 (6.200)b

22971 23071

β/1050 C m3 V2

a Measured in acetone. b Measured in CHCl . c Method: HRS, measured at 1064 nm, in CHCl , reference: p-nitroaniline 3 3 (β D 23 ð 1030 esu). d Calculated from the two-level model. e Method: HRS, measured at 1064 nm, in acetone, reference: p-nitroaniline (β D 25.9 ð 1030 esu).

trend, methylation of the ferrocene moiety resulted in a decrease of the β value (224 ð 1030 esu and 84 ð 1030 esu for 176 and 228, respectively). Furthermore, an enlargement of the π-system by introducing a CDC double bond also caused a reduction of the SHG activity, i.e. 182 ð 1030 esu for 177. On the other hand, when a thienyl group was part of the π-bridge, the β value rose to 672 ð 1030 esu for 230.

366

9.2.12

Ferrocenes: Ligands, Materials and Biomolecules

Azulene Complexes

On D-π-A systems with an azulene moiety as the electron accepting group some research was carried out by Herrmann et al. and the groups of Farrell, Heck and Manning (Figure 9.19, Table 9.20).84, 90 Although no ﬂuorescence was detected for the ferrocene derivatives 231–233, for compounds 232 and 233 it was not possible to determine NLO activity due to somewhat higher absorption in the range of the frequency doubled light.90 Azulenylium (158, 160) and guaiazulenylium (162–164)

Bridge

Azulene Group

Fe Compound

Bridge

Azulene Group CH3

23190

(E)-CH=CH CH3 H3C

23290

N

CH3 H 3C

23390

N

15884

none

15984

(E)-CH=CH

16084

(E,E)-(CH=CH)2

16184

(all E)-(CH=CH)3

16284

none

163

84

(E)-CH=CH

164

84

(E,E)-(CH=CH)2

16584

+

H 3C + CH3

H 3C

H 3C

(all E)-(CH=CH)3

2+

H 3C 23484

none

H 3C Ru

CH3 CH3

Figure 9.19 Compounds with acceptor groups based on azulene

Ferrocene-Based Electro-Optical Materials

367

Table 9.20 Linear optical and quadratic nonlinear optical response parameters of compounds with acceptor groups based on azulene β/1030 esu

Compound

λmax /nm (ε/103 M1 cm1 )

15884

716 (10.200)a

15984

798

(17.300)a

–

866

(22.900)a

810b

16084 84

161

920 (35.900)

16284

724

16384 16484

a

782 (21.600)a

1539b,d

(35.400)a

1373b,d (360)e

16584

859 (43.300)a

β0 /1030 esu

121

145c

301

451c

β0 /1050 C m3 V2

1500 β333 /1050 C m3 V2

53.8

–

– 220b

827

(13.500)a

326b

β/1050 C m3 V2

167

–

81.7

101c

37.5

571.3

821c

30.5

509.7 (134)

770c

–

285 (49.0)

(132)e –

90

f

231

482 (2.090) 502 (2.870)g

9.6h

23290

475 (4.600)a 480 (4.118)g 639 (1.460)i

j

23390

740 (0.250)k 639 (0.728)g

j

23484

714 (0.052)a

326b

121

134c

49.7

a Measured in CH Cl . b Method: HRS, measured at 1064 nm in CH Cl , reference: p-nitroaniline. c Calculated 2 2 2 2 from the two-level model. d Enhanced due to two-photon absorption ﬂuorescence. e Obtained from high frequency demodulation at λ D 1300 nm. f Measured in hexane. g Measured in acetone. h Method: HRS, measured at 1500 nm 1500 D in CHCl3 , concentration: 0.33–2.66 ð 105 mol L1 , reference: 4-(N ,N -dimethylamino)cinnamic aldehyde (β333

13.0 ð 1050 Cm3 V2 ). i Measured in formamide. j It was not possible to determine the hyperpolarisability due to somewhat higher absorption in the range of the frequency doubled light. k Measured in ether.

substituted compounds are closely related differing only in the degree of alkylation of the cationic terminus. The additional alkyl groups of the guaiazulenylium moiety diminish the electron accepting capability and corresponding to that the SHG efﬁciency of these compounds decreased. An improved electron withdrawing capability was achieved by coordination of a CpRuC fragment to the guaiazulenylium moiety of 162 to form the dicationic dinuclear complex 234, which was accompanied by a higher β value (220 ð 1030 esu and 326 ð 1030 esu, respectively). While complexes with extended π-conjugation (162–164) exhibited an increase of SHG activity (β0 rose from 101 ð 1030 esu to 770 ð 1030 esu) they also showed stronger two-photon absorption ﬂuorescence.84 9.2.13

Heterocyclic Compounds

Ferrocene was also linked to extended heterocyclic compounds such as corrole (235–247) and phthalocyanine (248–251) conjugates (Figures 9.20 and 9.21;

368

Ferrocenes: Ligands, Materials and Biomolecules Ar Bridge Fe

Compound

N N H ML n NH H N

Bridge

23591

O Ar MLn

Aromatic Ring Ar

none

not specified

91

p-C6H4Mes

23791

p-C6H4CH3

23891

p-C6H4-t-Bu

23991

p-C6H4-i-Pr

236

24091

p-C6H4Mes

none

24191

p-C6H4CH3

24291

p-C6H4-t-Bu

24391

p-C6H4Mes

24491

p-C6H4CH3

24591

p-C6H4-t-Bu

24691

p-C6H4CH3

Rh(CO)2

24791

p-C6H4-t-Bu

Figure 9.20 Compounds containing a large heterocyclic ring as acceptor group (I)

Compound

Ferrocene Derivative

24892 24992

Zn

N

N

Co

N

M

N

N

N

FcR

25092 25192

t-Bu

R= Metal M

Zn fcR2

N

N t-Bu

Co t-Bu

Figure 9.21 Compounds containing large heterocyclic rings as acceptor groups (II)

Ferrocene-Based Electro-Optical Materials

369

Table 9.21 Linear optical and quadratic nonlinear optical response parameters of compounds containing large heterocyclic rings as the acceptor groups Compound

λmax /nm (ε/103 M1 cm1 )

23591

a

23691 23791 91

238

23991

β/1030 esu

β/1050 C m3 V2

32.8b

12.2

700

(12.1)c

16.2b

6.01

700

(13.4)c

27.2b

10.1

c

b

10.6

700 (13.9)

28.5

700

(13.1)c

24091

703

(19.6)c

20.4b

7.57

24191

703 (14.7)c

19.9b

7.39

24291

703

(15.1)c

20.2b

7.50

704

(15.5)c

20.2b

7.50

704

(17.3)c

32.9b

12.2

c

b

11.0

24391 24491 91

245

704 (16.2)

26.8b

9.95

29.5

β/1030 esu

β0 /1050 C m3 V2

β/1030 esu

12.6

20.3f

7.54

5.57

37.9f

14.1

5.64

6.6f

2.5

10.8

10.8f

4.01

24691

637

(15.2)c

19.8b

24791

637 (14.0)c

31.8b

11.8

24892

119d .1

44.2

33.9e

24992

94.5d

35.1

15.0e

25092

88.8d

33.0

15.2e

43.1

29.2e

25192

116d

β0 /1050 C m3 V2

7.35

a No value reported. b Method: HRS, measured at 1064 nm in CH Cl , concentration: 105 to 106 mol L1 , reference: 2 2 p-nitroaniline (β D 18.7 ð 1030 esu). c Measured in CH2 Cl2 . d Method: HRS, measured at 1.06 µm in CHCl3 . e Method: EFISHG, measured at 1.06 µm in CHCl3 . f Method: EFISHG, measured at 1.9 µm in CHCl3 .

Table 9.21).91,92 Chandrashkar et al. altered the corrole–ferrocene system in three distinct ways: by varying the spacer group between the ferrocene and corrole moiety; by changing meso-alkyl substituents at the heterocyclic ring; and by coordinating different metal atoms (rhodium, iridium) to the heterocyclic unit. None of the above listed variations resulted in a consistent effect on the nonlinear responses. The reason given for this outcome was the nonplanar orientation of the corrole π-system, the ferrocene moiety and the spacer group. But compared to the structurally similar 18 π electron tetraphenylporphyrin derivatives,93 the 22 π electron possessing corrole ring showed a nonlinear response that was roughly 4–5 times larger.91,93 The combination of phthalocyanine and ferrocene building blocks led to compounds called dyads (248, 249) and triads (250, 251). These materials were investigated by Agull´o-L´opez, Torres et al.92 HRS measurements revealed that both dyads and triads presented similar dipole moments and quadratic responses. According to EFISHG

370

Ferrocenes: Ligands, Materials and Biomolecules

experiments the dipolar contribution to β was substantially lower. From the data presented no decisive trends were discernable concerning mono- or disubstituted ferrocenes (248, 249 and 250, 251, respectively) on the one hand side and phthalocyanine coordinated zinc (248, 250) or cobalt (249, 251) atoms on the other hand.92 Compounds containing far smaller heterocyclic units (252–256) were investigated by Toma et al. (Figure 9.22, Table 9.22).71 Although different solvents had to be used in order to determine the SHG response, it could be concluded that the introduction of a thiophene ring led to a higher increase in the SHG activity than the introduction of a CDC double bond.71

O

H N

Fe

25271 H N

S

N Fe

H

NH2

H 25371

25471 N

S

NH Fe

S

S

H

O

N

O

S Fe

S

O

S

S Fe

H

25671

25571

Figure 9.22 Compounds with small heterocyclic rings as acceptor group

Table 9.22 Linear optical and quadratic nonlinear optical response parameters of compounds with small heterocyclic rings as acceptor group β/1030 esu

β/1050 C m3 V2

β0 /1030 esu

β0 /1050 C m3 V2

530 (7.000)a 542 (3.200)b

173c

64.2

5d

2

25371

538 (5.100)a 545 (5200)b

135c

50.1

4d

1

25471

537 (4.300)a 553 (5.500)b

e

25571

508 (2.000)a 523 (3.000)b

72f

27

49d

18

25671

480 (3.400)a 488 (2.600)b

65f

24

10d

Compound

λmax /nm (ε/103 M1 cm1 )

25271

3.7

a Measured in acetone. b Measured in CHCl . c Method: HRS, measured at 1064 nm, in CHCl , reference: p-nitroaniline 3 3 (β D 23 ð 1030 esu). d Calculated from the two-level model. e SHG signal too low to be detected. f Method: HRS, measured at 1064 nm, in acetone, reference: p-nitroaniline (β D 25.9 ð 1030 esu).

Ferrocene-Based Electro-Optical Materials

9.2.14

371

Triarylcarbenium Compounds

Triarylmethanol derivatives with up to three donors, at least one of them being ferrocene, were studied by Screttas et al. (Figure 9.23).94 The triarylmethanol compounds (257–260) themselves showed only poor SHG activity but upon formation of carbenium ions in acid solution (257C –260C , including 261C , 262C ) the β values increased

Bridge

Substituent 2

Linker

Fe

Substituent 1 Compound

Bridge

Linker

Substituent 1

OH

25794 none

H

NMe2

+

257+ 94

OH

25894 none 258+ 94

+

Ph

NMe2

OH

25994 none 259+ 94

Substituent 2

NMe2

+

NMe2

OH 26094 260+ 94 26194 261+ 94

26294 262+ 94

+

1-naphthyl

Ph

OH

+

Fc

Ph

OH

+

Fc

Fc

Figure 9.23 Compounds with a sp3 -carbon/carbenium atom as part of the ‘π -bridge’

372

Ferrocenes: Ligands, Materials and Biomolecules Table 9.23 Linear optical and quadratic nonlinear optical response parameters of compounds with sp3 -carbon/carbenium or silicon atom(s) as part of the ‘π -bridge’ (part I) Compound

λmax /nm (ε/103 M1 cm1 )

25794

256 (49.500)a

257C 94

432

25894

357 (41.225)a

C 94

258

25994

(19.398)c

714 (4.276)

c

356

(59.200)a

259C 94

775

(3.075)c

26094

463 (1.920)a

260C 94 94

261

804

(2.340)c

461 (2.460)

a

β/1030 esu 18 š 1b 200

6.7 š 0.4

š 10d,e

10 š 1b 320 š 15 260

š 1b š 20d,g

55 š 10b 400

š 200d,h

119 š 5.6 3.7 š 4 96.5 š 7.4 20 š 3.7 148 š 74.2

–

261C 94

744

26294

459 (5.292)a

–

262C 94

1100 (7.437)c

1700 š 250d,j

(6.430)c

74.2 š 3.7 3.7 š 4

d,f

10

β/1050 C m3 V2

900 š 150d,i

334 š 55.7 631 š 92.8

a Measured in CHCl . b Method: HRS (harmonic light scattering (HLS) by Terhune and Maker), measured 3 at 1064 nm in CHCl3 , concentration: 0.01 mol L1 , reference: N -4-nitrophenyl-prolinol (NPP) powder. c Measured in CHCl :CF COOH D 114. d Method: HRS (harmonic light scattering (HLS) by Terhune 3 3 and Maker), measured at 1064 µm in CF2 HCOOH, reference: N -4-nitrophenyl-prolinol (NPP) powe der. Concentration: 6.86 ð 105 mol L . f Concentration: 5.81 ð 105 mol L1 . g Concentration: 3.94 ð 105 mol L1 . h Concentration: 5.38 ð 105 mol L1 . i Concentration: 3.22 ð 105 mol L1 . j Concentration: 2.64 ð 105 mol L1 .

up to 32 times, the highest value being β D 1700 ð 1030 esu for 262C (Table 9.23). A rise in β values was also observed on going from the mono (260C ) to di (261C ) and triferrocenyl carbenium compound (262C ), clearly demonstrating the superior electron donating ability of ferrocene compared to the also used donor of 4-dimethylamino-40 stilbenyl (257C –259C ). 9.2.15

Silicon-Containing Bridges

Regarding complexes featuring one or more silicon atoms in the linker between donor and acceptor, research was carried out by the groups of St¨uger, Pannell and Zyss/Zanello (Figure 9.24; Table 9.24).82, 95, 96 It turned out that the hyperpolarisability of the bis-silyl substituted ferrocenylene 275 was about eight times greater than the related mono-silyl ferrocene complex 274 (β0 D 57.0 ð 1030 esu and 7.0 ð 1030 esu, respectively). The nonlinear activity of compounds containing silicon atoms was generally similar to those of the corresponding substituted benzenes, exceptions being the above mentioned complexes 275 and 274 with the most electron withdrawing substituents (p-CHDC(CN)2 ).95, 96 The cyclohexasilane derivatives 276 and 277 were subjected to EFISHG measurements.82 It was claimed that, due to an increased donor capacity, 276 and 277 possessed higher β values compared to their open chain analogues.

Ferrocene-Based Electro-Optical Materials

Fe

CH3 Terminal Group Si CH3 n

Compound

n

26396

1

26496

1

26596

2

26696

3

Fe

Terminal Group CH3

373

NC

CH3 Si CH3 CH3 Si CH3

CN NC CN

27595

Ph

Fe

26796

4

26896

5

26996

6

27096

2

m-C6H4CF3

27196

2

p-C6H4Cl

27296

2

p-C6H4OCH3

27396

2

p-C6H4NMe2

27496

2

–CH=C(CN)2

Si Si Si Si Si Si

CN NC

27682

CN

Fe

27782

H2N

N O Fe

Si O O

27897

Fe

CN

Si Si Si Si Si Si

O Si N OO 27997

Fe

O Si N OO

28097

Figure 9.24 Compounds with silicon atom(s) as part of the ‘π -bridge’

Ferrocenylsilatranes (278–280) that were examined by Herrmann et al. also belong into the category of silicon containing compounds (Figure 9.24, Table 9.23).97 The silatrane moieties acted as stronger donors than the ferrocene unit in the same molecule. Their SHG activity was determined by hyper-Rayleigh scattering and was found to range between β333 D 13 ð 1030 esu and 23 ð 1030 esu. The compounds considered exhibited an unconventional behaviour in so far that the addition of a double bond did not increase the hyperpolarisability. 9.2.16

Fullerenes

Yamamoto et al. probed systems using [60]fullerene as the accepting group (Figure 9.25, Table 9.25).89 Their results were in accordance with generally accepted trends, i.e. the more extended the π-conjugation the higher the NLO response and trans-oriented double bonds produced better results than cis-oriented double bonds.

374

Ferrocenes: Ligands, Materials and Biomolecules Table 9.24 Linear optical and quadratic nonlinear optical response parameters of compounds with (a) silicon atom(s) as part of the ‘‘π -bridge’’ (part II) Compound

λmax /nm (ε/103 M1 cm1 )

26396

448 (–)a

1.26b

26496

456 (–)a

0.5b

0.2

26596

455

(–)a

0.6b

0.2

457

(–)a

0.4b

0.1

a

b

6.4

2.4

1.76b

0.653

26696 96

β0 /1030 esu

β0 /1050 C m3 V2

267

456 (–)

26896

456 (–)a

26996

456

(–)a

1.0b

0.37

457

(–)a

2.2b

0.82

456

(–)a

1.6b

0.59

0.26b

0.097

27096 27196 27296

456

(–)a

27396

455

(–)a

3.5b

1.3

27496

342 (–)a

7.0b

2.6

27595

(–)c

57.0d

21.2

436

27682

(17.500)e

368 364 (–)f 363 (–)g

h

27782

367 (17.300)e 363 (–)f 364 (–)g

h

27897

446 (–)g

13

4.8i

27997

446 (–)g

23

8.4i

28097

400 (–)a

23

8.5j

a Measured

in CHCl3 . b Method: Maker fringes, measured in acetone, concentration: 2 3 10 –10 mol L1 , β0 calculated using a two-level model. c Measured in CH2 Cl2 . d Method: EFISHG. e Measured in cyclohexane, concentration: 2 ð 105 mol L1 . f Measured in acetonitrile. g Measured in CH2 Cl2 . h In EFISHG studies the value of the ﬁrst hyperpolarisability was measured to be up to three times that of the open chain analogue. i β333 value, method: HRS, measured at 1300 nm, CH2 Cl2 , reference: 4-(N ,N -dimethylamino)cinnamic aldehyde (β333 D 15.2 ð 1050 C m3 V2 ). j β333 value, method: HRS, measured at 1300 nm, CHCl3 , reference: 4-(N ,N -dimethylamino)cinnamic aldehyde (β333 D 15.2 ð 1050 C m3 V2 ).

9.2.17

Allenylidene and Related Compounds

Ferrocenyl derivatives with electron accepting groups based on allenylidene and carbyne complexes were investigated by the groups of Fischer98 and Long99 (Figure 9.26, Table 9.26). The SHG efﬁciency of the allenylidene compounds 282–284 followed the general trend that an enlargement of the π-system should result in higher β values.98

Ferrocene-Based Electro-Optical Materials

Fe

Bridge

Compound 281

375

Bridge – –C–C– –

89

21389

21489

Figure 9.25 Compounds with fullerence as acceptor group Table 9.25 Linear optical and quadratic nonlinear optical response parameters of compounds containing fullerene as acceptor group Compound

λmax /nm (ε/103 M1 cm1 )

β/1030 esu

β/1050 C m3 V2

β0 /1030 esu

β0 /1050 C m3 V2

431 (4.900)a 432 (6.200)a 432 (2.600)a

374b 568b 173b

139 211 64.2

213c 317c 97c

79.1 118 36

21389 21489 28189

a Measured in CHCl at ambient temperature. b Method: HRS, measured at 1.06 µm in CHCl . c Calculated using the 3 3 Oudar–Chemla equation.

Bridge Fe

H3C H

N C C Cr(CO)5

Fe

NMe2 ML

Compound

Bridge

Compound

28298

– –C–C– –

28599

fac-Mn(dppm)(CO)3

28699

Ru(dppm)2Cl

28799

Os(dppm)2Cl

28398

28498

MLn

2

Figure 9.26 Compounds with allenylidene and ethynyl complexes as acceptor groups

The Kurtz-powder experiments (measured at 1064 nm with urea as reference) on the chiral carbyne complexes 285–287 revealed only negligible SHG responses.99 The chiral substituent had been introduced in an attempt to avoid a centrosymmetric arrangement in the solid state.

376

Ferrocenes: Ligands, Materials and Biomolecules

Table 9.26 Linear optical and quadratic nonlinear optical response parameters of compounds with allenylidene and carbyne complexes as acceptor groups Compound

λmax /nm

β/1030 esu

β/1050

(ε/103 M1 cm1 )

β0 /1050

C m3 V2

28298

519 (4.424)a

0b

28398

520 (4.421)a

20b

28498

(4.529)a

73b

522

β0 /1030 esu

C m3 V2 0c

7.4 27

9c

3

33c

12

a Measured in CH Cl . b Method: HRS, measured at 1500 nm, in CH Cl , reference: Disperse Red 1 (β D 70 ð 2 2 2 2 1030 esu), error: š10 %. c Value: š10 %.

9.2.18

‘Through-Space’ SHG Chromophores

Heck et al. studied whether cofacially ﬁxed sandwich complexes exhibit second harmonic generation.100 The compounds shown in Figure 9.27 were subjected to hyperRayleigh scattering (HRS) as well as to Kurtz-powder measurements. Although cyclic voltametry studies suggest ‘through space’ interaction between both metallocene moieties, no SHG intensity was observed. 9.2.19

Miscellaneous Compounds

Figure 9.28 and Table 9.27 highlight compounds which could not be dealt in one of the above mentioned categories.54, 55, 70, 101 The bis(ferrocenylalkenyl) substituted benzene ring (295) only exhibited observable SHG activity when it was coordinated +

LnM

Fe

PF6−

LnM = Cp∗Fe (288), Cp∗Ru (289), Cp∗Ir (290), (h6-C6H6)Ru (291), CpFe (292)

Figure 9.27 Cofacially ﬁxed sandwich complexes100

R H

Fe H R=H

Fe

Fe

O

(293)54

Fe 29570

Cl (294)54,55

Figure 9.28 Miscellaneous compounds

Cr(CO)3 29670

Fe

Ferrocene-Based Electro-Optical Materials

377

Table 9.27 Linear optical and quadratic nonlinear optical response parameters of miscellaneous compounds Compound

λ max /nm (ε/103 M1 cm1 )

29354

474 ( – )a

β/1030 esu

54,55

12b 3.4

29570

(0.45)c

457

29670

455 (0.31)c

SHG activity relative to urea

SHG activity relative to KH2 PO4 (KDP)

4.5 b

294

β/1050 C m3 V2

1.3 d,e

0.14d

297101

0.27f

298101

0.06f

299101

0.10f

a Solvent not stated. b Method: EFISHG. c Measured in THF. d Method: Kurtz-powder, measured at 1.3 µm, reference: urea. e Not determined unambiguously. f Method: EFISHG, measured at 1064 nm, reference: KDP.

by a Cr(CO)3 moiety (296) resulting in a SHG efﬁciency of 0.14 ð urea.70 Zhang et al. investigated charge-transfer complexes based on ferrocene and a Keggin structure (Hn XW12 O40 žmH2 O, X D P, Si, Ge). Their SHG efﬁciencies (I2ω ) were measured to be 0.27 ð IKDP for 297 ([FeCp2 H]3 PW12 O40 ) 0.06 ð IKDP for 298 ([FeCp2 H]4 SiW12 O40 ) and 0.10 ð IKDP for 299 ([FeCp2 H]4 GeW12 O40 ).101 9.2.20

Octupolar Compounds

It has been mentioned above that a major characteristic/precondition of SHG active compounds is a large difference between the dipole moments of the ground and excited state,8 the former not being equal to zero. Compounds having an octupolar symmetry (D3 , D3h , Td , and D2 ) exhibit nonzero quadratic hyperpolarisability β in spite of the absence of any ground state dipole moment.102–104 For such molecules the vector part βJD1 of β is cancelled out and only the octupolar contribution β JD3 adds to the β tensor. Octupolar compounds offer some advantages. They are more likely to crystallise in a noncentrosymmetric way, are expected to offer an improved nonlinearity/transparency trade-off, as well as giving an extension of the dimensionality of the NLO response.102, 103, 105, 185 Riant et al. designed chiral octupoles containing chiral ferrocenyl units as donors (Figure 9.29, Table 9.28).76 Compounds 300, 302 and 303 were two-dimensional octupoles, maintaining a noncentrosymmetric arrangement. The presence of a true octupole required electronic communication between the three ferrocene units. As a consequence, the SHG efﬁciency of the whole molecule had to be compared with the corresponding single ferrocene unit (β 0 of 301, 129 and 128, respectively). A deeper analysis of the experimental data showed that because the β0 values of the octupolar compounds 300, 302 and 303 were at least equal, if not higher

378

Ferrocenes: Ligands, Materials and Biomolecules

A

A

S

Fe

Compound

Accepting Group A

30076

S Fe

O

O

Fe

30276

NC

12976*

CN CN

NC

30376

A

S N

30176 *

S

N

CN CN CN

12876*

Ph

* Not an octupolar system but ferrocene derivative with single attached bridgeaccepting group unit.

Figure 9.29 Chiral octupolar compounds Table 9.28 Linear optical and quadratic nonlinear optical response parameters of chiral octupoles Compound 30076 30176,d 30276

λmax /nm (ε/103 M1 cm1 )

β/1030 esu

608 (40)a

350b

616

(5.4)a

120b 500b

712

(27)a

12976,d

724

(9.5)a

140b

30376

658 (39)a

500b

12876,d

663 (8.2)a

180b

β/1050 C m3 V2

130 44.5 186 52.0 186 66.8

β0 /1030 esu

β0 /1050 C m3 V2

187c

69.4

63c

23

190c

71

51c 231c 82c

19 85.7 30

a Measured in CHCl . b Method: Harmonic Light Scattering (HLS), measured at 1.91 µm in CHCl in various concen3 3 trations, reference: N -4-nitrophenyl-prolinol (NPP). c Calculated from the two-level model. d Not an octupolar system but the single ferrocene-thienyl bridge-accepting group unit.

than 3 ð β 0 , this result was a clear indication of a signiﬁcant electronic constructive interaction between the three ferrocene units through the benzene ring.76

9.3

THG Active Compounds

A variety of compounds possessing third order nonlinear optical properties has been developed and examined in recent years (Figure 9.30, Table 9.29).42, 70, 106–128 Unfortunately, their high diversity prevented an intended division into suitable subgroups and the various measuring conditions thwarted attempts to compare results. Therefore,

Ferrocene-Based Electro-Optical Materials O H N

Fe

N

N N

O H N

Fe

304107 O H N Fe H N O 305107

Fe N

309108 (X = −) 310108 (X = C6H4)

N M(OAc)2

N Fe

N N N

N M(OAc)2

O

2

311109,110

307107

(M = Hg) 308107 (M = Cd)

R

R N H

N R

(Cp2Fe)2C60 313112,113

R

N N

Fe

N

Fe

H H

n

H

H

Fe

320118

321118

R

N S

Bu n

323119,120

n

n

Fe 322118

Fc Fc

O

O

H m

m

Fe

Fe

Bu

H

n

Fe

319-n (n = 5, 10, 15, 20, 25)118

R 317115 (R = p-C6H4-OMe)

Ph

H

n

Ph

R

O

315115 (X = HCl) 316115 (X = −)

C

Ph

R

NH

314115 (X = Rh(CO)2)

318117 H

Fe

R = p-C6H4-OMe

N Pr

N HN

X O NH H N–

312111 (R = –CH=CH-Fc)

Fe

Fe

N N

O

O H N H N O

N

Fe

2

306107

X

X

N Zn(NO3)

N

N M

S

S

N N

R

Fe

N Fe

S

324118 (M = −, R = H) 325122 (M = Ni, R = H) 326121,122 (M = Pd, R = H) 327122 (M = −, R = H) 328123 (M = −, R = Ph) 329123 (M = Ni, R = Ph) 330123 (M = Pd, R = Ph) 331123 (M = −, R = Ph)

L 332124 [Ag(L2)](NO3)2 333124 [HgI2(L2)] 334125 [WS4(Cudppf)2] 335126 [MoS4(Cudppf)2] 336127 [HgI2(dppf)]

Figure 9.30 THG active compounds

379

Method

THG

THG

THG

THG

THG

THG

THG

THG

Z -scand

Z -scand

Compound

2142,106

2342,106

2442,106

2542,106

2642,106

2742,106

10542,106

10642,106

11270

11370

800 nm

800 nm

1064 nm 1907 nm

1064 nm 1907 nm

1064 nm 1907 nm

1064 nm 1907 nm

1064 nm 1907 nm

1064 nm 1907 nm

1064 nm 1907 nm

1064 nm 1907 nm

λ

9.92(0.5) ð 1014 esua 30.2(1.7) ð 1014 esub 40.3(5.7) ð 1014 esua 89.6(12.6) ð 1014 esub 27.3(3.0) ð 1014 esua 50.4(5.6) ð 1014 esub 31.0(4.3) ð 1014 esua 64.4(8.8) ð 1014 esub 24.8(3.6) ð 1014 esua 50.4(7.3) ð 1014 esub 20.2(3.4) ð 1014 esua 7.8(1.3) ð 1014 esub 18.6(1.2) ð 1014 esua 14.6(0.97) ð 1014 esub

18 ð 1019 cm1

7.1 ð 1019 cm1

9 ð 1019 cm1 7.3 ð 1019 cm1

6.9 ð 1019 cm1

6 ð 1019 cm1

15 ð 1019 cm1

Susceptibility χ (3) 15.2(1.4) ð 1014 esua 13.4(1.25) ð 1014 esub

Nonlinear refractive index n2

10.7 ð 1019 cm1

Molecular concentration

Table 9.29 Linear and cubic nonlinear optical response parameters of THG active compounds

γr 840 š 400 ð 1036 γi 770 š 200 ð 1036 jγ j1140 š 430 ð 1036

Insufﬁciently soluble

γr 6.05 ð 1034 c γi 8.60 ð 1034 jγ j10.5 ð 1034

γr 4.28 ð 1034 c γi 57.7 ð 1034 jγ j57.9 ð 1034

γr 4.67 ð 1034 c γi 13.2 ð 1034 jγ j14 ð 1034

γr 34.5 ð 1034 c γi 32.1 ð 1034 jγ j47.1 ð 1034

γr 4.50 ð 1034 c γi 4.60 ð 1034 jγ j6.40 ð 1034

γr 1.80 ð 1034 b γi 4.10 ð 1034 jγ j4.48 ð 1034

Hyperpolarisability γ /esu

380 Ferrocenes: Ligands, Materials and Biomolecules

Z -scand

Z -scand

Z -scand

Z -scand

Z -scan

Z -scan

Z -scan

Z -scan

Z -scan

Z -scan

Z -scan

Z -scan

14970

15070

29570

29670

297101

298101

299101

304107

305107

306107

107

308107

THG

THGf

310108

309

f

108

307

Z -scand

11470

W1

mol dm

m W

1

esu

0.85 š 0.3 ð 1034 0.76 š 0.1 ð 1034

1.5 ð 103 mol L1 1.1 ð 103 mol L1

1.91 µm

(continued overleaf )

0.9 š 0.35 ð 1034 0.8 š 0.07 ð 1034

3.10 ð 1028

3.12 ð 1028

2.46 ð 1028

1.53 ð 1028

7.8 ð 10 mol L 8.9 ð 103 mol L1

4.20 ð 1011 esu

3.72 ð 10

11

5.43 ð 1011 esu

esu

1.51 ð 1028

1.91 µm

1.72 ð 1017 2 W1

1.53 ð 10 2

6.51 ð

1011

8.97 ð 1011 esu

1

7.65 ð 104 mol dm1 e

6.45 ð 10

1 e

17

m2

2.22 ð 1017 m2 W1

2.67 ð

1017

3.67 ð 1017 m2 W1

4

mol dm1 e

1.22 ð 104 mol dm1 e

2.35

ð 104

3.27 ð 104 mol dm1 e

6.5 ð 1012 esu

3.1 ð 1012 esu

2.4 ð 1013 esu

γr 850 š 300 ð 1036 γi 95 š 30 ð 1036 jγ j860 š 300 ð 1036

γr 640 š 300 ð 1036 γi 30 š 20 ð 1036 jγ j640 š 300 ð 1036

γr 310 š 200 ð 1036 γi 50 š 30 ð 1036 jγ j310 š 200 ð 1036

γr 280 š 150 ð 1036 γi 30 š 30 ð 1036 jγ j280 š 150 ð 1036

γr 600 š 300 ð 1036 γi 0 š 50 ð 1036 jγ j600 š 300 ð 1036

3

532 nm

532 nm

532 nm

532 nm

532 nm

800 nm

800 nm

800 nm

800 nm

800 nm

Ferrocene-Based Electro-Optical Materials 381

Method

Z -scang

Compound

311109,110

χr 0.26 š 0.039 ð 1020 m2 V2 χi 2.0 š 0.3 ð 1020 m2 V2 χr 0.22 š 0.033 ð 1020 m2 V2 χi 3.4 š 0.6 ð 1020 m2 V2 χr 0.11 š 0.017 ð 1020 m2 V2 – χr 0.11 š 0.017 ð 1020 m2 V2 χi 1.9 š 0.3 ð 1020 m2 V2 χr 0.10 š 0.015 ð 1020 m2 V2 χi 0.05 š 0.26 ð 1020 m2 V2 χr 0.18 š 0.027 ð 1020 m2 V2 χi 0.16 š 0.024 ð 1020 m2 V2 χr 0.13 š 0.02 ð 1020 m2 V2 χi 0.24 š 0.33 ð 1020 m2 V2

4.5 š 0.68 ð 1020 cm2 /GW

4.0 š 0.6 ð 1020 cm2 /GW

1.9 š 0.29 ð 1020 cm2 /GW

2.0 š 0.3 ð 1020 cm2 /GW

1.7 š 0.26 ð 1020 cm2 /GW

3.1 š 0.47 ð 1020 cm2 /GW

2.2 š 0.33 ð 1020 cm2 /GW

540 nm

560 nm

580 nm

600 nm

620 nm

660 nm

Susceptibility χ (3)

520 nm

Nonlinear refractive index n2 χr 0.23 š 0.035 ð 1020 m2 V2 χi 2.8 š 0.42 ð 1020 m2 V2

500 nm

13 ð 103 mol L1

Molecular concentration 4.1 š 0.62 ð 1020 cm2 /GW

λ

Table 9.29 (continued ) Hyperpolarisability γ /esu

382 Ferrocenes: Ligands, Materials and Biomolecules

Z -scang

Z -scani

Z -scani

Z -scanl

Z -scanl

Z -scanl

Z -scanl

THGm

DFWM

DFWM

DFWM

DFWM

DFWM

DFWM

DFWM

DFWM

312111,h

313112,113

313112,113

314115

315115

316115

317115

318117

319-5118

319-10118

319-15118

319-20118

319-25118

320118

321118

322118

1.05 µm 1.2 µm 1.5 µm 1.8 µm 1.95 µm

527.5 nmk

527.5 nmj

532 nm

5.83 ð 108 cm2 W1 FOM: 2.28 ð 108 cm2 W1 4.61 ð 108 cm2 W1 FOM: 1.66 ð 108 cm2 W1 2.77 ð 108 cm2 W1 FOM: 1.05 ð 108 cm2 W1 0.43 ð 108 cm2 W1 FOM: 0.19 ð 108 cm2 W1

¾104 mol L1 ¾104 mol L1 ¾104 mol L1

0.60 ð 1018 m2 V2

1.61 ð 1018 m2 V2

1.26 ð 1018 m2 V2

1.62 ð 1020 m2 V2

3.05 ð 1020 m2 V2

2.92 ð 1020 m2 V2

2.15 ð 1020 m2 V2

1.6 ð 1018 m2 V2

9.43 ð 1012 esu 12.0 ð 1012 esu 19.8 ð 1012 esu 22.3 ð 1012 esu 21.1 ð 1012 esu

χi 2.4 ð 1012 esu

2 g L1 ¾104 mol L1

χr 1.2 ð 1012 esu χi 7.7 ð 1013 esu

2 g L1

0.5 ð 103 mol L1

(continued overleaf )

γi 5.6 ð 1031

γr 2.4 ð 1032 γi 1.8 ð 1031

Ferrocene-Based Electro-Optical Materials 383

DFWMn

DFWMe,o

DFWMe,o

DFWMe,o

DFWMe,o

DFWMe,o

DFWMo

DFWMo

DFWMo

DFWMo

Z -scane

Z -scane

Z -scane

Z -scane

Z -scane

323119,120

324122

325122

326122

326121

327122

328123

329123

330123

331123

332124

333124

334125

335126

336127

532 nm

532 nm

532 nm

532 nm

308 nm

308 nm

308 nm

308 nm

450 nm

450 nm

450 nm

450 nm

450 nm

450 nm

λ

mol dm3

W1

W1

2.48 ð

mol dm3

6.86 ð 1018 m2 W1

m2

1.35 ð 1017 m2 W1

6.1 ð

1018

8.02 ð 1019 m2 W1

m2

104

mol dm3

1.15

ð 1018

Nonlinear refractive index n2

1.3 ð 104 mol dm3

1.2 ð

104

1.1 ð 104 mol dm3

8.5 ð

104

Molecular concentration

2.87 ð 1012 esu

4.11

1.44 ð 1030

2.68 ð 1030

4.52 ð 1035

0.35 ð 1012 esu esu

306 ð 1035 esu

ð 1012 ð 1012

371 ð 1035

1.60 ð 1012 esu 2.14

3.11 ð 1035

Hyperpolarisability γ /esu

0.13 ð 1012 esu

2.73 ð 1013 esu

4.5 ð 1012 esu

5.12 ð 1012 esu

3.14 ð 1012 esu

1.12 ð 1013 esu

1.24 ð 108 esu

Susceptibility χ (3)

a Spin coated thin ﬁlm on PMMA at 1064 nm. b Spin coated thin ﬁlm on PMMA at 1907 nm. c Measured in CHCl . d Measured in THF, referenced to the nonlinear refractive index 3 of silica n2 D 3 ð 1016 m2 W1 . e Measured in DMF. f Measured in CH2 Cl2 or CHCl3 ; published information inconsistent. g Measured in CHCl3 . h The following values were calculated from Z -scan measurements, indicating strong NLO effects: β D 1.7 ð 109.87 cm W1 , σ D 2.85 ð 1046 cm4 s photon1 . i Measured in toluene. j Linear polarisation. k Circular polarisation. l Measured in CH Cl . m Spin coated polyazine thin ﬁlm. n Spin coated thin ﬁlm (0.12 µm thick), reference: liquid CS (χ(3) D 6.8 ð 1013 esu). o Reference: 2 2 2 CS2 (χ (3) D 6.8 ð 1013 esu).

Method

Compound

Table 9.29 (continued )

384 Ferrocenes: Ligands, Materials and Biomolecules

Ferrocene-Based Electro-Optical Materials

385

in the following text only exemplary cases are presented. It has to be mentioned that two reports are not included in Table 9.29. The ﬁrst concerned two-photon absorption cross section (TPA) values σ (2) of ferrocene containing smaragdyrin derivatives.114 The second described a photorefractive effect of polymer layers doped with, e.g., ferrocene.116 For some molecules both SHG and THG experiments were carried out in an attempt to elucidate the role of structural parameters.42, 70, 91, 106, 107, 115 The groups of Agull´oL´opez and Cano examined the THG properties of the complexes 21, 23–27, 105 and 106, most of which contain an tris(pyrazolyl)borato molybdenum fragment as the acceptor (Figures 9.1 and 9.6; Tables 9.2, 9.7 and 9.29).106, 107 The extension of the bridging π-system and the substitution of the aza for the stilbene bridging group caused an increase of χ (2) as well as χ (3) . In contrast, while the substitution of the [Mo(Tp)(NO)(Cl)] moiety for the nitro group led to a marked increase in χ (2) , it did not inﬂuence χ (3) signiﬁcantly. Peris, Humphrey and coworkers discovered for their compounds, e.g. 113 and 150, that increasing the acceptor strength led to a parallel enhancement of quadratic and cubic nonlinearities (Figure 9.16; Tables 9.16 and 9.29).70 Audebert et al. reported on the ferrocenyl substituted tetrazines 309 and 310 (Figure 9.30, Table 9.29).108 In that case, the introduction of a ferrocene moiety resulted in reduced γ values, which were ascribed to a possibly detrimental effect of a metal-to-ligand charge transfer. A more benign inﬂuence of ferrocene on THG properties was found when attached to fullerene (313).112, 113 A strong dependence of γ on the polarisation of the incident light was also observed, circular polarisation yielding higher γ values. A number of ferrocene containing polymers was studied, too (319–323 in Figure 9.30, Table 9.29).117–120 Sundararajan et al. reported that for polymers with ferrocene units as pendants an increased conjugation length was favourable in respect to χ (3) . Up to a certain chain length the same trend was observed for polymers with ferrocene as end group (319-5, 319-10, 319-15, 319-20, 319-25).118 Ferrocene was also used as a functional group in ligands of organometallic complexes.121–123, 125–127 One of those ligands was S-methyl-N -(ferrocenyl-1methylmethylidene)dithiocarbazate that was coordinated by transition metal atoms (325–327, Figure 9.30, Table 9.29).122 Compared to the uncoordinated ligand the χ (3) value for the copper complex 327 was not obviously enhanced. In contrast, the χ (3) values for the other two complexes (325 and 326) were one order of magnitude higher than that of the ligand. The same trend was found for complexes containing the analogous benzyl substituted ligand.123 There have been a number of reports about compounds with potential NLO properties.82, 128–135 They have not been dealt with in detail in the text above and the list of references given is not exhaustive. As the evidence presented in this chapter shows ferrocene is still a moiety of choice for NLOphores. During the last decade the established structure–property relationships for SHG active compounds have been used to develop new compounds and have been checked and modiﬁed by new results. Such relationships are not yet fully understood for THG active complexes and research into octupolar systems containing ferrocene is still in its infancy. It is not exaggerated

386

Ferrocenes: Ligands, Materials and Biomolecules

to state that ferrocene will play a key role in the development of nonlinear active compounds in the foreseeable future.

9.4

Ferrocene in Material Science (except NLO)

In recent years, ferrocene has not only been used as a donor moiety in NLOphores, it has also been employed in scientiﬁc ﬁelds of research such as, e.g., self-assembled monolayers, thin ﬁlms and artiﬁcial photosynthesis as well as electronic devices, data storage and nano tubes. In most cases, ferrocene was chosen because of its redox properties. Those are strongly affected by the ferrocene environment and can be easily detected/measured by cyclic voltametry, thus making ferrocene an excellent sensor for changes of all sorts in a variety of materials. To cover all subjects in detail is not possible within the scope of this chapter. For this reason, they will be only touched on in the following and the quoted literature will not be comprehensive. 9.4.1

Self-Assembled Monolayers

Ferrocene containing SAMs can be used for a number of purposes.67, 136–160 For example, Uosaki et al. did some research on electrochemically controllable SHG active SAMs.67 When the ferrocene derivative was oxidised, its SHG intensity increased. Upon reduction, it receded again. The origin for this behaviour was attributed to the changes in orientation of the SAM and of the hyperpolarisability of the molecule. Another ﬁeld of application are SAMs of chromophores for photovoltaic devices.136–140 In these systems ferrocene is usually used as an electron donor to promote long range electron transfer and to stabilise charge separation. 9.4.2

Thin Films

Ferrocene containing compounds have a also been used for the preparation of thin ﬁlms.162–164 Some of the NLO active compounds that have been mentioned in this chapter were used for that purpose, too.56, 67, 68, 72, 74, 117, 119, 120, 145, 160 As an example, Clays et al. incorporated 109 into a specially designed cell because of the reversible electrochemical switching of its SHG properties. It was shown that the interconversion between 109 and 109C could be repeated many times.68 Another example might be that of Mallik et al., who reported that ferrocene doped poly(methyl methacrylate) (PMMA) thin ﬁlms containing chloroform molecules exhibited photoswitching properties that were measured by photoconductivity measurements. The results were discussed on the basis of the formation of photoinduced charge-transfer complexes of ferrocene with chloroform molecules conﬁned in a PMMA thin ﬁlm.164 9.4.3

Artiﬁcial Photosynthesis

Ferrocene also plays a role in the development of systems for artiﬁcial photosynthesis.136–140, 161, 164–171 A general design of such systems is based on acceptor–sensitiser–donor (A–S–D) triad molecules in which ferrocene is incorporated as a donor. Upon excitation of the sensitiser moiety, an electron transfer from S to A occurs.

Ferrocene-Based Electro-Optical Materials

387

This is charge separation is supported/stabilised by a second electron transfer from the ferrocene donor D to S, resulting in a long-lived charge-separated state Až –S–DžC . The ultimate goal for applied chemical research is the incorporation of its chemical substances into technical devices. The same is true for ferrocene derivatives.67, 68, 74, 89, 110, 137, 138, 140, 142, 149, 152, 155, 172–174 The group of Huskens and Reinhoudt performed research on molecular printboards. For this, ferrocenyl functionalised dendrimers were absorbed at self-assembled monolayers. The dendrimers formed supramolecular assemblies but could be efﬁciently removed from the host surface by electrochemical oxidation of the ferrocene end groups.142 M¨uller-Meskamp et al. reported on the development of devices with nanoscale dimensions for computational use. Ferrocene derivatives were inserted into SAMs from solution. Upon annealing, the inserted molecules formed separated domains, which protrude as islands from the surrounding monolayer.156 9.4.4

Data Storage

Ferrocene derivatives have also been of interest for data storage.141, 156, 174, 175 The approach of the groups of Bocian, Kuhr and Lindsey was to construct molecular architectures comprised of multiple redox-active units, one of those being ferrocene.175 Those units were incorporated into SAMs and exhibited robust reversible electrochemistry, making them potential candidates for multibit molecular information storage elements. 9.4.5

Nanotubes

In the ﬁeld of nanotubes ferrocene has been incorporated into such tubes (forming pea pods) or attached to them.149, 176–182 The ﬁlling of the nanotubes could be accomplished from both the liquid and the vapour phases. Ferrocene molecules formed single molecular chains inside the tube and affected its properties. The metallocene was decomposed inside the nanotube to form new tubes inside the existing tubes and/or affect the tube’s properties in a different way. Since the ﬁrst carbon nanotubes ﬁlled with metallocenes were only reported in 2002, this area of research is still in its infancy.182

References 1. P. Harper, B. Wherrett (Eds), Nonlinear Optics, 1977, Academic Press, New York, USA. 2. Y.R. Shen, The Principles of Nonlinear Optics, 1984, John Wiley & Sans Inc., New York, USA. 3. B.E.A. Saleh, M.C. Teich, Fundamentals of Photonics, 1991, John Wiley & Sans Inc., New York, USA. 4. D.S. Chemla, J. Zyss (Eds), Nonlinear Optical Properties of Organic Molecules and Crystals, Vols. 1 and 2, 1987, Academic Press: Orlando, FL, USA. 5. J. Zyss, Molecular Nonlinear Optics: Materials, Physics and Devices, 1994, Academic Press: Boston, USA. 6. C. Bosshard, K. Sutter, P. Prˆetre et al. Nonlinear Optical Materials (Advances in Nonlinear Optics, Vol. 1), 1995, Gordon & Breach: Amsterdam, the Netherlands.

388

Ferrocenes: Ligands, Materials and Biomolecules

7. H.S. Nalwa, S. Miyata (Eds), Nonlinear Optics of Organic Molecules and Polymers, 1997, CRC Press: Boca Raton, FL, USA. 8. I.R. Whittall, A.M. McDonagh, M.G. Humphrey, Adv. Organomet. Chem., 1998, 42, 291–362. 9. Gmelin Handbook of Inorganic and Organometallic Chemistry, 8th Ed., Fe Organic Compounds, Vols. A1 (1974), A2 (1977), A3 (1978), A4 (1980), A5 (1981), A6 (1977), A7 (1980), A8 (1985), A9 (1989), A10 (1991), Springer, Berlin, Germany. 10. B.J. Coe, Comprehensive Coordination Chemistry II , 2004, 9, 621–687. 11. A. Togni, T. Hayashi (Eds), Ferrocenes, 1995, VCH, Weinheim, Germany. 12. B. Bildstein, Coord. Chem. Rev., 2000, 206–207, 369–394. 13. F. Zhang, M.H. Prosenc, J. Heck, J. Organomet. Chem., 2006, 691, 455–462. 14. S.K. Kurtz, T.T. Perry, J. Appl. Phys., 1968, 39, 3798–3813. 15. J.J. Wolff, R. Wortmann, Adv. Phys. Org. Chem., 1999, 32, 121–217. 16. B.F. Levine, C.G. Bethea, Appl. Phys. Lett., 1974, 24, 445–447. 17. B.F. Levine, C.G. Bethea, J. Chem. Phys., 1975, 63, 2666–2682. 18. K.D. Singer, A.F. Garito, J. Chem. Phys., 1981, 75, 3572–3580. 19. R.W. Terhune, P.D. Maker, C.M. Savage, Phys. Rev. Lett., 1965, 14, 681–684. 20. K. Clays, A. Persoons, Rev. Sci. Instrum., 1992, 63, 3285–3289. 21. E. Hendrickx, K. Clays, A. Persoons, Acc. Chem. Res., 1998, 31, 675–683. 22. G. Olbrechts, R. Stobbe, K. Clays, A. Persoons, Rev. Sci. Instrum., 1998, 69, 2233–2241. 23. K. Clays, G. Olbrechts, T. Munters, A. Persoons, Polym. Prepr., 1998, 39, 1131. 24. R. Wortmann, P. Kr¨amer, C. Glania et al. Chem. Phys., 1993, 173, 99–108. 25. J.L. Oudar, D.S. Chemla, J. Chem. Phys., 1977, 66, 2664–2668. 26. J.L. Oudar, J. Chem. Phys., 1977, 67, 446–457. 27. S. Barlow, H.E. Bunting, C. Ringham et al. J. Am. Chem. Soc., 1999; 121, 3715–3723. 28. S. Barlow, S.R. Marder, Chem. Commun., 2000, 1555–1562. 29. A. Yariv, IEEE J. Quantum Electron., 1978, QE14, 650–660. 30. D.G. Steel, R.C. Lind, J.F. Lam, Phys. Rev. A, 1981, 23, 2513–2524. 31. R.C. Lind, D.G. Steel, G.J. Dunning, Opt. Eng., 1982, 21, 190–198. 32. I. Biaggio, Phys. Rev. A, 2001, 64, 063813/1–13. 33. M. Sheik–Bahae, A.A. Said, E.W. van Stryland, Opt. Lett., 1989, 14, 955–957. 34. M. Sheik–Bahae, A.A. Said, T.H. Wie et al. IEEE J. Quantum Electron., 1990, 26, 760–769. 35. E.W. van Stryland, M. Sheik–Bahae, Opt. Eng., 1998, 60, 655–692. 36. A. Willetts, J.E. Rice, D.M. Burland et al. J. Chem. Phys., 1992, 97, 7590–7599. 37. J.J. Wolff, R. Wortmann, Adv. Phys. Org. Chem., 1999, 32, 121–217. 38. M. Malaun, R. Kowallick, A.M. McDonagh et al. Dalton Trans., 2001, 3025–3038. 39. B.J. Coe, T.A. Hamor, C.J. Jones et al. Dalton Trans., 1995, 673–684. 40. J.A. Campo, M. Cano, J.V. Heras et al. J. Mater. Chem., 1999, 9, 899–907. 41. C. L´opez-Garabito, J.A. Campo, J.H. Heras et al. J. Phys. Chem. B, 1998, 102, 10698–10706. 42. G. Rojo, F. Agull´o-L´opez, J.A. Campo et al. Synth. Met., 2001, 124, 201–203. 43. J. Heck, H.C. Brussaard, S. Dabek et al. Proc. SPIE Int. Soc. Opt. Eng., 1997, 3147, 53–61. 44. T. Meyer-Friedrichsen, C. Mecker, M.H. Prosenc, J. Heck, Eur. J. Inorg. Chem., 2002, 239–248. 45. H. Wong, R. Meyer-Friedrichsen, T. Farrell et al. Eur. J. Inorg. Chem., 2000, 631–346. 46. U. Behrens. H. Brussaard, U. Hagenau et al. Chem. Eur. J., 1996, 2, 98–103. 47. J. Mata, S. Uriel, E. Peris, et al. J. Organomet. Chem., 1998, 562, 197–202. 48. F. Zhang, M.H. Prosenc, J. Heck, J. Organomet. Chem., 2006, 691, 455–462. 49. U. Behrens, T. Meyer-Friedrichsen, J. Heck, Z. Anorg. Allg. Chem., 2003, 629, 1421–1430. 50. U. Hagenau, J. Heck, E. Hendrickx et al. Inorg. Chem., 1996, 35, 7863–7866.

Ferrocene-Based Electro-Optical Materials

389

51. T.J.J. M¨uller, A. Netz, M. Ansorge, Organometallics, 1999, 18, 5066–5074. 52. J. Heck, S. Dabek, T. Meyer-Friedrichsen, H. Wong, Coord. Chem. Rev., 1999, 190–192, 1217–1254. 53. J.A. Mata, E. Peris, I. Asselberghs et al. New J. Chem., 2001, 25, 1043–1046. 54. Y. Qian, Y. Sun, J. Liu, Gongneng Cailiao, 1999, 30, 604–605. 55. Y. Qian, Y-M. Sun, J-Z. Liu et al. Acta Chimica Sinica (Huaxue Xuebao), 1998, 56, 340–346. 56. L-H. Gao, K-Z. Wang, C-H. Huang et al. Chem. Lett., 1995, 1049–1050. 57. K-Z. Wang, C-H. Huang, D-J. Zhou et al. Solid State Commun., 1995, 93, 189–191. 58. O. Briel, K. S¨unkel, I. Krossing et al. Eur. J. Inorg. Chem., 1999, 483–490. 59. D.A. Davies, J. Silver, G. Cross, P. Thomas, J. Org. Chem., 2001, 631, 59–66. 60. I.S. Lee, Y.K. Chung, J. Mun, C.S. Yoon, Organometallics, 1999, 18, 5080–5085. 61. D.M. Shin, I.S. Lee, Y.K. Chung, Eur. J. Inorg. Chem., 2003, 2311–2317. 62. G.G.A. Balavoine, J-C. Daran, G. Iftime et al. Organometallics, 1999, 18, 21–29. 63. G. Iftime, G.G.A. Balavoine, J-C. Daran et al. C. R. Acad. Sci. Paris, S´erie IIc, Chimie, 2000, 3, 139–146. 64. N. Tsuboya, M. Lamrani, R. Hamasaki et al. J. Mater. Chem., 2002, 12, 2701–2705. 65. S.K. Pal, A. Krishnan, P.K. Das, A.G. Samuelson, J. Organomet. Chem., 2000, 604, 248–259. 66. H. Guo-Sheng, L. Yong-Min, W. Xiao-Li et al. Appl. Organomet. Chem., 2003, 17, 706–710. 67. T. Kondo. S. Horiuchi, I. Yagi et al. J. Am. Chem. Soc., 1999, 121, 391–398. 68. I. Asselberghs, K. Clays, A. Persoons et al. Chem. Phys. Lett., 2003, 368, 408–411. 69. M. Malaun, Z.R. Reeves, R.L. Paul et al. Chem. Commun., 2001, 49–50. 70. J.A. Mata, E. Peris, R. Llusar et al. Eur. J. Inorg. Chem., 2001, 2113–2122. 71. E. Stankovic, S. Toma, R.V. Boxel et al. J. Organomet. Chem., 2001, 637–639, 426–434. 72. V. Alain, A. Fort, M. Barzoukas et al. Inog. Chim. Acta, 1996, 242, 43–49. 73. V. Alain, M. Blanchard-Desce, C-T. Chen et al. Synth. Met., 1996, 81, 133–136. 74. Y-W. Cao, X-D. Chai, W-S. Yang et al. Thin Solid Films, 1996, 284–285, 859–862. 75. S.K. Pal, A. Krishnan, P.K. Das, A.G. Samuelson, J. Organomet. Chem., 2001, 637–639, 827–831. 76. V. Mamane, L. Ledoux-Rak, S. Deveau et al. Synthesis, 2003, 455–467. 77. A. Krishnan, S.K. Pal, P. Nandakumar et al. Chem. Phys., 2001, 265, 313–322. 78. A.J. Moore, A. Chesney, M.R. Bryce et al. Eur. J. Org. Chem., 2001, 2670–2687. 79. M. Blanchard-Desce, C. Runser, A. Fort et al. Chem. Phys., 1995, 199, 253–261. 80. J.A. Mata, E. Peris, S. Uriel et al. Polyhedron, 2001, 20, 2083–2088. 81. I. Janowska, J. Zakrzewski, K. Nakatani et al. J. Organomet. Chem., 2003, 675, 35–41. 82. C. Grogger, H. Rautz, H. St¨uger, Monatsh. Chem., 2001, 132, 453–464. 83. K.N. Jayaprakash, P.C. Ray, I. Matsuoka et al. Organometallics, 1999, 18, 3851–3858. 84. R. Farrell, T. Meyer-Friedrichsen, M. Malessa et al. Dalton Trans., 2001, 29–36. 85. T. Farrell, A.R. Manning, T.C. Murphy et al. Eur. J. Inorg. Chem., 2001, 2365–2375. 86. D. Naskar, S.K. Das, L. Giribabu et al. Organometallics, 2000, 19, 1464–1469. 87. R.D.A. Hudson, A.R. Manning, D.F. Nolan et al. J. Organomet. Chem., 2001, 619, 141–151. 88. R.D.A. Hudson, I. Asselberghs, K. Clays et al. J. Organomet. Chem., 2001, 637–639, 435–444. 89. N. Tsuboya, R. Hamasaki, M. Ito et al. J. Mater. Chem., 2003, 13, 511–513. 90. R. Herrmann, B. Pedersen, G. Wagner, J-H. Youn, J. Organomet. Chem., 1998, 571, 261–266. 91. R. Kumar, R. Misra, V. PrabhuRaja, T.K. Chandrashekar, Chem. Eur. J., 2005, 11, 5695–5707.

390

Ferrocenes: Ligands, Materials and Biomolecules

92. A. Gonz´alez-Cabello, D.G. Claessens, G. Martin-Fuch et al. Synth. Met., 2003, 137, 1487–1488. 93. A. Sen, P.C. Ray, P.K. Das, V. Krishnan, J. Phys. Chem., 1996, 100, 19611–19613. 94. C. Arbez-Gindre, B.R. Steele, G.A. Heropoulos et al. J. Organomet. Chem., 2005, 690, 1620–1626 95. X. Zhao, H.K. Sharma, F. Cervantes-Lee et al. J. Organomet. Chem., 2003, 686, 235–241. 96. H.K. Sharma, K.H. Pannell, I. Ledoux et al. Organometallics, 2000, 19, 770–774. 97. B. Pedersen, G. Wagner, R. Herrmann et al. J. Organomet. Chem., 1999, 590, 129–137. 98. N. Szesni, M. Drexler, J. Maurer et al. Organometallics, 2006, 25, 5774–5787. 99. M.C.B. Colbert, J. Lewis, N.J. Long et al. J. Organomet. Chem., 1997, 531, 183–190. 100. M. Malessa, J. Heck, J. Kopf, M.H. Garcia, Eur. Inorg. Chem., 2006, 857–867. 101. W-H. Zhang, Y-Z. Xu, S-Z. Liu et al. Acta Chimica Sinica, 2001, 12, 2176–2180. 102. J. Zyss, Nonlinear Opt., 1991, 1, 3–18. 103. J. Zyss, J. Chem. Phys., 1993, 98, 6583–6599. 104. J. Zyss. I. Ledoux, Chem. Rev., 1994, 94, 77–105. 105. M. Joffre, D. Yaron, R.J. Silbey, J. Zyss, J. Chem. Pys., 1992, 97, 5607–5615. 106. G. Rojo, F. Agull´o-L´opez, J.A. Campo et al. J. Phys. Chem. B, 1999, 103, 11016–11020. 107. G. Li, Y. Song, H. Hou et al. Inorg. Chem., 2003, 42, 913–920. 108. I. Janoswka, F. Miomandre, G. Clavier et al. J. Phys. Chem. A, 2006, 110, 12971–12975. 109. R. Rangel-Rojo, K. Kimura, H. Matsuda et al. Optics Communications, 2003, 228, 181–186. 110. R. Rangel-Rojo, H. Matsuda, K. Kimura et al. Proc. SPIE Int. Soc. Opt. Eng., 2003, 5212, 364–371. 111. C-Y. Zuo, Y-H. Zhou, J-Y. Wu, Y-P. Tian, Chinese Journal of Inorganic Chemistry, 2004, 20, 1018–1022. 112. A.N. Egorov, O.B. Mavritsky, A.N. Petrovsky, K.V. Yabukovsky, Laser Physics, 1995, 5, 1006–1013. 113. O.B. Mavritsky, A.N. Egorov, A.N. Petrovsky, K.V. Yabukovsky, Proc. SPIE Int. Soc. Opt. Eng., 1996, 2854, 254–265. 114. R. Misra, R. Kumar, T. Chandrashekar et al. J. Am. Chem. Soc., 2006, 128, 16083–16091. 115. S. Venkatraman, R. Kumar, J. Sankar et al. Chem. Eur. J., 2004, 10, 1423–1432. 116. A.V. Vannikov, A.D. Grishina, Y.G. Gorbunova et al. Doklady Physical Chemistry, 2005, 403, 137–141. 117. H.S. Nalwa, Materials Letters, 1997, 33, 23–26. 118. K. Dhanalakshmi, G. Sundararajan, J. Organomet. Chem., 2002, 645, 27–32. 119. J-K. Lee, S.I. Cho, Mat. Res. Soc. Symph. Proc., 2000, 598, BB 3.43.1–BB3.43.6. 120. J-K. Lee, S.I. Cho, Bull. Korean Chem. Soc., 2000, 21, 759–760. 121. J.Y. Wu, Y.P. Tian, F.X. Xie, S.S. Ni, Chinese Chemical Letters, 1999, 10, 251–254. 122. J-Y. Wu, Y-P. Tian, F-X. Xie, S-S. Ni, Chem. Res.Chin. Univ., 1999, 15, 218–225. 123. Y-P. Tian, Z-L. Lu, X-Z. You et al. Acta Chimica Sinica, 1999, 57, 1068–1074. 124. H. Hou, G. Li, Y. Song et al. Eur. J. Inorg. Chem., 2003, 2325–2332. 125. Y. Cai, Y. Song, H. Zheng et al. Chemistry Letters, 2002, 508–509. 126. Y-Y. Niu, T-N. Chen, S-X. Liu et al. Dalton Trans., 2002, 1980–1984. 127. Y-Y. Niu, Y-L. Song, S-X. Liu, X-Q. Xin, Z. Anorg. Allg. Chem., 2002, 628, 179–182. 128. M.D. Thomson, M. Novosel, H.G. Roskos et al. J. Phys. Chem. A, 2004, 108, 3281–3291. 129. A.M. Asiri, Appl. Organometal. Chem., 2003, 15, 907–915. 130. Q-B. Song, X-N. Li, T-H. Shen et al. Synth. Commun., 2003, 33, 3935–3941. 131. D-Y. Noh, H-J. Lee, H. Kang et al. Mol. Cryst. Liq. Cryst.. 2002, 376, 269–274. 132. K-Y. Kay, Y.G. Back, Chem. Ber./Recueil , 1997, 130, 581–584. 133. H. Kersten, P. Boldt, J. prakt. Chem., 1996, 338, 129–139. 134. P. Hu, K-Q. Zhao, H-B. Xu, Molecules, 2001, 6, M249.

Ferrocene-Based Electro-Optical Materials 135. 136. 137. 138. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 166. 167. 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 178. 179. 180.

391

K.R.J. Thomas, J.T. Lin, Y.S. Wen, J. Organomet. Chem., 1999, 575, 301–309. K. Uosaki, T. Kondo, X-Q. Zhang, M. Yanagida, J. Am. Chem. Soc., 1997, 119, 8367–8369. H. Imahori, M. Kimura, K. Hosomizu et al. Chem. Eur. J., 2004, 10, 5111–5122. M. Yanagida, T. Kanai, X-Q. Zhang et al. Bull. Chem. Soc. Jpn., 1998, 71, 2555–2559. H. Imahori, H. Norieda, H. Yamada et al. J. Am. Chem. Soc., 2001, 123, 100–110. T. Morita, S. Kimura, J. Am. Chem. Soc,, 2003, 125, 8732–8733. Q. Li, G. Mathur, S. Gowda et al. Adv. Mater., 2004, 16, 133–137. C.A. Nijhuis, F. Yu, W. Knoll et al. Langmuir, 2005, 21, 7866–7876. J. Watanabe, T. Morita, S. Kimura, J. Phys. Chem. B, 2005, 109, 14416–14425. E.J. Pacsial, D. Alexander, R.J. Alvarado et al. J. Phys. Chem. B, 2004, 108, 19307–19313. J. Hu, D.L. Mattern, J. Org. Chem., 2000, 65, 2277–2281. K. Kitagawa, T. Morita, S. Kimura, Langmuir, 2005, 21, 10624–10631. K. Hu, Z. Chai, J.K. Whitesell, A.J. Bard, Langmuir, 1999, 15, 3343–3347. J. Liu, J.J. Gooding, M.N. Paddon-Row, Chem. Commun., 2005, 631–633. Y-f. Chen, I.A. Banerjee, L. Yu et al. Langmuir, 2004, 20, 8409–8413. G. Valincius, G. Niaura, B. Kazekeviˇcien´e et al. Langmuir, 2004, 20, 6631–6638. M. P´eter, M.A. Hempenius, E.S. Kooij et al. Langmuir, 2004, 20, 891–897. R.A. Wassel, R. Fuierer, N. Kim, C.B. Gorman, Nano Lett., 2003, 3, 1617–1620. P.D. Beer, J.J. Davis, D.A. Drillsma-Milgrom, F. Szemes, Chem Commun., 2002, 1716–1717. J.F. Smalley, H.O. Finklea, C.E.D. Chidsey et al. J. Am. Chem. Soc., 2003, 125, 2004–2013. T. Kondo, T. Kanai, K. Uosaki, Langmuir, 2001, 17, 6317–6324. L. M¨uller-Meskamp, B. L¨ussem, S. Karth¨auser et al. Phys. Status Solidi A, 2006, 203, 1448–1452. S.W. Han, H. Seo, Y.K. Chung, K. Kim, Langmuir, 2000, 16, 9493–9500. S. Ye, Y. Sato, K. Uosaki, Langmuir, 1997, 13, 3157–3161. M.T. Crua˜nes, H.G. Drickamer, L.R. Faulkner, Langmuir, 1995, 11, 4089–4097. T. Weidner, B. Krohn, M. Trojtza et al. Proc. SPIE Int. Soc. Opt. Eng., 2006, 6106, 61061S–1-61061S-10. R. Sawada, M. Sakomura, K. Hirukawa, M. Fujihira, Colloid Surf., A, 2006, 284–285, 448–452. W.J. Shumate, D.L. Mattern, A. Jaiswal et al. J. Phys. Chem. B, 2006, 110, 11146–11159. D. Basak, B. Mallik, Chem. Lett., 2004, 33, 1524–1525. H. Imahori, K. Tamaki, Y. Araki et al. J. Am. Chem. Soc., 2002, 124, 5165–5174. H. Imahori, Y. Sekiguchi, Y. Kashiwagi et al. Chem. Eur. J., 2004, 10, 3184–3196. V. Biju, S. Barazzouk, K.G. Thomas et al. Langmuir, 2001, 17, 2930–2936. H. Imahori, K. Tamaka, D.M. Guldi et al. J. Am. Chem. Soc., 2001, 123, 2607–2617. ´ Herranz, B. Illescas, N. Martin et al. J. Org. Chem., 2000, 65, 5728–5738. M.A. F. Langa, P. de la Cruz, E. Esp´ıldora et al. J. Org. Chem., 2000, 65, 8675–8684. H. Imahori, H. Yamada, S. Ozawa et al. Chem. Commun., 1999,, 1165–1166. M. Sakomura, S. Lin, T.A. Moore et al. J. Phys. Chem. A, 2002, 106, 2218–2226. C. Sporer, I. Ratera, D. Ruiz-Molina et al. J. Phys. Chem. Solids, 2004, 65, 753–758. I. Ratera, D. Ruiz-Molina, F. Renz et al. J. Am. Chem. Soc., 2003, 125, 1462–1463. W. Yuan, L. Sun, H. Tang et al. Adv. Mater., 2005, 17, 156–160. D.T. Gryko, F. Zhao, A.A. Yasseri et al. J. Org. Chem., 2000, 665, 7356–7362. Y.F. Li, R. Hatakeyama, T. Kaneko et al. Nanotechnology, 2006, 17, 4143–4147. V.M. Garc´ıa-Su´arez, J. Ferrer, C.J. Lambert, Phys. Rev. B, 2006, 74, 205421–1-205421-11. E.L. Sceats, J.C. Green, J. Chem. Phys., 2006, 125, 154704–1-154704-12. N. Sun, L. Guan, Z. Shi et al. Anal. Chem., 2006, 78, 6050–6057. X-S. Wang, M.A. Winnik, I. Manners, Angew. Chem. Int. Ed., 2004, 43, 3703–3704.

392

Ferrocenes: Ligands, Materials and Biomolecules

181. L. Guan, Z. Shi, M. Li, Z. Gu, Carbon, 2005, 43, 2780–2785. 182. F. Stercel, N.M. Nemes, J.E. Fischer, D.E. Luzzi, Mat. Res. Soc. Symp. Proc., 2002, 706, 245–250. 183. D.R. Kanis, M.A. Ratner, T.J. Marks, J. Am. Chem. Soc., 1992, 114, 10338–10357. 184. N. Bloembergen, P.S. Pershan, Phys. Rev., 1962, 128, 606–622. 185. C.R. Moylan, S.A. Swanson, C.A. Walsh et al. Proc. SPIE Int. Soc. Opt. Eng., 1993, 2025, 192–201.

10 Ferrocene-Containing Polymers and Dendrimers Nicholas J. Long and Konrad Kowalski

10.1

Introduction

The advances in science and technology witnessed over the last few decades, and their effects on everyday life, are inextricably linked to the new types of materials that have became available. Foremost amongst them, polymers – both man-made synthetic and naturally-occurring – have had a vital role to play. Interest in metal-containing polymers has grown rapidly, largely due to their technological potential and improvements in synthetic methodology. The introduction of a metal centre into an organic, and usually conjugated, polymeric chain may introduce a range of properties (such as redox, magnetic, optical and electronic) that differ from those of conventional organic polymers. Ferrocene chemistry has once again been an integral part of the development of new synthetic organometallic polymers, and has introduced a raft of applications for these seemingly underdeveloped species. Inclusion of the ferrocene unit is attractive due its electron donating ability, reversible redox chemistry, steric properties and its established and facile functionalisation and derivatisation. This chapter focuses on the synthesis, properties and applications of ferrocenecontaining polymers. It will also provide information covering the burgeoning and fascinating area of ferrocene-containing dendrimers, with key literature referenced throughout the chapter. It should be pointed out that in 2004 an excellent book entitled ‘Synthetic Metal-Containing Polymers’ by Ian Manners was published.1 It covers the wider, more general area of metal-containing polymers, though of course there is Ferrocenes: Ligands, Materials and Biomolecules Edited by Petr Stepnicka  2008 John Wiley & Sons, Ltd

394

Ferrocenes: Ligands, Materials and Biomolecules

Fe

Fe

Fe n

Fe

n

Fe

Fe A

Fe

B

Fe

Fe n

C

Scheme 10.1 General classiﬁcation of ferrocene-containing polymers

signiﬁcant emphasis given to ferrocene-containing polymers and materials, as would be expected from the leader in this ﬁeld. A general classiﬁcation of ferrocene-containing polymers has recently been described by Hudson in an excellent review article.2 In this chapter, this has been modiﬁed for slightly more accuracy (Scheme 10.1) and, thus, the following groupings have been made: A Main chain polymers in which a 1,1 -substitution pattern of the ferrocene unit is present. B Side-chain polymers in which the ferrocene plays the role of a lateral group attached to main polymer chain. C A rare subclass of group B in which the ferrocene unit is inserted into the polymer chain via double substitution of one of its cyclopentadienyl ligands. Group A can be divided into two sub-groups: A1 Polyferrocenylenes – polymers where the ferrocene units in the main chain are connected directly by carbon−carbon bonds from the cyclopentadienyl rings. A2 Polyferrocenylenes – where the ferrocene units with a 1,1 -substitution pattern are connected indirectly by short/long spacers (Scheme 10.2). The ﬁrst ferrocene-containing polymer, poly(vinylferrocene), was synthesised in 1955 soon after ferrocene itself was discovered and reported. It was soon realised that the application potential of ferrocene-containing polymers could be linked to the high chemical, thermal, air and moisture stability of ferrocene itself, along with its reversible

Ferrocene-Containing Polymers and Dendrimers

395

Fe

Fe

Fe

n

A1 spacer Fe spacer Fe spacer Fe n

A2

Scheme 10.2 Polyferrocenylenes of class A1 and A2

one-electron oxidation to the reasonably stable ferrocenium cation. Initially, there was quite a challenge in transferring the considerable achievements in the synthesis and structural characterisation of σ - and π-bridged diferrocenes to high weight ferrocenecontaining polymers. To date, applications of ferrocene-bearing polymers include electrochemically-switchable wires, ceramic devices, biosensors, molecular magnets and nonlinear optically active materials, which will be covered later in the chapter. To examine the various physical properties of ferrocene-containing polymers a combination of techniques is employed, i.e. methods for low molecular weight compounds such as multinuclear NMR, ESR, IR, electrochemistry and spectroelectrochemistry and M¨ossbauer spectroscopy are combined with those belonging to polymers and macromolecules. The latter comprise Gel Permeation Chromatography (GPC) also known as Size Exclusion Chromatography (SEC), which gives information about the molecular weight distribution as well as values of weight-average molecular weight (MW ) and the number-average molecular weight (Mn ), MALDI–TOF mass spectrometry, ﬁeld emission scanning electron microscopy (FESEM), differential scanning calorimetry (DSC) and atomic force microscopy (AFM). Information on the morphology of polymers is accessible by powder X-ray diffraction (PXRD), also known as wide-range X-ray scattering (SAXS). In addition, historically old techniques such as melting temperature (Tm ) are useful for the detection of order–disorder transition. However, the fundamental problem that for many years blocked progress in the area of ferrocene-containing polymers was the lack of suitable synthetic methodology, along with problems in the characterisation of the often low molecular weight materials obtained. The status of synthetic methodology is discussed here.

396

10.2

Ferrocenes: Ligands, Materials and Biomolecules

Synthetic Methods Leading to Ferrocene-Containing Polymers

Ferrocene can be appended to the polymer in two ways. The ﬁrst is when the polymer is built from ferrocene-containing monomers. This approach opens possibilities for copolymerization of ferrocene monomers with ‘pure’ organic ones. The second is when an already existing polymer is modiﬁed by ferrocene (or a ferrocene-containing fragment). The practical realization of this synthetic methodology is covered by many types of chemical reactions but, in general, they originate from ‘traditional’ organic polymer chemistry. It is important to emphasize that progress in the synthesis of ferrocene-containing polymers has been strongly dependent on the synthesis of suitable low molecular weight monomeric precursors. The main routes to ferrocene-containing polymers are: free radical polymerization, addition polymerization, condensation polymerization, living polymerization and electropolymerization. Additionally in recent years, ring opening polymerization (ROP), which can be divided into thermal ROP, transition metal-catalyzed ROP and ringopening metathesis polymerization (ROMP), have come to the fore. Examples will be given of each of the polymerization processes, though detailed mechanistic insight is not possible due to space constraints.

10.3

Synthesis of Side-Chain Ferrocene-Containing Polymers (type B)

The ﬁrst example of a side-chain ferrocene-containing polymer and ﬁrst known ferrocene-containing polymer in general was made from vinylferrocene 1.3 The product poly(vinylferrocene) (PVFc, 2) was synthesized by a radical-initiated polymerization of the vinylferrocene monomer (Scheme 10.3). Molecular masses of polymers obtained via this method are generally less than 10 000, however polymers with MW > 105 (PDI > 1.3) are known. The ﬁrst detailed characterization of poly(vinylferrocene) was carried out by Pittmann.4 Azobis(isobutyronitrile) (AIBN), a peroxide initiator, was typically used in the synthesis of 2. However, it can cause side effects via the oxidation of the iron atom in ferrocene. More recently living anionic polymerization of 1 has been described.5 Based on this strategy, block copolymers of 2 with poly(methyl methacrylate) (PVFcb-PMMA, 3 in Scheme 10.4) or polystyrene (PVFc-b-PS) were obtained.6 Free radical polymerization has also been applied to the synthesis of ferrocene-containing

Fe

radical or ionic initiator

H2C CH n

Fe

1

2

Scheme 10.3 Polymerization of vinylferrocene

Ferrocene-Containing Polymers and Dendrimers RLi/THF/ −25 °C

Fc 1

397

PVFc i) CH2=CMe(CO2Me)/THF

ii) H2O

−70 °C CH2 R

CH

H2C

Fc

CH2 C CO2CH3

n

H m

3

Scheme 10.4 Living anionic polymerization of vinylferrocene using alkyllithium initiators

0.05 O

Fe

Me O

CO2Et

NC 4

Scheme 10.5 Ferrocene-containing NLO active polymer 10

methacrylate/methyl methacrylate copolymers 4 (Scheme 10.5).7 The molecular weight of these materials has been reported to be Mn = 3.0 × 104 by GPC, and the second harmonic generation properties are reported as comparable to quartz standards. The Wurtz synthesis has been used to form ferrocene-polysilanes 5 (Scheme 10.6) and via this method polymers containing a relatively low level of ferrocenyl units as side chains can be obtained.8 The ratio between methylphenylsilane and methylferrocenylsilane ranges from 6:1 to 27:1 and isolation of copolymer fraction with MW up to 3.9 × 105 has been achieved. It was interesting to note that the methylferrocenylsilane polymer exhibited greater photostability than its poly(methylphenylsilane) analogue – presumably due to the stabilization effect of the ferrocene moiety. The hydrosilylation reaction has been used to synthesize ferrocene–polysiloxane random copolymers 6 with molecular weights ranging from 5000 to 10 000 (Scheme 10.7). Such polymers have been applied as amperometric biosensors for the detection of glucose. Similar hydrosilylation strategies have also been applied to the preparation of liquid-crystalline ferrocene-containing polymers 7 (Scheme 10.8; see also Chapter 11).9 The molecular weights of these macromolecules were about MW = 2.5–3.1 × 104 with the PDI value 1.4–1.6. Thermal ring-opening polymerization of strained ferrocenyl-organocyclotriphosphazene monomers 8 leads to a ferrocene

398

Ferrocenes: Ligands, Materials and Biomolecules

SiMeCl2

+

Me

Me

Si

Si

Na/toluene/heat Fe

SiMeCl2 Fe

n

5

Scheme 10.6 Wurtz synthesis of ferrocene–polysilanes

Me3Si

Me

Me

Si O

Si

H

Me

O

SiMe3 n

CH2=CHR H2PtCl6 Cat.

R = Fc or Fc(CH2)7

Me Me3Si

Me

Si O

Si

(CH2)y

Me

O

SiMe3 n

Fc 6

Scheme 10.7 reactions

y = 2 or 9

Synthesis of ferrocene–polysiloxane random copolymers via hydrosilylation

side-chain polymer 9 (MW = 1.8 × 106 , PDI = 6.2) which represents an unusual structural motif due to bonding of the ferrocene unit to the main chain via a 1,1 binding mode (Scheme 10.9).10 In contrast to the above methodology, introduction of ferrocene to poly(methylphenylphosphazene) polymers can be achieved by a deprotonation of the poly(methylphenylphosphazene) polymer chain and subsequent addition of an electrophilic ferrocene derivative.11 Depending on the ferrocene reactant, various loading levels have been obtained. The molecular weights of the polymeric products 10 also differed, ranging from MW = 2.0 × 105 (R = H) and 1.5 × 105 (R = Me) (Scheme 10.10). Another synthesis of ferrocene side-chain polymers this time via ring opening metathesis polymerization (ROMP) of ferrocene-functionalized norbornenes in the presence of molybdenum alkylidene initiators, has been reported.12 The MW of obtained material (11) were, however, only of the order of 5090–9030 (PDI ≤ 1.2; Scheme 10.11).

Ferrocene-Containing Polymers and Dendrimers

Me Si

Me O

Si

H

Me

CH2=CHR O

Si

[Pt] Cat.

Me O

Si

CH2CH2

Me

399

O

Me n

n

R 7

CO2

OCnH2n+1

Fe a

(CH2)m

R=

O

2

CO2 2

b

(CH2)m

R=

CO2

O

OCnH2n+1

CO2 Fe 2

2

Scheme 10.8 Preparation of liquid–crystalline ferrocene–polysiloxane random copolymers

Fe R Fe

P

N

P

N

R P

N

1 mol% [NPCl2]3 250 °C

R P

R

N

R

P

N

R

P R

N n

R 8

Scheme 10.9 monomers

Thermal ring opening polymerization of strained ferrocenyl-triphosphazene

Ph P

9

1. 0.5 n BuLi N

O

2.

Me n

P

Ph N

P

Me

Fc R 3. H+

Ph

N

CH2 R

n

C OH Fc

(R = H or Me)

10

Scheme 10.10 Introduction of ferrocene-containing side chains to poly(methylphenylphosphazene) polymers

400

Ferrocenes: Ligands, Materials and Biomolecules

CO2Me

[Mo=CH(t-Bu)] cat. n

Fc

CO2Me

Fc 11

Scheme 10.11 bornenes

Ring opening metathesis polymerization of ferrocene-functionalised nor-

In the search for anticancer-active compounds, water-soluble ferrocene-containing polymers have a role to play. Thus, polyamide polymers with ferrocene as the pendant group have been synthesized by polycondensation and Michael type polyaddition reactions.13 For example, the reaction between polyamide polymer and N succinimidyl 4-ferrocenylbutanoate, Fc(CH2 )3 COOSuc afforded polymer 12 with O O

O

O N H

n

O

H N N H

H N

O

O

Fc

Fc

O

N H

(CH2)3COOSuc.

O N H

n

O

n

O O

O

N H

n

O

n

N N H

H N

O

O n

N H

n

O

O

N H n

12

Scheme 10.12 Synthesis of ferrocene-containing water soluble polyamide polymers: the reaction of polyamide polymer with N -succinimidyl 4-ferrocenylbutanoate, Fc(CH2 )3 CO2 Suc O

O

H N

H N

Fc(CH2)3CO2H

n

Fe

n

CONH R OH

a R = CH2 b R = (CH2)2O(CH2)2

CONH R OCO 13

Scheme 10.13 Synthesis of ferrocene-containing water soluble polyaspartamide or oligo(ethyleneoxide)-containing polyamides

Ferrocene-Containing Polymers and Dendrimers

401

35–70 % yield (Scheme 10.12).13 Another example of water-compatible ferrocene side-chain containing polymers is polyaspartamide or oligo(ethyleneoxide)-containing polyamides 13 (Scheme 10.13).13 These materials have been synthesized by reaction of the organic polymer with 4-ferrocenylbutanoic acid. The biological uses and applications of ferrocene-containing polymers are detailed in Chapter 13, so will not be covered here.

