ADVANCES IN SULFUR CHEMISTRY
Editor: CHRISTOPHER M. RAYNER School of Chemistry University of Leeds Leeds, England
VOLUME 2 • 2000
(j^ \
jAI PRESS INC. y Stamford, Connecticut
This Page Intentionally Left Blank
Copyright © 2000 byJAI PRESS INC 100 Prospect Street Stamford, Connecticut 06904 All rights reserved No part of this publication may be reproduced, stored on a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, filming, recording, or otherwise without prior permission in writing from the publisher. ISBN: 0-7623-0618-1 Manufactured in the United States of America
This Page Intentionally Left Blank
DEDICATION To Charlotte and Chios
This Page Intentionally Left Blank
CON I tN IS
LIST OF CONTRIBUTORS
ix
PREFACE Christopher M. Rayner
xiii
SYNTHESIS AND CHEMISTRY OF THIOACYLSILANES Bianca Flavia Bonini and Mariafrancesca Fochi
1
IODINE, A VERSATILE REAGENT IN CARBOHYDRATE CHEMISTRY: ACTIVATION OF THIOGLYCOSIDES AND GLYCOSYL SULFOXIDES K. P. Ravindranathan Kartha, Mahmoud Aloui, Peter Cura, Steven j. Marsh, and Robert A. Field
37
RECENT ADVANCES IN THE STEREOSELECTIVE SYNTHESIS OF CHIRAL SULFOXIDES Noureddine Khiar, Inmaculada Fernandez, Ana Alcudia, and Felipe Alcudia
57
CYCLIC SULFOXIDES AS CHIRAL AUXILIARIES IN ASYMMETRIC SYNTHESIS Steven M. Allin and Philip C. Bulman Page
117
RECENT ADVANCES IN THE CHEMISTRY OF a,p-UNSATURATED SULFOXIDES AND SULFONES Ian Forristal and Christopher M. Rayner ASYMMETRIC PUMMERER REARRANGEMENT AND RELATED REACTIONS Masato Matsugi, Norio Shibata, and Yasuyuki Kita SYNTHESES AND REACTIONS OF SULFINIMINES Ping Zhou, Bang-Chi Chen, and Franklin A. Davis
VII
155
215 249
viii CHIRAL SULFOXIMINES FOR DIASTEREOSELECTIVE AND ASYMMETRIC SYNTHESIS Stephen C. Pyne INDEX
CONTENTS
283 367
LIST OF CONTRIBUTORS
Ana Alcudia
Departamento de Quimica Organica y Farmaceutica Universidad de Sevilla Sevilla, Spain
Felipe Alcudia
Departamento de Quimica Organica y Farmaceutica Universidad de Sevilla Sevilla, Spain
Steven ISA. Allin
Department of Chemistry Loughborough University Loughborough, England
Mahmoud Aloui
School of Chemistry and Centre for Biomolecular Sciences University of St Andrews St Andrews, Scotland
Bianca Flavia Bon in i
Dipartimento di Chimica Organica Universita di Bologna Bologna, Italy
Bang-Chi Chen
Discovery Chemistry Bristol-Myers Squibb Pharmaceutical Research Institute Princeton, New Jersey
Peter Cura
School of Chemistry and Centre for Biomolecular Sciences University of St Andrews St Andrews, Scotland
LIST OF CONTRIBUTORS Franklin A. Davis
Department of Chemistry Temple University Philadelphia, Pennsylvania
Inmaculada Fernandez
Departamento de Quimicas Organica y Farmaceutica Universidad de Sevilla Sevilla, Spain
Robert A. Field
School of Chemistry and Centre for Biomolecular Sciences University of St Andrews St Andrews, Scotland
Mariafrancesca Fochi
Dipartimento di Chimica Organica Universita di Bologna Bologna, Italy
Ian Forristal
Department of Chemistry King's College London, England
K.P. Ravindranathan Kartha
School of Chemistry and Centre for Biomolecular Sciences University of St Andrews St Andrews, Scotland
Noureddine Khiar
Institute de Investigaciones Quimicas C.S.I.C. Sevilla, Spain
Yasuyuki Kita
Graduate School of Pharmaceutical Sciences Osaka University Osaka, Japan
Steven j. Marsh
School of Chemistry and Centre for Biomolecular Sciences University of St Andrews St Andrews, Scotland
Masato Matsugi
Graduate School of Pharmaceutical Sciences Osaka University Osaka, Japan
List of Contributors Philip C. Bulman Page
Department of Chemistry Loughborough University Loughborough, England
Stephen G. Pyne
Department of Chemistry University of Wollongong Wollongong, New South Wales, Australia
Christopher SA. Rayner
School of Chemistry University of Leeds Leeds, England
Norio Shibata
Graduate School of Pharmaceutical Sciences Osaka University Osaka, Japan
Ping Zhou
Chemical Sciences Wyeth-Ayerst Research Princeton, New Jersey
This Page Intentionally Left Blank
PREFACE
This second volume of Advances in Sulfur Chemistry is testament to the continuing importance of sulfur chemistry, and the tremendous progress that has been made in recent years. This volume concentrates on sulfur-based synthetic organic chemistry, however its impact in a range of areas is apparent from the content of the chapters, particularly with respect to molecules of biological and medicinal importance. Tremendous progress has been made in the area of stereocontrolled synthesis using organosulfur chemistry, and many of the most important areas of this research are included in this volume. Contributions have been written by a truly international range of authors, who are all experts in their own specialized field. I am very grateful to them all for agreeing to work with me on this project, and making my job as editor such an easy and enjoyable one. The first chapter, by Bianca Bonini and Mariafrancesca Fochi, is a detailed account of the chemistry of thioacylsilanes, ranging from methods for their synthesis, to a variety of synthetic applications of this fascinating functional group. The phenomenal current interest in the construction of complex carbohydrates for biological studies relies in many cases on the use of thioglycosides and the corresponding sulfoxides. The chapter by Robert Field, Ravindranathan Kartha, Mahmoud Aloui, Peter Cura, and Steven Marsh, describes recent advances in this area, particularly using iodine as a novel activator for glycosyl coupling reactions. Sulfoxides are now well established as probably the most important sulfur-containing functional group for stereocontrolled synthesis. The contribuxiii
PREFACE
xiv
tion from Noureddine Khiar, Imuaculada Fernandez, Ana Alcudia, and Felipe Alcudia describes recent advances in the synthesis of chiral sulfoxides, including sections on the two most widely used procedures, asymmetric sulfoxidation, and nucleophilic substitution of chiral sulfur derivatives. The chapter by Steven Allin and Philip Page goes on to describe the use of cyclic sulfoxides as stereocontrolling elements in asymmetric synthesis, whereas the subsequent chapter from Ian Forristal and myself concentrates on the chemistry of unsaturated sulfoxides and sulfones, again with a particular emphasis on stereocontrol. Chiral sulfoxides are also substrates for the asymmetric Punmierer rearrangements, which are described in the extensive chapter by Masato Matsugi, Norio Shibata, and Yasuyuki Kita. More recently, chiral nitrogen-containing sulfur compounds have found application in synthesis. The contribution from Ping Zhou, Bang-Chi Chen, and Franklin Davis describes the synthesis of enantiopure sulfinimines (TV-sulfinylimines) and applications in asynmietric synthesis. These versatile intermediates undergo a wide range of transformations, many of which allow a high degree of stereocontrol originating from the sulfinyl group. In the final contribution, Stephen Pyne gives an extensive account of the use of chiral sulfoximines for diastereoselective and asymmetric synthesis. Again, high levels of stereocontrol are typical for these versatile and interesting intermediates. Finally, I wish to thank my wife, Charlotte (a fellow sulfur chemist) for her help with proofreading many of the contributions. It is to her, and our beautiful new daughter Chloe that this volume is dedicated. Christopher M. Rayner Series Editor
SYNTHESIS AND CHEMISTRY OF THIOACYLSILANES
Bianca Flavia Bonini and Mariafrancesca Fochi
I. Introduction II. Synthesis of Thioacylsilanes A. Reaction of Acylsilanes with H2S/HCI B. Reaction of Acylsilanes with (Me3Si)2S in the Presence of Cobalt II Chloride C. Use of Lawesson's Reagent III. Spectroscopy IV. Thermal Stabihty V. Reactivity of Aromatic and Non-enethiolizable Thioacylsilanes A. Reaction with Organometallic Reagents B. Cycloaddition with 1,3-Dipoles C. [4+2] Cycloaddition with Dienes D. [4+2] Cycloaddition with Heterodienes E. Photoinduced Cycloadditions with Olefins F. Thioacylsilane 5-Oxides (Silyl Sulfines) VI. Thioacylsilanes as Synthetic Equivalents of Unstable Thioaldehydes VII. Thioacylsilanes Chiral at Silicon or at Carbon VIII. Enethiolizable Thioacylsilanes
Advances in Sulfur Chemistry Volume 2, pages 1-35. Copyright © 2000 by JAI Press Inc. AH rights of reproduction in any form reserved. ISBN: 0-7623-0618-1 1
2 3 3 4 4 6 7 8 8 . 9 10 12 12 13 14 19 20
2
BIANCA FLAVIA BONINI and MARIAFRANCESCA FOCHI
A. Synthesis of Z-a-SilylEnethiols B. Synthesis of Cyclic a-Silyl Vinyl Sulfides C. Synthesis of Open-Chain a-Silyl Vinyl Sulfides D. Synthesis of Unsaturated Silylated Thiolactones IX. Reactivity of a-Silyl Vinyl Sulfides A. Reaction with Electrophiles B. Reaction with Nucleophiles C. Reaction with Fluoride Ions X. The Use of Thioacysilanes as Spin Trapping Agents Acknowledgments References
20 21 22 24 25 25 27 29 31 32 33
I. INTRODUCTION The chemistry of thioacylsilanes started in 1981^ and has been reviewed twice since.^^'^ The early work on thioacylsilanes mostly concerned the remarkably high reactivity of the carbon-sulfur double bond of aromatic and nonenethiolizable derivatives toward nucleophilic additions, cycloaddition reactions, and oxidation to silylated sulfines. The various reaction modes, which are similar to those of thioketones,"^ allowed the synthesis of a large variety of compounds containing the Si-C-S unit. However, the presence of the silyl group bonded to the carbon-sulfur double bond attributes to thioacylsilanes features different from those of other thiocarbonyl derivatives. In fact, thioacylsilanes and their 5-oxides can be considered synthetic equivalents of unstable thioaldehydes and thioaldehyde 5-oxides, respectively, reflecting the wellknown ease of replacing the silyl group with a proton at the stage of the reaction products (silylated adducts and silyl sulfines). More recently, the synthesis and chemistry of enethiolizable thioacylsilanes was investigated and it was found that they undergo a complete and stereoselective enethiolization to Z-a-silyl enethiols,^*' with respect to alkyl thiones which are generally obtained as mixtures of thione and enethiol.'^ This behavior is consistent with the recent finding in acylsilane chemistry that a-silyl-substitution markedly stabilizes enols relative to their keto isomers in comparison with silicon-free analogues.^^'^ Several types of Z-a-silyl enethiols have been synthesized and used for further synthetic transformations, such as the stereoselective synthesis of Z-a-silyl vinyl sulfides and unsaturated silyl thiolactones. In particular, Z-a-silyl vinyl sulfides proved useful substrates for the stereoselective synthesis of vinyl sulfides, vinyl silanes, and thiofunctionalized enones. Finally, an interesting application of thioaryloyltriphenylsilanes is their use as spin trapping agents resulting from their ability to readily undergo addition with a wide variety of radicals including electrophilic oxygen and sulfur-centered species, nucleophilic alky Is, silicon.
Synthesis and Chemistry of Thioacylsilanes
3
germanium, tin, lead, and phosphorus as well as transition-metal-centered radicals, to give unusually persistent paramagnetic adducts.
IL SYNTHESIS OF THIOACYLSILANES In the same way, thioacylsilanes 1, like other thiocarbonyl derivatives, are prepared by thionation of the corresponding acylsilanes 2 that are in turn obtained in many ways^^'*' among which the most used are the dithiane methodology investigated by Brook^ and Corey^ and the procedure based on the nucleophilic silylation of an acylchloride.^"^^ Among the various thionation methods that have been described in the literature^^ for the transformation of carbonyl into thiocarbonyl function, we describe here three different procedures that have been applied to the synthesis of thioacylsilanes. A. Reaction of Acylsilanes with H2S/HCI
The acid-catalyzed reaction of acylsilanes 2 with hydrogen sulfide in ether at low temperatures allowed the synthesis of thermally unstable thioacylsilanes 1 (Scheme 1) in good yields (Table 1). Long reaction times lead to the disappearance of the blue color of thione 1, related to the addition of a second molecule of hydrogen sulfide to give colorless gemdithiols 3. Aromatic derivatives (R = Ar) are less prone to further addition of hydrogen sulfide, whereas with aliphatic derivatives (R = alkyl), g^m-dithiols are generally the final reaction products and can be isolated and fully characterized.^^ g^m-Dithiols 3 can be converted into thioacylsilanes 1 by neutralization of the thionation solution with solid sodium hydrogen carbonate. With this procedure, enolizable acylsilanes 2 (R = R^CH2) are stereoselectively transformed into Z-asilyl enethiols 4 (vide infra). Although in principle cycloalkyl silylthiones bearing a hydrogen atom at the a-C might give enethiolization, compounds lh-1 were prepared in very good yields (Table 1) and were found to be stable enough to be characterized.^ In particular, the
^ R ^ i 2
H2S/HCI *
*
f
H2S _
R ^ i
NaHCOa 1
R = R^CH2 SH 4
Si
Scheme 1,
HS
SH
^
^i 3
BIANCA FLAVIA BONINI and MARIAFRANCESCA FOCHI Table 1. Synthesis of Thioacylsilanes 1 with H2S/HCI 1
R
Si
a
Ph Ph
SiPh3
85
13,14
b
SiMe3
58
13,14 13,14
Yield (%)
Ref.
c
P-CH3-C6H4
SiMe3
70
d
m-CI-C6H4
SiMe3
55
14
e
3,5-f-Bu-C6H3
SiPh3
60
15
f g h i
CH3
SiMe3
a
16
r-Bu
SiMe3
90
16
C-C3H5
SiMe2Ph
85
9
C.C3H5
SiMe3
75
9
SiMe2Ph
95
9
k 1
C-C4H7 c-CeHii
SiMe2Ph
82
9
Myrtanyl
SiMe2Ph
77
9
m
Ph
SiMePha-Np^
95
17
J
Notes: ^Trapped in situ. ''a-Np, a-naphthyl.
cyclopropyl derivatives Ih and li can be stored at -20 °C for several months without noticeable decomposition. B. Reaction of Acylsilanes with (Me3Si)2S in the Presence of Cobalt II Chloride
Ricci and co-workers showed that it is possible to transform a wide range of acylsilanes 2 into the corresponding thioacylsilanes 1 by reaction of bis(trimethylsilyl)sulfide in the presence of CoCl2-6H20 under very mild conditions and in good yields (Table 2)^^ (Scheme 2). C. Use of Lawesson's Reagent
Lawesson's reagent [2,4-bis(4-methoxyphenyl)-2,4-dithioxo-P^,P'^-1,3,2,4dithiaphosphetane] is a versatile and efficient reagent for the conversion of carbonyl to thiocarbonyl derivatives in boiling toluene.^^ While the high temperature required prevents the use of this method for thermally unstable thioacylsilanes, the more stable derivatives have been prepared in excellent yields (Table 3, Scheme 3). Worth mentioning is that the thionation of ferrocenoyltrimethylsilane and ferrocenoyl dimethylphenylsilane, in THF could be carried out at room temperature, affording in a few minutes quantitative yields of the corresponding thioacylsilanes It and lu. Enolizable acylsilanes, treated with Lawesson's reagent in boiling toluene, gave enethiols contaminated with disulfides'^ (Scheme 4).
Synthesis and Chemistry of
Thioacylsilanes
ff CoCi26H20^ R-'^SiMes (MeaSOaS ={"^811^163 (Me3Si)2S 2
t R'^^SiMea R i 1
Scheme 2.
Table 2,
Synthesis of Thioacylsilanes 1 with (Me3Si)2S
R
Yield (%)
f
Me
n
Me(CH2)5 Ph
b o
3-MeO-C6H4
P
4-MeO-C6H4
q r
2-thienyl
30 64 92 74 66 58 59
2-furyl
A. ^
LR. ,
^'
X
solvent. A
R
Si
Scheme 3.
Table 3,
Synthesis of Thioacylsilanes 1 with Lawesson's Reagent
R s
2,4,6-CH3-C6H2
Si
Solvent
T
Yield (%)
Ret
SiMe3
Toluene
110 °C
68
14
SiMe3
THF
rt
96
20
SiMe2Ph
THF
rt
98
20
4 ^ t
4 ^ u
Fa
^^-O^Si,MesPh
LR. solvent. A
Scheme 4,
E
r
1
6
BIANCA FLAVIA BONINI and MARIAFRANCESCA FOCHI
III. SPECTROSCOPY The interesting and unusual spectroscopic properties of acylsilanes have received considerable attention.^^'^^ Unfortunately, a similar systematic study for thioacylsilanes has not been done because of their low thermal stability. These compounds are blue materials [^max(^^U) = ^^8 nm for lb, 692 nm for la,^^ and 612 nm for Ig^^], and are generally liquids except thiobenzoyltriphenylsilane la, 2,4,6-trimethyl-thiobenzoyltrimethylsilane Is, and 3,5-dir^rr-butylthiobenzoyltriphenylsilane le which are solids. A ^-^C-NMR and a theoretical investigation on (Me3C)2C=X and (Me3C)(Me3Si)C=X, where X = O, S, have been reported (Table 4),^^ showing that the large downfield shifts of the carbonyl and thiocarbonyl carbons, observed on silylation, correspond to a remarkable increase in negative charge in the same atoms. The replacement of a r^r/-butyl group by SiMe3 going from di-tert-buiyl ketone to rerr-butanoyl trimethylsilane and from di-^^r/-butyl thioketone to r^r^thiobutanoyl trimethylsilane Ig increased A^ .^^^^ and deshielded the quaternary carbon. Furthermore, an inversion in polarization of the C=S bond and a decrease of the HOMO-LUMO energy differences were observed in the rerr-thiobutanoyl trimethylsilane with respect to di-r^rr-butyl thioketone.
Table 4, Compound
^^C Chemical Shifts^ and Total Charges^ of Carbonyl and Thiocarbonyl Compounds 6c=x (pRf^^)
9c/ '^^~
9x/ ^^~
9c/ ^^~
218
+237
-307
+276
249
-45
-299
+258
278
-99
-25
+152
316
-429
+5
+120
^Bu
MeaSi ^Bu f-Bu t-Bu MeaSi
Notes: ^\n CHCI3, from TMS. ^MNDO, fully optimized geometry.
Synthesis and Chemistry of Thioacylsilanes
7
This finding is in agreement with aZ? initio calculations on the silylated thioaldehyde HCSSiMcg.^^'^^
IV. THERMAL STABILITY Arylsilylthiones have limited thermal stability and they are slowly transformed into trimers (1,3,5-trithianes) on standing.^"^ Both the a- (cis,trans) 5 and the P- (cis, cis) 6 isomers were isolated in ca. 90% overall yield, from the decomposition of thiobenzoyltrimethylsilane l b and thiobenzoyltriphenylsilane la, with the P-diastereoisomers being the major products (Scheme 5). An X-ray analysis showed a chairlike conformation for the P-form with the silyl groups all equatorial. The decomposition of thioacetyl trimethylsilane If^^ (Scheme 6) deserves special comment. Open chain dimeric products 7 and 8 were isolated from the decomposition mixture. Their formation is accounted for by assuming enethiolization of the thione If and a subsequent thiophilic and carbophilic addition of the enethiol form to another molecule of the same thioacylsilane. Product 7 was the major product derived from thiophilic addition. Remarkably, it has been reported that methylthioketones such as methyl r^rr-butyl thioketone do not show any tendency to exist in the enethiol form."*^ Dimeric products were also obtained during the thionation of bromoacetyldimethylphenylsilane.^^ The behavior of methylsilylthione (Scheme 6) and the quantitative enethiolization of alkyl silyl thiones (Scheme 1) are ascribed to the presence of the silyl group, which favors enethiolization in comparison to nonsilylated derivatives."*^ A deviating behavior during the thionation was observed for the bicyclic acylsilane 9. The ultimate product turned out to be the ^n(io-tetrahydro-2trimethylsilylcyclopenta[b]thiopyran 10 whose formation was rationalized by a [3,3]-sigmatropic rearrangement of the initially formed thione^ (Scheme 7). Similar retro thio-Claisen rearrangement has been observed during thionation of ^AiJ(?-2-acetylnorborn-5-ene.^^
Ar... _^Sv.^^^ ..Ar
An.. ^ / S , ^ ..Ar
Si-q
S i ^
pSi
Si""'^Ar 5
pSi
Sr
''Ar 6
a-(cis,trans)
p-(c/s,c/s)
Ar = Ph, Si = SiMes; Ar = Ph, Si = SiPha Scheme 5.
BIANCA FLAVIA BONINI and MARIAFRANCESCA FOCHI
X,..-
H3C
SiMea
SH
J.,
H2C^ "SiMea
If
SiMea
SiMeaSH
CHa 7
7:8 = 3:1
8
Scheme 6.
[3.3]
H2S/HCI EtaO -20''C
(XX 10 50%
encfo:exo = 6A Scheme 7.
V. REACTIVITY OF AROMATIC AND NON-ENETHIOLIZABLETHIOACYLSILANES A. Reaction with Organometallic Reagents
It is well documented that thioketones generally undergo thiophilic addition with organolithium reagents,^^ but carbophilic addition has also been observed.^^ Thioacylsilanes react with organolithium reagents only at the sulfur, probably because of the stabilizing effect of the silyl group on the intermediate a-silyl carbanion, affording a-silyl sulfides 11 in good to excellent yields^^'^^ (Scheme 8). a-Silyl sulfides 11 were used for further synthetic applications, for instance by
H+/H2O
R^i 1 R = Ph, t-Bu Ri = Alk, Ar Si = SiMea, SiMe2Ph, SiPha Scheme 8.
SR^ R^SI 11 60-90%
Synthesis and Chemistry of
Thioacylsilanes
SR^
T
SR^
SR^
T
J
1)CsF,CH3CN
R-'^^Si
2)E
R-'^^E
11
"^
FT
27-73%
R = Ph Si = SiMea, SiMePha-Np E = 4-CH3-C6H4-CHO, EtCHO
/ = 1,4-addition
Scheme 9.
performing a fluoro-desilylation reaction in the presence of carbon electrophiles such as aldehydes and cyclohexenone. The yields were moderate to good when anhydrous conditions were used to minimize protiodesilylation^^ (Scheme 9). The reaction of thiocyclopropanoyl dimethylphenylsilane Ih with organolithium reagents takes a deviant course as a result of enethiolization, and the initially formed enethiolate is trapped in the presence of methyl iodide^^ (Scheme 10). B. Cycloaddition with 1,3-Dipoles
Cycloaddition between thiocarbonyl derivatives as hetero dipolarophiles and various 1,3-dipoles provides easy entry to five-membered thiaheterocycles. The reaction of thiobenzoyltrimethylsilane l b and thiobenzoyltriphenylsilane l a with benzonitrile oxides, diphenylnitrilimine, and benzonitrile-4-nitrobenzylide gave regiospecifically 5//-l,4,2-oxathiazoles 12a,b, 1,3,4-thiadiazoline 13, and 4,5-dihydrothiazole 14^^ (Scheme 11). The regiochemistry of this reaction was assigned through the protiodesilylation of the adducts {vide infra). The reaction of thioacylsilanes with diaryldiazomethanes is a route to silylated thiiranes 15, with the subsequent desulfurization of the products 15 with triphenylphosphine affording trisubstituted vinyl silanes 16 in excellent yields^^ (Scheme 12). Oxidation of thiiranes 15 (Ar = Ph, Ar^ = 4-CH3O-C6H4, Si = SiMe3) with MCPBA at -20 °C led to the corresponding thiirane-5-oxide 17, for which an X-ray analysis showed an anti configuration of the silyl group and the oxygen.^^
S®Li®
RLi
^^^^^^
M. '
"
SCH3
-^^ >=<.
SiMezPh
Scheme 10.
,^
r^..
.
IX
'
SiMesPh
10
BIANCA FLAVIA BONINI and MARIAFRANCESCA FOCHI
S
X
A r - / ^ ^ '
N—X
X >^^
—-—-
1a.b Ar X 12a Ph O 12b 4-CI-C6H4 O 13 Ph NPh 14 Ph 4-NO2-C6H4CH
Si Yield (%) SiMea 81 SiPha 95 siMes 65 SiiVlea 20
Scheme 11.
C. [4+2] Cycloaddition with Dienes
[4+2] Cycloaddition of thioacylsilanes 1 with open-chain dienes and cyclopentadiene represents an easy and high-yielding approach to silyldihydrothiopyrans 18 and silylated thiabicyclo[2.2.1]heptenes 199.1632,33 (Scheme 13). The diastereoselectivity of the reaction of thiobenzoyltrimethylsilane l b with cyclopentadiene is temperature dependent^^: at 0 °C only the S-endo trimethylsilyl adduct is formed while at 25 °C an 8:1 mixture of S-endo and 3-exo trimethylsilyl2-thiabicyclo[2.2.1]hept-5-enes (19: R = Ph, Si = SiMe3) is obtained. The regiochemistry of the cycloaddition of thioacylsilanes to dienes was explored in the case of r^rr-butanoyItrimethylsilane Ig-^"*: the reaction with 2-substituted-1,3butadienes was found to proceed with no significant regioselectivity at 80 °C, whereas the reaction with 1-substituted-1,3-butadienes affords single regioisomers as a mixture of diastereoisomers thus suggesting that the regioselectivity is controlled by steric hindrance (Scheme 14).
PPhs Ar -|g Ar^
f
. ^ W , J^O
®'V-V^''—I
1
15 50-100%
Ar=Ph.p-CH3-C6H4 Ar' = Ph, P-CH3-C6H4, P-CH30-C6H4 Si = SiMea, SiPhs Scheme 12.
80-100% I MCPBA,
^S^-/^*^
C^^2Cl2 -20°C Ar-" V ' A r ' 6 17
Synthesis and Chemistry of Thioacylsilanes R' n
I1
R' n
11 D1 D»
Si R
60-96%
,R
76-78%
18
o Si
19
R = Ph, CH3, t-Bu, C-C3H5, C-C4H7, c-CeHu, R^ = H, CH3 Si = SiMes, SiMe2Pii
H
Scheme 13,
t-B\
t-B CeHe, A
MesSl
MesSi'
Ratio
t
R = CH3 R = 0SiMe3
.
80 55
f-BLT^SiMes ig R^ R^ ^
f-Bi
MesSi R^ = CH3, R2 = H R^ = OCH3, R^ = 0SiMe3 Scheme 14,
84% 76%
O.Y. (%) 20 45
85 68
BIANCA FLAVIA BONINI and MARIAFRANCESCA FOCHI
12
D. [4+2] Cycloaddition with Heterodienes
[4+2] Hetero cycloaddition of thiones shows great potentiality for synthetic approaches in heterocyclic chemistry. Among heterodienes, a-nitrosostirene 20, generated in situ via dehydrobromination of a-bromo acetophenone oxime, reacted with thiobenzoyltrimethylsilane lb and with thiobenzoyltriphenylsilane la giving 4ff-l,5,2-oxathiazines 21.^^'^^ Propenoyltrimethylsilane 22 reacted as a heterodiene with the same thioacylsilanes affording 4H' 1,3-oxathiins 23."^^ In both cases excellent yields were obtained (Scheme 15). E. Photoinduced Cycloadditions with Olefins
Photocycloadditions of aromatic^^^"^ and aliphatic^^^"^ thiones with olefins have been extensively reported. Photoinduced reactions of thiobenzoyltriphenylsilane la with electron-poor olefins such as acrylonitrile, methylacrylate, and cis- and rraA25-l,2-dichloroethene gave silyl thietanes 24,25a, and 25b in a regio- and highly stereoselective manner^^ (Scheme 16). The reaction times are considerably shorter than those required for aromatic, aliphatic and a,P-unsaturated thioketones, further demonstrating the high reactivity of the thioacylsilanes with respect to thioketones. Thietane formation, most likely, proceeds via Sj excitation involving a 1,4diradical species 26 as the initial intermediate. Ring closure is governed by steric effects, with the dominant product being the one with the bulky triphenylsilyl group and the adjacent electron-withdrawing group trans to each other. In contrast, the reaction with vinyl ethers gave thietanes without any regio- or stereocontrol. a-Methyl styrene and 2,3-dimethyl-but-2-ene afforded the open
Ph 20
!r°Y-si Ph--" v ^
Si = SiMe3, 91% Si = SIPha, 90%
21
1 _ Ph^^Si 1a,b
MeaSiv^
r°
MesSiv^^O \x^^
Vx--^ 23
22 Scheme 15,
Ph Si = SiMe3, 100% Si = SiPh3, 75%
Synthesis and Chemistry of Thioacylsilanes
13 EWG
hv, -70°C
'^C^l^'^-SiPhs
EWG EWG = CN, COsMe
24 63-65%
V
CI, S Ph^^SiPha 1a
Ph
Ph
-CU
hv, -40°C 25a 70%
c,/^c.
26
'''^o^i:^^^-siPh3
hv, -40X
H 25b 70% Scheme 16.
chain products 27 and 28, whose formation is explained by assuming the intermediacy of a 1,4-diradical 29 which then undergoes a 1,5-hydrogen shift (Scheme 17). F. Thioacylsilane S-Oxides (Silyl Sulfines)
The oxidation of a large variety of thiocarbonyl compounds to the corresponding 5-oxides has been studied by Zwanenburg."^ Like normal thiones, thioacylsilanes Me
Me
Me
Me
hv/25°C
Me^ Me^ M e^^ Me^
1^ Ph SiPha
27 45%
X.
Ph-^ ^SiPha
Ph-^^S-^T PhaSi R^ 29
Me Ph,
-< Ph.
hv/25°C
Y H SiPha 28 85%
Scheme 17.
R^ = R 2 = H , R3=Ph
R^ = R2=R3=Me
14
BIANCA FLAVIA BONINI and MARIAFRANCESCA FOCHI
s/
s/ 1
V 'o
s/ E
Z 30 50-97%
R = Ph. P-CH3-C6H4, m-CI-C6H4, 2,4,6-CH3-C6H2, t-Bu, C-C3H5, C-C4H7, c-CeHn,
Si = SiMes, SiMeaPh, SiPhs Scheme 18.
1 can be converted into silyl sulfines 30^'^^^'^^ (Scheme 18), with the oxidation being generally stereoselective since only the kinetically preferred ^-isomers are obtained. In contrast, substrates with a comparable steric hindrance of the R and the silyl groups, e.g., thiomesitoyltrimethylsilane Is and rerr-thiobutanoyltrimethylsilane Ig, gave a mixture of E- and Z-isomers in ratios of 66:34 and 60:40, respectively.^'* The geometry of the silyl sulfmes 30 was determined by LIS measurements. During chromatographic purification on silica gel, a partial protiodesilylation to the corresponding thioaldehydes 5-oxides was observed {vide infra).
VI. THIOAC VLSI LANES AS SYNTHETIC EQUIVALENTS OF UNSTABLE THIOALDEHYDES Thioacylsilanes are potential synthetic equivalents of unstable thioaldehydes, related to their well-known ease of replacing the silyl group with a proton by means of a fluoride ion at the stage of the reaction products. This synthetic equivalence was explored in a large variety of reaction products of thioacylsilanes. Thus, the a- and (3-trimers of thiobenzoyltrimethylsilane 5 and 6 were converted stereospecifically to the a- and P-trimers of thiobenzaldehyde^"^ by reaction with tetrabutyl ammonium fluoride (TBAF) (Scheme 19). The desilylated products 31 and 32 (Scheme 20) were obtained by the protiodesilylation of a number of thioacylsilane adducts and the corresponding sulfones obtained by oxidation of the cycloadducts with oxone (potassium hydrogen persulfate). Compounds 31 are formally derived from unstable thioaldehydes and the cyclic sulfones 32 from thioaldehyde 5,5-dioxide (sulfenes) (Scheme 20). It should be noted that sulfenes produced by dehydrochlorination
Synthesis and Chemistry of Thioacylsilanes
5 a-(cis,trans)
15 An.. ^^S>^ ,,Ar vr\ pH ^><-^
TBAF ^ THF r.t.
a-form
^ a, •
•X
Ar...^S^^^.,Ar H^ PH
TBAF
hr ''Ar P-form Scheme 19.
of methane sulfonyl chlorides with a base, do not undergo a [4+2] cycloaddition reaction."^^ The protiodesilylation of cycloadducts 12a, 13, and 14 presented some problems, because the initially formed H-substituted heterocycles underwent easy ring fragmentation (products 33 and 34) or aromatization (product 35)^° (Scheme 21). The decomposition products allowed the assignment of the regiochemistry of the cycloaddition. Fluoride-induced desilylation of silylated thiiranes 15 with tetraethyl ammonium fluoride (TEAF) resulted in a concomitant desilylation and desulfurization^^ (Scheme 22). The protiodesilylation with TBAF/THF or CSF/CH3CN of Diels-Alder cycloadducts 18 with R = Ar is found to occur easily at room temperature giving 36 generally in good yields^^; on the contrary, alkyl-substituted derivatives are more reluctant to undergo protiodesilylation. Thus, drastic conditions (TBAF/toluene/110 °C) are required for the r^rr-butyl derivative (18: R = r-Bu, Si = SiMe3), ^^
1^
cycloaddition^
RJ ^
F- ,
R J ^ 31
Oxone
R Si' 32 Scheme 20.
16
BIANCA FLAVIA BONINI and MARIAFRANCESCA FOCHI
J ^ J < ^ Si
-
^
^
^
p J C ^ ^ H
.d«™,pos,,lo„p,od„c.
Yield (%) 12a X = O
33
PhCN + (PliC0S)2
1 3 X = N-Ph
34
50
PhCN + PliCSNHPli
14 X = CH-C6H4-p-N02
35
12
XTX. Ph^^S ^Ph
Scheme 21.
while the methyl (18: R = Me, Si = SiMej)^^ and the cycloalkyl derivatives (18: R = cycloalkyl, Si = SiMe2Ph) fail to undergo desilylation with fluoride ions. The corresponding 5,5-dioxides 37 can be instead protiodesilylated at room temperature^ giving 38 in excellent yields (Scheme 23). Protiodesilylation of endo- and e;co-trimethylsilyl-3-phenyl-2-thiabicyclo[2.2.1]hept-5-enes 19, ^n^o-trimethylsilyl-^jco-5-oxide 39, and the corresponding 5,5-dioxide 40 with fluoride gave the endo-ph^nyl derivatives through a stereoconvergent reaction. These results were rationalized by hypothesizing a common carbanionic intermediate^^ (Scheme 24). It was demonstrated that desilylation of the silyl thietanes 24 occurs with prevalent configuration inversion at the carbon bearing the silicon group.^^ The inversion probably occurs because in this way the less crowded product is obtained (Scheme 25). Synthesis and Reactivity of Thioaldehyde S-Oxides
Protiodesilylation of thioacylsilane 5-oxides 30 (silyl sulfmes) offers a unique mild synthetic path to thioaldehyde 5-oxides 40 (monosubstituted sulfines), which
^'\y^' Ar
s
TEAF/CH3CN Ar
^\ Ar
15 Ar = Ar^=Ph, Si = SiMea, SiPha Scheme 22.
f"^' Ar^
Synthesis and Chemistry of
17
Thioacylsilanes
R! Si
H R-
^
18
^R
36 Yields (%)
R^ = H, CH3 Si = SiMea, SiMe2Ph
R = Ar R = t-Bu
50-97 30
R = CH3, CycloalkyI
Si
02
^S02 ^Si
Oxone
R - ^ \ R
18
38
37
81-86%
R^ = H, CH3 Si = SiMes, SiMeaPh R = CycloalkyI Scheme 23.
CSF/CH3CN/H2O
W-^S(0)n ^--^-^Ph
'S(0)n
or CSF/DMSO/H2O
SiMes 19n = 0 39n = 1 40 n = 2 Scheme 24.
EWG ^Cz
EWG I
Ph 5C
^O^^S
^'^'^
CSF/DMSO/H2O'
/^^^-S
ph /
EWG
H.^3X_3
24 EWG = CN EWG = C02Me
ratio: 1 ratio: 1 Scheme 25.
2 10
H ^Ph
18
BIANCA FLAVIA BONINI and MARIAFRANCESCA FOCHI
\=S0
- ^
\=SO
SI
H
30
40 Scheme 26.
cannot be obtained through the oxidative route because of the instability of thioaldehydes. Aromatic and ahphatic non-enethiolizable thioaldehyde 5-oxides are obtained by fluoro-desilylation with TBAF in THF/H2O at 40 °C or with CsF in CH3CN at room temperature^'^"^'^^ (Scheme 26). A detailed study*'* of the stereochemistry of fluoro-desilylation (Table 5) showed that the removal of silicon is a stereospecific process occurring with retention of configuration. Accordingly, from the ^-isomers of thioacylsilane 5-oxides 30, the Z-thioaldehyde 5-oxides 40 were obtained. The myrtanyl derivative (entry 9) is the first enantiomerically pure thioaldehyde 5-oxide ever prepared. In contrast, loss of stereochemical integrity was observed during the desilylation of tert-h\iiy\ trimethylsilyl sulfine (entries 4 and 5) and mesityltrimethylsilylsulfine (entries 10, 11, and 12). It was demonstrated that the loss of stereospecificity results from a fluoride-induced equilibration of thioaldehyde 5-oxides after the desilylation.
Table 5. Synthesis and Stereochemistry of Thioaldehyde S-Oxides 30 Entry
R
40 E:Z
Si
E:Z
Yield (%)
Ref.
1
Ph
SiMe3
E
75
Z
14
2
p-CH3-CeH4
SiMe3
E
85
Z
14
3
m-O'CeH^
SiMe3
E
63
z
14
4
t-Bu
SiMe3
60:40
44
30:70
14
5
t-Bu
SiMe3
E^
69^
23:77
14
6
Cyclopropyl
SiMe2Ph
E
27
Z
9
7
Cyclobutyl
SiMe2Ph
E
42
Z
9
8
Cyclohexy!
SiMe2Ph
E
60
Myrtanyl
SiMe2Ph
E
100^
z z
9
9
Mes
SiMe3
66:33
90
5:95
14
11
Mes
SiMe3
E^
90
5:95
14
12
Mes
SiMe3
Z^
90
5:95
14
10
Notes: ^Obtained after chromatography. ^[a]D = -40°(c2.01 CeHe).
9
19
Synthesis and Chemistry of Thioacylsilanes
R
O
fC
fO R
+
40
41 Scheme 27,
Thioaldehyde 5-oxides were subjected to 1,4-cycloaddition with a variety of 1,3-dienes. Unexpectedly, the reactions of buta-l,3-diene and 2,3-dimethylbuta1,3-diene with Z-mono-arylsulfines afforded cisltrans mixtures of the corresponding dihydrothiopyran 5-oxides 41, in which the relative amounts of the two isomers depended on the initial diene/sulfine ratio"^^ (Scheme 27). The behavior of monosubstituted sulfmes is in contrast to that of unsymmetrically disubstituted analogues in the cycloaddition reaction, which occurred with retention of the stereochemistry."^ The deviant result was attributed to a Z to £ isomerization of sulfmes during the cycloaddition and to a surprisingly high reactivity of the E-thioaldehyde 5-oxides compared with the Z-isomers."^^
VII. THIOACYLSILANES CHIRAL AT SILICON OR AT CARBON Mazzanti and co-workers studied the reactions of (/?)-(-)-thiobenzoyl methyl-anaphthylphenylsilane Im with dienes and organometallics^^ in order to test the ability of this thione to transfer chirality to the carbon a to silicon.
S
^^A.
X
Si* Ph
Ph"^Si*
[a]D =-132.2 (c = 0.18C6D6)
I Ph^^^Si*
TBAF
87% 50% de
RM,
SR
Ph sr= H
Ph I .Si—
M^' V iNapht
51% ee
, Ph^^Si*
Ph^^Si*
R = Me, n-Bu, f-Bu, Ph, p-tol; M = Li R = Me, n-Bu, Ph, p-tol; M = MgBr E = H. D Scheme 28.
R = Me E=D de = 48%
V R = Me E=D ee = 45%
20
BIANCA FLAVIA BONINI and MARIAFRANCESCA FOCHI
PhMe2Si^^S **' [a]D = -372
^"^^SiMe2Ph
^-"^SiMeaPh
64% de = 78%
70% de = 81%
^""^ H 84% de = 38%
Scheme 29.
Appreciable diastereoselectivity (50%) was found in the cycloaddition with buta-l,3-diene at -78 °C and the protiodesilylation of the diastereomeric mixture of the adducts with TBAF gave dihydrothiopyran with 51% ee (Scheme 28). Similar diastereoselectivity was observed in the reaction with organolithium and Grignard reagents which gave a-silyl sulfides in moderate yields with medium to good levels of asymmetric induction (de = 40-76%).'*^ As an example, the derivative with E = D gave with protiodesilylation a-deuterio benzyl methyl sulfide with 45% ee. These results lead to the conclusion that in both cases protiodesilylation occurs stereospecifically without loss of the induced chirality (Scheme 28). (15,25,55)-(+)-6,6-Dimethylbicyclo[3.1.1 ]heptan-2-yl-dimethylphenylsilyl thioketone 11 was used as a model compound for chiral at carbon acylsilanes. The de value obtained in the reaction of 11 and buta-l,3-diene was 78%, but the protiodesilylation performed on the corresponding sulfone proceeded with a considerable loss of optical purity^ (Scheme 29).
VIII. ENETHIOLIZABLE THIOACYLSILANES A. Synthesis of Z-a-Silyl Enethiols
As indicated in Scheme 1, enolizable acylsilanes were transformed by thionation, into Z-a-silyl enethiols 4 (Scheme 30). The reaction gave excellent yields independently of the substitution at silicon and the nature of the R^ group, which can be either an alkyl or an aryl group and can contain a range of functional groups including chloro, bromo, and COOH (Table 6). The Z-stereochemistry was assigned to the enethiols 4 by NOE experiments."^^
<x. Si
1)H2S/HCI 2) NaHCOa or L.R. toluene A Scheme 30.
4
^i
Synthesis and Chemistry of Thioacylsilanes
21
Table 6. Synthesis of Z-a-Silyl Enethlols 4 4
R'
Si
a b c d e f
CH3CH2 (CH3)2CH (CH3)2CH Ph CI(CH2)2 CI(CH2)3 CI(CH2)5 Br(CH2)4 Br(CH2)5 Br(CH2)5 Br(CH2)9 Br(CH2)n H02C(CH2)2 H02C(CH2)3 H02C(CH2)4 H02C(CH2)6
SiMe2Ph SiMe2Ph SiMe3 SiMe3 SiMe2Ph SiMe2Ph SiMe3 SiMe2Ph SiMe2Ph SiMe3 SiMe2Ph SiMe2Ph SiMe2Ph SiMe2Ph SiMe2Ph SiMe2Ph
g h i
J k
1 m n 0
P
Yield (%)
Ref.
97 93 94 95 87 100 98 100 100 98 100 96 75 82 85 82
46 47 48 46 18 49 50 18 18 50 18 18 51 51 51 51
It is interesting to note that nonsilylated enethiols have only been prepared as mixtures with their isomeric thioketones."*^'^ The first selective synthesis of aliphatic enethiols was performed by Metzner^^ by deprotonation of enethiolizable thioketones 42 with LDA and reaction of the enethiolates with trimethylsilylchloride. The subsequent methanolysis of silyl vinyl sulfides 43 afforded enethiols devoid of isomeric thioketones. Treatment of the enethiolates with various proton sources afforded instead mixtures of thioketones and enethiols (Scheme 31). B. Synthesis of Cyclic a-Silyl Vinyl Sulfides
a-Silyl enethiols 4 bearing a good leaving group in the co position (4e-l, Table 6) have been used for a convenient synthesis of silylated sulfur heterocycles, via an
MeaSiCI
\[S^^NR3
^QCQ ^ R2
R2
43 LDA/-78°C R ^ ^ A R 3
H 42
_
•^30
R,
SH i R2
Sd le/ne 31.
SH
SSiMea J MeOH
Ri
^ ^5
22
BIANCA FLAVIA BONINI and MARIAFRANCESCA FOCHI
intramolecular cyclization promoted by solid sodium hydrogen carbonate (Scheme 32, path a). The same products could also be obtained via a one-pot reaction, performed without isolation of the intermediate enethiols 4, by thionation of the enolizable acylsilane bearing the leaving group in the co position and treatment of the solution with solid sodium hydroxide (Scheme 32, path b). Excellent to quantitative yields of 2-silylthiacycloalk-2-enes 44 were obtained in the one-pot synthesis of 5- (n = 2), 6- (n = 3), and 7- {n = 4) membered cyclic compounds.^^'"^^'^^ Low yields were instead obtained in the one pot synthesis of meso- (ring size in the range 8 to 11) and macrocycles (ring size > 12), reflecting the fact that in these cases the reactions were more sluggish and prone to competing side reactions, especially dimerization. These difficulties were bypassed following path a under high dilution conditions, i.e., by slowly adding the enethiol to a suspension of solid sodium hydroxide or cesium carbonate in diethyl ether (Scheme 32). In this way^^'^^ the compounds with the 8- (n = 5), 12- (n = 9) and 14- (n = 11) membered rings were synthesized in good yields. C. Synthesis of Open-Chain a-Silyl Vinyl Sulfides
The simplest compound in this series, 1-phenylthio-l-trimethylsilyl ethylene 45, was conveniently prepared by Magnus and co-workers either by reaction of a-lithio vinyl phenyl sulfide 46 with trimethylsilylchloride,^^ or by addition of arylsulfeny 1chloride to vinyl trimethylsilane followed by elimination of hydrogen chloride^^ (Scheme 33). The presence of a substituent in the P-position of the a-silyl vinyl sulfides introduces the problem of geometry in the resulting olefins. As yet, relatively few synthetic methods for P-substituted a-silyl vinyl sulfides are known and some of them afford the olefins as a mixture ofE- and Z-isomers. For example, the Peterson olefination of bis(trimethylsilyl)alkylthiomethyllithium 47 with aldehydes^^ and the treatment of the sulfoxides 48 with LDA and trimethylchlorosilane^^ gave 1)H2S/HCI 2) NaHCOa
x^\_/^"
n = 5,9,11 O 2 X = CI. Br Si = SiMes, SiMe2Ph
base
-*-
Si
44 50-65%
4
path a
Base = NaOH, CsCC^
path b
'- ..J:)
1)H2S/HCI 2)NaOH ' n = 2.3.4
Si-^^S^ ^
Scheme 32,
67-100%
Synthesis and Chemistry of Thioacylsilanes
==/
JiDA^
=/'"'
SPh
23
MeaSiCI
SPh
/^^
97%
siMea
46
/^
PhSCI^
CI
SiMea
45
SiMea
X
92%
SPh 94% Scheme 33,
the products 49 and 50, respectively, as a mixture oi E- and Z-isomers (Scheme 34). On the contrary, the reaction of l-phenylthio-l-trimethylsilyl-2-propene 52^^ with alkyllithium afforded the Z-1-phenylthio-l-trimethylsilyl alkenes 51 and the reaction of 1-methoxy-l-phenylthio-l-trialkylsilylalkanes 54^^ provided a highly stereoselective synthesis of Z-1-phenylthio-l-trialkylsilylalkenes 53 (Scheme 35). A large variety of Z-a-silyl vinyl sulfides 55 were prepared starting from Z-a-silyl enethiols 4a-d with two different procedures"^^'"^^ (Scheme 36). According to path a (Scheme 36), products 55 were stereoselectively obtained by reacting 4 with halides in acetone in the presence of dry K2CO3 at room temperature. The
RS
SiMea
X. MeaSi 1-'
RICHO
"
K
47 R = CHa, Ph R^ = H. CHa, n-CeHia. n-CsHu, CH(C2H5)2. Ph. - H ^ .
p^
^SiMea
53-86%
49
~ ^ 3 ^ 0
or ^MPkA
R
i l i e ^ _ 2)MeaSiCI 48
SPh
\ = / ^glMea 50
R = H, C2H5,Ph, CH=CH2 Scheme 34.
70-80%
24
BIANCA FLAVIA BONINI and MARIAFRANCESCA FOCHI ^.^^^^SMez
RLi
^--S^SiMes
SPh
^ ^
Li-*-SPh
^-^^SiMea E
52
SPh
51 45-78%
E = aldehydes and ketones R
SPh
^
R
SPh
. , /""SiMe2R^ MeO
^^^... „i SiMeaR
53 0-95%
54 R = H, Alk, Ph rf = CH3. Ph. f-Bu Scheme 35.
yields of the reaction were very good (ranging from 80 to 85%) except in the case of halides containing an ethoxy group in the a- or P-position. In these cases the yields were increased by performing the reaction in ether using triethylamine as a base. The alternative path b (Scheme 36) involves a base (DBU)-catalyzed Michaeltype addition of enethiols 4 to olefins bearing an electron-withdrawing group to afford products 55 in excellent yields (70-90%). D. Synthesis of Unsaturated Silylated Thiolactones The chemistry of thiolactones is particularly interesting because of the biological activity associated with a number of these derivatives.^^ More recently, a method has been pursued^ ^ for the preparation of silylated unsaturated thiolactones 56 having a five- to ten-membered ring (Scheme 37). a-Carboxy acylsilanes 57,
^ ' R1
Acetone / K2CO3 S H path a <••'• Si
K />^^ 'si R^ = H, CH=CH2, Ph, 55 80-85% EWG(COORgCN.COR4) EDG (OEt)
path b
4a-d THF/DBU/r.t. R^= Et, /-Pr, Ph Si = SiMes, SiMe2Ph n = 1,2,3
Si 55 70-90%
Scheme 36.
Synthesis and Chemistry of Thioacylsilanes
II 1 CI^'T^fr^CI
(PhMe2Si)2CuCNLi2 THF-78X — o x '
25
Jl X H0'^);r^SIMe2Ph
1)H2S/HCI^ 2) NaHCOa*
57 40-43%
X
^_/^'^^^P^
PPE/CHCI3
HO^>)?Cr^SH
I
r.t.or40x' ^ ^ ^ ^
" = 2.3.4.5.7 SiMe2Ph
4m-p 75-86%
56 52-80% Scheme 37.
prepared in moderate yields by nucleophilic silylation of acyldichlorides with bis(dimethylphenylsilyl)lithium cyano cuprate, were subjected to thionation and treatment with NaHC03 giving excellent yields of co-carboxy-a-silyl enethiols 4m-p (Table 6). Subsequent cyclization with polyphosphate ester (PPE) afforded thiolactones 56 in good to moderate yields depending on the ring size.
IX. REACTIVITY OF a-SILYL VINYL SULFIDES A. Reaction with Electrophiles
Z-a-Silyl vinyl sulfides are intriguing species since the silyl and the thioether functions exert an opposing polarization on the olefmic bond (Scheme 38). The reactivity of 1-phenylthio-l-trimethylsilyl ethylene 45 with various electrophiles was investigated by Ager^° (Scheme 39). The reaction of 45 with bromine followed by treatment with sodium methoxide in methanol gave vinyl bromide 58. The reaction with acylchlorides in the presence of titanium tetrachloride or aluminum trichloride afforded enones 59 arising from an attack of the electrophile controlled by the sulfur. Magnus and co-workers^^^'^^ also found that the reaction of cyclic a,P-unsaturated acid chlorides in the presence of silver tetrafluoroborate leading to P-mercapto phenyl-substituted cyclopentenones 60 via Nazarov cyclization, is dominated by the nucleophilicity of the thioenolether functionality which directs the attack of the acylchloride in the P-position (Scheme 40). This kind of annulation reaction has been used to synthesize (±)-hirsutene.^^*' ®
e
e ® SiRa Scheme 38.
SR
26
BIANCA FLAVIA BONINI and MARIAFRANCESCA FOCHI /SPh \iMe3
_ ^
^
^^'^
SPh Br/^TSiMes
MeONa MeOH
,
/^''^ =^^^
45
58 ^^^
^
RCOCI TiCU AICI3
SiMea 45
^^^\ ,. Q i / V - n MesSi R^ O 59
R Me Ph n-Bu
AICI3 TiCU 70% 75% 63%
68% 78% 56%
Scheme 39,
Cyclic a-silyl vinyl sulfides 44 were shown to be very suitable substrates for the synthesis of thioannulated cyclopentenones.^^ Indeed, treatment of 44 with 3,3-dimethylacryloyl chloride in the presence of AgBF^ afforded bicyclic enones 61. While the reaction proceeded with excellent yields in the case of Si = SiMe3, yields were very low for Si = SiMe2Ph (Scheme 41). The reaction of 2-trimethylsilylthiacyclohex-2-ene 44 (n = 2) with cyclopentenoylchloride in the presence of AgBF^ gave the tricyclic enone 62 in 45% yield^^ (Scheme 42). The reaction of cyclic a-silyl vinyl sulfides 44 (R* = Me, n = 3,4) with acylchloride in the presence of AICI3, gave products 63 and 64, the latter still containing the trimethylsilyl group.^° Protiodesilylation of the reaction mixture, with TBAF in boiling THE afforded products 63. Substrates 44 containing the dimethylphenylsilyl group gave, in some cases, the competitive formation of phenylketones 65, through the attack of the acyl group on the phenyl ring of the silyl moiety (Scheme 43). Products 55 with a methyl group bonded to the sulfur have been tested as model compounds in the reaction with acylchlorides^^ (Scheme 44). Also in these reactions the yields of enones 66 were higher in the case of the starting materials containing the trimethylsilyl group, than in the case of the dimethylphenylsilyl group.
SPh SiMea
CrI
0 AgBF4 ^
y SPh
45
60 35% Scheme 40.
"^SPh
Synthesis and Chemistry of
M-y^'
CI
27
Thioacylsilanes
AgBF4 (1.5 equiv) "S"" 61
44
Si = SiMe3 Yield (%) n=2 83 :3 92 n=4 90 n=5 91
Scheme 41,
Both cyclic and open chain a-silyl vinyl sulfides undergo attack of the electrophiles in the P-position in agreement with the results of Ager and Magnus. It is worth mentioning that the enone 66 did not contain the silyl group in contrast to the result of Ager^^ (Scheme 39). A possible explanation^^ of this result is depicted in Scheme 45. The carbonyl group cis to the silicon would coordinate a molecule of aluminum trichloride giving an intermediate 67 in which the chlorine is in a suitable position for the protiodesilylation. B.
Reaction w i t h Nucleophiles
The addition of an alkyllithium reagent to 1-phenyl-1-trimethylsilyl ethylene 45 was used by Ager to prepare aldehydes^^ and ketones^^ according to Scheme 46. The initially formed acylanion equivalents 68 gave, by quenching with water, 1-phenylthio-1-trimethylsilyl alkanes 69 or, by alkylation, the ketone equivalents 70. Oxidation, thermal rearrangement, and hydrolysis converted products 69 and 70 into the corresponding aldehydes and ketones. It is known that vinyl sulfides couple with Grignard reagents in the presence of Ni(II)-phosphine complexes, to give olefins with predominant retention of configuration.^"^ With the same procedure, a-silyl vinyl sulfides 55 provided disubstituted vinyl silanes 71"*^ (Scheme 47). The reaction was carried out with a large excess of Grignard reagent in the presence of Ni(PPh3)2Cl2 as catalyst. Generally high yields of products 71 were obtained with a high degree of selectivity. Butylmagnesiumbromide gave on the contrary a large amount of product 72 deriving from the hydrogenolysis reaction.
cx *
Ql
AgBF4
Scheme 42,
BIANCA FLAVIA BONINI and MARIAFRANCESCA FOCHI
28
^^/r^" f^
R^OCI/AlCb 3eq. ^ '^'" R2=Me, Ph
n-
R' = Me
^^)r^"
s/
"^ f ~ ^ " TBAF^ 63 "*" R ^ M e z S i ^ S ^
63 20-47%
64 46-15%
44 R^OC.A,Cb3eq.^R.
n = 3,4
^^^^^
R2=Me, Ph,^^>=^
63 35-68% Scheme
M
Riv
SCH3
43.
SCH3
R2C0CI
AICI3 Sequiv
SI
O
65 60-15%
+
^
-1
PhCOR^
Q
R^
66
55 Ri = Et, Ph Si = SiMes, SiMeaPh R2 = Me, Ph,
65
Si =SiMe3 78-100% Si = SiMeaPii 20-97%
\ = y Scheme 44.
ft'
sff
M
^oc, AICI3
SiMeaR
1^ ^sh' j^^ p / ^ > R ^ S^e2R Ok e^Ch>
67 Scheme 45.
e(s!f "
^
Synthesis and Chemistry of
SPh
=<
SiMe3
Thioacylsilanes
RLi/TMEDA^ Et20,0°C
45
29
SPh
SPh 1)MCPBA /—( "TT: ^ H20>r / \ 2) A J > ^ f< ^^SiMea 3, ^ ^ o
SiMea
SPh
68
D
L.. .
SiMea 70
1)MCPBA -*2) A 3)H20
R" "CHO
R
COR'
Scheme 46.
C.
Reaction w i t h Fluoride Ions
The cleavage of the silicon-vinyl carbon bond by fluoride ions is known to be difficult^^ and few examples of this reaction have been reported.^^ The presence of an anion-stabilizing substituent on the carbon bearing the silyl group facilitates the Si-C bond cleavage. In fact, the Z-a-silyl vinyl sulfides 55, obtained according to Scheme 36, were desilylated in good to quantitative yields either by reaction with CsF in moist dimethyl sulfoxide (DMSO) at 40 °C or with TBAF in boiling THF giving vinyl sulfides 73 in a stereospecific way"^^'"^^ (Scheme 48). A deviant result was observed during the desilylation of Z-a-dimethylphenylsilyl vinyl sulfides containing an electron-withdrawing group (3 to the sulfur (55: n = 2, Si = SiMe2Ph, R^ = EWG). A migration of the phenyl group from the silicon to the adjacent carbon occurred, giving products 74 in moderate yields (Scheme 49). It was demonstrated"^^ that the formation of 74 occurs by a retro-Michael reaction of the carbanion 75, originated by a fluoride-induced deprotonation a to the electronwithdrawing group. The loss of the olefin 76 results in the thione 78 via the enethiolate 77. The phenyl group migrates from the silicon to the adjacent carbon, giving the thiolate 79 which, by Michael addition to the olefin 76, affords product 74 after desilylation (Scheme 50).
R\ ^ S j
Ni(PPh3)2Cl2
55
R2
Si
Si 71 60-94%
R^ = Et, Ph Si = SiMes, SiMeaPh R2=Me, Et, Bu Scheme 47.
72
30
BIANCA FLAVIA BONINI and MARIAFRANCESCA FOCHI S^..^R2
. _^^
.^v
R\
AorB .
55
S
R2
^ ^ ^ V ^ 73
R^ = Et, hPr, Ph SI = SIMe2Ph, SiMes R 2 = H . EWG,
A = CsF/DMSO/40°C B = TBAF/THF/reflux
EDG
n = 1,2.3 Scheme 48.
^\_/^-^--EWG ^SiMesPh
_IBAF__ THF/H2OA
p
Ph 1^ "^ '
^EWG
74 48-60%
EWG = COOEt, COOMe, CN, COMe Scheme 49.
- -^^^EWG
SiMe2Ph
SiMe2Ph 75 R^
^ .
S
V<.'SiMe2PH
'SIMe2Ph
^
*
^^^ (^
Si—Me
^Ph^ 'Me
77
78 R^
S®
^^TT^sii SiMe2F
<^^EWG 76
o
R!
-^
Ph
^'^^-^'''^EWG Ph
74
79 Scheme 50.
+ SiMe2F2
Synthesis and Chemistry of Thioacylsilanes
31
X. THEUSEOFTHIOACYSILANESASSPIN TRAPPING AGENTS Organic free radicals and radical ions are most conveniently studied by means of electron spin resonance (ESR) spectroscopy, a tool that can provide direct information on the electronic and molecular structure of the species under investigation.^^ On the other hand, the applicability of this technique is often limited by the short lifetime of free radicals; indeed, these species are normally transient and very labile and under the reaction conditions they often fail to reach a steady-state concentration larger than the sensitivity threshold typical of ESR spectrometers. This inconvenience can be bypassed in most cases through the use of spin trapping, a technique through which transient radicals are converted to persistent (or longer-lived) spin adducts (SA) by reaction with appropriate substrates (the spin traps, ST) added to the reaction system.^^ The vast majority of successful spin traps designed since the introduction of the technique belong to the nitroso and nitrone compounds, with the resulting spin adducts being dialkyl- or aryl alkyl nitroxides.^^ Since the late 1970s it has become evident that also thiocarbonyl compounds can readily undergo homolytic addition by a large variety of organic radicals through a regiospecific thiophilic attack'^^"'^^ (Scheme 51). The main drawback in the use of thiocarbonyl compounds as spin traps was represented by the fact that in most cases the resulting spin adducts either were as transient as the attacking radicals (aliphatic thioketones and dithioesters) or were characterized by very complex ESR spectra (thiobenzophenone and its derivatives). It was only after the introduction of thiobenzoyltriphenylsilane l a that the use of thiocarbonyl compounds in spin trapping experiments acquired some practical value. Alberti, Pedulli, and others^'^'^^ showed that like other thiocarbonyl compounds, la regiospecifically undergoes thiophilic radical addition to afford very persistent spin adducts. The general structure is outlined in Scheme 52. The unusual persistence of the spin adducts from la has been attributed to a combination of two factors.'^^ First, the large bulk of the triphenylsilyl group provides a sort of "protective umbrella" to the radical center, thus preventing it from undergoing further reactions. Second, the presence of the electron-withdrawing triphenylsilyl group and the electron-donating SXR' group simultaneously bound to the radical center provides a capto-dative stabilization^^ of the radical, enhancing its persistence. The spin adducts from l a exhibit ESR spectra whose hyperfme pattern originates from the interaction of the unpaired electron with the five aromatic protons of the
R2C=S ST
+
•XR'
^
Scheme 51.
R2C—SXR' SA
32
BIANCA FLAVIA BONINI and MARIAFRANCESCA FOCHI SXR' SiPha
+
-XR'
r
T
• SiPha
1a Scheme 52.
thiobenzoyl moiety and with the magnetically active nuclei of the XR' fragment: the spectra, although sometimes complex, are therefore very informative as to the nature of the trapped radical. The introduction of two tert-bniy\ groups in positions 3 and 5 of the aromatic ring provides a solution to the spectral complexity by reducing the number of aromatic hydrogens interacting with the unpaired electron.^^ Indeed, 3,5-d\-tertbutylthiobenzoyltriphenylsilane, le, leads to spin adducts with simpler ESR spectra and its use is favored in those experiments where the trapping of more than one transient radical is foreseen. In general, la and le are very reactive toward nucleophilic carbon-centered radicals (e.g., alkyls) as well as radicals centered at a phosphorus atom such as phosphinyls, phosphonyls, and thiophosphonyls, which can be readily identified from the knowledge of the spectral parameters of their adducts. ^^'^^'^^'^^ They are also good scavengers for radicals centered at a group 14 element (•SiR3, •GeR3, •SnR3) or at a transition metal atom [•Mn(CO)5, •Re(CO)5], and are almost unrivaled in the trapping of electrophilic oxygen- and sulfur-centered radicals (e.g., alkoxyls and thiyls).^^''^'*-'^^'^^ There are instead no reports of the trapping of nitrogen-centered radicals by la and le. As afinalpoint, it should be emphasized that the detection of spin adducts in the presence of thioacylsilanes does not necessarily imply the occurrence of a radicalbased reaction because paramagnetic species identical to the spin adducts can sometimes be formed through reactions other than spin trapping which involve single electron transfer processes. This has been recently shown to be the case for Grignard reagents that spontaneously react with thiocarbonyl compounds having an appropriately low reduction potential.'^^
ACKNOWLEDGMENTS The authors wish to thank all of the colleagues listed in the references, in particular Angelo Alberti, Mauro Comes-Franchini, Gaetano Maccagnani,^ GermanaMazzanti, Alfredo Ricci, Paolo Zani, and Binne Zwanenburg and the many students who contributed to the work reported herein. ^Deceased on March 11, 1989.
Synthesis and Chemistry of Thioacylsilanes
33
REFERENCES 1. Bonini, B. F ; Maccagnani, G.; Mazzanti, G.; Sarti, S.; Zanirato, P. J. Chem. Soc, Chem. Commun. 1981, 822. 2. (a) Bonini, B. F. Phosphorus, Sulfur and Silicon 1993, 74, 31; (b) Bonini, B. F ; Fochi, M. Rev. on Heteroatom Chem. 1997,16,47. 3. Duus, F In Comprehensive Organic Chemistry; Jones, D. N., Ed.; Pergamon: Oxford, 1979, Vol. 3. 4. (a) Demuynck, C.; Demuynck, M.; Paquer, D.; Vialle, J. Bull. Soc. Chim. Fr. 1966, 3366; (b) Paquer, D.; Vialle, J. Bull. Soc. Chim. Fr. 1969, 3327; (c) Paquer, D.; Vialle, J. Bull Soc. Chim. Fr. 1969,3595; (d) Paquer, D.; Vialle, J. Bull. Soc. Chim. Fr. 1971,4407; (e) Kresge, A. J.; Meng, Q. / Am. Chem. Soc. 1998,120, 11830. 5. (a) Kresge, A. J.; Tobin, J. B. / Am. Chem. Soc. 1990,112, 2805; (b) Kresge, A. J.; Tobin, J. B. J. Org. Chem. 1993, 58, 2652. 6. (a) Ricci, A.; Degl'Innocenti, A. Synthesis 1989, 647; (b) Bulman Page, R C ; Klair, S. S.; Rosenthal, S. Chem. Soc. Rev. 1990,19, 147. 7. Brook, A. G.; Duff, J. M.; Jones, R F ; Davis, N. R. J. Am. Chem. Soc. 1967, 89,431. 8. Corey, E. J.; Seebach, D.; Freedman, R. J. Am. Chem. Soc. 1967, 89, 434. 9. Bonini, B. F ; Busi, F ; de Laet, R. C.; Mazzanti, G.; Thuring, J. W. J. F ; Zani, R; Zwanenburg, B. J. Chem. Soc, Perkin Trans. 1 1993, 1011. 10. Bonini, B. F ; Comes-Franchini, M.; Mazzanti, G.; Passamonti, U.; Ricci, A.; Zani, P. Synthesis 1995, 92. 11. Cappenicci, A.; Degl'Innocenti, A.; Faggi, C.; Ricci, A.; Dembech, P.; Seconi, G. J. Org. Chem. 1988,55,3612. 12. Brillon, D. Sulfur Rep. 1992,12, 297. 13. Barbaro, G.; Battaglia, A.; Giorgianni, P.; Maccagnani, G.; Macciatelli, D.; Bonini, B. F ; Mazzanti, G.; Zani, R J. Chem. Soc, Perkin Trans. 11986, 381. 14. Barbaro, G.; Battaglia, A.; Giorgianni, P.; Bonini, B. F ; Maccagnani, G.; Zani, P. J. Org. Chem. 1990, 55, 3744. 15. Alberti, A.; Benaglia, M. J. Organomet. Chem. 1992,434, 151. 16. Bonini, B. F ; Mazzanti, G.; Zani, R; Maccagnani, G. J. Chem. Soc, Perkin Trans. 1 1989, 2083. 17. Bonini, B. F ; Maccagnani, G.; Mazzanti, G.; Zani, R J. Chem. Soc, Chem. Commun. 1988, 365. 18. Bonini, B. F ; Comes-Franchini, M.; Fochi, M.; Mazzanti, G.; Ricci, A. Tetrahedron 1996, 52, 4803. 19. Ricci, A.; Degl'Innocenti, A.; Cappenicci, A.; Reginato, G. J. Org. Chem. 1989, 54,19. 20. Bonini, B. F ; Comes-Franchini, M.; Fochi, M.; Mazzanti, G.; Ricci, A. Synlett. In press. 21. Barbarella, G.; Bongini, A. Tetrahedron 1989,45, 5137. 22. Vedejs, E.; Perry, D. A. J. Am. Chem. Soc 1983,105, 6999. 23. Carisi, P.; Mazzanti, G.; Zani, P.; Barbaro, G.; Battaglia, A.; Giorgianni, P. J. Chem. Soc, Perkin Trans. 11987, 2647. 24. Bonini, B. F ; Foresti, E.; Maccagnani, G.; Mazzanti, G.; Zani P. J. Chem. Soc, Perkin Trans. 1 1988,1499. 25. Beslin, R; Lagain, D.; Vialle, J. J. Org. Chem. 1980,45, 2517. 26. Beak, R; Warley, J.W. J. Am. Chem Soc 1972, 94, 597. 27. Ohno, A.; Nakamura, K.; Vohama, M.; Oka, S.; Yamabe,T.; Nagata, S. Bull Chem. Soc Jpn. 1975, 48, 3718. 28. Bonini, B. F ; Masiero. S.; Mazzanti, G.; Zani, R Tetrahedron Lett. 1991, 32, 815. 29. Bonini, B. F Unpublished results. 30. Bonini, B. F ; Maccagnani, G.; Mazzanti, G.; Atawa, G. A. L. A.; Zani, R Heterocycles 1990, 31, 47.
34
BIANCA FLAVIA BONINI and MARIAFRANCESCA FOCHI
31. Bonini, B. F ; Foresti, E.; Leardini, R.; Maccagnani, G.; Mazzanti, G. Tetrahedron Lett. 1984,25, 445. 32. Bonini, B. F ; Lenzi, A.; Maccagnani, G.; Barbaro, G.; Giorgianni, P.; Macciantelli, D. J. Chem. Soc, Perkin Trans. 11987, 2643. 33. Bonini, B. F ; Masi, E.; Masiero, S.; Mazzanti, G.; Zani, P. Tetrahedron 1996, 52, 3553. 34. Kang, K. T.; Park, C. H.; Yoon, U. C. Bull. Korean Chem. Soc. 1992,13, 41. 35. Bonini, B. F ; Foresti, E.; Maccagnani, G.; Mazzanti, G.; Sabatino, P.; Zani, P. Tetrahedron Lett. 1985,26,2131. 36. Bonini, B. F ; Masiero, S.; Mazzanti, G.; Zani, R Tetrahedron Lett. 1991, 52, 2971. 37. (a) Ohno, A.; Ohnishi, Y; Fukuyama, M.; Tsuchihashi, G. J. Am. Chem. Soc. 1968, 90, 7038; (b) Ohno, A.; Ohnishi, Y; Tsuchihashi, G. J. Am. Chem. Soc. 1969, 97, 5038; (c) Ohno, A.; Ohnishi, Y; Tsuchihashi, G. Tetrahedron Lett. 1969, 283. 38. (a) Lawrence, A. H.; Liao, C. C ; de Mayo, R; Ramamurthy, V. J. Am. Chem. Soc. 1976, 98, 2219; (b) Devanathan, S.; Ramamurthy, V. J. Org. Chem. 1988, 53, 741; (c) Pushkara Rao, V; Ramamurthy, V. J. Org. Chem. 1988,53, 327. 39. Bonini, B. F ; Comes-Franchini, M.; Fochi, M.; Mazzanti, G.; Ricci, A., Zani, P.; Zwanenburg, B. J. Chem. Soc, Perkin Trans. 1 1995, 2039. 40. Zwanenburg, B. Rec. Trav. Chim. Pays-Bas 1982,101, 1. 41. King, J. F ; Rathore, R. The Chemistry ofSulphonic Acid and Esters and Their Derivatives; Patai, S.; Rappoport, Z., Eds.; John Wiley & Sons: New York, 1994, Chapter 17, p. 697. 42. Bonini, B. F ; Mazzanti, G.; Zani, P., Maccagnani, G.; Barbaro, G., Battaglia, A.; Giorgianni, P. J. Chem. Soc, Chem. Commun. 1986, 964. 43. Barbaro, G.; Battaglia, A.; Giorgianni, P.; Bonini, B. F ; Maccagnani, G.; Zani, P. J. Org. Chem. 1991,56,2512. 44. Zwanenburg, B.; Thijs, L.; Broens, J. B.; Strating, J. Rec Trav. Chim. Pays-Bas 1972, 91, 443. 45. Bonini, B. F ; Maccagnani, G.; Masiero, S.; Mazzanti, G.; Zani, P. Tetrahedron Lett. 1989, 30, 2677. 46. Bonini, B. F ; Comes-Franchini, M.; Fochi, M.; Mazzanti, G.; Peri, F ; Ricci, A. J. Chem. Soc, Perkin Trans. 1 1996, 2803. 47. Bonini, B. F ; Comes-Franchini, M.; Fochi, M.; Mazzanti, G.; Ricci, A. J. Chem. Soc, Perkin Trans. 11997,3211. 48. Bonini, B. F Unpublished results. 49. Bonini, B. F;Comes-Franchini, M.; Mazzanti,G.; Ricci, A.; Rosa-Fauzza, L.; Zani, P. Tetrahedron Lett. 1994, 35, 9227. 50. Bonini, B. F ; Comes-Franchini, M.; Fochi, M.; Mazzanti, G.; Ricci, A. Tetrahedron 1997, 53, 7897. 51. Bonini, B. F ; Comes-Franchini, M.; Fochi, M.; Mazzanti, G.; Ricci, A. Synlett 1999, 486-488. 52. Le Nocher, A. M.; Metzner, R Tetrahedron Lett. 1992, 33, 6151. 53. Harirchian, B.; Magnus, P. J. Chem. Soc, Chem. Commun. 1977, 522. 54. Cooke, F ; Moerck, R.; Schwindenam, J.; Magnus, R J. Org. Chem. 1980, 45, 1046. 55. Grobel, B. T; Seebach, D. Chem. Ber. 1977,110, 852. 56. Miller, R. D.; Hassig, R. Tetrahedron Lett. 1984, 25, 5351. 57. Kyler, K. S.; Watt, D.S. / Org. Chem. 1981, 46, 5182. 58. Mandai, T.; Kohama, M.; Sato, H.; Kawada, M.; Tsuji, J. Tetrahedron 1990,46, 4553. 59. (a) Wang, C. L. J.; Salvino, J. M. Tetrahedron Lett. 1984, 25, 5243; (b) Panetta, J. A.; Rapoport, H. J. Org. Chem. 1982, 47, 2626; (c) Khan, S. A.; Erickson, B. W. J. Am. Chem. Soc 1984,106, 798; (d) Araujo, H. C ; Mahajan, J. R. Synthesis 1978, 228. 60. Ager, D. J. Tetrahedron Lett. 1982, 23, 1945. 61. (a) Magnus, R; Quagliato, D. A.; Huffman, J. C. Organometallics 1982,1, 1240; (b) Magnus, R; QuagHato, D. A. J. Org. Chem. 1985, 50, 1621. 62. Ager, D. J. Tetrahedron Lett. 1981,22, 587.
Synthesis and Chemistry of Thioacylsilanes
35
63. Ager, D. J. Tetrahedron Lett. 1983, 24,95. 64. (a) Wenkert, E.; Ferreira, T. W.; Michelotti, E. L. J. Chem. Soc, Chem. Commun. 1979, 637; (b) Okamura, H.; Miura, M.; Takei, H. Tetrahedron Lett. 1979, 43; (c) Fiandanese, V.; Marchese, G.; Naso, F; Ronzini, L. J. Chem. Soc, Chem. Commun. 1982, 647; (d) Fiandanese, V; Marchese, G.; Naso, F ; Ronzini, L. J. Chem. Soc, Perkin Trans 11985, 1115. 65. (a) Chan, T. H.; Mychajlowskij, W. Tetrahedron Utt. 1974, 3479; (b) Chan, T. H.; Lau, R W. K., Li, M. R Tetrahedron Lett. 1976, 2667. 66. Oda, H.; Sato, M.; Morizawa, Y; Oshima, K.; Nozaki, H. Tetrahedron Lett. 1983, 24, 2877. 67. (a) Camngton, A.; McLachlan, A. D. Introduction to Magnetic Resonance, Harper & Row: New York, 1969; (b) Wertz, J. E.; Bolton, J. R. Electron Spin Resonance: Elementary Theory and Practical Applications, McGraw-Hill: New York, 1972. 68. Perkins, M. J. In Advances in Physical Organic Chemistry, Gold, V.; Bethell, D., Eds.; Academic Press: London, 1980, Vol. 17, p. 1, and references therein. 69. Janzen, E. G.; Haire, D. L. Adv. Free Radical Chem. 1990,1, 253. 70. Scaiano, J. C ; Ingold, K. U. / Am. Chem. Soc 1976, 98, 4727. 71. Alberti, A.; Colonna, F R; Pedulli, G. F Tetrahedron 1980, 36, 3043. 72. Adeleke, B. B.; Chen, K. S.; Wan, J. K. S. / Organomet. Chem. 1981, 208, 317. 73. McGimpsey, W.G.; Depew, M.C.; Wan, J.K.S. Phosphorus Sulphur 1984, 21, 135. 74. Alberti, A.; Bonini, B. F ; Pedulli, G. F Tetrahedron Lett. 1987, 28, 3737. 75. Alberti, A.; Benaglia, M.; Bonini, B. F ; Pedulli, G. F J. Chem. Soc, Faraday Trans. 1 1988, 84, 3347. 76. Viehe, G.; Janousek, Z.; Merenyi, R.; Stella, L. Ace Chem. Res. 1985,18, 917. 77. Alberti, A.; Benaglia, M.; Bonini, B. F; PedulU, G. F Phys. Med 1989, 2-4, 151. 78. Alberti, A.; Benaglia, M.; Vismara, E. Res. Chem. Intermed 1989,11, 117. 79. Alberti, A.; Benaglia, M.; Macciantelli, D.; Marcaccio, M.; Olmeda, A.; Pedulli, G. F ; Roffia, S. y. Org. Chem. 1997, 62, 6309.