10.4 10.4.1

Synthesis of Main-Chain Ferrocene-Containing Polymers Synthesis of Polyferrocenylenes (type A1)

Heating ferrocene in the presence of stoichiometric amounts of di-t-butyl peroxide yields a complex mixture of oligomeric and polymeric materials with paramagnetic species (impurities) among them. Free radical mechanisms for this reaction have been proposed, leading to several different ferrocene-containing free radicals (homoannular 1,2-, and 1,3-, and heteroannular 1,1 -disubstituted ferrocene species). They all have a similar reactivity hence a mixture of products is formed. However, soluble polymer fractions with molecular weights <7000 have been isolated.14–18 The much more controlled step-growth polycondensation reaction of 1,1 -dilithioferrocene•TMEDA complex with 1,1 -diiodoferrocene at <25 ◦ C has been described19 and leads to soluble, spectroscopically characterizable, heteroannular, amorphous and diamagnetic polyferrocenylene 14 with Mn < 4000 (Scheme 10.14). Another variant of this reaction involves dehalogenation of starting dihaloferrocenes by magnesium (Scheme 10.14).20, 21 A similar strategy was involved in the synthesis of n-hexyl-substituted polyferrocenylenes 15. This method was based on the reaction of dihexylfulvalene dianion with [FeCl2 (THF)2 ] (Scheme 10.15). Via this method, the material was not of high molecular weight but a fraction of poly(1,1 -dihexylferrocenylene) (15) with Mn ≈ 5000 and PDI = 1.2 was isolated in trace yields. Interestingly, conductance properties were reported for this material. Li TMEDA

Fe I

Li

X

Mg

Fe

Fe

− MgBr

I

Fe X

n

14

X = Br, I

Scheme 10.14 Preparation of A1 type polymers

To complete the synthetic methods in this section, copper-mediated polycondensations of diiodoferrocenes and bis(stannyl) ferrocenes at 150–180 ◦ C should be mentioned (Scheme 10.16). This reaction yields low molecular weight but well characterized polyferrocenylenes 16 that contain substituents on the cyclopentadienyl rings.22

402

Ferrocenes: Ligands, Materials and Biomolecules 2−

R

R

FeCl2(THF)2 Fe R R n

R = C6H13

15

Scheme 10.15 Preparation of A1 type polymers by direct synthesis from dihexylfulvalene anion precursor and iron salt

R

R

I +

Fe I

SnMe3 Fe

Fe

R

R

R Cu powder

SnMe3

150 − 180 °C

R n

16

Scheme 10.16 Preparation of A1 type polymers by copper-mediated polycondensation reaction of diiodoferrocenes and bis(stannyl) ferrocenes

10.4.2

Synthesis of Polyferrocenylenes (type A2)

Synthetic methods for the preparation of polyferrocenes with one-carbon spacers (17 in Scheme 10.17) have been reviewed elsewhere23 and comprise polycondensation reactions of aldehydes with ferrocene as well as cationic polymerization of carbinols. Similar synthetic strategies were used in the synthesis of polyferrocenes with phosphorus spacers. When melted (or at 80–170 ◦ C in solution), ferrocene reacted with PPhCl2 , P(O)PhCl2 or P(S)PhCl2 in the presence of a Lewis acid catalyst to give low molecular weight materials 18–20 (Scheme 10.17) with Mn < 6000 and generally below 3500.24, 25 The McMurry reaction has been used to synthesize

H C Fe

P

R

Fe

n

17

Fe

Ph

S

P

P Fe

Ph

19

R = H, CH3, C6H5

Scheme 10.17 Examples of A2 type polymers

Ph

n

n

n

18

O

20

Ferrocene-Containing Polymers and Dendrimers

403

O R

Zn/TiCl4

R

Fe

Fe

R

R O n

21 R = C2H5 , C6H13, C12H25

Scheme 10.18 Preparation of A2 type polymers by the McMurry reaction

poly(ferrocenylenevinylene) polymers 21 (Scheme 10.18) with Mn = 3000–10 000 and E-geometry in respect to the vinylene fragment.26 10.4.3

Synthesis of Polyferrocenylenes (type C)

A rare example of C-type ferrocene-containing polymers 22 has been synthesized by Plenio et al. via a palladium-catalyzed polycondensation process (Scheme 10.19). Electrochemical studies of this material indicated electronic communication between iron atoms mediated by the ethynyl bridge.27 R

R Pd cat., CuI, i-Pr2NH

I Fe

Fe

n

R = CH2OMe, CH2NMe2

22

Scheme 10.19 Synthesis of type C polymers via the palladium-catalyzed Sonogashira reaction

10.4.4

Ring Opening Polymerization (ROP)

Ring opening polymerization has become an extraordinarily useful methodology for the preparation of well-deﬁned high molecular weight ferrocene-containing polymers and therefore will be described separately in this subchapter (it is discussed more extensively in Manners’ book).1 Strained cyclic ferrocene derivatives called [1]ferrocenophanes are the usual starting monomers for ROP. They possess two Cp ligands connected (bridged) by single atom. This causes adoption of a highly energetic, bent conformation (the cyclopentadienyl rings are not parallel to each other). Despite ferrocenophanes being known since 1975,28 the ﬁrst example of their application as monomers in ROP dates from 199229 and featured the thermal ROP of silylferrocenophane 23 yielding high molecular weight polyferrocenylsilanes (PFSs, 24) (Scheme 10.20). Monomeric [1]ferrocenophanes possess tilt angles of 16–21◦ between

404

Ferrocenes: Ligands, Materials and Biomolecules

R Fe

Si

R′

R′ Si

heat Fe

R n

23

24 R, R′ = Me or Ph

Scheme 10.20 Thermal ring opening polymerization of [1]silylferrocenophanes

the cyclopentadienyl ring planes and strain energies of ca. 70–80 kJ mol−1 .30, 31 Thermal ROP of silicon-bridged [1]ferrocenophanes is not limited to just methyl (alkyl) or phenyl (aryl) substituents attached to the silicon atom. [1]Dichlorosilaferrocenophane 25 provides possibilities for the substitution of the chlorine atoms by alkoxy or amino functions and ROP of such modiﬁed monomers produces polyferrocenylalkoxysilanes 26 (Scheme 10.21).32

RO OR ROH/Et3N

Cl Fe

OR

Si

Fe Et2O

Cl

Si

ROP Fe

Si OR

n

25

26

Scheme 10.21 Preparation of polyferrocenylalkoxysilanes via ring opening polymerization of [1]dialkoxysilylferrocenophanes

Experience accumulated on ROP of [1]silaferrocenophanes has been transferred to other ferrocenophanes. For example, [1]ferrocenophanes with germanium (27),33 phosphorus (28),34 sulfur (29),35 boron (30)36 or tin (31)37 (Scheme 10.22) as bridging atoms have been also used as monomers for ROP. In addition, less strained [2]metallocenophanes (with two bridging atoms between the Cp rings) can be used R′

R

R Fe

Fe

Ge R 27

P

R′

Fe R

28

S

Fe

B

NRR′

R Fe

Sn

R

R 29

30

Scheme 10.22 Selected examples of [1]heteroferrocenophanes

31

Ferrocene-Containing Polymers and Dendrimers

CY2

CY2CY2

ROP

Fe

405

Fe

CY2 n

32

33 Y = H or Me

Scheme 10.23 Ring opening polymerization of [2]ferrocenophanes

as monomeric precursors for ROP. Ethylene [2]ferrocenophane 32 undergoes ROP at 300 ◦ C to yield polymeric materials 33 (Scheme 10.23). In cases where the monomer contains CMe2 CMe2 as the bridging group, ROP gives soluble products that can be separated into two fractions: oligomeric (MW = 4800) and polymeric (MW = 9.6 × 104 ). ROP also allows the synthesis of copolymers, i.e. thermal ring opening copolymerization of silicon [1]ferrocenophane 34 with cyclic tetrasilanes yields polyferrocenylsilaner-polysilane 35 (Scheme 10.24).38

Me

Me Me Me Fe

Si Me

+

Ph

Si Si Ph

Me

Si Si Me

150 °C

Fe

Me Me Ph Me Ph Si

Si

Si

Si

Si

Me Ph Me Ph

Ph Ph n

34

35

Scheme 10.24 Preparation of polyferrocenylsilane-r -polysilane via thermal ring opening copolymerization of [1]dimethylsilylferrocenophane (34) and cyclic tetrasilane

Anionic ROP of ferrocenophanes is also known. In the presence of lithioferrocene or n-BuLi as anionic initiators, silicon-bridged [1]ferrocenophane 34 undergoes polymerization.39 In cases when high purity 34 and solvent were used, ‘living’ anionic ROP could be achieved.40, 41 The reaction comprises three stages: initiation, propagation and termination. Once formed, the anionic polymer can be terminated by a range of capping molecules to yield polyferrocenyl silanes 38 (Scheme 10.25). As long as the initiation stage is fast and no chain transfer or uncontrolled termination occurs, polyferrocenylsilanes with very narrow polydispersities (PDI < 1.10) are accessible. The Mn of such materials can be up to ca.120 000 and are dependent on the monomer to initiator ratio that limits the number of chain propagation sites.39, 41 Cationic ROP has also been reported for [1]stanna-42 and C-S-bridged [2]ferrocenophanes.43

406

Ferrocenes: Ligands, Materials and Biomolecules

(i) initiation

− Me Fe

RLi

Si

Fe

−Li+

Me

R Si Me Me

34

R = Me

36

(ii) propagation

n

−

Fe

Si

Me

Me

Me Si

Me Fe

Fe

Fe

R Si Me Me

R Si Me Me

− n

36

37

(iii) termination Me

Me Si

Fe

Fe

R Si Me Me

−

− Cl−

Me3SiCl

n

37

Me

Me Si

Fe

Fe

R Si Me Me

SiMe3 n

38

Scheme 10.25 Mechanism of anionic ‘living’ ring opening polymerization of 34

Another type of ROP known for strained ferrocenophanes is transition metalcatalyzed ROP. The application of transition metals as catalysts in the synthesis of ferrocene-containing polymers from monomeric strained ferrocenophanes offers some advantages over thermal or anionic methodology, i.e. relatively low reaction temperatures, extremely pure monomers and solvents are not required, and copolymerization of [1]ferrocenophanes with other monomers is allowed. Transition metal-catalyzed ROP of silicon and germanium-bridged [1]ferrocenophanes has been reported featuring both silicon and germanium atoms in the main chain of the polymer.44, 45 Transition metal-catalyzed ring opening metathesis polymerisation

Ferrocene-Containing Polymers and Dendrimers

407

(ROMP) of metallocenophanes can be regarded as an expansion of metal-catalyzed ROP. Vinylene-bridged [2]ferrocenophane 39 undergoes ROMP in the presence of a molybdenum catalyst to produce an insoluble orange polymer 40 (Scheme 10.26). However, partially soluble block copolymers 41 were also obtained from 39 and norbornene via the ROMP technique. By changing the catalyst:norbornene:39 ratio, it was possible to obtain soluble materials with Mw = 1710 and Mw = 3000 (for 1:10:10 ratio) and a higher Mw of 21 000 when a 1:50:10 ratio was applied.

Ph 1) [Mo]

Fe

Fe

2) PhCHO PhMe2C 39

n

40

1) m NBE, [Mo] 2) n x 39 3) PhCHO

Ph Fe CMe2Ph n m

41

Scheme 10.26 Molybdenum-catalyzed ring opening metathesis polymerization of [2]ferrocenophane 39 (NBE = norbornene)

ROMP has also been applied to the synthesis of high molecular weight soluble materials 42 (Mw > 300 000) starting from monomeric [4]ferrocenophane 43 (Scheme 10.27).46 Another example of ROP is the atom abstraction-induced ROP of chalcogenido-bridged metallocenophanes. Via atom (sulfur) abstractioninduced ring opening polymerization of [3]trithiaferrocenophane 44, well-deﬁned poly(ferrocenylenepersulﬁde)s 45 (Scheme 10.28) has been reported.47 The yellowcoloured material obtained proved to be soluble only when the ferrocene units were substituted with alkyl groups, and in addition, the rate of the desulfurization reaction and the nature of the resultant polymers has been reported to be solvent dependent.48–50 These fascinating materials are photosensitive in solution but stable in solid state.

408

Ferrocenes: Ligands, Materials and Biomolecules

[W cat]

Fe

Fe

n

43

Scheme 10.27 phane 43

42

Tungsten-catalyzed ring opening metathesis polymerization of [4]ferroceno-

R

R S S

Fe

S

Fe − S=PBu3

S

R'

S

PBu3 25 °C

R′ n

44

45

R = R′ = H, t − Bu R = n−Bu, R′ = H

Scheme 10.28 cenophane

10.4.5

Atom abstraction-induced ring opening polymerization of [3]trithiaferro-

Face-to-Face and Multidecker Polyferrocenes Obtained by Condensation Routes

Face-to-face ferrocene polymers are structurally interesting examples of ferrocenecontaining polymers. Their structure is deﬁned by locking of the ferrocene units in close proximity – the planes of the cyclopentadienyl rings are very near to each other, thus the ferrocene units stack to form pillar-like structures 46 bridge

Fe

Fe

Fe

bridge

bridge 46

Scheme 10.29 Face-to-face ferrocene-containing polymers

Ferrocene-Containing Polymers and Dendrimers

409

(Scheme 10.29). A detailed classiﬁcation of face-to-face metallocene polymers has been previously published.1, 51 1,8-Diiodonaphthalene has been successfully applied as a bridge in the synthesis of face-to-face ferrocene polymers. The suitable starting material 1,1 -bis(chlorozinc)ferrocene has been prepared by dilithiation of ferrocene followed by transmetallation with anhydrous zinc(II)chloride. Then, the 1,1 -dimetallated ferrocene and 1,8-diiodonaphthalene were reacted under palladium-catalyzed coupling conditions.52 The presence of moisture in commercially available etheral ZnCl2 solution effected formation of compound 47 (Scheme 10.30) in reasonable yields. A change in the ZnCl2 :dilithioferrocene ratio from 2:1 to 1:1 effected oligomer formation. However, despite much effort, the complete separation of reaction mixtures into pure oligomeric products has not been achieved. TLC separation indicated several fractions. The ﬁrst contained monomeric 47, the second consisted principally of the oligomers 48 and 49, whilst fraction three consisted with oligomers 50 and 51 (Scheme 10.30). Electrochemical measurements of oligomer 51 showed that four redox active iron centres generate three waves at a scan rate of 100 mV s−1 in methylene chloride. At lower scan rates, the redox reaction became irreversible along with the appearance of a

I

Fe Fe

Fe

Fe

Fe

Fe

47

Fe

48

Fe

Fe

50

49

Fe

Fe

Fe

51

Scheme 10.30 Ferrocenylnaphthalene mono-, di- and trimers

Fe

I

53

O

1. Na[N(SiMe3)2] 2. FeCl2 Fe

I CpCu Me2S

CpCu Me2S

Scheme 10.31 Multi-step synthesis of the monomer 54

I

LiAlH4

I

I

Fe

54

I

410 Ferrocenes: Ligands, Materials and Biomolecules

Ferrocene-Containing Polymers and Dendrimers

411

strong cathodic wave at 0.466 V vs. saturated calomel electrode (SCE). The difference in half-wave potentials for successive iron atom oxidations in 51 shows a progressive diminution in comparison to that observed in 47 and 50. Additionally, the ﬁrst and second oxidation half-wave potentials for the measured oligomers show a progressive decrease with growth of the chain length but this change diminishes sharply with increasing chain length. The above effects indicate an electronic interaction between iron centres of the partially oxidized species but the synthetic strategy suffers from the reagents’ water sensitivity and the low solubility of polymeric products. To overcome these difﬁculties a new synthetic strategy has been developed – the coupling reaction of cyclopentadienylcopper-dimethyl sulﬁde with 1,8-diiodonaphthalene gives 1,8bis(cyclopentadienyl)naphthalene 52.53 To increase the solubility of the polymer product, a long aliphatic octyl carbon chain was introduced onto the cyclopentadienyl rings. It was achieved by condensation of cyclopentadiene with 2-octanone followed by reduction of the intermediate fulvene with lithium aluminium hydride to give the cyclopentadienide. This was converted to its copper salt and then coupled with 1,8-diiodonaphthalene to give 53 in 72 % overall yield. Conversion of this to the iodoferrocene through treatment with sodium bis(trimethylsilyl)amide and ferrous chloride followed by coupling with cyclopentadienylcopper–dimethyl sulﬁde gave the building block 54 in 71 % yield (Scheme 10.31). Polymerization of 54 was achieved by treatment with Na[N(SiMe3 )2 ] followed by iron(II) chloride addition. Polymerization of 54 at room temperature performed over 10 days, followed by separation of low molecular mass products gave the dark purple polymer 55 in 71 % yield (Scheme 10.32). Polymer 55 is soluble in THF with Mn = 14 363 and Mw = 18 398 (bimodal GPC trace peaks at 26 574 and 17 584). Polymerization of an analogous branched monomer gave polymer 56 (Scheme 10.32) which was partially soluble in THF. The GPC analysis of this species shows a bimodal molecular weight distribution with components of 139 000 molecular weight.53 Synthesis of the ferrocene-containing polymers with 3,7-dimethyloctyl and decyl substituents have been also described51 and exhibit interesting material properties.

10.5

Properties of Ferrocene-Containing Polymers

This section details with the applications of ferrocene-containing polymers and features examples from all the key areas, though some of these will be covered in more depth elsewhere in this book. The focus is on actual applied examples rather than on compounds that only show promise. 10.5.1

Nonlinear Optical Activity

Ferrocene-containing polymers have been recognised as promising nonlinear optical (NLO) materials and combine advantages resulting from the ferrocenyl chromophore and the polymeric backbone (see also Chapter 9). Ferrocene itself is an effective NLO chromophore54 due to its high thermal/photochemical stability, rich chemistry (many derivatives are easily available) and stable oxidized form (ferrocenium cation). There is a general rule that to observe second harmonic generation, the NLOphore(s) must

C8H17

C6H13

C6H13

Fe

Fe

Fe

C8H17

C6H13

56

Fe

55

Fe

C6H13

C8H17

C6H13

n

Fe

n

Fe

C8H17

C6H13

C6H13

Scheme 10.32 Face-to-face polymers 55 and 56

C6H13

C8H17

C6H13

Fe

C6H13

Fe

Fe

C6H13

C6H13

C8H17

412 Ferrocenes: Ligands, Materials and Biomolecules

Ferrocene-Containing Polymers and Dendrimers

413

be oriented noncentrosymmetrically in the investigated material. Such an orientation is possible to obtain in noncentrosymmetric crystals, and in the case of NLOphores (including ferrocene NLOphores) within polymers a noncentrosymmetric environment can be generated by thermal-electric poling. This technique is based on warming the polymeric material up to glass transition (Tg ) temperatures. Heating is accomplished by applying an external electric ﬁeld and the procedure depends on the NLOphore microenviroment undergoing a structural reorganization. The sample is then cooled down with the electric ﬁeld still applied and at room temperature the electric ﬁeld is removed. Over time though, the NLOphore alignment comes back to the original positions and SHG signal decay. Nevertheless, second harmonic generation provides an excellent way of probing the molecular environment of the polymer matrix in close proximity to the NLOphore. To describe polymer relaxation behaviour below the glass transition temperature (Tg ) the Williams–Watts (WW) relation is often applied55 and, as examples, polymers 57–59 have been investigated (Scheme 10.33).56 Polymers 57 and 59 displayed the highest SHG signals when poling temperatures of Tg + 25 ◦ C were employed. At this temperature, the polymer is mobile enough to allow the NLOphores to readily orientate into the required noncentrosymmetric macroscopic structure. Applied temperature did not decompose the polymers. The polymers were also investigated in an attempt to establish time dependent decay of χ (2) value. The smaller NLOphore from 59 showed a lower initial SHG signal but with a better long term stability in comparison to the larger NLOphore in 57, i.e. the signal for 59 was observed for hours compared to seconds for 57. This indicates that molecular structure (packing) has a role to play in the different behaviour of the χ (2) for 59 and 57. The observed behaviour can be rationalized by the difference in density of chain packing for both ferrocene polymers. Since possessing a less sterically demanding chromophore, 59 can pack more efﬁciently and provide a more stable microenvironment for the ferrocene NLOphores, whereas such packing in the more bulky chromophore polymer 57 seems to be more difﬁcult. The study was the ﬁrst empirical probe of the orientational and relaxation behaviour of ferrocene-based organometallic NLOphores in a polymer

0.05 Me O

Fe

O

R NC 57 R= 4−BrC6H4 58 R= 4−Py 59 R= CN

Scheme 10.33 Ferrocene-containing nonlinear optical active polymers

414

Ferrocenes: Ligands, Materials and Biomolecules

environment by the use of χ (2) nonlinear optical spectroscopy. It clearly proved that NLOphore size has an effective inﬂuence on polymer chain packing and on SHG signal decay. The ﬁrst organometallic polymer to undergo successful alignment by corona onset poling and then exhibit the χ (2) NLO property of SHG was ferrocene poly(methyl methacrylate) 4 (a similar polymer to 57 and 59).57 A thin ﬁlm (in range of 2–3 µm) of polymer 4 was subjected to corona onset poling at 126 ◦ C for 30 minutes, cooled to 88 ◦ C for 30 minutes and then cooled to 40 ◦ C with the ﬁeld constantly on. SHG measurements were performed using the 1064 nm fundamental output of a Nd:YAG laser. It was found that the ferrocene-containing copolymer displayed an SHG efﬁciency approximately four times that of inorganic standard. 10.5.2

Biological Applications

It has been well documented that ferrocenium cations and selected ferrocene derivatives exhibit well-deﬁned biological activity58, 59 and much more detailed information is covered in Chapter 13. The most important problem to overcome in the bioorganometallic chemistry of ferrocene is its low solubility in water, which is obviously the common medium for biological applications. To address this problem, Neuse et al. reported the synthesis60–62 and antineoplastic activity tests63 of the water-soluble ferrocene polyaspartamide polymers 60–64 (Scheme 10.34). The in vitro tests were performed over a period of 72 hours against HeLa cell lines at concentrations up to 50 µg Fe ml−1 . The highest activity, with IC50 values of 2 and 7 µg Fe ml−1 , were determined for polymers 62–60 containing predominantly tertiary amine side functions that are protonizable at physiological pH. Lower activities of IC50 values 45 and 60 were exhibited by poly(ethylene oxide) ferrocene polymers 63 and 64. Impressive antiproliferative activity of some water-soluble ferrocenium species against murine and tumour lines is known. Neuse and coworkers have used ferrocene-functionalized polyamines to force the metallocene into close contact with the polymer backbone and thus probe bioactivity on the accessibility of the bioﬁssionable linkages.13, 64 10.5.3

Redox-Active Polymers

Electrochemical investigation of systems containing two linked ferrocenes is an important area of modern organometallic chemistry.65 Iron–iron interactions in such systems can be probed by a combination of electrochemical and spectroscopic techniques and such investigations become even more interesting when performed on ferrocenecontaining polymers. It is thought that such large redox-active systems can behave as molecular nanoscale wires and conducting devices. In Section 1.4.3, the synthesis of a rare class of ferrocene-containing polymer was mentioned. In fact this class represents the ﬁrst optically active ferrocene acetylene polymers.27 (S)-1-Iodoferrocene-2carbaaldehyde was used as the starting material, and was accessible in ten gram-scale due to the Kagan procedure.66 Further synthetic methodology featured a carbaldehyde reduction followed by hydroxyl group alkylation and Sonagashira coupling reaction leading to monomer 65. This monomer was polymerized under palladium conditions in a DMF/(i-Pr)2 NH mixture as the solvent (Scheme 10.35). The polymeric species was fractionated by chromatography to give the main product fraction 66

O

O

NH

OH

NH

x

O

H N

R2

NH

x

O

O

H N

PEO

NH

Fc(CH2)3C(O)NH

O

O

O

H N

63 R1 =

62 R1 =

Fc(CH2)3C(O)NH

y

y

61 R1 =

NH

NH

w

OH

O

N

N

N

H2N

O

H N

O

NH

NH

R2 =

R2 =

R2 =

R2 =

z−w

64 PEO =

O

O

10

O 18

O 2

x/y = 9

O 1

x/y/w/z−w = 6.42 : 0.67 : 1 : 0.25

O

O

O

x/y = 3

x/y = 3

Scheme 10.34 Water-soluble ferrocene-containing polyaspartamide polymers possessing in vitro anticancer activity

O

H N

1 O R

H N

60 R1 =

O

x/y = 6.7

Ferrocene-Containing Polymers and Dendrimers 415

416

Ferrocenes: Ligands, Materials and Biomolecules

OCH3

OCH3 I

H

Pd coupling Fe

Fe

n

65

66

Scheme 10.35 Palladium-catalyzed reaction of (Rp )-1-(ethynyl)-2-methoxymethyl-3-iodoferrocene (65)

in 45 % yield along with two smaller fractions with higher molecular weights. As the polymerization started from an enantiopure (ee > 98 %) planar chiral ferrocene, consequently the resulting polymer was optically active [α]D 20 = −198.0◦ per ferrocene unit. Based on GPC–LALLS (gel permeation chromatography with a low angle light scattering detector) measurements, the molecular mass of the polymer was estimated as up to 10 kDa. The polymeric material displays a very weak and broad IR absorption band at 2217 cm−1 resulting from the carbon−carbon triple bond and the electrochemistry was characterized by reversible redox processes. A squarewave voltammogram of the polymer showed two oxidation peaks at 0.65 and 0.80 V vs. Ag/AgCl in acetonitrile. This 145 mV separation between the two peaks corresponds closely to the redox separation of the ﬁrst E1 and second E2 potentials in the model dimer compound. For the tetramer 67, cyclic voltammetry and square-wave voltammetry shows four peaks, however the last one (at the highest potential) is noticeably smaller that the previous, suggesting some instability of the tetracation [67]4+ (Scheme 10.36). OCH3

OCH3

OCH3

OCH3 SiEt3

Fe

Fe

Fe

Fe

67

Scheme 10.36 Optically and redox-active ferroceneacetylene tetramer 67

The polymers and oligomers were investigated by UV/Vis spectroscopy and a linear relationship was found between λmax and 1/n (n = number of ferrocene units) in the oligomer. The metal–metal charge transfer bands of the mixed-valence oligoferrocene cations were examined by UV/Vis–NIR spectroscopy. The resonance exchange integrals calculated on the basis of spectral information from the metal–metal charge

Ferrocene-Containing Polymers and Dendrimers

Hex

Hex

1. 2eq n−BuLi or 2eq LiN(SiMe3)2 2. FeI2

Hex

Fe

417

Hex

(isomers) n

68

70

S Hex

Hex

1. 2eq n−BuLi or 2eq LiN(SiMe3)2 2. FeI2

S Hex

Fe

Hex

(isomers) n

71

69

Scheme 10.37 The Synthesis of poly(1,1 -ferrocenylenearylene)s by direct reaction of iron diiodide with organic ligands (Hex = n-hexyl) Hex Hex

Fe

Fe

Hex

Hex 72

Scheme 10.38 Cyclic dimer 72 (Hex = n-hexyl)

transfer (MMCT) bands were between 290–552 cm−1 . The synthesis of poly(1,1 ferrocenylenearylene)s 70 and 71 have been described by the reaction of the organic precursors 68 and 69 with iron(II) iodide with 71 and 50 % yields (Scheme 10.37).67 FAB mass spectrometry of 70 revealed peaks corresponding to masses of cyclic dimer (912 g mol−1 ) 72 (Scheme 10.38), trimer (1368 g mol−1 ) and tetramer (1824 g mol−1 ). Dimer 72 has been isolated by high-temperature sublimation from 70. Formation of cyclic dimer and trimer in 71 were conﬁrmed by GPC analysis and it seems that cyclization competes with the polymerization reaction under these conditions. The molecular weight data for the samples of 70 and 71 gave the following results: Mn = 4 × 103 , Mw = 4.2 × 104 , PDI = 10.5, DP = 8.8 and Mn = 3.6 × 103 , Mw = 5.3 × 104 , PDI = 14.6, DP = 7.8. Thermogravimetric analysis of 70 and 71 indicated a high thermal stability of these polymeric materials with decomposition temperatures (indicated by 5 % mass loss) of 406 and 440 ◦ C respectively. Cyclic voltammetry of 70 and 71 indicated moderate electronic interaction between the metal centres (as indicated by two reversible waves of equal intensity). The ﬁrst

418

Ferrocenes: Ligands, Materials and Biomolecules

wave (at −0.24 and −0.20 V vs. ferrocene/ferrocenium for 70 and 71 respectively) was followed by a second wave at higher potential (−0.07 for 70 and −0.01 V for 71). It is known that separation of the two oxidation waves (E) reveals an electronic interaction between the iron centres (or indeed other metals) in the polymer backbone. That electronic interaction can be mathematically given by the comproportionation constant (Kc ) according to the equation: Kc = exp(FE/RT) = exp(E/25.69) where T = 298 K and E is given in mV. Based on this, the Kc values for 70 and 71 are 750 and 1630 respectively and indicate class II mixed-valence systems. Incorporation of ferrocene to conjugated polymers, followed by deposition on gold electrodes along with full electrochemical, spectroscopic and microscopic (STM) characterization has been described by Liu et al.68 Thus, three types of ferrocene-grafted poly(p-phenylene-ethylene) polymers 73–75 have been synthesized by Sonagashira coupling (Scheme 10.39). In polymers 73 and 74, the ferrocene group is attached to the main chain by ethylene oxide tethers, unlike 75 where it is appended via methylene tethers. GPC indicated Mn = 34 020 and polydispersity 2.29 for 73, Mn = 27 000 and polydispersity 1.93 for 74, Mn = 17 300 and polydispersity 1.66 for 75. The three polymers exhibited similar redox behaviour. For example, the cyclic voltammogram of 75 recorded during potential cycling between −0.1 and +1.2 V vs. ferrocene/ferrocenium, shows two pairs of redox waves, the ﬁrst fully reversible wave peak at 0.55 V, corresponding to the oxidation of the ferrocene groups and the second irreversible wave peak at 1.02 V being related to the oxidation/reduction (p-doping/dedoping) of the polymer backbone. Following this, the authors deposited (immobilized) thin ﬁlms of 73–75 on a single gold crystal electrode (Au (1 1 1)surface). All the polymers have thioacetylterminated groups thus ﬂat orientation on the gold surface was predicted and expected. This assumption was conﬁrmed by both electrochemical and STM image data. Such topography can be regarded as typical for thiol self-assembly on Au (1 1 1). ˚ The cross-section provides information that the polymer ﬁlm of 73 is about 1.5–3.5 A thick and conﬁrms that it lies ﬂat on the gold surface. The cyclic voltammetric study OR1 O

OR2

OR1

S

S

O

n

R1O

R2O

R1O

73

R1/R2 = O(CH2CH2O)2Me/O(CH2CH2O)2C(O)Fc

74

R1/R2 = O(CH2CH2O)3Me/O(CH2CH2O)3C(O)Fc

75

R1/R2 = (CH2)9Me/(CH2)10C(O)Fc

Scheme 10.39 Ferrocene-grafted poly(p-phenylene-ethynylenes) with hydrophilic ethylene oxide or hydrophilic alkyl side chains

Ferrocene-Containing Polymers and Dendrimers

419

of the same polymer thin ﬁlm on a gold electrode showed that current is proportional to applied scan rate, indicating a kinetically fast redox couple strongly bound to an electrode surface. The potential peak-to-peak separation was less than 23 mV at a scan rate of 500 mV s−1 , indicating a very fast charge transfer and counterion movement through the ﬁlm along with very fast charge transfer from the ﬁlm to the electrode. The authors applied Laviron’s formalism to describe more precisely the electrontransfer rate (Ks ) of immobilized ferrocene polymers.69 Electron transfer of ferrocene from the thin ﬁlm of polymer 73 was calculated as being 31.69 ± 4.1 Hz for eight determinations when the scan rates were 100–900 mV s−1 . Surface coverage () of redox-active centres on the surface was found to be 1.5 ± 0.15 × 10−10 mol cm−2 . The electron transfer rate of the ﬁlm of 74, where the ethylene linker is longer by one ethylene oxide unit compared to 73, was 26.64 ± 3.5 Hz for seven determinations, thus being a little slower compared to 73. Interestingly, the ﬁlms of polymer 75 exhibit asymmetric surface waves and a kinetically slow redox couple strongly bound to the gold surface (the peak-to-peak potential separation is 129 mV even at 95 mV s−1 scan rate). The electron transfer rate of ferrocene from the ﬁlm of 75 was 0.37 ± 0.05 Hz for seven determinations – a value clearly smaller than for 73 and 74. An explanation of the experimental behaviour can be based on the fact that incorporation of hydrophobic alkyl side chains in 75 causes poorer solvation and results in irreversible electrochemical behaviour of the ﬁlm and affects (hinders) the ferrocene oxidation. The authors reported that an immobilized ﬁlm of polymer 75 displayed higher formal potential (E0 = 536 mV vs. Ag/AgCl reference) than that of polymer 73 (E0 = 384 mV). This is explained as being due to the effect of the hydrophilic ethylene oxide side chain in 73 that can stabilize the ferrocenium ion by electron donating properties, though such stabilization is hindered in 75 by the hydrophobic chain. From a commercial standpoint, thin ﬁlms of polymers 73–75 were remarkably stable and displayed the same current response even after more than 100 potential scan cycles. In addition, they did not lose their stability after one month. One of the early ﬁndings in the characterization of the polyferrocenylsilanes was that their cyclic voltammograms indicate the presence of two reversible oxidation waves due to the ferrocene/ferrocenium redox process (Figure 10.1).29, 70, 71 This is unusual when compared to the single oxidation wave found in many other ferrocene polymers (such as polyvinylferrocene and polyferrocenylalkynes72 ) and it is suggested that the initial oxidation at neighboring species is a consequence of signiﬁcant interactions between the iron atoms. Therefore, as one iron centre is oxidized, the nearby iron sites become more difﬁcult to oxidize and, thus, have to do so at a higher redox potential. The bridging linker between the ferrocene units is crucial in allowing a transfer (or not) of electron density and studies on a range of ferrocenylsilanes of different lengths indicate that these materials are class II materials, i.e. they show mixed-valence behaviour. Polyferrocenylsilanes exhibit unusual electrochromic behaviour, with a colour change from orange (neutral) to blue (oxidized).73 It is interesting to note that, in general, polyferrocenylsilanes are insulating, but when partially oxidised or doped by iodine for example, the materials can become strongly semiconducting. It has been shown that iodine doping is partially reversible and at high iodine concentrations an [I5 ]− species is formed. This intermediate could be an important factor in facilitating

420

Ferrocenes: Ligands, Materials and Biomolecules

Figure 10.1 The cyclic voltammogram of polyferrocenyldimethylsilane accumulated (0.1 M Bu4 N[PF6 ] in CH2 Cl2 ) at scan rates of 500 and 1000 mV−1 , showing two reversible oxidation waves separated by ca. 0.24 V (Reproduced with kind permission of the author)

metal–metal interactions and charge transport.74, 75 Devices have been manufactured and used in applications where lower values of conductivity are required. They can be used as charge dissipation materials with potential application as protective coatings for dielectrics.76 To facilitate charge transfer between ferrocene groups and a conjugated polymer backbone, a range of poly(thiophenes) with pendant vinylene ferrocene or cyanovinylene ferrocene units have been formed.77 The rationale was that by altering the redox properties of the main chain with respect to the pendant groups, polymers with a large range of electron donor/acceptor main chain/pendant groups could be easily prepared and their electronic/optical properties analysed. A range of ferrocenecontaining polythiophenes were synthesized via nickel(0)-mediated copolymerisation of ferrocene-functionalized thiophene and brominated 3-butylthiophene. Charge transfer between the main chain and the pendant ferrocene groups was evidenced by large photoluminescence (PL) quenching. With these hybrid polymers it was possible to fabricate organic p/n photocells that showed photoresponse over an entire visible window. The inclusion of the ferrocene units was found to dramatically increase short-circuit current density. During their search for new electrochemical glucose sensors, Foulds and Lowe described the preparation of electrodes covered by electropolymerized ferrocene– polypyrrole polymers and copolymers78 (see also Chapter 8). Electropolymerization conditions allowed ferrocene–polypyrrole to be doped by glucose oxidase. The entrapment of enzymes in polypyrrole ﬁlms provides a simple method of enzyme immobilization – an important component of certain biosensors.79 Ferrocene has also proved to be an efﬁcient electron transfer mediator between the reduced form of glucose oxidase and electrode surfaces.68, 80–82 In this context, [(ferrocenyl)amidopropyl]pyrrole (76), {[(ferrocenyl)amidopentyl]amidopropyl}pyrrole (77) and pyrrole have been used as suitable starting materials (Scheme 10.40).

Ferrocene-Containing Polymers and Dendrimers

O

O N

N H

76

N Fe

Fe

H N

N H

421

O

77

Scheme 10.40 Structural formulae of the ferrocenepyrrole monomers 76 and 77

Two types of electrodes were prepared: the ﬁrst did not contain glucose oxidase molecules trapped in the structure of the electrodeposited ferrocene polypyrrole polymer ﬁlm, whilst the second featured enzyme molecules trapped within the polymer. Anodic electropolymerization reactions of ferrocenyl–pyrrole monomers from aqueous sodium perchlorate solution was based on successive potential scans between 0 and +1.0 V vs. SCE at 100 mV s−1 ; and the electropolymerization process was monitored by observing the reduction in the wave peak height of the starting monomer and the appearance of peaks attributable to the immobilized ferrocene polymer. Firstly, the cyclic voltammetry of monomers 76 and 77 was examined in solution in the presence of glucose, and under the experimental conditions neither glucose nor glucose oxidase exhibited any observable electrochemistry. However, after adding glucose oxidase (glucose was present in the sample), to the samples, disappearance of the oxidation peak along with a large current ﬂow at oxidizing potentials were clearly observed, indicating the catalytic regeneration of ferrocene from the ferrocenium cation by the reduced enzyme. From this detailed analysis the authors found second-order homogenous rate constants (ks ) for the oxidation of reduced glucose oxidase with 76 and 77 to be 6.7 × 10−5 L mol−1 s−1 and 21.0 × 10−5 L mol−1 s−1 respectively. In the next step, the catalytic mediation of glucose oxidase (entrapped in the polymers of 76 and 77/copolymers of 76 and 77 with pyrrole ﬁlms) was investigated. Addition of glucose to the samples lead to an enhanced anodic current and a decreased cathodic current. At slow scan speeds (<10 mV s−1 ) the cathodic peak disappeared completely and the anodic peak appeared as a plateau. Reduced glucose oxidase thus appears to regenerate ferrocene from the ferrocenium cation within the polymer ﬁlm. Noticeably, the rate of reduction of the immobilized ferrocenium cation is markedly slower than that for the soluble ferrocene species. This can be explained via the steric hindrance created by the reduced mobility of the redox groups in the immobilized state. Polymers of 76 and 77 and their copolymers with pyrrole were catalytically active with glucose oxidase but with differing rates. Polymer ﬁlms electropolymerized with 76 were enzymatically regenerated faster than those with 77. The experimentally derived ratio of anodic current in the presence of enzyme (ik ) to the peak anodic current in its absence (id ) at a given scan rate may be taken as a measure of the catalytic efﬁciency of the modiﬁed electrodes.83 For scan speeds of 10 mV s−1 and homopolymer ﬁlms, the ratio (ik /id ) was relatively large (7.2 for 76 polymer and 2.75 for 77 polymer) for very thin ﬁlms with surface coverage ∼10−10 mol cm−2 . However, the ﬁgures fall rapidly to 1 with thicker ﬁlms (surface coverage ∼10−8 mol cm−2 ). Such

422

Ferrocenes: Ligands, Materials and Biomolecules

experimental observations can be interpreted as the enzyme being too large to diffuse into the ﬁlms to any signiﬁcant extent and that a decrease in catalytic efﬁciency with increasing ﬁlm thickness must be a function of the limited charge propagation throughout the ﬁlm. Overall, the catalytic efﬁciencies (ik /id ) for the homogenous reactions of 76 and 77 with glucose oxidase were 16 and 18, respectively, at 10 mV s−1 , which are signiﬁcantly greater than those for even the most efﬁcient non-ferrocenyl polymeric ﬁlms. Another interesting approach to the synthesis of water-soluble ferrocene polymers useful in glucose biosensing applications was described recently by Liu et al.84 Obtained after a multistep synthesis, the bromoalkyl polymer 78 (Mn = 15 000 g mol−1 , polydispersity = 1.86) was reacted with (dimethylaminomethyl)ferrocene in DMF solution at 60 ◦ C (Scheme 10.41). The post-polymerization quaternization reaction allowed attachment of ferrocene units to the poly(p-phenyleneethynylene) (PPE) backbone and thus the formation of polymer 79. The polymer was readily soluble in common organic solvents but not water, but addition of a β-cyclodextrine (β-CD) saturated water solution to dilute 79 in chloroform caused the polymer to transfer from the organic to the water phase overnight (Scheme 10.42). The phase transfer and formation of the water-soluble polymer 80 can be explained by strong complexation of side-chain ferrocene groups with β-CDs. Polymer 80 displays maximum absorption and emission peaks at 440 and 500 nm and is red-shifted compared with polymer 79. The authors selected polymer 80 and glucose oxidase to examine how a water-soluble polymeric redox mediator can facilitate electron transfer between an enzyme and the electrode.85 The electrostatically-bound polymer 80 and glucose oxidase complex were attached to a cystamine monolayer gold electrode via glutaraldehyde cross-linker methodology.85 When glucose is not present in the sample, immobilized oxidase glucose gives no response and the only observable electrochemistry belongs to that of the polymer itself. Cyclic voltammetry displays the linear relationship of peak current versus square root of the scan rate for the glucose biosensors and indicates the propagation of charge in the volume of the polymer network by a diffusion-like process. Mobile redox-active ferrocene units allow electron transfer since the rate of electron transferring collisions increases with the mobility of the tethered redox centres.86 After addition of glucose to the solution, the cyclic voltammetric responses of the glucose biosensor change dramatically via an increase in the oxidation current and a decrease in the reduction current (Figure 10.2). This change clearly indicated that the polymer-bound ferrocene/ferrocenium groups allow close contact with the FAD/FADH2 centres of the glucose oxidase and a resultant ﬂow of electrons from the enzyme to the electrode. The mechanism of the process comprizes a reduction of the glucose oxidase by the glucose (after its diffusion into the ﬁlm), then the reduced enzyme is oxidized/reactivated by ferrocenium units. The electrons are transported through the ferrocene polymer ﬁlm to the electrode surface. The disappearance of a reduction peak indicates that the ﬁlm is homogeneously maintained in the reduced state by the transfer of electrons from GOx (FADH2 ) to the ferrocenium sites, resulting in a bioelectrocatalytic oxidation current. The long ﬂexible hydrophilic side arms allow close interaction between the ferrocene moieties and the FAD/FADH2 centres of the glucose oxidase, opening a way to better electrical wiring between the relays themselves.

Ferrocene-Containing Polymers and Dendrimers

OR1 O

OR2

423

OR1

S

S

O

n

R1O

R2O

R1O 78

O

R1 :

O

O

O

R2 :

O

O

Br O

DMF / 60 °C FcCH2NMe2 Fc N

O

O

Br −

O

O

O

O

O

O

O O O

+

O

O

O

S

S

O

n

O

O

O

O

O

O

O O

O

O O

O

O

+

N

Br −

Fc 79

Scheme 10.41 Poly(p-phenyleneethynylene) ferrocene biosensing

polymers

useful

in

glucose

424

Ferrocenes: Ligands, Materials and Biomolecules 79

beta-cyclodextrin Fc +

N

O

O

O

O

O

O O O

O

O

O

Br −

O

O

O

S

S

O

n

O

O

O

O

O

O

O O

O

O O

O

O

Br −+ N Fc 80

Scheme 10.42 Organic to water phase transfer of polymer 79 mediated by β-CD addition

10.5.4

Ceramic and Magnetic Materials

The ﬁrst report on the pyrolysis of a resin containing poly(acetylferrocene) resulting in formation of a glassy matrix containing iron particles was published in the late 1970s.87 Further works from the ﬁrst half of 1990s showed that upon pyrolysis of high-molecular weight polyferrocenylsilanes at 1000 ◦ C, ceramic materials containing ferromagnetic α-Fe crystallites embedded in a C/SiC/Si3 N4 matrix were formed.88, 89 Latterly, a successful preparation of ceramics containing α-Fe2 O3 nanocrystals from a hyperbranched polyferrocenylsilane precursor has also been described.90 In 2000, a report by Manners

Ferrocene-Containing Polymers and Dendrimers

425

0.5 0.0

Current (mA)

−0.5 −1.0 −1.5

No glucose 4 mM glucose 6 mM glucose 8 mM glucose 11 mM glucose 15 mM glucose

−2.0 −2.5 −3.0 0.6

0.5

0.4 0.3 0.2 Potential (vs. Ag/AgCl)

0.1

0.0

Figure 10.2 Cyclic voltammetric responses of a glucose biosensor in the absence and presence of glucose in 0.1 M phosphate buffer (pH 7.0) containing 0.1 M NaNO3 at scan rate of 10 mV/s. Reprinted with permission from Ref. 86. Copyright 2003, American Chemical Society.

et al.91 illustrated how high molecular weight poly(ferrocenylsilanes) (formed from the ring opening polymerization of strained [1]silaferrocenophanes) are excellent starting materials in pyrolysis-induced ceramic fabrication. An important factor in ceramic synthesis from ferrocene-containing polymers is the ceramic yield. It strongly limits the utility, bulk properties and shape retention in the resulting ceramic. In this respect, cross-linked polymers are important starting materials in ceramic synthesis because of the reduction in the amount of volatile decomposition products. Thus Manners and Ozin et al. chose spirocyclic [1]silaferrocenophane 81 as a cross-linked ironcontaining polymer precursor to demonstrate shape retention in the resulting ceramic. The thermally-induced ROP reaction of 81 gives a cross-linked preceramic polymer with three possible silicon microenvironments 82–84 (Scheme 10.43). The network of the cross-linked preceramic polymer 82–84 has been characterized by powder X-ray diffraction (PXRD), solid state magic-angle spinning (MAS) multinuclear NMR, thermogravimetric analysis and pyrolysis mass spectrometry. Heating of the monomer was accompanied with a small weight loss (90 % ceramic yield) and contraction (10 %) but the overall shape was retained. The ceramic material formed was more dense, mechanically hard and attracted to a bar magnet. Scanning electron microscopy (SEM) of the bulk ceramic along with high-resolution transmission electron microscopy (TEM) revealed a smooth and nonporous texture, which was conﬁrmed by nitrogen adsorption measurements at −196 ◦ C, and high-resolution transmission electron microscopy (TEM) showed the presence of Fen nanoclusters in the ceramic product. Magnetic measurements conﬁrmed that the ceramic material contained superparamagnetic iron clusters. The zero-ﬁeld-cooled magnetic susceptibility versus temperature curve was consistent with the presence of superparamagnetic

426

Ferrocenes: Ligands, Materials and Biomolecules

Si

Fe

81

thermal ROP

Fe

Fe

Fe Fe Si

Si

Si Fe Fe

82

83

84

Scheme 10.43 The synthesis of cross-linked preceramic polymer by thermally–induced ring opening polymerization of spirocyclic [1]silaferrocenophane precursors

clusters with a blocking temperature (Tb ) of around 300 ◦ C. When the superparamagnetic sample was cooled below Tb in the absence of a magnetic ﬁeld, the magnetic dipoles of the clusters locked in a random orientation. As the sample was warmed in a magnetic ﬁeld, the dipoles slowly aligned. Above Tb , thermally induced ﬂuctuations of the magnetic dipoles dominated, and the clusters behaved as superparamagnets. All the evidence seemed to exclude the possibility of retention of the ferrocene sandwich structure – the ferrocene units being decomposed and releasing iron for the cluster aggregation. Under a nitrogen atmosphere, the ceramic yield was greater than 90 %, and the only volatile components that were detected were assigned to traces of ferrocene. Starting with the poly(ferrocenylsilanes) 82–84, there is an observation that with increasing pyrolysis temperature 600 ◦ C, the formation of superparamagnetic iron clusters are formed and dispersed within an amorphous ‘carbosilane’ matrix, which gradually transforms 700 ◦ C into larger clusters embedded in a composite ‘carbosilanegraphite’ matrix. Above 900 ◦ C, small amounts of silicon carbide (SiC) and silicon nitride (Si3 N4 ) are formed. With an increase in the pyrolysis temperature, an increase in Fen particle size was observed by PXRD and TEM. Magnetization measurements performed for ceramics obtained below and above 900 ◦ C showed that in ceramics

Ferrocene-Containing Polymers and Dendrimers

427

prepared below 900 ◦ C the iron nanoclusters are superparamagnetic, whereas in ceramics prepared at higher temperatures the larger iron clusters formed are ferromagnetic. Thus, by varying the pyrolysis temperature and type of the polymer precursor it was possible to control the magnetic properties of the resulting ceramic.91 In an attempt to gain a deeper understanding of ceramic formation from these ferrocene polymers, the same authors systematically varied pyrolysis temperatures and times under nitrogen to study the growth of α-Fe nanoparticles within their surrounding matrix.92 Three types of ceramic samples were studied: disk shaped, powder and ﬁlm, which were prepared by pyrolysis of disk-shaped polymer samples, powdered polymer samples and freestanding polymer ﬁlm samples respectively. Pyrolysis temperatures were varied from 500–1000 ◦ C and pyrolysis times were varied from 0 h (heating the sample to the set temperature and then immediately allowing it to cool) to 24 h at selected temperatures. The signal pattern of PXRD of the ceramics prepared using variable temperatures (from 500–1000 ◦ C for 2 h) shows signiﬁcant evolution of signals. The starting pattern of signals, which is characterized for poly(ferrocenylsilanes) ˚ and 2θ = 13◦ , d = 5.84 A, ˚ corre82–84 (two sharp peaks at 2θ = 11◦ , d = 6.98 A sponding to preferred Fe–Fe distances in the network) is lost with increasing temperature, and explained by a breaking of the polymer network and short-range order along with aggregation of iron atoms into α-Fe nanoparticles. At 600 ◦ C the (110) reﬂection ˚ and this sharpens at 700 ◦ C of α-Fe emerged as a broad halo (2θ = 45◦ , d = 2.03 A) indicating the presence of crystalline α-Fe nanoparticles. The peak assigned to the (110) reﬂection of α-Fe becomes narrower and more intense as the pyrolysis temperature increases. The formation of large α-Fe particles is attributed to enhanced iron mobility in the matrix above 700 ◦ C. Presumably above this temperature Si−C and C−C bond cleavage readily occurs facilitating matrix rearrangement and iron mobility. Crystallization of the matrix surrounding the α-Fe nanoparticles was ﬁrst observed at ˚ 700 ◦ C with the appearance of the (002) reﬂection of graphite at 2θ = 26◦ , d = 3.37 A. Appearance of a graphite crystallization signal was detected at the same temperature that crystalline α-Fe appeared (700 ◦ C). Progressive narrowing of the (002) reﬂection of graphite was observed with increasing temperature, and crystallinity improved up to 1000 ◦ C. The temperature at which graphite initially formed (700 ◦ C) was relatively low in comparison to the 1500–2100 ◦ C which is required to form pyrolytic graphite from thermal decomposition of hydrocarbons.93 The low temperature formation and crystallization of graphite in this system may indeed be catalyzed by α-Fe nanoparticles. The temperature dependence of the α-Fe particle size was also monitored using TEM. For the sample prepared at 600 ◦ C, no α-Fe particles were visible, and the matrix appeared homogeneous. By 700 ◦ C, there were distinct small and large α-Fe particles present as well as graphitic ribbons. As the temperature was further increased, the α-Fe particles visibly increased in size whilst maintaining a distribution of particle sizes. The ceramic materials prepared at ≥600 ◦ C were attracted to a bar magnet and the correlation between magnetic properties of the materials with α-Fe particle size was investigated. Therefore samples prepared at 650, 850 and 1000 ◦ C were analyzed by SQUID magnetometry. Overall, the magnetic saturation was gradual, consistent with the behaviour of superparamagnetic particles. The ceramic prepared at 1000 ◦ C displayed room temperature hysteresis and a small remnant magnetization (1 emu g−1 ) consistent with a soft ferromagnet, thus indicating that the α-Fe particles had become

428

Ferrocenes: Ligands, Materials and Biomolecules Fe Fe

Fe

Fe

Fe Fe Fe Fe Fe

Fe

Fe Fe Fe

Fe

Fe

Fe Fe

Fe Fe Fe

(i)

Fe Fe

Fe

Fe

Fe

Fe Fe

Fe Fe Fe

Fe Fe

Fe Fe

(ii)

Fe Fe Fe Fe Fe Fe Fe

Fe Fe Fe Fe Fe Fe Fe FeFe Fe

Fe

Fe

(i) iron atom release from ferrocene polymer

(ii) nucleation and growth of iron anoparticles

Figure 10.3 The proposed mechanism of magnetic ceramic formation (Based on a ﬁgure from Ref. 92)

large enough to display ferromagnetism. SQUID magnetometry thus demonstrated a transition from small superparamagnetic α-Fe particles to larger ferromagnetic αFe particles at ca. 900 ◦ C. Therefore, by varying the pyrolysis temperature, the αFe particle size and magnetic properties of the resulting ceramics could be tuned. Based on detailed multitechnique studies, a model of ceramic formation from ferrocene–polymeric precursors was proposed and shows how the genesis of magnetic ceramics from cross-linked polyferrocenylsilanes depends on the pyrolysis conditions. Visualization of the process is shown in Figure 10.3. Upon heating the cross-linked polymer 82–84, expansion of the network occurs and further heating results in iron atom ‘release’ from the ferrocene moieties. UVVis/NIR spectroscopy showed the disappearance of ferrocene units by 650 ◦ C, with the polymer matrix being dismantled as nucleation and growth of the α-Fe particles occurs. Disappearance of polymer order before 600 ◦ C was detected by PXRD and coincided with the initial observation of α-Fe particles. At 700 ◦ C, the α-Fe particles appeared to catalyze the formation of graphitic ribbons in the ceramic; both species then continued to grow in size with temperature. The α-Fe particles form with a bimodal particle size distribution, possibly due to the formation of larger particles at grain boundaries and defect sites as a result of faster intergrain iron atom diffusion. Magnetic measurements demonstrated that as the α-Fe particles became larger, they underwent the expected transition from superparamagnetic to ferromagnetic behaviour. Particles formed below 900 ◦ C contained α-Fe particles that were smaller than a single Weiss domain and behaved as superparamagnets, whereas larger particles prepared at 1000 ◦ C conferred ferromagnetic properties to the ceramic. Pyrolysis of poly(ferrocenylsilanes) can take place inside mesoporous silica MCM41 leading to the new mesoporous–ceramic modiﬁed materials.94, 95 Such pyrolytic methodology of ceramic formation inside structurally well-deﬁned mesoporous materials aims to control the particle size of the ceramic formed. Mesoporous silica MCM-41 was obtained for ﬁrst time by Kresge et al.96 ; it possesses an internal architecture of ordered channels arranged in a hexagonal lattice. Uniform channel sizes readily varied ˚ in diameter, making this an attractive host material. Manners and between 20 and 100 A Ozin94, 95 loaded mesoporous silica MCM-41 with different amounts of strained silylferrocene monomers 34 and 81 by vapor deposition techniques. With a full loading of

Ferrocene-Containing Polymers and Dendrimers

429

34 nearly all of the hydroxyl sites were reacted with monomer to form ring opened species and it appeared that the ring opened and oligomeric products were bound to the surface SiO3 units inside the channels of MCM-41. Of course, some material was also deposited on the exterior surface of MCM-41 but the surface area inside MCM-41 is much greater than the outside. Similarly, spectroscopic methods provided evidence of ring opened monomer 81 and its oligomers tethered to the walls of the MCM-41 host with the silacyclobutane rings intact. The schematic model for channels of MCM-41 loaded with 34 is shown in Figure 10.4.95 When 34 was studied by DSC, the ferrocenophane melted near 75 ◦ C and then underwent exothermic ROP. DSC of fully loaded MCM-41 with 34 revealed a broad exotherm corresponding to ROP of 34 between 75 and 200 ◦ C. The absence of an endothermic melt transition for 34-MCM-41 indicated that the free monomer was not crystalline inside the channels in contrast to bulk 34. In the case of ROP polymerization of 81 it was found that this monomer undergoes thermal ROP of the silacyclobutane ring at 240 ◦ C in the bulk. For both polymerized 34 and 81 inside the MCM-41 channels, the newly polymeric MCM-41 material was characterized by multinuclear NMR along with UV-Vis/NIR techniques. All the peaks in the NMR spectrum were broad indicating that the species are immobile as expected for cross-linked polymers and monomers pinned to the channels. Polymerized 34-MCM-41 was heated at 900 ◦ C for two hours under a slow ﬂow of nitrogen to yield ceramic materials. During that heating, the starting yellow powder was transformed into a ﬁne black powder that was attracted to a magnet. A small amount of ferrocene sublimed from the material during pyrolysis and overall, ceramic yields of 69–86 % were obtained. Compared to the bulk pyrolysis of the poly(ferrocenylsilane), the iron particles in polymerized 34-MCM-41 are substantially smaller and, indeed, these α-Fe particles were small enough to be superparamagnetic. Additionally, neither graphite nor α-Si3 N4 were observed in the pyrolyses. The pyrolysis of 81-MCM-41 yielded ceramic in 73–85 %, and although the silica underwent further polymerizationinduced contraction during heating, the hexagonal mesoporous structure of the host was maintained up to 1000 ◦ C. PXRD indicated that the ferrocenylsilane guest had already undergone a large structural transformation by 500 ◦ C. In the PXRD pattern of the sample prepared at 1000 ◦ C, peaks assigned to both high temperature γ -Fe and low temperature α-Fe nanoparticles were apparent as well as a peak due to magnetite (Fe3 O4 ). TEM images of the 81-MCM-41 ceramic showed very small iron nanoparticles that were conﬁned to single channels (when the pyrolysis temperature was 600 ◦ C), but at 1000 ◦ C the particles are much larger and nearly round. Overall, there is compelling evidence that the iron nanoparticles are inside the MCM-41 channels and form a new and exciting class of material. In related work, the same team formed a range of microspheres from polyferrocenylsilanes.97 Small spherical, colloidal particles are of interest for a range of applications, including catalysis and chemical sensing, and introduction of a redox-active ‘coat’ opens up other exciting possibilities. By precipitation polymerisation methodology and varying the reaction conditions, a variety of morphologies were obtained. Upon oxidation, the microspheres undergo electrostatic self-assembly with silica spheres to form core-shell composite particles. The oxidation and therefore charge of the materials can be controlled, and variation of the pyrolysis conditions leads to ceramics with tunable magnetic properties from a superparamagnetic state to a ferromagnetic one.

430

Ferrocenes: Ligands, Materials and Biomolecules

Figure 10.4 Graphical representation of MCM-41 channels loaded with 34. Reprinted with permission from Ref. 95. Copyright 2000, American Chemical Society.

Poly(vinylferrocene) (1) has found application as a material for AFM tip modiﬁcation. Such modiﬁed tips have been used to study adhesion/adhesion forces by controlling the degree of adhesion through the selective oxidation or reduction of the polymer ﬁlm.98 The AFM tip before (left) and after (right) coating with 1 is shown in Figure 10.5. Similarly, a newly-developed technique termed ‘redox probe microscopy’ (RPM) has been described. This is based on a metal-coated SPM probe which is modiﬁed with a redox-active ﬁlm (for example a ferrocene polymer) so that tip-sample interactions, such as adhesion, can be modulated by the electrochemical potential.99

Ferrocene-Containing Polymers and Dendrimers

431

Figure 10.5 (A) Schematic representation of a gold substrate and gold coated AFM tip, modiﬁed with a thin ﬁlm of poly(vinylferrocene) (PVF). The tip and ﬁlm were mounted inside a glass ﬂuid cell in an atomic force microscope and held under potential control. (B) SEM image of a clean, gold-coated AFM tip. (C) SEM image of a gold-coated AFM tip modiﬁed with a 35 nm thick ﬁlm of PVF. Reprinted with permission from Ref. 98. Copyright 1996, American Chemical Society.

This tool has been shown to be useful in mechanical manipulation of objects at the micrometer scale. The manipulator tip was based on commercially available silicon ˚ chromium adhesion nitride (Si3 N4 ) and then coated by thermal evaporation of a 25 A ˚ layer of gold. Finally, the tip was chemically modiﬁed with layer followed by a 500 A 1 by its electrochemical deposition from dichloromethane. Surface manipulation of adsorbates through electrochemically controlled/modulated adhesion is via the tip’s interaction (through adhesive effects) with an adsorbate controlled by either strong or weak interactions depending on the oxidation state of the redox center and the nature of the adsorbate. The adsorbate is assumed to bind with moderate (intermediate) strength to the substrate surface. At a potential where the redox centre is in the reduced (or ground Fe2+ ) state and its interaction with the adsorbate is weak, the tip cannot overcome the adsorbate/substrate interaction and thus cannot displace/move the adsorbate. On the other hand, at a potential where the redox center is in the oxidation state that interacts strongly with the adsorbate, it should be possible to bind in a controlled fashion and deliberately move and place the adsorbate in any arbitrarily desired location by a series of potential steps.99 The process is shown in Figure 10.6. In practice this approach has been described to reproducibly bind and reposition 15 µm, sulfonic acid terminated chromatography beads. By using a programmed sequence of potential steps, the beads can be moved and placed in any location. Importantly, since the beads exhibit some interaction with the substrate, once moved and placed in a desired position, they remain at that location. While bound to the tip, the bead can be moved to a new region/location on the sample. Once the desired new location is reached, the potential of the tip is returned to a value where the poly(vinylferrocene) ﬁlm is once again neutral resulting in the ‘release’ of the bead from the tip’s surface. As the manipulations can be carried out at room temperature, the generated structure remains indeﬁnitely stable, and it is hoped that novel structures can be generated by this technique.

432

Ferrocenes: Ligands, Materials and Biomolecules A.

B.

C.

D.

E.

F.

Figure 10.6 Schematic depiction of the manipulation of a molecular adsorbate with an RPM tip which can be switched between strongly binding (dark grey) and weakly binding (light grey). By controlling the movement and redox state of the tip, it can be used to move and reposition adsorbates. The adsorbate is assumed to bind with moderate strength to the substrate. Reprinted with permission from Ref. 99. Copyright 2001, American Chemical Society.

An interesting application of AFM-based single-molecule force spectroscopy (SMFS) into the investigation of the elastic/mechanical properties of single chains of poly (ferrocenyldimethylsilane) (34) and poly(ferrocenylmethylphenylsilane) (85) in their normal and oxidized forms has been described by Zhang et al.100 The single-chain elasticity of the two polymers is similar to that of conventional polymers, such as poly(N isopropylacrylamide) and poly(acrylamide), even though they contain ferrocene units. This result supports the previous understanding that the ferrocene iron atom acts as a freely rotating ‘molecular ball bearing’ in polyferrocenylsilane materials.101, 102 The two polymers investigated show similar elasticity in their normal forms, though bearing different side groups. However, in the oxidized forms the difference in their enthalpic elasticity due to steric effects is ampliﬁed, thus allowing a differentiation of the effect of the side groups on single-chain elasticity between 85 and 34. Applications of block copolymers featuring polyferrocenylsilane have recently been published.103, 104 A method of fabricating nanotextured silver surfaces using a template of the self-assembled block copolymer polystyrene-b-polyferrocenylsilane has been reported.103 The block copolymer was used to produce periodic arrays of inorganic oxide nanostructures, which when used as a template produced a roughened silver surface with periodically ordered nanoscale features. The nanotextured silver surfaces exhibit very high surface-enhanced Raman scattering (SERS) and it is hoped that the simple method for producing SERS substrates offers commercial possibilities and the fabrication of sensor chips. From polyferrocenylsilane–blockxy –polysiloxane diblock copolymers, carbon nanotubes with small and tunable diameters have been formed.104 The iron-containing nanostructures produced from thin ﬁlms of the block copolymers are catalytically active for the initiation and growth of high density, small diameter carbon nanotubes (CNTs) – indeed, the tube diameter and density can be tuned by adjusting the chain lengths of the block copolymer. CNTs with a diameter of around one nanometer have been synthesized and it has been shown that the yield of defect-free CNTs is sensitive to catalyst size. Thus uniform-sized catalyst-containing nanostructures are crucial for producing consistent and high quality CNTs with a minimum number of defects and amount of amorphous carbon.

Ferrocene-Containing Polymers and Dendrimers

10.6

433

Ferrocene-Containing Dendrimers

As ferrocene-containing dendrimers are not covered elsewhere in this book, their emergence as a new type of ferrocene-containing polymer will be detailed here. In simple terms, dendrimers can be thought of as a combination of ‘classical’ hyperbranched polymers and ‘typical’ small organic molecules. More precisely, dendrimers can be deﬁned as globular, branched, fractal-like macromolecules with three-dimensional shape, synthesized by iterative/repetitive procedures. Differentiation between dendrimers and hyperbranched polymers can be based on the methods of preparation. Polymers are usually prepared by one-step, single-pot reactions, yielding materials with high or very high molecular mass distributions. The synthetic methodologies applied to the synthesis of dendrimers are generally referred to as divergent105, 106 and convergent107 and will not be discussed in detail here. Rationale for the introduction of ferrocene within dendrimeric structures is generally the same as for the synthesis of ferrocene-containing polymers – a chance to introduce the unique properties of the ferrocene unit (redox, magnetic, optical and electronic properties) into an organic superstructure. The position of the ferrocene unit in dendrimeric structures has been classiﬁed as follows (Scheme 10.44): A ferrocene as the dendrimer core; B ferrocene as terminal (capping) groups; C1,n ferrocene locked by a 1,n-mode (n = 2, 3 or 1 ) of substitution inside the dendrimer’s branches. Due to the limited space in this chapter, only a few representative examples of ferrocene-containing dendrimers are focused on and their properties described. The selection is rather arbitrary but for more details relevant review articles are available.108–111 A most interesting feature of ferrocene-containing dendrimers is their redox behaviour and, due to the globular-like architecture of dendrimers, they can be thought of as protein mimics and electrochemical investigations can shed new light on metalloenzymes modes of action. Ferrocene ether–amide dendrimers 86–88 are representative examples of type A in our classiﬁcation. They have been prepared via a convergent coupling strategy (Scheme 10.45).112 Electrochemical measurements on 86–88 indicated increasing redox potential for the higher generation dendrimers over the lower ones, i.e. the redox potentials (measured in dichloromethane as a solvent and given in Volts relative to a Ag/AgCl reference electrode) are: 0.875 for 86, 0.914 for 87 and 0.953 for 88. The so-called ‘dendritic effect’ caused by the branched shell is different in different solvents. In weakly polar, non-hydrogen bond forming solvents, branched fragments increase E1/2 of the encapsulated ferrocene, but in methanol this effect is rather weak. This can be explained as the ability of methanol to form branched chains of hydrogen bonds and therefore open up the ferrocene centres to electronic communication with the environment. The effect of the electrolyte has been also studied and there is a signiﬁcant inﬂuence on the redox potential by the total polarity of the medium. Despite shielding of the redox-active ferrocene centers by the organic branches, cyclic voltammograms of both 87 and 88 are reversible, suggesting a lack of electron transfer disruption between the electrode and ferrocene moiety.

434

Ferrocenes: Ligands, Materials and Biomolecules Fc Fc Fc

Fc Fc Fe

Fc

Fc

Fc

Fc

Fc Fc

Fc

A Fc

Fc Fc Fc B

Fe

Fe

Fe

Fe

Fe

Fe

Fe

Fe

C1, n

Scheme 10.44 General classiﬁcation of ferrocene-containing dendrimers

Ferrocene-Containing Polymers and Dendrimers

435

O OMe

O MeO

O O

O N H

O

Fe

H N

MeO

Fe

H N

O O

O

O O

N H

O O

OMe O

MeO 86

87

O

O MeO

OMe O

O

MeO

O

O

OMe

O

O

O

OMe

O

O MeO O O

HN O

O

H N

O O

O

O

MeO

O

O

OMe

O O

O O

OMe

O

O

O

O

O

NH

H N

MeO

OMe

O

O

O

O

NH

O

O

O

MeO O

Fe

O

O N H

O H N

OMe

O

MeO O

O

N H

O

O

O

O MeO

OMe

O O

O

MeO

MeO

O MeO 88

Scheme 10.45 Ferrocene ether–amide dendrimers of class A

The relationship between dendritic architecture and redox properties of ferroceneencapsulated dendrimers is complex. In contrast to 86–88, in the asymmetric dendrimers 89–91 (Scheme 10.46), increased branching levels favours oxidation of the ferrocene units,113 as illustrated by their decreasing redox potentials. For example, in dichloromethane, E1/2 potentials quoted in Volts relative to Ag/AgCl reference electrode are as follows: 0.63 for 89, 0.60 for 90 and 0.54 for 91. However, in the case of dendrimers 92 and 93 (Scheme 10.47), polyether linear branches do not affect the oxidation potentials and there is little decrease in the electrochemical reversibility.114

436

Ferrocenes: Ligands, Materials and Biomolecules O O O O

O

O

O

O

N H

O

O

O

O N H

O N H

O

Fe

O

O

HN

O

Fe

O O

NH

O

O

89

90

O O

O O

O O

O

O

O

O

OH O O

O

O

O

O

O

O

NH O

O

N H

O

O O

H N O

O

O

O

NH

O

O

O

O N H

O

O

O

N H

NH

O O

Fe O NH

O

NH

O

O O

O

O

NH

O

O

NH

O

O

HN

O

O

HN

O

O

O

O

O

O O

O

O O

O O

O

O

O

O

O

O

O

O

O O

O

O

O

91

Scheme 10.46 Structures of the asymmetric ferrocene-containing dendrimers

Ferrocene-Containing Polymers and Dendrimers

437

O O

O O

O

O

O

Fe

O

O O

O O

O

O

O

O

O

O

O 92

O

O O O

O O

O

O

O O

O

O O O

O

O

O

O

O O

Fe

O

O

O O

O O

O

O

O

O

O

O O O O

O

O

O

O 93

O O

Scheme 10.47 Chemical structures of the class A polyether ferrocene dendrimers

In dichloromethane, E1/2 potentials quoted in Volts are 0.460 for 92 and 0.466 for 93 (the potentials are quoted relative to ferrocene/ferrocenium reference). Beautiful, rosette-like class B poly(propylenimine) ferrocene dendrimers with 4, 8, 16, 32 and 64 ferrocenyl moieties have been described.115 For all the family of these dendrimers, a single redox process takes place, implying a simultaneous multi-electron transfer of all the ferrocene capping groups at the same potential. In addition, solvent dependent redox behaviour for the dendrimers is observed. Scheme 10.48 shows class B dendrimers 94–97, all of which exhibit electronic communication between the iron centers.116 Signiﬁcant electronic interaction is illustrated in the values of the comproportionation constant Kc . The Kc values are as follows: 1104 for 94, 1630 for 96, ∼ =507 for 95 and 97. These numbers conﬁrm the presence of partially oxidized class II mixedvalence species according to the Robin–Day classiﬁcation.117 Dendrimers 95 and 96 have also been used as materials for electrode modiﬁcation.116 Another interesting feature of some ferrocene-containing dendrimers is their liquid–crystalline behaviour.118, 119 More information about ferrocene-containing liquid crystalline materials is presented in Chapter 11.

438

Ferrocenes: Ligands, Materials and Biomolecules Fc

Fc Fc

Si

Si

Si Fc

Fc

Si Fc

Fc

Fc

Si

Si

Si

Si

Si

Fc

Fc

Si

Si

94

95

Fc

96

Si

Si

Fc

Fc

Fc Fc

Fc

Si

Si

Fc

Fc

Si

Si

Fc

Si Fc

Si

Si

Si

Fc Si

Si

Si Fc

Fc Si

Si

Fc

Fc

Si

Si

Fc

Si

Si Fc

Fc

Fc

Fc

97

Scheme 10.48 Ferrocene dendrimers of class B

Fc

Ferrocene-Containing Polymers and Dendrimers

439

The interesting possibility of control/switching of the ferrocene oxidation potential by protonation or N -methylation of alkylamine fragments in dendrimers 98–101 was reported by Kaifer et al. (Scheme 10.49).120 From cyclic voltammetric studies, it seemed that as long as the nitrogen atoms in the organic branches remained nonprotonated (nonalkylated), all the ferrocene centers behaved as noninteracting and equivalent redox centers. However, electronic differentiation of the ferrocene centers between two different iron oxidation states became possible after protonation (alkylation) of the nitrogen atoms. Fc

Fc Fc

Fc N

N

N+

N Fc

Fc

Fc

Fc

98

99

Fc Fc

H N

Fc N

Fc Fc

Fc Fc

N

N

N

N

Fc Fc

Fc

N

N

Fc

N

N

N

N

Fc Fc

Fc

Fc 100

101

Scheme 10.49 Ferrocene dendrimers of class B with alkylamine branches available for protonation and alkylation

Ferrocenyl silicon-based dendrimers 102–105 (Scheme 10.50) have been applied as electron mediators in amperometric sensing of glucose.121 The steady-state amperometric response of carbon paste electrodes containing 102, 103, 104, 105 and glucose oxidase as a function of the glucose concentration revealed the fast response of the system to the addition of glucose. This response can be correlated to the size of the dendrimer, the number of ferrocene units and the nature of dendritic branches. For equimolar amounts of the ferrocene, the eight ferrocenes contained in the long branches of dendrimers 103 and 105 mediate electron transfer more efﬁciently than the short branched four ferrocene dendrimers 102 and 104. Additionally, dendrimers 104 and 105, in which the ferrocenyl units are attached to the branches by ﬂexible methylene bridges, are recognized as being more effective at mediating electron transfer between reduced glucose oxidase and the carbon paste electrode in comparison to the equal numbered ferrocene-containing dendrimers 102 and 103.

440

Ferrocenes: Ligands, Materials and Biomolecules Fc Fc Si Si

Fc

Si

Si

Fc Si

Fc

Si

Si

Fc

Si

Si

Si

Fc

Si

Si

Si

Si

Fc

Si

Fc Si

Si

Si

Fc

Fc

Fc 102

103 Fc Si Si Fc Fc

Si

Fc

Fc Si

Si

Si

Si Si

Si Fc

Si

Si

Fc

Si Si

Si Fc

Si Si

Fc

Si

Si

Fc

Fc Fc 104

105

Scheme 10.50 Chemical structures of the class B ferrocenyl silicon-based dendrimers used for glucose sensing

Ferrocene-containing dendrimers can play a role as supramolecular redox-active sensors useful in the recognition of small inorganic anions122, 123 and metal cations123 (see also Chapter 8). To be exact, dendrimers such as 106, obtained by so-called ‘click’ chemistry, are able to sense selected oxo anions (ATP2− , H2 PO4 − ) and metal cations (Pd2+ , Pt2+ , Cu2+ , Cu+ ) (Scheme 10.51). The sensing act is observed by changes in cyclic voltammograms of the dendrimer. For oxo anions, the new wave appears at a less positive potential compared to the potential of initial (dendrimer without addition of anion) wave. This suggests that the dendrimer–oxo anion ‘supramolecular complex’ undergoes easier oxidation compared to the dendrimer alone. In the case of analogous metal cation dendrimer complexes, the new waves appear at more positive potential compared to initial waves.

Ferrocene-Containing Polymers and Dendrimers Fc

Fc N

N

N

Fc

441

N

N

N

N N N

Si Si

Si

N Fc

N

Si

Si

N

N

N

Si

Si N Fc N

N

Si

Si

N

N

N

Fc N

N

N Fc

Fc N

N

N

Fc N

106

Scheme 10.51 Chemical structure of the class B ferrocenyl–triazine dendrimer applied for sensing of ATP2− , H2 PO4 − anions and Pd2+ , Pt2+ , Cu2+ , Cu+ cations

The application of dendrimer 107, featuring 48 peripheral ferrocenes, for carbon monoxide gas detection has been reported (Scheme 10.52).124 The sensing device built around this dendrimer is able to generate an analytical signal 50 seconds after the ﬂow of carbon monoxide is turned on, and was found to be very sensitive and selective to carbon monoxide gas. The response of the sensor, as measured by its conductance as a function of the increasing concentration of carbon monoxide gas, displayed a linear character up to 40 % carbon monoxide volume concentration, but above that concentration the sensor becomes saturated. The C1,1 type of ferrocene-containing dendrimers 108 have been reported (Scheme 10.53)125 and the introduction of chiral ferrocenes into phosphorus-containing dendrimers of type C1,2 have also been recently described.126

10.7

Conclusions and Outlook

As witnessed by the wide range of examples detailed in this chapter, the area of ferrocene-containing polymers is one full of excitement and surprises, though there remains great scope for further development.

442

Ferrocenes: Ligands, Materials and Biomolecules Fc

R

Fc O

Si

O Si R

R Si

Fc

Si

O

Si Si

Si

O

R

O O

Fc

Si

Si

107

O O Fc

Si O

Si

R=

O

Si

Si

O O O

Si

Fc

O

Si O

O Si

Si

O

O

Si

Fc

Si O

O Fc

O

Si O

Si

Si

O

Fc

O

Si

O

Si Fc

Si

Si Fc

Fc

Scheme 10.52 Chemical structure of the class B ferrocenyl dendrimer used in the sensing of carbon monoxide gas

S Fe

C N N P H

CHO

O Fe

S P O

S H C N N P

Fe O 2

108

Scheme 10.53 An example of the C1,1 class of ferrocene dendrimers

2

3

Ferrocene-Containing Polymers and Dendrimers

443

The ROP and related methods, pioneered by Manners and co workers, have given rise to a huge raft of new polymers and subsequently from these, materials. The application of the ROP methodology is now mature, but there is the promise of new materials, via copolymerization and blending and the development of ‘smart’ polymers. Synthetic methodology in this area remains a challenge and thus the application of these ferrocene-containing polymers remains underdeveloped. The ﬁeld is ripe for another synthetic breakthrough, such as ROP, to produce further good quality polymers with true polymer properties and characteristics. There remain too many examples of large oligomers, rather than polymers, with limited synthetic methodology resulting in niche polymers or ‘one-off examples’. Ferrocene-containing dendrimers are an exciting and emerging new class of compound and their properties and application have still yet to be harnessed. There is certainly room for further synthetic creativity to produce hitherto unknown combinations of elements within the polymeric structures – and with this further and new applications and uses found. With improvements in the formation of the polymers themselves, better scale up procedures and a greater understanding of their stability and reactivity, further dramatic ﬁndings and exciting applications are sure to result.

References 1. I. Manners, Synthetic Metal-Containing Polymers, Wiley-VCH Verlag GmbH, Weinheim, Germany (2004). 2. R.D.A. Hudson, J. Organomet. Chem., 2001, 637–639, 47–69. 3. J.F.S. Arimoto, A.C. Haven Jr, J. Am. Chem. Soc., 1955, 77, 6295–6297. 4. C.U. Pittmann Jr, J.C. Lai, D.P. Vanderpool et al. Macromolecules, 1970, 3, 746–754. 5. O. Nuyken, V. Burkhardt, T. Pohlmann, M. Herberhold, Makromol. Chem., Macromol. Symp., 1991, 44, 195–206. 6. O. Nuyken, V. Burkhardt, C. Hubsch, Macromol. Chem. Phys., 1997, 198, 3353–3363. 7. M.E. Wright, E.G. Toplikar, R.F. Kubin, M.D. Seltzer, Macromolecules, 1992, 25, 1838–1839. 8. K.H. Pannell, J.M. Rozell, J.M. Ziegler, Macromolecules, 1988, 21, 276–278. 9. R. Deschenaux, I. Jauslin, U. Scholten et al. Macromolecules, 1998, 31, 5647–5654. 10. H.R. Allcock, J.A. Dodge, I. Manners, G.H. Riding, J. Am. Chem. Soc., 1991, 113, 9596–9603. 11. P. Wisian-Neilson, R.R. Ford, Macromolecules, 1989, 22, 72–75; P. Wisian-Neilson, C. Zhang, K.A. Koch, Macromolecules, 1998, 31, 1808–1813. 12. D. Albagli, G. Bazan, M.S. Wrighton, R.R. Schrock, J. Am. Chem. Soc., 1992, 114, 4150–4158; D. Albagli, G. Bazan, R.R. Schrock, M.S. Wrighton, J. Phys. Chem., 1993, 97, 10211–10216. 13. M.G. Meirim, E.W. Neuse, G. Caldwell, J. Inorg. Organomet. Polym., 1998, 8, 225–236. 14. V.V. Korshak, S.L. Sosin, V.P. Alekseeva, Dokl. Akad. Nauk SSSR, 1960, 132, 360–365. 15. V.V. Korshak, S.L. Sosin, V.P. Alekseeva, Vysokomol. Soedin, 1961, 3, 1332–1339. 16. A.N. Nesmeyanov, V.V. Korshak, V.V. Voevodski et al. Dokl. Akad. Nauk SSSR, 1961, 137, 1370–1375. 17. H. Rosenberg, E.W. Neuse, J. Organomet. Chem., 1966, 6, 76–85. 18. A.N. Nesmeyanov, V.N. Drozd, V.A. Sazonova et al. Izv. Akad. Nauk SSSR, Otd. Khim. Nauk , 1963, 667–669. 19. E.W. Neuse, L. Bednarik, Macromolecules, 1979, 12, 187–195.

444 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63.

Ferrocenes: Ligands, Materials and Biomolecules K. Sanechika, T. Yamamoto, A. Yamamoto, Polym. J., 1981, 13, 255–261. T. Yamamoto, K. Senechika, A. Yamamoto et al. Inorg. Chim. Acta, 1983, 73, 75–82. P. Park, A.J. Lough, D.A. Foucher, Macromolecules, 2002, 35, 3810–3818. E.W. Neuse, H. Rosenberg, J. Macromol. Sci. Rev., Macromol. Chem., 1970, C4, 1–8. E.W. Neuse, G.J. Chris, J. Macromol. Sci., Chem., 1967, A1, 371–375. C.U. Pittman Jr, J. Polym. Sci., Polym. Chem. Ed., 1967, 5, 2927–2937. T. Itoh, H. Saitoh, S. Iwatsuki, J. Polym. Sci., Polym. Chem., 1995, 33, 1589–1596. H. Plenio, J. Hermann, S. Sehring, Chem. Eur. J., 2000, 6, 1820–1829. A.G. Osborne, R.H. Whiteley, J. Organomet. Chem., 1975,101, C27–C28; M. Herberhold, Angew. Chem. Int. Ed. Engl., 1995, 34, 1837–1839. D.A. Foucher, B.Z. Tang, I. Manners, J. Am. Chem. Soc., 1992, 114, 6246–6248. W. Finckh, B.Z. Tang, D.A. Foucher et al. Organometallics, 1993, 12, 823–829. I. Manners, Adv. Organomet. Chem., 1995, 37, 131–168. P. Nguyen, G. Stojcevic, K. Kulbara et al. Macromolecules, 1998, 31, 5977–5983. D.A. Foucher, M. Edwards, R.A. Burrow et al. Organometallics, 1994, 13, 4959–4966. C.H. Honeyman, D.A. Foucher, F.Y. Dahmen et al. Organometallics, 1995, 14, 5503–5512. R. Rulkens, D.P. Gates, D. Balaishis et al. J. Am. Chem. Soc., 1997, 119, 10976–10986. A. Berenbaum, H. Braunschweig, R. Dirk et al. J. Am. Chem. Soc., 2000, 122, 5765–5774. F. J¨akle, R. Rulkens, G. Zech et al. Chem. Eur. J., 1998, 4, 2117–2128. R. Rulkens, R. Resendes, A. Verma et al. Macromolecules, 1997, 30, 8165–8171. R. Rulkens, A.J. Lough, I. Manners, J. Am. Chem. Soc., 1994, 116, 797–798. R. Rulkens, Y.Z. Ni, I. Manners, J. Am. Chem. Soc., 1994, 116, 12121–12122. Y.Z. Ni, R. Rulkens, I. Manners, J. Am. Chem. Soc., 1996, 118, 4102–4114. T. Baumgartner, F. J¨akle, R. Rulkens et al. J. Am. Chem. Soc., 2002, 124, 10062–10070. R. Resendes, J.M. Nelson, A. Fischer et al. J. Am. Chem. Soc., 2001, 123, 2116–2126. N.P. Preddy, H. Yamashita, M. Tanaka, Chem. Commun., 1995, 2263–2264. T.J. Peckham, J.A. Massey, M. Edwards et al. Macromolecules, 1996, 29, 2396–2403. R.W. Heo, F.B. Somoza, T.R. Lee, J. Am. Chem. Soc., 1998, 120, 1621–1622. P.F. Brandt, T.B. Rauchfuss, J. Am. Chem. Soc., 1992, 114, 1926–1927. C.P. Galloway, T.B. Rauchfuss, Angew. Chem., Int. Ed. Engl., 1993, 32, 1319–1321. D.L. Compton, T.B. Rauchfuss, Organometallics, 1994, 13, 4367–4376. D.L. Compton, P.F. Brandt, T.B. Rauchfuss et al. Chem. Mater., 1995, 7, 2342–2349. M. Rosenblum, H.M. Nugent, K-S. Jang et al. Macromolecules, 1995, 28, 6330–6342. R. Arnold, S.A. Matchett, M. Rosenblum, Organometallics, 1988, 7, 2261–2266. H.M. Nugent, M. Rosenblum, J. Am. Chem. Soc., 1993, 115, 3848–3849. N.J. Long, Angew. Chem., Int. Ed. Engl., 1995, 34, 21–38. J.M. Hodge, A.R. Berens, Macromolecules, 1982, 15, 762. M.E. Wright, E.G. Toplikar, H.S. Lackritz, J.T. Kerney, Macromolecules, 1994, 27, 3016–3022. M.E. Wright, E.G. Toplikar, R.F. Kubin, M.D. Seltzer, Macromolecules, 1992, 25, 1838–1839. D.R. van Staveren, N. Metzler-Nolte, Chem. Rev., 2004, 104, 5931–5985. G. Jaouen (Ed.), Bioorganometallics: Biomolecules, Labelling, Medicine, Wiley-VCH, Verlag GmbH, Weinheim, Germany (2006). M.G. Meirim, E.W. Neuse, G. Caldwell, J. Inorg. Organomet. Polym., 1997, 7, 71–91. M. de L. Machado, E.W. Neuse, A.G. Perlwitz, S. Schmitt, Polym. Adv. Technol., 1990, 1, 275–285. E.W. Neuse, A.G. Perlwitz, A.P. Barbosa, J. Appl. Polym. Sci., 1994, 54, 57–63. G. Caldwell, M.G. Meirim, E.W. Neuse, C.E.J. van Rensburg, Appl. Organomet. Chem., 1998, 12, 793–799.

Ferrocene-Containing Polymers and Dendrimers 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108.