This Page Intentionally Left Blank
IODINE, A VERSATILE REAGENT IN CARBOHYDRATE CHEMISTRY: ACTIVATION OF THIOGLYCOSIDES AND GLYCOSYL SULFOXIDES
K. R Ravindranathan Kartha, Mahmoud Aloui, Peter Cura, Steven J. Marsh, and Robert A. Field
I. II. III. IV. V. VI. VII. VIII.
Glycobiology and the Challenge for Synthetic Chemistry Glycosyl Donors Thioglycosides Iodine as a Thiophilic Reagent Iodine-Promoted Glycosylation Chemoselective Activation Glyco-amino Acid Synthesis Activation of "Disarmed" Thioglycosides A. Iodine plus Additive B. Interhalogens IX. Tuning Donor Reactivity X. Iodine as a Lewis Acid A. Activation of Glycosyl Sulfoxides
Advances in Sulfur Chemistry Volume 2, pages 37-56. Copyright © 2000 by JAI Press Inc. All rights of reproduction in any form reserved. ISBN: 0-7623-0618-1 37
38 38 39 39 41 42 43 44 45 46 50 51 51
38
K. P. RAVINDRANATHAN KARTHA et al.
B. Synthesis of Thioglycosides and Glycosyl Iodides XI. Future Prospects Acknowledgments References
53 54 54 55
I. GLYCOBIOLOGY AND THE CHALLENGE FOR SYNTHETIC CHEMISTRY It is becoming increasingly clear that carbohydrates are not simply the biological energy stores they were once thought to be. The past 20 years has highlighted a plethora of roles for sugars in precise structural, recognition, and signaling events ranging from determination of blood group types in man and evasion of the host immune response by the African trypanosome, to bacterial signaling for root nodule formation in leguminous plants.^""^ In many cases, the carbohydrate structures concerned are either heterogeneous or occur naturally in such small quantities that determination of structure-function relationships is very problematic indeed. There is clearly a need for the synthesis of oligosaccharide structures to facilitate biological studies. However, in contrast to the relatively straightforward, automated methods available for peptide and nucleic acid synthesis, the preparation of oligosaccharides has remained in the dark ages until relatively recently. While advances have been made in methods for glycoside synthesis over the past 20 years,^ few methods have found universal favor and the old adage of a different method for the synthesis of each glycosidic linkage still remains uncomfortably true.
II. GLYCOSYL DONORS Perhaps the most widely established activated building blocks for glycoside synthesis are the glycosyl halides, glycosyl trichloroacetimidates, and thioglycosides.^ Recent developments have highlighted the potential of glycals^ and glycosyl sulfoxides,^ which also offer much scope for improved coupling strategies (Fig. 1). With the exception of the thioglycosides and glycals, which require chemospecific activation and are routinely stable to functional group transformations, the other glycosyl donor reagents mentioned are inherently labile.
glycosyl halide
glycosyl trichloroacetimidate
thioglycoside
Figure 1.
glycal
glycosyl sulfoxide
Iodine in Carbohydrate Chemistry
39
III. THIOGLYCOSIDES Chasing the "holy grail" of shelf-stable building blocks that might ultimately be compatible with a solid-phase glycosylation approach,^ we were particularly attracted to the thioglycosides. The preparation and application of thioglycoside building blocks in oligosaccharide synthesis have recently been reviewed by Norberg^ and Garegg.*^ Typical methods for the activation of this class of building blocks make use of reagents such as mercuric cyanide, methyl triflate, dimethyl(methylthio)sulfonium triflate (DMTST), iodonium dicollidine perchlorate (IDCP), //-iodosuccinimide (NIS)/triflic acid, and NIS/silver nitrate.^'^^ Most of these reagents have drawbacks related to their toxicity, cost, smell, or difficulty of handling. Most are also light- and moisture-sensitive. Practical alternatives would therefore be welcome.
IV. IODINE AS A THIOPHILIC REAGENT We were attracted to the use of iodonium ion-based activation of thioglycosides and we were drawn to conclude that molecular iodine might serve as an effective source of this, or related, thiophilic species (Fig. 2). Inspection of the literature identified work by Hiskey and Tucker^ ^ who reported the use of iodine to cleave a THP-protected thiol in 1962. Even earlier work by Ratner and Clarke in 1937 demonstrated that thiazolidines can also be cleaved using iodine.^^ Both of these reactions give rise to disulfides (Fig. 3). It is interesting to note that Hiskey and Tucker comment that their approach was "suggested by the earlier work of Bonner,"^^ who in 1948 demonstrated the conversion of acetylated
ROg
O OR
Figure 2.
cx..^ "O
^S
NH2 RS-SR
CO2R H
'CO2R
Figure 3.
K. P. RAVINDRANATHAN KARTHAet al.
40
CH2OAC
CHjOAc AcO
AcO' Br Figure 4.
l2 + TrtC104
^ P 30?
© e
^=
CHjOBn f
BnO-*''T^ *OR
RC)H
BnOi
+ Trt I CHzOBn BnO-^-^^V BnO
1 Me' Me Figure 5.
Ph-I0 + Tf20
)Tf. TfO
^
ROH
^ 'R
- r ^R Ph
A
^
Ar-IFj
O
N ^
^
© "AT
I AT
Figure 6.
Iodine in Carbohydrate Chemistry
41
thioglucoside to the corresponding glycosyl bromide on treatment with bromine (Fig. 4). However, neither Hiskey and Tucker nor Bonner comment on the reactivity of thioglycosides with iodine. It might have been expected that the less electrophilic iodine would be less likely than bromine to activate acetylated thioglycosides. The tetrahydropyranyl thioacetal is more reactive than a thioglycoside since it does not possess electron-withdrawing functionality. It is therefore perhaps not surprising that iodine was not considered an obvious reagent for the direct activation of thioglycosides, although it has found use in conjunction with silver oxide in the Koenigs-Knorr glycosylation reactions with glycosyl halides.^'* More recent studies have shown that iodine in conjunction with trityl perchlorate, which presumably forms iodonium perchlorate in situ, efficiently activates glycosyl dimethylphosphinothioates for glycosylation chemistry (Fig. 5).^^ Hypervalent iodine species have also been shown to promote glycosylation reactions with thioglycoside donor units,^^ and to effect the conversion of thioglycoside to glycosyl fluorides (Fig. 6).^^
V. IODINE-PROMOTED GLYCOSYLATION In our hands iodine was indeed found to be an effective promoter of the alcoholysis of unprotected thioglycosides.^^ Acetylated, "disarmed"^^ thioglycosides proved unreactive toward iodine, whereas benzylated, "armed"^^ thioglycosides were found to undergo efficient glycosylation of both simple and sugar alcohols in the presence of iodine (Fig. 7).^^
^"VcH^OBn
UXt-O BnOAi.**^\. -SMe BnO
I2. ROH *CH2CI2
BnO.\i:^A^OR BnO
ROH = "?CH20Bn HO
92% a:3-1.2:1
83% a:P-2:l Figure 7.
BnO V ^ ^ ^ ^ O M e BnO
91% a:P -1.2:1
K. P. RAVINDRANATHAN KARTHAet al.
42
II \
//\ I-I
Me
^ 1® 7^ I®I-^®Me ^«S'
®
Me
W^S^Me
Prospective transformations of a thioglycoside in the presence of iodine and an acceptor alcohol might give rise to a variety of species (Fig. 8) that could have an impact on the efficiency and stereochemical outcome of the reaction, such as alkylsulfenyl iodides, H-I, and glycosyl iodides. To date we have been unable to identify the formation of glycosyl iodides from thioglycosides on treatment with iodine under a variety of conditions (see also Section X.B.). The glycosyl species which reacts with the acceptor alcohol could have a profound effect on the stereochemical outcome of the reaction. Traditional glycoside coupling chemistry is based around the cyclic oxocarbonium ion, with resulting loss of anomeric stereochemistry at the transition state/reaction intermediate stage. If one could identify glycosyl species that will undergo clean Sj^2 chemistry at the anomeric center, with concomitant inversion of stereochemistry, one might have a general approach to glycoside synthesis.
VI. CHEMOSELECTIVE ACTIVATION It is clear from the work of van Boom^^ that esterified thioglycosides are much less reactive than their benzylated counterparts, thus permitting the chemospecific activation of an "armed" thioglycoside donor (with IDCP) and its coupling to a "disarmed" thioglycoside acceptor. The "disarmed" thioglycoside center is carried through one round of coupling and is then available for subsequent activation under more forcing conditions (e.g., NIS/TfOH) without the need to manipulate the anomeric leaving group (Fig. 9).
Iodine in Carbohydrate Chemistry
43
CH20Bn
S"0
CH2OH
CHjOBn
JIX:P
BnO
- *
0--V
BzO-\^v
BzO-A-'^'^^SEt BzO
_,
BzO 84% a:P7:l Figure 9.
BnOCHaOBn BnO
h^
BnO-\-^
HO-V-'^^^^^SMe AcO
71% a:p 1:0 figure 10.
We have demonstrated^^ the selectivity of iodine in the chemospecific activation of an "armed" thioglycoside in the synthesis of the galactose-containing disaccharide shown in Figure 10. This disaccharide is a fragment of the so-called Gallili antigen, which is of interest in connection with xenotransplantation since it is responsible for the rejection of porcine tissue by man.
VII. GLYCO-AMINO ACID SYNTHESIS The synthesis of glycopeptides as mimics of oligosaccharides^^ or as a basis for the development of anticancer vaccines^^ is currently topical. While many of the traditional glycosylation procedures have been explored for the synthesis of glycoamino acid building blocks, few methods give uniformly reliable results. Taking this into account, and noting the cost of the limited number of commercially available glyco-amino acid building blocks (typically $450/100 mg), we have investigated the iodine-promoted glycosylation of Fmoc-protected serine units.^^ Glycosylation with acetobromogalactose gives the expected |3-glycoside in nearquantitative yield. Glycosylation of the same acceptor with an "armed" thiogalactoside, however, is less straightforward. In the absence of an acid scavenger, such as potassium carbonate, the thermodynamically favored a-glycoside is formed. In
K. P. RAVINDRANATHAN KARTHAet al.
44
BnOCH20Bn BnO
I2.CH2CI2
NH-Fmoc
R = Me, Bn
I2, CH2CI2 K2CO3
BnOcH20Bn BnO^^^A BnOl
BnOCH20Bn BnO-V^--^*^-' BnO 80-90% a:P 1:3.5
80-90% a:p3:l Figure 11.
contrast, in the presence of potassium carbonate, P-glycoside is the dominant product (Fig. 11). We assume that the P-glycoside is the kinetic product and that it undergoes an acid-catalyzed epimerization in situ. It is conceivable that H-I is responsible for effecting this process (see also Fig. 8 and Section X). Attempts to glycosylate Boc-protected serine were very disappointing. This is perhaps not surprising since glycosylation reactions tend to be acidic. In fact, we find that iodine is an effective reagent for the cleavage of the Boc group, although we assume this reflects in situ generation of H-I, as reported for the sodium iodide-mediated cleavage of the Boc group^"* and the iodine-methanol cleavage of trityl ethers.^^
VIII. ACTIVATION OF "DISARMED" THIOGLYCOSIDES Perhaps the principal cost in solution-phase oligosaccharide synthesis is manpower, related to the routine but laborious need for chromatography after each round of glycoside coupling. In addition to developments in solid-phase methods,^ there is also great interest in the development of one-pot multiple coupling procedures,^^ methods for the orthogonal activation of building blocks,^^ and the development of blockwise syntheses.^^ Each of these approaches requires the availability of "tuned" glycosyl donors and/or promoters to allow chemoselective activation. In the context of our own work, we faced the challenge of identifying a practical alternative to the aggressive NIS/TfOH combination to activate "disarmed" thioglycosides.
Iodine in Carbohydrate Chemistry
45
A. Iodine plus Additive
To standardize reagents, we chose to stick with an iodine-based reagent in the hope that an "additive" might increase its potency as a thioglycoside activator (as noted above, iodine alone does not activate "disarmed" thioglycosides). The CANpromoted oxidation of acetamidophenyl glycosides attracted our attention (Fig. 12),^^ as did the DDQ-mediated oxidative cleavage of methoxybenzy 1 ethers.^^We therefore investigated the effect of CAN on thioglycoside activation. On its own, this reagent did not activate "armed" methoxybenzyl thioglycosides. However, in combination with iodine (I2-CAN has previously been used for the oxidative aromatization of cyclic enones^^), methanolysis rates increased dramatically (approximately 70-fold) (Fig. 13).^^ The mechanism of the l2-CAN-mediated glycosylation remains to be established. It is not clear if CAN plays an oxidative or an electrophilic role in thioglycoside activation. Sulfur-centered radical cation intermediates have previously been proposed for electrochemical glycosylation reactions using thioglycosides (Fig. 14).^^'^"^ Similar species have also been proposed for the chemical single-electron oxidation of thioglycosides with the stable tris-(4-bromophenyl)ammonium radical cation,^^ although an electrophilic mode of activation by this agent has recently been suggested by Mehta and Pinto.^^ Attempts to perform glycosylation reactions with acetylated, "disarmed" methoxybenzyl thioglycosides proved unsatisfactory because of partial deacetylation of the donor and product under the acidic conditions generated by use of CAN. We therefore moved to a milder (i.e., lower redox potential) oxidizing agent, namely, DDQ (redox potentials: Ce^^ = 1.61 V, DDQ = 1.00 V).^^ The I2-DDQ combination proved particularly effective at activating acetylated glycosyl halides for the glycosylation of simple alcohols.^^ However, this reagent combination was
^=s^<
-o-
^^a.
CAN
NHAc
^OR DDQ
OMe
MeCN H2O
HO^^OR
HoO
NAc
CHO
^'OMe
OMe
Figure 12,
OMe
A^O
K. P. RAVINDRANATHAN KARTHAet al.
46
OMe BnO ll2,CAN,MeCN
:§,Me BnO
^'^VcHiOBn I
OMe
BnO V--'^^^S BnO I MeOH BnOI uJttoOBn BnO BnO
OMe
Figure 13.
much less effective in attempted disaccharide syntheses with both "disarmed" glycosyl halides and "disarmed" thioglycosides.-^^ An alternative activation system was therefore sought. B. Interhalogens It occurred to us that the interhalogens might serve as a more potent source of electrophilic iodine than I2 itself. The relative electrophilic character of such
SPh
^ ^ * S P . * ^ PhS' \ PhS-SPh Figure 14.
ROH
^ ^ ^ ^ O R
Iodine in Carbohydrate Chemistry
47
compounds (I-Cl > I-Br > y was established many years ago.^^ We have demonstrated that, in the absence of an acceptor alcohol, treatment of thioglycosides with I-Br gives rise to the corresponding glycosyl bromides in excellent yields. The reaction gives exclusively the thermodynamically favored axial bromide and is compatible with a wide range of common protecting groups, including those that are acid labile such as methoxybenzyl ethers and isopropylidene acetals (Fig. 15)."^^ In contrast to iodine, I-Br is a very effective promoter of disaccharide syntheses from both "disarmed" glycosyl halides and "disarmed" thioglycosides (Fig. 16)."^^ Care needs to be taken with this reagent since it is a very effective Lewis acid and can promote acyl migration reactions (see also Section X)."^^ Comparative studies on the reaction of an "armed" thiogalactoside with I2, I-Br, and I-Cl in the absence of an acceptor alcohol give an interesting set of results (Fig. 17)."*^ Reaction with iodine results in thioglycoside epimerization, a process previously observed in van Boom's group,"*^ but we have been unable to detect formation of glycosyl iodides. As indicated earlier in this section, reaction of thioglycosides with I-Br gives rise to the thermodynamically more stable axial bromide. In contrast, reaction with I-Cl gives rise to the anomeric chloride resulting from Sj^2 displacement of the sulfur-based leaving group. Prolonged reaction times, or a large excess of I-Cl eventually results in epimerization to the more stable axial
WA««.SR R = Me. Et. Ph
OMe
'^'^?CH20Ac
•o
AcoX««^
CHzOBn 0-
OMe
^"Ofir
BnOCH20Bn \y'?CH20MBn
"
-QACO,
BnO^Si^'^'TCH^OAc
MBnol Br Figure 15.
K. P. RAVINDRANATHAN KARTHAet al.
48
"^^VcHjOAc AcOVi^^ AcO^^
I-Br,MeCN,70%
AcO^ rCHjOH AcO V ^ ^ O M e AcO
I>Br,MeCN.659^
"^^VcHjOAc AcQ W i ^ V O AcO AcO"
AcO ICH2OAC
OMe AcO
AcO^^^>SMe AcO Figure 16,
chloride, although reaction of thioglycosides with I-Cl does provide a practical route to the kinetically favored equatorial anomeric chlorides."*^ Using crossover experiments (Fig. 18) with selectively ^^C-enriched thioglycosides and analysis by mass spectrometry, Boons and Stauch were able to demonstrate that the IDCP-promoted epimerization of thioglycosides is an intermolecular process."^^ As iodine is a milder thioglycoside activator than IDCP, we have been able to directly monitor iodine-promoted thioglycoside activation, epimerization, and glycosylation reactions by ^H NMR spectroscopy."^^ "^ We have been able to show that
I-I
«"?CTl20Bn BnO-S-^^^-X-^SMe BnO
^"?CH,OBn BnO
^"?CH20Bn I-Br
BnOAi,.i.«^
BnoJ^
I-Cl
^"VcHaOBn BnOA--^»^--Cl BnO
Figure 17.
Iodine in Carbohydrate Chemistry
49
CHiOBn
CHjOBn
BnO
BnO
BDCP
CHaOBn
CHjOBn SCH3
BnO
BnO Figure 18.
iodine-promoted thioglycoside epimerization is also an intermolecular process."^^ As with iodine-promoted glycosylation reactions, the rate of epimerization is markedly solvent dependent. The reaction is perhaps ten times faster in acetonitrile than in dichloromethane, suggesting a role for solvent stabilization of a charged, or partially charged, reaction intermediate. If one can find ways to effect Sj^2 reactions of glycosyl halides with oxygen nucleophiles (e.g., sugar alcohols), selective access to both sets of anomeric halides could prove invaluable. Other studies which employ Sj^2 chemistry at glycosidic centers include the halide-assisted glycosylation developed by Lemieux et al.,"*^ the use of a participating solvent such as acetonitrile,"^^ and exploitation of torsional control of anomeric reactivity."^^ This latter approach has been very successfully exploited by Crich and Sun in the synthesis of P-mannosides (Fig. 19)."^^
CHaOBn
©0
CH20Bn
CHjOBn Bnl Bn'
ROH Br
BnO
CHzOBn
CH20Bn BnO BnO
promoter BnOJ BnOJ^
MeCN
^^„
BnOBnO
CH20Bn BnO
BnO
III
C Me
Ph-"-^0
BnO-X*^"^ SPh
PhSOTf
Ph-^O
BnoX-J^ OTf
Figure 19.
ROH
Ph'^O
OBn
3nO-X-J^^OR
50
K. P. RAVINDRANATHAN KARTHAet al.
IX, TUNING DONOR REACTIVITY There is much current interest in tuning the reactivity of donor building blocks for glycoside synthesis. This can be achieved through alteration of the protecting groups on the sugar ring,'^^'*^ or by changing the steric^^ or electronic characteristics^^ of the leaving group. A range of glycosy 1 donor reactivities also requires a corresponding range of promoter reactivities to permit activation of unreactive ("latent") thioglycosides, for instance. There has been some degree of success in the glycosylation of "latent" nitrothiophenyl glycosides with thioethyl glycosides (Fig. 20),^^ but results have not been entirely encouraging to date. Roy and co-workers^^ have investigated the relative reactivity of various thioglycosyl donors having ethyl, phenyl, orpam-substituted phenyl groups with electrondonating (NAc) or electron-withdrawing (NO2) substituents. Comparative studies using diacetone galactose as a standard glycosyl acceptor showed a decrease in donor reactivity from ethyl > phenyl > /7-acetamidophenyl > p-nitrophenyl. In the latter situation, when the thioglycosyl donor was also equipped with "disarming" CH2OBZ ^SEt NIS
BzO
TFOH
CH2OH
BzO
BzO
NO2
NO2
'^7%
Figure 20.
I2 MeOH
BnoS:^^^^ OMe
OBn
R group 4.N02Ph Ph Bn 4-MeOPh Me
BnOv^O^^
OBn Time for complete methanolvsis > 1200 mins - incomplete 260 mins 90 mins 60 mins 20 mins Figure 21.
1 1 1 1 1 1
Iodine in Carbohydrate Chemistry
BnO
51
DBn
MeOH BnO •^-^'•^A-— OBn
\ ^ ^ N 0 2
Promoter
1 ^2 IBr h DDQ h DDQ ICl IBr ICl
Solvent DCM DCM DCM MeCN DCM MeCN MeCN
Promoter
BnoS^J^i
OMe
OBn
Time for complete methanolvsis >12(X) mins - incomplete 1200 mins 900 mins 80 mins 25 mins 14 mins 2 mins
1 1 1 1 1 1 1 1
Figure 22.
ester protecting groups, they were found to be inert or inactive toward common thiophilic promoters. Given that we find iodine to be a relatively mild promoter for thioglycoside activation, we have investigated its reaction with a panel of thioglycosides of varying reactivity (Fig. 21).^^ We also have made a systematic study of the ability of iodine-based promoters to effect activation of a "latent" nitrothiophenyl galactoside in an attempt to ascertain the relative reactivity of the various promoter systems we have studied (Fig. 22).^^ It is clear that the cheap, readily available promoters we have reported vary in their reactivity by several orders of magnitude.
X. IODINE AS A LEWIS ACID As noted at several points in previous sections, iodine-based reagents can act as Lewis acids. This has been known for many years, and has been exploited in the formation^"^ and cleavage^^ of isopropylidene acetals of carbohydrates. In recent work, we have shown^^ that iodine-acetic anhydride is an effective combination for the acetylation or partial acetylation of sugars; this represents a practical alternative to the conmionly used pyridine-acetic anhydride combination. The potential of iodine-based reagents to activate oxygen as well as sulfur centers is clear. A. Activation of Glycosyl Sulfoxides Perhaps the most potent class of glycosyl donor reported to date are the glycosyl sulfoxides, introduced by Yan and Kahne,^ which are typically activated at very low
52
K. P. RAVINDRANATHAN KARTHA et al.
1.^ HO
^ _
^, HO
'
OMe
OMe I
\
ffl ©OMe
Figure 23.
temperature with triflic acid or triflic anhydride. We noted Trost and Miller's work on the iodine-promoted Pummerer reaction (Fig. 23)^^ and we have investigated iodine-promoted glycosyl sulfoxide activation (Fig. 24).^^ Preliminary studies show that iodine-promoted glycosylation of sugar alcohols with a maAzn(9-configured sulfoxide at room temperature gives rise to disaccharides in moderate to good yield (unoptimized). The stereochemical outcome of these
^OJn^
OBn BnO—1-\-|0
i©
room temperature, 2 - 4 h
^*^
ROH = OH BzO-T-^O^ BzO*.X.--r-A
75% a:P 1:2
0„ BnO-T-i^q BnOA^--rA 50% a:p 1:3.3 unoptimised Figure 24.
HO/OBz
1^0
BnOA^-^A^O^j^g BnO 50% a:p 1:4.3
Iodine in Carbohydrate Chemistry
53
reactions is noteworthy since the P-mannosides are the predominant products even though the reactions have been carried out at ambient temperature. Presumably this stereochemical preference arises from at least some Sj^2 character in the displacement of the anomeric a-sulfoxide. B. Synthesis of Thioglycosides and Glycosyl iodides
Thioglycosides are routinely synthesized by the Lewis acid-promoted reaction of a per-acetylated sugar with a mercaptan or its 5-trimethylsilyl derivative.^'^^ We have noted (Fig. 25) that iodine can replace other Lewis acids in both of these procedures.^^ While the iodine-thioalkyltrimethylsilane procedure gives rise to the kinetic thioglycosides (i.e., 1,2-trans for an acetylated sugar) in good yield (typically > 75%), the iodine-mercaptan reaction initially gives the 1,2-rran^-thioglycoside but on prolonged reaction the l,2-c/5'-thioglycoside is obtained. We assume that H-I generated in situ is capable of effecting this epimerization process. The outcome of the iodine-mercaptan reaction is really quite remarkable given that one might have expected disulfide formation and iodine consumption to occur instantaneously. In an attempt to find a cheaper alternative to thioalkyltrimethylsilanes, we have also studied thioglycoside formation from per-acetylated sugars using iodine in combination with hexamethyldisilane and dimethyldisulfide (the latter two reagents are approximately 8 and 500 times cheaper than thioalkyltrimethylsilanes) (Fig. 26).^^ In dichloromethane, the P-thioglycoside is obtained, whereas in acetonitrile, the a-thioglycoside is favored. The active species in the thioglycosideforming reaction remain to be established, but based on observations that iodine in conjunction with HMDS, which generates trimethylsilyl iodide in situ, is an M + EtSH =CiH-I + EtS-I
M + RS-SiMe3
:5::Me3Si-I + RS-I
Q OAc Figure 25.
K. P. RAVINDRANATHAN KARTHAet al.
54
I-I
+ MeaSi-SiMea
: ^ McaSi-I
I-I
+ MeaSi-SiMea : ^ McaSi-I
V^'^AA/^OAC
^
n
^
SMe
I I-I
+ MeS-SMe :;5i MeS-I Figure 26.
effective combination for the conversion of glycosyl acetates to glycosyl iodides,^ both TMS-I and MeS-I could realistically be involved. Interest in glycosyl iodides, which for many years were thought to be too unstable to be useful reagents, has recently been revived and reviewed by Gervay.^^ As noted in Section VIII.B on glycosyl halide formation, selective access to both thioglycoside anomers, and by oxidation the corresponding sulfoxides, could prove invaluable if procedures for Sj^2-type chemistry with oxygen nucleophiles at the anomeric center can be established.
XI. FUTURE PROSPECTS The work described herein focuses on the iodine- and interhalogen-promoted activation of thioglycosides and glycosyl sulfoxides. However, there is every reason to expect that such reagents could find applications in natural product synthesis more generally. Thioacetal and dithioketal chemistry has been used in the synthesis of C-glycosides,^^ Strychnos alkaloids,^^ and brevetoxin^; sulfoxide chemistry has also been exploited in the synthesis of giberellins,^^ and P-lactams.^^ In addition, dithioketals have been used in macrocyclization reactions^^ and thioacetals have been exploited as "electroauxiliaries" for inter- and intramolecular C-C bond formation.^^
ACKNOWLEDGMENTS This paper is dedicated to the memory of Professor R. W. Hay who, in collaboration with Professor R. J. Ferrier, made some early contributions to the use of thioglycosides in
Iodine in Carbohydrate Chemistry glycoside synthesis: Ferrier, R. J.; Hay, R. W.; Vethaviyasar, N. Carbohydrate 55.
55 Res. 1973,27,
REFERENCES 1. 2. 3. 4. 5.
6. 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.
Vsirki, A. Glycobiology 1993, 3, 91. Rudd, P. M.; Dwek, R. A. Grit. Rev. Biochem. Mol. Biol. 1997, 52, 1. Sears, R; Wong, C.-H. Cell. Mol Life Sci. 1998, 54, 223. Drickamer, K.; Taylor, M. E. Trends Biochem. Sci. 1998, 23, 321. For an overview of topical methods in carbohydrate chemistry, readers are referred to: Khan, S. H.; O'Neill, R. A. Modem Methods in Carbohydrate Synthesis; Harwood Academic Publishers: Amsterdam, 1996. Seeberger, R H.; Danishefsky, S. J. Ace. Chem. Res. 1998, 31, 685. Yan, L.; Kahne, D. J. Am. Chem. Soc. 1996,118, 9239. Osbom, H. M. I.; Kahn, T. H. Tetrahedron 1999, 55, 1807. Norberg, T. See Reference 5, Chapter 4. Garegg, R J. Adv. Carbohydr. Chem. Biochem. 1997, 52, 179. Hiskey, R. G.; Tucker, W. R J. Am. Chem. Soc. 1962, 84, 4794. Ratner, S.; Clarke, H. T. J. Am. Chem. Soc. 1937, 59, 200. Bonner, W. A. J. Am. Chem. Soc. 1948, 70, 770, 3491. Reviewed in: Goldschmidt, H. R.; Perlin, A. S. Can J. Chem. 1961, 39, 2025. Yamanoi,T; Nakamura, K.;Sada, S.;Goto, M.; Furusawa,Y;Takano, M.; Fujioka, A.;Yanagihara, K.; Satoh, Y; Hosokawa, H.; Inazu, T. Bull. Chem. Soc. Jpn. 1993, 66, 2617. Fukase, K.; Kinoshita, I.; Kanoh, T; Nakai, Y; Hasuoka, A.; Kusomoto, S. Tetrahedron 1996, 52, 3897. Caddick, S.; Gazzard, L.; Motherwell, W. B.; Wilkinson, J. A. Tetrahedron 1996, 52, 149. Kartha, K. R R.; Aloui, M.; Field, R. A. Tetrahedron Lett. 1996, 37, 5175. Mootoo, D. R.; Konradsson, R; Udodong, U.; Fraser-Reid, B. J. Am. Chem. Soc. 1988,110, 5583. Fraser-Reid, B.; Wu, Z. F; Udodong, U. E.; Ottoson, H. J. Org. Chem. 1990, 55, 6068. Veeneman, G. H.; van Boom, J. H. Tetrahedron Lett. 1990, 31, 275. StHilaire, R M.; Lowary, T. L.; Meldal, M.; Bock, K. J. Am. Chem. Soc. 1998,120, 13312. Kuduk, S. D.; Schwarz, J. B.; Chen, X. T; Glunz, R W; Sames, D.; Ragupathi, G.; Livingston, R O.; Danishefsky, S. J. / Am. Chem. Soc. 1998,120, 12474. Kartha, K. R R.; Ballell, L.; Field, R. A. Unpublished observations. Ham, J.; Choi, K.; Ko, J.; Lee, H.; Jung, M. Protein Peptide Lett. 1998,5, 257 and references cited therein. Wahlstrom, J. L.; Ronald, R. C. J. Org. Chem. 1998, 63, 6021. Raghavan, S.; Kahne, D. J. Am. Chem. Soc. 1993,115, 1580. Ito, Y; Kanie, O.; Ogawa, T. Angew. Chem. Int. Ed. Engl. 1996,35,2510; Kanie, O.; Ito, Y; Ogawa, T J. Am. Chem. Soc. 1994,116,12073. Grice, R; Ley, S. V.; Pietruszka, J.; Osbom, H. M. L; Priepke, H. W. M.; Warriner, S. L. Chem. Eur. J. 1997, 3, 431. Fukase, K.; Yasukochi, T; Nakai, Y; Kusumoto, S. Tetrahedron Lett. 1996, 37, 3343. Mathew, L.; Sankararaman, S. J. Org. Chem. 1993, 58, 7576. Horiuchi, C. A.; Fukunishi, Kajita, M.; Yamaguchi, A.; Kiyomiya, H.; Kiji, S. Chem. Lett. 1991, 192L Aloui, M.; Kartha, K. P. R.; Field, R. A. UnpubUshed observations. Amatore, C ; Jutand, A.; Meyer, G.; Bourhis, R; Machetto, F ; Mallet, J. M.; Sinay, R; Tabeur, C ; Zhang, Y M. J. Appl. Electrochem. 1994, 24, 725; Singy, R Phosphorus, Sulfur Silicon 1994, 95-96, 89.
56
K. P. RAVINDRANATHAN KARTHA et al.
34. Balavoine, G.; Berteina, S.; Gref, A.; Fischer, J.-C; Lubineau, A. / Carbohydr. Chem. 1995,14, nil, 1237. 35. Zhang, Y.-M.; Mallet, J.-M.; Sinay, P. Carbohydr. Res. 1992, 236, 73. 36. Mehta, S.; Pinto, B. M. Carbohydr Res. 1998, 310, 43. 37. CRC Handbook of Chemistry and Physics, 65th ed., CRC Press, 1985; Walker, D.; Hiebert, J. D. C/iem./?^v. 1967,67, 153. 38. Kartha, K. P R.; Aloui, M.; Field, R. A. Tetrahedron Lett. 1996, 37, 8807. 39. White, E. P; Robertson, P W. J. Chem. Soc. 1939, 1509. 40. Kartha, K. P R.; Field, R. A. Tetrahedron Lett. 1997, 38, 8233. 41. Kartha, K. P R.; Cura, P; Aloui, M.; Readman, S. K.; Rutherford, T. J.; Field, R. A. Manuscript in preparation. 42. Boons, G.-J.; Stauch, T. Synlett 1996, 906. 43. Veeneman, G. Ph.D. thesis, Leiden, The Netherlands, 1996. 44. Rutherford, T. J.; Kartha, K. P R.; Readman, S. K.; Cura, P; Field, R. A. Tetrahedron Lett. 1999, 40, 2025. 45. Lemieux, R. U.; Hendriks, K. B.; Stick, R. V.; James, K. J. Am. Chem. Soc. 1975, 97, 4056. 46. Sinay, P Pure Appl. Chem. 1991, 63, 519. 47. Fraser-Reid, B.; Wu, Z. F; Andrews, C. W; Skowronski, E.; Bowen, J. P J. Am. Chem. Soc. 1991, 113, 1434. 48. Crich, D.; Sun, S. X. J. Org. Chem. 1997, 62, 1198; J. Am. Chem. Soc. 1997,119, 11217; J. Am. Chem. Soc. 1998,120, 435. 49. Douglas, N. L.; Ley, S. V; Lucking, U.; Warriner, S. L. J. Chem. Soc, Perkin Trans. 1 1998, 65. 50. Guertsen, R.; Holmes, D. S.; Boons, G.-J. J. Org. Chem. 1997, 62, 8145. 51. Sliedregt, L. A. J. M.; van der Marel, G. A.; van Boom., J. H. Proc. Indian Acad. Sci. 1994,106, 1213; Sliedregt, L. A. J. M.; Zegelaarjaarsveld, K.; van der Marel, G. A.; van Boom, J. H. Synlett 1993, 335. 52. Roy, R.; Andersson, F O.; Letellier, M. Tetrahedron Lett. 1992, 33, 6053; Cao, S. D.; HemandezMateo, F ; Roy, R. J. Carbohydr. Chem. 1998,17, 609. 53. Cura, P.; Aloui, M.; Kartha, K. P. R.; Field, R. A. Manuscript in preparation. 54. Kartha, K. P R. Tetrahedron Utt. 1986, 27, 3415. 55. Szarek, W. A.; Zamojski, A.; Tiwari, K. N.; Ison, E. R. Tetrahedron Lett. 1986, 27, 3827. 56. Kartha, K. P R.; Field, R. A. Tetrahedron 1997, 53, 11753. 57. Trost, B. M.; Miller, C. H. J. Am. Chem. Soc. 1975, 97, 7182. 58. Marsh, S. J.; Field, R. A. Unpublished observations. 59. Kartha, K. P R.; Field, R. A. J. Carbohydr Chem. 1998,17, 693. 60. Kartha, K. P R.; Field, R. A. Carbohydr Utt. 1998, 3, 179. 61. Reviewed in: Gervay, J. In Organic Synthesis : Theory and Applications, Vol. 4; JAI Press, 1998, pp.121-153. 62. Craig, D.; Pennington, M. W; Warner, P Tetrahedron Lett. 1995, 36, 5815; Craig, D.; Payne, A. H.; Warner, P Tetrahedron Lett. 1998, 39, 8325. 63. Amat, M.; Alvarez, M.; Bonjoch, J.; Casamitjana, N.; Gracia, J.; Lavilla, R.; Garcias, X.; Bosch, J. Tetrahedron Lett. 1990,31, 3453; Magnus, P ; Sear, N. L.; Kim, C. S.; Vicker, N. J. Org. Chem. 1992, 57, 70. 64. Nicolaou, K. C ; Duggan, M. E.; Huvang, C.-K. J. Am. Chem. Soc. 1986,108, 2468. 65. Mander, L. N.; Mundill, P H. C. Synthesis 1981, 620. 66. Kaneko, T.; Okamoto, Y; Hatada, K. J. Chem. Soc, Chem. Commun. 1987, 1511. 67. Trost, B. M.; Sato, T. J. Am. Chem. Soc 1985,107, 719. 68. Yoshida, J.-I.; Sugawara, M.; Tatsumi, M.; Kise, N. J, Org. Chem. 1998, 63, 5950.
RECENT ADVANCES IN THE STEREOSELECTIVE SYNTHESIS OF CHIRAL SULFOXIDES
Noureddine Khiar, Inmaculada Fernandez, Ana Alcudia, and Felipe Alcudia
I. Introduction II. Asymmetric Oxidation of Sulfides A. Diastereoselective Oxidations B. Enantioselective Oxidations III. Nucleophilic Substitution on Chiral Sulfur Derivatives A. Sulfite Methodology B. rra«5-2-Phenylcyclohexanol in the Synthesis of Sulfmate Esters C. Aminosulfites D. N-Sulfinyl Oxazolidinones E. DAG (Diacetone-D-glucose) Methodology IV. Sunmiary and Perspectives Acknowledgments References
Advances in Sulfur Chemistry Volume 2, pages 57-115. Copyright © 2000 by JAI Press Inc. All rights of reproduction in any form reserved. ISBN: 0-7623-0618-1 57
58 60 60 61 77 78 81 84 85 89 109 110 110
58
NOUREDDINE KHIAR et al.
I. INTRODUCTION In the ever-expanding field of asymmetric synthesis, the development of new and efficient chiral transformations is a continuous challenge for organic chemists. Conceptually, an ideal chiral reaction is one able to give both enantiomers of a molecule of interest, with high enantioselectivity and with a minimal change in the synthetic procedure. The interest in both isomers of a chiral molecule is dictated either by their biological properties or by their suitability as chiral controller. The sophistication in asymmetric synthesis has reached a point where good chiral auxiliaries are not those which can be used in a single stereocontrolled transformation, but rather those able to induce complete stereocontrol in a plethora of chemical transformations. Chiral sulfoxides developed at the beginning of the 1980s are acquiring renewed interest at the end of the 1990s as they respond to the latter. Sulfoxides have been used in the control of numerous asymmetric reactions, including Michael addition,^ aldol condensation,^ addition to imines,^ carbonyl reduction,"* radical addition,^ the Heck reaction,^ and the Diels-Alder reaction.^ Recently, chiral sulfoxides have also been shown to be good ligands for metal-catalyzed enantioselective reactions such as the catalytic Diels-Alder reaction,^ addition of organozinc to aldehydes^ and to vinyl sulfoxides,^^ or in Pd-catalyzed allylic substitution.^^ The effectiveness of the sulfinyl group as chiral controller is a result of two factors: 1. The large stereoelectronic differences between the three substituents at the sulfinyl sulfur: the lone pair of electrons, the oxygen atom, and two alkyl or aryl groups 2. The high optical stability of chiral sulfoxides'^ The activation parameters of the pyramidal inversion have been determined for various dialkyl, diaryl, and alkyl aryl sulfoxides. They vary between 35 and 42 kcal/mol for A//*, and - 8 and +4 cal/(molK) for A5^. These values indicate that, in most cases, the thermal stereomutation of sulfoxides occurs at a significant rate only at about 200 °C. There are a few exceptions, such as benzyl and allyl sulfoxides, whose racemization is raised at 130-150 and 50-70 °C, respectively. Additionally, a number of biologically significant molecules have in their structure a chiral nonracemic sulfinyl group. Among these molecules (Fig. 1), it is worth noting the new immunosuppressive thiaspirane sulfoxide Nuphar alkaloid type l a and lb,'"' the gastric antisecretory omeprazole 2,'^* the new potassium channel activators of Rhone Poulenc aprikalim 3,'^ the cyclic hexapeptide waiakeamide 4,'^ the ACAT inhibitor 5,'^ and the potent human immunodeficiency virus type 1 protease inhibitor 6}^ The development of new and efficient routes to chiral nonracemic sulfoxides with high enantiomeric purity has been a subject of constant interest over the last two
Synthesis of Chiral Sulfoxides
59 OMe
OMe
H Me"
1b
o^/:r ^N
6
V
ou.-~^ o
Figure 1. Structure of biologically active sulfoxides.
decades. There are two main methodologies (excluding resolution) for the preparation of chiral sulfoxides: 1. Asymmetric oxidation of prochiral sulfides 2. Nucleophilic substitution on chiral sulfur derivatives (the so-called Andersen method) The purpose of this article is to present recent developments in the preparation of optically pure sulfoxides using both methods, mainly from 1990 to the present. Emphasis has been given to the bibliographic impact of each method. An application section is included after each route, especially in the case of variation in the Andersen methodology, where important advances have been achieved. It is not the aim of this article to review the chemistry of chiral sulfoxides—several excellent review articles have appeared on this subject, from the seminal review by Solladie^^ in 1981 to other recent reviews.^^ The literature has been surveyed up to January 1999. The preparation and utilization of chiral sulfoxides in asymmetric synthesis have been the subject of valuable comprehensive as well as specialized accounts which should be consulted for details and considered as complementary to this article.
60
NOUREDDINE KHIAR et al.
IL ASYMMETRIC OXIDATION OF SULFIDES Asymmetric oxidation of sulfides to sulfoxides is undoubtedly the most straightforward route to optically pure (o.p.) sulfoxides. It is thus not surprising that various groups have been trying to develop efficient methods toward this end. Several o.p. sulfoxides have been obtained, by either diastereoselective or enantioselective oxidation of sulfides. A. Diastereoselective Oxidations Most reports on diastereoselective oxidation of sulfides are substrate-directed. Diastereoselectivity has been achieved by either steric- or neighboring-group participation.^^ Incipient hydrogen bonding between the substrate hydroxyl group and the incoming percarboxylic acid has been evoked to explain the high diastereoselectivity observed in the oxidation of 10-^jcf?-hydroxy-bornyl- derivatives 7 and 9 (Scheme 1). The oxidation of 9 with m-CPBA in MeOH occurs without stereoselectivity. Type-10 sulfoxides have been shown to be good chiral dienophiles, and chiral acceptors in asymmetric Diels-Alder and Michael addition reactions, respectively. The same results have been obtained recently in the oxidation of phenyl thiogalactopyranosides 12 (Scheme 2). The level of the diastereoselective oxidation by m-CPBA has been shown to be dependent mainly on the substituents at C-2.^^ These sulfoxides have been prepared in order to study the effect of the chirality at the sulfinyl in the acid-catalyzed glycosidation reaction, using sulfinyl glycosides as glycosyl donors. While a modest diastereoselective cleavage has been observed in
mCPBA
/-BuO
S
f-BuCf
mOPBA ^ CH2CI2
MeO
Scheme 1,
Synthesis of Chiral Sulfoxides
61
T l
mCPBA
11
0
OH
OH
12 Ri, Ra - -CMe2-, R3 - Ac R2, R3 «-CMe2-, Ri « H
13
Scheme 2.
the hydrolysis with triflic acid, total diastereoselection has been achieved in the reaction with P-galactosidase. (5s)-sulfinyl galactopyranoside is completely hydrolyzed by the enzyme, whereas the (/?s)-diastereoisomer is recovered optically pure. Using the stereodirecting effect shown by the proton of an amide function, several authors have prepared various o.p. sulfoxides with high biological interest, such as the P-lactamic compounds, penicillin and cephalosporin^^ (Table 1). Steric effects were responsible for the complete diastereocontrol observed in the oxidation of various 6-halopenicillins by dimethyl dioxirane (DMD). Only one of the two possible diastereomeric sulfoxides has been obtained in each case^^ (Table 1). Scheme 3 shows that perborate oxidation of optically active sulfide 16 affords, with moderate diastereoselectivity (78% de), the (/?)-sulfoxide Yl}^ designed as chiral ligand for catalytic asymmetric synthesis. B. Enantioselective Oxidations The development of reactions for the enantioselective oxidation of prochiral sulfides is a formidable synthetic challenge, as sulfides are examples of nonfunctionalized substrate. Lacking functional groups, these compounds are unable to
Sulfoxidation of P-Lactam Compounds 29
Table / .
R3 H s ^ ^CH3 CH3 R22 44«- _- p^ S
R3 H p RR22<S - ^pSs ^ . C. H 3
(O) (O)
' 'CH3 502R1
C02R1 14 Compound
Rj
14a 14b 14c
H H Bn
15 R2
PhOCH2CONH PhOCH2CONH Br
Note: ^DMD, dimethyl dioxirane.
R3
(Oxidant)
H H F
H2O2 K2S2O8 DMD^
Yield (%)
S/R
90 S(only) 85 20/1 quantitative R{on\y)
62
NOUREDDINE KHIAR et al.
a
Me
NMe2 S-^®
16
Me
NaB03.4H20 AcOH
f^Y^^^^^ ' ^ ^Os - M e 17
Scheme 3.
coordinate with the oxidant to form the highly ordered and rigid transition-state geometries which are a prerequisite for most enantioselective transformations that occur with high diastereoselectivity. In nonfunctionalized substrate, it is necessary to use an enzymatic approach—or to develop a reagent that acts in the same manner as enzymes—in order to control the molecular recognition. Both approaches have been used to oxidize prochiral sulfides efficiently. Biological Sulfoxidation Impressive advances have been made in biological sulfoxidation in the last decade.^^ The results augur that this approach will be of some synthetic interest in the future, instead of being an intellectual curiosity. Both isolated enzymes and whole cells have been used in the enantioselective oxidation of prochiral sulfides. Enzyme-catalyzed sulfoxidation. The first use of isolated enzyme in the enantioselective oxidation of prochiral sulfides was by Light et al.^^ at the beginning of the 1980s. Both enantiomers of ethyl p-tolyl sulfoxide were obtained using pig liver FAD-dependent mono-oxygenase, purified P-450 isozyme, and flavin-containing cyclohexanone mono-oxygenase from Acinetobacter. More recently, Colonna extended the application of cyclohexanone mono-oxygenase from Acinetobacter calcoaceticus (CYMO) to the biosulfoxidation of other sulfides. Several works of Colonna's group^^ have shown that the mono-oxygenase is the most effective general enzyme for the enantioselective synthesis of sulfoxides. An interesting feature of the Italian approach is the use of a second enzyme to regenerate NADPH, thereby allowing the use of NADPH in catalytic quantity, lowering the cost of the operation. The stereochemical outcome of the enzymatic reactions has been shown to be highly dependent on the sulfide structure. Thus, for alkyl aryl sulfides, the optical purity of the products ranges from 99% ee for the (/?)-methyl phenyl sulfoxides to 93% ee for (5)-ethyl p-fluorophenyl sulfoxides. The method is particularly suitable for the preparation of (R)-tert-h\My\ methyl sulfoxide (99% ee), as well as for the enantioselective oxidation of dithioacetals (Table 2). Another important class of enzymes able to catalyze sulfoxidation is the peroxidases. Of the peroxidases developed to date, the most versatile is a chloroperoxidase from Caldariomycesfumago (CPO), isolated in 1961 by Shaw and Hager,^^ and used for the first time in enantioselective sulfoxidation by Kobayashi et al.,^^ although in low
Synthesis of Chiral Sulfoxides
63
Table 2, Enantioselective CYMO-Catalyzed Oxidation of Sulfides Sulfide PhSMe p-FC6H4SMe PhSEt PhS-/-Pr f-Bu-S-Me
Q r\
Q Me
Me
Yield (%)
ee (%)
Sulfoxide Configuration
88 91 86 93 98
99 92 47 3 99
R R R S R
81^
>98
R
94a
>98
R
92a
>98
R
Note: ^Monosulfoxides.
ee (13%). A detailed study by Colonna's group^^ of CPO-catalyzed enantioselective sulfoxidation showed that H2O2 is the best oxidant, promoting the synthesis of o.p. sulfoxides with high yield and selectivity. However, substantial uncatalyzed oxidation of the sulfide (10-30%) was observed, lowering the enantioselectivity. Recently, Sheldon's group^^ has reported experimental conditions with no uncatalyzed reaction, and enantiopure sulfoxides were obtained in water as well as in tert-butyl alcohol/water mixture (Table 3). Wong's group-'^ investigated the chloroperoxidase-catalyzed oxidation of p-substituted alkyl phenyl sulfides by hydrogen peroxide or racemic alkyl hydroperoxides as oxidant in aqueous buffer. Slow addition of H2O2 to the reaction mixture afforded nearly enantiopure sulfoxides (97-99% ee). Interestingly, when racemic alkyl hydroperoxides were used as oxidant, optically active alcohols and alkyl hydroperoxides were obtained (Scheme 4). Recently, a vanadium-containing bromoperoxidase (VBrPO), from the alga Corallina officinalis, has been shown to catalyze the stereoselective oxidation of some aromatic bicyclic sulfides in high ee (up to 91%).^^ The enantioselectivity observed was not the result of a kinetic resolution, as no overoxidation to sulfone was detected. The VBrPO, which does not oxidize methyl p-tolyl sulfide, has the interesting characteristic of producing the (5)-bicyclic sulfoxide 19, with the opposite stereochemistry to that obtained with heme-containing chloroperoxidase (CPO) from Caldariomycesfumago (Scheme 5).
64
NOUREDDINEKHIARetal. Table 3. Oxidation of Sulfides by CPO and H202^^ t-BuOH/Buffer (50:50,v/v)
Sulfide
Conversion Entry 1 2 3 4 5 6 7 8 9 10 11
^1
Ph Ph Ph p-Tol p-MeOC6H4 m-MeOCeH4 o-MeOC6H4 />BrC6H4 m-BrC6H4 2-ThJenyl 2-(1,3-Thiazolyl)
^2
Me Et Pr Me Me Me Me Me Me Me Me
(%) 73 52 1 66 50 19 2 46 22 91 80
Buffer Conversion
eer%; 99 99 60 99 99 99 99 99 99 99 99
(%) 100 83 3 83 53 37 3 15 11 100 100
ee (%) 99 99 27 99 99 99 99 99 99 99 99
Various other heme-peroxidases were found to catalyze the enantioselective sulfoxidation of alkyl aryl sulfides. These included horseradish peroxidase (HRP),"^"*'^^ cytochrome c peroxidase (CcP),^^ microsome peroxidase (MP),^^ lactoperoxidase (LPO),^^ and dioxygenase.^^ However, their turnover numbers (TON) and enantioselectivities were much lower than those observed with CPO (Table 4). Microbiological oxidation. Biological oxidation using whole cells has employed mainly fungi, and strains in the oxidation of sulfides to sulfoxides.
Scheme 4.
Synthesis of Chiral Sulfoxides
65 O I
CO ^=^ 00 18
19
enzyme CPO: quantltave, 99% ee, R enzyme VBrPO: 99%, 90% ee, S Scheme 5.
While the enantioselectivities achieved with microorganisms are less spectacular than those with isolated enzymes, some of these approaches give specific sulfoxides with high ee (above 80%). Fungi such as Helminthosporium^ and Mortierella isabellina^^ have been shown to oxidize phenyl and benzyl alkyl sulfides, /7-alkylbenzyl methyl sulfide, and isocyanate sulfides in a complementary manner. In the oxidation of methyl aryl sulfides, Helminthosporium gives the S-sulfoxide as the major isomer, while M. isabellina produces the R enantiomer"^^ with modest to good ee and chemical yields. Holland's group"^ has shown that the most versatile biotransformation using whole cell biocatalyst is the one using the fungus species NRRL 4671. From analysis of the sulfoxidation of a large number of substrates (> 90), they recently proposed a predictive model for chiral sulfoxidation by the fungus. The model (Fig. 2), developed from energy-minimized (MM"*") structures of substrates produced by Hyperchem, is able to explain the stereochemical inversion seen for sulfoxidation of some phenyl alkyl sulfides, such as phenyl vinyl and phenyl hexyl sulfide. Baker's yeast (Saccharomyces cereviseae NRC 2335) wasfirstused by Buist and Marecak"*^ in the enantioselective oxidation of fatty acid analogues with 70% ee. A later work by Roberts and co-workers,"^ using S. cereviseae NCYC 73, succeeded
Table 4. Comparative Values of Peroxidases for the Oxidation of Methyl Phenyl Sulfide to Sulfoxide Enzyme CPO^ HRP*' LPO^
MP-ir
Reaction Time (min) 60 60 105 45
Yield (%) 100 95 40 45
ee (%) (ConfigJ
TON^
98 (R) 46 (S) 52 (R) 3(S)
6.3x10^ 29 57 3
Notes: *rON (turnover number), mole of product produced per mole of enzyme used. ^C?0, chloroperoxidase. ^R?, horseradish peroxidase. *'LP0, lactoperoxidase. ®MP, microsome peroxidase.
66
NOUREDDINE KHIAR et al, Table 5. Biotransformation of Sulfides, RTS-R2, to Sulfoxides with Helminthosporiuni^^ /?2
/?! Ph P-Br-C6H4 P-NC-C6H4 p-MeO-C6H4 p-MeS-C6H4 P-BU-C6H4CH2 p-(/-Pr)CeH4CH2 P-CIC6H4CH2 P-O2N-C6H4CH2 p-MeO-C6H4CH2
Et Me Me Me Me Me Me Me Me Me
Yield (%) 40 69 80 83 64 74 77 71 95 86
ee (%)
Config. at S
84 90 92 80 80 90 80 90 92 80
S S S S S S S S S S
in the oxidation of methyl p-tolyl sulfide with 92% ee and 60% yield. In 1995, Allen et al."^^ reported the microbial oxidation of aryl alkyl and diary 1 sulfides to o.p. sulfoxides by selected strains of the bacterium Pseudomonas putida UV4 to give /^-sulfoxides with high ee, while P. putida NCIMB 8859 preferentially produced 5-sulfoxides. Chemical Enantioselective Sulfoxidation
The importance of the enantioselective chemical oxidation of sulfides has long been known. Nevertheless, it was not until the early 1980s that various approaches began to be developed simultaneously. Until very recently, two methods were used in the oxidation of sulfides'^^''^^: those based on the modified Sharpless asymmetric epoxidation,"*^ and those based on chiral oxaziridines."*^ While these methods lead
a: binding leading to (S) -phenyl ethyl sulfoxide, 84% ee. b: binding leading to (R)-pheny\ hexyl sulfoxide, 25 % ee. Figure 2. Model for sulfoxidation by Helminthosporium.
Synthesis ofChiral Sulfoxides
67
to high enantioselectivities, the reactions suffer from the use of stoichiometric amounts of the chiral auxiliary and Ti(0-/-Pr)4. Recent advances aim to solve this problem by developing efficient catalytic reactions. Various catalytic systems have been reported, based on the use of (salen) manganese, (salen) vanadium, and titanium complexes, with only moderate success. Oxidation in the presence ofctiiral titanium tartrate (modified Sharpless method). Inspired by the Sharpless asymmetric epoxidation^^ of allylic alcohols with hydroperoxides in the presence of chiral titanium complex [diethyl tartrate (DET) and Ti(0-/-Pr)J, Kagan and co-workers'*^ and Modena and co-workers'*^ developed almost at the same time two variations of this reaction leading to o.p. sulfoxides with high enantiomeric purity. The Orsay system. A good example of serendipity is the discovery by Kagan and co-workers'*^ at Orsay that 1 mol of water was necessary to produce the active catalyst able to oxidize prochiral sulfides to sulfoxides with high ee. Optimization of the stoichiometry of the titanium complex permitted the determination of the combination Ti(0'i'Fv)J(R,R )-DET/H20 (1/ 2/ 1) at -20 °C in CH2CI2 as the optimal conditions to achieve high enantioselectivity. Table 6 shows some representative results obtained for the oxidation of several thioethers with tert-butyl hydroperoxide (TBHP) under these conditions.^^'^^ The Orsay group continued working intensively on the optimization of their system to make it catalytic without losing the enantioselectivity of the stoichiomet-
Table 6, Asymmetric Oxidation of Sulfide Ar-S-R by r-BuOOH in the Presence of Ti(0-/-Pr)V(+)-DET/H20 in a 1:2:1 Ratio Isolated Yield Entry
Ar
R
(%)
1 2 3 4 5 6 7 8 9 10 11 12 13 14
p-Tolyl p-Tolyl p-Tolyl 1-Naphthyl 2-Naphthyl 2-Naphthyl 9-Anthracenyl o-Tolyl p-MeOC6H4 o-MeOC6H4 Phenyl Phenyl Phenyl 2-Pyridyl
Methyl Ethyl n-Butyl Methyl Methyl n-Propyl Methyl Methyl Methyl Methyl c-Propyl CH2CI CH2CN Methyl
90 71 75 98 88 78 33 77 71 70 73 60 85 63
ee {%)
Ref.
89 74 75 89 90 24 86 89 86 84 95 47 34 77
50 50 50 51 50 50 51 51 50 50 51 51 51 50
68
NOUREDDINE KHIAR et al.
ric condition. The most improvement achieved was the change of TBHP with cumene hydroperoxide (CHP). Not only could the amount of titanium complex be reduced to 0.2 mol equiv, but the asymmetric induction was significantly increased for all of the sulfides^^ (Table 7). The exact mechanism of enantioselective oxidation of sulfide with the watermodified Sharpless catalyst is still unknown. However, there is a very good correlation between absolute configuration of tartrate and the sulfoxides formed. Thus, Kagan and Rebiere proposed that the absolute configuration as R for the sulfoxide formed can be made by taking L and S groups as large and small, respectively^^ (Scheme 6). Figure 3 shows the transition state proposed by Kagan and co-workers^^ for methyl phenyl sulfide oxidation. A recent systematic study of the role played by titanium alkoxide, 2-propanol, and molecular sieves (MS) has permitted the development of an efficient catalytic system furnishing chiral sulfoxides with high ee.^"* This catalyst has a new composition: Ti(0-/-Pr)4/(/?,/?)-DET//-PrOH (1 / 4 / 4), in the presence of 4A MS, which is a combination of the Modena"*^ and Sharpless systems. Using this new system, the Orsay group achieved the highest enantioselectivity in catalytic asymmetric oxidation of sulfides by a nonenzymatic method (Table 8). A tentative catalytic cycle (Scheme 7) has been proposed for the oxidation of sulfides with the new system. In this mechanism, the active species is the monomeric titanium complex 21, formed from the dimer titanium compound 20 by the acdon of 2-propanol. The alcohol also has a beneficial effect by displacing the sulfoxide formed, inducing the formation of 21, thereby permitting the catalytic cycle.^"* Complexes based on titanium excess tartrate combination (the Padova system). In 1984, the same year the Orsay group developed their system, a group in Padova, Italy, headed by Modena,"^^ developed a different system, able to oxidize sulfides to sulfoxides with high selectivity, also based on a modification of the Sharpless catalyst. The Padova group used TBHP in the presence of 1 mol equiv of Ti(0-i-FT)^/(R,R )-DET, 1/4 combination. The reactions were performed at -20
Table 7. Enantiomeric Excess (%) in Asymmetric Oxidation of Sulfides by Various Hydroperoxides in the Presence of Ti(0-/-Pr)V(/?,/?)-DET/H20 in a 1:2:1 Ratio Sulfide Me-S-(p-tolyl) Me-S-Co-anisyl) Me-S-phenyl Me-S-(n-octyl) Me-S-benzyl
Cumene Hydroperoxide
t-BuOOH
Ph^COOH
96 93 93 80 61
89 74 88 53 35
16
32
Synthesis ofChiral Sulfoxides
Large'^ \ma\\ (L) (S)
69
(^^-DET
Large^ "^Small (L) (S)
(L)=Ar. (S)=Alkyl Scheme 6.
°C in toluene or 1,2-dichloroethane.'*^ The results were similar in both yield and selectivity to those obtained using the water-modified reagent (Table 9). It was hypothesized from the beginning that identical species could be involved in the two systems. The role of an exce^ of tartrate was to bring an uncontrolled amount of water into the system, based on the observation that the Padova system produces racemic sulfoxides by addition of molecular sieves. The procedure with an excess of DET gives better results in the asynmietric oxidation of 1,3-dithiolane (Table 9). This latter finding has been elegantly applied by the same group to the resolution of racemic ketones, through their transformation into 1,3-dithiolanes, asymmetric monosulfoxidation followed by diastereomeric separation, and regeneration of the parent ketone.^^ Some applications. Most of the groups who have used the modified Sharpless oxidation for the preparation of a desired sulfoxide have preferred to use the water-modified titanium system. Page and co-workers^^^ found, during their investigation for the synthesis of a chiral acyl anion equivalent 1,3-dithiane-l-oxide, that the introduction of a polar group in the 2 position of 1,3-dithiane was responsible for the high enantiomeric excess (99% ee) observed. In this regard it is interesting to note that Aggarwal's group has found that the Modena modification was superior to Kagan's for the enantioselective oxidation of the 2-carbethoxy-l,3-dithiane.^^^ Davis et al.^^ prepared both enantiomers of p-anisyl methyl sulfoxide 24 by oxidation with cumene hydroperoxide in the presence of the titanium complex, and
Figure 3, Transition state of Kagan sulfoxidation.
70
NOUREDDINE KHIAR et al. Table 8. Asymmetric Oxidation of Sulfides (R1SR2) by CHP in the Presence of Ti(0-/-Pr)4/f/?,/?;-DET//-PrOH, 1:4:4, and Molecular Sieves
Entry
^1
R2
1 2 3 4 5 6 7 8 9 10 11
Phenyl p-Tolyl p-Anisyl o-Anisyl o-Nitrophenyl Phenyl p-Tolyl p-Tolyl o-Anisyl Benzyl n-Octyl
Me Me Me Me Me CH=CH2 Et n-Butyl Phenyl Me Me
Yield (%)
Bio-^--r°^ ,Ti(0APr)2
Boy^o'
20 UprOH
EtO,^^
EtO ^ > ?
HOAPr
O ^O^pOAPr OhPr
R(X)H
21
R/
^RU
ROH 2/-PrOH
2 /-PrOH
OAPr 22
ee (%) (R) 91.2 95.6 92.1 89.3 75.0 55.4 78.1 25.0 6.2 90.3 70.7
81 77 73 72 51 58 68 70 64 72 69
Oi-Pr 5 . Rs' " L 23
Synthesis of Chiral Sulfoxides
71
Table 9. Asymmetric Oxidation of Sulfides by TBHP in the Presence of Ti(0-/-Pr)4/(+)-DET in a 1:4 Ratio Sulfide
Yield (%)
p-Tol-S-Me p-Tol-S-f-Bu P-CIC6H4-S-CH2CH2OH PhCH2-S-Me 2-Ph-2-Me-1,3-dithiolane 2-Ph-1,3-dithiolane 2-f-Bu-l,3-dithiolane 2-f-Bu-2-Me-l ,3-dithiolane
60 99 41 70 66^ 76' 82^ 61^
Diastereomer Ratio
ee (%)
97/3 94/6 99/1 99/1
88 34 14 46 83 76 70 68
Note: ^Monooxidation.
used them as chiral auxiliary and a building block for the synthesis of a cardiovascular drug 25 (Scheme 8). Thomas and co-workers^^ have described the oxidation of methylthio-substituted tricarbonyl-(Ti^-arene)chromium(0) complex 26 with CHP in the presence of Ti(0-/-Pr)4/DET/H20 combination to give the corresponding sulfoxide in 9 0 95% ee, in a similar approach to that used by Kagan^^ for the oxidation of aryl ferrocenyl sulfide 28 (Scheme 9). With regard to the effort to render the system catalytic, it is worth noting the work of Uemura,^^ who has succeeded in developing a catalytic system using /?-(+)binaphthol 30 instead of DET as chiral source. The highest ee's were obtained with TBHP at 25 °C and 2 mol% of the catalyst, and sulfoxide was obtained with R absolute configuration and up to 96% ee (Scheme 10).
MeO
^ ^
MeO^^^^ 24
MeO"^==^^
OMe
/
Scheme 8.
72
NOUREDDINE KHIAR et al. Me R - ^ ^ ^ S '
^ CHP.Ti(0/Pr)4.W-DET/H20
Cr(C0)3 2g
27 R - H. Me, MeO
vSy Fe
28
V. •
CHP. Ti{0/Pr)4. (+)-DET/ HjO ^
V"* Fe
2^
R«Me. Ph. ;Molyl
^'<^^>^ yield « 60-73% ee > 90-95%
\
yield « 78-86% ^^^gg,/^
Scheme 9.
Metallo-(salen)'Catalyzed oxidation. In 1986, Pasini and co-workers^^ developed chiral oxotitanium(IV)-Schiff base complex as catalyst for the oxidation of methyl phenyl sulfide. While the catalytic activity of the Pasini system was excellent (catalyst : substrate ratio, 1:1000 to 1:1500), the enantioselection was unfortunately low (< 20% ee) for catalyst 31 (Fig. 4). In the same year, Fujita's group^-' reported the asymmetric oxidation of aryl methyl sulfide by hydroperoxides (TBHP, CHP) and an optically active catalyst formed by a Schiff base-oxovanadium(IV) complex 32, giving (5)-sulfoxides in low ee (up to 40%) (Fig. 4). Later, they developed^ a more promising approach using 33, a binuclear Schiff base-titanium(IV) complex (4 mol% equiv) to catalyze the asymmetric oxidation of methyl phenyl sulfide by trityl hydroperoxide in methanol at 0 °C. The (/?)-methyl phenyl sulfoxide was obtained with 60% ee. Using the clear homology of epoxidation of olefin and the oxidation of sulfide, Jacobsen and co-workers^^ and Katsuki and co-workers^^'^^ applied their system developed for the asymmetric epoxidation of simple olefin to the asynmietric oxidation of prochiral sulfides.
^QrgH(0.2eq) f-BuOOH (2 eg). ^^<^ Me^ ^Ar
30/?
ri(OAPr)4(0.1 eq), H2O(2.0eq)
Scheme 10.
^5^ ^^ ^^ yield « 32-88% ee « up to 96%
Synthesis of Chiral Sulfoxides R.
73
...R
r=K^H=, .
a-_ « .
R » Me. Ph
0-°NL°-W
31 31
O^ 32
\
/
R s OMe, OEt, f-Bu
a
f-Bu
V C . : N ^ - ^
CP55
OH N^^f-Bu 33
Figure 4. Chiral Schiff bases used in catalytic asymmetric sulfoxidation.
In 1992, Jacobsen's group^^ reported on the use of chiral (salen) manganese complex for the oxidation of sulfide with H2O2. Unlike the excellent results obtained with this system in the epoxidation of conjugated olefin, only moderate enantioselectivity has been obtained for oxidation of alkyl aryl sulfides. A new (salen) manganese(III) complex 35 was synthesized,^^ which showed better catalytic asymmetric induction in the oxidation of sulfides (up to 90% ee) with PhIO instead of the H2O2 used by Jacobsen. The asymmetric induction, which is better for aryl sulfides with an electron-withdrawing group than with an electron-donating one, has been shown not to proceed via a kinetic resolution, but by overoxidation of the formed sulfoxide (Scheme 11).
./Ph / ^ ^^
Me
^^,^^^ ^ PhlO(1eq.
Ar« 0-O2NC6H4, o-BrC6H4, p-MeOC6H4 Scheme 11.
Me (0.01 eq.) Ar
^Me
yield « 45-74% ee « 40-90%
NOUREDDINE KHIAR et al.
74
Recently, Bolm and Bienewald^^ reported on an efficient catalytic system formed in situ from VO(acac)2 and ligand type 34 (Fig. 4). This system, which is able to catalyze sulfoxide formation in a concentration of 0.01 mol%, has the advantage of proceeding under simple reaction conditions. Unfortunately, the enantioselectivity is in general moderate (53-70% ee), with the exception of the 2-phenyl-l,3dithiane, where only the trans isomer was obtained in 85% ee. The main methodologies developed until now for enantioselective oxidation of sulfides are effective only in the oxidation of alkyl aryl sulfoxides. Dialkyl sulfoxides on the other hand are generally oxidized with only poor selectivity. In an attempt to solve this problem, Schenk's group^^ recently reported a stereoselective oxidation of metal-coordinated thioethers with DMD. The prochiral thioether is first coordinated to a chiral ruthenium complex by reaction with the chloride complexes [CpRu[(5,5)-chiraphos]Cl], 36. Diastereoselective oxygen transfer from DMD produces the corresponding sulfoxides in high yield and selectivity. The chiral sulfoxides 37 are liberated from the complexes by treatment with sodium iodide. Several o.p. aryl methyl sulfoxides have been obtained by this method in moderate to high ee (Scheme 12). The authors have recently applied this methodology to the synthesis of (/?)- and (5)-sulforafane^^ 40, although it suffers from difficulties in the experimental procedure and from the use of expensive chiral auxiliaries in stoichiometric amount (Scheme 13).
Phg I P—Ru^^,
RSMe NH4PF6
Phz
A-
36
Phz I O P—Ru-^l...>Me
PFs"
37 Scheme 12.
PF6
Synthesis of Chiral Sulfoxides
75
DMD. acetone 0"C
iRul
Ru 38
NCS 40 (sulforaphane) Scheme 13.
Oxidation by chiral oxaziridines. For more than a decade, D a v i s ' s group"*^'^^"^^ has been working on the stoichiometric asymmetric oxidation of prochiral sulfides. In a series of elegant and important papers, they have demonstrated that their approach is one of the best methods in the synthesis of chiral sulfoxides. This research has yielded four generations of chiral oxaziridines 4 1 - 44 exhibiting different stereoselectivities as a result of their dissimilar active-site structures (Fig. 5). The most effective and general (camphorylsulfonyl)oxaziridine developed to date for the asymmetric oxidation of sulfide is 44. This oxaziridine gives a large number of sulfoxides not only in high enantioselectivity, but also in a predictable manner. The oxidation of sulfide is generally conducted in CCI4 or CH2CI2 at 20 °C by treatment with 1 equiv of the oxaziridine (Table 10). The highest ee's (> 90%) were observed for those sulfides in which the groups (R^-S-Rs) were sterically very different.^^ As with Kagan's reagent, a group size difference effect is observed for the oxidation of Ar-S-R. Although Davis and Kagan reagents gave similar results for aryl alkyl sulfides, the former is generally better for the oxidation of dialkyl sulfides.^^
•°- .Ar , < N - \ H
41 Figure 5.
42
43
44
Chiral oxaziridines used in the asymmetric oxidation of sulfides.
NOUREDDINE KHIAR et al.
76
Table 10. Asymmetric Sulfoxidation of Sulfides (R1SR2) with Oxaziridine 44 R^
^^2
p-Tolyl p-Anisyl p-Tolyl c-Propyl 2-Naphthyl 9-Anthranyl 9-Anthranyl Me PhCH2 Me
Me Me PhCH2
Con fig. at S
Yield (%) 95 74 94 90 84 90 60 84 80 60
Ph Me Me /-Pr t-Bu f-Bu n-octyl
S S S S S S S S S S
ee (%) >95 80 88 92 94 95 94 94 94 45
An active-site model has been proposed to explain the high asymmetric oxidation of sulfide to sulfoxides'^ (Fig. 6). The model consists of three pockets, A, B, and C, where pocket B, defined by the two chlorine atoms and the phenylsulfonyl group, is responsible for the high enantioselectivity exhibited for the oxidation of sulfides RL-S-RS- The absolute stereochemistry of the final sulfoxides is predicted in terms of a simple steric model, which involves minimization of nonbonded interaction between the RL and Rg groups of the sulfides (RL-S-RJ) and the active site surface of the oxaziridine in an orientative planar transition state. Sulfoxidation by chiral peroxides. The first attempts to use enantioselective hydrogen transfer from optically active hydroperoxides derived from sugars in the asymmetric oxidation of prochiral sulfides appeared in 1997.^' Hydroperoxides 45 and 46 were obtained in 65 and 75% yield from the corresponding a-methyl glycosides by oxidation with hydrogen peroxide. Asymmetric oxidation of methyl phenyl and methyl p-tolyl sulfides gave the corresponding sulfoxides in a modest 25% ee. Recently, Adam et al.'^ have reported the use of enantiomerically pure hydroperoxides such as 47- 49 as source in oxygen transfer to alky I aryl sulfides. These hydroperoxides were obtained easily by HRP-catalyzed kinetic resolution.'^
C
A
•• 1 ^«
V
^
Figure 6. Top view of active-site model for (+)-44.
Synthesis of Chiral Sulfoxides
77
BnO^ OOH
47
OOH
48
49
50H
45: R^ « OBn. Rg = H 46: Ri « H. Rg « OBn
Figure 7. Chiral hydroperoxides used in the asymmetric oxidation of sulfides.
The best results were obtained with (5)-(-)-phenylethyl hydroperoxide 47 at -20 °C in CCI4, which afforded (5)-sulfoxides with low to modest enantioselectivity and low yield. A time profile of the oxidation of methyl p-tolyl sulfide with 47 showed that the asymmetric induction in the sulfoxidation was rather low (< 20%), demonstrating that the enantioselectivity obtained is related to a concomitant kinetic resolution of the sulfoxide formed.
III. NUCLEOPHILIC SUBSTITUTION ON CHIRAL SULFUR DERIVATIVES Undoubtedly, the most widely used method for the synthesis of o.p. sulfoxides is the nucleophilic addition of metal organic reagent to an electrophilic sulfur with preestablished chirality, and the subsequent displacement of the sulfoxide. The reason is that either a good kinetic resolution of the sulfmyl chloride, generated first, or high separation factor of the intermediate diastereoisomers formed, permits the sulfinylating agent to be obtained in 100% de. The widely used Andersen method^^ is the most popular adaptation of this strategy. Developed at the beginning of the 1960s, it is based on the nucleophilic substitution on diastereomerically pure (-)-(5)-menthyl sulfinates with Grignard reagents. This substitution occurs with complete inversion of configuration at the sulfinyl sulfur, as demonstrated by chemical correlation^^ and ORD studies.^^"^^ The esterification reaction of (-)-menthol with /?-toluenesulfinyl chloride, using pyridine as base, occurs without any stereoselectivity; thus, several fractional crystallizations are needed to accede to a diastereomerically pure sulfmate. The seminal work of Solladie on the epimerization of sulfmate esters 50 and 51 in acidic medium in order to displace the equilibrium by precipitation of the less soluble isomer 51, permitting its isolation in 80-90% yield,^"^ constitutes a real breakthrough in the synthesis of sulfoxides (Scheme 14). The Andersen methodology has been used to prepare a large number of o.p. p-tolyl alkyl and p-tolyl aryl sulfoxides, and the use of other organometallic reagents, even highly functionalized ones, has allowed the synthesis of a wide variety of enantiomerically pure sulfoxides.^^ Nevertheless, the classical Andersen methodology suffers from the considerable drawback of not being general; in fact.
78
NOUREDDINE KHIAR et al.
p-ToK ^ONa I a) SOCb/benzene, O^C I b) Menthol/Py/ether, r.t.
e p-TohZ-^OMenthyl i^) 50
crystallization in acetone/HCI, -20°C = -^ ^ NSHCI
, H2O y 7
O .»..-S^ p.^^ OMenthyl ^^^ 51
80%
CI
. I
•'";S—OMenthyl p-Tol^l CI Scheme 14.
dialkyl sulfoxides cannot be obtained by this method. The starting alkanesulfinates have not been available epimerically pure at sulfur (for instance, the menthyl methanesulfinates are oils), and attempts to separate the epimers have not succeeded. Most of the methods developed recently are aimed at solving this problem. With the exception of Solladie's work on the epimerization of sulfinate esters, until recently no significant advances on the Andersen methodology have been made since its development at the beginning of the 1960s. Various approaches appeared at the beginning of the 1990s which solve most of the problems associated with the synthesis of o.p. sulfoxides. These include the cyclic sulfite^^ developed by Kagan, Evans's sulfinyl oxazolidinones,^^ Whitesell's chlorosulfinate,^^ and our DAG methodology^^ These methods will be discussed in detail, showing their scope and limitations, and, whenever possible, the applications they have generated in the literature. A. Sulfite Methodology^^ This is the second important contribution of Kagan's group in the synthesis of chiral sulfoxides. The method was reported in 1989 for the synthesis of tert-h\iiy\ sulfoxides,^ and the full paper on the generalization of the method was published in 1991.^^ The approach is based on the synthesis and use of an o.p. cyclic sulfite in the synthesis of various sulfoxides by two successive condensations of two organometallic reagents, RjM and R2M. Thus, the sulfoxides are produced in three separate steps: the formation of cyclic sulfite, synthesis of sulfinate esters, and transformation of sulfinates to chiral sulfoxides. The chiral diol 52^^ (Scheme 15), obtained from ethyl tartrate in one step (75%), was used to obtain the intermediate five-membered ring cyclic sulfite. The reaction
Synthesis ofChiral Sulfoxides
79
MeltpPh
^^'^'-^^"^
Hdc>H
CHzClg.EtaN
^ H ^ J ^ ^ , , ^H^Vi'ph 0,^,0
52
0.^.0 O^ 54
53 Scheme 15.
was not stereoselective, and gave a 1:1 mixture of trans- and c/^-sulfite, 53 and 54, when the traditional conditions were used^^: slow addition of thionyl chloride over the diol and Et3N dissolved in CH2CI2 at room temperature. A simple change in the experimental conditions, that of adding Et3N slowly into the CH2CI2 solution of diol 52 and thionyl chloride at -40 °C, enhanced the selectivity to 90:10 toward the isomer 53 with trans stereochemistry. This was obtained optically pure in 70% yield after crystallization in hexane. This sulfite was found to react cleanly with various organometallic reagents to give the corresponding intermediate sulfinate esters. Interestingly, the cyclic sulfites have been shown to generate the intermediate sulfinate with both large and small organometallics. This is in contrast to open sulfites, where small organometallics gave symmetric sulfoxides.^^ The trans structure of the starting sulfite 53 was originally determined, based on the transformation to sulfoxides with known absolute configuration, by assuming a double inversion of configuration in both successive reactions with organometallics RjM and R2M. This assignment has been confirmed recently by an X-ray analysis of the major sulfite 53, establishing for the first time that the monosubstitution on sulfites with organometallics takes place with complete inversion of configuration at the sulfinyl sulfur. The regioselective ring opening of cyclic sulfite with two potential leaving groups is closely related to the steric volume of the organometallic. Accordingly, when Rj is bulky, such as r-Bu or mesityl, regioselective cleavage gives mainly sulfinate 55 in high selectivity (80 and 76%, respectively), whereas when Rj is small, such as Ma Ph H'VH'Ph O OH S-Ri
0^1
RM —=
Ri
R2M
Ri
55
.R2 V o^ V 57
Ph „
'H^^)—^Ph
53
HO
\)
RiV O' "•• 56 Scheme 16.
^R2
*S.
-^'^ 58
80
NOUREDDINE KHIAR et al. Table 11. Synthesis of Chiral Sulfinates 55 and 56 from Sulfite 53 (Scheme 16) Entry
R^M
56/S5 Ratio
1 2 3 4 5 6 7 8 9 10 11 12
MeLi MeMgl EtMgBr n-OctMgBr f-BuMgBr r-BuMgCI f-BuLi BnMgCI BnMgBr H2C=CHMgCI MesitylMgBr PhMgBr
75/25 80/20 92/8 95/5 5/95 10/90 a 70/30 55/45 95/5 12/88 50/50
Note ^Only di-f-butylsulfoxide is obtained.