445

M.G. Meirim, E.W. Neuse, G. Caldwell, J. Inorg. Organomet. Polym., 1999, 9, 221. S. Barlow, D. O’Hare, Chem. Rev., 1997, 97, 637–669. O. Riant, O. Samuel, T. Flessner et al. J. Org. Chem., 1997, 62, 6733–6745. G.E. Southard, M.D. Curtis, Organometallics, 2001, 20, 508–522. C. Xue, Z. Chen, Y. Wen et al. Langmuir, 2005, 21, 7860–7865. E. Laviron, J. Electroanal. Chem., 1979, 101, 19–28. D.A. Foucher, C.H. Honeyman, J.M. Nelson et al. Angew. Chem., Int. Ed. Engl., 1993, 32, 1709–1711. M.T. Nguyen, A.F. Diaz, V.V. Dement’ev, K.H. Pannell, Chem. Mater., 1993, 5, 1389–1394. N.J. Long, A.J. Martin, R. Vilar et al. Organometallics, 1999, 18, 4261–4269. M.T. Nguyen, A.F. Diaz, V.V. Dement’ev, K.H. Pannell, Chem. Mater., 1994, 6, 952–954. L. Espada, K.H. Pannell, V. Papkov et al. Organometallics, 2002, 21, 3758–3761. L. Bukueva, E.H. Sargent, R. Resendes et al. J. Mater. Sci., Mater. Electr., 2001, 12, 21–25. R. Resendes, A. Berenbaum, G. Stojevic et al. Adv. Mater., 2000, 12, 327–330. L. Tan, M.D. Curtis, A.H. Francis, Macromolecules, 2002, 35, 4628–4635. N.C. Foulds, C.R. Lowe, Anal. Chem., 1988, 60, 2473–2478. N.C. Foulds, C.R. Lowe, Faraday Trans., 1986, 82, 1259–1264. M.J. Green, H.A.O. Hill, Faraday Trans., 1986, 82, 1237–1243. P.N. Bartlett, R.G. Whitaker, J. Electroanal. Chem., 1987, 224, 37–48; J. Hodak, R. Etchenique, E.J. Calvo et al. Langmuir, 1997, 13, 2708–2716. J.E. Frew, M.J. Green, H.A.O. Hill, J. Chem. Technol. Biotechnol., 1986, 36, 357–363. C.P. Andrieux, J.M. Saveant, J. Electroanal. Chem., 1978, 93, 163–168. C. Xue, Z. Chen, F-T. Luo et al. Biomacromolecules, 2005, 6, 1810–1815. I. Willner, E. Katz, Angew. Chem., Int. Ed., 2000, 39, 1180–1218. F. Mao, N. Mano, A. Heller, J. Am. Chem. Soc., 2003, 125, 4951–4957. S. Yajima, M. Omori, Nature, 1977, 267, 823–825. B-Z. Tang, R. Petersen, D.A. Foucher et al. Chem. Commun., 1993, 523–524. R. Petersen, D.A. Foucher, B-Z. Tang et al. Chem. Mater., 1995, 7, 2045–2053. Q. Sun, J.W.Y. Lam, K. Xu et al. Chem. Mater., 2000, 12, 2617–2624. M.J. MacLachlan, M. Ginzburg, N. Coombs et al. Science, 2000, 287, 1460–1463. M. Ginzburg, M.J. MacLachlan, S.M. Yang et al. J. Am. Chem. Soc., 2002, 124, 2625–2639. C.L. Mantell, Carbon and Graphite Handbook , Interscience Publishers, New York, USA (1968), pp. 162–170. M.J. MacLachlan, P. Aroca, N. Coombs et al. Adv. Mater., 1998, 10, 144–149. M.J. MacLachlan, M. Ginzburg, N. Coombs et al. J. Am. Chem. Soc., 2000, 122, 3878–3891. C.T. Kresge, M.E. Leonowicz, W.J. Roth et al. Nature, 1992, 359, 710–712. K. Kulbaba, A. Cheng, A. Bartole et al. J. Am. Chem. Soc., 2002, 124, 12522–12534. J.E. Hudson, H.D. Abruna, J. Am. Chem. Soc., 1996, 118, 6303–6304. D.J. Diaz, J.E. Hudson, G.D. Storrier et al. Langmuir, 2001, 17, 5932–5938. W. Shi, S. Cui, Ch. Wang et al. Macromolecules, 2004, 37, 1839–1842. K. Kulbaba, I. Manners, Macromol. Rapid Commun., 2001, 22, 711–724. I. Manners, Chem. Commun., 1999, 857–865. J. Lu, D. Chamberlin, D.A. Rider et al. Nanotechnology, 2006, 17, 5792–5797. J.Q. Lu, D.A. Rider, E. Onyegam et al. Langmuir, 2006, 22, 5174–5179. D.A. Tomalia, H. Baker, J.R. Dewald et al. Polym. J., 1985, 17, 117–132. G.R. Newkome, Z. Yao, G.R. Baker, V.K. Gupta, J. Org. Chem., 1985, 50, 2003–2004. C. Hawker, J.M. Frechet, Chem. Commun., 1990, 1010–1013. G.R. Newkome, E. He, Ch.N. Mooreﬁeld, Chem. Rev., 1999, 99, 1689–1746.

446

Ferrocenes: Ligands, Materials and Biomolecules

109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126.

C.M. Casado, I. Cuadrado, M. Mor´an et al. Coord. Chem. Rev., 1999, 185-186, 53–80. A-M. Caminade, J-P. Majoral, Coord. Chem. Rev., 2005, 249, 1917–1926. M. Wagner, Angew. Chem. Int. Ed., 2006, 45, 5916–5918. D.L. Stone, D.K. Smith, P.T. McGrail, J. Am. Chem. Soc., 2002, 124, 856–864. C.M. Cardona, A.E. Kaifer, J. Am. Chem. Soc., 1998, 120, 4023–4024. D.K. Smith, Perkin Trans., 1999, 1563–1566. I. Cuadrado, M. Mor´an, C.M. Casado et al. Organometallics, 1996, 15, 5278–5280. I. Cuadrado, C.M. Casado, B. Alonso et al. J. Am. Chem. Soc., 1997, 119, 7613–7614. M.B. Robin, P. Day, Adv. Inorg. Chem. Radiochem., 1967, 10, 247–271. T. Chuard, M-T. Beguin, R. Deschenaux, C. R. Chimie., 2003, 6, 959–962. B. Dardel, R. Deschenaux, M. Even, E. Serrano, Macromolecules, 1999, 32, 5193–5198. J. Alvarez, T. Ren, A.E. Kaifer, Organometallics, 2001, 20, 3543–3549. J. Losada, I. Cuadrado, M. Mor´an et al. Anal. Chim. Acta, 1997, 338, 191–198. Ch. Va´elrio, J-L. Fillaut, J. Ruiz et al. J. Am. Chem. Soc., 1997, 119, 2588–2589. C. Ornelas, J.R. Aranzaes, E. Cloutet et al. Angew. Chem., Int. Ed., 2007, 46, 872–877. Ch. Kim, E. Park, Ch.K. Song, B.W. Koo, Synth. Met., 2001, 123, 493–496. C.O. Turrin, J. Chiffre, D. de Montauzon et al. Macromolecules, 2000, 33, 7328–7336. C-O. Turrin, J. Chiffre, D. de Montauzon et al. Organometallics, 2002, 21, 1891–1897.

11 Ferrocene-Containing Thermotropic Liquid Crystals∗ Robert Deschenaux

11.1

Introduction

Thermotropic liquid crystals play a crucial role in everyday life as they have found widespread applications in the manufacture of, e.g., watches, calculators, mobile telephones, notebook computers, thermometry, speciﬁc oils and pigments. Further applications are expected in the future. Obviously, the achievement of this goal requires the design of liquid–crystalline materials with novel properties. The development of functional liquid crystals with tailor-made properties is a current challenge in materials science. Incorporation of active subunits into liquid crystals should give rise to such new, highly efﬁcient materials, which combine the properties of the subunits with those of liquid crystals (organization, anisotropy). If speciﬁc properties are to be exploited (e.g. magnetic, redox, optical and catalytic properties), the choice of the subunit and how it is connected to the liquid crystal are of prime importance. Owing to its remarkable electrochemical properties (fast and reversible one electron transfer process) and three-dimensional structure, which offers the possibility of synthesizing a great variety of derivatives, ferrocene is a unit of choice for the design of thermotropic functional liquid crystals. The scope of this chapter is to highlight with selected examples some relevant results we have obtained in the ﬁeld of ferrocene-containing thermotropic liquid crystals. Those examples will demonstrate the wide applications of ferrocene in this area. ∗ Speciﬁc abbreviations used throughout this chapter are given at the end of the chapter before the Reference List.

Ferrocenes: Ligands, Materials and Biomolecules  2008 John Wiley & Sons, Ltd

Edited by Petr Stepnicka

448

Ferrocenes: Ligands, Materials and Biomolecules

For details and in-depth discussion regarding ferrocene-containing thermotropic liquid crystals, reviews are available elsewhere.1–3

11.2

Ferrocene-Containing Side-Chain Liquid–Crystalline Polymers

Attention was focused on side-chain polysiloxanes and polymethacrylates, which it was anticipated could be prepared from vinyl- and methacrylate-containing ferrocene monomers, respectively, following well-established procedures developed for organictype monomers. Polysiloxane 14 (Scheme 11.1) was prepared by grafting the corresponding vinyl monomer (structure not shown) onto commercially available polyhydrosiloxane following a standard procedure [toluene, 70 ◦ C, 24 h, PtCl2 (cod)]. Polymethacrylates 25 and 36 (Scheme 11.1) were prepared by free-radical polymerization (THF, AIBN, 50 ◦ C) of the corresponding methacrylate monomers (structures not shown). Polymers 1–3 showed good solubility in common organic solvents (dichloromethane , chloroform, tetrahydrofuran), good thermal stability (no decomposition was detected up to 250 ◦ C) and narrow molecular weight distribution (Mw /Mn : 1.4–1.6). All polymers gave rise to the formation of smectic phases. Polymers 1 and 3 showed SmA and SmC phases, and polymer 2 showed a SmC phase. The mesomorphic behavior of 1–3 is in agreement with the linear structure of the pendant monomeric units, the tendency of which is to organize in a parallel fashion, i.e. into lamellar phases. The effective synthesis of 1–3 as well as their thermal and mesomorphic properties conﬁrmed that ferrocene can be readily incorporated into liquid–crystalline side-chain polymers as organic-type monomers.

11.3

Liquid–Crystalline Ferrocenes with Planar Chirality

Symmetry breaking operations and chirality have held centre stage in the fundamental investigations and applications of liquid crystals.7, 8 The search for new chiral effects in mesomorphic systems prompted us to synthesize unsymmetrically 1,3-disubstituted ferrocene-containing liquid crystals,9 where the different substituents at the 1- and 3-positions generate structures with planar chirality (Figure 11.1). Ferrocene (S)-4 (Scheme 11.2) was obtained with an enantiomeric excess of 98 %,12 and its absolute conﬁguration determined on the basis of circular dichroism (CD) spectra.13 Smectic C∗ and SmA phases were observed for (S)-4. The spontaneous polarization of (S)-4 was determined and a value of 2.8 nC cm−2 was obtained. This low value is in agreement with the structure of (S)-4, which carries two organic fragments that are differentiated only by the length of the terminal alkyl chains and the orientation of the external ester functions. Planar chirality is of interest as the chiral center is located at the ferrocene itself. Such structures might help to clarify the role of ferrocene in the formation, nature and stability of liquid–crystalline phases. Note that planar chirality was elegantly exploited by Malthˆete and coworkers, who reported optically-active butadiene–tricarbonyliron liquid crystal complexes.14 The spontaneous polarization of one of the complexes was determined, and a value of 32 nC cm−2 was

CH3

CO2

CH2 C

CH3

x

CO2

x

SiO

H2C

CH2 C

(H3C)3SiO

CH3

(CH2)6 O

(CH2)6 O

Si(CH3)3 x (CH2)10 O

CO2

O2C

O2C

2

Fe

1

Scheme 11.1

CO2

CO2

O2C

3

Fe

Tg : 30; SmC 210 I

CO2

Cr 136 SmC 140 SmA 183 I

CO2

Tg : not detected; SmC 154 SmA 185 I

CO2

CO2

CO2

Fe

OC18H37

OC18H37

Ferrocene-Containing Thermotropic Liquid Crystals 449

450

Ferrocenes: Ligands, Materials and Biomolecules R1

R1 Fe

R2

R2

Fe

Figure 11.1 Planar chiral 1,3-disubstituted ferrocene derivatives (R1 = R2 ) (Planar chirality: chirality resulting from the arrangement of out-of-plane groups with respect to a reference plane, called the ‘chiral plane’.10 The conﬁguration of the 1,3-disubstituted ferrocenes described in this chapter is assigned by applying the planar nomenclature.11 )

H21C10O

O2C

O2C

CO2

O2C

OC18H37

Fe

(S)-4

Cr 170 SmC* 198 SmA 202 I

Scheme 11.2

obtained.14b Also, as most of the studies on chiral liquid crystals are developed with compounds having central chirality, planar chirality is an alternative way to obtain optically-active materials. Planar chiral 1,3-disubstituted ferrocenes are valuable building blocks for the preparation of optically-active side-chain polymers. The chirality is located on the ferrocene unit, and the polymerizable group is located in one of the organic fragments connected to ferrocene. Tuning of the liquid–crystalline properties of the monomers and polymers can be achieved by structural modiﬁcations of the organic part only. This concept is appealing since the modiﬁcation of the organic motifs will not alter the chirality at the ferrocene. Poly[(R)-5] and poly[(S)-6] (Scheme 11.3) were prepared by grafting either (R)-5 or (S)-6 (Scheme 11.3) onto commercially available polyhydrosiloxane (x = 15–18 %) in the presence of [platinum(0)-1,3-divinyl-1,1,3,3-tetramethyldisiloxane].13 Poly[(R)5] and poly[(S)-6] were soluble in common organic solvents and showed good thermal stability and narrow molecular weight distribution (Mw /Mn : 1.3–1.6). The structure of the monomers was chosen in order to generate either SmC∗ (from the linear-type monomer) or N∗ (from the laterally-branched monomer) phases. A polyhydrosiloxane with a low content of SiH units was used to avoid decomposition of the polymers at elevated temperatures (the clearing temperature increases with the number of mesogenic units). The absolute conﬁguration was established on the basis of CD spectra.13 Monomer (R)-5 showed enantiotropic SmA and N∗ phases and a monotropic SmC∗ phase. Monomer (S)-6 displayed a N∗ phase. The mesomorphic properties of (R)-5 and (S)-6 are in agreement with their structure: linear-type mesogens give rise to smectic phases whereas laterally-branched mesogens tend to exhibit nematic behavior. For poly[(R)-5], the melting and clearing points were clearly identiﬁed by polarized optical microscopy and differential scanning calorimetry. On heating, poly[(S)-6] gave

O2C

O2C

H21C10O

H21C10O

O2C

O2C O2C

O2C

Cr 115 N* 120 I

CO2

O

(CH2)4

O (CH2)9

OC6H13

HC CH2

Cr 175 (SmC* 171) SmA 274 N* 278 I

CO2

Scheme 11.3

(S)-6

Fe

(R)-5

Fe

HC CH2

Ferrocene-Containing Thermotropic Liquid Crystals 451

O2C

O2C

H21C10O

H21C10O

O2C

O2C

poly[(S)-6]

Fe

poly[(R)-5]

Fe

O2C

(CH2)4

Scheme 11.3 (continued )

(Tm : 82, Tc : 94)

CO2

O

CH3 x

OC6H13

O Si

O2C

(CH3)3Si

Tm : 171, Tc : 302

CO2

CH3 O Si x

O Si(CH3)3 CH3 1-x

CH3 O Si

O Si O Si(CH3)3 CH3 1-x

CH3

O (CH2)9

(CH3)3Si

452 Ferrocenes: Ligands, Materials and Biomolecules

Ferrocene-Containing Thermotropic Liquid Crystals

453

a series of broad transitions between 100 and 115 ◦ C, which corresponded to melting and crystallization processes; poly[(S)-6] displayed a monotropic mesomorphic behavior. Poly[(R)-5] and poly[(S)-6] retained the liquid–crystalline phases displayed by their respective monomer. Therefore, their liquid–crystalline properties are in agreement with the structure of the pendent motifs (poly[(R)-5]: linear-type mesogens; poly[(S)-6]: laterally-branched mesogens). With respect to possible applications as a new type of chiral dopants for display applications, the helical twisting power (HTP) of (S)-6 and poly[(S)-6] were investigated in a Grandjean–Cano wedge.15 As a nematic host, E7 (multicomponent nematic mixture, Merck) was used. The solubility of both materials in E7 was poor, and the prepared mixtures contained about 1 % of either (S)-6 or poly[(S)-6]. Both materials induced a helix in nematic E7. Regular disclination lines became visible shortly after ﬁlling the Grandjean–Cano cells and after relaxation of the texture. Disclination lines separated regions with a different number of half pitches according to the local thickness in the Grandjean-Cano wedge. For (S)-6 (1.06 % mixture, 40 µm pitch) and poly[(S)-6] (1.06 % mixture, 60 µm pitch), HTP values of 2.4 and 1.6 µm−1 were obtained, respectively. These values are rather small in comparison with commercially available chiral dopants that reach values of −11 µm−1 (S811 from Merck), but are in agreement with the structures of (S)-6 and poly[(S)-6], in which the ferrocene unit is functionalized by two substituents with similar length and polarity.

11.4

Redox-Active Liquid–Crystalline Ferrocenes

Ferrocene has found interesting applications as an electroactive building block for the design of switchable molecular aggregates,16 redox-active receptors,17 redox-active polymeric ionomers18 and conducting and magnetic materials.19 With the aim of designing switchable liquid crystals based on the ferrocene/ferrocenium redox couple, the peralkylated ferrocenes 720 and 821 (Scheme 11.4) were synthesized. The peralkylated ferrocene moiety was selected as the redox-active center in order to use the easy oxidation of this unit in comparison with less alkylated ferrocenes. Compounds 7 and 8 were readily oxidized with silver(I) tosylate into the corresponding ferrocenium derivatives 9 and 10 (Scheme 11.4), respectively. Compounds 7 and 8 did not display mesomorphism. As in the case of other monosubstituted ferrocenes,1–3 the organic part used here cannot thwart the size of the organometallic core, and only isotropic ﬂuids formed when the samples melted. However, the ferrocenium derivatives 9 and 10 exhibited liquid–crystalline behavior: a monotropic SmA phase was observed for 9 and a monotropic Colr phase was obtained for 10. The fact that the oxidized species gave rise to mesomorphism is an indication that the formation of liquid–crystalline phases depends on structural factors and ionic interactions. The SmA (monomolecular organization) nature of the mesophase displayed by 9 is due to the head-to-tail orientation of the molecules within the mesophase. The Colr phase obtained for 10 was found to have a centered rectangular symmetry. The discrepancy between the molecular volume of the ﬂexible tail with that of the rigid rod is responsible for the formation of the columnar phase.

454

Ferrocenes: Ligands, Materials and Biomolecules CO2

CO2

OC10H21

Fe Cr 154 I

7

CO2

O2C

OC18H37

Fe 8

H 3C

SO3

Cr 134 I

9

H 3C

SO3

CO2

Fe +

OC10H21

Cr 132 (SmA 83) I

CO2

Fe + 10

CO2

O2C

OC18H37

Cr 123 (Colr 101) I

CH3 CH2 C

x CO2 (CH2)6 O

CO2

CO2

O2C I3

Tg : ~70; N ~160 I

−

Fe + 11

Scheme 11.4

The temperature-dependent magnetic susceptibilities of 10 were found to be in agreement with those expected for weakly interacting ferrocenium cores. No magneticﬁeld-induced orientation was observed upon cooling the sample from the isotropic liquid down to the Colr phase. The redox properties of the ferrocene unit were also used to control the liquid– crystalline organization of side-chain liquid–crystalline polymers: the reduced polymer 3 (Scheme 11.1) gave rise to SmC and SmA phases whereas the oxidized polymer 11 (Scheme 11.4) (3 was oxidized with iodine) showed a N phase.22 These results conﬁrm that liquid–crystalline switches should, in principle, be available by incorporating ferrocene/ferrocenium into mesomorphic materials.

Ferrocene-Containing Thermotropic Liquid Crystals

11.5

455

Ferrocene-Containing Liquid–Crystalline Dendrimers

Dendrimers represent a class of materials that combine unique features (well-deﬁned macromolecular structure, monodispersity, low viscosity) with remarkable properties (encapsulation, catalysis, chiroptical properties).23 Functionalized dendrimers, i.e. dendrimers incorporating active or reactive functions, are considered as new materials with potential applications, such as for the preparation of macromolecular libraries.24 With the view to combining the properties and features of dendrimers with the properties of ferrocene, ferrocene-containing liquid–crystalline dendrimers were designed.25, 26 They were prepared by applying a convergent synthetic methodology. Therefore, they were isolated as monodisperse macromolecules. One example of ferrocene-containing liquid–crystalline dendrimers is illustrated by second generation dendrimer 12 (Scheme 11.5), in which cholesterol acts as liquid–crystalline promoter.26 A broad-range SmA phase was observed. Dendrimer 12 was analyzed by cyclic voltametry, which revealed a reversible oxidation process with twelve electrons being transferred (E o = 0.93 V vs. ferrocene in dichloromethane). In 12, the ferrocene units are directly connected to the dendritic core; however, they can also be located at the periphery of the structure.26

11.6

Liquid–Crystalline Ferrocene–Fullerene Dyads

The development of multifunctional materials represents a ﬁeld of intense research activity. In fact, the role played by two or more compounds may well be performed by one single multicomponent molecule (each component can be considered as an active subunit). Electroactive and/or photoactive liquid crystals belong to this class of materials and are of fundamental and technological interest. Photoinduced electron transfer reactions in fullerene–ferrocene dyads were investigated.27 The results are interesting for the development of optical and electronic supramolecular devices. Indeed, we have shown that electron transfer in liquid–crystalline ferrocenes can be used to generate or modify mesomorphism (see Section 11.4). Therefore, photoinduced electron transfer in fullerene–ferrocene liquid crystals could be used to control the liquid–crystalline properties because of the presence of either the ferrocene (light off) or ferrocenium (light on) species. On the basis of the above considerations, liquid–crystalline fullerene–ferrocene dyads 1328 (Scheme 11.6), 1429 (Scheme 11.7), 1530 (Scheme 11.8), 1631 (Scheme 11.9) and 1731 (Scheme 11.9) were designed The second generation liquid–crystalline poly(aryl ester) dendrimers in 14–17 were synthesized by applying a convergent approach, and functionalization of [60]fullerene was achieved via the Bingel addition reaction32 (for 13 and 14) or the 1,3-dipolar cycloaddition reaction33 (for 15–17). Compounds 14–17 were thus obtained as monodisperse macromolecules. In 13 and 14, the ferrocene unit is located in the mesogenic part. However, in 15–17, ferrocene is independent of the liquid crystal promoter. Compounds 13–17 were found to be soluble in common organic solvents and thermally stable.

CholO2C

Chol =

CholO2C

CholO2C

O

CholO2C

(CH2)10 O2C

O (CH2)10 O2C

Fe

Fe

Fe

O

O

O

O

O

O

Fe

O O

O

O

O

O (CH2)10 O2C

O (CH2)10 O2C

O (CH2)10 O2C

O (CH2)10 O2C

CholO2C

CholO2C

O

O

Fe

O

O

O

Fe

O

O O

O

O

O

O

O

O

O

O

O

12

O

O

O

O

O

O

O

O

Scheme 11.5

Tg : 52; SmA 169 I

O

O

O

O

O

O

Fe

O

O

O

O

O

O

O

O

O

O

O

O

O

O

O

O

O

O

O

O

O

O

Fe

O O

Fe

CO2Chol

CO2Chol

CO2Chol

CO2Chol

CO2Chol

CO2Chol

CO2 (CH2)10 O

CO2 (CH2)10 O

Fe

CO2 (CH2)10 O

CO2 (CH2)10 O

Fe

CO2 (CH2)10 O

CO2 (CH2)10 O

Fe

456 Ferrocenes: Ligands, Materials and Biomolecules

O2C

O (CH2)10 O2C Fe CO2

13

O2C

Cr 66 SmA 118 I

CO2 (CH2)6 O

Scheme 11.6

O (CH2)6 O2C

Fe

CO2 (CH2)10 O

CO2

Ferrocene-Containing Thermotropic Liquid Crystals 457

O

O

O C

2

2

O C

2 10

(CH )

2 10

(CH )

2

O C

O2C

2

O C

2

O C

Fe

Fe

O

O

(CH2)10 O 2C

O

O

CO 2

O

O

2

O C

CO2

2 10

(CH )

Fe

O

O

Fe

O

O 2

O C O 2 6

(CH ) 2

O C 2

CO 2 6

(CH )

O

CO2

O

Scheme 11.7

Tg : not detected; SmA 157 I

O

O

14

O

O

O

O

O

O

O

O

O

O

O

O

O

O

Fe

O

O

O

Fe

2

CO

O

O

O

O

2

CO

2

O

(CH2)10

2

O

O C

O C

2 10

(CH )

Fe

Fe

2

CO

CO

2

2

CO2

CO

10

(CH2)10

(CH2)

O

O

2

CO2

CO

458 Ferrocenes: Ligands, Materials and Biomolecules

15

N

O

O

HN Fe

Tg : 40; SmA 135 I

CO2 (CH2)10 O

O

CO2

O

O (CH2)10 O

O (CH2)10 O

O (CH2)10 O

O

O

O

O (CH2)10 O

Scheme 11.8

O

O

O

O

CO2

CO2

CO2

CO2

CN

CN

CN

CN

Ferrocene-Containing Thermotropic Liquid Crystals 459

Fe

Fe

Fe

Fe

17

16

N

O

N

O

CO2

CO2

Scheme 11.9

CO2 (CH2)10 O

Tg : 34; SmA 168 I

Fe

Tg : 47; SmA 171 I

CO2 (CH2)10 O

O

O

O

O

O (CH2)10 O

Fe

O

O (CH2)10 O

CO2

CO2

CO2

CO2

O (CH2)10 O

O (CH2)10 O

O

O

O

O (CH2)10 O

O (CH2)10 O

O

O (CH2)10 O

O

O (CH2)10 O

O

O

O

O

O

O

CO2

CO2

CO2

CO2

CN

CN

CN

CN

CN

CN

CN

CN

460 Ferrocenes: Ligands, Materials and Biomolecules

Ferrocene-Containing Thermotropic Liquid Crystals

461

All compounds gave rise to the formation of SmA phases in agreement with the nature of the liquid–crystalline promoter, i.e. cholesterol in 13 and 14, and a cyanobiphenyl derivative in 15–17. Compound 14 showed a higher clearing temperature than 13. This result shows that liquid–crystalline dendrimers tend to stabilize the mesophases because of the higher number of mesogenic units in their structure (8 and 2 mesogenic units in 14 and 13, respectively). It is remarkable that for 17, a second generation

N

Fe

Fe Fe

Fe Fe

N

Fe

N

Fe

Fe

Fe

Fe Fe

Fe

Fe

N

Fe

N

Fe

Fe

Fe

Fe Fe

Fe

Fe

Fe

Fe

Fe

N

Figure 11.2 Postulated supramolecular organization of 17 within the SmA phase

462

Ferrocenes: Ligands, Materials and Biomolecules

liquid–crystalline dendrimer can thwart the bulkiness of the ferrocene-based dendrimer and induce such a broad mesomorphic domain. This is, most likely, due to the supramolecular organization of 17 within the SmA phase: the molecular units organize into a bilayer smectic structure with alternate sublayers of cyanobiphenyl units, [60]fullerene units and ferrocene units (Figure 11.2). Such a supramolecular organization is driven by steric constraints resulting from the difference of the cross˚ 2 ), ferrocene (45 A ˚ 2 ) and the four mesogenic sectional area of fullerene (90–100 A 2 ˚ per mesogenic unit). Photoinduced electron transfer from ferrocene groups (22–25 A to [60]fullerene was identiﬁed for 13 and 15–17 but with short lifetimes for the charge-separated state (e.g. 560 ns for 15 in THF).

11.7

Conclusion

The results presented in this chapter show that ferrocene-containing liquid crystals have reached a high degree of complexity and sophistication that can be used for the elaboration of mesomorphic materials with speciﬁc functions and properties. Further developments in our group will focus on the synthesis of polyfunctional materials for which the supramolecular organization can be controlled by design. Such materials are of interest for the development of nanotechnologies by the ‘bottom-up’ approach. The author acknowledges the Swiss National Science Foundation for ﬁnancial support.

11.8

Abbreviations

Cr crystal Colr rectangular columnar phase SmC smectic C phase SmC∗ chiral smectic C phase SmA smectic A phase N nematic phase N∗ chiral nematic phase I isotropic liquid Tg glass transition Tm melting point Tc clearing point For monotropic mesophases the transitions are given in parentheses. Temperatures are given in ◦ C.

References 1. R. Deschenaux, J.W. Goodby, Ferrocene-Containing Thermotropic Liquid Crystals, in Ferrocenes: Homogeneous Catalysis, Organic Synthesis, Materials Science, A. Togni and T. Hayashi (Eds), VCH, Weinheim, Germany (1995), pp. 471–495.

Ferrocene-Containing Thermotropic Liquid Crystals

463

2. B. Donnio, D. Guillon, R. Deschenaux, D.W. Bruce, Metallomesogens, in Comprehensive Coordination Chemistry II , J.A. McCleverty and T.J. Meyer (Eds), Elsevier, Oxford, UK (2003), Vol. 7, pp. 357–627. 3. D.W. Bruce, R. Deschenaux, B. Donnio, D. Guillon, Metallomesogens, in Comprehensive Organometallic Chemistry III , R.H. Crabtree and D.P. Mingos (Eds), Elsevier, Oxford, UK (2007), Vol. 12, pp. 195–293. 4. R. Deschenaux, I. Jauslin, U. Scholten et al. Macromolecules, 1998, 31, 5647. 5. R. Deschenaux, V. Izvolenski, F. Turpin et al. Chem. Commun., 1996, 439. 6. R. Deschenaux, F. Turpin, D. Guillon, Macromolecules, 1997, 30, 3759. 7. J.W. Goodby, J. Mater. Chem., 1991, 1, 307. 8. J.W. Goodby, R. Blinc, N.A. Clark et al. Ferroelectric Liquid Crystals – Principle, Properties and Applications, Gordon and Breach, Philadelphia and Reading (1991). 9. (a) R. Deschenaux, J. Santiago, Tetrahedron Lett., 1994, 35, 2169; (b) R. Deschenaux, I. Kosztics, B. Nicolet, J. Mater. Chem., 1995, 5, 2291. 10. E.L. Elliel, S.H. Wilen, Stereochemistry of Organic Compounds, John Wiley & Sons Inc. New York, USA (1994). 11. G. Wagner, R. Herrmann, Chiral Ferrocene Derivatives. An Introduction, in Ferrocenes: Homogeneous Catalysis, Organic Synthesis, Materials Science, A. Togni and T. Hayashi (Eds), VCH, Weinheim, Germany, (1995) pp. 173–218. 12. T. Chuard, S.J. Cowling, M. Fernandez-Ciurleo et al. Chem. Commun., 2000, 2109. 13. J. Brettar, T. B¨urgi, B. Donnio et al. Adv. Funct. Mater., 2006, 16, 260. 14. (a) L. Ziminsky, J. Malthˆete, J. Chem. Soc., Chem. Commun., 1990, 1495; (b) P. Jacq, J. Malthˆete, Liq. Cryst., 1996, 21, 291. 15. P. Oswald, P. Pieranski, in Les Cristaux Liquides, Vol. 1, Gordon and Breach, Paris, (2000), pp. 378–392. 16. J.C. Medina, I. Gay, Z. Chen et al. J. Am. Chem. Soc., 1991, 113, 365. 17. P.D. Beer, E.L. Tite, A. Ibbotson, J. Chem. Soc., Dalton Trans. , 1991, 1691. 18. A. Wiesemann, R. Zentel, G. Lieser, Acta Polym., 1995, 46, 25. 19. A. Togni, Ferrocene-Containing Charge-Transfer Complexes. Conducting and Magnetic Materials in Ferrocenes: Homogeneous Catalysis, Organic Synthesis, Materials Science, A. Togni and T. Hayashi (Eds), VCH, Weinheim, Germany (1995), pp. 433–469. 20. R. Deschenaux, M. Schweissguth, A.-M. Levelut, Chem. Commun., 1996, 1275. 21. R. Deschenaux, M. Schweissguth, M.-T. Vilches et al. Organometallics, 1999, 18, 5553. 22. F. Turpin, D. Guillon, R. Deschenaux, Mol. Cryst. Liq. Cryst., 2001, 362, 171. 23. (a) J.M.J. Fr´echet, D.A. Tomalia, Dendrimers and other Dendritic Polymers, John Wiley & Sons Ltd, Chichester, UK (2001); (b) G.R. Newkome, C.N. Mooreﬁeld, F. V¨ogtle, Dendrimers and Dendrons: Concepts, Syntheses, Applications, Wiley-VCH Verlag GmbH, Weinheim, Germany (2001). 24. A.W. Freeman, L.A.J. Chrisstoffels, J.M.J. Fr´echet, J. Org. Chem., 2000, 65, 7612. 25. R. Deschenaux, E. Serrano, A.-M. Levelut, Chem. Commun., 1997, 1577. 26. T. Chuard, M.-T. B´eguin, R. Deschenaux, C. R. Chimie, 2003, 6, 959. 27. D.M. Guldi, M. Maggini, G. Scorrano, M. Prato, J. Am. Chem. Soc., 1997, 119, 974. 28. (a) R. Deschenaux, M. Even, D. Guillon, Chem. Commun., 1998, 537; (b) M. Even, B. Heinrich, D. Guillon et al. Chem. Eur. J., 2001, 7, 2595. 29. B. Dardel, R. Deschenaux, M. Even, E. Serrano, Macromolecules, 1999, 32, 5193. 30. S. Campidelli, E. V´azquez, D. Milic et al. J. Mater. Chem., 2004, 14, 1266. 31. S. Campidelli, L. P´erez, J. Rodr´ıguez-L´opez et al. Tetrahedron, 2006, 62, 2115. 32. (a) C. Bingel, Chem. Ber., 1993, 126, 1957; (b) J.-F. Nierengarten, A. Herrmann, R.R. Tykwinsky et al. Helv. Chim. Acta, 1997, 80, 293; (c) X. Camps, A. Hirsch, J. Chem. Soc., Perkin Trans. 1 , 1997, 1595. 33. (a) M. Prato, M. Maggini, Acc. Chem. Res., 1998, 31, 519; (b) N. Tagmatarchis, M. Prato, Synlett., 2003, 768.

12 Crystal Engineering with Ferrocene Compounds Dario Braga, Marco Curzi, Stefano Luca Giaffreda, Fabrizia Grepioni, Lucia Maini, Anna Pettersen and Marco Polito

12.1

Introduction

Making crystals by design is the paradigm of crystal engineering.1 The main goal of the crystal engineering endeavour is that of being able to obtain collective supramolecular properties2 from the convolution of the physical and chemical properties of the individual building blocks with crystal periodicity and symmetry.3 These collective properties and functions (e.g. chirality, conductivity, magnetism, nanoporosity etc) or applications (e.g. NLO, polymorphism, molecular trapping and sensing etc) depend on the aggregation via intermolecular bonds of the component units (molecules or ions). Given that the aim is the construction and exploitation of solid materials,4 crystal engineering is essentially a branch of solid state chemistry, precisely supramolecular solid state chemistry. Two main sub-areas of crystal engineering can be envisaged, namely those of coordination networks5, 6 and of molecular materials.7–9 This is a practical subdivision, though all possible intermediate situations are possible. Indeed the chemistry of coordination networks can be regarded as periodical coordination chemistry since it exploits the possibility of divergent ligand–metal coordination, as opposed to the more traditional convergent coordination chemistry operated by chelating polydentate ligands9 (Figure 12.1). The possibility of exploiting coordination networks for

Ferrocenes: Ligands, Materials and Biomolecules  2008 John Wiley & Sons, Ltd

Edited by Petr Stepnicka

466

Ferrocenes: Ligands, Materials and Biomolecules

+

M M

+

metal centre diverging ligand metal centre chelating ligand

M+ M

M M

M

M M

M M

coordination compound coordination network

Figure 12.1 The relationship between molecular (left) and periodical (right) coordination chemistry: the use of bidentate ligand spacers allows construction of periodical coordination complexes

practical applications (such as absorption of molecules, reactions in cavities etc) critically depends on whether the networks contain large empty spaces (channels, cavities etc)10 or whether the network is close packed because of interpenetration and selfentanglement.6, 11 The possibility of a sponge-like behaviour to accommodate/release guest molecules should also be taken into account.12, 13 While in periodical coordination chemistry it is useful to focus on the knots and spacers in order to describe the topology of the network, when dealing with molecular materials what matters most are the characteristics of the component molecules or ions and the type of interactions (van der Waals, hydrogen bonds, π-stacking, ionic interactions, ion pairs etc) holding these building blocks together.13, 14 The metalcontaining coordination-directed self-assembly is an extremely active area of research, especially for the synthesis of metal–organic coordination networks, ﬁnite (triangles, squares, etc)15 or inﬁnite networks.6, 16 Very many relevant books and reviews address this subject.17, 18, 19, 20 This chapter is devoted to the use of ferrocene compounds in crystal engineering, giving references to related systems. It is important to point out, however, that these compounds represent a small subset of the classes of coordination compounds used in crystal engineering. It should be recalled that the structure of ferrocene itself has been determined several times and at different temperatures25, 26 because the molecule possesses conformational freedom giving rise to conformational crystal polymorphism.21 Indeed, structural ﬂexibility, a key characteristic of organometallic molecules,22 plays a particularly important role in organometallic polymorphism.23 Structurally nonrigid organometallic molecules are likely candidates for the formation of conformational polymorphs. Besides the scientiﬁc relevance, crystal polymorphism has important practical implications.24

Crystal Engineering with Ferrocene Compounds

12.2 12.2.1

467

Polymorphism, Isomerizations and Phase Transitions Ferrocene: Early Steps in Crystal Engineering

On approaching crystal polymorphism it is necessary to distinguish between those polymorphs that interconvert as a function of temperature via a phase transition (enantiotropic systems) and those that do not (monotropic systems). Besides, the existence of solvate forms, which contain the same molecule or ions but are cocrystallized with solvent molecules, must be taken into account. Ferrocene crystals have been investigated by a number of scientists, though the most systematic investigations were covered out by J. D. Dunitz and collaborators. For this molecule, one room temperature disordered25 and two low temperature ordered crystalline forms are known.26 At the molecular level the ferrocene molecules differ only in the relative orientation of the two cyclopentadienyl rings. At the crystal level they differ in the relative orientation of the molecules (Figure 12.2), so that the phase transition mechanism requires only low-energy reorientation of the rings and a limited motion of the molecules in the crystal structure.27 Many crystals of sandwich organometallic molecules are known to undergo phase transitions, and for some the formation of plastic phases characterized by short range orientational disorder and long range order has been established. For instance, ferrocene derivatives FcCHO28 and FcC(O)Me29 as well as salts of the type [Fe(η5 -C6 H5 F)(η5 C5 H5 )][A] [A D AsF6 , PF6 , SbF6 and BF4 ]30 are all known to undergo order–disorder phase transitions. The phase transitional behaviours of the ferrocenium and cobaltocenium salts [M(η5 -C5 H5 )2 ][PF6 ] (M D Co, Fe) have also been recently studied in an effort to understand the factors controlling phase transition temperatures and differences between the cobalt and iron species.31 The room temperature ordered monoclinic crystal (Form I) of the two salts transforms, below 252 K, into another ordered monoclinic crystal (Form II) with different relative orientation of the two independent [Co(η5 -C5 H5 )2 ]C cations, and into a semi-plastic system (Form III) containing ordered PF6 anions and orientationally disordered [Co(η5 -C5 H5 )2 ]C cations above 314 K. Although isomorphous at room temperature with crystalline [Co(η5 -C5 H5 )2 ][PF6 ], the ferrocenium salt [Fe(η5 -C5 H5 )2 ][PF6 ]32 shows a different phase transitional behaviour, with the two phase transitions taking place at 213 and 347 K, i.e. 39 and 33 K below and above those of crystalline [Co(η5 -C5 H5 )2 ][PF6 ].

(a)

(b)

(c)

Figure 12.2 Representations of the molecular conformations in the structures of the three polymorphic forms of ferrocene: (a) monoclinic, (b) triclinic, (c) orthorhombic phase

468

Ferrocenes: Ligands, Materials and Biomolecules

Co100 −80

−60

−40

−20

0

20

40

60

80

−40

−20

0

20

40

60

80

−40

−20

0

20

40

60

80

−40

−20

0

20

40

60

80

−60

−40

−20

0

20

40

60

80

−60

−40

−20

0 T/°C

20

40

60

80

Co90:Fe10 −80

−60

Co75:Fe25

−80

−60

Co50:Fe50 −80

−60

Co25:Fe75

−80

Fe100

−80

Figure 12.3 DSC thermograms (heating cycle) showing how the RT LT and RT HT transitions ‘diverge’ from the values observed in the case for [Co(η5 -C5 H5 )2 ]C as the percentage of [Fe(η5 -C5 H5 )2 ]C in the mixture increases

Crystal Engineering with Ferrocene Compounds

469

The close structural similarity between cations [M(η5 -C5 H5 )2 ]C (M D Co, Fe) generated the idea of growing crystals from solutions containing mixtures of the two cations. It has been discovered that, in the solid state, the two cations are fully miscible in the whole range of composition and that the composition is the same as that of the water solutions from which the mixed crystals are precipitated, e.g. the mixed salts can be formulated as [(η5 -C5 H5 )2 Cox Fe1x ][PF6 ] (with 0 < x < 1).33 Moreover, the phase transition behaviour depends linearly on the composition following Vegard’s rule.34 The temperature variation of the two solid–solid transitions with composition is shown in Figure 12.3. This is a clear demonstration that the phase transition behaviour of the crystalline material [Cox Fe1-x (η5 -C5 H5 )2 ][PF6 ] can be tuned through changing the [Co(η5 -C5 H5 )2 ]C /[Fe(η5 -C5 H5 )2 ]C molar ratio, because the [Co(η5 -C5 H5 )2 ]C and [Fe(η5 -C5 H5 )2 ]C cations form fully miscible solid solutions. The linear response of physical properties with composition is typical of inorganic alloys. Thus, the mixed crystal [Cox Fe1x (η5 -C5 H5 )2 ][PF6 ], though composed of molecular ions and soluble in water, possesses the features of a random Ax B1x alloy. Structurally similar systems, however, show very different phase transitional behaviours. The compound [Ru(η5 -C5 H5 )(η6 -C6 H6 )][PF6 ]35 does not show a low-temperature phase transition on decreasing the temperature down to 223 K on the DSC and down to 100 K on the diffractometer, but undergoes an order–disorder phase transition on increasing the temperature to 332.5 K. The bis-benzene chromium analogue, [Cr(η6 C6 H6 )2 ][PF6 ], on the other hand, even though it crystallizes in a manner that is strictly related to that of the low temperature phases of cobalt and iron, does not appear to undergo phase changes either on cooling or on heating. Information on the phase transitional behaviour of several metallocene or metallobenzene hexaﬂuorophosphate salts, is collected in Table 12.1. The related compound [Fe(η6 -C6 H6 )(η5 -C5 H5 )][AsF6 ] undergoes phase transitions between three different crystal forms.30, 36 Variable temperature solid state NMR measurements have shown that rotation of the entire cation takes place in a cubic phase above 310 K, while in the intermediate β-phase the rotational motion is restricted to Table 12.1 Phase transition behaviour and thermodynamic data for the family of complexes [M(η5 -C5 H5 )][PF6 ] (M D Co, Fe), [Cr(η6 -C6 H6 )2 ][PF6 ] and [Ru(η5 -C5 H5 )(η6 -C6 H6 )][PF6 ] Species [Fe(η5 -C5 H5 )2 ][PF6 ] 5

[Co(η -C5 H5 )2 ][PF6 ] [Ru(η5 -C5 H5 )(η6 -C6 H6 )][PF6 ] [Cr(η6 -C

6 H6 )2 ][PF6 ]

Phase transition (T/K)

H/kJ mol1

LT ! RT (213.1)a

1.95

RT ! HT

4.50

(347.1)a

LT ! RT (251.8)

a

1.27

RT ! HT (313.9)a

3.06

No RT ! LT transitionb

–

RT ! HT (332.5)

4.16

No RT ! LT

a DSC, heating cycle. b DSC, cooling cycle down to 223 K.

a

transitionb

–

470

Ferrocenes: Ligands, Materials and Biomolecules

a 90Ž in-plane reorientation. Below 270 K the crystal is in a low-symmetry phase, in which whole-body rotation does not take place though the rings execute jumping motion that persists down to 200 K. Transition from a rotational jumping state to a whole-body reorientation has also been detected from the M¨ossbauer spectra of the [PF6 ] salt of the same complex.30, 36 Preformed crystals of structurally similar species have been used to separate different crystals that crystallize concomitantly (concomitant polymorphs).37 Precipitation of [Fe(η5 -C5 H5 )2 ]C as its [AsF6 ] salt generates two types of crystals: a trigonal phase (Fe–T) and a monoclinic phase (Fe–M), which proved difﬁcult to separate out. From another experiment, it was known that crystallization of the congener [Co(η5 C5 H5 )2 ][AsF6 ] only leads to a trigonal form (Co–T) isomorphous with Fe–T. In addition, the monoclinic phase Fe–M is isomorphous with the room temperature, monoclinic phase of the pair [Fe(η5 -C5 H5 )2 ][PF6 ] and [Co(η5 -C5 H5 )2 ][PF6 ]. The relationship between crystalline Fe–T and Fe–M is shown in Figure 12.4. It is worth noting that, besides the differences in packing arrangements of the forms, the cyclopentadienyl ligands are eclipsed in Fe–T, while they are all staggered in Fe–M. To drive the crystallization process towards the formation of the separate polymorphs, heteromolecular seeding was used. Crystals of trigonal [Co(η5 -C5 H5 )2 ][AsF6 ] were used to grow the trigonal form of [Fe(η5 -C5 H5 )2 ][AsF6 ], while crystals of monoclinic [Fe(η5 C5 H5 )2 ][PF6 ] were used to obtain the monoclinic form of [Fe(η5 -C5 H5 )2 ][AsF6 ].38 An interesting case of crystal polymorphism is represented by the boronic acid derivative (4-C5 H4 N)fcB(OH)2 , which has been isolated in three forms, the two anhydrous forms I and II and the monohydrate form III. The three forms are compared

[Fe(C5H5)2][AsF6] (aq)

[Co(C5H5)2][AsF6]

trigonal

[Co(C5H5)2][PF6]

trigonal + monoclinic

monoclinic

[Fe(C5H5)2][AsF6]

Figure 12.4 Precipitation of [Fe(η5 -C5 H5 )2 ][AsF6 ] salt generates a trigonal phase (Fe–T) and a monoclinic phase (Fe–M), which can be separated out by heteromolecular seeding with isomorphous crystals of trigonal [Co(η5 -C5 H5 )2 ][AsF6 ] and of monoclinic [Fe(η5 -C5 H5 )2 ][PF6 ]. The crystals are sufﬁciently robust to undergo a full cycle of four phase transitions directly on the diffractometer, Fe–T ! Fe–M ! Fe–C ! Fe–M ! Fe–T

Crystal Engineering with Ferrocene Compounds

471

Figure 12.5 Hydrogen-bonded dimers formed by (B)OHžžžN interactions in crystalline forms I (top), II (middle) and III (bottom)

472

Ferrocenes: Ligands, Materials and Biomolecules

in Figure 12.5.39, 40 In crystalline II the molecules form dimers via (B)OHžžžN bonds, which are then linked in a secondary pattern by the (B)OHžžžO(B) lateral bonds. This arrangement leads to eclipsing of the B(OH)2 group over the pyridyl group. In II, on the other hand, the primary motif appears to be the boronic acid ring based on (B)OHžžžO(B) bonds while dimers are formed by the lateral OH groups and the pyridyl acceptors. The conformation of the two ligands is cisoid. In crystalline III a third, almost intermediate, topology is observed. Crystalline fc(COOH)2 is known in three polymorphic forms. A monoclinic (form I) and a triclinic (form II) crystal form were determined decades ago;40, 41 a third polymorphic form III has been obtained more recently.42 All three crystal forms are based on dimers [fc(COOH)2 ]2 of doubly hydrogen bonded carboxylic rings and do not interconvert via a phase transition. The simplest way to compare the three crystal structures is by looking at molecular layers formed by these supramolecular dimers. A representation of a section of the three forms is shown in Figure 12.6. Recently, we have shown that interconversion of form II into form I can be obtained via an indirect route.43 Both crystals react with vapours of ammonia to yield the same ammonium salt [NH4 ]2 [fc(COO)2 ] which, upon ammonia removal, convert exclusively into form I as represented in Figure 12.7. Finally, it is worth mentioning that the metastable room temperature phase of crystalline 1,10 -diformylferrocene (fc(CHO)2 ) on heating undergoes a transition at ca. 311 K. On cooling, however, the reverse process leads to a new phase.44 Subsequent cycles of heating and cooling show that the new phase reversibly switches between the room temperature and high temperature phases, without reverting to the initial phase. This behaviour suggests that the ﬁrst room temperature phase is a kinetic product of the crystallization process: on heating, the sample undergoes an order ! disorder phase transition to the plastic phase, which transforms on cooling to a thermodynamically more stable form. Once this latter phase is formed, the initial phase

Figure 12.6 An in-plane projection of the packings in (top) form I (monoclinic), (middle) form II (triclinic) and (bottom) form III (triclinic) of crystalline fc(COOH)2

Crystal Engineering with Ferrocene Compounds

473

Figure 12.6 (continued )

Form I monoclinic

hydrated NH 3(vap)

hydrated NH 3(vap)

Form II triclinic

− NH3

Figure 12.7 Form II and form I of fc(COOH)2 react with ammonia to yield the ammonium salt, which converts only to form I of the acid upon ammonia removal (salts prepared from methyl- and dimethylamine react similarly)

can no longer be obtained, unless the compound is redissolved and recrystallised. The H of transition (14.0 kJ. mol1 ) is comparable with the value (12.1 kJ. mol1 ) found for the mono-formyl derivative FcCHO, which shows a mesophase between 316 K and the melting point (396 K).28 On further heating crystalline fc(CHO)2 undergoes polymerization.

474

12.3 12.3.1

Ferrocenes: Ligands, Materials and Biomolecules

Crystal Engineering with Ferrocene Building Blocks Hydrogen Bonded Networks Templated by Organometallic Sandwich Units

The assembly of ferrocenium and ferrocenium-like cations in network structures of polycarboxylic acids has been thoroughly exploited via a combination of redox and acid-base processes (Table 12.2).45–50 The oxidation by molecular oxygen of the neutral complexes [Co(η5 -C5 H5 )2 ], [Cr(η6 -C6 H6 )2 ], [Fe(η5 -C5 Me5 )2 ] and [Co(η5 C5 Me5 )2 ] to [Co(η5 -C5 H5 )2 ]C , [Cr(η6 -C6 H6 )2 ]C , [Fe(η5 -C5 Me5 )2 ]C and [Co(η5 C5 Me5 )2 ]C , respectively, generates the strongly basic anion O2 , which is able to fully or partially deprotonate the polyprotic acid, depending on the stoichiometric ratio. Since the oxidation products are not suitable for coordination by the COO() groups, self assembly of the polycarboxylic acid is forced with formation of one, two or three-dimensional homo-ionic framework structures around the organometallic cations. The interaction between organic framework and organometallic cations is based on charge assisted CHžžžO bonds between cations and anions. When oxalic acid was used with [Fe(η5 -C5 Me5 )2 ]C and [Cr(η6 -C6 H6 )2 ]C , the crystalline compounds [Fe(η5 -C5 Me5 )2 ][HC2 O4 ]ž[H2 C2 O4 ]0.5 and [Cr(η6 -C6 H6 )2 ] [HC2 O4 ]ž[H2 O] were prepared and isolated.45 All crystal structures are characterized by the presence of a columnar aggregation of the radical cations [Fe(η5 -C5 Me5 )2 ]C and [Cr(η6 -C6 H6 )2 ]C . Earlier studies of crystals containing the radical anions51 tetracyanoethylene (TCNEž ) and tetracyano-p-quinodimethane (TCNQž ) or other similar anions together with the radical cations [Fe(η5 -C5 Me5 )2 ]C , as well as other sandwich cations (e.g. [Cr(η6 -C6 Mex H6x )2 ]C ), have led to the formulation of the linear-chain paradigm, which postulates the formation of sequences of the type A() –C(C) –A() – C(C) between radical anions and cations in crystals.8, 52 Linear arrangements of the type C(C) –C(C) –C(C) –C(C) surrounded by chains of A() –A() –A() –A() anions, or Table 12.2 Principal ferrocene complexes used for the construction of hydrogen bonded networks Organometallic complex

Formula

Ref.

Ferrocene-1,10 -dicarboxylic acid

fc(COOH)2

40–42, 53–55

1,10 -bis(ethenyl-4-pyridyl)

fcfCHDCH(C5 H4 N-4)g2

56–58

1,10 -bis(L-Ala-L-Pro-OEt)

ferrocene

ferrocene

Ferrocene-1,10 -diboronic acid [10 -(4-pyridyl)ferrocenyl]boronic

acid

fc(L-Ala-L-Pro-OEt)2

62

fc[B(OH)2 ]2

39, 63

(C5 H4 N-4)fc[B(OH)2 ]

39

1,10 -bis(diphenylhydroxymethyl)ferrocene

fcfC(OH)(Ph2 g2

64, 65

rac -2-(diphenylphosphino)ferrocenylmethanol

[Fe(η5 -C

5 H5 5 H3 -1(CH2 OH)-2-PPh2 )]

65

9-[N -(ferrocenylmethyl)carbamoyl] anthracene

[Fe(η5 -C5 H5 )(C21 H16 NO)]

66

)(η5 -C

Crystal Engineering with Ferrocene Compounds

475

intercalated between layers of anions, are not easily obtained, because of the tendency of the ions of a given sign to be surrounded by ions of the opposite sign. In crystalline compound [Fe(η5 -C5 Me5 )2 ][HC2 O4 ]ž[H2 C2 O4 ]0.5 the hydrogen oxalate anions and the neutral oxalic acid molecules form linear chains (Figure 12.8). The chains run parallel to the columns formed by the [Fe(η5 -C5 Me5 )2 ]C cations, and contain deca-atomic dimers formed by two hydrogen oxalate anions joined together via () O-HžžžO() interactions. The anions are not arranged in the usual head-to-tail chain or carboxylic ring, but adopt a side-to-side arrangement. The hydrogen oxalate dimers are bridged together by one molecule of oxalic acid. The hydrogen oxalate/oxalic acid chain can thus be described as a (H2 C2 O4 )-[(HC2 O4 )(HC2 O4 )]2 -(H2 C2 O4 )[(HC2 O4 )(HC2 O4 )]2 sequence of ions bridged by neutral molecules. These chains form the backbone of the crystal structure and allow the cations to pile up, as shown in Figure 12.8. A twisted conformation is adopted by the hydrogen oxalate anion in the bis-benzene chromium salt [Cr(η6 -C6 H6 )2 ][HC2 O4 ]ž[H2 O]. The twisted hydrogen oxalate anions do not interact with each other, but are linked in the chains by means of water bridges, which form complex 14-membered rings based on four hydrogen bonding interactions. Crystalline [Cr(η6 -C6 H6 )2 ][HC4 O4 ] has been obtained by reaction of squaric acid (3,4-dihydroxy-3-cyclobutene-1,2-dione, H2 C4 O4 ) with [Cr(η6 -C6 H6 )2 ] in THF.46 The ﬂat shape and small dimensions of the squarate anion [HC4 O4 ] leads to its intercalation between the ﬂat benzene ligands of the paramagnetic cation [Cr(η6 C6 H6 )2 ]C , with formation of one-dimensional C(C) –A() –C(C) –A() aggregates, comprised of alternating cation donors (C) and anion acceptors (A). The anions selfassemble into chains linked by () OHO() interactions and intercalates between the ˚ When the reaction is carried out in water, benzene ligands (π –π distance 3.375 A). the hydrated crystalline material [Cr(η6 -C6 H6 )2 ]2 [C4 O4 ]ž6H2 O is instead obtained. [Cr(η6 -C6 H6 )2 ]2 [C4 O4 ]ž6H2 O contains layers of organometallic cations intercalated with layers of water molecules, hydrogen bonded to squarate dianions. Contrary to most organic salts of [Cr(η6 -C6 H6 )2 ]C and [Co(η5 -C5 H5 )2 ]C , which are yellow, crystals of [Cr(η6 -C6 H6 )2 ][HC4 O4 ] are orange in colour. Reﬂectance spectra measured on the crystalline material [Cr(η6 -C6 H6 )2 ][HC4 O4 ] show the presence of an intense tail, which was assigned to a charge transfer transition via the [Cr(η6 -C6 H6 )2 ]C / [HC4 O4 ] π-stacking interactions. When the size of hydrogen bonding organic units exceeds that of the cationic carbocyclic ligands, intercalation is no longer possible. This is demonstrated by the use of ﬂat and large diacids, such as phthalic and terephthalic acid.47 The architecture of crystalline [Co(η5 -C5 H5 )2 ]4 [C6 H4 (COOH)(COO)]2[C6 H4 (COO)2 ]ž4H2 O resembles a brick wall, with large rectangular channels occupied by pairs of columns of [Co(η5 C5 H5 )2 ]C cations. Honeycomb-type organic frameworks can be obtained by using the organic acids D,L- and L-tartaric acid [HO2 CCH(OH)CH(OH)CO2H] forming the achiral [Co(η5 C5 H5 )2 ][(D,L-HO2 CCH(OH)CH(OH)CO2)(D,L-HO2 CCH(OH)CH(OH)CO2H)] and chiral [Co(η5 -C5 H5 )2 ][L-HO2 CCH(OH)CH(OH)CO2] crystals.48 The cobaltocenium cations in both crystals are encapsulated within hexagonal and square organic honeycomb-frameworks, respectively. Similarly, trimesic acid, [C6 H3 -1,3,5-(CO2 H)3 ], has

476

Ferrocenes: Ligands, Materials and Biomolecules

Figure 12.8 [Fe(η5 -C5 Me5 )2 ][HC2 O4 ]ž[H2 C2 O4 ]0.5 : linear chains of [(HC2 O4 )(HC2 O4 )]2 žžž (H2 C2 O4 )žžž[(HC2 O4 )(HC2 O4 )]2

Crystal Engineering with Ferrocene Compounds

477

been used to construct [Co(η5 -C5 H5 )2 ]C f[(C6 H3 (COOH)3 ][C6 H3 (COOH)2 (COO)]g ž 2H2 O.49 Binaphthol, (R)-(C)-1,10 -bi-2-naphthol [(R)-(C)-(HOC10H6 C10 H6 OH)] has been used as a chiral building block in the construction of chiral organic–organometallic crystals.50 The reaction of the neutral molecule with [Co(η5 -C5 H5 )2 ] in ether yields the supramolecular salts [Co(η5 -C5 H5 )2 ][(R)-(C)-(HOC10H6 C10 H6 O)]ž[(R)-(C)-(HOC10 H6 C10 H6 OH)], and [Co(η5 -C5 H5 )2 ][(R)-(C)-(HOC10H6 C10 H6 O)]ž[(R)-(C)-(HOC10H6 C10 H6 OH)]0.5 , depending on the stoichiometric ratios between the binaphthol and the organometallic sandwich compound. In [Co(η5 -C5 H5 )2 ][(R)-(C)-(HOC10H6 C10 H6 O)]ž [(R)-(C)-(HOC10H6 C10 H6 OH)] the neutral molecules act as bridges between monoanions, thus forming a chain system in which anions and neutral molecules alternate, as observed before for [Fe(η5 -C5 Me5 )2 ][HC2 O4 ]ž[H2 C2 O4 ]0.5 . The OHžžžO hydrogen bonds along the chain are of two types: the intermolecular OHžžžO() bond ˚ respectively], and between the neutral spacer and the anion [2.667(7) and 2.612(7) A, the intramolecular hydrogen bond between the OH group of the anion and the depro˚ Because of the enantiomerically pure nature of the tonated oxygen atom [2.427(7) A]. component, the chain is chiral as in the case of L-tartaric acid. In summary, the aggregation of partially deprotonated polycarboxylic acids or polyalcohols follows a sort of aufbau hierarchy: (i) priority goes to the maximization of the number of hydrogen bonding interactions between strong donors and acceptors, (ii) a compromise is also required between space ﬁlling and charge equalization, (iii) weaker CHžžžO and similar interactions are also optimized when all strong interactions have been accommodated. Even though the stability of the aggregate also depends on the topological features of the counterions, () OHžžžO() hydrogen bonding interactions between building blocks of like charges (as it is generally the case of the compounds described in this section) contribute to crystal stability by decreasing the anion–anion (or cation–cation) repulsions. 12.3.2

Hydrogen Bonded Networks Formed by Sandwich Units

In this section the focus is on the supramolecular bonding of sandwich complexes, which themselves carry hydrogen bonding donor/acceptor groups (Table 12.2). These complexes participate in the hydrogen bonding network allowing construction of hybrid organic–organometallic and organometallic–organometallic structures. As will be apparent in the following, a large number of hydrogen bonded networks use ferrocenyl complexes. Ferrocene-1,10 -dicarboxylic acid fc(COOH)2 has been widely used as a hydrogen bonding building block in the preparation of hybrid organic–organometallic and of organometallic–organometallic crystal architectures. A similar strategy was used to produce mixed-metal crystalline materials, the organometallic cation is used to template the anionic hydrogen bonded network formed by this organometallic acid. The crystalline salts [Co(η5 -C5 H5 )2 ][fc(COOH)(COO)], [Co(η5 -C5 H5 )2 ][fc(COOH)(COO)]ž H2 O, [Cr(η6 -C6 H6 )2 ]2 f[fc(COOH)(COO)]2[fc(COOH)2 ]g and [Cr(η6 -C6 H6 )2 ][fc (COOH)(COO)]žH2 O have been prepared and structurally characterized.53 The four species contain different electronic and spin metal centres: 18-electron iron(II) and

478

Ferrocenes: Ligands, Materials and Biomolecules

Figure 12.9 [Co (η5 -C5 H5 )2 ][fcCOOH)(COO)]: the anions [fc(COOH)(COO)] form parallel chains enclosing the [Co(η5 -C5 H5 )2 ]C cations

cobalt(III) metal atoms are present in [Co(η5 -C5 H5 )2 ][fc(COOH)(COO)] and [Co(η5 C5 H5 )2 ][fc(COOH)(COO)]žH2 O, whereas 18-electron iron(II) and paramagnetic 17electron chromium(I) are present in f[Cr(η6 -C6 H6 )2 ]g2 f[fc(COOH)(COO)]2[fc (COOH)2 ]g and [Cr(η6 -C6 H6 )2 ][fc(COOH)(COO)]žH2 O. The crystalline ediﬁces are held together by the complementary contribution of neutral OHžžžO and/or negatively charged OHžžžO() hydrogen bonding interactions between the acid moieties and of charge assisted CHδC žžžOδ bonds between cations and anions. In crystalline [Co(η5 C5 H5 )2 ][fc(COOH)(COO)] the [fc(COOH)(COO)] mono anions form chains via symmetric OžžžHžžžO interactions between ligands in transoid conformation (Figure 12.9). Crystalline [Cr(η6 -C6 H6 )2 ]2 f[fc(COOH)(COO)]2[fc(COOH)2 g shows the presence of pairs of [Cr(η6 -C6 H6 )2 ]C cations in the packing. Acid fc(COOH)2 has been widely employed as a building block in supramolecular chemistry, because the carboxylic groups can easily react and form hydrogen bonds with suitable bases such as bis-amidines or diamines. Moreover, an acid:base stoichiometric ratio of 2:1 allows partial deprotonation of the acid, thus affording species that show the simultaneous presence of homo-ionic OH() žžžO() and heteroionic NH(C) žžžO() interactions, as in the compounds [C8 H16 N4 ][fc(COOH)(COO)]2 and [C10 H20 N4 ][fc(COOH)(COO)]2, obtained by reacting fc(COOH)2 with the bisamidines [C8 H14 N4 ], and [C10 H18 N4 ], respectively.53 Figure 12.10 shows the twodimensional network in [C8 H16 N4 ][fc(COOH)(COO)]2: the ionic arrangement can be described as composed of chains of OH() žžžO() interacting fc(COOH)(COO) anions joined by bis-amidines bridges; the crystalline structure can also be described as

Crystal Engineering with Ferrocene Compounds

479

Figure 12.10 The two-dimensional network in [C8 H16 N4 ][fc(COOH)(COO)]2

formed of chains of singly deprotonated acids interacting with the protonated bisamides in dihapto mode. In a related approach, Glidewell et al. have reported54 the preparation of nine adducts formed between fc(COOH)2 and the organic nitrogen compounds: methylamine, 1,4diazabicyclo[2.2.2]octane, 4,40 -bipyridyl, morpholine, octylamine, piperidine, dicyclohexylamine, tris(2-aminoethyl)amine, and 2-(40 -hydroxyphenyl)ethylamine (tyramine). All the crystal structures show strong Nžžž(H)žžžO hydrogen bonds, and the proton transfer depends on the nature of the nitrogen compound. In the reactions with 1,4-diazabicyclo[2.2.2]octane (stoichiometric ratio 2/1) the molecular components are linked into ﬁnite three-component aggregates by strong OHžžžN hydrogen bonds.55 The strategy of using strong Nžžž(H)žžžO hydrogen bonds to construct the network has been exploited also when the organometallic moiety carries the pyridyl groups. For example, the complex 1,10 -bis[2-(4-pyridyl)ethenyl]ferrocene has been cocrystallized with 1,3-dihydroxybenzene (resorcinol) and 1,3,5-trihydroxybenzene (phloroglucinol)56 to form hydrogen bonded molecular clips (Figure 12.11), which lock the conformation of the organometallic fragment in a dipolar arrangement, a prerequisite for a nonlinear optical molecules. The same ferrocenyl complex was also cocrystallized with 4-alkyloxybenzoic acids, 4-Cn H2nC1 OC6 H4 COOH (n D 6, 7, 8).57 The structures contain supramolecular arrays, in which one molecule of fcfCHDCHC5 H4 N-4)g2 and two molecules of alkyloxybenzoic acids are held together by strong OHžžžN hydrogen bonding. Self-assembly with terephthalic acid, isophthalic acid, phthalic acid and trimesic acid has also been reported.57 When 1,10 -bis[2-(4-pyridyl)ethenyl]ferrocene is cocrystallized with trimesic acid in the ratio of 2:3 a cage-like supramolecular array is formed (Figure 12.12). In the case of isophthalic acid, proton transfer from the carboxylic group to the pyridyl nitrogen takes place with formation of the cation [fc(CHDCHC5 H4 N-4)(CHDCHC5 H4 NH)]C . The cations are assembled via two NHžžžN hydrogen bridges between protonated and neutral pyridine groups and form a dicationic cyclic unit, in which two

480

Ferrocenes: Ligands, Materials and Biomolecules

Figure 12.11 The molecular clip formed by fc(CHDCHC5 H4 N)2 and resorcinol

Figure 12.12 The supramolecular array obtained with trimesic acid and fc(CHDCH C5 H4 N)2

Crystal Engineering with Ferrocene Compounds

481

b axis

Figure 12.13 Two-dimensional layered structure of the cocrystal between fc(CHDCH C5 H4 N)2 and 1,10 -binaphthol assembled through the NžžžHO and CHžžžπ interactions (view down the a axis); note the perfect alignment of molecular dipole moments along the b axis

pyridines are stacked via π –π interaction. These supramolecular units, in turn, pile up to form a columnar structure. 1,10 -[2-(4-Pyridyl)ethenyl]ferrocene has been used for NLO applications.58 In the crystal structure of the complex, the two substituents on the cyclopentadienyl rings adopt a typical antiperiplanar conformation, while upon cocrystallization with 1,10 binaphthol in methanol and ethanol the two substituents of the ferrocene adopt a synclinal conformation.58 The use of optically pure 1,10 -binaphthol leads to a noncentrosymmetric packing arrangement (Figure 12.13), with all the molecular dipoles aligned in the same direction, i.e. the crystal is polar. Ferrocene systems bearing dipeptide chains have been thoroughly investigated, since ferrocenes are recognized as organometallic scaffolds for molecular receptors and peptide mimetic models.59, 60 Conformational enantiomers based on the torsional twist about the Cp(centroid)–Fe–Cp(centroid) axis have been studied in the case of 1,10 disubstituted ferrocenes.61 Generally, conformational enantiomers easily interconvert because of the low energy barrier involved in the cyclopentadienyl ring twisting. The introduction of peptide chains into ferrocene induces conformational enantiomerization by restricting the torsional twist through chirality organization based on the intramolecular hydrogen bondings.62 The complex [Fe(η5 -C5 H4 -L-Ala-L-Pro-OEt)2 ] ˚ shows a helical molecular arrangement in the solid state with turns of 14.91 A. Another interesting hydrogen bonded system is that obtained when the boronic group is used over an organometallic ferrocenyl moiety. The diboronic acid derivative fc[B(OH)2 ]2 has been investigated by two independent groups.39, 63 It forms chains of hydrogen-bonded ferrocenyl moieties in a transoid conformation (Figure 12.14). The derivative fc(C5 H4 N)(B(OH)2 ), obtained from fc[B(OH)2 ]2 , carries both a B(OH)2 unit59 and a C5 H4 N unit has been isolated in three crystal forms.39 These compounds will be described later in the section devoted to polymorphism.

482

Ferrocenes: Ligands, Materials and Biomolecules

Figure 12.14 The structure of fc[B(OH)2 ]2 : (top) chains of hydrogen-bonded ferrocenyl moieties in a transoid conformation; (bottom) criss-crossing of the chains in the crystal structure and establishment of hydrogen-bonding cross-links with lateral protons

Other hydrogen bonded adducts have been obtained by reacting 1,10 - bis(diphenylhydroxymethyl)ferrocene with a wide range of amines, particularly heteroaromatic amines and diamines,55, 64 while the use of the hydroxyl group as a hydrogen bonding donor has been reported in the crystal structures of ferrocenylmethanol derivatives bearing phosphorus substituents such as rac-2-(diphenylphosphino)ferrocenylmethanol, [Fe(η5 -C5 H5 )(η5 -C5 H3 -1-(CH2 OH)-2-PPh2 )], rac-2-(diphenylphosphinoyl)ferrocenylmethanol [Fe(η5 -C5 H5 )(η5 -C5 H3 -1-(CH2 OH)-2-P(O)Ph2 )] and rac-2-(diphenylthiophosphoryl)ferrocenylmethanol [Fe(η5 -C5 H5 )(η5 -C5 H3 -1-(CH2 OH)-2-P(S)Ph2 )], and with β-aminoalcohol FcCH2 NHCMe2 CH2 OH.65 Hydrogen bonding interactions

Crystal Engineering with Ferrocene Compounds

483

between amido groups have been observed in the crystal structure of N -(ferrocenylmethyl)anthracene-9-carboxamide [Fe(η5 -C5 H5 )(C21 H16 NO)].66 The neutral diamagnetic organometallic dicarboxylic acid [Cr0 (η6 -C6 H5 COOH)2 ] and the paramagnetic zwitterion [CrI (η6 -C6 H5 COOH)(η6 -C6 H5 COO)] (obtained as a cocrystal with ammonium hexaﬂuorophosphate) both form hydrogen bonded dimers in their solids, while the paramagnetic dicarboxylic acid cation, [CrI (η6 -C6 H5 COOH)2 ]C , crystallizes with formation of homo-ionic chains.67 The water soluble dicarboxylic cationic acid [Co(η5 -C5 H4 COOH)2 ]C 68 has proven to be an extremely versatile building block for the construction of hydrogen bonded networks. What is more, the networks formed by the complex in cationic or zwitterionic form can be made react in solid–solid and solid–gas processes (see below). Removal of one proton from [Co(η5 -C5 H4 COOH)2 ]C leads to formation of the neutral zwitterion [Co(η5 -C5 H4 COOH)(η5 -C5 H4 COO)], while further deprotonation leads to formation of the dicarboxylate mono anion [Co(η5 -C5 H4 COO)2 ] . The cationic form [Co(η5 C5 H4 COOH)2 ]C has been characterized as the hexaﬂuorophosphate or chloride salt, as well as in cocrystals with urea and with the zwitterionic form [Co(η5 -C5 H4 COOH)(η5 C5 H4 COO)] in f[Co(η5 -C5 H4 COOH)(η5 -C5 H4 COO)] [Co(η5 -C5 H4 COOH)2 ]g[PF6 ].69 When aqueous solutions of f[Co(η5 -C5 H4 COOH)(η5 -C5 H4 COO)][Co(η5 -C5 H4 COOH)2 ]g[PF6 ] are treated with alkali metal or ammonium hydroxides MOH (M D K, Rb, Cs, and NH4 ) in 1:1 stoichiometric ratio, the acid cation [Co(η5 -C5 H4 COOH)2 ]C is partially deprotonated and the zwitterion [Co(η5 -C5 H4 COOH)(η5 -C5 H4 COO)] is formed. Upon crystallization [Co(η5 -C5 H4 COOH)(η5 -C5 H4 COO)] forms a series of nearly isomorphous supramolecular aggregates with the inorganic salts MPF6 (M D K, Rb, Cs, NH4 ).69 In all these crystalline materials the cations are encapsulated via either MC žžžO interactions (M D K, Rb, Cs) or NHžžžO hydrogen bonds (M D NH4 ), within a strongly nucleophilic cage formed by four molecules of [Co(η5 -C5 H4 COOH)(η5 -C5 H4 COO)]. The ‘walls’ of the cage comprise two dimeric units of [Co(η5 -C5 H4 COOH)(η5 -C5 H4 COO)], held together by two OH – O hydrogen bonds and by two CH – O bonds, these latter involving the hydrogen atoms of the cyclopentadienyl rings and the free lone pairs on the carboxylic oxygens. Alternatively, the same compounds can be prepared by treating aqueous solutions of the zwitterion with a stoichiometric amount of the appropriate [PF6 ] salts, i.e. MPF6 (M D K, Rb, Cs, and NH4 ) or by mechanochemical mixing of the crystalline solids.70 The use of sandwich complexes as supramolecular building blocks has been recently extended to mono and bis-amido derivatives of [Co(η5 -C5 H4 COOH)2 ]C . The prototype of this class of complexes is the carboxyl–amide [Co(η5 -C5 H4 CONHC5 H4 N)(η5 C5 H4 COOH)]C .71 The complex [Co(η5 -C5 H4 CONHC5 H4 NH)(η5 -C5 H4 COO)][PF6 ] possesses an eclipsed conformation of the ligands in the solid state. X-ray diffraction shows that the carboxylic group is deprotonated, while the nitrogen atom of the pyridine group is protonated. As a consequence of the formal proton transfer from the carboxylic group to the pyridine, the global ionic charge of the complex does not change, but the complex acquires a zwitterionic nature. In the crystal, two cations [Co(η5 -C5 H4 CONHC5 H4 NH)(η5 -C5 H4 COO)]C are linked together via a bifurcate NH(C) žžžO() hydrogen bond, forming the dimer shown in Figure 12.15.

484

Ferrocenes: Ligands, Materials and Biomolecules

Figure 12.15 (top) The dimeric unit f[Co(η5 -C5 H4 CONHC5 H4 NH)(η5 -C5 H4 COO)]C g2 in crystalline [Co(η5 -C5 H4 CONHC5 H4 NH)(η5 -C5 H4 COO)][PF6 ], formed via bifurcate N-H(C) žžžO() hydrogen bonding interactions. (bottom) The ferrocene-1,10 -dicarboxylic acid molecule and the diamido molecule [Co(η5 -C5 H4 CONHC5 H4 NH)2 ] in [Co(η5 C5 H4 CONHC5 H4 N)2 ][fc(COOH)2 ][PF6 ] are linked via OHžžžN hydrogen bonds

The cationic bis-amide[Co(η5 -C5 H4 C(O)NHC5 H4 NH)2 ] ion has been used in the formation of a cocrystal with fc(COOH)2 .71 In crystalline [Co(η5 -C5 H4 CONHC5 H4 N)2 ][fc(COOH)2 ][PF6 ] the two moieties are linked by an OHžžžN hydrogen bond forming a sort of dimer that recalls the one observed in crystalline [Co(η5 -C5 H4 CONH C5 H4 NH) (η5 -C5 H4 COO)] [PF6 ].71 12.3.3

Mechanochemical Preparation of Hydrogen Bonded Adducts

All reactions described thus far, with the notable exception of the complexes formed from [Co(η5 -C5 H4 COOH)(η5 -C5 H4 COO)] and the salts MPF6 , take place in solution where an appropriate solvent is used. In recent years it has been shown, however, that direct reaction between solid reactants is a viable alternative way to prepare novel molecular crystals containing ferrocenyl moieties. This approach has been successfully used for the preparation of whole classes of novel adducts. It is useful to remember that reactions involving solid reactants or occurring between solids and gases avoid the recovery, storage and disposal of solvents, hence they are of interest in the ﬁeld of ‘green chemistry’, where environmentally friendly processes are actively sought.72 Furthermore, solvent-less reactions often lead to very pure products and reduce the formation of solvate species.73, 74 In spite of these investigations, solid- state processes involving organometallic systems have only recently begun in a systematic way.73, 74

Crystal Engineering with Ferrocene Compounds

NH2

485

N N

H2N

H2N

HN NH

NH2

NH2 CO3

H2N NH2

2

Figure 12.16 Grinding of solid fc(COOH)2 with the solid bases 1,4-diazabicyclo[2.2.2]octane, guanidinium carbonate, 1,4-phenylenediamine, piperazine and trans-1,4cyclohexanediamine generates quantitatively the corresponding adducts [HC6 H12 N2 ][fc (COOH)(COO)], [C(NH2 )3 ]2 [fc(COO)2 ]ž2H2 O, [HC6 H8 N2 ][fc(COOH)(COO)], [H2 C4 H10 N2 ][fc(COO)2 ] and [H2 C6 H14 N2 ][fc(COO)2 ]ž2H2 O

Manual grinding of the ferrocene-1,10-dicarboxylic acid with solid nitrogen containing bases, namely 1,4-diazabicyclo[2.2.2]octane, 1,4-phenylenediamine, piperazine, trans-1,4-cyclohexanediamine and guanidinium carbonate, generates quantitatively the corresponding organic–organometallic adducts (Figure 12.16).75 The case of the adduct [HC6 N2 H12 ][fc(COOH)(COO)] is particularly noteworthy, because the same product can be obtained in three different ways: by reaction of solid fc(COOH)2 with vapours of 1,4-diazabicyclo[2.2.2]octane (which possesses a small but signiﬁcant vapour pressure); by reaction of solid fc(COOH)2 with solid 1,4-diazabicyclo[2.2.2]octane, i.e. by cogrinding of the two crystalline powders; and by reaction in MeOH solution of the two reactants. It is also interesting to note that the base can be removed by mild treatment, regenerating the structure of the starting dicarboxylic acid. Bis-substituted pyridine/pyrimidine ferrocenyl complexes have also been obtained by a mechanically-induced Suzuki coupling reaction76 in the solid state starting from

486

Ferrocenes: Ligands, Materials and Biomolecules

Figure 12.17 The metallo–macrocycles produced by reaction of [fc(4-C5 H4 N)2 ] and the salts Zn(CH3 COO)2 (left) and Cd(NO3 )2 (right)

ferrocene-1,10 -diboronic acid (fc[B(OH)2 ]2 ).39 The ligand fc(4-C5 H4 N)2 , obtained by both solution and solid state methods, was then used to prepare a whole family of hetero-bimetallic metallo-macrocycles by reaction with AgNO3 , Cd(NO3 )2 , Cu(CH3 COO)2 , Zn(CH3 COO)2 and ZnCl2 ; the complexes [fc(4-C5 H4 N)2 ]2 Ag2 (NO3 )2 ž1.5 H2 O, [fc(4-C5 H4 N)2 ]2 Cu2 (CH3 COO)4 ž3H2 O, [fc(4-C5 H4 N)2 ]2 Cd2 (NO3 )4 žCH3 OHž 0.5C6 H6 , [fc(4-C5 H4 N)2 ]2 Zn2 (CH3 COO)4 and [fc(4-C5 H4 N)2 ]2 Zn2 Cl4 were obtained (examples are shown in Figure 12.17).77 Reaction of mechanochemically prepared [fc(4-C5 H4 N)2 ] with fc(COOH)2 has led to formation of the supramolecular adduct fc(4-C5 H4 N)2 žfc(COOH)2 .77 More recently, the same building block has been used in the preparation of supramolecular metallomacrocycles with dicarboxylic acid of variable aliphatic chain length.78 The supramolecular macrocyclic adducts of general formula fc(4C5 H4 N)2 ž2[HOOC(CH2 )n COOH] with n D 4 (adipic acid), n D 6 (suberic acid), n D 7 (azelaic acid), n D 8 (sebacic acid) have been obtained quantitatively by kneading powdered samples of the crystalline organometallic and organic reactants with drops of MeOH (for n D 4, 6, 7) and by direct crystallization from MeOH for n D 8 (sebacic), while the adduct with n D 5 (pimelic) represents an isomeric open chain alternative to the macrocycle. All complexes, with the exception of that involving pimelic acid, share a common structural feature, namely the formation of supramolecular macrocycles constituted of two organometallic and two organic units linked in large tetramolecular units by OHžžžN hydrogen bonds between the carboxyl groups of the dicarboxylic acids and the nitrogen atom of the ferrocenyl complex. The structures of fc(4-C5 H4 N)2 ž2[HOOC(CH2 )n COOH], where n D 6–8 are shown in Figure 12.18. It can be appreciated how the even/odd alternation of carbon atoms in the organic spacers is accommodated by the twist of the cyclopentadienyl–pyridyl groups and by the eclipsed or staggered juxtaposition of the organic moieties Other results have been obtained by using the zwitterion sandwich complex [CoIII 5 (η -C5 H4 COOH)(η5 -C5 H4 COO)].69 Thanks to its amphoteric behaviour the complex undergoes reversible gas-solid reactions with the hydrated vapours of a variety of acids (e.g. HCl,79 CF3 COOH, CCl3 COOH, CHF2 COOH, HBF4 , HCOOH)80–82 and

Crystal Engineering with Ferrocene Compounds

487

adipic (4)

suberic (6)

azelaic (7)

sebaic (8)

Figure 12.18 The supramolecular structures of the macrocycles in solid adducts fc(1C5 H4 N)2 žHOOC(CH2 )n COOH, where n D 4 (a), 6 (b), 7 (c) and 8 (d), showing the hydrogen bond links between the two outer organometallic molecules and the inner organic spacers

bases (e.g. NH3 ,79 NMe3 , NH2 Me)79 as well as solid–solid reactions with crystalline alkali and ammonium salts of the formula MX (M D K, Rb, Cs, and NH4 ; X D Cl, Br, I, PF6 , though not in all permutations).80 These products could also be obtained by solution methods, as discussed before. Similar behaviour is shown towards other volatile acids. Exposure of the zwitterion to vapours of CF3 COOH and HBF4 , for instance, quantitatively produces the corresponding salts of the cation [Co(η5 -C5 H4 COOH)2 ]C , namely. [Co(η5 -C5 H4 COOH)2 ] [CF3 COO] and [Co(η5 -C5 H4 COOH)2 ][BF4 ] (Figure 12.19). Exposure of the solid zwitterion to vapours of CHF2 COOH quantitatively produces the corresponding salts of the cation [Co(η5 -C5 H4 COOH)2 ][CHF2 COO]. The zwitterion also reversibly absorbs formic acid from humid vapours forming selectively a 1:1 cocrystal, [Co(η5 -C5 H4 COOH)(η5 -C5 H4 COO)][HCOOH] (Figure 12.19), from which the starting material can be fully recovered by mild thermal treatment. Contrary to the other compounds of this class, no proton transfer from the adsorbed acid to the organometallic moiety has been observed (Figure 12.19). Hence, the reaction between [Co(η5 -C5 H4 COOH)(η5 -C5 H4 COO)] (solid) and HCOOH(vapour) would be more appropriately described as a special kind of solvation rather

488

Ferrocenes: Ligands, Materials and Biomolecules

[CoIII(h5-C5H4COOH)(h5-C5H4COO)]

HBF4(vap) (54 % in DEE), 16 h

443 K, 30 min, vacuum

[CoIII(h5-C5H4COOH)2][BF4]

[CoIII(h5-C5H4COOH)(h5-C5H4COO)]

HCOOH (vap)

, vacuum

[CoIII(h5-C5H4COOH)(h5-C5H4COO)][HCOOH]

Figure 12.19 (top) Exposure of the solid zwitterion [CoIII (η5 -C5 H4 COOH)(η5 -C5 H4 COO)] to vapours of HBF4 quantitatively produces the corresponding salt [Co(η5 -C5 H4 COOH)2 ][BF4 ], while (bottom) exposure to vapours of HCOOH yields the cocrystalline material [Co(η5 C5 H4 COOH)(η5 -C5 H4 COO)][HCOOH]; both products revert back to the solid zwitterion after mild thermal treatment (DEE D diethyl ether)

Crystal Engineering with Ferrocene Compounds

489

than as a heterogeneous acid–base reaction as also conﬁrmed by 13 C CP MAS NMR spectroscopy.82 The behaviour of the zwitterion towards ammonia and other bases is similar to that towards HCl but obviously opposite in terms of proton exchange. In the case of ammonia the neutral system transforms into the hydrated ammonium salt [NH4 ][Co(η5 C5 H4 COO)2 ]ž3H2 O. Finally, on concluding the section on functionalized sandwich compounds capable of hydrogen bond formation, the structures of trovacenyl boronic acid [V(η7 -C7 H7 )(η5 C5 H4 B(OH)2 )] and the carboxylic acid [V(η7 -C7 H7 )(η5 -C5 H4 COOH)] should also be pointed out.73 12.3.4

Hydrogen-Bonded Networks Formed by Coordination Compounds

It have been seen in the previous sections that functionalized ferrocenyl complexes form a large portion of the organometallic complexes investigated for their supramolecular bonding capacity. This popularity extends also to the use of ferrocenyl complexes for the preparation of coordination networks and ‘complexes of complexes’ in which the ferrocenyl derivative is used as a ligand. Some recent results are summarized in Table 12.3. The ferrocenyl complex 1,10 -bis(diphenylphosphino)ferrocene (dppf) has been widely used in supramolecular chemistry,83 but has only recently begun to be exploited in the formation of coordination networks (see Chapter 2). For example the reaction of dppf with silver triﬂate has been reported to yield the two dimensional (2-D) coordination network fAg4 (SO3 CF3 )4 (dppf)2 g.84 Heteropolynuclear organometallic coordination networks have been constructed by using fc(COOH)2 as a spacer.85 When the diacid reacts with d-block transition metal

Table 12.3 Coordination networks and ‘complexes of complexes’ in which the ferrocenyl derivative is used as a ligand Organometallic building block

Metal ion

Formula of the resulting system

Ref.

AgC

fAg4 (SO3 CF3 )4 (dppf)2 g

84

2

Zn2C

NaZn3 [fc(COO)2 ]2 (OH)3 (H2 O)

85

fc(COO)2 2

Cd2C

Cd[fc(COO)2 ](DMF)2 (H2 O)

85

dppf fc(COO)2 fc(COO)2

2

La

fLa2 [fc(COO)2 ]3 (CH3 OH)4 g1

86

fc(COO)2

]2

Eu3C

[Eu2 [fc(COO)2 ]3 (H2 O)5 ]1

86

fc(COO)2

2

Gd3C

fGd2 [fc(COO)2 ]3 (CH3 OH)2 (H2 O)3 g1

86

fAg[fc(S-4-C5 H5 N)2 ](CH3 CN)2 g[PF6 ]

87

3C

fc(S-4-C5 H5 N)2

AgC

fc(S-4-C5 H5 N)2

Mn2C ,

fM[fc(S-4-C5 H5 N)2 ][CH(COCF3 )2 ]2 g

88

fc(S-2-C5 H5 N)2

CuC

fCu[fc(S-2-C5 H5 N)2 ]g[PF6 ]

88

Cu2C

fCu[Co(η5 -C

89

[Co(η5 -C

5 H4 COO)2

]2

Cu2C ,

Zn2C

5 H4 COO)2 ]2 g(CH3 OH)2

490

Ferrocenes: Ligands, Materials and Biomolecules

ions, the ligand fc(COO)2 2 generally adopts a cisoid conformation, with formation leading to the formation of the discrete molecular architectures, but in the case of zinc and cadmium the coordination polymers NaZn3 [fc(COO)2 ]2 (OH)3 (H2 O) and Cd[fc(COO)2 ](DMF)2 (H2 O) have been obtained (DMF D N ,N 0 -dimethylformamide). The conformation of the cyclopentadienyl rings in NaZn3 [fc(COO)2 ]2 (OH)3 (H2 O) is syn-periplanar and this special conformation plays an important role in determining the unusual topological motif of NaZn3 [fc(COO)2 ]2 (OH)3 (H2 O) (Figure 12.20). Heteropolynuclear organometallic compounds based on fc(COOH)2 have also been constructed.86 Interaction of the acid with La3C , Eu3C and Gd3C ions affords the 2-D networks fLa2 [fc(COO)2 ]3 (CH3 OH)4 g1 , fEu2 [fc(COO)2 ]3 (H2 O)5 g1 and fGd2 [fc (COO)2 ]3 (CH3 OH)2 (H2 O)3 g1 , respectively, in which the transoid conformation of the ferrocene moiety provides opportunities to form inﬁnite 2-D networks. In addition to this, π žžžπ interactions between the ferrocene moieties were also found to stabilize the supramolecular architectures in the solid state. Complex fLa2 [fc(COO)2 ]3 (CH3 OH)4 g1 , for example, contains a 2-D- network structure (Figure 12.21). The two independent ferrocenedicarboxylate units adopt a different conformation: syn-periplanar and anti -periplanar. The two ligands in syn-periplanar conformation coordinate two lanthanum atoms forming a metallacycle. The other coordination sites of the lanthanum metals are occupied by methanol and the oxygen atoms of the fc(COO)2 unit in anti -periplanar conformation. The latter oxygen atoms act as bridges between the metallacycles, generating an inﬁnite 3-D network.

Figure 12.20 The ion organization in crystalline NaZn3 [fc(COO)2 ]2 (OH)3 (H2 O) viewed along the b axis (top) and c axis (bottom)

Crystal Engineering with Ferrocene Compounds

491

Figure 12.21 Two-dimensional layer of complex fLa2 [fc(COO)2 ]3 (CH3 OH)4 g1 showing the different conformations and coordination modes of the ligands

Crystalline fEu2 [fc(COO)2 ]3 (H2 O)5 g1 , and fGd2 [fc(COO)2 ]3 (CH3 OH)2 (H2 O)3 g1 are isostructural and contain a 2-D coordination network constructed from the lanthanide–organometallic layers. The two independent ferrocene ligands adopt a syn-clinal eclipsed and anti -periplanar conformation and each carboxylate group coordinates to a different europium atom generating a two-dimensional brick wall structure. The ferrocene-based bidentate ligands, 1,10 -(4-dipyridinethio)ferrocene and 1,10 -(2dipyridinethio)ferrocene have been used by Mochida and coworkers to construct coordination networks.87 To obtain coordination polymers, these bidentate ligands were combined with metal ions bearing two coordination sites, silver(I), copper(I) and M(hfac)2 (M D ﬁrst row transition metal, hfac D hexaﬂuoroacetylacetonate). M(hfac)2 building blocks have been employed in the construction of coordination networks.88 The network in fAg[1,10 -(4-dipyridinethio)ferrocene](PF6)gn is constituted of chains formed by silver ions bridged by 1,10 -(4-dipyridinethio)ferrocene ligands. A relatively ˚ is present between adjacent chains. The reacshort AgžžžAg distance of 3.2670(8) A tion of 1,10 -(4-dipyridinethio)ferrocene with M(hfac)2 (M D manganese, copper, zinc) yielded the respective isomorphous complexes fM(hfac)2 [1,10 -(4-dipyridinethio)ferrocene]gn (M D Mn, Cu, Zn). The structures show a one-dimensional straight chain, which consists of the alternate linkage of M(hfac)2 units and the ligand. The ligand 1,10 -(2-dipyridinethio)ferrocene also afforded a one-dimensional coordination network in the linear coordination with copper(I) centres. The monoanion [Co(η5 -C5 H4 COO)2 ] was used as a unique monoanionic dicarboxylate ligand with copper(II) to construct a new two-dimensional coordination

492

Ferrocenes: Ligands, Materials and Biomolecules

Figure 12.22 A view of the packing in crystalline Cu[Co(η5 -C5 H4 COO)2 ]2 ž2MeOH (MeOH molecules are not shown for clarity)

polymer.89 The crystal structure of Cu[Co(η5 -C5 H4 COO)2 ]2 ž2MeOH is shown in Figure 12.22. It can be seen that four oxygen atoms of the carboxylate groups from the dianion [Co(η5 -C5 H4 COO)2 ] bind to the copper(II) centre to form a distorted square planar geometry. Each [Co(η5 -C5 H4 COO)2 ] anion adopts a conformation between anticlinal–eclipsed and antiperiplanar and connects two copper(II) centres to yield a ˚ 2 ), which trap two-dimensional structure with small square cavities (about 4 ð 4 A methanol molecules. When the dianionic 1,10 -ferrocenedicarboxylate is used with cadmium(II) in a 1:1 ligand to metal ratio, neutral coordination frameworks are obtained, such as f[Cd[fc(COO)2 ](DMF)2 (H2 O)g.85

12.4

Conclusions

Papers and results reported in this chapter represent only a small selection, arbitrary and personal, of the plethora of papers and reviews that have been published on issues related to supramolecular organometallic solid state chemistry or organometallic crystal engineering. Though ‘born organic’ crystal engineering is now expanding rapidly in the neighbouring areas of coordination chemistry (coordination networks, mainly) and organometallic chemistry (metal containing molecular materials). The reasons for this interest are manifold, but it cannot be denied that the advent of powerful tools for the structural characterization of complex systems such as some of those described above, has been a crucial step in the development of the ﬁeld. Undoubtedly, organometallic

Crystal Engineering with Ferrocene Compounds

493

chemistry has beneﬁted from its birth of the ‘power of crystallography’. Without the increase in accuracy and speed of structural data collection and treatment very little would still be known about the structure of complex systems and, hence, of their functions and properties. This is even more appreciable in the ﬁeld of crystal engineering, where traditional analytical tools are of limited assistance. In most cases it is only when the crystal structure of the product is known, that the chemist can rationalize the building up process, the sequence that leads from the selected building blocks, whether molecules or ions, to the supramolecular aggregation and, eventually, reproduce the same sequence successfully. Special attention should be paid to the possibility of multiple choices in the ultimate result of a crystallization process, i.e. to crystal polymorphism, the occurrence of more than one crystal structure for the same compound. Rather than representing the nemesys of crystal engineering, crystal polymorphism represents, if purposely investigated, an important route to information about supramolecular bonding and about the role of crystallization conditions. In general, the temperature tempendance of a given crystal structure should also be routinely investigated. This is also a problem of crystal polymorphism, though associated with solid–solid phase transitions. The knowledge of the thermal stability of a given crystal engineering product is of paramount importance. Finally, examples have been provided of ‘non-solution’ preparation of ferrocenyl based crystalline materials. We believe, that the awareness that crystals can be made from crystals without the intermediacy of a solvent is a useful notion in organometallic crystal engineering and, more generally, in organometallic chemistry.

Acknowledgments We acknowledge ﬁnancial support by MUR and by the University of Bologna. We thank the Swedish Research Council (VR) and PolyCrystalLine for postdoctoral grants to A. Pettersen and M. Curz; respectively.

References 1. a) D. Braga and F. Grepioni, Making Crystals by Design, Wiley-VCH Verlag GmbH, Wienstein, Germany, (2006); b) G.R. Desiraju, Crystal Engeneering: The Design of Organic Solids, Elsevier (1989). 2. a) G.R. Desiraju, Angew. Chem., Int. Ed. Engl., 1995, 34, 2311–2327; b) D. Braga and F. Grepioni, Acc. Chem. Res., 2000, 33, 601–608. 3. D. Braga, F. Grepioni and A.G. Orpen, Crystal Engineering: from molecules and crystals to materials, Kluwer Academic Publishers, The Netherlands (1999). 4. J.M. Lehn, Supramolecular Chemistry: Concepts and Perspectives, VCH, Weinstein, Germany (1995). 5. a) A.J. Blake, N.R. Champness, P. Hubberstey et al. Coord. Chem. Rev., 1999, 183, 117–138; b) S.R. Batten, B.F. Hoskins and R. Robson, Chem. Eur. J., 2000, 6, 156–161; c) B. Rather and M.J. Zaworotko, Chem. Commun., 2003, 830–831; d) B. Moulton, H. Abourahma, M.W. Bradner et al. Chem. Commun., 2003, 1342–1343; e) M. Fujita, Chem. Soc. Rev., 1998, 27, 417–425; f) B. Olenyuk, A. Fechtenkotter and P.J. Stang, Dalton Trans., 1998,

494

6. 7.

8. 9. 10.

11.

12.

13. 14.

Ferrocenes: Ligands, Materials and Biomolecules 1707–1728; g) L. Pan, M.B. Sander, X. Huang et al. J. Am. Chem. Soc., 2004, 126, 1308–1309; h) G. Ferey, M. Latroche, C. Serre et al. Chem. Commun., 2003, 2976–2977; i) F.A. Cotton, C. Lin and C.A. Murillo, Dalton Trans., 2001, 499–501; j) F.A. Cotton, C. Lin and C.A. Murillo, Chem. Commun., 2001, 11–12; k) W. Mori and S. Takamizawa, J. Solid State Chem., 2000, 152, 120–129; l) L. Carlucci, G. Ciani, D.M. Proserpio and S. Rizzato, CrystEngComm, 2002, 4, 121–129; m) J. Lu, B. Moulton, M.J. Zaworotko and S.A. Bourne, Chem. Commun., 2001, 861–862. B. Moulton and M.J. Zaworotko, Chem. Rev., 2001, 101, 1629–1658. a) D.W. Bruce, D. O’Hare (Eds), Inorganic Materials, John Wiley & Sons Ltd Inc., (1992); b) N.J. Long, Angew. Chem., Int. Ed. Engl., 1995, 34, 21–38; c) T.J. Marks and M.A. Ratner, Angew. Chem., Int. Ed. Engl., 1995, 34, 155–173; d) D.R. Kanis, M.A. Ratner and T.J. Marks, Chem. Rev., 1994, 94, 195–242; e) O. Khan, Molecular Magnetism, VCH, Weinheim, Germany, (1993); f) D. Gatteschi, Adv. Mater., 1994, 6, 635–645; g) M.D. Ward, Science, 2003, 300, 1104–1105; h) J.S. Miller, Angew. Chem. Int. Ed., 2003, 42, 27–29; i) T.E. Vos, Y. Liao, W.W. Shum et al. J. Am. Chem. Soc., 2004, 126, 11630–11639; j) E.B. Vickers, I.D. Giles and J.S. Miller, Chem. Mater., 2005, 17, 1667–1672; k) L. Toma, R. Lescouezec, J. Vaissermann et al. Chem. Eur. J., 2004, 10, 6130–6145; l) E. Pardo, K. Bernot, M. Julve et al. Chem. Commun, 2004, 920–921. J.S. Miller and A.J. Epstein, Chem. Eng. News, 1995, 73, 30–41. D. Braga, Chem. Commun., 2003, 2751–2754. a) N.L. Rosi, M. Eddaoudi, J. Kim et al. Angew. Chem. Int. Ed., 2002, 41, 284–287; b) O.M. Yaghi, H. Li, C. Davis et al. Acc. Chem. Res., 1998, 31, 474–484; c) H. Li, M. Eddaoudi, M. O’Keeffe and M. Yaghi, Nature, 1999, 402, 276–279; d) M. Eddaoudi, J. Kim, N. Rosi et al. Science, 2002, 295, 469–472; e) N.L. Rosi, M. Eddaoudi, J. Kim et al. CrystEngComm, 2002, 4, 401–404; f) N.L. Rosi, J. Eckert, M. Eddaoudi et al. Science, 2003, 300, 1127–1130; g) M.E. Braun, C.D. Steffek, J. Kim et al. Chem. Commun, 2001, 2532–2533. a) S.R. Batten and R. Robson, Angew. Chem. Int. Ed., 1998, 37, 1461–1494; b) L. Carlucci, G. Ciani and D.M. Proserpio, CrystEngComm, 2003, 5, 269–279; c) L. Carlucci, G. Ciani and D.M. Proserpio, Coord. Chem. Rev., 2003, 246, 247–289; d) S.A. Barnett and N.R. Champness, Coord. Chem. Rev., 2003, 246, 145–168. a) S. Takamizawa, E-i. Nakata, T. Saito et al. CrystEngComm, 2004, 6, 197–199; b) S. Takamizawa, E-i. Nakata and T. Saito, Angew. Chem. Int. Ed., 2004, 43, 1368–1371; c) S. Takamizawa, E-i. Nakata and T. Saito, CrystEngComm, 2004, 6, 39–41; d) S. Takamizawa, E-i. Nakata, H. Yokoyama et al. Angew. Chem. Int. Ed., 2003, 42, 4331–4334; e) E. Le Fur, E. Demers, T. Maris and J.D. Wuest, Chem. Commun., 2003, 2966–2967; f) D. Laliberte, T. Maris and J.D. Wuest, J. Org. Chem., 2004, 69, 1776–1787; g) O. Saied, T. Maris and J.D. Wuest, J. Am. Chem. Soc., 2003, 125, 14956–14957. K. Biradha, CrystEngComm, 2003, 5, 374–384. a) A. Nangia, CrystEngComm, 2002, 4, 93–101; M.W. Hosseini and A. De Cian, Chem. Commun., 1998, 727–733; b) D. Braga and F. Grepioni, Chem. Commun., 1996, 571–578; c) C.B. Aakeroy and K.R. Seddon, Chem. Soc. Rev., 1993, 22, 397–407; d) H.W. Roesky and M. Andruh, Coord. Chem. Rev., 2003, 236, 91–119; e) A.M. Beatty, Coord. Chem. Rev., 2003, 246, 131–143; f) E.A. Bruton, L. Brammer, F.C. Pigge et al. New J. Chem., 2003, 27, 1084–1094; g) D. Braga, F. Grepioni, K. Biradha et al. J. Am. Chem. Soc., 1995, 117, 3156–3166; h) M.J. Zaworotko, Chem. Soc. Rev., 1994, 23, 283–288; i) M.J. Calhorda, Chem. Commun., 2000, 801–809; j) D. Braga, L. Maini, M. Polito and F. Grepioni, Struct. Bonding, 2004, 111, 1–32; k) C. Janiak, Angew. Chem., Int. Ed. Engl., 1997, 36, 1431–1434; l) S. George, A. Nangia, M. Bagieu-Beucher et al. New J. Chem., 2003, 27, 568–576; m) M. Nishio, CrystEngComm, 2004, 6, 130–158; n) J. Hulliger, Chem. Eur. J., 2002, 8,

Crystal Engineering with Ferrocene Compounds

15.

16.

17. 18. 19.

20.

21.

22. 23.

24.

25. 26. 27.

28. 29. 30.

495

4578–4586; o) L. Brammer, M.D. Burgard, M.D. Eddleston et al. CrystEngComm, 2002, 4, 239–248; p) L. Brammer, Dalton Trans., 2003, 3145–3157; q) A.D. Burrows, Struct. Bonding, 2004, 108, 55–96; r) A.M. Beatty, CrystEngComm, 2001, 243–255. a) D. Philp and J.F. Stoddart, Angew. Chem., Int. Ed. Engl., 1996, 35, 1155–1196; b) B.J. Holliday and C.A. Mirkin, Angew. Chem. Int. Ed., 2001, 40, 2022–2043; c) S. Leininger, B. Olenyuk and P.J. Stang, Chem. Rev., 2000, 100, 853–907; d) M. Fujita, Struct. Bonding, 2000, 96, 177. a) R.W. Saalfrank, H. Maid, F. Hampel and K. Peters, Eur. J. Inorg. Chem, 1999, 1859–1867; b) R. Robson, Dalton Trans., 2000, 3735–3744; M. Eddaoudi, D.B. Moler, H. Li et al. Acc. Chem. Res., 2001, 34, 319–330; c) O.R. Evans and W. Lin, Acc. Chem. Res., 2002, 35, 511–522; d) B. Chen, M. Eddaoudi, S.T. Hyde et al. Science, 2001, 291, 1021–1023; e) K. Biradha and M. Fujita, Angew. Chem. Int. Ed., 2002, 41, 3392–3395; f) G.R. Desiraju, Nature, 2001, 412, 397–400. E.R.T. Tiekink and J.J. Vittal, Frontiers in Crystal Engineering, John Wiley & Sons Ltd., (2005). D. Braga, F. Grepioni and G.R. Desiraju, Chem. Rev., 1998, 98, 1375–1405. a) Dalton Trans., 2000, 3705–3998, Proceedings of the Dalton Discussion on Inorganic Crystal Engineering; b) D. Braga, G.R. Desiraju, J.S. Miller et al. CrystEngComm, 2002, 4, 500–509; c) M.D. Hollingsworth, Science, 2002, 295, 2410–2413. a) M. Oh, G.B. Carpenter and D.A. Sweigart, Organometallics, 2002, 21, 1290–1295; b) M. Oh, G.B. Carpenter and D.A. Sweigart, Angew. Chem. Int. Ed., 2001, 40, 3191–3194; c) M. Oh, G.B. Carpenter and D.A. Sweigart, Angew. Chem. Int. Ed., 2002, 41, 3650–3653; d) M. Oh, G.B. Carpenter and D.A. Sweigart, Chem. Commun., 2002, 2168–2169; e) M. Oh, G.B. Carpenter and D.A. Sweigart, Organometallics, 2003, 22, 2364–2366; f) M. Oh, G.B. Carpenter and D.A. Sweigart, Angew. Chem. Int. Ed., 2003, 42, 2026–2028; g) M. Oh, G.B. Carpenter and D.A. Sweigart, Acc. Chem. Res., 2004, 37, 1–11; h) M. Oh, J.A. Reingold, G.B. Carpenter and D.A. Sweigart, Coord. Chem. Rev., 2004, 248, 561–569. a) W.C. McCrone, Polymorphism, in Physics and Chemistry of the Organic Solid State, D. Fox, M.M. Labes and A. Weissemberg (Eds), Interscience, New York, USA; b) (1965); b) J. Bernstein, Polymorphism in Molecular Crystals, Oxford University Press, UK (2002). D. Braga, Chem. Rev., 1992, 92, 633–665. a) D. Braga and F. Grepioni, in Topics in Organometallic Chemistry, J.M. Brownand, P. Hofmann (Eds), Springer Verlag, Berlin, Germany, (1999); b) D. Braga and F. Grepioni, Chem. Soc. Rev., 2000, 29, 229–238. a) J. Bernstein, R.J. Davey and J-O. Henck, Angew. Chem. Int. Ed., 1999, 38, 3441–3461; b) S.R. Byrn, Solid-State Chemistry of Drugs, Academic Press, (1982); c) N. Blagden and R. Davey, Chem. Br., 1999, 35, 44–47. P. Seiler and J.D. Dunitz, Acta Cryst., 1979, B35, 2020–2032. a) F. Takusagawa and T.F. Koetzle, Acta Cryst., 1979, B35, 1074–1081; b) P. Seiler and J.D. Dunitz, Acta Cryst., 1979, B35, 1068–1074. a) J.D. Dunitz, Acta Cryst., 1995, B51, 619–631; b) D. Braga and F. Grepioni, Organometallics, 1992, 11, 711–718; c) J.D. Dunitz, Forty years of Ferrocene, in Organic Chemistry: Its Language and its State of Art , D. Kisakurek (Ed.), Verlag Helvetica Chimica Acta, Basel, Switzerland, (1993). M.F. Daniel, A.J. Leadbetter, R.E. Meads and W.G. Parker, Faraday Trans. 2 , 1978, 74, 456–465. K. Sato, M. Katada, H. Sano and M. Konno, Bull. Chem. Soc. Jpn., 1984, 57, 2361–2365. a) B.W. Fitzsimmons and W.G. Marshall, Dalton Trans., 1992, 73–76; b) B.W. Fitzsimmons and I. Sayer, Dalton Trans., 1991, 2907–2909.

496

Ferrocenes: Ligands, Materials and Biomolecules

31. a) D. Braga, L. Scaccianoce, F. Grepioni and S.M. Draper, Organometallics, 1996, 15, 4675–4677; b) F. Grepioni, G. Cojazzi, S.M. Draper et al. Organometallics, 1998, 17, 296–307. 32. a) R. Martinez and A. Tiripicchio, Acta Cryst., 1990, C46, 202–205; b) R.J. Webb, M.D. Lowery, Y. Shiomi et al. Inorg. Chem., 1992, 31, 5211–5219; c) M. Sorai and Y. Shiomi, Thermochim. Acta, 1986, 109, 29–44. 33. D. Braga, D. Paolucci, G. Cojazzi and F. Grepioni, Chem. Commun., 2001, 803–804. 34. L. Vegard and H. Dale, Z. Kristallogr., 1928, 67, 148–161. 35. F. Grepioni, G. Cojazzi, D. Braga et al. Dalton Trans., 1999, 553–558. 36. B.W. Fitzsimmons and A.R. Hume, Dalton Trans., 1980, 180–185. 37. D. Braga, G. Cojazzi, D. Paolucci and F. Grepioni, CrystEngComm, 2001, 3, 159–161. 38. The preparation of the salt [Ru(η6 -C6 H6 )2 ][BF4 ]2 from the solvate form [Ru(η6 C6 H6 )2 ][BF4 ]2 žMeNO2 is, instead, an example of how controlled desolvation can be exploited to prepare an elusive unsolvate crystal: D. Braga, G. Cojazzi, A. Abati et al. Dalton Trans., 2000, 3969–3975. 39. D. Braga, M. Polito, M. Bracaccini et al. Organometallics, 2003, 22, 2142–2150. 40. G.J. Palenik, Inorg. Chem., 1969, 8, 2744–2749. 41. F. Takusagawa and T.F. Koetzle, Acta Cryst., 1979, B35, 2888–2896. 42. D. Braga, M. Polito, D. D’Addario and F. Grepioni, Cryst. Growth Des., 2004, 4, 1109–1112. 43. D. Braga, F. Grepioni, M. Polito et al. Organometallics, 2006, 25, 4627–4633. 44. D. Braga, F. Paganelli, E. Tagliavini et al. Organometallics, 1999, 18, 4191–4196. 45. D. Braga, M. Eckert, M. Fraccastoro et al. New J. Chem., 2002, 26, 1280–1286. 46. D. Braga, L. Maini, L. Prodi et al. Chem. Eur. J., 2000, 6, 1310–1317. 47. D. Braga, A. Angeloni, L. Maini et al. New J. Chem., 1999, 23, 17–24. 48. a) D. Braga, A. Angeloni, F. Grepioni and E. Tagliavini, Chem. Commun., 1997, 1447–1448; b) D. Braga, A. Angeloni, F. Grepioni and E. Tagliavini, Organometallics, 1997, 16, 5478–5485. 49. D. Braga, A. Angeloni, E. Tagliavini and F. Grepioni, Dalton Trans., 1998, 1961–1968. 50. F. Grepioni, S. Gladiali, L. Scaccianoce et al. New J. Chem., 2001, 25, 690–695. 51. J.S. Miller and A.J. Epstein, Chem. Commun., 1998, 1319–1325. 52. a) J.S. Miller, A.J. Epstein and W.M. Reiff, Chem. Rev., 1988, 88, 201–220; b) E. Coronado, J.R. Galan-Mascaros, C.J. Gomez-Garcia and V. Laukhin, Nature, 2000, 408, 447–449; c) E. Coronado, J-R. Galan-Mascaros, C-J. Gomez-Garcia et al. Chem. Eur. J., 2000, 6, 552–563; d) M.B. Zaman, M. Tomura, Y. Yamashita et al. CrystEngComm, 1999, 1, 36–38. 53. a) D. Braga, L. Maini and F. Grepioni, Angew. Chem. Int. Ed., 1998, 37, 2240–2242; b) D. Braga, L. Maini and F. Grepioni, J. Organomet. Chem., 2000, 593–594, 101–108; c) D. Braga, L. Maini, F. Grepioni et al. New J. Chem., 2000, 24, 547–553. 54. C.M. Zakaria, G. Ferguson, A.J. Lough and C. Glidewell, Acta Cryst., 2002, B58, 786–802. 55. M. Zakaria Choudhury, G. Ferguson, A.J Lough and C. Glidewell, Acta Cryst., 2002, 58, m1–4. 56. D.M. Shin, I.S. Lee and Y.K. Chung, Eur. J. Inorg. Chem., 2003, 2311–2317. 57. I.S. Lee, D.M. Shin and Y.K. Chung, Cryst. Growth Des., 2003, 3, 521–529. 58. I.S. Lee, Y.K. Chung, J. Mun and C.S. Yoon, Organometallics, 1999, 18, 5080–5085. 59. T-a. Okamura, K. Sakauye, N. Ueyama and A. Nakamura, Inorg. Chem., 1998, 37, 6731–6736. 60. a) T. Saji and I. Kinoshita, Chem. Commun., 1986, 716–717; b) J.C. Medina, I. Gay, Z. Chen et al. J. Am. Chem. Soc., 1991, 113, 365–366; c) J.C. Medina, T.T. Goodnow, M.T. Rojas et al. J. Am. Chem. Soc., 1992, 114, 10583–10595; d) P.D. Beer, Chem. Commun., 1996, 689–696; e) P.D. Beer, A.R. Graydon, A.O.M. Johnson and D.K. Smith, Inorg.

Crystal Engineering with Ferrocene Compounds

61.

62.

63. 64.

65. 66. 67. 68. 69. 70. 71. 72. 73.

74.

75. 76. 77. 78. 79. 80. 81. 82. 83.

497

Chem., 1997, 36, 2112–2118; f) R.S. Herrick, R.M. Jarret, T.P. Curran et al. Tetrahedron Lett., 1996, 37, 5289–5292; g) M. Kira, T. Matsubara, H. Shinohara and M. Sisido, Chem. Lett., 1997, 89–90; h) H-B. Kraatz, J. Lusztyk and G.D. Enright, Inorg. Chem., 1997, 36, 2400–2405. a) G. Cerichelli, B. Floris and G. Ortaggi, J. Organomet. Chem., 1974, 76, 73–79; b) A. Togni, T. Hayashi (Eds), Ferrocenes: homogeneous catalysis, organic synthesis, materials science, VCH, Weinheim, Germany(1995). a) A. Nomoto, T. Moriuchi, S. Yamazaki et al. Chem. Commun., 1998, 1963–1964; b) T. Moriuchi, A. Nomoto, K. Yoshida and T. Hirao, J. Organomet. Chem., 1999, 589, 50–58; c) T. Moriuchi, A. Nomoto, K. Yoshida et al. J. Am. Chem. Soc., 2001, 123, 68–75; d) T. Moriuchi, K. Yoshida and T. Hirao, J. Organomet. Chem., 2003, 668, 31–34. C. Bresner, S. Aldridge, I.A. Fallis and L.L. Ooi, Acta Cryst., 2004, E60, m441–m443. a) M. Zakaria Choudhury, G. Ferguson, A.J. Lough and C. Glidewell, Acta Cryst., 2002, 58, m5–7; b) C.M. Zakaria, G. Ferguson, A.J. Lough and C. Glidewell, Acta Cryst , 2001, C57, 687–689; c) C.M. Zakaria, G. Ferguson, A.J. Lough and C. Glidewell, Acta Cryst , 2001, C57, 914–917. a) P. Stepnicka and I. Cisarova, New J. Chem., 2002, 26, 1389–1396; b) P. Stepnicka, I. Cisarova and J. Ludvik, J. Organomet. Chem., 2004, 689, 631–638. J.F. Gallagher, P.N. Kelly, P.T.M. Kenny and A.J. Lough, Acta Cryst , 2003, C59, m552–m554. D. Braga, L. Maini, F. Grepioni et al. Organometallics, 2001, 20, 1875–1881. D. Braga, L. Maini, M. Polito and F. Grepioni, Organometallics, 1999, 18, 2577–2579. D. Braga, L. Maini, M. Polito et al. Chem. Eur. J., 2000, 6, 4227–4235. a) D. Braga, L. Maini, M. Polito and F. Grepioni, Chem. Commun, 2002, 2302–2303; b) D. Braga, L. Maini, S.L. Giaffreda et al. Chem. Eur. J., 2004, 10, 3261–3269. D. Braga, M. Polito and F. Grepioni, Cryst. Growth Des., 2004, 4, 769–774. P. Anastas and J. Warner, Green Chemistry: Theory and Practice, Oxford University Press, UK 1998. a) C. Elschenbroich, O. Schiemann, O. Burghaus and K. Harms, J. Am. Chem. Soc., 1997, 119, 7452–7457; b) C. Elschenbroich, M. Wolf, J. Pebler and K. Harms, Organometallics, 2004, 23, 454–459. a) K. Tanaka and F. Toda, Chem. Rev., 2000, 100, 1025–1074; b) K. Tanaka and F. Toda, Solvent-Free Organic Synthesis, Wiley-VCH Verlag GmbH, Weinheim, Germany, (2003); c) D. Bradley, Chem. Br., 2002, 38, 42–45; d)G.W.V. Cave, C.L. Raston and J.L. Scott, Chem. Commun, 2001, 2159–2169; e) G. Rothenberg, A.P. Downie, C.L. Raston and J.L. Scott, J. Am. Chem. Soc., 2001, 123, 8701–8708; f) F. Toda, CrystEngComm, 2002, 4, 215–222; g) L.R. Nassimbeni, Acc. Chem. Res., 2003, 36, 631–637. a) D. Braga, L. Maini, M. Polito et al. Chem. Commun., 2002, 2960–2961; b) D. Braga, L. Maini, M. Polito et al. Chem. Eur. J., 2003, 9, 4362–4370. D. Braga, D. D’Addario, M. Polito and F. Grepioni, Organometallics, 2004, 23, 2810–2812. D. Braga, M. Polito, D. D’Addario et al. Organometallics, 2003, 22, 4532–4538. D. Braga, S.L. Giaffreda and F. Grepioni, Chem. Commun., 2006, 3877–3879. D. Braga, G. Cojazzi, D. Emiliani et al. Chem. Commun., 2001, 2272–2273. D. Braga, G. Cojazzi, D. Emiliani et al. Organometallics, 2002, 21, 1315–1318. D. Braga, L. Maini, M. Mazzotti et al. CrystEngComm, 2003, 5, 154–158. D. Braga, L. Maini, M. Mazzotti et al. Chem. Commun., 2002, 2296–2297. a) G. Bandoli and A. Dolmella, Coord. Chem. Rev., 2000, 209, 161–196; b) M.C. Gimeno, P.G. Jones, A. Laguna and C. Sarroca, Dalton Trans., 1995, 1473–1481; c) S.P. Neo and T.S.A. Hor, J. Organomet. Chem., 1994, 464, 113–119; d) T.S.A. Hor, S.P. Neo, C.S. Tan et al. Inorg. Chem., 1992, 31, 4510–4516.

498

Ferrocenes: Ligands, Materials and Biomolecules

84. X.L. Lu, W.K. Leong, T.S.A. Hor and L.Y. Goh, J. Organomet. Chem., 2004, 689, 1746–1756. 85. G. Dong, M. Hong, C-Y. Duan et al. Dalton Trans., 2002, 13, 2593–2594. 86. G. Dong, Y-t. Li, C-y. Duan et al. Inorg. Chem., 2003, 42, 2519–2530. 87. R. Horikoshi, T. Mochida and H. Moriyama, Inorg. Chem., 2002, 41, 3017–3024. 88. a) R. Horikoshi, T. Mochida and H. Moriyama, Inorg. Chem., 2001, 40, 2430–2433; b) G. Mago, M. Hinago, H. Miyasaka et al. Inorg. Chim. Acta, 1997, 254, 145–150; c) YB. Dong, M.D. Smith, R.C. Layland and H-C. zur Loye, Inorg. Chem., 1999, 38, 5027–5033. 89. M. Kondo, Y. Hayakawa, M. Miyazawa et al. Inorg. Chem., 2004, 43, 5801–5803.

13 The Bioorganometallic Chemistry of Ferrocene∗ Nils Metzler-Nolte and Mich`ele Salmain

13.1

Introduction

The discovery of ferrocene1, 2 and the elucidation of its remarkable structure3, 4 is often considered as the starting point for modern organometallic chemistry. In the time since, ferrocene chemistry has had a major impact on a variety of ﬁelds including catalysis, materials chemistry and electrochemical sensors, as described in previous chapters of this book. The excellent stability of the ferrocene framework in aqueous, aerobic media, the accessibility of a large variety of derivatives and its favourable electrochemical properties have also made ferrocene and its derivatives very popular molecules for biological applications and for conjugation with biomolecules. In recent years, bioorganometallic chemistry has developed as a rapidly growing and maturing area which links classical organometallic chemistry to biology, medicine and molecular biotechnology.5–7 This chapter is focused on bioconjugates of ferrocene with amino acids and peptides (Section 13.2), enzymes and proteins (Section 13.3), DNA, RNA and PNA (Section 13.4), carbohydrates (Section 13.5), and hormones and others (Section 13.6). In Section 13.7, the physiology and medicinal applications of ferrocene are described. This chapter has been limited to conjugates in which ferrocene is covalently bonded to the biomolecule, and their applications. Systems where free ferrocene serves only as an electron mediator between enzymes and an electrode are not considered. ∗ Speciﬁc abbreviations used throughout this chapter are given at the end of the chapter before the Reference List.

Ferrocenes: Ligands, Materials and Biomolecules  2008 John Wiley & Sons, Ltd

Edited by Petr Stepnicka

500

Ferrocenes: Ligands, Materials and Biomolecules

This chapter covers the literature from 1957, when Schl¨ogl reported the ﬁrst ferrocenyl amino acids,8 through to 2006. It is based largely on our earlier review article of 2004,9 massively extended by the new literature which has been published meanwhile. The medicinal chemistry of ferrocene (section 13.7) has not been reviewed before. However, the bioorganometallic chemistry of ferrocene, as it presents itself in all its facettes today, is already too extensive to be covered in a single review. Therefore, this chapter does not try to be comprehensive. We have concentrated on what we consider major developments and deliberately omitted material that seemed peripheral to us. Naturally, this is a personal choice and implies no judgement on the scientiﬁc quality of the work. The reader is referred to our review article for a more comprehensive treatment and further references on some sub-sections.9 In addition, there is available a recent monograph entitled Bioorganometallics.10 Further information on the subject is also available from an introductory book chapter.11 More specialised reviews cover related aspects of ferrocene chemistry, for instance on applications of the ferrocene compounds in glucose biosensors12 bioelectronics13, 14 and on ferrocenecontaining nucleic acids.15 Finally, there are a number of interesting related articles in Comprehensive Organometallic Chemistry III in Volumes 1 and 12.16, 17

13.2 13.2.1

Conjugates of Ferrocene with Amino Acids and Peptides Ferrocene-Based Amino Acids: Ferrocenylalanine, 1 -Aminoferrocene-1-carboxylic Acid and Others

The ﬁrst two ferrocene amino acid analogues were reported by Schl¨ogl as early as 1957, only six years after the initial reports on the synthesis of ferrocene.1, 2 They are: DL-ferrocenylalanine (DL-Fer, 1);8, 18, 19 and DL-ferrocenylphenylalanine 28 (Scheme 13.1). While 2 has never been used after its initial report, ferrocenylalanine 1 has been subject of extensive research. Some time after the original report, several other papers reported on improved and stereoselective syntheses of this amino acid.20–23 Two strategies have been adapted to obtain 1 in enantiomerically pure form: either via resolution of the racemic mixture or via stereospeciﬁc synthesis. Resolving the racemic mixture can be achieved by fractional crystallisation of the diastereomeric brucine salt;22 or kinetic resolution by stereoselective de-acylation of the acylated racemate using the enzyme acylase.23 Ferrocenylalanine can be stereospeciﬁcally synthesised with up to 94 % ee via asymmetric hydrogenation of the corresponding Z-conﬁgured dehydro acetamido derivative 3 (Scheme 13.2), to give N -acetyl-protected L-1 (4) followed by deprotection.21 Another elegant method for obtaining enantiomerically pure H2N

H2N CO2H

CO2H

Fe

Fe

1

2

Scheme 13.1

The Bioorganometallic Chemistry of Ferrocene

O

501

O

HN

HN CO2H

CO2H [{RhCl(cod)}2] / H2

Fe

Fe

Asymmetric hydrogenation

chiral phosphine 3

4, Ac-Fer BocHN CO2H

FcI

+

BocHN IZn

6

[Pd2(dba)3] CO2H

7

Fe

Pd-catalysed cross-coupling

(o-Tol)3P 5, Boc-Fer

Scheme 13.2

Boc-protected ferrocenylalanine (5) consists of a palladium(0)-catalysed cross-coupling between FcI (6) and the serine-derived organozinc reagent 7 (Scheme 13.2).20 Ferrocenylalanine alone has been subjected to biological tests. In a bacterial growth assay with the phenylalanine-requiring bacterium Leuconostoc mesenteroides DL-Fer did not support growth of L. mesenteroides in the absence of Phe, nor was it able to inhibit the growth of this bacterium in the presence of Phe.24, 25 Furthermore, DLFer was tested for toxicity against Chinese Hamster Ovarian (CHO) cells.25 Only at relatively high concentrations of 0.45 mM or higher, DL-Fer was toxic against CHO cells. DL-Fer was also tested for its function to serve as a substrate or inhibitor of phenylalanine hydroxylase and aromatic L-amino acid decarboxylase. For phenylalanine hydroxylase, DL-Fer was a noncompetitive inhibitor (Ki = 0.89 mM) with respect to L-Phe and a mixed inhibitor with respect to the cofactor (DMPH4 ).24, 25 DL-Fer was found to be a competitive inhibitor (Ki = 7.2 mM) of aromatic L-amino acid decarboxylase with respect to Phe. A variety of peptides that contain 1 as an unnatural amino acid has been synthesised. A number of derivatives with peptides that do not have a particular biological function were prepared: cyclo(D-Fer-L-prolyl), cyclo(L-Fer-L-prolyl)22 and polypeptides containing repeating L-Fer-[Glu(OBzl)]4 or L-Fer2 -[Glu(OBzl)]4 units.23 The second, much larger class comprises modiﬁed biogenic peptides in which 1 has been speciﬁcally substituted for a phenylalanine (Phe) residue. Compared to Phe, the cylindrically shaped 1 is more bulky and lipophilic. Owing to these differences, results from binding assays for the modiﬁed peptide in comparison to those for the natural peptide could reveal important information about substrate-receptor interactions. A summary of all Fer-derivatives of naturally occurring peptides is shown in Table 13.1. The ﬁrst biogenic peptide that was modiﬁed with 1 is the pentapeptide [Leu5 ]Enkephalin, which has the primary sequence H-Tyr-Gly-Gly-Phe-Leu-OH (8 in Scheme 13.3). [Leu5 ]-Enkephalin was ﬁrst isolated as a mixture with its position 5 methionine analogue ([Met5 ]-Enkephalin) by Hughes et al. from pig brain in 1975.39 The

502

Ferrocenes: Ligands, Materials and Biomolecules Table 13.1 Ferrocene derivatives of peptides with a biological function

Entry

Peptide name/group (Primary sequence)

Abbreviation

Amino acid sequence

Ref.

1

Enkephalin, Enk (Tyr-Gly-Gly-PheLeu; [Leu5 ]Enkephalin)

[DL-Fer4 ]-Enk

H-Tyr-Gly-Gly-Fer-LeuOHb

26

[D-Fer4 ]-Enk

H-Tyr-Gly-Gly-Fer-LeuOHc

27

H-Tyr-Gly-Gly-Fer-LeuOHd

28

H-Tyr-Gly-Gly-Fer-LeuOHc

27

H-Tyr-Gly-Gly-Fer-LeuOHd

28

2a 2b

[L-Fer4 ]-Enk

3a 3b 4

FcCO-Enk-OH

FcCO-Tyr-Gly-Gly-PheLeu-OH and FcCO-Tyr-Gly-GlyPhe-Leu-NH2 i

29

5

[p-CC-Fc-Phe4 ]-Enk

H-Tyr-Gly-Gly-Phe(pCC-Fc)-Leu-OH

30

6

[p-CC-CR2 NHCOFc-Phe4 ]Enk

H-Tyr-Gly-Gly-Phe(pCC-CR2 -NHCOFc)Leu-OH

30

[DL-Fer7 ]-SP

H-Arg-Pro-Lys-Pro-GlnGln-Fer-Phe-Gly-LeuMet-NH2 b, e

31

8

[DL-Fer8 ]-SP

H-Arg-Pro-Lys-Pro-GlnGln-Phe-Fer-Gly-LeuMet-NH2 b, e

31

9

[DL-Fer7 ,Ile8 ]-SP

H-Arg-Pro-Lys-Pro-GlnGln-Fer-Ile-Gly-LeuMet-NH2 b, e

31

10

FcCO-SP

FcCO-Arg-Pro-Lys-ProGln-Gln-Phe-PheGly-Leu-Met-NH2

32

[DL-Fer5 ]-BK

H-Arg-Pro-Pro-Gly-FerSer-Pro-Phe-ArgOHb, f

31

[DL-Fer9 ]-BK

H-Arg-Pro-Pro-Gly-PheSer-Pro-Fer-ArgOHb, f

31

7

11

12

Substance P, SP (Arg-Pro-Lys-ProGln-Gln-Phe-PheGly-Leu-Met)

Bradikinin, BK (H-Arg-Pro-Pro-GlyPhe-Ser-Pro-Phe-ArgOH)

The Bioorganometallic Chemistry of Ferrocene

503

Table 13.1 (continued ) Entry

Peptide name/group (Primary sequence)

Abbreviation

Amino acid sequence

Ref.

13

Angiotensin II (Asp-Arg-Val-TyrIle-His-Pro-Phe)

AT II

14

Saralasin (Sar-Arg-Val-Tyr-IleHis-Pro-Val)a

[Sar1 , Fer8 ]-AT II

H-Sar-Arg-Val-Tyr-IleHis-Pro-Fer-OHb, c, g

33, 34

15

Nuclear Localisation Sequence, NLS (Pro-Lys-Lys-LysArg-Lys-Val)

FcCO-NLS

FcCO-Lys-Pro-Lys-LysLys-Arg-Lys-ValNH2 h, i

35, 36

16

HIV-TAT (Gly-Arg-Lys-Lys-ArgArg-Gln-Arg-Arg-Arg)

FcCO-TAT

FcCO-Lys-Gly-Arg-LysLys-Arg-Arg-Gln-ArgArg-Arg-OHh, i

36

H-Lys-Gly-Fer-Gln-GlyOHk

31

FcCO-Arg-Trp-Arg-TrpArg-NH2 i, l

37

FcCO-Trp-Arg-Trp-ArgTrp-NH2 i, l, m

37, 38

17 18

Artiﬁcial antimicrobial peptides

19

a Sar: Sarcosine = N -Methylglycine. b Diastereomeric mixture. c Separation by preparative HPLC. d Prepared from enantiomerically pure D-Fer. e Could not be separated by preparative HPLC. f could be separated by preparative HPLC but absolute conﬁguration could not be assigned. g The diastereomeric mixtures as well as separated isomers were applied to biological testing. h Also ﬂuorescein thiourea derivative on Lys1 . i Also cobaltocenium derivative prepared. k Used for chymotrypsin digestion studies; also cymantrenyl-alanine derivative. l See Section 13.7.6. m Several shorter peptide derivatives were also prepared but found to be less active.

H Tyr Gly Gly Phe Leu OH

8

H

H N

O N H

H N O OH

Fc

O N H

H N

O OH

O 9

Scheme 13.3 Amino acid sequence of [Leu5 ]-Enk and structure of [Leu5 ,Fer4 ]-Enk

enkephalins are neuropeptides with an action similar to morphine, exhibiting high afﬁnity for the opioid receptor. These pentapeptides were detected in various human tissues, such as the human brain, the human spinal ﬂuid and in blood plasma.40 Three reports on the synthesis of [Fer4 , Leu5 ]-Enkephalin, in which the Phe residue in position 4 has been replaced by 1 (9, entries 1–3 in Table 13.1, see also Scheme 13.3), appeared almost simultaneously in the literature.26–28 The compounds were prepared on a solid support, by employing classical Merriﬁeld techniques. Diastereomerically

504

Ferrocenes: Ligands, Materials and Biomolecules

pure [D-Fer4 ] and [L-Fer4 ] enkephalins were obtained either by HPLC puriﬁcation of the racemic [DL-Fer4 , Leu5 ]-Enkephalin mixture27 or by performing the solid phase peptide synthesis with enantiomerically pure L-Fer and D-Fer.28 Either diastereomer displays a much reduced afﬁnity for the enkephalin receptor compared to [Leu5 ]enkephalin,26, 27 although both are signiﬁcantly more potent than many other position 4 analogues.41 Interestingly, the [D-Fer4 ]-diastereomer was signiﬁcantly more potent than the L-analogue.26, 27 Substance P (SP) is an undecapeptide with two consecutive Phe residues in positions 7 and 8 (entry 7, Table 13.1). The peptide was ﬁrst discovered by von Euler and Gaddum in 1931,42 but it took another four decades before its amino acid sequence was determined by Chang et al.43, 44 SP is involved in several important physiological processes, such as contraction of the smooth muscles, pain transmission and activation of the immune system.45 Tartar and coworkers synthesised the following three SPderivatives: [DL-Fer7 ]-SP (10), [DL-Fer8 ]-SP (11), and [DL-Fer7 , Ile8 ]-SP (12) (entries 7–9 in Table 13.1).31 As discussed above for enkephalin, one or both Phe residues were replaced by 1. Racemic 1 was used for the syntheses, resulting in diastereomeric mixtures, which could not be resolved by preparative HPLC. Results from binding assays show a decreased activity of the ferrocenylalanine substituted peptides compared to native SP in the following order: [DL-Fer8 ]-SP > [DL-Fer7 ]-SP > [DL-Fer7 , Ile8 ]-SP.31 The same group also prepared ferrocenylalanine Bradykinin analogues.31 The nonapeptide Bradykinin (BK, entry 11 in Table 13.1) is a tissue hormone involved in the blood clotting process. When the hormone is released from the precursor kininogen, it triggers a lowering of the blood pressure via dilatation of the blood vessels.46, 47 Either of the positions 5 or 9 Phe residues has been substituted by DL-Fer, resulting in conjugates 13 and 14 (entries 11 and 12 in Table 13.1). Although separation of the diastereomers was accomplished by HPLC, it was not possible to assign their absolute conﬁguration. Binding studies revealed a signiﬁcant decrease of activity upon substitution of Phe by Fer. The two diastereomers of [Fer5 ]-BK (13) and [Fer9 ]-BK (14) showed activities in the range 1.7–7 × 10−2 relative to native BK. Considering the low activities of the diastereomeric [Fer]-BK analogues, it is likely that the ferrocenyl moiety interacts poorly with the receptor. The family of Angiotensin II (AT II) related compounds has attracted the attention of Tartar and coworkers as well. Angiotensin II, an octapeptide having Phe in position 8, strongly increases blood pressure in mammals (entry 13 in Table 13.1).48 Several AT II derivatives have found medicinal applications, such as the agonist [Asn1 , Val5 ]AT II (Hypertension), which has been used to normalise blood pressure as quickly as possible after a shock or collapse. The antagonist [Sar1 , Val8 ]-AT II (Saralasin), on the other hand, has been used for the diagnosis of AT II-dependent forms of hypertonia. Fer-modiﬁed [Sar1 ]-Angiotensin II ([Sar1 , Fer8 ]-AT II, 15) has been prepared (entry 14 in Table 13.1).33, 34 The synthesis of 15 was performed with DL-1 and the diastereomers were separated by HPLC for physiological testing only. Either diastereomer showed approximately 1 % activity relative to [Sar1 ]-AT II in an assay with rabbit aorta strips.34 The diastereomeric mixture of 15 showed an activity of 6.1 % relative to [Sar1 ]-AT II in a binding assay to puriﬁed bovine adrenocortical membranes.33

The Bioorganometallic Chemistry of Ferrocene

505

Chymotrypsin hydrolysis investigations were performed on the peptides H-Lys-GlyPhe-Gln-Gly-OH (16) and H-Lys-Gly-Fer-Gln-Gly-OH (17, entry 17 in Table 13.1).31 Chymotrypsin belongs to the class of serine proteases and is present in the human pancreas. The high selectivity of the enzyme Chymotrypsin for the scission of peptidic bonds next to the aromatic amino acids Phe, Trp or Tyr can be rationalised on the basis of the X-ray crystal structure of bovine Chymotrypsin.49, 50 Upon the action of Chymotrypsin, the Phe-containing pentapeptide 16 was hydrolysed (Km = 3.85 mM), whereas the Fer-congener 17 was not hydrolysed at all. Interestingly, the analogue containing cymantrenylalanine (cymantrene = [CpMn(CO)3 ]) was also hydrolysed.31 1 -Aminoferrocene-1-carboxylic acid hydrochloride (Fca or 18·HCl in Scheme 13.4) constitutes one of the simplest ferrocene-based amino acids. The ﬁrst derivative of Fca was prepared by Janda and coworkers and used for the generation of catalytic antibodies (see also Section 13.2.3). Its synthesis was later reported independently by two groups. A ﬁrst preparation by lithiation of 1 -amino-1-bromoferrocene, followed by quenching with solid carbon dioxide did not yield pure 18.51 At about the same time, a synthesis which yielded 18 in rather low yield was published,52 along with a careful structural comparison of derivatives of 18 in the solid state and in solution. An improved synthesis was recently reported by Heinze and coworkers, and the compound can now be considered readily available. Scheme 13.4 summarises the synthesis of 18 according to Heinze.53 Rapic and coworkers were also able to obtain N -protected derivatives of 18 for solid phase synthesis starting from fc(CO2 H)2 (19).54 These peptides show interesting intramolecular hydrogen bonds. Their synthesis, properties and applications are further discussed in Section 13.2.3. O

6

N

Cu-phthalimide Fe

O

82 %

(2,5-C6H3Cl2)COCl

N2H4

KOtBu, H2O

O

74 %

Fe

99 %

NHAc Fe

NH2

Cl

NHAc Fe

95 %

NHAc Fe

HCl

O

70 %

NH2 • HCl Fe

O

98 % OH

Cl

Ac2O

18 • HCl (Fca • HCl)

OH

Scheme 13.4 Optimised synthesis of Fca 18 according to Heinze and coworkers

Several other ferrocenylalanine analogues have also been reported. Most importantly, these include 1,1 -ferrocenyl-bis-alanine (20 in Scheme 13.5), which will be discussed in more detail below. The related compound 1,2-ferrocenyl-bis-alanine (21) was readily obtained as the bis(N -formyl) derivative but decomposed upon deprotection (Scheme 13.5).55 The β-amino acids ferrocenyl-β-alanine (22) and 1,1 -ferrocenyl-bisβ-alanine (23) could be synthesised in an enantiomerically pure form (Scheme 13.5).56 No attempts were made to hydrolyse the methyl esters to free amino acids. A few more

506

Ferrocenes: Ligands, Materials and Biomolecules CO2R

H2N CO2H Fe

CO2R NHCHO

Fe CO2H H 2N

NH2

NHCHO Fe

21 (R = Et, H)

CO2Me

22

20 NH2 Fe

23

CO2Me NH2 CO2Me

(CH2)m CO2R Fe (CH2)n NHR′ 24 (m = 0, n = 3) 25 (m = 3, n = 0)

Scheme 13.5

ferrocene derivatives have been synthesised which formally have an amino acid type structure, but their chemistry was not explored further.57–59 Other organometallic amino acids 24a and 25a (R = R = H in both cases) were reported by Rapic’s group.60 Both compounds were prepared in several steps and thoroughly characterised. The X-ray single crystal structures of 24b (R = Me, R = Ac) and 25d (R = Me, R = Boc) were reported, along with the preparation of several other derivatives with varying protecting groups. These compounds may prove to be more ﬂexible building blocks for peptides with a ferrocene moiety incorporated into the peptide chain than for instance 18 (see Section 13.2.3). The difunctionalised derivative 20 (Scheme 13.5) has been used in the synthesis of peptide mimetics.20, 55, 61–65 The methods for obtaining 20 in optically pure form are identical to those for 1: Pd(0)-catalysed cross-coupling of fcI2 (26) with the serine derived organozinc reagent 7 shown in Scheme 13.2;20 and asymmetric hydrogenation of the corresponding bis-didehydro amino acid derivatives.61–64 With the latter method, an enantiomeric excess higher than 99 % could be achieved.63, 64 Via a sophisticated synthetic route, Frejd and coworkers synthesised an enantiomerically pure 1,1 -ferrocenyl-bis-alanine derivative 27 that contains orthogonal protecting groups (Scheme 13.6).62–64, 66 Optically pure 27 was deprotected by hydrogenation into the MeO2C

MeO2C H2, Pd/C (5 %)

Fe

PyAOP

Fe

MeOH

27

NH Fe CO

DIPEA, DMF CO2H

CO2Bn BocHN

CO2Me

NH2

NHCBz

BocHN 28

Scheme 13.6

29

NHBoc

(+ Dimer and Trimer)

The Bioorganometallic Chemistry of Ferrocene

507

peptide mimetic (S,S)-28, which has a free amino group and a free carboxylic acid moiety. Reaction of 28 with the peptide coupling reagent PyAOP yielded lactam 29, along with two macrocyclic peptides (Scheme 13.6).66 Frejd and coworkers then used lactam 29 as a substitute for the H-Phe-Phe-OH unit which was incorporated into a structural mimic of substance P (30, Scheme 13.7).63 However, CD-spectroscopic studies indicated that the constraints imposed by the ferrocenophane moiety prohibited the peptide from adopting the characteristic α-helical secondary structure found for native SP. No biological activity data for this SP-mimic are available.63 O NH O H-Arg-Pro-Lys-Pro-Gln-Gln N H

Gly-Leu-Met-NH2

30

Fe

Gly(2)

O Gly(3)

HN 31

N O H

H2N

O HN

Fc(1) for Tyr

Fc(4) for Phe O

HN

Leu(5) NH2

Fe O

Scheme 13.7

In another interesting paper, Frejd and coworkers prepared a [Leu5 ]-Enkephalin mimetic in which the H-Tyr-Gly-Gly-Phe-subunit has been replaced by a Gly-Glylooped 1,1 -ferrocenyl-bis-alanine residue (31, Scheme 13.7).64 The ferrocene rings of this constrained compound constitute a substitute for the aromatic phenol (Tyr) and phenyl (Phe) rings. NMR spectroscopic studies show that 31 is a model for native [Leu5 ]-enkephalin in the single-bend conformation, which is stabilised by a β-turn.67, 68 Unfortunately, no biological activity data for this peptide mimetic is available either. 13.2.2

Amino Acid and Peptide Conjugates Featuring Mono-Substituted Ferrocene Unit

Imines from ferrocenecarbaldehyde (FcCHO, 32) and amino acids or amino acid esters can be readily prepared (Scheme 13.8). Various solvents have been used for this synthetic transformation, such as EtOH,69 MeOH70 or CHCl3 .71 The reactions can be performed around room temperature70 or at slightly elevated temperatures around 60 ◦ C.70, 71 One paper reports the synthesis of 33 even under solvent-free conditions at room temperature.72 An overview of all reported imines between 32 and amino acids or amino acid esters is given in Table 13.2.

508

Ferrocenes: Ligands, Materials and Biomolecules R1 FcCHO

NaBH3CN

OR2

H2N O

32

R2 = OH or OMe −H2O

R1

R1 N Fe

OR2 O

NaBH4 or

Fe

N H

H2, Pd/C (5 %)

OR2 O

37, R2 = OH or OMe

33

Scheme 13.8 Table 13.2 Constitution of imines 33 from 32 and amino acid derivatives Compound

Ref.

FcCH=N(Gly-OH) FcCH=N(DL-Ala-OH) FcCH=N(Ala-OMe)

69, 70

FcCH=N(Ala-OEt) FcCH=N(DL-Val-OH)

72

FcCH=N(DL-Val-OEt) FcCH=N(Leu-OH)

69

FcCH=N(Leu-OEt) FcCH=N(Ile-OMe)

69

FcCH=N(Glu-OH) FcCH=N(DL-Tyr-OH) FcCH=N(Met-OH) FcCH=N(His-OMe) FcCH=N(Lys(Nε =C-Fc)-OMe)

69 72

Compound FcCH=N(Gly-OEt) FcCH=N(Ala-OH) FcCH=N(DL-Ala-OEt)

Ref. 69, 70 70 69

FcCH=N(β-Ala-OH) FcCH=N(Val-OH)

69

FcCH=N(DL-Leu-OH) FcCH=N(Leu-OMe)

69

FcCH=N(DL-Ile-OH) FcCH=N(DL-Asp-OH)

69

FcCH=N(Phe-OMe) FcCH=N(Tyr-OH)

71 69

72

FcCH=N(Arg-OH) FcCH=N(Ser-OMe)

72

[FcCH=N(Cys(S-)-OH)]2

69

69 70 72 69 69 69, 71

70 71 69 69 72

These imines are quite stable compounds in the absence of aqueous acids. They can be reduced with NaBH4 or H2 /Pd to yield the more stable secondary amine derivatives as described below. Amino acid ferrocenylimines have been used as ligands for transition metal ions. Beck and coworkers prepared palladium(II) complexes with imines from 32 and the amino acids Ala, Gly, Val, and Leu as ligands.70 The composition of the isolated

The Bioorganometallic Chemistry of Ferrocene R

509

O

R

Fc

O

N

[PdCl4]2−

N

Fc

ONa 33

O Pd

R = H, CH3, CH(CH3)2 or CH2CH(CH3)2

O

N

O

R

Fc

34

N R

Fc N

O

CHRCOOR′

Pd O

Pd(OAc)2 Fe

O

O

OR′

O

33 35

R′OOCRHC

Fe

R = H, R′ = Et R = Me, R′ = Me

Pd N

Scheme 13.9

complexes depends on the palladium(II) salt and the type of amino acid derivative, as illustrated in Scheme 13.9. Reaction of Na2 [PdCl4 ] with imines from 32 and the free carboxylic acid of Ala, Leu and Val gives mononuclear complexes with a ligand/palladium ratio of 2/1 (34, Scheme 13.9) while the reaction between palladium(II) chloride and FcCH=NCH2 CO2 H (from Gly) yields a dinuclear chloro-bridged palladium(II) complex. The use of Pd(OAc)2 leads to an ortho-metallated acetato-bridged species 35 (Scheme 13.9). Several of these complexes, including an ortho-palladated derivative, have been structurally characterised by X-ray crystallography. In a later paper, Beck and coworkers prepared imines from tri, tetra and hexasubstituted benzenes with p-ethynyl phenylalanine methyl ester via Sonogashira coupling.73 The molecular structure of the hexaferrocene derivative 36 is depicted in Scheme 13.10. Ferrocenylmethyl (Fem) derivatives 37 have been prepared via reduction of the imines 33 from 32 and amino acids with NaBH4 or H2 /Pd. Alternatively, these amines can be synthesised directly in a one-pot-procedure from 32, the amino acid derivative and the appropriate reducing reagent (Scheme 13.8). Up to now, a considerable number of Fem derivatives have been reported (Table 13.3). Some of these compounds, in particular the glycine derivative Fem-Gly-OH (38a), found application in peptide synthesis as lipophilic amino acid residues, with masked amide moieties. In a stoichiometric reaction between glycine methyl ester and 32, small amounts of the bimetallic compound (Fem)2 Gly-OMe (39) were isolated in addition to the expected mono-Fem derivative 38b.74 Beck and coworkers also used Fem-Pro-OH (40) and Fem-Ala-OH (41) as ligands for palladium(II).70 The reactions with Na2 [PdCl4 ] gave mono and bimetallic complexes with structures relating to the ones shown in Scheme 13.9.

510

Ferrocenes: Ligands, Materials and Biomolecules CO2Me N

Fc MeO2C

Fc

N CO2Me N Fc

Fc N MeO2C N Fc

CO2Me

Fc

N CO2Me

36

Scheme 13.10 Table 13.3 Fem amino acid derivatives Compound Fem-Gly-OH (38a) Fem-Ala-OH (41) Fem-Pro-OH (40) Fem-Phe-OMe Fem-Phe-O(t -Bu) Fem-Leu-OMe Fem2 -Gly-OMe (39)

Ref.

Compound

75 70, 75 70 71 75 71 74

Fem-Gly-OMe (38b) Fem-Ala-OMe Fem-Met-OMe Fem-Phe-OH (43) Fem-Leu-O(t -Bu) Fem-Val-O(t -Bu)

Ref. 75 75 71 71, 75 75 71

Metzler-Nolte and coworkers explored the peptide-coupling chemistry of Fem amino acid derivatives. Coupling of amino acids and peptides to the C-terminus of Fem amino acid esters proceeds readily after hydrolysis of the ester functionality and HBTU activation.71, 76–78 L-Leu, as well as L- and D-Ala, were successfully used for this coupling reaction. NMR investigations in combination with molecular modelling was used to elucidate the structure of these conjugates in solution and explain the differences in reactivity.71 These ferrocenyl benzylamine derivatives were later applied for labelling the C-terminus of amino acids and peptides.78 Starting from the Fem dipeptide 42, a number of bimetallic derivatives such as 44 and 45 have been prepared (Scheme 13.11).71, 76 Fem derivatives were also used as electrochemically active labelling groups for the detection of biomolecules, in particular amino acids and peptides. In addition, they can be used as protecting groups during peptide synthesis.

The Bioorganometallic Chemistry of Ferrocene

511

O H2N

N H

O N Fe

42

O

OH

OH O

Cr OC

C O

CO 43

H N O

Cr OC

C O

CO

N H

Fe

O N H

O

O

N Fe

Fe

45

N H

H N

O N H

O N Fe

44

Scheme 13.11

Eckert and coworkers introduced the Fem group as a lipophilic residue to mask the highly polar glycine amide moiety during peptide synthesis in solution.74, 75, 79 Peptide coupling of Fem-modiﬁed glycine 38a to other amino acids or peptide fragments proceeds readily by employing the standard carbodiimide method used for peptide synthesis in solution.80, 81 There are several advantages of this masked glycine derivative for peptide synthesis in solution. Firstly, the intermediates and products show increased solubility in apolar organic solvents due to the lipophilic Fem rest. This results in complete reactions and improved yields. In addition, the chromatographic puriﬁcation over silica gel is facilitated, since the products can be eluted with less polar solvent mixtures. Secondly, the ferrocene chromophore facilitates product identiﬁcation and isolation during chromatographic puriﬁcation. Finally, the growing peptide is more stable against racemisation during the coupling steps when it is assembled in step-wise fashion starting from the C-terminus. Removal of the Fem group is accomplished by treatment with TFA/β-thionaphtol in dichloromethane at room temperature for 2–4 h after assembly of the complete peptide sequence.74, 75, 79 The usefulness of

512

Ferrocenes: Ligands, Materials and Biomolecules

Fem–Gly derivatives for peptide synthesis in solution is illustrated by the successful synthesis of [Leu5 ]-enkephalin and H-(Gly)6 -OH.74 The Fem group can also be adapted for the synthesis of masked asparagine residues from aspartic acid79 or it may serve as a protecting group for the cysteine thiol functionality.82, 83 The compound H-Cys(S-Fem)-OH (46) can be conveniently prepared by reacting cysteine with FcCH2 OH (47) in aqueous acetone in the presence of a catalytic amount of acid. This side-chain protected cysteine can be transformed into the N -terminus Boc or Fmoc protected derivatives. Similarly to the Fem–Gly derivative, reaction with TFA/β-thionaphtol in CH2 Cl2 liberates the asparagine, while the Fem moiety can be cleaved from Cys by TFA or by soft heavy metal ions, such as Ag+ or Hg2+ .82, 83 The usefulness of the Asn(Fem)-protected residue in combination with the Fem–Gly derivative was demonstrated for the synthesis of the 24–31 octapeptide sequence of human β-endorphin.79 The Cys(Fem) was successfully used in the synthesis of the tripeptide glutathione (GSH).82 Amino acid derivatives of ferrocenecarboxylic acid (FcCO2 H, 48) were already reported in the ﬁrst paper by Schl¨ogl.8 Up to now, they probably represent the largest class of ferrocene bioconjugates if simple amino acids are included in the deﬁnition of ‘biomolecules’. The most extensively explored way to couple the ferrocene moiety to amino acids and peptides is via amide formation between 48 and the terminal amino group. The ﬁrst amino acid and dipeptide derivatives of this kind, FcCO-GlyOMe, FcCO-Gly-OH and FcCO-Gly-Leu-OEt were reported by Schl¨ogl as early as in 1957.8 To make coupling of a carboxylic acid with an amino group possible under mild conditions at room temperature, the acid needs to be activated. There are three strategies to achieve this (Scheme 13.12): 1) transformation into FcCOCl (49);8, 84 2) transformation into the succinimide 50 or benzotriazole ester 51, with the use of DCC or EDC in conjunction with N -hydroxysuccinimide (HOSu) or 1-hydroxybenzotriazole (HOBt);85 or 3) activation of the acid in situ by HBTU or TBTU.86 (COCl)2 or SOCl2

4

HO-Su HOBt

FcCOCl (49) FcCO2Su (50) FcCO2Bt (51)

Amino acid ester, Base

O HBTU or TBTU Amino acid ester, Base

CO2R1 Fe

HN R2

Scheme 13.12

In a reaction of Boc-Fca-OH (52) and HOBt in the presence of EDC, Kraatz and coworkers isolated N -oxide 53 as a rearrangement product in addition to the expected activated Boc-Fca compound 54 (Scheme 13.13).87 Formation of the thermodynamic

The Bioorganometallic Chemistry of Ferrocene NH-Boc Fe

NH-Boc HOBt, EDC

O

Fe

OH

52

NH-Boc +

O N

N

53

513

N+

Fe

O−

O O N

N

N

54

Scheme 13.13

product 53 is favoured in polar solvents, while the kinetic product 54 is obtained in highest yield in dry dichloromethane. The active ester 54, but not the rearranged amide 53, reacts with glycine to form the desired amide.87 Beck and coworkers reported another interesting method to transform ferrocenoyl-amino acids into the corresponding dipeptide derivatives via oxazolone formation with the use of a carbodiimide.88 A large number of mono-substituted ferrocenoyl amino acid and dipeptide derivatives are known (Table 13.4). Several of these derivatives have been structurally characterised by X-ray crystallography. In the solid state, inter-molecular hydrogen bond interactions are a dominant feature, typically resulting into zig-zag,89, 90 sheetlike91 and helical packing arrangements.86, 91 Unlike the 1,n -disubstituted derivatives discussed in Section 13.2.3, most of these derivatives do not have an ordered structure in solution. The group of Kraatz studied the interaction of several ferrocenoyl-dipeptides (FcCOGly2 -OEt, FcCO-Ala2 -OBn, FcCO-Phe2 -OMe, FcCO-Leu2 -OMe, FcCO-Val2 -OMe, FcCO-Leu-Phe-OMe and FcCO-Val-Phe-OMe) with 3-amino-pyrazole (Apzl), 3amino-5-methyl-pyrazole (3-AMP), and 3-triﬂuoroacetylamido-5-methylpyrazole (3TFAc-AMP).91, 92 In solution, these compounds form 1:1 adducts stabilised by hydrogen bonds as shown for 55 in Scheme 13.14. From the results of 1 H NMR titration experiments, the binding constants between these pyrazole derivatives and the dipeptides in CDCl3 were determined to range from 8 to 27 M−1 . Lower binding constants were found in solvents of higher polarity such as acetonitrile or acetone. These values correspond to moderate binding energies of 5–7 kJ mol−1 . Recently, Kraatz published an overview of his group’s work on ferrocenoyl peptides.93 An interesting result is that the ferrocene/ferrocenium redox-potential shifts towards less positive values when the attached peptide becomes more helical. On going from FcCO-Pro-OH to FcCO-Pro4 -OBn, a decrease of the redox potential of about 30 mV occurs.93, 94 In another paper, Kraatz and coworkers determined the redox potential of ferrocenoyl dipeptides in a large variety of solvents, concluding that the redox potential could be correlated with the hydrogen donor ability α of the Kamlet–Taft formalism.95 The peptide derivatives FcCO-Phe-OMe (56) and FcCO-Ala-Phe-OMe (57) were subjected to 57 Fe M¨ossbauer spectroscopic investigations.86 The values for the isomer shift and quadrupole splitting are in the range reported for several other ferrocene derivatives that have electron withdrawing substituents on the cyclopentadienyl ring,96, 97 and resemble those of ferrocene-PNA derivatives.98 The peptide fragment appears to have little inﬂuence on the electric ﬁeld gradient of the iron nucleus.

514

Ferrocenes: Ligands, Materials and Biomolecules Table 13.4 Ferrocenoyl amino acid and dipeptide derivatives

Compound

Ref.

Compound

Ref.

Ferrocenyl amino acid derivatives 99a

FcCO-Gly-OMe

FcCO-Ala-OMe

88

FcCO-Gly-OEt

8, 85, 88

FcCO-Ala-OEt

88

FcCO-Gly-OBn

100a

FcCO-Ala-OBn

85a

FcCO-Gly-OH

8, 88

FcCO-Ala-OH

88

FcCO-Gly-OSu

101b, c

FcCO-Asp-OH

95a

FcCO-Cys(SBn)-OMe

85c

FcCO-Asp(OBn)-OBn

FcCO-Met-OEt

102

FcCO-Pro-OH

FcCO-Pro-OBn

85

FcCO-Phe-OMe

84, 86, 88,

104c

95 94, 103a

FcCO-Pro-OMe

84

FcCO-Phe-OH

88, 103

FcCO-Phe-OtBu

101

FcCO-Phe-OBn

85

FcCO-Tyr-OBn

85

FcCO-Val-OMe

84, 105a

103

FcCO-DL-Val-OH

103

FcCO-Lys(Nε -CO-Fc)OH

103

FcCO-Glu(OBn)-OBn

85a

FcCO-His-OMe

106

FcCO-His(Nε -CO-Fc)-OMe

106a

FcCO-Gly-NHC2 H4 S]2

107

FcCO-His(Nε -CO-Fc)-OMe

108

[FcCO-Ala-NHC2 H4 S]2

109c

FcCO-DL-Leu-OHd d

Ferrocenoyl dipeptide derivatives FcCO-Gly-Gly-OEt

88, 91, 92a

FcCO-Gly-Gly-OMe

104c

FcCO-Gly-Ala-OMe

88

FcCO-Gly-Pro-OEt

110

FcCO-Ala-Ala-OBn

91, 92

FcCO-Ala-Ala-OMe

88

FcCO-Ala-Pro-OEt

89a , 90, 111

FcCO-Ala-Phe-OEt

104c

FcCO-Ala-Phe-OMe

86a

FcCO-Asp(OBn)-Asp(OBn)OBn

95

FcCO-Asp(OBn)-Glu(OEt)OEt

95

FcCO-Asp(OBn)-Cys(SBn)OMe

95

FcCO-Pro-Gly-OEt

110

FcCO-Gly-Leu-OEt

8

FcCO-Leu-Phe-OMe

91, 92a a

FcCO-Pro-Pro-OBn

94

FcCO-Ala-Pro-NHPya, e, f

112

FcCO-Phe-Phe-OMe

FcCO-Phe-Leu-OBn

104c

FcCO-Phe-Ser-OEt

104c

FcCO-Val-Phe-OMe

91

FcCO-Val-Val-OMe

91

FcCO-Leu-Leu-OEt

104c

FcCO-Leu-Leu-OMe

91

[FcCO-Pro-Pro-NHC2 H4 S]2

108

91, 104c

a X-ray crystal structure determined. b OSu = succinimidyl c Not well characterised. d Racemic mixture used. e NHPy = amide of 2-amino-pyridine. f Complexation of PdCl2 to the pyridine nitrogen atom has been is also reported.

The Bioorganometallic Chemistry of Ferrocene R

515

OBn

O N H

HN R Fe

O H

O

H N N

N H

55

Scheme 13.14

The ferrocenoyl derivatives of oligo-peptides for which no biological function has been reported are summarised in Table 13.5. As described above, most simple ferrocenoyl amino acid and peptide derivatives do not adopt a ﬁxed conformation, the exceptions being FcCO-Pro-Pro-Phe-OH and FcCO-(Pro)n -OBn (n = 2–4).94 The former reveals a structural motif characteristic for a β-turn in proteins. The two Pro residues are connected in a cis-fashion and a strong hydrogen bond is present between the Phe ˚ amide nitrogen atom N3 and the ferrocenyl oxygen atom (N•••O contact = 2.853 A). In the X-ray crystal structure of FcCO-(Pro)4 -OBn, the amide bonds between the Pro residues are trans-conﬁgured and the compound has a left-handed helical conformation characteristic for polyproline II.113, 114 X-ray crystal structures of the related compounds FcCO-(Pro)3 -OBn and FcCO-(Pro)2 -OBn also show a left-handed helical orientation, with trans-conﬁgured proline linkages.94 NMR-spectroscopic data conﬁrm that the solid state conformation of these proline compounds is maintained in CDCl3 and CD3 CN solutions.94 To possibly discover conformational changes upon introduction of the lipophilic ferrocene group, a 21-mer helical peptide containing six ferrocenoyl groups attached to lysine-Nε -amino groups has been prepared.115, 116 The spectroscopic investigations revealed a helical conformation similar to that of the unmodiﬁed peptide. Another paper reported the reaction of a L-lysine polymer of 126.2 kDa molecular weight with 49.117 In this case, the conjugate was obtained with 6 % of the lysine residues carrying a ferrocenoyl tag. In the same paper, a 52.1 kDa lysine copolymer, having the amino acid molar ratio Lys:Ala:Glu:Tyr of 34:45:14:7, was reacted with 49. This resulted in derivatisation of 14 % of the lysine NH2 groups. These two water-soluble ferrocenoyl polymers were used for the construction of an amperometric glucose-sensing electrode, by immobilising them together with the enzyme Glucose Oxidase (GOD) on the surface of a glassy carbon electrode. The anti-proliferative properties of ferrocene-containing polymers are further discussed in Section 13.7.2. In addition to the compounds treated in Section 13.2.2, a number of ferrocenoyl peptide derivatives have been reported, in which either the peptide alone or the conjugate have a biological function. These peptides are included in Table 13.1. Most of the peptide derivatives presented thus far have been synthesised in solution. For the preparation of larger peptides, solid phase peptide synthesis (SPPS) is the method of choice. To develop SPPS for ferrocene peptide conjugates, a few precautions are necessary. In particular, the original Merriﬁeld synthesis uses Boc chemistry as the N -terminal protecting group, which is removed by concentrated TFA. Moreover, cleavage from

516

Ferrocenes: Ligands, Materials and Biomolecules Table 13.5 Ferrocenoyl oligo-peptide conjugates Ferrocenoyl tri, tetra and oligopeptide derivatives

Compound

Ref.

Compound

Ref.

94

FcCO-Pro-Pro-Phe-OH

94a

FcCO-Gly-Phe-Leu-OHc

103

FcCO-Pro-Pro-Pro-OBn

94a

FcCO-Gly-Gly-Tyr(OBn)Arg(NO2 )-OMe

118

FcCO-Gly-Gly-Tyr(OBn)Arg(NO2 )-OH

FcCO-Gly-Gly-Tyr-Arg-OH

118

FcCO-(SP)

FcCO-Pro-Gln-Phe-Phe-Gly-LeuMet-NH2 c

103c

[FcCO-(Pro)x -NHC2 H4 S]2

108d

FcCO-(Gly)4 -NHC2 H4 SH

119

[FcCO-(Pro-Pro-Gly)y NHC2 H4 S]2

109e, f

FcCO-Pro-Pro-Pro-OBn

a

118 32b

a X-ray crystal structure presented in this reference. b SP is an undecapeptide (see Table 13.1). c This is the 4–11 amino acid fragment of SP. d x = 3–6. e Not well characterised. f y = 1–3.

the resin is performed by HF. Ferrocene or its derivatives will generally not withstand these conditions and this is probably the reason why most of the Fer peptide conjugates prepared by Tartar and coworkers in the 1980s were poorly characterised or decomposed directly during preparation (see Section 13.2.1 and Table 13.1). A ferrocenoyl labelled undecapeptide Substance P (58, entry 10 in Table 13.1) has been prepared via classical Merriﬁeld solid phase synthesis methods.32 In contrast to the Fer derivatives presented in Section 13.2, no serious oxidation or decomposition was observed upon cleavage of the ferrocenoyl conjugate 58 from the resin with HF. The higher robustness of 58 has been attributed to the increase in oxidation potential of the ferrocene moiety, a consequence of the electron withdrawing effect of the amide group. Fortunately, milder methods for SPPS are available nowadays. The Fmoc group which is commonly employed as a temporary protecting group, can be removed by dilute piperidine. A variety of linkers is commercially available and permits cleavage under mildly acidic (e.g. dilute TFA, or even acetic acid) or even basic conditions (concentrated ammonia in MeOH). We have adopted such methodology for the preparation of numerous novel ferrocene peptide conjugates. All of those were obtained in good yield and without signiﬁcant signs of decomposition. The same SPPS methods can also be used for the synthesis of ferrocene PNA conjugates as described later in Section 13.4.4. The Metzler-Nolte group has used enkephalin as a model peptide for the synthesis of new metal–peptide bioconjugates. In particular, the chemistry was established for different ferrocene derivatives. C-terminal carboxamides are obtained by using the HMB resin, from which the product can be cleaved by concentrated NH3 in MeOH. Alternatively, the Rink amide linker produces the same product, but upon acid cleavage. It is indeed possible to use concentrated acids such as TFA in cleavage mixtures for ferrocene peptides. In this case, a large excess of phenol must be added to prevent excessive oxidation of ferrocene. The free C-terminal carboxylic acid is produced from Wang resin, which again can be cleaved by a strong acid. Finally, should strong acids

The Bioorganometallic Chemistry of Ferrocene

517

be avoided, more labile linker such as the trityl group may be employed. By using the different methods described here, a large number of organometallic peptide conjugates can be prepared for biological studies.29 The same group has recently prepared metallocene derivatives for the study of cellular uptake and subcellular localisation. Conjugates were synthesised by FmocSPPS using ferrocenoyl or cobaltocenium groups as organometallic labels (Scheme 13.15).35, 36 All conjugates were readily puriﬁed by preparative HPLC and characterised. To enable visualisation inside living cells, an additional ﬂuorescence tag was added on an orthogonally protected lysine residue. TAT–metallocene conjugates (59, entry 16 in Table 13.1, Scheme 13.15) were prepared to study how the lipophilic metallocene will inﬂuence cellular uptake. Nuclear localisation was studied by binding the metallocene to the simian virus SV40 nuclear localisation sequence (NLS, 60 in Scheme 13.15, entry 15 in Table 13.1). Only the ferrocene conjugate of 59 is readily taken up by HepG2 cells and accommodated mainly in the cytoplasm. Both the ferrocene-NLS and cobaltocenium-NLS conjugates 60, on the other hand, are readily taken up but localised in the nucleus of the cells.35 In addition, a variety of ferrocene peptide conjugates with anti-bacterial activity has been prepared by SPPS (Table 13.1).37, 38 Their biological activity is described in Section 13.7.3. O M

Lys Gly Arg Lys Lys Arg Arg Gln Arg Arg Arg NH2 S HN Fluorescein

TAT conjugates (59): M = Fe, Co+

O Lys Pro Lys Lys Lys Arg Lys Val OH M

S HN Fluorescein

NLS conjugates (60): M = Fe, Co+

Scheme 13.15

The Sonogashira reaction is a palladium-catalysed cross-coupling of terminal alkynes and halogenated arenes.120–122 The Metzler-Nolte group has developed a practical and ﬂexible two-step method for labelling the C- or N -terminus of amino acids and peptides, based on Sonogashira coupling.123–125 Synthetic methods used to attach a marker to the C-terminus are shown in Scheme 13.16. The ﬁrst step for introducing a label on the C-terminus consists of functionalising a Boc-protected amino acid or Boc-protected dipeptide at its C-terminus by amide formation with propargylamines 61 (R = H (a), Et (b)) or with p-iodoaniline (62). The synthesis starts with the preparation of amides 63 and 64 from FcCO2 H (48) and 62 or 61, respectively, via standard solution phase peptide synthesis methods.126 In the second step, the Sonogashira coupling is performed, yielding the ferrocene functionalised amino acids and peptides such as 65 and 66. Amide 64b proved to be superior to its congener 64a because of its signiﬁcantly enhanced solubility and easier puriﬁcation of conjugates like 66b.