Et, n-octyl, or vinyl, the sulfinate 56 is the major product (> 80%). Moderate selectivity was obtained with MeMgl (60%) and poor selectivity was obtained in the cases of benzyl and phenyl sulfinate (40 to 0%, see Table 11). The o.p. sulfinate was obtained by crystallization, with a yield of isolated product claimed to be in the range of 60 to 80%. Finally, the sulfinates were transformed to o.p. sulfoxides by treatment with either Grignard or organolithium reagents. The presence of a free hydroxy 1 group in the sulfinate esters involves the use of two molar equivalents of the organometallic in THF at room temperature. Various dialkyl, alkyl aryl, and diaryl sulfoxides have been obtained in quantitative yield and in 100% ee (Table 12). The sulfite method resolves some of the limitations of the traditional methodology in the synthesis of some dialkyl sulfoxides with high ee. The method is particularly suitable for the synthesis of r^rf-butyl sulfoxides. The synthesis of both isomers of a given sulfoxide can be achieved via the Orsay route by permutation of the Rj and R2 in organometallics involved in the two substitution steps, and has been done in the synthesis of o.p. R and S methyl octyl sulfoxide and benzyl ethyl sulfoxide. This is possible only when both Rj and R2 are either small or bulky—when one of the groups is small and the other bulky, the permutation leads to the same sulfoxide. The (/?)-isobutyl lactate is commercially available, and thus can be used for the synthesis of the other enantiomer of sulfite, and then to sulfoxides 57 and 58 (Rj small and R2 bulky). The sulfite method is an ideal diastereoselective route to o.p. sulfoxides, especially when group permutation is possible. Thus, a single intermediate is used in an enantiodivergent approach to both sulfoxides.
Synthesis of Chirat Sulfoxides
81
Table 12. Synthesis of Enantiomerically Pure Sulfoxides 57 and 58 from Sulfinates 55 and 56, Respectively, and Organometallic R2M (Scheme 16) Entry
Sulfinate (R-i)
^2
1 2 3 4 5 6 7 8 9 10 11 12 13 14
55(r-Bu) 55(r-Bu) 55(t-Bu) 55(f-Bu) 55(t-Bu) 55(r-Bu) 55(f-Bu) 55(Mesityl) 55(Mesityl) 56(Me) 56(Et) 56(Et) 56(n-Octyl) 56(PhCH2)
MeLi PhLi n-BuLI H2C=CHMgCI 1-((2-CH2)C5H4N)Li PhCH2MgBr Ph(CH2)2MgBr MeLi PhMgBr n-OctMgBr PhLi PhCH2MgBr MeMgl EtMgBr
Configuration of Sulfoxide R 5 R R R R R R R R R R S S
However, this method suffers severely from the tedious experimental conditions leading to the sulfoxides from diol 52. Several crystallizations are required—the first to purify the trans sulfite, a second to purify the hydroxy sulfinate, and,finally,a column chromatography to purify the sulfoxide. This may be the reason why there is no application of this method in the literature, apartfromthat by the same group in the sulfinylation of ferrocene,^"^ getting the o-lithium derivative to react with various electrophiles to afford chiral ferrocenes (Scheme 17,59 R = HOCMe2, Me, Ph2P). B. frans-2-Phenylcyclohexanol in the Synthesis of Sulfinate Esters^^
As part of a program examining the utility of chiral auxiliary trans-l-phcnylcyclohexanol 60, introduced by Whitesell's group in 1985,^^ the same group has investigated the use of this alcohol for the synthesis of chiral sulfoxides. The reaction of 60 with an excess of alkane- or arenesulfinyl chloride affords sulfinate esters 61'(R) and 61-(S) in good yields and moderate selectivity [(410): 1]. The diastereomeric sulfinates were separated either by crystallization or by column chromatography. Two arene- and two alkanesulfinates were prepared optically pure by the reported method (Scheme 18). The major isomer of the methanesulfinate was used in the synthesis of o.p. (R) methyl /7-tolyl sulfoxide 62 in 76% yield, while the p-toluenesulfinate was used to obtain the (5)-/7-phenoxyphenyl/7-tolyl sulfoxide 63 in 70% yield. In order to get better stereochemical control and to circumvent the use of sulfinyl chlorides in the synthesis of sulfinate esters, the reaction of chlorosulfite esters of
82
NOUREDDINEKHIARetal. Q ^/-Bu H'")—^Ph q OH
Ferrocenyllithium
yHT
>.•
I
n-BuLi
i
Fe
Fo
f-BuMgCI Me
Ph
H"7—^Ph
o o . / 'o
O, ^^Bu
53
Scheme 17,
rra«^-2-phenylcyclohexanol with nucleophiles was investigated.^^ Reaction of 60 with thionyl chloride gave a mixture of chlorosulfmates 64 and 65 in 1:1 and 2:1 ratio at room temperature and -78 °C, respectively. The mixture of chlorosulfite esters 64 and 65 underwent reaction with Grignard and organolithium reagents to form sulfmate esters with de similar to those of the chlorosulfmates. Reaction of the diastereomeric mixture of chlorosulfite esters 64 and 65 with 0.9,0.5, and 0.25
O I
^ ' ^ - ^ ^ ^=^A- ^=^«'t' 61-W
<^^yl
61-rs; Chromatogr. separation or crystallization
Me"
(R « Me)
62
p-(PhO)C6H4MgBr (R = p.Tolyl)
•V 63 Scheme 18.
Synthesis ofChiral Sulfoxides
y ~ ^ O H
SOCI2
83
P=r;8'\?' . /=^pg'\--
60
65
64 Me2Zn ether, -78«C
O I
o I .s
Z^^^^^'S^ . /inrZ^pf V"^ 66
(2 :98)
67
Scheme 19.
equiv of dialkylzinc reagent (Me, Et, i-Pr) afforded the corresponding sulfmate esters with 92, 90, and 80% de, respectively. The explanation for both chlorosulfinates leading to the same sulfmate esters is that the chlorosulfinates 64 and 65 are in a rapid equilibrium relative to their reaction with the organozinc reagent, and the rates of reaction of the two diastereoisomers are significantly different (Scheme 19). The chlorosulfite method, which is claimed to be indicated for the synthesis of alkanesulfmates, works only in the case of methanesulfmate. The levels of control of diastereomeric excess with arylzinc and any arylmetal were in general very low. Additionally, reaction of chlorosulfite esters with alcohols and amines produced sulfinic acid derivatives in good yield but with low selectivity (< 2:1). The usefulness of the approach in producing methanesulfinates in high de was demonstrated by the synthesis of both isomers of sulforaphane 68, a naturally occurring sulfoxide that has been shown to stimulate the production of carcinogendetoxifying enzymes^^ (Scheme 20). O 'Me
p=^.
ph
Me*"*
:•
»C=S 6B'(S)
Ph
I Me
/ Me" Scheme 20,
"N»C=S SS-(R)
84
NOUREDDINEKHIARetal.
C. Aminosulfites The use of aminosulfites in the synthesis of o.p. sulfoxides was described for the first time in 1973 by Wudl and Lee,^^ using ephedrine as chiral auxiliary. In 1991, Benson and Snyder reported a modification of the Wudl and Lee procedure, obtaining o.p. sulfoxides in high yield by sequential displacement reactions of organometallic reagents on the 1,2,3-oxathiazolidine-S-oxide (aminosulfite 70), obtained from ephedrine and thionyl chloride.^^ In the modified procedure, it was found that allowing the diastereoisomers 70 and 71 to equilibrate by storing them at 0 °C for 24 h in the presence of EtgNHCl increased the ratio of 70:71 from 4.4:1 (Wudl conditions) to 9:1, thereby enabling 70 to be isolated in 70% yield. The addition offi-eshlyprepared Grignard reagents to 70 in toluene led to the production of sulfinamides 72 in good yield and diastereoselectivity. Addition of phenyl magnesium bromide produced only diphenyl sulfoxide, and no phenyl organometallic was able to give the intermediate phenyl sulfinamide 72 in good yield, one of the considerable drawbacks of the method for the synthesis of diaryl sulfoxides. It has been shown that addition of AlMcj to the intermediate sulfinamide 72, followed by addition of the Grignard reagent, was necessary to give the corresponding sulfoxides 73 in good yields and enantioselectivity (> 99% ee). The tert-buty\ sulfinamide 72 (R = t-Bu, entry 7 in Table 14) has been shown to be unreactive, and no tert-buty\ sulfoxide could be produced under any conditions with any Grignard reagent. Snyder's modification of the Wudl method is suitable for the synthesis of dialkyl and alkyl aryl sulfoxides in high ee. Both enantiomeric sulfoxides may be produced, either by reversing the order of organometallic displacement or by using the (15, 2/?)-(+)-enantiomer of ephedrine, which is commercially available. Compared with
HO^NHCHa
soCfed.aeq). EtaN
CfiHg CHg
CH2CI2. O^'C, 24h
69 (1R,2S)-(-)-ephedrine
•^ O
V
O^ "NMe
CeHs 70
R'MgX
CH3 C e H T c H a 71
Me,
RM, -40*'C Toluene. 5 h.
AIMea
5h, r.t. 73
^S' crystallization O' "NMe ^ 70
CH2CI2 30 min., r.t. CeHs
CH3
HO,
NMe
CfiHs
CH3
72 Scheme 21.
Synthesis ofChiral Sulfoxides
85
Table 13, Synthesis of Ephedrine Sulflnamides 72 from Oxathiazolidine 70 Entry
/?of72
RM
1 2 3 4 5 6 7 8
CH3 n-C^h^ CH2=CH CH2=CHCH2 (CH3)3C (CH3)2CH CH3CH2 C6H5
CH3MgBr n-C4H9MgCI Vinyl-MgBr Allyl-MgBr (CH3)3CMgCI (CH3)2CHMgBr CH3CH2MgBr CeHsMgBr
Yield (%)
de (%)
64 65 94 84 84 91 50 a
>99 75 96 97 89 98 61
—
Note: ^Diphenylsulfoxide was the main product.
Kagan's sulfite, the method has the advantage of regioselectivity, but the hmitation of producing tert-buiyl or aryl phenyl sulfoxides. D.
/V-Sulfinyl Oxazolidinones^^
In a broad program of using chiral oxazolidinones in asymmetric synthesis, ^^ Evans's group published a paper in 1992 on the synthesis and utilization of A^-sulfmyl oxazolidinones as new sulfmylating agent.^^ Two chiral auxiliaries were used in the study: oxazolidinones derived from (4/?, 55)-norephedrine 74^^^ and (45)-phenylalanine 75.^^^ The corresponding N-sulfmyl oxazolidinones 77 and 78 were obtained either by sulfinylation of the metallated oxazolidinone or by oxidation of the derived N-sulfenamides (Table 15). The reaction of the lithiated oxazolidinone derived from (4/?, 55)-norephedrine 74 and (45)-benzyloxazolidone 75 with arenesulfmyl chloride 76 in THF at -78 °C gave the corresponding A^-sulfmyl oxazolidinones 77 and 78 in modest diastereoselectivity (32-54% de) in favor of the R diastereomer (Scheme 22).
Table 14. Synthesis of Sulfoxides 73 from Sulfinamides 72 (Scheme 21) Entry 1 2 3 4 5 6 7
Sulfinamide 11 R:Me R:Me R:Me R:Me R: GH2=CH R: CH2CH=CH2 R: r-Bu
Note: ^Only sulfinamide was recovered.
R:tA PhMgBr CeFsLi n-BuMgCI f-BuMgCI PhMgBr PhMgBr PhMgBr
Sulfoxide Config. S S 5 5 S S a
ee (%) >99 60 >99 >99 >99 >99
86
NOUREDDINEKHIARetal. Table 15. Methods of Synthesis of N-Sulfinyloxazolidinones 77 and 78 (Schemes 22 and 23) Isolated Yield
N-Sulfinyl Oxazolidinone
Method of Synthesis
de (%)
77a 77b 78a 78b 78b 78c 78d
acylation acylation acylation acylation oxidation oxidation oxidation
32 54 46 44 42 16 16
Note
%/?S
%5s
69 61 9 20 68 a 35
1 4 61 50 28 33 49
^(/?s)-78c was unstable to chromatographic purification.
In the oxidation of the A^-sulfenamides 79, the m-chloroperbenzoic acid proved to be the best oxidant, yielding the A'^-sulfinyl oxazolidinones 78b-d in good yields (72-96%) as 1.4-2.5:1.0 mixtures of diastereoisomers, which were readily purified by chromatography (Scheme 23). The absolute configurations of the N-sulfmyl oxazolidinones 77 and 78 were determined by a combination of X-ray crystallography and chemical correlation of the derived sulfoxides obtained by displacement with Grignard reagents. Although the diastereoselective formation of A^-sulfinyl oxazolidinones is poor to modest, these compounds have been found to be efficient sulfmyl transfer O
A
O
HN O \—I ArSOCI ^ % , P^ 76a ^* 76b
X O
n-BuLi
O ..^
Ar'
m
O O A 11 .<^^N^0 \—l Bxf major 78a (5) 78b (S)
76a 76b
75 a : Ar« p-Tol, b: Ar» Ph
Scheme 22.
Ph minor 77a (S) 77b (S)
o o
AP
ArSOCI
O N O
major 77a (fl) 77b (fl)
O
HN
O
^BuLi
Ar minor 78a (A^ 78b («)
Synthesis ofChiral Sulfoxides 0 A
A
\
87 Q
0
^CPBA^ /
0 R....J5
0 11
CH2CI2
Brf
Brf
Brf major 78 W
79
minor 78 (S)
b:R-Ph:c:R-Me:d:R. -f-Bu Scheme 23.
reagents. They react rapidly with a variety of Grignard reagents to give the corresponding aryl alkyl and dialkyl sulfoxides in high yield (78-92%) and enantioselectivity (> 90%) (Table 16). Evans's group has used their A/^-sulfinyl oxazolidinone for the synthesis of (S)-tert-buty\ (phenylsulfinyl)acetate with an 81% yield using the Reformatsky reagent, derived from tert-butyl bromoacetate and activated zinc.^^^ Moreover, A^-sulfinyl oxazolidinones have been shown to be good intermediates for the synthesis of chiral sulfmate esters and sulfmamides with excellent ee. In all cases, the absolute configuration of the sulfoxide obtained is in agreement with the fact that nucleophilic displacement occurs with inversion of configuration at the sulfur center in the starting A^-sulfinyl oxazolidinone. In a competition experiment between chiral N-sulfinyl oxazolidinone and Andersen's menthyl sulfmate ester, it has been shown that the former is at least two orders of magnitude more reactive than the latter. This finding is being used to avoid some of the problems involved in sulfmate esters, related to the nature of the alkoxide leaving group in the nucleophilic substitution.
Table 16. Synthesis of Sulfoxides, R1SR2/ from N-Sulfinyloxazolidinones 78 Entry
^1
^2
1 2 3 4 5 6 7 8 9 10 11
Me Et /-Pr r-Bu Bn Me Me Me Me t-Bu t-Bu
p-Tol p-Tol p-Tol p-Tol p-Tol Ph r-Bu Bn Octyl Me n-Bu
Yield (%)
ee(%)
Configuration atS
90 90 91 88 86 87 78 82 78 92 91
99 98 97 97 99 90 93 91 100 100 100
S S S S S R R R R S S
88
NOUREDDINE KHIAR et al
O o ;•
OH
^
HO'.../\..»NHTs Ph-,^
HOf 80
79
^SMe
Mannostatin A
Scheme 24.
Garcia Ruano and co-workers^^ have recently developed a new sulfmyl transfer agent 79 (Scheme 24) obtained from menthyl p-toluene sulfinate. This sulfinylating agent, as in the case of the A^-sulfinyl oxazolidinones 77 and 78 of Evans, has an amide anion as leaving group, with the corresponding increase in reactivity. It has been used in the synthesis of 80,^^^ a precursor of the glycosidase inhibitor Mannostatin A.^^ Recently, Oppolzer's group reported on the synthesis and use of a new sulfinylating agent, ^^^ the A^-sulfinyl sultam 82, as part of a broad program on the use of the versatile bornane-10,2-sultam 81 in asymmetric synthesis. ^^^ The condensation of/7-TolSOCl with 81 in THF, using dimethylaminopyridine (DMAP) as catalyst, gave the A^-(p-tolylsulfinyl)bornane-10,2-sultam as a 6.2:1 diastereomeric mixture. Crystallization of the mixture from Et20/hexane afforded pure 82 in 77% yield. X-ray analysis showed the absolute configuration at the sulfinyl sulfur to be (/?). The reaction has been shown to be kinetically controlled, in contrast to the results obtained when n-BuLi was used instead of DMAR In the latter case, the reaction was under thermodynamic control, in accord with the result obtained by Evans with A^-sulfinyl oxazolidinone (Scheme 25). Compound 82 was shown to be a good sulfinylating agent. It reacts with various Grignard and Reformatsky reagents to give enantiomerically pure sulfoxides in high yield (Table 17), together with bornane-10,2-sultam 81, which can be recovered (> 91% yield) and reused.
1)DMAP, p-TolSCXJI, THF, r.t. 2) crystallization
RM.
\C
.i7
Me
^
83
Scheme 25.
Synthesis of Chiral Sulfoxides
89
Table 17, Synthesis of Optically Active Alky I (or Aryl) p-Tolyl Sulfoxides 83 from Sulfinamide 82 and Organometallic RM (Scheme 25) RM
Yield (%)
MeMgBr /-PrMgCI n-BuMgCi BnMgCI VinylMgCI (Z)-1 -PropenylMgBr (6-1-PropenylMgBr 2-PropenylMgBr AllylMgCI 2-ThienylMgBr 3-FurylMgBr 1 -PentynylMgBr BrMgCHjCOOf-Bu
93 92 97 91 95 80 79 90 96 89 89 85 83
Config.
ee(%)
R R R R R R R R R S S R R
99 99 97 >99 96 99 96 >99 >99 99 99 >98 >99
Sulfinyl sultam 82 was also used in the synthesis of enantiopure sulfinimines 85, useful precursors in the synthesis of enantiomerically pure amine, as well as a- and P-aminoacid derivatives. ^^^ Interestingly, the addition of one equivalent of water to the sulfinylated HMDS 84 prior to the addition of the aldehyde was necessary to convert enolizable aldehyde into enantiomerically pure sulfinimines 85, which cannot be obtained by the Davis procedure. Thus, both enolizable and nonenolizable aldehydes can afford enantiomerically pure aryl and alkyl sulfinimine 85 in good yield (Scheme 26). E. DAG (Diacetone-D-glucose) Methodology®^ Our own contribution in this area was the development, at the beginning of the 1990s, of a general method (which we named the DAG methodology) for the
82
84
Method A: i) 1 eq. HgO In THF, -78«C. ii) 1.1 eq. RCHO, -20 to 5«C Method B: I) 1.1 eq. RCHO. -20fiC. ii) 2 eq. CsF, -20 »C to r.t. Scheme 26,
85
90
NOUREDDINE KHIAR et al.
synthesis of both isomers of a large number of dialkyl, diaryl, and alkyl aryl sulfoxides. In a broad program aimed at the synthesis of sulfoxides with biological activity, as well as new chiral controllers for stoichiometric and catalytic asymmetric synthesis, ^^^ we were seeking a general route producing both isomers of a large number of sulfoxides. As we have seenfromthe preceding sections, such a general method was lacking in the literature at the end of the 1980s, when we started this program. It seemed to us that the Andersen approach was the candidate of choice m developing a general approach for the synthesis of o.p. sulfoxides. Since its appearance, there had been no systematic work to determine the optimal conditions, such as the best inducer of chirality, or the best solvent, base, and temperature for producing the intermediate chiral sulfmate. Thus, we decided to have a close look at the Andersen method in order tofindthem. An effective Andersen-like approach is one leading to the diastereomerically pure intermediate sulfinate esters, either by kinetic resolution or by physical separation. This could be achieved straightforwardly by using another inducer of chirality instead of the widely used menthol, which did not give any selectivity in the sulfinate formation step. This approximation had already been used by other authors, but the results were not encouraging. Mikolajczyk et al. used (-)-cholesterol for the synthesis of cholesteryl methanesulfinate, but although they obtained optically pure (-)-(S) and (+)-(/?)-diastereoisomers, the yields were very poor (3.5 and 0.7%, respectively).^^^ A review of the literature showed that the glucose derivative diacetone-D-glucose (DAG) was one of the most successful chiral controllers in a variety of processes,^^^ including arenesulfinate esters (up to 52% de).^^^ This behavior is not surprising as the secondary hydroxyl function (the hydroxyl group at C-3 position) is flanked by two functionalities that are very different from both steric and stereoelectronic points of view: a hydrogen atom at C-2 and a D-glyceraldehyde chiral backbone at C-4. With regard to the sulfinyl chlorides needed for sulfinate ester synthesis, a large number of them are readily accessible by oxidation of the corresponding disulfide according to the Young and Herrmann procedure.^*"* However, in the case of the more hindered tert-buiyl chloride, Netscher and Prinzbach's method gives better
Table 18. Synthesis of Optically Active C5j-N-Sulfinimines 85 from Sulfinamlde 82 (Scheme 26) Entry 1 2 3 4 5 6
R n-Bu n-Bu Ph Ph r£>MeCH==CH rf>PhCH=C H
Method
Yield (%)
ee (%)
A B A B A B
73 0 0 84 70 74
>99.5
>99.5 >99.5 98
Synthesis of Chiral
91
Sulfoxides
^Pr2NEt Toluene/-78«C
OA/ |
py THF/-78«C
Diacetone- D-glucose 86 fSs^-Alkanesulfinate 88
fHsM'kanesulfinate 87
Py THF/-78»C OH Diacetone-L-glucose R« Me, Et, Pr, /-Pr, p-Tol
CSs>AIKanesulfinate 90
89 Scheme 27.
results. Oxidation oftert-buiyl disulfide with hydrogen peroxide in acetic acid gives the corresponding r^rr-butyl thiosulfinate in quantitative yield, which on treatment with chlorine affords the desired r^r^butanesulflnyl chloride.^^^ The reaction of DAG 86 with alkane and arenesulfmyl chloride in THF at -78 °C, using pyridine as base, gave the corresponding sulfmate esters 87 and 88 in high yield and selectivity (Tables 17 and 18). The results left no doubt that DAG was the inducer of choice for the synthesis of o.p. arenesulfinate, as the diastereoselectivity (ranging from 86/14 to >98:<2) was the highest ever obtained in the synthesis of sulfmates. Having obtained a good inducer of chirality, we started the optimization of other parameters of the reaction, such as base, solvent, and temperature, in order to find conditions able to give a single isomer for all sulfinate esters. The Base Effect
The first parameter to optimize was the nature of the base used. While there were precedents in the literature for the change in the starting chiral alcohol, to the best of our knowledge there was no report on the effect of the base on the stereocourse of the formation of sulfinate esters. In the literature on the asymmetric synthesis of sulfinate esters, statements such as "asymmetric synthesis of sulfinate esters is achieved by the reaction of sulfinyl chloride with a chiral secondary alcohol in the
92
NOUREDDINE KHIAR et al.
presence of pyridine" at least indicate that pyridine is the best—if not the only— base which works in this reaction. We were delighted tofindthat the reaction of methanesulfinyl chloride with DAG in THF, using Hunig's base (/-Pr2NEt) as catalyst, gave a single isomer. Surprisingly this isomer has the opposite configuration at the sulfinyl sulfur to the one obtained using pyridine as base (see Table 19, entries 1 and 2). The reactions of the other sulfinyl chlorides (n-Pr, /-Pr, p-Tol) with DAG using /-Pr2NEt as base led to the formation of (5)-sulfinate esters as the major isomers with high selectivity (89 to >95%) instead of (/?-)sulfinate esters predominantly obtained when pyridine was used as base (70 to > 95% de. Table 20). To the best of our knowledge, this is the first report that the stereocourse of the formation of sulfinate esters depends not only on the chiral alcohol but also on the nature of the base. With the same inducer of chirality, DAG, it is possible to obtain optically pure (iS)-methanesulfinate 88, instead of the R diastereomer 87, simply changing the base from Py to /-Pr2NEt. Using DAG and /-PrjNEt is equivalent to using Py and changing the inducer of chirality from the cheap and commercially
Table 19, Influence of the Base and Solvent on the Stereochemistry of the Reaction of DAG with Methanesulfinyl Chloride O II
Base
DAG + Me^.s^CI Solvent (-782C)
••v° Me'^^ODAG
+ °^-" Me-^ ^ODAG
Base
Solvent
Pyridine
THF Toluene Et20 CH2CI2 CH3CN/THF THF THF
93/7
Toluene THF CH2CI2 THF Toluene THF Toluene THF Toluene
^2/^98^ <2/^98^ <2/^98^ 37/63
DMAP Imidazole /-PriNEt
Collidine MejNPh (DMA) NEt3
Note: ^Only one isomer was detected by ^H NMR.
(R)/(S)
83/17 84/16 76/24 73/27 78/22 82/18
17/83 42/58 37/63 17/83
<2/^98^
Synthesis ofChiral Sulfoxides
93
Table 20. Reaction of DAG with Different Sulfinyl Chlorides
0 DAG
0.
Base Solvent (-78^0)
—^
R"
+ s:: ODAG s
Solvent
/^s
Pyridine
THF
/-PrjNEt
Toluene
93/7 <2/>98 86/14 <2/>98 85/15 4/96 >98/<2 <2/>98 86/14 6/94
Entry
R
Base
1 2 3 4 5 6 7 8 9 10
Me Me Et Et
Pyridine
THF
/-Pr2NEt
Toluene
n-Pr
Pyridine
THF
n-Pr
/-Pr2NEt
Toluene
/-Pr
Pyridine
THF
/-Pr
/-Pr2NEt
Toluene
p-Tol
Pyridine
THF
p-Tol
/-Pr2NEt
Toluene
R" "ODAG R Yielcf (%)
87 90 85 90 75 80 56 50 84 87
Note: ^Diastereomerically pure compounds after purification of the major isomers by c.c. or recrystallization.
available DAG to the diacetone derivative of the expensive, nonnatural (-)-L-glucose 89 (Scheme 27, R = Me). The results in Tables 19 and 20 indicate that the stereochemistry of the sulfmate esters formed can be predetermined depending on the base used. In addition to the high diastereoselection of the reaction, the sulfmate ester diastereomers are well resolved to be easily separated by column chromatography, or by crystallization as most of them are crystalline. Moreover the diastereomeric excess can be easily and accurately determined in each case by ^H NMR analysis of the crude mixtures, because of the differences in chemical shifts of H-1 and H-2 in the two diastereomeric sulfmate esters ( Table 21). All of the sulfinate esters obtained up to now are stable and can be stored for several days in the refrigerator without any decomposition. An interesting and common feature of all of these sulfmate esters is the sign of specific rotation: (/?)-sulfmates are dextrovoiaiovy, while (5)-sulfmates are /^vorotatory. Generalization of the Base Effect The intriguing and useful base effect on the stereocourse formation of sulfinate esters of DAG raises several questions. The first and most important is the origin of such effect. In order to answer this question it is essential to know the stereoelectronic prerequisite of a base so as to direct the main formation ofR- or 5-sulfmate. For this study, and taking into account the major sulfmate formed in this reaction.
94
NOUREDDINE KHIAR et al. Table 21. Specific Rotation and ^H NMR Data of Alkane and Arenesulfinates, RS(0)ODAG
Entry
R
1 2 3 4 5 6 7 8 9 10
Me Me Et Et n-Pr n-Pr /-Pr /-Pr p-Tol p-Tol
Config. at S R S R S R S R S R S
Mo
+17(c 4.4, Me2CO) -60(c 2.7, MejCO) +12(c1.8, MejCO) -63(c 4.3, Me2CO) +6.4(cl.1,Me2CO) -33(c 0.5, Me2CO) +11(c2.9, Me2CO) -50(c 0.3, Me2CO) +10(c1.7, Me2CO) -125(c0.4, Me2CO)^
H-1 (ppm) 5.91 5.91 5.89 5.88 5.91 5.88 5.89 5.90 5.86 5.92
H-2 (ppm) 4.73-4.80 4.62 4.65-4.82 4.64 4.67-4.83 4.85 4.78 4.60 4.7 4.84
Note: ^Diastereomerlc mixture containing 86/14 of S/R sulfinates.
the base can be classified as one of two types: (1) type A or pyridine like bases, which include Py, DMAP, and imidazole, give mainly the (/?)-methanesulfinate with 56-86% de, and (2) type B or i-PrJ^Et-like bases, which include /-Pr2NEt, Et3N, collidine, and DMA, give mainly the (5)-methanesulfmate with 16 to >95% de. Thus, aliphatic amines, including much-hindered tertiary amines, behave like /-Pr2NEt, and bases having a nonhindered heterocyclic nitrogen act like pyridine. At first sight, this effect might be thought to reflect the relative basicity of the catalyst. However, there is no good correlation between the strength of the base and its behavior in the reaction. For example, pyridine and dimethylaniline (DMA) with similar p^^ values (5.20 and 5.06, respectively) afford different stereoisomers (Table 19). The opposite results obtained using pyridine and Et3N suggested the effect could be related to the hybridization type of the nitrogen atom in the amine. However, the results obtained with collidine, having an sp^ nitrogen and acting as an /-Pr2NEt-like base, ruled out this hypothesis and indicated that the only meaningful factor is the bulkiness of the amine. Structure of the Secondary Alcohol
The obvious question is whether the achiral stereodirecting base effect observed with DAG is a particular case of this alcohol or general behavior of secondary chiral carbinols. In order to answer this question and to get a better insight into the mechanism of the reaction, the reactivity of different chiral carbinols with methanesulfinyl chloride was tested using the optimal conditions previously determined for DAG: (1) in the presence of /-Pr2NEt in toluene at -78 °C, and (2) with pyridine in THE at-78 °C.^^^ The results obtained, summarized in Table 22, show that the stereocourse of this reaction is tightly bound to the nature of the base used. The differences in chemical
Synthesis of Chiral Sulfoxides
95
shifts allow determination of the ratio ofR and S sulfinates by ^H NMR analysis of the crude mixture. The higher de was obtained with dicyclohexylidene-D-glucose (DCG) as predicted from its structural similarity to DAG. /?-Sulfmate is obtained as the major isomer with pyridine (88% de, entry 3), while 5-sulfmate is the only isomer detected with /-Pr2NEt (de > 96%, entry 4). (+)-Isopinocanpheol, (-)-menthol, and methyl (5)-(-)-lactate show a similar behavior to DCG but with markedly lower de. On the other hand, (-)-borneol, (+)-isomenthol, (-)-cholesterol, and (/?)-3,3-dimethyl-2-hydroxy-Y-butyrolactone yield mainly the 5-sulfmate as the major isomer with pyridine (entries 5, 7, 11, and 15) and the /?-sulfmate with /-Pr2NEt (Table 22, entries 6, 8,12, and 16). Surprisingly, the lowest de (4 and 6%, entries 11 and 12) were obtained with (-)-cholesterol, which was the first chiral
Table 22. Reaction of Methanesulfinyl Chloride with Different Chiral Secondary Alcohols
9 Me" "CI
Entry 1
Solvent (-782C)
Alcohol Diacetone-D-glucose
2 3
18
Pyridine
87
7/93
86
/-Pr2NEt
90
>98/<2
>96 88 >96
Pyridine
61
74/26
48
/-PrzNEt
68
38/62
24
(IS, 2/?, 5/?)-(+)-lsomenthol
Pyridine
86
65/35
30
/-Pr2NEt
95
40/60
20
(IS, 2/?, 35, 5/?)-(+)Isopinocampheol
Pyridine
80
46/54
8
/-Pr2NEt
92
65/35
30
(-)-Cholesterol
Pyridine
>95
52/48
4
/-Pr2NEt
>95
47/53
6
(1/?, 2S, 5/?)-(-)-Menthol
Pyridine
>95
28/72
44
/-Pr2NEt
>95
71/29
42
(/?)-(-)-3,3-Dimethyl-2-hydroxyY-butyro lactone
Pyridine
70
61/39
22
/-Pr2NEt
74
49/51
2
Methyl (S)-(-)-lactate
Pyridine
83
39/61
22
/-Pr2NEt
87
54/46
8
[(1 S)-encyo]-(-)-Borneol
16 17
de (%)
6/94
14 15
Diast. Ratio Ss/Rs
>98/<2
12 13
Yield (%)
53
10 11
Rs
92
8 9
Me
/-Pr2NEt
6 7
Base^
Me"^"OR^
Pyridine
Dicyclohexylidene-D-glucose
4 5
P
o ..
Base
Note: *The solvents used were THF with pyridine and toluene with /-PrjNEt.
96
NOUREDDINE KHIAR et al.
alcohol used in the synthesis of o.p. methanesulfinates en route to o.p. methyl alkyl sulfoxides. The results of Table 22 prove unambiguously that the stereochemical outcome of sulfmate ester synthesis is highly dependent on the base used, and that we are dealing with a general stereochemical behavior in the asymmetric synthesis of sulfmate esters. Origin of the Diastereoselectivity in ttie DAG
Metliodology
Although the exact mechanism of the reaction is difficult to resolve, from our accumulated data we can propose a mechanism which explains the experimental results obtained and enables the prediction of the stereochemistry of the sulfinate formed, based on the steric volume of the amine. The experimental data clearly demonstrate that the diastereoselective formation of a sulfinate is base-dependent. Therefore, the proposed mechanism has to include the effect of the base on the stereocourse of the reaction. Some additional circumstances have to be taken into account: (1) the reaction of sulfmyl chlorides with alcohols in the presence of bases does not proceed via a sulfme intermediate^ ^^; (2) Mislow has proved that the reaction of chiral alcohols with sulfmyl chlorides in the presence of a base is kinetically controlled^^^'^^; and (3) on first inspection, one could imagine the extreme case where DAG, under the influence of the base, reacts with only one of the enantiomeric sulfmyl chlorides. However, this is not the case, because the same yield and ee are obtained when 1.2 or 2.0 equiv of MeSOCl are used. Accordingly, thefirststep of the process could be an equilibrium reaction involving the sulfmyl chloride and the base. The racemic sulfinamide formed would be the active sulfur species that interacts with the sugar derivative. In order to explain the diastereoselectivity observed in this process, we propose the formation of an intermediate sulfurane that can undergo pseudorotation during the reaction^ ^^ (Scheme 28). As shown in Scheme 28, the bulky and electronegative /-Pr2NEt and the incoming alcohol occupy the apical positions to form sulfuranes C and D. Sulfurane C evolves directly into the (5)-sulfmate. The formation of sulfurane D is less favorable because of a destabilizing interaction between the alkyl or aryl group (R) and C-5 of the sugar ring. Alternatively, the formation of (5)-sulfmate as the sole isomer can be explained by the simultaneous formation of sulfuranes C and D. By three consecutive pseudorotations, sulfurane D gives sulfurane G, which evolves into the (5)-sulfmate. When Py is used as the base, sulfuranes E and F are formed; the incoming alcohol is in an apical position and the less bulky base in an equatorial position. Sulfurane E is less favored because of the interaction between the alkyl or aryl group (R) and C-5 of the sugar ring. By two successive Berry pseudorotations, the more favored sulfurane F gives sulfurane I, where the leaving group is in an apical position, and I evolves into the (/?)-sulfinate. As in the case of i-Pr2NEt, it is also possible that
R,
,ODAG
p .... 0
w5)
+
F
R,
Scheme 28,
,ODAG
98
NOUREDDINE KHIAR et al.
sulfuranes E and F are formed simultaneously, and that, in a single pseudorotation, sulfurane E gives sulfurane H, which yields the (/?)-sulfmate (Scheme 28). Although the present experimental data do not allow us to confirm the exact mechanism, there is good qualitative agreement between the experimental data and the mechanism proposed in Scheme 28. Scope and Limitations
The method seems to be appropriate for the synthesis of a large number of o.p. dialkyl, alkyl aryl, and diaryl sulfoxides. The reaction of a Grignard reagent with the intermediate sulfmate ester leads to the corresponding sulfoxide in high yield and selectivity. Tables 23 and 24 show the large number of sulfoxides obtained. Most of these sulfoxides have been used to determine the absolute configuration of the intermediate sulfinates, by assuming that the substitution step occurs with complete inversion of configuration at the sulfur. Of the large number of sulfoxides prepared by the DAG method, special attention should be given to the methyl sulfoxides present in various biologically significant molecules, and to the tert-buiy\ sulfoxides. Recently, it has been shown that optically active alkyl tert-huiyl sulfoxides react with acyclic a,p-unsaturated esters with much higher selectivity than the more commonly used p-tolyl sulfoxide.^^^ Similar behavior has been observed in the aldol-type reaction between sulfinyl anions and aldehydes.^^^ tert-Buiyl sulfoxides can be obtained by the DAG methodology in two ways: either by reacting the tert-buiy\ Grignard reagent with diastereomerically pure sulfinates, or by reacting a Grignard reagent with
Table 23, Synthesis of Optically Active Sulfoxides, MeS(0)R, from DAG MethanesuIfinates and R'MgX Methanesu Ifinates Entry
(Config.atS)
R: In R'MgX
Sulfoxide
Yield {%r
1 2 3 4 5
R R R R R
p-Tol Ph PhCH2 n-Pr f-Bu
84 78 83 66 95
6 7 8 9 10 11
S S S S S S
p-Tol Ph PhCH2 n-Pr r-Bu vinyl
90 80 83 69 95 37
Note:
^ i e l d after flash chromatography.
Coniig. (ee %) R{^00)
R{^oo) /?(100)
R{^oo) /?(100) 5(100) 5(100) 5(100) 5(100) 5(100) 5 (100)
Synthesis of Chiral Sulfoxides
99
Table 24. Synthesis of Optically Active Sulfoxides, RS(0)R', from DAG Alkane- or Arenesulfinates, RS(0)ODAG and R'MgX^ Sulfmate Entry
R
1 2 3 4 5 6 7 8 9 10 11
Et Et n-Pr n-Pr /-Pr /•-Pr
p-Tol p-Tol r-Bu f-Bu f-Bu
Sulfoxide
Config.atS R! R S R S R S R S R S S
p-Tol p-Tol p-Tol p-Tol p-Tol p-Tol Et Et Me
Ph p-Tol
Yield (%) 96 90 88 89 98 89 87 80 60 85 87
Config. (ee %) R{99) 5(99) /?(100) SdOO) /?(100) 5(100) 5(70)^ /?(100) 5(100) 5(90) R{93)
Notes: ^All reactions were carried out by adding 1.2-1.3 equiv. of the Grignard reagent over 0.05 M solution of the sulfinate in toluene at 0 °C. ^Obtained from a crude sulfinate wih 70% de.
rerf-butanesulfmates.^^^ The obtaining of both r^rr-butanesulfinate esters optically pure is of great importance, as they can lead not only to r^rr-butyl sulfoxides but also to other /^rr-butylsulfinyl derivatives with high synthetic value, as we will see in the following section. Some Applications The first applications of the DAG methodology have been made in our group and were aimed mainly at the synthesis of biologically active compounds bearing a chiral methylsulfinyl group. In 1993, we developed a general method for the asynmietric synthesis of P-amino-y-hydroxysulfoxide, a constituent of the antibiotic sparsomycin.^^^ Sparsomycin is a unique antibiotic isolated from the fermentation broth of Streptomyces sparsogenes var. sparsogenes^^^ and Streptomyces cuspidosporus}^^ The structure of sparsomycin was assigned in 1970,^^^ and its biosynthesis was elucidated recently. ^^^ Sparsomycin exhibits antibiotic activity against a variety of gram-positive bacteria, and it shows potent antitumor activity against the KB human epidermoid carcinoma cell in tissue culture.^^^ Three total syntheses of sparsomycin have been reported in the literature, ^^^"^^^ but no asymmetric approach has yet been developed. Sparsomycin can be viewed as an amide between the known acid 91 and the highly functionalized P-amino-y-hydroxysulfoxide 92. The synthesis of A^-cyclohexyl analogues of 95 is presented in Scheme 30. Treatment of 1 equiv of the 3-oxazoline 93^^^ with 2.2 equiv of LDA and (/?)- or
100
NOUREDDINEKHIARetal. O CO2H
HN
o \ o^ ;.
O
SCHj
91
(S)-(+)-Sparsomycin HgN"
^ " ^ ^ CH3 92
Scheme 29.
(5)-methanesulfinate of DAG gives o.p. methyl sulfoxides 94/? and 945, respectively. The asymmetric induction in the reduction of the C = N double bond was studied in detail, using various reducing systems in different solvents and at different temperatures. The presence of an acid (ZnCl2, MgC104, or AcOH) was necessary for an efficient reduction of the C=N double bond. The high stereochemical outcome was obtained using DIBAL in the presence of ZnCl2 at -78 °C in THF. Under these conditions, o.p. N-cyclohexyl derivatives 95(7?^, S^ and 9S{SQ,R^ were obtained as single isomers from 945 and 94/?, in 80 and 75% yield, respectively. This study has shown that the methylsulfmyl group is as good a chiral controller as the widely used p-tolylsulfmyl group. The DAG methodology has been used recently in the synthesis of both isomers of oxisuran, as an application of a general route to the synthesis of o.p. a-methylsulfmyl ketones.^^^ Oxisuran 96, (methylsulfmyl)methyl-2-pyridyl ketone, is a synthetic immunosuppressive drug used in organ and tissue transplants to suppress cell-mediated
N
RHN
C);.
DIBALyZnClz
6J^^^^CHa
80%
N-Cyclohexyl-95-(/?c,%)
94.^5;
93
s
r \
o
^Z'
^ y^o^^Jk^i^ "CH3
..
DIBAUZnCl2
Me" ^ODAG
75%
RHN
q^CH3
N-Cyclohexyl-95<Sc.«s) Scheme
30.
Synthesis of Chiral Sulfoxides
101
immunity, as well as to promote graft acceptance without inhibiting humoral antibody formation.^^^ While various syntheses of racemic oxisuran have been reported in the literature, ^^'^ no asymmetric synthesis of o.p. oxisuran has been developed. The usual approach for the synthesis of P-ketosulfoxides by condensation of an a-sulfmyl anion and an ester^^^ cannot be used for the synthesis of oxisuran and analogues. Condensation of the potassium enolate of methyl 2-pyridyl ketone with (5)-methanesulfinate (88, R = Me, Scheme 27) gave the desired P-ketosulfoxide in 70% yield and only 33% ee, because of the epimerization at the sulfmyl sulfur in the condensation step (Scheme 31). The epimerization of the sulfmyl sulfur during the condensation of an enolate with o.p. sulfmate esters, especially in the case of cyclic ketones, has been reported in the literature. The problem has been solved either by using an acid-catalyzed reaction of an enol silyl ether of a cyclic ketone with optically active menthyl p-toluenesulfmate,^^^ or by using a magnesium enolate instead of the lithium enolate of acyclic ketone with (-)-(5)-0-menthyl/7-toluenesulfmate.^^^ In our case, we chose to use the A^//-dimethylhydrazone derivatives of the starting ketone, known to give the corresponding sulfoxides with complete inversion of configuration."« The reaction of 2 equiv of A^,A^-dimethylhydrazone 97, obtained by standard procedure, with 2 equiv of n-BuLi gave the corresponding a-lithioderivative, which on reaction with methanesulfmates 87 (R = Me) and 88 (R = Me) yielded o.p. sulfoxides 985 and 98/?, respectively, in high yield. The configurational assignment of the a-sulfmyl hydrazone obtained was made assuming that the condensation step occurs with complete inversion of configuration at the sulfmyl sulfur, as is the case for many Andersen-type reactions. Reaction of compounds 985 and 9SR with copper chloride in aqueous THF^-^^ led to the first synthesis of o.p. 965 and 96R oxisuran, respectively (Scheme 32). The generality of the method was demonstrated by the preparation of various oxisuran bioisosters where the pyridyl moiety was replaced by phenyl, furyl, and thienyl moieties. The optical purities of these products were determined by proton NMR spectroscopy using the chiral shift reagents Eu(hfc)3^'*^ and (-)-A^-(3,5-dinitrobenzoyO-a-phenylethylamine,^"*^ following conditions established by the study of racemic mixtures of the P-ketosulfoxides.
II
1)KHMDSm
r
I
_ 96
yield: 70%.ee: 33% Scheme 31.
102
NOUREDDINE KHIAR et al.
g^CHa
cuCt O '7 O
t^
96-cs; CH3
97
I
fvBuLi
87(R"Me) 5.CH3 II
N
i V
«
^
o (^V 96-r/?;
98-W Scheme 32.
Using the DAG methodology, Noiret et al.^"^^ synthesized the two enantiomeric sulfoxides 99(R) and 99(5) (Scheme 33) in high yield and selectivity, in order to study the mechanism of the in vivo desaturation of oleic acid.^"*^ These sulfoxides, 99(R) and 99(5), which mimic the homoallylic position of sulfur at Sj3 oleic acid, were used as models to determine the best chiral shift agent for analyzing the stereochemistry of dialkyl sulfoxides with complex spin systems'*^ generated in the oxidation of thio-oleic acid formed in in vivo experiments. The DAG methodology has also been used for the synthesis of dithioacetal mono-5-oxides 100, important chiral formyl analogues, and the 3-ethylsulfmylmethyl-4,5-dihydro-4,5-diphenylisoxazoles 101 (Fig. 8), important synthetic intermediates.^"^ In an interesting example of the first asymmetric Heck reaction,^ using sulfoxides as chiral auxiliaries, Carretero et al.^"^^ have recently used a new chiral sulfoxide, obtained by the DAG methodology. The palladium-catalyzed arylations of 4-arylsulfinyl-2,3-dihydrofurans 102 have shown that the stereochemical outcome of the reaction is highly dependent on the substitution of the sulfoxide. Thus, independently of electronic substitution of the ary 1 iodide, different aryl sulfoxides
Pyridine/loluene
'v ;
AODAG
n-Bu
n-Bu
^ 99-(R)
\
;•
X
n-Bu
Scheme 33.
^'^^-^^MgBr
ODAG
99'(S)
Synthesis of Chiral Sulfoxides
103
R=:Me, Et
P^
100
101
Figure 8, Chiral methyl and ethyl sulfoxides of interest as synthetic intermediates.
lead predominantly to one isomer with moderate selectivity (34-56% de), while (9-dimethylaminophenyl sulfoxide 102 [Scheme 34, Ar = c?-(Me2N)C^H4] affords the isomer 103 with high stereocontrol (70-88% de). For the synthesis of o.p. compounds by the reported method, the important (5)-<9-(A^,A^-dimethylamino)phenyl methyl sulfoxide was obtained using the DAG methodology. The 2-phenyl-3-(arylsulfmyl)-2,5-dihydrofurans obtained, 103, which are also vinyl sulfoxides, undergo a second Heck reaction leading to 3,5-diaryl-2,3-dihydrofurans 104 as single isomers in excellent yields. Subsequently, 103 has been reduced by Ni Raney leading to o.p. dihydrofuran 105 (Scheme 34). While high stereoselection has been achieved in radical reactions which occur in a-position^"^ to a center substituted with a chiral auxiliary, diastereofacial control in the addition of achiral radicals to the P carbon is, in general, difficult to achieve. ^"^^ In connection with this, Toru et al. reported extremely high P-stereoselection in the addition of tertiary, secondary, and even primary alkyl radicals to chiral a-sulfmyl cyclopentanones in 1993.^"^^ The effectiveness of the diastereoselective addition of achiral radicals has been shown to depend on the size of the substituent at the sulfmyl sulfur. Bulky chiral arylsulfmyl groups show excellent diastereoselectivities (> 98:< 2).
V -
Me'' ''ODAG
PK
Ar«o-(Me2N)C6H4
g
^\
.SOAr
t X r ^ a™ ^^
a) ArMgX; b) (i) LDA; (ii) ethylene oxide; c) (I) LDA; (ii) HCOOEt; d) MsCI/EtaN; e) Phl/Pd(OAc)2/Ag2C03/ phosphine ligand; f) Zn; g) H2, Ni-Raney Scheme 34,
Q105
NOUREDDINEKHIARetal.
104
SOaNa pj
1)CIS03H H2O
"'xy"'''''" NaOH, NazCOa
Sf
PH
3) NaC
Pr''
1) SOCI2 2) DAG, Py 3) Chromatogr. separation
O
^^^Yl^^*^
MeMgl/ether-THF
S ^ Pr' 107
QfiC. r.t.
'"'\
y^"
Pr' 106-^S;
Scheme 35.
The best chiral controller has been shown to be the 2,4,6-triisopropylphenylsulfinyl group, obtained by the DAG methodology. The two diastereomers of the DAG 2,4,6-triisopropylbenzenesulfinate were separable by silica gel flash-column chromatography, giving the (5)-sulfmate 106(5) as an oil and the (/?)-sulfmate 106(/?) as a solid^"*^ (Scheme 35). Treatment of 106(5) with methyl magnesium iodide affords the R isomer of triisopropylphenyl methyl sulfoxide 107. The reaction of (5)-2-[(2,4,6-triisopropylphenyl)sulfmyl]-2-cyclopentenone 110a, obtained from 106(5) and lithiumacetal 109, with an alkyl radical gave a single diastereomer in all cases studied even with ethyl radicals. In the same study the mesityl sulfoxide 110b obtained from (5)-2,2-diphenyl-l,2-dihydroxypropyl2-0-(2,4,6-trimethylphenyl)sulfmate 108 prepared by the sulfite methodology was shown to be less effective in the diastereoselective addition of small alkyl radicals (Table 25, entries 1 and 5). Neverthless, 110b has the advantage of giving both
0^6.^ \ J
1)106-rs;or108-rs:;
O ^ / . R' /'''v^'S:^^^
A- R B B-EtaB/RI
2) H2SO4
109
110 a,b
a: R' - APr. b: R *- Me
Scheme 36.
9\ ^A>^...*SOAr R llla,b
O JL^SOAr 'R ll2a,b
Synthesis of Chiral Sulfoxides
105
diastereoisomers in the addition of tert-buiyl radical in the presence and absence of Lewis acid (Table 25, entries 12-14). Recently, Toru's group reported on the radical P-addition to the 4- and 5-methyl2-[(2,4,6-triisopropylphenyl)sulfinyl]-2-cyclopentenone.^^^The starting cyclopentenones 117 and 118 were synthesized by condensation of the vinyl lithium 113 and 114 with DAG 2,4,6-triisopropylbenzenesulfinate 106 to afford 115 and 116, respectively, as a 1:1 diastereomeric mixture in 93% yield. Removal of the acetal group by treatment with acidic silica gel gave the desired cyclopentenones 117 and 118 in 99% yield. The P-addition of alkyl radicals to 4-methyl-2-(arylsulfmyl)-2-cyclopentenone 117 has been shown to occur in a completely stereocontrolled manner. Of a mixture of (4R)- and (45)-117, only (4/?)-117 reacts with t-Bu and /-Pr radicals to give the trans adducts 119a and 119b in 99% yield, while (45)-117 remained entirely unreacted. The stereochemical outcome of the reaction shows that the alkyl radical approaches from the side opposite to the aryl moiety in an antiperiplanar orientation to the carbonyl and sulfoxide bond. The 2,4,6-triisopropylphenyl group on sulfur plays a critical role, as it effectively shields the olefin face at the P-position by one of the isopropyl groups. This was confirmed by the 1:1 diastereomeric mixture obtained in the reaction of 4-methyl-2-(/7-tolylsulfinyl)-2-cyclopentanone with the tert'buiyl radical. In the case of (45)-117, attacks of an alkyl radical from both re and si faces are inhibited by the isopropyl group on the phenyl ring and by the 4-methyl group on the cyclopentanone ring. Addition of r^rf-butyl radical to a 1:1 diastereomeric mixture of {5R)- and (55)-methyl-2-[(2,4,6-triisopropylphenyl)sulfinyl]-2-cyclopentenone 118 gave stereoselectively the cis adduct 120 as a single isomer
Table 25, Radical P-Addition to Cyclopentenone 110, Scheme 37 Entry
Enone
1 2 3 4 5 6 7 8 9 10
110a 110a 110a 110a 110b 110b 110b 110b 110a 110a 110a 110b 110b 110b
n 12 13 14
Method R A B B B A B B B A A A B B B
Et /-Pr c-Hex r-Bu Et /-Pr c-hex t-Bu Et Et Et f-Bu f-Bu f-Bu
Lewis Acid
EtAICIj TiCl2(0-,f-Pr)2 EtAICl2 TiCl2(0--f-Pr)2
Yield (%) 98 99 87 97 94 99 89 99 95 95 60 99 99 19
111.112 >98:<2 >98:<2 >98:<2 >98:<2 94:6 >98:<2 >98:<2 >98:<2 >98:<2 21:79 42:58 >98:<2 <2:>98 <2:>98
106
NOUREDDINE KHIAR et al.
"'
1,3-114
' 115-116
"7-"»
Ar-2.4,6-rAPr)3C6H2 113.115 and 117: R^ - Me, Rg - H 114.116 and 118: Ri - H. Rg - Me
Scheme 37.
together with unreacted (55)-118 (Scheme 39). These results have been rationalized based on the results obtained in the reaction of 5-methyl-2-cyclopentenone, which gives the 2,4-cis isomer preferentially. In the reaction of (5/?)-118, the 3,5-cis compound 120 should be formed preferentially because the opposite re face is completely shielded by the bulky 2,4,6-triisopropylphenyl group in the preferred conformation of the sulfoxide. In contrast, (55)-118 would not give the addition product because the 2,4,6-triisopropylphenyl group shields the preferential re face. As part of their work on the asymmetric synthesis of biologically significant molecules using chiral sulfinyl derivatives, Garcia Ruano, Fernandez et al.^^^ have recently found that the tert-butyl sulfinyl group is the best inducer of chirality in the synthesis of o.p. aziridines. The DAG methodology has been successfully applied for the synthesis of the starting A^-/^r^butylsulfinimine, giving access to both isomers optically pure (Scheme 40). The synthesis of aziridines was performed using either dimethylsulfonium methylide (reagent A) or dimethyloxosulfonium
EtaB. Rl (4R)'U7
f4S>117
A r » 2,4,6-lriisopropylphenyl a: R = t-Bu, b: R «i-Pr. c: R « Et
Scheme 38.
Synthesis of Chiral Sulfoxides O
107
O I •Ar O
(SRyUB O
O
O
O
EtsB, /-Bui
O I
Bu'
?^Ar
r5s;-ii8
120
If ••
(recovered)
f5S;-118 Ar» 2,4,6-triisopropylphenyl Scheme 39.
(reagent B or C) as methylene transfer reagent. In all of the cases studied, the r^r/-butylsulfinyl group induced the highest stereocontrol (>90%) as compared with thep-tolyl and the naphthyl derivatives. Optically pure A^-sulfmylaziridines 124 (S^, S) and 124 (R^, R) were obtained in 75% isolated yield after a single crystallization in hexane. Interestingly, the use of the DAG methodology allowed the authors to double-tune the diastereoselection of the aziridination reaction, either by the chirality at the sulfmyl sulfur, or by the nature of the methylene transfer reagent used, the former being more efficient. In a project aimed at diastereoselective alkylations of A^-acylsulfmamide enolate,^^^ Ellman et al. found that the more sterically hindered ferr-butanesulfinamide 126 provided a level of diastereoselection higher than that of//-acyl derivatives of arenesulfmamides.^^^ For the synthesis of the starting chiral r^rr-butanesulfmamide 126, Ellman's group used the DAG methodology. ^^"^ Later, they developed a catalytic method using the technology developed by Bolm et al. based on the use of chiral Schiff base-vanadium complexes. After extensive work optimizing the catalytic system, they found that the use of 0.25 equiv of the 1.LHMDS
EtaN
'\P
'j'
f-Bu* ODAG 2. RCHO/CsF ^ - B U ^ ^ ' N ' ^ R (S) 119S, R=Ph 120S. R=E-PhCH=CH
XP THF
1.LHMDS
f-Bu'^'GDAG 2. RCHO/CsF ^Bu^ "N ^^
Scheme 40,
R
119S, R»Ph 120S. R=E-PhCH=CH
108
NOUREDDINE KHIAR et al.
Table 26. Reaction of N-Sulflnylimines with Dimethylsulfoniurn Methylide (n = 0, Reagent A) and Dimethyloxosulfonium Methylide (n = 1, Reagents B and C)
k
CH2 = S(0)nMe2
/ Reagenf N-Sulfinyl asyUR) Aziridines Ratio Yield (%) de (%)
Entry R^ 1 2 3 4 5 6 7 8 9 10 11
p-Tol p-Tol p-Tol p-Tol Naphthyl Naphthyl f-Bu t-Bu
R14
R14
n = 0, 1
— — — — — — — —
—
NBu
r-Bu f-Bu
— —
Ph Ph (aPhCH==CH (aPhCH==CH Ph Ph Ph Ph Ph (6PhCH==CH (aPhCH==CH
121 121 122 122 123 123 124 124 124 125 125
A B A B A B A B B A C
40/60 73/27 42/58 60/40 23/77 83/17 15/85 95/5 5/95 18/82 83/17
73 95 82 85 75 80 70 85 85 72 42
20 46 16 20 54 66 70 90 90 64 66
Note: ^Generated in situ from trimethylsulfonium iodide with NaH in DMSO for A, from trimethyloxosulfonium iodide with NaH in toluene for B, and in THF as solvent with n-BuLi as base for C.
EtaN/Toluene f-BuSOCI (86:14) 1) chromatogr. separation 2) UN(TMS)2 3) KF/AI2O3 t-Bu
NH2
126-^S; Scheme 41.
Synthesis of Chiral Sulfoxides
109
(^Bo'S^S-'-^" 128
^26'(R) Scheme 42,
catalyst formed by ligands 127, V0(acac)2, is able to oxidize r^rr-butyldisulfide 128 in 1 mol scale with H2O2 in CHCI3 affording thiosulfinate 129 with 91% ee in 92% conversion and isolated in 88% yield. Careful crystallization of thiosulfinate ester 129 twice from hexane affords {R)-129 with > 99% ee and 52% recovery (Scheme 42). Conversion of 129 to the desired sulfmamide requires the use of LiNH2 in ammonia as solvent, as other usual solvents do not give the reaction. Thiosulfinate 129 has been used for the synthesis of various chiral sulfinyl derivatives such as sulfoxides, sulfinamides, and sulfinimines in good yield, tertButanesulfinyl ketimines have been successfully used for the asymmetric synthesis of a,a-dibranched amines^^*^ as well as for the synthesis of o.p. amino acids.^^^^
IV, SUMMARY AND PERSPECTIVES The previous sections demonstrate without any doubt that the synthesis of chiral sulfoxides has experienced an enormous expansion in the last decade from both the quantitative and qualitative points of view. Nowadays it is possible to produce a large number of optically pure sulfoxides with tailored structural properties employing relatively simple and very efficient synthetic procedures. The easiness and generality of the new methods have opened the way for the utilization of chiral sulfoxides in new chiral transformations. In addition to the classical uses of sulfoxides as chiral controllers in a few chemical transformations, the new applications on record concern virtually every aspect of asymmetric synthesis. The wide range of structures reported to date demonstrates that the basis for the synthesis of this interesting class of molecules is now established and further challenges are ready to be undertaken in the near future. An area which will profit from the new development is the synthesis of biologically significant molecules with an optically
110
NOUREDDINE KHIAR et al.
pure dialkyl sulfoxide. The excellent results achieved in the asymmetric synthesis of nonnatural amino acids, in the P-radical addition, in the aziridination and Heck reactions demonstrate the benefit of using sterically and stereoelectronically different sulfinyl derivatives from the widely used/7-tolyl sulfoxide. The effectiveness of the recent synthetic methodologies anticipates the future utilization of chiral sulfoxides in solid phase and combinatorial chemistry, for the preparation of libraries of chiral small molecules with or without a chiral sulfmyl group. The different ways of chiral sulfoxides of efficiently chelating transition metals remain almost unexplored and deserve thorough investigation in connection with transition metal-catalyzed enantioselective transformations. Additionally, chelating and hydrogen bonding capabilities of sulfoxides in combination with the chirality of the sulfinyl group augur interesting results in relation to their use as chiral host molecules in molecular recognition.
ACKNOWLEDGMENTS The authors thank the DGICYT of Ministerio de Educacion y Ciencia (Spain) for financial support (grants PB 97-0731 and PB 96-0820).
REFERENCES 1. (a) Lacote, E.; Delouvrie, B.; Fensterbank, L.; Malacria, M. Angew. Chem. Int. Ed. Engl. 1998, 37, 2116. (b) Lee. A. W. M.; Chan, W. H.; Mo, T. Tetrahedron Lett. 1997,38, 3001. (c) Konno, T.; Yamazaki, T; Kitazume, T. Tetrahedron 1996, 52, 199. (d) Kuenzer, H.; Thiel, M.; Peschke, B. Tetrahedron Lett. 1996,57,1771. (d) Marco, J. L.; Femdndez, I.; Khiar, N.; Femdndez, R; Romero, A. / Org. Chem. 1995, 60, 667. 2. (a) Aggarwal, V. K.; Franklin, R. J.; Rice, M. J. Tetrahedron Lett. 1991, 32, 7743. (b) Aggarwal, V. K.; Davies, I. W.; Maddock, J.; Mahon, M. F ; Molloy, K. C. Tetrahedron Lett. 1990,31, 135. 3. (a) Bravo, R; Zanda, M.; Zappala, C. Tetrahedron Lett. 1996, 37, 6005. (b) Tsuchihashi, G.; Iriuchijima, S.; Maniwa, K. Tetrahedron Lett., 1973, 14, 3389. (c) Ronan, B.; Marchalin, S.; Samuel, O.; Kagan, H. B. Tetrahedron Lett. 1988, 29, 6101. (d) Pyne, S. G.; Dikic, B. J. Chem. Soc, Chem. Commun. 1989,826. (e) Pyne, S. G.; Dikic, B. / Org. Chem. 1990,55,1932. (0 Pyne, S. G.; Boche, G. / Org. Chem. 1989, 54, 2663. (g) Bravo, R; Viani, F; Zanda, R; Fokina, N.; Kukhar, V. R; Soloshonok, V. A.; Sishkin, O. V.; Struchkov, Y. T. Gazz. Chim. Ital. 1997,62, 2555. 4. (a) Butlin, R. J.; Linney, I. A.; Mahon, M. F ; T^e, H.; Wills, M. J. Chem. Soc, Perkin Trans. 1 1996, 95. (b) Carreno, M. C ; Garcia Ruano, J. L.; Martin, A. M.; Pedregal, C.; Rodriguez, J. H.; Rubio, A.; Sanchez, J.; Solladi^, G. / Org. Chem., 1990,55, 2120. (c) Miyashita, K.; Nishimoto, M.; Murafujf, H.; Murakami, A.; Obika, S.; Ishida, T.; Imanishi, T. J. Chem. Soc, Chem. Commun. 1996, 2535. (d) Solladi^, G. Pure Appl. Chem. 1988, 60, 1699. (e) Page, R C. B.; Prodger, J. C ; Hursthouse, M. B.; Mazid, M. / Chem. Soc, Perkin Trans. J 1990, 167. 5. (a) Mase, N.; Wake, S.; Watanabe, Y; Toru, T. Tetrahedron Lett. 1998, 39, 5553. (b) Toru, T; Watanabe, Y; Mase, N.; Tsusaka, M.; Hayakawa, T.; Ueno, Y Pure Appl. Chem. 1996, 68, 711. (c) Zahouily, M.; Caron, G.; Carrupt, R-A.; Knouzi, N.; Renaud, R Tetrahedron Lett. 1996, 37, 8387. (d) Renaud, R Chimia 1997,51, 236. (e) Beckwith, A. L. J.; Hersperger, R.; White, J. M. / Chem. Soc, Chem. Commun. 1991, 1151. (0 Snyder, B. B.; Wan, B. Y-F; Buckman, B. O.; Foxman, B. M. J. Org. Chem. 1991, 56, 328. (g) Renaud, R; Bourquard, T.; Gerster, M.; Moufid, N. Angew. Chem. Int. Ed Engl. 1994,33, 1601.
Synthesis of ChiraI Sulfoxides
111
6. (a) Shibasaki, M.; Boden, C. D. J.; Kojima, A. Tetrahedron 1997, 53, 131\. (b) Loiseleur, O.; Hayashi, M.; Schmees, N.; Pfaltz, A. Synthesis 1997, 1338. 7. (a) Arai, Y.; Masuda, T.; Masaki, Y. Chem. Pharm. Bull. 1998,46, 107. (b) Lee, A. W. M.; Chan, W. H. Top. Curr. Chem. 1997,190, 103. (c) Garcia Ruano, J. L.; Carretero, J. C ; Carreno, M. C ; Martin, L. C ; Urbano, A. Pure Appl. Chem. 1996, 68, 925. (d) Padwa, A.; Cochran, J. E.; Kappe, C. O. J. Org. Chem. 1996, 61, 3706. (e) SoUadi^, G.; Carreno, M. C. Organosulfur Chem. 1995, 1. (0 Arai, Y; Koizumi, T. Sulfur Rep., 1993, 75, 41. 8. (a) Lautens, M.; Edwards, L. G.; Tarn, W. J. Am. Chem. Soc. 1995,117, 10276. (b) Edwards, L. G.; Lautens, M.; Lough, A. J. J. Chem. Crystallogn 1997, 27, 471. 9. Carrefio, M. C ; Garcia Ruano, J. L.; Maestro, M. C ; Martin, L. M. Tetrahedron: Asymmetry 1993, 4, 111. 10. Houpis, I.; Molina, A.; Dorziotis, I.; Reamer, R. A.; Volante, R. R; Reider, R I. Tetrahedron Lett. 1997,35,7131. 11. Hiroi, K.; Hiroshi, O.; Arinaga, Y. Chem. Utt. 1995, 1099. 12. Mikolajczyk, M.; Drabowicz, J. Top. Stereochem.; Allinger, N. L.; Eliel, E. L.; Wilen, S. H., Eds.; New York, 1982, Vol. 13, p. 333. 13. Yoshikawa, M.; Murakami, T.; Ishikado, A.; Wakao, S.; Murakami, N.; Yamahara, J.; Matsuda, H. Heterocycles 1997. 46, 301. 14. (a) Lindberg, R; Norberg, R; Alminger, T.; Brandstrom, A. J. Med. Chem. 1986, 29,1327. (b) Im, W. B.; Sih, J. C ; Blakeman, D. R; Macgrath, J. R J. Biol. Chem. 1985,260, 4591. 15. Palfreyman, M. N. Potassium Channels and Their Modulators; Evans, J. M., Ed.; Taylor Francis: London, 1996, p. 55. 16. Mau, C. M. S.; Nakao, Y; Yoshida, W. Y; Scheuer, R J.; Kelly-Borges, M. J. Org. Chem. 1996, 61, 6302. 17. Pitchen, P.; France, C. F ; Mcfarlane, I. M.; Newton, C. N.; Thompson, D. M. Tetrahedron Lett. 1994,35, 485. 18. Jungheim, L. N.; Shepherd, T. A.; Boxter, A. J.; Burgess, J.; Hatch, S. D.; Lubberhusen, R; Wiskerchem, M.; Muesing, M. A. J. Med. Chem. 1996, 39, 96. 19. Solladi^,G.5yn//i^m 1981,185. 20. (a) Carreno, M. C. Chem. Rev. 1995, 95, 1717. (b) Hua, D. H. Adv. Heterocycl Nat. Prod. Synth. 1996,5,151. (c) Abu-Yousef, I. A.; Harpp, D. N. Sulfur Rep. 1997,20,1. (d) Kita, Y Phosphorus, Sulfur Silicon Relat. Elem. 1997,120-121, 145. (e) AUin, S. M. Organosulfur Chem. 1998,2,41. (f) Westwell, A. D.; Rayner, C. M. Organosulfur Chem. 1998, 2, 157. 21. (a) De Lucchi, O. Bull. Soc. Chim. Belg. 1988, 97, 679. (b) Annunziata, R.; Cinquini, M.; Cozzi, F; Farina, S.; Montanari, V. Tetrahedron 1987, 43, 1013. (c) Escher, B. M.; Haynes, R. K.; Kremmydas. S.; Ridley, D. D. J. Chem. Soc, Chem. Commun. 1988, 137. 22. Khiar, N.; Alonso, I.; Rodriguez, N.; Femandez-Mayoralas, A.; Jim^nez-Barbero, J.; Nieto, O.; Cano, F; Foces-Foces, C ; Martin-Lomas, M. Tetrahedron Lett. 1997,38, 8267. 23. (a) Davis, M.; Wu, W.-Y Aust. J. Chem. 1986, 39, 1165. (b) Danelon, G. O.; Mata, E. G.; Mascaretti, O. A. Tetrahedron Lett. 1993, 34, ISll. (c) Quallich, G. J.; Lackey, J. W. Tetrahedron Lett. 1990, 31, 3685. 24. (a) Shimazaki, M.; Takahashi, M.; Komatsu, H.; Ohta, A.; Kaji, K.; Kadoma, Y Synthesis 1992, 555. (b) Shimazaki, M.; Komatsu, H.; Ohta, A.; Kadoma, Y Synthesis 1992, 957. 25. (a) Holland, H. L. Chem. Rev. 1988. 88, 473. (b) Holland, H. L.; Brown, R M.; Larsen, B. G. Tetrahedron: Asymmetry 1994, 5, 1129. (c) Holland, H. L.; Rand, C. G.; Viski, R; Brown, F M. Can. J. Chem. 1991, 69, 1989. 26. Light, D. R.; Waxman, D. J.; Walsh, C. T. Biochemistry 1982, 21, 2490. 27. (a) Colonna, S.; Gaggero, N.; Pasta, P.; Ottolina, G. J. Chem. Soc, Chem. Commun. 1996, 2303. (b) Pasta, R; Carrea, G.; Holland, H. L.; Dallavalle, S. Tetrahedron: Asymmetry 1995, 6, 933. (c) Carrea, G.; Redigolo, B.; Riva, S.; Colonna, S.; Gaggero, N.; Battistel, E.; Bianchi, D. Tetrahedron:
112
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.
NOUREDDINE KHIAR et al. Asymmetry 1992, 3, 1063. (d) Ottolina, G.; Pasta, P.; Carrea, G.; Colonna, S.; Dallavalle, S.; Holland, H. L. Tetrahedron: Asymmetry 1995,6, 1375. Shaw, R D.; Hager, L. R J. Biol Chem. 1961, 256, 1626. Kobayashi, S.; Nakaro, M.; Kimura, T.; Schaap, A. R Biochemistry 1987, 26, 5019. (a) Colonna, S.; Gaggero, N.; Casella, L.; Carrea, G.; Pasta, R Tetrahedron: Asymmetry 1992, 3, 95. (b) Colonna, S.; Gaggero, N.; Manfredi, A.; Casella, L.; Gullotti, M.; Carrea, G.; Pasta, P. Biochemistry 1990,29, 10465. van Deurzen, M. R J.; van Rantwijk, F; Sheldon, R. A. Tetrahedron 1997, 55, 13183. Fu, H.; Kondo, H.; Ichikawa, Y; Look, G. C ; Wong, C. H. J. Org. Chem. 1992,57, 7265. Andersson, M.; Willetts, A.; Allenmark, S. J. Org. Chem. 1997,62, 8544. Ozaki, S.; Ortiz de Montellano, R R. / Am. Chem. Soc. 1995,117, 7056. Galzigna, L.; Rizzoli, V.; Schiappelli, M. R; Rigobello, M. P.; Scarpa, M.; Rigo, A. FreeRad. Biol. Med 1996, 20, 807. Miller, V. R; DePillis, G. D.; Ferrer, J. C ; Mauk, A. G.; Ortiz de Montellano, R R. / Biol. Chem. 1992, 267, 8936. Colonna, S.; Gaggero, N.; Carrea, G.; Pasta, R Tetrahedron Lett. 1994, 35, 9103. Colonna, S.; Gaggero, N.; Richelmi, C ; Carrea, G.; Pasta, R Gaz. Chim. Ital. 1995,125, 479. Lee, K.; Brand, J. M.; Gibson, D. T. Biochem. Biophys. Res. Commun. 1995, 272, 9. Holland, H. L.; Brown, F M.; Laksmaiah, G.; Larsen, B. G.; Patel, M. Tetrahedron: Asymmetry 1997, 8, 683. Madesclaire, M.; Fauve, A.; Metin, J.; Carpy, A. Tetrahedron: Asymmetry 1990, 7, 311. (a) Holland, H. L.; Brown, F M.; Larsen, B. G. Tetrahedron: Asymmetry 1995. 6, 1561. (b) Abushanab, E.; Reed, D.; Suzuki, F ; Sih, C. J. Tetrahedron Lett. 1978, 37, 3415. Buist, R H.; Marecak, D. M. J. Am. Chem. Soc. 1992. 114, 5073. (a) Beecher, J.; Brackenridge, 1.; Roberts, S. M.; Tang, J.; Willetts, A. J. J. Chem. Soc, Perkin Trans. 1 1995. 1641. (b) Tang, J.; Brackenridge, I.; Roberts, S. M.; Beecher, J.; Willetts, A. J. Tetrahedron 1995. 57, 13217. Allen, C. C. R.; Boyd, D. R.; Dalton, H.; Sharma, N. D.; Haughley, S. A.; McMordie, R. A. S.; Mcmurray, B. T.; Sheldrake, G. N.; Sproule, K. J. Chem. Soc, Chem. Commun. 1995, 119. (a) Pitchen, R; Kagan, H. B. Tetrahedron Lett. 1984. 24, 1049. (b) Kagan, H. B.; Diter, R Organosulfur Chem. 1998,2, 139. Di Furia, F ; Modena, G.; Seraglia, R. Synthesis 1984, 325. Johnson, R. A.; Sharpless, K. B. Catalytic Asymmetric Synthesis', Ojima, I., Ed.; VCH: New York, 1993, p. 227. Davis, F A.; Chen, B. C. Chem. Rev. 1992, 92, 919. Pitchen, R; Deshmukh, M. N.; Dunach, E.; Kagan, H. B. J. Am. Chem. Soc 1984. 106, 8188. Dunach, E.; Kagan, H. B. New J. Chem. 1985, 9, 1. (a) Zhao, S. H.; Samuel, O.; Kagan, H. B. Tetrahedron 1987. 43, 5135. (b) Zhao, S. H.; Samuel, O.; Kagan, H. B. Org. Synth. 1989, 68, 49. Kagan, H. B.; Rebiere, F Synlett 1990, 643. Brunei, J. M.; Kagan, H. B. Bull. Soc Chim. FK 1996. 755. 1109. Bortolini, O.; Di Furia, F ; Licini, G.; Modena, G.; Rossi, M. Tetrahedron Lett. 1986. 27, 6257. Bortolini, O.; Di Furia, F ; Licini, G.; Modena, G. Rev. Heteroatom. Chem. 1988, 7, 66. (a) Page, R C. B.; Namwindwa, E. S.; Klair, S. S.; Westwood, D. Synlett 1990, 547. (b) Page, R C. B.; Namwindwa, E. S. Synlett 1991, 80. (c) Page, R C. B.; Wilkes, R. D.; Barkley, J. V.; Witty, M. J. Synlett 1994,547. (d) Page, R C. B.; Wilkes, R. D.; Witty, M. J. Org. Prep. Proced. Int. 1994. 26, 702. (e) Page, R C. B.; Wilkes, R. D.; Namwindwa, E. S.; Witty, M. J. Tetrahedron 1996. 52, 2125. (f) Aggarwal, V. K.; Evans, G.; Moya, E.; Dowden, J. J. Org. Chem. 1992, 57, 6390. Davis, F A.; Kern, J. R.; Kurtz, L. J.; Pfister, J. R. J. Am. Chem. Soc 1988, 770, 7873. (a) Griffiths, S. L.; Perrio, S.; Thomas, S. E. Tetrahedron: Asymmetry 1994, 5. 545. (b) Griffiths, S. L.; Perrio, S.; Thomas, S. E. Tetrahedron: Asymmetry 1994, 5, 1847.