Fe

Boc N H

Boc N H

H N

O

O

O

H N

O

O

N H

H

N C H Et2

64b

R2

R2

HN C Et2

R1

R1

+ I 67

I+ H

H + I

C NH R2

O

C N H O

R1 O

64a (R = H) 64b (R = Et)

63

NH

O

Boc N H

Boc N H

THF, ∆, 4 h

Pd(PPh3)2Cl2 CuI, NEt3

THF, ∆, 4 h

Pd(PPh3)2Cl2 CuI, NEt3

THF, ∆, 4 h

Scheme 13.16

OCH3

Fe

Fe

Pd(PPh3)2Cl2 CuI, NEt3 O

O

Fe

R1

R1

O

R2

O

R2

O

68

N C H Et2

N H

HN C Et2

H N

H N

C NH R2

O

C N H O

R1 O

66a (R = H) 66b (R = Et)

65

NH

O

OCH3

Fe

Fe

518 Ferrocenes: Ligands, Materials and Biomolecules

The Bioorganometallic Chemistry of Ferrocene

519

This method is not restricted to C-terminal labelling, but via a slight modiﬁcation can be adapted for the introduction of a marker on the N -terminus (Scheme 13.16, bottom). The amide 67 of p-iodobenzoic acid and H-Leu-OMe was reacted with 64b under identical Sonogashira coupling conditions to yield N -terminally labelled 68.124 This method was later extended to label the naturally occurring pentapeptide enkephalin (Enk, see above) by the same method at the N -terminus.30 This two-step labelling procedure is very attractive for several reasons. First of all, both steps proceed in good yields (up to 95 %). Secondly, commercial reagent-grade solvents can be employed in both steps while the use of deoxygenated solvents is required only for the Sonogashira coupling reaction. Thirdly, the method is versatile and of wide scope as it tolerates a variety of functional groups, including alcohols (Ser), thioethers (Met), esters and amides.125 This work has recently been extended to the labelling of peptides at internal positions.30 The unnatural amino acid (p-iodophenyl)alanine (p-I-Phe) was incorporated into a number of dipeptides. Subsequently, Sonogashira coupling with ferrocenyl alkynes such as 64b and ethynylferrocene (69) was performed to afford ferrocene– dipeptides 70–77 in good yields (Scheme 13.17). Finally, the [p-I-Phe4 ]-Enk derivative (78) was prepared by solid phase peptide synthesis. After cleavage from the resin and HPLC puriﬁcation, 78 was reacted with 64b under Sonogashira conditions to yields the conjugate 79 in almost quantitative yields (Scheme 13.18).30 In addition to the Sonogashira procedure, several other methods for the derivatisation of the C-terminus have been developed. One of these, amide formation between Fem-(p-methyl)benzylamine and activated amino acids has already been presented in Section 13.2.2. Eisenthal and coworkers prepared two ferrocenoyl derivatives 80 of the channel forming peptide alamethicin (ALM) produced by the fungus Trichoderma viride.127 One of the derivatives (80a) was prepared from 48, whereas the other (80b) was synthesised from 19 (Scheme 13.19). In both cases, the ferrocenoyl moieties were linked via ester formation to the C-terminus of the 20-mer peptide in solution. In the reduced (neutral) form, both ALM derivatives 80 form voltage-dependent channels in planar lipid bilayers at positive potentials, with conductance properties similar to unmodiﬁed alamethicin. While in the lipid bilayer form, oxidation of 80b results in a shorter lived channel whereas the oxidation of 80a causes a time-dependent elimination of channel openings, which can be restored by increasing the trans-bilayer potential. Pretreatment of the ferrocenoyl peptides with oxidising agents alters their single-channel properties in a qualitatively similar manner. The most obvious method for C-terminal derivatization is probably the reaction of (activated) amino acids or peptides with aminoferrocene. Surprisingly, this reaction has only very recently been explored and a handful of compounds of the general formula Boc-Aaa-NHFc have been prepared in the Metzler-Nolte group (Aaa = Gly, Leu, Phe, Val, Cys(Acm) and Tyr(t-Bu)). The X-ray single crystal structure of the glycine derivative has been reported.128 An unconventional procedure for labelling the N -terminus of amino acids has been developed by Jaouen and coworkers.129 The method consists of the synthesis of a pyrylium salt 81 from FcLi (82) and 2,6-dimethyl-γ -pyrone (83) and subsequent

Boc

H N

R

O

N H

O

OMe

I

+

69

Fe

64b

Fe

O N H

H

H

Scheme 13.17

THF / NEt3 (1:1) 6 h / RT / Ar

Pd(PPh3)2Cl2 CuI

THF / NEt3 (1:1) 6 h / RT / Ar

Pd(PPh3)2Cl2 CuI

Boc

Boc

H N

H N

O N H

N H

O

O

OMe

OMe

R = CH3 R = CH(CH3)2 R = CH2C6H5 R = CH2C6H4OH

R

R

O

Fe

70 71 72 73

74 75 76 77

From 64b From 69

Fe

N H

O

520 Ferrocenes: Ligands, Materials and Biomolecules

H 3C

O

H N

FmocHN

O

O

N H

OH

O

H N

O

Wang-Linker

N H

79

O

H N

O

Fe

H 3C

Scheme 13.18

OH

N H

O

SPPS, then Cleavage O

O N H

OH

O

H N

O N H

64b, Sonogashira Coupling

H N

78

O

H N

O

I

OH

The Bioorganometallic Chemistry of Ferrocene 521

522

Ferrocenes: Ligands, Materials and Biomolecules

Ac-Aib-Pro-Aib-Ala-Aib-Ala-Gln-Aib-Val-Aib-Gly-Leu-Aib-Pro-Val-Aib-Aib-Glu-Gln-Phol O

Aib = a-aminoisobutyric acid, Phol = phenylalaninol, R = H (80a) or CO2H (80b)

Fe R

Scheme 13.19 Structures of two alamethicin ferrocene derivatives 80

FcLi

+

LiO O

+ O

O HPF6

O

82

Fe

Fe

83

81

PF6

OEt + N

O

PF6

H-b-Ala-OEt

Fe 84

Scheme 13.20

acidiﬁcation (Scheme 13.20). This pyrylium salt reacts with amines, in this case βalanine, to yield pyridinium salt 84. To the best of our knowledge, this method has not been adapted for the introduction of a ferrocene tag to larger biomolecules. In two papers and a patent, Eckert and Koller described a variety of ferrocene compounds for derivatisation of the N -terminus of amino acids and the lysine-NH2 moieties of the protein BSA.101, 130, 131 The objective was to investigate which of the reagents was best suited for labelling purposes and subsequent electrochemical detection of the conjugates by HPLC–ECD (HPLC with Electrochemical Detection). Reagents for HPLC–ECD derivatisation that were investigated by Eckert and Koller include, for example, FcSO2 Cl (85), FcCOCl (49), (FcCO)2 O (86), (Fc(CH2 )2 CO)2 O (87) and (ferrocenylmethyl)isocyanate (88). Of these derivatives, 87 was found to be best suited on the basis of labelling results and the sensitivity for electrochemical detection of the corresponding conjugates.130 However, some results in the literature are contradictive, because Koppang et al. found 85 and 49 to be good derivatisation reagents for HPLC–ECD,103 whereas Eckert and Koller found these to be unsuitable.130 In a related paper, Shimada et al. compared FcNCS (89) and FcCH2 NCS (90) for glycine and 4aminobutyric acid labelling.132 The latter displayed higher reactivity and favourable electrochemical properties and was adapted for 4-aminobutyric acid determination in biological samples by HPLC–ECD. The labelling methods mentioned thus far are suitable for the derivatisation of amino and carboxylic acid moieties of peptides and amino acids. However, selective labels for the SH group of cysteine have also been reported.133–135 For the labelling of glutathione (GSH, 91), ferrocenylethylmaleimide (92), (iodoacetyl)ferrocene (93)

The Bioorganometallic Chemistry of Ferrocene NH2

FcCH2OH (47)

H N

HOOC

O N H

O

S

O

O

HOOC

N H SH

O

NH2

O

H N

92

OH

HOOC

O N H

O

O

GSH (91)

S

OH O

O NH2

H N

HOOC O Fc

O

Fc

96

H N

OH

N

Fc NH2

523

93

O OH

N H

Fc

94

O

S

I O

N

O

O 95

Fc

Scheme 13.21

and ferrocenylmethanol (47) turned out to be very effective to yield 94, 95 and 96, respectively (Scheme 13.21).133, 135, 136 A related maleimide derivative has also been used to label the hexapeptide Ac-Arg-Arg-Ala-Ser-Leu-Cys-OH.134 This labelled hexapeptide was applied for the detection of serine phosphorylation by protein kinase A via electrochemical methods. The use of 92 for the labelling of Cys residues in proteins such as Cytochrome P450 is described in Section 13.3.1. Compound 93 has also been used for the labelling of BSA (Section 3.2) and a 5 -thiolated oligonucleotide (see Section 4).135 It should be noted that (chloromercuri)ferrocene (FcHgCl, 97) and N -chloromethyl ferrocenecarboxamide (FcCONHCH2 Cl, 98) have also been used for the introduction of a ferrocene marker on Cys-SH groups in proteins (see Section 13.3). Section 13.2.2 also describes the use of the Fem residue as a protecting group for the cysteine SH moiety during peptide synthesis. Toniolo, Maggini and coworkers prepared a nonapeptide with a 310 -helical conformation, in which two tyrosine phenol rings were esteriﬁed with acid 48.137–139 These synthetic helical peptides were shown to have a hydrophobic cavity, which can serve as an efﬁcient host for [60]fullerene. Microcystins are a class of low-molecular weight cyclic peptide hepatoxins. These microcystins are produced by cyanobacteria (blue-green algae) in eutropic lakes and drinking water reservoirs, particularly in global regions that are hot and relatively dry.140 Because these substances are hazardous to human beings, cattle and wildlife, it is important to develop methods for rapid detection of these toxins. Lo, Lam and coworkers recently developed an elegant method that makes the detection of a particular microcystin, namely microcystin-LR (99), possible via a quick electrochemical measurement.141 This method is based on the speciﬁc derivatisation of the exocyclic double bond of the α,β-unsaturated carbonyl moiety of 99 with 6-ferrocenylhexanethiol

524

Ferrocenes: Ligands, Materials and Biomolecules COOH HN OCH3

O

O

Ph

NH

O

O

CH2

O O

NH

NH

HN

+ Fc(CH2)6SH (100)

HN

HN H2N

N

O

NH

O HO

99 COOH N

HN OCH3

O

O

Ph

NH

O

O

NH

NH

HN

HN

HN H 2N

O O

S(CH2)6Fc

NH

O

O HO

101

Scheme 13.22

(100) (Scheme 13.22), and the subsequent electrochemical detection of the redoxlabelled conjugate 101. The limit of detection was found to be ca. 18 ng, which is in the range of other common detection methods, such as thin-layer chromatography and HPLC with UV-detection. It must be noted that much lower levels of microcystins can be detected with more expensive methods, such as ﬂuorescence spectroscopy and HPLC–MS. The electrochemical method, however, is quick, speciﬁc and inexpensive. 13.2.3

Amino Acid and Peptide Conjugates of 1,n -disubstituted Ferrocene

In this section, only peptidic ferrocene derivatives and related compounds will be considered. It is noteworthy that no peptide derivatives with a 1,2- or a 1,3- substitution pattern at the ferrocene moiety have been reported (but see note added in proof at the end of this chapter). On the other hand, a large number of amino acid and peptide derivatives with a 1,n -substitution pattern are known (Table 13.6). A few Tyr derivatives of ferrocene were reported in which the amino acid is connected via a methylene group to one cyclopentadienyl ring.142 If only direct peptide linkages to the Cp ring are taken into account, then three different classes of compounds need to be considered (Scheme 13.23). Systems derived from ferrocene-1,1-dicarboxylic acid (19, Fcc) and 1,1 -diaminoferrocene (102, Faa) will give rise to a parallel orientation of the peptide strands, whereas compounds derived from Fca (18, see Section 13.2.1 and Scheme 13.4) will lead to an anti-parallel orientation. The Fca-based system resembles

The Bioorganometallic Chemistry of Ferrocene

525

Table 13.6 Overview of the reported 1,1 -type bis(amino acid) and bis-peptide derivatives Fcc type Entry

fc(COX)(COY) X

Y

Helical chirality

Conformation, Available Datad

Reference

1a

Gly-OH

Gly-OH

P M

(1,2 ); X, CD (1,5 ); X, CD

164

2

Gly-OEt

Gly-OEt

P M

(1,4 ); X, CD (1,3 ); X, CD

164

3

Gly-NH2

Gly-NH2

P M

(1,2 ); X, CD (1,5 ); X, CD

143

4

Ala-OMe

Ala-OMe

P

(1,2 ); CD

165

5 6 7

Phe-O-Me

D-Phe-OMe Phe-OMe

Phe-O-Me

D-Phe-OMe Ala-O-Me

P

(1,2 ); X, CD

86, 165

M

(1,5 );

CD

165

P

(1,2 );

CD

165

8

D-Phe-OMe

Ala-OMe

P M

(1,2 ); CD (1,5 ); CD

165

9

Cys(SBz)-OMe

Cys(SBz)-OMe

P

(1,3 ); X, CD

151

10

Glu(OEt)-OEt

Glu(OEt)-OEt

P

(1,2 ); CD

166

11

His-OMe

His-OMe

P

(1,2); CD

167

M

(1,4 );

X

158

P

(1,2 ); X

105

P

(1,2 );

X, CD

110

P

(1,2 );

X

111

P

(1,2 ); X

111

P

(1,2 );

Xe

86

P

(1,2 );

X, CD

90

12 13 14 15 16 17 18 19

Pro-OMe Val-OMe Gly-Pro-OEt Gly-Leu-OEt Gly-Phe-OMe Ala-Phe-O-Me Ala-Pro-OMe Ala-Pro-OEt

Pro-OMe Val-OMe Gly-Pro-OEt Gly-Leu-OEt Gly-Phe-OMe Ala-Phe-OMe Ala-Pro-OMe Ala-Pro-OEt

P

(1,2 ); X, CD

111 111

20

D-Ala-D-ProOEt

D-Ala-D-ProOEt

M

(1,5 );

21

Ala-Pro-OPr

Ala-Pro-OPr

P

(1,2 ); X, CD

90

P

(1,2 );

X, CD

90

P

(1,2 );

X, CD

148

22 23

Ala-Pro-OBz Ala-Pro-NHpy

Ala-Pro-OBz Ala-Pro-NHpy

X, CD

24

Ala-ProNHpy (PdCl2 )

Ala-Pro-NHpy

P

1,2 ); X, CD

148

25

Ala-D-Pro-NHPy

Ala-D-Pro-NHPy

P

(1,2 ); X, CD

168

(continued overleaf )

526

Ferrocenes: Ligands, Materials and Biomolecules Table 13.6 (continued ) Fcc type

Entry

fc(COX)(COY) X

Helical chirality

Conformation, Available Datad

Reference

Y

26

D-Ala-Pro-NHPy

D-Ala-Pro-NHPy

M

(1,5 ); X, CD

168

27

Pro-Ala-NHPy

Pro-Ala-NHPy

P

(1,2 ); X, CD

169

28

D-Pro-D-AlaNHPy

D-Pro-D-AlaNHPy

M

(1,5 ), X, CD

169

29

Pro-D-Ala-NHPy

Pro-D-Ala-NHPy

M

(1,4 ); X, CD

169

30

D-Pro-Ala-NHPy

D-Pro-Ala-NHPy

P

(1,3); X, CD

169

M?

(1,5 );

X, CD

150

31 a

Met-Met-OMe

Met-Met-OMe

32

Gly-NH2 -(CH2 )S-

Gly-NH2 -(CH2 )S-

P M

(1,2 ); X, CD (1,5 ); X, CD

156

33

NH-(CH2 )2 -NHCO-ImNH-(β-Ala)CO-Im

NH-(CH2 )2 -NHCO-Py-CONH-Im-CONH-Py(CH2 )2 N(CH3 )2

P

(1,2 ); CD

170

34a

Gly-Val-NH2 (CH2 )-S-

Gly-Val-NH2 (CH2 )-S-

P M

(1,2 ); X (1,2 ); X

171

35b

NH-(β-Cyclodextrin)

b) -(CH2 )CH2

P

(1,2); CD

172

36

[Glu63 OEt64 ] (Dendrimer (G6))

[Glu63 OEt64 ] (Dendrimer (G6))

P M

(1,3 ); CD (1,4 ); CD and calculations

166

37a

–

His(Nπ )-

P M

(1,2 ); CD, X (1,5 ); CD, X

167

Fca type fc(NHX)(COY) X

Helical chirality

Conformation, Available Data

Reference

Y

38

H

C6 H3 Cl2

P

(1,1 ) calculations

173

39

Me

Me

–

(1,n ), no data

174

40

Et

Me

–

(1,n ), no data

174

41

Ac

Val-OMe

P

(1,2 ); CD

175

The Bioorganometallic Chemistry of Ferrocene

527

Table 13.6 (continued ) Fca type

42 43

fc(NHX)(COY) X

Helical chirality

Conformation, Available Data

Reference

Y

Ac

Ile-OMe

P

(1,2 ); CD

175

P

(1,2 )

176

Ala-NH-Ac

Ala-OMe

Calculations 44 45

D-Ala-Boc Ala-Boc

Ala-OMe Ala-Ala-OMe

M

(1,2 ); CD

177

P

(1,2 );

X, CD

178

46

Ala-Ala-Boc

D-Ala-D-AlaOMe

P

(1,2 );

CD

177

47

Ala-Gly-Val-Ac

Ala-Gly-Val-Leu

P

(1,2); CDf

179

–

(1,n ),

no data

174

–

(1,n ), no data

174 173

48 49

EtSCH2 CH2 (EtO)2 P(O)

Me Me

50c

-(N -)Phtalimide

C6 H3 Cl2

P

(1,3 ),

51

Ala-Boc

Ala-Fca-Me

P

(1,2); X, CD

180

52

D-Ala-Boc

D-Ala-Fca-Me

M

(1,2); X, CD

180

X

53

Boc

(CH2 S2 CH2 )Fca-Boc

M

(1,5 );

X (Fca1) (1;1 ); X (Fca2)

181

54

Boc

NH-(CH2 )2 -S2 (CH2 )2 -NHFc2-NH-Boc

M

(1,5 ); Fc1 X (1,1 ); Fc2 X

181

55

Thioctic acid (Thc)

Gly-Gly-Arg-Tyr

M

(1,4 )

182

Faa type fc(NHX)(NHY) X

Helical chirality

Conformation, Available Data

Reference

Y

56

Ala-Boc

Ala-Boc

M

(1,1 ); X, CD

183

57

D-Ala-Boc

D-Ala-Boc

P

(1,1 ); X, CD

183

58

CO-FcCOCOCH3

[CO-Fc-NH]n COFcCOCH3 ; nmax = 4

–

No data

184

a Cyclic ferrocenoyl peptide. b Group is directly attached to Cp ring (not strictly an Fcc derivative). c The nitrogen atom is part of the phtalimide ring. d ‘X’ denotes X-ray structure, ‘CD’ denotes a structure derived from CD-spectroscopic data. e Also Cobaltocenium derivative reported. f Only derivative prepared by SPPS in this table.

528

Ferrocenes: Ligands, Materials and Biomolecules R

O N H Fe

H N

H N

...

N H

O

Fe

O

R

O

O

... N H

N H

H N

...

O

O

... N H

R O Fca (‘Ferrocene amino acid’)

R O Fcc (‘Ferrocene diacid’)

O

Fe

H N

O

R

H N

N H H N

...

...

O R Faa (‘Ferrocene diamine’)

Scheme 13.23

turn structures in naturally occuring proteins most closely. On the other hand, there is also an interest in modelling parallel peptide strands. Therefore, investigations of the synthesis and in particular the structures in solution and the solid state of all three classes of compounds were performed. After discussing individual compounds in the following sections ﬁrst, a general classiﬁcation will be presented later in this section. An overview of the known Fcc-type derivatives is given in Table 13.6. In 1996, Herrick, Jarret, Curran and coworkers found that compounds of the general formula fc(CO-Aaa-OMe)2 (Aaa = Pro), have an ordered structure in CH2 Cl2 and CHCl3 (103a, Scheme 13.24).84 This ordered structure is comprised of two symmetrically equivalent hydrogen bonds between the amide NH and the methyl ester carbonyl moiety of the other strand. Irrespective of the peptide orientation, which gives rise to a parallel orientation (see Scheme 13.23), they suggested that such compounds may serve as peptide turn mimetics. This suggestion was originally based on rather scarce experimental data but has been multiply conﬁrmed later. A number of studies were carried out with a particular emphasis on the structural properties. In addition to singlecrystal X-ray diffraction analysis, numerous spectroscopic techniques were employed in solution (IR, NMR and CD spectroscopy). O MeO

N O

Fe

H N O 103a

R

O

R

N H

H

Fe

O

OMe R

O

...

O

H N

O

H N

R

... N H

103, ‘Herrick conformation’

Scheme 13.24 Original formulation by Herrick et al. of hydrogen bonding interactions in 1,n -disubstituted ferrocene peptides (left) and general representation of the ‘Herrick conformation’ (right)

The ordered conformation proposed by Herrick et al. and shown for 103 in Scheme 13.24 is indeed observed for the valine derivative fc(CO-Val-OMe)2 (104) in the crystalline state, although the hydrogen bond interactions are quite weak (N•••O contacts

The Bioorganometallic Chemistry of Ferrocene

529

˚ 105 The difference in hydrogen bond acceptor strengths between amide of 3.25 A). carbonyl moieties and ester carbonyl groups is nicely illustrated by the X-ray crystal structures of 104 and fc(CO-Gly-NH2 )2 (105).105, 143 Both compounds display an ordered conformation with the two symmetrical hydrogen bonds, but the N•••O contacts ˚ vs. 3.25 A). ˚ This type of arrangeare considerably shorter for 105 than 104 (2.88 A ment with two symmetrical hydrogen bonds which was originally proposed by Herrick, Jarret and Curran has later been generalised and termed the ‘Herrick conformation’ (Scheme 13.24, right). In addition to the 1,1 -disubstituted amino acid derivatives, a large number of dipeptide ester derivatives were prepared (Table 13.6). Hirao and coworkers were the ﬁrst to report and to investigate compounds of general composition fc(CO-Ala-Pro-OR)2 (106), with (R = Me (a), Et (b), n-Pr (c), Bn (d)).89, 90 As a representative example, the X-ray crystal structure of 106a is depicted in Figure 13.1.90 The structure reveals an ordered conformation related to 103 (Scheme 13.24), having intramolecular hydrogen bonds between the Ala–NH and the Ala–CO of another strand, with N•••O contacts ˚ and 3.04 A. ˚ All these derivatives with varying R groups display a simiof 2.91 A lar type of ordered structure in the solid state as well as in solution, even in polar media, such as MeCN and CDCl3 /DMSO-d6 (9/1 v/v) mixtures, as derived by combining the results from NMR, IR and CD spectroscopic investigations.90 Hirao and coworkers also prepared and structurally characterised the D-amino acid containing compound fc(CO-D-Ala-D-Pro-OEt)2 (106e).111 The fact that this compound is in an enantiomeric relationship to 106b has been illustrated by CD spectroscopy. Stereochemical aspects and their consequences for CD spectroscopy are further discussed in Section 13.2.3. Proline is the only naturally occurring amino acid that is derived from a secondary amine, thus no NH proton is available for hydrogen bonding in Pro-containing

Figure 13.1

530

Ferrocenes: Ligands, Materials and Biomolecules

peptides. Consequently, fc(CO-Pro-OMe)2 (107 in Scheme 13.25) does not form any intra-molecular hydrogen bonds. Accordingly, the CD spectrum of 107 does not give any indication of an ordered structure in solution. It is noteworthy that the compound fc(CO-Gly-Pro-OEt)2 (108) has the same type of arrangement as the above-mentioned Ala−Pro derivatives 106 in the solid state as well as in solution.110 Moriuchi, Nagai and Hirao recently reported the synthesis and structural characterisation of fc(COPro-Ala-NHpy)2 derivatives (109 in Scheme 13.25), where NHPy is the amide of 2-aminopyridine. In these compounds, the Pro residue is closest to the ferrocene core and hence no NH proton for Herrick-type hydrogen bonding is available. Instead, a ten-membered ring is formed via a hydrogen bond between the Cp–CO carbonyl and the Ala−NH on the other ring. The chirality of the metallocene (and hence the sign of the CD signal in the ferrocene region around 480 nm) is determined by the chirality of the ﬁrst amino acid (Pro) on the ferrocene, making 109a similar to 109c. Likewise, 109b and d show the same metallocene chirality, but are mirror images to the other pair. The chirality of the second amino acid (Ala) in 109 does not inﬂuence the core chirality but rather determines the direction of the NHpy groups, which are bent towards the ferrocene for derivatives a and b, or pointing away from it in c and d. The work of Hirao’s group has been summarised elsewhere.144–146 O N

O

O

Fe

O

O N

OMe Fe

N H

O O

MeO N

H N

N 107

O

H N N

O O

N N H

109 a: L-Pro-L-Ala b: D-Pro-D-Ala c: L-Pro-D-Ala d: D-Pro-L-Ala

Scheme 13.25

In addition to the above-mentioned compounds, which possess one NH group per peptide strand, peptide derivatives with two NH amide moieties have been prepared. These include fc(CO-Gly-Leu-OEt)2 (110), fc(CO-Gly-Phe-OEt)2 (111)111 and fc(COAla-Phe-OMe)2 (112).86 The amide NH moieties of the second amino acid are not oriented in such a way that allows for the formation of intramolecular hydrogen bonds but, instead, they are involved in intermolecular hydrogen bond interactions. The strength of the hydrogen bonds in dichloromethane and trichloromethane solution of 112 was investigated in relation to that of its positively charged cobaltocenium analogue.86 The 57 Fe M¨ossbauer spectroscopic parameters of several mono- and disubstituted peptide ferrocene derivatives were compared to those of ferrocene.86, 147 While the isomer shifts for all compounds are very similar, the values for the quadrupole splitting

The Bioorganometallic Chemistry of Ferrocene

531

(EQ ) vary in the following order: FcH (EQ = 2.42 mm s−1 )97, 147 > ferrocenoyl derivatives (such as 56 and 57, EQ = 2.32 mm s−1 ) > 1,1 -disubstituted derivatives (such as fc(CO-Phe-OMe)2 and 112, EQ = 2.27 mm s−1 ).96 Ferrocene peptides may also serve as ligands to other transition metal fragments. Hirao and coworkers prepared the compound fc(CO-Ala-Pro-NHPy)2 (113), where NHPy is the amide of 2-amino-pyridine (see also Scheme 13.25). This compound was subsequently transformed into trans-[PdCl2 (113)] (114), in which the pyridine nitrogen atoms coordinate to the palladium atom.148 X-ray crystal structures for both compounds were reported. Both compounds 113 and 114 adopt the ‘Herrick conformation’ in the solid state as well as in solution, with the Ala−NH group being involved in intramolecular hydrogen interactions with the Ala−CO moiety of another strand. The Metzler-Nolte group has used ferrocene peptide conjugates with sulfurcontaining amino acids (Met and Cys) as ligands for metal carbonyl fragments. Following the structural characterisation of Met-containing ferrocene-1,1-dicarboxylic acid derivatives,149, 150 Cys-containing derivatives were investigated in more detail. Reaction of 19 with protected cysteine yields 115 which, after deprotection, reacts with [Fe2 (CO)9 ] to yield the bis-cysteine bridged Fe2 (CO)6 cluster 116 Scheme 13.26. Although known for quite some time, there is recently renewed interest in thiolate iron carbonyls as model compounds for hydrogenase enzymes. Compound 116 is the ﬁrst model complex of this kind that combines the iron–carbonyl core with an electron relay in the form of ferrocene. This compound has been thoroughly characterised by spectroscopic methods, including IR spectro-electrochemistry. Interestingly, oxidation of the iron carbonyl moiety has a signiﬁcant inﬂuence on the carbonyl bands of neighbouring peptide bonds, whereas the oxidation of the ferrocene core has no such effect.151 To fully understand the properties and requirements for ferrocene peptides as model systems for the hydrogenase enzyme family, extensive molecular modelling was carried out. To this end, a new force ﬁeld was developed and implemented in the CHARMM molecular modelling program package in collaboration with the group of Jeremy Smith in Heidelberg. The parameter set for ferrocene was derived from automatic frequency matching with the results from DFT calculations and extensively tested against known X-ray crystal structures. It was then applied to small cysteinecontaining ferrocene oligopeptides which may serve as minimal model systems for the active site of Ni−Fe hydrogenase enzymes.152 An important point is that an ordered structure will only occur if the terminal amino groups of the amino acids and/or peptides form amides with a carboxylic acid O

O H N fc(CO-Cys-Trt-OMe)2

Fe

O

O

SH

O

SH N H

115

H N + Fe2(CO)9 − 3 CO

Scheme 13.26

O

S

O

S

Fe

O O

O

N H 116

O O

CO Fe Fe CO

CO CO CO CO

532

Ferrocenes: Ligands, Materials and Biomolecules

N

OEt

O N

O

O

O

O

O

O

H

NH

HN O

Fe

N

O

O O

O

N

NH

HN

O

O

117

NH

Fe

HN H

O

O OEt

O

O

O

118

O

Scheme 13.27

that is directly attached to the cyclopentadienyl ring. The groups of Hirao and Beck prepared compounds 117 and 118 with different linkers (Scheme 13.27), and they found that these do not exhibit an intramolecular hydrogen-bonded conformation in solution.111, 153 Han et al. prepared macrocyclic compounds from fc(COCl)2 (119) and pseudocystine derivatives with various linkers between the sulfur atoms.154 The synthesis yielded ferrocenophane derivatives 120 along with cyclodimers 121 (Scheme 13.28). The redox potential for the ferrocene/ferrocenium couple of 120 depends on the nature MeO O Cl Fe

O O

HN CO2Me +

CO2Me

H2NCHCH2SRSCH2CHNH2

S

O O

Fe

R S

Cl

HN R = (CH2)2, (CH2)4, p-CH2-C6H4-CH2

O

119

120

MeO +

MeO

OMe O R

HN Fe

O O

S

S

S R

O

Scheme 13.28

MeO

NH

S

HN 121

O O O NH O OMe

Fe

The Bioorganometallic Chemistry of Ferrocene

533

of the anion. In particular, the addition of ﬂuoride resulted in a pronounced potential shift of about 120–150 mV to more positive values. From the occurrence of this shift, it is anticipated that halide ions occupy a position in the binding pocket of the ferrocenophane. In a subsequent report,155 cystine and cystine di- and tripeptides [(AA)n -Cys]2 (AA = Gly, Ala, Leu, Met, Pro; n = 0, 1, 2) were synthesised and reacted with 119 to form ferrocenophanes with cavities of different size, capable of accommodating alkaline and alkaline earth cations. CD spectroscopy and cyclic voltammetry were used to determine binding constants. For one derivative, a high binding afﬁnity for Mg2+ and Ca2+ , along with a high preference for Ca2+ over K+ , was observed, which was even better than the values for the natural ionophore valinomycin. Recently, other macrocyclic amino acid cystamine derivatives fc(CO-Aaa-NHC2H4 S)2 were reported (Aaa = Ala, Val, Leu).156 These compounds show strong intramolecular hydrogen bonding and structures related to 103 as discussed above. Their electrochemical properties were also investigated in solution as well as in an immobilised form.157 Recently, Kraatz and coworkers prepared compounds of the general formula fc(COPron -OBn)2 (n = 1–4).158 As observed for monosubstituted analogues FcCO-Pron OBn,94 the proline chains of the 1,1 -disubstituted compounds display the left-handed helical arrangement characteristic for polyproline II. For fc(CO-(Pro)2 -OBn)2 , the ordered conformation adapted by the other 1,1 -disubstituted derivatives presented above is not observed, because the Pro-amide linkages lack NH moieties. The redox potential of the ferrocene/ferrocenium couple was found to shift to a small extent to less positive values on going from fc(CO-Pro-OBn)2 to fc(CO-Pro4 -OBn)2 , a trend similarly to that observed for the monosubstituted derivatives.94 Interestingly, it was reported that the synthesis of fc(CO-Pron -OBn)2 with n = 2, 3 or 4, yielded a second class of compounds, i.e. the monosubstituted derivatives fc(CO-Pron -OBn)(COOBt).158 These were further reacted with a pepstatin analogue to obtain the conjugates 122 (Scheme 13.29).159 Pepstatin is a potent naturally occurring inhibitor of aspartic proteases, including HIV-1 protease and pepsin, and conjugates 122 were tested for their ability to inhibit such enzymes.159–162

HO

H N

O

O

Fe

O

O N H

H N

OH

O

O N H

H N

OH

O OH

O

N n

122, n = 3 (a) or 4 (b)

Scheme 13.29

Sekine and coworkers used ferrocene as a scaffold for pyrrole-imidazole polyamides.163 These molecules form hairpin-type structures and were designed for selective DNA binding to GCG/CGC sequences. CD titration expriments gave binding constants in the sub-micromolar range, indicating their potential as redox-active gene sensors. 1 -Aminoferrocene-1-carboxylic acid (Fca, 18) constitutes one of the simplest ferrocene-based amino acids. Its synthesis according to Heinze and coworkers is

534

Ferrocenes: Ligands, Materials and Biomolecules NH2 Fe

O

NH2 • HCl Fe

OH 18

O

NHAc Fe

NMe2 52

O

NH-Boc Fe

OMe 123

O

NH-Fmoc Fe

OH 124

O OH

125

Scheme 13.30

presented in Scheme 13.4 (Section 13.2.1). Several protected derivatives of 18 were also prepared. The X-ray single crystal structures of the N -acetyl methyl ester (AcNHfcCO2 Me, 123)185 and N -Boc protected acid (BocNHfcCO2 H, 52) were published.186 Some of the reported Fca derivatives are shown in Scheme 13.30. Ferrocene-1,1 -diamine (Faa) is another interesting starting material which was recently made available for amino acid and peptide conjugates.177, 183, 187 Starting from 123, a number of amino acid and peptide derivatives were prepared in which 18 serves as one amino acid (Table 13.6). An extensive structural and spectroscopic collaborative study was recently published.177 In collaboration with Rapic’ group, the Metzler-Nolte group has published the ﬁrst Fca-oligopeptide derivative, Boc-Ala-Fca-Ala-Ala-OMe (126).178 Peptide 126 was prepared in solution by modiﬁed peptide coupling reactions and puriﬁed by thin layer chromatography. As revealed by X-ray crystallography, the conﬁguration of the peptide strands is stabilised by two intra-molecular hydrogen bonds, which enforce a P -helical conformation at the metallocene. CD spectroscopy indicates that this conformation is preserved even in solution. By using a combination of Boc- and Fmoc solid phase synthesis, the octapeptide Ac-Val-Gly-Ala-Fca-Ala-Gly-Val-Leu-NH2 (127) was prepared from 52. According to CD spectroscopy, this peptide shows an ordered helical structure in solution which is stabilised by intramolecular hydrogen bonding.179 Heinze and coworkers later reported the synthesis of tripeptides containing Fca by SPPS methods using the Fmoc-derivative 125.176 13.2.4

Stereochemical Considerations

One intriguing aspect of the ferrocene–peptide conjugates described above is their stereochemistry. We have classiﬁed all possible stereoisomers, and suggested a general nomenclature which deﬁnes all possible conformations for this class of compounds.187 As already noted, there is only a very small barrier of rotation for the two Cp rings against each other. This feature has led to the description of ferrocene as a molecular ball bearing.188 This also means that 1,n -disubstituted ferrocenes may exist in a great number of conformations, which are roughly classiﬁed into ﬁve categories from 1,1 to 1,5 , depending on the relative positions of the two substituents (Scheme 13.31). Two of those are pairs of helically chiral enantiomers, which can be described by the usual P /M nomenclature. If there are sufﬁciently strong intramolecular (interstrand) hydrogen bonds, then a single isomer can be isolated, and the deﬁnition of conﬁgurational isomers applies. If, as in the above examples, enantiomerically pure amino acids are attached to the ferrocene core by peptide bonds, then in fact the chirality of the amino

The Bioorganometallic Chemistry of Ferrocene

ω 1,1': 36° < ω < −36° 1,2': 36° < ω < 108° 1,3': 108° < ω < 180° 1,4': 180° < ω < −108° = 1,3' 1,5': −108° < ω < −36° = 1,2'

535

1,1'

1,2' (P helical)

Enantiomers

1,5' (M helical)

1,3' (P helical)

Enantiomers

1,4' (M helical)

Scheme 13.31

L

diastereomers

Fe

L,P-1,2',L

L

Fe L

L L,M-1,2',L

enantiomers diastereomers

D D,P-1,2',D

diastereomers

diastereomers

Fe D

Fe D

D D,M-1,2',D

Scheme 13.32

acids will determine the helical chirality of the ferrocene. For example, in all cases of a ‘Herrick conformation’, L amino acids will induce P chirality at the metallocene. All conceivable isomers and their stereochemical relation are also shown in Scheme 13.32. The interesting feature here is that the helical chirality is ‘locked in’ by a self-assembly process, solely based on the amino acid chirality. The energy difference between the (preferred) L,P -1,2 ,L isomer and its less favoured L,M-1,2 ,L diastereomer has been calculated to be around 17 kJ mol−1 in the case of the ‘Herrick conformations’.150 This compound also exempliﬁes the order and assignment of stereochemical denominators that we use. At this point, an interesting difference is observed between derivatives of fc(CO2 H)2 (Fcc, 19) and Fca (18). While both amino acids have equal inﬂuence on the helical chirality of the ferrocene core in the Fcc

536

Ferrocenes: Ligands, Materials and Biomolecules

Mq / deg mM−1 cm−1

case,165 only the amino acid attached to the N -terminus of Fca determines the helical chirality of the ferrocene core in this class of compounds.177 This ﬁnding, which is readily deduced from the CD spectra of suitably selected compounds, can be rationalised on the basis of the available X-ray structures. This point will be discussed below with relation to structures in naturally occurring peptides. CD spectroscopy is a highly valuable tool for stereochemical analysis of ferrocene peptides,187 the induced Cotton effect in the region of ferrocene-based transitions around 480 nm being particularly useful. Depending on the helical chirality of the metallocene, this band is either positive (P helical) or negative (M helical), regardless of the ferrocene derivative (Fcc, Fca, Faa, or others) stereochemistry. The intensity of this band also gives an indication whether a strong hydrogen bonding pattern such as the ‘Herrick conformation’ exists or not. As an example, four CD spectra from our own work are shown in Figure 13.2. For the Fcc derivatives fc(CO-Ala-OMe)2 (128) with ‘Herrick’ type hydrogen bonding, bands of the same strong intensity but opposite sign are observed. Positive bands arise from L amino acid derivatives which induce P helical chirality in the ferrocene moiety (L-Ala-P -Fcc-L-Ala, 128a), and a negative band is observed for the D-Ala-M-Fcc-D-Ala derivative (128b). For fc(COPro-OMe)2 (107), which has only proline residues and thus no possibility for hydrogen

l / nm

Figure 13.2 CD spectra of 128a, 128b, 107 and 129 illustrating the different shapes and intensities of the peptide and ferrocene-based (around 480 nm) bands (See text for compound numbers and explanations)

The Bioorganometallic Chemistry of Ferrocene

537

bonding, a very weak band is observed. The residual intensity of this band is probably due to the presence of the chiral amino acids, which makes this compound overall chiral, although the open 1,4 conformation (ω = 123◦ ) is observed in the solid state.158 Up to this point, only compounds with same amino acid chirality were considered. This raises the interesting question: which helical conformer will be observed if two amino acids of different chirality are attached to the metallocene? The CD spectrum of fc(CO-D-Phe-OMe)(CO-L-Ala-OMe) (129) is also shown in Figure 13.2. It has a much weaker, negative band around 480 nm, probably resulting from a mixture of M and P ‘Herrick conformers’ in solution with a slightly higher inducing power of the (sterically more demanding) D-Phe residue.165 The structures discussed above resemble turn structures in peptides. They do, however, provide a much greater structural variety. Two principally different types of turn structures in peptide chemistry are shown in Scheme 13.33. They differ in their hydrogen bonding patterns. If the carbonyl group is closer to the peptide N -terminus than the hydrogen-bonded amide proton, the direction is deﬁned as N → C (Scheme 13.33, left). Depending on the size of the hydrogen bonded ring, γ -turns (7-membered ring) and β-turns (10-membered ring) form, whereas 13-membered rings are present in αhelices. The ‘reverse turns’ (C → N, Scheme 13.33, right) rvs-γ (5-membered ring), rvs-β (8-membered rings) and rvs-α (11-membered ring) are energetically disfavoured and rarely observed in protein structures.189 i+2

i+2

O i+3

O i+1 HN

N N H H β γ H N O α

i

O O i+1 i+4

O i+3

N N H βH γ NH α O

C-terminus

N-terminus

C-terminus

N-terminus N

C

N

C

Scheme 13.33 Helical chirality of 1,1 -disubstituted ferrocenes, deﬁning 1,n substitution pattern (1, 1 to 1,5 ) and stereochemical relation between enantiomeric pairs

An additional complexity occurs because the two peptide strands in ferrocene turn mimetics may be aligned parallel and anti-parallel (Scheme 13.23). Hence, the turn classiﬁcation has to be extended. Classical parameters such as torsion angles may not be meaningful in the ferrocene peptides discussed here. Therefore, we propose to use a modiﬁed nomenclature based on geometrical considerations. For example, two 10membered rings are formed in the Herrick conformation shown in Scheme 13.24 if the two ipso-C atoms of the Cp rings are thought to be directly connected (Scheme 13.34). Also, if each carbonyl group is counted as one whole amino acid, then the carbonyl group hydrogen-binds to the NH proton of the i+3 ‘amino acid’. Both 10membered ring and i to i+3 hydrogen bonding classify this kind of hydrogen bonding

538

Ferrocenes: Ligands, Materials and Biomolecules O

Me MeO

7

9

N 10H O

O

1

O

Fe

8

6 5

HN4

O

3 2

OMe

Me

Stereochemistry: P-1,2' (ferrocene chirality) L,L (amino acid chirality) H-bond pattern: two 10-membered rings (b-turn)

Fe 5

6

Me 3 4 O N N7 H H 2 8 Me Ala O 1 O NH OMe OtBu

Stereochemistry: P-1,2' (ferrocene chirality) L,L,L (amino acid chirality) H-bond pattern: 10-membered ring (b-turn) and 8-membered ring (rvs-b-turn)

Scheme 13.34 Schematic relation of amino acid chirality (denoted D/L) and helical chirality of ferrocene core (denoted P/M)

as β-turn-like. In contrast, the Fca tetrapeptide 126 (Scheme 13.34, right) has a β-turn (10-membered ring) and a rvs-β-turn (8-membered ring) in one and the same molecule, provided we count Fca as one normal amino acid, i.e. three atoms.178 13.2.5

Applications

Galka and Kraatz reported a series of disulﬁde-bridged prolyl cystamine derivatives of the general formula fc(CO-(Pro)n -NHC2 H4 S)2 , where n = 0–6.108 Self-assembled monolayers on a gold electrode were prepared from these compounds, and the electron transfer properties between the ferrocenoyl moiety and the surface of the gold micro-electrode were investigated. A through-bond mechanism of the electron transfer process was suggested on the basis of a signiﬁcant deviation from Marcus-type behaviour, although kET showed a distance-dependence. Kraatz et al. also prepared the glycyl cystamine derivative fc(CO-Gly-NHC2 H4 S)2 .107 This compound crystallises in a supramolecular helical assembly consisting of two different helices. It is very interesting to note that the high ﬂexibility of 1,1 -disubstituted ferrocene acid derivatives, along with their propensity to form strong hydrogen bonds, has been exploited well before the work of Herrick et al. In their work on antibodies that catalyse the Diels–Alder cycloaddition reaction, Janda and coworkers used Fcc derivatives as antigens. They did in fact also report the ﬁrst synthesis of a derivative of Fca, as mentioned above.190 As a spectacular result, the X-ray single crystal structure of the catalytic antibody 13G5 was solved alone and in a complex with hydrogen-bonded 1carboxy-1 -(dimethylcarbamoyl)ferrocene (130).191 The overall view of this antibody with bound 130 is shown in Figure 13.3. Only the P-helical conformation is observed in the antibody-bound ferrocene compound, which is held in place only by hydrogen bonding. The idea of helical chirality (and the ease of interconversion between the two enantiomers) of 1,1 -disubstituted ferrocene compounds was thus used and also implicitly mentioned by Janda and coworkers. Kraatz’s group has recently shown that ferrocene-containing peptides may serve as collagen mimics.192 Compounds of the formula FcCO-(Pro-Hyp-Gly)n -Cys (n = 4, 6–9) were synthesised by solid phase synthesis. CD spectrosocpic studies show that these compounds form collagen-like triple helices. Addition of the N -terminal ferrocene group increases the thermal stability of the collagen mimics. The same group has also structurally characterised the ﬁrst synthetic model for β-barrel structures,

The Bioorganometallic Chemistry of Ferrocene

539

Figure 13.3 X-ray structure of the catalytic antibody 13G5 with bound ferrocene derivative 133 (PDB code 1A3L) (Graphics were prepared with the iMol program on a MacBook Pro)

which form by self-assembly of cyclopeptides fc(CO-Gly-Val-CSA)2 (131) and fc(COGly-Ile-CSA)2 (132, CSA = cysteamine).171 The β-barrel model has eight β-strands arranged almost parallel to the axis of the channel with an internal pore diameter of ˚ ca. 8 A.

13.3

Conjugates of Ferrocene with Proteins

Salmain and Jaouen recently published a review article on the side-chain selective and covalent labelling of proteins with transition organometallic complexes.193 Their article, however, is more concerned with organometallic labels different from ferrocene and their side chain selectivity. The main impetus for ferrocene labelling of proteins was the application in electrochemical biosensors. There are numerous examples for labelling of the protein itself, derivatives thereof, or cofactors such as heme groups or biotin. We will concentrate more on the different biosensor formats that were used in this section. 13.3.1

Glucose Oxidase and Other Redox Proteins

Glucose oxidase (GOD) from Aspergillus niger is a 150–180 kD dimeric glycoprotein with the carbohydrate part of the enzyme constituting about 16–24 % of the molecular

540

Ferrocenes: Ligands, Materials and Biomolecules

weight.194 The enzyme contains two tightly bound FAD cofactors, which are essential for activity.195 Dissociation of the two subunits only occurs under strongly denaturating conditions (SDS) and is accompanied by the loss of the FAD cofactors.196 The sequence of the 583 amino acid protein has been determined for the cloned Aspergillus niger gene,197, 198 and the crystal structure of the partially deglycosylated enzyme has been ˚ resolution.196, 199 The enzyme possesses 32 NH2 groups (15 lysine solved to 2.3 A units per monomer and two N-terminal amino groups).197, 198, 200 GOD catalyses the oxidation of β-D-glucose to δ-D-gluconolactone via a twoelectron process (Equation (13.1)). The FAD cofactor serves as the electron donor and is reduced to FADH2 . In the next step, FADH2 is regenerated by oxidation with dioxygen, yielding hydrogen peroxide (Equation (13.2)): GOD(FAD) + β-D-glucose −→ GOD(FADH2 ) + δ-D-gluconolactone (13.1) GOD(FADH2 ) + O2 −→ GOD(FAD) + H2 O2

(13.2)

Frequent monitoring of the glucose level in blood is essential for patients suffering from diabetes, hence there is a need for simple devices for its monitoring. The development of biosensor devices for this application has been recently reviewed by Wang12 and appears as one of the very successful applications of ferrocene in life sciences. As early as 1984, Cass and coworkers established the basis for an oxygen-insensitive amperometric glucose biosensor using GOD immobilised on an electrode and fcMe2 as redox mediator (Figure 13.4, conﬁguration A). Rather than being regenerated by oxygen, GOD(FADH2 ) is reoxidised into GOD(FAD) by the oxidised ferrocenium form of the mediator according to Equation (13.3).201 GOD(FADH2 ) + 2Fc+ −→ GOD(FAD) + 2Fc + 2H+

(13.3)

The mediator in its reduced form (ferrocene) is then reoxidised at the electrode giving rise to a detectable current (Equation 13.4). 2Fc − 2e− −→ 2Fc+

(13.4)

One of the drawbacks of this setup is that the diffusional mediator could leak out, leading to biosensor deterioration. This is why alternative biosensor conﬁgurations (numbered B, C and D in Figure 13.4) have been proposed and will be mentioned all along this chapter. In the mid 1980s, the research groups of Hill, Heller and Bartlett independently suggested conjugating the redox mediator to GOD200, 202 to construct glucose biosensors of conﬁguration B in Figure 13.4. Several additional reports by the groups of Bartlett203 , Heller204–207 and others208–211 followed in the next decade. The rationale behind the derivatisation of GOD was to establish direct electrochemical communication between the enzyme and the electrode as the FAD cofactors are deeply buried in the core of the enzyme with the surrounding part of the protein acting as an insulating shell. Acids 48 and FcCH2 CO2 H (133) were mainly employed as labelling reagents, in combination with the activating agent EDC. Coupling reactions were performed under various conditions, resulting in conjugates with varying Fc-to-GOD ratios and variable enzymatic activities; an overview is given in Table 13.7. The reported results are

The Bioorganometallic Chemistry of Ferrocene

541

Figure 13.4 Amperometric glucose biosensor conﬁgurations: (A) the presence of a diffusional (1) or non diffusional (2) mediator; (B) covalent tethering of the mediator to GOD; (C) coimmobilisation of GOD with a redox polymer; (D) covalent tethering of the mediator to biotin (1a), avidin (Av) (1b) or antibody (2)

542

Ferrocenes: Ligands, Materials and Biomolecules Table 13.7 Overview of ferrocene derivatives of GOD

Entry

Ferrocene derivative

Coupling reagent

Denaturing conditions

Fc:GOD ratio

Ref.

1

48

IBCF

none

8

200

2

48

EDC

none

8.5

210

3

48

EDC

3 M urea

12 ± 1

4

48

EDC

3 M urea

13

5

48

EDC

3 M urea

3.6, 8.5, 13, 29 ± 3a

215

6

48

EDC

3 M urea

18.2

210

7

48

EDC

3 M urea

9.2

218, 219

8

48

EDC

400 MPa

7.4

210

9

48

EDC/HOSu

none

5 ± 1, 12a

215

5

211 215

202, 204–206 203, 229

10

48

EDC/HOSu

noneb

11

19

EDC/HOSu

none

7.5

12

133

EDC

3 M urea

13 ± 1

204–206

13

133

EDC

3 M urea

22

203, 229

14

133

EDC/HOSu

none

10

215

15

Fc(CH2 )3 COOH

EDC

3 M urea

29

229

16

134

EDC/HOSu

none

21

209, 226

17

134

EDC/HOSu

none

24

224, 225, 227

18

32

NaBH4 c

none

4.2

210

32

NaBH4

c

3 M urea

8.1

210

NaBH4

c

none

27.5

210

19 20

32

a Various concentrations of the acids resulted in conjugates with a different Fc/GOD ratio. b Performed in inverse micelles of the protein in a H2 O:MeCN ratio of 1:75. c NaBH4 is not a coupling reagent, but a reducing reagent to transform the labile imine into a secondary amine.

somewhat contradictive, because derivatisation performed under apparently identical conditions but by different groups led to conjugates with different Fc-to-GOD ratios. This is exempliﬁed by comparing entries 3–7 as well as entries 12 and 13 in Table 13.7. In addition, the derivatisation of GOD with N -(ferrocenylmethyl)aminocaproic acid, (FcCH2 NH(CH2 )5 CO2 H, 134) by Willner, Katz and coworkers led to conjugates with Fc-to-GOD ratios of 21 or 24 (entries 16 and 17). In other reports, the Fc-to-GOD ratio was not given.13, 212 For the derivatisation, it was found that denaturation of the protein in the presence of 3 M urea results in a signiﬁcant loss of activity of the reconstituted enzyme, even when no 48 was added.208, 211 More suitable conditions for obtaining conjugates with higher catalytic activity required activation of the carboxylic acid groups with EDC and HOSu, without addition of urea. Labelling of GOD with 133 turned out to be more

The Bioorganometallic Chemistry of Ferrocene

543

favourable with respect to derivatisation with 48, because the resulting conjugates exhibited higher enzyme activity. Coupling of ferrocenyl redox species to GOD was also done during the immobilisation process on the electrode. A non-organised multilayer of ferrocene-tethered GOD was constructed on the surface of a gold electrode by cross-linking GOD and 2-aminoethylferrocene with glutaraldehyde.213 Alternatively, a multilayered network of GOD was assembled in a step-wise manner on a gold electrode by making use of the cross-linking agent DIDS (4,4 -diisothiocyanato-trans-stilbene-2,2 -disulfonic acid disodium salt). Ferrocene groups were then introduced into this network by reaction of 134 in the presence of EDC, HOSu and 1 M urea.214 Both biosensor conﬁgurations gave a good amperometric response to glucose (Figure 13.4, conﬁguration B2). However, for the design of reagentless glucose sensors, these GOD–Fc conjugates did not exhibit satisfying properties. First of all, repeated electrochemical cycles were shown to bring about a gradual decrease in the peak current, which can be interpreted by slow decomposition of the covalently attached redox centers. The second unsatisfying fact was revealed by the work by Mikkelsen, English and coworkers, who determined the rates for intramolecular electron transfer for some of the conjugates listed in Table 13.7.215 Despite the difference in the number of attached ferrocene moieties, the values for electron transfer rate between the conjugates and the electrode span a narrow range between 0.16 and 0.90 s−1 . This suggests that the location of the ferrocenyl groups rather than their number is rate-determining,215 in accordance with the Marcus electron transfer theory.216, 217 An analysis of the recently determined crystal structure of GOD showed that the separations between the lysines and the FAD cofac˚ 199 These two drawbacks make clear that an alternative approach is tor were all >23 A. necessary towards electrochemically wired GOD derivatives with excellent mediation between the FAD cofactor and the electrode. Heller and coworkers tried to improve the electronic communication between the FAD cofactor and the electrode by using ferrocenyl derivatives with a long alkyl chain as labelling reagents, and, in addition, these were attached to the carbohydrate part instead of the protein part of the enzyme.207 In the ﬁrst step, GOD was oxidised with NaIO4 to yield an average of 6.4 aldehyde groups on the outer surface of the enzyme. Subsequently, methylamino-ferrocenyl residues were introduced by reacting with amines 135 (n = 2, 3, 4, 6, 8, 10) and subsequent NaBH4 reduction (Scheme 13.35). By coulometric analysis, the number of introduced ferrocene moieties was estimated to be around one for each derivative. The length of the spacer turned out to play a crucial role for the mechanism of electric communication between the enzyme and the electrode. When attached to longer chains of more than 10 bonds, the ferrocenyl moiety can penetrate the enzyme to a sufﬁcient depth to allow fast intramolecular electron transfer between the ﬂavin centers and the electrode. When attached to short chains of less than ﬁve bonds, on the other hand, the electron transfer process was primarily intermolecular, with another modiﬁed GOD molecule acting as a conventional diffusing mediator.207 The results obtained with short chains are consistent with those obtained by derivatisation with various (short chain) carboxylic acid derivatives presented above. A few years later, Karube et al. also introduced aldehyde groups on the sugar residues of GOD and, subsequently, connected ferrocenyl moieties on adipic acid

544

Ferrocenes: Ligands, Materials and Biomolecules H

NaIO4

O 1) FcCH2NH(CH2)nNH2, 135

O

n = 2, 3, 4, 6, 8, 10

GOD

GOD

O

H

2) NaBH4

H CH2-N-(CH2)n-N-CH2– H H

Fe

CH2-N-(CH2)n-N-CH2 H H

Fe

GOD CH2-N-(CH2)n-N-CH2H H

Fe

Scheme 13.35

hydrazide spacers to the enzyme.218–220 Also in these cases, direct electrical communication between the enzyme and the electrode took place. In addition, a mixture of the modiﬁed GOD derivative with organic solvent was used as the ink component for the screen-printing technique. Katz, Willner and coworkers devised a different strategy for decreasing the distance between the FAD cofactor and the redox active ferrocene moiety. Instead of attaching ferrocene moieties randomly on the periphery of the enzyme, they reconstituted GOD from apo–GOD in the presence of a semi-synthetic FAD cofactor 136 to which a tail bearing a ferrocenyl moiety was tethered (Scheme 13.36).221, 222 By performing electrochemical and kinetic analyses, the rate for electron transfer between the modiﬁed enzyme and the electrode was found to be 40 s−1 .221, 222 This value is about 45 times higher than the highest observed rate constant for GOD randomly substituted on the periphery with ferrocene moieties (k = 0.9 s−1 ). Clearly, the enzyme reconstituted with the ferrocenyl-modiﬁed FAD cofactor exhibits superior electrical communication with respect to the randomly derivatised conjugates. Chen et al. modiﬁed GOD on the gene level to attach a Lys10 tail via a (SerGly)5 peptide linker to the C-terminus and expressed the recombinant protein in the yeast Pichia pastroris.223 Because translation proceeds from the N - to C-terminus, it was anticipated that the Lys10 tail would not be buried in the core of enzyme. In addition, the X-ray crystal structure of native GOD showed the C-terminus to be located at the surface of the wild-type (wt) protein, whereas the N -terminus is buried deeply in the core of the protein.199 Finally, the added Lys residues are hydrophilic and should prefer a position on the surface over a position in the hydrophobic core of the protein. Indeed, these lysine-NH2 moieties were found to be accessible for labelling. After derivatisation with 48 and EDC in 2 M urea, the modiﬁed Lys10 -GOD derivative displayed 90 % activity relative to wt-GOD. Furthermore, the Lys10 -GODFc conjugate was found to have a better electrical availability and a higher lifetime than the ferrocenyl-conjugated GOD controls lacking the Lys10 tail.223

The Bioorganometallic Chemistry of Ferrocene

545

O H 3C

N

H 3C

N

N H H H H H

C C C C C

N

H O

H OH OH OH H

HN

HO H CH2 -CH2-N-C-(CH2)5-N-CH2

O O O P O P O O− O−

Fe

N

N N

N

O OH OH

136

CONH2 O −O P O

+

N O OH OH

O

HN N

−O

P O O

N O

R N

R= Fc(CH2)nC(O)CH2 –

N

(144; n = 0: a; n = 2: b; n = 3: c) Fc(CH2)2NHC(O)CH2CH2C(O)CH2 –

OH OH

(145)

Scheme 13.36 Ferrocene-modiﬁed cofactors used for bioassays (top: FAD, bottom: NAD)

The group of Katz and Willner adapted ferrocenyl-modiﬁed GOD to probe antigen–antibody associations at the surface of an electrode to lead to a model amperometric immunosensor.13, 209, 212, 224–227 The principle of this immunosensor is outlined schematically in Figure 13.5. In the absence of antibody, Fc-GOD can easily exchange electrons with the electrode. Formation of interfacial antigen–antibody complexes gradually hinders electron transfer, yielding a reduction of the bioelectrocatalytic current. The ﬁeld of bioelectronics has been reviewed by this group.14, 206 Aizawa et al. prepared GOD derivatives that were conjugated with both ferrocenyl moieties and digoxin.228 The authors used two different procedures for ferrocenyl labelling: the 48/EDC route and labelling via imine formation with 32 followed by reduction. Conjugates with various Fc/digoxin/GOD ratios were prepared, and these were adapted to a homogeneous electro-enzymatic immunoassay. D-Amino Acid Oxidase (DAAO) was one of the ﬁrst ﬂavoproteins to be discovered in the 1930s.230, 231 The enzyme is found in numerous organisms, such as yeast, insects, birds and mammals.232 It catalyses the oxidative deamination of D-amino

546

Ferrocenes: Ligands, Materials and Biomolecules

Figure 13.5 Principle of amperometric immunosensor for DNP based on inhibition of electron transfer from/to GOD in the presence of speciﬁc anti-DNP antibody

acids according to Equations (13.5) to (13.7). DAAO(FAD) + H2 NCH(R)CO2 H −→ DAAO(FADH2 ) + HN=C(R)CO2 H (13.5) HN=CH(R)CO2 OH + H2 O −→ RC(O)CO2 H DAAO(FADH2 ) + O2 −→ DAAO(FAD) + H2 O2

(13.6) (13.7)

In the ﬁrst step, the amino acid is reduced to an imine, which subsequently hydrolyses to yield a ketoester. Similar to the regeneration of glucose oxidase, FADH2 is regenerated by dioxygen to return the DAAO in the FAD-state. In vitro, DAAO is reactive towards a wide range of neutral and basic D-amino acids, but the highest reactivity is displayed towards amino acids that possess a hydrophobic side chain. The amino acids D-aspartic acid and D-glutamic acid are no substrates for DAAO, but are instead oxidatively deaminated by D-aspartate oxidase. A detailed review on DAAO recently appeared in the literature.233 D-Amino Acid Oxidase from pig kidney (pkDAAO) was the ﬁrst DAAO to be obtained in pure form in 1973.234 About 15–20 years later, the isolation and puriﬁcation of DAAO from the yeasts Rhodoturula gracilis (rgDAAO) and Trigonopsis variabilis (tvDAAO) were reported.235, 236 Mammalian pkDAAO is a monomeric 347 amino acid protein, containing one FAD cofactor, with a molecular mass of 39.6 kDa whereas rgDAAO is a homodimeric enzyme of approximately 80 kDa, with each monomer consisting of 368 amino acids. Furthermore, the rgDAAO contains a noncovalently bound FAD cofactor.237 Protein X-ray crystal structures of pkDAAO238–242 and rgDAAO243, 244 have been reported. Heller and Degani could bind 3 ± 1 ferrocenoyl moieties on pkDAAO by reaction with 48 in the presence EDC and 3 M urea.204, 206 The conjugate showed about 25 % activity relative to the native enzyme. Katz, Willner and coworkers reconstituted apo–pkDAAO in the presence of the ferrocenyl-modiﬁed cofactor 136. The reconstituted enzyme was reported to display only about 20 % of activity relative to the native pkDAAO.222 For both conjugates, it was demonstrated that direct electrochemical communication took place between the electrode and the modiﬁed enzyme. The addition of D-alanine to either conjugate resulted in an electrochemical response, with the magnitude of the current depending on the concentration of D-alanine. A linear

The Bioorganometallic Chemistry of Ferrocene

547

relationship between the D-alanine concentration and the current allows for use of the modiﬁed enzymes in amperometric D-amino acid sensors.204, 206, 222 Cytochrome P450cam (CyP450cam ) from the soil bacterium Pseudomonas putida is a monooxigenase with a molecular mass of 45 kD, which catalyses the stereospeciﬁc hydroxylation of camphor to 5-exo-hydroxycamphor.245 For this transformation, it relies on two redox partners that transfer electrons from NADH to CyP450cam , namely putidaredoxin and putidaredoxin reductase. Like all members of the CyP450 family, CyP450cam contains a heme subunit, with the iron atom being coordinated by a planar tetradentate heme ligand and by a cysteinato-S atom. A variety of crystal structures of CyP450cam have been reported by Poulos and coworkers, including the substrate-free enzyme,246, 247 the enzyme complexed with its natural substrate camphor248 and with several other substrate analogues.249, 250 In addition, crystal structures of CyP450cam inhibited by carbon monoxide251 as well as by several other inhibitors have been reported.252, 253 The electrochemical and spectroscopic properties of CyP450cam have been studied and correlated with the results from protein X-ray crystallography.254 In a series of papers, Hill, Di Gleria, Wong and coworkers prepared Cytochrome P450cam derivatives modiﬁed with ferrocenyl-succinimide residues.133, 255, 256 The maleimide moiety has a high afﬁnity and selectivity for the sulfhydryl group of cysteine (see Section 13.2.4). The highlight of these papers is undoubtedly the protein ˚ resolution of the CyP450cam (C334A) mutant to X-ray single crystal structure at 2.2 A which two ferrocenylethylsuccinimide groups were covalently attached after reaction of maleimide 104.255 The exposed Cys334, which is at the surface of the wt protein, was altered to an alanine via site-directed mutagenesis to prevent dimerization of the enzyme via disulﬁde formation. The overall structure of the modiﬁed enzyme is depicted in Figure 13.6. This X-ray crystal structure analysis revealed that one of the ferrocenylethylsuccinimide groups is tethered to Cys136 on the periphery of the enzyme, whereas the other succinimide moiety is covalently linked to Cys85, occupying a position in the binding pocket of the enzyme. The Cys85-bound ferrocenyl moiety turned out to be an irreversible inhibitor of the enzyme, displacing camphor from the binding pocket during the derivatisation reaction with 104. The heme iron atom, however, is still accessible for small molecules such as carbon monoxide. This could by veriﬁed by the appearance of a band at 448 nm in the UV spectrum upon reduction of the enzyme and exposure to carbon monoxide.254 Another mutant of CyP450cam was prepared via site-directed mutagenesis.256 All of the surface cysteines (Cys58, Cys85, Cys136 and Cys334) were changed to alanines and, in addition, Lys 344 on the periphery of the enzyme was changed into a cysteine. Subsequently, this K344C mutant was modiﬁed with N -ferrocenylmaleimide (N .B. this is a different reagent from 104). In contrast to the doubly labelled enzyme presented above, this conjugate was found to exhibit about 80 % activity compared to wt CyP450cam , as determined by NADH turnover rates.256 At this stage it should be noted that attempts have been made to modify bakers’ yeast CyP450 with FcHgCl (97).257–259 Instead of attachment of the FcHg+ group to Cys102, a protein monomer modiﬁed at Cys102 with a HgCl+ and a protein dimer formed, in which the mercury atom bridges thiolate groups of two Cys102 residues.

548

Ferrocenes: Ligands, Materials and Biomolecules

Figure 13.6 X-ray structure of Cytochrome P450cam with two molecules 104 bound to the peripheral Cys136 (bottom left) and Cys85, which occupies a position in the heme-binding pocket of the enzyme (PDB code 1GJM) (Graphics were prepared with the iMol program on a MacBook Pro)

Horseradish peroxidase (HRP) is a monomeric heme-containing glycoprotein, consisting of 308 amino acid residues.260 The activity of the enzyme relies on the presence of Ca2+ ions, which regulate the structure of the protein.261, 262 The X-ray crystal struc˚ resolution was recently reported.263 The ture of recombinant HRP determined at 2.15 A enzyme reacts with alkylperoxides in the following manner, according to a modiﬁed ping-pong mechanism Equations (13.8) to (13.10): HRP + ROOH −→ HRP-O(state 2+ ) + ROH +

+

HRP-O(state 2 ) + Dred −→ HRP-O(state 1 ) + Dox HRP-O(state 1+ ) + Dred −→ HRP + Dox + H2 O

(13.8) (13.9) (13.10)

For HRP, only a narrow range of molecules can act as an oxidant, whereas a large variety of enzymes or molecules can act as the reductant (Dred ). Several ferrocene derivatives have been shown to be very suitable electron-donor substrates Dred .264–268 Tsai and Cass prepared HRP derivatives with a ferrocenyl moiety covalently attached to oxidised mannose residues of the main carbohydrate chain.269 The strategy for the modiﬁcation is outlined in Scheme 13.37. First, aldehyde groups are generated by NaIO4 oxidation of the terminal mannose moieties. Next, imines are formed between the aldehyde group and amine 137, followed by reduction with NaBH3 CN to yield the corresponding secondary amine HRP conjugate 138. The kinetic parameters for this HRP-ferrocene conjugate are, in the case of hydrogen peroxide as the substrate, in the same range as those for unmodiﬁed HRP. With linoleic hydroperoxide as the substrate, however, differences between ferrocene-modiﬁed HRP and wt HRP were

The Bioorganometallic Chemistry of Ferrocene

549

CH2OH

1) CH2OH NaIO4 O

POD

HO

O O

Fe

POD

HO

137

H3C

CH2OH

OH OH OH

H3C

OH C CH2 NH2 H

Fe

2) NaBH3CN

O

138

CH2OH

NH2 H2N NaBH3CN OH

CH3

H3C

CH2OH

FcCOOH 48 EDC/HOSu OH

O NH HN

HRP

O NH CH2 HO CH O

SBP

O NH HN

SBP

POD = HRP or SBP H2N

NH2

HN O

NH2 Fc

143

Scheme 13.37

observed. The modiﬁed enzyme shows a signiﬁcantly lower apparent Km value and a signiﬁcantly higher apparent Vmax compared to the native enzyme. An explanation for the difference could be that the surface of the modiﬁed enzyme is more lipophilic than that of the unmodiﬁed enzyme. An enzyme electrode was constructed by adsorbing the ferrocene-HRP conjugate on a printed carbon electrode. This enzyme electrode was suitable for determination of hydrogen peroxide in the range 1–50 µmol/L and quantiﬁcation of linoleic hydroperoxide in the range 5–100 µmol/L. Ryabov and coworkers used a different method to introduce an electrochemically active ferrocene moiety to HRP.270 This strategy consists in the derivatisation of hemin chloride with FcCH2 NH2 (139), by using EDC/HOSu as the coupling reagent mixture (Scheme 13.38). In addition to two different isomeric derivatives 140 and 141, also the disubstituted conjugate 142 was obtained. The mono and disubstituted compounds could be separated by column chromatography, but separation of the two different monosubstituted positional isomers 140 and 141 was not possible. Reconstitution of apo–HRP in the presence of either the mixture of 140 and 141 or 142 alone yielded a catalytically active enzyme only with the mixture of the monosubstituted (140/141) hemin chloride derivatives. Detailed kinetic analysis in conjunction with molecular modeling, starting from the X-ray crystal structure of recombinant HRP263 suggests that the ferrocenyl rest is likely situated on the surface of the active site. It is then believed that a more hydrophobic binding pocket is created on the modiﬁed enzyme. This might explain the altered kinetics of the modiﬁed enzyme compared to native HRP in the case of ferrocene compounds serving as electron donors. O’Fagain and coworkers recently reported the covalent attachment of ferrocenyl units to soybean peroxidase (SBP) by a multistep approach.271 First, oxidation of the carbohydrate residues of SBP (18 % of its 44 kD molecular weight) with NaIO4 afforded aldehyde groups. Treatment with ethane-1,2-diamine converted these aldehydes into primary amines, which were further conjugated 48 in the presence of EDC

550

Ferrocenes: Ligands, Materials and Biomolecules

N

N Fe Cl

N

N

HO2C

CO2H

EDC / HOSu, DMF

N

FcCH2NH2 (139)

N

140: X = NHFc, Y = OH 141: X = OH, Y = NHFc 142: X = Y = NHFc

Fe Cl N

O

X

N

Y

O

Scheme 13.38

and HOSu after reduction of the imines by NaBH3 CN (Scheme 13.37). This procedure yielded conjugates 143 with a Fc-to-SBP ratio of 1.5 with identical catalytic activity with respect to native SBP. The more favorable electroactive properties of Fc-SBP allowed for construction of a sensitive enzyme biosensor for hydrogen peroxide. Kijima and coworkers constructed an alcohol dehydrogenase-based amperometric biosensor for ethanol assay.272 Similar to glucose oxidase, ADH operates with the redox cofactor NAD+ /NADH2 according to Equation (13.11). ADH(NAD+ ) + RCH2 OH −→ ADH(NADH2 ) + RCHO + H+

(13.11)

As with GOD, electrical contacting between NAD-dependent enzymes and electrodes is difﬁcult, hence the necessity for electron relays (Equations (13.12) and (13.13)). ADH(NADH2 ) + 2Fc+ −→ ADH(NAD+ ) + 2Fc + H+ 2 Fc −→ 2Fc+ + 2e− (to the electrode)

(13.12) (13.13)

Following the work published by Willner on the FAD cofactor labelled by a ferrocene entity,222 the authors synthesised several Fc-NAD derivatives by grafting ferrocenyl groups to the exocyclic N-6 group (144 and 145 in Scheme 13.36). The enzymatic

The Bioorganometallic Chemistry of Ferrocene

551

activity of the ADH(NAD-Fc) enzyme was somewhat lower than that of the native enzyme. A bioelectrocatalytic current was measured in the presence of ethanol, ADH and Fc-NAD and the electron transfer efﬁciency was dependent of the spacer length between the NAD core and the redox mediator. 13.3.2

Antibodies, Other Non-Redox Proteins and Dendrimers

The ﬁrst reports on synthetic polypeptides and proteins labelled with the ferrocene moiety date back to 1966 and 1967.273–276 FcNCS (89),273–275 Fc(CH2 )3 CO2 H/carbodiimide275 and FcSO2 Cl (85)276 were used as labelling reagents. In 1967, (4-ferrocenylphenyl)isothiocyanate and (3-carboxy-4-ferrocenylphenyl)isothiocyanate (CFPI, 146 in Scheme 13.39) were introduced for conjugation with primary amino groups.277–279 These compounds were applied for histochemical purposes. The usefulness of this technique was ﬁrst demonstrated by labelling various tissues, such as the placenta,280–283 and several proteins.282–284 Later on, the more polar derivative 146 found application for the labelling of immunoglobulins, which will be described in the following section. S NCS Fe

H2N

IgG Fe

CO2H

N H

N H

IgG

CO2H

146

Scheme 13.39

The labelling of antibodies, in particular immunoglobulins G (IgG), with 146 was reported in a series of papers by Franz, Wildf¨uhr, Wagner and coworkers in the late 1960s and early 1970s (Scheme 13.39).280, 284–289 The CFPI-labelled IgGs were used for immunohistochemical investigations. In a series of papers, the concept of this method, together with a detailed description on how labelling of IgGs could be achieved, was presented.280, 286, 289 In subsequent papers, 146-labelled IgGs were adapted for the localisation and visualisation of: (1) cell wall antigens, in this case the M protein, in the bacterium Streptococcus pyogenes;285 (2) surface antigens on the parasite Toxoplasma gondii ;288 and (3) GRAFFI-virus-induced surface antigens in rat GRAFFI leukemia cells287 In addition to derivatisation with CFPI, examples of IgG labelling with 97,259, 290 48,291–297 133,298 and 32299, 300 have also been reported. Yasuda prepared Femmodiﬁed IgG antibody via imine formation with 32 and subsequent NaBH4 reduction.299 However, these labelled IgGs were not well characterised and their properties not determined. In another paper by Yasuda and Yamamoto, ferrocene labels were introduced in an indirect manner on the immunoglobulin, by using ovalbumin as a ferrocene-label-carrying molecule (see Section 13.3.2).290 First, hen egg ovalbumin was reacted with N -acetylhomocysteine thiolactone, which resulted in amide formation between protein NH2 groups and the C-terminus of N -acetylcysteine. In this way, about nine new sulfhydryl groups were introduced on the protein ovalbumin.

552

Ferrocenes: Ligands, Materials and Biomolecules

Subsequently, ferrocene moieties were covalently attached through Fc-Hg-S bonds by treatment with 97. Next, aldehyde groups were introduced on the carbohydrate part of the ovalbumin derivative via reaction with NaIO4 . In the last step, the ovalbumin-HgFc conjugate was covalently attached to IgG via imine formation, followed by reduction with NaBH4 . Some of these electroactive antibody conjugates were employed as part of electrochemical immunoassays for model or clinically-important analytes. A solution-phase, ﬂow-though, direct assay was set up for HCG and mouse IgG (mIgG) using labelled anti-HCG and anti-mIgG antibodies. Separation of excess labelled antibody from the antigen–antibody complex was achieved on the basis of the difference of pIs between both species, using on-line capillary electrophoresis297 or cation exchange chromatography291–294 . Amperometric detection of the ferrocene tag was carried out in an electrochemical cell (Figure 13.7, top). A calibration curve for HCG was established within the range between 20 and 2000 mIU/mL. For the hapten-type analyte histamine, a solution phase competitive assay was set up using labelled antihistamine antibody and histamine-BSA conjugate. Separation of free and bound fractions of the labelled antibody was performed using a micro-fabricated on-chip multi-channel matrix column device (Figure 13.7, bottom). Histamine could be reliably quantiﬁed in the range of concentrations between 250 and 2500 ng/mL.294 This setup was recently adapted to the assay of the A1c form of hemoglobin (HbA1c ), which is the glycated form of hemoglobin (Hb) and a biological marker for diabetes.301 Anti-Hb antibody (recognising both forms of Hb) labelled with ferrocenyl entities was mixed with Hb and HbA1c to give two antigen–antibody complexes. On-line afﬁnity chromatography on boronate-activated agarose selectively retained the anti-Hb–HbA1c complex while other species (including the free labelled antibody) were eluted and detected electrochemically. A calibration curve for HCG was established within the range between 200 and 2000 µg/mL. Other immunoassays employed antibodies labelled by ferrocenyl entities. Jiang and coworkers labelled an anti-2,4-dichlorophenoxyacetic (2,4-D) antibody by reaction

Figure 13.7 Schematic diagram of the electrochemical ﬂow immunoassay: A) direct mode for antigens; B) Competitive mode for haptens

The Bioorganometallic Chemistry of Ferrocene

553

with the N -hydroxysuccinimide ester of 3-(ferrocenyl)propionic acid (147).302 The Fcto-antibody ratio of the conjugate was not determined. Alternatively, the same reagent was reacted with HRP and the resulting protein was conjugated to the antibody by reductive amination after oxidation of the sugar residues of HRP with NaIO4 . The Fc-to-HRP and HRP-to-antibody molar ratios were estimated by UV–Visible measurements. A solid phase inhibition-type immunoassay of the pesticide acid (2,4-D) was set up with labelled anti-2,4-D antibody. Quantiﬁcation of iron borne by the bound fraction of labelled antibody was carried out with ICP–MS using the dynamic reaction cell technique. A calibration curve for 2,4-D was established in the range between 0.1 and 1000 ng/mL with a LOD of 0.044 and 0.055 ng/mL for the Fc-HRP-IgG and Fc-IgG reagents, respectively. Mak and coworkers encapsulated manually ground ferrocene microcrystals via a layer-by-layer technology to which was adsorbed anti-mIgG antibody303 and achieved extremely high Fc-to-antibody ratios (104 –105 ). A solid phase immunometric assay of mIgG was set up and the amount of bound labelled antibody was quantiﬁed electrochemically after addition of DMSO as a releasing solution. A solid phase electrochemical sandwich immunoassay for HCG using ferrocenelabelled antibodies was reported recently.304 The antibody was labelled by reaction with 48.291 In this assay, the capture antibody was directly adsorbed onto a glassy carbon electrode after activation with EDC. A calibration curve was established for assaying HCG in human serum in the range between 50 and 3200 IU/L. Signal ampliﬁcation was achieved in an amperometric immunosensor operating in the sandwich format developed by Yasuzawa and coworkers.305 The antitransferrin detection antibody 15C2H3 was labelled with a conducting ferrocene-containing polymer. Quantitative analysis of transferrin was carried out by measuring the oxidation current of ferrocene, which is proportional to the transferrin concentration. In a series of interesting papers, Moiroux and coworkers labelled IgG with PEG-Fc molecules to yield conjugate 148 (Scheme 13.40).306–308 They assembled successive layers of modiﬁed IgG molecules and anti-IgG antibodies on electrodes and determined several physical properties, such as the diffusion of the redox probes and the dynamics of the chain. Electron transfer to and from GOD-antibody conjugates positioned at the outermost part of the immunological constructions was studied as a function of the ﬁlm thickness (which is determined by the number of antigen–antibody layers). This construct represents an alternative way of wiring a redox enzyme, here GOD, to an electrode (Figure 13.4, conﬁguration D2). Bovine serum albumin (BSA) is a transport protein with a molecular weight of 66.7 kD, consisting of 583 amino acids, of which 59 are lysine residues.309 In 1969, Schl¨ogl et al.310 reacted BSA with ferrocene imidocarboxylic acid ethyl ester (149 in Scheme 13.41) to obtain a conjugate that, according to the authors, contained

O IgG

O

O

O

76

O 148

Scheme 13.40

N H

Fe

554

Ferrocenes: Ligands, Materials and Biomolecules

BSA

EtO BSA

NH2

+

HN

Fe

NH HN

Fe

149

Scheme 13.41

approximately 60 ferrocenyl groups. This value should be taken with appropriate care, because it has been determined by gravimetry and not by more accurate techniques such as atomic absorption spectrometry (AAS) or spectrophotometry.208 In two papers, Mizutani and Asai introduced Fem groups on BSA, via the reaction with 32 followed by NaBH4 reduction.311, 312 Under denaturating conditions in the presence of 6 M urea, about 40 of the 60 amino groups of BSA were derivatised, whereas, without added urea, only 20 Fem rests were introduced.312 These ferrocenylBSA conjugates were applied as efﬁcient diffusing macromolecular mediators between the enzyme GOD and the electrode (Figure 13.4, conﬁguration A2). Shinohara and coworkers reacted BSA with 147 in the presence of EDC, which resulted in formation of two types of conjugates: proteins that possessed either 23 or 5 ferrocenyl units.313 The authors showed these conjugates to be efﬁcient macromolecular mediators between fructose dehydrogenase and an electrode. The conjugate with higher ferrocene-to-BSA ratio was found to be more effective, which was explained by the higher number of redox centres attached. In an interesting paper, Kunugi et al. explored the reaction of BSA with 32 followed by NaBH4 reduction under different conditions. They investigated the differences between the denaturation of the protein via high pressure (500 MPa) or via addition of urea during the labelling process. Conjugates with higher Fc-to-BSA ratio were obtained by applying a 500 MPa pressure. However, the conjugates prepared via denaturation with urea were better mediators between GOD and the electrode, despite the lower number of ferrocenyl residues attached.210 Aizawa and coworkers prepared ferrocene and digoxin double-conjugated BSA derivatives for application in a homogeneous electrochemical immunoassay.314 In total, 39 Fem groups and 8 digoxin molecules were connected to BSA via imine formation and subsequent NaBH4 reduction. This derivative was used as an antigen in an immunoassay of anti-digoxin antibody. Binding of the antibody resulted in a decrease of the electrochemical activity of the BSA derivative. This is caused by the diminished electrochemical availability of the ferrocene units, since the bulky antibody shielded them. The measured peak currents were inversely related to the concentration of antibody, in this way allowing the quantiﬁcation of the antibody. Lo and coworkers have also used 93 for the labelling of sulfhydryl-modiﬁed BSA.135 Before reaction with 93, BSA was treated with N -succinimidyl-S-acetylthioacetate to increase the number of free thiol groups. After labelling, the ferrocene-to-BSA ratio was about 40:1, as estimated from UV absorbance spectra. Another BSA-Fc conjugate obtained from reaction of BSA with 48 after activation with EDC was used to construct an amperometric biosensor for hydrogen peroxide.315 HRP together with BSA-Fc and carbon nanotubes were entrapped in a sol–gel matrix of ormosil deposited at the

The Bioorganometallic Chemistry of Ferrocene

555

surface of a glassy carbon electrode. The resulting sensor could be used for ﬂow injection analysis of hydrogen peroxide within the range of concentrations from 0.02 to 4.5 mM. Avidin (AV) is a tetrameric glycoprotein of molecular weight 66 kD found in egg white. It is able to bind four biotin molecules with very high afﬁnity (Kd = 10−15 M).316 AV contains nine lysines per monomer (total of 40 amino groups). Streptavidin (SAV) produced by the bacterium Streptomyces avidinii is also able to up to four molecules of biotin with high afﬁnity (Kd = 10−13 M). SAV contains eight lysines per monomer, i.e. a total of 36 amino groups. Although these proteins are not genetically related, their sequences are rather similar and the conserved amino acid residues are mostly conﬁned to six homologous segments.317, 318 Crystal structures have been reported for free and biotin-complexes form of AV319, 320 and SAV.321, 322 Padeste and coworkers coupled ferrocenyl moieties to the amino groups of AV and SAV by means of EDC coupling chemistry using N -(ferrocenylmethyl)aminocaproic acid (150), acetylated N -ferrocenylmethyl glycine (151) and the oligo-glycine derivative 152 as ferrocenyl source (Scheme 13.42).323–328 glycine

FcCHO

NaBH4

32 NaBH4

H N

H2N O

Fe

N H

Fe

N H

COOH

AcOAc

Fe

N Ac

COOH

151 COOH

6

H N

COOH

6

O

AcOAc

Fe

N Ac

H N

COOH

6

H 152

Scheme 13.42 Synthesis of ferrocenyl carboxylic acid derivatives 151 and 152 for the labelling of AV and SAV

By controlling the reaction conditions, AV and SAV conjugates with Fc-to-protein ratios between 3 and 35 were obtained. Chemical modiﬁcation of AV/SAV did not alter their binding capacity towards biotin. Covalent immobilisation of these labelled proteins on the surface of a gold electrode yielded electroactive platforms with wiring capacities (Figure 13.4, conﬁguration D1b). Immobilisation of the biotinylated redox enzymes lactate oxidase and GOD and the enzyme mimic microperoxidase 11 led to amperometric biosensors for detection of lactate, glucose327 and hydrogen peroxide,323 respectively. The same group devised a DNA sensor making use of Fc-SAV as electrochemical tracer (Figure 13.8, conﬁguration A).325, 326 A 15-mer PNA probe was attached via its 5 end to a gold electrode via a ﬂexible polyethylene glycol linker. Biotinylated 12-mer complementary DNA sequence was hybridised to the capture probe. This was followed by addition of Fc-SAV. Ferrocene present on the surface of the electrode was measured by square wave voltammetry (SVW), allowing detection of the target down to 10 pM.

556

Ferrocenes: Ligands, Materials and Biomolecules

Figure 13.8 Electrochemical DNA sensors using avidin labelled with ferrocene markers

Wang and coworkers reported another strategy to label SAV with ferrocene moieities.329, 330 It consisted of treating commercial gold nanoparticles coated by SAV with 6-ferrocenylhexanethiol (153). Analysis of the conjugate revealed that an average of 127 ferrocene residues and 1.7 SAV molecules were bound to each particle, i.e. a much higher Fc-to-SAV ratio with respect to direct labelling of SAV (see above). A DNA sensor operating in the sandwich format with a biotinylated signaling probe or the direct format with a biotinylated target (Figure 13.8, conﬁguration B) was subsequently built. Ferrocene was detected amperometrically using cyclic voltammetry (CV) and the charge under the anodic wave was correlated with the concentration of target in the range down to 2 pM for the sandwich assay and 0.25 pM for the direct assay. The same Fc-SAV-Au reagent was used in an attempt to quantify the number of sulfhydryl groups in surface-conﬁned glutathione (GSH) or metallothionein (MT) (Figure 13.9).329, 331 Anne, Moiroux and coworkers were the ﬁrst to report the synthesis of ferrocenyl derivatives of biotin 154 and 155 (Scheme 13.43).332, 333 Compound 155 was included in the construction of a supramolecular architecture based on the avidin – biotin association aimed at the wiring of GOD (Figure 13.4, conﬁguration D1a). The high ﬂexibility and length of the PEG spacer arm allowed fast oxidation of FADH2 cofactor by ferricinium entities, hence more efﬁcient electron wiring with respect to Fc-GOD. Tanaka and coworkers prepared the biotin–ferrocene conjugate 156 (Scheme 13.43) by reaction of biotin hydrazide with 32 in the presence of NaBH4 .334 This tracer was

Figure 13.9 The quantiﬁcation of surface-conﬁned protein/peptide sulfhydryl groups

The Bioorganometallic Chemistry of Ferrocene

557

O NH

HN Fe

H N O

O

O x

R

NH H

O

H N

O

S O

S O

Fe N H

69

155

O NH H

HN H

z

y

154, R = H or Me

O HN H

H N

O

H N

S O

N H

Fe

156

Fe

H N

O O

O 157

N H

O

H HN

NH S

H

Scheme 13.43 Chemical structure of biotin-ferrocene derivatives

used as part of a homogeneous electrochemical binding assay of biotin on the principle initially described by Limoges, Degrand and coworkers335, 336 for the immunoassay of haptens (see Section 13.6.4). Compound 156 was sensitively detected by SWV on a Naﬁon-coated electrode after accumulation of the ferricinium species at 600 mV. Addition of avidin led to the formation of avidin/biotin-Fc complexes, which resulted in the decrease of the anodic signal as the high molecular complex was unable to penetrate in the ﬁlm. Further addition of free biotin displaced the binding of biotinFc to avidin, which resulted in an increase anodic current. A calibration curve was established for biotin concentrations ranging from 10 nM to 10 µM. Mosbach and Schuhmann followed a different approach toward an afﬁnity-based biotin sensor. They prepared the ferrocene-biotin conjugate 157 (Scheme 13.43).337, 338 The afﬁnity assay was based on the fact that the measured current for a microelectrode is correlated to the diffusion coefﬁcient of the redox species. Binding of the electroactive conjugate 157 to streptavidin results in a change in the diffusion coefﬁcient of the electroactive label on account of the increase in molecular mass. Suggestions for signal ampliﬁcation were given in this paper, and the generality of the idea was demonstrated.