Synthesis of Chiral Sulfoxides
113
60. Diter, P.; Samuel, O.; Taudien, S.; Kagan, H. B. Tetrahedron: Asymmetry 1994, J, 549. 61. (a) Komatsu, N.; Nishibayashi, Y; Sugita, T.; Uemura, S. Tetrahedron Lett. 1992. 33, 5391. (b) Komatsu, N.; Hashizume, M.; Sugita, T.; Uemura, S. J. Org. Chem. 1993, 58, 4529. 62. Colombo, A.; Marturano, G.; Pasini, A. Gazz. Chim. Ital. 1986,116, 35. 63. Nakajima, K.; Kojima, M.; Fujita, J. Chem. Lett. 1986, 1483. 64. (a) Nakajima, K.; Sasaki, C ; Kojima, M.; Aoyama, T.; Ohba, S.; Sayto, Y; Fujita, J. Chem. Lett. 1987, 2189. (b) Sasaki. C ; Nakajima, K.; Kojima, M.; Fujita, J. Bull. Chem. Soc. Jpn. 1991. 64, 1318. 65. Palucki, M.; Hanson, R; Jacobsen, E. N. Tetrahedron Lett. 1992. 33, 7111. 66. Noda, K.; Hosoya, N.; Irie, R.; Yamashita, Y; Katsuki, T. Tetrahedron 1994. 50, 9609. 67. Noda, K.; Hosoya, N.; Yanai, K.; Irie. R.; Katsuki, T. Tetrahedron Lett. 1994. 35,1887. 68. Bolm, C ; Bienewald, R Angew. Chem. Int. Ed. Engl. 1995. 34, 2640. 69. Schenk, W. A.; Frisch, J.; Adam, W.; PrechU, F Angew. Chem. Int. Ed. Engl. 1994,33, 1609. 70. Schenk, W. A.; DUrr, M. Chem. Eur. J. 1997, 3,713. 71. Davis, F A.; Sheppard, A. C. Tetrahedron 1995,45, 5703. 72. Davis, F A.; Reddy, R. T.; Weismiller, M. C. J. Am. Chem. Soc. 1989, III, 5964. 73. Davis, F A.; Towson, J. C ; Weismiller, M. C ; Lai, S.; Carroll, R J. J. Am. Chem. Soc. 1988,110, 8477. 74. Davis, F A.; Weismiller, M. C ; Murphy, C. M.; Reddy, R. T.; Chen, B. C. J. Org. Chem. 1992, 57,1214. 75. Davis, F A.; Reddy, R. T.; Han, W.; Carroll, R J. J. Am. Chem. Soc. 1992. 114, 1428. 76. Rossi, C ; Faure, A.; Madasclaire, M.; Roche, D.; Davis, F A.; Reddy, R. T. Tetrahedron: Asymmetry 1992, 3,629. 77. Hamann, H. J.; Hoft, E.; Mostowicz, D.; Mishnev, A.; Urbanczyk-Lipkoswska, Z.; Chmielewski, M. Tetrahedron 1997,53, 185. 78. Adam, W.; Korb, M. N.; Roschmann, K. J.; Saha-MoUer, C. R. J. Org. Chem. 1998. 63, 3423. 79. (a) Adam, W.; Hoch, U.; Lazarus, M.; Saha-Moller, C. R.; Schreier, R J. Am. Chem. Soc. 1995, 117, 11898. (b) Adam, W; Korb, M. N. Tetrahedron: Asymmetry 1997, S, 1131. 80. (a) Andersen, K. K. Tetrahedron Lett. 1962,93. (b) Andersen, K. K.; Gaffield, W.; Papanikolaou, N. E.; Foley, J. W; Perkins, R. I. J. Am. Chem. Soc. 1964, 86, 5637. (c) Andersen, K. K. Int. J. Sulfur Chem. W1\, 6, 69. 81. (a) Alexrod, M.; Bickert, R; Jacobus, J.; Green, M.; Mislow, K. J. Am. Chem. Soc. 1968, 90,4835. (b) Nishio, M.; Nishihata, K. J. Chem Soc, Chem. Commun. 1970, 1485. (c) Juge, S.; Kagan, H. B. Tetrahedron Utt. 1975, 2733. 82. Mislow, K.; Temay, A. L.; Melillo. J. T. J. Am. Chem. Soc. 1963, 85, 2329. 83. Mislow, K.; Green, M. M.; Laur, P.; Melillo, J. T.; Simmons, T.; Temay, A. L. J. Am. Chem. Soc. 1965, 87,1958. 84. (a) Mioskowski, C ; Solladi6, G. Tetrahedron 1980, 36, 227. (b) Mioskowski, C ; Solladi^, G. Tetrahedron Lett. 1975, 3341. (c) Solladi6. G.; Hutt, J.; Girardin, A. Synthesis 1987, 173. 85. Andersen, K. K. The Chemistry of Sulfones and Sulfoxides; Patai, S.; Rappaport, Z.; Stirling, C, Eds.; John Wiley and Sons: New York, 1988, Chapter 3. Drabowicz, J.; Kielbasinski, P.; Mikolajczyk, M. Ibid. Chapter 8. Posner, G. M. Ibid. Chapter 16. 86. Rebifere, F ; Samuel, O.; Ricard, L.; Kagan, H. B. J. Org. Chem. 1991, 56, 5991. 87. Evans, D. A.; Faul, M. M.; Colombo, L.; Bisaha, J. J.; Clardy, J.; Cherry, D. J. Am. Chem. Soc. 1992,114, 5977. 88. Whitesell, J. K.; Wong, M.-S. J. Org. Chem. 1994, 59, 597. 89. Femdndez, L; Khiar, N.; Llera, J. M.; Alcudia, F J. Org. Chem. 1992,57, 6789. 90. Rebidre, F ; Kagan, H. B. Tetrahedron Lett. 1989, 30, 3659. 91. (a) Rebi^re, F ; Riant, O.; Kagan, H. B. Tetrahedron: Asymmetry 1990, 1, 199. (b) Devant, R.; Malher, U.; Braon, M. Chem. Ber 1988,121, 397. 92. Van Woerden, H. F Chem. Rev. 1963, 557 and references therein.
114
NOUREDDINE KHIAR et al.
93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103.
Mikolajczyk, M.; Drabowicz, J. Synthesis 1974, 124. Rebidre, E; Riant, O.; Ricard, L.; Kagan, H. B. Angew. Chem. Int. Ed. Engl. 1993, 32, 568. Whitesell, J. K.; Wong, M.-S. / Org. Chem. 1991,56,4552. Whitesell, J. K.; Chen, H. H.; Lawrence, R. M. J. Org. Chem. 1985, 50,4663. Zang, Y. S.; Talalay, R; Cho, C. G.; Posner, G. H. Proc. Natl. Acad. Sci. USA 1992, 89, 2399. Wudl, F; Lee, T. B. K. J. Am. Chem. Soc. 1973, 95, 6349. Benson, S. C ; Snyder, J. K. Tetrahedron Lett. 1991, 32, 5885. Evans, D. A.; Kaldor, S. W; Jones, T. K.; Clardy, J.; Stout, T. J. / Am. Chem. Soc. 1990,112,7001. Evans, D. A.; Mathre, D. J.; Scott, W. L. J. Org. Chem. 1985, 50, 1830. Evans, D. A.; Gage, J. R. Org. Synth. 1989, 68, 77. (a) Mioskowski, C ; Solladid, G. J. Chem. Soc, Chem. Commun. 1977, 162. (b) Solladi^, G.; Matloubi-Moghadam, F. J. Org. Chem. 1982, ^7, 91. Alonso, R.; Garcia Ruano, J. L.; Noheda, R; Zarzuelo, M. M. Tetrahedron: Asymmetry 1995, 6, 1133. Bueno, A. B.; Carreno, M. C ; Garcia Ruano, J. L.; Gdmez Array^s, R.; Zarzuelo, M. M. J. Org. Chem. 1997, 62, 2139. (a) Yoshikuni, Y. Trends Glycosci. Glycotechnol. 1991, 3, 184. (b) Yamamoto, T.; Aoyagi, T.; Nakamura, H.; litakay, Y / Antibiot. 1989, 42, 1008. Oppolzer, W; Froelich, O.; Wiaux-Zamar, C ; Bernardinelli, G. Tetrahedron Lett. 1997,38,2825. Oppolzer, W. Pure Appl. Chem. 1990, 62,1241. (a) Davis, F A.; Szewczyk, J. M.; Reddy, R. E. J. Org. Chem. 1996, 61, 2222. (b) Davis, F A.; Portonovo, R S.; Reddy, R. E.; Chiu, Y.-H. J. Org. Chem. 1996, 61,440; (c) Hua, D. H.; Lagneau, N.; Wang, H.; Chen, J. Tetrahedron: Asymmetry 1995, 6, 349. (d) Hose, D. R. J.; Mahon, M. F ; Molloy, K. C ; Raynham, T.; Wills, M. / Chem. Soc, Perkin Trans. 11996, 691. (e) Mikolajczyk, M.; Lyzwa, R; Drabowicz, J.; Wieczorek, M. W; Blaszczyk, J. Chem. Commun. 1996, 1503. (0 Fujisawa, T.; Kooriyama, Y; Shimizu, M. Tetrahedron Lett. 1996,37, 3881. (g) Tang, T. P; Ellman, J. A. J. Org. Chem. 1999, 64, 12. (h) Cogan, D. A.; Ellman, J. A. J. Am. Chem. Soc 1999, 121, 268. (i) Garcia Ruano, J. L.; Fernandez, I.; del Prado Catalina, M.; Hermoso, J. A.; Sanz-Aparicio, J.; Martinez-Ripoll, M. J. Org. Chem. 1998, 63, 7151. Khiar, N.; Fernandez, L; Alcudia, F Tetrahedron Lett. 1993, 34, 123. Andersen, K. K.; Bujnicki, B.; Drabowicz, J.; Mikolajczyk, M.; O'Brien, J. B. J. Org. Chem. 1984, 49, 4070. (a) Ridley, D. D.; Smal, M. A. J. Chem. Soc, Chem. Commun. 1981, 505. (b) Ridley, D. D.; Smal, M. A. Aust. J. Chem. 1982, 35, 495. Hultin, R G.; Earle, M. A.; Sudharshan, M. Tetrahedron 1997, 53, 14823. Young, J. H.; Herrmann, R. Tetrahedron Lett. 1986, 27, 1493. Netscher, T.; Prinzbach, H. Synthesis 1987, 683. Femdndez, I.; Khiar, N.; Roca, A.; Benabra, A.; Alcudia, A.; Espartero, J. L.; Alcudia, F Tetrahedron Lett. 1999,40, 2029. Heering, A.; Jaspers, M.; Schwermenn, I. Chem. Ben 1979,112, 2903. Berry, R. S. J. Chem. Phys. 1960, 32, 933. Casey, M.; Manage, A. C ; Nezhat, L. Tetrahedron Lett. 1988, 29, 5821. Pyne, S. G.; Boche, G. J. Org. Chem. 1989, 54, 2663. Khiar, N.; Femdndez, I.; Alcudia, F Tetrahedron Lett. 1994, 35, 5719. Khiar, N.; Fernandez, I.; Alcudia, F; Hua, D. H. Tetrahedron Utt. 1993, 34, 699. Argoudelis, A. D.; Herr, R. R. Antimicrob. Agents Chemother 1962, 780. Higashide, E.; Hasegawa, T.; Shibata, M.; Mizuno, K.; Akaike, H. Takeda Kenkyusho Nempo 1966, 25, 1; Chem. Abstr 1967, 66, 54238. (a) Wiley, R F ; MacKellar, F A. J. Am. Chem. Soc 1970, 92, 417. (b) Wiley, P F ; MacKellar, F A. J. Org. Chem. 1976, 41, 1858.
104. 105. 106. 107. 108. 109.
110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125.
Synthesis of Chiral Sulfoxides
115
126. (a) Parry, R. J.; Li, Y; Gomez, E. E. J. Am. Chem. Soc. 1992,114, 5946. (b) Parry, R. J.; Hoyt, J. C ; Li, Y. Tetrahedron Utt. 1994, 35, 7497. 127. Owen, S. R; Dietz, A.; Camiene, G. W. Antimicrob. Agents Chemother 1962, 772. 128. Ottenheijm, H. C ; liskamp, R. M. J.; van Nispen, S. P J. M.; Boot, H. A.; Tijhuis, M. W. J. Org. Chem. 1981,46, 3273. 129. Helquist, R; Shekhani, M. S. J. Am. Chem. Soc. 1979,101, 1057. 130. Hwang, D. R.; Helquist, R; Shekhani, M. S. J. Org. Chem. 1985, 50, 1264. 131. Hua, D. H.; Khiar, N.; Zhang, E; Lambs, L. Tetrahedron Lett. 1992, 33, 7751. 132. El Ouazzani, H.; Khiar, N.; Fernandez, I.; Alcudia, F J. Org. Chem. 1997, 62, 287. 133. (a) Freedman, H. A.; Fox, A. E.; Shavel, J.; Morrison, G. C. Proc. Soc. Exp. Biol. Med. 1972,139, 909. (b) Husberg, B.; Penn, I. Proc. Soc. Exp. Biol. Med. 1974,145, 669. (c) Fox, A. E.; Gawlak, D. L.; Ballantyne, D. L., Jr.; Freedman, H. H. Transplantation 1973, 75, 389. (d) Briziarelli, G.; Abruptyn, D.; Tomaben, J. A.; Scwartz, E. Toxicol. Appl. Pharmacol. 1976, 36, 49. (e) Krazmer, J. S.; Daddona, P E.; Dalke, A. R; Kelly, W. N. Biochem. Pharmacol. 1983, 32, 805. 134. (a) Garcia Ruano, J. L.; Pedregal, C.; Rodriguez, J. H. Tetrahedron 1987, 43,4407. (b) Alcudia, F; Fernandez, L; Llera, J. M.; Trujillo, M.; Zorrilla, F Magn. Reson. Chem. 1988, 26, 687. (c) Alvarez-Ibarra, C.; Cuervo-Rodriguez, R.; Femdndez-Monreal, M. C ; Ruiz, M. P. J. Org. Chem. 1994, 59, 7284. 135. Kunieda, N.; Nokami, J.; Kinoshita, M. Chem. Utt 1974, 379. 136. Hiroi, K.; Matsuyama, N. Chem. Lett. 1986, 65. 137. Carreno, M. C.; Garcfa Ruano, J. L.; Rubio, A. Tetrahedron Lett. 1987, 28, 4861. 138. (a) Banfi, L.; Colombo, L.; Gennari, C ; Annunziata, R.; Cozzi, F Synthesis 1982, 829. (b) Wills, M.; Linney, I. D.; Lacey, C ; Mahon, M. F ; Molloy, K. C. Synlett 1991, 836. 139. Corey, E. J.; Knapp, S. Tetrahedron Lett. 1976, 3667. 140. Eraser, R. R.; Petit, M. A.; Saunders, J. K. J. Chem. Soc, Chem. Commun. 1971, 1450. (b) Davis, F A.; Billmers, J. M. / Org. Chem. 1983,48, 2672. 141. Deshmukh, M.; Dunach, E.; Juge, S.; Kagan, H. B. Tetrahedron Lett. 1984, 25, 3467. 142. Gautier, N.; Noiret, N.; Nugier-Chauvin, C ; Patin, H. Tetrahedron: Asymmetry 1997, 8, 501. 143. (a) G^nard, S.; Patin, H. Bull. Soc. Chim. Fr 1991,128, 397. (b) Gruiec, R.; Noiret, N.; Patin, H. Bull. Soc. Chim. Fr 1994,131, 699. 144. Arroyo-G(3mez, Y; L6pez-Sastre, J. A.; Rodriguez-Amo, J. F ; Sdntos-Garcia, M.; Sanz-Tejedor, M. A. J. Chem. Soc, Perkin Trans. 1 1994, 2177. 145. Diaz Bueno, N.; Alonso, I.; Carretero, J. C. J. Am. Chem. Soc 1998,120, 7129. 146. (a) Damm, W.; Giese, B.; Artung, J.; Hasskerk, T.; Houk, K. N.; Hutter, O.; Zipse, H. J. Am. Chem. Soc 1992,114, 4067. (b) Porter, N. A.; Rosenstein, I. J.; Breyer, R. A.; Bruhnke, J. D.; Wu, W.; McPhail, A. T. J. Am. Chem. Soc 1992, 114, 7664. (c) Porter, N. A.; Bruhnke, J. D.; Wu, W; Rosenstein, L J.; Breyer, R. A. / Am. Chem. Soc 1991,113, 7788. 147. (a) Renaud, P; Bjorup, P; Carrupt, P-A.; Schenk, K.; Schubert, S. Synlett 1992, 211. (b) Stack, J. G.; Curran, D. P; Rebek, J., Jr.; Ballester, P J. Am. Chem. Soc 1991,113, 5918. (c) Renaud, P; Schubert, S. Synlett 1990, 625. 148. Tom, T; Watanabe, Y; Tsusaka, M.; Ueno, Y J. Am. Chem. Soc 1993,115, 10464. 149. Mase, N.; Watanabe, Y; Ueno, Y; Toru, T. J. Org. Chem. 1997, 62, 7794. 150. Mase, N.; Watanabe, Y; Toru, T J. Org. Chem. 1998, 63, 3899. 151. Garcia Ruano, J. L.; Fernandez, I.; del Prado Catalina, M.; Alcudia Cruz, A. Tetrahedron: Asymmetry 1996, 7, 3407. 152. (a) Backes, B. J.; Ellman, J. A. J. Am. Chem. Soc 1994, 776,11171. (b) Backes, B. J.; Virgilio, A. A.; Ellman, J. A. J. Am. Chem. Soc 1996,118, 3055. 153. Liu, G.; Cogan, D. A.; Ellman, J. A. J. Am. Chem. Soc 1997, 779, 9913. 154. Cogan, D. A.; Liu, G.; Kim, K.; Backes, B. J.; Ellman, J. A. J. Am. Chem. Soc 1998,120, 8011.
This Page Intentionally Left Blank
CYCLIC SULFOXIDES AS CHIRAL AUXILIARIES IN ASYMMETRIC SYNTHESIS
Steven M. Allin and Philip C. Bulman Page
I. Introduction II. Six-Membered Sulfur Heterocycles A. 1,3-Dithiane 1-Oxide B. 1,3-Dithiane Dioxides III. Five-Membered Sulfur Heterocycles A. 1,3-Dithiolane-l-Oxide References
117 118 118 147 148 148 150
I. INTRODUCTION The stereocontrolled introduction of chirality into organic molecules remains an important challenge to the synthetic organic chemist. To produce, at will, only one of two possible enantiomeric compounds is more cost-effective than traditional resolution procedures, and considerably more elegant.
Advances in Sulfur Chemistry Volume 2, pages 117-153. Copyright © 2000 by JAI Press Inc. Allrightsof reproduction in any form reserved. ISBN: 0-7623-0618-1 117
118
STEVEN M. ALLIN and PHILIP C. BULMAN PAGE
While many enantioselective processes utilizing optically pure acyclic sulfoxide derivatives have been studied extensively among the synthetic community, their cyclic counterparts have, by comparison, received little attention. Cyclic sulfoxide systems which have received the most attention are five- and six-membered ring derivatives, and it is the chemistry of such compounds which forms the subject of this discussion. The applications described below outline the effectiveness of cyclic chiral sulfoxides as stereocontrol elements, and highlight the ready removal of the sulfoxide group after its contribution to the synthetic scheme. In all cases, the sense of stereochemical induction can be rationalized and predicted on the basis of steric, stereoelectronic, and/or chelation control factors.
II. SIX-MEMBERED SULFUR HETEROCYCLES A. 1,3-Dithiane 1-Oxide
Since 1987 our group has been concerned with the design, synthesis, development, and more recently the application of dithiane 1-oxide derivatives as asymmetric building blocks for organic synthesis. This review focuses on the development of highly diastereoselective reactions, principally carried out at the acyl side chain of 2-acyl dithiane 1-oxide derivatives (1). In the early stages of the project, we reasoned that the sulfoxide unit might be expected to influence the transition state geometry of the 2-acyl side chain, perhaps by chelation to a metal counterion, and hence control the stereochemistry of a wide range of functional group transformations. Indeed, a chelation control model of the reactivity of the 2-acyl dithiane 1-oxide systems has allowed us to rationalize, and predict, the stereochemical outcome of most of the reactions studied so far. These predictions have, in many cases, been confirmed by X-ray structure determination of the relative stereochemistries within product structures.^""* The dithiane 1-oxide (DiTOX) moiety (2) fulfills the following criteria for an ideal chiral auxiliary: 1. DiTOX and its 2-substituted derivatives are readily prepared, generally stable, and relatively inexpensive. 2. No experimentally difficult chemistry is involved in their preparation. "O
,O
^5-|iL-«' Syrj-(1)
' ? ^1 O
M-i^' Anti'(l)
Cyclic Sulfoxides
119 "O
o (RH2)
'Q
o (5)-(2)
3. DiTOX systems are amenable to stereoselective preparation for both sulfoxide enantiomers. 4. The DiTOX system has been shown to induce high levels of stereoselectivity for a range of reaction types based on carbonyl group reactivity. 5. The DiTOX auxiliary can be readily removed, in high yield, without loss of stereochemical integrity at the newly created asymmetric center(s). 6. Both absolute configurations are available. Interestingly, deprotection (hydrolysis) of the heterocyclic auxiliary exposes a synthetically useful carbonyl group. This is possible in our system since the auxiliary is bonded to the carbonyl group by a carbon atom rather than a heteroatom. Such hydrolyses are well established for 1,3-dithiane derivatives as a result of their ubiquity as synthons for umpolung reactivity of the carbonyl group (Fig. 1).^ Preparation ofRacemic l-Acyl-l-alkyl-l ^3-dithiane 1-Oxide Systems Our early studies centered on diastereoselective transformations of racemic 2-acyl-2-alkyl-l,3-DiTOX systems, typically prepared as shown in Scheme 1. Acyl dithianes (5) may be prepared by reaction of the 2-lithio derivative of the 2-alkyl dithiane (3) with a desired aldehyde to give alcohols (4) which are oxidized using Swern conditions.^ Racemic sulfur oxidation to yield {±)-anti and {±)-syn isomers, 6 and 7, respectively, is accomplished with aqueous sodium periodate. The syn and anti diastereoisomers are readily separated by flash column chromatography; the major {anti) isomer generally being the more polar. Interestingly, the anti isomers also display a discrete signal in their ^H NMR spectra (ca. 5 1.7 ppm) corresponding to dithiane ring protons at C-5. This signal appears at higher field for the syn isomer and is sometimes masked by other resonances.
Figure 1. Umpolung reactivity of 13-dithiane.
120
STEVEN M. ALLIN and PHILIP C. BULMAN PAGE OH
o
^ k J ^ ^ * RkA^|:7 R
R
(6), (±yanti
(7), (±ysyn
Reagents: i, BuLi, THF,-78; °C; aldehyde; ii, DMSO, TFAA, CH2CI2, -50 °C; EtsN; iii,NaI04,MeOH. Scheme 1.
Unless otherwise stated, raceniic 2-acyl-2-alkyl-DiTOX derivatives were used as substrates for the applications described in this review. The preparation of enantiomerically pure 2-acyl-2-alkyl-DiTOX systems are described in a subsequent section. Development of the DiTOX Asymmetric Building Block Diastereoselective control in the addition ofgrignard reagents to ketones.^
The addition of organometallic reagents to ketones bearing a chiral grouping directly attached to the carbonyl group has been extensively studied by others.^ Stereoselectivities are often not high^ unless one of the substituents adjacent to the ketone is capable of chelation with the organometallic reagent.^ We expected that the sulfoxide unit of the DiTOX auxiliary could influence the course of organometallic attack at the carbonyl group of a 2-acyl-l,3-dithiane 1-oxide system by chelation control. Thus, the reactions of syn and anti 2-propionyl-2-methy 1-1,3dithiane 1-oxide substrates, 8 and 10, respectively (prepared as described in Scheme 1), with methylmagnesium iodide were investigated (Scheme 2). The diastereoisomeric ratios for the product alcohols are summarized in Table 1. As expected, the diastereoselectivities show a dependence on temperature and solvent. ^^ The major product diastereoisomers, 9 and 11, respectively, are shown in Scheme 2. For the syn substrate 8, the approach of the organometallic nucleophile is controlled by the steric bias imposed by the bulky DiTOX ring (Fig. 2). The nucleophile approaches the prochiral carbonyl group from the direction of the
Cyclic Sulfoxides
121
MeMgl
.! Men \ ^ ^
MeMgl
(10) Scheme 2.
relatively small 2-methyl substituent, giving rise to excellent diastereoselectivities at low temperatures (entries b, c). For the anti substrate 10, although the chelated transition state relies solely on the 2-methyl substituent to exert any steric hindrance toward the approach of the Grignard reagent (Fig. 3), good stereoselectivities are achieved at low temperature (entry h). Literature precedent indicated that THF was the solvent of choice for stereoselective Grignard additions^^ our systems were found to behave accordingly, with higher product diastereoselectivities observed in THF than in diethyl ether. Although most reactions of syn and anti substrates listed in Table 1 gave products corresponding to the Cram-type chelated transition states described above, in one instance substrate 8 gave a product ratio which violated the expected pattern (entry
Table 1, Addition of Methylmagnesium Iodide to syn and anti 2-Propionyl-2-methyl-l,3-dithiane 1-Oxide Substrates Entry a
b c
d e
f g h i
Solvent
Temp r O
Product
Yield (%)
Ratio of Isomers^
8-syn 8 8 8 8
THF THF THF Et20 Et20
25 -20 -78 -20 -78
9 9 9 9 9
70 95 95 96 33
4: 1 ca. 25 : 1 Exclusive^ 1 : 1.4 1 :1
10-ant/ 10 10 10
THF THF THF Et20
25 -20 -78 -78
11 11 11 11
92 96 95 96
3:1 7:1 15: 1 3:1
Substrate
fslotes: ^Determined by 250-MH2 ^H NMR. ^ Other isomer undetectable by HPLC or NMR.
122
STEVEN M. ALLIN and PHILIP C. BULMAN PAGE Mg:;;0
C\" ^Me
<
1 Nu©
Figure 2. Chelated transition-state model for syn substrate.
Nu© M9
^
o'/mTr-^
^^"^ Me
Me
Figure 3. Chelated transition-state model for anti substrate.
d). In this case an alternative chelated transition state containing an axial sulfoxide may apply (Fig. 4), or possibly an open transition state^ or dipolar system (Fig. 5).^^ The major product diastereoisomer 12 from addition of methy Imagnesium iodide to syn substrate 8 was isolated by recrystallization, and the structure solved by X-ray analysis. The structure was found to be in accordance with the expected chelated transition state (Fig. 2) and approach of the nucleophile from the least hindered face of the carbonyl group. Highly diastereoselective reduction of ketones.^ The stereoselective reduction of ketones has been previously reported using chirally modified hydride
(12) ?;:^Mg
.Mg
Me
Me k
Figure 4. Syn substrate—axial sulfoxide.
Cyclic Sulfoxides
1 23 ~0
c
()l>Me
^b Mg
Figure 5. Syn substrate—dipolar transition state.
reagents^^ and by chiral auxiliary approaches.^"^ We turned our attention to this important synthetic transformation and were pleased to fmd that racemic 2-acyl-2alkyl-l,3-dithiane 1-oxide substrates underwent reduction to the corresponding secondary alcohols with extremely high levels of diastereoselectivity (Scheme 3, Table 2)? In our study, the reduction reactions were carried out using di-isobutyl aluminum hydride (DIBAL) or DIBAL/ZnCl2 mixtures in THF at low temperature (-78 °C). These reaction conditions and reagents were known to provide high levels of stereoselectivity in reduction of acyclic P-ketosulfoxides.^^ Transition states for reduction according to our usual model of chelation-controlled 2-acyl 1,3-dithiane 1-oxide reactivity, together with steric approach control were proposed to rationalize the high levels of observed stereoselectivity. Previous work by Solladie suggests that ketone reduction by the DIBAL7ZnCl2 system does indeed involve such chelated transition states.^^ As can be seen from Table 2, very high stereoselectivities could be observed for both syn and anti substrates, depending on the 2-alkyl substituent (R). In the absence of ZnCl2, a nonchelated chairlike transition state was anticipated, following the Solladie model, with intramolecular hydride transfer. This process was expected to lead to an opposite sense of selectivity to that observed for the chelation-controlled model (with DIBAL/ZnC^). This reversal in stereoselectivity was indeed observed "O
O
"O
Syn
Scheme 3.
OH
STEVEN M. ALLIN and PHILIP C. BULMAN PAGE
124
Table 2, Diastereoselective Reduction of 2-Acyl-2-alkyl-1,3-dithiane 1-Oxide Substrates Entry
Ratio of Isomers^
R
a b c d e f
syn syn syn syn syn syn
Me Me Et Et Ph Ph
DIBAL DIBAL/ZnCl2 DIBAL DIBAiyZnCl2 DIBAL DIBAL/ZnCl2
45 75 25 83 25 83
exclusive 7:1 exclusive exclusive exclusive 1.7:1
g h i
and anti anti anti anti anti anti anti
Me Me Et Et 'Pr •Pr Ph Ph
DIBAL DIBAL/ZnCl2 DIBAL DIBAL/ZnCl2 DIBAL DIBALyZnCl2 DIBAL DIBAL/ZnCl2
40 85 50 42 21 81 42 80
exclusive exclusive 10.5:1 36:1 exclusive exclusive exclusive 6.3:1
J k 1 m n
Reagent
Yield (%)
Substrate
Comment Opposite sense to a Opposite sense to c Same sense as e Opposite sense to g Same sense as i Opposite sense to k Opposite sense to m
Note: ^Determined by ^H and/or ^^C NMR spectroscopy; exclusive diastereoselectivity indicates that minor isomer was not detected.
for some substrates (see Table 2); however, in several cases the selectivity was found to have the same sense as the chelation-controlled method. While we cannot fully explain this rather curious feature using our present rationale, the effect of the size of the R group on transition state conformation may be a factor. We have not speculated further, as our simple model of acyl DiTOX reactivity does not take into account the role of solvent, electrostatic, and aggregation effects; it has nonetheless remained a useful predictive working model throughout our studies. Figures 6 (syn substrates, via major reactive conformation 13) and 7 (anti substrates, via major reactive conformation 14) were postulated as transition-state
(13) Figure 6.
Cyclic Sulfoxides
125
(14) Figure 7.
"9 pjOH
(15)
(16)
models to explain the highly stereoselective intramolecular hydride transfer in DiTOX substrates. The sense of stereoselectivity expected through application of our transition-state models was borne out by single crystal X-ray structure determination. Structure 15 shows the product of DIBAL reduction of yy/z-2-propionyl-2methyl-l,3-dithiane 1-oxide (entry a, Table 2), and 16 shows the product of DIBAL/ZnCl2 reduction of aAzn'-2-propionyl-2-isopropyl-DiTOX (entry 1, Table 2). Stereoselective conjugate addition of lithium organocuprate reagents to a^^'Unsaturated 2'acyl-2-alkyl'1,3-dithiane 1-oxide substrates. A number of methods for the asymmetric control of conjugate addition of organocopper reagents to a,P-unsaturated acyl derivatives have been developed.^^^^ We were able to demonstrate interesting levels of diastereoselectivity in conjugate addition of lithium organocuprate reagents to racemic 2-acyl-2-alkyl DiTOX substrates (Scheme 4). As a result of our earlier investigations, we anticipated that rapid complexation should occur between the organometallic reagent and the enone substrate involving
R'gCuLi
Scheme 4.
STEVEN M. ALLIN and PHILIP C. BULMAN PAGE
126
Table 3. Conjugate Addition to a,P-Unsaturated 2-Acyl-2-alkyl-1,3-dithiane 1 -Oxide Substrates Substrate syn syn syn syn syn syn anti anti anti anti anti anti
Yield (%)
Reagent
Ratio of Isomer^
Me Me Et Et Ph Ph
Bu2CuLi Ph2CuLi Bu2CuLi Ph2CuLi Bu2CuLi Ph2CuLi
84 95 73 67 80 70
4.3:1 2.3:1 10.5:1 3.2:1 6.6:1 3.4:1
Me Me Et Et Ph Ph
Bu2CuLi Ph2CuLi Bu2CuLi Ph2CuLi Bu2CuLi Ph2CuLi
84 83 50 87 75 60
2.0:1 1.2:1 2.0:1 4.0:1 2.0:1 2.3:1
Note: ^Determined by ^H and/or '^C NMR spectroscopy.
bidentate coordination of the sulfoxide and carbonyl group oxygen atoms to the metal counterion. Table 3 shows our results for syn and anti DiTOX substrates. The proposed chelated transition-state models are analogous to those previously presented. Syn substrates containing axial sulfoxide units would not be expected to show much selectivity; in the case of anti substrates, no chelation is possible in conformations containing axial sulfoxides. For syn substrates, in the equatorial sulfoxide conformation 17, the bulk of the dithiane ring effectively shields one face of the 7i-system, the other face being exposed unless a very large 2-alkyl group is present. For anti systems such as 18 in the equatorial sulfoxide conformation, only the 2-alkyl substituent is available to hinder reagent approach, and selectivity should rise as this group becomes larger.
4°
- C¥ S
(17)
Cyclic Sulfoxides
12/
^ R
(18)
While such transition-state models have helped us to rationalize the patterns of selectivity observed in other reactions of 2-acyl-l,3-dithiane 1-oxide substrates, such clear trends are not found in conjugate addition reactions (Table 3). One simple explanation for the poorer levels of stereoselectivity may be bond rotation within the acyl substituent, allowing the enone moiety to attain conformations other than those shown in 17 and 18. Stereoselective functionalization of enolates derived from 2-acyl-2' all
STEVEN M. ALLIN and PHILIP C. BULMAN PAGE
128
Table 4. Alkylation of Enolates of 2-Acyl-2-alkyl-l ,3-dithiane 1 -Oxide Substrates Substrate
Temp (°C)
syn
Me
syn
Et
syn
'Pr
syn
Ph
syn
*Bu
anti
Me
anti
Et
anti
'Pr
anti
Ph
Ratio of Isomers^
-78 -100 -100 -100 -78
25:1 20:1 3:1 1:1
—
-100 •100 -100 -78 -78
anti
2.6:1 exclusive^ 25:1 1:1.3
—
Notes: ^Determined by ^H NMR spectroscopy. ^Minor isomer not detected by 250-MH2 NMR spectroscopy.
From Table 4 it is apparent that a major controlling factor in governing the levels of product diastereoselection is the relative size of the 2-alkyl substituent (R). For syn substrates, the highest levels of diastereoselectivity are observed with a methyl group as the 2-substituent, whereas for anti substrates, ethyl gives the highest levels. Application of our usual chelated transition state models allows us to rationalize the effect of the 2-alkyl substituent. For syn systems, conformations containing axial sulfoxides (e.g., 19), which could occur for very small (axial) 2-alkyl substituents (e.g., proton), would be expected to provide only low levels of selectivity. In the alternative, equatorial, sulfoxide conformation 20, one face of the enolate is effectively shielded by the bulk of the dithiane ring, the other face being exposed unless a very large 2-alkyl substituent is present. Stereoselectivity is therefore expected to become poorer as the relative size of the (equatorial) 2-alkyl substituent is increased, an effect which can be observed in the results outlined in Table 4 on moving from R = Me to R = Ph. Li .LI
.^
(19)
A-
" '^'t'
Cyclic Sulfoxides
129 .Li
3-ii^
(20)
For anti substrates, if the 2-alkyl substituent is axial, reasonable except for very large groups (e.g., -Bu, Ph), the sulfoxide can adopt the equatorial chelated conformation 21; conversely to syn substrates, however, although one face of the enolate is again partially shielded by the 2-alkyl substituent, the bulk of the dithiane ring is distant from the reacting center, and stereoselectivity is expected to be governed solely by the size of the 2-alkyl substituent. The stereoselectivity is therefore expected to improve as the relative size of the 2-alkyl substituent increases, as is indeed observed in Table 4 on moving from 2-methyl to 2-ethyl substituents. In the axial sulfoxide conformation 22, which may occur for anti substrates with very bulky 2-alkyl substituents which may prefer an equatorial orientation (e.g., phenyl, tert-buiyl), the acyl substituent is required to be axial. No
^5^...<=^ ^ • 'ofH-Q
X '
(21)
.s?;
X: (22)
^
^
STEVEN M. ALLIN and PHILIP C. BULMAN PAGE
130
chelation is possible and stereoselectivity is expected to be low. Hence, a fall in diastereoselectivity is observed on moving from the 2-ethyl to 2-phenyl substituent (Table 4). It is important to note that on applying the transition-state models described above, the absolute sense of stereochemical induction expected at the newly created chiral center is predicted to be the same for both syn and anti substrates for either sulfoxide conformation:
syn sulfoxide This prediction was verified by X-ray structure determination of one of the major product diastereoisomers produced in the anti series. The structure was found to be consistent with our proposed transition-state models, and the intermediate formation of a chelated Z-enolate. From these and other studies we have developed a rule of thumb for prediction of stereochemical induction during enolate derivatization: for both syn and anti substrates, the relative stereochemistry at the newly created asymmetric center is opposite to that of the sulfoxide moiety when the structure is drawn as shown above. Asymmetric Mannich reactions?^ Our enolate alkylation methodology has been subsequently extended to include asymmetric Mannich reactions. The Mannich reaction can be viewed as an imino analogue of the aldol reaction and is a very common synthetic method for the preparation of P-aminoketones. Although methods for stereocontrol of the aldol reaction are well documented, including diastereofacial selectivity in reactions of chiral enolates,^^ stereocontrol in Mannich reactions appears to have received relatively little attention.^^"^^ In our studies, we employed the 2-propionyl-2-ethyl-l,3-dithiane 1-oxide substrates, since our previous work on enolate alkylation had demonstrated optimum levels of stereocontrol with an ethyl group as 2-substituent. ^^ Mannich reactions were first performed using commercially available Eschenmoser's salt as the aminoalkylating agent, and enolates derived from syn and anti
1)1.1 eq LHMDS,-78°C. THF •
ii) 1.1. eq Eschenmoser's salt -78°C to 25'*C, THF
Scheme 6.
NMe2
Cyclic Sulfoxides Table 5.
131
Mannich Reaction of Enolates of 2-Propionyl-2-ethyl DiTOX Substrates Using Eschenmoser's Salt as Aminoalkylating Agent
Substrate
Metal
Temp (°C)
Ratio of Isomers^
syn syn syn
Li Zn B
-78 -78 -78
6:1 6:1 4:1
anti anti anti
Li Zn B
-78 -78 -78
1.6:1 1.6:1 1.4:1
Note: ^Determined by ^H NMR spectroscopy.
2-propionyl-2-ethyl-l,3-ciithiane 1-oxides under a variety of reaction conditions and using a range of metal counterions (Scheme 6, Table 5). Table 5 shows that no significant increase in product diastereoselectivity was observed on variation of the metal counterion. Given that a larger and perhaps less reactive electrophile might discriminate between the faces of the prochiral enolate to a higher degree than does Eschenmoser's salt, and so lead to an increase in diastereoselectivity, we chose to employ the benzotriazole-based aminoalkylating agents 23-25 pioneered by Katritzky (Table 6).^^ We were pleased to isolate the desired aminoalkylated products in good yields and, in most cases, with extremely high diastereoselectivity. The use of benzotriazole derivative 24 results in the stereoselective introduction of a primary amine equivalent. The high levels of diastereoselectivity were rationalized through our usual transition-state models for enolate derivatization. X-ray analysis of the product obtained on reaction of the anti enolate with benzotriazole derivative 23 confirmed that the sense of induced stereoselectivity for the major product isomer was as predicted from the transition-state model, and is shown below, 26. From the high levels of product diastereoselectivities observed when employing benzotriazole derivatives 23 and 24, we reasoned that addition of 1 molar equivalent of benzotriazole to a stirred solution of Eschenmoser's salt at room temperature prior to reaction
a>
^NHPh
(23)
a> a> ^NHCOPh
(24)
^NMe2
(25)
132
STEVEN M. ALLIN and PHILIP C. BULMAN PAGE
Table 6. Mannich Reaction of Enolates of 2-Proplonyl-2-ethyl-l ,3-dithiane 1-Oxide Substrates Using Benzotriazole Derivatives as Aminoalkylating Agents Substrate anti syn anti syn anti syn
Benzotriazole 23 23 24 24 25 25
Temp C'C) -78 -78 -78 -78 -78 -78
Yield (%)
Ratio of Isomers^
72 61 81 72
>48:1^ >54:l'' 36:1 >40:1^
—
—
64
1:1
Notes: ^Determined by ^ H NMR spectroscopy. ^Minor isomer not detected by 400-MHz NMR spectroscopy.
with the lithium enolate might provide an increase in diastereoselectivity by forming a similar "masked" iminium ion in situ. Under these conditions, an increase in diastereoselectivity was indeed observed for the anti substrate (up to 3:1), although no improvement was seen with the syn isomer. Subsequently, the benzotriazole-based equivalent of Eschenmoser's salt, 25, was prepared in our laboratory. Curiously, we found this substrate not only to be considerably less reactive than 23 or 24, but also less reactive than the Eschenmoser's salt/benzotriazole system, suggesting that 25 is not formed in situ by mixing these two reagents. One possible explanation is provided by considering the probable reactive intermediates involved: fragmentation of 25 with effective loss of a benzotriazole anion must necessarily give rise to a reactive, and therefore less selective, iminium salt, while concomitant proton loss from 23 or 24 could give rise to a less reactive neutral imine by similar, but more facile, fragmentation. Stereoselective enolate bromination as an approach to a-halo carboxylic acids and a-aminoketones?^ a-Haloketones are useful synthetic intermediates,^^ and may be derived from enolates by treatment with sources of electrophilic halide. This methodology has been applied by others^"^'^^ as a stereoselective approach to chiral a-aminoacids.
(26)
Cyclic Sulfoxides
133
-O O ^. f Et n
I) 1.1 eq LHMDS, -78X. THF
T
li)1.5eq N-l Sl-bromosuccinimide
r
^ f Et
Scheme 7.
Table 7. Diastereoselective Bromlnation of 2-Propionyl-2-ethyl-1,3-clithiane 1 -Oxides Substrate syn anti syn anti
Metal
Reagent
Li Li B B
NBS NBS NBS NBS
Yield (%) 59 73 87 92
Ratio of Isomer^ 1:2.5 1:1 1.67:1 5.5:1''
Notes: ^Determined by ^H NMR spectroscopy, ^yn material—see text.
Our now favored 2-propionyl-2-ethyl-l,3-dithiane 1-oxide substrates were deprotonated using LHMDS at -78 °C in THF, and were treated with solid A^-bromosuccinimide. A selection of results is presented in Scheme 7 and Table 7. The sense of induced stereoselectivity for all reactions carried out on each substrate was assessed in each series on the basis of *H NMR evidence. The structure of the minor isomer from bromination of the lithium enolate derived from the syn substrate was determined by X-ray crystallographic analysis; the relative stereochemistry is as shown in 27. The changes in stereoselectivity observed on the change in counterion (see Table 7) may result simply from the butyl groups carried on the boron atom altering the reacting conformation or the sterically controlled approach of the electrophile. A particularly curious but entirely reproducible result is that obtained from the boron enolate derived from the anti propionyl substrate. In this case, the halogenated product isolated has the syn configuration around the dithiane moiety, with only a trace of anti material remaining. An isomerization from anti-io-syn has therefore taken place under the reaction conditions, perhaps the result of an equilibration process. Such anti-io-syn isomerization could not, however, be in-
(27)
134
STEVEN M. ALLIN and PHILIP C. BULMAN PAGE
duced to take place under a range of conditions, including treatment with NBS,-'^ with either the starting material or brominated and material prepared using a lithium enolate. We have previously observed syn-io-anti equilibration in acyl dithiane oxides on treatment with trifluoroacetic anhydride.^^ a-Haloketones are themselves useful synthetic intermediates,^^ and, given the ready conversion of acyl dithiane oxides into the corresponding acids,^^ the 2-ethyl2-(2-haloacyl)-1,3-dithiane 1-oxides can be regarded as protected a-halocarboxylic acids, compounds which have found use in the synthesis of a variety of products including herbicides and pharmaceuticals.'*^ We envisaged one potential application of the 2-ethyl-2-(2-haloacyl)-l,3dithiane 1-oxides (27) as precursors to a-aminated products, by using nitrogenbased nucleophiles. Ammonia, benzylamine, and tetramethyl guanidinium azide were all unsuccessful as nucleophiles in displacement reactions, resulting either in racemization at the halogenated chiral center or in protiodebromination of the substrate. Further, we were surprised to isolate in excellent yield the 1,2-diketone 28 from attempted sodium azide displacement. A transformation of a-azidoketone to diketone has, however, been reported in the literature.^^"*^ Our most successful introduction of nitrogen using a nucleophilic amination procedure was realized using potassium phthalimide in DMSO solution at 30-40 °C over 12-18 h. A selection of the results is given in Table 8. It is clear that we obtain a dramatic change in stereochemistry in the isolated product mixtures; the ratio of isomers in each case has fallen to ca. 2:1, and the same isomer predominates regardless of the stereochemistry of the starting material at the brominated center. We interpret this observation as a result of equilibration of the asymmetric center through enolization resulting from deprotonation after displacement by excess phthalimide anion under the reaction conditions, or through attack by displaced bromide anion. Overall the chemistry does provide the nucleus of a high yielding approach to chiral a-aminoketones, but the reduction in stereochemical integrity following the nitrogen displacement step invariably limits the synthetic application. Diastereoselective enolate amination as an approach! to a-aminol<e' tones."^^ We have demonstrated that the Mannich reaction is successful for the highly stereoselective introduction of |3-aminoketone moieties (vide supra^ "Asymmetric Mannich Reactions^). The diastereofacially selective electrophilic amination of enolates is attractive as a complementary approach to the asymmetric
(28)
Cyclic
Sulfoxides
135
Table 8. Phthalimide Anion Displacement Reactions with 2-(2-Bromoacyl)-2-ethyl-l,3-dithiane 1-Oxide Substrates
KNphth Nphth
DMSO, 4Cf C
major isomer Substrate
Isomer
Yield (%)
Ratio of Isomers^
syn syn
Major Minor
91 95
2:1 2:1
Note: ^Determined by 'H NMR spectroscopy.
preparation of a-aminoketones, and is commonly used in the preparation of a-aminoacids."*^"^^ Because of the problems encountered during our nucleophilic amination as an approach to a-aminoketones, we investigated an alternative electrophilic amination procedure; which would, if successful, actually provide a more direct approach to these target systems.
NBOC NHBOC
major isomer i; II filBOC NHBOC
major isomer Reagents: i) LHMDS (1.1 eq.), -78°C, THF ii) DBAD (1.1 eq., THF, -78°C, 15 min; HOAc, -78°C Scheme 8.
136
STEVEN M . ALLIN and PHILIP C. B U L M A N PAGE Table 9. Diastereoselectivity of Electrophilic Amination of 2-Acyl-2-alkyl-1,3-dithiane 1 -Oxides
Substrate anti anti anti syn syn
R
R!
Yield (%)
Me Me Me Me Me
Me Et Ph Me Et
69 48 37 76 42
Ratio of Isomers^ 2:1 >99:1 2.7:1 3:1 12:1
Note: ^ Determined by ^H NMR spectroscopy.
We chose to employ di-tert-buty\ azodicarboxylate (DEAD) as the electrophilic aminating reagent.'^"^^ This reagent offers several advantages: it is a stable, crystalline solid available commercially; methods for removal of the t-Boc protecting groups under mild nonracemizing conditions are well documented, and they are complementary to the known methods for N-N bond cleavage."^"^^ The corresponding lithium enolates were generated from syn and anti 2propionyl-2-ethyl-l,3-dithiane 1-oxide substrates in dry THF solvent at -78 °C using LHMDS, and were added via cannula to precooled solutions of DEAD in dry THF at -78 °C. Interestingly, a diastereoselectivity of only 2:1 was observed with the anti substrate if the reaction mixture was allowed to reach room temperature over 12 h before quenching the reaction with saturated aqueous ammonium chloride solution. If the reaction mixture was quenched at -78 °C with glacial acetic acid after only 10-15 min reaction time with DEAD, the resulting diastereoselectivity was much improved to >99:1; only one product isomer being detectable by 400-MHz ^H NMR spectroscopy (Scheme 8). Further results are given in Table 9. It is interesting to note that the effect of the 2-alkyl substituent closely parallels the results obtained in our studies of enolate alkylation. The major isomer proved to have the same relative stereochemistry from both the -78 °C quench and room-temperature quench. The low-temperature acetic acid quench may prevent loss of stereochemical integrity at the new asymmetric center.
Scheme 9,
Cyclic Sulfoxides
137
Table 10. Diastereoselective Cycloaddition Reactions of Danishefsky's Diene with 2-Formyl-2-methyl-l ,3-clithiane 1 -Oxide Substrate
Lewis Acid
Solvent
Yield (%)
ZnCl2 MgBr2
THF THF
exclusive
Ratio of Isomers^
MgBr2
toluene
MgBr2
CH2CI2
82 80 15 60
MgBr2
Et20
—
—
anti
ZnCl2
anti
MgBr2
THF THF
53 20
3.3:1
syn syn syn syn syn
2.1:1 exclusive exclusive
2:1
Note: ^Determined by ^H NMR spectroscopy.
which may occur at higher temperatures. The pattern of diastereoselectivity was rationalized on the basis of our usual chelation control models. Chelation-mediated facially selective cycloaddition reactions.^^ The Diels-Alder reaction is an extremely useful synthetic tool; the reaction displays excellent regio- and stereoselectivity, and these properties have been exploited in the synthesis of many natural product systems.^^ We aimed to develop facially selective Diels-Alder reactions using our DiTOX methodology; such methodology has been the goal of many research groups, and several useful chiral auxiliaries have been developed to accomplish this aim.^^ Syn 2-formyl-2-methyl-l,3-dithiane 1-oxide undergoes efficient cycloaddition reaction with Danishefsky's diene with excellent levels of diastereoselectivity in the presence of magnesium bromide at -78 °C (Scheme 9, Table 10). Chelation control models which are similar to those described previously by us, and others,^"* were proposed to rationalize the observed stereoselectivity. The proposed model 29 for the syn system is shown below; the structure of the major product diastereoiM*
^ A|r"; (29)
STEVEN M. ALLIN and PHILIP C. BULMAN PAGE
138
(30)
somer 30 was confirmed by X-ray single crystal analysis and conformed to the proposed model. As is apparent from Table 10, yields of the cycloadducts and levels of stereoselectivity are highly dependent on several factors including solvent, reaction temperature, and especially Lewis acid. One might expect solvents such as petroleum ether to favor chelated transition states by virtue of their less polar nature, but evidently the solvent effect is more complex than this. Surprisingly, only two of the Lewis acids examined gave isolable products. Yields and levels of product diastereoselectivity were generally lower for the anti substrate than for the syn isomer. Regio- and stereoselective 1,3'dipolar cycloaddition reactions.^^ 1,3Dipolar cycloadditions provide a convenient and useful method of preparation of a wide range of five-membered ring heterocycles,^^ often producing a high degree of stereocontrol as a consequence of a concerted mechanism.^^ We have recently investigated the reactions of nitrile oxides with 2-alkyl-2-crotonyl-l,3-dithiane 1-oxide substrates. The reactions proved to be remarkably regioselective, with only the 5-acyl dihydroisoxazoles being isolated, as highlighted in Scheme 10 for the ^3^rt-2-methyl-2-crotyl-l,3-dithiane 1-oxide substrate. The product diastereoselectivity (a:b) was found to be relatively low for all substrates (up to 5:1), with the syn substrates favoring the formation of isomer a, while the anti substrates tend to favor formation of isomer b, suggesting that it is the stereochemistry at the 2-position of the dithiane unit which is exerting the greatest influence over the stereochemical course of the reactions. This observation is interesting since it contrasts directly with the pattern of stereoselectivities found in our other investigations of dithiane oxides as stereocontrol elements, where the sulfoxide is the principal controlling factor.
RCNO EtjO
Cyclic Siulfoxides
139
~0 o + ' Et II
Me I
Me ^Ph
NBS/H2O
Ph •
k^^^^S
O-N
//
70%
O-N
Scheme 11.
As with our other acyl dithiane oxide systems, the thioacetal moiety can be readily removed by hydrolysis, in this case without affecting the dihydroisoxazoline ring (Scheme 11). In order to achieve chemodifferentiation of the two ketone groups, carbonyl reduction may be carried out prior to NBS-mediated hydrolysis. Reduction with L-Selectride was found to be highly efficient and stereoselective, producing only one diastereoisomer of the product alcohol (Scheme 12). Derivatization of DiTOX: acylation. In our early work, preparation of 2acyl-l,3-dithiane 1-oxide substrates relied on a lengthy procedure from 1,3dithiane. Lately, however, we have made considerable advances, having solved the unexpectedly difficult problem of acylation of DiTOX itself. Acylation is efficiently achieved using N-acyl imidazoles with over 2 equiv of base to yield the desired 2-acyl-l,3-dithiane 1-oxides in good yields after protic workup (Scheme 13).^^ We are also now able to prepare 2-acyl-2-alkyl-l,3-dithiane 1-oxides, our most commonly used substrates, in a one-pot application of this procedure.^^ Using mixed base conditions, we were able to isolate the desired DiTOX derivatives in moderate to good yields (Table 11), and with excellent levels of diastereoselectivity. Interest-
o OH Me + f Me I : : i S ^ : J L > \ .Ph
OH
Ph
//
66%
O-N
Scheme 12.
O
o
l.NHMDS (l.l eq), THF, - 7 8 ^ 15 min. 2. BuLi (1.1 eq), -78^C, 15 min.
S 3.RCOimid(l.leq),-7^Ctor.t 4. HjO^ Scheme 13.
Me
NBS/H2O
~0
O
^S^JL^ k.^S
STEVEN M. ALLIN and PHILIP C. BULMAN PAGE
140
Table 11. One-Pot Generation of 2-Acyl-2-alkyl-1,3-cllthiane 1 -Oxides O
l.NHMDS (1.1 eq), THF, -78°C, 15 min.
O S
2. BuLi (1.1 eq), -78^C, 15 min. 3. RCamid (1.1 eq), -7tfC to r.t., 2 hr 4. R'l (2.0 eq.), -78°C to r.t., 16 h R'
Me Me Me Et Et Bu Bu 4-te/t-Bu-Ph
Me Et CH2=CHCH2 Me CH2=CHCH2 Me CH2=CHCH2 Me
Yield (%) 65 54 73 73 75 71 66 67
Selectivity (syn/anti) 7: 1 exclusive^ exclusive exclusive 15:1 exclusive 4: 1 exclusive
Note: ^Minor Isomer not detectable by 400-MHz ^H NMR spectroscopy.
ingly, the isomer formed predominantly has the syn configuration; this route is therefore complementary to our earlier route involving sulfur oxidation as the final synthetic step (which provides predominantly anti material). In addition, we have discovered that the syn isomers are cleanly converted into anti by low-temperature treatment with trifluoroacetic anhydride by equilibration at sulfoxide sulfur.^ Perkin ring synthesis using DiTOX. Anions derived from DiTOX undergo efficient Perkin ring synthesis on treatment with dihaloalkanes to provide cycloalkane rings of up to seven members (Table 12).^^ One-pot stereocontrolled cycloalkanone synttiesis.^^ An application of the methodology described above allows a one-pot stereocontrolled cycloalkanone synthesis. We were pleased to find that deprotonation followed by sequential treatment with an A^-acyl imidazole and a diiodoalkane, led to the corresponding haloalkylated material, formed exclusively with syn stereochemistry. Further treatment with NHMDS gave, in two cases, cyclization to carbocyclic products with sufficiently high diastereoselectivity that the minor isomer could not be detected by 400-MHz ^H NMR spectroscopy (Table 13). Two new asymmetric centers and two new C-C bonds are therefore each formed in these one pot cyclization reactions with extremely high stereoselectivity. Curiously it is the seven- and eightmembered ring compounds which are most readily formed; reaction with 1,3-diiodopropane gave preferential elimination of HI to provide only syn-2-a\\y\-
Cyclic Sulfoxides
141 Table 12. Perkin Ring Synthesis Using DiTOX
C
l.NHMPS (2.2 eg). THF, -20°C. 30 min. ^ S v J
S 2.Br(CH2)„Br(2.0eq),-78°Ctor.t.,16h
L^^^S
n
Yield (%)
4 5 6
75 79 81
y "
2-propionyl-l,3-dithiane 1-oxide, while 1,6-diiodohexane gave the haloalkylated material but did not undergo cyclization to the nine-membered ring. The relative stereochemistry indicated in the product structures in Table 13 are as predicted from earlier enolate alkylation studies"^'^^ and from knowledge of a favored syn intermediate haloalkylated species (vide infra). Presumably, conformations adopted by the intermediates are such that cyclization is favored only for a limited range of ring sizes. Asymmetric sulfoxidation of l-acyl-l ,3-dithianes: Preparation of optically pure DiTOX substrates. The preliminary investigations of 1,3-dithiane derivatives as asymmetric building blocks and chiral auxiliaries described in this review Table 13. One-Pot Stereocontrolled Cyclizations Using DiTOX
:?
l.NHMDS (2.2 eq), THF, -TS'^C, 15 min;. " (^ ^ 2. BuLi(1.1 eg), -78°C, 15 min.;
/ ' ^ ' .
3. RCHjCamid (1.1 eq), -7^C to r.t, 2 h; ^-..^/^ 4.1(CH2y (10 eq), -78°C to r.t, 16 h; 5. NHMDS (1.1 eq), -78*'C to r.t, 16 h
Me Me Me Me Et Et
n
Yield (%)
3 4 5 6 4 5
a
72 74 _b
60 63
Notes: ^syn-2-allyl-2-propionyl-l,3-clithiane 1-oxide isolated (78%). '^Uncyclized haloalkylated material isolated (64%).
Selectivity (syn/anti)
— exclusive exclusive
— >4: 1 exclusive
142
STEVEN M. ALLIN and PHILIP C. BULMAN PAGE
employed the racemic DiTOX substrates for diastereoselective transformations. To pursue syntheses of nonracemic target compounds, it was necessary for us to produce DiTOX systems in the enantiomerically pure sulfoxide series. Remarkably, there is a noticeable lack of general methods for the asymmetric preparation of chiral sulfoxides from sulfides. The most satisfactory method would be a generally applicable enantioselective sulfoxidation reaction which would allow the preparation of sulfoxides from any prochiral sulfide with high ee's and in which the sulfoxide would be amenable to enantioselective preparation in both senses. Several approaches to the enantioselective oxidation of sulfides have been reported,^^ including enzymatic approaches,^^ use of optically pure oxidants,^^ and several modifications of the Sharpless epoxidation procedure.^^'^ The success of these procedures is somewhat substrate dependent; for example, dialkyl sulfides and more complex substrates can give unpredictable results. 1,3-Dithiane itself is oxidized with only ca. 20% ee; optically pure DiTOX has, however, been obtained by resolution.^^ Oxidation of simple 2-substituted dithianes using modified Sharpless conditions gave poor results (ca. 10-20% ee). We subsequently recognized that, in common with the Sharpless epoxidation itself, such modifications might require the presence of a dipolar grouping within the molecule. Indeed, a paper by Modena and co-workers has reported the enantioselective sulfoxidation of a range of P-hydroxy sulfides and derivatives in up to 80% ee.^^ Accordingly, we examined 2-acyl dithianes as substrates and, after some adjustment of reaction conditions and workup conditions, we were pleased to isolate acyl dithiane sulfoxides in up to 97% ee and in high yields, with the and isomer predominating (Scheme 14).
"O
a
O
^ I >L ^ r ^R
k,,^S
'^
O
(+)-H(-)5V/7
(minor) •
(major) Reagents: (+)-diethyl tartrate (2.0 eq), Ti (OiPr)4 (1.0 eq), H2O (1.0 eq), tert-butyl hydroperoxide (1.1 eq), CH2CI2, -20°C, ca 1-3 days Scheme 14,
Cyclic Sulfoxides
143 (+)-DET
SmalK ^Large
"O Small^^^Large
Scheme 15.
Kagan has proposed a rule of thumb for predicting the absolute configuration at sulfur in these sulfoxidation procedures'^: the sulfide substrate is drawn as a two-dimensional representation with the sulfur lone pairs pointing upwards, the larger, or perhaps coordinating, alkyl group pointing to the right and downwards, and the smaller alkyl group pointing to the left and downwards (Scheme 15). Using (+)-tartrate, the oxygenation then normally occurs from the front. For 2-substituted acyldithianes this results in the /^-configuration at sulfoxide sulfur. Enantiomerically pure 2-acyl-2-alkyl-l,3-dithiane 1-oxide substrates could then be obtained through recrystallization. We have subsequently extended our studies to include the enantioselective synthesis of a wide range of 2-substituted-1,3-dithiane 1-oxides,'^ including 2-heterosubstituted-1,3-dithiane 1 -oxides.'^ In addition to the enantioselective preparation of 1,3-dithiane 1 -oxides, our group has been concerned with the development of novel methods for the catalytic asymmetric oxidation of other prochiral sulfides; our currently preferred system employs an enantiomerically pure sulfonylimine and commercially available hydrogen peroxide.^^ Applications of the DiTOX asymmetric building block: enantioselective synthesis of (R)-(-)'2^6-dimethylheptanoic acid. The first application demonstrating the use of DiTOX units as chiral auxiliaries was reported in 1994.^^ We described the two-step enantioselective synthesis of (/?)-(-)-2,6-dimethylheptanoic acid (31), a natural product derivative containing a carboxylic acid function substituted at the a-carbon atom, a feature common to many analgesic compounds. Our synthetic route is outlined in Scheme 16. (1/?, 2/?)-(+)-anr/-2-Propanoyl-2ethyl-1,3-dithiane 1-oxide was prepared by enantioselective sulfur oxidation as described in the preceding section of this review. Enolate alkylation proceeded without complication in 57% yield to give the optically pure a-alkylated product. Simple base-mediated deacylation led directly to the desired a-alkyl carboxylic acid in 39% yield, without loss of stereochemical integrity; the 2-ethyl-1,3-dithiane HO2C
(31)
STEVEN M. ALLIN and PHILIP C. BULMAN PAGE
144 .1 Et
i)
j ^
Reagents: i) LHMDS (1.1 eq), THF, DMPU (10 eq), -78X; ii) 4-methyl iodopentane, -78°C to r.t.; HaO^ Scheme 16.
1-oxide auxiliary is recoverable in optically pure form. This simple synthesis has paved the way for further application of the DiTOX asymmetric building block. Enantioselective synthesis of a-arylpropanoic acids. a-Arylpropanoic acids are an important class of compounds, well known for their anti-inflammatory activity; a number of methods have been developed for the racemic and enantioselective synthesis of this class of compound.^^ Several of these compounds are successfully marketed, with perhaps the most well-known example being ibuprofen (32). The acyl dithiane oxide substrates 33a-d used in this study were prepared by methods described in this review. Scheme 17 highlights our synthetic route to the target compounds 36a-d; the yields and product enantioselectivities are given in Table 14. In this case, removal of the 1,3-dithiane 1-oxide units to reveal the carboxylic acid could not be accomplished using the base-induced cleavage employed previously,^ ^ but was readily achieved through a two-step procedure involving hydrolysis to furnish the a-diketones which, remarkably, retained their
HO' CH3 (32)
Cyclic Sulfoxides
145 CI
j-fi
j~/y
CH3
(36a)
HO'
6H3
CH3
(36b)
(36d)
Stereochemical integrity, followed by oxidative cleavage by aqueous sodium periodate in methanol. Enantioselective synthesis ofa-hydroxyketones7^ a-Hydroxyketones are an important structural feature of many biologically active molecules.^^ Compounds containing this functionality have also been reported to control the stereochemistry in several different transformations.^^ Optically pure 1,3-ciithiane 1-oxide substrates 37 and 38 were prepared by our standard methods. We have previously described the stereoselective reduction of 2-acy 1-2-alky 1-1,3-ciithiane 1 -oxides with DIB AL,^ and normally observe a reversal of selectivity on addition of zinc chloride. In this case (Scheme 18), THF solutions of the substrates were treated at -78 °C with either DIBAL or DIBAL/ZnCl2 reducing systems. As expected, the DIBAL and DIBAL/ZnC^ reducing systems
i):li)
^s/^
(33)
Ho^rO-^^
-K-R
CHa (36)
i) LHMDS (1.1 eq.), THF, -78°C; ii) Mel (1.5 eq.); iii)MBS (8eq.).acetone-water(97:3), r.t.;iv)NaI04(2 eq),MeOH, 20''C Scheme 17.
146
STEVEN M. ALLIN and PHILIP C. BULMAN PAGE Table 14. Preparation of a-ary I propanoic Acids (36a-d)
Substrate
Yield of 34 (%)
Yield of 35 (%)
Yield of 36 (%)
ee (%)
77 84 80 70
96 98 81 97
80 77 68 79
93 90 87 81
33a 33b 33c 33d
DIBAL
"O OH +r p h :
58%
(37)
"o
o DIBAL
(38)
ZnCb 84%
Scheme 18.
gave products of opposite stereoselectivity, and in most cases only one product diastereoisomer was observed. Hydrolysis of the 1,3-dithiane 1-oxide moieties of the product alcohols under our standard conditions using NBS/acetone/water gave the corresponding a-hydroxyketones 39 and 40 in excellent yields (Scheme 19).
O ^ OH +f Ph =
OH
NBS 3%H20/acetone 84% "O
Ph
(39)
OH
NBS 3%H20/acetone 76%
OH Ph
(40) Scheme 19.
Cyclic Sulfoxides
147
An excellent enantioselectivity of 93% was observed for products 39 and 40; in the case of the 2-methyl analogues, a degree of racemization was observed. B. 1,3-Dithiane Dioxides Addition to Aldehydes
Aggarwal has demonstrated the success of a variety of electrophilic reactions involving the C2 symmetric species, 1,3-dithiane dioxide^^^"^ This substrate is easily prepared in 60% yield from oxidation of 1,3-dithiane with sodium metaperiodate, and is isolated as a mixture of diastereoisomers in a ratio of 95 : 5 in favor of the fran5-l,3-dithiane-5,5-dioxide product. Aggarwal has discovered that the anion derived from 1,3-dithiane dioxide undergoes rapid reaction with a range of aldehydes leading to the formation of diastereoisomeric mixtures in good yield (Scheme 20).^^^ Lithiated 2-chloro-l,3-dithiane-l,3-dioxide undergoes an analogous addition reaction with aldehydes to give product mixtures displaying high diastereoselectivities (Scheme 21).'^^ Enantioselective synthesis ofa-hydroxy acid derivatives. Recently, Aggarwal has reported an enantioselective approach to the synthesis of a-hydroxy acid derivatives using rran^-1,3-dithiane-1,3-dioxide. For example, reaction of transl,3-dithiane-l,3-dioxide with an aromatic aldehyde liberates the alcohol which is protected as the tetrahydropyranyl (THP) ether; the resulting product may then be subjected to a Pummerer reaction, using trifluoroacetic anhydride, to give a thiolester. Transthiolesterification of this product using LiSEt gives the ^ ^ ^ .1 1.
(i)py/THF (1.5:1): (ii) BuLi,-45«C;
r'^/'^o-
(iii) RCHO.-78**C; (iv) HCI (2mol.r'')
^ ^ ^ +1 L
j / +1
^ 1+
••rf^^^V^'^o-rf^^^N^'^oJ + R'^f^^^ H - S ^ OH ^^ <* OH " v^ R
66:34 Scheme 20.
( ^ ^
T CI
(i)LHMDS,(fC °
(ii)PhCHO
^.^^
?"
^ _ 4
O^i
(iii) HCI/EtOH
OjL, 8-92
Scheme 21.
?"
STEVEN M . ALLIN and PHILIP C. B U L M A N PAGE
148
(i)NHMDS, PhCHO •
THF. py
-Q^.^S Ph-^OH
(li) TFAA, py (ill) LiOH, EtSH THF, H2O \
..OH
\
(iv) PPTS, EtOH
Y^Ph
(V) LiOH. 4202 THF, H2O
oOTHP
Y^Ph
Scheme 22.
ethylthiolester as shown in Scheme 22. Subsequent hydrolysis furnishes the corresponding a-hydroxy acid in 86% yield and 97% enantiomeric excess (Scheme 22).'^^
III. FIVE-MEMBERED SULFUR HETEROCYCLES A. Addition
to
1,3-Dithiolane-1-Oxide
Aldehydes
The aldol addition of certain optically active a-sulfmyl enolates to carbonyl compounds affords the corresponding adducts in up to 98% enantiomeric excess and in good chemical yields.^^^'^''^^ Modena has shown that the magnesium enolate derived from (~)-rraAz^-2-A^,A^-diethylacetamide-l,3-dithiolane-5-oxide, which is obtained in high ee by using the enantioselective oxidation pioneered by DiFuria and Modena,^ ^ undergoes aldol-type addition with w<9-butanal to furnish the alcohol as a single diastereoisomer.^^ The relative stereochemistry of the product, determined to be IR, 2R, VR from X-ray diffraction studies, is presumed to originate from the rigid transition state in which both the enolate and aldehyde oxygen atoms are coordinated to the magnesium atom (Scheme 23). 2'All
1-Oxides
Maycock has recently reported the use of an optically pure 2-alkyl-2-acyl-l,3dithiolane 1-oxide to synthesize optically active a-hydroxyketone derivatives.^^ The acyl dithiolane 1-oxide was prepared by an enantioselective sulfoxidation procedure. Interestingly, Maycock has recently reported enhanced diastereo- and
Cyclic Sulfoxides /
149 \
(i) Ti (IV), (+)-DET
><; H
/
(ii) TBHP. DCE, -20-0 '^CONEti
H^
(iii). Bu*MgCI, THF.-78'C
-
> <
790/0
^CONEtj
s. .s-o H o ^ , ;
-
(iv) MeCHzCHO. 82%
\ +
^ J ^ " ^coNEt,
"
Me
Scheme 23.
I
V+ -
DIBAL/-78°C
J^
V
-
3 steps
.Ar^^
MeA-^«
b U
HO .1
Me'
j^
^^^
^^^^^ occ
Scheme 24.
NH2
6 steps CO2H
Me
a^ ^
^
^ ^
^v
BF3.0Et2 -78 '^C, 5 mjn
BF3.0Et2
Scheme 26.
150
STEVEN M. ALLIN and PHILIP C. BULMAN PAGE
enantioselectivity in the formation of acyldithiolane sulfoxides by the asymmetric oxidation (modified Sharpless conditions) of their corresponding silyl enol ethers.^"* Reduction of the carbonyl group using DIBAL at low temperatures afforded the desired alcohol with complete diastereoselectivity. Deprotection yielded the protected a-hydroxyketone in good yield and with > 98% ee (Scheme 24). 1^3'Dithiolane Dioxide
Cycloaddition reaction of tranS''2'methylene-I^S-dithiolane dioxide. Aggarwal has shown that the rran5-2-methylene-l,3-dithiane dioxide, which is prepared in six steps from anthranilic acid, reacts well with cyclopentadiene under Lewis acid conditions and in highly diastereoselective fashion to furnish a [4 + 2] cycloadduct (Scheme 25).^^ He has proposed that these cyclic alkenyl sulfoxides represent potential chiral ketene equivalents, offering several advantages over other ketene equivalents such as a-acetoxyacrylonitrile^^ and nitroethane^^: Aggarwal's species offer low steric bulk with two activating groups present at the same carbon atoms. This methodology has recently been extended to include a simpler dithiolane 1,3-dioxide derivative which undergoes highly stereoselective Diels-Alder reactions giving adducts as single diastereoisomers (Scheme 26).^^
REFERENCES 1. Page, P. C. B.; Westwood, D.; Slawin, A. M. Z.; Williams, D. J. / Chem. Soc, Perkin Trans. 1 1989,1158. 2. Page, P. C. B.; Prodger, J. C ; Hursthouse, M. B.; Mazid, M. / Chem. Soc, Perkin Trans. 1 1990, 167. 3. Page, P. C. B.; Prodger, J. C. Synlett 1990, 460. 4. Page, P C. B.; Slawin, A. M. Z.; Westwood, D.; Williams, D. J. J. Chem. Soc, Perkin Trans. 1 1989, 185. 5. Page, P C. B.; Prodger, J. C ; van Neil, M. B. Tetrahedron 1989, 45, 7643. 6. Tidwell, T. T. Synthesis 1990, 857. 7. Curtin, D. Y; Harris, E. E.; Meislich, E. K. J. Am. Chem. Soc 1952, 74, 2901; Cram, D. J.; Abd Elhafez, R A. J. Am. Chem. Soc 1952, 74, 5828. 8. See Morrison, J. H.; Mosher, H. S. Asymmetric Organic Reactions', Prentice-Hall: New York, 1971. 9. Cram, D. J.; Kopecky, K. R. J. Am. Chem. Soc 1959, 81, 2748; also Eliel, E. L. \n Asymmetric Synthesis; Morrison, J. D., Ed.; Academic Press: London, 1986, Vol. 2, Chapter 5. 10. Eliel, E. L.; Frye, S. V. / Am. Chem. Soc 1988,110,484; Eliel, E. L.; Morris-Natschke, S. / Am. Chem. Soc 1984,106,2937; Mukaiyama, T. Tetrahedron 1981,37,4111; Asami, M.; Mukaiyama, T. Chem. Lett. 1983, 93. 11. Cram, D. J.; Wilson, D. R. J. Am. Chem. Soc 1963, 85, 1249; Still, W. C ; McDonald, J. H. Tetrahedron Lett. 1980, 21, 1031. 12. Comforth, J. W.; Comforth, R. H.; Mathew, K. K. J. Chem. Soc 1959, 112. 13. Grandbois, E. R.; Howard, S. J.; Morrison, J. D. In Asymmetric Synthesis', Morrison, J. D., Ed; Academic Press: London 1983, Vol. 2, Chapter 3.
Cyclic Sulfoxides
151
14. Eliel, E. L. \n Asymmetric Synthesis; Morrison, J. D., Ed.; Academic Press: London 1983, Vol. 2, Chapter 5. 15. Solladi6, G.; Demailly, G.; Greek, C. Tetrahedron Lett, 1985, 26, 435; Carreno, M. C ; Garcia Ruano, J. L.; Martin, A. M.; Pedregal, C ; Rodriguez, J. H.; Rubio, A.; Sanchez, J.; Solladi^, G. J. Org. Chem. 199^, 55,2X20. 16. Posner, G. H.; Kogan, T. R; Hulce, M. Tetrahedron Lett. 1984, 25, 383. 17. Mukaiyama, T.; Takeda, T.; Fujimoto, F Bull. Chem. Soc. Jpn. 1978, 57, 3368. 18. Ledyendecker, F ; Jesser, F ; Rubland, B. Tetrahedron Lett. 1981, 22, 3601. 19. Procter, G. Asymmetric Synthesis; Oxford University Press: London, 1996. 20. Lutomski, K. A.; Meyers, A. I. In Asymmetric Synthesis; Morrison, J. D.; Ed.; Academic Press: London 1984, Vol. 3, Chapter 3. 21. Evans, D. A.; Britton, T. C ; Dorow, R. L.; Dellario, J. F J. Am. Chem. Soc. 1986,108, 6395, and references contained therein. 22. Page, R C. B.; Klair, S. S.; Westwood, D. J. Chem. Soc, Perkin Trans. 11989, 2441. 23. Heathcock, C. H.; Buse, C. T.; Kleschick, W. A.; Pirrung, M. C ; Sohn, J. E.; Lampe, J. / Org. Chem. 1980,45,1066. 24. Page, R C. B.; AUin, S. M.; CoUington, E. E.; Carr, R. A. E. /. Org. Chem. 1993, 58, 6902. 25. See Heathcock, C. H. Aldrichimica Acta 1990, 23, 99. 26. Seebach, D.; Betschart, C ; Schiess, M. Helv. Chim. Acta 1989, 67, 1593. 27. Katritzky, A. R.; Harris, R A. Tetrahedron 1990,46, 987. 28. Kunz, H.; Pfrengle, W. Angew. Chem. Int. Ed Engl. 1989, 28, 1067. 29. Kunz, H.; Schanzenbach, D. Angew. Chem. Int. Ed. Engl. 1989,28, 1068. 30. Nolen, E. G.; AUocco, A.; Broody, M.; Zuppa, A. Tetrahedron Lett. 1981, 22, 3601. 31. Katritzky, A. R.; Rachwal, S.; Rachwal, B. J. Chem. Soc, Perkin Trans. 1 1987, 799; Katritzky, A. R.; Rachwal, S.; Hitchings, G. J. Tetrahedron 1991,47, 2683. 32. Page, R C. B.; McKenzie, M. J.; AUin, S. M.; CoUington, E. W.; Carr, R. A. E. Tetrahedron 1995, 57, 1285. 33. De Kimpe, N.; Verhe R. In The Chemistry ofa-Haloketones, a-Haloaldehydes and a-Haloimines; Wiley: New York, 1988. 34. Evans, D. A.; EUman, J. A.; Dorow, R. L. Tetrahedron Lett. 1987, 28, 1123. 35. Oppolzer, W.; Pedrosa, R.; Moretti, R. Tetrahedron Lett. 1986,27, 831. 36. Sulfoxides are known to undergo epimerization on treatment with NCS: Satoh, T.; Oohara, T.; Ueda, Y; Yamakawa, K. Tetrahedron Utt. 1988, 29, 313; Satoh, T; Oohara, T.; Ueda, Y; Yamakawa, K. J. Org Chem. 1989,54, 3130; Drabowicz, J. J. Org Chem. 1986, 57, 831. 37. Page, R C. B.; Shuttleworth, S. J.; McKenzie, M. J.; Schilling, M. B.; Tapolczay, D. J. Synthesis 1995, 73. 38. De Kimpe, N.; Verhe, R. In The Chemistry ofa-Haloketones, a-Haloaldehydes and a-Haloimines; Wiley: New York, 1988. 39. Page, R C. B.; AUin, S. M.; CoUington, E. W.; Carr, R. A. E. Tetrahedron Lett. 1994, 35, 2607. 40. Duhamel, L.; Angibaud, R; Demurs, J. R.; Valnot, J. Y Synlett 1991, 807. 41. Edwards, O. E.; Purushothaman, K. K. Can. J. Chem. 1964, 42, 712; Raap, R. Tetrahedron Lett. 1969, 3493; Manis, R; Rathke, M. W J. Org. Chem. 1980, 45,4952; Watthey, J. W H.; Stanton, J. L.; Desai, M.; Barbiarz, J. E.; Finn, B. M. J. Med. Chem. 1985,28, 1511. 42. Page, R C. B.; AUin, S. M.; CoUington, E. W; Carr, R. A. E. Tetrahedron Lett. 1994, 35, 2427. 43. Diels, O.; Behncke, H. Ben 1924,57, 653. 44. Evans, D. A.; Evrard, D. A.; Rychnovsky, S. D.; Fruh, T.; Whittingham, W. G.; DeVries, K. M. Tetrahedron Lett. 1992, 33,1189; Evans, D. A.; Britton, T. C ; EUman, J. A.; Dorow, R. L. J. Am. Chem. Soc 1990,112,4011. 45. Oppolzer, W; Tamura, O. Tetrahedron Lett. 1990, 31, 991. 46. Trimble, L. A.; Vederas, J. C. J. Am. Chem. Soc 1986,108, 6397. 47. Oppolzer, W.; Moretti, R. Helv. Chim. Acta 1986, 69, 1923.
152 48. 49. 50. 51. 52.
STEVEN M. ALLIN and PHILIP C. BULMAN PAGE
Evans, D. A.; Britton, T. C ; Dorow, R. L.; Dellaria, J. F. Tetrahedron 1988,44, 5525. Evans, D. A.; Britton, T. C ; Dorow, R. L.; Dellaria, J. F. J. Am. Chem. Soc. 1986,108, 6395. Mellor, J. M.; Smith, N. M. J. Chem. Soc, Perkin Trans. 11984, 2927. Page, R C. B.; Prodger, J. C. Synlett 1991, 84. Carruthers W. A. In Cycloaddition Reactions in Organic Synthesis', Pergamon: Oxford, 1990; Fringuelli, F ; Taticchi A. In Dienes in the Diels Alder Reaction, Wiley: New York, 1990. 53. Paquette L. A. In Asymmetric Synthesis; Morrison, J. D., Ed.; Academic Press: London, 1989, Vol. 3, Chapter 7. 54. Danishefsky, S. J.; Pearson, W. H.; Harvey, D. F ; Maring, C. J.; Springer, J. P. J. Am. Chem. Soc. 1985,107, 1256; Midland, M. M.; Koops, R. W. J. Org. Chem. 1990, 55, 5058. 55. Page, P C. B.; Purdie, M.; Lathbury, D. Tetrahedron 1997, 57, 1061. 56. Torssell, K. B. G. In Nitrite Oxides, Nitrones and Nitronates in Organic Synthesis, VCH: Weinheim, 1988; Kozikowski, A. R Ace. Chem. Res. 1984, 17, 410; Kanemasa, S.; Tsuge, O. Heterocycles 1990, 30,719. 57. Huisgen, R. Angew. Chem. Int. Ed. Engl. 1963,2,633; Houk, K. N.; Firestone, R. A.; Munchausen, L. L.; Mueller, R H.; Arison, B. H.; Garcia, L. A. J. Am. Chem. Soc. 1985,107,7227; Christi, M.; Huisgen, R. Tetrahedron Lett. 1968, 5209. 58. Page, R C. B.; Gareh, M. T.; Porter, R. A. Tetrahedron Lett. 1993, 34, 5159. 59. Page, R C. B.; Shuttleworth, S. J.; Schilling, M. B.; Tapolczay, D. J. Tetrahedron Lett. 1993, 34, 6947. 60. Page, R C. B.; Shuttleworth, S. J.; McKenzie, M. J.; SchilHng, M. B.; Tapolczay, D. J. Synthesis 1995, 73. 61. Colonna, S.; Gaggero, N.; Manfredi, A.; Casella, L.; GuUotti, M. J. Chem. Soc, Chem. Commun. 1988,1451. 62. Davis, F A.; Thinmia Reddy, R.; Weismiller, M. C. / Am. Chem. Soc, 1989, 111, 5964; Davis, F A.; McCauley, J. R; Chattopadhyay, S.; Harakai, M. E.; Towson, J. C ; Watson, W. H.; Tavanaiepour, I. J. J. Am. Chem. Soc 1987.109,3370; Davis, F A.; Towson, J. C ; Weismiller, M. C ; Lai, S.; Carroll, R J. J. Am. Chem. Soc 1988,110, 8477. 63. Pitchen, R; Dunach, E.; Desmukh, M. N.; Kagan, H. B. J. Am. Chem. Soc. 1984,106, 8188; Zhao, S.-H.; Samuel, O.; Kagan, H. B. Tetrahedron 1987,43, 5135; Dunach, E.; Kagan, H. B. Nouv. J. Chim. 1985, 9, 1; Kagan, H. B.; Dunach, E.; Nemecek, E.; Pitchen, R; Samuel, O.; Zhao, S.-H. PureAppl. Chem. 1985,57, 1911. 64. Bortolini, O.; DiFuria, F ; Licini, G.; Modena. G.; Rossie, M. Tetrahedron Lett. 1986, 27, 6257; DiFuria, F ; Modena, G.; Seraglia, R. Synthesis 1984, 1049. 65. Bryan, R. F ; Carey, F A.; Dailey, O. D.; Maher, R. J.; Miller, R. W. J. Org. Chem. 1978,43, 90. 66. Conte, V; DiFuria, F ; Licini, G.; Modena, G. Tetrahedron Lett. 1989, 30, 4857. 67. Pitchen, R; Dunach, E.; Desmukh, M. N.; Kagan, H. B. J. Am. Chem. Soc 1984, 106, 8188; Pitchen, R; Kagan, H. B. Tetrahedron Lett. 1984,1049; Dunach, E.; Kagan, H. B. Nouv. J. Chim. 1985, 9, 1; Kagan, H. B.; Dunach, E.; Nemecek, E.; Pitchen, R; Samuel, O.; Zhao, S.-H. Pure Appl. Chem. 1985,57, \9n\Yi2ig2in,'^.B. Phosphorus Sulfur \9U, 27,127; Zhao, S.-H.; Samuel, O.; Kagan, H. B. Tetrahedron 1987, 43, 5135; Baldenius, K.-U.; Kagan, H. B. Tetrahedron: Asymmetry 1990,1, 597. 68. Page, R C. B.; Namwindwa, E. S.; Klair, S. S.; Westwood, D. Synlett, 1990, 457, Page, R C. B.; Namwindwa, E. S. Synlett 1991, 80', Page, R C. B.; Heer, J. R; Bethell, D.; Collington, E. W; Andrews, D. M. Tetrahedron: Asymmetry 1995, 5, 2911. 69. Page, R C. B.; Wilkes, R. D.; Barkley, J. V; Witty, M. J. Synlett 1994, 547; Page, R C. B.; Wilkes, R. D.; Namwindwa, E. S.; Witty, M. J. Tetrahedron 1996,52, 2125. 70. Page, R C. B.; Heer, J. R; Bethell, D.; Collington, E. W; Andrews, D. M. Tetrahedron Lett. 1994, 35, 9629; Page, R C. B.; Heer, J. R; Bethell, D.; Collington, E. W; Andrews, D. M. Tetrahedron: Asymmetry 1995,6, 2911; Page, R C. B.; Heer, J. R; Bethell, D.; Collington, E. W; Andrews, D. M. Synlett 1995, 773.
Cyclic Sulfoxides
15 3
71. Page, P. C. B.; AUin, S. M.; Klair, S. S.; CoUington, E. W.; Carr, R. A. E. Tetrahedron Lett. 1994, 55, 2607. 72. Shen, T. Y. Angew. Chem. Int. Ed. Engl. 1972, 77, 460; Rieu, J. P; Boucherle, A.; Cousse, H.; Mouzin, G. Tetrahedron 1986, 42, 4095; Sonawane, H. R.; Bellur, N. S.; Ahuja, J. R.; Kulkami, D. G. Tetrahedron: Asymmetry 1992, 5, 163. 73. Page, P C. B.; Purdie, M.; Lathbury, D. Tetrahedron 1996, 57, 8929. 74. Murahashi, S. I.; Saito, T.; Hanoaka, H.; Murakami, Y; Naota, T.; Kumobayashi, H.; Akutagawa, S. J. Org. Chem. 1993,55, 2929; Cain, C. M.; Simpkins, N. S. Tetrahedron Lett. 1987, 25, 3723; Paquette, L. A.; Hin, H. S.; Coghlan, M. J. Tetrahedron Lett. 1987, 25, 5017. 75. Nakata, T; Tanaka, T.; Oishi, T. Tetrahedron Lett. 1983,24, 2653; Trost, B. M.; Urabe, H. J. Org. Chem, 1990,55, 3982; Paterson, I.; Wallace, D. J.; Velazquez, S. M.; Tetrahedron Lett. 1984, 55, 9083; Paterson, I.; Wallace, D. J. Tetrahedron Lett. 1984, 35, 9087. 76. (a) Aggarwal, V. K,; Davis, I. W; Maddock, J.; Mahon, M. F ; MoUey, K. C. Tetrahedron Utt. 1990, 57, 135; (b) Aggarwal, V. K.; Davis, I. W; Maddock, J.; Mahon, M. R; Molley, K. C. J. Chem. Soc, Perkin Trans. 11992,662; (c) Aggarwal, V. K.; Franklin, R. J.; Rice, M. J. Tetrahedron Utt. 1991,32,1143. 11. Aggarwal, V. K.; Worrall, J. M.; Adams, H.; Alexander, R. Tetrahedron Lett. 1994, 35, 6167; Aggarwal, V. K.; Boccardo, G.; Worrall, J. M.; Adams, H.; Alexander, R. J. Chem. Soc, Perkin Trans. 11997,11. 78. Aggarwal, V. K.; Thomas, A.; Franklin, R. J. J. Chem. Soc, Chem. Commun. 1994, 1653. 79. (a) Mioskowski, C ; Solladi6, G. J. Chem. Soc, Chem. Commun. 1977,162; (b) Mioskowski, C.; Solladi^, G. Tetrahedron 1980,36,227; (c) Solladi6, G.; Frechou, C ; Demailly, G. Nouv. J. Chim. 1985, 9, 22. 80. Annunziata, R.; Cinquini, M.;Cozzi,F; Montanari, F ; Restelli, A. J. Chem. Soc, Chem. Commun. 1983,1138. 81. DiFuria, F ; Modena, G.; Seragha, R. Synthesis 1984, 325. 82. Corich, M.; DiFuria, F ; Lincini, G.; Modena, G. Tetrahedron Lett. 1992, 33, 3043. 83. Barros, M. T.; Leitao, A. J.; Maycock, C. D. Tetrahedron Lett. 1995,36, 6537. 84. Barros, M. T; Leitao, A. J.; Maycock, C. D. Tetrahedron Lett. 1997,55, 5047. 85. Aggarwal, V. K.; Lightowler, M.; Lindell, S. D. Synthesis 1992,730. 86. Bartlett, P D.; Tate, E. B. J. Am. Chem. Soc 1956, 78, 2473. 87. McMurry, J. E.; Melton, J. J. Org. Chem. 1973,55, 4367. 88. Aggarwal, V. K.; Drabowicz, J.; Grainger, R. S.; Gultekin, Z.; Lightowler, M.; Spargo, P. L. J. Org. Chem. 1995, 60, 4962.
This Page Intentionally Left Blank
RECENT ADVANCES IN THE CHEMISTRY OF a,p-UNSATURATED SULFOXIDES AND SULFONES
Ian Forristal and Christopher M. Rayner
I. Introduction II. Nucleophilic Additions to a,P-Unsaturated Sulfoxides and Sulfones A. Conjugate Addition of Carbon Nucleophiles B. Conjugate Addition of Heteroatom Nucleophiles III. Electrophilic Additions to a,P-Unsaturated Sulfoxides and Sulfones IV. Pericyclic Reactions of a,P-Unsaturated Sulfoxides and Sulfones A. [2+2] Cycloadditions B. [3+2] Cycloadditions C. [4+2] Cycloadditions V. Rearrangements Involving a,P-Unsaturated Sulfoxides and Sulfones A. Pummerer Reactions B. [3,3]-Sigmatropic Rearrangements VI. Miscellaneous Reactions of a,P-Unsaturated Sulfoxides and Sulfones A. Epoxidation and Cyclopropanation Reactions B. Metal-Catalyzed Reactions References
Advances in Sulfur Chemistry Volume 2, pages 155-213. Copyright © 2000 by JAI Press Inc. All rights of reproduction in any form reserved. ISBN: 0-7623-0618-1 155
....
156 157 157 170 176 179 179 181 185 196 196 200 201 201 206 209
156
IAN FORRISTAL and CHRISTOPHER M. RAYNER
I. INTRODUCTION The use of enantiomerically pure sulfoxides to direct the absolute stereochemistry of emerging chiral centers has been the primary feature of numerous publications over the last 20 years.^^"* The efficacy of the sulfoxide in diastereoselective auxiliary-induced reactions is mainly related to the steric and stereoelectronic differences existing between the substituents of the stereogenic sulfur atom: a lone electron pair, an oxygen, and two different carbon ligands, which are able to differentiate the diastereotopic faces of a proximal or even a remote reaction center. Besides the high configurational stability of the sulfmyl group, ^J~*^ the existence of the increasing number of efficient methods to obtain homochiral sulfoxides^ as well as their synthetic versatility has led to a substantial growth of the use of these chiral starting materials. Many natural products contain the sulfoxide functionality (Fig. 1). One example is the antifungal agent sparsomycin (1), a total synthesis of which has been reported.^ The usiltoxins (2a,b) are highly functionalized cyclic peptides and they are important anticancer drug leads because of their potent antimitotic activity. Recently a stereocontrolled synthesis of the sulfoxide side chain was reported."* Also a variety of biologically interesting a,P-unsaturated sulfoxides and sulfones have been synthesized (Fig. 2). Cephams (3) bearing olefmic sulfoxide and sulfone side chains are potential P-lactamase inhibitors.^ Vinyl sulfone cysteine proteinase inhibitors (4) also exhibit antimalarial effects.^ Plantavax® (5) is a well-established fungicide.^ The synthesis and pharmacological evaluation of 6 and other related vinyl sulfone-based anticancer agents have been reported.^ a,P-Unsaturated sulfoxides have also been used extensively in asymmetric synthesis as versatile chiral reagents with the sulfmyl group playing the role of chiral auxiliary.^ The related a,P-unsaturated sulfones are widely used as building blocks in synthetic organic chemistry. ^° The utility of these substrates as starting materials to synthesize highly functionalized systems has been exploited in both cyclic and acyclic vinyl sulfones. Thus, the chemistry of a,P-unsaturated sulfoxides and sulfones has developed into an important area of organic chemistry. R
0
0 r"e
yj-^trC-"""'
rv^s^^^'^"- "o"^'" 1'^ (1)
(2a) R = Me UstiloxinA (2b) R='Pr UstiloxJnB
Figure 1.
a,P-Unsaturated Sulfoxides and Sulfones
Vs
157
O J—H^ 0 \
(3) n=1,2 R = Me,Ar ;BU
O R (4) Ri, R2 = amino acid side chains
(6)
Figure 2.
This review continues from one entitled "The Chemistry of a,P-Unsaturated Sulfoxides" which was recently published in a monograph of organosulfur chemistry.^' In addition, the coverage has been extended to include the analogous a,P-unsaturated sulfones. The focus of the present review are recent advances in the chemistry of a,P-unsaturated sulfoxides and sulfones, especially those which have been published since 1993. The synthesis of a,P-unsaturated sulfoxides and sulfones has already been reviewed extensively by Rayner^ and hence will not be covered here. Only the chemistry of vinyl (alkenyl) sulfoxides and sulfones will be considered. The chemistry of other a,P-unsaturated sulfoxides and sulfones such as dienyl, allenyl, and propargyl (alkynyl) sulfoxides and sulfones is beyond the scope of this review. Considerable emphasis has been placed on stereo- and enantioselective reactions, reflecting the current interest in this area.
II. NUCLEOPHILIC ADDITIONS TO a,P-UNSATURATED SULFOXIDES AND SULFONES A. Conjugate Addition of Carbon Nucleophiles Intermolecular Additions
A highly stereoselective, nickel-catalyzed P-addition of organozincates to optically pure vinylic sulfoxide (7) played a crucial role in the synthesis of the phosphodiesterase IV inhibitor L-765,527 (CDP-840) (9) (Scheme 1).^^ The syn-
158
IAN FORRISTAL and CHRISTOPHER M. RAYNER
•pTol PhaZnMgBr. THF ^5-7% Ni(acac)2 -25°C MeO'
MeO' OCp
(7)
(8) OCp
92o//ee
OCp
Cp = cyclopentyl
Scheme 1.
thetic utility of chiral sulfoxides, both as a stereocontrolling element and related to their ease of desulfurization, is aptly demonstrated by this report. Hydrocyanation of a-trifluoromethyl-P-sulfmylenamines proceeds mainly to give the syn diastereoisomer in high yield but modest stereoselectivity (30% de)}^ Enantiomerically pure (£)-2-halo- (10) and (Z)-2-halovinyl sulfoxides (12) react with anions derived from diethyl malonate through an addition/elimination sequence to give enantiopure 2-malonyl sulfoxides 11 and 13 with retention of stereochemistry of the double bond (Scheme 2).^^ 2,3-Dibromo-l-(phenylsulfonyl)-l-propene (15) undergoes a similar addition/elimination reaction with anions of 1,3-dicarbonyl compounds (14) (Scheme 3). The addition/elimination product after decarboxylation (16) under-
o RCH(C02Et)2
f +^pTol X
(10)
NaH.THF, 12h 56-98% yield EtOsC
X = B r , I.OMs
C02Et O !
RCH(C02Et)2 Bu"Li,THF, 12h 63-73% yield
(^r (13)Et02C
Scheme 2.
C02Et
a,fi'Unsaturated
Sulfoxides and Sulfones
C^^)
159
S02Ph
SOzPh CH2
<> (17) ^ S O z P h
(16) -•
Scheme 3.
went further transformation to yield a 3-(phenylsulfonyl)methyl cyclopentenone (17). This can undergo alkylation reactions to yield functionalized cyclopentenones.^"* Other tandem conjugate addition/annulation reactions have been reported by both Smith and Fuchs^^ and Hassner and co-workers.^^ They used conjugate additions to vinyl sulfones with subsequent intramolecular Sj^2 displacement of suitable leaving groups by the intermediate anionic species resultingfromthe initial conjugate addition. A review by Fuchs and co-workers, highlighting "new synthetic methods exploiting vinyl sulfones," includes examples of the trapping of such anionic intermediates, resulting from conjugate additions, with various electrophiles.^^ Dominguez and Carretero developed highly stereoselective conjugate additions of (£)-Y-oxygenated-a,P-unsaturated phenylsulfones with organolithiums and organocuprates (Scheme 4). The y-hydroxy derivative (18a) underwent conjugate addition with an organocuprate to give addition product (19) with high anti selectivity.^^ Protection of the y-hydroxy group as the methoxymethyl ether (18b) resulted in a reversal of stereoselectivity. Chelation between the protecting group and the metal ion of the organolithium nucleophile resulted in the syn product (20) OH
Me3CuLi2, Et20, 25°C R =H
Ph02S'
'Pr Me anti
(19) Ph02S' (18a) R = H (18b) R = MOM
R = MOM i. "BuLi, THF,-78°C il. TMSCI iii. MeLi. -78°C iv. KF, MeOH
OMOM Ph02S
Pr Me
(20) Scheme 4,
88% de yield = 89%
syn
> 96% de yield = 85%
IAN FORRISTAL and CHRISTOPHER M. RAYNER
160
being formed. ^^ However, the a-position must first be protected as the silyl derivative prior to conjugate addition with methyllithium, otherwise deprotonation occurs. The same group utilized this methodology for the construction of polypropionate chains. ^^'^^ Sugars are a useful source of chirality in organic synthesis. Isobe and Jiang have synthesized D-glucose derivatives with a vinyl sulfone conjugated to the heteroatom at the C-1 position of the carbohydrate and a secondary alcohol at the C-2 position (21a). Lithium alkyls and acetylides undergo highly diastereoselective conjugate additions to these compounds (Scheme 5).^^'^^ Simple switching of the syn-anti selectivity was achieved by protection or nonprotection of the 2-hydroxy group. The unprotected alcohol (21a) underwent addition through P-chelation control to yield the anti diastereoisomers (22a^3a), whereas the protected alcohol (21b) resulted in the syn diastereoisomers (22b,23b). Since protection of the free alcohol had blocked the P-chelation between the p-hydroxy and the metal ion of the nucleophiles, the selectivity had been successfully switched to an a-chelation-controlled product. This methodology has been exploited in the synthesis of tautomycin.^^ Toru and colleagues have investigated radical-mediated conjugate additions to a,P-unsaturated sulfoxides possessing an activating group at the a-position. They reported highly diastereoselective P-additions of alkyl radicals to chiral 2-(arylsul-
TBDMS i.Li = : SIMe R=H p«chelation
(22a) O R i ioaPh TBDMSO—y Me jmiDVj \ Me syn: anti i- MeMgBr^ " ^ ^ ^ ^ T ^ ^ ^ 4:96r^^ ii. KF (23a) OR^ S02Ph^^'''°
TBDMSO—X
M (21a)R = H ^^^ (21b)R = TMS
S02Ph
Li i ^ SiMe^ ii. KF R = TMS a - chelation
.m^
TBDMSO—A
|||
syn: anti >100 :1
(22b) 0R^!^ ^02Ph TBDMSO—\ Me I Duwioyj \ Me syn: anti \ r\ ' 8j1 i. MeLi 72% ii. KF
(23b) ORi
Scheme 5.
iozPh
a,P-Unsaturated Sulfoxides and Sulfones
161
^Bul, R3B, air CH2CI2, 0°C 0.01 mol/L
(25)
^Bu*
(26)
^Bu*
Ar = 2,4,6-trimethylphenyl Lewis acid
EtAICl2
yield (%)
ratio (25): (26)
99
>98 : <2
99
< 2 : >98
Scheme 6.
finyl)-2-cyclopentenone (24) (Scheme 6)}^ Radical P-addition gave diastereoisomer 25 in both excellent yield and selectivity, whereas in the presence of a Lewis acid there was a reversal in observed stereoselectivity. Now the other diastereoisomer (26) was formed, again with excellent yield and selectivity. The observed change in stereoselectivity can be rationalized by consideration of the conformation of the 2-(arylsulfmyl)-2-cyclopentenone (24) (Fig. 3). The sulfinyl and carbonyl moieties are normally arranged in an anti periplanar orientation (27). The bulky aromatic substituent on the chiral sulfmyl group shields one face of the alkene and thereby controls the facial selectivity of the reaction. In the presence of the Lewis acid the sulfmyl and carbonyl moieties are locked in a syn orientation (28) as a result of chelation between the two moieties and the metal. Thus, the opposite face of the alkene is shielded and P-addition results in the other diastereoisomer being formed. These results are analogous to the pioneering work of Posner, who initially developed highly stereoselective conjugate additions of carbon nucleophiles to chiral 2-(arylsulfinyl)-2-cycloalkenones.^^ This methodology has been extended to include novel diastereomer differentiating radical P-additions in which the two
IAN FORRISTAL and CHRISTOPHER M. RAYNER
162
OH '
^Bul, R3B,
O !
OH
OH
•t^f-pTol +
CH2CI2 -78°C 0.01 mol/L
(29)
0
(^°)
"^^"^
0 T ^^+^pTol
(31)
Lewis acid
yield (%)
ratio (30): (31)
EtAICl2
75
>98 :<2
Scheme 7.
diastereoisomers are kinetically separated. Additions of alkyl radicals to a diastereomeric mixture of (4K)' and (^5)-methyl-2-[(2,4,6-triisopropylphenyl)sulfinyl]-2-cyclopentenones gave the diastereomerically pure addition product derived from the (4K) isomer, while the (^5) isomer remained unreacted.^^ Radical P-additions to acyclic chiral a,P-unsaturated sulfoxides resulted in nonstereoselective addition products or unexpected Pummerer rearrangement products.^^ This illustrates the difficulty in achieving good stereoselectivity for conjugate additions to acyclic as compared to cyclic a,P-unsaturated sulfoxides. Recently there was a report of a highly stereoselective reaction of an acyclic a,P-unsaturated sulfoxide. Hydrogenation of an a-sulfmyl radical generated from an alkyl radical addition to a-(l-hydroxyethyl)-vinyl sulfoxide (29) gave the addition/hydrogenation product 30 with high diastereoselectivity (Scheme 7).^^ Radical addition to Y-hydroxy-(a-methylsulfmyl)-a,P-unsaturated sulfone (32) was observed in excellent yield and high stereoselectivity for the syn diastereoisomer (33) (Scheme 8).^^'^^
OH
SMe
i. Pr^OH, hw, PhsCO. 97 SO2PTOI
SO2PT0I
ii. Ra / Ni, Eton. 84%
(32)
(33)
Scheme 8.
syn: anti >95:5
a,^-Unsaturated Sulfoxides and Sulfones
163
SN2' Additions
The nucleophilic opening of vinyl oxiranes by Sj^2' reactions with organocuprates is a powerful synthetic tool for the stereoselective construction of carboncarbon bonds. Marino and co-workers showed that enantiomerically pure epoxy vinyl sulfoxides undergo highly regio- and stereoselective Sj^2' ring openings with cyanocuprate reagents (Scheme 9).^^ Conjugate addition to substrate 34b gave exclusive formation of the anti product 35b, whereas its diastereoisomer 37c gave predominately the syn product 38c (Table 1). The stereochemical outcome of these additions may be rationalized in terms of a "matched" situation for 34a,b in which the sulfmyl and vinyl epoxide functionalities display cooperative stereodirecting capabilities. On the other hand, the remarkable reversal of selectivity found for 37a~c, affording 1,4-syn products, suggests a "mismatched" situation in which the sulfinyl group can override the highly anti-sclectiwQ stereochemical pathway associated with Sisj2' displacements of vinyl oxiranes with organocuprates. Thus, the anti-syn stereochemical course of the process is primarily controlled by the chiral sulfur atom. Enantiomerically pure epoxy vinyl sulfoxides also undergo a
R2^
'
R2>.
?
"RCu"
VTol (34)
VS-TOI
^^
-78 ° C ^ r t EtgO
Ri^-R R2\
I?
^
"RCu"
-78 °C->rt (37)
^OH
>Tol
(35)
Rl"^R
,0H
T
l?S,n
'
f?
(36) Anti
• ' ^ ^
EtgO
R^^'R
(38)
"R (39)
Scheme 9. Table 1. Reaction of Cyanocuprate Reagents with Epoxy Vinyl Sulfoxides Product Ratio rieia
Substrate
R2CU
35
36
96 100
4
34a 34b
R^ = R2 = n-Bu R^ = Ph, R2 = n-Bu
MeCuCNLi n-BuCuCNLi
37a 37b 37c
R^ = R2 = n-Bu R^ = R2 = n-Bu R^ = Ph, R2 = n-Bu
MeCuCNLi EtCuCNLi n-BuCuCNLi
38
39
(%) 91 87
— 85 100 91
15
— 9
68 70 78
164
IAN FORRISTAL and CHRISTOPHER M. RAYNER
base-induced rearrangement to give hydroxy 2-sulfinyl dienes with remarkable geometric selectivity. Once again, the observed stereoselectivity is completely controlled by the absolute configuration of the chiral sulfur atom.^^ Acyclic stereocontrol remains a challenging problem in synthesis. While enantiomerically pure sulfoxides are valuable synthetic intermediates for enantiocontroUed carbon-carbon bond formation by conjugate addition in cyclic cases, their usefulness for such alkylations in acyclic cases has not been firmly established. Moreover, most sulfoxide directed alkylation protocols utilize the valuable sulfur auxiliary just once, which limits the synthetic versatility of the process. Marino et al. have recently reported Sj^2' displacements of acyclic ally lie mesyloxy vinyl sulfoxides with organocopper reagents (Scheme 10)."^^ In addition to the excellent observed stereoselectivities, the newly created chiral center is adjacent to a vinyl sulfoxide which should allow for subsequent chirality transfer operations. On treatment with organocopper nucleophiles, both sulfoxide diastereoisomers 40b and 43b underwent Sf^2' displacements with high Z selectivity to yield products 42b and 45b, respectively (Table 2). The oxidation state on the sulfur was varied
THF
M e ^ P h (41)
Me-'^^Ph (42)
f
(0)n "R2CU"
-78 X->rt
^ ^ S > ^
>- E t - ^ ^ V ^ %-Tol ^
THF
{^)n
. Y^
^ . ^ v ^ ^S^
Me'''*^Ph (44)
^ ^
^p-Tol
Me^^^Ph (45)
Scheme 10. Table 2, Reaction of Organocopper Reagents with Ally lie Mesylates Product Ratio Substrate
R2CU
41
42
6 6 93
94 94 7
40a 40b 40c
n=0 n= 1 n=2
MeCuCNMgBr MeCuCNLi Me2CuCN(MgBr)2
43a 43b 43c
n=0 D=1 n=2
MeCuCNMgBr MeCuCNMgBr Me2CuCN(MgBr)2
44
45
Yield (%) 61 81 81
6 6 91
94 94 9
58 80 76
a,P'Unsaturated
Sulfoxides and Sulfones
165
X
••
st =
i. MS2O, pyridine, 0°C, 84% St
^S. ii. MeCuCNLi. THF. 76% O^ >Tol (46)
^
^
^ 1
r HT y
HJ H
St x^S.^ (47)
^
100% de
J
Scheme 11.
and similar Sj^2' displacements of the corresponding sulfides 40a, 43a and sulfones 40c, 43c were investigated. The vinyl sulfides displayed high stereoselectivity toward the Z isomers 42a, 45a, as was found for the sulfoxides. However, the corresponding sulfones 40c, 43c allowed for a reversal of stereochemistry. They displayed a very high E selectivity for this process, to yield products 41c, 44c. Thus, the absolute configuration of the displacement products was controlled by adjusting the oxidation level of the sulfur auxiliary. Overall the stereochemical outcome of these processes is primarily controlled by the configuration of the allylic mesylate. Allylic mesyloxy sulfinyl steriod 46 also underwent highly stereoselective Sj^2' displacement when treated with a cyanocuprate to yield 47 (Scheme 11). In this manner, a formal synthesis of the plant growth regulator brassinolide was achieved.^"^ Efficient regio- and stereocontrolled Sj,j2' alkylations of oxabicyclic vinyl sulfones with concomitant cleavage of the oxygen bridge have been developed. A range of 7-oxabicyclo[2.2.1]heptenyl and 8-oxabicyclo[3.2.1]octenyl sulfones underwent an overall syn Sjsj2' opening when treated with organolithium reagents."^^ Applications of this methodology toward the synthesis of natural products have been investigated by the same group (Scheme 12). The 7-oxabicyclo[2.2.1]heptenyl
Me
PhS02 ^v/^^Me
-78°C.94%
Me^^j^^Me (49) OH
' O
Me
"TBS ^ " OBz
(50) OH
P h S 0 2 ^ ^ ^ ^ 0 B n ^^^j ^^p P h S 0 2 > ^ ^ ^ x ^ 0 B n ^ ^ ^ ^ O B n -78°C. 85% (51)
Me^^j^OBn (52) OH
Scheme 12.
MeO^^^k^OBn M e ^ ^ ^ O B n (53)
NH2
IAN FORRISTAL and CHRISTOPHER M. RAYNER
166
sulfone (48) underwent Sj^2' opening with methyllithium to give 49 as the exclusive product. This was readily converted into a polypropionate chain (50), with a high level of stereocontrol.'^^ Another 7-oxabicyclo[2.2.1]heptenyl sulfone (51) underwent a similar Sj^2' opening with methyllithium, again with complete stereocontrol to give 52. Further transformation led to the important fragment of the alkaloid pancratistatin (53). A recent paper highlighted the synthetic applications of functionalized cyclohexenylsulfones obtained by Sj^2' ring opening of such oxabicyclic compounds with lithium acetylides.^^ Intramolecular Additions
The enantioselective synthesis of the tetrahydroisoquinoline (/?)-(+)-carnegine (56) has been reported.^^ The crucial step consisted of an acid-catalyzed cyclization of P-amino-a,p-unsaturated sulfoxide (54) onto the indole ring to give a single diastereomeric product (55) (Scheme 13). An asymmetric reductive cyclization utilizing the intramolecular conjugate of enolates onto a,P-unsaturated sulfoxides has recently been reported."^^ When 57 was treated with L-Selectride®, an intramolecular Michael addition of the resulting ester enolate (58) onto the homochiral vinyl sulfoxide gave functionalized cyclohexane (59) as a single isomer (Scheme 14). Jones and colleagues had previously developed reductive cyclizations of vinyl sulfones using hydridoaluminates. In this example the vinyl sulfone was initially reduced and the anionic species generated by conjugate addition of the hydride onto the vinyl sulfone underwent intramolecular acylation."^^ Reaction of lithiated allyl sulfone 60 with cyclopentenone gave the y-1,4-addition product (61). On treatment with LiOH the enolate intermediate 62 underwent a completely diastereoselective intramolecular conjugate addition onto the vinyl
^''''^^^^^p^
/NH
j p A CHCI3. 0°C. 65% Ar= P-NO2C6H4
MeO"
^^ (56)
N
Me
(RH+ycarnegine Scheme 13.
a,^-Unsaturated
Sulfoxides and Sulfones
^ - s j ^ ^ UX02Me L . ; , . . . ^ Li(s-Bu)3BH
167
H r^C02Me
OMe
-30^C. 1h then10°C.2h •^-^^^v''n Q--^w^/p.jol
^ " " ^ ^
O " ^ >^'p-Tol (59)
(57)
100% de
Scheme 14.
sulfone, which resulted in the formation of the bicyclo[2.2.1]heptane derivative 63 (Scheme 15). It should be noted that cyclization using potassium r^rr-butoxide gave a mixture of two diastereoisomeric bicyclic products, thus confirming that stereoselective formation of the 6-^nJo-methylsulfonyl substituted bicyclo[2.2.1]heptan-2one 63 was related to the presence of the lithium ion."*^ Pyne et al. have reported a similar sequence using a variety of five- and six-membered cyclic enones.'*^ In recent years, methods for ring construction based on intramolecular addition of radicals to activated carbon-carbon double bonds have gained great attendon, because of the mild conditions required for the radical generafion and the wide functional group tolerance usually displayed by this type of reaction. As 6- and, particularly, 5-exo closures are very favored processes, six- and five-membered rings have been prepared by this methodology. Tetrahydrofuran derivatives have been synthesized using this protocol. A S-exo-trig radical cyclization of optically
© (60)
®\-\
PhOgS'
Scheme 15,
168
IAN FORRISTAL and CHRISTOPHER M. RAYNER
II X. P-T0I--S
/
r.j/ (64) S O s r Me'
TTMSS, AIBN. PhH. A, 6h
2/ ^ 0\ ^y Me'
1
pTol
(1R. 3R) """
79% yield 1 >96%de
(65)
^Me Scheme 16.
pure P-alkoxy vinyl sulfoxide 64 gave the functionalized tetrahydrofuran derivative 65 in good yield and excellent diastereoselectivity (Scheme 16)."*^ Intramolecular 6-^;cc?-radical cyclization of 7-bromo-3-methoxy-l-methylthiol-(p-tolylsulfonyl)-l-heptene (66) exhibited a highly efficient 1,2-asymmetric induction to yield the trans ring-closure product 67 (Scheme l?)."^^ Also reported are radical cyclizations of Y-oxygenated-a,P-unsaturated sulfones, often with very high observed stereoselectivity."^^'"^ Carretero and co-workers reported that the a-sulfonyl radicals resulting from such radical additions to Y-oxygenated-a,P-unsaturated sulfones are useful intermediates for the generation of a second carbon-carbon bond via intramolecular addition to a suitably located carbon-carbon double bond."^^'"^^ The highly functionalized acyclic Y-hydroxy vinyl sulfone 68 underwent a novel "cascade" process, based on two sequential radical cyclizations, affording the bicyclic compounds 69 and 70 in good yield (Scheme 18). The results can be rationalized as follows: the first cyclization of substrate 68 gave a mixture of a-sulfonyl radical intermediates 71 and 72 (Scheme 19). These then underwent a second fully stereoselective cyclization to give the cis- and trans-fused bicyclic products 69 and 70, respectively, as single isomers. Such sequential transformations, in which two carbon-carbon bonds are formed in a single step, are attractive methods for enhancing the efficiency of organic synthesis. As Y-hydroxy-a,P-unsaturated phenyl sulfones can be prepared in enantiomerically pure form,"^^ this procedure should be readily applied to the synthesis of enantiomerically pure bicyclic products.
Br«
^Ph P—/
SMe i. "BuaSnH, AIBN. PhH \
_
(67) Scheme 17.
/
•
"
trans
82% yield 48% de ^
SOzPh
169
a,p-Unsaturated Sulfoxides and Sulfones
^
r
S02Ph ^S02Ph
O^^V^
SOzPh BuaSnH, AIBN,
"SOzPh
^eHe. A
H
S
W
(69) 37% ?02Ph '
(68) H
V.
^S02Ph
(70)
44%
Scheme 18.
.S02Ph
0x-^^^S02Ph
. r^Y^S02Ph ^"""^ (71) I ^—^
+
^SOgPh trans
c/s (72)
(68)
fii
SOzPh
^^S02Ph
l_,
° ^ S O , P h
S02Ph
w
^S02Ph (69)
H
CIS' fusion
Scheme 19.
^OzP*^ ^SOaPh
(70)
H
trans- fusion
IAN FORRISTAL and CHRISTOPHER M. RAYNER
170
B. Conjugate Addition of Heteroatom Nucleophiles Intermolecular Additions
There are relatively few conjugate additions of heteroatom nucleophiles to a,P-unsaturated sulfoxides and sulfones reported in the literature, unlike conjugate additions of carbon nucleophiles. Despite their small number, such additions have been used in the synthesis of several natural products. Conjugate addition of J/-valine to phenyl vinyl sulfoxide resulted in an intermediate used in the synthesis of pentafluorophenylthiohydantoin derivatives."^^ One of the early steps in the total synthesis of (±)-deethylibophyllidine consisted of conjugate addition of an amine to phenyl vinyl sulfoxide.^^'^^ Many classes of natural products contain P-amino acid derivatives as fragments. The conjugate addition of ammonia to tert-buiy\-(E)' 2-[(5)-/7-tolylsulfmyl]cinnamate (73), followed by successive reduction of the sulfinyl group with samarium(II) iodide, proceeded smoothly to give (R)-tert-buiy\ P-amino-P-phenylpropionate (74) with good optical purity (Scheme 20). The six-membered hydrazine, piperidazine, underwent conjugate addition and subsequent cyclization with 73 to yield 75, with complete control of the stereochemistry. This was converted to the natural product (5)-celacinnine (76) in five steps.^^ Solid-phase organic synthesis (SPOS) has been revolutionary in the quest to produce libraries of structurally diverse molecules both quickly and cheaply. The utility of benzyl and aryl vinyl sulfones in the "traceless linker" synthesis of amines
i. NH3, THF, rt '0*Bu
p-Tor
•
O^Bu
ii. Sml2, MeOH H2N Ph (R,E)-{73)
Ph' NH2 (S)-(74) 74% ee
KO^u. THF ii. Sml2, MeOH
Ph' (S)-(75)
100% ee
(76) Scheme 20.
{S}' celacinnine
a,p-Unsaturated Sulfoxides and Sulfones
171
Br
Br Nu0
Nu
r (77)
°' W
0 S' 02
(78)
Br Nu
Nu
0
(79)
(80) Scheme 21.
Table 3. Reaction of a-Bromo Vinyl Sulfone 77 with Heteroatom Nucleophiles Nu NaOMe BnSH (5)-PhCHMeNH 2
Product
% Yield
r£:Z)
80a 80b 80c
52 77 53
75:25 93:7 21 : 79
has been described.^^ The vinyl sulfone group reacts efficiently with secondary amines, via conjugate addition, and the resin-bound tertiary amine products can be quaternized through alkylation. Subsequent deamination gave the corresponding tertiary amines and the regenerated vinyl sulfone was recycled. Evans and Taylor have recently reported that a-bromo-a,|3-unsaturated sulfones (77) undergo a novel tandem conjugate addition/Ramberg-Backlund rearrangement (Scheme 21).^'* A variety of heteroatom nucleophiles were used in the synthesis of allylic ethers (80a), sulfides (80b), and amines (80c) (Table 3). This Michael-induced Ramberg-Backlund (MIRE) reaction is believed to proceed via an initial conjugate addition of the heteroatom nucleophile onto the vinyl sulfone (77—>78), followed by proton exchange (78—>79) and Ramberg-Backlund rearrangement (79-»80). A chemoselective addition of allylic alcohols (82) to 3-halogenovinyl sulfones (81) has been accomplished using KF-basic alumina as the basic medium.^^ The resulting adducts (83) can be stereoselectively cyclized by a radical process, affording 2,4-disubstituted tetrahydrofurans (84) (Scheme 22, Eq. 1). The same group has recently reported an analogous protocol using allylic amines (85) for the synthesis of 2,4-disubstituted pyrrolidines (87) (Scheme 22, Eq. 2).^^
172
IAN FORRISTAL and CHRISTOPHER M. RAYNER
PhSOi^ ""^^ " X (81) KF-AI2O3
.Me
X BuaSnH, EUB,
(84) 2h. rt
[Eqn- 1]
O2, -78''C SOzPh
HC^^-^ (82)
70%
SOoPh
78%
trans : cis 8 6 : 14
(83)
Me
PhSOi^ ^
(81)
BuaSnH. AIBN,
KF-Al203^
• •
2h. rt
H
S02Ph
HoN (85)
OQHQ, A
N^ I
75%
(87)
[Eqn.2]
u
S02Ph
70%
trans: cis 85:15
(86)
X = CI, Br Scheme 22.
The facial selectivity for the reduction of a-(fIuoroalkyl)-P-sulfinylenamine (88) with K-Selectride® was controlled by the sulfoxide, and proceeded with high diastereoselectivity to yield product 89 (Scheme 23).^^ Intramolecular Additions Intramolecular nucleophilic addition of alkoxides to vinyl sulfoxides (90) provided a route to P-alkoxysulfoxides (91) (Scheme 24). The cw-product was formed with up to 18 : 1 selectivity.^^ Alkoxide nucleophiles undergo an addition/elimination protocol with P-iodo vinyl sulfoxides to yield tetrahydrofurans and tetrahydropyrans, which contain either an endo- or an ^jc<3-cyclic double bond.^^'^ A sugar-derived vinyl sulfone (92) was found to undergo a Michael-initiated ring closure (MIRC) process to build up a chiral polysubstituted oxolan system with high stereoselectivity, yielding the all-^^'M stereoisomer (93) (Scheme 25).^^ The observed MIRC selectivity demonstrates that the S-exo-thg process is strongly O II p-Tor ' ^ ^
NHo "CF3
K-selectride THF, 75%'
(88)
O p-Tor"^^ (89)
Scheme 23.
NH2 "CF3 86% de
a,p-Unsaturated
Sulfoxides and Sulfones
173
ACOL
Qy ii. MeONa, MeOH iii. AC2O, pyridine Ao>0. Dvridine
^ N ^ ^ S - ^ Q r\
,OAC
O
"1^ i
Bu
(93)
Scheme 24.
favored over the 6-exo-ing option, which is in agreement with Baldwin's general conclusions.^^ Carretero and co-workers reported a one-step synthesis of functionalized dioxaspiro[4.5]decanes from P-phenylsulfonyl dihydrofurans and y-lactones.^^'^ The method is based on the acylation of the anion derived from 94 with y-butyrolactone to afford the intermediate alkoxide 95, which undergoes intramolecular conjugate addition to the vinyl sulfone moiety to give the 4-phenyIsulfonyl-1,6-dioxaspiro[4.5]decan-10-one (96) as a single isomer (Scheme 26). The spirocyclization product 96 was converted into its thermodynamically more stable isomer 97 on treatment with lithium hydroxide. Thus, a quantitative and complete epimerization, at the carbon bearing the phenyl sulfonyl group (C-4), was achieved. These 4-phenylsulfonyl-l,6-dioxaspiro[4.5]decan-10-ones, on treatment with sodium amalgam, underwent desulfonylation and unexpected transformation into 1,6-dioxadecalins.^^ The stereoselective synthesis of 2,5-dialkyl-3-(phenylsulfonyl) tetrahydrofurans (99) via cyclization of (Z)-sulfonyl-substituted homoallylic alcohol (98) has been reported.^^ A highly a/i/Z-stereoselective 5-endo-ing cyclization reaction was observed (Scheme 27, Eq. 1). A similar 5-endo-ing cyclization of (£)-vinylic sulfone (100) led to the virtually exclusive formation of the 2,5-5>'AZ-disubstituted-3(phenylsulfonyl) pyrolidine (101) (Scheme 27, Eq. 2).^^ Also reported is a highly stereoselective synthesis of c/5-2,6-disubstituted tetrahydropyrans from intramolecular addition of an alkoxide onto a vinyl sulfone.^^ Allylic trichloroacetimidates, generated in situ from cyclic Y-hydroxy-a,Punsaturated sulfones, undergo intramolecular conjugate addition to the vinyl sulfone moiety to afford oxazolines. Acid hydrolysis of the oxazolines generated
AcO, i. 90%CF3CO2H g^j ii. MeONa, IVIeOH iii. AC2O. pyridine
^N^^^S^" O2
,OAc
.-0-. O
(93) Scheme 25.
Bu ^T^ QAC
174
IAN FORRISTAL and CHRISTOPHER M. RAYNER QPhOzS
Ph02 i. "BuLi, THF. -78°C
//
> •
O
'Oc
(94)
r ° ,S02Ph
V-^
^ °
(96)
>96:4
SOzPh
(97)
LiOH. THF >H20
I
Scheme 26.
vicinal cw-amino alcohol derivatives of five- and six-membered rings.^^ A highly stereoselective S-exo-ihg intramolecular carbamate cyclization was used for the preparation of ^^^n 2-amino alcohol derivatives.^^ The y-hydroxy vinyl sulfone 102 was converted to the corresponding imide 103 which underwent cyclization, and partial hydrolysis, to give the /mAZ^-Oxazolidinone 104 (Scheme 28). Following
PhS02v^^x\^'Pr
*BuOK (1 eq.), *BuOH (10 eq.) PhSOz [Eqn. 1]
J^ Ph" (98) PhS02
OH (Z) - isomer Bu p^ NHDPP
(100)
THF (0.032M). rt 83%
freshly ground NaOH (3 eq.) 1.4-dioxane (0.1M). rt 70%
(E)' isomer
Scheme 27.
Ph^ ^O (99)
'IPr
2,5'anti : 2,5-syn 10 : 1
PhSOo
..-D-.
[Eqn. 2]
PM^'^N^^'BU DPP (101) 2,5-anti : 2,5-syn 1 : >20
a.fi-Unsaturated Sulfoxides and Sulfones
175 O
R
O
XX
Cl3C(0)NC0
O'^^N'^CCIa H
K2CO3, THF, rt
89 - 92% yield
P^OzS'
(103) K2CO3, MeOH, CH2C/I2 80 - 94% yield O
U BnN'^0 ) Ph02S-<' (105) E
A.
i. BnBr, K2CO3, THF, A -<-
1 \
ii. "BuLi, THF, -78°C lii.E"
E* =RCO CI, RCHO, RX, MeS02SMe
HN
\ i Ph02S—' \ trans-{^04)
59 - 98% yield
82-92% de
Scheme 28,
benzylation, these systems could be further functionalized by lithiation and subsequent alkylation to yield products 105. Naturally occurring polyhydroxylated indolizidine and pyrrolizidine alkaloids exhibit diverse and important physiological properties, spurring the intense efforts to develop an efficient synthesis of these compounds. The pyrrolidine intermediates 107a,b of such alkaloids can be formed by highly stereoselective intramolecular conjugate additions of nitrogen moieties onto the vinyl sulfone unit of N-substituted Y-oxygenated-a,P-unsaturated sulfones (106a,b) (Scheme 29)?^'^^^ The stereoselectivity of the pyrrolidine synthesis was highly dependent on the bulkiness of the
RO ) ^ ^ ^ SOzPh
j TFA, CH2CI2
^^^^^^^^^^2^^ BOC
ii. Et3N.-78°C
(106a) R= H (106b) R = TIPS
RO
^'^ S02Ph
W
V-N,^^
RQ ^'•^^^ S02Ph
C02Me
(107a) R = H (96%) cis -107a / trans - 107a = 81 /19 (107b) R = TIPS (93%) c/s-107b / frans-107b =10/90 Scheme 29.
176
IAN FORRISTAL and CHRISTOPHER M. RAYNER
A
THF. 0.3eq.*BuOK,
O
OTBS SOsPh
OTBS SOzPh {;i/E
=1:2)
(109)
(108)
=0
o TBSO, e^A. y^ I
TBSO, 6 J ^ . J l I
+
I
TBSO.
II
' Ph
32%
SOzPh
(111) 45%S02Ph
(112)
SOzPh
(110)
Scheme 30.
Y-oxygenated function. The free alcohol gave predominately cw-pyrrolidines 107a whereas the tri-Z^c^-propylsilyl ether (OTIPS) derivatives led to the trans isomers 107b. Piperidine rings can also be synthesized using the same methodology. They undergo further transformation into the related 1,2,9-trihydroxylated quinolizidines.'^^ The enantioselective synthesis of (-)-sedacryptine, a piperidine alkaloid, has been achieved via the "double intramolecular conjugate addition" of a carbamate group onto a vinyl sulfone and then an enone (Scheme 30). The first conjugate addition of 108 proceeded in a syn-1,3 fashion. The successive cyclization of the resulting carbamate anion 110, which was formed from carbanion 109 via proton transfer, gave a mixture of stereoisomeric products 111 and 112. Both of these isomers were converted into the target natural product.^^
III. ELECTROPHILIC ADDITIONS TO a,P-UNSATURATED SULFOXIDES AND SULFONES Electrophilic additions to vinyl sulfoxides and sulfones allow for further functionalization of these compounds at the a-position. Deprotonation of the a-hydrogen, under appropriate conditions, gives rise to a-vinyl anions which can be trapped by electrophiles. This protocol has been employed in the synthesis of a diverse range of compounds, such as 34 and 37, 40 and 43, and 46 (see Schemes 9, 10, and 11, respectively). Both enantiomers of optically pure propargylic alcohols (115) were conveniently prepared by the reaction of the a-vinyl anion of 2-(trimethylsilyl) vinyl
a,^-Unsaturated Sulfoxides and Sulfones
\77
i. 2 eq. LDA, THF,-100°C
-^- p-ToK+
p-Tol'^+
(113)
OH
ii. 2eq. "C5H11CHO SiMe'3 THF,-100°C
"C5H11 p.ToK+ SiMea
SiMea
r/^;-(ii4)
rs;-(ii4)
rs;.(ii4) : r^;-(ii4) 73 : 27 (Method A )
rs;-(ii4)
Method A : X = H, yield = 83%
QH
NaH. THF. 0°C
[Method B ] Toluene, A
"C5H11
Method B : X =SiMe3, yield = 87%
rs;-(ii5) ^ Scheme 31.
/7-tolyl sulfoxide (113) with aldehydes and either subsequent desilylsulfinylation or thermal elimination of the sulfmyl group (Scheme 31)7^ Electrophilic addition proceeded with moderate stereoselectivity, showing preferential formation of the (5)-114 isomer. Each isomer was readily separated by chromatography and the (5)-114 isomer was subjected to desilylsulfinylation or thermal elimination of the sulfmyl group, to give the corresponding (5)-propargylic alcohols (115) in enantiomerically pure form (>99% ee). Jin and Fuchs reported that vinyl sulfones, using basic phase-transfer catalyst conditions, were regiospecifically alkylated at the a-position7^ No P-elimination products were observed in systems capable of undergoing anion-promoted P-elimination. He also reported that y-methoxy vinyl sulfones 116 can be converted to the corresponding P-substituted enones 119 using this protocol (Scheme 32).^^ On reaction with r^rr-butyllithium, 116 is converted to the y-methoxy allylsulfonyl anion 117, which was regiospecifically trapped by a variety of electrophiles to provide the enol ether 118. On hydrolysis the P-substituted enone 119 was obtained.
(116)
(117)
(118) Scheme 32.
(119)
IAN FORRISTAL and CHRISTOPHER M. RAYNER
178
PhOjSx /
\
Vv°
sec-BuLl.THF. ^.
Li
ti
-78°C (121)
Smio
P
^ THF, HMPA
PhOsS. (122)
(123)
V
Scheme 33, Table 4. Reaction of a-Vinyl Anion (121) with Electrophiles Entry a
b c
d e
Yield (%) of Yield (%) of Z/E ratio of 122 123 123
Electrophile Mel CH2=CHCH2Br PhCHjBr
Me CH2=CHCH2 PhCH2
90 46 50 52 41
42 85 75
88: 12 99:1 94:6
— —
— —
Several synthetic applications of lithiated vinyl sulfones have been reported by Ndjera and co-workers.^°'^^ The reaction of (£)-^-(3-tosyl-2-propenyl)morpholine (120) with 5ec-butyllithium led to the vinylic y-aminated organolithium species (121) (Scheme 33). This reacted with alkyl halides and carbonyl compounds to afford alkylated compounds 122a-c or unsaturated aminoalcohols 122d~e, respectively (Table 4). Only 1,2-addition was observed when a,P-unsaturated compounds were used as electrophiles (entry e, Table 4). The corresponding alkylated compounds 122a-c were stereoselectively reduced with samarium(II) iodide to the y-alkylated allylic amines 123a-c with mainly Zconfiguration. Vinyl sulfone 124a, derived from 3-buten-l-ol, and its MOM derivative 124b were regio- and stereoselectively lithiated at the 4-position to afford intermediates 125 and 127, respectively (Scheme 34). The protected derivative 124b can be alkylated at the vinylic position and both intermediates can also react with carbonyl compounds to give 1,5-diols (Table 5). An interesting application of this protocol was utilized in the total synthesis of (±)-pumiliotoxin.^^ Intramolecular acylation of the a-anion of vinyl sulfone 128 gave enaminone 129 which was readily converted to the desired alkaloid pumiliotoxin 130 (Scheme 35).
179
a,/^'Unsaturated Sulfoxides and Sulfones
PhOzSv
THF.
^OH
ii. H2O (124a)
-^®°^
(125)
n-BuLi. Ph02S, THF. "OMOM
^ .78°C
(127) ^
.+ Ph02Sv i. E ^^^-cT T H P ^
^OMOM
^ E (126b)
Scheme 34. Table 5.
Reaction of a-Vinyl Anions (125,127) with Electrophiles
Anion
Electrophiles
125 127 125 127
Mel Mel PhCHO PhCHO
E Me Me PhCHOH PhCHOH
Yield (%)
— 72 72 73
(130) (±)- Pumiliotoxin Scheme 35.
IV. PERICYCLIC REACTIONS OF a,P-UNSATURATED SULFOXIDES AND SULFONES A.
[ 2 + 2 ] Cycloadditions
2,3-Dihydrothiophene-1,1-dioxide (2-sulfolene) (131) underwent a photochemical [2+2] cycloaddition with maleic anhydride to yield cycloadduct 132 which contained a ci^y-fused cyclobutane (Scheme 36).^^ Simple reactions of the anhydride function provided access to a wide range of novel bi- and tricyclic sulfones 133-135 containing the novel 2-thiabicyclo[3.2.0]heptane-2,2-dioxide ring system. Lewis acid-containing enolates, such as dimethylaluminum enolates, are extremely good partners for [2+2] cycloadditions with the otherwise relatively unre-
180
IAN FORRISTAL and CHRISTOPHER M . RAYNER ,0 HjOorROH
O2
r-^
o (132)
RNH2 I yT'-.
^(133) ( R = H.Me.Et.'Pr.CH2Ph)
^COzH
S" O2
^COzR
.H2O
/"'
^CONHR
S' O2
(134) Scheme 36.
active phenyl vinyl sulfoxide. When phenyl vinyl sulfoxide (137) was reacted with l-dimethylaluniinumoxy-l,3-cyclohexadiene (136), instead of the expected [4+2] Diels-Alder adduct (139), a mixture of two diastereomeric [2+2] cycloadducts (138) was cleanly formed (Scheme 37).^"^ High c/^-stereoselectivity between the hydroxyl and the sulfoxide moieties was observed. This is probably a direct consequence of the preassociation of the reactants, caused by a strong aluminumsulfoxide complexation, and strong steric interactions in the cyclic transition state. From a preparative point of view, this protocol is one of the best ways to prepare functionalized cyclobutanols, with the remaining sulfoxide moiety giving a powerful handle for further diversification. An unusual nickel-catalyzed [2+2+2] homo-Diels-Alder reaction of norbomadiene (140) with (5)-(-)-/7-tolyl vinyl sulfoxide (141), which yielded deltacyclane (142), has been reported (Scheme 38).^^ A high degree of stereoselectivity was observed, with mainly the ^;cc?-product being formed.
0AIMe2
O II "^Ph
OH
SOPh
Et20. THF, not -30°C
(136)
(137)
64%
H (138) f cis: trans ] 67:33 Scheme 37.
HO
L (139) sOPh
a,P-Unsaturated Sulfoxides and Sulfones
{UO)//J^//
181
(5-10%)Ni(COD)2. P(0Ph)3
+
cr SOp-Tol
exo: endo >19: 1
^ci 690/,
p-TolOS
(R^,Ss)'(U2)
(S)^{U^) Scheme 38. B. [3+2] Cycloadditions
a,P-Unsaturated sulfoxides and sulfones have been exploited as dipolarophiles in 1,3-dipolar cycloadditions, reacting with a variety of 1,3-dipoles such as nitrile oxides, nitrones, and diazomethane. In the following sections, 1,3-dipolar cycloadditions of vinyl sulfoxides and sulfones with each dipole will be considered. Reaction with Nitrile Oxides
Carretero et al. reported that enantiopure Y-oxygenated-a,P-unsaturated sulfone (143) underwent a completely regioselective 1,3-dipolar cycloaddition with ace-
Me-^N-O OMOM Toluene or CH2CI2 PhOaS'" ^'^^
'Pr
rt, 2-4 days
(143) ee>98%
PhOgS Me
I-
OMOM 'Pr
PhOzS
+ Me
OMOM
4^
&
anti -(144)
-O syn - (144)
yield = 65% conversion = 68% anti: syn = >15 : 1
Na - Hg anti -(144)-
98
%
O OH anti -(146)
M©
ill 6 anti -(145)
(147) r^Ho OH syn-anti : anti-anti] 82:14 Scheme 39.
''Pr
IAN FORRISTAL and CHRISTOPHER M. RAYNER
182
Ph, jD EtOaS'
H /P
Ph
.0
silica gel
Ph-^N^o^
0 ^ 1 ^ 85% EtOzS ^s02Et (149)
SOjEt (148)
Scheme 40.
tonitrile N-oxidc to yield isoxazolines (144), with the sulfonyl group at the C-4 position (Scheme 39).^^ This reaction also proceeded with a high anti stereoselectivity. The phenyl sulfonyl group of the isoxazoline (144) could readily be removed by reduction with sodium amalgam and the product (145) could be transformed into the enantiomerically pure P-hydroxyketone (146) or 1,3-aminoalcohol (147). 4,5-Diethylsulfonylfuran-2(5//)-one (148) was shown to be a highly reactive dipolarophile, undergoing 1,3-ciipolar cycloaddition with benzonitrile oxide at room temperature. The reaction was highly regioselective giving isoxazoline (149), which was readily aromatized to the furoisoxazole (150) by chromatography on silica gel (Scheme 40).^^ The furoisoxazole (150) through its annelation reactions with different benzoquinone monoketals, is a useful synthon for the preparation of heterocyclic anthraquinones.^^ Bravo and colleagues reported that the chiral methyl enol ether of (/?)-3-fluoro-l-(p-tolylsulfmyl)-2-propanone underwent an asymmetric 1,3-dipolar cycloaddition with nitrile oxides, with high diastereoselectivity, to yield chiral 4,5-dihydroisoxazoles.^^ Reaction with Nitrones
1,3-Dipolar cycloaddition of nitrones to alkenes has been widely utilized for the synthesis of many nitrogen-containing natural products. Indeed, in this process, up to three stereogenic centers are built up in a single step, often in a highly stereoselective manner. Louis and Hootele recently reported the first highly selective 1,3-dipolar cycloaddition between an a,p-unsaturated sulfoxide and a cyclic ni-
Et20
o (151)
X
>-Tol Ph (152)
rt 7-10 days 97%
Scheme 41.
a,fi-Unsaturated Sulfoxides and Sulfones
183
trone.^^ Cycloaddition of 2,3,4,5-tetrahydropyridine-l-oxide (151) to (Z)-(/?)-vinyl sulfoxide (152) proceeded in high yield to give isoxazolidines 153 and 154 with complete exo selectivity and excellent asymmetric induction (Scheme 41). The obtained isoxazolidines are versatile synthons and were converted into a variety of compounds, with retention of stereochemistry. Reductive cleavage of the N - O bond of the isoxazolidine nucleus unmasked the 1,3-amino alcohol moiety. Thus, isoxazolidines can be viewed as direct precursors of y-^mino alcohols. Hence, this highly diastereoselective 1,3-dipolar cycloaddition/N-O bond cleavage/desulfurization sequence allowed for the asymmetric synthesis of naturally occurring piperidine alkaloids; (+)-sedridine, (-)-hygroline, and (-)-(25)-A^-carbomethoxypelletierine.^' 5-Trifluoromethylisoxazolidines were synthesized by 1,3-dipolar cycloadditions of l,l,l-trifluoro-3-phenylsulfonylpropene and various acyclic nitrones, with a high degree of regio- and stereoselectivity.^^ The cycloaddition adducts were converted into the corresponding trifluoromethylated 5>'n-3-amino alcohols by desulfonylation with sodium amalgam, followed by reductive cleavage of the N ~ 0 bond by catalytic hydrogenation. The same group also reported asymmetric 1,3dipolar cycloaddition of optically active trifluoromethylated a,P-unsaturated aryl sulfones, which contained a chiral A^,A^-dialkylaminoethyl group on the ortho position, with acyclic nitrones. The corresponding isoxazolidines were obtained with excellent regioselectivity (>98%) and moderate stereoselectivity (36-56% The lower levels of stereocontrol which are often observed in the 1,3-dipolar cycloaddition of acyclic nitrones, as opposed to cyclic nitrones, could be accounted for by the possibility of interconversion of the nitrone geometry. One innovative solution to this problem is Aggarwal's recently reported 1,3-dipolar cycloadditions of the C2-symmetric cyclic alkenyl sulfoxide (l/?,3/?)-2-methylene-l,4-dithiolane 1,3-dioxide (155) with acyclic nitrones.^"^ The presence of a C2 symmetry element in 155 means that the exo/endo approaches of 155 to a dipole are symmetry related and therefore identical, thereby reducing the number of possible transition states in the reaction. 1,3-Dipolar cycloaddition of 155 with nitrones 156a-c resulted in single diastereomeric 4,4-disubstituted isoxazolidine products (157a-c) (Scheme 42). Likewise 1,3-dipolar cycloadditions with other acyclic nitrones yielded single diastereomeric products. The observed stereoselectivity can be rationalized by considering the possible transition states for the reaction (Fig. 4). Because of the C2 symmetry element in the dipolarophile, only two transition states are possible, leading to the diastereomeric 4-substituted isoxazolidine products 157 and 158. Transition state TS 1 leads to the observed product 157. The alternative transition state TS 2 suffers from steric and/or electronic repulsions between the phenyl ring of the nitrone and the sulfmyl oxygen; in TS 1 the phenyl group approaches over a sulfinyl lone-pair and the oxygen of the second sulfoxide over the smaller hydrogen atom (Fig. 4).
IAN FORRISTAL and CHRISTOPHER M. RAYNER
184
O"
r^.
Svy^^O +
R^+^O
CH2CI2
Ijj
^r/
Ph (156a-c)
(155)
(157a-c) a R = 'BU77% b R = Ph 64% c R = Me 86% J
Scheme 42.
preferred (157) observed
TS1
Ph
H
I
•
/ *
x=x / Ph
steric and electronic repulsion
'?^t
(158)
not observed
TS 2 Figure 4.
Reaction with Diazomethane Garcia Ruano and co-workers reported that cycloadditions of diazomethane to (5j)-5-ethoxy-3-/7-tolylsulfinylfuran-2(5//)-ones 159a, 161a and their corresponding 4-methyl derivatives 159b, 161b, proceeded with quantitative yields to give enantiomerically pure 3//,6//,3a,6a-dihydrofuro[3.4-c]pyrazol-4-ones 160a, 162a and 160b, 162b, respectively (Scheme 43).^^ The sulfmyl group at C-3 strongly increases both the reactivity and the 7i-facial selectivity of the reaction. The dipole
a,p'Unsaturated Sulfoxides and Sulfones
TolOS CH2N2 R OEt 159a :R = H 159b:R = Me
161a :R = H 161b:R = Me
96%
185
/P
->-
160a :R = H 160b : R = Me
162a :R = H 162b:R = Me Scheme 43.
approach is determined by the configuration at the sulfinyl group. Pyrolysis of pyrazolines 160a, 162a gave the methyl derivatives 159b, 161b in excellent yield. C. [4+2] Cycloadditions
During the past two decades, the asymmetric Diels-Alder reaction has become one of the most powerful tools in asymmetric synthesis as a result of its capacity to create up to four chiral centers in one step, often in a highly stereoselective manner. In the following sections, recent advances in this area using vinyl sulfoxide and vinyl sulfone dienophiles will be considered. It should be noted that, although beyond the scope of this review, many asymmetric Diels-Alder reactions of chiral sulfinyl-1,3-dienes have been reported.^^ Intermolecular Cycloadditions to Vinyl Sulfoxides
The ability of the sulfinyl group to control the 7i-facial selectivity in the asymmetric Diels-Alder reaction has provided impetus for the use of enantiomerically pure a,P-unsaturated sulfoxides as dienophiles. Generally dienophiles which contain a sulfinyl moiety as the sole activating group, show poor reactivity and 7i-facial selectivity. One exception is Ronan and Kagan's report that aryl vinyl sulfoxides can be efficiently activated to achieve highly stereoselective Diels-Alder reactions by transformation into the corresponding alkoxysulfoxonium salts.^^ Most of the reported asymmetric Diels-Alder reactions involving chiral vinyl sulfoxides, are those in which additional electron-withdrawing groups have been introduced at the
186
IAN FORRISTAL and CHRISTOPHER M. RAYNER
double bond. This has the property of both increasing their dienophilic reactivity and restricting the conformational mobility around the C-S bond, which leads to improved 7C-facial selectivity. In the past decade Carretero et al. have made significant advances in this area using a variety of activated chiral vinyl sulfoxides; containing either one, two, or three additional activating groups.^^ The Diels-Alder cycloadditions of several dienophiles based on the 2-sulfinylcyclopentenone skeleton have been investigated (Fig. 5). The Diels-Alder reaction of (5)-2-/7-tolylsulfmyl cyclopentenone (163a) with Dane's diene (166), catalyzed by EtAlC^, was used to synthesize the steroid skeleton of perhydro-cyclopenta[a]phenanthrenes in a single step.^^ The cycloaddition proceeded with complete regio-, endo-, and 7C-facial selectivity to yield a single cycloadduct. The dienophilic behavior of (5)-2-/7-tolylsulfinyl butenolide (163b) has been described.^^ Its reactivity was also quite low, requiring the use of high pressure or catalyst to reach high yields. The 7i-facial selectivity of its reactions with cyclopentadiene was very high in the presence of EtAlCl2, but the endo/exo selectivity was only moderate. An opposite situation was observed with acyclic dienes, which reacted with total endo selectivity but moderate 7i-facial selectivity. The corresponding 5-substituted butenolides 164 and 165 also underwent asymmetric Diels-Alder reactions with cyclopentadiene.^^ In these substrates, both the configuration at C-5 and at sulfur were important. In uncatalyzed reactions the 7i-facial selectivity was controlled predominately by the 5-alkoxy group, whereas in reactions catalysed by ZnBr2 it was controlled by the sulfmyl moiety. The Diels-Alder reactions of a variety of acyclic chiral vinyl sulfoxide dienophiles, bearing additional electron-withdrawing groups at the double bond, have been investigated (Fig. 6). The asymmetric Diels-Alder reaction of enantiopure (5)-benzyl 2-p-tolylsulfmylacrylate (167) with furan, under high pressure, afforded mainly a 2 : 1 mixture of endo adducts. These were then stereoselectively transformed into (+)-shikimic acid and (-i-)-5-ep/-shikimic acid.^^^ However, the poor control of the endo/exo selectivity and the low reactivity were the two main problems restricting the general synthetic usefulness of such a-sulfmylacrylate dienophiles.
(163a) X = CH2 (163b) X = 0 Figure 5.
a,p-Unsaturated Sulfoxides and Sulfones O
187
O
f
9
f
Tor ^
^ Tor Nj^
^
y
Tor
XOzMe (167)
0
t
jor >r^ ^TOI
EtOgC
(168)
0
ft
COsEt
(169)
EtOzC^^COsEt (170)
Figure 6.
In order to overcome these limitations, the asymmetric Diels-Alder reactions of (5)-benzyl methyl 2-p-tolylsulfmyl maleate (168) were investigated.^^^'^^^ It was reasoned that the additional electron-withdrawing group at the double bond would lead to an improvement in its reactivity and endo selectivity, compared to the corresponding sulfmylacrylates. Nevertheless, the second ester group did not confer the expected effect to the dienophile and thus the reactivity of the maleates was scarcely modified and their endo selectivity was only slightly improved with respect to those of the acrylates. However, the two main contributions of the studies of such maleate dienophiles were the strong increase in reactivity in the presence of TiCl4 and the fact that the endo selectivity was substantially improved in the reactions with acyclic dienes. Recently a comparative study of the asymmetric Diels-Alder reactions of both (5)-benzyl 2-/7-tolylsulfmylacrylate (167) and (5)-benzyl methyl 2-/?-tolylsulfmyl maleate (168) was carried out. ^^ Consequently, improved mechanistic models were developed in order to explain the behavior of such sulfmyl maleate and acrylate dienophiles in asymmetric Diels-Alder reactions. ^^ It was postulated that conformational equilibrium around the C-S bond must be completely shifted toward the rotamer with the sulfmyl oxygen in an s-cis arrangement (the most stable from an electrostatic point of view), making favored approach of the diene from the less hindered upper face supporting the lone electron pair (Fig. 7). The chelation of the sulfmyl and carbonyl oxygens with metals shifts the conformational equilibria
ZnX2
/ \ ^S-, •
BnO- ->r%> ^ s-trans
TiCU
2"^2
.X
O ^ s-cis
A^'
X^ (167)
Figure 7.
TiCU
J<
Js: •
1^ ^ " ° ^ % / s-trans
188
IAN FORRISTAL and CHRISTOPHER M. RAYNER
toward the s-trans rotamer, determining the inversion of the observed 7i-facial selectivity (Fig. 7). Taking into account that the results obtained from maleates and acrylates were identical under thermal and ZnX2-catalyzed conditions, but the opposite under TiCl4 catalysis, it was proposed that the second ester group must have been responsible for the observed difference in the last case. The formation of the chelate involving the two ester groups (instead of only one of them and the sulfmyl group) could explain the observed results. Thus, the results of this comparative study clearly indicated that (1) the 71-facial selectivity of the Diels-Alder reactions of vinyl sulfoxides must be explained by assuming a steric approach control of the diene on the less hindered face of the dienophile (that supporting the lone pair at sulfur), taking into account the relative reactivity of the rotamers around the C-S bond, in addition to their populations; (2) both the endo and the 7i-facial selectivities of the cycloadditions of vinySulfoxides with acyclic dienes are substantially higher than with cyclic ones. In order to further study the relationship between additional electron-withdrawing groups at the double bond and dienophile reactivity, cycloadditions of 169 and 170 (Fig. 6) were carried out. The reaction of the sulfmyl trialkoxycarbonyl ethene 169 with cyclopentadiene revealed that this dienophile exhibited a lower reactivity than maleate 168, despite the additional ester group existing in this triester.^^^ Also, the C2 symmetric Z?w-sulfoxide 170 showed unexpectedly low reactivity, requiring high pressures (13 kbar) in order to react with cyclopentadiene.^^ It was suggested that the lack of reactivity for such sulfmyl dienophiles reinforces the assumption that the sulfmyl group could act as a modulator of electron density, and thereby dienophile reactivity, of the double bond. Thus, in light of these recent studies the influence of the sulfmyl group on the dienophilic reactivity of the double bond, ranging from withdrawing to donating electron character, is dependent on the electronic effect of other groups attached to it.^^ The Diels-Alder reaction of 168 with piperylene 171 was investigated, in order to study the endo selectivity of the cycloaddition (Scheme 44).^°^ The resulting
,ZnX2
d
b
OBn Tol
BnO'
V OMe
OMe s-c/s (168)
$ - trans
Figure 8.
a,p-Unsaturated Sulfoxides and Sulfones
O
^
Tor'W"'^"
(171)
I
189
Me [I
le
^ (X
r "COjBn
COsMe Catalyst
(168)
Me C02Bn^V.^C02Bn
(173)
(172)
(174)
Scheme 44.
ra6/e 6. Diels-Alder Reactions of 168 with Piperylene Cycloaddition Conditions Entry 1 2
Catalyst Eu(fod)3 TiCU
Product MZ
TC'C)
t (h)
Yield (%)
ee (%)
0 -78
144 24
25 31
>96 >96
Product ^74 Yield (%) 59 41
ee (%) 38 696
adduct (172) was not stable at room temperature. It underwent a spontaneous, nonregioselective, sulfinyl elimination to yield the corresponding 1,3-cyclohexadiene (173) and 1,4-cyclohexadiene (174). An interesting finding concerned the optical purity of the cyclohexadienes. The ee's of the 1,3-cyclohexadienes (173) were higher than 96% regardless of the catalyst used, whereas that of the 1,4-cyclohexadiene (174) was dependent on the catalyst. Cycloaddition using TiCl4 gave optically pure 174 (entry 2, Table 6), but with Eu(fod)3174 was obtained with an ee of only 38% (entry 1, Table 6). These results were explained by assuming that the sulfinyl group controlled the 71-facial selectivity. Therefore, only one face of the dienophile was accessible to the diene, which resulted in only one exo (176) and one endo (175) adduct being formed (Scheme 45). The syn character of the elimination of the sulfinyl moiety would determine that adduct endo (175) would evolve into a mixture of 1,3-cyclohexadiene (173) and 1,4-cyclohexadiene (174a). Adduct ^jco (176) would evolve into 1,4-cyclohexadiene (174b), the enantiomer of 174a. This evolution justified the high optical purity of all 1,3-cyclohexadienes, as they can only be derived from adduct endo (175). Also, it relates the optical purity of 1,4-cyclohexadiene to the degree of endo selectivity of the cycloaddition. Thus, the high optical purity of the cyclohexadienes obtained with TiCl4 suggested that the reaction proceeded with complete endo selectivity, whereas the moderate ee of the 1,4-cyclohexadiene (174) (entry 1, Table 6) obtained in the presence of Eu(fod)3 indicated that the reaction proceeded with only moderate endo/exo selectivity. Carmen Carreno and co-workers have investigated Diels-Alder reactions of a wide variety of benzoquinone-based chiral vinyl sulfoxide dienophiles (Fig. 9).
IAN FORRISTAL and CHRISTOPHER M. RAYNER
190
.Me ^.--^C^SOTol
COoMe (175) endo
C02Bn
Scheme 45.
They reported that the Diels-Alder reaction of (5)-2-(/7-tolylsulfinyl)-l,4-benzoquinone (177a) with cyclic dienes gave endo cycloaddition products, resulting from reaction of the unsubstituted C^-C^ double bond. However, with acyclic dienes dienophile 177a underwent cycloaddition on the sulfinyl-substituted C2-C3 double bond exclusively. ^^^ These cycloadditions occurred with a high degree of 7i-facial selectivity. Thus, the chemoselectivity was mainly related to the cyclic or acyclic structure of the diene. However, the Diels-Alder reactions of the 3-substituted-(5)2-(p-tolylsulfmyl)-l,4-benzoquinones (177b,c), with both cyclic and acyclic dienes, took place exclusively on the unsubstituted C5-C5 double bond.^^^ Optically active 4a,5,8,8a-tetrahydronaphthoquinones were obtained with moderate to good diastereoselectivity. It should be noted that these products contain a vinyl sulfoxide Boa
(177a) R = H (177b) R = Et (177c) R = CI Figure 9.
a,p'Unsaturated
Sulfoxides and Sulfones
191
moiety. Thus, they can act as dienophiles and undergo further Diels-Alder cycloaddition reactions.^^'^^° The related A/'-(r^rr-butoxycarbonyl)-3-/7-tolyIsulfinyl-1 -benzoquinone-4-imine (178) underwent cycloaddition exclusively on the sulfinyl-substituted dienophile double bond with /ra/z^'-piperylene.^^^ With cyclopentadiene its Diels-Alder reactions took place on the double bonds C2-C3 or €5-0^, depending on the reaction conditions, with total endo selectivity and high 7i-facial diastereoselectivity. The same group recently reported the enantioselective Diels-Alder approach to the tetracyclic skeleton of angucyclinones from the chiral dienophile, (S)-2-(p-tolylsulfinyl)-l,4-naphthoquinone (179) and racemic vinylcyclohexene (180) (Scheme 46).^^^ The ee value of greater than 97% for 181 indicated excellent diastereoselectivity for the Diels-Alder reaction between 179 and f+)-180. Optically active f-)-180 was recovered, which indicated that a kinetic resolution had occurred during cycloaddition. Thus, the sulfmyl group on the quinone framework promoted a double induction in Diels-Alder cycloaddition leading to an efficient kinetic resolution of (±)-l-r^rr-butyldimethylsilyloxy-3-vinylcyclohex-2-ene (180). Maleimide dienophile 182 bearing a (2-e;cd?-hydroxy-10-bornyl)sulfmyl group as a chiral auxiliary showed high diastereoselectivity in the Diels-Alder reaction with cyclopentadiene, under chelation-controlled conditions, for the endo adduct 183 (Scheme 47).^^^ The Lewis acid (ZnCl2) played a role not only as a reaction promoter but also as a chelating agent of the sulfinyl oxygen with one imidocarbonyl group, which resulted in a rigid Diels-Alder transition state of the dienophile. The sulfmyl group, which had served as a chiral auxiliary in the Diels-Alder cycloaddition, was then employed as an efficient control element to effect the diastereoselective reduction of the imidocarbonyl group in adduct 183 to yield the y-hydroxy lactam. Desulfinylation followed by alkylation of the y-hydroxy moiety gave the y-ethoxy lactam 185. Treatment with a Lewis acid, following by either an organocuprate or an ally 1 silane, resulted in A^-acyliminium addition directed by the bicyclo[2.2.1]-heptene moiety to give the corresponding y-functionalized lactams 186 with a high degree of stereocontrol.
TBS(
(±)-(180) (2 equiv)
(181) O
(+)-(179) O
Scheme 46,
(-H180)
IAN FORRISTAL and CHRISTOPHER M. RAYNER
192
•
T (182) ^
>Bn
o^Clhr
/-NBn Q (183) Product Ratio
/-NBn O (184) Yield (%)
[ 183:184 (97: 3) 100
J
iH j^^^y\^^ •'-'^^'^^^'^ ^ / t T v o
i. NaBH4. EtOH (183)
I
ii. Sml2. HMPA, THF. *BuOH iii. PPTS, EtOH
) EtO
I
NBn
^•^
1
ii. organocuprate or allyl silane
\
(185)
r
I NBn (186) R2 = alkyl. allyl. aryl
Scheme 47.
Chiral (Z)-l-(alkylsulfinyl)-2-nitroalkenes underwent Lewis acid-promoted Diels-Alder reactions, with complete control of the diastereoselectivity as well as endo selectivity, to yield single cycloadducts.^ ^"^ Enantiopure {-ytrans-hQnzo[d ]dithiine-5,y-dioxide underwent Diels-Alder cycloadditions with a series of cyclic dienes, affording adducts with diastereoselectivities ranging from fair to high.^^^ Ketene equivalents have found widespread use as partners in Diels-Alder reactions for the construction of cyclic, fused, and bridged unsaturated ketones. However, ketene equivalents based on simple vinyl sulfoxides are poor dienophiles and show low levels of diastereocontrol. Recently, Aggarwal and colleagues reported that (l/?,3/?)-2-methylene-l,3-dithiolane 1,3-dioxide (187) was a highly reactive and highly selective chiral ketene equivalent. ^^^ Diels-Alder cycloaddition of this
^
\
(187) EtCN BF3.0Et2 ^O
-78°C, 20 min 74%
(188)
Scheme 48.
i. PBr3, CH2CI2 ^ii. CuCl2,Si02, H2O, CH2CI2 90%
(189)
a,fi-Unsaturated Sulfoxides and Sulfones
193
C2-syinmetric cyclic alkenyl sulfoxide (187), with cyclopentadiene, proceeded under mild conditions to give a single diastereomeric adduct (188) in excellent yield (Scheme 48). Other acyclic dienes also gave single diastereomeric adducts often without the necessity of using Lewis acids. The ^w-sulfoxide moiety can be readily deprotected using a two-step sequence of sulfoxide reduction followed by hydrolysis of the dithiolane to give the enantiomerically pure norbomenone (189). Intermolecular Cycloadditions to Vinyl Sulfones
Cyclic Y-oxygenated-a,P-unsaturated sulfones 190a,b were used as dienophiles in Diels-Alder cycloadditions with Danishefsky's diene 191 (Scheme 4g)}^'^^^^^ When the secondary hydroxy group was protected as its benzyl ether 190a, cycloaddition gave a mixture (1 : 1.2) of adducts 192a and 193a. It was reasoned that a more bulky protecting group might lead to higher diastereofacial selectivity in the cycloaddition. With the triisopropylsilyl ether 190b a more stereoselective process was observed, although the improvement was only moderate, and a mixture (1 : 2.5) of adducts 192b and 193b was obtained. Subsequent reduction of the sulfone moiety of adduct 193, to the corresponding sulfide, and 1,3-dipolar cycloaddition with a functionalized nitrile oxide gave the hydrobenzothiophene subunit of the sesquiterpene breynolide.