558

Ferrocenes: Ligands, Materials and Biomolecules

Dendrimers are a relatively new range of polymers with a well-deﬁned hyperbranched architecture (see Chapter 10).339 Their size together with several other unique properties such as monodispersity and a high number of terminal functional groups make them very attractive for a wide range of biological applications,340 such as drug delivery, diagnostic imaging,341 neutron capture therapy342 and gene delivery343 . Of particular note in this context is the fact that the size and globular shape of the poly(amidoamine) (PAMAM) and poly(propylenimine) (PPI) dendrimers are closely related to that of proteins.344 PAMAM and PPI dendrimers are synthesised by a divergent approach deﬁning successive generations. Kim, Kwak and coworkers reported the tethering of ferrocenyl entities to PAMAM generation four dendrimers by reductive amination with 32.345–349 They subsequently constructed a bioelectronic platform by covalent immobilisation of these electroactive dendrimers on gold electrodes to form a monolayer. A reagentless glucose biosensor was built by alternating layer-by-layer deposition of periodate-activated GOD and FcPAMAM. The assay sensitivity was correlated to the number of Fc entities tethered to PAMAM and the number of deposited layers. This indicated that the bioelectronic platform efﬁciently shuttled electrons between GOD and the electrode (Figure 13.4, conﬁguration C2).347 The same electroactive monolayer, decorated with biotin rests, was used to build up a biosensor for AV. Shuttling of electrons to and from GOD in solution was inhibited by formation of interface avidin–biotin complexes (Figure 13.10, top). Hence, the bioelectrocatalytic signal resulting from the oxidation of glucose by GOD decreased in the presence of increasing amounts of AV in the range between 1.5 pM and 10 nM.348 As the biotin–avidin association is very strong and is difﬁcult to dissociate, the authors replaced AV by an antibiotin antibody. Electron wiring was again inhibited by binding

Figure 13.10 PAMAM-Fc conjugates as a bioelectronic platform for measuring avidin or anti-biotin antibodies

The Bioorganometallic Chemistry of Ferrocene

559

of the antibody to the biotinylated electroactive platform, but the effect could this time be reversed by adding free biotin (Figure 13.10, bottom).349 The ferrocene-PAMAM platform was also used to construct a DNA sensor operating in the sandwich format.345 The capture probe was covalently immobilised onto this platform using a hetero-bifunctional cross-linking agent. Hybridisation of target DNA was followed by addition of a biotinylated signalling probe complementary to target. Avidin-ALP (alkaline phosphatase) conjugate bound to the signalling probe catalysed the formation of the electroactive label p-aminophenol (AP) by hydrolysis of p-aminophenylphosphate (APP). AP was electrocatalytically oxidised into p-quinone imine (QI) by mediation of ferrocene in Fc-PAMAM dendrimer, resulting in signal enhancement (Figure 13.11). The versatility of the Fc-PAMAM biotinylated platform combined to the electrocatalytic detection of AP was recently demonstrated for the setup of an immunosensor for the antibiotin antibody.346 By using a secondary antibody labelled by ALP, a calibration curve for antibiotin was obtained in the range between 0.1 and 100 µg/mL (Figure 13.12). In a related work, AP was produced by reduction of p-nitrophenol (NP) with NaBH4 catalysed by gold nanoparticles decorated by the detection antibody.350 Very sensitive

Figure 13.11 An electrochemical DNA sensor conﬁguration using PAMAM-Fc dendrimers

Figure 13.12 An enzyme-ampliﬁed immunosensor with mediation by PAMAM-Fc

560

Ferrocenes: Ligands, Materials and Biomolecules

Figure 13.13 Schematic representation of an electrochemical immunosensor for PSA mediated by PAMAM-Fc

immunosensors for mIgG and PSA operating in the sandwich format were set up thanks to the ampliﬁcation of the electrocatalytic signal by back reduction of QI to AP by NaBH4 (Figure 13.13). A calibration curve was established for both analytes between 1 fg/ml (for mIgG) or 10 fg/mL (for PSA) and 10 µg/mL. Alonso and Garcia Armada designed several reagentless aerobic glucose biosensors using a layer of PPI dendrimers bearing peripheral ferrocenyl groups.351–353 These electroactive dendrimers were synthesised as depicted in Scheme 13.44. A layer of these dendrimers was deposited on the electrodes, followed by electrostatic adsorption of GOD. The mixed cobaltocenium, ferrocene dendrimers efﬁciently mediated electron transport from/to GOD while allowing the determination of the oxygen variation during the enzymatic reaction.351 The polymethylferrocene dendrimers mediated the reduction of hydrogen peroxide produced when oxygen is used as mediator of GOD catalytic oxidation of glucose (aerobic conditions).352, 353 The ﬁrst report of a protein, in this case ovalbumin, labelled with the ferrocene moiety dates back to 1967.276 Ovalbumin, the main protein of egg white, is a glycoprotein with a molecular mass of 45 kD.354 The primary amino acid sequence of hen ovalbumin has 385 residues, 20 of which are lysines,355, 356 and the N -terminus of the protein is acetylated.357 The crystal structure of uncleaved ovalbumin was solved ˚ resolution.358 In a short communication, Peterlik reported the reaction of 85 at 1.95 A with ovalbumin to yield conjugates in which, on average, 8.6 of the total 20 lysines were derivatised.276 The protein papain found in unripe fruit (papayas) of the tree Carica papaya belongs to the class of cysteine proteinases.359 This protein consists of a single chain of 212 amino acids, of which seven are cysteines.360–362 The X-ray crystal structure of the protein revealed that six of these cysteines form disulﬁdes, whereas Cys25 is located in the active site.363 This Cys25 is essential for enzyme activity.359 Douglas and coworkers performed experiments to investigate whether 49 or 97 could irreversibly

The Bioorganometallic Chemistry of Ferrocene

561

Fe

PPI Gx

NaBH4

Fe

(NH2)n +

PPI Gx

NH

n

CHO

x = 1, n = 4 x = 2, n = 8 x = 3, n = 16 x = 4, n = 32

O

PPI Gx

(NH2)n

+

Fe

+ COCl

Co+ COCl

PPI Gx HN

Fe

∗ NH n/2

O n/2

Co+

Scheme 13.44 Tethering of different ferrocenyl and cobaltocenium units to PPI dendrimers

inactivate papain.364 The latter reagent turned out to be completely unsuitable, whereas 49 rapidly inactivates papain. Kinetic experiments indicated that the inhibition occurs via a two-stage process, involving initial complexation of 49 by the enzyme, followed by covalent attachment via thioether formation. Papain was also shown to be irreversibly inactivated by reaction with N -(ferrocenylmethyl)maleimide (158) by siteselective alkylation of the Cys25. Formation of the covalent Fc-papain adduct was conﬁrmed by ESI–MS analysis.365 A ferrocenyl entity was covalently attached to the active site of copper-free azurin (Apo-azurin) by reaction of 2-[(methylsulfonyl)thio]ethylferrocene (FcCH2 CH2 SSO2 Me, 159) with its Cys112. Electrochemical studies along with molecular modeling showed that the Fc moiety was encapsulated within the hydrophobic protein core, stabilising the oxidised ferricinium form.366 Poly-L-lysine polymers were modiﬁed with Fc groups by reaction with 49. The polymers acted as electron mediators between electrodes and the reduced form of GOD. Amperometric glucose-sensing electrodes were constructed by the simultaneous immobilisation of the polymeric mediator and GOD near the surface of a glassy carbon electrode with a semipermeable membrane.117 Staphylococcus aureus protein A is a protein that presents high afﬁnity for the constant fragment of IgGs of various mammalian species367 and is extensively used for antibody puriﬁcation purposes368 . Labelling of protein A with ferrocenyl entities was achieved by reaction with 32 in the presence of NaBH4 .369 Conjugates with Fc-toprotein ratios of 2 to 12 were obtained. Their afﬁnity for goat IgG decreased steadily when the Fc-to-protein ratio increased.

562

13.4

Ferrocenes: Ligands, Materials and Biomolecules

Conjugates of Ferrocene with DNA, RNA and PNA

Compared to ferrocene derivatives of amino acids and proteins described in Section 13.2, this is a relatively young and less explored ﬁeld in the bioorganometallic chemistry of ferrocene. It is interesting to note, however, that most of the chemistry and applications of ferrocene are rather similar. Imine and amide formation with 32 or 48, respectively, as well as Sonogashira coupling are the preferred methods for covalent attachment of ferrocene to (oligo)nucleotides. A potentially very promising application is as simple and inexpensive gene sensors.370 For such devices, electrochemically active DNA derivatives for sequence speciﬁc detection of DNA oligomers are required. Unsurprisingly, most efforts are concentrating on ferrocene derivatives for this purpose. As will be discussed later (Section 13.4.3), a variety of detection schemes are conceivable and there does not seem to be a gold standard yet. Grinstaff and coworkers have summarised metallo-oligodeoxynucleotides371 and recent developments in redox probes for DNA.372 The subject of this section has recently been reviewed.9, 15 13.4.1

Ferrocene Derivatives of Nucleobases, Nucleosides and Nucleotides

The ﬁrst work on ferrocene derivatives of nucleosides dates back to 1991.373 This work by a French group has not been cited frequently and thus some ‘ﬁrst’ or ‘unprecedented’ claims in later papers are unjustiﬁed. Gautheron and coworkers used a variety of different palladium catalysed C−C coupling reactions to synthesise ferrocene nucleosides and derivatives thereof.373 TMS-protected 5-iodouridine, 8-bromoadenosine and 2 -deoxyuridine cleanly reacts with [Cp2 Zr(Cl)CH=CHFc] (derived from 69 and Schwartz’ reagent) and [PdCl2 (C6 H5 CN)2 ] as a catalyst. After workup with aqueous MeOH, the unprotected uridine, 2 -deoxyuridine and adenosine derivatives 160–162 were obtained in good yield (Scheme 13.45). A reaction is also possible with ethynylferrocene (69) itself and 5-iodouracil, iodouridine, 2-deoxyiodouridine and bromo-adenosine (Sonogashira coupling),374 yielding the substituted nucleobase 5-(ferrocenylethynyl)uracil (163), 5-(ferrocenylethynyl)uridine (164) and 5-(ferrocenylethynyl)deoxyuridine (165) as well as the (ferrocenylethynyl)adenosine (169; Scheme 13.46) . The 5-ethynyl uracil derivatives can be cyclised in the presence of strong base, e.g. NEt3 , yielding ferrocenyl derivatives 166–168. The mechanism, as well as the exact conditions required for this transformation, were controversial.373, 375, 376 Two additional unexpected side products were found, all of which formed exclusively from 163. An improved mechanism has been suggested which satisfactorily explains the range of products observed so far. In line with this, the fully protected nucleoside 170, for which cyclisation is not possible, yields the Sonogashira product 171 in 79 % yield. Related cyclisation reactions with purely organic substrates have previously been reported in the literature.377–380 The ferrocenyl substituent stabilises positive charge on the α-ethynyl carbon atom,381 thus facilitating an attack of the nucleophilic oxygen atom to form the ﬁve-membered ring. Houlton and coworkers also re-investigated this cyclisation reaction under very mild conditions382 and the crystal structures of protected derivatives of 165 and 168. The formation of cyclisation product 168 could be completely suppressed by performing the Sonogashira reaction at room temperature

The Bioorganometallic Chemistry of Ferrocene O I

HN

2. MeOH, H2O

N R

O

1. [Cp2Zr(Cl)C=CHFc], Pd cat.

O

Fc

HN O

N R' HO

HO

OH OH Ribose (160) NH2

Me3SiO

Br N

N

N

1. [Cp2Zr(Cl)C=CHFc], Pd cat. HO

O Me3SiO

OH H Deoxyribose (161)

NH2 N

N

O

O

R' =

N

563

N

N

O

2. MeOH OSiMe3

Fc

162

OH OH

Scheme 13.45

for four hours. Under these conditions, 5 -DMT-protected 165 could be isolated in 75 % yield. Cleavage of the anomeric bond in the ferrocenyl adenosine derivative 169 could be affected by a solution of anhydrous hydrogen chloride gas in chloroform to yield 172.373 The synthesis of ferrocenylethynyl-purines was also reported by Hocek, ˇ epniˇcka and coworkers.383 The triple bond could be successfully reduced to yield Stˇ ferrocenylethyl-purines in an elegant way. Along with electrochemical studies, a number of compounds were characterised crystallographically. Selected compounds were subjected to cytotoxicity assays.373, 383 Another approach to metal nucleobase derivatives was published by Houlton et al.,384, 385 who used (ferrocenylmethyl)trimethylammonium iodide ([FcCH2 NMe3 ]I, 173) as a convenient source of the ferrocenylmethyl cation to alkylate various nucleobases and nucleobase derivatives. A complete series of cytosine, thymine, uracil, guanine and adenine derivatives was obtained; a representative example is shown in Scheme 13.47 for 9-(ferrocenylmethyl)adenine (174).385 The ﬁrst preparation of 174 was already reported in 1980 by Chen, who actually obtained a mixture of 174 along with the isomeric N6 -ferrocenylmethyl adenine and 7-ferrocenylmethyl adenine.386 The crystal structure of 174 reveals a combination of hydrogen bonding to Watson-Crick as well as Hoogsteen sites. Compound 175, which has planar chirality, represents a different approach to metallocene derivatives of nucleobases. It was synthesised regio- and stereospeciﬁcally by an eight-step procedure starting from ferrocenecarbaldehyde protected with a chiral auxiliary.387 The key intermediate 1-amino-2-cyanoferrocene (176) was cyclised with trimethyl orthoformate and p-toluenesulfonic acid, followed by treatment with a solution of NH3 in MeOH to yield 175 (Scheme 13.48). This compound represents an

564

Ferrocenes: Ligands, Materials and Biomolecules Fc O

1. FcC CH, Pd cat.

I

HN

O

HN

2. MeOH, H2O

N R

O

O

O

Me3SiO

N Br 1. FcC CH, Pd cat.

N

O Me3SiO

163 164 165

N R 166 167 168

NH2

NH2

N

HN O

N R

R= H Ribose Deoxyribose

N

NEt3, ∆

Fc

N

N HO

NH2

Fc N

N

HCl

O

2. MeOH

N

N

Fc

172

OH OH

OSiMe3

N

N

169 O BnN BnO

O

N

O

Fc

O I 1. FcC CH, Pd cat., CuI, NEt3, DMF

BnN BnO

OBn OBn

N

O O

OBn OBn

170

171

Scheme 13.46 Palladium-catalysed synthesis of ferrocenylated nucleosides NH2 Adenine, H2O, ∆

[FcCH2NMe3]I

N

N

N

173

N Fc

Scheme 13.47 H 2N

CN NH2 Fe

1. (MeO)3CH, TsOH 2. NH3, MeOH

176

N N Fe 175

Scheme 13.48

174

The Bioorganometallic Chemistry of Ferrocene

565

electrochemically active adenine analogue, in which the electro-active metallocene is actually part of the nucleobase itself. 13.4.2

Ferrocenylated Oligonucleotides

Although the main interest in ferrocene oligonucleotide derivatives is for electrochemical DNA sensors, a few papers deal mainly with synthetic aspects and these will be discussed ﬁrst. We have recently published a detailed analysis of melting temperatures of ferrocene ODNs with complementary DNA and RNA.9 In one instance, the Mitsunobu reaction of FcCH2 OH (47) with 5 -O-(4,4 -dimethoxytriphenylmethyl)-3 -O-acetyl-thymidine yields the coupling product 177, which was readily converted to the phosphoramidite 178 (Scheme 13.49). Puriﬁed 178 was used as a building block in automated oligonucleotide synthesis of DNA 16-mers and 17-mers.388 Four different metallated oligomers were synthesised, differing in the position of the metal label. A series of melting studies with complementary DNA and RNA strands were performed. In all cases, drastic reduction in stability of the duplices was observed except when the metallated T was almost terminal in a B-DNA duplex. Even the formation of triple helices was observed with the third strand bearing the metallated T. CD spectroscopic studies and CD melting point studies were further used to supplement structural information on the duplices and triplices formed in this study. This work represents the most comprehensive study to date on the inﬂuence of the site of ferrocene substitution on stability and structure of DNA•RNA duplexes, dsDNA and triplex formation. O Fc

DMTO

O Fc

N O

N

DMTO

N O

O OAc

177

HN

N

O P N(iPr)2

178

Fc

N DMTO

O

O O

Ac

N

O

CN

O

O P N(iPr)2

CN

179

Scheme 13.49 Ferrocenylated T monomers 177 and 178 and C monomer 179. Compounds 178 and 179 can be used as modiﬁed T and C monomers in automated oligonucleotide synthesis

Acid 48 has been attached to the amino group of a 5 -terminal hexylamine of ODNs. In 1991, the purpose of these conjugates was to use stable isotopes of iron for multiplexed detection of nucleotides of different sequence by resonance ionisation spectroscopy (RIS), either induced by ion sputtering (sputter-initiated RIS or SIRIS) or laser desorption (laser atomisation RIS or LARIS).389–91 This idea is somehow related to the very ﬁrst metalloimmunoassays on ferrocenyl derivatives of steroids reported by Cais, in which AAS detection was used (see Section 13.6.4). Although the method

566

Ferrocenes: Ligands, Materials and Biomolecules

worked with high spatial resolution on electrophoresis gels and the sensitivity was comparable to autoradiography with 32 P, its use is very limited. Firstly, a relatively high background of 56 Fe was found in carriers like Nylon membranes, making the use of 57 Fe necessary for good sensitivity.389 Secondly, a sophisticated and expensive device for detection is necessary. Finally, the advent of DNA chips has largely overcome the need for multiplexed analysis in traditional sequencing. In most other works, ferrocenylated nucleotides were synthesised by Sonogashira coupling of ethynylferrocenes to iodo-U or bromo-A. Two synthetic strategies have been followed: transformation of ferrocenylated nucleotides into phosphoramidites; and subjecting them to automated oligomer synthesis or derivatisation with ethynylferrocene (69) after assembly of the oligomer. Metzler-Nolte et al. have found that propargylamides 64 are particularly suitable ferrocene derivatives for Sonogashira coupling on amino acids and peptides (see Scheme 13.16). These compounds are conveniently prepared on a large scale and the X-ray crystal structure of 64b has been reported.392 Compound 64a was used by Grinstaff et al. for the labelling of oligonucleotides by Sonogashira coupling.371, 393, 394 These authors reported improved yields and purity of the conjugates if the derivatisation is carried out on the column, rather than in solution after cleavage of the oligonucleotide. Similar ﬁndings were made for labelling of DNA by Sonogashira coupling with a tris(bipyridine)Ru propargylamine derivative.395, 396 In their work, Grinstaff et al. found that the structure of dsDNA at room temperature is not signiﬁcantly altered by incorporation of an ferrocenylethynylA or -U, regardless of the position of the label. On the other hand, a decrease in melting temperature Tm of up to 9 ◦ C for a 16-mer is observed if an A in the middle of one strand is substituted by ferrocene. Again, the effect is smaller if U is substituted and if the label is close to the end of the duplex.394 During their efforts to incorporate 69 directly into U monomers by Sonogashira coupling, Yu and coworkers observed the same cyclisation reaction in the presence of DIPEA that had already been reported by Gautheron and coworkers.373, 375 Both isomers 165 and 168 (see Scheme 13.46) could be separated, transformed into phosphoramidites and used as building blocks in automated DNA oligomer synthesis. On the other hand, standard DNA deprotection/coupling requires basic conditions and cleavage of (ferrocenylated) ODNs from the solid support is achieved by concentrated ammonia. It is, therefore, not surprising that only 168 is found in ODNs, regardless of the starting material and this point is indeed convincingly demonstrated in this paper. However, another aspect of this work really is surprising. A number of hybridisation experiments with 15-mer ODNs were carried out using ferrocenyl-U in the fourth position.375 If (cyclised) Fc-U is opposite to A, Tm is 48.0 ◦ C and the decrease in melting temperature Tm is −5.7 ◦ C compared to a perfect match. However, if G is placed opposite to Fc-U, the melting temperature is 56.5 ◦ C, which is even higher than the perfect A-T match just mentioned. On the other hand, Tm to a perfect G-C match is −4.2 ◦ C, which seems a reasonable number. It can be concluded that cyclised Fc-U behaves more like dC and prefers to hybridise to G instead of A. This is an important ﬁnding because it underscores the necessity for comprehensive studies. Houlton’s group reported the synthesis of the 5-(ferrocenylethynyl)deoxycytidine monomer 179 (Scheme 13.49) by palladium-catalysed cross-coupling between 69 and 5-iododeoxycytidine.397 This compound was further incorporated into a 12-mer ODN

The Bioorganometallic Chemistry of Ferrocene

567

by solid phase synthesis on both CPG and silicon wafers. Hybridisation with complementary ODN on the silicon wafer shifted the reduction potential by +34 mV. A general approach for introducing a label such as ferrocene has been published by Saito.398 The aldehyde-containing universal base 3-formylindole 2 -deoxynucleoside is introduced in an ODN during solid phase synthesis at any desired position. The ODN is then treated with a hydrazine derivative, such as FcCONHNH2 , to yield the postsynthetically modiﬁed ferrocene ODN. Unfortunately, no additional data, e.g. melting temperatures with complementary DNA or electrochemical characterisation, were presented. On the other hand, this approach is quite universal and deserves further attention. In all cases reported so far, including the ones in Section 13.4.3, the ferrocene label is introduced either on the monomer stage or after assembly of the oligomer sequence at the 5 end. An interesting alternative for the labelling of the 3 end of an oligonucleotide was presented by Anne and coworkers. These authors used the dideoxynucleotide triphosphate Fc-ddUTP (180 in Scheme 13.50) in an enzymatic reaction to extend the 3 terminus of an oligonucleotide.399 The building block 180 was prepared from 5-iodouridine and a ferrocenyl alkyne, followed by several standard organic synthesis steps in an overall yield of about 10 %. The enzyme terminal deoxynucleotidyl transferase (TDT) was then used to extend 5 -p-(dT)10 at the 3 end. Rapid and quantitative transformation was observed by HPLC with a 10-fold excess of Fc-ddUTP. O O −

−

−

O− O O N O P O P O P O O O O O O 180

N H

HN

−

Fc

O3P O (dT)10 OH, TDT 5' 3'

−

−

HN

O O

N H

Fc

O− O O N − O3P O (dT) 10 O P O P O P O O O O O O

Scheme 13.50 Enzymatic labelling of ODN with the Fc conjugated dideoxynucleotide triphosphate FcddUTP

13.4.3

Applications of Ferrocene-Labelled DNA and RNA in Gene Sensors with Electrochemical Detection

The interaction or hybridisation of two complementary oligonucleotide strands usually takes place in a strictly speciﬁc manner. Gene sensors capitalise on this speciﬁc interaction. Strand I, called the (capture) probe, with a known sequence, interacts with strand II (or the target), giving rise to a readable analytical signal. If probe and target sequences are not fully complementary, no hybridisation occurs, thus no signal is measured. Practically, this ‘digital’ response is hardly ever achieved. The kinetics

568

Ferrocenes: Ligands, Materials and Biomolecules

of hybridisation is an issue, as well as the sensitivity of the system to one or more mismatches. A variety of signal transduction methods have been devised, involving optical, gravimetric and electrochemical readouts. Optical DNA sensors based on ﬂuorescence are very sensitive and allow massive parallel analysis with high density arrays. Given their cost, these techniques are not applicable to wide-scale genetic testing. This opens a space for the development of easy-to-use, fast and inexpensive analytical devices. These characteristics are indeed met by electrochemical DNA sensors. The advent of electrochemical DNA biosensing technologies is linked to the set up of point-of-care gene analysers that may help physicians to quickly establish diagnostics as regards genetic (single-point mutations (SNP) analysis) and infectious diseases. With respect to the now well-established ﬂuorescent-based gene chips, electrochemical DNA sensors also allow multiplexed analysis, although the electrode arrays are limited to a few hundred at most.400 Several reviews in recent years discussed individual aspects of electrochemical DNA sensors such as miniaturisation,401 DNA probe immobilisation402 and detection methods.400, 403–406 In DNA or RNA oligomers, the guanine nucleobases may be electrochemically oxidised. This property has been exploited for so-called indicatorfree (direct) electrochemical DNA sensors.400, 405 Unfortunately, the high potentials required for DNA oxidation generates signiﬁcant background currents.403 Not surprisingly, because of its well-behaved redox properties, many gene sensor conﬁgurations employing ferrocene or ferrocenyl derivative as redox label have been reported. All the DNA sensors making use of ferrocene are amperometric, i.e. they rely on the measurement of the current associated with the oxidation of ferrocene at constant potential. Two main conﬁgurations are encountered, whether the redox species is covalently linked to a nucleobase or to a hybridisation indicator (intercalator). In the former case, the ferrocenyl entity can be linked to the capture probe or the target, or to a DNA sequence complementary of the capture probe or the target. In the latter case, the hybridisation indicator is a molecule that binds preferentially to dsDNA resulting from the hybridisation of the probe with the target. In addition to DNA sequence detection and SNP analysis, ferrocene-based DNA electrochemical platforms have found other related applications that will be reviewed at the end of this section. The very ﬁrst reports of ferrocene-labelled DNA applied to electrochemical gene detection concerned solution phase assays. Ferrocene was attached via a ﬂexible linker to the 5 end of a DNA oligomer, as exempliﬁed for the ferrocenylated (dT)12 ODN (181 in Scheme 13.51). This electrochemically active probe DNA was hybridised to the complementary DNA, and the dsDNA was electrochemically detected by HPLC– ECD.407–410 This technique is related to the work of Eckert et al., who used ferrocene labels to make peptides amenable to electrochemical detection in HPLC.411, 412 In the HPLC–ECD system, Takenaka et al. achieved a sensitivity of down to 1 fmol of DNA when a ferrocenylated T12 -mer was hybridised with an excess of poly(dA).409 In a more realistic case, the same probe was hybridised with a plasmid containing a choline transporter gene (CTG) fragment of 3693 bp. The CTG promoter region contains one A13 sequence, and the whole CTG fragment was detected at a minimal concentration of 20 fmol. A better sensitivity was achieved using a ferrocenylated mixed sequence 20-mer as the probe,409 and this difference was attributed to the

The Bioorganometallic Chemistry of Ferrocene O Fe

N H

569

O− 5' 3' O P O TTT TTT TTT TTT – OH O

181

Scheme 13.51

higher melting temperature of duplexes containing the longer probe. On the other hand, a higher melting temperature might imply a decreased mismatch sensitivity, but this aspect was unfortunately not investigated. The thermodynamics of triplex formation with a ferrocenylated ODN has been studied in detail.407 A huge enthalpic gain upon triplex formation is largely canceled by entropic effects, resulting in a gain of free energy of 2–3 kcal mol−1 , compared to triplex formation of an unmodiﬁed ODN. A similar enthalpy–entropy compensation has been observed before in unmodiﬁed ODN triplices, suggesting that triplex formation with a ferrocenylated ODN is governed by the same major inﬂuences as are unmodiﬁed ODNs and that there is no unusual ferrocene-speciﬁc effect, at least in these systems. Finally, electrochemical data for the ferrocene moiety in the triplex are very favourable. There is only a slight sequence dependence of the redox potential of 30 mV. HPLC–ECD detection of this triple helix is possible at the femtomole level, which compares favorably to the sensitivity of radioisotopic or enzyme-linked colorimetric assays.407, 413 In later work, the stability and structure of a 16-mer DNA triplex containing a 3-N -Fem thymidine residue in the third strand were compared to an unmodiﬁed triplex of the same sequence by differential scanning calorimetry, CD spectroscopy, and molecular modeling. In this case, the ferrocene nucleotide does not disrupt the global geometry of the triplex but lowers the apparent pKa value of neighboring cytosines by making them more accessible to the solvent.414 In their work on sequence-speciﬁc DNA detection, Fang et al. appended an amine function to the 5 end of single strand calf thymus DNA by reaction with 1,2diaminoethane. Covalent attachment of the ferrocene tag was achieved by reaction with 48 in the presence of EDC.415 Target (complementary single strand calf thymus) DNA was accumulated on a glassy carbon electrode. The electrochemical response after addition of ferrocenylated probe measured by DPV yielded a limit of detection of 5 nM. They also reported the coating of glassy carbon electrodes with chitosan ﬁlm and a slightly different ferrocene-labelled DNA oligomer.416 This positively charged polymer forms very stable complexes with negatively charged DNA. Still the DNA hybridises with the complementary probe carrying a ferrocene label. A linear relationship between the anodic peak current measured by DPV and the amount of immobilised DNA over more than two orders of magnitude was measured with a limit of detection of 2 nM.416 A model system making use of a 5 -ferrocenylated capture was set up to electrochemically detect single point mutation on the human haemochromatosis gene.417 The ferrocene phosphoramidite 183 was attached to the 5 end of an 18-mer oligonucleotide by solid phase synthesis (Scheme 13.52). The resulting redox probe was hybridised with the 60mer target in appropriate buffer. Treatment of this duplex with T7 exonuclease, which displays a 5 -3 exonuclease activity, cleaved the terminal

570

Ferrocenes: Ligands, Materials and Biomolecules FcCO2H 48

1. oxalyl chloride 2. H2N(CH2)6OH

FcC(O)NH(CH2)6OH

NC(CH2)2OPCl[N(i-Pr)2] (182) O

Fe

N H

O 5

P N

O

CN 183

Scheme 13.52 Synthesis of 5 -ferrocenylated phosphoramidite 183

nucleotide at the 5 end of the probe. The generated Fc-mononucleotide was selectively detected at the electrode because of its higher diffusion coefﬁcient with respect to the labelled duplex. Letsinger et al. were the ﬁrst to immobilise ferrocenylated ODNs in a self-assembled redox-active monolayer on a gold surface.418 (6-Hydroxyhexyl)ferrocene was reacted with (2-cyanoethyl)-N ,N -diisopropylchlorophosphoramidite (182) to yield the corresponding ferrocenyl amidite 184, which was coupled to thymidine modiﬁed CPG. In addition to the monomer 185, a 5 -ferrocenylated (dT)14 ODN (186) and ferrocenylated thymidine monomer with a 3 -thiol group (187) were synthesised (Scheme 13.53). Compound 187 was adsorbed on the surface of gold electrodes as a monolayer. The CV of a gold substrate modiﬁed in this manner exhibited a reversible wave at slightly more positive potential than 184 in solution. In addition to applications as electrochemical DNA sensors, such self-assembled DNA monolayers with electrochemically active groups may provide information on the mechanism of electron transfer through DNA as well as on the ﬂexibility of short DNA. Kraatz and coworkers used differences in the ferrocene/ferrocenium redox potential and kET values to differentiate between different modes of electron transfer between a gold electrode and ferrocene-labelled dsDNA, in particular intrastrand ET versus interstrand crossing mechanisms.419 In a later paper, the properties of ferrocene-labelled ssDNA adsorbed on gold electrodes were studied.420 When ssDNA with a ferrocenyl probe is bound to a gold electrode, it may be expected to adopt an unordered coil conformation. If, on the other hand, a complementary ODN is added, the coil will open up and the DNA double helix will form. These conformational changes result in a change in the mobility of the ferrocenyl reporter group, which in turn alters its electron transfer properties, as shown by Anne et al. in a very interesting publication.421 This hybridisation-induced physical change of DNA was again observed for 5 -ferrocenylated DNA chemisorbed on a special interface. The labelled ssDNA 20-mer capture probe adopts a ﬂexible conformation enabling electron transfer between the redox label and the electrode, whereas the dsDNA is more rigid, switching off the redox peak.422 This behaviour is reminiscent of the so-called molecular beacons, in which ﬂuorescence emitted by a probe is switched off by a nearby quencher. Upon binding of the

The Bioorganometallic Chemistry of Ferrocene

571

O

(CH2)6 Fe

O− O P O O

HN N

O O

O

OH 185 (CH2)6

O (CH2)6 O P Fe

Fe

CN

O− O P O O

HN O

N

O

N(iPr)2

O 186

O

184

O

(CH2)6 Fe 187

O− O P O O

HN O P O− O N O

HN O

N

O

O O P O− O HS

O

3

12

O O O P O− O

HN O

N

O OH

Scheme 13.53

complementary ODN, probe and quencher are spatially separated and a ﬂuorescence signal can be picked up. Grinstaff and coworkers and Heeger and coworkers have simultaneously translated this principle to DNA detection systems in which the probe is an immobilised, ferrocene-labelled ODN with self-complementary 5-base sequences at the 5 and 3 ends.423, 424 In the absence of a target, the ODN adopts a stem-loop (or hairpin) conformation, which brings the redox label at close proximity to the electrode surface, enabling facile electron transfer (Figure 13.14, conﬁguration A). Upon binding of a complementary ODN, the stem loop opens up whereupon an intermolecular dsDNA is formed; the distance between the ferrocene and the gold surface is signiﬁcantly altered and the peak current decreases. The same principle was recently used by Jenkins and coworkers who made use of screen-printed electrodes. In these conditions, the assay sensitivity was much increased as it reached 0.11 pM for a 24-mer target representative of Agrobacterium tumefaciens strain C58, which is commonly used for plant gene transformation.425 In contrast to the molecular beacons with ﬂuorescent detection described above, this is rather a ‘light-off’ device. Unfortunately, such systems are more susceptible to false-positive responses. Grinstaff and coworkers devised a related setup (reverse molecular beacon), where the capture probe is a DNA-PEG-DNA triblock carrying the redox ferrocenyl entity at the 5 end (Figure 13.14, conﬁguration B).426 The two DNA

572

Ferrocenes: Ligands, Materials and Biomolecules

Figure 13.14 Different conﬁgurations of electrochemical DNA sensors where the ferrocene label is covalently bound to the capture probe

sequences are complementary of adjacent portions of target DNA. In the absence of target, the 5 -terminal ferrocenyl reporter tag is repelled from the surface by electrostatic repulsion, making it electrochemically inaccessible. Upon target hybridisation, the distance between the redox tag and the electrode is reduced, giving rise to an amperometric signal. Conversely to the hairpin capture probes described above, this biosensor conﬁguration is a ‘signal-on’ method. The estimated limit of detection of this prototype is 200 pM. Another approach, by Inouye and coworkers, was speciﬁcally dedicated to the detection of SNP achieving a quasi ‘on-off’ response.427 The fully conjugated ferrocenyl nucleoside analogue in its phosphoramidite form 188 was synthesised in several steps

The Bioorganometallic Chemistry of Ferrocene

573

H

Fe Ac

I

HN Ac

i

+

Fe

N H

N

N H

N

Ac

HN Ac

OH

OH

ii

HN Ac Fe N

O NC

O P

HN Ac N(iPr)2 188

Scheme 13.54 Synthesis of ferrocenylated nucleoside analogue 188: i) [Pd(PPh3 )4 ]/ Cu(OAc)2 , ii) 182/DMPA/DIPEA

(Scheme 13.54) and coupled by solid phase synthesis to the 5 end of the 15-mer capture probe. Duplex DNAs resulting from hybridisation of the capture probe (which was 3 thiolated) with fully complementary or single base mismatch 16-mer targets (with SNPs at different positions along the sequence) were self-assembled on a gold electrode (Figure 13.14, conﬁguration C). In SVW, a strong anodic peak corresponding to ferrocene oxidation was observed only when both strands were fully matched or when SNP was located at one of the extremities of the target. The mechanism for electron transfer was identiﬁed as hole transport enabled by π-conjugation between the Fc tag and the base pairs. This DNA probe behave as a ‘signal-on’ sensor because hybridisation produced an electrochemical response. Picomolar sensitivity is, in most cases, not enough for direct detection of pathogen DNA without ampliﬁcation. This is why Willner and coworkers devised an original strategy to amplify the ﬁnal amperometric signal (Figure 13.14, conﬁguration D).428 A 5 -thiolated 27-mer capture probe complementary to a portion of the 7229 base length M13Ø viral cyclic DNA was assembled on a gold electrode. Hybridisation with the capture probe was followed by enzymatic replication of target DNA in the presence of polymerase and a mixture of dNTP including a ferrocenyldeoxyuridine triphosphosphate (193a, Scheme 13.56), originally synthesised by Wlassoff and King429 , the capture probe operating as primer. An average of 350 ferrocenyl entities per replica was estimated from electrochemical analysis. These ferrocene units were

574

Ferrocenes: Ligands, Materials and Biomolecules

used as mediator of the biocatalytic oxidation of glucose by GOD, which brought an additional ampliﬁcation step. With these conditions, a limit of detection of 0.1 pM of viral DNA was reached using DPV as electrochemical method. Several strategies have been proposed to incorporate one or several Fc tags onto target DNA for gene detection purposes. Chronologically, the ﬁrst one relies on the use of ferrocenylated PCR primers. In the work published by Takenaka and coworkers, two DNA fragments (a 0.97 kb DNA fragment from the oncogene v-myc and a 0.51 kb fragment from exon 48 of the human dystrophin gene) were PCR-ampliﬁed using ferrocene-modiﬁed and unmodiﬁed primers of 25 bp and 26 bp length.430 In both cases, correct PCR products were obtained and the efﬁciency of ampliﬁcation with the ferrocenylated primers was about half of that of the unmodiﬁed primers. Under low numbers of PCR cycles, DNA ampliﬁcation proceeds exponentially. Thus, a quantitative response was observed by HPLC–ECD over at least two orders of magnitude, and as little as 0.1 fmol of oncogene v-myc DNA was detected reliably. This experiment suggests that ferrocenylated primers can be used for quantitative PCR analysis coupled with HPLC–ECD. An additional advantage of this method is the fact that no reference is required, in contrast to conventional quantitative PCR, in which the sample and a reference DNA are co-ampliﬁed in one vial. Excellent sensitivity was also achieved by coupling capillary gel electrophoresis (CGE) to electrochemical detection of ferrocenylated ODNs. This technique has been pioneered by Kuhr and coworkers.431, 432 Sinusoidal voltammetry (SV) was recommended for selective identiﬁcation of the DNA ampliﬁcation products following PCR. This technique has also been proposed for electrochemical detection of native amino acids and peptides.433 A strategy for so-called ‘four colour DNA sequencing’ has been developed by this group.431 As an example, the T3 PCR primer was covalently 5 -modiﬁed with four different ferrocene derivatives 189–192 (Scheme 13.55). Their redox potentials span a range of 230 mV, but the minimal difference between 189 HO

O

Adenine

O− O P O O

O Fc

N H

O O (CH2)4 Fc O P O− O−

O

O− O P O O

Fc N H

191

O O P O− O−

190

189

Adenine O

HO

Adenine O

Adenine O

O O P O− O−

Scheme 13.55

O O O P O− O−

Fe

O NMe2

192

The Bioorganometallic Chemistry of Ferrocene

575

and 190 is only 34 mV. This difference, which would be difﬁcult to differentiate reliably in a normal CV, is readily discernible by SV. When coupled to CGE separation, the discriminating power of the technique was shown to be applicable to ‘low resolution’ DNA sequencing.431 By combining the single-base PCR extension technique of a ferrocenylated primer with CGE and electrochemical detection, a novel system for single nucleotide polymorphism (SNP) analysis was established. The unextended 20-mer primer could be separated from the 21-mer extension product, thus demonstrating a single-base resolution separation of a DNA oligomer with electrochemical detection.432 Another strategy relies on the incorporation of ferrocenyl tags by use of ferrocene deoxyribonucleotide triphosphate in the course of target ampliﬁcation by PCR. Wlassoff and King prepared two ferrocene dUTP derivatives 193a and 193b and tested them as substrates of standard DNA polymerases (Scheme 13.56).429 Interestingly, only 193a was a good substrate, whereas the homologous 193b was very poorly incorporated. These authors also convincingly demonstrated the higher sensitivity of electrochemical versus UV detection for HPLC analysis of ferrocenylated ODNs.429 In a related publication, the extension of such detection systems to RNA has been described.434 The electrochemically active RNA monomer 193c (Scheme 13.56) was synthesised and incorporated into RNA oligomers in place of U by two different RNA polymerases. Increasing the 193c:U ratio did indeed produce more heavily labelled transcripts, as shown by gel electrophoresis and an increased response in SWV of the RNA transcripts. These ferrocene-labelled RNA oligomers could be immobilised on gold electrodes (SAM), and the system was successfully used for the electrochemical detection of very small amounts of RNA. An electrochemical ‘two-colour’ assay is proposed by using 193c-modiﬁed RNA oligomers along with anthraquinone-labelled UTP monomers.434, 435

O HN O− O− O− − N O P O P O P O O O O O O OH R

H N

n

Fc

O

n = 0, R = H (193a) n = 1, R = H (193b) n = 0, R = OH (193c)

Scheme 13.56 Ferrocenylated (deoxy)uridine triphosphate

Takenaka and coworkers recently published another route for labelling nucleic acids. Their strategy makes use of the reaction of carbodiimide 194a and 194b with the imino moiety of thymine, guanine and uracil bases (Scheme 13.57).436, 437 The stability of duplexes formed by hybridisation of Fc-tagged ssODN with its complementary probe was studied. Electrochemical detection of the hybridisation of Fc-tagged ODN with a complementary capture probe chemisorbed on a gold electrode was achieved with a limit of detection of 50 nM (50 fmol). As both labels 194a and b are oxidised at different potentials, this system may be used for gene expression analysis.

576

Ferrocenes: Ligands, Materials and Biomolecules H N

n

Fe

N +

N C N +

O

N

n = 0 (194a) n = 2 (194b) O

N DNA

O

N H

O

N N

O N H

O + N

N H

Fe n

Scheme 13.57 Reaction of ferrocenylcarbodiimides 194a and 194b with thymine bases in DNA

Ihara and coworkers published a three-ODN system,438 later termed ‘sandwich assay’ (Figure 13.15, conﬁguration A).408 Firstly, an ODN capture probe is immobilised on a gold electrode surface, for example by several phosphorothioate units. The target DNA binds to the immobilised capture probe if the complementary sequence is present. In addition, the target DNA carries a binding sequence complementary to the ferrocene-containing probe DNA, thus giving a ferrocene-labelled target DNA. If the target DNA also binds to the immobilised capture probe ODN, the ferrocene is brought close to the electrode and a signiﬁcant anodic peak due to oxidation of the ferrocene is observed in the DPV of the system. If G is replaced with C, thus generating a single CC mismatch in the target DNA sequence, binding to the immobilised probe becomes much weaker. As a result, fewer ferrocene molecules are held in the vicinity of the electrode and, consequently, only a very small anodic peak is observed in the DPV.438 This system was later characterised in detail, including a thorough analysis of the ODN-modiﬁed surface by IR spectroscopy and a quartz crystal microbalance (QCM) study.439 This detection scheme was more recently reﬁned by Yu and coworkers. In contrast to most other groups, which used 5 -terminal substitution of nucleotides with ferrocene, this group synthesised 2 -ferrocenylnucleotides such as 195 and 196 (Scheme 13.58).440 These compounds, which represent the ﬁrst ferrocenyl-RNA derivatives, are more versatile chemically because they may be incorporated into an oligonucleotide strand at any position and even ODNs with multiple ferrocenyl incorporation are feasible. Whereas the synthesis of the adenosine derivative 195 was straightforward and the corresponding phosphoramidite could be obtained in reasonable yield, the cytidine derivative 196 was obtained together with the 3 -ferrocenylated isomer in a 2:1 ratio. However, the isomers were separated on silica in the form of their 5 -DMT derivatives and could be directly used in solid phase DNA synthesis. In thermal melting studies of mixed sequence 15-mer ODNs, replacement of either A with 195 or C with 196 did not produce a signiﬁcant effect on the melting temperature of the duplices. However, if 196 was replaced with its 3 isomer, the melting temperature of the duplices decreased by 4 ◦ C, comparable to introduction of a single GG mismatch at the same position. The CV of a 15-mer ODN containing 195 showed a reversible wave virtually at the

The Bioorganometallic Chemistry of Ferrocene

577

Figure 13.15 Different conﬁgurations of electrochemical DNA sensors for which the ferrocene label is covalently bound to the signalling probe (sandwich hybridisation assay)

same position as that for a water-soluble ethylferrocene derivative under the same experimental conditions. To alter the electrochemical properties of DNA oligomers in dependence of the DNA sequence, a second A monomer 197 with a different ferrocene derivative attached to the 2 position was prepared (Scheme 13.58).441 The difference in the redox potentials of 195 and 197 of 170 mV is a consequence of the dimethylcarbamoyl substituent at the Cp ring in the latter case. Sequence-dependent electrochemical detection now works

578

Ferrocenes: Ligands, Materials and Biomolecules NHiBu

NHPac N DMTO

N

N

N

O O (iPr)

2N

P

O

DMTO

195

CN

(iPr)

2N

P

DMTO

N

O O O

Fc 3

O

NHCOPh N

N

N O

O

O

Fc 3

O

CN

196

(iPr)

2N

P

N N

O O 197

CN

Fe

O NMe2

Scheme 13.58

as follows. A capture probe (23-mer ODN) is immobilised on a gold surface. The target ODN is known to contain the sequence complementary to the capture probe and binds with high afﬁnity. Finally, the signalling probe binds to another part of the target ODN. Two different 16-mer ODN signalling probes I and II were used; they were extended with three molecules of 195 or 197, respectively, at the 5 end. In addition, both signalling probes differ in sequence only in one base. Depending on the exact sequence of the target DNA, one probe will bind preferentially, bringing different ferrocene derivatives with different redox potentials close to the surface. This difference is easily detected by alternating current voltammetry (ACV). This scheme uses ferrocene probes similar to those of the work of Kuhr discussed above and clearly holds a lot of promise, although the difference in stability (melting temperature) was not reported, nor was any further mismatch sensitivity investigated. More fundamental studies of ODN immobilisation on gold electrodes were also performed using capture probe and target labelled by the two different ferrocene tags.442 Recently, the development of a versatile platform for molecular diagnostics on microarrays based on this chemistry was described in more detail.443 It makes use of a disposable chip with 14 gold microelectrodes on which are co-assembled thiolated capture probes, thiolated PEG (as insulators) and thiolated poly(phenylacetylene) (as molecular wires). This particular interface was designed from a previous report on electron transfer for immobilised ferrocene through molecular wires.444 A sandwich assay format was set up with electrochemically active signalling probes containing 2 -ferrocenylated adenosine (Figure 13.15, conﬁguration A). The system has been successfully tested for sequence-speciﬁc electrochemical detection of PCR-ampliﬁed DNA without the need for further puriﬁcation. The usefulness was further demonstrated for SNP analysis on different genes,443, 445, 446 gene expression monitoring443 and viral pathogen detection.443, 447 This platform is currently commercialised by Osmetech.448 Another signaling probe was described by Mirkin, Nguyen and coworkers.449 ROMP copolymerisation of functionalised norbornene derivatives (Scheme 13.59) followed by coupling to the 5 end of ODN by reaction of the phosphoramidite residues gave polymeric ODN with a high level of ferrocene tags (198). Duplexes resulting from their hybridisation with the target were highly stable and presented sharp melting curves. A sandwich hybridisation assay was carried out by immobilising a 12mer capture probe on a gold electrode followed by addition of a 27mer target and the polymeric signalling

The Bioorganometallic Chemistry of Ferrocene

Et

m

i. Grubbs catalyst

O

O

ii.

O

n

579

Et

O 3

Fc

3

Fe iii.

O

iv. 182 O N P O OH

CN

198

Scheme 13.59 Synthesis of a ferrocenylated polymer for the labelling of ODN

probe including 18mer ODN strands (Figure 13.15, conﬁguration B). Electrochemical detection was done by AC voltammetry and the limit of detection was 0.1 nM of target. A slight modiﬁcation of the sandwich hybridisation format was recently reported by Shen and coworkers.450 In this case, the capture and the signalling probes were designed to display complementary base sequences (at the 5 end for the capture probe and at the 3 end for the signalling probe). Moreover, both capture and signalling probes were complementary to immediately adjacent base sequences of the target and the 5 end of the signalling probe was phosphorylated (Figure 13.15, conﬁguration C). After hybridisation of the target and signalling probe, a ligase was added to repair the nick. This was followed by thermal denaturation which removed the target. Rehybridisation yielded a hairpin DNA, which brought the ferrocenyl tag at close proximity of the electrode, favouring the electron transfer process. A quantitative assay of target was established in the range from 3.4 pM to 0.14 µM with a limit of detection of 1 pM. Two examples of electrochemical DNA sensors are based on the competitive hybridisation of target to dsDNA labelled by a Fc tag immobilised on gold electrodes.451, 452 Both sensors operate in a ‘signal-off’ mode. In the work by Kim, Yang and coworkers the signalling probe consisted of a 12mer ODN with two Fc tags appended at the 5 end. Initially, this signalling probe formed a duplex with the 12-mer capture probe. Upon addition of target under thermal stringency, the signalling probe is displaced, inducing a decrease of the anodic current resulting from ferrocene oxidation (Figure 13.16). Very good sensitivity (subnanomolar) and mismatch sensitivity are achieved without the need for additional washing steps. With the same principle, Hartwich and coworkers developed an integrated electrochemical DNA microarray biosensor for the detection and discrimination of food pathogens. This technology is commercialised under the name of EDDA (electrically detected displacement assay).453 As already mentioned in the introduction to this section, electrochemical transduction of DNA hybridisation events was also performed without the need to label any of the ODNs involved in the reaction, but rather by adding an electroactive compound that will interact preferentially with dsDNA (Figure 13.17, conﬁguration A). Takenaka’s group has successfully used the ferrocene-modiﬁed naphthalenediimide (199)

580

Ferrocenes: Ligands, Materials and Biomolecules

Figure 13.16 Electrochemical DNA sensor operating in the competitive hybridisation mode

Figure 13.17 Electrochemical DNA sensor conﬁgurations using the threading intercalator 199

The Bioorganometallic Chemistry of Ferrocene

581

O N

N H

N

N

Fe

O

N

O

O

N

O

O

N

O

O

N

O

N

199

N

Fe

H N

N

O

H N O

Fe

O

Fe

N H

200

Fe N

O

N

O

O

N

O

N Fe 201

Scheme 13.60 Structures of electroactive threading intercalators proposed by Takenaka

as a threading intercalator (Scheme 13.60).410, 454 To assess this approach, the ﬁrst experiments dealt with the hybridisation of synthetic ODNs and a gene fragment on a plasmid. This system proved to be chemically very robust and highly sensitive, reaching a detection limit of 10 zmol DNA with DPV.410 The sensitivity of the system could be enhanced by coupling it to an enzymatic reaction like the glucose oxidation using glucose oxidase (GOD).455 Another bis-ferrocenyl intercalator 200 was also prepared and was shown to bind preferentially to DNA–RNA heteroduplexes.456 Efforts were

582

Ferrocenes: Ligands, Materials and Biomolecules

also put into the detection of SNPs by this technique. In fact, direct mismatch detection may be linked to the number of intercalated molecules or the rate of electron transfer. In a perfect match situation, one molecule of 199 is intercalated between every second base.457 It is assumed that this number will decrease if the local geometry is perturbed by a single base mismatch and less electrons will be transferred less quickly as a consequence. This approach has been tested on the detection of mutations on the lipoprotein lipase gene. The aim was to set up a tool to discriminate heterozygous from homozygous deﬁciencies. PCR amplicons were prepared from chromosomal DNA and hybridised in the presence of 199 to immobilised capture probes corresponding to every situation met. Perfect match was detected by an increase of anodic signal, whereas no signal increase was observed in case of mismatch.458 Evidently, this system needs careful calibration and probably optimisation for every single application, specially when dealing with ‘real-life’ samples.459 To improve this situation, a new assay format was recently proposed by the same team (Figure 13.17, conﬁguration B).460 Several improvements were brought to the system. Firstly, a multi-array chip (called ECA) was implemented, where different capture probes were immobilised on each electrode. This allowed screening for more mutations. Secondly, ampliﬁcation of gene sequences was performed by asymmetric PCR (to yield one strand in excess over the other) using a special primer. The primer contained a sequence of several bases at its 5 end that was complementary to the portion of DNA immediately adjacent to the mutation point. Consequently, the PCR products have a self-looped form. Thirdly, the mutation site on the capture probes was located at the 3 end. When the PCR product hybridises with the capture probes, dsDNA forms. If both strands fully match, a ligase will be able to repair the nick; if not, denaturation will destroy the duplex and remove the target from the probe. Discrimination of both situations was achieved by addition of 199 in an efﬁcient manner. The Japanese start-up company TUM-gene currently manufactures and commercialises ECA chips and the ECA chip reader to perform genetic testings.461 Several other applications make use of DNA probes in association with ferrocene. Detection of DNA damage was carried out by hybridisation of a ferrocenylated probe ODN with immobilised ODNs on a graphite electrode.462 The amount of hybridised probe DNA is determined as the anodic peak current of ferrocene by differential pulse voltammetry (DPV). If the sample DNA is deliberately damaged, for example by the action of hydroxyl radicals, less probe binds and a smaller signal is detected. This experimental set up was then used to assess the protection efﬁciency of various hydroxyl radical scavengers, and the results were found to correlate well with the literature. A nicked hairpin loop DNA tethered to a gold electrode was used for the electrochemical detection of DNA ligation.463 In its principle, capture probes including sequences of complementary bases were assembled on the electrode surface, resulting in self-looped structures. ODN sequences carrying a ferrocene tag at the 5 end and complementary to the single strand part of the capture probes were hybridised, giving rise to a nicked double strand structure. Ligase, if present, catalyses nick repairing, which keeps the ferrocene tag at close proximity to the electrode after a denaturation/renaturation cycle. If not, denaturation of the hybrid allows washing away of the ferrocenylated ODN, resulting in signal loss.

The Bioorganometallic Chemistry of Ferrocene

583

Takenaka and coworkers synthesised a novel bis-ferrocenyl naphtalenediimide 201 (Scheme 13.60) that is able to bind to tetraplex DNA.464 It provided the basis for an electrochemical telomerase assay. A telomere primer sequence was assembled on a gold electrode. Telomerase catalysed the extension of the telomere primer in the presence of a mixture of dNTP. The extended telomere DNA adopted a tetraplex structure to which 201 bound. A correlation between the number of cancer cells (expressing telomerase) and DPV signal was observed in the range between 40 and 140 cells/µL. The same group made use of their threading intercalator in association with PCR gene ampliﬁcation to detect gene methylation.465 To discriminate a DNA sequence containing a methylated cytosine from a native DNA sequence, the sample is treated by bisulﬁte, allowing conversion of cytosine into thymine. PCR ampliﬁcation of these DNA sequences yields two different PCR products. The one originating from native DNA contains thymines instead of cytosines while the other one contains cytosines instead of methylated cytosines. These sequences are then differentiated with appropriate capture probes immobilised on an ECA chip as previously described.458 The last application of electrochemical DNA sensors is related to the ﬁeld of aptamers. Aptamers are oligonucleotides selected from combinatorial libraries on the basis of their high afﬁnity for target proteins.466 As regards analytical applications, they present noticeable advantages over antibodies as they can be produced synthetically and readily labelled with a variety of markers.467 O’Sullivan and coworkers devised an aptamer-based electrochemical sensor for the quantiﬁcation of thrombin.468, 469 This sensor operates as in the reverse molecular beacon mode with the aptamer carrying a ferrocene tag at the 5 end assembled on a gold electrode by an Au−S bond. Before contact with thrombin, the immobilised aptamer adopts a coil-like structure, hindering efﬁcient electron transfer between the redox label and the electrode. Association of thrombin induces a change of conformation of the aptamer to a tetraplex structure, which brings the Fc tag closer to the electrode. The resulting increase of anodic current measured by DPV can be correlated to the concentration of thrombin in the range between 5 and 40 nM with a limit of detection of 0.5 nM. The sensor is regenerable up to 25 times. 13.4.4

Ferrocene Derivatives of PNA Monomers and Oligomers

Peptide Nucleic Acids (PNA) are a class of DNA analogues in which the ribose phosphate ester backbone is replaced by a pseudo-peptide backbone.470–472 (Scheme 13.61) The nucleobases are linked to this backbone via a carboxymethylene linker.473, 474 PNA binds to complementary DNA or RNA oligomers according to Watson–Crick rules with high stability475–480 and has found applications in molecular biology and as antisense agents.470, 472, 481 The Metzler-Nolte group has prepared the ﬁrst organometallic derivatives of PNA monomers. The T-PNA ferrocene derivative 202 has been prepared (Scheme 13.62), which is similar to the peptide derivative 42 (see Section 13.2.2). The N -Fmoc protecting group can be readily removed under mild conditions and coupling to a second T-PNA monomer is possible, yielding the ferrocenylated PNA dimer 203. Acid 48 could also be coupled to the amino group of PNA monomers using HBTU as the

584

Ferrocenes: Ligands, Materials and Biomolecules B DNA:

B −O 3'

O O

O

5'

O P

−O

O O

O P

O

O n

B

B O

PNA:

N

N H

O

O

O

N

N H

N H n

Scheme 13.61 Comparison of chemical structures of DNA and PNA

N

Fe O

N

O O

NH N

N

Fe

O

NH

O

N

N

O

O

O

NH

NH O

Fmoc

O

NH N

202

203

N

O

O NH Fmoc

O

HN

HN O

N O

N O

O Fe

204

N H

Z

N

O

N O

O OCH3

Fe

N H

O

N

OCH3

205

Scheme 13.62 Ferrocene derivatives of PNA monomers

coupling agent to give compounds 204 (T) and 205 (Z-protected C).482 A ferrocenyl uracil PNA monomer was recently reported by Gasser, Spiccia and coworkers.483 Baldoli, Maiorana and coworkers reported the synthesis of several ferrocenyl derivatives of a chiral tyrosine PNA monomer (as racemic mixtures or in enantiomerically pure form). Monoferrocenylated compound 206 (Scheme 13.63) was synthesised in

The Bioorganometallic Chemistry of Ferrocene O

O

O

O

HN

HN O

N O

CbzHN

HN O

N O

O

N

OMe

206

O

O

N

CbzHN

O

O

N

N H

207

Fc

585

O

OMe

Fc

Scheme 13.63 Ferrocene-labelled tyrosine PNA monomers 206 and dimer 207 O HN N

O

O Fc

ZHN

O H2N

Fc O N H

209

Fc

Fc

O

N

O

O

N O

CbzHN

O

Fc

Fc 208

HN O

OMe O

O

O O

O

N

HN

O

O

OMe Fc O

H N

H N O

210

O

Fc

O

O

O

O

Fc

Fc

Fc

N O

O N H

N H

N

O OMe

O Fc

211

Scheme 13.64 Tris-ferrocenyl PNA monomers 209–211

three steps from protected tyrosine. Grafting of the ferrocene entity to the phenolic group was achieved by a Mitsunobu reaction with 47.484 Condensation with another thymine PNA monomer gave dimeric species 207 (Scheme 13.63). Trisferrocenyl derivatives 209–211 were synthesised from amine 208 (Scheme 13.64).485 The electrochemical behaviour of these PNA derivatives was studied in great detail and it was concluded that, in spite of their bulkiness, they displayed interesting redox properties.486 The three ferrocenyl moieties of the trisferrocenyl PNA compounds behave as independent redox centres, resulting in current intensity multiplied by three with respect to the monoferrocenyl PNA compound. Using DPV as analytical technique, the limit of detection reached 10−8 M.485 The main achievements obtained

586

Ferrocenes: Ligands, Materials and Biomolecules

by this group on organometallic derivatives of PNA were recently summarised in a review article.487 The ﬁrst bimetallic derivatives of PNA monomer 213 were prepared in the MetzlerNolte group.77 This compound was prepared by coordination of the Pt(PPh3 )2 fragment to the PNA alkyne 212 (Scheme 13.65). Like all PNA monomers and single-stranded oligomers, compound 213 exists as a mixture of cis-/trans-isomers at the tertiary amide bond, which was elegantly shown by 31 P NMR. O

O

HN O

HN O

N O

O Fe

N H

N

O

N O

O N H

H Fe

N H

N

O N H Ph3P

212

H Pt

PPh3

213

Scheme 13.65 A bimetallic PNA derivative 213

Finally, ferrocene was also incorporated into PNA oligomers by solid phase peptide synthesis.488 The PNA heptamer Fmoc-tggatcg-Gly was prepared on a solid support by standard Fmoc PNA synthesis methods. After deprotection of the last Fmoc group, the heptamer was reacted with activated 48 on the resin. Cleavage from the resin was achieved by methanolic ammonia with simultaneous removal of the exocyclic protecting groups. Only FcCO-tggatcg-Gly-NH2 (214) was observed in reverse phase HPLC of the crude reaction product (>90 %). To increase stability and solubility in water, the related cobaltocenium carboxylic acid was incorporated into PNA oligomers.489 After initial studies on smaller oligomers to optimise coupling and deprotection conditions, the decamer CpCo+ C5 H4 CO-acc ctg tta t-Lys-OH was synthesised by solid phase synthesis techniques, puriﬁed by preparative HPLC and characterised by MALDI–TOF mass spectrometry. Interaction of this conjugate with complementary DNA was also studied.489 Metallocene PNA oligomers were deposited on Au micro-electrodes and their electrochemical properties were found to be very attractive for electrochemical DNA sensors.489, 490

13.5

Conjugates of Ferrocene with Carbohydrates

Many ferrocene sugar derivatives reported so far are Fem derivatives or esters, thioesters or amides obtained from 48, while the 1,1 -disubstituted ferrocene derivatives are mostly derivatives of the diacid 19. Typical reactions are the same as for derivatives described in previous sections, i.e. alkylation with 173, Schiff base formation with 32 followed by reduction to the amine and reaction with activated ferrocenecarboxylic

The Bioorganometallic Chemistry of Ferrocene

587

acids. Some ferrocene carbohydrate derivatives were tested for biological activity, in particular antimalarial and antiproliferative properties. 13.5.1

Carbohydrate Derivatives of Monosubstituted Ferrocene

Well over two dozen carbohydrate derivatives of monosubstituted ferrocene were reported, with glucose and glucosamine being the most widely studied sugar derivatives. A rather complete table is found in our earlier review.9 Only selected examples from that review and more recent work491, 492 are described herein. The very ﬁrst sugar derivatives of ferrocene were reported in 1961.493 Compound 215b was prepared from alcohol 47 and 2,3,4,6-tetra-O-acetyl-α-D-glucopyranosyl bromide in the presence of silver(II) oxide and calcium sulfate, followed by base hydrolysis of the acetyl protecting groups (Scheme 13.66). This compound was used in a kinetic investigation of hydrolysis of the glucosidic bond.381, 493, 494 Similarly, kinetic results from

OR

OR O

RO RO

Fc

OR

O

RO RO

O

S Fc

OR

R = Ac (215a), H (215b)

R = Ac (219a), H (219b) OR

OR O

RO RO

RO RO

OR

OR O

O

RO RO

OR

OR

HN

N

N

Fc R = Ac (217a), H (217b)

Fc R = Ac (216a), H (216b)

OCH2Ph

OR

OR O

RO RO

R′ N Fc

OR R′ = H (220), CH2Fc (221)

N N

O

RO RO

O

N N

R′′ O

Fe

Fe R′′ N

OR R′′ = RO RO

Fc

O

N N

O

PhCH2O PhCH2O

OR n = 2 (222), 3 (223), 5 (224) R = CH2Ph (a), H (b), Ac (c)

N

226

N n

O

Fc

Fc R = Ac (218)

Fc 225

OCH2Ph

OR′

N R′′ O

O

RO RO

OR NH Fc O

227

R = Ac, H

R, R′ = Ac (228a) R, R′ = H (228b) R = H, R′ = PO3H (229)

OR

Scheme 13.66 Various ferrocene carbohydrate derivatives

588

Ferrocenes: Ligands, Materials and Biomolecules

hydrolysis of a galactose derivative helped to shed light on the mechanism of the enzyme β-galactosidase.495 In a comprehensive preparative study, Adam and Hall reported the synthesis of a wide variety of ferrocene sugar derivatives by different methods.496–498 Glucosamine derivatives 216 and 217 were obtained by Schiff base formation of glucosamine derivatives with 32,497 followed by reduction.499 From this reaction, the N ,N -bis(Fem) derivative 218 could also be isolated,499 which parallels the isolation of (Fem)2 GlyOMe (39) in amino acid chemistry described in Section 13.2.3. The same compounds were also readily prepared by reaction of glucosamine with FcCH2 OTs.497 Thiosugar derivatives such as 219 were prepared with the same reagents (Scheme 13.66). More thiosugar derivatives were also reported by a Spanish group.492 A few glucopyranosylamine derivatives such as 220 and 221 (R = Ac) were obtained from glucopyranosylamine under retention of the stereochemistry at the anomeric carbon atom.500 Careful hydrolysis with Amberlite IRA 400 (OH) resin in methanol yielded the unprotected sugars (R = H). In their earlier study, Adam and Hall were unable to obtain the deprotected derivatives without decomposition.497 Only a relatively stable thiosugar derivative could previously be deacetylated with NaOMe in methanol. The preparation of three imino sugar derivatives (of glucosamine, galactosamine and mannosamine) from ferrocene was also reported by Schneider and Wenzel in 1979.501 Catalytic hydrogenation in ethanol in the presence of PtO2 did not just reduce the C=N double bond but also the amino sugar aldehyde group, yielding ferrocene methyl derivatives of amino sorbit, amino dulcit and amino mannit. However, only mass spectrometric data for these compounds were reported.501 In a recent study, Orvig and coworkers extended the range of available compounds by preparation of some 6-N and 6-O-ferrocenoyl carbohydrates and fc(CO2 H)2 derivatives. A number of compounds in this study were characterised by X-ray crystallography. All compounds were relatively nontoxic in two cell lines by the MTT cytotoxicity test. Only two compounds showed moderate antimalarial activity against two different strains of P. falciparum, one of them resistant against the common antimalarial drug chloroquine (see Table 13.9 and Scheme 13.94).491 Another class of water soluble ferrocene carbohydrate conjugates was reported by Robinson and coworkers. They prepared benzyl derivatives 222a–224a which could be deprotected by careful catalytic hydrogenation, yielding the fully unprotected 222b– 224b.499 Compound 222b was also prepared by a simpler route via the acetylated derivative 222c which was successfully deacetylated using the IRA 400 (OH) resin. The anomerically pure β-D-glucopyranosyl derivative 225 was obtained in a C−C bond forming reaction in 31 % yield via the Friedel–Crafts alkylation reaction from an anomeric mixture (α:β = 4:1) of benzyl protected 1-O-acetyl-D-glucopyranose.502 The copper-catalysed [2 + 3] cycloaddition reaction between alkynes and azides (‘Click chemistry’)503, 504 has also been successfully employed for the synthesis of novel ferrocenyl carbohydrates under mild conditions.492 The cycloaddition of propargyl glycosides and azidomethyl and bis(azidomethyl)ferrocene yields 1,2,3-triazoles as exempliﬁed for the glucose derivatives 226 and 227. Likewise, the reaction of 69 with azido-sugars yields isomeric ferrocene carbohydrate conjugates. Depending on the substitution pattern of the cyclopentadienyl rings, these compounds span a range of ca. 300 mV for the reversible electrochemical oxidation of the ferrocene unit.492

The Bioorganometallic Chemistry of Ferrocene

589

Kraatz et al. reported the synthesis of 228b and 229 directly from activated 48 and unprotected glucosamine and glucosamido-2 -phosphate.505 These water soluble ferrocene derivatives were subjected to electrochemical studies in aqueous solution at different pH values. The oxidation of 228b is fully reversible between pH 2 and 9. At pH 12, however, the oxidation becomes irreversible and a mechanism of decomposition is proposed. 13.5.2

Carbohydrate Derivatives of Ferrocene-1,1 -dicarboxylic Acid

The 2,3-(ferrocene-1,1-dicarbonyl)-O-α-D-glucopyranoside (230) has been the key starting material for a number of derivatives, as summarised in Scheme 13.67. Applications as chiral matrices were evaluated by Itoh and coworkers,506–508 and medicinal applications were proposed by the groups of Itoh509, 510 and Keppler,511 and Orvig.491 Compound 230 is readily prepared by reaction of fc(COCl)2 (119) with 4,6-benzylidene-O-methyl-α-D-glucopyranoside to yield 231, followed by deprotection (Scheme 13.67).509, 511 Reaction of 230 with 49 or 119 yields 232510 and 233,511 respectively. From 230, the disaccharides 234 and 235 and the trisaccharide 236 were obtained in about 50 % yield.510 Itoh’s group has also prepared ferrocenoyl derivatives 237–239 of ellagitannins (Scheme 13.68),509 which constitute the main curative and palliative ingredient in various traditional herbal medicines. The chiral biphenyl derivative 238 as well as the binaphthyl derivative 239 were synthesised in enantiomerically pure form.509, 510 All ferrocene derivatives 230–239 as well as a few other ferrocenecarbohydrates were tested for their antimalarial activity against a chloroquine-sensitive P. falciparum strain and were found to be less active than quinine.510 Only (SD)-238 showed an EC50 value in the high nM range (see Table 13.9). The same set of compounds was found to have very little cytotoxicity, and similar nontoxicity was found by Keppler and coworkers for 230 in four other tumor cell lines.511 More recently, Orvig and coworkers reported antimalarial and cytotoxicity data for a range of monoand disubstituted ferrocene carbohydrate conjugates.491 Credi and coworkers have recently published an extensive study on ferrocenecontaining carbohydrate dendrimers.512 Compounds 240–243 (Scheme 13.69) were prepared by coupling of the ferrocenecarboxylic acid chlorides with amino group of the protected sugars. Deprotection of the benzyl or acetyl derivatives (240–243a) was claimed to be quantitative. Electron transfer reactions of the ferrocene core were thouroughly investigated for all compounds. 13.5.3

Other Carbohydrate Derivatives

In a few publications, cyclodextrins were covalently derivatised with ferrocene derivatives.513–515 Most cases were concerned with electrochemical detection or studies of electron transfer. CD spectroscopy is a powerful tool for characterisation and molecular modelling was used to explore the structures of the ferrocene-cyclodextrin conjugates.516, 517 Interestingly, ferrocene also binds noncovalently into the cavity of β-cyclodextrin.518 This host-guest interactions was used to obtain peptide nanotubeferrocene cyclodextrin self-assembled monolayers on gold surfaces.519 So far, Vasella’s group has prepared all of the chemically unusual ferrocene sugar derivatives that we are aware of.520–22 Cyclopentadienyl C-glycosides have been

590

Ferrocenes: Ligands, Materials and Biomolecules Ph

O O O

O O O

OH

O OCH3

O Fe

O

O HO

O

O

O Fe

O O 232

OH

Fe

O O O

235

O

OCH3 Fe

Fe

O O

O

O

230

O

O O

O O O O

OCH3 233

Fe

Fe

OMe

OMe O

MeO MeO

OMe O

O O

OMe O O

MeO

O

234

O

MeO MeO

HO O

+

O OCH3

O

O OMe O

MeO MeO

O O 236

Fe

Fe

Scheme 13.67 O

O O

O Fc

O

O O

MeO MeO

Fc

OMe MeO

O

O Fc

O O

OCH3

Fe

O

O OAc HO O

O

O

HO O OAc AcO AcO

O O

OCH3

OH 237

231

O

O O

O

O

OH

O

HO

Fe

O OH

O

O

O O O O

OCH3

OMe OMe

238

239

Scheme 13.68

Fc

O OCH3

O OCH3

O OCH3

The Bioorganometallic Chemistry of Ferrocene

591

OR OR O

RO RO

O

OR O

NH

OR

O

RO RO

O

NH

Fe

N H

O O

OR OR

RO

OR

Fc O

O

R = Bz (240a), H (240b)

R = Bz (241a), H (241b) OR

OR O

RO RO

OR O

OR

RO RO

NH

OR

Fc O

OR R = Ac (242a), H (242b)

RO RO

RO

OR

O

O

N H

O OR

O

O

O O

OR

O

RO RO

O O

RO RO

O OR

RO RO

OR

NH

Fe

O O RO

OR O

O O

O O

OR OR OR OR OR OR OR OR

RO

OR R = Ac (243a), H (243b)

Scheme 13.69

prepared as latent fulvenes.520 Lithium aluminium hydride reduction, followed by silyl ether formation gave the fully protected cyclopentadienyl mannitols 244a. In a similar fashion, reaction with PhLi, followed by silyl ether formation gave 244b. Lithiation of 244, followed by addition of FeCl2 or a Cp*Fe synthon, then gave the ferrocene derivatives 245a,b and 246. Separation of the three diastereomers of 245b, which formed in a nonstatistical ratio, by preparative HPLC afforded stereochemically pure products (Scheme 13.70). The conﬁguration of these stereoisomers was deduced from 1 H NMR data. Related sequences of reactions gave the two metallocenes 247 and 248 derived from ribose and arabinose, respectively. It is of interest to note that stereochemically almost pure compounds 247b were obtained in tetrahydrofuran, in contrast to a low diastereomeric excess for 245b. Annulated ferrocene derivatives of carbohydrates were reported by the same group.521, 522 The dimesylate of 1,3,4,6-tetra-O-methyl-D-mannitol (249) reacted with CpNa to a spiro[4,4]nona-1,3-diene. Thermolysis of this compound afforded three isomeric tetrahydroindenes, which could be transformed into the annulated C2 -symmetric ferrocene 250a by lithiation and reaction with FeCl2 (Scheme 13.71).522 C1 -symmetric annulated ferrocenes were obtained from 2,3:5,6-di-Oisopropylidene-D-mannitol (251) or the corresponding tetra-O-methyl derivative.521 In this case, a mixture of 252, 253 and 254 was obtained with a slight preference for 252a and 253b, respectively, over the statistical ratio (Scheme 13.72). The isopropylidene derivatives could be deprotected by HCl in MeOH, yielding 252c–254c. All these deprotected derivatives were quite soluble in water (at least 100 g/L). In a related fashion, 250b and c were obtained (Scheme 13.72, left).521 Compound 252a was readily converted into the diiodo derivative (252e) via chloromercuration (252d, Scheme 13.71, right).

592

Ferrocenes: Ligands, Materials and Biomolecules

O O R O O OR′

H Fe

Fe

O O R O O OR′

O O H O O OR′

H

245, R′ = SiMe2(t-Bu) R = H (a), Ph (b)

H

246, R′ = SiMe2(t-Bu)

R

O Ph

R

R

S

O O O

O O O O H

Fe

H Fe

O Ph

R

O O O O H

O O O

247, R = H (a), Ph (b) (Stereochemistry applies to b only)

248

S

H

Scheme 13.70 R 1O

O O

OR1

R 2O

OR2

O O R R

O

Fe

R 2O

OR2

R 1O

O Fe

O

O

OR1

250, R1 = R2 = Me (a), Bn (b) R1, R2 = PhCH (c)

R = H (252a)

HgCl (252d)

I (252e)

Scheme 13.71 R1O R2O 2 R O

R1O R2O OR2

R2O

OR1

R1O

Fe

R1O

OR2

R1O R2O R2O

Fe

OR2 252

OR1

R1, R2 = Me2C (a) R1 = R2 = Me (b) R1 = R2 = H (c)

253

Scheme 13.72

OR2 Fe

OR1 OR2 OR1

R1O

OR1 OR2 OR1 254

The Bioorganometallic Chemistry of Ferrocene

13.6

593

Ferrocene Conjugates with Other Biomolecules

In this section, ferrocene conjugates with other molecules of biological relevance are covered, such as ligands for endogenous receptors, lipids and drugs. In this context, the section closes with electrochemical immunoassays. The focus is on molecules with importance for humans; no effort was made to cover biomolecules that are important only in lower organisms, such as plants (e.g. growth regulators) and insects (e.g. pheromones). 13.6.1

Steroid Hormones

Peptide hormones such as enkephalin have already been covered in Section 13.2. The ﬁrst reports of steroid hormones labelled with the ferrocene moiety appeared in the literature in 1977.523, 524 Riesselmann and Wenzel esteriﬁed 48 with the 3OH group of estradiol and estrone.523 Subsequently, metal exchange reactions with 103 RuCl afforded the radioactive ruthenocene estradiol and estrone conjugates. These 3 derivatives were used to study the organ distribution and clearance properties in mice. Later, a number of schemes for immunoassay using metals as labels were reported, including Metalloimmunoassays (MIA) with AAS detection,524, 525 Carbonyl Metalloimmunoassay (CMIA) with IR detection,526, 527 and various amperometric detection schemes (see section 13.6.4) have also been explored. In particular, the steroid hormone estradiol 255 has been subject of extensive labelling with the ferrocene group (Scheme 13.73 and Table 13.8). Two additional ferrocene conjugates, 17α-ferrocenyl-17β-estradiol (256)528–530 17α-ferrocenylethynyl17β-estradiol (257)531, 532 (Scheme 13.74), were also prepared and found to have a relative binding afﬁnity (RBA) of 8 % and 28 % to the estrogen α-receptor (ERα). The ERα is the ﬁrst receptor subtype that was discovered and cloned.533 In 1996, a second estrogen receptor, termed ERβ, was identiﬁed and cloned.534 Thus far, only the afﬁnity of 257 for the ERβ has been determined to be 37 % of estradiol. R2 R3 R1 R1

= OH, R2 = OH, R3 = H (estradiol, 255)

Scheme 13.73 Estradiol and derivatives (see Table 8 for residues R1 − R3 )

Two reports on ferrocene derivatives of the female sexual hormone estrone 258a, namely 258b and c, were reported in the literature (Scheme 13.75, left).531, 536 Shimada et al. also reported labelling of the position 3 glucoronides of estradiol and estrogen with the Fem moiety for HPLC–ECD investigations,538 as well as derivatisation of hydroxysteroids, such as estrogen and dehydroepiandrosterone, with various ferrocene reagents for the same purpose.539 Only one example of a ferrocene-labelled conjugate 259b of the male sexual hormone testosterone 259a has been reported (Scheme 13.75, right).523, 535

594

Ferrocenes: Ligands, Materials and Biomolecules Table 13.8 Ferrocene derivatives of estradiol (Refer to Scheme 13.73 for residue numbering) R1

R2

R3

Ref.

FcCO2

OH

H

523

OH

FcCO2

H

523, 535

FcCO2

FcCO2

H

535

OH

FcCH2 NHC(O) (CH2 )2 CO2

H

524, 525, 536

OH

FcCH2 NHC(O) (CH2 )2 CO2

FcCH2 NHC(O) (CH2 )2 CO2

525, 536

FcCH2 NHC(O) CH2 O

OH

H

536

FcCH2 NHC(O) CH2 O

OH

OH

536

OH

NHC(O)(CH2 )2 C(O) NHCH2 Fc

H

525

CH3 O

OHa

FcCH=Nb

537c

CH3 O

OHa

FcCH2 NHb

537c

a The 17-α ferrocenyl substituted derivatives 256 and 257 (Scheme 13.74) are not included in this table. b Compound with the inverted geometry at this carbon atom was also reported. c Single crystal X-ray structure is reported.