R(>--<s^S02 (190a-b)
i. mesitylene, A, 48 hr
+
ii. PPTS, CeHe. A, 24 hr
SO2
RO-(193)
a R= Bn (1 : 1.2) b R = TIPS (1 : 2.5)
I (191) OMe Scheme 49.
F3BO
^^
o=s^ (194)^
^^
e7
BF3.0Et2 (cat) CHCI3
0-
-
(195)
0
Scheme 50.
J
(196)
194
IAN FORRISTAL and CHRISTOPHER M. RAYNER
One of the most interesting examples of a,P-unsaturated sulfone cycloaddition is by long-range activation of sulfonyl dienophile (194), via its oxosulfoxonium salt (195), to yield a mixture of two of the four possible adducts (196) (Scheme 50).^^^ This oxosulfoxonium salt (195) is formed by a reversible Lewis acidcatalyzed reaction between the sulfone moiety and a suitably placed epoxide. This dramatically increased the electron-withdrawing ability of the dienophile. The corresponding vinyl sulfone without the epoxide functionality did not undergo cycloadditon, even at elevated temperatures or prolonged reaction times. Intramolecular Cycloadditions to Vinyl Sulfones The intramolecular Diels-Alder (IMDA) reaction continues to be the subject of widespread research effort in the contexts of synthetic methodology and strategy, and of total synthesis. Craig and co-workers reported that trienes possessing internally activated vinylic sulfones undergo the IMDA reaction with high or complete selectivity for the ci.y-fused products.^^^ This methodology allowed for the synthesis of bicyclo[4.4.0] and -[4.3.0] systems, possessing a bridgehead sulfonyl group, with excellent selectivity. Thermolysis of triene 197 gave in high yield a single bicyclo[4.3.0] cycloadduct 198 (Scheme 51). The stereochemical outcome of this process can be rationalized in terms of the preference for an ^jco-oriented phenylsulfonyl group. Heating a solution of triene 199 using the standard procedure gave a more complex product mixture. In addition to the expected c/^-fused bicyclo[4.4.0] system 200, a 1 : 3 mixture of two bicyclo[4.3.0] cycloadducts 201 and 202 was obtained (Scheme 52). It was proposed that under the thermal conditions required for the IMDA reaction to occur, triene 199 undergoes E- to Z-isomerization followed by a [1,5]-hydrogen shift and Z- to £-isomerization to the homologue 205 possessing a terminal methyl group on the diene moiety. Triene 205 undergoes the IMDA reaction to give 202; cyclization prior to the final Z- to £-isomerization gives rise to the by-product 201 (Scheme 52). In recent years the temporary tethering of dienes and dienophiles has emerged as an effective strategy for the control of regio- and stereoselectivity in Diels-Alder reactions. The products of overall mfennolecular processes may be accessed while exploiting the benefits of m^mmolecularity during C-C bond formation. An effective tethering strategy must combine high IMDA reactivity and stereoselectiv-
S02Ph
PhMe, 180^0. 4.5 h^ 88%
(197)
Scheme 51,
^02Ph(198)
a,P'Unsaturated Sulfoxides and Sulfones
195
PhMe. 180°C. 36h (200)
CD^
S02Ph [200:201 : 202 = 69:8:23]
SOzPh (201)
sOzPh (202)
SOzPh
(204)
[4+2] Z - ^ E-
"SOzPh (205) Scheme 52,
ity with efficient methods for introduction and removal post-cycloaddition of the linking functionality. One such example is the IMDA reaction of ether-tethered triene. Cyclization of 206, proceeded with complete selectivity, to give the trans bicyclo[4.4.0] cycloadduct 207 (Scheme 53).^^^ Other related sulfone-activated trienes showed moderate trans selectivity in similar IMDA reactions. The detethering of one such trans cycloadduct (208), containing a tertiary ether linkage, was carried out. An El process, mediated by a Lewis acid, effected selective C - 0 bond cleavage to give the most stable of the alternative regioisomeric carbocations, and finally olefin (209) (Scheme 54),
PhMe, 165°C, 5h
y
SOaPh ! -Me
^
(207)
92% (206) Scheme 53.
IAN FORRISTAL and CHRISTOPHER M. RAYNER
196 SOoPh H I
SOsPh FeCl3(0.1 eq) AC2O, 0°C, 30 min AcO' • 65%
(208)
(209) Scheme 54.
V. REARRANGEMENTS INVOLVING a,p-UNSATURATED SULFOXIDES AND SULFONES A. Pummerer Reactions
The Pummerer rearrangement of sulfoxides with acid anhydrides has been extensively utilized as a method for the synthesis of a-substituted sulfides. When a,P-unsaturated sulfoxides are used, the initial formed oxysulfonium ion may undergo two different pathways: the additive Pummerer reaction or the vinylogous Pummerer reaction. The following sections will consider examples from both pathways. Additive Pummerer Reactions
For the intramolecular additive Pummerer reaction, following formation of the oxysulfonium ion (211), nucleophilic attack occurs at the electrophilic P-carbon of the (9-activated substrate producing a saturated (3-functionalized thionium species (212). Trapping with a second nucleophilic agent affords a product (213) formally derived by sequential attack of two nucleophiles on an a,P-dication (Fig. 10).
^Nu-{^)
n (210)
AC20 S-^0 I R
/ f/A/^z^ • ^ N U - K ) ^
r> .S-^OCOCHa
(211)
^Nu
1 ^^--^^SR
K)
(212)
(213) Figure 10.
a,P-Unsaturated Sulfoxides and Sulfones
197
(215) 9^^^^^ <^OMe pTol
ZnCl2. THF
MeO'
"pTol (216)
(214)
OTBDMS 78% yield 78% ee
Scheme 55,
Treatment of chiral, nonracemic vinyl sulfoxides (214) with O-silylated ketene acetal (215) in the presence of a catalytic amount of zinc chloride resulted in an enantioselective additive Pummerer-type reaction, affording the corresponding enantiomerically enriched methyl-4-siloxy-4-sulfenylbuyrate (216) (Scheme 55).^^^ This is the overall addition of the enolate equivalent to the vinyl sulfoxide. The finding that thionium ions may serve as electrophiles in electrophilic substitution chemistry has greatly extended the synthetic range of the Pummerer reaction. Padwa and Kuethe used intramolecular versions of this process in the preparation of nitrogen-containing heterocycles. Vinyl amido sulfoxide 217 underwent an additive Pummerer reaction, on treatment with triflic anhydride, to yield product 220 (Scheme 56).^^^ The critical step in this transformation involves a
s'^ o NBu Tf20
MeO.
63% M7\ 1 . . (217) oMe
N*Bu MeO.
. H"^
MeO.
(219) OMe
(220) 6Me Scheme 56.
IAN FORRISTAL and CHRISTOPHER M. RAYNER
198
6'exo-ihg cyclization of intermediate 218 to give 219, which is ultimately converted to 220. Tethered alkenes can also be utilized in the cyclization step of these cascade reactions. Vinylogous Pummerer Reactions
For the intramolecular vinylogous Pummerer reaction pathway, an electrophilic thionium ion intermediate (222) is formed by y-proton loss from 221 followed by sulfoxide S-O bond scission. This unsaturated thionium ion (222) is then intercepted by a nucleophile at the y-position to yield a vinyl sulfide product (223) (Fig. 11). Padwa and Kuethe have also used vinylogous Pummerer reactions of amido sulfoxides in the preparation of nitrogen-containing heterocycles. Vinyl amido sulfoxide (224) underwent an additive Pummerer reaction, on treatment with trifluoroacetic anhydride, to yield product 226 (Scheme 57).^^-^ The a-thiocarbocation 225 generated from the Pummerer reaction of A^-methyl-A^-phenyl-2[2-(toluene-4-sulfmyl)phenyl]acetamide (224) underwent a Friedel-Crafts reaction at the y-carbon with the tethered aromatic ring. Reductive removal of the
r
Ar
cr° L
TFAA
UQ/
(224)
liie
(225)
SAr Ra/Ni
(227)
(226) Scheme 57.
-H*
a,P-Unsaturated Sulfoxides and Sulfones
199
phenylthio group from the resulting product (226), using Raney nickel, provided a new method for the synthesis of 3-phenyl-substituted oxindoles (227). Miscellaneous Pummerer Reactions (/?)-Fluoropyruvaldehyde A^,5-ketal 233 has been prepared from a-(fluoroalkyl)p-sulfmylenamine (/?)-(Z)-228, through a new self-immolative tandem sequential process, consisting of a Pummerer reaction, promoted by trifluoroacetic anhydride, followed by a 1,2-migration of the /7-tolylthio group (231-^233), triggered by addition of silica gel (Scheme 58).^^"^"^^^ Each transfer of stereogenic center, from sulfur to the a-carbon and then to the P-carbon, occurred with a high degree of stereocontrol (up to 88% ee). Cis geometry between the sulfinyl and the amino groups of the starting enamine (R)-22S was necessary for achieving a high level of stereocontrol, since neighboring group participation by the N-Cbz group prevented racemization of the sulfinyl center by trifluoroacetic anhydride (228—>231). NMR studies have shown that imine 231 was an intermediate product of the Pummerer rearrangement. The reaction of phenyl vinyl sulfoxide 234 with isobutene, in the presence of trifluoroacetic anhydride, yielded the Z?^-alkylated product 238 (Scheme 59).^^^ It was suggested that this reaction proceeded by a different mechanism than the usual additive Pummerer mechanism. The alkene reacts with the electrophilic sulfur atom of intermediate 235, giving, after loss of a trifluoroacetate ion and a proton, the sulfonium ion 236. Thio-Claisen rearrangement of the ion then gives the thonium ion 237 which reacts with a further molecule of isobutene to give the product 238. 9F3 NHCbz
^k^c
Tolp" + ^ ^
TFAA
CF3 CF. L
(R)-(228)
(230)
Cbz
p - T o l S ^ . NHCbz 0HCr^CF3 (R)-{233) 62% yield 82% ee
-TFA p-TolS
CF3 (232)
Scheme 58.
p-TolS
NCbz
(R)-(231)
IAN FORRISTAL and CHRISTOPHER M. RAYNER
200
^OCOCFa II Ph' (234)
TFAA > •
(235)
/ T ^
(236)
\ [3.3]
>SPh (238)
(237)
SPh Scheme 59.
B. [3,3]-Sigmatropic Rearrangements
Reaction of methyl alk-1-ene sulfinate 239 with an ally lie Grignard reagent gave substrate 240. This underwent [3,3]-sigmatropic rearrangement to yield the y-unsaturated thioaldehyde-5-oxide 241 under exceptionally mild conditions (Scheme 60).^^^ Other similar substrates, derived from the reaction of methyl alk-2-ene sulfmates with vinylic Grignard reagents, also underwent similar [3,3]sigmatropic rearrangements. The first example of an asymmetric thio-Claisen rearrangement directed by a sulfmyl group has been reported.^^^ The substrate employed for the rearrangement was the racemic ketene dithioacetal 244, containing a sulfmyl group at the double bond, which readily underwent [3,3]-sigmatropic rearrangement at room temperature to yield the y-unsaturated-a-sulfmyl dithioester 245 (Scheme 61). The asymmetric induction of this procedure was extremely effective, with very high observed diastereoselectivity.
O II MeO (239)
-78°C->+18°C
MgCI
C4H9O S
THF 'C4H9 (240)
(241)
73% E:Z
5:95 Scheme 60.
201
a,P'Unsaturated Sulfoxides and Sulfones i. MeLi
"X ArS
SMe (243)
SMe
i. LDA ii. R2
SMe
ff rx
CH2CI2 <
86-98% de 42-63% yield
R2
31-^^
20°C (245) ,1^2
SMe (244)
Scheme 61.
OH
CH3C(OEt)3
(246)
"S02Ph
C2H5CO2H (cat) xylene A Ph02S
DBU. THF R
^^ (249)
^^
OEt
i8h
r.t
75-94%
R
" ^ (248)
O ^^
OEt
65-89%
Scheme 62.
Ketene acetals 247, prepared in situ from racemic y-hydroxy vinyl sulfones 246, underwent Claisen rearrangements to give 3-phenylsulfonyl esters 248. These compounds were converted to the (2£,4F)-dienoic esters 249 by a base-assisted elimination of benzenesulfmic acid (Scheme 62).^^^
VI. MISCELLANEOUS REACTIONS OF a,p-UNSATURATED SULFOXIDES AND SULFONES A. Epoxidation and Cyclopropanation Reactions Epoxidation Reactions
Enantiopure sulfmyl and sulfonyl oxiranes are versatile synthetic intermediates. Mori and colleagues reported that oxiranyllithium compounds, generated from
202
IAN FORRISTAL and CHRISTOPHER M. RAYNER SOzPh
02Ph
LiOO^Bu
OR
OR
SOjPh +
^
THF (251) ^r syn
Pr (250a) R = H (250b) R = TIPS
(251) ^,
anti
r a R == H
: yield =65%, (syn : anti) (25 :1) bR == TIPS : yield =61%, (syn: anti) (1 :
1
25) J
V ._
Scheme 63.
epoxy sulfones by deprotonation, react with a-alkoxy alkyl triflates to give new substituted epoxides in high yields.^^^ This methodology was subsequently used for the reiterative synthesis of fran^-fused tetrahydropyrans.^^^ Thus, several groups have developed asymmetric protocols for the synthesis of this class of compounds. Nucleophilic epoxidation of a-(l-hydroxyalkyl)-a,P-unsaturated sulfone 250a, using lithium r^rr-butyl hydroperoxide, proceeded with high facial diastereoselectivity to give the syn epoxide 251a (Scheme 63).^-^'^ However, epoxidation of the corresponding triisopropylsilyl ether 250b led to a reversal of diastereoselectivity, giving the anti epoxide 251b. Thus, the stereochemical course of these epoxidations was controlled merely by the appropriate choice of protecting group. The observed syn selectivity for the unprotected alcohol can be accounted for by coordination of the lithium cation to the alcohol. This would allow delivery of the reagent from the same face as the hydroxy group. Jackson and colleagues also reported similar nucleophilic epoxidations of Y-oxygenated-a,P-unsaturated sulfones, also with high diastereoselectivity, to yield Y-oxygenated epoxy sulfones. Once again a reversal of diastereoselectivity, dependent on the protecting group, was observed. The free alcohol 252a gave the syn product 253a, whereas the MOM ether derivative 252b gave the anti isomer 253b (Scheme 64).^^^
OP
*BuOOH, 'SOzPh THF, KH
OP PH
SOgPh
SOsPh (253) anti
(253) syn
(252a) R=H (252b) R=MOM
a P= H
: yield =66%, (syn: anti) (25 :1)
b p == MOM : yield =67%, (syn: V
Scheme 64.
anti)
(1 : 25) -J
203
a,P'Unsaturated Sulfoxides and Sulfones
><^S02Ph
SOsPh LiOO^Bu
90% yield 100% de J
THF (254)
(255) anti Scheme 65.
The epoxidation of cyclic vinyl sulfone 254 proceeded with high diastereoselectivity, only the anti epoxide 255 was formed as a single isomer (Scheme 65).^^^ Aggarwal and co-workers have used Jackson's metal peroxide oxidants for the epoxidation of ketene thioacetals. This resulted in highly diastereoselective epoxide formation (>20 : 1 selectivity).^^^ Carmen Carreiio and colleagues showed that epoxidation of (5)-(2-p-tolylsulfonyl)-2-cyclohexan-l-ol and its OAc and OMOM derivatives, with lithium tert-butyl hydroperoxide, proceeds with high stereoselectivity to give the syn epoxy alcohols.^^^ Pradilla et al. have shown that simple p-tolyl vinyl sulfoxides undergo nucleophilic epoxidation with metal alkyl peroxides to give enantiopure sulfinyl oxiranes.^^^ This process takes place with fair to excellent diastereoselectivities. The same group recently reported the epoxidation of diastereomeric hydroxy vinyl sulfoxides, bearing an additional stereocenter adjacent to the reactive carbon-carbon double bond. Hydroxy vinyl sulfoxides 256 and 258 underwent epoxidation with lithium tert-buty\ peroxide with high anti selectivity. However, when potassium tert'butyl peroxide was used, only hydroxy vinyl sulfoxide 256 showed anti
OH
O ^p-Tol
Bu"
QH .
^ Bu-^
(256)
KOO^Bu (2 eq.) ^THF, 0°C
p-Tol Bu"
55%
syn: anti 0 : 100
(257) anti
O p-Tol
KOO^Bu (3 eq.) ^ THF, 0°C
ErV^VTol
(258)
Bu^
60%
(259) syn
Scheme 66.
\syn: antl\ 8 0 : 20
IAN FORRISTAL and CHRISTOPHER M. RAYNER
204 OTBS
TBSO n S^^ A Ph
^tc NaOO^Bu (4eq.)
• n-Pr
-20°C,THF
A Ph
(261)
78%
syn: anti 11 : 89
NaOO^u (4eq.) -20°C, THF
"'^^
80% Scheme 67.
selectivity to yield epoxide 257. Its diastereoisomer 258 showed a reversal in facial selectivity, with the syn isomer 259 being formed (Scheme 66).^^^ For the epoxidation of y-oxygenated vinyl sulfoxides 260 and 262, with sodium r^rr-butyl peroxide, an even greater reversal in stereoselectivity was observed. Epoxidation of 260 gave the anti isomer 261, whereas 262 gave the syn isomer 263 (Scheme 67).^^^ This remarkable reversal of facial selectivity may be understood in terms of a "mismatched" situation and underlines that a chiral sulfmyl functionality is an extremely powerful chiral controller. Cyclopropanation Reactions
The stereoselective construction of functionalized cyclopropanes with high optical purity is of great importance, since a wide range of natural products and currently used insecticides contain the cyclopropane ring in a chiral environment.
MeO' (264)
Ph'"^(0)n
(265) P h ^ ^ ( O ) n
n^
(266) P h ^ ^ ( O ) „
- sulfoxide 12 :1 (S)-• sulfoxide 1 :3 - sulfone 1.1 1
Scheme 68.
a,p-Unsaturated Sulfoxides and Sulfones
Ph2SCMe2,
EtO-.t EtO'
r^Y^p^^'
205
E t c . , II • EtO^
VTol
THF, rt 65%
(267)
II
(268) Me 100% de
Scheme 69,
Methylenation of unsaturated sulfoxides 264 using trimethylsulfoxonium iodide and sodium hydride led to the methylene bridged derivative 265 of estradiol (Scheme 68).^"^ Good stereoselectivity was observed, but it was dependent on the chirality of the sulfoxide moiety. Mikolajczyk and colleagues recently designed a new type of activated chiral vinyl sulfoxide, namely, a-phosphoryl vinyl /7-tolyl sulfoxide (267). The phosphoryl group not only activated the double bond but also made possible further reactions, for instance, the Horner-Wittig reaction. These chiral sulfoxides were found to be good Michael acceptors as well as Diels-Alder dienophiles, but the asymmetric induction in these reactions was not very high. However, they underwent asymmetric cyclopropanation reactions. When sulfoxide 267 was treated with diphenylsulfonium isopropylide, the corresponding cyclopropane 268 was formed as a single diastereoisomer (Scheme 69).^'*^ The structurally related optically active a-acyl vinyl p-tolyl sulfoxide 269 underwent asymmetric cyclopropanation. Michael addition of the carbanion of bromomalonate to 269 and the subsequent intramolecular alkylation yielded the corresponding optically active a-acyl-cyclopropane 271, with a high degree of diastereoselectivity (Scheme 70).^"^^ It was proposed that the stereochemical outcome of the reaction can be rationalized by transition state 270, in which there is chelation of the oxygen atom of the carbonyl and sulfmyl groups to the metal cation.
f?3
^Xl
(Et02C)2CHBr >Tol NaH. THF. 0°C
(269)
45%
Et02C^B^ C02Et (270)
Scheme 70.
'pTol
ViX 0 2 E t (271) ^OgEt 67% stereoselectivity
206
IAN FORRISTAL and CHRISTOPHER M. RAYNER
if
1? CI
C>"
THF, -78°C
(272)
>99% de 84% yield
(273) Scheme 71.
The intramolecular alkylation would occur from the downward side of the chiral sulfinyl group, opposite the bulky p-tolyl group. Optically active vinyl sulfoxide 272, bearing a leaving group at the y-position, was stereoselectively transformed into a chiral cyclopropane (273) by means of a Michael-induced ring closure (MIRC) reaction (Scheme 71).^'*^ This reaction required a chloride as the leaving group for high diastereoselectivity. The a-sulfmyl anion intermediate, as a result of the diastereoselective Michael addition of the nucleophile from the re face, underwent intramolecular alkylation with stereocontrol of the two contiguous asymmetric centers. This methodology was used to synthesize similar bicyclo[4.1.0]heptane derivatives, which were utilized for a diastereoselective total synthesis of (-)-solavetivone.^'^ B. Metal-Catalyzed Reactions Palladium-Mediated Reactions
Palladium(0)-mediated intramolecular addition of aryl and vinyl halides to polarized olefins is a well-known and useful synthetic protocol. Zin and Fuchs extended this methodology to include vinyl sulfones.^"^^ Oxygen alkylation of phenol 275 with cyclic y-bromo vinyl sulfone 274, using phase transfer conditions, gave aryl iodide 276. Treatment of this system with 5% rerra^w(triphenylphos-
(274) Br\/^v^S02Ph 10%("BuUNl, U^ J ^ +
50% KOH (aq), ^
5»/o Pd(PPh3)4, <^
-^^
-^^^^
(275)
OH
^
'
(276) Scheme 72.
^
a,p-Unsaturated
Sulfoxides and Sulfones
207 .S02Ph^^2{dba)3.Hexn^
SOoPh CICOzEt. Hex^s
Hex">
Py. DMAP. (278)
icOoEt
'^''^
'
dppe. (279)
0
R = OEt, 79% yield JL R = Me, 67% yield ^
^'"v^^^-p;^^
THF.A
0
^^^^ ^
>^(280)
^
Hex JjHe Jj—T
frans; c/s >98 : <2 SOzPh
(282a-b)
(281)
Scheme 73.
pine)palladium(O) in the presence of silver nitrate effected intramolecular conjugate-addition/reductive elimination to generate an annulated vinyl sulfone (277) (Scheme 72). Omission of the silver nitrate produced a mixture of vinyl and allyl sulfones. In a later communication a comparision of vinyl organolithium and organopalladium reagents for intramolecular conjugate additions to vinyl sulfones was made.*^ For very hindered systems it was found that only the organolithium protocol resulted in annulation. One of the most synthetically useful processes involving palladium-catalyzed reactions is the nucleophilic substitution of allylic alcohols and derivatives via their 7i-allylpalladium complexes. Carretero and colleagues showed that y-oxygenated vinyl sulfones underwent palladium-catalyzed allylic substitution with carbon nucleophiles, to give new y-substituted vinyl sulfones (Scheme 73).^"^^ They used Tsuji's method, namely, converting the y-hy^roxy vinyl sulfone 278 into the corresponding carbonate 279. Formation of ay-substituted vinyl sulfone (281) was observed on addition of malonate nucleophiles to the 7i-allylpalladium complex 280. rhe reaction was completely regioselective, with exclusive attack of the nucleophile at the y-position. However, for both P-keto ester and 1,3-diketone nucleophiles, tetrasubstituted dihydrofurans 282a,b were obtained. These results indicated that a tandem process, based on an initial y-allylic substitution followed
7% Pd(OAc)2, 14% PPhg.
OTf
^^x'"^::N;^^C02Me
BuaSn'
(283)
'J^
^SOzPh
(284) Scheme 74.
THF.65°C
C02Me
f'
^^/'^^^^^SOzPh Ph ^ (285)
208
IAN FORRISTAL and CHRISTOPHER M. RAYNER
by cyclization via an intramolecular conjugate addition of its enol tautomer, or enolate, to the vinyl sulfone had taken place. The cyclizations occurred in transstereoselective manner. The synthesis of electron-deficient diene 285 was achieved by the Stille coupling of P-trifluoromethanesulfonyl-a,P-unsaturated sulfone (284) with a P-stannyl-a,Punsaturated ester (283) (Scheme 74). ^^^ Similarly, the preparation of a diverse range of enantiomerically pure 1- and 2-sulfmyl dienes has been achieved via Stille coupling of halovinyl sulfoxides and vinyl stannanes.^"^^'^^^ Enantiomerically pure 1- and 2-sulfmyl dienes have been used extensively in asymmetric Diels-Alder reactions. ^^ Iron-Mediated Reactions Enders and colleagues reported that y-oxygenated vinyl sulfones underwent iron-mediated allylic substitutions with complete chirality transfer.*^^'^^^ y-Oxygenated vinyl sulfone 286 was converted into the 1-phenylsulfonyl-substituted planar chiral tetracarbonyl(Ti^-allyl)iron tetrafluoroborate complex (288). Nucleophilic addition of various functionalized copper-zinc reagents to 288 provided, after oxidative removal of the tetracarbonyliron group, highly functionalized y-substituted alkenylsulfones (289) of high enantiomeric purity (Scheme 75). The reaction was highly regio- and stereoselective retaining the double bond geometry and with virtually complete chirality transfer ( C - 0 to C-C) with respect to the starting material. A similar iron-mediated allylic substitution of y-oxygenated vinyl sulfones with allyltrimethylsilane was used to synthesize the chiral building blocks of (/?,/?)-6,12-dimethylpentadecan-2-one, the female sex pheromone of the banded cucumber beetle.^^^
SQzPh '• '^Q2(CO)9> CO, n-hexane^ li. recrystallisation, 65%
lAe^^^^^r-^^S02Ph (287) OBn^®^^^^^ > 99% de and ee HBF4, EtzO. 96%
Me^^^^^^SOzPh ^ (289) IH2CH2FG
i. FG(CH2)2Cu(CN)ZnBr ii.CAN/H20
Me^^-v^S02Ph (288) ©,^e(C0)4 ® F 4 [ > 99% de and ee
FG = CN, 81%. >96%ee FG = P0(0Et)2, 83%, >96% ee Scheme 75,
a,^-Unsaturated Sulfoxides and Sulfones
209
REFERENCES 1. For reviews, see: (a) Posner, G.; Rappoport, Z.; Stirling, C. J. M. In The Chemistry ofSulphones and Sulphoxides, Patai, S., Ed.; Wiley: New York, 1988, p. 823. (b) Solladi6, G. Synthesis 1988, 185. (c) Walker, A. J. Tetrahedron: Asymmetry 1992, 3, 961. (d) Sulfur Reagents in Organic Synthesis; Metzner, P.; Thuillier, A., Eds.; Academic Press: London, 1993. (e) Garcia Ruano, J. L. Phosphorus, Sulfur Silicon Relat. Elem. 1993, 74, 233. (0 Solladi^, G.; Carreno, M. C. In Organosulfur Chemistry: Synthetic Aspects; Page, P. C. B., Ed.; Academic Press: London, 1995, Vol. 1, p. 1. (g) Carmen Carreno, M. Chem. Rev. 1995, 95, 1717. (h) Aversa, M. C ; Barattucci, A.; Bonaccorsi, P; Gianetto, P Tetrahedron: Asymmetry 1997,5,1339. (i) Westwell, A. D.; Rayner, C. M. In Organosulfur Chemistry: Synthetic and Stereochemical Aspects; Page, P. C. B., Ed.; Academic Press: London, 1998, p. 157. (j) Mislow, K.; Siegel, J. J. Am. Chem. Soc. 1984, 106, 3319. (k) Solladi6, G. Synthesis 1981,185. 2. (a) Rayner, C. M. Contemp. Org. Synth. 1994, /, 191. (b) Rayner, C. M. Contemp. Org. Synth. 1995, 2, 409. (c) Rayner, C. M. Contemp. Org. Synth. 1996, 3, 499. (d) Baird, C. P; Rayner, C. M. J. Chem. Soc, Perkin Trans. 1 1998, 1978. 3. Ottenheijm, H. C. J.; Liskamp, R. M. J.; van Nispen, S. P J. M.; Boots, H. A.; Tijhuis, M. W. J. Org. Chem. 19SI, 46, 3213. 4. Hutton, C. A.; White, J. M. Tetrahedron Utt. 1997, 38, 1643. 5. Farina, V.; Hauck, S. I.; Firestone, R. A. Bioorg. Med. Chem. Lett. 1996, 6, 1613. 6. (a) Rosenthal, P J.; Olson, J. E.; Lee, G. K.; Palmer, J. T.; Klaus, J. L.; Rasnick, D. Antimicrob. Agents Chemother 1996,40,1600. (b) Palmer, J. T.; Rasnick, D.; Klaus, J. L.; Bromme, D. J. Med. Chem. 1995, 38,3\93. 7. Caputo, R.; Ferreri, C ; Guaragna, A.; Palumbo, G.; Pedatella, S. J. Chem. Soc, Perkin Trans. 1 1995,1971. 8. Li, C ; Mahadevan, A.; Arasappan, A.; Phillips, J. R.; Merriman, R. L.; Tanzer, L. R.; Fuchs, P. L. Bioorg. Med. Chem. Lett. 1994,4, 1585. 9. (a) Chiral Sulfur Reagents: Applications in Asymmetric and Stereoselective Synthesis; Mikolajczyk, M.; Drabowicz, J.; Kielbasinski, P, Eds.; CRC Press: Boca Raton, 1997. (b) Li, A.-H.; Dai, L.-X.; Aggarwal, V. K. Chem. Rev. 1997, 97, 2341. 10. (a) Fuchs, P L.; Braish, T. F. Chem. Rev. 1986, 86, 903. (b) Sulfones in Organic Synthesis, Simpkins, N. S., Ed.; Pergamon Press: Oxford, 1993, and references cited therein. 11. Houpis, I. N.; Molina, A.; Dorziotis, I.; Reamer, R. A. Tetrahedron Lett. 1997, 58, 713L 12. Bravo, P.; Capelli, S.; Meille, S. V.; Seresini, P.; Volonterio, A.;Zanda, M. Tetrahedron:Asymmetry 1996,7,2321. 13. Marino, J. P; Laborde, E.; Deering, C. F ; Paley, R. S.; Ventura, M. P J. Org. Chem. 1994, 59, 3193. 14. Padwa, A.; MuUer, C. L.; Rodriguez, A.; Watterson, S. H. Tetrahedron 1998, 54, 9651. 15. Smith, D. C ; Fuchs, P L. J. Org. Chem. 1995,60, 2692. 16. Yechezkel, T.; Ghera, E.; Ostercamp, D.; Hassner, A. J. Org. Chem. 1995, 60, 5135. 17. Jin, Z. D.; Vandort, P C ; Fuchs, P L. Phosphorus, Sulfur Silicon Relat. Elem. 1994, 95-6, 1. 18. Dominguez, E.; Carretero, J. C. Tetrahedron Lett. 1993, 34, 5803. 19. Carretero, J. C ; Dominguez, E. J. Org. Chem. 1993, 58,1596. 20. Dominguez, E.; Carretero, J. C. Tetrahedron 1994, 50,7557. 21. Isobe, M.; Jiang, Y. Tetrahedron Lett. 1995, 36, 567. 22. Jiang, Y; Isobe, M. Tetrahedron 1996,52, 2877. 23. Jiang, Y; Ichikawa, Y; Isobe, M. Tetrahedron 1997,53, 5103. 24. Mase, N.; Watanabe, Y; Ueno, Y; Torn, T. J. Org. Chem. 1997, 62, 7794. 25. Posner, G. H. Ace Chem. Res. 1987, 20, 72. 26. Mase, N.; Watanabe, Y; Torn, T. J. Org. Chem. 1998, 63, 3899. 27. Mase, N.; Watanabe Y; Ueno, Y; Tom, T. J. Chem. Soc, Perkin Trans. 11998, 1613.
210 28. 29. 30. 31.
IAN FORRISTAL and CHRISTOPHER M. RAYNER
Mase, N.; Wake S.; Watanabe, Y; Tom, T. Tetrahedron Lett. 1998, 39, 5553. Ogura, K.; Kayano, A.; Sumitani, N.; Akazome, M.; Fujita, M. J. Org. Chem. 1995, 60, 1106. Kayano, A.; Akazome, M.; Fujita, M.; Ogura, K. Tetrahedron 1997,53,12101. Marino, J. P.; Anna, L. J.; Fernandez de la Pradilla, R.; Martinez, M. V.; Montero, C ; Viso, A. Tetrahedron Lett. 1996, 37, 8031. 32. Fernandez de la Pradilla, R.; Martinez, M. V.; Montero, C ; Viso, A. Tetrahedron Lett. 1997, 38, lllZ. 33. Marino, J. P; Viso, A.; Lee, J. D.; Fernandez de la Pradilla, R.; Fernandez, P ; Rubio, M. B. 7. Org. Chem. 1997, 62,645. 34. Marino, J. P; de Dios, A.; Anna, L. J.; Fernandez de la Pradilla, R. J. Org. Chem. 1996, 61, 109. 35. Arjona, O.; de Dios, A.; Fernandez de la Pradilla, R.; Plumet, J.; Viso, A. J. Org. Chem. 1994,59, 3906. 36. Acena, J. L.; Arjona, O.; Leon, M.; Plumet, J. Tetrahedron Lett. 1996, 37, 8957. 37. Acena, J. L.; Arjona, O.; Iradier, F ; Plumet, J. Tetrahedron Lett. 1996, 37, 105. 38. Arjona, O.; Borrallo, C ; Iradier, F ; Medel, R.; Plumet, J. Tetrahedron Lett. 1998, 39, 1911. 39. Lee, A. W. M.; Chan, W. H.; Tao, Y; Lee, Y K. J. Chem. Soc, Perkin Trans. 1 1994, 477. 40. Yoshizaki, H.; Tanaka, T; Yoshii, E.; Koizumi, T; Takeda, K. Tetrahedron Lett. 1998, 39, 47. 41. Jones, D. N.; Maybury, M. W. J.; Swallow, S.; Tomkinson, N. C. O. Tetrahedron Lett. 1993, 34, 8553. 42. Collins, M. A.; Jones, D. N. Tetrahedron 1996, 52, 8795. 43. Dong, Z.; Pyne, S. G.; Skelton, B. W.; White, A. H. Synlett 1997, 103. 44. Zahouily, M.; Joumet, M.; Malacria, M. Synlett 1994, 366. 45. Ogura, K.; Kayano, A.; Fujino, T; Sumitani, N.; Fujita, M. Tetrahedron Lett. 1993, 34, 8313. 46. Adrio, J.; Carretero, J. C ; Arrayas, R. G. Synlett 1996, 640. 47. Adrio, J.; Carretero, J. C. Tetrahedron 1998, 54, 1601. 48. Carretero, J. C ; Dominguez, E. J. Org. Chem. 1992, 57, 3867. 49. Lawrence, R. M. Tetrahedron Lett. 1994,35, 3161. 50. Catena, J.; Vails, N.; Bosch, J.; Bonjoch, J. Tetrahedron Lett. 1994, 35, 4433. 51. Bonjoch, J.; Catena, J.; Vails, N. J. Org. Chem. 1996, 61, 7106. 52. Matsuyama, H.; Itoh, N.; Yoshida, M.; lyoda, M. Phosphorus, Sulfur Silicon Relat. Elem. 1997, 120-1, 475. 53. Kroll, F E. K.; Morphy, R.; Rees, D.; Gani, D. Tetrahedron Lett. 1997, 38, 8573. 54. Evans, P; Taylor, R. J. K. Synlett 1997, 1043. 55. Giovannini, R.; Petrini, M. Chem. Commun. 1997, 1829. 56. Giovannini, R.; Petrini, M. Synlett 1998, 91. 57. Amone, A.; Bravo, P.; Capelli, S.; Fronza, G.; Meille, S. V; Zanda, M.; Cavicchio, G.; Crucianelli, M. J. Org. Chem. 1996, 61, 3375. 58. Mandai, T; Ueda, M.; Kashiwagi, K.; Kawada, M.; Tsuji, J. Tetrahedron Lett. 1993, 34,111. 59. Short, K. M.; Ziegler, C B., Jr. Tetrahedron Lett. 1995, 36, 355. 60. Edwards, G. L.; Muldoon, C. A.; Sinclair, D. J. Tetrahedron 1996, 52, 7779. 61. Marot, C ; RoUin, P Tetrahedron Lett. 1994, 35, 8377. 62. (a) Baldwin, J. E. J. Chem. Soc, Chem. Commun. 1976, 734. (b) Baldwin, J. E.; Thomas, R. C ; Kruse, L. I.; Silberman, L. J. J. Org. Chem. 1977, 42, 3846. 63. Carretero, J. C ; Diaz, N.; Rojo, J. Tetrahedron Lett. 1994, 35, 6917. 64. Carretero, J. C ; Rojo, J.; Diaz, N.; Hamdouchi, C ; Poveda, A. Tetrahedron 1995, 57, 8507. 65. Carretero, J. C ; Diaz, N.; Molina, M. L.; Rojo, J. Tetrahedron Lett. 1996, 37, 3179. 66. Craig, D.; Ikin, N. J.; Matthews, N.; Smith, A. M. Tetrahedron Lett. 1995, 36, 7531. 67. Craig, D.; Jones, P S.; Rowlands, G. J. Synlett 1997, 1423. 68. Craig, D. C ; Edwards, G. L.; Muldoon, C. A. Synlett 1997, 1319. 69. Li, C ; Fuchs, P L. Synlett 1994, 629. 70. de Bias, J.; Carretero, J. C ; Dominguez, E. Tetrahedron Lett. 1994,35, 4603.
a,p-Unsaturated Sulfoxides and Sulfones
211
71. Carretero, J. C ; Arrayas, R. G. J. Org. Chem. 1995, 60, 6000. 72. Carretero, J. C ; Arrayas, R. G.; de Gracia, I. S. Tetrahedron Lett. 1996, 57, 3379. 73. Carretero, J. C ; Arrayas, R. G.; de Gracia, I. S.; Adrio, J. Phosphorus, Sulfur Silicon Relat. Elem. 1997,120-1, 347. 74. Carretero, J. C ; Arrayas, R. G. J. Org. Chem. 1998,63, 2993. 75. Carretero, J. C ; Arrayas, R. G.; de Gracia, I. S. Tetrahedron Lett. 1997, 38, 8537. 76. Akiyama, E.; Hirama, M. Synlett 1996,100. 77. Kusuda, S.; Kawamura, K.; Ueno, Y.; Toru, T. Tetrahedron Lett. 1993, 34, 6587. 78. Jin, Z. D.; Fuchs, R L. Tetrahedron Lett. 1996, 37, 5249. 79. Kim, S. H.; Jin, Z. D.; Fuchs, R L. Tetrahedron Lett. 1995, 36,4537. 80. Ibanez, R L.; Ndjera, C. Tetrahedron Lett. 1993, 34, 2003. 81. Caturla, F ; Ndjera, C. Tetrahedron 1997, 53, 11449. 82. Back, T. G.; Nakajima, K. Tetrahedron Lett. 1997, 38, 989. 83. Aitken, R. A.; Cadogan, J. I. G.; Gosney, I. J. Chem. Soc, Perkin Trans. 1 1994, 1983. 84. Bienayme, H.; Guicher, N. Tetrahedron Lett. 1997, 38, 5511. 85. Lautensori, M.; Edwards, L. G.; Tam, W.; Lough, A. J. / Am. Chem. Soc. 1995,117, 10276. 86. de Bias, J.; Carretero, J. C ; Dominguez, E. Tetrahedron: Asymmetry 1995, 6, 1035. 87. Alguacil, R.; Farina, F ; Martin, M. V. Tetrahedron 1996, 52, 3457. 88. Alguacil, R.; Farina, F ; Martin, M. V.; Paredes, M. C. Tetrahedron Lett. 1995, 36, 6773. 89. Bravo, R; Bruche, L.; Merli, A.; Fronza, G. Gazz. Chim. Ital. 1994,124, 275. 90. Louis, C ; Hootele, C. Tetrahedron: Asymmetry 1995, 6, 2149. 91. Louis, C ; Hootele, C. Tetrahedron: Asymmetry 1997, 8, 109. 92. Tsuge, H.; Okano, T.; Eguchi, S. J. Chem. Soc, Perkin Trans. 11995, 2761. 93. Tsuge, H.; Okano, T.; Eguchi, S.; Kimoto, H. J. Chem. Soc, Perkin Trans. 1 1997, 1581. 94. Aggarwal, V. K.; Grainger, R. S.; Adams, H.; Spargo, R L. J. Org. Chem. 1998, 63, 3481. 95. Garcia Ruano, J. L.; Fraile, A.; Martin, M. R. Tetrahedron: Asymmetry 1996, 7, 1943. 96. Ronan, B.; Kagan, H. B. Tetrahedron: Asymmetry 1991, 2, 75. 97. Garcia Ruano, J. L.; Carretero, J. C ; Carmen Carreno, M.; Cabrejas, L. M. M.; Urbano, A. Pure Appl. Chem. 1996, 68, 925. 98. Alonso, I.; Carretero, J. C ; Garcia Ruano, J. L.; Cabrejas, L. M. M.; Solera, I. L.; Raithby, P. R. Tetrahedron Lett. 1994, 35, 9461. 99. Carretero, J. C ; Garcia Ruano, J. L.; Cabrejas, L. M. M. Tetrahedron 1997, 53, 14115. 100. Carretero, J. C ; Garcia Ruano, J. L.; Lorente, A.; Yuste, F Tetrahedron: Asymmetry 1993,4,177. 101. Adrio, J.; Carretero, J. C ; Garcia Ruano, J. L.; Cabrejas, L. M. M. Tetrahedron: Asymmetry 1997, 8, 1623. 102. Alonso, I.; Carretero, J. C ; Garcia Ruano, J. L. /. Org. Chem. 1994, 59, 1499. 103. Carretero, J. C ; Garcia Ruano, J. L.; Cabrejas, L. M. M. Tetrahedron Lett. 1994, 35, 5895. 104. Carretero, J. C ; Garcia Ruano, J. L.; Cabrejas, L. M. M. Tetrahedron: Asymmetry 1997, 8, 2215. 105. Carretero, J. C ; Garcia Ruano, J. L.; Cabrejas, L. M. M. Tetrahedron 1995, 51, 8323. 106. Carretero, J. C ; Garcia Ruano, J. L.; Cabrejas, L. M. M. Tetrahedron: Asymmetry 1997, 8, 409. 107. Carmen Carreno, M.; Garcia Ruano, J. L.; Toledo, M. A.; Urbano, A.; Remor, C. Z.; Stefani, V.; Fischer, J. J. Org. Chem. 1996, 61, 503. 108. Carmen Carreno, M.; Garcia Ruano, J. L.; Toledo, M. A.; Urbano, A. Tetrahedron Lett. 1994, 35, 9759. 109. Carmen Carreno, M.;GarciaRuano,J.L.;Urbano,A.;Hoyos,M.A.y. Org. Chem. 1996,67,2980. 110. Carmen Carreno, M.; Garcia Ruano, J. L.; Urbano, A.; Lopez-Solera, M. I. J. Org. Chem. 1997, 62, 976. 111. Bartolome, J. M.; Carmen Carrefio, M.; Urbano, A. Tetrahedron Lett. 1996, 37, 3187. 112. Carmen Carreno, M.; Urbano, A.; Fischer, J. Angew. Chem. Int. Ed. Engl. 1997, 36, 1621. 113. Arai, Y.; Matsui, M.; Fujii, A.; Kontani, T; Ohno, T.; Koizumi, T; Shiro, M. J. Chem. Soc. Perkin Trans. 1 1994, 25.
212
IAN FORRISTAL and CHRISTOPHER M. RAYNER
114. Fuji, K.; Tanaka, K.; Abe, H.; Matsumoto, K.; Harayama, T.; Ikeda, A.; Taga, T.; Miwa, Y; Node, M. / Org. Chem. 1994, 59, 2211. 115. Cecchet, E.; Di Furia, F; Licini, G.; Modena, G. Tetrahedron: Asymmetry 1996, 7, 369. 116. Aggarwal, V. K.; Drabowicz, J.; Grainger, R. S.; Gultekin, Z.; Lightowler, M.; Spargo, P. L. J. Org. Chem. 1995, 60, 4962. 117. Martin, S. F; Daniel, D. Tetrahedwn Lett. 1993, 34, 4281. 118. Martin, S. F ; Anderson, B. G.; Daniel, D.; Gaucher, A. Tetrahedron 1997, 53, 8997. 119. Zanon, J.; Lucchini, V.; Pasquato, L.; De Lucchi, O. Chem. Commun. 1996, 709. 120. Clasby, M. C ; Craig, D.; Slawin, A. M. Z.; White, A. J. R; Williams, D. J. Tetrahedron 1995, 57, 1509. 121. Ainsworth, R J.; Craig, D.; Reader, J. C ; Slawin, A. M. Z.; White, A. J. R; Williams, D. J. Tetrahedron 1996,52,695. 122. Shibata, N.; Fujimori, C ; Fujita, S.; Kita, Y Chem. Pharm. Bull. 1996,44, 892. 123. Padwa, A.; Kuethe, J. T. / Org. Chem. 1998, 63, 4256. 124. Bravo, R; Crucianelli, M.; Fronza, G.; Zanda, M. Synlett 1996, 249. 125. Bjorsvik, H. R.; Bravo, P.; Crucianelli, M.; Volonterio, A.; Zanda, M. Tetrahedron: Asymmetry 1997, 8, 2817. 126. Bravo,R;Crucianelli,M.; Volonterio, A.; Zanda, M.P/io5p/iorMj, Sulfur Silicon Relat. Elem. \991, 120-1, 353. 127. Volonterio, A.; Zanda, M.; Bravo, P.; Fronza, G.; Cavicchio, G.; Crucianelli, M. J. Org. Chem. 1997,62,8031. 128. Harvey, J. N.; Vieche, H. G. J. Chem. Soc, Chem. Commun. 1995, 2345. 129. Baudin, J. B.; Commenil, M. G.; Julia, S. A.; Wang, Y Bull. Soc. Chim. FK 1996,133, 515. 130. Alayrac, C ; Fromont, C ; Metzner, R; Anh, N. T. Angew. Chem. Int. Ed. Engl. 1997, 36, 371. 131. Giovannini, R.; Marcantoni, E.; Petrini, M. Tetrahedron Lett. 1998, 39, 5827. 132. Mori, Y; Yaegashi, K.; Iwase, K.; Yamamori, Y; Furukawa, H. Tetrahedron Lett. 1996, 37, 2605. 133. Mori, Y; Yaegashi, K.; Iwase, K.; Furukawa, H. J. Am. Chem. Soc. 1996,118, 8158. 134. Jackson, R. F W; Standen, S. R; Clegg, W; McCamley, A. J. Chem. Soc. Perkin Trans. 1 1995, 141. 135. Jackson, R. F W; Standen, S. R; Clegg, W J. Chem. Soc, Perkin Trans. 1 1995, 149. 136. Aggarwal, V. K.; Worrall, J. M.; Alexander, R. Phosphorus, Sulfur Silicon Relat. Elem. \WJ, 120-1,35\. 137. Bueno, A. B.; Carmen Carreno, M.; Garcia Ruano, J. L. Tetrahedron Lett. 1993, 34, 5007. 138. (a) Fernandez de la Pradilla, R.; Castro, S.; Manzano, R; Priego, J.; Viso, A. J. Org. Chem. 1996, 61, 3586. (b) Fernandez de la Pradilla, R.; Castro, S.; Manzano, P.; Martin-Ortega, M.; Priego, J.; Viso, A.; Rodriguez, A.; Fonseca, I. J. Org. Chem. 1998, 63,4954. 139. Fernandez de la Pradilla, R.; Manzano, P.; Priego, J.; Viso, A.; RipoU, M. M.; Rodriguez, A. Tetrahedron Lett. 1996,37, 6793. 140. Kunzer, H.; Thiel, M.; Peschke, B. Tetrahedron Utt. 1996,37, 1771. 141. Midura, W. H.; Krysiak, J. A.; Wieczorek, M. W; Majzner, W R.; Mikolajczyk, M. Chem. Commun. 1998, 1109. 142. Hiroi, K.; Arinaga, Y Chem. Pharm. Bull. 1994,42, 985. 143. Takemoto, Y; Ohra, T.; Sugiyama, K.; Imanishi, T.; Iwata, C. Chem. Pharm. Bull. 1995,43, 571. 144. Takemoto, Y; Kuraoka, S.; Ohra, T; Yonetoku, Y; Iwata, C. Tetrahedron 1997,53,603. 145. Jin. Z. D.; Fuchs, R L. Tetrahedwn Lett. 1993, 34, 5205. 146. Lee, S. W; Fuchs, R L. Tetrahedron Lett. 1993,34, 5209. 147. Alonso, I.; Carretero, J. C ; Garrido, J. L.; Magro, V; Pedregal, C. J. Org. Chem. 1997, 62, 5682. 148. Houpis, I. N.; DiMichele, L.; Molina, A. Synlett 1993, 365. 149. Paley, R. S.; de Dios, A.; Estroff, L. A.; Lafontaine, J. A.; Montero, C ; McCulley, D. J.; Rubio, M. B.; Ventura, M. R; Weers, H. L.; Fernandez de la Pradilla, R.; Castro, S.; Dorado, R.; Morente, M. / Org. Chem. 1997, 62, 6326.
a,fi-Unsaturated Sulfoxides and Sulfones
213
150. Paley, R. S.; Weers, H. L.; Fernandez, P.; Fernandez de la Pradilla, R.; Castro, S. Tetrahedron Lett. 1995,36, 3605. 151. Enders, D.; von Berg, S.; Jandeleit, B. Synlett 1996, 18. 152. Enders, D.; Jandeleit, B.; Raabe, G. Angew. Chem. Int. Ed. Engl. 1994, 33, 1949. 153. Enders, D.; Jandeleit, B.; Prokopenko, O. F. Tetrahedron 1995,57, 6273.
This Page Intentionally Left Blank
ASYMMETRIC PUMMERER REARRANGEMENT AND RELATED REACTIONS
Masato Matsugi, Norio Shibata, and Yasuyuki Kita
I. Introduction 216 II. Pummerer Rearrangement 216 A. Mechanistic Interpretations 216 B. Stereoselective Pummerer Rearrangement: Some Examples 217 III. Planning for Asymmetric Punmierer-Type Reactions 219 A. A Consideration from the Mechanism 219 B. Silicon-Induced Pummerer-iype and Polonovski-TVpe Rearrangements . 220 C. Silicon-Induced Additive Pummerer-TVpe Rearrangement 224 IV. Asymmetric Pummerer-TVpe Rearrangement 224 A. Highly Enantioselective Pummerer-TVpe Rearrangement Induced by 0-Silylated Ketene Acetals 224 B. Reaction Mechanism 227 C. Enantiospecific Pummerer-TVpe Rearrangement Induced by 0-Silylated Ketene Acetals 230 D. Asymmetric Additive Pummerer-TVpe Rearrangement 231 E. Effect of Substituent Groups on Silyl Function 233
Advances in Sulfur Chemistry Volume 2, pages 215-248. Copyright © 2000 by JAI Press Inc. All rights of reproduction in any form reserved. ISBN: 0-7623-0618-1 215
216
MASATO MATSUGI, NORIO SHIBATA, and YASUYUKI KITA
F. Carbon-Carbon Bond Formation via Pummerer-TVpe Reaction 234 G. Asymmetric Pummerer Rearrangement Induced by Ethoxy Vmyl Esters . . . 236 H. Control of Enantioselectivity 237 V. Asynmietric Intramolecular Punrnierer-TVpe Cyclization 239 VI. Asynmietric Punmierer Reaction Using Selenoxides 244 VII. Outlook for Asymmetric Punmierer Rearrangement 246 References 246
I. INTRODUCTION In 1909, Pummerer reported that treatment of a-phenylsulfinyl acetic acid with dilute sulfuric acid results in a rearrangement in which the oxygen group attached to sulfur is cleaved and benzenethiol, glyoxylic acid, and bis(thiophenoxy) acetic acid are obtained (Fig. 1).^ The generality of this reaction was confirmed by Horner et al., who proposed to call it "Pummerer rearrangement."^ The Pummerer rearrangements have received considerable attention both mechanistically and as a synthetically useful process for the a-functionalized sulfides.'^ Since then many useful applications have been reported for this reaction,"^ although there are few successful reports on an asymmetric Punmierer rearrangement of optically active sulfoxides except for a limited number of cases.^ The stereoselective Pummerer rearrangement of optically active sulfoxides is a self-immolative asymmetric transformation and is of considerable interest, because of its potential to provide a means of synthesizing chiral, nonracemic a-substituted sulfides. In this article, we wish to describe the details of some recent advances in the asymmetric Punrunerer rearrangement.
9 i+
d-H2S04 pruH
^
Ph-SH +
OHC—CO2H
+
PhS^ V-CO2H
Figure 1.
II. PUMMERER REARRANGEMENT A. Mechanistic interpretations The generally accepted mechanism^ of Pummerer rearrangement is the one in which there is an initial attack on the sulfoxide oxygen atom by an electrophilic species, e.g., protonation or acylation. Acylation is followed by proton abstraction by a base from the a-carbon atom of the sulfoxide to form an ylide, which rapidly eliminates an acetate anion to form the a-sulfonium salt. Addition of acetate anion to the sulfonium intermediate completes the formation of the a-functionalized sulfide. Ylide formation from sulfoxonium salts is well recognized, and this aspect
Asymmetric Pummerer Rearrangement
stepi
o"
AcgO
OAc sulfonium salt
step ii
217
OAc ^=S
It
-R^^-p.*-
OAc
OAc
"OAc
yiide
OAc sulfurane
F/gi/re 2.
of the mechanism has received considerable experimental support.^ Furthermore, Oae's many excellent studies of the Pummerer rearrangement using ^^O-tracer experiments showed intramolecular rearrangements to be involved,^ and now a general mechanism for the Pummerer rearrangement consisting of four sequential elemental reactions can be considered^: (i) acylation of the oxygen of sulfoxide, (ii) abstraction of a-hydrogen, (iii) cleavage of S-O bond, and (iv) rearrangement of the acetoxy function (Fig. 2). It is known that the rate-determining step of this rearrangement varies from step i to step iv of the mechanism, depending on the reaction conditions. B. Stereoselective Pummerer Rearrangement: Some Examples
A few examples of constructing the asymmetric a-carbon of the sulfur atom effectively by Pummerer rearrangement are as follows (Fig. 3): Treating cyclic sulfoxides with acetic anhydride results in highly selective Pummerer rearrangements.^^ Similarly, Pummerer rearrangement of phenylsulfenylcyclopropane derivative gave the products with moderate selectivities. Acetoxy anion approached the a-carbon from the backside of the abstracted hydrogen site.^^ Figure 3 suggests
Ac^O
MKJ^^
AC2O
Mi
S-
M^^\_>OAc
OAc 85 - 90% de A(^0 I.
o
NaOAc 72%
Ph^""
Ac(5 76
24
Figure 3.
MASATO MATSUGI, N O R I O SHIBATA, and YASUYUKI KITA
218 3iMe3
CX
I
3Ph
benzene
o
o^:.
Ph SiMea
"OSiMeg
'SPh
95%
100
benzene
^
OSiMea ^
r > > / ^3Ph ^^ "OSiMea
95% 84
16
Figure 4,
that the reactions were mainly controlled kinetically. In addition, the sila-Pummerer rearrangement induced by the transfer of a silyl function from the a-asymmetric carbon of sulfoxide was reported (Fig. A)}^ Additionally, in the cases using thiazoline oxide or thiopyrane-1-oxide as the substrate, highly diastereoselective Pummerer rearrangements were achieved (Fig. 5).^^'^"* As Figure 5 shows, highly stereoselective Punmierer rearrangements could be achieved in the cyclic systems. On the other hand, the stereoselectivity of the Pummerer rearrangement using acyclic sulfoxides was quite low (Fig. 6).*^ Because of the racemization in the reaction process, the stereoselectivity was ineffective.
C02Me 0~S+/N-COPh PAR2
Bu^MegSiOTf
Bu^Me^Siqt
E^aN CH2CI2
CO^Me
C02Me
S.^N-COPh
S^.N-COPh
R1^-R2 single isomer
Accr-^—^+ CN
AC2O TsOH CH2CI2
A( EtOjC
Accr^—-^
IN
93%
""^ 100
Figure 5.
^X^c OAc
Asymmetric Pummerer Rearrangement O" I
219
AC2O
p-Tol
OAc
R = CN: 29% ee = COgEt: 29% ee = COPh: 0.5% ee = P0(0Me)2: 24% ee
Figure 6.
III. PLANNING FOR ASYMMETRIC PUMMERER-TYPE REACTIONS A. A Consideration from the Mechanism
The difficulty of achieving the high enantioselectivity with the rearrangement in acyclic systems is probably related to the formation of an achiral sulfurane intermediate and a planar sulfonium ion intermediate during the course of the reaction (Fig. 7).^^ It was reported that an addition of 1,3-dicyclohexylcarbodiimide (DCC) as an effective scavenger of acetic acid increased enantioselectivity, although the chemical yields significantly decreased (Fig. 8).^^ Nevertheless, the Pummerer rearrangement of chiral benzyl tolylsulfoxide in the presence of DCC gave only the
OAc and OAc sulfonium intermediate
sulfurane
Figure 7. OAc
Pi
i N>Tol OAc sulfurane
OAc R = CN, PO(OMe)2, <30% ee
p-Tol ACgO, DCC OAc Figure 8.
R = COgEt, 70% ee, 10% cy R = Ph, racemic
MASATO MATSUGI, NORIO SHIBATA, and YASUYUKI KITA
220
racemic adduct.^^ It is anticipated that increasing the enantioselectivity can be achieved by inhibition of achiral structure formation. That is to say, if the reaction proceeds without the formation of sulfurane and sulfonium intermediates, a highly enantioselective intramolecular rearrangement will be carried out. The key to achieving the high enantioselectivity in an acyclic system with Pummerer rearrangement is to inhibit the formation of the achiral structure. B. Silicon-Induced Pummerer-Type and Polonovski-Type Rearrangements
The reactions using ketene acetal derivatives as the reagents for acylation and silylation were generally carried out in an inert solvent and usually brought to completion at a low temperature over a short period to give the desired products in high yields.^^ These reagents provide the mildest and most direct and effective silylation of alcohols, carboxylic acids, mercaptans, and amides in the absence of base, acid, or catalyst and allow for easy isolation of the products in almost quantitative yields (Fig. 9).^° In these reactions, concentrated reaction mixtures give pure silylated products without aqueous workup and are only accompanied by a volatile ester as a by-product. These silylating reagents are to be expected for a reagent of the Pummerer-type rearrangement under mild conditions. To date, intramolecular silicon-induced Punmierer-type thermal rearrangements of atrimethylsilylsulfoxides, leading to a-siloxy sulfides (so-called sila-Punmierer or silyl Pummerer rearrangements), have been reported.^* However, there is no successful silicon-induced Pummerer rearrangement of normal sulfoxides not possessing an a-silyl group. As an example, when some silylating agents were reacted with sulfoxides, the eliminated products were obtained predominantly over the rearrangement products, a-siloxysulfides.^^ On the other hand, the use of an effective silylating reagent, O-silylated ketene acetal (SKA), for the Pummerer-type rearrangement of sulfoxides provided a new methodology which leads to a-siloxysulfides from normal sulfoxides, not bearing an a-silyl group.^^ The Pummerer-type
PE
NuH
NuE
R3R2C=^
OR^
R 3 R 2 ^ ^ ^
R^ = Me or Et R2 = H or Me, Fp = H or Me, E = CO^R or SiM^ or SiM^tBu NuH = ROH, RCCbH, RSH, RCONHR' Figure 9.
R^R^CHCOgR^
Asymmetric Pummerer Rearrangement
221
rearrangement of sulfoxides with SKA proceeds smoothly in either the presence or absence of a catalytic amount of zinc iodide in acetonitrile to give the corresponding a-siloxysulfides in good yields (Fig. 10).^'* The role of zinc iodide might be considered for the generation of the Reformatsky-type reagent such as IZnCH2C02Me and ISiMe2Bu^ although this has not as yet been ascertained. The reaction of sulfoxides with SKA presumably occurs via the sulfonium intermediate shown in Figure 10. Initial silicon transfer from the reagent to the sulfoxide and subsequent abstraction of the a-hydrogen by the generated ester enolate anion presumably form a sulfonium intermediate, which rearranges by the usual Pummerer pathway, giving a-siloxysulfides. In contrast to the rearrangement, when sulfoxides are treated with SKA, bearing the trimethylsilyl group instead of the f^r/-butyldimethylsilyl group, the carbon-carbon bond-forming products are obtained in moderate yields (Fig. 10).^^ Similarly, a cyclic hemithioacetal sulfoxide readily reacted with trimethylsilyl enol ether to cause a Pummerer reaction to give a-carbon-carbon bond-forming product in excellent yields (Fig. 11).^^ These results contrast with those from the reaction of sulfoxides with SKA where Pummerer-type products predominate, which can be reasonably explained by
O I R^s:^p^ +
pSIMe2Bu*
=KOMe
f^Y)^-Ph
(SKA) + ^S>
Th
cat. Znl2 MeCN
—
ri
0SiMe2Bu*
"OSiMegBu*
Rx^Stp
+
pSiMea MeHC=< OMe
R^^S>
- ..X
Ph cat. Znig MeCN
Ph
Me^ ^C02Me Figure 10.
( O-SiMea OMe
OSiMea
222
MASATO MATSUGI, NORIO SHIBATA, and YASUYUKI KITA pSiMe^
=(.
°-i^o
LX^OMe
O
Ph
S O
PhAAA/OMe
TMSOTf 2,6-lutidine CH2CI2 -78 to 25 °C 95%
"OMe
Figure 11.
considering a common initial intermediate; the carbon-carbon bond-forming reaction is greatly facilitated by a strong silicon-oxygen affinity as compared with a carbon-oxygen affinity in the ordinary Punmierer-type reaction. In the case of having a bulky tert-h\xiy\ group on the silicon, the strong steric hindrance causes the siloxy anion to attack the carbon atom of the sulfonium intermediate rather than the silicon atom of the starting reagent (SKA), as shown in Fig. \0}^ On the other hand, in the Polonovski reaction of A^-oxides using SKA, carboncarbon bond-forming product was selectively obtained even if the substrate was allowed to react with r^rf-butyldimethylsilyl ketene acetal in acetonitrile.^^ It has become apparent that the choice of silyl moiety in SKA influences the course of the reaction. Specifically, in the case of SKA bearing a r^rr-butyldimethylsilyl group, a (methoxycarbonyl) methyl adduct was preferentially converted to a cyanomethyl adduct, whereas in the case of SKA bearing an 0-/^r/-butyldiphenyl-
Table 1.
Jr6^
02Me
jOSiMegBu^
Bn
IDMe
^
CXTS-OBnMOM
t)Me Znig CH3CN
QHDBn
\Pathb .aPh,BuiDSiPh
|;^G=(^?<|
Oo:;Bn
OMe
Entry
\J-OBn
SIR^R^R^
Path a (%)
Path b (%)
SiMesBu^ SiPh2Bu^ SiMej
24 0
24 54 No reaction
Asymmetric Pummerer Rearrangement
223
silyl group, only a cyanomethyl compound was obtained (Table 1). The reaction did not proceed with SKA bearing a trimethylsilyl group. The following mechanism is proposed to account for this interesting behavior: The oxygen atom of the A^-oxide is first silylated by SKA, and then the hydrogen atom on the a-carbon to the nitrogen is removed by the in situ-generated ester anion. The nucleophilic addition of SKA to the iminium intermediate forms the (methoxycarbonyl)methyl adduct (path a). It is thought that the cyanomethyl compound is formed by nucleophilic addition of acetonitrile activated by zinc iodide through path b. The formation of the cyanomethyl compound was inhibited by another solvent such as propionitrile. A convenient explanation of the control of this reaction course by the difference in the silyl part is depicted in Figure 12. Namely, it is suggested that the attack of the siloxy anion on the unreacted SKA is depressed by the steric repulsion when using (9-f^r/-butyldiphenylsilyl ketene acetal whose silyl part is fairly bulky. Consequently, the enolate anion derived from SKA as a nucleophile is not formed (Fig. 12). As can be seen from Table 1, O-r^rf-butyldiphenylsilyl ketene acetal only acts during the activation of the ^-oxide to the iminium ion intermediate, and the nucleophilic attack by the enolate anion does not occur. Therefore, it would be
BuVhaSiO' ^K
^
^^
/I
\ I+—CH2—C«N ZrK| >SiPh2Bu* V ^ i OMe -OBn H2C=TC=N^2r\^ CN
Bn Figure 12.
pSiPf^Bu* OMe nudeophile '-Me
^*CN.rt
)
.O-SiMea
R OMe nucleophile R = H, Me. OMe, Ph Figure 13,
•'l!^
OBn
224
MASATO MATSUGI, NORIO SHIBATA, and YASUYUKI KITA
OSiMegBu^
?^
OMe
RgOCCR-HC /
^
"osiRa
^ P-SiR3 MeHC=(^ OMe
pSiMeg MeHC=( OMe
route a "^
route b R3=Me3
JL
S
OSiMe^Bu^
I s
M e 0 2 C ( M e ) H C r ^ ""Ph CH(Me)CC^Me
Figure 14.
feasible to introduce various nucleophilic species on the a-carbon to the nitrogen by employing this remarkable feature. Consequently, a regioselective carbon-carbon bond formation on the a-carbon atom of the heterocyclic nitrogen in the A^-oxide could be achieved in good yields by addition of only the expected nucleophile (Fig. 13). C. Silicon-Induced Additive Pummerer-Type Rearrangement
SKAs are also effective in an additive Pummerer-type reaction.^^ Vinyl sulfoxides, on treatment with bulky SKA, undergo an additive Pummerer reaction to give Y-siloxy-y-phenylthioesters (route a. Fig. 14), while with the less bulky SKA, the reaction undergoes a double carbon-carbon bond-forming reaction to give 3phenylthioadipates (route b). From these reactions, it was assumed that various types of ketene acetal-type reagents are available for the Pummerer-type rearrangement. Furthermore, perhaps the elaborate reactivity can be controlled by the molecular design of the reagents.
IV. ASYMMETRIC PUMMERER-TYPE REARRANGEMENT A. Highly Enantioselective Pummerer-Type Rearrangement Induced by O-Silylated Ketene Acetals
First, the normal Pummerer reaction of syn- and an//-sulfoxides with hot acetic anhydride was examined. It was found that both gave the same 80:20 ratio of
Asymmetric Pummerer Rearrangement
>Tol (±)-syn R2
^""""^ Ac^O, reflux
225
OSiMegBu^
PK
•p-Tol OAc
0"
R^^^VTOI (±)-antl
0SiMe2BJ syn 80
OSiMesBu^ anti 20
Figure 15.
diastereomeric a-acetoxysulfides. These results indicated that the acetoxy anion attacked the sulfonium intermediate from the less hindered site in the Felkin-Ahnn model (Fig. 15). Next, chiral and racemic syn- and anr/-P-substituted sulfoxides were examined with SKA in the presence of a catalytic amount of zinc iodide in acetonitrile. These substrates were convenient for identifying the results of the stereochemistry. All reactions proceeded under mild conditions with high chemical yields and a remarkably high degree of stereospecificity. It was surprising that extremely high retention occurred in all P-siloxy, P-acylamino-, P-alkylamino, P-alkyl-, and P-arylsubstituted sulfoxides, and of course, in both racemic and nonracemic sulfoxides (Fig. 16).^^ Namely, contrary to the foregoing findings obtained for the normal Pummerer reaction, an//-a-siloxysulfides were predominantly obtained from the an//-sulfoxides, and ^^'Az-a-siloxysulfides were obtained from the ^^^n-sulfoxides. These stereochemical results suggested that the stereo-
pSiM^Bu* R2
O"
R'"^^>Tol
=KOMe (SKA) Znig, MeCN
R'^^>TOI OSiMegBu*
sy/i-sulfoxides
R2
g-
s^'/i-sulfides
R^^^>TOI OSiMezBu* »
flrtri-sulfides R2
SKA
>Tol
>Tol 0SiMe2Bu*
Znl2, MeCN
^/i-sulfides
on/i-sulfoxides Figure 16.
OSiMegBu* «
fl/iri-sulfides
Table 2.
Q-
SKA
R-CH2-SLpTol S*
0-
N
.h-)
b
S*
cat. Znb, MeCN
v+ R- CH2-S-pTol R*
m
QSiM9Bu' R- CH- S - ~ T O I
OSiMeBu'
SKA
v
b
R-CH-S-pTol
cat. Zn12, MeCN
R*
Sulfoxides Entry
Config.
R
Sulfides Conditions
ee (%)
field (%)
[a1Dl
*
Reported Results Config.
1
S
COZEt
60 "C, 4 h
87
75
+35.8
S
2 3
R
s
COZEt CONMe,
60 "C, 4 h 60 "C, 12 h
86 88
72 65
-34.8 -28.9
4
R R
CONMe, Ph
60 "C, 12 h rt, 3 h rt, 3 h
88
69 87 66 61
+24.6 -30.2 +29.6
R S R R S R
5 6
s
7
R
2-Py 2-Py
rt, 4
h
70 83 82
+28.8
ee (%I
Yield (%)
70
10
65 0
35 10
Asymmetric Pummerer Rearrangement Me
Ph""^^
"^p-Tol
227 OSiM^
(cat. Znig), MeCN
Me
C02Me anti
65-93% (80-100% de) Figure 17.
chemistry of the reaction is mainly determined by the stereochemistry on the sulfur atom rather than the steric factor on the P-position. This is the first report of the highly stereoselective chiral transfer reaction using Pummerer rearrangement in the acyclic system.^ ^ To ascertain the effect of the sulfoxide itself, the reaction of SKA with optically pure sulfoxides, which have a stereogenic center on only the sulfur atom, was examined next. Known chiral, nonracemic sulfoxides were treated with SKA in the absence of a catalyst in acetonitrile to give the corresponding chiral, nonracemic a-siloxysulfides. In each case, the optical purity and chemical yields of the Pummerer adducts were greater than those from Oae's approach^^ (Table 2). It was clarified that the rearrangement proceeded with high enantioselectivity by only an asymmetric center on the sulfur atom in these Pummerer-type rearrangements. Because it had been difficult to achieve the high enantioselectivity to date, it was noteworthy that sulfoxides having the benzyl group gave a high selectivity (70% ee). However, the reaction of the sulfoxides with the less bulky SKA gave the carbon-carbon bond-forming products instead of the rearrangement product with high diastereoselectivity^^ as explained in Figure 10^^ (Fig. 17). B. Reaction Mechanism
While the general scheme of the Pummerer rearrangement is believed to be that shown in the four sequential elemental reaction steps in Figure 2, it is of interest to know which step causes the asynmietric induction. Wolfe and Kazmaier studied the diastereotopic selectivity of the deprotonation step of the syn- and anti-adeuteriobenzylmethylsulfoxides under normal Pummerer conditions.^^ According to their findings, little selectivity was observed because of the competing epimerization at the sulfur via the sulfurane intermediate. We investigated the reaction of deuteriobenzylmethylsulfoxides with SKA and found that the deprotonation of the a-proton occurred with high diastereoselectivity (Fig. 18).^"* These results suggested that the rearrangement product was produced by selective abstraction of the sulfmyl prV'R hydrogen in the substrate.
MASATO MATSUGI, NORIO SHIBATA, and YASUYUKI KITA
228
pSiMe^Bu^
~0
=<
OMe (SKA)
D
cat.Znl2, MeCN 0 ° C- rt, 1 - 6 h
D
R Me Bu*
{±)'Syn (sy/7:anf/=85:15) (±)- syn (syn:anti=96'A)
O" Ph^^^St,
SKA
Ph
Bu*
Ph^^S-R
OSiMe^Bu*
Yield (%) 36 69
OSiMefeBu*
88:12 95:5
S-R
D R Me
H
OSiMefeBu* Yield (%) 49 73
catZnla, MeCN 0°C-rt, 1-12h
(±)- anti (syn:anf/=3:97) (±)- anti (syn:anf/=4:96)
Ph^^S-R
Ph^^S-R
H
OSiMe^Bu^
32:68 5:95
Figure 18.
The following mechanism is proposed to explain these results: Silylation of optically pure sulfoxides with SKA affords the intermediate A, which may yield an anion intermediate B through abstraction of an ann'-periplanar hydrogen from the opposite face of the sulfoxide oxygen by a generated ester enolate anion. The siloxy group may then be forced to migrate to the a-position via one of the following three mechanisms: (a) intimate ion pair mechanism, (b) radical dissociation-recombination mechanism,^^ or (c) direct carbanion attack (Fig. 19). It is easier to understand the reaction mechanism of the asymmetric Pummerer rearrangement of a-deuterio cyclic sulfoxides than that of acyclic sulfoxides. The reaction of rigid cyclic sulfoxides with SKA proceeded via a trans E2-type elimination to give a-siloxysulfides with extremely high retention of the stereochemistry of the starting sulfoxides; rran5'-l-thiadecalin l-aj«*a/-oxide gave equatorial a-siloxysulfide equatorial-oxido and rra/i5'-l-thiadecalin l-equatorial-oxidc gave l-ojc/aZ-a-siloxysulfide (Fig. 20).^^ In contrast to these findings with SKA, treatment of rmn^-l-thiadecalin l-ojcfaZ-oxide and fran^-l-thiadecalin 1-equatorial-oxide with acetic anhydride in the absence or presence of DCC gave the equatorial a-acetoxysulfide in each case (Fig. 21).^^ While the mechanism of the siloxy group migration may involve either (i) an intimate ion pair mechanism or (ii) a radical dissociation-recombination mechanism, that of the acetoxy group probably implies a sulfonium ion.
~0
©
SKA
0-SiMesBu^ ^ i
0-SiMet2Bu^ p-Si^
RI*^S-R^ O-SiMe^Bu'
R2
-CHgCOgMe A OSiMegBu^
OSiMe^Bu^
Me02CCHi
^H(D)^
H(D)s A
/
OSiMe^Bu*
route a
OSiMe2Bu'
0SiMe2Bu*
0-SiMegBu'
routeb
R//,n_»..».>*.
H^R
ffl.,^.
^T
.
O-SiMe^Bu^ B route c Hs' Figure 19.
SKA
ion or radical dissociationrecombination
Bu*Me2Si-0
/:::CZ:SXJZ:H
atp-face ^
^ (87 % of H) (72% of D)
*- /C^^I!2^XZoSiMe^Bu* , „ ^ , .^. H (89% of H)
.MeOaCHgC ~
MeOgCHgC
ion or radical dissociationrecombination
^
SKA
1
0 O-l ' ' SD( 8 1 % 0 f H ) (87% of D)
at a-face Bu*Me2Si-0
J
?\gure 20.
229
/X::ZsCZ.D(89%ofD) OSIMe^Bu*
MASATO MATSUGI, NORIO SHIBATA, and YASUYUKI KITA
230 O
AC2O (DCC)
^ ^ ^ ^ ^
«-
AC2O
L—J^S^ OAc
(DCC)
H Figure 21.
C. Enantiospecific Pummerer-Type Rearrangement Induced by O-Silylated Ketene Acetals
Thefirsthighly asymmetric silicon-induced Pummerer-type rearrangement using SKA has been described. However, the extent of the asymmetric transformation never exceeded 90% ee. To develop the optimal asymmetric transformation of the sulfoxides, it is quite important to determine in which step(s) racemization occurs. The deprotonation step of the a-methylene protons plays a significant role in the stereoselectivity; therefore, the substrates which have two stereogenic centers at the a-carbon and the sulfur atom were examined. As a result, surprisingly, enantiospecific transfer was observed in the reaction of the above substrate with SKA."^^ Thus, the treatment of ^^^/i-sulfoxide with SKA in the presence of a catalytic amount of zinc iodide in THF gave enantiomerically pure a-siloxy sulfide, and the reaction of anti-sulfoxidQ with SKA likewise gave the pure one. Interestingly, the stereochemistry of the sulfoxide had no effect on the configuration of the product (Fig. 22). The introduction of a stereogenic center a- to a sulfoxide dramatically
(S)
Bu^Me2SiO
-^ 2-Pyv^S-p-Tol
2-Py^^a +\'.'P-Tol ^9 2-Pyv^S^^^^^
MecAoSiMegBu* (SKA)
R tlSiMegBu^
A"
R {CsSsH-)-anti
Bu^MegSiO R^^~+\"P"Tol
\ 2-Pyv^S-p-Tol Bu^MegSiO
2-Py^^Q
R ^ ? \ ;>Tol
-o 2-Pyv^S^
SKA
R {CsSpH')-syn R = Me R = Et
R^ OSiMegBu^
I
B
Bu*Me2SiO
'l>Jo\ B' 2-py^^?\;:.
Figure 22.
(S) R
"OSiMegBu^
Asymmetric Pummerer Rearrangement
231
ra6/e 3.
~9
MeO
2-Pyv^S>, ^ , 1 * ^ ™ R Run
OSiMegBu* —
cat.Znl2.THF
Sulfoxides
2- Py>s:^S-p-Tol R OSiMegBu* N
Products
Yield (%)
V'
ee(%) >99
1 2
3 4
5 6
0
2"PyOs^ ^ ,
I P-Tol p (Cs Ss)-(-)-am/
2-Pyi^S-p-Tol R OSiMegBu*
H
" R (Cs S^)-(-)-syn
p (Cfl Sfl)-(+)-anf/
(S) 2-Py^S-p-Tol R '0SiMe2Bu^
R = Me : 70 R = Et: 61
>99
R = Me : 49 R = Et: 72
>99 >99
R = Me : 69 R = Et: 57
>99 >99
improved the enantioselectivity from 88% ee to > 99% ee. The results showed that the deprotonation step is the most important for high enantiomeric purity, and an optimal asymmetric Pummerer reaction of chiral nonracemic acyclic sulfoxides was accomplished by controlling this step. The synthesis of enantiomerically pure quaternary substituted carbon compounds as well as complete asymmetric transfer in the Pummerer rearrangement is especially noteworthy (Table 3). D. Asymmetric Additive Pummerer-Type Rearrangement
The Pummerer-type rearrangement of vinyl sulfoxide has been named an additive Pummerer rearrangement because both the addition to the double bond and the 0SiMe2Bu' ?; ^s.+,
OMe(SKA) THF Figure 23.
MeO^C^/^/S-R
232
MASATO MATSUGI, NORIO SHIBATA, and YASUYUKI KITA
SKA >Tol
(S)
Q-
Mep^C^S.^^^,
cat.ZnCl2,THF,rt,2d
OSiMegBu^ 78% (78% ee)
SKA MeC^C^S.^^^,
'p-Tol
cat.ZnCl2.THF,rt, 1d
OSiMesBu^ 78% (80% ee)
Figure 24.
OSIMezBu* O-
OMe (SKA) catZnCIa THF
OSiMezBu*
c> -
MeOfeC^
OSiMezBu^
"•S-k^
(«) MeOfeCv
"CHgCOgMe
Figure 25.
reduction of the sulfur occur at the same time (Fig. 23). Reported examples have been achieved using acetic anhydride as the reagent, and there are a few reports on the application to the asymmetric Pummerer rearrangement.^^ Treatment of the chiral nonracemic vinyl sulfoxide with SKA as the reagent in the presence of a catalytic amount of zinc chloride (or zinc iodide) gave the additive Punmierer-type product in moderate optical yields (Fig. 24)."^ This additive Pummerer rearrangement may proceed via intermediate A and B as shown in Figure 25. When vinyl sulfoxide bearing methyl function at the a-position was employed, cyclopropylsulfide was obtained with a diastereoselectivity of 35% de (Fig. 26). Another special example is y-lactone synthesis by the reaction of vinyl sulfoxide and dichloroketene (Fig. 27).^^
O ^ S^
SKA
Me
ZnCb MeCN
(p-SiM€feBu* ^rM Me • CHgCOgMe Figure 26.
MeO^C,- H . y \
S-iP-Tol Me
65% (35%de)
Asymmetric Pummerer Rearrangement
233
CI ^
^
•
;
'A>TOI
^
CI-
THF
R ^ ^
S-p-Tol 95% ee r\
Q^
Oj
•C
o
1^
^'-f ^^o-
^,
\\
_^ci^c-_o
CI,
_
•
p o
Figure 27.
E. Effect of Substituent Groups on Silyl Function
As previously described, the effect of substituent groups on the silyl function is an important factor in terms of determining the reaction course. Table 4 shows the ratio of the products in the Pummerer reaction using various SKAs.^^ These results suggest that carbon-carbon bond formation preferentially occurs when using a small silyl function such as the trimethylsilyl function. This tendency was observed in another substrate which has asymmetric carbon at the P-position of the sulfur atom (Table 5).^-^ Interestingly, the yyn-selectivity of the rearrangement product
Table 4. 0SiX3
OH Ph^
OMe
-"^>Tol
COgMe
0SiX3
Znl2. MeCN C-C bond formation S/X3 SiMe3 SiEt3 SiMe2Bu^ SiMe2thexyl Si/Pr3 SIPh2Bu^
C-0 bond formation
Yield (%)
Ratio C-C/C-O
96 86 86 82 74 —
93/7 86/14 5/95 13/87 6/94 5/95
234
MASATO MATSUGI, NORIO SHIBATA, and YASUYUKI KITA Table 5. 0SiX3
Bu^MepSiO
0
JL s-
0SiMe2Bu^
OSiMegBu*
OMe COgMe
Znig, MeCN
C-C bond formation
Yield (%)
Ratio C-C/C-O
C-C Bond Formation antiisyn
74 92 85 46 41
>99 / 1 20/80 13/87 1/>99 1 / >99
92/8 91/9 84/16 79/21 —
Ph-'"^^%Tol
1 2 3 4Entry 5
SiMej SiMe2Bu^ SiMe2thexyl Si/Pr3SiX^ SiPh2Bu^
Ph^^^>Tol OSiXa C-0 bond formation C-O Bond Formation antiisyn 5/95 12/88 22/78 54/46
decreased when the substituent function on the silyl group was bulky. This result is via the sulfonium ion route, namely, the siloxy anion attacked the sulfonium intermediate which is formed by elimination of siloxy anion (Fig. 28).
0SiMe2Bu* .OSiXg (Hfjr Y Hs p-Tol
0SiX3
R = Bu^MeaSiO:
PVT/ X T > T o l Hs^ HR "CHgCOsMe
Bu^MegSiO^ OSiXg I _+
H
0SiMe2Bu^
HT OSiXg syn
OSIMegBu^ Ph'-^^VTol H bsiXa anti
Figure 28. F. C a r b o n - C a r b o n Bond Formation via Pummerer-Type Reaction
An attempt at carbon-carbon bond formation using chiral sulfoxide was also studied. The syn and anti nonracemic chiral sulfoxides which have asymmetric
Asymmetric Pummerer Rearrangement Bu^MegSiO
235
O
.A^s:> T o l
P}rr ^^ syn
Me
OSiMe^
Me
OMe
OSiMe^Bu*
Bu*Me2SiO
Ph"N^^>Tol
B
or Znig, MeCN
0"
X^i
PK ^ ^
'^MT^CO^Me anti
>Tol
88 -94% (> 99% de)
anti Figure 29.
carbon at the (3-position of the sulfur atom were examined. In both cases, anti products were preferentially formed in spite of the differences in the relative configuration of the employed substrates. The stereochemistry of the carbon-carbon bond-forming product was mainly controlled by only an asymmetric center on the P-position of the sulfur atom (Fig. 19)?^ These results were also interpreted via the sulfonium ion route. It is thought that the eclipse conformation is preferred over other conformations because of the electronic interaction between a sulfonium ion and a heteroatom on the P-position. As a result, the anti product was preferentially formed over the syn product as shown in Fig. 30. On the other hand, an attempt at asymmetric carbon-carbon bond formation using nonracemic chiral sulfoxide having chirality on only the sulfur atom failed probably because of the achiral sulfonium intermediate (Fig. 31).^^
Me Bu^MegSiO
pSiMes
O Me
OMe
Bu^Me2SiO !
^OSiMea 1+
EiCB or E2 elimination
Ph''''^r> >Tol H
syn
^-~CMe2C02Me Bu^eaSiO ^^
^
>Tol ^^'"'
pSiMegBu*
^'OSiMeg
Me \
Oj-SiMe3
Me
OMe
0SiMe2Bu*
Me/^COgMe anfAselectivity
Figure 30,
236
MASATO MATSUGI, NORIO SHIBATA, and YASUYUKI KITA
O"
L >Tol
Me
OSIMe^
Me
OMe
MZ^COgMe
Znig, MeCN
R = Ph: 86%, 0% ee R = COgMe: 48%, 0% ee Figure 31. G. Asymmetric Pummerer Rearrangement Induced by Ethoxy Vinyl Esters
The main reason for the high asymmetric induction of the silicon-induced Pummerer-type reactions seems to be the absence of sulfurane formation. Therefore, a novel asymmetric Pummerer reaction was carried out using a similar type of acyl-inducing reagent, ethoxy vinyl ester (EVE),"^^ which is known to be a powerful acylating reagent for active hydrogen compounds such as alcohols, amines, and carboxylic acids. It was found that EVE brought about a highly asymmetric Pummerer rearangement of chiral sulfoxides to give a-acetoxysulfides."*^ Treatment of sulfoxides with EVE in refluxing 1,2-dichloroethane, benzene, or toluene gave chiral a-acetoxysulfides in very high ee. The observed optical and chemical yields were higher than those of the reported asymmetric Pummerer rearrangement using an acid anhydride. Although the asymmetric induction of the present asymmetric Pummerer rearrangement is slightly lower than that of the silicon-induced type, it is quite interesting that the asymmetric induction is increased by preventing the formation of the sulfurane intermediate even in refluxing toluene (Table 6). Table 6. Me
OEt
toluene, reflux
OAc
R
Config.
% ee (% Yield)
C02Et CONMe2 Ph P(0)(0Me)2
R R R S
>Tol
Substrate Config. R R R S
71 (42) 84 (39) 20 (64) 69 (38)
Reported Method % ee (% Yield) 25 (22) 21 (51) 0(10) 24 (73)
Asymmetric Pummerer Rearrangement
237
H. Control of Enantioselectivity
In these reactions, the enantioselectivities might be controlled by the elimination ability of the acyloxy functions in EVE. It was considered that the stereoselectivities were mainly affected by the abstraction process of the a-hydrogen of the sulfmyl group. Therefore, the ease of continuous elimination of the acyloxy function might be an important factor also regarding the stereoselectivity. Asymmetric Pummerer reactions of enantiomerically pure sulfoxides were examined with EVE bearing various acyloxy functions to confirm the effect of the acyloxy substituent. These EVE bearing various acyl moieties were readily prepared by the reaction of ethoxyacetylene with the corresponding carboxylic acid in the presence of a catalytic amount of Bennett's complex."^ Treatment of sulfoxide with EVE in refluxing toluene gave the optically active a-acetoxysulfides in high enantiomeric excess. It is noteworthy that the enantiomeric excess of the products was influenced by the acyloxy function. That is, the electron-donating ability and steric bulkiness of the acyloxy function affect the enantiomeric excess of the products. Consequently, the highest enantioselectivity (81% ee) could be obtained by an EVE bearing a trimethoxyphenyl group as the R function (Table 7)."^^ A similar trend in enantioselectivity was observed in the cyclization of an optically active sulfoxide. A Pummerer-type ring closure reaction of enantiomerically pure sulfoxide is shown in Table 8. In this case, the chirality of the optically pure sulfoxide was transferred in only 6-11 % ee using acetic anhydride as a reagent, and in the case of using DCC as a reagent, up to 23-30% ee was obtained with the opposite configuration."*^ On the other hand, treatment of the optically pure sulfoxide with EVE at 100 °C or in refluxing toluene gave the optically active cyclic sulfide in moderate enantiomeric excess (Table 8). Similarly to the results obtained by the reaction of acyclic sulfoxide with EVE bearing various types of acyloxy functions, the increase in the electron-donating ability of EVE has a tendency to increase the enantiomeric excess of the cyclic products. The use of EVE bearing a methoxyphenyl group as the R function gave the product in relatively high enantioselectivity (44% ee). In addition, it was proved Table 7.
Me
R
MeOp,^ MeO^'
Me
Ph
CH2CI
ee (%)
81
76
71
69
66
54
53
yield (%)
19
37
47
35
20
63
39
MASATO MATSUGI, NORIO SHIBATA, and YASUYUKI KITA
238
Table 8. R
W
oOEt
———^— » toluene
R
MeO-^
P^
44 91
40 95
ee (%) yield (%)
o
33 82
Me
CH2CI
AC2O
12 86
3 92
6-11 91-95
by comparison of the sign of the product's specific rotation that the configuration of the product in the case of EVE was opposite to that in the case of acetic anhydride. The mechanism by which the product's configuration becomes opposite in this case remains to be elucidated. A convenient explanation"*^ of this effect of the acyloxy function on enantioselectivity is depicted in Figure 32. It could be assumed that the major part of this reaction proceeded via a five-membered ring transition state (cyclic mode) and/or a three-membered ring transition state (sliding mode)."^ The cleavage of the S-O bond was inhibited when using EVE bearing a powerful electron-donating R group, and hence high enantioselectivity was achieved because of the acceleration of the rearrangement via an intramolecular process. (The powerful electron-donating R group inhibited the route via a dissociation ion pair model with the disappearance of stereoselectivity.) It has become apparent that
R <^
I High electron-donating ability
OEt
'O
(EVE)
>Tol
A,
Jp^^3Et ==y-^ ^
R 0^0
I No sulfurane formation
°I
R^>v*^S-p-Tol
O Figure 32.
Asymmetric Pummerer Rearrangement
239
various EVE were effective reagents for promoting the asymmetric Pummerer reaction and that the enantioselectivity was influenced by the electron-donating ability of the R group in EVE. Consequently, effective chirality transfer from an optically active sulfoxide to the prochiral a-carbon of the sulfur atom was achieved using electron-rich EVE. This methodology will be applied to the synthesis of a biologically active substance through the optically active S'jO-acetals.'*^
V. ASYMMETRIC INTRAMOLECULAR PUMMERER-TYPE CYCLIZATION The Pummerer intermolecular cyclization induced by SKA is particularly useful in natural product chemistry because of its mild conditions. co-Carbamoylsulfoxides undergo an intramolecular Pummerer-type cyclization with SKA in acetonitrile in the presence of a catalytic amount of zinc iodide to give sulfenyl-^V-heterocycles including four- to seven-membered a-sulfenyl lactams in good to excellent yields under nearly neutral conditions."^^ P-Amido sulfoxides react with SKA to give phenylthioazetidine-2-ones (Fig. 33)."^^ This silicon-induced intramolecular Pummerer-type cyclization is also effective for the synthesis of carbapenem antibiotics such as (+)-PS-5 and (+)-thienamycin (Fig. 34). The oxidation of phenylthiolactam with m-chloroperbenzoic acid gives the corresponding sulfoxides, which when treated with SKA selectively give the rran^-azetidine-2-one esters. This reaction presumably proceeds via an acyliminium intermediate shown in Figure 35; initial silyl transfer from SKA to the sulfoxide oxygen and subsequent elimination of the phenylsulfmyl group by the electron-donating effect of the nitrogen atom would give an acyliminium intermediate, which would then undergo nucleophilic attack by the generated ester enolate at the 4-position to give carbon-carbon bond-forming products. Similarly, the starting sulfoxide reacted with silylated heteronucleophiles (Y-SiMe3) to give the corresponding rran^-4-heterofunctionally substituted azetidine-2-ones in high yields (Fig. 36).^^ The asymmetric version of an intramolecular Pummerer-type cyclization is quite useful for the synthesis of optically active heterocyclic compounds. Only a few examples of these types of reactions have been reported, and the ee yields were low. An example of an asymmetric intramolecular Pummerer cyclization was reported
9 +S-Ph NHR
MeO'^OSiMe^Bu* T (SKA) ^^^" ZnCl2, CH2CI2
r^Q_ph ^ ^ ^^ d^
0
NR
O Figure 33.
,NS-Ph
O 4-7 membered lactam
MASATO MATSUGI, NORIO SHIBATA, and YASUYUKI KITA
240
(R)
^S-A>Tol p V ^ N R
1)LiHMDS/Etl (76%) 2)m-CPBA (100%)
1)LiAIH4(68%) 2)Na/liq.NH3(81%) 3) TBDMSOTf 2,6-lutidine(100%)
Et, '1—[
(0)n i ^ol ''(—/ /-NCHPt^
^^
SKA cat. Znig (75%)
J-NCHPr^
n=0 n=1
CH2CH2OR' known
El^ SCH2CH2NHAC C02H
R=CHPh2, R'=H R=R*=H R=R'=SiMe2Bu^
(+)-PS-5
Figure 34.
(9)n
V
S-Ph
SKA cat. Znl2
0-SiMe2Bu*
Cs-Ph
/-NR O n=0 n=1
^
^',.
^"CHgCOaMe
R\
^CH2C02Me / - NNR F trans
Figure 35.
by Stridsberg and Allenmark in 1974^^ and Wolfe et al. in 1979^^ for the synthesis of chiral y-lactones (< 30% ee). The stereoselectivity in this reaction was improved to 67% ee by Kaneko et al. in 1987^^ using trimethylsilyltrifluoro-methanesulfonate (TMSOTf)/diisopropylethylamine (Fig. 37). Interestingly, SKA were also very effective reagents for these intramolecular-type cyclizations regarding the stereoselectivities. The reactions of S- and /^-sulfoxides with SKA in the presence of a catalytic amount of zinc chloride in dichloromethane gave the corresponding 4Rand 4-5-P-lactams in more than 80% ee (Table 9).^"^ These results demonstrate that the stereo-induction is influenced by the absolute configuration of the sulfoxides. The present Pummerer-type cyclization shows the highest optical induction of all previously examined methods. The following mechanism is proposed to explain the results, and a transition state is postulated for the reaction of chiral sulfoxide bearing amide moiety with SKA. Silylation of the oxygen atom of sulfoxide affords the silylated-intermediate. Thus, this intermediate may yield a chiral pseudoisothiazolone derivative through axial attack on the sulfur by the amino anion.
Asymmetric Pummerer Rearrangement ?.
241
Y-SiMe3
—^
<J-NR
11
O
frans
Y = OR",OCOR",e
Figure 36.
DCC or Ac^O •
0 < 30% ee 0~ 1+ S-Ph
TMSOTf iPrgNEt
S-Ph /-NH 0
/-NH2 0
67% ee (76% cy) Figure 37.
generated by proton abstraction with the ester enolate anion, and elimination of the siloxy ligand. The hydrogen neighboring the sulfur atom is then removed by the siloxy anion, and the amide ligand migrates from the a-face to give the (/?)-cyclized product (Fig. 38). The mechanism of penicillin biosynthesis from the Arnstein tripeptide, 6-(L-aaminoadipoyl)-L-cysteinyl-D-valine (ACV), has been extensively studied and reviewed by many chemists. Most of the biosynthetic mechanisms have been ascertained by Baldwin and Bradley using an excellent enzymatic technique.^^ However, the first step in the biosynthesis of penicillin, conversion of the Arnstein tripeptide to a cw-jJ-lactam intermediate, is still a fascinating mechanistic problem. Although Baldwin et al. recently proposed a mechanism involving an iron-bound thioaldehyde formation route via a Pummerer-type cyclization, the intermediate for this mechanism has not been identified. The mechanism of selective formation of the cw-p-lactam ring is still also unknown (Fig. 39).^^ These types of biomimetic reactions have been chemically studied. An example of an unsuccessful intramolecular Pummerer cyclization of the sulfoxide involving a cysteine moiety under standard Pummerer conditions was reported by Wolfe et al.^^ Although Kaneko reported the conversion of the very simple 3-phenylsulfmyl propionamide into a P-lactam with TMSOTf/triethylamine,^^ a successful biomimetic synthesis of
MASATO MATSUGI, NORIO SHIBATA, and YASUYUKI KITA
242
Table 9. 0" f+ ^S-p-Tol
J-NHR
MecAoSiMe^Bu* (SKA)
^ ^.sS-p-Tol /-NR 0
cat. ZnCl2, CH2CI2
vJ
(S)
9"
^^S-p-Tol
SKA
/XR
•
cat. ZnCl2, CH2CI2
i-NHR
Products Entry
Sulfoxides
R
(5) (5)
CH(Me)Ph CH(Me)Ph CH(Me)Ph CH2Ph CH2Ph CHPh2 CHPh2 CH2-a-BrPh
1 2 3 4 5 6 7 8
(/?)
(5) (/?)
(S) (/?)
(5)
Conditions
Config .
lalo
ee(%)
Yield (%)
R R S R S R S R
-98.8
60 82 85 80 82 80 83 83
71 96 89 54 54 84 90 59
0 °C, 1 d 0 °C, 3 d 0 °C, 3 d 5 °C, 6 d 5 °C, 6 d 15°C, 2d 15°C, 2d 5 °C, 7 d
-98.8 +116 -73.2 +75.2 -37.0 +38.4 -59.2
P-lactams from Arnstein tripeptide derivatives had not yet been carried out (Fig. 40). The intramolecular Pummerer-type cyclization of the closely related (/?)-Arnstein tripeptide analogues with SKA predominantly gave the cw-P-lactams and the (5)-Arnstein tripeptide analogues gave a mixture of cw and trans P-lactams.^^^ It is noteworthy that cw-P-lactams were preferentially obtained from the (/?)-substrate considering the fact that the 3-amino-P-lactam moiety of naturally occurring
o.- n t ^SiMe2Bu^
(S) "Q^ / • + 'S-p-Tol J-NHR
SKA^
/ 0SiMe2Bu^ ( ^
H.
7^S-p-Tol
/-NR
O R ) A
I CH2C02Me
Figure 38.
B
Asymmetric
Pummerer
Rearrangement
243
H2O N.
COg"
^
.SH
IPNS
J-NH
""T^s, o
/
02
HO2C
HOgC?
^
Amstein tripeptide (ACV)
j-HN--^^
Fe=0
Pummerer -type rearrangement
O^Te^ -|-HN,
-^-HN,
/-N O HO2CJ
HO2C
602H
c/s-P-lactam
isopenicillin N
F/gi/re 39.
AcHIW
+ S~Ph
^
J—NHCHgPh O
Me
AcHIS^^
Ac^O or TFAA
+S-Ph
^SPh
J—NCh^Ph
EtgN
J-NH
41%
^NH2
"*"
J—NH
O
O^ a s : trans = 2 . 7 : 1
Figure 40. (^s)
-o V
O
'
1 MeOr^OSiMebBu^ (SKA)
^
cat. ZnCl2. CH2CI2
O^V-C MeO^C
MeO^CJ
cis 3.1-11 : 1 (Ss) RHN^^
+S-Ph
RHN^^
SKA
^SPh ZHN.
Q ^ N H / MeO^C
cat.ZnCl2.CH2Cl2
d ^ ^ V - ( ^ MeO^C
Figure 41.
R= C02Bn
O
MASATO MATSUGI, NORIO SHIBATA, and YASUYUKI KITA
244
HN-AA
HN~AA 1.NH
IPNS
ACV
H.
NH ^
H H. . N H / ^ ^
,
V
Fe 2+ H ^ , . H20^|^Vsp2i6 Me HIS 270
<-9 h
MeO/,.jJ^uHiS2i4 Me H20^1.^Asp2i6 nIS 270
HN-AA
Ms illin N -• isopenicillin
-Q QNX^T Me
HN-AA
5
X.j^O^UxHiS2l4 Fe: Me' HpO H 2 0 ^ i ^ A s p 2 i 6 "2* Hjs 270
niS 270
AA = L-a-aminoadlpoyI Figure 42.
penicillin has a cw-orientation. The present results provide useful information on thefirstkey step in penicillin biosynthesis (Fig. 41).^^*' Very recently, Baldwin and his co-workers unraveled the intimate details of penicillin biosynthesis using the substrate-bound crystal structure of isopenicillin synthetone (IPNS)^ (Fig. 42).
VI. ASYMMETRIC PUMMERER REACTION USING SELENOXIDES The Pummerer-type reaction is not limited to the use of sulfoxides as the substrates (such as ^-oxide in Section III). For example, arylselenoxide bearing a difluoromethyl moiety is reacted with acetic anhydride and tetrahydrofuran to give the a-substituted product in excellent yield (Fig. 43).^^ This process might occur through the selenonium compound which is considered as an intermediate of the Pummerer-type reaction. Until recently, simple chiral selenoxides were little studied, being first reported by F. A. Davis et al. in 1983.^^ The principal difficulty in studying and preparing chiral selenoxides in high enantiomeric excess is their configurational lability. In earlier studies, Davis et al. demonstrated that chiral alkyl arylselenoxides racemize in the presence of moisture via the formation of an achiral hydrate which is strongly acid catalyzed (Fig. 44).^^ Although successfiil asynmietric Pummerer rearrangement using chiral selenoxide has not been achieved to date, a few applications of asymmetric rearrangements were reported. The asynmietric
Asymmetric Pummerer
245
Rearrangement
O
Se'V.F F
AC2O
Q
87%
AcO
Figure 43.
O
OH
H3O*
HaO^ :-Se
A/
OH' Figure 44.
oxidation of (E)- and (Z)-aryl cinnamyl selenides with (+)-oxaziridine affords optically active 1-phenyl allyl alcohol via a concerted [2,3]-sigmatropic selenoxide-selenate rearrangement. The extent of chirality transfer (41-62% ee) as well as the endo/exo transition state geometry is highly dependent on the structure of the
.Ph
J-
PhSex
40% ee
Figure 45.
M e 0 2 C > ^ / \ * ^ Se—Ph
<::>s^Se-Ph
OSiMe^Bu^
1
ih ^ MeO
OSiMe^Bu^
(SKA)
Figure 46.
31% yield 13% ee
246
MASATO MATSUGI, NORIO SHIBATA, and YASUYUKI KITA
O^^
SKA
Se-Ph CDCb
* 25% ee
k.A.*^Se-Ph
CDCb 51% yield 9% ee
Figure 47.
allylic selenide (Fig. 45).^^ Although it was found that the asymmetric Pummerertype reaction of phenyl vinyl selenoxide was observed by one-pot reaction (asymmetric oxidation continuous Pummerer rearrangement),^ the extent of the chiral transfer was not clarified because of the lability of the prepared selenoxide in situ (Fig. 46).^^ On the other hand, it is confirmed that the asymmetric Pummerer rearrangement of pyridylmethylphenylselenoxide (25% ee) proceeds with 74% ee chiral transfer to give the corresponding product in 9% ee (Fig. 47).^^
VII. OUTLOOK FOR ASYMMETRIC PUMMERER REARRANGEMENT Asymmetric Pummerer rearrangement is a very attractive reaction as previously described. In particular, the reactions induced by SKA work well, and may be synthetically exploited in many cases. The results described here demonstrate that the stereoselective a-deprotonation of the sulfoxide is a prerequisite process for asymmetric induction in the Pummerer reaction. Since many kinds of synthetic and enzymatic preparative methods of optically pure sulfoxides have been developed, the present Pummerer-type reaction will be applicable to many other chiral sulfoxides with one a-substituent, chiral vinylsulfoxides and chiral co-carbamoylsulfoxides, thus leading to enantioselective syntheses of many new bioactive compounds in the near future. It is expected that the construction of elaborate selective reactions such as biomimetic synthesis catalyzed by enzymes will be developed.
REFERENCES 1. Pummerer, R., Chem. Ber. 1909,42, 2282; 1910, 43, 1401. 2. (a) Homer, L.; Kaiser P. Justus Liebigs Ann. Chem. 1959, 626, 19. (b) Homer, L. Justus Liebigs Ann. Chem. 1960, 657, 198. 3. Reviews, see: (a) Durst, T. Adv. Org. Chem. 1969, 6, 285. (b) Numata, T.; Oae, S. Yuki Gosei Kagaku Kyokai Shi 1977, 55, 726. (c) Marino, J. P. In Topics in Sulfur Chemistry, Senning, A., Ed.; Thieme: Stuttgart, 1976, Vol. 1, p. 1. (d) Numata, T. Yuki Gosei Kagaku Kyokai Shi 1978,36, 845. (e) Oae, S.; Numata, T. In Isotopes in Organic Chemistry; Buncel, E.; Lee, C. C , Eds.;
Asymmetric Pummerer Rearrangement
4.
5.
6. 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.
247
Elsevier: New York, 1980, Vol. 5, Chapter 2. (0 Oae, S.; Numata, T; Yoshimura, T. In The Chemistry of the Sulphonium Group; Stirling, C. J. M.; Patai, S., Eds.; John Wiley & Sons: New York, 1981, Part 2, Chapter 15. (a) McCormick, J. E.; McElhinney, R. S. J. Chem. Soc, Chem. Commun. 1969, 171. (b) Glue, S. Kay, I. T.; Kipps, M. R. Chem. Commun. 1970, 1158. (c) Numata, T.; Oae, S. Tetrahedron Lett. 1977, 1337. (d) Numata, T.; Itoh, O.; Oae, S. Chem. Lett. 1971, 909. (e) Masuda, T.; Numata, T. Furukawa, N.; Oae, S. J. Chem. Soc, Perkin Trans. 2 1978, 1302. (f) Itoh, O.; Numata, T. Yoshimura, T.; Oae, S. Bull. Chem. Soc. Jpn. 1983, 56, 266. (g) Oae, S.; Itoh, O.; Numata, T. Yoshimura, T. Bull. Chem. Soc. Jpn. 1983,56,270. (h) Bhupathy, M.; Cohen, T. Tetrahedron Lett. 1987, 28, 4793. (i) Tokitoh, N.; Igarashi, Y; Ando, W. Tetrahedron Lett. 1987, 28, 5903. (j) Gonzalez, F. S.; Mendoza, P G. Carbohydn Res. 1988, 227. (a) Mikolajczyk, M.; Zatorski, A.; Grzejszczak, S.; Costisella, B.; Midura, W. J. Org. Chem. 1978, 43, 2518. (b) Numata, T.; Itoh, O.; Oae, S. Tetrahedron Lett, 1979,1869. (c) Numata, T.; Itoh, O.; Yoshimura, T.; Oae, S. Bull. Chem. Soc. Jpn. 1983,56, 257. Kennedy, M.; Mckervey, M. A. In Comprehensive Organic Synthesis, Pattenden, G., Ed.; London, 1991, Vol. 2, p. 193. Numata, T.; Oae, S. Tetrahedron Lett. 1977, 1337. Numata, T.; Itoh, O.; Oae, S. Chem. Lett. 1977, 909. Numata, T.; Oae, S. Yuki Gosei Kagaku Kyokai Shi WJl, 35, 726. Glue, S.; Kay, I. T.; Kipps, M. R. Chem. Commun. 1970, 1158. Masuda, T.; Numata, T.; Furukawa, N.; Oae, S. J. Chem. Soc, Perkin Trans. 2 1978, 1302. Bhupathy, M.; Cohen, T. Tetrahedron Lett. 1987, 28, 4793. Tokitoh, N.; Igarashi, Y; Ando, W. Tetrahedron Lett. 1987, 28, 5903. Gonzalez, F S.; Mendoza, P G. Carbohydn Res. 1988,183, 227. (a) Numata, T.; Itoh, O.; Oae, S. Tetrahedron Lett. 1979,1869. (b) Numata, T.; Itoh, O.; Yoshimura, T; Oae, S. Bull. Chem. Soc Jpn. 1983, 56, 257. Numata, T.; Oae, S. Yuki Gosei Kagaku Kyokai Shi 1977, 35, 726. (a) Numata, T.; Itoh, O.; Oae, S. Tetrahedron Lett. 1979,1869. (b) Numata, T.; Itoh, O.; Yoshimura, T.; Oae, S. Bull. Chem. Soc Jpn. 1983, 56, 257. Itoh, O.; Numata, T.; Yoshimura, T.; Oae, S. Bull. Chem. Soc Jpn. 1983, 56, 266. (a) Kita, Y Yakugaku Zasshi 1986,106,269. (b) Kita, Y; Tamura, O.; Tamura, Y Yuki Gosei Kagaku KyokaiShi19S6,44, 1118. (a) Kita, Y; Yasuda, H.; Sugiyama, Y; Furukawa, F ; Haruta, J.; Tamura, Y Tetrahedron Lett. 1983, 24, 1273. (b) Kita, Y; Haruta, J.; Fujii, T; Segawa, J.; Tamura, Y Synthesis 1981, 451. (c) Kita, Y; Haruta, J.; Segawa, J.; Tamura, Y Tetrahedron Lett. 1979, 4311. Brook, A. G. Ace Chem. Res. 1974, 7, 77. Olah, G. A.; Narang, S. C. Tetrahedron 1982, 38, 2225. Kita, Y; Yasuda, H.; Tamura, O.; Itoh, F ; Tamura, Y Tetrahedron Lett. 1984, 25, 4681. Kita, Y; Yasuda, H.; Tamura, O.; Itoh, F ; Tamura, Y Chem. Pharm. Bull. 1987, 35, 562. (a) Kita, Y; Yasuda, H.; Tamura, O.; Itoh, F; Tamura, Y Chem. Pharm. Bull. 1985, 33, 4235. (b) Mori, I.; Bartlett, A.; Heathcock, C. H. J Org. Chem. 1990, 55, 5966. Magunus, P; Mitchell, I. S. Tetrahedron Lett. 1998,39, 9131. Kita, Y; Shibata, N. Synlett 1996, 289. Kita, Y; Gotanda, K.; Fujimori, C ; Murata, K.; Matsugi, M. J Org. Chem. 1997, 62, 8268. Kita, Y; Tamura, O.; Miki. T; Tamura, Y Chem. Pharm. Bull. 1987, 35, 562. Kita, Y; Shibata, N.; Yoshida, N. Tetrahedron Lett. 1993, 34, 4063. Wolfe, S.; Kazmaier, P M. Can. J Chem. 1979, 57, 2388, 2397. Kita, Y; Shibata, N.; Yoshida, N.; Fujita, S. / Chem. Soc, Perkin Trans. J 1994, 3335. Kita, Y; Shibata, N.; Fujita, S.; Gyoten, M.; Fujimori, C. 45th Meeting of Kinki Branch of Pharmaceutical Society of Japan, Dll-5, Kinki University, 1995. Kita, Y; Shibata, N.; Yoshida, N.; Fukui, S.; Fujimori, C. Tetrahedron Lett. 1994, 35, 2569.
248
MASATO MATSUGI, NORIO SHIBATA, and YASUYUKI KITA
35. (a) Schollkopf, U. Angew. Chem. Int. Ed. Engl. 1970, 9, 763. (b) Hoffmann, R.; Ruckert, T.; Buckner, R. Tetrahedron Lett. 1993,34, 297. (c) Tomooka, K.; Igarashi, T.; Nakai, T. Tetmhedwn Lett. 1993, 34, 8139; 1994,50, 5927. 36. Kita, Y; Shibata, N.; Yoshida, N.; Fukui, S.; Fujimori, C. Tetrahedron Lett. 1994, 35, 2569. 37. Oae, S.; Itoh, O.; Numata, T.; Yoshimura, T. Bull. Chem. Sac. Jpn. 1983, 56, 270. 38. Kita, Y; Shibata, N.; Fukui, S.; Fujita, S. Tetrahedron Lett. 1994, 55, 9733. 39. Reviews, see: (a) Block, E. Reaction of Organosulfur Compounds', Academic Press, New York, 1980. (b) Kosugi, H. Yuki Gosei Kagaku Kyokai Shi 1987,45, All. 40. Shibata, N.; Fujimori, C ; Fujita, S.; Kita, Y Chem. Pharm. Bull. 1996, 44, 892. 41. (a) Kosugi, H. Yuki Gosei Kagaku Kyokai Shi 1987,45, All. (b) Kosugi, H.; Kitaoka, M.; Tagami, K.; Uda, H. / Org. Chem. 1987,52,1078. (c) Marino, J. R; Perez, A. D. J. Am. Chem. Soc. 1984, 106, 7643. 42. Kita, Y; Maeda, H.; Omori, K.; Okuno, T.; Tamura, Y J. Chem. Soc, Perkin Trans. J 1993, 2999. 43. Shibata, N.; Matsugi, M.; Kawano, N.; Fukui, S.; Fujimori, C ; Gotanda, K.; Murata, K.; Kita, Y. Tetrahedron: Asymmetry 1997, 8, 303. 44. Kita, Y; Maeda, H.; Omori, K.; Okuno, T.; Tamura, Y Synlett 1993, 273. 45. Stridsberg, B.; Allenmark, S. Acta. Chem. Scand. 1974, B28, 591; 1976, B30, 219. 46. (a) Numata, T.; Itoh, O.; Yoshimura, T.; Oae, S. Bull. Chem. Soc. Jpn. 1983,56,151. (b) Numata, T.; Oae, S. Tetrahedron Lett. 1979, 1869. 47. Kita, Y; Shibata, N. Synlett 1996, 289. 48. (a) Kita, Y; Tamura, O.; Miki, T; Tamura, Y Tetrahedron Lett. 1987,28,6479. (b) Kita, Y; Tamura, O.; Shibata, N.; Miki, T Chem. Pharm. Bull. 1990, 38, 1473. 49. (a) Kita, Y; Tamura, O.; Miki, T.; Tono, H.; Shibata, N.; Tamura, Y Tetrahedron Lett. 1989, 30, 729. (b) Kita, Y; Tamura, O.; Shibata, N.; Miki, T. J. Chem. Soc, Perkin Trans. 11989, 1862. (c) Kita, Y; Shibata, N.; Tamura, O.; Miki, T Chem. Pharm. Bull. 1991, 39,1115. 50. (a) Kita, Y; Shibata, N.; Tohjo, T; Yoshida, N. J. Chem. Soc, Perkin Trans. 11992,1795. (b) Kita, Y; Shibata, N.; Yoshida, N.; Tohjo, T Tetrahedron Lett. 1991, 32, 2375. 51. Stridsberg, B.; Allenmark, S. Acta Chem. Scand. 1974, B28, 591; 1976, B30, 219. 52. Wolfe, S.; Kazmaier, P M.; Aukis, H. Can. J. Chem. 1979, 57, 2404. 53. Kaneko, T; Okamoto, Y; Hatada, K. J. Chem. Soc, Chem. Commun. 1987, 1511. 54. Kita, Y; Shibata, N.; Kawano, N.; Tohjo, T; Matsumoto, K. Tetrahedron Lett. 1995, 36, 115. 55. Baldwin, J. E.; Bradley, M. Chem. Rev. 1990, 1079. 56. Queener, S. W.; Neuss, N. In Chemistry and Biology of ^-Lactam Antibiotics', Morin, R. B.; Gorman, M., Eds.; Academic Press: New York, 1982, Vol. 3, pp. 22-81. 57. (a) Wolfe, S.; Kazmaier, P M.; Aukis, H. Can. J. Chem. 1979,57, lAll. (b) Wolfe, S.; Bowers, R. J.; Hasan, S. K.; Kazmaier, P M. Can. J. Chem. 1981, 59, 406. 58. Kaneko, T J. Am. Chem. Soc 1985,107, 5409. 59. (a) Kita, Y; Shibata, N.; Kawano, N.; Tohjo, T; Fujimori, C ; Ohishi, H. J. Am. Chem. Soc 1994, 116, 5116. (b) Shibata, N.; Kita, Y Chem. Heterocycl Compd 1998, 1463. 60. Roach, P L.; Clifton, I. J.; Hensgens, C. M. H.; Shibata, N.; Schofield, C. J.; Hajdu, J.; Baldwin, J. E. Nature 1997, 387, 827. 61. Uneyama, K.; Maeda, K.; Tokunaga, Y; Itano, N. J. Org. Chem. 1995, 60, 370. 62. Davis, F. A.; Billmers, J. M.; Stringer, O. D. Tetrahedron Lett. 1983, 24,3191. 63. Davis, K A.; Reddy, R. T J. Am. Chem. Soc 1992,57, 2599. 64. Davis, F A.; Stringer, O. D.; McCauley, J. P Tetrahedron 1985, 41, AlAl. 65. Unpublished data.
SYNTHESES AND REACTIONS OF SULFINIMINES
Ping Zhou, Bang-chi Chen, and Franklin A. Davis
I. Introduction II. Preparation of Sulfmimines A. From Sulfenimines B. From a-Aminosulfenimides C. From Sulfmyl Halides D. From Sulfmates E. From Thiosulfmates F. From Sulfmamides G. From Halosulfoximides H. From a-Azidosulfoxides I. From A^-Sulfmylamides J. From Sulfonylazides III. Reactions A. Eliminations B. Oxidations C. Halogenations D. Reductions E. Michael Additions
250 250 250 253 253 254 256 256 259 259 259 260 261 262 262 263 263 265
Advances in Sulfur Chemistry Volume 2, pages 249-282. Copyright © 2000 by JAI Press Inc. Allrightsof reproduction in any form reserved. ISBN: 0-7623-0618-1 249
250
PING ZHOU, BANG-CHI CHEN, and FRANKLIN A. DAVIS
F. [1+2] Cycloadditions G. [3+2] Cycloadditions H. Hetero Diels-Alder Reactions IV. Conclusions Acknowledgments Note Added in Proof References
273 275 277 278 278 279 279
I. INTRODUCTION The sulfur bonding imines 1 are versatile intermediates in organic synthesis.^~^ Among them, sulfmimines (thiooxime 5-oxides, A^-sulfmyl imines, A^-alkylidenesulfinamides, lb) displayed unique reactivity and stereoselectivity as a result of the existence of the chiral electron-withdrawing sulfmyl group."*"^ Like sulfoxides, sulfmimines undergo thermo elimination to give sulfenic acids (RSOH). Sulfmimines also undergo Michael-type reactions with a variety of nucleophiles and more importantly the sulfur stereogenic center makes it possible to control many of these reactions in a highly diastereoselective manner. On the other hand, many molecules having sulfmimine structure units show interesting biological properties.^"^ The purpose of this chapter is to give a general summary of the chemistry of sulfinimines with particular attention to the application of enantiopure sulfmimines in asymmetric synthesis.
II. PREPARATION OF SULFINIMINES A. From Sulfenimines
Chemoselective oxidation of sulfenimines 2 is a useful method for the preparation of sulfmimines 3.^^"^^ The oxidation was usually carried out with m-CPBA (3-chloroperbenzoic acid) in a biphasic mixture of chloroform and aqueous sodium bicarbonate and a variety of sulfmimines have been prepared in this way. Sulfmimines 3 were usually obtained as a single £-isomer despite the fact that their precursors 2 may exist as mixtures of E- and Z-isomers.^^ Other oxidants which have been demonstrated to be effective for oxidation include MMPP (magnesium monoperoxyphthalate)^^ and chlorine. ^"^'^^ Various sulfmimines with other oxidizable groups in the molecules can also be obtained in good yields by this method.^^'^^ For example, chemoselective oxidation of sulfenimine 4 with m-CPBA gave sulfmimine 5 in 95% yield.^^ Gordon et al. (0)n R^ 1a, n=0, sulfenimine b, n=1,sulfinimine c, n=2, sulfonimine
Syntheses and Reactions of Sulfinimines
251 O
m-CPBA/CHOa HzO/NaHCO^
Ar'
Bu-t
Bu-t O
m-CPBA/ChtCl2
^ t-Bu'^^^N'^Ph
^N^Ph
t-Bu
R^
rt/5min. 95%
found that oxidation of 7-sulfeniniinocephalosporin 6 with m-CPBA is highly selective affording 7 in 52% yield as a diastereomeric mixture, with none of the sulfone being detected.^^ Dehydrophenylalanine sulfinamide 10 was isolated in 31% yield as a single isomer of undetermined geometry at the double bond on treatment of 8 with m-CPBA.^^ The product is thought to be formed by rapid isomerization of the initially formed sulfmimine 9 which could not be isolated. Optically active sulfinimines were prepared by Yang and co-workers by oxidation of nonracemic sulfenimines.^^ Thus, treatment of 11 with m-CPBA or MMPP afforded sulfinimines 12 in 83-99% yield. The diastereoselectivity, however, was highly dependent on the R group in the chiral auxiliary. When R = H, diastereopure sulfinimine (R^)-12 was obtained. Another approach to the enantiomerically enriched sulfinimines reported by Davis and co-workers is the enantioselective oxidation of sulfenimines with an asymmetric oxidizing reagent, (-)-//-(phenylsulfonyl)(3,3-dichlorocamphoryl)p-MePh^
p-MePh^ O '>\
S
m-CPBA/CH2Cl2
' \
^S
^
K^. "^r- „)=Q<
C02CH2CCI3
CO2CH2CCI3
Ph>
Ph
1
m-CPBA/CH2Cl2
p-MePhSN'^^^^C02Et
p-MePhlS(0)N'*^C02Et
p-MePhlS(0)NH
Y
C02Et
10 (31%)
252
PING Z H O U , BANG-CHI CHEN, and FRANKLIN A. DAVIS Ph
Ph
.xTv^S
m-CPBAorMMPP
— ( •
i^TN^^v'^ •
Y^OR
Ph
•
( •
Y^OR
11
J^Ty^^'h^' +
"^i ( -•
Y^C
(^).12
vJ
(^H2
oxaziridine (14).^^"^^ Thus, oxidation of sulfenimine 13 with (-)-14 afforded sulfinimine (R^yiS in 85-91% ee, which on recrystallization were obtained enantiomerically pure. The antipode sulfmimines (S^yiS can be readily prepared by the use of epimeric (+)-A^-(phenylsulfonyl)(3,3-dichlorocamphoryl)oxaziridine, (+)-14. A related reaction is the oxidation of isothiazoles to the corresponding oxides. Treatment of 3-amino-5-phenylisothiazole 16 with persulfuric acid gave 17 in 60% yield.^"^ Nitric acid in acetic anhydride has also been used.^^ In another example, oxidation of 18 gave 19 in 99% yield.^^ Noteworthy is the fact that the other sulfur atoms are unaffected by this oxidation. Oxidation of 3-amino-l,2-benzisothiazoles 20 using nitric acid was reported to give 21 in 48% yield.^^ On the other hand, oxidation of benzisothiazole 22 with m-CPBA afforded sulfinimine 23 which showed significant in vivo antipsychotic activity.^ The latter reaction was run at -78 °C in methylene chloride to minimize overoxidation. Inferior results were obtained when nitric acid was used as the oxidant.
1
^^^ -4 1
X^ci
13
O \
('^HS
(.).14
o II
Ph^/^^N
H2O2/H2SO4
M ,
rt/2h NH2
Ph-^S.^
•
M ,
60%
^^
NH2
16 O \\
// 18
S Ph
m-CPBA/ChtCl2 ^ 99%
\JF
S >h
Syntheses and Reactions ofSulfinimines
253
N.S-0
21
20
m-CPBA/CH2Cl2
N^c-P
B. From a-Aminosulfenimides
Oxidation of N-benzoylsulfenimide 24 with 1 equiv of m-CPBA in methylene chloride gave 25, which on further oxidation gave the heterocyclic sulfinimine 27 in 64% yield.^^'^^ This oxidative elimination may be the result of the decomposition of the intermediate A^-oxide 26 as shown.
11 ,S-|sj''''^"Ph
ff^CPBA/CH2Cl2
24
C. From Sulfinyl Halides
Condensation of sulfinyl chlorides and NH ketimines gave, as expected, the corresponding sulfinimines.^"*'^^ For example, treatment of arylsulfinyl chlorides
PING ZHOU, BANG-CHI CHEN, and FRANKLIN A. DAVIS
254 O II
Ar-^Xl
O
EtaN/Et^O
>='NH
81-85%
Ph
29
28
R
30 Ar=p-MePh,/>BrPh R = Ph, CF3
O II
F3Q )=NH F3C
O
CsF
VaC'^^N^^CFa
32
31
CF3
33
28 with ketimines 29 in the presence of triethylamine gave sulfinimines 30 in 81-85% yield. Reaction of trifluoromethanesulfinyl fluoride (31) and the hexafluoroacetone derived ketimine 32 is reported to give 33 in the presence of CsR^^ Aminonitrile 35 reacts with 2-nitrobenzenesulfinyl chloride (34a) to give 36 in 55-62% yield.^^ 4-Nitrobenzenesulfinyl chloride 34b reacts in a similar way.^^ With LiHMDS sulfmyl chloride 37 gave A^,A^-bis(trimethylsilyl) sulfinamide 38 in situ. Subsequent addition of an aldehyde and CsF affords sulfinimines 39 in good yield.^^ This "one-pot" procedure is suitable for the preparation of both alkylidene and arylidene sulfinamides. CI
x ^ " - " . 34a, X = 2-NO2 b, X = 4-NO2
9
*^0
Nss — N 'p
'^-N^N'"
55-73y<
35
36
0
RCHO/CsF
LiHMDSmHF
Ar*^
AK^^CI
H N
R
TMS 38
37
39 R
D.
% Yield
Ph
65
n-Pr
48
From Sulfiinates
Enantiopure sulfinimines 43 were prepared for the first time by Cinquini and co-workers from the commercially available Andersen reagent, (l/?,25,5/?)-(-)menthyl (5)-/?-toluenesulfinate (40) or (15,2/?,55)-(+)-menthyl (/?)-/7-toluenesulfi-
Syntheses and Reactions ofSulfinimines
255
ArCN 41 RLI or RMgBr orDIBAL-H
O
R
O R
Ar
p-MePtf "O'
p^MePrt'' - M ^ ^ A 42
(S5)-43
(5s)-40
nate (40) and imino-metallo reagents 42 in moderate yields."^^"^^ The reaction is highly stereoselective, taking place at the chiral sulfur atom in an Sj^2 fashion. The imino-metallo reagents 42 are usually prepared in situ via the reaction of aromatic nitriles 41 with organolithium or Grignard reagents or DIBAL and for this reason the method is limited. A method that makes available aromatic and aliphatic aldehyde derived sulfmimines 47, for thefirsttime, was recently introduced by Davis and co-workers.^"^'^^ This one-pot procedure entails treatment of the Andersen reagent 40 with LiHMDS to generate 44 which subsequently reacts with the lithium methoxide by-product to produce silyl sulfinamide anion 46. Reaction of 46 with the aldehyde in a Peterson-type olefination reaction affords the sulfinimine 47 in >96% ee. This method was highly effective for the preparation of arylidene sulfinamides 47 (R = aryl) which were usually obtained in 60-76% yield although the alkyl counterparts
>.?
p-MePh^
N
p-MePh'^^"
Lia
TMS (Ss)-44
O
H
RCHO
p-MePh^''^"N^
R
(Ss)-47 (>96% ee)
% Yield Ph p-N02Ph 3-Pyriclyl E-MeCH=CH n-Pr ^Bu
76 74 64 60 64 61
p-MePlf
N 46
256
PING ZHOU, BANG-CHI CHEN, and FRANKLIN A. DAVIS LiHMDSn-HF
"^c
I
X
'~a'
(S.)-49
c*'^'
44%
48
47 (R = alkyl) were obtained in more moderate yields (60-65%). Bis-sulfinimines 49 can be prepared in a similar way from isophthalaldehyde (48).^'^ In a variation of this method, Garcia Ruano and workers prepared 5-alkyl sulfmimines from the diacetone-D-glucose derived sulfinate 50."^^ Subsequent treatment with LiHMDS, aldehyde, and CsF afforded 5-r^rf-butyl sulfmimines 51 in enantiomerically pure form.^^
1.UHMDS 2. RCHO/CsF
r\ . \jO f-Bu*^
LA m H N
R
51 R
% Yield
Ph PhCH=CH
70 56
E. From Thiosulfinates
Reaction of (R)-(+ytert-butyl r^rr-butanethiosulfmate (52) with lithiated imine 53 gave 54 as a single isomer in 92% yield.-^^ The thiosulfmate 52 was prepared by asymmetric oxidation of rerr-butyldisulfide with H2O2 in the presence of a chiral vanadium complex. Me
O
S I ^•S.^J< 52
^ ® ^ ^^^
THF/-78°aih
. '^'g^ A >f ^N^Ph
53
54 92%, 100%ee
F. From Sulfinamides
Analogous to sulfmates, sulfinamide 55 reacts with the lithiated imines 42 to give 56 as a single isomer in good yield."^^""*^ In a similar manner, camphorsultam-derived (R^)-57 reacts with LiHMDS in THF to give (S^y44 which reacts with aldehydes in the usual way to give (S^y47 in 65-84% yields and 98- > 99.5% ee."*^ Interestingly, better yields of enolizable sulfinimines were obtained only if 1 equiv of water was present.
Syntheses and Reactions of Sulfinimines y^Me
^
r\4 •>='"' ;s^s
257
THF
Bu-f
f-BuCONK 42
55
Me i-Pr f-Bu
78 54 52
O LiHMDS/THF , *- b-MePft'
"N
x-TMS
TMS (^)-44
O
H
RCHO
p-MePrt'
N
Condensation of benzenesulfinamide (58) with reactive carbonyl compounds such as hexafluoroacetone (59) gave hemiaminal 60 in 74% yield."^ Subsequent treatment with trifluoroacetic anhydride and pyridine afforded 62 in 42% yield. Elimination of TfOH in 61 was proposed to account for this. Bravo et al. treated (55)-(+)-p-toluenesulfmamide (63), prepared by hydrolysis of 44,^^ with triphenylphosphine in the presence of DEAD to give the AT-sulfmyl iminophosphorane 64 in 92% yield.'*^ The Staudinger, aza-Wittig reaction of 64 with methyl or ethyl trifluoropyruvate afforded the unstable sulfmimine 65. Attempts to purify the imino sulfinimines by flash chromatography resulted in hydrolysis.
O Ph" ^NH2
F3C
O
F3C ^ . ,
Ph-^^N^CFs I H
F3C 59
58
Tf20/pyr
CI-I2CI2
O
FQC
61
O
CF3
62 (42%)
258
PING ZHOU, BANG-CHI CHEN, and FRANKLIN A. DAVIS :S
PPha/DEAD^
p-MePh^'*^"NH2
92%
:^9
CFaCOCO^R^
.^O
CO2R
p-MePh'^''^^N=PPh3 PUHMO^C p-MePh^'* "N
(Ss)-63
(Ss)-64
CF3
(Ss)-65 R=Me, Et
A particularly effective method for the asymmetric synthesis of both aldehydeand ketone-derived sulfinimines recently introduced by Davis and co-workers is the condensation of (5)-(+)-/7-toluenesulfmamide (63) with aldehydes and ketones using activated 4-A molecular sieves or titanium ethoxide [Ti(OEt)4].'*^ This procedure avoids the problem of removing the menthol by-product of the one-pot procedure (see Section II.D) which is sometimes problematic.^^ Importantly, this methodology affords ketone derived sulfinimines 66 which are difficult to prepare by other means. In a similar way, (/?)-r^rr-butanesulfinamide (67) reacts with aldehydes in methylene chloride in the presence of magnesium sulfate to give sulfinimines (R^)-6S in 90-96% yields."*^ With Ti(OEt)4 the reaction has recently been extended to ketones.'*'^^ Heating 2-aminobenzenesulfinamides 69 with ortho esters or the acetal of DMF affords heterocyclic sulfinimines such as 70."*^ Q •^n ,.Sv + RCHO p-MePh^' NH2 (S8)-63
4A molecular selves • orTi{OEt)4
Q II • V V '^.c ^ p-MePh^' N R (^).47
R/Conditions
% Yield
Ph/4A MS PhAri(OEt)4 f-Bu/4A MS f-Bu/Ti(0Et)4
52 99 30 89
O
O
•^M />MePh^' NH2 (S8)-63
TKOEVCHoCU
Me
:
%Yield
Ph f-Bu
O :/,,!! > ^
O H (R') RCH0/MgS04
"NH2 (fl)-67
://.^
T
orRC(0)RVTi(OEt)4' (Hs)-68 100%ee. 77-96%
60% 40%
Syntheses and Reactions of Sulfinlmines
259 H
NH2
1
R1C(OR2)3
•
R^ ^^NH2 S
R-^ S'"
22-87%
O 70
II
o
G. From Halosulfoximides
Sulfoximides, A^- or a-halo, rearrange to sulfmimines on treatment with base."^^"^^ Thus, N-halosulfoximides 71 or a-halosulfoximides 72 react with 1,5-diazabicyclo-[5.4.0]-undec-5-ene (DBU) or K2CO3 to afford sulfinimines 74 in good to excellent yield. The formation of 74 is suggested to occur via rearrangement of an intermediate thiazirine 5-oxide 73.
Ar-S-CHgAr-NX 71
O II
o Ar-S-CHXAr'-J II NH
H I
Ar-S-CHAil N 73
H. From a-Azidosulfoxides
Jarvis et al. found that heating a-azidobenzyl phenyl sulfoxide (75) at 70 °C resulted in the formation of 78 in 33% yield.^^ In addition to the sulfmimine, minor amounts of benzonitrile (15%), benzaldehyde (20%), PhS02SPh, and PhSSPh were detected. A radical mechanism involving radical pairs 76 and 77 is suggested and was supported by an NMR CIDNP effect.
O
O
Ph-V N3 75
O -N2
ecu reflux/1.5h
[ 76
H
Ph-'^- (-"'^
tf 77
78 (33%)
I. From /V-Sulfinylamides
Amidine 79 reacts with A^-sulfmyl p-toluenesulfonamide (80) to give 81 in 94% yield. Subsequent acid hydrolysis in refluxing acetic acid gave the benzothiadizine 82.4«
260
PING ZHOU, BANG-CHI CHEN, and FRANKLIN A. DAVIS H
CL
CI
XJ U '
TsNSO
79
80
94% CI^'^^^^^S^'^ N.^PhMe-p
CI
81 CIv^^^^^N HOAc/reflux 80%
Tl
T
O O
Ph iT
" ci^^^s-^ ^' >o ? 82
A variety of heterocyclic sulfinimines 85 have been prepared by a [4+2] cycloaddition of alkenes 83 with ^-sulfinylurethanes (84, R = Me, Et).^^"^^ The reaction was found to proceed with the retention of the configuration of the alkene. Alkynes 86 react with 84 in a similar manner.^'^^ (A^-Sulfinylaniino)azines 89 cycloadd as heterodienes with 4-epoxy-1,4-dihydronaphthalenes (88, R = H, Me) in refluxing benzene to give trans and cis-exo adducts 90 which differ in their configuration at sulfur.^^ With alkenes 91 (A^-sulfinylamino)azine 92 affords 94 which results from rearomatization of the initially formed sulfmimine cycloadduct 93.^^~^ J. From Sulfonylazides
Himbert reported that ynamine 95 reacts with arylsulfonylazides 96 to give 2-oxo-3-siloxy N-sulfmyl-3-butenamides 98 in fair to good yields.^^'^ The reaction was thought to involve initial formation of diazo compound 97 which gives on elimination of N2 and migration of one of the sulfonyl oxygens, the sulfmimine.
OyR R" 'R-" 83
R^" § O
II
O
84
85
R1
ill 86
R2-
-S'
II
II
o
o 84
87
Syntheses and Reactions
261
ofSulfinimines
90 R^ = H, Me
) 91 a,X = N,Y=:CH b.X = CH,Y = N c, X = Y= CH
R
ill . N
Et. ArS02N3Et
««
N2
-N2
N-Sf Ar
^\
>=0
Et
N-S Ar
97
95 R = Ph3SiO-C=CPh2
•3f'
III. REACTIONS Sulfinimines are multifunctionalized molecules capable of reacting at the sulfinyl and imino groups in a highly regioselective manner (Fig. 1)."^ The latter reactions are of considerable synthetic importance because they can lead to chiral amine derivatives as a consequence of the sulfur stereogenic center. Most chiral imines suffer from low electrophilicity toward organometallic reagents resulting in no
^
o
Stereogenic center Activating group Sulfinimine
R2
Z-M R^^?-N^R3
HgN^^R^
I
H Sulflnamide Figure 1.
Primary amine
262
PING ZHOU, BANG-CHI CHEN, and FRANKLIN A. DAVIS
reaction or predominant side-reactions (i.e., reduction or deprotonation to form aza enolates). Other problems include synlanti isomerization and moisture sensitivity of imines which lead to poor diastereoselectivities and moderate or low yields, respectively. When primary amines are the objective, removal of the A^-auxiliary often causes epimerization or destruction of the product. By contrast, the A^-sulfinyl group in sulfinimines is a superior chiral imine auxiliary because the electron-attracting sulfmyl group activates the C=N bond toward nucleophilic addition permitting reactions to proceed at lower temperatures. The sulfmyl auxiliary also exerts powerful stereodirecting effects, resulting in addition of enolates and organometallic reagents to both enolizable and nonenolizable sulfinimines with high asymmetric induction. Epimerization of newly created carbon stereocenters in the sulfinamide diastereomers is inhibited because the sulfinyl group stabilizes anions at nitrogen. In contrast to aliphatic imines, aliphatic sulfinimines are stable and not particularly susceptible to deprotonation or self-condensation. Moreover, unlike other imine ^V-auxiliaries, the sulfmyl group in the product sulfinamide is easily removed under comparatively mild conditions in addition to being a versatile amine protecting group which can be used for further elaboration of the product. A.
Eliminations
Early investigations by Davis and co-workers demonstrated that arylsulfinimines 39 undergo thermo-elimination to produce sulfenic acids 99, key intermediates in biological transformations, when heated for 24 h at 77-110 °C. These thermolytically generated sulfenic acids can be trapped as silyl sulfenates 101 or vinyl sulfoxides 103.^^'^^ B. Oxidations
Although no report has appeared on the direct oxidation of sulfinimines, studies have indicated that sulfonimines 106 were probably obtained on oxidation of sulfenimines 104 with excess m-CPBA via the intermediate sulfinimines 105.^^'^^ O H ^Sv < ^ Ar- ^ N ^ R
^
f ArSOH 1 + I g^ ]
R-CN ,00
39 TMSCITTMSBNH / A ArSOTMS
\
^=_Q02Et \
102 ^
101
COgEt ) = <
Ar~S,
H 103
Syntheses and Reactions of Sulflnlmines ^
263
excess m-CPBA
O
H
u
I 105
104
107 A/-Sulf onyloxazi ridine
106
Further oxidation of sulfonimines 106 gave A^-sulfonyloxaziridines 107, a versatile class of neutral, aprotic, chiral oxidizing reagents.^^'^^ C. Halogenations Reports on the halogenation of sulfinimines are rare. Reaction of the hexafluoroacetone-derived sulfinimine 108 with chlorine reportedly affords sulfoimidoyl chloride 109 in 58-60% yieldJ"^
F3C
S-Ar
>=N ""3^
CI2
F3C Cl-
\.Ar
108 Ar=Ph, 60% Ar=p-PhMe, 58%
D. Reductions With Hydrides Asymmetric reduction of (55)-sulfmimine 110 with diisobutylaluminum hydride (DIBAL) afforded a diastereomeric mixture of sulfmamides 111 in 92% yield and in a ratio of 96:4.^"^ Use of sodium boron hydride, lithium aluminum hydride, or lithium alkoxyaluminum hydride resulted in lower optical yields.^^'^"^'^^ The sulfmyl group can be removed by treating 111 with trifluoroacetic acid (TFA) and methanol to give a-phenylethyl amine (112). In a similar manner, sulfinimine 113 reacts with DIBAL in THF at room temperature in the presence of zinc bromide to give {R^,R,S)-\1^ and (/^^,/?,/?)-114 in 94% yield and in a ratio of 96:4. In the absence of ZnBr2, {R^,R,R)'\\^ becomes the major product."^^'"*^ Reduction of (/?)-(-)-A^-[l-(triethoxymethyl)ethylidene]-/?-toluenesulfinamide (115), prepared by addition of MeLi to triethoxyacetonitrile followed by the Andersen reagent 40, gave a 95% yield of 116 as a single diastereoisomer on
264
PING ZHOU, BANG-CHI CHEN, and FRANKLIN A. DAVIS '-K^
9*^3
p-MePh^'*^"N
DIBALyTHF/-30°C
Ph
92%
(S,)-110
O H3C ^
O H3C „^
p-MePh^'* "N^^Ph p-MePh^'' "N^'^H H H (S8.S)-111
96:4
(Ss.f?)-111
I TFA/MeOH
HgN^^Ph (S)-112 ioo%ee, 92%
H H
I
ir^t f-BuCONH''^^Me^
""""
f-BuCONH^ ^Me^
(«..«)-113
f-BuCONH
(/=?..aS).114
HH
Me («../?.^-114
Conditions
ratio RS/RR
DIBAL-ZnBr2n"HF/rt DIBAiyTHF/-23°C
Yield
96:4 7:93
94% 98%
treatment with 9-borabicyclo[3,3,l]nonane (9-BBN).^^ Other hydride reagents gave poorer de's. Hydrolysis of the ortho ester on sihca gel followed by removal of the ^-sulfinyl group resulted in formation of (D)-(/?)-alanine ethyl ester (117). With Phosphines
The sulfinyl group in heterocyclic sulfinimine 118 was selectively reduced with 1 equiv of tributylphosphine affording benzothiadiazine 119 in 71% yield."^^ Excess tributylphosphine produced benzothiazole 120 in 58% yield. O
CHq
115
. 9
^^°'''
reflux
r-H
116
(PY N^^^i^^ fr\'' 1 . ^ g , N
9^3
I L ^ g ^ N
"7
^ 58%
r r Vph l l ^ ^ s ' ^
o 118
119
120
Syntheses and Reactions of Sulfinimines
265
E. Michael Additions With Oxygen ISIucleophiles
Sulfinimines are excellent Michael acceptors. For example, hexafluoro acetonederived sulfmimine 62 reacts readily with methanol to give adduct 121 in 90% yield.^0 With Sulfur Nucleophiles
Thiols also react with sulfinimines in a Michael fashion. Thus, thiophenol gave the thiol adduct 122 in 92% yield with sulfmimine 62."^ With Nitrogen Nucleophiles
Amines also react with 62 to give the corresponding amine adducts in good yield."*^ However, with 123 w-butylamine gave the corresponding w-butyl imine 125 in 85% yield.^^ Apparendy, the initially formed amine adduct 124 eliminates benzenesulfmamide (58) with the formation of 125. On the other hand, 123 reacts with phenylhydrazine in ethanol at 50 °C to give the stable hydrazine adduct in 60% yield (124, n-Bu = PhNH).^°
CH3OH
{i
JL^OCHa
(90%)
O II
CF3 I '^
H 121
62
O FqC ii VSPh -S..
PhSH (920/0)
Ph-'
N--CF3
H 122
O
O n
H />BuNH2 EtOH
58 (89%)
Ph'
124
123
^S^ Ph^ NH2
H j^NHBu-n
+
A>BuN
PING ZHOU, BANG-CHI CHEN, and FRANKLIN A. DAVIS
266
With Phosphorus Nucleophiles
Lefebvre and Evans found that lithium diethyl phosphite [(EtO)2POLi] adds to (53)-A^-benzylidene-/7-toluenesulfinaniide (126) to give (53,5)-128 in 85% yield and 84% de7^ The diastereomeric excess was improved to 93% de by using the sodium salt and to >97% with lithium diisopropyl phosphite. Transition state 127, involving 5i-face attack of the nucleophile, was proposed to account for the favored formation of 128. a-Aminobenzyl phosphonic acid 129 was obtained on hydrolysis of 128.^^ Similar results were reported for the addition of diamido phosphite to 126.^^ With Carbon Nucleophiles
Cyanides. As an extension of the Strecker synthesis, addition of cyanide to sulfmimines is expected to give a-amino acids. However, no reaction was observed on treatment of sulfmimines with common cyanide reagents such as KCN or TMSCN.^^ On the other hand, (S^)-47 reacts with diethyl aluminum cyanide (Et2AlCN) to give a-amino nitriles 130 in good yield, but modest de's 36-42%.^^ Significantly, addition of ethyl alkoxy aluminum cyanide [Et(R'0)AlCN], prepared by treatment of Et2AlCN with isopropyl alcohol (R'OH), resulted in dramatic improvement in the diastereoselectivity, e.g., from 36-42% to 82-86%.^^ Simple crystallization of the amino nitriles affords a diastereomerically pure product 130 (>96% de) in good yield. The enhanced de's are attributed to the reduced Lewis acidity of Et(R'O)AlCN versus Et2AlCN which makes it more selective. Acidcatalyzed hydrolysis of 130 not only removes the sulfmyl auxiliary, but also hydrolyzes the nitrile group, affording the enantiomerically pure (>95% ee) aamino acids 131 in good yield. Importantly, racemization of sensitive aryl glycines was not detected. The product stereochemistry is consistent with complexation of H S.
- ^
p-MePh^'' "N
THF/-78°C
Ph
:4
+(RO)2POM
>P(0)(OR)2
p-MePh^'' "N
^^Ph
(Ss)-126 (Ss.SH 28
JI^P(0)(0H)2 129
LI Na Li
127
Et Et Pr-i
Yield
de
85% 80% 82%
84% 93% 97%
Syntheses and Reactions of Sulfinimines
267
EtzAICN
O
H
•4 i^N
36-42%de
T
EtzAICN/l-PrOH (S,)-47
(Sg,S)-130
82-86%de
R = Ph, ^Bu
JuCOgH
6NHa 70-86%
Et
(S)-131
N 132
the aluminum reagent with the sulfmy 1 oxygen activating the imine for intramolecular cyanide addition via a chairlike transition state 132.^^ Davis and Fanelli applied the sulfmimine mediated asymmetric Strecker synthesis to the enantioselective synthesis of the racemization-prone (/?)-(4-methoxy-3,5dihydroxyphenyOglycine (134) from 133.^^ This amino acid is the central amino acid of the clinically important glycopeptide antibiotic vancomycin as well as related antibiotics. Organolithium and Grignard reagents. Sulfinimines also react with organolithium and Grignard reagents to give amine derivatives.^^'-^"*'"*^'^^ For example, reaction of the isobutyl aldehyde-derived sulfmimine {R^)-135 with methylmagnesium bromide gave (R^,S)-136 in 96% de and 99% yield. Acid hydrolysis afforded enantiomerically pure (5)-methyl i-butyl amine (137) in 97% yield."^^ Ethylmagnesium bromide adds to the trifluoropyruvate sulfmimine (S^)'65 to give sulfmamides 138 in 70% yield and in aratio of 73:27."^^ The major diastereomer (S^yR)-13S can be obtained in a diastereomerically pure form by flash chromatog-
:..?
p-MePh^ "N 2 steps
OBn (8s)-133
O
W
><^^N (fl^)-135
MeMgBr^ CH2CI2 96%de, 99%
'•'±
^-s
(^,S)-136
Me
HCl ^ 'iJi^ HCI-htN
H
Me
97% (S)-137
268
PING Z H O U , BANG-CHI CHEN, and FRANKLIN A. DAVIS .^^
902Et
EtMgBr/rHF/-70OC
:,9
J^aEt
H ^^3 (S,.«)-138. ^3.2^
(S^)-65
0
COaEt H
CF3
(S„S)-138
CO2H
(^-139
raphy and further transformed into nonracemic a-trifluoromethyl amino acid (/?)-139.^^ The reaction of allylmagnesium bromide with sulfinimines is of particular importance because it is highly stereoselective.^^'^^ For example, the acetophenone derived sulfinimine (55)-110 and allylmagnesium bromide react to afford 140 in 98% yield as a single isomer. Elaboration of this adduct gave P-amino acid 141.-''^ The remarkably high de was attributed to a chairlike six-membered transition state 242 34,76 sjjniiarly, allylmagnesium bromide reacts with 115 to give 143 as a single diastereoisomer in 95% yield which was transformed into (5)-2-amino-2-methyl4-butenoic acid (144)7^ Davis and Andermichael recently described a new method for the asymmetric synthesis of 3-substituted-l-(2//)-isoquinolones which are important chiral building blocks for alkaloid synthesis.^"' This procedure involved the highly diastereoseO :<"
CH3 I
.^ss^v^MgBr
O H3C
—
p-MePrt''""N^ ^Ph
'^s J^"""-^ -
Et2O/0°C, 2.5 hp-MePrt' (98%)
(^)-iio
"N'^Ph
H3C
Jk^COpH
HgN^^Ph
140
Ph
I
141
n
<> Mg-Br
0'° 142
O
CH3
p-MePrr "N^^C(OEtb 115
CH2=CHChtMgBr Et2O/0''C
•p-MePi/
91%
HoN
95% 144
^
269
Syntheses and Reactions ofSulfinimines
Phv^N^g,\PhMe-p
'^V'^^v'^®
MeO^
^•'-P^'^Q°^
UA^NEt2 ^
p-MePft'* "N Ph (Ss)-126
145
0 146 >97% de. 70% OH
1) TFA 2)^-BuU
MeO^^^^^^>s^Ph T i l
(90%)
iT
^ ^,
MeO.>^ ^'''^^ ^ - " ^ ^ P h
H ^ ^ 1 ^X^As^NMe
^ ^^
0 (S)-147
(3f?,4S)-148
lective addition of the lateral lithiated amide of 145 to {S^)-126 affording sulfinamide 146 in 70% yield. Removal of the sulfmyl group with TFA/MeOH and subsequent treatment of the amine with f-BuLi gave the isoquinolone 147 which was readily transformed into (3/?,45)-(-)-4-hydroxy-3-phenyltetrahydroisoquinoline 148 via hydroxylation and reduction. Enolates. Asymmetric addition of enolates to enantiopure sulfmimines is an important method for the preparation of P-amino esters.^^'^"*'^^ For example, treatment of (55)-sulfmimine 47 with the sodium enolate of methyl acetate in ether afforded P-amino ester 149 in 84-85% yield and in >98% de.^^'^'^ After removal of the N-sulfmyl group, P-amino esters 150 were obtained in >90% yield.^"^ The P-amino esters were further elaborated into the Taxol C-13 side chain 151a,^^ its 85 fluoro analogue 151b,' (+)-2-phenylpiperidine (152a), and (+)-dihydropinidine (152b).^^
O
ONa
H
p-MePrt'*^"N
R
OMe
O
Et20/-78°C
®^**/'*' >98%de p-MePft'
H N
COgMe
(§s.^-149
(^-47 R = Ph.MeCH=CH-
O
X
NH2 O OMe
Ph^^NH O OMe Ph' X
,AA,
151a, X=OH 151b, X=F
(fl)-150
o^M-^c
152a:R=Ph, R' = H 152b:R = n-Pr, R = Me
PING ZHOU, BANG-CHI CHEN, and FRANKLIN A. DAVIS
270
Addition of the lithium enolate of ethyl acetate to sulfmimine 153 gave 154 with a diastereomeric ratio (d.r.) of 89:11 .^^ Separation of the diastereoisomers by flash chromatography and deprotection with TFA/EtOH afforded the P-amino ester 155 in >97% ee and 68% overall yield. (5)-Ethyl P-amino-3-pyridinepropanoate (155) is a key component of the peptidomimetic 156 for the Arg-Gly-Asp-Phe sequence of fibrinogen. The diastereoselectivity for the reaction of 157, which possesses a 2-methyl-l,3dioxolanyl group, with enolates generated from tert-buty\ acetate was found to be highly dependent on the reaction conditions.^^ For example, the lithium enolate gave (53,5)-158 (72% de) while the titanium enolate afforded (5,,/?)-158 (92% de). A non-chelation-controlled transition state was used to explain the preferential formation of (53,5)-158 while a six-membered chairlike transition state containing a four-membered metallocycle and/or a seven-membered counterpart was proposed for the formation of the (S^,R)-15S. Treatment of (55,5)-158 with TEA gave P-amino acid 159 in 70% yield. Enantiopure bis-P-amino acids can be prepared from chiral bis-sulfinimines.^^ Bis-sulfmimine (S^,S^)-160 and the sodium enolate of methyl acetate react to give 161 as a diastereomeric mixture. The major isomer (S^,R,S^,R)-161 can be isolated by preparative reverse-phase HPLC in 46% yield. Hydrolysis of (S^,R,S^,R)-161 gave bis-P-amino ester (/?,/?)-162 in >97% ee and 86% yield.^"^ Dienolates. Garcia Ruano and co-workers reported that A^-2-methoxynaphthyl sulfmimine (S^yi63 reacts with the lithium dienolate of 3-butenoic methyl ester (164) to afford a-ethylidene-P-sulfmylamino ester 165 as a single isomer in 82% yield.^ In the presence of ZnBr2, a-vinyl-P-sulfmylamino esters 166 were obtained in 90% as a diastereomeric mixture in a ratio of 30:70. Both (S^y2S,3R)166/(S^y2R,3R)-166 can be converted to the same a-ethylidene-P-amino ester 168 via deprotection of the A^-sulfmyl group and subsequent base promoted epimerization of the a-chiral center.^
CH3C02Et/LiHMDS
{Ss.R)^^5A
NH2
.-H
TFA
COoEt
EtOH >97%ee. 68%
155
NH 156
271
Syntheses and Reactions of Sulfinimines NH2
159 70% TFA COgBu-f
. O THF/HMPA M=LJ
^9
Y
,^
„^v
^
OM
HJ " iL O
(38.S)-158 96%de, 68%
p-MePrt
I
L^ (Ss)-157
O
O
THF M=Ti(0Pr-/)2
p-MePrt'' " N
C02Bu-f
HJ
^ X l
(Ss.f^-158 92%de. 89%
O *^S-N p-MePK
\__-^^\_-/ \ = /
o
CH3C02Me/NaHMDS
N-S-"i:
THF/-78°C/7h
(§5.Ss)-160 ^C02Me
Me02C>,
:-S-NH PhMe-d^MePh'
:-S-NH p-MePh'' (^.^,^.^-161
70:30 46%
i
H2N
TFA/MeOH 0°C/4h
\ = / (afl)-i62 >97%ee, 86%
NH2
^ = /
HN-S"': PhMe-p {Ss,R,Ss,S)-^B^
272
PING Z H O U , BANG-CHI CHEN, and FRANKLIN A. DAVIS O H
Y^
N
OMe
Ph
164
OMe (S5)-163 LDAn"HF/-78°C
[ LDA/ZnBr2rrHF/-78°C
82%
O
Ph
O
O
OMe OMe Me (Ss,3R)-165
OMe
30:70
(Ss,2S.3/?)-166
TFA/0°C
Ph
O
CFgCOzH-HzN^^Y^OMe
Ph
O
H i OMe "^
OMe
{Ss.2R,3R)-^66
1
TFA/0°C
Ph
O
CF3C02H.H2N'^S^OMe
(2S.3/?)-167
Oxime carbanions. Addition of the dianion generated from oxime 169a and rt-BuLi to 170 gives 171a in 69% yield as a mixture of diastereomers.^^ The anion generated from the corresponding oxime methyl ether reacts similarly to give 171b, but the yield was only 34%.^^ a-Sulfonyl carbanions. Balasubramanian and Hassner described a short stereoselective synthesis of the alkaloid (5)-anatabine (175) by treatment of the lithium dianion of 4-phenylsulfonyl cis- but-2-en-l-ol (172) with (5,)-173.^^ Four diastereomers were produced and the major product, 174, was isolated by flash column chromatography and recrystallization in 45-57% yield and >95% de. Elaboration of 174 gave (5)-175. a-Phosphonate carbanions. Nonracemic P-aminophosphonic acids can be prepared in high enantiomeric purity by addition of a-phosphonate carbanions to
Syntheses and Reactions of
Sulfinimines
273 ..OR
1. n-BuLi
Jl
2.
O
p-MeOPh'
H
171a, R=H b, R=Me
170
169a. R=H b, R=Me
Ph
p-MeOPh'
N ' "PhOMe-p
Ph'
OH 1.LiHMDS/-78°C/0.5h
O 0=S.p^
p-MePh^
II ^V^
H Ar
N
S p-MePh'^'* "N
(Ss)-173 172
174
-100to-60°C/1.5h
Q
^ I
{MeO)2P(0)Me LiHMDS/THF
O
1^ P(0)(OMe^
52% (^,fl).176
(Ss)-126
Q
^ P(0)(OMet
M
TFA/MeOH 3h,rt ^ / " ^
H 10.3:1
(Ss.S)-176
aq. HCI/HOAc
78% Y ' NH2 O
J^P^-OMe Ph' OMe (-)-(«)-177
NH2 O
.,A^P,-OH "
OH (+).(fi)-178
sulfinimines. Mikolajczyk et al. prepared A^-sulfinyl P-aminophosphonate 176 from (5)-126 and the lithium anion of diethyl methanephosphonate with a d.r. of 10.3:1.^^'^"^ The major diastereomer (S^,R)-176, after separation, can be converted to the corresponding P-aminophosphonate 177 or to (+)-p-amino-P-phenylethane phosphonic acid 178, whose absolute configuration was established as (R) by X-ray crystallography. The preferential formation of {S^,Ryil6 was rationalized assuming nucleophilic attack of a-phosphonate carbanion on the S-cis conformation ( S = 0 and C = N syn coplanar) of 126 and is anti to the large jp-tolyl sulfinyl group. F. [1 + 2 ] Cycloadditions With Sulfur
Ylides
A^-Sulfinylaziridines 181 are formed as mixtures of diastereoisomers from (5^)179 and dimethyloxosulfonium methylide (180) 38,95.96 The diastereoselectivity can
274
PING Z H O U , BANG-CHI CHEN, and FRANKLIN A. DAVIS O is."