OH C C

OH

Fe

Fe HO

256

HO

257

Scheme 13.74 Two ferrocenyl-17α-estradiol derivatives O R

O R2 R1

O

R1 = OH, R2 = H (estrone, 258a) R1 = OH, R2 = CHFc (258b) R1 = OCH2C(O)NHCH2Fc, R2 = H (258c)

R = H (testosterone, 259) R = C(O)Fc (259a)

Scheme 13.75

The Bioorganometallic Chemistry of Ferrocene

R2 R1

R1

R2

Reference

FcCO2 FcCH2CO2 Fc(CH2)4CO2 FcCH2NH FcCH2N(Ac) FcCH2NH

H H H H H OH

535 540 540 537 537 537

595

Scheme 13.76

A variety of ferrocene cholesterol derivatives have been prepared, as shown in Scheme 13.76.535, 537, 540 More recently, a steroid derivative with some resemblance to cholesterol has been reported by Coutoli-Argyropoulou et al. via Sonogashira coupling.541 In his PhD work, Koller synthesised a variety of steroid derivatives in which a ferrocene nucleus replaced the A-ring.542 Several of these derivatives were screened for antibacterial and pepsin-inhibiting activity. At concentrations of 100 µg/mL, all of these compounds exhibited antibacterial properties, whereas only one was found to display signiﬁcant pepsin-inhibiting activity. 13.6.2

Lipids

A few papers have been published on ferrocene conjugates of lipids. Nambara and coworkers reported the determination of ten fatty acids by HPLC–ECD after derivatisation with 3-bromoacetyl-1,1 -dimethylferrocene (260).543 The method showed a detection limit as low as 0.5 pmol and was suitable for direct determination of various fatty acids in human serum. More recently, Hailes and coworkers prepared ferrocenyl conjugates with fatty acids.544, 545 To increase solubility, oligo(oxo-ethylene) spacers were inserted between the ferrocene and the fatty acid. A thorough electrochemical investigation was performed.545 Moreover, different binding sites on human serum albumin (HSA) could be identiﬁed depending on the length of the oxo–ethylene spacer.544 A conjugate of ferrocene with glycerol has also been reported.546, 547 Sokolov and Troitskaya used the dimeric ferrocenyl-organopalladium complex 261 to synthesise optically active glycerol derivatives 262 via a diastereoselective carbonylation reaction (Scheme 13.77). A ferrocenylcholine derivative has been prepared and tested for NMe2 NMe2 Fe

Pd Cl Cl Pd

O Fe

+

CH2OH HC OBn CH2OH

Me2N

CO

CH2O HC OBn CH2OH 262

261

Scheme 13.77

Fe

596

Ferrocenes: Ligands, Materials and Biomolecules

its ability to inhibit the hydrolysis of butyrylcholine by the enzyme Horse Serum Butyrylcholinesterase.548 The Ki of the ferrocenylcholine conjugate was determined to be 9.63 × 10−6 L/mol. Another very interesting application uses ferrocene-containing cationic lipids as transfection agents.549 The cationic, two-tailed lipid bis(11-ferrocenylundecyl)dimethylammonium bromide (BFDMA, 263 in Scheme 13.78) can be reversibly oxidised, changing the overall charge of 263 from +1 (reduced) to +3 (oxidised). Whereas reduced 263 forms sub-micrometer vesicles in aqueous solution, oxidised 263 only forms smaller, micellar aggregates. Transfection experiments were performed with 263 as the transfection agent and a plasmid construct that encoded for an enhanced green ﬂuorescent protein (EGFP). Fluorescence microscopy images of the transfected cells clearly showed high (for reduced 263) or very low transfection yields when the oxidised 263 was employed. The results thus indicate an interesting principle of electrochemical activation for DNA transfection.549 H3C Fc

CH3 N+

9

Fc 9

263red

− 2 e− +2

e−

+

H3C Fc

CH3 N+

9

Fc

+

9

263ox

Scheme 13.78 A redox-activated ferrocenyl transfection agent

13.6.3

Drugs

Several drugs were labelled by a ferrocene entity, with the view of using them in metalloimmunoassays with electrochemical detection (see following section). The labelling of the antiarrhythmic drug lidocaine was performed by acylation of p-aminolidocaine with 1 ,3-dimethylferrocenoyl chloride to yield 264 (Scheme 13.79).550 Moise and coworkers introduced a ferrocenoyl entity into the amine-containing drugs amphetamine, metamphetamine, desipramine, nortriptyline and norfenefrine and the biogenic amine histamine by acylation with 49.551 Limoges, Degrand and coworkers reported the synthesis of a cationic ferrocenyl derivative of the antiepileptic drug phenytoin 265 (Scheme 13.79) in two steps.552 The sodium salt of phenytoin was alkylated with dibromopropane to which was added FcCH2 NMe2 . A ferrocene entity was attached to theophylline, a drug prescribed in the treatment of respiratory diseases.553 Condensation of ferrocenylalkyl carboxylic acids with 5,6diamino-1,3-dimethyluracil yielded ferrocenylalkyl-theophylline conjugates 266 with different alkyl chain lengths (Scheme 13.79). Functionalisation of the unsubstituted cyclopentadienyl ring is also possible either by Friedel–Crafts acetylation or by methylation. A Mannich reaction yields tertiary amino derivatives 267 (Scheme 13.79), although as a mixture of isomers which could not be readily separated. Finally, a ferrocenoyl entity was attached to the immunosuppressive drug FK506 (Tacrolimus) by reaction of ferrocenoylamido ethylamine with an N -succinimidyl ester derivative of FK506 to give compound 268 (Scheme 13.79). This compound was used in an electrochemical receptor binding assay (see following section).554

The Bioorganometallic Chemistry of Ferrocene

Fe

H N

Ph Ph

H N

O

N

Me O

O Me

O

N

N H

597

+

NMe2 Br− Fc

264 H N

n

Fe

N N H3C

265 +

Me3NH2C

O

O

CH3 N

N CH3

O

HN

Fe

N

O

N

CH3

267

266 (n = 1, 2, 3, 4, 8)

OMe OMe

O H N

Fc O

O N H

O

O

O

O

OH O N

O HO

268 OMe OH

Scheme 13.79 Ferrocenyl tracers of lidocaine 264, phenytoin 165, theophylline 267 and Tacrolimus 268

13.6.4

Ferrocene-Based Immunoassays with Electrochemical Detection

The general principle of immunoassays relies on the observation that, in a system containing the analyte and a speciﬁc antibody, the distribution of the analyte between the bound and free forms is quantitatively related to the total analyte concentration.555 In the original assay format (radioimmunoassay or RIA) designed by Yalow and Berson in the early 1960s,556 the distribution of the analyte between the bound and free forms was monitored by adding a known amount of radioisotopically labelled analyte (called the tracer) to yield analyte–antibody and tracer–antibody complexes. After both forms of analyte/tracer are separated from each other by an adequate method, one of the forms is quantiﬁed accordingly by radioisotope counting and a standard curve established by plotting the concentration of free or bound form of tracer versus the concentration of analyte in standard samples (Figure 13.18). This type of assay is denominated by the adjective ‘competitive’ as analyte and tracer compete for binding to antibody. It is also referred to as ‘heterogeneous’ as it

598

Ferrocenes: Ligands, Materials and Biomolecules

Figure 13.18 The principle of competitive immunoassays with a radio-labelled tracer

requires a separation step. Less commonly, some assays do not require separation of the free from the bound fraction of tracer as the physicochemical properties of the tracer change upon binding to antibody. These assays are denominated as ‘homogeneous’. A few years after the ﬁrst radioimmunoassay was described, Cais proposed a new type of immunoassay that he coined Metalloimmunoassay (MIA). The radiolabel is replaced by a metal atom embedded in an organometallic complex such as ferrocene and AAS was proposed as the detection method.524, 557 Because of endogenous contamination of samples by iron, the objective never came to completion. The ﬁrst indication of using ferrocene as an electrochemical tag in an immunoassay dates from the late 1970s when Weber and Purdy reported the labelling of morphine with a ferrocene entity. The fact that the redox behaviour of this tracer in the presence of antimorphine antibody was altered opened the way to an homogeneous electrochemical immunoassay.558 In fact, the ﬁrst complete immunoassays employing ferrocene as label were independently published by two groups in 1986. Both are homogeneous. In the work reported by McNeil and coworkers and dealing with the assay of the antiarrhythmic drug lidocaine,550 the ferrocenylated tracer 264 mediates electron transfer from GOD to the electrode whereas addition of antilidocaine antibody inhibits it, yielding a decrease of the bioelectrocatalytic current in the presence of glucose. Further addition of free lidocaine reverses this behaviour because of the competitive binding of the analyte to the antibody (Figure 13.19, conﬁguration A). The same principle was employed by Robinson and coworkers to assay the hormone thyroxine using tracer 269 (Scheme 13.80). The limit of detection reached 15 nM in this case.559 Slightly after these pioneering reports, Yao and Rechnitz published another immunoassay of thyroxine using a similar reagent 270.560 Another difference was that this time GOD and the antithyroxine antibody were coadsorbed on the electrode instead of being in solution. With this format, a limit of detection of 15 nM was also reached. This approach was resumed a few years ago by Forrow and coworkers to set up an assay for theophylline, using the cationic tracer 267 (Scheme 13.79).553 A calibration curve was established for concentrations of analyte ranging from 10 to 100 µM. The cationic tracer 267 was preferred to another ferrocene labelled theophylline derivatives 266 because of the lower nonspeciﬁc binding to the antibody. The same principle was recently exploited by Kobatake and coworkers to set up a homogeneous

The Bioorganometallic Chemistry of Ferrocene

599

Figure 13.19 Electrochemical immunoassay conﬁgurations using ferrocenylated haptens as tracers or ferrocenylated enzyme substrates: (A) homogeneous assay, (B) homogeneous-like assay with Naﬁon-coated electrode, (C) enzyme-ampliﬁed homogeneous-like assay

600

Ferrocenes: Ligands, Materials and Biomolecules I HO

I O NHR2

R1

I

HO2C

R1 = I, R2 = FcCO (269) or FcCH2 (270) R1 = H, R2 = FcCO (271)

Scheme 13.80

electrochemical receptor binding assay aimed at drug screening, using the ferrocenyl derivative of Tacrolimus 268 as electron mediator and the protein FKBP12 as protein receptor.554 Another form of electrochemical immunoassay was proposed by Degrand, Limoges and coworkers.335 Its principle is shown schematically in Figure 13.19, conﬁguration B. It is based on the electrochemical detection of a cationic electroactive label combined with NaﬁonTM -coated carbon electrodes. Naﬁon is a polyanionic perﬂuorosulfonated polymer that, because of its structure and charge, is able to operate a selection on the molecules contained in the solution in contact with it, so that only the cationic small molecules (i.e. the free form of the electroactive tracer) are able to penetrate the ﬁlm and accumulate at the electrode surface. This is an elegant approach for separating the bound fraction of tracer from the free while accumulation leads to ampliﬁed anodic current which in turn results in a homogeneous-like immunoassay. The cationic ferrocenyl tracer of phenytoin 265 together with a cobaltocenium tracer of the antiepileptic drug phenobarbital were used in a simultaneous dual electrochemical immunoassay.561 The analytical range of the resulting assay was 0.2–50 µM for both analytes with a limit of detection of 0.5 µM. More recently, Wang and coworkers devised an heterogeneous competitive electrochemical immunoassay of tri-iodothyronine (T3 ) using the ferrocenylated tracer 271, for which the free fraction of tracer was separated from the bound fraction by capillary electrophoresis.297 A calibration curve was established for concentrations of T3 ranging from 2 to 8 µg/mL with a limit of detection of 1 µg/mL. In the early 1970s, the ﬁrst immunoassays using enzymes as reporter groups (ELISA) were reported.562 For competitive ELISAs, the tracer is an enzyme chemically conjugated to the analyte that catalyses the conversion of a substrate to an easily detectable product. Due to the high turnover of enzymes, signal ampliﬁcation results from this assay format, enabling highly sensitive assays to be developed accordingly. The most popular methods for quantifying enzyme products today are spectrophotometric. However some examples of amperometric detection of enzyme products have also been reported in the literature.563 This strategy is advantageous to colorimetric assays when dealing with cloudy or even opaque samples. Two categories of ELISA using ferrocene derivatives have been reported in the literature. One of them uses simple ferrocene compounds as redox mediators of oxidoreductases (GOD or HRP). The other approach, taken by Limoges, Degrand and coworkers, makes use of specially designed enzyme substrates including a ferrocenyl moiety. For reasons of limited space, only the second one will be reviewed herein.

The Bioorganometallic Chemistry of Ferrocene O Fe

H N

OH P OH O Me

Fe O

O 272

HO Me

601

O P OH

273

Scheme 13.81 Ferrocenylated substrates of ALP

These authors synthesised the phosphate derivative 272 (Scheme 13.81) and tested it as a substrate of ALP, one of the most commonly used enzymes in ELISA.564 ALP was able to catalyse the hydrolysis of this compound to yield the corresponding alcohol. The enzymatic product (in the oxidised ferrocenium form) was selectively entrapped and accumulated under potential within the Naﬁon ﬁlm deposited on the carbon electrode. This resulted in a very sensitive electrochemical detection of ALP with a limit of detection of 0.02 U/L (0.028 pM) by SWV. An electrochemical ELISA was set up for the antiepileptic drug phenytoin, where the antiphenytoin antibody was adsorbed on the wall of plastic tubes and a conjugate of phenytoin and ALP was used as tracer565 (Figure 13.19, conﬁguration C). A calibration curve was established in the range between 10 and 100 µM of phenytoin. The same authors synthesised another ALP substrate 273 (Scheme 13.81) and used it to set up an afﬁnity sandwich assay of avidin for which a biotin–BSA conjugate was adsorbed on the wall of plastic tubes and biotin–ALP was used as tracer.566 A calibration curve for avidin was set up in the range between 0.5 and 100 µg/mL with a limit of detection of ca. 5 nM.

13.7

Medicinal Chemistry of Ferrocene and Selected Ferrocene Derivatives

In view of their relatively easy synthesis and benign properties under physiological conditions, it is perhaps not surprising that numerous ferrocene compounds were synthesised and tested for their biological activity. In fact, the physiological chemistry of ferrocene itself is of interest as it was the ﬁrst organometallic compound for which antiproliferative properties were reported. This report sparked the development of organometallic anticancer compounds.567, 568 Although ferrocene-derived anticancer compounds still comprise a signiﬁcant portion of all medicinally relevant ferrocene research, many more indications were investigated. Synthetically, ferrocene often simply replaces an organic group, mostly phenyl, in an existing drug to give the new compound. This concept is called bioisosterism and is frequently employed in medicinal chemistry. Although it is ball-shaped rather than ﬂat, most researchers consider ferrocene bioisosteric to phenyl rings. Frequently, such bioisosteric replacement has little effect on the biological activity, as for instance in ferrocene penicillin derivatives. It may, in some cases, lead to a modiﬁed activity, as will be shown for the ferroquine example below. The real challenge, however, is to disclose new modes of action,

602

Ferrocenes: Ligands, Materials and Biomolecules

which are solely dependent on the nature of the metallocene and cannot be achieved by purely organic compounds. One such example is the ferrocifen series of compounds discussed in Section 13.7.3. There may, however, be other examples where the mode of action has not been discovered yet, or was simply – but incorrectly – assumed in analogy to the organic drug. Compounds described in this section are of course related to bioconjugates described in the chemically related previous sections of this chapter. Drugs that were labelled with a ferrocene entity solely for bioanalytical purposes have been described in Section 13.6.3. Our choice of compounds has been limited mostly to ferrocene compounds that were synthesised with a concept in mind. Not included are ‘random’ ferrocene compounds, that were (maybe inter alia) tested for biological activity. There is a justiﬁcation for this negligence. Although, of course, random screening has the potential for the discovery of novel drugs with spectacular activity, there are, to the best of our knowledge, no such examples in the ferrocene ﬁeld. 13.7.1

Physiological Chemistry of Ferrocene and Derivatives

Ferrocene by itself is not a particularly toxic compound. It can be injected, inhaled or taken orally. Like most xenobiotics, it is degraded in the liver by cytochromes. Because of its aromatic character, a metabolism related to benzene can be expected and was indeed found experimentally. As shown in experiments with rats that were orally given a single dose of ferrocene in sesame oil, ferrocene is hydroxylated enzymatically in the liver and excreted urinally in the form of conjugates to sulfate (minor product) and glucuronic acid as the main product.569 In vitro, intact liver microsomes, NADPH and molecular oxygen were found necessary for the hydroxylation of ferrocene. This process was inhibited in vitro by CO, but signiﬁcantly stimulated in vivo by pretreatment of the rats with phenobarbital. These ﬁndings give a conclusive evidence that the hydroxylation of ferrocene is carried out by cytochrome P450 enzymes, much like benzene and indeed many other hydrocarbons are. On the other hand, hydroxyferrocene is rather unstable and decomposes in aqueous solution, ﬁnally releasing solvated iron atoms. It is thus expected that at least some of the iron is released from ferrocene upon liver metabolism. After oral administration of 59 Fe-labelled ferrocene, <10 % of the dose was excreted in the faeces within three days, compared to >93 % of 59 FeSO4 in rats, guinea pigs or mice. After ﬁve days, between 5 and 40 % of the total dose was excreted, depending on the species.570 Using radioactive labelled 59 Fe and 103 Ru metallocene carboxylic acids, retention of radioactivity in blood and the presence of free iron in urine of mice indicates that the ferrocene derivative is degraded in vivo while the ruthenium analogue is not. The effect of ferrocene and a number of alkyl derivatives on rats which were made anaemic by an iron-free diet and regular bleeding was investigated in some detail. In these animals, a single oral dose of 100 mg kg−1 of ferrocene produced an increase in haemoglobin of about 65 %. This value was reached three weeks after dosing and was only slightly lower than the increase reached with the same dose of iron in the form of the well-established haematinic Imferon, an iron-dextran complex. On the other hand, all of a group of ﬁve mice died after a single oral dose of ferrocene corresponding to 250 mg kg−1 . Ferrocene is more toxic when injected intraperitoneally as an aqueous suspension. Thus, seven out of ten mice died after a single injection corresponding

The Bioorganometallic Chemistry of Ferrocene

603

to 100 mg kg−1 . The toxicity of ferrocene was also tested in dogs which were fed daily up to 300 mg kg−1 for six months, or even 1 g kg−1 (sic!) up to three months.571 While no acute toxicity or even deaths were observed, massive chronic iron overload was diagnosed and the dogs recovered afterwards. The ferrocene-induced hepatic iron overload could be reduced after the removal of large quantities of iron by repeated venesection.571 Similar experiments with ferrocene derivatives revealed the following trends. Ferrocene mono and dicarboxylic acids and mono and disulfonic acids were at least as toxic as ferrocene. Like benzoic acid or benzene sulfonic acid, they are excreted in the urine, presumably undegradated but conjugated. Acyl derivatives were generally more toxic than the alkyl and phenyl derivatives tested, and disubstituted ferrocenes were found to be less toxic than the corresponding monosubstituted compounds. The higher homologues with seven or more carbon atoms in the side chain are not well absorbed, while the very small derivatives (with one to three carbon atoms in the substituent) were at least as toxic as ferrocene and partially excreted urinally. 1,1 -Dineopentylferrocene (274) was chosen as an example of a relatively non-toxic and well-absorbed ferrocene derivative. When given as an aqueous dispersion to rats, most of the nonabsorbed ferrocene was excreted within three days in the faeces, corresponding to less than half of the total dose. A maximum concentration of iron in the brain and the liver was reached 2–3 days after dosing. According to this study, a similar picture was obtained with mice, guinea pigs, dogs and monkeys. The toxicity of 274 depends drastically on the formulation. A single dose of 1000 mg kg−1 was well tolerated by mice as an aqueous suspension, but killed the animals when the ferrocene was dissolved in 1 mL of oil (cod liver, olive or corn oil, but not in medicinal parafﬁn). Likewise, rats kept on a normal diet with 3 % fat survived thirteen daily doses of 274 as aqueous suspension, each corresponding to 500 mg kg−1 Fe. However, this treatment was fatal to three out of ﬁve mice when the diet consisted exclusively of peanuts (48 % fat). Acetylferrocene, on the other hand, is a rather toxic compound. The oral LD was <5 mg/kg for female rats, between 5 and 50 mg/kg for male rats, and between 10 and 100 mg/kg for monkeys.572 Its toxicity appeared to be delayed, with most mortality occurring on the third day after dosing. The mechanism by which monkeys are less susceptible than rats to the toxicity of acetylferrocene is not clear. Ethylferrocene, on the other hand, is relatively nontoxic. No deaths were observed at doses of 500 mg kg−1 .572 Coarse tremors and convulsions were the symptom of ferrocene poisoning, eventually killing the animal.573 An increased sensitivity to electroshock was observed, which was maximal on the day following a single dose and vanished within days, depending on the dose. Again, both actions are potentialised by fat. The mechanism by which the fat increases the toxicity remained unclear in this study. Although fat does increase the amount of absorbed ferrocenes somewhat, this cannot explain the increased toxicity of the metallocenes because administering a higher dose leads to the same iron loading without the same toxic effects.573 Ferrocene is a relatively volatile compound. It can thus also be incorporated by inhalation. Dahl and coworkers studied the fate of ferrocene in the respiratory tract of rats with relation not only to cytochrome P450 but a host of other enzymes.574, 575 The physiological effects of ferrocene inhalation have also been studied dose-dependent in two strains of genetically well-characterised mice. Overall, the mean iron lung burden

604

Ferrocenes: Ligands, Materials and Biomolecules

was slightly increased and histochemical stains indicated the presence of iron in the target tissues, but no exposure-related clinical, functional or haematological parameters were signiﬁcantly altered. Evidently, accumulation of iron in the lung did not cause an inﬂammatory response or any functional impairment up to the doses applied.576, 577 13.7.2

Ferrocene Compounds for the Study of Iron-Related Disorders

Uptake of iron, mainly in the form of iron(II) ions, into the human body is a tightly regulated process, with iron-responsive elements working on the RNA level. In the light of the ferrocene metabolism described above, it should be possible to use ferrocene, or its derivatives, for dietary iron supplements (‘haematinic’). Since iron comes in a ‘disguised’ form and is only later released in the liver upon degradation of ferrocene, the usual iron uptake mechanisms will not regulate ferrocene uptake. In that sense, ferrocene could be described as a prodrug form of iron. By using the same argument, and because natural iron excretion is not regulated in any way, higher doses of ferrocene will result in iron overload, again bypassing the regulated iron uptake system. Both aspects have been investigated and used to elucidate molecular mechanisms and effects of iron-related disorders. The metabolism of a new ferrocene haematinic, (thiodiethylidene)ferrocene (275, Scheme 13.82), on rats, dogs and monkeys was studied in detail. In this case, the monooxigenase systems was mildly induced, while the cytochrome system and microsomal proteins were unaffected.578 Valerio and Petersen have recently characterised the cellular, biochemical and molecular aspects of hepatic iron overload in mice.579, 580 Taken together, their data support the hypothesis that peroxidation of cellular membrane lipids is an important mechanism in the toxicity of excess hepatic iron. Ethane exhalation as a direct and α-tocopherol and ubiquinol concentrations in liver and plasma as indirect markers of this process were proposed.581 Interestingly, the same idea was put forward by Osella and coworkers in relation to ferrocene cytotoxicity, as will be discussed below. When male Wistar rats were fed with high doses of ferrocene for several weeks, symptoms of hepatic iron loading, principally in lysosomes, were observed that resemble those determined in humans suffering from haemochromatosis.582, 583 Haemochromatosis is a fairly frequent (one in 400 in Western societies) genetically encoded disease which leads to unhealthy high levels of iron in the body and iron oxide deposits in the liver, under the skin (therefore the name) and elsewhere.584 If untreated, the individual will eventually die. It can be reliably treated by iron chelation therapy and venesection. Several iron storage proteins such as ferritin, haemosiderin and prehaemosiderin were characterised by M¨ossbauer spectroscopy and other methods. In these studies, slight difference were observed in ferrocene-treated rats as compared to humans affected by haemochromatosis.582, 583 Still, ferrocene loading presents a good model to study haemochromatosis-related effects in a controlled animal model. One compound that has received particular attention is (3,5,5-trimethylhexanoyl) ferrocene (TMH-ferrocene, 276 in Scheme 13.82).585, 586 When fed to cultured hepatocytes587, 588 or whole animals, 276 produces an iron overload that closely mimics hepatocellular iron overload in humans. Like ferrocene, TMH-ferrocene is not recognised as iron ions and the regulated iron uptake mechanisms are bypassed. The

The Bioorganometallic Chemistry of Ferrocene O

Fe

275

S

O

605

CO2Na • 4 H2O

Fe

Fe

276, TMH-ferrocene

278, Ferrocerone

Scheme 13.82

same amount of TMH-ferrocene is absorbed in iron-deﬁcient or iron-loaded rats,589 independent from the dose.586 Obviously, TMH-ferrocene does not bind to the usual iron transport proteins in plasma such as transferrin. It is, however, degraded in the liver by cytochrome P450 , which may be induced by phenobarbital in the usual way, to release iron.590 Enzyme induction on various systems has been studied in detail in chick embryo liver cells.591 Subsequently, the free iron atoms are then processed and stored in the same way as dietary iron ions. The effect of experimental iron overload by TMHferrocene was studied on various organs such as bone marrow,585 liver,592–594 heart 595 and brain.596, 597 The different effects of ferrocene and two derivatives, namely 276 and 1,1 -bis(3,5,5-trimethylhexanoyl)ferrocene (277) were studied by 59 Fe labelling in rats. The bioavailability of iron from 276 (whole body retention, 48 % from a 5 mg iron dose) was twice as high as from ferrocene and six times higher than from 277 and ferrous sulfate.586 The TMH-ferrocene model has also been used to investigate the treatment of iron overload.598, 599 Combined biochemical and histological evidence indicates that the additional exposition to ethanol in iron-loaded Wistar rats, again using the TMH-ferrocene model, could modulate the organ damage and oxidative stress.600 The compound 278 (Ferrocerone, Scheme 13.82) was developed and clinically approved in the USSR to treat anemia iron-deﬁciency. It is orally available and the considerations above for circumventing the regulations in iron uptake apply. This compound is, in fact, to the best of our knowledge the ﬁrst marketed drug containing the ferrocene moiety.601 13.7.3

Anticancer Activity of Ferrocene and Derivatives

An antiproliferative effect has been demonstrated for bent metallocenes such as titanocene dichloride by K¨opf and K¨opf-Maier.602–604 Simple ferrocenium salts were the ﬁrst iron compounds for which an antiproliferative effect on certain types of cancer cells was demonstrated.605 The mechanism of action has not yet been elucidated and several targets including nuclear DNA the cell wall and the enzyme topoisomerase II were proposed. Osella et al. showed that ferrocenium salts may generate hydroxyl radicals in physiological solutions.606 An earlier report suggests that these radicals damage the DNA in a Fenton-type reaction.607 The cytotoxic effect of decamethylferrocenium tetraﬂuoroborate (279) was correlated to the production of 8-oxoguanine, the initial product of DNA oxidation. Direct evidence for hydroxyl and superoxide radicals stems from ESR and spin-trapping experiments. Finally, a synergistic effect between 279 and the iron-dependent antitumour drug bleomycin was observed.608 Topoisomerase II inhibition was proposed as a cellular target by other workers.609, 610 However, it

606

Ferrocenes: Ligands, Materials and Biomolecules

might also be envisaged that the cell wall was the cellular target. This ﬁnding relates to the toxic effects, namely peroxidation of membrane lipids, of excess hepatic iron described in Section 13.7.2. It may thus be an interesting aspect for future investigations whether the antiproliferative properties of ferrocene or ferrocenium salts are in fact due to the formation of iron ions. This may be in line with the ﬁnding that the ferrocenium cation is the most unstable cation compared to fcMe2 and even perdeuterated ferrocene, and yet showed the highest cytostatic effect in cultures of Yoshida ascites tumour cells.611 In addition, there are conﬂicting reports on whether or not the redox state of the iron atoms (Fe(II) in ferrocene or Fe(III) in ferrocenium salts) is crucial for cytotoxicity. One could argue that once the compound is internalised into cells, the Fe2+ /Fe3+ equilibrium is established due to the biologically available redox couples such as glutathione. Free radicals (hydroxide and peroxide) may form in the course of this reaction which contribute to the antiproliferative activity. If this assumption is true, then issues of bioavailability are obviously more important than the redox state of the initial compound. Also, the redox potential of the ferrocene compound has to be in the biological window, but its exact value may not be as crucial as other factors such as stability and bioavailability. In a more general sense, redox activity is a property that is of course not unique to metal compounds, but certainly more frequently encountered in this class of compounds. It is thus interesting to correlate the redox properties of metal compounds with electron transfer, oxidative stress, the formation of reactive oxygen species, and generally the redox status of cells.612, 613 This seamingly obvious correlation has only been tentatively touched, and it is in fact difﬁcult to study, the exact ‘redox potential’ or even ‘redox status’ of a whole cell. A mechanism where anticancer ferrocenes are activated by a redox pathway has very recently been suggested by Jaouen’s work and is discussed below.613 Neuse and coworkers found signiﬁcantly enhanced cytotoxicity when ferrocenes were bound to polymeric supports such as polyaspartamide.614–618 Smaller ferrocenyl polyamines were also tested by Brynes and coworkers at almost the same time, but with limited success.619 Neuse et al. claim that enhanced water solubility may be a crucial factor for activity. Along the same lines, the nature of the counterion greatly inﬂuences the cytotoxicity of ferrocenium salts. While the poorly soluble heptamolybdate is inactive, ferrocenium salts with good aqueous solubility such as the picrate and trichloroacetate also display high antitumour activity.620 The work on polymer-bound ferrocene as anticancer drugs has recently been reviewed by Neuse.621 A number of ferrocene derivatives have been tested for their antiproliferative activity in a nonsystematic way.612, 622–626 An interesting approach is found in a ferroceneacridine conjugate, which serves to bring the ferrocene close to DNA by intercalation of the acridine moiety.624 Indeed, signiﬁcant antiproliferative activity is observed for this conjugate. Borylated ferrocenes have been tested for their activity in Boron Neutron Capture Therapy (BNCT) by Wagner and coworkers.622 This work presents very interesting organ distribution studies. In particular, one derivative was found to penetrate the blood-brain barrier, which is of importance for the treatment of brain tumors. Another interesting ferrocene derivative was recently synthesised by Schmalz and coworkers. Compound 280 (Scheme 13.83) is an example of a nucleoside analogue of ferrocene.627 Several derivatives showed promising antiproliferative activity with LD50

The Bioorganometallic Chemistry of Ferrocene

607

NH2 N N O t-Bu Fe 280

Scheme 13.83

values in the low µM range, although not as high as the iron-tricarbonyl nucleoside analogues from the same group.628, 629 A different approach has been to substitute phenyl rings in established drugs by ferrocenyl groups. This has given rise to a group of derivatives of the anticancer drug tamoxifen, called ferrocifens from Jaouen’s group.613, 630 Tamoxifen (281) is the frontline chemotherapeutic agent for patients with hormone-dependent breast cancer; its active metabolite is hydroxy-tamoxifen (282 in Scheme 13.84). In general, breast tumours can be divided into two groups that are distinguished by the presence (ER(+)) or absence (ER(−)) of the estrogen receptor. About two thirds of all cases belong to the ER(+) type, which is susceptible to hormone therapy by selective estrogen receptor modulators (SERMs) such as 281. These patients have signiﬁcant improved chances for successful treatment compared to the ER(−) group of patients.613 The antiproliferative action of tamoxifen arises from the competitive binding to the estrogen receptor ERα subtype, thus repressing estradiol-mediated DNA transcription in the tumour tissue.631 Unfortunately, expression of the ERα may become down-regulated under tamoxifen treatment, turning the drug ineffective. The activity of ferrocifens (283) against ER(+) cancer cell lines is thought to follow the same principle. Originally, it was proposed that the combination of antiestrogenic activity of tamoxifen with ferrocene cytotoxicity would be an advantageous R

R

Fe O(CH2)2N(CH3)2 Tamoxifen 281 (R = H)

O(CH2)nN(CH3)2 Ferrocifens 283 (R = H; n = 2, 3, 4, 5, 8)

Hydroxy-tamoxifen 282 (R = OH)

Scheme 13.84 Tamoxifen 281, the active metabolite hydroxy-tamoxifen 282, and ferrocifen derivatives 283

608

Ferrocenes: Ligands, Materials and Biomolecules

strategy. Along this reasoning, a number of ferrocene derivatives of tamoxifen were prepared (Scheme 13.84), which display interesting structure–activity relationship. Firstly, replacement of the phenyl group by ferrocene reduces receptor afﬁnity by about 60 %. Secondly, increasing the length of the dimethylamino-alkyl chain has an adverse effect on receptor binding. In addition, it also changes the bioavailability and determines whether estrogenic or anti-estrogenic activity is observed in an animal experiment so that an optimum value seems to be around n = 4. Generally, the Z isomers bind more strongly to the ERα than the E isomers in in vitro tests. However, there is rapid isomerisation under physiological conditions. Ferrocifens were shown to be effective anti-estrogens in MCF-7 breast cancer cells lines (ER(+)) and against estrogen-dependent tumour xenografts in nude mice. Recently, however, 283 was shown to be active against the ER(−) MDA-MB231 tumour cell line, which lacks the molecular target of 281 and 282. This indicates a new and different mode of action. Interestingly, the ruthenocene analogue of 283 also acts as an antiestrogen in ER(+) breast cancer cells but lacks the antiproliferative effect of ferrocifen against ER(−) cell lines.632 Other organometallic fragments in place of the ferrocenyl group were also tested, but found to be inactive.630, 633 This suggests two different modes of action for ferrocifen. In addition to tamoxifenlike binding to the ERα receptor, the second pathway must critically dependent on the properties of ferrocene. In an elegant study, redox activation has been proposed as the second mode of action.634 The active metabolite hydroxyferrocifen is readily oxidised, yielding a quinone methide intermediate which is activated for nucleophilic attack by nucleophiles. Quinone methides of 4-hydroxytamoxifen are known to be stable for hours under physiological conditions. Adducts of such tamoxifen metabolites with gluthatione and nucleobases are thought to be responsible for its general toxicity and mutagenic potential,635–637 and related chemistry is likely to apply to the activated ferrocifens. Extensive structure–activity relationship studies in correlation with electrochemical properties support this hypothesis. It is particularly noteworthy that redox activity of the metallocene is the key for additional biological activity which exceeds that of a purely organic analogue. The mode of action of ferrocifen is thus substantially different from the other very promising ferrocene-based drug ferroquine discussed below, where ferrocene is also crucially important but no conclusive evidence for a metal-speciﬁc mode of action was found. Once this redox-activation mode of action was established, it is clearly not dependent on the tamoxifen-related substructure. Very recently, the same group has presented work on ferrocenyl diphenols and unconjugated phenol derivatives that also have good antiproliferative activity, presumably via a related mechanism of activation and formation of similar intermediates.638, 639 As an agonist of the µ, δ and κ-opoid receptors, the morphian alkaloid etorphine has been shown to inhibit breast cancer cell proliferation. Schottenberger, Laus and coworkers prepared ferrocenyl derivatives of etorphine (Scheme 13.85) and their binding afﬁnities towards µ, δ and κ-opoid receptors were measured by radioassays.640 One of the stereiosomers of compound 285 kept a very good afﬁnity to these receptors as compared to etorphine itself. Unfortunately, no antiproliferative assays were carried out with this compound.

The Bioorganometallic Chemistry of Ferrocene

609

HO

HO

O

O

N

N O

O

HO

HO etorphine

Fe

285

Scheme 13.85 Etorphine and its ferrocenyl derivative 285

13.7.4

CNS-Active Drugs

The ﬁrst report dealing with ferrocenylated derivatives of pharmacologically active molecules dates from the early 1960s when Loev and Flores described the synthesis of ferrocene analogues of the psychoactive drug amphetamine (286) and the anticonvulsivant phenytoin (287, Scheme 13.86).641 O HN

O HN

NH

NH

Ph NH2

Amphetamine

Fe

Phenytoin

286 O

N Cl

Ph O

NH2

Fe

N

O

287 O

N N

R

Fe

Diazepam

288, R = H or I

Scheme 13.86 Amphetamine, phenytoin, diazepam and their ferrocenylated analogues

The idea was to evaluate how the replacement of their phenyl substituent(s) by the organometallic moiety ferrocene affected their biological activity. This concept was later referred to as bioisosteric replacement by Maryanoff.642 Unfortunately, the results were quite disappointing as the ferrocene analogues totally lacked the activity of the parent compounds. Nonetheless, other ferrocene derivatives designed from molecules with known pharmacological activity were prepared during the 1960s but,

610

Ferrocenes: Ligands, Materials and Biomolecules

again, the biological results were disappointing.643, 644 Replacement of the phenyl substituent at the 5-position of the psychoactive drug diazepam by a ferrocenyl moiety (288, Scheme 13.86) also induced a complete loss of its pharmacological properties in vivo.645 These early results come as no surprise in the light of today’s knowledge about structure–activity relationships (SAR) for these classes of compounds. The concept of bioisosterism was recently and more successfully applied by Gmeiner and coworkers in the area of CNS active drugs.646 From the lead compound BP 897, a known selective agonist of the D3 subtype dopamine receptor, they synthesised a series of ferrocenic analogues 289 where the naphthyl substituent is replaced by a ferrocenyl entity. Two representative examples are shown in Scheme 13.87.

Cl O

N N H

R′

Cl

O

N Fe

BP 897

N H

N

R

N

289a: R = OMe, R′ = H 289b: R = R′ = Cl

Scheme 13.87 The dopamine receptor agonist BP 897 and its ferrocene analogues

The binding afﬁnity of the molecules was measured on ﬁve subtypes of dopamine receptors, two subtypes of serotonin receptors and one adrenaline receptor. It appears that the binding proﬁle of the complexes was substantially modiﬁed with respect to BP 897. Subnanomolar inhibition constants were measured for compound 289a with the D4.4 dopamine and the 5-HT1A serotonin receptors and for compound 289b with the D3 and D4.4 dopamine receptors. 13.7.5

Anti-Inﬂammatory and Analgesic Drugs

Maryanoff and coworkers synthesised several ferrocenyl analogues of the nonsteroidal antiinﬂammatory drugs tolmedin, fenbufen, ﬂurbiprofen and fenclofenac. Unfortunately, none of them had the biological activity of the parent compounds.642 13.7.6

Antimicrobial Activity of Ferrocene Derivatives

In the mid seventies, Epton, Marr and coworkers published a series of ferrocenylated penicillins and cephalosporines in which the ferrocene moiety replaced the phenyl substituent of penicillin G and the thiophene substituent of cephalotin (Scheme 13.88).647–650 It was found that the distance between the ferrocene moiety and the β-lactam ring had a marked inﬂuence on the antibacterial activity of the compounds. Compound 290b displayed the highest antibacterial activity of the series and compound 290d was the best β-lactamase inhibitor. The ferrocenyl cephalosporines 291 were less active than the parent compound cephalotin.

The Bioorganometallic Chemistry of Ferrocene H N

R S

O

H N

O

N O

S N

O

COOH

COOH

Penicilline G S

H N O

R S O

O

N O

290a: R = Fc 290b: R = FcCH2 290c: R = FcCHMe 290d: R = FcCMe2

S O

N O

COOH Cephalotin

H N

O

COOH

611

O

291a: R = Fc 291b: R = FcCH2 291c: R = FcCHMe 291d: R = FcCMe2

Scheme 13.88 Penicillin G, cephalotin and their ferrocene analogues

The Metzler-Notle group has recently prepared ferrocene-peptide bioconjugates with signiﬁcant antibiotic activity.37, 38 A variety of naturally occuring peptides with antimicrobial activity have been known for quite some time. Their mode of action is thought to be membrane interaction with the bacterial membranes, although details remain elusive.651 A number of small peptides with particular antibacterial activity in mind were synthesised (entries 18 and 19 in Table 13.1).37, 38 The most active peptides were pentapeptides with a mixed Trp−Arg sequence. The most active compounds had an activity of 7.1 µM, which was even slightly better than the naturally occuring pilosulin 2, which has 20 amino acids. Moreover, the presence of an N -terminal metallocene not only increases activity compared to nonmetallic derivatives. It also changes the speciﬁcity of the conjugate against Gram-negative or Gram-positive bacteria.37 Given the great health concerns that arise from multidrug resistant strains of the Gram-positive S. aureus, these metal–peptide derivatives may be of great potential. Malaria is a tropical disease caused by parasites of the genus Plasmodium that infects humans when bitten by infected mosquitoes of the Anopheles genus. More than 40 % of the world’s population lives in areas with the risk of malaria. According to the World Health Organisation (WHO), 300 to 500 million clinical cases occur every year. This disease kills 1.2 million people per year mostly in Sub-Saharan Africa. Death is caused by one of the four species of Plasmodium, namely P. falciparum.652 A number of drugs is available to treat infected patients, the oldest one being quinine (Scheme 13.91) and the most important one being chloroquine (CQ) (Scheme 13.90). Unfortunately, the widespread distribution of chloroquine led to the appearance of resistant strains, which are now common in all endemic areas around the world. Therefore, new drugs, with good accessibility, safety and affordability need to be developed rapidly.653–656 Since the mechanisms of action and resistance seem to be independent, the 4aminoquinoline class of antimalarial compound is still attractive for developing new therapeutic agents. Several series of ferrocenic analogues of established antimalarials were synthesised with the idea that the presence of ferrocene would potentialise their biological properties towards P. falciparum and possibly overcome drug resistance. The chloroquine analogues 292a–c were the ﬁrst compounds to be reported.657 The synthesis of 292a is depicted in Scheme 13.89.

612

Ferrocenes: Ligands, Materials and Biomolecules HO N

OHC

NMe2 Fe

NMe2 i

Fe

NMe2

H 2N ii

Fe

iii HN

NMe2 Fe

Cl

N 292a

Scheme 13.89 Synthesis of Ferroquine 292a: i) NH2 OH, OH− ; ii) LiAlH4 ; iii) K2 CO3 , 4,7-dichloroquinoline

Their antimalarial activity together with that of chloroquine was measured on different strains of P. falciparum (Table 13.9).657, 658 On the chloroquine-sensitive (CQ(+)) strain SGE2, compound 292a was shown to be as effective as CQ. On the chloroquineresistant strain (CQ(−)) FCM6 (note: a parasite strain is considered as CQ(−) when the IC50 of CQ is higher than 100 nM), compounds 292a–c were 5 to 20 times more effective than CQ, with 292a being the most effective of the series. Other ferrocenyl compounds belonging of the same series were synthesised in order to delineate structure–activity relationships. Changing the substituents on the tertiary amino group had little to no inﬂuence on the effectiveness towards CQ(−) strains (compounds 292d, 292e, 292f, 292l, 292m and 292n) except when the new substituent was Fem.659–661 Interestingly, compounds 292a, 292l, 292m and 292n also displayed selective antiviral activity on human SARS coronavirus.659 Moving the methylene dimethylamino group from one Cp ring to the other (compound 292o) did not change the antimalarial activity on the CQ(+) strain but decreased it slightly on the CQ(−) strain.662 The position of the ferrocene substituent along the side chain was also probed. Compound 293a, the closest analogue of CQ in structure with the Fem moiety grafted to the tertiary amine group, was twice less efﬁcient than 292a on the Dd2 CQ(−) strain but still much more efﬁcient than CQ.660, 663 Compounds 293b–t displayed variable efﬁciencies on different strains of P. falciparum. Many of them had IC50 lower than 100 nM, including the W2 strain, which has an IC50 = 452 nM for CQ. However compound 292a was still the most efﬁcient. Moving the position of the ferrocenyl to the 4-aminoquinoline group (compound 295) yielded a dramatic decrease of the antimalarial activity.663 The distance between the aminoquinoline group and ferrocenyl methylene dimethylamino moiety was also extended (compounds 294). This resulted in slightly less efﬁcient antimalarial agents in vitro.664 The bisquinoline compound 296 was also synthesised. It was shown to be less potent than 292a on CQ(+) and CQ(−) strains.665

Cl

Chloroquine (CQ)

N

HN

NEt2

Cl

Cl

Cl R

2

Fe

R2

N

N

Cl

Cl

293r 293s 293t

293q

293o 293p

293k 293l 293m 293n

293h 293i 293j

293b 293c 293d 293e 293f 293g

+

H

N

H H H H H H H H H H H H H H H H Me Me Me

1

R1

1

2

H Me Et

H Me Me Me Me Et Et Et Et Fem Fem Fem Fem

H H

H

R

n

N R2

293a

R

compound R

N

HN

N H

+

HN

0 1 2 4 1 1 1

2 4 0 1 2 4

295

+

N

HN

Fe

Et2 N + H

294i

Cl

2 C4H5O6−

N

H N

N

N

N

N

Fe

6

0 1 2 4 0 1

296

N

Cl

NMe2

4

−C(O)NHCH2Ph

3

−C(O)NHCH2Ph

−C(O)NHCH2Ph

294h

Cl

294e 294f

2

6

3 4

2

n

NMe2

294g

Cl

294c

HN

H

294b

Fe

H −C(O)NHCH2Ph

H H

294a

294d

R1

N

R

nN

compound

Cl

HN

n

Fe

2 C4H5O6−

Fe

Scheme 13.90 Ferrocenylated analogues of chloroquine (CQ)

292p

N

HN

292o

N

HN

N R1

Fem CH2CH2OH CH2CH2OH CH2CH2OH

Me Fem

Et H H Me Et

Fem

Et

292k 292l 292m 292n

Me

H

t

Bu H

H

Me Me Et Et −(CH2)5− H Et H Me

R

1

Fe

292h 292i 292j

292d 292e 292f 292g

compound 292a = FQ 292b 292c

N

HN

Fe

The Bioorganometallic Chemistry of Ferrocene 613

614

Ferrocenes: Ligands, Materials and Biomolecules Table 13.9 In vitro activity of selected ferrocenyl compounds on CQ(+) and CQ(−) P. falciparum strains Compound

IC50 on CQ(+) strain (nM)

Chloroquine

29 ± 26

IC50 on CQ(−) strain (nM) 436 ± 330

292a

20 ± 14

23 ± 16

292b

55 ± 29

97 ± 29

292c

81 ± 31

70 ± 48

292d

19 ± 5

16 ± 2

292e

13 ± 1

23 ± 2

292f

38 ± 13

29 ± 1

292g

21 ± 7

54 ± 9

292m

13

7

292l

19

49

Ref. 658

661

662

293a

27 ± 15

53 ± 5

295

143 ± 93

789 ± 120

663

294a

33

37

662

294b

51 ± 4

37 ± 2

664

294e

21 ± 2

37 ± 7

296

110 ± 27

62 ± 12

665

5±1

9±1

674

10 ± 3

32 ± 7

2H-Arteminisin 299a 299b Atovaquone 301a

12 ± 2

14 ± 2

600 ± 200

700 ± 350

5000 ± 400

2500 ± 300

302a

40

nd

302b

80

nd

302c

80

nd

303a

4600

303b

5100

304 (SD)-238 305

15200 ± 2000

1060 ± 30

600 4900

675 676

677, 678 679 510

6100

491

Efforts were put into the disclosure of the mechanism of the action of the ferroquine. The question was whether it followed the same mechanism as CQ and correlatively to explain why 292a was active on CQ(−) strains. First, as 292a possesses planar chirality, the two enantiomers were synthesised in pure form by enzymatic resolution. Both enantiomers had the same antimalarial activity in vitro as the racemic compound.666 Second, like CQ, 292a was shown to be a strong inhibitor of hemozoin

The Bioorganometallic Chemistry of Ferrocene

Fe HO

HO

N H

NR1R2 CF3

N

CF3

N

F3C

F3C

Mefloquine

R1

R2

297a

Me

Me

297b

Et

compound

Et −(CH2)5−

297

−(CH2)2N(Me)(CH2)2−

297d H

H HO

H

Fe

N

HO O

O

NR1R2 N

N Quinine

compound

R1

298a 298b

Me Me −(CH2)5−

H

H

H

O O O H

O O

O O O H

H

O

H

O OMe

OH

Artemisinin

2H-Artemisinin

Artemether H

H O O

O O O H

O H

H

O

O

O H

R2

O H

H

H

O

Fe

H N

O

O O

299a

Fe

299b

Scheme 13.91 Ferrocenyl derivatives of quinine, meﬂoquine and artemisinin

615

616

Ferrocenes: Ligands, Materials and Biomolecules

O

O

N H R

O

R = Ac, strychnobrasiline R = H, Na-deacetylstrychnobrasiline

N H O

O

Fc 300

Scheme 13.92 Strychnobrasiline and its ferrocenyl derivative

formation. The main difference between 292a and CQ lies in their lipophilicity and their pKa . This was hypothesised to have a consequence on its interaction with the protein PfCRT involved in CQ resistance.667 Another question that may arise is the precise contribution of the ferrocenyl moiety to the antimalarial activity of 292a, for instance whether its redox properties play any role. The very good antiplasmodial activity towards both CQ(−) and CQ(+) P. falciparum strains of compound 292p (an analogue of ferroquine with the Fc moiety replaced by a phenyl group) and the ruthenocene analogue of 292a,662 indicates that the metallocene might act through a modiﬁcation of shape, volume, lipophilicity, basicity and electronic proﬁle of the parent molecule, which eventually alters its pharmacodynamic behaviour.667 Given the fact that ferroquine (292a) not only displays potent activity in vitro on P. falciparum laboratory strains, but also on ﬁeld isolates from Gabon,668–670 Senegal671 and Cambodia672 together with in vivo schizontocidal activity on murine models infected by P. vickei strains657, 666, 673 , this molecule is being developed in the laboratories of Sanoﬁ–Aventis and is currently under phase IIb clinical trials. On the basis of the results obtained with ferroquine, the known antimalarials quinine, meﬂoquine680 and artemisinin674 were also derivatised with a ferrocenyl unit (compounds 297–299 in Scheme 13.91). Ferrocenyl analogues of meﬂoquine 297 and quinine 298 were much less active compounds than the parent drug whatever the strain was. Modiﬁcation of artemisinin led to compounds 299 with 1.2 to 5.5 times lower activity on laboratory strains (Table 13.9). To reverse the drug resistance observed with chloroquine, it may be useful to associate drugs called chemosensitisers. The natural compound strychnobrasiline possesses such properties. There was an interest in tethering a ferrocene motif onto it (Scheme 13.92).681 The resulting compound 300 displayed moderate activity in vitro on a CQ(−) strain but no such properties were observed in vivo. Atovaquone is administered in association with pronaguil for the treatment and the prevention of malaria. It inhibits the mitochondrial electron transport through the cytochrome c reductase complex.656 Several analogues of atovaquone with the ferrocenyl motif in place of the chlorophenyl substituent were synthesised (301 in Scheme 13.93). Their antimalarial activity was somehow disappointing as it was ﬁve to ten times lower than that of the parent compound (Table 13.9). However, very interesting results were obtained on the parasite Toxoplasma gondii as ferrocenyl atovaquone 301a–c analogues had very good activity in vitro on both ATO(−) and ATO(+) T. gondii strains.675

The Bioorganometallic Chemistry of Ferrocene

617

O Cl

OH R N

O

Fc

O OH O Atovaquone

compound

R

301a 301b

(CH2)5Me (CH2)6Me

301c

(CH2)7Me

Scheme 13.93 Atovaquone and its ferrocenyl analogues O N R N

R

I−

303a

compound

S

N

Fe

O 302a

Fe Me

Br

302b

303b

O2N

302c

H N

Fe

N 304

Fe

OAc O

AcO AcO OAc AcO AcO

OAc

S O

O OAc

Fe

S O

305

Scheme 13.94

Four other series of ferrocenyl compounds with in vitro antimalarial activity were published in the literature (Schemes 13.68 and 13.94). The benzimidazolium series 302 showed promising activity on a CQ(+) strain, with an IC50 only two to four times higher than CQ and artemether (Table 13.9).676 The other compounds in Scheme 13.94 were much less active with IC50 values in the micromolar range.491, 510, 677–679 The antiparasital activity of several benzimidazolium compounds was measured on the Leishmania infantum parasite strain L1, which is responsible for another tropical disease called Leishmaniasis. The most effective compounds were at least 25 times less active than the drug of choice PX-6518.682

618

Ferrocenes: Ligands, Materials and Biomolecules F

OH

N N

HO

N

N N

F

N N

N

N N

Fe

N

N

Fluconazole

306 O

O

O

O N

N

N

Fe

Cl

N

N

N 307a

Triadimefon

O

O X

Fe N 307b

Fe

N

N

307c

X N

N

N OH

OH

O

O N

X

N N Triadimenol

Cl

Fe

N

X N

N 308

Scheme 13.95 Ferrocenyl analogues of common fungicides

A ferrocene analogue of the fungicide ﬂuconazole was synthesised by Biot and coworkers where the diﬂuorophenyl substituent was replaced by a ferrocene entity (306 in Scheme 13.95).683 Its fungicidal property was tested on ﬂuconazole sensitive and ﬂuconazole insensitive yeast strains. No activity was noticed on any of the strains. Several ferrocene analogues of fungicides triadimefon and triadimenol were synthesised (307a–c and 308 in Scheme 13.95).684–688 None of them had signiﬁcant antifungal activities when compared to the parent compounds, but series 307a and 307b had good plant regulatory activities.686, 688 13.7.7

Insecticidal Properties of Ferrocene Derivatives

The insecticide tebufenozide, a nonsteroidal synthetic agonist of the moulding hormone 20-hydroxyecdysone, was chemically modiﬁed so as to introduce a ferrocenyl entity in place of one or the other of its aromatic substituents (309 and 310 in Scheme 13.96).689 Compound 309 conserved an excellent larvicidal activity as compared to control molecule, while it was much lower for 310. The ferrocene analogue 311 of the diﬂubenzuron family of insecticides, acting as chitin biosynthesis inhibitors,690 were synthesised (Scheme 13.96). They displayed

The Bioorganometallic Chemistry of Ferrocene

O

H N

Fe

H N

N

O

O

N

O

309 Cl

O N H

O

N H

Fe

O N H

311

O

O

N N

NH

COOMe N

O

NH

N N

N N R3

R2

310, X = H or 4-Cl Cl

N H

Fe

F Diflubenzuron

R1

X N

O

O

Tebufenozide F

H N

O

619

X

O COOMe

Cl Cl

RH 3421: R1 = H, R2 = Me, R3 = COOMe PH 60-41: R1 = Cl, R2 = R3 = H

N

Fe OCF3

N

N 312

Indoxacarb

Scheme 13.96 Insecticides and their ferrocenic analogues

poor insecticidal activity on Oriental armyworms.691 Finally, the ferrocene analogues 312 of the indoxacarb family of insecticides also comprising compounds RH 3421 and PH 60-41, acting as sodium channels modulators,690 were synthesised (Scheme 13.96).684 Their bactericidal properties were weak. No data on their insecticidal properties were given. 13.7.8

Herbicidal Properties of Ferrocene Derivatives

The herbicidal property of the cyanoacrylate derivative 313, that operates as photosystem II electron transport inhibitor,692 was tentatively improved by replacing the aromatic substituent by a ferrocenyl moiety (Scheme 13.97).693 Compounds 314 and 315 displayed comparable herbicidal properties with respect to the parent compound. Cl Fe O NH

O CN 313

OEt

Fe

O NH

O CN 314

OEt

O NH MeS

O CN

315

Scheme 13.97 Cyanoacrylate herbicide 312 and ferrocenic analogues

OEt

620

13.8

Ferrocenes: Ligands, Materials and Biomolecules

Acknowledgements

The authors are grateful to all coworkers for their contribution to Bioorganometallic chemistry. NMN is grateful to Antonio Pinto for literature searches and editorial support, and to Nicole Ray for general support of this project and our group. Dr Srecko Kirin has critically inﬂuenced the development of Section 13.2.3 of this chapter. C. Biot is gratefully acknowledged for his critical proof-reading of section 13.7.6. A number of colleagues have freely shared preprints and preliminary information, which is also gratefully acknowledged.

13.9

Abbreviations

Standard three letter codes for amino acids are used throughout. Unless speciﬁcally noted, stereochemistry implies pure L amino acids. Peptides are consistently written from N - to C-terminus in standard peptide nomenclature. In a peptide, ‘Gly’ corresponds the fragment ‘HNCH2 CO’. For example, H-Gly-NH2 is glycine carboxamide (H2 NCH2 CONH2 ) while Ac-Gly-Ala-OH is N -acetylated glycyl alanine. According to common convention, the same four letter code is used for PNA as for DNA, small letters however indicate PNA oligomers. A AA AAS ACV ADH Ala ALM ALP AP APP Asn AT AV Bn BK BNCT Boc BSA C CD CFPI CGE CHO cells CMIA CNS

Adenine (nucleobase) or Adenosine (nucleoside) Amino Acid Atomic Absorption Spectroscopy Alternating Current Voltammetry Alcohol Dehydrogenase Alanine Alamethicin Alkaline Phosphatase p-Aminophenol p-Aminophenolphosphate Asparagine Angiotensin Avidin Benzyl Bradykinin Boron Neutron Capture Therapy t-Butyloxycarbonyl Bovine Serum Albumin Cytosine (nucleobase) or Cytidine (nucleoside) Circular dichroism 3-carboxy-4-ferrocenylphenylisothiocyanate Capillary Gel Electrophoresis Chinese Hamster Ovarian cells Carbonyl Metalloimmunoassay Central Nervous System

The Bioorganometallic Chemistry of Ferrocene

Cp CPG CQ CV CyP450 2,4-D dA DAAO DCC ddUTP DMAP DNP DPV dsDNA dT EDC ELISA Enk ER ET FAD Faa Fca Fcc Fem Fer Fmoc G Gln Glu Gly GOD GSH Hb HBTU HCG HOBt HOSu HPLC HPLC–ECD HRP HSA IBCF ICP–MS

621

Cyclopentadienide anion (C5 H5 − ) Controlled Pore Glass Chloroquine Cyclic Voltammetry/Cyclic Voltammogram Cytochrome P450 2,4-Dichlorophenoxyacetic acid Deoxy adenosine (DNA/RNA monomer) D-Amino Acid Oxidase Dicyclohexyl carbodiimide Dideoxy uridine triphosphate 4-(Dimethylamino)pyridine Dinitrophenol Differential Pulse Voltammetry double stranded DNA Deoxy thymidine (DNA monomer) N -(3-dimethylaminopropyl)-N -ethylcarbodiimide Enzyme-linked Immunoassay Enkephalin Estrogen receptor Electron Transfer Flavin adenine dinucleotide 1,1 -Diaminoferrocene 1 -Aminoferrocene-1-carboxylic acid Ferrocene-1,1 -dicarboxylic acid Ferrocenylmethyl Ferrocenylalanine Fluorenylmethoxy carbonyl Guanine (nucleobase) or Guanosine (nucleoside) Glutamine Glutamate Glycine Glucose Oxidase Glutathione Hemoglobin O-(Benzotriazol-1-yl)-N ,N ,N ,N -tetramethyluronium hexaﬂuorophosphate Human Chorionic Gonadotropin 1-Hydroxybenzotriazole N -Hydroxysuccinimide High performance (pressure) liquid chromatography High performance (pressure) liquid chromatography with electrochemical detection Horseradish peroxidase Human Serum Albumin Isobutyl chloroformate Inductively Coupled Plasma Mass Spectrometry

622

Ferrocenes: Ligands, Materials and Biomolecules

IgG LARIS LD Leu LOD Lys MEI MIA mIgG MT NAD NADP NLS NP ODN Pac PAMAM PCR PEG Phe pkDAAO PNA PPI Pro PSA PyAOP QCM QI RBA rgDAAO RIA RIS ROMP Sar SAR SAV SBP SDS SERM SIRIS SNP SP ssDNA SV SWV

Immunoglobulin G Laser Atomisation Resonance Ionisation Spectroscopy Lethal Dose Leucine Limit of Detection Lysine Morpholino ethylisocyanide Metalloimmunoassay Mouse Immunoglobin G Metallothionein Nicotinamide Adenine Dinucleotide Nicotinamide Adenine Dinucleotide Phosphate Nuclear Localisation Sequence p-Nitrophenol Oligo deoxy nucleotide (DNA oligomer) Phenoxyacetyl Poly(amidoamine) Polymerase Chain Reaction Poly(ethylene glycol) Phenylalanine pig kidney D-Amino Acid Oxidase Peptide Nucleic Acid Poly(propylenimine) Proline Prostate Speciﬁc Antigen 7-Azabenzotriazol-1-yloxytris(pyrrolidino)phosphonium hexaﬂuorophosphate Quarz Crystal Microbalance Quinone imide Relative binding afﬁnity D-Amino Acid Oxidase from Rhodoturula gracilis Radio Immunoassay Resonance Ionisation Spectroscopy Ring Opening Metathesis Polymerisation Sarcosine (N -Methylglycine) Structure–Activity Relationship Streptavidin Soybean Peroxidase Sodium n-dodecyl-sulfate Selective Estrogen Receptor Modulator Sputter-Initiated Resonance Ionisation Spectroscopy Single Nucleotide Polymorphism Substance P single stranded DNA Sinusoidal Voltammetry Square wave voltammogram

The Bioorganometallic Chemistry of Ferrocene

T TBTU TCBoc TFA Tm tvDAAO Tyr U UTP Val WHO wt Z

623

Thymine (nucleobase, only in DNA and PNA, not RNA) or Thymidine (nucleoside) O-(Benzotriazol-1-yl)-N ,N ,N ,N -tetramethyluronium tetraﬂuoroborate Trichloro-t-butyloxycarbonyl Triﬂuoro acetic acid Melting temperature (measure of stability of oligonucleotide duplices) D-Amino Acid Oxidase from Trigonopsis variabilis Tyrosine Uracil (nucleobase, RNA only) or Uridine (nucleoside) Uridine triphosphate Valine World Health Organization wild-type Benzyloxy-carbonyl

References 1. T.J. Kealy, P.L. Pauson, Nature, 1951, 168, 1039–1040. 2. S.A. Miller, J.A. Tebboth, J.F. Tremaine, J. Chem. Soc., 1952, 632–635. 3. G. Wilkinson, M. Rosenblum, M.C. Whiting, R.B. Woodward, J. Am. Chem. Soc., 1952, 74, 2125–2126. 4. E.O. Fischer, W. Pfab, Z. Naturforsch., 1952, 7b, 377–379. 5. G. Jaouen, A. Vessi`eres, I.S. Butler, Acc. Chem. Res., 1993, 26, 361–369. 6. R.H. Fish, G. Jaouen, Organometallics, 2003, 22, 2166. 7. N. Metzler-Nolte, Angew. Chem., 2001, 113, 1072–1076. 8. K. Schl¨ogl, Monatsh. Chem., 1957, 88, 601–620. 9. D.R. van Staveren, N. Metzler-Nolte, Chem. Rev., 2004, 104, 5931–5985. 10. G. Jaouen (Ed.), Bioorganometallics, Wiley-VCh Verlag GmbH, Weinheim, Germany (2006). 11. K. Severin, N. Metzler-Nolte, Bioorganometallic Chemistry, in Concepts and Models in Bioinorganic Chemistry, H-B. Kraatz, N. Metzler-Nolte (Eds.), Wiley-VCh Verlag GmbH, Weinheim, Germany (2006), pp. 113–136. 12. J. Wang, Electroanalysis, 2001, 13, 983–988. 13. I. Willner, B. Willner, Bioelectrochem. Bioenerg., 1997, 42, 43–57. 14. I. Willner, E. Katz, Angew. Chem. Int. Ed., 2000, 39, 1180–1218. 15. T.S. Zatsepin, S.Y. Andreev, T. Hianik, T.S. Oretskaya, Russ. Chem. Rev., 2003, 72, 537–554. 16. R.H. Crabtree, D.M.P. Mingos (Eds.), Comprehensive Organometallic Chemistry III , Elsevier, Amsterdam, The Netherlands (2007). 17. N. Metzler-Nolte, Bioorganometallic Chemistry, in Comprehensive Organometallic Chemistry III , G. Parkin (Ed.), Elsevier, Amsterdam, The Netherlands (2007), pp. 883–920. 18. C.R. Hauser, J.K. Lindsay, J. Org. Chem., 1957, 22, 1246–1247. 19. J.M. Osgerby, P.L. Pauson, J. Chem. Soc., 1958, 656–660. 20. R.F.W. Jackson, D. Turner, M.H. Block, Synlett , 1996, 862–864. 21. H. Brunner, W. K¨onig, B. Nuber, Tetrahedron Asym., 1993, 4, 699–707.

624 22. 23. 24. 25. 26. 27. 28. 29. 30. 31.

32. 33.

34. 35. 36. 37. 38. 39. 40.

41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60.

Ferrocenes: Ligands, Materials and Biomolecules J. Pospisek, S. Toma, I. Fric, K. Blaha, Coll. Czech. Chem. Commun., 1980, 45, 435–441. M. Kira, T. Matsubara, H. Shinohara, M. Sisido, Chem. Lett., 1997, 89–90. R.P. Hanzlik, P. Soine, W.H. Soine, J. Med. Chem., 1979, 22, 424–428. W.H. Soine, Diss. Abstr. Int., 1979, 39, 3849. E. Cuignet, C. Sergheraert, A. Tartar, M. Dautrevaux, J. Organomet. Chem., 1980, 195, 325–329. E. Cuignet, M. Dautrevaux, C. Sergheraert et al. Eur. J. Med. Chem., 1982, 17, 203–206. R. Epton, G. Marr, G.A. Willmore et al. Int. J. Biol. Macromol., 1981, 3, 395–396. U. Hoffmanns, PhD Thesis, Ruprecht-Karls-Universit¨at, Heidelberg, Germany, 2005. U. Hoffmanns, N. Metzler-Nolte, Bioconjugate Chem., 2006, 17, 204–213. J.C. Brunet, E. Cuignet, M. Dautrevaux et al. Use of Organometallic amino acids as a tool for studying interactions in Phenylalanine-containing Peptides, in Peptides 1980. Proceedings of the sixteenth European Peptide Symposium, K. Brunfeldt (Ed.), Scriptor, Copenhagen, Denmark (1981), pp. 603–607. P. Hublau, C. Sergheraert, L. Ballester, M. Dautrevaux, Eur. J. Med. Chem., 1983, 18, 131–133. A. Tartar, A. Demarly, C. Sergheraert, E. Escher, Structural and Conformational Considerations: Angiotensin in Proceedings of the 8th American Peptide Symposium, V.J. Hruby, D.H. Rich (Eds.), Pierce Chemical, Rockford, Ill. (1983), pp. 377–380. P. Maes, A. Ricouart, E. Escher et al. Coll. Czech. Chem. Commun., 1988, 53, 2914–2919. F. Noor, A. W¨ustholz, R. Kinscherf, N. Metzler-Nolte, Angew. Chem. Int. Ed., 2005, 44, 2429–2432. F. Noor, N. Metzler-Nolte, 2006, in preparation. J. Chantson, M.V. Varga Falzacappa, S. Crovella, N. Metzler-Nolte, ChemMedChem, 2006, 1, 1268–1274. J. Chantson, M.V. Varga Falzacappa, S. Crovella, N. Metzler-Nolte, J. Organomet. Chem., 2005, 690, 4564–4572. J. Hughes, T.W. Smith, H.W. Kosterlitz et al. Nature, 1975, 258, 577–579. V. Clement-Jones, G.M. Besser, Opiod Peptides in Humans and their Clinical Signiﬁcance in The Peptides: Analysis, Synthesis, Biology, S. Udenfried, J. Meienhofer (Eds.), Academic Press, New York, USA (1984), pp. 323–389. J.S. Morley, Annu. Rev. Pharmacol. Toxicol., 1980, 20, 81–110. U.S. von Euler, J.H. Gaddum, J. Physiol., 1931, 72, 74–87. M.M. Chang, S.E. Leeman, H.D. Niall, Nature New Biol., 1971, 232, 86–89. G.W. Tregear, H.D. Niall, J.T. Potts Jun. et al. Nature New Biol., 1971, 232, 87–89. M. Otsuka, K. Yoshioka, Physiol. Rev., 1993, 73, 229–308. N-E. Rhaleb, G. Drapeau, S. Dion et al. British J. Pharmacol., 1990, 99, 445–448. J.E. Taylor, F.V. DeFeudis, J.P. Moreau, Drug Dev. Res., 1989, 16, 1–11. J.M. Saavedra, Endocrine Rev., 1992, 13, 329–380. D.M. Blow, Acc. Chem. Res., 1976, 9, 145–152. T.A. Steitz, R. Henderson, D.M. Blow, J. Mol. Biol., 1969, 46, 337–348. I.R. Butler, S.C. Quayle, J. Organomet. Chem., 1998, 552, 63–68. T. Okamura, K. Sakauye, N. Ueyama, A. Nakamura, Inorg. Chem., 1998, 37, 6731–6736. K. Heinze, M. Schlenker, Eur. J. Inorg. Chem., 2004, 2974–2988. L. Barisic, V. Rapic, V. Kovac, Croat. Chem. Acta, 2002, 75, 199–210. S. Kaluz, S. Toma, Collect. Czech. Chem. Commun., 1988, 53, 638–642. M. Adamczyk, R.E. Reddy, S.D. Rege, Synth. Commun., 2000, 30, 1389–1400. H. Dialer, W. Steglich, W. Beck, Tetrahedron, 2001, 57, 4855–4861. H. Dialer, K. Polborn, W. Ponikwar et al. Chem. Eur. J., 2002, 8, 691–699. D. Kunz, R. Fr¨ohlich, G. Erker, Organometallics, 2001, 20, 572–574. L. Barisic, V. Rapic, H. Pritzkow et al. J. Organomet. Chem., 2003, 682, 131–142.

The Bioorganometallic Chemistry of Ferrocene 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79.

80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102.

625

A-S. Carlstr¨om, T. Frejd, J. Org. Chem., 1990, 55, 4175–4180. B. Basu, S.K. Chattopadhyay, A. Ritzen, T. Frejd, Tetrahedron Asym., 1997, 8, 1841–1846. S. Maricic, A. Ritzen, U. Berg, T. Frejd, Tetrahedron, 2001, 57, 6523–6529. S. Maricic, U. Berg, T. Frejd, Tetrahedron, 2002, 58, 3085–3093. A-S. Carlstr¨om, T. Frejd, Synthesis, 1989, 414–418. S. Maricic, T. Frejd, J. Org. Chem., 2002, 67, 7600–7606. G.D. Smith, J.F. Grifﬁn, Science, 1978, 199, 1214–1216. D. Picone, A. D’Ursi, A. Motta et al. Eur. J. Biochem., 1990, 192, 433–439. A.M. Osman, M.A. El-Maghraby, K.M. Hassan, Bull. Chem. Soc. Jpn., 1975, 48, 2226. D. Freiesleben, K. Polborn, C. Robl et al. Can. J. Chem., 1995, 73, 1164–1174. A. Hess, J. Sehnert, T. Weyherm¨uller, N. Metzler-Nolte, Inorg. Chem., 2000, 39, 5437– 5443. N.S. Khruscheva, N.M. Loim, V.I. Sokolov, Russ. Chem. Bull., 1999, 48, 578–580. B. Kayser, J. Altman, H. N¨oth et al. Eur. J. Inorg. Chem., 1998, 1791–1798. H. Eckert, B. Forster, C. Seidel, Z. Naturforsch., 1991, 46b, 339–352. H. Eckert, C. Seidel, Angew. Chem., 1986, 98, 168–170. J. Sehnert, A. Hess, N. Metzler-Nolte, J. Organomet. Chem., 2001, 637–639, 349–355. A. Hess, PhD Thesis, Ruhr-Universit¨at, Bochum, Germany, 1999. A. Hess, O. Brosch, T. Weyherm¨uller, N. Metzler-Nolte, J. Organomet. Chem., 1999, 589, 75–84. H. Eckert, Y. Kiesel, C. Seidel et al. New Strategy For Peptide Synthesis Using Highly Lipophillic And Chromophoric Groups. Synthesis Of Octapeptide Sequence (24–31) HLYS-ASN-ALA-TYR-LYS-LYS-GLY-GLU-OH of Human β-Endorphin, in Chemistry of Peptides and Proteins, W. Voelter, E. Bayer, Y.A. Ovchinnikov, T. Ivanov (Eds.), Walter de Gruyter, Berlin, New York (1986), pp. 19–28. M. Bodanszky, A. Bodanszky, The Practice of Peptide Synthesis, Springer-Verlag, Berlin, Germany (1984). J. Jones, The Chemical Synthesis of Peptides, Clarendon Press, Oxford, UK (1991). A.S.J. Stewart, C.N.C. Drey, Perkin Trans. I , 1990, 1753–1756. C.N.C. Drey, A.S.J. Stewart, Selective Protection of Cysteine with Ferrocene Derivatives, in Peptides 1986 , D. Theodoropoulos (Ed.), Walter de Gruyter, Berlin (1987), pp. 65–68. R.S. Herrick, R.M. Jarret, T.P. Curran et al. Tetrahedron Lett., 1996, 37, 5289–5292. H-B. Kraatz, J. Lusztyk, G.D. Enright, Inorg. Chem., 1997, 36, 2400–2405. D.R. van Staveren, T. Weyherm¨uller, N. Metzler-Nolte, Dalton Trans., 2003, 210–220. K.A. Mahmoud, Y-T. Long, G. Schatte, H-B. Kraatz, Eur. J. Inorg. Chem., 2005, 173–180. W. Bauer, K. Polborn, W. Beck, J. Organomet. Chem., 1999, 579, 269–279. A. Nomoto, T. Moriuchi, S. Yamazaki et al. Chem. Commun., 1998, 1963–1964. T. Moriuchi, A. Nomoto, K. Yoshida, T. Hirao, J. Organomet. Chem., 1999, 589, 50–58. P. Saweczko, G.D. Enright, H-B. Kraatz, Inorg. Chem., 2001, 40, 4409–4419. P. Saweczko, H-B. Kraatz, Coord. Chem. Rev., 1999, 190–192, 185–198. H-B. Kraatz, M. Galka, Metal Ions Biol. Systems, 2001, 38, 385–409. H-B. Kraatz, D.M. Leek, A. Houmama et al. J. Organomet. Chem., 1999, 589, 38–49. M.V. Baker, H-B. Kraatz, J.W. Quail, New J. Chem., 2001, 25, 427–433. R.M.G. Roberts, J. Silver, J. Organomet. Chem., 1984, 263, 235–241. N.N. Greenwood, T.C. Gibb, M¨ossbauer Spectroscopy, Chapman and Hall, London, UK (1971). A. Hess, N. Metzler-Nolte, Chem. Commun., 1999, 885–886. J.F. Gallagher, P.T.M. Kenny, M.J. Sheehy, Inorg. Chem. Commun., 1999, 2, 200–202. J.F. Gallagher, P.T.M. Kenny, M.J. Sheehy, Acta Crystallogr., 1999, C55, 1257–1260. H. Eckert, M. Koller, Z. Naturforsch., 1990, 45b, 1709–1714. M. Wenzel, I-H. Park, Applied Radiot. Isot., 1986, 37, 491–495.

626

Ferrocenes: Ligands, Materials and Biomolecules

103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116.

R.L. Cox, T.W. Schneider, M.D. Koppang, Anal. Chim. Acta, 1992, 262, 145–159. J.F. Gallagher, P.T.M. Kenny, M.J. Sheehy, Inorg. Chem. Commun., 1999, 2, 327–330. M. Oberhoff, L. Duda, J. Karl et al. Organometallics, 1996, 15, 4005–4011. H.S. Mandal, H-B. Kraatz, J. Organomet. Chem., 2003, 674, 32–37. I. Bediako-Amoa, R. Silerova, H-B. Kraatz, Chem. Commun., 2002, 2430–2431. M.M. Galka, H-B. Kraatz, ChemPhysChem, 2002, 356–359. H-B. Kraatz, Macromol. Symp., 2003, 196, 39–44. T. Moriuchi, A. Nomoto, K. Yoshida, T. Hirao, Organometallics, 2001, 20, 1008–1013. T. Moriuchi, A. Nomoto, K. Yoshida et al. J. Am. Chem. Soc., 2001, 123, 68–75. T. Moriuchi, K. Yoshida, T. Hirao, J. Organomet. Chem., 2001, 637–639, 75–79. P.M. Cowan, S. McGavin, Nature, 1955, 176, 501–503. I.Z. Steinberg, W.F. Harrington, A. Berger et al. J. Am. Chem. Soc., 1960, 82, 5263. A. Wieckowska, R. Bilewicz, A. Misicka et al. Chem. Phys. Lett., 2001, 350, 447–458. M. Pietraszkiewicz, A. Wieckowska, R. Bilewicz et al. Mat. Science Eng. C , 2001, 18, 121–124. S. Iijima, F. Mizutani, S. Yabuki et al. Anal. Chim. Acta, 1993, 281, 483–487. K. Plumb, H-B. Kraatz, Bioconjugate Chem., 2003, 14, 601–606. S. Sek, R. Moszynski, A. Seipiol et al. J. Electroanal. Chem., 2003, 550–551, 359–364. K. Sonogashira, Y. Tohda, N. Hagihara, Tetrahedron Lett., 1975, 50, 4467–4470. N. Hagihara, K. Sonogashira, Y. Tohda, Synthesis, 1977, 777–778. N. Hagihara, S. Takahashi, Y. Kuroyama, K. Sonogashira, Synthesis, 1980, 627–630. O. Brosch, T. Weyherm¨uller, N. Metzler-Nolte, Inorg. Chem., 1999, 38, 5308–5313. O. Brosch, T. Weyherm¨uller, N. Metzler-Nolte, Eur. J. Inorg. Chem., 2000, 323–330. O. Brosch, PhD Thesis, Ruhr-Universit¨at, Bochum, Germany, 1999. J. Meienhofer, The Mixed Carbonic Anhydride Method of Peptide Synthesis, Academic Press, London, UK (1979). J.D. Schmitt, M.S.P. Sansom, I.D. Kerr et al. Biochemistry, 1997, 36, 1115–1122. J.L. Jios, S.I. Kirin, N.N. Buceta et al. J. Organomet. Chem., 2007, in press. K.L. Malisza, S. Top, J. Vaissermann, B. Caro et al. Organometallics, 1995, 14, 5273– 5280. H. Eckert, M. Koller, J. Liq. Chromatogr., 1990, 13, 3399–3414. M. Koller, German Patent 39193178-A1 (1990). K. Shimada, Y. Kawai, T. Oe, T. Nambara, J. Liq. Chromatogr., 1989, 12, 359–371. K. Di Gleria, H.A.O. Hill, L-L. Wong, FEBS Lett., 1996, 390, 142–144. Y. Katayama, S. Sakakihara, M. Maeda, Anal. Sci., 2001, 17, 17–19. K.K-W. Lo, J.S-Y. Lau, D.C-M. Ng, N. Zhu, Dalton Trans., 2002, 1753–1756. B. Misterkiewicz, M. Salmain, G. Jaouen, Tetrahedron Lett., 2004, 45, 7511–7513. A. Bianco, F. Gasparrini, M. Maggini et al. J. Am. Chem. Soc., 1997, 119, 7550–7554. D.M. Guldi, M. Maggini, G. Scorrano et al. J. Inf. Recording, 1998, 24, 33–39. A. Bianco, C. Corvaja, M. Crisma et al. Chem. Eur. J., 2002, 8, 1544–1553. T.N. Duy, P.K.S. Lam, G.R. Shaw, D.W. Connell, Rev. Environ. Contam. Toxicol., 2000, 163, 113–186. K.K-W. Lo, D.C-M. Ng, J.S-Y. Lau et al. New J. Chem., 2003, 27, 274–279. E. Coutouli-Argyropoulou, C. Sideris, G. Kokkinidis, J. Organomet. Chem., 2006, 691, 3909. A.S. Georgopoulou, D.M.P. Mingos, A.J.P. White et al. Dalton Trans., 2000, 2969–2974. T. Moriuchi, T. Hirao, J. Synth. Org. Chem. Jpn., 2001, 59, 1195–1203. T. Moriuchi, T. Hirao, Chem. Soc. Rev., 2004, 294–301. T. Moriuchi, T. Hirao, Ferrocene-Peptide Bioconjugates, in Topics in Organometallic Chemistry: Bioorganometallic Chemistry, Springer, Heidelberg, Germany (2006), pp. 143–175. B. Corain, B. Longato, G. Favero et al. Inorg. Chim. Acta, 1989, 157, 259–266.

117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 140. 141. 142. 143. 144. 145. 146. 147.

The Bioorganometallic Chemistry of Ferrocene

627

148. T. Moriuchi, K. Yoshida, T. Hirao, Organometallics, 2001, 20, 3101–3105. 149. X. de Hatten, T. Weyherm¨uller, N. Metzler-Nolte, J. Organomet. Chem., 2004, 689, 4856. 150. S.I. Kirin, U. Schatzschneider, X. de Hatten et al. J. Organomet. Chem., 2006, 691, 3451–3457. 151. X. de Hatten, E. Bothe, K. Merz et al. PNAS , 2007, submitted. 152. X. de Hatten, Z. Cournia, I. Huc et al. Chem. Eur. J., 2007, 13, 8139–8152. 153. H. Dialer, W. Steglich, W. Beck, Z. Naturforsch., 2001, 56B, 1084–1087. 154. Q-W. Han, X-Q. Zhu, X-B. Hu, J-P. Cheng, Chem. J. Chin. Univ., 2002, 23, 2076–2079. 155. H. Huang, L. Mu, J. He, J-P. Cheng, J. Org. Chem., 2003, 68, 7605–7611. 156. S. Chowdhury, G. Schatte, H-B. Kraatz, Dalton Trans., 2004, 1726–1730. 157. I. Bediako-Amoa, T.C. Sutherland, C-Z. Li et al. J. Phys. Chem., 2004, B108, 704–714. 158. Y. Xu, P. Saweczko, H-B. Kraatz, J. Organomet. Chem., 2001, 637–639, 335–342. 159. Y. Xu, H-B. Kraatz, Tetrahedron Lett., 2001, 42, 2601–2603. 160. R.E. Babine, S.L. Bender, Chem. Rev., 1997, 97, 1359–1472. 161. H. Umezawa, T. Aoyagi, H. Morishima et al. J. Antibiotics, 1970, 23, 259–262. 162. I. Katoh, T. Yasunaga, Y. Ikawa, Y. Yoshinaka, Nature, 1987, 329, 654–656. 163. K. Seio, M. Mizuta, T. Terada, M. Sekine, Tetrahedron Lett., 2004, 45, 6783–6786. 164. F.E. Appoh, T.C. Sutherland, H-B. Kraatz, J. Organomet. Chem., 2004, 689, 4669. 165. S.I. Kirin, D. Wissenbach, N. Metzler-Nolte, New J. Chem., 2005, 1168–1173. 166. F.E. Appoh, D.S. Thomas, H.B. Kraatz, Macromolecules, 2006, 39, 5629–5638. 167. S. Chowdhury, G. Schatte, H-B. Kraatz, Eur. J. Inorg. Chem., 2006, 2006, 988–993. 168. T. Moriuchi, T. Nagai, T. Hirao, Org. Lett., 2005, 7, 5265–5268. 169. T. Moriuchi, T. Nagai, T. Hirao, Org. Lett., 2006, 8, 31–34. 170. K. Seio, M. Mizuta, T. Terada, M. Sekine, J. Org. Chem., 2005, 70, 10311–10322. 171. S. Chowdhury, D.A.R. Sanders, G. Schatte, H-B. Kraatz, Angew. Chem. Int. Ed., 2006, 45, 751–754. 172. Y. Han, K. Cheng, K.A. Simon et al. J. Am. Chem. Soc., 2006, 128, 13913–13920. 173. K. Heinze, M. Beckmann, J. Organomet. Chem., 2006, 691, 5576–5584. 174. M.A.K. Khan, D.S. Thomas, H-B. Kraatz, Inorg. Chim. Acta, 2006, 359, 3339–3344. 175. K. Heinze, M. Beckmann, Eur. J. Inorg. Chem., 2005, 3450–3457. 176. K. Heinze, U. Wild, M. Beckmann, Eur. J. Inorg. Chem., 2007, 617–623. 177. L. Barisic, M. Cakic, K.A. Mahmoud et al. Chem. Eur. J., 2006, 12, 4965–4980. 178. L. Barisic, M. Dropucic, V. Rapic et al. Chem. Commun., 2004, 2004–2005. 179. L. Barisic, V. Rapic, N. Metzler-Nolte, Eur. J. Inorg. Chem., 2006, 4019–4021. 180. S. Chowdhury, G. Schatte, H-B. Kraatz, Angew. Chem. Int. Ed., 2006, 45, 6882–6884. 181. M.A.K. Khan, Y.T. Long, G. Schatte, H-B. Kraatz, Anal. Chem., 2007, 79, 2877–2884. 182. K.A. Mahmoud, H-B. Kraatz, Chem. Eur. J., 2007, 13, 5885–5895. 183. S. Chowdhury, K.A. Mahmoud, S. Gabriele, H-B. Kraatz, Org. Biomol. Chem., 2005, 3, 3018–3023. 184. K.A. Mahmoud, H-B. Kraatz, J. Inorg. Organomet. Polymer Mats, 2006, 16, 201–210. 185. G. Pavlovic, L. Barisic, V. Rapic, I. Leban, Acta Crystallogr., 2002, E58, m13–m15. 186. G. Pavlovic, L. Barisic, V. Rapic, V. Kovac, Acta Crystallogr., 2003, C59, m55–m57. 187. S.I. Kirin, H-B. Kraatz, N. Metzler-Nolte, Chem. Soc. Rev., 2006, 35, 348–354. 188. J.C. Medina, C. Li, S.G. Bott et al. J. Am. Chem. Soc., 1991, 113, 366–367. 189. C.M. Venkatachalam, Biopolym., 1968, 6, 1425–1434. 190. J.T. Yli-Kauhaluoma, J.A. Ashley, C-H. Lo et al. J. Am. Chem. Soc., 1995, 117, 7041–7047. 191. A. Heine, E.A. Stura, J.T. Yli-Kauhaluoma et al. Science, 1998, 279, 1934–1940. 192. S.K. Dey, H-B. Kraatz, Bioconjugate Chem., 2006, 17, 84–89. 193. M. Salmain, G. Jaouen, C. R. Chimie, 2003, 6, 249–258.

628

Ferrocenes: Ligands, Materials and Biomolecules

194. H.M. Kalisz, H.J. Hecht, D. Schomburg, R.D. Schmid, Biochim. Biophys. Acta, 1991, 1080, 138–142. 195. J.H. Pazur, K. Kleppe, Biochemistry, 1964, 3, 578–583. 196. M.N. Jones, P. Manley, A. Wilkinson, Biochem. J., 1982, 203, 285–291. 197. K.R. Frederick, J. Tung, R.S. Emerick et al. J. Biol. Chem., 1990, 265, 3793–3802. 198. M. Kriechbaum, H.J. Heilmann, F.J. Wientjes et al. FEBS Lett., 1989, 255, 63–66. 199. H.J. Hecht, H.M. Kalisz, J. Hendle et al. J. Mol. Biol., 1993, 229, 153–172. 200. European Patent Appl. 84303090.9 (1984). 201. A.E.G. Cass, G. Davies, G.D. Francis et al. Anal. Chem., 1984, 56, 667–671. 202. Y. Degani, A. Heller, J. Phys. Chem., 1987, 91, 1285–1289. 203. P.N. Bartlett, V.Q. Bradford, R.G. Whitaker, Talanta, 1991, 38, 57–63. 204. Y. Degani, A. Heller, J. Am. Chem. Soc., 1988, 110, 2615–2620. 205. A. Heller, Acc. Chem. Res., 1990, 23, 128–134. 206. A. Heller, Y. Degani, Direct electrical communication between chemically modiﬁed enzymes and metal electrodes: III. electron-transfer relay modiﬁed glucose oxidase and Damino-acid oxidase, in Redox Chemistry and Interfacial Behaviour of Biological Molecules, G. Dryhurst, K. Niki (Eds.), Plenum Pub. Corp., New York, USA (1988), pp. 151–170. 207. W. Schuhmann, T.J. Ohara, H-L. Schmidt, A. Heller, J. Am. Chem. Soc., 1991, 113, 1394–1397. 208. A. Badia, N.H.H. Thai, A.M. English et al. Anal. Chim. Acta, 1992, 262, 87–90. 209. R. Blonder, E. Katz, Y. Cohen et al. Anal. Chem., 1996, 68, 3151–3157. 210. S. Kunugi, Y. Murakami, K. Ikeda, N. Itoh, Int. J. Biol. Macromol., 1992, 14, 210–214. 211. A.D. Ryabov, A.M. Trushkin, L.I. Baksheeva et al. Angew. Chem., 1992, 104, 788–790. 212. R. Blonder, I. Ben-Dov, A. Dagan et al. Biosens. Bioelectronics, 1997, 12, 627–644. 213. S. Kuwabata, T. Okamoto, Y. Kajiya, H. Yoneyama, Anal. Chem., 1995, 67, 1684–1690. 214. A. Riklin, I. Willner, Anal. Chem., 1995, 67, 4118–4126. 215. A. Badia, R. Carlini, A. Fernandez et al. J. Am. Chem. Soc., 1993, 115, 7053–7060. 216. H.B. Gray, J.R. Winkler, Annu. Rev. Biochem., 1996, 65, 537–561. 217. R.A. Marcus, N. Sutin, Biochim. Biophys. Acta, 1985, 811, 265–322. 218. R. Nagata, S.A. Clark, K. Yokoyama et al. Anal. Chim. Acta, 1995, 304, 157–164. 219. R. Nagata, K. Yokoyama, S.A. Clark, I. Karube, Biosens. Bioelectronics, 1995, 10, 261–267. 220. R. Nagata, K. Yokoyama, H. Durliat et al. Electroanalysis, 1995, 7, 1027–1031. 221. E. Katz, A. Riklin, V. Heleg-Shabtai et al. Anal. Chim. Acta, 1999, 385, 45–58. 222. A. Riklin, E. Katz, I. Willner et al. Nature, 1995, 376, 672–675. 223. L-Q. Chen, X-E. Zhang, W-H. Xie et al. Biosens. Bioelectronics, 2002, 17, 851–857. 224. E. Katz, B. Willner, I. Willner, Biosens. Bioelectronics, 1997, 12, 703–719. 225. F. Patolsky, B. Filanovsky, E. Katz, I. Willner, J. Phys. Chem., 1998, B 102, 10359–10367. 226. R. Blonder, S. Levi, G. Tao et al. J. Am. Chem. Soc., 1997, 119, 10467–10478. 227. I. Willner, A. Doron, E. Katz et al. Langmuir, 1996, 12, 946–954. 228. T. Suzawa, Y. Ikariyama, M. Aizawa, Anal. Chem., 1994, 66, 3889–3894. 229. P.N. Bartlett, R.G. Whitaker, M.J. Green, J. Frew, Chem. Commun., 1987, 1603–1604. 230. H.A. Krebs, Biochem. J., 1935, 24, 1620–1644. 231. O. Warburg, W. Christian, Biochem. Z., 1938, 298, 150–168. 232. M. Friedman, Agric. Food Chem., 1999, 47, 3457–3459. 233. M.S. Pilone, Cell Mol. Life Sci., 2000, 57, 1732–1747. 234. B. Curti, S. Ronchi, U. Branzoli et al. Biochim. Biophys. Acta, 1973, 327, 266–273. 235. M.P. Simonetta, M.A. Vanoni, P. Casalin, Biochim. Biophys. Acta, 1987, 914, 136–142. 236. L. Pollegioni, S. Buto, W. Tischer et al. Biochem. Mol. Biol. Int., 1993, 31, 709–717. 237. M. Pilone Simonetta, L. Pollegioni, P. Casalin et al. Eur. J. Biochem., 1989, 180, 190–204.

The Bioorganometallic Chemistry of Ferrocene

629

238. A. Mattevi, M.A. Vanoni, F. Todone et al. Proc. Natl. Acad. Sci. USA, 1996, 93, 7496–7501. 239. F. Todone, M.A. Vanoni, A. Mozzarelli et al. Biochemistry, 1997, 36, 5853–5860. 240. H. Mizutani, I. Miyahara, K. Hirotsu et al. J. Biochem., 1996, 120, 14–17. 241. R. Miura, C. Setoyama, Y. Nishina et al. J. Biochem., 1997, 122, 825–833. 242. T. Mizuno, I. Miyahara, K. Hirotsu et al. J. Biochem., 2000, 128, 73–81. 243. S. Umhau, L. Pollegioni, G. Molla et al. Proc. Natl. Acad. Sci. USA, 2000, 97, 12463–12468. 244. L. Pollegioni, K. Diederichs, G. Molla et al. J. Mol. Biol., 2002, 324, 535–546. 245. M.H. Gelb, D.C. Heimbrook, P. M¨alk¨onen, S.G. Sligar, Biochemistry, 1982, 21, 370–377. 246. T.L. Poulos, B.C. Finzel, I.C. Gunsalus et al. J. Biol. Chem., 1985, 260, 16122–16130. 247. T.L. Poulos, B.C. Finzel, A.J. Howard, Biochemistry, 1986, 25, 5314–5322. 248. T.L. Poulos, B.C. Finzel, A.J. Howard, J. Mol. Biol., 1987, 195, 687–700. 249. R. Raag, T.L. Poulos, Biochemistry, 1989, 28, 917–922. 250. R. Raag, T.L. Poulos, Biochemistry, 1991, 30, 2674–2684. 251. R. Raag, T.L. Poulos, Biochemistry, 1989, 28, 7586–7592. 252. T.L. Poulos, A.J. Howard, Biochemistry, 1987, 26, 8165–8174. 253. R. Raag, H. Li, B.C. Jones, T.L. Poulos, Biochemistry, 1993, 32, 4571–4578. 254. E.J. Mueller, P.J. Loida, S.G. Sligar, Twenty-ﬁve Years of P450cam Research, in Cytochrome P450: Structure, Mechanism, and Biochemistry, P.R. Ortiz de Montellano (Ed.), Plenum Press, New York, USA (1995), pp. 83–124. 255. K. Di Gleria, D.P. Nickerson, H.A.O. Hill et al. J. Am. Chem. Soc., 1998, 120, 46–52. 256. K-W.K. Lo, L-L. Wong, H.A.O. Hill, FEBS Lett., 1999, 451, 342–346. 257. N.M. Kostic, Comments Inorg. Chem., 1988, 8, 137–162. 258. A.J. Lukes, N.M. Kostic, J. Inorg. Biochem., 1992, 46, 77–85. 259. K. S¨unkel, T. Kießling, J. Organomet. Chem., 2001, 637–639, 796–799. 260. K.G. Welinder, Eur. J. Biochem., 1979, 96, 483–502. 261. Y. Shiro, M. Kurono, I. Morishima, J. Biol. Chem., 1986, 261, 9382–9390. 262. A.T. Smith, N. Santama, S. Dacey et al. J. Biol. Chem., 1990, 265, 13335–13343. 263. M. Gajhede, D.J. Schuller, A. Henriksen et al. Nature Struct. Biol., 1997, 4, 1032–1038. 264. V.N. Goral, A.D. Ryabov, Biochem. Mol. Biol. Int., 1998, 45, 61–71. 265. A.D. Ryabov, V.N. Goral, J. Biol. Inorg. Chem., 1997, 2, 182–190. 266. J.E. Frew, M.A. Harmer, H.A.O. Hill, S.I. Libor, J Electroanal. Chem. Interfac. Electrochem., 1986, 210, 1–10. 267. M.H. Smit, A.E.G. Cass, Anal. Chem., 1990, 62, 2429–2436. 268. J.C. Vidal, M.A. Yague, J.R. Castillo, Sensors and Actuators, 1994, B 21, 135–141. 269. W-C. Tsai, A.E.G. Cass, Analyst , 1995, 120, 2249–2254. 270. A.D. Ryabov, V.N. Goral, L. Gorton, E. Cs¨oregi, Chem. Eur. J., 1999, 5, 961–967. 271. N. Carolan, R.J. Forster, C. O’Fagain, Bioconjugate Chem., 2007, 18, 524–529. 272. T. Kijima, T. Suzuki, T. Izumi, J. Biosci. Bioeng., 2003, 96, 585–587. 273. C.B. Carpenter, T.J. Gill III, L.T. Mann Jr, J. Immun., 1967, 98, 236–250. 274. T.J. Gill III, L.T. Mann Jr, J. Immun., 1966, 96, 906–912. 275. L.T. Mann Jr, J. Labelld. Compd., 1967, 3, 87–97. 276. M. Peterlik, Monatsh. Chem., 1967, 98, 2133–2134. 277. H. Franz, Z. Chem., 1967, 7, 427–428. 278. H. Franz, Z. Chem., 1967, 7, 235–236. 279. H. Franz, Naturwissenschaften, 1967, 54, 339. 280. H. Franz, Histochemie, 1968, 12, 230–239. 281. H. Franz, Histochemie, 1968, 16, 220–226. 282. D.E. Allen, D.D. Perrin, J. Histochem. Cytochem., 1974, 22, 919–928. 283. A.J. Swift, J. Histochem. Cytochem., 1975, 23, 390–391.

630

Ferrocenes: Ligands, Materials and Biomolecules

284. 285. 286. 287. 288. 289. 290. 291. 292. 293. 294. 295. 296.

W. Wildf¨uhr, E. Mehner, Z. ges. Hyg., 1969, 15, 616–618. B. Wagner, M. Wagner, Zbl. Bakt. Hyg., I. Abt. Orig., 1972, A 222, 468–483. H. Franz, W. Wildf¨uhr, Z. Immun.-Forsch., 1971, 142, 334–341. B. Micheel, D. Bierwolf, A. Randt et al. Acta biol. med. germ., 1971, 27, 639–649. W. Wildf¨uhr, H. Franz, Zbl. Bakt., I. Abt. Orig., 1971, 216, 532–540. H. Franz, J. Mohr, Z. Inn. Med., 1969, 24, 673–675. K. Yasuda, N. Yamamoto, Acta Histochem. Cytochem., 1975, 8, 215–219. T-K. Lim, S. Imai, T. Matsunaga, Biotechnol. Bioeng., 2002, 77, 758–763. T-K. Lim, T. Matsunaga, Biosens. Bioelectronics, 2001, 16, 1063–1069. T-K. Lim, N. Nakamura, M. Ikehata, T. Matsunaga, Electrochemistry, 2000, 11, 872–874. T-K. Lim, H. Ohta, T. Matsunaga, Anal. Chem., 2003, 75, 3316–3321. S. Kossek, C. Padeste, L. Tiefenauer, J Mol. Recogn., 1996, 9, 485–487. L.X. Tiefenauer, S. Kossek, C. Padeste, P. Thi´ebaud, Biosens. Bioelectronics, 1997, 12, 213–223. J. Wang, A. Ib´anez, M.P. Chatrathi, Electrophoresis, 2002, 23, 3744–3749. M.J. Choi, S.Y. Kim, J. Choi, I.R. Paeng, Microchem. J., 1999, 63, 92–99. K. Yasuda, J. Electron Microsc., 1978, 27, 361. M. Okochi, H. Ohta, T. Tanaka, T. Matsunaga, Biotechnol. Bioengineer., 2005, 90, 14–19. T. Tanaka, S. Tsukube, K. Izawa et al. Biosens. Bioelectron., 2007, 22, 2051–2056. A.P. Deng, H.T. Liu, S.J. Jiang et al. Anal. Chim. Acta, 2002, 472, 55–61. W.C. Mak, K.Y. Cheung, D. Trau et al. Anal. Chem., 2005, 77, 2835–2841. M. Akram, M.C. Stuart, D.K.Y. Wong, Electroanalysis, 2006, 18, 237–246. M. Yasuzawa, H. Hamada, K. Oga, Y. Fuchiwaki, Intern. J. Modern Phys. B, 2003, 17, 1211–1216. A. Anne, C. Demaille, J. Moiroux, J. Am. Chem. Soc., 2001, 123, 4817–4825. A. Anne, C. Demaille, J. Moiroux, J. Am. Chem. Soc., 1999, 121, 10379–10388. N. Anicet, A. Anne, C. Bourdillon et al. Faraday Discuss., 2000, 269–279. http://www.expasy.org/uniprot/P02769 H. Falk, M. Peterlik, K. Schl¨ogl, Monatsh. Chem., 1969, 100, 787–788. F. Mizutani, M. Asai, Denki Kagaku, 1988, 56, 1100–1101. F. Mizutani, M. Asai, Amperometric Glucose Electrode with the use of Ferrocene-modiﬁed BSA as a Mediator, in Proceedings of the MRS International Meeting on Advanced Materials, M. Doyama, S. Somiya, R.P.H. Chang, I. Karube, (Eds.), Materials Research Society, Pittsburgh, Pennsylvania, USA (1989), pp. 147–152. H. Shinohara, T. Kusaka, E. Yokota et al. Sensors and Actuators, 2000, B65, 144–146. T. Suzawa, Y. Ikariyama, M. Aizawa, Bull. Chem. Soc. Jpn., 1995, 68, 165–171. V.S. Tripathi, V.B. Kandimalla, H.X. Ju, Biosens. Bioelectron., 2006, 21, 1529–1535. N.M. Green, Methods Enzymol., 1990, 184, 51–67. E.A. Bayer, M. Wilchek, J. Chromatogr., 1990, 510, 3–11. M. Wilchek, E.A. Bayer, Trends Biochem. Sci., 1989, 14, 408–412. O. Livnah, E.A. Bayer, M. Wilchek, J.L. Sussman, Proc. Natl. Acad. Sci. USA, 1993, 90, 5076–5080. L. Pugliese, A. Coda, M. Malcovati, M. Bolognesi, J. Mol. Biol., 1993, 231, 698–710. W.A. Hendrickson, A. P¨ahler, J.L. Smith et al. Proc. Natl. Acad. Sci. USA, 1989, 86, 2190–2194. P.C. Weber, D.H. Ohlendorf, J.J. Wendolowski, F.R. Salemme, Science, 1989, 243, 85–88. C. Padeste, A. Grubelnik, L. Tiefenauer, Biosens. Bioelectronics, 2000, 15, 431–438. B. Steiger, C. Padeste, A. Grubelnik, L. Tiefenauer, Electrochimica Acta, 2003, 48, 761–769. J.Y. Liu, S.J. Tian, L. Tiefenauer et al. Anal. Chem., 2005, 77, 2756–2761. J.Y. Liu, L. Tiefenauer, S.J. Tian et al. Anal. Chem., 2006, 78, 470–476.

297. 298. 299. 300. 301. 302. 303. 304. 305. 306. 307. 308. 309. 310. 311. 312.

313. 314. 315. 316. 317. 318. 319. 320. 321. 322. 323. 324. 325. 326.

The Bioorganometallic Chemistry of Ferrocene

631

327. C. Padeste, B. Steiger, A. Grubelnik, L. Tiefenauer, Biosens. Bioelectron., 2003, 19, 239–247. 328. C. Padeste, B. Steiger, A. Grubelnik, L. Tiefenauer, Biosens. Bioelectron., 2004, 20, 545–552. 329. A.J. Baca, F.M. Zhou, J. Wang et al. Electroanalysis, 2004, 16, 73–80. 330. J. Wang, J.H. Li, A.J. Baca et al. Anal. Chem., 2003, 75, 3941–3945. 331. O.A. Jimenez, S. Chikneyan, A.J. Baca et al. Environ. Sci. Technol., 2005, 39, 1209–1213. 332. N. Anicet, A. Anne, J. Moiroux, J.M. Sav´eant, J. Am. Chem. Soc., 1998, 120, 7115–7116. 333. A. Anne, Tetrahedron Lett., 1998, 39, 561–564. 334. S. Tanaka, K. Yoshida, H. Kuramitz et al. Anal. Sci., 1999, 15, 863–866. 335. B. Limoges, C. Degrand, P. Brossier, R.L. Blankespoor, Anal. Chem., 1993, 65, 1054–1060. 336. S. Rapicault, B. Limoges, C. Degrand, Anal. Chem., 1996, 68, 930–935. 337. M. Mosbach, W. Schuhmann, Sensors and Actuators, 2000, B70, 145–152. 338. M. Mosbach, T. Laurell, J. Nilsson et al. Biosens. Bioelectron., 2001, 16, 611–620. 339. G.R. Newkome, C.N. Mooreﬁeld, F. V¨ogtle, Dendrimers and dendrons, Wiley-VCH Verlag GmbH, Weinheim, Germany (2004). 340. S-E. Stiriba, H. Frey, R. Haag, Angew. Chem. Int. Ed., 2002, 41, 1329–1334. 341. D.A. Tomalia, L.A. Reyna, S. Svenson, Biochem. Soc. Trans., 2007, 35, 61–67. 342. G. Wu, R.F. Barth, W. Yang et al. J.M. Backer, Anticancer Agents Med. Chem., 2006, 6, 167–184. 343. C. Dufes, I.F. Uchegbu, A.G. Schatzlein, Adv. Drug Deliv. Rev., 2005, 57, 2177–2202. 344. D.A. Tomalia, A.M. Naylor, W.A. Goddard, Angew. Chem. Int. Ed. Engl., 1990, 29, 138–175. 345. E. Kim, K. Kim, H. Yang et al. Anal. Chem., 2003, 75, 5665–5672. 346. S.J. Kwon, E. Kim, H. Yang, J. Kwak, Analyst , 2006, 131, 402–406. 347. H.C. Yoon, M.Y. Hong, H.S. Kim, Anal. Chem., 2000, 72, 4420–4427. 348. H.C. Yoon, M.Y. Hong, H.S. Kim, Anal. Biochem., 2000, 282, 121–128. 349. H.C. Yoon, D. Lee, H.S. Kim, Anal. Chim. Acta, 2002, 456, 209–218. 350. J. Das, M.A. Aziz, H. Yang, J. Am. Chem. Soc., 2006, 128, 16022–16023. 351. B. Alonso, P. Garcia Armada, J. Losada et al. Biosens. Bioelectron., 2004, 19, 1617–1625. 352. M.P. Garcia Armada, J. Losada, M. Zamora et al. Bioelectrochemistry, 2006, 69, 65–73. 353. J. Losada, M. Zamora, P. Garcia Armada et al. Anal. Bioanal. Chem., 2006, 385, 1209–1217. 354. G. Taborsky, Adv. Protein Chem., 1974, 28, 1–210. 355. L. McReynolds, B.W. O’Malley, A.D. Nisbet et al. Nature, 1978, 273, 723–728. 356. A.D. Nisbet, R.I. Saundry, A.J.G. Moir et al. Eur. J. Biochem., 1981, 115, 335–345. 357. K. Narita, J. Ishii, J. Biochem., 1962, 52, 367–373. 358. P.E. Stein, A.G.W. Leslie, J.T. Finch, R.W. Carrell, J. Mol. Biol., 1991, 221, 941–959. 359. G. Lowe, Tetrahedron, 1976, 32, 291–302. 360. S.S. Husain, G. Lowe, Biochem. J., 1970, 116, 689–692. 361. S.S. Husain, G. Lowe, Biochem. J., 1969, 114, 279–288. 362. R.E.J. Mitchel, I.M. Chaiken, E.L. Smith, J. Biol. Chem., 1970, 245, 3485–3492. 363. J. Drenth, J.N. Jansonius, R. Koekoek et al. Nature, 1968, 218, 929–932. 364. K.T. Douglas, O.S. Ejim, K. Taylor, J. Enz. Inhib., 1992, 6, 233–242. 365. P. Haquette, M. Salmain, K. Svedlung et al. ChemBioChem, 2007, 8, 224–231. 366. H.J. Hwang, J.R. Carey, E.T. Brower et al. J. Am. Chem. Soc., 2005, 127, 15356–15357. 367. H. Hjelm, J. Sjodahl, J. Sjoquist, Eur. J. Biochem., 1975, 57, 395–403. 368. R. Lindmark, K. Thoren-Telling, J. Sjoquist, J. Immunol. Methods, 1983, 62, 1–13. 369. M. Okochi, H. Ohta, T. Taguchi et al. Electrochim. Acta, 2005, 51, 952–955. 370. E.K. Wilson, Chem. Eng. News, 1998, 47–49.

632

Ferrocenes: Ligands, Materials and Biomolecules

371. 372. 373. 374. 375. 376.

A.E. Beilstein, M.T. Tierney, M.W. Grinstaff, Comments Inorg. Chem., 2000, 22, 105–127. X. Hu, S.S. Lee, M.W. Grinstaff, Methods Enzymol., 2002, 353, 548. P. Meunier, I. Ouattara, B. Gautheron et al. Eur. J. Med. Chem., 1991, 26, 351–362. K. Sonogashira, Y. Tohda, N. Hagihara, Tetrahedron Lett., 1975, 50, 4467–4470. C.J. Yu, H. Yowanto, Y. Wan et al. J. Am. Chem. Soc., 2000, 122, 6767–6768. E. Coutouli-Argyropoulou, M. Tsitabani, G. Petrantonakis et al. Org. Biomol. Chem., 2003, 1, 1382–1388. M.J. Robins, P.J. Barr, Tetrahedron Lett., 1981, 421. M.J. Robins, R.S. Vinayak, S.G. Wood, Tetrahedron Lett., 1990, 31, 3731. M.J. Robins, P.J. Barr, J. Org. Chem., 1983, 48, 1854. C. McGuigan, C.J. Yarnold, G. Jones et al. J. Med. Chem., 1999, 42, 4479. A. Appel, F. J¨akle, T. Priermeier et al. Organometallics, 1996, 15, 1188–1194. A.R. Pike, L.C. Ryder, B.R. Horrocks et al. Chem. Eur. J., 2002, 8, 2891–2899. ˇ epniˇcka, J. Ludvik et al. Chem. Eur. J., 2004, 10, 2058–2066. M. Hocek, P. Stˇ A. Houlton, C.J. Isaac, A.E. Gibson et al. Dalton Trans., 1999, 3229–3234. C. Price, M. Aslanoglu, C.J. Isaac et al. Dalton Trans., 1996, 4115–4120. S-C. Chen, J. Organomet. Chem., 1980, 202, 183–189. R.S. Glass, N.Y.T. Stessman, Tetrahedron Lett., 2000, 41, 9581–9584. E. Bucci, L. De Napoli, G. Di Fabio et al. Tetrahedron, 1999, 55, 14435–14450. H.F. Arlinghaus, N. Thonnard, M.T. Spaar et al. Anal. Chem., 1991, 63, 402–407. K.B. Jacobson, H.F. Arlinghaus, H.W. Schmitt et al. Genomics, 1991, 9, 51–59. F.V. Slopp, G.M. Brown, R.A. Sachleben et al. New J. Chem., 1994, 18, 317–326. O. Brosch, T. Weyherm¨uller, N. Metzler-Nolte, Eur. J. Inorg. Chem., 2000, 323–330. A.E. Beilstein, M.W. Grinstaff, Chem. Commun., 2000, 509–510. A.E. Beilstein, M.W. Grinstaff, J. Organomet. Chem., 2001, 637–639, 398–406. S.I. Khan, A.E. Beilstein, M.T. Tierney et al. Inorg. Chem., 1999, 38, 5999–6002. S.I. Khan, M.W. Grinstaff, J. Am. Chem. Soc., 1999, 121, 4704–4705. A.R. Pike, L.C. Ryder, B.R. Horrocks et al. Chem. Eur. J., 2005, 11, 344–353. A. Okamoto, K. Tainaka, I. Saito, Tetrahedron Lett., 2002, 43, 4581–4583. A. Anne, B. Blanc, J. Moiroux, Bioconjugate Chem., 2001, 12, 396–405. J.J. Gooding, Electroanalysis, 2002, 14, 1149–1156. R.C. McGlennen, Clin. Chem., 2001, 47, 393–402. M.I. Pividori, A. Merkoci, S. Alegret, Biosens. Bioelectron., 2000, 15, 291–303. P. de-los-Santos-Alvarez, M.J. Lobo-Castanon, A.J. Miranda-Ordieres, P. Tunon-Blanco, Anal. Bioanal. Chem., 2004, 378, 104–118. T.G. Drummond, M.G. Hill, J.K. Barton, Nature Biotechnol., 2003, 21, 1192–1199. K. Kerman, M. Kobayashi, E. Tamiya, Meas. Sci. Technol., 2004, 15, R1–R11. J. Wang, Anal. Chim. Acta, 2002, 469, 63–71. T. Ihara, Y. Maruo, S. Takenaka, M. Takagi, Nucl. Acids Res., 1996, 24, 4273–4280. S. Takenaka, Bull. Chem. Soc. Jpn, 2001, 74, 217–224. S. Takenaka, Y. Uto, H. Kondo et al. Anal. Biochem., 1994, 218, 436–443. S. Takenaka, K. Yamashita, M. Takagi et al. Anal. Chem., 2000, 72, 1334–1341. H. Eckert, M. Koller, J. Liquid Chromatogr., 1990, 13, 3399–3414. H. Eckert, M. Koller, Z. Naturforsch., 1990, 45b, 1709–1714. S. Beck, H. K¨oster, Anal. Chem., 1990, 62, 2258–2270. L. Petraccone, E. Erra, A. Messere et al. Biophys. Chem., 2003, 104, 259–270. C. Xu, P. He, Y. Fang, Anal. Chim. Acta, 2000, 411, 31–36. C. Xu, H. Cai, P. He, Y. Fang, Analyst , 2001, 126, 62–65. S.C. Hillier, S.E. Flower, C.G. Frost et al. Electrochem. Commun., 2004, 6, 1227–1232. R.C. Mucic, M.K. Herrlein, C.A. Mirkin, R.L. Letsinger, Chem. Commun., 1996, 555–557. Y-T. Long, C-Z. Li, T.C. Sutherland et al. J. Am. Chem. Soc., 2003, 125, 8724–8725.

377. 378. 379. 380. 381. 382. 383. 384. 385. 386. 387. 388. 389. 390. 391. 392. 393. 394. 395. 396. 397. 398. 399. 400. 401. 402. 403. 404. 405. 406. 407. 408. 409. 410. 411. 412. 413. 414. 415. 416. 417. 418. 419.

The Bioorganometallic Chemistry of Ferrocene 420. 421. 422. 423. 424. 425. 426. 427. 428. 429. 430. 431. 432. 433. 434. 435. 436. 437. 438. 439. 440. 441. 442. 443. 444. 445. 446. 447. 448. 449. 450. 451. 452. 453. 454. 455. 456. 457. 458. 459. 460. 461. 462. 463. 464. 465. 466.

633

M. Chahma, J.S. Lee, H-B. Kraatz, J. Electroanal. Chem., 2004, 567, 283–287. A. Anne, A. Bouchardon, J. Moiroux, J. Am. Chem. Soc., 2003, 125, 1112–1113. F.J. Mearns, E.L.S. Wong, K. Short et al. Electroanalysis, 2006, 18, 1971–1981. C. Fan, K.W. Plaxco, A.J. Heeger, Proc. Natl. Acad. Sci. USA, 2003, 100, 9134–9137. C.E. Immoos, S.J. Lee, M.W. Grinstaff, Chembiochem, 2004, 5, 1100–1103. D.M. Jenkins, B. Chami, M. Kreuzer et al. Anal. Chem., 2006, 78, 2314–2318. C.E. Immoos, S.J. Lee, M.W. Grinstaff, J. Am. Chem. Soc., 2004, 126, 10814–10815. M. Inouye, R. Ikeda, M. Takase et al. Proc. Natl. Acad. Sci. USA, 2005, 102, 11606–11610. F. Patolsky, Y. Weizmann, I. Willner, J. Am. Chem. Soc., 2002, 124, 770–772. W.A. Wlassoff, G.C. King, Nucleic Acids Res., 2002, 30, e58. Y. Uto, H. Kondo, M. Abe et al. Anal. Biochem., 1997, 250, 122–124. S.A. Brazill, P.H. Kim, W.G. Kuhr, Anal. Chem., 2001, 73, 4882–4890. S.A. Brazill, W.G. Kuhr, Anal. Chem., 2002, 74, 3421–3428. S.A. Brazill, P. Singhal, W.G. Kuhr, Anal. Chem., 2000, 72, 5542–5548. D.A. Di Giusto, W.A. Wlassoff, S. Giesebrecht et al. Angew. Chem. Int. Ed., 2004, 43, 2809–2812. G.C. King, D.A. Di Giusto, W.A. Wlassoff et al. Hum. Mutat., 2004, 23, 420–425. K. Mukumoto, T. Nojima, S. Sato et al. Anal. Sci., 2007, 23, 115–119. K. Mukumoto, T. Nojima, S. Takenaka, Tetrahedron, 2005, 61, 11705–11715. T. Ihara, M. Nakayama, M. Murata et al. Chem. Commun., 1997, 1609–1610. M. Nakayama, T. Ihara, K. Nakano, M. Maeda, Talanta, 2002, 56, 857–866. C.J. Yu, H. Wang, Y. Wan et al. J. Org. Chem., 2001, 66, 2937–2942. C.J. Yu, Y. Wan, H. Yowanto et al. J. Am. Chem. Soc., 2001, 123, 11155–11161. N.S. Swami, C.F. Chou, R. Terberueggen, Langmuir, 2005, 21, 1937–1941. R.M. Umek, S.W. Lin, J. Vielmetter et al. J. Mol. Diagn., 2001, 3, 74–84. S. Creager, C.J. Yu, C. Bamdad et al. J. Am. Chem. Soc., 1999, 121, 1059–1064. R.M. Umek, S.S. Lin, Y-P. Chen et al. Mol. Diagn., 2000, 5, 321–328. S.H. Bernacki, D.H. Farkas, W.M. Shi et al. Arch. Pathol. Lab. Med., 2003, 127, 1565–1572. S.D. Vernon, D.H. Farkas, E.R. Unger et al. BMC Infectious Diseases, 2003, 3, Art no. 12; doi:10.1186/1471233411831112. http://www.osmetech.com/ J.M. Gibbs, S.J. Park, D.R. Anderson et al. J. Am. Chem. Soc., 2005, 127, 1170–1178. Z.S. Wu, J.H. Jiang, G.L. Shen, R.Q. Yu, Hum. Mutat., 2007, 28, 630–637. K. Kim, H. Yang, S.H. Park et al. Chem. Commun., 2004, 1466–1467. P. Liepold, H. Wieder, H. Hillebrandt et al. Bioelectrochem., 2005, 67, 143–150. http://www.frizbiochem.com S. Takenaka, Y. Uto, H. Saita et al. Chem. Commun., 1998, 1111–1112. S. Takenaka, Y. Uto, M. Takagi, H. Kondo, Chem. Lett., 1998, 989–990. S. Sato, S. Fujii, K. Yamashita et al. J. Organomet. Chem., 2001, 637, 476–483. K. Yamashita, M. Takagi, H. Kondo, S. Takenaka, Anal. Biochem., 2002, 306, 188–196. K. Yamashita, A. Takagi et al. Bioconjugate Chem., 2002, 13, 1193–1199. H. Miyahara, K. Yamashita, M. Kanai et al. Talanta, 2002, 56, 829–835. J. Wakai, A. Takagi, M. Nakayama et al. Nucleic Acids Res., 2004, 32, e141; doi:110.1093/nar/gnh1141. http://www.tum-gene.com C. Xu, H. Cai, P. He, Y. Fang, Fresenius J. Anal. Chem., 2000, 367, 593–595. G. Zauner, Y.T. Wang, M. Lavesa-Curto et al. Analyst , 2005, 130, 345–349. S. Sato, H. Kondo, T. Nojima, S. Takenaka, Anal. Chem., 2005, 77, 7304–7309. S. Sato, K. Hokazono, T. Irie et al. Anal. Chim. Acta, 2006, 578, 82–87. M. Rimmele, Chembiochem, 2003, 4, 963–971.

634

Ferrocenes: Ligands, Materials and Biomolecules

467. S. Tombelli, M. Minunni, M. Mascini, Biosens. Bioelectron., 2005, 20, 2424–2434. 468. J.L. Acero-Sanchez, E. Baldrich, A.E.G. Radi et al. Electroanalysis, 2006, 18, 1957–1962. 469. A.E. Radi, J.L.A. Sanchez, E. Baldrich, C.K. O’Sullivan, J. Am. Chem. Soc., 2006, 128, 117–124. 470. P.E. Nielsen, Acc. Chem. Res., 1999, 32, 624–630. 471. P.E. Nielsen, M. Egholm, R.H. Berg, O. Buchardt, Science, 1991, 254, 1497–1500. 472. E. Uhlmann, A. Peyman, G. Breipohl, D.W. Will, Angew. Chem. Int. Ed., 1998, 37, 2797–2823. 473. K.L. Dueholm, M. Engholm, C. Behrens et al. J. Org. Chem., 1994, 59, 5767–5773. 474. S.A. Thomson, J.A. Josey, R. Cadilla et al. Tetrahedron, 1995, 51, 6179–6194. 475. M. Egholm, O. Buchardt, L. Christensen et al. Nature, 1993, 365, 566–568. 476. M. Leijon, A. Gr¨aslund, P.E. Nielsen et al. Biochemistry, 1994, 33, 9820–9825. 477. S.C. Brown, S.A. Thomson, J.M. Veal, D.G. Davis, Science, 1994, 265, 777–780. 478. T. Ratilainen, A. Holm´en, E. Tuite et al. Biochemistry, 1998, 31, 12331–12342. 479. T. Ratilainen, A. Holm´en, E. Tuite et al. Biochemistry, 2000, 39, 7781–7791. 480. M. Eriksson, P.E. Nielsen, Q. Rev. Biophys., 1996, 29, 369–394. 481. H.J. Larsen, T. Bentin, P.E. Nielsen, Biochim. Biophys. Acta, 1999, 1489, 159–166. 482. A. Hess, N. Metzler-Nolte, Chem. Commun., 1999, 885–886. 483. G. Gasser, M.J. Belousoff, A.M. Bond, L. Spiccia, J. Org. Chem., 2006, 71, 7565–7573. 484. C. Baldoli, L. Falciola, E. Licandro et al. J. Organomet. Chem., 2004, 689, 4791. 485. C. Baldoli, C. Rigamonti, S. Maiorana et al. Chem. Eur. J., 2006, 12, 4091–4100. 486. C. Baldoli, E. Licandro, S. Maiorana et al. J. Electroanal. Chem., 2005, 585, 197. 487. C. Baldoli, P. Cerea, C. Giannini et al. Synlett , 2005, 1984–1994. 488. J.C. Verheijen, G.A. van der Marel, J.H. van Boom, N. Metzler-Nolte, Bioconjugate Chem., 2000, 11, 741–743. 489. A. Maurer, H-B. Kraatz, N. Metzler-Nolte, Eur. J. Inorg. Chem., 2005, 3207–3210. 490. A. Maurer, H-B. Kraatz, N. Metzler-Nolte, 2007, unpublished results. 491. C.L. Ferreira, C.B. Ewart, C.A. Barta et al. Inorg. Chem., 2006, 45, 8414–8422. 492. J.M. Casas-Solvas, A. Vargas-Berenguel, L.F. Capitan-Vallvey, F. Santoyo-Gonzalez, Org. Lett., 2004, 6, 3687–3690. 493. A.N. de Belder, E.J. Bourne, J.B. Pridham, J. Chem. Soc., 1961, 4464–4467. 494. P.M. Collins, W.G. Overend, B.A. Rayner, Perkin Trans. II , 1973, 310–314. 495. C.C. Jones, M.L. Sinnott, I.J.L. Souchard, Perkin Trans. II , 1977, 1191–1198. 496. M.J. Adam, L.D. Hall, Chem. Commun., 1979, 865–866. 497. M.J. Adam, L.D. Hall, Can. J. Chem., 1980, 58, 1188–1197. 498. M.J. Adam, L.D. Hall, J. Organomet. Chem., 1980, 186, 289–296. 499. J.S. Landells, J.L. Kerr, D.S. Larsen et al. Dalton Trans., 2000, 1403–1409. 500. J.L. Kerr, J.S. Landells, D.S. Larsen et al. Dalton Trans., 2000, 1411–1417. 501. M. Schneider, M. Wenzel, J. Labelld. Compd. Radiopharm., 1979, 18, 293–301. 502. G. Grynkiewicz, J.N. BeMiller, Carbohydrate Res., 1984, 131, 273–276. 503. R. Huisgen, Angew. Chem. Int. Ed. Engl., 1963, 2, 565–598. 504. H.C. Kolb, M.G. Finn, K.B. Sharpless, Angew. Chem. Int. Ed., 2001, 40, 2004–2021. 505. M. Chahma, J.S. Lee, H-B. Kraatz, J. Organomet. Chem., 2002, 648, 81–86. 506. S. Shirakami, T. Itoh, Tetrahedron Asym., 2000, 11, 2823–2833. 507. M. Takeuchi, T. Mizuno, S. Shinkai et al. Tetrahedron Asym., 2000, 11, 3311–3322. 508. T. Itoh, S. Shirakami, J. Synth. Org. Chem. Jpn., 2002, 60, 232–239. 509. T. Itoh, S. Shirakami, Y. Nakao, T. Yoshida, Chem. Lett., 1998, 979–980. 510. T. Itoh, S. Shirakami, N. Ishida et al. Bioorg. Med. Chem. Lett., 2000, 10, 1657–1659. 511. C.G. Hartinger, A.A. Nazarov, V.B. Arion et al. New J. Chem., 2002, 26, 671–673. 512. P.R. Ashton, V. Balzani, M. Clemente-Leon et al. Chem. Eur. J., 2002, 8, 673–684. 513. I. Suzuki, Q. Chen, A. Ueno, T. Osa, Bull. Chem. Soc. Jpn., 1993, 66, 1472–1481.

The Bioorganometallic Chemistry of Ferrocene

635

514. A. Ueno, I. Suzuki, T. Osa, Makromol. Chem., Rapid Commun., 1987, 8, 131–134. 515. A. Ueno, F. Moriwaki, T. Matsue, T. Osa, Makromol. Chem., Rapid Commun., 1985, 6, 231–233. 516. H-J. Thiem, M. Brandl, R. Breslow, J. Am. Chem. Soc., 1988, 110, 8612–8616. 517. F.M. Menger, M.J. Sherrod, J. Am. Chem. Soc., 1988, 110, 8606–8611. 518. N. Kobayashi, M. Opallo, Chem. Commun., 1990, 477–479. 519. Y.f. Chen, I.A. Banerjee, L. Yu et al. Langmuir, 2004, 20, 8409–8413. 520. P. Vedso, R. Chauvin, Z. Li et al. Helv. Chim. Acta, 1994, 77, 1631–1659. 521. A. Vasella, Z. Li, Helv. Chim. Acta, 1996, 79, 2201–2218. 522. C. Widauer, B. Bernet, A. Vasella, Synth. Commun., 1998, 28, 593–602. 523. B. Rießelmann, M. Wenzel, Hoppe-Seyler’s Z. Physiol. Chem., 1977, 358, 1353–1357. 524. M. Cais, S. Dani, Y. Eden et al. Nature, 1977, 270, 534–535. 525. M. Cais, L’ actualite chimique, 1979, September, 14–26. 526. M. Salmain, A. Vessi`eres, P. Brossier et al. J. Immun. Meth., 1992, 148, 65–75. 527. A. Vessi`eres, M. Salmain, P. Brossier, G. Jaouen, J. Pharm. Biomed. Anal., 1999, 21, 625–633. 528. A. Vessi`eres, C. Vaillant, M. Gruselle et al. Chem. Commun., 1990, 837–839. 529. D. Vichard, M. Gruselle, G. Jaouen et al. J. Organomet. Chem., 1994, 484, 1–8. 530. D. Osella, O. Gambino, G.C. Dutto et al. Inorg. Chim. Acta, 1994, 218, 207–210. 531. M. Wenzel, C. Klinge, J. Labelld. Compd. Radiopharm., 1994, 34, 981–987. 532. D. Osella, C. Nervi, F. Galeotti et al. Helv. Chim. Acta, 2001, 84, 3289–3298. 533. S. Green, P. Walter, V. Kumar et al. Nature, 1986, 320, 134–139. 534. G.G.J.M. Kuiper, E. Enmark, M. Pelto-Huikko et al. Proc. Natl. Acad. Sci. USA, 1996, 93, 5925–5930. 535. K. Hoffmann, B. Rießelmann, M. Wenzel, Liebigs Ann. Chem., 1980, 1181–1185. 536. M. Cais, E. Slovin, L. Snarsky, J. Organomet. Chem., 1978, 160, 223–230. 537. R. Krieg, R. Wyrwa, U. M¨ollmann et al. Steroids, 1998, 63, 531–541. 538. K. Shimada, E. Nagashima, S. Orii, T. Nambara, J. Pharm. Biomed. Anal., 1987, 5, 361–368. 539. K. Shimada, S. Orii, M. Tanaka, T. Nambara, J. Chromatogr., 1986, 352, 329–335. 540. D. Heydenhauss, G. Jaenecke, H. Schubert, Z. Chem., 1973, 13, 295. 541. E. Coutouli-Argyropoulou, M. Tsitabani, G. Petrantonakis et al. Org. Biomol. Chem., 2003, 1, 1382–1388. 542. A. Koller, Diss. Abstr. Int. B, 1975, 35, 5820. 543. K. Shimada, C. Sakayori, T. Nambara, J. Liq. Chromatogr., 1987, 10, 2177–2187. 544. I. Tranchant, A.C. Herve, S. Carlisle et al. Bioconjugate Chem., 2007, 18, 199–208. 545. I. Tranchant, A.C. Herve, S. Carlisle et al. Bioconjugate Chem., 2006, 17, 1256–1264. 546. L.L. Troitskaya, V.I. Sokolov, Russ. Chem. Bull., 1985, 34, 1548. 547. L.L. Troitskaya, V.I. Sokolov, J. Organomet. Chem., 1985, 285, 389–393. 548. W.H. Kim, Diss. Abstr. Int. B, 1978, 39, 1774. 549. N.L. Abbott, C.M. Jewell, M.E. Hays et al. J. Am. Chem. Soc., 2005, 127, 11576–11577. 550. K. Di Gleria, H.A.O. Hill, C.J. McNeil, M.J. Green, Analytical Chemistry, 1986, 58, 1203–1205. 551. I. Lavastre, J. Besancon, P. Brossier, C. Moise, Appl. Organomet. Chem., 1991, 5, 143–149. 552. A.L. Bordes, B. Schollhorn, B. Limoges, C. Degrand, Appl. Organom. Chem., 1998, 12, 59–65. 553. N.J. Forrow, N.C. Foulds, J.E. Frew, J.T. Law, Bioconjugate Chem., 2004, 15, 137–144. 554. H. Funabashi, Y. Tanaka, Y. Imamura et al. Biosens. Bioelectron., 2006, 21, 1675–1683. 555. E.P. Diamandis, T.K. Christopoulos, Past, present and future of immunoassays, in Immunoassay, E.P. Diamandis, T.K. Christopoulos (Eds.), Academic Press, San Diego, USA (1996), pp. 1–3.

636

Ferrocenes: Ligands, Materials and Biomolecules

556. 557. 558. 559. 560. 561. 562. 563.

R.S. Yalow, S.A. Berson, J. Clin. Invest., 1960, 39, 1157–1175. M. Cais, Methods Enzymol., 1983, 92, 445–458. S.G. Weber, W.C. Purdy, Anal. Lett., 1979, 12, 1–9. G.A. Robinson, G. Martinazzo, G.C. Forrest, J. Immunoassay, 1986, 7, 1–15. T. Yao, G.A. Rechnitz, Biosensors, 1988, 3, 307–312. A.L. Bordes, B. Limoges, P. Brossier, C. Degrand, Anal. Chim. Acta, 1997, 356, 195–203. B.K. Van Weeman, A.H.W.M. Schuurs, FEBS Lett., 1971, 15, 232–236. P. Treloar, J. Kane, P. Vagdama, Electrochemical immunoassays, in Principles and practice of immunoassays, 2nd Ed., C.P. Price, D.J. Newman, (Eds.), MacMillan, London, UK (1997). A. Le Gal La Salle, B. Limoges, C. Degrand, J. Electroanal. Chem., 1994, 379, 281–291. A. Le Gal La Salle, B. Limoges, C. Degrand, P. Brossier, Anal. Chem., 1995, 67, 1245–1253. B. Limoges, C. Degrand, Anal. Chem., 1996, 68, 4141–4148. P. K¨opf-Maier, H. K¨opf, Chem. Rev., 1987, 89, 1137–1152. P. K¨opf-Maier, H. K¨opf, Struct. Bond., 1988, 70, 105–185. R.P. Hanzlik, W.H. Soine, J. Am. Chem. Soc., 1978, 100, 1290–1291. L. Golberg, L.E. Martin, Life Sciences, 1964, 3, 1465–1474. R.A. Yeary, Toxicol. Appl. Pharmacol., 1969, 15, 666–676. H.W. Leung, D.W. Hallesy, L.D. Shott et al. Toxicol. Lett., 1987, 38, 103–108. J.L. Madinaveitia, British J. Pharmacol. Chemother., 1965, 24, 352–359. A.R. Dahl, T.J. Briner, Toxicol. Appl. Pharmacol., 1980, 56, 232–239. A.R. Dahl, W.M. Hadley, Crit. Rev. Toxicol., 1991, 21, 345–372. K.J. Nikula, J.D. Sun, E.B. Barr et al. Fundam. Appl. Toxicol., 1993, 21, 127–139. J.D. Sun, A.R. Dahl, N.A. Gillett et al. Fundam. Appl. Toxicol., 1991, 17, 150–158. J.T. Yarrington, K.W. Huffman, G.A. Leeson et al. Fundam. Appl. Toxicol., 1983, 3, 86–94. L.G. Valerio Jr, D.R. Petersen, Toxicol Lett , 1998, 98, 31–39. L.G. Valerio Jr, D.R. Petersen, Exp. Mol. Pathol., 2000, 68, 1–12. B. Dresow, C. Albert, I. Zimmermann, P. Nielsen, Hepatol., 1995, 21, 1099–1105. R.J. Ward, A.L. Florence, D. Baldwin et al. Eur. J. Biochem., 1991, 202, 405–410. R.J. Ward, M. Ramsey, D.P. Dickson et al. Eur. J. Biochem., 1994, 225, 187–194. R.J. Ward, L.C. Kuhn, P. Kaldy et al. Eur. J. Biochem., 1994, 220, 927–931. A. Braumann, U. Wulfhekel, J. Duellmann, P. Nielsen, Acta Anatomica, 1992, 144, 285–295. P. Nielsen, H.C. Heinrich, Biochem. Pharmacol., 1993, 45, 385–391. E.E. Cable, J.R. Connor, H.C. Isom, Am. J. Pathol., 1998, 152, 781–792. B. Magens, J. Dullmann, K. Schumann et al. Acta Histochem., 2005, 107, 57–65. B. Scheiber-Mojdehkar, I. Zimmermann, B. Dresow, H. Goldenberg, J. Hepatol., 1999, 31, 61–70. E.E. Cable, H.C. Isom, Drug Metabol. Dispos., 1999, 27, 255–260. R.W. Lambrecht, J.W. Cable, J.A. Pepe, H.L. Bonkovsky, Biochem. Pharmacol., 1994, 47, 1669–1676. P. Nielsen, J. Duellmann, U. Wulfhekel, H.C. Heinrich, Int. J. Biochem., 1993, 25, 223–232. T. Hertz, J. Duellman, P. Nielsen, U. Wulfhekel, Cells Hep. Sinusoid , 1993, 4, 407–409. P. Prunescu, C.C. Prunescu, Rev. Roumaine Biol. Ser. Biol. Animale, 2004, 47, 65–70. A. Braumann, U. Wulfhekel, P. Nielsen et al. Acta Anatomica, 1994, 150, 45–54. E.A. Malecki, E.E. Cable, H.C. Isom, J.R. Connor, Biol. Trace El. Res., 2002, 86, 73–84. R.J. Ward, D. Dexter, A. Florence et al. Biochem. Pharmacol., 1995, 49, 1821–1826. P. Nielsen, M. Furtjes, B. Dresow et al. Drugs Today, 1992, 28, 45–53. V. Eybl, D. Kotyzova, M. Kolek et al. Toxicol. Lett., 2002, 128, 169–175.

564. 565. 566. 567. 568. 569. 570. 571. 572. 573. 574. 575. 576. 577. 578. 579. 580. 581. 582. 583. 584. 585. 586. 587. 588. 589. 590. 591. 592. 593. 594. 595. 596. 597. 598. 599.

The Bioorganometallic Chemistry of Ferrocene 600. 601. 602. 603. 604. 605. 606. 607. 608. 609. 610. 611. 612. 613. 614. 615. 616. 617. 618. 619. 620. 621. 622. 623. 624. 625. 626. 627. 628. 629. 630. 631. 632. 633. 634. 635. 636. 637. 638. 639. 640. 641. 642. 643.

637

S.C. Beutelspacher, N. Serbecic, P.H. Tan et al. J. Nutr. Sc. Vitaminol., 2004, 50, 78–86. A.N. Nesmeyanov, L.G. Bogomolova, V. Viltchevskaya et al. US Patent, 119 356 (1971). H. K¨opf, P. K¨opf-Maier, Angew. Chem., 1979, 91, 509. P. K¨opf-Maier, B. Hesse, R. Voigtl¨ander, H. K¨opf, J. Cancer Res. Clin. Oncol., 1980, 97, 31–39. P. K¨opf-Maier, H. K¨opf, Drugs Future, 1986, 11, 297–320. P. K¨opf-Maier, H. K¨opf, E.W. Neuse, Angew. Chem., 1984, 96, 446–447. D. Osella, M. Ferrali, P. Zanello et al. Inorg. Chim. Acta, 2000, 306, 42–48. H. Tamura, M. Miwa, Chem. Lett., 1997, 26, 1177–1178. G. Tabbi, C. Cassino, G. Cavigiolio et al. J. Med. Chem., 2002, 45, 5786–5796. G. Mokdsi, M.M. Harding, J. Inorg. Biochem., 2001, 83, 205–209. Y.N. Vashisht Gopal, D. Jayaraju, A.K. Kondapi, Arch. Biochem. Biophys., 2000, 376, 229–235. M. Wenzel, Y. Wu, E. Liss, E.W. Neuse, Z. Naturforsch., 1988, 43c, 963–966. P. Kovavic, W.J. Popp, J.R. Ames, M.D. Ryan, Anti-Cancer Drug Design, 1988, 3, 205–216. U. Schatzschneider, N. Metzler-Nolte, Angew. Chem. Int. Ed., 2006, 45, 1504–1507. G. Caldwell, M.G. Meirim, E.W. Neuse, C.E.J. van Rensburg, Appl. Organomet. Chem., 1998, 12, 793–799. E.W. Neuse, Macromol. Symp., 2001, 172, 127–138. E.W. Neuse, F. Kanzawa, Appl. Organomet. Chem., 1990, 19–26. J.C. Swarts, D.M. Swarts, D.M. Maree et al. Anticancer Res., 2001, 21, 2033–2038. M.T. Johnson, E. Kreft, D.D. N’Da et al. J. Inorg. Organomet. Polymers, 2003, 13, 255–267. V.J. Fiorina, R.J. Dubois, S. Brynes, J. Med. Chem., 1978, 21, 393–395. P. K¨opf-Maier, H. K¨opf, E.W. Neuse, J. Cancer Res. Clin. Oncol., 1984, 108, 336–340. E.W. Neuse, J. Inorg. Organomet. Polymers, 2005, 15, 3–32. L. Weissﬂoch, M. Wagner, T. Probst et al. BioMetals, 2001, 14, 43–49. A. Houlton, R.M.G. Roberts, G. Silver, J. Organomet. Chem., 1991, 418, 107–112. C-W. Ong, J-Y. Jeng, S-S. Juang, C-F. Chen, Bioorg. Med. Chem. Lett., 1992, 2, 929–932. A. Ferle-Vidovic, M. Poljak-Blazi, V. Rapic, D. Skare, Cancer Biotherapy & Radiopharmaceuticals, 2000, 15, 617–624. N. Motohashi, R. Meyer, S.R. Gollapudi, K.R. Bhattiprolu, J. Organomet. Chem., 1990, 398, 205–217. P. James, J. Neud¨orﬂ, M. Eissmann et al. Org. Lett., 2006, 8, 2763–2766. D. Schlawe, A. Majdalani, J. Velcicky et al. Angew. Chem. Int. Ed., 2004, 43, 1731–1734. J. Velcicky, A. Lanver, J. Lex et al. Chem. Eur. J., 2004, 10, 5087–5110. G. Jaouen, S. Top, A. Vessieres et al. Curr. Med. Chem., 2004, 11, 2505–2517. A.K. Shiau, D. Barstad, P.M. Lorisa et al. Cell , 1998, 95, 927–937. P. Pigeon, S. Top, A. Vessi`eres et al. J. Med. Chem., 2005, 48, 2814–2821. S. Top, E.B. Kaloun, A. Vessi`eres et al. J. Organomet. Chem., 2002, 643–644, 350–356. E. Hillard, A. V`essieres, L. Thouin et al. Angew. Chem. Int. Ed., 2006, 45, 285–290. P.W. Fan, F. Zhang, J.L. Bolton, Chem. Res. Toxicol., 2000, 13, 45–52. F. Zhang, P.W. Fan, X. Liu et al. Chem. Res. Toxicol., 2000, 13, 53–62. I. Terashima, N. Suzuki, S. Shibutani, Cancer Res., 1999, 59, 2091–2095. E. Hillard, A. Vessi`eres, F. Le Bideau et al. ChemMedChem, 2006, 1, 551–559. A. Vessi`eres, S. Top, P. Pigeon et al. J. Med. Chem., 2005, 48, 3937–3940. G. Laus, J. Schuetz, H. Schottenberger et al. Helv. Chim. Acta, 2003, 86, 3274–3280. B. Loev, M. Flores, J. Org. Chem., 1961, 26, 3595. B.E. Maryanoff, S.L. Keeley, F.J. Persico, J. Med. Chem., 1983, 26, 226–229. F.D. Popp, E.B. Moynahan, J. Med. Chem., 1970, 13, 1020–1021.

638

Ferrocenes: Ligands, Materials and Biomolecules

644. 645. 646. 647. 648. 649. 650. 651. 652. 653. 654. 655. 656.

F.D. Popp, S. Roth, J. Kirby, J. Med. Chem., 1963, 6, 83–85. R. Kalish, T.V. Steppe, A. Walser, J. Med. Chem., 1975, 18, 222–224. K. Schlotter, F. Boeckler, H. Hubner, P. Gmeiner, J. Med. Chem., 2005, 48, 3696–3699. E.I. Edwards, R. Epton, G. Marr, J. Organomet. Chem., 1975, 85, C23–C25. E.I. Edwards, R. Epton, G. Marr, J. Organomet. Chem., 1976, 122, C49–C53. E.I. Edwards, R. Epton, G. Marr, J. Organomet. Chem., 1976, 107, 351–357. E.I. Edwards, R. Epton, G. Marr, J. Organomet. Chem., 1979, 168, 259–272. E. Breukink, B. de Kruijff, Nature Rev. Drug Discov., 2006. A.M. Thayer Chevn. Eng. News 2005, 83, 69–82. C. Biot, K. Chibale, Infect. Disord. Drug Targets, 2006, 6, 173–204. B. Mordmuller, P.G. Kremsner, Curr. Mol. Med., 2006, 6, 247–251. R.A. Sanchez-Delgado, A. Anzellotti, Mini-Rev. Med. Chem., 2004, 4, 23–30. J. Wiesner, R. Ortmann, H. Jomaa, M. Schlitzer, Angew. Chem. Int. Ed., 2003, 42, 5274–5293. C. Biot, G. Glorian, L.A. Maciejewski et al. J. Med. Chem., 1997, 40, 3715–3718. O. Domarle, G. Blampain, H. Agnaniet et al. Antimicrob. Agents Chemother., 1998, 42, 540–544. C. Biot, W. Daher, N. Chavain et al. J. Med. Chem., 2006, 49, 2845–2849. C. Biot, W. Daher, C.M. Ndiaye et al. J. Med. Chem., 2006, 49, 4707–4714. C. Biot, L. Delhaes, C.M. N’Diaye et al. Bioorg. Med. Chem., 1999, 7, 2843–2847. P. Beagley, M.A.L. Blackie, K. Chibale et al. Dalton Trans., 2003, 3046–3051. C. Biot, L. Delhaes, H. Abessolo et al. J. Organomet. Chem., 1999, 589, 59–65. K. Chibale, J.R. Moss, M. Blackie et al. Tetrahedron Lett., 2000, 41, 6231–6235. C. Biot, J. Dessolin, I. Ricard, D. Dive, J. Organomet. Chem., 2004, 689, 4678–4682. L. Delhaes, C. Biot, L. Berry et al. Chembiochem, 2002, 3, 418–423. C. Biot, D. Taramelli, I. Forfar-Bares et al. Mol. Pharm., 2005, 2, 185–193. B. Pradines, T. Fusai, W. Daries et al. J. Antimicrob. Chemother., 2001, 48, 179–184. C. Atteke, J.M. Ndong, A. Aubouy et al. J. Antimicrob. Chemother., 2003, 51, 1021–1024. A. Kreidenweiss, P.G. Kremsner, K. Dietz, B. Mordmuller, Am. J. Trop. Med. Hyg., 2006, 75, 1178–1181. B. Pradines, A. Tall, C. Rogier et al. Trop. Med. Int. Health, 2002, 7, 265–270. P. Chim, P. Lim, R. Sem et al. Ann. Trop. Med. Parasitol., 2004, 98, 419–424. C. Biot, Curr. Med. Chem. Anti-Infective Agents, 2004, 3, 135–147. L. Delhaes, C. Biot, L. Berry et al. Bioorg. Med. Chem., 2000, 8, 2739–2745. A. Baramee, A. Coppin, M. Mortuaire et al. Bioorg. Med. Chem., 2006, 14, 1294–1302. J. Howarth, K. Hanlon, Tetrahedron Lett., 2001, 42, 751–754. X. Wu, E.R.T. Tiekink, I. Kostetski et al. Eur. J. Pharm. Sci., 2006, 27, 175–187. X. Wu, P. Wilairat, M.L. Go, Bioorg. Med. Chem. Lett., 2002, 12, 2299–2302. C.A. Molyneaux, M. Krugliak, H. Ginsburg, K. Chibale, Biochem. Pharmacol., 2005, 71, 61–68. C. Biot, L. Delhaes, L.A. Maciejewski et al. Eur. J. Med. Chem., 2000, 35, 707–714. D. Razaﬁmahefa, L. Pelinski, M.T. Martin et al. Bioorg. Med. Chem. Lett., 2005, 15, 1239–1241. J. Howarth, K. Hanlon, Bioorg. Med. Chem. Lett., 2003, 13, 2017–2020. C. Biot, N. Francois, L. Maciejewski et al. Bioorg. Med. Chem. Lett., 2000, 10, 839–841. J. Fang, Z. Jin, Z. Li, W. Liu, J. Organomet. Chem., 2003, 674, 1–9. J.X. Fang, Z. Jin, Y. Hu et al. Appl. Organomet. Chem., 2006, 20, 813–818. Z. Jin, Y. Hu, A.H. Huo et al. J. Organomet. Chem., 2006, 691, 2340–2345. Z. Jin, A.H. Huo, T. Liu et al. J. Organomet. Chem., 2005, 690, 1226–1232.

657. 658. 659. 660. 661. 662. 663. 664. 665. 666. 667. 668. 669. 670. 671. 672. 673. 674. 675. 676. 677. 678. 679. 680. 681. 682. 683. 684. 685. 686. 687.

The Bioorganometallic Chemistry of Ferrocene 688. 689. 690. 691. 692. 693.

J.B. Liu, T. Liu, H. Dai et al. Appl. Organomet. Chem., 2006, 20, 610–614. R.Q. Huang, Q.M. Wang, J. Organomet. Chem., 2001, 637, 94–98. M. Beckmann, K.J. Haack, Chem. unserer Zeit , 2003, 37, 88–97. L. Chen, Q.M. Wang, R.Q. Huang et al. Appl. Organomet. Chem., 2005, 19, 45–48. Q. Wang, H. Li, Y. Li, R. Huang, J. Agric. Food Chem., 2004, 52, 1918–1922. H.K. Sun, Q.M. Wang, R.Q. Huang et al. J. Organomet. Chem., 2002, 655, 182–185.

639

Index

References to diagrams and other images are given in italic type. Reference to tables are given in bold type. adsorption control, on AFM probe tip 431 alamethicin (ALM) 519 alcohols chiral, synthesis 136–137 as donor ligand 24 as ligands 11–13 aldehyde, coordination properties 190 allenylamine, cyclisation, and dppf 89 allylsilanes 73 amination, and dppf 82–87, 88 amines as donor compound 8–9 synthesis 6–11, 92 histamine 552 PAMAM 558–560 pyrrole-imidazole polyamines 533–534 Ugi’s amine 237–240, 238 in synthesis of 1,3-disubstituted ligands 260 see also amino acids Ferrocenes: Ligands, Materials and Biomolecules  2008 John Wiley & Sons, Ltd

amino acids C-terminus labelling 517–519 conjugates, as ligands 531 ferrocene conjugates 524–534 chirality 534–538 Faa 527, 528 Fca 526–527 Fcc 525–526, 528–529 ferrocene derivatives 507–524 ferrocenecarboxylic acid 512–513 ferrocene-based 500–507 N-terminus labelling 519–522 phosphino- 25 sensor compounds 308 Sonogashira coupling 566 aminoferrocene 6–11 ammonium, sensing 292 Angiotensin II 504 animal studies iron-related disorders 604–605 toxicity 602–603

Edited by Petr Stepnicka

642

Index

anions complexed, and ferrocene sensor oxidation potential 285–286 sensor compounds 298, 440 applications 301 dendrimers 311 receptors 298–303 thin ﬁlms and monolayers 311–313 anti-cancer drugs 22, 23, 24, 400, 601, 605–608 anti-inﬂammatory drugs 610 antibiotics 24, 91, 610–618 antibodies and GOD conjugates, effect on electron transfer 545–546 labelling 551 structural determination 538–539, 539 applications amino acid and peptide conjugates 538–539 anion sensor compounds 301 cyclometallated dimethylaminomethylferrocenes 22 Fem derivatives 510 ferrocene ceramics in atomic force microscopy 431–432 nanofabrication 432 ferrocenyl selenoether ligands 19–20 Josiphos ligands 210, 211–214, 211 liquid crystals, redox-active 453 Mandyphos/Ferriphos ligands 224 monosubstituted phosphor-donor ferrocenes 14 nanostructures 387 polymers 414 side-chain 397

protein conjugates 551 self-assembling monolayers 386 Tanaphios ligands 222–223 thin ﬁlms 386 see also catalysis aptamers 583 aqueous environments 301 aryl group-containing compounds selenides, synthesis 95 sulﬁdes, synthesis 95 synthesis using dppf 69–70, 70, 73–74, 87 triﬂates 89 see also individual aryl groups; Schiff bases arylation, dppf complex-catalysed 73–74, 87 Asperillus niger 539–540 asymmetric compounds see chiral compounds atomic force microscopy (AFM), tip 430–431, 431 atovaquone 616 avidin, sensor compounds 558–559 azomethines see Schiff bases barbiturates, sensor compounds 304–305 bases, chiral 256–257 Bayliss-Hillman reaction 14 benzamides, synthesis 96 biaryls, synthesis using dppf 69–70, 70 biomolecules anionic 298 ferrocene sensors 287–288 neutral, sensor compounds 304–309 see also amino acids; DNA; enzymes; peptides; PNA biotin, sensor compounds 557 BIPHEP ligands, analogues 227 bite angle 118, 119–121, 120, 121, 123

Index

BoPhoz ligands 217–218 P-chiral derivatives 229 Boron Neutron Capture Therapy (BNCT) 606 BPPFA (1,2-disubstituted phosphine ligand) 207–209 BPPFOH (1,2-disubstituted phosphine ligand) 207–209 Bradykinin 504 C-terminus labelling, amino acids 517–519 cadmium in crystalline coordination networks 490 as metal centre in monosubstituted phosphor-donor compounds 17 calcium, ﬂuorescence detection with ferrocene sensors 297 cancer, drugs 22, 23, 24, 400, 601, 605–608 capillary gel electrophoresis, in gene sequencing 574 capture probe 567 carbapenem antibiotics 91 carbohydrates 587 ferrocene 1,1 -dicarboxylic acid derivatives 589 monosubstituted ferrocene derivatives 587–589 other derivatives 589–592 sensor compounds 308 carbon as donor in unsymmetric ligands, with phosphorus 180–184 fullerenes 455–462 nanotubes 387, 432 as spacer in monosubstituted ferrocene donor ligands 20–23 carbon monoxide, sensor compounds 441

643

carbonyl metal immunoassay (CMIA) 593 carbonylation, catalysis 96–100 carboxylates, sensor compounds 301 carboxylic acid, derivatives, indole 96–97 catalysis 243–251 arylation 126–127, 128 asymmetric ring opening, dppf 106, 107 Baylis-Hillman reaction 14 C-B coupling, dppf 82 C-C coupling chiral 137 and dppf copper-catalysed 78 Heck reaction 72–73 palladium-catalysed 73–77 Suzuki reaction 69–71 rhodium-catalysed 77–78, 79, 80 and dppf analogues 122–125, 137 and oxazolines 242 zinc compound-catalysed 78 carbonylation 96–100 chiral ferrocene oxazolines 243–251 cooligomerisation and copolymerisation 129–130 cyanation 100–101 with cyclopalladated ferrocenes 23 with dppf 80–82 germylation 80, 81 glucose oxidase mediation 421 hydrogenation 101–103, 127–129, 135–137, 136 hydrosilylation, and oxazolines 242, 252 indolisation 127 isomerisation 23, 104–105 terminal propargylic alcohols 104

644

Index

catalysis (continued ) Josiphos ligands 212–214, 213 oligomerisation and polymerisation 105–106, 127–129 oxidation 101, see also redox chemistry palladium-catalysed 15 phosphorus-donor ferrocenes 13–14 reduction, and oxazolines 424 substitution reactions 101, 103 Suzuki reaction 23, 25–26, 193–195 cations ammonium 292 complexed, and ferrocene sensor oxidation potential 285 sensor compounds receptors 290–298 thin ﬁlms and monolayers 313 CD spectroscopy 529, 536–537 cellular uptake assays 517 central nervous system, drugs 609–610 ceramics 424–432 application in atomic force microscopy 431–432 chalcogen-donor compounds 17–20 with phosphorus 190–198 see also oxygen-donor compounds; selenium, donor compounds CHARMM modelling software 531 chelates chiral disubstituted phosphor-donor 205–207 with planar, axial and central carbon chirality 226–228 with planar, central carbon and central phosphorus chirality 228–230

with planar and central carbon chirality 207–226, 208 containing dioether ligand 155 phosphines, see also dppf symmetrical, containing cyclic dithioether ligand 158 see also monosubstituted ferrocenes chemical shifts, ferrocenylphosphines 122, 123 chiral compounds amino acid conjugates 534–538 bases 256–257 chelates disubstituted diphosphines 205–207 axial-chiral 231 planar-chiral 207–226, 208, 226–228, 228–230, 231 central-chiral carbon 207–226, 208, 228–230, 231 phosphorus 226–228, 228–230 polydentate 266–268 dendrimers 436 dppf analogues, symmetrical 130 dppf complexes 75 and electro-optical response 375 ferrocenes, disubstituted, synthesis 178–180 ferrocenyl sulfoxides 18 ferrocenyldialkylphosphines 14 ligands, with bridged cyclopentadienyl rings 261 liquid crystals 448–453 octupole-symmetric 378 synthesis oxazolines 240–253 via Ugi’s amine and related groups 237–240

Index

chloroquine 611 analogues 613 cholesterol 595 circular dichroism (CD) spectroscopy 536–537 CNS, drugs affecting 609–610 CNT see nanotubes collagen, mimic compounds 538–539 colorimetry 303 competitive immunoassays 597–598 coordination chemistry aldehydes 190 in anti-malarial drugs 612 bidentate ligands, dithiolato framework 163–169 crystals 466, 489–492 dppf 34–35, 36–60, 37, 61–62 changes 60, 64 derivatives 62–69 dppf analogues, planar chiral 133–134 monosubstituted compounds chalcogen-donor 18–19 phosphorus-donor 14–15 unsymmetric achiral diphosphines 119–122 copper, as centre in sulphur-donor symmetric ligands 157 cross-coupling see catalysis crown ethers complexes, as cation sensor 290–294 as electrostatic sensor compounds 281–282 cryptands 152–153 complexes, as cation sensor compounds 294 as ion-pair sensor compounds 310 crystal engineering, overview 465–466 crystals hydrogen-bonded formed from coordination compounds 489–492

645

formed from ferrocenes 477–484 templated 474–477, 474 polymorphism 466, 467, 470–473 cyclometallated compounds (dimethylaminomethyl)ferrocene 22–23 palladium 12–13 cyclopentadienyl rings bridged 261–266, 262 conjugation 11 Cytochrome P450 547, 548 D-amino acid oxidase (DAAO) 545–547 data storage 387 dendrimers classiﬁcation 433, 434, 435, 438 and Josiphos ligands 215–216 liquid-crystalline 455 poly(amidoamine) (PAMAM) 558–560 poly(propyleneimine) (PPI) 560, 561 as sensor compounds 311 derivatives, oxygen-substituted 12–13 diamino framework ligands 144 Diels-Alder reaction 252 dienes, in allylic amine synthesis 92 diiron complexes, as acceptor groups 358 dinucleosides, detector compounds 306 dipeptide ester derivatives 529 diphenyl disulﬁde, as phenylchalcogenolate source in aryl sulphide and selenide synthesis 95 diphosphines, in ruthenium-catalysed hydrogenation reactions 102 DNA ferrocene conjugates 562–565 ferrocene labelling 565–567

646

Index

DNA (continued ) sensor compounds 313, 567–583 for pathogenic DNA 573–574 with threading intercalator 579–583 donor compounds see ligands, donor compounds dppf advantages 34 analogues chiral for C-C and C-P coupling 137 for hydrogenation 135–137, 136 for oligomerisation and polymerisation 127–129 symmetrical for arylation catalysis 126–127, 128 for C-C coupling catalysis 122–125 for C-heteroatom coupling catalysis 125–126 for indolisation catalysis 127 compounds bismuth 58 chromium 57 cobalt 42, 65 coordination mode changes 60 copper 54, 57, 59, 67 gallium 57 gold 56, 57, 59, 66 iridium 43 iron 37 lead 58 manganese 36, 57 molybdenum 36, 57 nickel 44–45 osmium 41–42, 60 palladium 45–50, 57–58, 60, 65, 66 platinum 51–53, 58–59

rhenium 36–37 rhodium 42–43, 60, 66 ruthenium 37–41, 57, 59, 60, 65 silver 54–56, 60, 65, 67 technetium 36 tin 57, 58 tungsten 36, 57, 59 bi- and polymetallic 57–59 in crystalline coordination networks 489 derivatives 62–69 chalcogen 68 oxygen 65 selenium 67, 68 sulphur 66 preparation 34–35 reactivity 37, 61–62 drugs anti-cancer 22, 23, 24, 400, 601, 605–608 anti-inﬂammatory 610 anti-malarial 611–612 antibiotics 24 iron disorders 605 labelling 596 dyads 369–370 electro-optical materials, overview 319–321; see also optical properties; second harmonic generation; third harmonic generation electron transfer DNA assays 570 sensor compounds 282–284 electrostatic interactions ferrocene sensor compounds 282–290, 290 ELISA (immunoassays involving enzymes as reporter groups) 600–601 enzymes alkaline phosphatase (ALP) 601

Index

D-amino acid oxidase (DAAO) 545–547 glucose oxidase (GOD) 539–545 catalytic mediation 421 ferrocene derivatives 542 horseradish peroxidase (HRP) 548–549 esterone 593 estradiol 593, 594 ethers 12 diether ligands 155 ferrocenyl selenyl ligands 19–20 synthesis 103

FAD cofactors 540 Ferriphos ligands 223–225 ferrocenecarboxylic acid, amino acid derivatives 512–513 ferrocenophanes 261–266, 403 ferrocenylalanine 500–506 ferrocenylmethyl (Fem) 509–512, 510 ferrocenylphosphines 1,1 -disubstituted non-chiral, symmetrical, synthesis 118–119 1,1 -disubstituted non-chiral asymmetrical, synthesis 119 coordination chemistry 119–122 ferrocifens, anticancer activity 607–608 ferroquine 612–615, 614 ﬂavin adenine dinucleotide (FAD) 540 ﬂuconazole 618 ﬂuorescence spectroscopy 296–298, 303 and DNA sensing 568 ﬂuorides, sensor compounds 302 four colour DNA sequencing 574–575 frequency doubling and tripling, electro-optical 320–321 fullerenes 455–462 optical properties 374, 375 fungicides 618

647

gel permeation chromatography 395 gene detection 574–583 see also DNA; oligonucleotides germylation 80, 81 glucose, sensor compounds 420–422, 423, 425, 439, 515, 540–543, 541, 560 glucose oxidase (GOD) and avidin-biotin complexes 556 catalytic mediation 421 ferrocene conjugates 539–545 ferrocene derivatives 542 and gene sensor sensitivity 581 gold as metal centre in dppf compounds 56, 57, 59, 66 in monosubstituted ferrocenes 20 in nanoparticles 314–315 guanine, oxidisation, and DNA sensing applications 568

haemoglobin 552 halogenated derivatives 4–6 Hartwig-Buchwald coupling, dppf analogues 125–126 Hdpf 191–195 Heck reaction 72–73 hemilability, phosphane ether ligands and 153 herbicides 619 Herrick conformation 528, 531, 534–535 heterocyclic compounds carbenes 9–10 catalysis of formation 76–79, 80 non-linear optical properties 367–370, 368, 369, 370 precursors, synthesis 76 synthesis, and dppf 78 histamine 552 homogeneous immunoassays 598 hormones, thyroxin 598

648

Index

horseradish peroxidase (HRP) 548–549 HRS 320 hydrazines, substituted, in amination reactions 85 hydroamination 87–88 hydrogenation catalysis 101–103, 127–129, 135–137, 136 Josiphos ligands 211–212 ketones and quinolenes 242 using BoPhoz ligands 218 using functionalised Josiphos ligands 216 using Tanaphios ligands 222 using Walphos ligands 219 hydrosilyation catalysis 242, 252 in polymer synthesis 397–398 hydroxyferrocene, synthesis 11–12 hyper-Rayleigh scattering (HRS) 320 imidates 76 imino groups, in electro-optical compounds 354–355 immunoglobin G (IgG) 551 insecticides 618–619 iodine, as additive in dppf analogue hydrogenation reactions 135 ions thermodynamic effects in ferrocene sensor compounds 285–286 see also anions; cations iridium, as metal centre with Josiphos ligands 212 iron complexes as acceptor groups 358 optical properties 353, 359 as metal centre in dppf compounds 37 iron-related disorders 604–605 isocyanoferrocene 8–9

isomers NLO systems, of monosubstituted ferrocenes 360–361 see also stereochemistry Josiphos ligands 209–214 immobilised and analogues 214–217 P-chiral derivatives 229 ketones arylation 126–127, 128 hydrogenation 242 lab-on-chip (LOC), gene sensing 578, 582 large molecules ferrocene sensors 287–288 as hydrogen bonding units in crystals 475 peptides 515 synthesis using ROMP 407 Lawesson’s reagent 17 Leishmanaisis 617 ligands amino acid conjugates as 531 bidentate, in crystal networks 491 with carbon spacer 20–23 disubstituted, 1,3,- 257–260 donor compounds carbon, with phosphorus 180–184 chalcogen 26 nitrogen 239–253 in chelates 144–151 disubstituted 144–151 in monodentate complexes 20–23 oxygen 24 in bidentate chelates 151–154 monosubstituted compounds 11–13, 24 phosphorus 25–26, see also dppf

Index

sulphur in bidentate chelates 154–160 with phosphorus 190–198 in hydrogen-bonded crystals 489–492 phosphine-ether 156 symmetric, dppf see dppf lipids 595–596 liquid crystalline compounds 447–448 dendrimers 437, 455 fullerene dyads 455–462 liquid-crystalline polymers 397, 399 lithiation in preparing sulphur substituted ferrocenes 18 in synthesis of planar chiral ferrocenes 264 lithium, in monosubstituted ferrocenes 5, 19 magnetic materials 424–432 malaria, drugs 611–612 malonates, arylation 126–127 Mandyphos ligands 223–225, 230 manganese 57 as metal centre in monosubstituted phosphor-donor compounds 17 Markownikoff product 86 meﬂoquin 616 melting temperature analysis 395 mercury, as centre in sulphur-donor symmetric ligands 157 metabolisation, ferrocene 602 metal centres dppf analogues, and bite angle 119–122 see also individual metals metallation 1,3-disubstituted ferrocenes 258 in ferrocene synthesis 4 lithiation, in ferrocene synthesis 4–5

649

mercuration, in ferrocene synthesis 5–6 monosubstituted ferrocenes, chalcogen-donor compounds 18–20 of Ugi’s amine 237–240 metalloimmunoassay (MIA) 593, 598 microcystins 523–524 microspheres, polyferrocenylsilane 429 molecular switches 309–310 monolayers 311–313 applications 386 cystamine derivatives 538 of oligodeoxynucleotides 570 monosubstituted ferrocenes synthesis 4–6 chalcogen-donor 17–20 nitrogen-donor 6–11 with carbon spacer 20–23 oxygen-donor 11–13 with carbon spacer 24 phosphorus-donor 13–17 with carbon spacer 25–26 mutations, sensing 569–570 N-terminus labelling, amino acids 519–522 Naﬁon 600 nanoparticles, and sensor compounds 314–315 nanostructures nanotubes 387, 432 and sensor compounds 311 dendrimers 311 nanoparticles 314–315 thin ﬁlms and monolayers 311–313 neutral compounds, sensor compounds 286–287, 304–309 nitrogen donor compounds 6–11 donor ligands 239–253 bidentate 144–151 as cation sensor 295 with phosphorus 184–190

650

Index

norbornene 15 nucleic acids ferrocene derivatives 562–565 labelling 567–583 see also DNA, PNA ODN 569 oleﬁns hydroamination, and dppf 86 hydrogenation 135 using ligands displaying axial, planar and central chirality 231 polymerisation 147 oligomerisation, catalysis 9, 105–106 oligonucleotides 569 aptamers 583 in DNA sensing 576–578 ferrocene derivatives 565–567 in monolayers, as sensor compounds 312–313 see also DNA optical properties 320–321 azulene complexes 367 compounds with cyano units 345, 346 diiron complex-based compounds 359 effect of structural parameters 385 ﬂuorene-based compounds 364 heterocyclic compounds 369, 370 compounds with imino groups 355–357 indane-1,3-dione 365 isomeric NLO systems 362–363 nitro-containing complexes 339–341 polyene-bridged compounds 348–352 polymers 411–414 pyridine-containing complexes 333–335 sesquilvalenes 328–329 silicon-containing compounds 372, 374

THG-active compounds 380–384 triarylcarbenium compounds 372 tris(pyrazolyl)borato complexes 324–325 other compounds 377 see also second harmonic generation; third harmonic generation ovalbumin 560 oxazoline group 240–253 oxidation catalysis 101 effect of substituents on oxidation potential 284 see also redox chemistry oxygen-donor compounds 151–154, 253–256 monosubstituted 11–13 palladium as catalyst in methoxycarbonylation, ethene 98 in oligonucleotide derivative synthesis 566–567 as metal centre dppf analogues, cooligomerisation and copolymerisation 129–130 in dppf reactions 62–68, 69–77, 87, 94–95 Josiphos ligands 213 in PPF and BPPF chiral ligands 208 in sulphur-dentate symmetric ligands 157, 159 as metal centre in dppf analogues 121–122 PAMAM dendrimers 558 particles nanoparticles 314–315 polyferrocenylsilane 429 size, effect on NLO responses 337

Index

penicillin, ferrocene derivatives 610–611 peptide nucleic acids (PNA) 583–586 peptides ferrocene containing 501 ferrocene derivatives 501–504 ferrocenyl 513–514, 514 labelling 519 Sonogashira coupling 566 phase transitions crystals 467–472, 468 phenylazoferrocene 10–11 phosphorus donor ligands 205–207, 266–267 with carbon 180–184 with chalcogens 190–198 in chiral chelates 226–228, 228–230 monosubstituted ferrocenes 13–17, 25 with nitrogen 180–184 photosynthesis 386–387 phthalides, synthesis 96 phthalocyanine groups, electro-optical properties 367–369 physiological chemistry 602–604 pinch and catch 18–19 PingFer 229–230, 230 platinum as metal centre monosubstituted ferrocenes nitrogen-donor 20 with carbon spacer 22–23 in sulphur-donor symmetric chelates 157 PNA 583–586 polarity, crystals 481 polyferrocenylenes 401–403 polymerisation catalysis 105–106, 127–129 ring-opening 403–407, 406 metathesis (ROMP) 398, 399, 400, 406–407 and Suzuki coupling 105–106

651

polymers analytic techniques 395 classiﬁcation 394 face-to-face and multideck 408–411 optical properties 385 polyamide 400–401 redox-active 414–424 side-chain liquid-crystalline 448 with planar chirality, liquid crystalline 448–453, 449 poly(vinyl)ferrocene 394–395, 396–397 as AFM probe tip 430–431 powder X-ray diffraction 395 PPI dendrimers 558 proline 529–530 proteins ferrocene conjugates antibodies 551–553 avidin and streptavidin 555–556 biotin 556–557 BSA 553–555 ovalbumin 560 redox proteins alcohol dehydrogenase 550–551 CyP450 547 DAAO 545–547 GOD 539–545 HRP 548–549 SBP 549–550 inactivation, papain 560–561 Staphylococcus aureus protein A 561 see also enzymes pyridine complexes 330–336 pyrolysis, of ceramic ferrocenes 424, 428 pyrrole-imidazole polyamines 533–534

652

Index

quinine, ferrocine derivatives 615 quinolines, hydrogenation 242 radioimmunoassay (RIA) 597–601 reaction coupling efﬁciency (RCE) 283–284 redox chemistry anticancer drugs 606 dendrimers 433 effect of substituents 284 ferrocene sensor compounds 282–284, 289–290 liquid crystals 453, 453–454 monosubstituted ferrocenes, nitrogen-donor 21 polymers 414–424 redox-switchable hemilabile ligand (RHL) 12–13, 13 rhodium as metal centre Josiphos ligands 213 TRAP ligands 225 as metal centre in dppf analogues, for hydrogenation 127–129 as metal centre in phosphor-donor compounds 193 ring-opening, using Tanaphios ligands 222 ring-opening metathesis polymerisation (ROMP) 398, 399, 400, 406–407 ring-opening polymerization (ROP) 403–407, anionic 406 ruthenium, as metal centre with Josiphos ligands 212 SAM see monolayers sandwich assay (gene detection) 576–579 Schiff bases 186 cyclopalladated 23 ligands 9, 10

as linking groups in electro-optical compounds 353 second harmonic generation 320–321 compounds allenylidene and related compounds 374–375, 376 azulene complexes 366–367 barbiturate acid derivatives 341 containing Schiff bases 353 ﬂuorene and related compounds 363–365 fullerenes 374, 375 heterocyclic compounds 367–370, 368 µ-carbyne diiron complexes 353 with nitrile functions 343–347 nitro-containing complexes 336–341, 336–337 polyene-bridged compounds 347–352 pyridine complexes 330–336 sesquilvalenes 325–330, 330–332 through-space chromophores 376 triarylcarbenium 371–372, 371 tris(pyrazolyl)borato complexes 321–325, 322–323 other 376–377 monodentate amine ligands 8 stereochemistry and symmetry 357–363, 377–378 selenium donor compounds 172 in monosubstituted ferrocenes 19 see also chalcogen-donor compounds dppf derivatives 35 sensor compounds amino acids 308 anion 440 dendrimers 311

Index

thin ﬁlms and monolayers 311–313 avidin 558–559 barbiturates 304–305 biomolecules 287–288 biotin 557 carbohydrates 308 cation 290–298, 440 complexed 285 crown ethers as 290–294 thin ﬁlms and monolayers 313 dinitrophenol 546 DNA 313, 559, 570–571, 572 electrochemical 579–583 sequence-speciﬁc 569 ﬂuorescent 308 glucose 420–422, 423, 425, 439, 515, 540–543, 541, 560 ion-pair 309–310, 310 large molecules 287–288 neutral compounds 304–309 nucleobases 305 oligonucleotides as 312–313 redox centres 289–290 thermodynamics 282–287 urea 304 SHG see second harmonic generation silicon, ferrocene derivatives 8–9 silver as metal centre in monosubstituted ferrocenes, nitrogen-donor 20 in sulphur-donor symmetric ligands 157 solid-phase synthesis 484–489 peptides (SPPS) 515–516 Sonogashira coupling 517–519, 518 in oligonucleotide derivative synthesis 566 SPPS 515–516 Staphylococcus aureus protein A 561 stereochemistry ferrocene-peptide conjugates 534–538 SHG-active compounds 357–363, 377–378

653

see also chiral compounds steroid hormones 593–595 streptavidin 555–556 substance P 504 sulphur-donor compounds 17–20 in bidentate chelates 154–160 dppf 66 with phosphorus 190–198 Suzuki coupling palladium dppf analogue complexes 124, 125 and polymerisation 105–106 symmetric compounds dendrimers 435 ligands dppf see dppf structure 142–143 dialkoxo framework 153–154 diamido framework 147–151 diamino framework 144 diether framework 151–153 diimino framework 144–145 diphosphoraneimino framework 145 diselenoether framework 169–171 diselenolato framework 171–172 ditellurolato framework 173 dithiocarbamate framework 154–155 dithioether framework 155–160 dithiolato framework 162–169 mixed thioether/disulﬁde framework 160–161 mixed thioether/thiolate framework 161–162 silicon-containing bridges 372–373 octupolar, NLO properties 377–378 symptoms, ferrocene poisoning 603

654

Index

synthesis amino acids, ferrocene analogues 500–501 biaryls, and dppf 70 chiral dppf analogues asymmetrical 130–132 P-chiral 133 symmetrical 130 crystals, solid-state 484–489 disubstituted ferrocenes, 1,2- 18 dppf, derivatives 63–64 dppf complexes 35 ligands 1,3-disubstituted 257–260, 259 monosubstituted, general routes 4–6 Mandyphos/Ferriphos ligands 223–225 monosubstituted ferrocenes, phosphorus-donor 25 polymers face-to-face and multideck 408–411 main-chain ferrocene 401–411 redox-active 414–424 side-chain ferrocene 396–401 polyphosphine ferrocenes 266–267 Tanaphios ligands 220–221, 220–222 TRAP ligands 225 unsymmetrically disubstituted ferrocenes 178–180 Walthos ligands 218 tamoxifen, ferrocene derivatives 607–608 Tanaphios ligands 220–223 target, gene sensor compounds 567

tellurium 173 in monosubstituted ferrocenes, chalcogen-donor 20 thermodynamics, ferrocene sensors 282–287 THG see frequency tripling thin ﬁlms applications 386 of sensor compounds 311, 311–313 thioimidates 76 third harmonic generation 378–386, 379 threading intercalator 579–583 titanium, as metal centre in bidentate ligands 147 toxicity 602, 603 ferrocenes bound to polymers 606 hydroxyferrocene 602 toxins, bacterial 523–524 trans effect, and reductive elimination rates of dppf analogues 126 transition metals, in chalcogen ligands 19–20 TRAP ligands 225, 226 triads 369–370 tris(pyrazolyl)borato complexes 321–325, 322–323 Ugi’s amine 237–240, 238 in synthesis of 1,3 disubstituted ligands 260 urea, sensor compounds 304 voltammetry differential pulse, limit of detection 585–586 in sensor molecule reactions 283 sinusoidal (SV), in gene detection 574

Index

Walphos ligands 218–219 Williams-Watts (WW) relation 413 W¨urtz synthesis 397 zinc in cation sensor compounds 296 in crystalline coordination networks 490

655

in monosubstituted phosphor-donor compounds 17 zirconium, as metal centre in bidentate ligands 147 zwitterions adduct formation in crystals 486–487, 488 sensor compounds 310