O
H I H3C
(Ss)-179
O
y.ph
Ph^
CH2 (SsS).181
180
(Ssfl)-181
R
Yleld%
Ratio S,S/S,R
p-MePh t-Bu
62 85
75:25 95:5
be improved by using r^rr-butylsulfinyl as the chiral directing group.^^'^^ Sulfonium ylides react similarly.^^^^ With a-Haloenolates The Darzens' type reaction between sulfinimines and the a-haloenolates is a versatile method for preparing aziridines with diverse ring and nitrogen substituents."* For example, the lithium enolate of a-bromoacetate and enantiopure sulfinimines 47 react to give the corresponding cis A^-sulfinylaziridine-2-carboxylic esters (S^,S,S)-1S2 in 94-98% de and 60-74% yield.^^'^^ A chairlike transition state 184 was used to explain the cis selectivity. Highly regio- and stereoselective ring-opening methodology was employed in efficient asymmetric syntheses of 5>'AZ-P-phenylserine (185),^^ and the antibiotic thiamphenicol (186)^^ft-omthese types of aziridines. QLI UL
O H :
THF
OMe Br
(^-47 R = Ph, p-MeOPh, /-Pr
•^ f>MePh'
p-MePrf
'C02Me H (S^.S,S)-182
OH
K^
NH2 185
H3C
(^,S,f^-183
Syntheses and Reactions ofSulfinimines
275
As an extension of this chemistry, A^-sulfmyl aziridine 188, prepared from (R)-1S7, was utilized in the asymmetric synthesis of protein kinase C inhibitor D-erythro-sphingosine 189^^ and in the first enantioselective synthesis of the marine cytotoxic antibiotic (/?)-(-)-dysidazirine (190).^^ This latter result constitutes the first general method for preparing nonracemic 2//-azirines, the smallest of the unsaturated nitrogen heterocycles.^^'^^^ Using the lithium enolate of a-bromopropionate, trans and cis A^-(p-toluenesulfinyl)-2-methyl-2-carbomethoxy aziridines 191 were prepared from (5)-47.^^^~^^^ The ^-isomers predominate. Regio- and stereoselective aziridine ring-opening leads to efficient asymmetric syntheses of (-)-a-methylphenylalanine (192),^^^ (+)-a-methyl-P-phenylserine (193),^^^ (+)-2-methyl-3-phenylpropanoic acid (194a),^^^ and (-)-2-methyl-3-aminopentanoic acid (194b).^^^ With a-Halophosphonate Carbanions Addition of the lithium anion of chloromethylphosphonate to sulfinimine 126 gave a-chloro-P-aminophosphonates 195 in a ratio of 59:41 and 98% total yield.^^ The diastereomeric products can be separated and each converted to the corresponding aziridine-2-phosphonates 196, new building chiral blocks for the enantioselective synthesis of a-aminophosphonates 197 and azirinyl phosphonates 198.^^ G. [3+2] Cycloadditions With Allylsulfones Balasubramanian and Hassner described that the [3+2] cycloaddition of the allylsulfone anion of 199 and (5)-126 gives 2-aryl-3-pyrroline derivatives 200 in 70% yield and a ratio of 88:12.^^^ Separation of the diastereomers led to optically
R
O •''•Q
H JL
^
D
BrCHgCOgMe 67%
Q p-MePh^
(^)-187 R = n-Ci2H25
H 188
I
OH OH
N
NH2
^ COaMe
189 31% 190 42%
276
PING ZHOU, BANG-CHI CHEN, and FRANKLIN A. DAVIS O
H
f
THF
R = Ph,Et O
H 'COaMe Me
.•Me C02Me
Z-(^,S.S)-191
E-(S^,aS)-191
E/Z(% yield) R = Ph95:5(84%) R = Et90:10(76%)
NH2
OH C02Me PhMe^
NH2
Mi
p-MePrt'
l^e
NH2
(+)-194a:R = Ph (-)-194b:R = Et
(-)-193
(+).192
O
pA^C02H
.C02Me
O
(EtO)2P(0)ChfeCI/ . 9 LIHMDS/rHF ,.S^ - ^ p-MePtt' NH ^ P h -78^C/30min
H
"N
+ p-MePrt''^"NH
p^A^P(0)(OEt^
p^A!/P(0)(OEtt
98% CI {Ss,R.R)-^9S
(§5)-126
CI SI
.0... 59:41
(^,R,S)-195
I NaH Ph
o
°
NHg
p-MePft^ =-P(0)(OEtt (^,a/=l)-196
{Ss,R,S)-^96
(S)-198
76%
75%
9.0 o p n
1.LDA. .1000c
Ph^ -^H
P((5)(0Ett
(OEtt
(Ri^^97
O u
?!:H
r^^NCT p-MePrt"
hoEXh
P h ^ ^
I NaH
Ph-S;; ^ W
9.P Me
Ph-S(
Me W
Ph*
^MePrt^'^^N^Ph ^^.p^s:;;; 199
(^)-126
^MePtr^^o
(5B.'^-200
(S8.S)-200
88:12 (70%)
Syntheses and Reactions of Sulfinimines
277
Ph O
i?
H
p-MePrt''^"N^^Ar
^^__/> a „
THF/20h .78to4°C
.
0=\ (^)-173
pi. y ^
NH
NT i'v,R Me02C
OMe 201
.
202
Ar=Ph.p-N02Ph R = CHaPh, Me
%de 55-80%, 90-96®/
/
Ph^
KOH ^Ph 203
pure 2(/?)-pyrrolines 200 which were desulfonated with Na(Hg) to give the enantiomeric enriched 3-arylpyrrole derivative. With Azomethine Ylides
Azomethine ylides 201 undergo 1,3-dipolar cycloadditions with sulfinimines such as 173 to give tetrahydroimidazoles 202 in 55-80% yields and 90-96% de.^^ Reduction of 202 (R = PhCH2-) with LAH followed by hydrolysis afforded the diamino alcohol 203 in good yield. H. Hetero Diels-Alder Reactions Intermolecular
Tietze and Schuffenhauer explored the intermolecular hetero Diels-Alder reaction of ethyl vinyl ether and sulfinimine 204 and found that at high pressures (11 kbar) and long reaction times (48 h at room temperature), tetrahydropyridines 205 and 206 were obtained in 96% yield and in a ratio of 1.7:1 .^^^ The ^jco-adducts were not observed. Phenyl vinyl sulfide reacts similarly, but simple alkenes failed. Theoretical calculations and the experimental results suggest that while the cycloaddition is concerted, it is highly asynchronous. Ph
Ph
Ph
CN CH2Cl2/rt/48h
Excr
11 kbar I
p-MePhrf'^O 204
96%
p-MePH^f^O 205
p-MePrrf'^O 1.7:1
206
278
PING ZHOU, BANG-CHI CHEN, and FRANKLIN A. DAVIS
CN
1.MeLi/-78°C 2. AcCI or Mel
209a, R=Ac, 56% b, R=Me, 60%
Intramolecular
The intramolecular Diels-Alder reaction of 207 gave 208 as a niixture of four diastereomers.^^^ The products can be transformed into A^-acetyl or A^-methyl derivatives 209 by treating 208 with methyllithium followed by reaction with acetyl chloride or methyliodide.
IV. CONCLUSIONS The A/'-sulfmyl group in sulfmimines is a superior imine auxiliary because it activates the C=N bond for addition, is highly stereodirecting, and is easily removed in the product sulfmamide. The converse is true for most other A^-imine auxiliaries. Furthermore, the sulfinyl group in the product is a useful amineprotecting group and can be used for further elaboration of the product. The chemistry of sulfinimines outlined herein demonstrates that sulfinimines are useful chiral building blocks for the asymmetric synthesis of amine derivatives. In addition, new sulfinimine-derived chiral building blocks, including ^-sulfinyl aziridine 2-carboxylates, 2//-azirine carboxylates, and isoquinolones are expected to play important roles in the asymmetric synthesis of novel biologically active a- and P-amino acids and alkaloids. It is hoped that this chapter will stimulate continued interest in the application of sulfmimines (thiooxime 5-oxides, A^-sulfmyl imines) for the synthesis of amine derivatives.
ACKNOWLEDGMENTS It is a pleasure to acknowledge the important efforts of our co-workers whose names appear in the references. Our own contributions to this chapter were supported by the National Science Foundation, the National Institutes of Health, and the Petroleum Research Fund administrated by the American Chemical Society.
Syntheses and Reactions of Sulfinimines
279
NOTE ADDED IN PROOF Since submitting this Chapter a number of papers have appeared which are particularly relevant to the subject. Ellman has provided additional details of the condensation of rerr-butanesulfmamide (67) with aldehydes and ketones. ^^^ Aldehyde and ketone derived camphor sulfenimines are oxidized with m-CPBA to give good yields of the corresponding sulfinimines.^^ A one-pot method for the asymmetric synthesis of rerr-butanesulfinyl-protected amines has been described. ^^^ This procedure involves condensation of 67 with ketones followed by the in situ reduction of the intermediate sulfinimines with NaBH^ (66-86%, drs 90:10 to 97:3). The sulfinimine mediated asymmetric Strecker synthesis has been employed in syntheses of P-fluoro a-amino acids^^^ and (2/?,35)-alloisoleucine*^^ via the addition of [EtAl(0-/-Pr)CN] to the sulfinimines derived from chiral a-fluoro aldehydes and (5)-(+)-2-methylbutanal, respectively. Davis and McCoull described the asymmetric synthesis of a-amino acids by treatment of Grignard reagents with ethyl (/?)-(-)-A^-(r^rf-butanesulfinyl)iminoacetate, derived from 67 and ethyl glyoxylate, in the presence of BF30Et2 (drs 83:17 to 99:1).^^^ Li, in a series of papers, reported the addition of lithium (a-carbalkoxyvinyl)cuprates to chiral p-toluenesulfinimines 47 to afford P-mono and P,P-disubstituted a-(aminoalkyl)acrylates Baylis-Hillman adducts.^^"*"^^^ Titanium enolates add to r^rr-butanesulfinimines of aldehydes and ketones to give P-amino acids in good yield and high de's.^^^ (+)-(/?)-3-Amino-3phenyl propanoic acid (L-P-phenylalanine), prepared by addition of the sodium enolate of methyl acetate to sulfinimine 47, was used in the first total synthesis of astin G, a cyclopentapeptide.^^^ Full details of the one-step aza-Darzens reaction of sulfinimines 47 with lithium a-bromoenolates to give diversely substituted cisand /ran^-N-sulfinylaziridine 2-carboxylate esters has appeared.^^^ This paper discloses a new method for the selective removal of the N-sulfinyl group with Grignard reagents affording the l//-aziridines in good to excellent yields. In a related aza-Darzens reaction Davis and McCoull prepared enantiopure N-sulfinylaziridine 2-phosphonates from 47 and chloromethylphosphonate anions. ^^^'^^^ Regioselective phase transfer hydrogenolysis of these aziridines resulted in the symmetric synthesis of a-amino phosphonates in excellent yield and ee (>99%). The first enantioselective synthesis of an azirinyl phosphonate was accomplished by Swern oxidation of l//-aziridine 2-phosphonates.^^^ The Lewis acid promoted addition of glycine iminoester enolates to /?-toluenesulfinimines 47 to give A^-sulfinyimidazolidines with good selectivity has been reported. ^^^ Reduction of the imidazolidine adducts gave vicinal diamines.
REFERENCES 1. Claus, P. K. In The Chemistry ofSulfenic Acids and Their Derivatives^ Patai, S., Ed.; John Wiley & Sons: New York, 1990, pp. 723-741.
280
PING ZHOU, BANG-CHI CHEN, and FRANKLIN A. DAVIS
2. Tillett, J. G. In The Chemistry of Sulfinic Acids, Esters and Their Derivatives, Patai, S., Ed.; John Wiley & Sons: New York, 1990, pp. 603-622. 3. Craine, L.; Raban, M. Chem. Rev. 1989, S9, 689. 4. Davis, F. A.; Zhou, R; Chen, B.-C. Chem. Soc. Rev. 1998. 27, 13. 5. Davis, F. A.; Portonovo, R S.; Reddy, R. E. Phosphorus, Sulfur Silicon Re lat. Elem. 1997,120-121, 291. 6. Davis, F. A.; Reddy, R. E. In Enantioselective Synthesis of ^-Amino Acids; Juaristi, E., Ed.; Wiley-VCH: New York, 1997, pp. 127-138. 7. Friedman, A. J. J. Agric. Food Chem. 1983, 31, 127. 8. Friedman, A. J.; Hopfinger, A. J. J. Agric. Food Chem. 1983, 31, 135. 9. Cipollina, J. A.; Ruediger, E. H.; New, J. S.; Wire, M. E.; Shepherd, T. A.; Smith, D. W; Yevich, J. R J. Med Chem. 1991, 34, 3316. 10. Davis, F A.; Friedman, A. J.; Kluger, E. W /. Am. Chem. Soc. Wl^, 96, 5000. 11. Almog, J.; Barton, D. J. R.; Magnus, R D.; Norris, R. K. J. Chem. Soc, Perkin Trans. 11974, 853. 12. Davis, F A.; Kluger, E. W. J. Am. Chem. Soc. 1976, 98, 302. 13. Knorr, R.; Ferchland, K.; Mehlstaubl, J.; Hoang, T. R; Bohrer, R; Ludemann, H.-D.; Lang, E. Chem. Ben 1992,125,2041. 14. Shermolovich, Y. G.; Talanov, V. S.; Dolenko, G. N.; Markovskii, L. N. Zh. Org. Khim. 1980,16, 964; Chem. Abstn 1980, 93, 167797. 15. Shermolovich, Y. G.; Talanov, V. S.; Dolenko, G. N.; Markovskii, L. N. J. Org. Chem. SSSR1980, 843. 16. Schaumann, E.; Bolte, O.; Behr, H. J. Chem. Soc, Perkin Trans. 11990, 182. 17. Oae, S.; Shinhama, K.; Fujimori, K.; Kim, Y. H. Bull. Chem. Soc Jpn. 1980, 53, 775. 18. Gordon, E. M.; Chang, H. W; Cimarusti, C. M.; Toeplitz, B.; Gougoutas, J. Z. / Am. Chem. Soc 1980,102, 1690. 19. Gordon, E. W; Pluscec, J. J. Org. Chem. 1979,44, 1218. 20. Yang, T.-K.; Chen, R.-Y; Lee, D.-S.; Peng, W.-S.; Jiang, Y.-Z.; Mi, A.-Q.; Jong,T.-T. J. Org. Chem. 1994,59,914. 21. Davis, F A.; Reddy, R. T; Reddy, R. E. J. Org. Chem. 1992, 57, 6387. 22. Davis, F A.; Reddy, R. T.; Han, W; Reddy, R. E. Pure Appl. Chem. 1993, 65, 633. 23. Davis, F A.; Reddy, R. E.; Szewczyk, J. M.; Reddy, G. V.; Portonovo, R S.; Zhang, H.; Fanelli, D.; Reddy, R. T.; Zhou, R; Carroll, R J. J. Org. Chem. 1997, 62, 2555. 24. Walsh, R. J. A.; Wooldridge, K. R. H. J. Chem. Soc, Perkin Trans. 1 1972, 1247. 25. Bruno, A.; Purrello, G. Gazz. Chim. Ital. 1966, 96, 1009. 26. Nishiwaki, T.; Kawamura, E.; Abe, N.; lori, M. J. Chem. Soc, Perkin Trans. 11980, 2693. 27. Boeshagen, H.; Geiger, W; Medenwald, H. Chem. Ber. 1970,103, 3166. 28. Burgess, E. M.; Penton, H. R., Jr. J. Am.Chem. Soc. 1973, 95, 279. 29. Burgess, E. M.; Penton, H. R., h.J.Org. Chem. 1974, 39, 2885. 30. Mews, R.; Kricke, R; Stahl, I. Z. Naturforsch. 1981,36b, 1093. 31. Reid, W; Dietschmann, H.; Erie, H.-E. Synthesis 1980, 619. 32. Zhou, R Ph.D. dissertation, Drexel University, Philadelphia, 1994. 33. Annunziata, R.; Cinquini, M.; Cozzi, F J. Chem. Soc, Perkin Trans. 1 1982, 339. 34. Hua, D. H.; Miao, S. W; Chen, J. S.; Iguchi, S. / Org. Chem. 1991, 56, 4. 35. Cinquini, M.; Cozzi, F J. Chem. Soc, Chem. Commun. 1977, 502. 36. Davis, F A.; Reddy, R. E.; Szewczyk, J. M.; Portonovo, R S. Tetrahedron Lett. 1993, 34, 6229. 37. Adamczyk, M.; Reddy, R. E. Tetrahedron: Asymmetry 1998, 9, 3919. 38. Garcia Ruano, J. L. G.; Fernandez, I.; Catalina, M. P.; Cruz, A. A. Tetrahedron: Asymmetry 1996, 7, 3407. 39. Cogan, D. A.; Liu, G.; Kim, K.; Backes, B. J.; Ellman, J. A. J. Am. Chem. Soc 1998,120, 8011. 40. Hose, D. R. J.; Raynham, T.; Wills, M. Tetrahedron: Asymmetry 1993, 4, 2159. 41. Hose, D. R. J.; Wills, M.; Raynham, T. Tetrahedron Lett. 1994, 35, 5303.
Syntheses and Reactions of Sulfinimines
281
42. Hose, D. R. J.; Mahon, M. F ; Molloy, K. C ; Raynham, T.; Wills, M. J. Chem. Soc, Perkin Trans. 7 1996,691. 43. Oppolzer, W.; Froelich, O.; Wiaux-Zamar, C ; Bemardinelli, G. Tetrahedron Lett. 1997,38, 2825. 44. Burger, K.; Albanbauer, J.; Kafig, P.; Penninger, S. Liebigs Ann. Chem. 1977, 624. 45. Bravo, R; Crucianelli, M.; Vergani, B.; Zanda, M. Tetrahedron Lett. 1998, 39,1111. 46. Davis, F A.; Zhang, Y; Andemichael, Y; Fang, T; Fanelli, D. L.; Zhang, H. / Org. Chem. 1999, 64, 1403. 47. (a) Liu, G.; Cogan, D. A.; Ellman, J. A. J. Am. Chem. Soc. 1997, 119, 9913. (b) Cogan, D. A.; Ellman, J. A. J. Am. Chem. Soc. 1999,121, 268. 48. Finch, N.; Ricca, S., Jr.; Werner, L. H. J. Org. Chem. 1980,45, 3416. 49. Yoshida, T.; Naruto, S.; Uno, H.; Nishimura, H. J. Chem. Soc, Chem. Comm. 1982, 106. 50. Yoshida, T.; Naruto, S.; Uno, H.; Nishimura, H. Chem. Pharm. Bull. 1982, 30, 2820. 51. Yoshida, T.; Naruto, S.; Uno, H.; Nishimura, H. Chem. Pharm. Bull. 1982, 30, 4346. 52. Jarvis, B. B.; Nicolas, R E.; Midiwo, J. O. J. Am. Chem. Soc. 1981,103, 3878. 53. Hoerhold, H.-H.; Eibisch, H. Chem. Ber. 1968,101, 3567. 54. Hoerhold, H.-H. Angew. Chem. Int. Ed Engl. 1967, 6, 357. 55. Hoerhold, H.-H. Angew. Chem. 1967, 79, 312. 56. Scherer, O. J.; Schmitt, R. Chem. Ber. 1968,101, 3302. 57. Carpanelli, C ; Gaiani, G.; Sancassan, F Gazz. Chim. Ital. 1982,112, 469. 58. Carpanelli, C ; Gaiani, G.; Sancassan, F Gazz. Chim. Ital. 1984,114, 399. 59. Carpanelli, C ; Gaiani, G.; Sancassan, F Gazz. Chim. /to/. 1985,115, 265. 60. Himbert, G.; Pfeifer, K.-R; Finkele, C. E. Phosphorus, Sulfur Silicon Relat. Elem. 1991, 59, 125. 61. Pfeifer, K.-R; Himbert, G. Tetrahedron Lett. 1990, 31, 5725. 62. Hanson, R; Wren, S. A. C. J. Chem. Soc, Perkin Trans. 2 1987, 197. 63. Hanson, R; Wren, S. A. C. J. Chem. Soc, Perkin Trans. 1 1990, 2089. 64. Beechen, H. Chem. Ber. 1967,100, 2159. 65. Frank, D.; Himbert, G.; Regitz, M. Chem. Ber. 1978, 111, 183. 66. Himbert, G. Liebigs Ann. Chem. 1979, 1828. 67. Davis, F A.; Friedman, A. J.; Nadir, U. K. / Am. Chem. Soc 1978,100, 2844. 68. Davis, F A.; Friedman, A. J. J. Org. Chem. 1976,41, 897. 69. Davis, F A.; Rizvi, S. Q. A.; Ardecky, R.; Gosciniak, D. J.; Friedman, A. J.; Yocklovich, S. G. J. Org. Chem. 19S0,45, \650. 70. Davis, F A.; Kaminski, J. M.; Kluger, E. W; Freilich, H. S. J. Am. Chem. Soc 1975, 97, 7085. 71. Davis, F A.;Lamendola,J.,Jr.;Nadir, U.; Kluger,E.W.; Sedergran,T. C ; Panunto,T. W; Billmers, R.; Jenkins, R.. Jr.; Turchi, I. J.; Watson, W. H.; Chen, J. S.; Kimura, M. J. Am. Chem. Soc 1980, 102, 2000. 72. Davis, F A.; Sheppard, A. Tetrahedron 1989,45, 5703. 73. Davis, F A.; Chen, B.-C. Chem. Rev. 1992, 92, 919. 74. Markovskii, L. N.; Talanov, V. S.; Shermolovich, Y G. J. Org. Chem. USSR (Engl. Transl.) 1984, 20, 316; Zh. Org. Khim. 1984, 20, 353. 75. Cinquini, M.; Cozzi, F J. Chem. Soc, Chem. Commun. 1977, 723. 76. Hua, D. H.; Lagneau, N.; Wang, H.; Chen. J. Tetrahedron: Asymmetry 1995, 6, 349. 77. Lefebvre, I. M.; Evans, S. A., Jr. J. Org. Chem. 1997, 62, 7532. 78. Mikolajczyk, M.; Lyzwa, P.; Drabowicz, J. Tetrahedron: Asymmetry 1997, 8, 3991. 79. Davis, F A.; Reddy, R. E.; Portonovo, R Tetrahedron Lett. 1994, 35,9351. 80. Davis, F A.; Portonovo, R; Reddy, R. E.; Chiu, Y-H. J. Org. Chem. 1996, 61, 440. 81. Davis, F A.; Fanelli, D. L. J. Org. Chem. 1998, 63, 1981. 82. Moreau, R; Essiz, M.; Merour, J.-Y; Bouzard, D. Tetrahedron: Asymmetry 1997, S, 591. 83. Davis, F A.; Andermichael, Y W. Tetrahedron Lett. 1998, 39, 3099. 84. Jiang, J.; Schumacher, K. K.; Joullie, M. M.; Davis, F A.; Reddy, R. E. Tetrahedron Lett. 1994, 55,2121.
282
PING ZHOU, BANG-CHI CHEN, and FRANKLIN A. DAVIS
85. 86. 87. 88. 89. 90.
Davis, F. A.; Reddy, R. E. Tetrahedron: Asymmetry 1994, 5, 955. Davis, F. A.; Reddy, R. E.; Szewczyk, J. M. J. Org. Chem. 1995, 60,1031. Davis, R A.; Szewczyk, J. M. Tetrahedrvn Lett. 1998, 59, 5951. Davis, F A.; Szewczyk, J. M.; Reddy, R. E. J. Org. Chem. 1996, (57, 2222. Fujisawa, T.; Kooriyama, Y.; Shimizu, M. Tetrahedrvn Lett. 1996, 37, 3881. Ruano, J. L. G.; Fernandez, I.; Catalina, M. R; Hermoso, J. A.; Sanz-Aparicio, J.; Martinez-Ripoll, M. J. Org. Chem. 1998, 63,1151. Kaiser, A.; Wiegrebe, W. Monatsh. Chem. 1996,127, 397. Balasubramanian, T.; Hassner, A. Tetrahedron: Asymmetry 1998, 9, 2201. Mikolajczyk, M.; Lyzwa, P.; Drabowicz, J.; Wieczorek, M. W.; Blaszczyk, J. J. Chem. Soc, Chem. Commun. 1996, 1503. Mikolajczyk, M.; Lyzwa, R; Drabowicz, J. Phosphorus, Sulfur Silicon Relat. Elem. 1997, 120-121, 357. Davis, R A.; Zhou, R; Liang, C.-H.; Reddy, R. E. Tetrahedrvn: Asymmetry 1995, 6, 1514. Garcia Ruano, J. L. G.; Fernandez, I.; Hamdouchi, C. Tetrahedron Lett. 1995, 36, 295. Davis, F A.; Zhou, R; Reddy, V. G. J. Org. Chem. 1994,58, 3243 Davis, F A.; Zhou, R Tetrahedron Lett. 1994, 35, 7525. Davis, F A.; Reddy, G. V. Tetrahedrvn Lett. 1996, 37,4349. Davis, F A.; Reddy, G. V.; Liu, H. / Am. Chem. Soc. 1995,117, 3651. Davis, F A.; Liang, C.-H.; Liu, H. J. Org. Chem. 1997, 62, 3796. Davis, F A.; Liu, H.; Reddy, G. V. Tetrahedron Lett. 1996, 37, 5473. Davis, F A.; Reddy, G. V.; Liang, C.-H. Tetrahedron Utt. 1997, 38, 5139. Davis, F A.; McCouU, W. Tetrahedron Lett. 1999,40, 249. Balasubramanian, T.; Hassner, A. Tetrahedron Lett. 1996, 37, 5755. Viso, A.; Pradilla, R. F ; Guerrero-Strachan, C ; Alonso, M.; Martinez-Ripoll, M.; Andre, I. J. Org. Chem. 1991,62,2316. Tietze, L. F ; Schuffenhauer, A. Eun J. Org. Chem. 1998, 1629. Liu, G.; Cogan, D. A.; Owens, T. D.; Tang, T. R; Ellman, J. A. J. Org. Chem. 1999, 64, 1278. Li, Y; Zhu, W.; Cheng, Z ; Yang, G. Syn. Commun. 1999, 29, 6245. Borg, G.; Cogan, D. A.; Ellman. J. A. Tetrahedron Lett. 1999,40, 6709. Davis, F A.; Srirajan, V.; Titus, D. D. / Org. Chem. 1999, 64, 6931. Portonovo, P; Liang, B.; Joullie, M. M. Tetrahedron: Asymmetry, 1999,10, 1451. Davis, F A.; McCouU, W. J. Org. Chem. 1999,64, 3396. Li, G.; Wei, H-X.; Whittlesey, B. R.; Batrice, N. N. / Org. Chem. 1999, 64, 1061. Li, G.; Wei, H-X.; Hook, J. D. Tetrahedron Utt. 1999,40,4611. Wei, H-X.; Hook, J. D.; Fitzgerald, K. A.; Li, G. Tetrahedron: Asymmetry, 1999,10, 661. Tang, T P; Ellman, J. A. J. Org. Chem. 1999, 64, 12. Schumacher, K. K.; Hauze, D. B.; Jiang, J.; Szewczyk, J.; Reddy, R. E.; Davis, F A.; Joullie, M. M. Tetrahedron Utt. 1999,40, 455. Davis, F A.; Liu, H.; Zhou, P; Fang, T; Reddy, G. V.; Zhang, Y / Org. Chem. 1999, 64, 7559. Davis, F A.; McCoull, W. Tetrahedron Utt. 1999,40, 249. Davis, F A.; McCoull, W; Titus, D. D. Org. Utt. 1999, 7, 1053. Viso, A.; Femadez de La Pradilla, R.; Garcia, A.; Alonso, M.; Guerrero-Strachan, C ; Fonseca, I. 5yn/ett 1999,1543.
91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122.
CHIRAL SULFOXIMINES FOR DIASTEREOSELECTIVE AND ASYMMETRIC SYNTHESIS
Stephen G. Pyne
I. Introduction 11. Alkyl Sulfoximines A. Synthesis B. Lithiated Sulfoximines C. Other Reactions III. Applications of P-Hydroxy Sulfoximines to Asymmetric Synthesis A. Resolution of Racemic Chiral Cyclic Ketones B. Synthesis of Alkenes via Reductive Elimination C. Asymmetric Synthesis of Alkenes with Axial Chirality D. Directed Simmons-Smith Cyclopropanations E. Directed Osmylations F. Enantioselective Reactions IV. Allylic Sulfoximines A. Synthesis B. Lithiated Allylic Sulfoximines C. Rearrangements to Allylic Sulfmamides and Related Reactions
Advances in Sulfur Chemistry Volume 2y pages 283-366. Copyright © 2000 by JAI Press Inc. All rights of reproduction in any form reserved. ISBN: 0-7623-0618-1 283
284 284 284 288 313 313 313 314 314 316 316 316 317 317 318 326
284
STEPHEN G. PYNE
D. Ally lie Substitution Reactions with Organometallic Reagents V Vinyl Sulfoximines A. Synthesis B. Michael Reactions C. Cross-Coupling Reactions D. Cycloaddition Reactions VI. a-Sulfonimidoyl Ketones and Esters A. Synthesis B. Reactions VII. Sulfoximines as Ligands for Asymmetric Synthesis VIII. Conclusions Acknowledgments References
337 338 338 340 350 352 355 355 356 359 362 . 363 363
r. INTRODUCTION This chapter reviews the applications of chiral sulfoximines to asymmetric and diastereoselective synthesis and as ligands in catalytic asymmetric synthesis.^ A number of studies have been performed on racemic substrates but only one enantiomer has been shown to assist the reader. In some schemes the S = N and/or the S = 0 bond of the sulfoximine has been shown as a single bond. This is only for reasons of clarity.
11. ALKYL SULFOXIMINES A. Synthesis
Enantiomerically pure (+)-(5)-l and (~)-(/?)-S-methyl-5-phenylsulfoximine 1 are readily available via the resolution of racemic 1 with (+)- or (-)-lO-camphorsulfonic acid, respectively.^'^ The resolution of these compounds using 0.6 molar equivalent of (+)- or (-)-lO-camphorsulfonic acid was reported in 1997 by Gais."^ This method is suitable for their large-scale resolution and their TV-methyl, A^-tosyl, and A^-silyl derivatives 2a-f are readily prepared from 1 in either racemic or optically active forms (see Eq. 3 for details). Johnson first reported the preparation of racemic sulfoximines from the reactions of phenyl sulfonimidates^ or sulfonimidoyl fluorides^ with organolithium and Grignard reagents (Eq. 1). The related
p Ph»-S-Me NH (+)-(S)-1
0
II Phi«-S-"iMe il
NH (-)-(^-l
o II
Ph—S-Me II NR 2
a; R = Me b; R = I s c; R = I M S d; R = Me2Bu*Si e; R = MeaPhSi f; R = Bu^PhgSi
Chiral Sulfoximines
285
sulfonimidoyl chlorides are not useful since they are generally reduced to the corresponding sulfinamides^; however, sulfonimidoyl chlorides react with ethylaluminum chloride to give 5-ethylsulfoxiniines in good yields.^
Ar—S-OPh
•
Ar—S-R^ -<
Ar—S-F
II
II
II
NR
NR
NR
(1) ^^
Phenyl sulfonimidates^ or sulfonimidoyl fluorides^ are readily prepared from sulfonimidoyl chlorides, which are available in two steps from arenesulfmyl chlorides (Eq. 2). Treatment of the arenesulfmyl chloride with an amine and base (e.g., triethylamine) gives the corresponding sulfmamide which on chlorination with chlorine,^ tert-buty\ hypochlorite,^ or A^-chlorobenzotriazole^^'^^ gives the corresponding sulfonimidoyl chloride (Eq. 2).
base ArSOCI + RNHg
O M "Cr" • Ar—S-NHR • NaOPh ^
^^y^ ^
O II Ar—S-OPh II
NR
O II Ai—S-CI ^^
NaF I
MeCN t O II Ar—S-F II
,^, (2)
NR
These methods have recently been extended to the preparation of enantiomerically pure sulfoximines from optically active sulfonimidates. For example, Pyne^* prepared the optically active diastereomeric sulfoximines 4 and 5 from the reaction of the phenyl sulfonimidate 3 and methyllithium. Compound 3 was prepared from (+)-norephedrine as a 1.8:1 mixture of diastereoisomers that could not be separated chromatographically. The sulfoximines 4 and 5 were readily separated by column chromatography in isolated yields of 14 and 28%, respectively. A related method to prepare optically active N-(5)-l-phenylethyl-5-methyl-5-phenylsulfoximines has also been reported.*^ Regellin has reported that enantiomerically pure sulfoximines 8 and 9 can be prepared via the nucleophilic substitution of the cyclic sulfonimidates 6 and 7, respectively, with 2 molar equivalents of an organolithium or Grignard reagent. Yields were excellent and ranged from 85 to 97%. These reactions were not successful with Bu^MgBr. The diastereomeric sulfonimidates 6 and 7 were prepared from 0-trimethylsilyl valinol and could be separated by column chromatography.^^ A more convenient synthesis of the diastereomerically pure cyclic sulfonimidates 6 and 7 has been developed. The diastereomeric sulfmamides 10 and 11 are more
286
STEPHEN G. PYNE ^
R
Ph- •S=N-f-''^® 1MeLi. ether ^ M e U P^^'^"Me Me-'^^jj^'P'^L-Me PhO ^ ^ p ^ 2. column chrom. MeO
p^*4^^, OMe
MeO^^Ph H
readily separated by crystallization than the cyclic sulfonimidates 6 and 7. Treatment of these separated compounds with tert-buiyl hypochlorite, a reaction that gives the corresponding sulfonimidoyl chlorides with retention of configuration, followed by base treatment at -78 °C gave the diastereomeric pure cyclic sulfonimidates 6 and 7, respectively, with inversion of stereochemistry at the stereogenic sulfur.^^ An alternative method for preparing sulfoximines is from the imination of sulfoxides with hydrazoic acid^^ (prepared in situ from sodium azide and sulfuric acid in chloroform) or 0-mesitylsulfonylhydroxylamine (MSH; CAUTION potentially explosive)^^'^^ to afford A^-unsubstituted sulfoximines. The former method, however, is not suitable for sulfoximines in which one or more 5-alkyl substituents
EtNMeo OTMS
ft
H
' OTMS (70% overall)
O
V
Vs
6
OH
8
O Ar«'"j^^Q
.^S. Ar'
LBu^OCI 2. KF. 18-crown-6 3. silica gel
O RM
Ar—S-^ OH
Ar = p-Tolyl M = Li or MgX R = alkyl, allyl, vinyl, phenyl
Chiral Sulfoximines
287 O I.Bu'OCI,-78°C 2. DBU, -78 °C
A.-.L,
V
A.* " 10
OH
LBu^OCI,-78°C 2. DBU, -78 °C
O II
,
" JH 11
can readily undergo heterolysis of the C-S bond under the acidic conditions (e.g., benzyl, allyl, Bu^ or some secondary alkyl groups). The MSH method on optically active sulfoxides gives optically active sulfoximines with retention of configuration.^^ The copper(0)-catalyzed reactions of tosyl azide^^'^^ or chloroamine-T^^ with sulfoxides gives A^-tosyl sulfoximines (Eq. 3). The A^-tosyl group in these compounds can be cleaved by acid hydrolysis,^^ photolysis,^^ or reduction.^^'^"^ The most efficient method seems to be reduction with sodium anthracenide,^'* although this method does not work for A^-tosyl-5-methyl-5-phenylsulfoximine. The yV-unsubstituted sulfoximines can be readily A^-silylated,^^"^^ methylated (HCOOH/HCHO^^'2^'2^ or Me3OBF430), alkylated,^^ methoxycarbonylated,^^ tosylated,^^ trifluoromethylsulfonylated,^^ or nitrosylated.^"* A method for preparing N-Boc sulfoximines from sulfoxides using r^rr-butyloxycarbonyl azide (CAUTION potentially explosive) and iron(II) chloride has been reported.^^ ^
HN3 or MSH
Ri—S-R2
TSN3 Cu(0)
^
base / R^X
^ R^—S-R2
NH
•-
^ R^—S-R2
II ,
NR3
or TsN(Na)CI O
H-'orhvorlH] (3)
R^-S-R2
n NTs
A^-Trimethylsilyl-5-alkylsulfoximines can be readily deprotonated with strong base (n-butyllithium or LDA) and alkylated with alkyl halides, aldehydes, and
288
STEPHEN G. PYNE
epoxides to give, after mild acid hydrolysis, chain extended N-unsubstituted sulfoximines (Eq. 4)}^ O II RCHp—S-Ph NTMS
1.A7-BuLJ •
O , II RCH(R^)—S-Ph
"^'^ ^ 3. H30-^
...
NH
Sulfoximines can also be prepared from the oxidation of sulfilimines. The oxidizing agents sodium periodate/ruthenium dioxide,^^ alkaline hydrogen peroxide,^^ and m-chloroperbenzoic acid anion^^ have been successfully utilized. N-Q?Tolylsulfonyl)sulfilimines have been oxidized to their corresponding sulfoximines in high yields using dimethyldioxirane in acetone.^^ Oxidation of (-)-(5)-5-(/7tolyO-S-methyl-A^-i/^-tolylsulfonyOsulfilimine 12 (ee 80%) gave the corresponding (-)-(/?)-sulfoximine 13 with complete retention of configuration at the sulfur atom. In 1998 Gais reported the synthesis of enantiomerically pure cyclic sulfoximines 19 and 20 from (+)-(5)-5-methyl-5-phenylsulfoximine 14 via A^-alkylation with THP protected 2-bromoethanol and 3-chloropropanol, THP hydrolysis, and then O-tosylation to give 17 and 18, respectively. Base-promoted cyclization of 17 and 18 gave enantiomerically pure cyclic sulfoximines 19 and 20, respectively."^ B. Lithiated Sulfoximines Structural Studies The X-ray crystal structures of lithiated sulfoximines were reported in 1986/87 by Gais. Lithiated (^-N,5-dimethyl-5-phenylsulfoximine crystallized as its tetramethylethylenediamine (tmeda) complex as a chiral tetramer of structure [(S)'Nmethyl-5'-phenylsulfonimidoyl)methyllithium]4-2(tmeda) with approximately C2 symmetry."^^ Two of the lithium cations of the tetramer were coordinated to a tmeda molecule and the O atoms of two different carbanionic species. The other two lithium cations were found to be coordinated to the N atoms of three different sulfonimidoyl carbanionic species and to one C atom (the a-carbon) of each of these carbanionic species. These lithium cations were thus found to form four-membered chelate rings involving the atoms, Li-C^^-S-N. A later study was successful
equiv.j ^Tol (S).(.) 12 (e.e. 80%)
acetone. 0-25 ^'C (90%)
^""^o' ifi'^aW? (e.e. 80/o)
Chiral Sulfoximines
289
O ^ .NH
1.KH, DME, n-Bu4Br 2. RX. 25 °C ^
Me (S)-14
or,OH Me
PK-
(S)-15(n = 1) (S)-16(n = 2)
3. HCI(aq) RX = Br(CH2)20THP or RX = CI(CH2)30THP
TsCI. EtgN CH2CI2 25 °C KO*Bu/n-BuLi
prf V.()„
THF, -78 to 25 "C
On Pft
.OTs
Me (S)-17(n = 1) (S).18(n = 2)
(S)-19(n = 1) (S)-20 (n = 2)
in producing crystals in the absence of the tmeda donor ligand. In this study the X-ray crystal structure of lithiated racemic trimethyl[(A^-(trimethylsilyl)-5-phenylsulfoniniidoyl)methyl]silane [(TMS)2CHS0(NTMS)Ph]' showed a chiral tetramer structure with C2 symmetry consisting of two (/?), (R) and two (5), (5) diastereomers.'*^ As was found above, the lithium cation formed four-membered chelate rings involving the atoms, Li-C^^^-S-N and L i - N - S - 0 . In THF solution, however, ^^C NMR experiments suggest that the lithiated sulfonimidoyl carbanion forms a THF-solvated aggregate having little or no C^-Li contact. Earlier solution ^^C NMR studies in THF concluded that C^^ in lithiated PhSO(NMe)Me was hybridized intermediate between pyramidal and planar.'*^ More recently, the gas-phase structures of lithiated A^,5,5-trimethylsulfoximine have been calculated by ab initio methods."*^ It was found that a Li-C^^-S-N four-membered chelate 21 was the most stable isomer and the complex involving a Li-N-S-O four-membered chelate (22) was only 1.4 kcal/mol higher in energy. The alternative complex involving a Li-C^j^-S-O four-membered chelate was 6.1 kcal/mol higher in energy.
Me I ^ H2C^ v ' M e 21
Li— O I N Me'/ CH2 22
STEPHEN G. PYNE
290
A single-crystal X-ray structural analysis of dilithiated racemic 5-ethyl-A^methyl-5-phenylsulfoximine/tmeda complex has been reported."^^ The unit cell contains two clusters, each composed of two sulfoximine dianions with the (R) configuration and two sulfoximine monoanions with the (5) configuration together with three tmeda molecules. The dianion shows two Li-C contacts with the ortho-C of the 5-phenyl substituent, as shown in 23a, rather than the a-carbon. The monoanionic sulfoximine moiety shows Li-C^^ contacts and the Li-substituent adopts a gauche conformation with respect to the N- and O-substituents of the sulfoximine group as shown in 23b. This conformation is most likely favored by a stabilizing n^-o* interaction between the nonbonding orbital (n^) on the a-carbon atom and the a* orbital of the S-Ph bond. Furthermore, the C^^ methyl substituent is anti to the A^-methyl substituent. Reactions with Alliylating Reagents
Alkylation of (+)-(S,5)-A^-methyl-5-phenyl-5-(/7-tolylsulfinylmethyl) sulfoximine 24 under phase-transfer conditions (50% aqueous NaOH/CH2Cl2/benzyltriethylammonium chloride) gave monoalkylated products with extremely high diastereoselectivity (100%). Its (5,/?)-diastereomer, however, is much less diastereoselective (diastereoselection ca. 80:20). In contrast, ethylation of the corresponding sulfone analogue of 24, (A^-methyl-5-phenyl-5-[/7-tolylsulfonylmethyl]sulfoximine), resulted in a 50:50 mixture of diastereomeric products; clearly the diastereoselectivity in these reactions is primarily determined by the sulfmyl moiety rather than by the sulfoximine group.'^^ Alkylation of the a-lithiated derivatives of the enantiomerically pure cyclic sulfoximines 19 and 20 with primary alkyl halides gave in high diastereomeric purities the corresponding alkylated Me 2 nBuLi, PhSO(NMe)CH2Me tmeda, LigO
K Li T P
N S
1 1 ^ Me H 23b
121'
Chiral Sulfoximines
291
Q O II n . Ph-S-CHg-Sr • '* NMe 24
Rx
0 0 n • " . >^ Ph-S-CH(R)-Sr-
Tni phase transfer H ^Tn\ ^0" NMe ^o' (100% diastereoselection)
products 25 and 26."^ Further treatment of the products with base and then trifluoroacetic acid (TFA) gave the corresponding epimeric products 27 and 28, respectively. While treatment of lithiated 26 with a different alkylating agent (R^X) gave the a,a-disubstituted adducts 29 in high diastereomeric purities. Thus, all reactions are diastereoselective and electrophilic attack occurs from the same side as the sulfoximine oxygen or anti to the 5-phenyl substituent. The stereochemical outcomes of these reactions were rationalized as arisingfromattack by the electrophile (RX or TFA) on the lithiated sulfoximine that would be expected to prefer the conformation 30. This conformation was expected to be favored because of a stabilizing n^-c* interaction between the nonbonding orbital on the a-carbon atom and the a* orbital of the S-Ph bond. Consequently, the S-Ph substituent is pseudoaxial. Attack on X would be expected from the topside of the molecule because of possible electrophilic assistance from the lithium cation and steric shielding of the bottom face by the S-Ph group. Racemic, dilithiated 5-ethyl-A^-methyl-5-phenylsulfoximine 23a has been treated with a variety of electrophiles. Treatment with 2 molar equiv of iodomethane 1./7.BuLI,THF 2. RX,.78°C
^./r^ X '
^ >n
(yields 57-94%) de 89-98% (S)-19(n = 1) {S)-20 (n = 2)
1.n-BuLi,THF 2. TFA
^
Ph^ V ( ) n R^^^ 25(n = 1) 26 (n = 2)
1. n-BuLi, THF I (yields 42-96%)
2. R^X,-78°C I (n = 2)
*
R^ R 29
>v
'
(yields 90-91%) Ph" V ( )n de 64-90% pf ^^
de>98%
27(n = 1) 28 (n = 2)
STEPHEN G. PYNE
292
gives the a,a-dimethylated adduct 31 in 85% yield.^"^ Surprisingly, only traces (< 3%) of c>rr/io-methylated products, that could arise from methylation of 23a, could be detected by GC/MS. To explain this result it has been suggested that a-methylation occurs initially followed by a translithiation from the ortho-position to the a-carbon. Similar reactions with bis-electrophiles gave cyclic sulfoximines 32, including cyclopropane (11% yield) and cyclobutane derivatives (27% yield). Interestingly, treatment of the dianion with 2 equiv of ethyl chloroformate gave the heterocycle 33 in which a bond has formed between the ortho-C and the electrophilic reagent. A study on the alkylation reactions of different (5)-A^-substituted sulfoximines 34 has been reported by Trost."*^ In contrast to our findings on the reactions of lithiated sulfoximines with aldehydes and imines'^^"^^ (see the following sections "Reactions with Aldehydes and Ketones" and "Reactions with Imines") the diastereoselectivity of the benzylation of lithiated 34 was not significantly dependent on the steric demand of the A^-substituent. The best diastereoselectivity (d.r. = 9:1) was found with the A^-nitrosulfoximine derivative 34 (R = NO2). Trost has suggested that the enhanced diastereoselectivity found using A^-nitrosulfoximines is related to "its participation as a coordinating substituent for lithium.""*^ The stereochemical outcomes of these alkylation reactions, for reasons that have been suggested above (see structure 30) and published previously,^^ can be rationalized by evoking alkylation of the lithium-chelated intermediate 36 from the less hindered face (and to the 5-phenyl substituent) with possible electrophilic assistance from the lithium cation.
,.
r^—Me
2 nBuLi, THF PhS0(NMe)CH2Me
0-^^—^
2 CICOgEt (63%)
MeN^ P Q
XCH2(CH2)nCH2X (11-81%)
MeS^^Me Me ^Me
31 32. m = 2-5
Chiral Sulfoximines
293
TMS S:'Miph
1. LIN(TMS)2/THF TMl 2. PhCHal ^.
(S)-34
PhCHgl TMS
rv
S:""Ph \ - "^ NR PhCH2 H ' " (R, S)-35
FRT= Ts: dr. = 4 : 1 R == 2,4,6 (POaCgHgSOg-: d.r = 4 : 1 R == Bu^PhgSi-; d. r. = 1 : 1 R == NO2: d. r. = 9 : 1
Reactions with Aldehydes and Ketones: Synthesis offi-Hydroxy Sulfoximines Addition to aldehydes. The reactions of lithiated 2a with aldehydes gives P-hydroxy sulfoximines with modest diastereoselectivity (Table 1).^^ The chromatographic resolution of the diastereomeric adducts was difficult; however, reductive cleavage of the diastereomeric product mixture gave secondary alcohols in 25-46% ee. For example, treatment of lithiated (+)-(5)-2a of 85% enantiomeric purity (ee) with benzaldehyde gave a mixture (3:1) of diastereomeric adducts. Reductive desulfurization of this mixture gave (+)-(/?)-1-phenylethanol in 37% ee. The stereochemical outcome of these reactions can be rationalized as arising from a chelated boat conformation analogous to 41 in Scheme 1. More recently, Hwang^^*' and Pyne"*^^ have reported much higher diastereoselectivities employing the ^V-silylated analogues of 2a. While racemic A^-trimethylsilyl5-methyl-5-phenyl sulfoximine 2d showed a similar diastereoselection to 2a in its condensation reactions with aldehydes, the sterically more hindered r^rr-butyldimethylsilyl, methyldiphenylsilyl, and r^rr-butyldiphenylsilyl derivatives 2d, 2e, and 2f exhibited much improved product diastereoselections. Table 2 clearly demonstrates the effect of the steric demand of the A^-silyl group on the diastereoselectivity of the 1,2-addition reaction of the lithiated sulfoximines 2(a, c-f) with pivaldehyde. Progressing from the least sterically demanding A^-trimethylsilyl derivatives 2d to the highly sterically demanding A^-rerr-butyldiphenylsilyl derivative 2f the product diastereoselection increased dramatically from 71:29 to 94:6. The A^-methyl sulfoximine 2a was slightly better than its A^-trimethylsilyl counterpart. Lithiated 2f showed high product diastereoselection with five representative aldehydes as shown in Table 3.^*^ The diastereoselectivity was found to be inde-
STEPHEN G. PYNE
294 OH
J.CH2 'v*Ph rrt il
j ^
(+)-(S)-2a
1./7-BuLi,THF^ 2. RCHO
^
v'
OH
37a '^'^® RaneyNi. ^
OH I
O II
R
H
^CH.
(ff)-38 (25-46% e.e.)
3;^NMe
Table 1. Reaction of Lithiated 2a with Aldehydes R of Aldehyde
ee (%) of 2a
Ph
Diastereoselection 37a:37b
85 85 92 95
n-C(,Hu Bu' Bu^
38 (% ee)
Yield (%)
37 25 30 46
75:25 60:40 71:29 74:26
78 70 69 65
pendent of the aldehyde (RCHO) substituent R. A similar trend was found with the related sulfoximines 2a and 2c (Table 4). From X-ray structural analysis the major adduct between lithiated 2c and acetaldehyde was determined to have the 2R*,SR* relative stereochemistry, identical to
Table 2. Effect of the N-Substituent of 2 on the Diastereoselection for the Condensation of 2 with Pivaldehyde
OH
0
il Ph^S'^CH-il i . n,,«riin -.. 11 2"-' + ou urnj NR
Sulfoximine^ R Me SIMej SiMe2Bu* SiMePh2 SiBu^Ph2
0
major diastereoJH^'^^Ph Bu'^:?\ NR isomer H + ^ >• minor OH 0 diastereoisomer Bu* NR
Diastereoselection 74:26 71:29 89:11 89:11 94:6
Reference Johnson^^ Hwang25b Hwang25b Hwang25^ Pyne^«^
Note: ^In some cases racemic 2 was employed, in these cases the products were also racemic.
Chiral Su Ifox imines
295
Table 3. Reaction of Lithiated 2f with Aldehydes (RCHO)
?'H 2f (racemic)
1./7-BuLi.THF
•* i H
2. RCHO
O II CHa'^CPh 39b
,
^SiBu'Ph^
R of Aldehyde
Yield {%r
39a:39bl^
Et
82 79 74 89 86
92:8 96:4 93:7 91:9 94:6
Bu' Pr' Ph
Notes:
C H , V Ph gla NSiBu'Pha
OH I H*i R
Entry
S
^After purification by column chromatography. ^Determined by ^H N M R spectroscopy (400 M H z ) on the crude reaction mixture.
Table 4. Effect of the R Substituent of the Aldehyde on Product Diastereoselection in the Condensation Reactions of 2 major ^'''Ph diastereoisomer
P OH I
Ph-^S-^CHoLi + R C H O
R
W-(S)-2 la
R
O II
minor diastereoisomer NR^ Diastereoselection
Me
Ph
Me
Bu^
2.8:1
SiMe3
Ph
2.8:1
SiMe3
Bu^
3:1
2.5:1
SiBu*Ph2
Ph
91:9
SiBu*Ph2
Bu^
94:6
Note:
^In some cases racemic 2 was employed, in these cases the products were also racemic.
296
STEPHEN G. PYNE
that found by Johnson for the reaction of 2a with aldehydes.^^ Cyclic chair transition states for the reaction of 2c-e with aldehydes, that involving chelation of both the aldehyde and sulfoximine oxygens or the aldehyde oxygen and the sulfoximine nitrogen by lithium cation, have been proposed by Hwang.^^*' In the case of lithiated 2f, chelation to the highly sterically hindered sulfoximine nitrogen would seem highly unlikely. The four possible chair- and four possible boat-like transition states for the reaction of lithiated 2f with aldehydes that involve chelation of both the aldehyde and sulfoximine oxygens by lithium cation are shown in Scheme 1. The chair transition states 40, 42, 44, and 46 suffer from severe 1,3-pseudodiaxial-like interactions and the boatlike transition states 43 and 45 suffer from a flagpole interaction between the aldehyde substituent (R) and the sulfoximine oxygen and therefore these transition states would seem energetically unlikely. The preference for the diastereomeric adduct 39a over 39b can be readily accounted for by considering the competing boat transition states 41 and 47. Transition state 47 would be expected to be energetically less favorable than 41 if one considers the steric interaction between the solvent ligand on the lithium cation and the sterically demanding NSiBu'Ph2 group in 47 and that between the solvent ligand (L) on the lithium cation and the less sterically demanding SPh group in 41. One would expect that as the steric demand of the A^-substituent of the sulfoximine was increased, the transition state 47 would then be destabilized relative to that of 41 and a higher product diastereoselection would result. This is indeed the case. Furthermore, the difference in free energy between transition states 41 and 47 would be expected to be largely independent of the steric demand of the aldehyde substituent (R) since in these two transition states R experiences little steric interaction with the large substituents (Ph, Bu^Ph2SiN) on sulfur. These studies have been extended to the reaction of lithiated racemic N-tertbutyldiphenyl-5-benzyl-5-methyl sulfoximine 48 and its SPh analogue 49."^^ The results of the reaction of lithiated 48 with various aldehydes are reported in Table 5. In each case studied all four possible racemic diastereomeric products were formed. In the case of benzaldehyde a much higher diastereoselectivity could be realized if the aldehyde was precomplexed with BF3 etherate prior to addition to lithiated 48. The major (50s) and the second most prominent diastereomeric products (51a) had the syn {J. ^ - 1-8-2 Hz) and anti (7^ 2 = 9.3-10 Hz) relative stereochemistry, respectively, while the former diastereoisomer showed a SMe resonance at lower field relative to the latter. The relative 15*, 2/?*, S5* stereochemistry of 50s (R = Et) was unequivocally determined by single-crystal X-ray analysis."*^ An estimate of the dihedral angle between Cl-Hl and C2-H2 in 50s (R = Et) from the structural analysis ((t)i 2 ca. 62°) and the value of the H1,H2 coupling constant {^^ 2 ~ ^-^ ^^) ^^ deuterochloroform suggest a similar conformation for this compound in the solid state and in solution. The relative stereochemistry of 51a (R = Ph) was determined from an experiment in which diastereomerically pure 50s (R = Ph) was first treated with 2 equiv of
Chiral Sulfoximines
297 H
"vZ' -CH2i.g^ NSiBu'Pha LI
L Ph
\J 40
u. . I L PhgBu'SiNO
41*
39a
Ph-^Sf-CH/ PhoBu'SiN L
H
o
42
43
CH2j,g/, NSiBu'Pha
R
'
H-^o
Lr/0 -2-
,-,
L Ph 44
39b 45 H
„ °\rLi-"0 I L PhgBu'SiNO
R J
46
Ph-^si-CH^ PhoBi/SiN L
47*
Scheme 1. (L = solvent ligand molecules)
n-BuLi (-78 °C, 1 h) and the resulting dianion was then quenched with water. This reaction produced a 65:35 mixture of 50s (R = Ph) and 51a (R = Ph), respectively. This result clearly indicated that 50s (R=Ph) and 51a (R = Ph) differ only in relative stereochemistry at CI. An analogous experiment with 1 equiv of «-BuLi resulted in unchanged 50s (R = Ph) and clearly indicated that interconversion of 50s and
STEPHEN G. PYNE
298
PhCHg—Sl
\
R2
48 R^ = SiBu^Phg, R^ = Me 48aR^ = S i B u W 2 , R ^ = M e 49 R^ = SiBu^Phg, R^ =Ph
51a via a retroaldol type process was not occurring at -78 °C. The relative stereochemistry of the two minor diastereoisomers 51s and 50a were based on the chemical shift of their respective SMe groups and the magnitude of/j 2The reaction between lithiated A^-rer^butyldiphenyl-5-benzyl-5-phenylsulfoximine 49 and benzaldehyde gave only three out of the possible four diastereomers in the ratio of 82:14:4 and in good yield. The major diastereoisomer had the syn relative stereochemistry (Jj 2 = ^-^ ^^)Table 5. Reactions of Lithiated 48 and Aldehydes R ; ^^
Hi 1.r7-BuLi,THF.-78°C 48
O^ ^J^SIBu^Phg
Ph
syn-508 R^ = OH, R^ = H anfZ-SOa R^ = H, R^ = OH
2. RCH=0 R ; ^^ R'^^
(\
/JSiBu^Phg
C^ PHT
Me
H^
anfA51aR^=0H, R2 = H syn-5l8 R^ = H, R2 = OH Diastereoselection Aldehyde PhCHO PhCHOBFj EtCHO Pr'CHO BuHlHG
Yield (%)
50s:
51a
51!;:
50a
95 95 96 91 90
48: 67: 77: 70: 75:
26 28 18 16 16
10 1 4 6 1
16 4 1 8 8
Chiral Sulfoximines
299
^ C > ^ ^^Sx. Ph^ y ^ Me H" Ph 508
1. n-BuLi (2 equiv), ^CV ^^Sx. • Ph"^ ^ C ^ Me THF,-78°C 2.H2O
R^ ^^ (50s:51a = 65:35) 508 R^ = H. R2 = Ph 51a R^ = Ph. R 2 = H
While the stereochemistry of the major diastereomeric adducts from the reaction of lithiated 48 and carbonyl compounds can be rationalized as arising from cyclic boat transition states (52a,b), the transition state 52b, which is analogous to the transition state 41 (Scheme 1) proposed for the reaction of 2f with aldehydes, appears unlikely because of a number of severe 1,2-steric interactions, in particular the SMe group and the benzylic phenyl group are eclipsed in 52b. Indeed, when the aldehyde is precomplexed with BF3, a cyclic transition state cannot occur. We suggest that the structure of lithiated 48, as shown by structure 48a (only the monomeric species is considered) in Scheme 2, may be similar to that of lithiated benzyl phenyl sulfone.^^ One would expect the benzylic carbon of 48a to be close to planar and the phenyl substituent to be and to the bulky A^-/^r^butyldiphenylsilyl moiety. The nonbonding orbital at the benzylic carbon would be approximately coplanar with the S-CH3 a bond as a result of a stabilizing n^-o* s-c interaction. Electrophilic attack on 48a should occur from the less hindered diastereoface, i.e., anti to the S-CH3. An open transition state 53 in which R of the aldehyde and the phenyl substituent of 48a are anti is consistent with the stereochemical outcome. The major anti diastereoisomer 51a most likely arises from an open transition state (involving attack of the aldehyde from the same diastereoface of 48a as the SMe group) while the minor syn and anti diastereoisomers (51s and 50a) most likely arise from a chelated chair transition state in which R of the aldehyde (RCHO) is pseudoequatorial. Consistent with this proposal is the observation that when PhCHO BF3 was employed, the yield of 51a was essentially unaffected while the combined yield of 50s and 51a decreased to about 5%^^
H. "~^0.-Li-/° PhOv"^'_Z>Me H
NSiBu'Pha 52a
Phg'BuSiN Me .< 52b
•• Ph
300
STEPHEN G. PYNE Li r^iBu^Ph2
"^ R
.Cr
Me
Me 50s Scheme 2.
Addition to ketones. In 1982, Johnson and Stark^^ reported the condensation reactions of (+)-(5)-A^,5-dimethyl-5-phenylsulfoximine (2a) with various aldehydes and prochiral ketones. The reaction of lithiated 2a with phenyl aryl ketones (PhCOR, R = Me, Et, n-Pr, n-Bu, and c-C^Hj j) gave a mixture of two diastereomeric P-hydroxy sulfoximine adducts 54 with modest diastereoselectivity. Unfortunately, the diastereoselectivities of all of these reactions were not documented. While these diastereoselectivities were modest, the diastereomeric adducts 54 could be readily separated by column chromatography in good overall yields. The resulting diastereomerically pure adducts could be converted to chiral tertiary alcohols in high enantiomeric purity (87%-100%). For example, the higher R^ diastereoisomer 54 (R = Et) from the reaction of lithiated 2a and ethyl phenyl ketone was converted to enantiomerically pure (+)-(5)-2-phenyl-2-butanol 55 by reductive desulfurization with Raney nickel.^^ The reaction of lithiated (-^y(S)-N-tertbutyldiphenylsilyl-5-methyl-5-phenylsulfoximine 2f with prochiral methyl ketones (RCOMe) gives a mixture of diastereomeric P-hydroxy sulfoximine adducts 56 and 57. The diastereoselectivity increased as the steric demand of the R group
r ^fa.n
^- ^"BuLi, THF. 0 °C, 15 min
^ ^^ ^
2. PhCOR. 25 °C
R
OH
O
1
IJ
P^l^i^^ R 54
Diastereoselection Yield(%)
Me
67:33
85
Et
60:40
88
\.r. NMe
Chiral Sulfoximines
301 OH I
Et
O II
Raney Ni
Ph NMe 54(2S,SS) 2 V
OH I Prt*f^CH3 Et
(+HS)-55 (100% e.e.)
in RCOMe increased. In each case the major diastereomer could be isolated diastereomerically pure after purification of the crude reaction mixture by column chromatography or recrystallization. The relative (25, S5) stereochemistry of the major adducts was determined by single-crystal X-ray structural analysis.^"^ The stereochemical outcome of the above reactions was rationalized as arising from the two competing boat transition states 58a and 58b. The difference in free energy between 58a and 58b, and hence the diastereoselectivity, would be expected to increase as the steric demand of the R group of the ketone increases as a result of an increasing flagpole interaction between R and the sulfoximine oxygen in 58b. Competing chair transition states (e.g., 59) were thought to be less favorable for steric reasons.^'* The reaction of lithiated (+)-(5)-2f with racemic 2-alkylcyclohexanones gave three diastereomeric products, 60 and 61; the latter product was obtained as a mixture of diastereoisomers. The preference for the formation of 60 was rationalized as occurring via the favored boat transition state 62. The reaction of lithiated 48 with cyclohexanone proceeded with high diastereoselectivity (94:6) but the yield was low (60%) and starting materials were always recovered, probably as a result of a competing proton transfer reaction between the two reactants. The stereochemical assignment of the major (63a) and minor diastereoisomers (63b) from this reaction was based on their respective SMe chemical shifts and by analogy with the reaction of 2f with aldehydes."^^
1.n-BuLi,THF. 2. RCOMe
::"Ph NSiBu'Phg
78 °C
56
(S)-2f OH
[RT = Et. 56 :57 = 80: 20 (69%)* R == Pi^.56: 57 = 79 : 21 (43%) R == Ph,56 57 = 91 : 9 (65%) IRJ= Bu».56 :57 == 98 : 2 (63%) yields refer to yield of pure 56
O NSiBu'Phg
57
302
STEPHEN G. PYNE Me
R
PI, ^^S—NS1R3 H
o
"^ju_s—NSiR3
Ph
H
58a
Ph
Me-^—• Ph
58b
59
Reactions with Imines
When a THF solution of lithiated racemic 2a was quenched with A^-benzylideneaniline at -78 °C, a 50:50 mixture of the two possible diastereoisomeric adducts 64 (R = Me) and 65 (R = Me) was obtained in 96% yield."*^ In contrast, the analogous reaction of the A^-butyldiphenylsilylsulfoximine 2f gave an 88:12 mixture of the diastereomeric products 64 (R = SiBu^Ph2) and 65 (R = SiBu'Ph2), respectively."*^ It is again apparent that a highly sterically demanding substituent is required at the sulfoximine nitrogen to ensure high diastereoselectivity in these reactions. The diastereoselectivities for the reaction of lithiated 2f with other imines of structure RCH=NPh are presented in Table 6. For these reactions the product diastereoselection progressively decreases as the steric demand of the substituent R increases. When R was relatively small (R = Et or Bu\ entries 1 and 2), high product diastereoselection (95:5) was observed, whereas when R was sterically demanding (R = Bu^ entry 6), the reaction proceeded with modest diastereoselectivity (79:21). When R was intermediate in size, that is, Pr^, phenyl, or 2-furyl, the product diastereoselection was consistently 90:10 (entries 3-5). The relative 2/?*,S5*
a a:
1.nBuLi,THF,-78°C 2.
R 60 NSiBu'Phz (major diastereomer)
(S)-2f
(R == Me. Bu^)
-A
H
5—-NSiRa Ph
H 62
M
g^^
NSiBu'Phg
(minor diastereomers)
Chiral Sulfoximines
303
Hi 1.n-BuLi,THF,-78°C
Ph
J 82.13 i
63a
48 2. cyclohexanone
K Ph
Hi
^
63b
fSI^QSl
Stereochemistry of 64 (R=Et) was unequivocally determined by a single-crystal X-ray structure analysis."*^^ Since the imines must have the (£)-geometry, only four possible chelated cyclic transition states are available for the reaction of lithiated 2f and imines, two chair (67 and 69) and two boat transition states (68 and 70).^^ The two possible chair transition states suffer from severe 1,3-pseudodiaxial-like interactions and therefore would seem energetically unlikely. Clearly the preference for the diastereomeric adduct 64 over 65 suggests that the boat transition state 68 is favored over its boat counterpart 70. This would seem likely when one considers the steric interaction between the solvent ligand on the lithium cation and the sterically demanding NSiBu^Ph2 group in 70 and that between the solvent ligand (L) on the lithium cation and the less sterically demanding SPh group in 68. One would expect that as the steric demand of the A^-substituent of the sulfoximine was decreased from SiBu^Ph^
NHPh
O
II Ph-^S-CHgLi
+PhCH=NPh
-78 ^C THF
Ph
^
NR
64
NR 2a R = Me 2f R = SiBu^Phg
(2RSS)
NHPh
Sulfoximine Diastereoselection 64:65 Ph^ 2a 2f
O
50:50 88:12
H
O
NR 65 (2S,SS)
304
STEPHEN G. PYNE Table 6. Reactions of Lithlated 2f with Imines 66 NHPh O I II R 1.n-BuLi,-78°C
^ 64
NSiBu^Ph2
(2/^SS)
2. RCH=NPh{66)
+ NHPh O I
R\i H
Entry
R of Imine 66^
CH 1 NSIBu^Phg 65 (2S.SS) Yield (%)^
Diastereoselection 64:65^
1
Et
68
94:6
2
Bu'
76
95:5
3
2-furyl
90
90:10
4
Pr'
70
90:10
5
Ph
90
88:12
6
Bu^
24
79:21
Notes: ^Reaction temperature -45 °C for a period of 2 h except for entries 3 and 5 (-78 °C, 1 h). ^After column chromatography. '^Determined by ' H N M R spectroscopy (400 MHz) on the crude reaction mixture.
R
TL
PhaBu'SiN . . - - ^
67
Ph L Ph O
68
f H. LI *-Ph
69
/Sr-Cf^2 PhoBu'SiN •Ph
70
Chiral Sulfoximines
305
to Me, transition states 68 and 70 would then be closer in energy and a lower product diastereoselection would result. This is indeed the case. The results of the reaction of lithiated 48 with imines 66 or the imine BF3 complex are presented in Table 7.^^'^^ In each case examined, only two of the four possible racemic diastereoisomeric products were formed. While the reaction of lithiated 48 with imines (Table 7, entries 1, 4, 5, and 7) proceeded with moderate product diastereoselection, the analogous reactions with imineBF3 complex gave the adducts 71 in consistently high diastereoselectivity (Table 7, entries 2, 3, and 6).^^ The relative 15*, 25*, S5* stereochemistry of 71 (R = Et) and 71 (R = Pr") were unequivocally determined by a single-crystal X-ray structure analysis.^^*^^ An estimate of the dihedral angle between Cl-Hl and C2-H2 in 71 (R = Et) and 71 (R=Pr^) from the structured analysis ((^^ ^ ^^- ^^ ^^^ 175°, respectively) and the value of the HI, H2 coupling constant (^i 2= 3.2 and 6.4 Hz, respectively) from the ^H NMR analysis of 71 (R = Et, Pr*) in deuterochloroform solution suggest that these compounds adopt a similar conformation in the solid state and in solution. The relative stereochemistry of the major diastereomeric adducts 71 (R = Ph, 2-furyl) were assigned by analogy with those of 71 (R = Et) and 71 (R = Pr"). The structural analysis clearly shows that the reaction of lithiated 48 with aldehydes and imines occurs in the same stereochemical sense with respect to the configuration at the stereogenic center at CI but in the opposite stereochemical sense with respect to the configuration at the stereogenic center at C2.
Table 7. Reactions of Lithiated 48 with Imines 66 1. /7-BuLi.THF,-78°C 48
"• 2. R^CH=NPh (66)
PhNH. 1^
^S\Bu^Ph2
H2
ct^ 1
^Me
. .4
Ph
or 66.BF3 71 Entry
R^ oflmine(ee)
1 2 3 4 5 6 7 8
Ph Ph 2-furyl Et Bu' Bu' Pr' Pr'
Additive
Temp rC)
—
-78 -78 -78 -45 -45 -78 -45 -78
BF3 BF3
— — BF3
— BF3
Yield (%) 60 86 82 66 51 70 58 0^
Diastereoselection^ 79:21 95:5 95:5 82:18 82:18 96:4 50:50
Notes: ^Determined by ^H NMR (400 MHz) spectroscopy on the crude reaction mixture. ''Complex mixtures of reaction products resulted.
—
306
STEPHEN G. PYNE
The open transition state 72a is consistent with the stereochemical outcome. The alternative transition state 72b, in which the A^-phenyl substituent of the imine is and to the benzylic phenyl group of the sulfoximine, suffers from severe steric interactions. Treatment of lithiated sulfoximines 48a and 49 with A^-benzylideneaniline BF3 complex gave the desired adducts with only moderate to good diastereoselectivity (Table 8)."*^^ The lithiated sulfoximine 49 failed to give adducts with other acyclic imines even when the imines were precomplexed with BF3 etherate. This is possibly a consequence of the increased steric demand and the resonance stabilizing effect of the 5-phenyl group of 49. The reaction of lithiated 48 with the 3,4-dihydro-6,7-dimethoxyisoquinoline BF3 complex 73 gave the 1-benzyltetrahydroisoquinoline 74a in a highly diastereoselective fashion (diastereoisomeric ratio 92:8). The relative stereochemistry of the major diastereomer of 74a was tentatively assigned by analogy with 71."^^ In contrast, the reaction of lithiated 48a and 3,4-dihydro-6,7-dimethoxyisoquinolineBF3 complex 73 gave all four possible racemic diastereomeric products in a ratio of 40:30:16:14 (Table 9). The stereochemistry of the two major diastereomeric compounds could be tentatively assigned on the basis of their ^H NMR spectra data, with the two major diastereoisomers assigned the 15*, S5* relative stereochemistry on the basis of the downfield chemical shift of their SMe groups (5 2.68 and 2.74, respectively). In contrast, the reaction of lithiated 49 with 73 proceeded in a highly diastereoselective fashion, although the yield was poor (40%, Table 9). The reactions of lithiated 48, 48a, and 49 with 73 appeared to have occurred in the same stereochemical sense as judged from their similar ^H NMR spectra."^^ Reactions iv/f/i Enones: 1,2- versus 1^4'Addition and the Synthesis of Cyclopropanes
Johnson first described the cyclopropanation of chalcone using lithiated A^-tosyl 5-alkyl-5-phenylsulfoximines in 1973.^^ In one example, an optically active (ee 49%) cyclopropane [(15, 25)-(2-phenylcyclopropyl) phenyl ketone] was prepared from the reaction of chalcone and lithiated (/?)-A^-tosyl-5-methyl-5-phenylsulfoximine (ee 84%) at room temperature for 12 h. More recently a solid-state version of this reaction was reported.^^ Treatment of a mixture of powdered chalcone, (+)-Ntosyl-5-methyl-5-phenylsulfoximine 2b, and KOH in the solid state at 70 °C gave optically active phenylcyclopropyl phenyl ketone 75a in poor yield (19%) and Li
Li
- —P
\\
«. ^N
i'-^H
I
r)$iBu'Ph2
Pfi
I
" 72a
"
,NfeiBu'Ph2 72b
Chiral Sulfoximines
307
Table 8. Reactions of Lithiated 48, 48a, and 49 with N-Benzylideneanirme-BF3 Lithiated Sulfoximine 48 48a 49
Yield (%) 86 85 55
Diastereoselection 95:5 83:17 88:12
optical purity (14% ee). The yields could be enhanced to 94% using Bu^OK at room temperature, but the optical purity of 75a was still low (24% ee). The use of an optically active host molecule had an adverse affect on the optical purity of 75a. Pyne and Dong found that the reaction of optically active lithiated (5)-A^-tosyl-5methyl-5-phenylsulfoximine (5)-2b (ee 99%) with enone 76a at -78 °C gave exclusively the 1,2-adduct 77 as a 58:42 diastereomeric mixture in quantitative yield.^^'^^ When this reaction was performed at room temperature, the optically active and diastereomerically pure cyclopropane 75a was isolated in 88% yield. The enantiomeric purity of 75a ([a]p27 -388° (c 0.05, acetone)) was judged to be 99% based on the reported specific rotation of enantiomerically pure 75a (lit.^^ [a]D25 +390.5° (c 1.0, acetone)). Treatment of 77 with LDA at -78 °C followed by warming the reaction mixture to room temperature for 1 h gave the diastereomerically pure cyclopropane 75a in 60% yield. Surprisingly, oxirane products, which could potentially arise from nucleophilic displacement of the sulfonimidoyl group by the alkoxide in 78 (see
Table 9. Diastereoselectivities for the Adducts from Lithiated 48, 48a, and 49 with 73 1.n-BuU,THF,-78°C 48/49
74a R^ = SIBu^Phg, R2 = Me 74b R^ = SIBuH^eg, R^ = Me 74c R^ = SiBu^Phg, R^ = Ph Lithiated Sulfoximine 48 48a 49
Yield (%) 43 55 40
Diastereoselection 92:8 40:30:16:14 92:8
308
STEPHEN G. PYNE KOH(s), 70°C (S)-2b
%
^
3^^XlAp^
Ph
75a(l9%, e.e. 14%)
the next section), could not be detected in the crude reaction mixture. This experiment indicated that at room temperature the kinetically favored anionic 1,2-adduct 78 is in equilibrium with the anionic 1,4-adduct 79 and that the latter undergoes intramolecular displacement of the sulfonimidoyl group (to give 75a) at a much faster rate than the former anion that could give rise to an oxirane. The reaction of racemic 2b with enone 76b gave the cyclopropane 75b in high diastereomeric purity (d.r. = 98 : 2 from GC analysis) in 95% yield. Treatment of (/?)-carvone with racemic lithiated 2b gave a mixture of the diastereomeric 1,2-adducts 80 at -78 °C and the diastereomeric cyclopropanes 81a,b, and the double addition product 82 as a single diastereoisomer at room temperature. The diastereoselectivity in the case of 81 was similar to that obtained when (5)-2b was employed. Compound 81 has been prepared as a single diastereoisomer by Corey and Chaykovsky.^ Racemic and optically active (5)-A^-tosyl-5-butyl-5-phenylsulfoximines were prepared by alkylation of lithiated racemic 2b or (5)-2b^*''' (ee 97%) with bromopropane, respectively. Treatment of lithiated racemic A^-tosyl-5-butyl-S-phenylsul-
>A^^R
CH2 Li
" Ph. J TsN^
76
^
Li-(S)-2b THF -78^
a; R = Ph b; R = Me
THF -78 °C tort
TsN' 77 (100%. d.r. = 58:42)
75a (92%, d.r. = 9 9 : 1, e.e. 99 %) 75b (95%, d.r. = 98 : 2)
Chiral Sulfoximines
77
309
LDA,..THF -78 ^ to rt
OLi H ^fv4 ; ^ Ph"'^^^^'^^ ^ Ph--.i ••
76+ Li-2b
/•^o TsN 78 Li"^ "p H Ph 75a -^ (60 %, d.r. = 99 : 1)
79
(f?)-carvone + fao-Li-2b
THF -78*0
THF -78 ^ to rt
Ha
Ph.
^^Sr^\
.OH
Sf
Me
Me
81a 80 (84%, d.r. = 54 : 46) O
82(25%,d.r. = >99:<1)
(81a-•-81b, 42%; 81a :81b = 73: 27)
81b
310
STEPHEN G. PYNE
foximine 83 with the acyclic enones 76a-c at -78 °C gave mixtures of 1,2- and 1,4-adducts. The latter were formed in high diastereomeric purities (d.r. = 98-96 : 2-4) while the former were formed as diastereoisomeric mixtures. The relative stereochemistry of 84a was determined by X-ray diffraction.^^ When these reactions were performed at room temperature, the cyclopropanes 85a-c could be isolated in high diastereomeric purities. Optically active 85a and 85c were obtained from the reaction of (5)-2b with 76a and 76c, respectively. The enantiomeric purity of 85c was determined to be 98% from ^H NMR studies using chiral shift reagents, while that of 85a could not be determined in this manner. However, the ee of 85a was thought to be high based on its diastereomeric purity and the magnitude of its specific rotation relative to that of 85c. Synthesis ofOxiranes, Oxetanes, and Aziridines
The reaction of the sodium salt of (R)-A^-(/7-tolylsulfonyl)-5-methyl-5-phenylsulfoximine 2a (84% ee) with acetophenone at room temperature gave (-)-(5)-2methyl-2-phenyloxirane.^^The (5)-oxirane product must arise from the collapse of
Li ^i^-^'V:::^^^p2
-
r
Pr
P h •*——M^
TsN
76
83
THF -78 °C tort
THF -78 °C H
R2
R*^
I
o
TsN 84a (68%, d.r. = 98 : 2) 84b (56%, d.r. = 98 : 2) 84c (62%. d.r. = 9 6 : 4)
85a (90%, d.r. = 99 : 1 ) 85b (58%. d.r. = 9 9 : 1 ) 85c (68%, d.r. = 96 : 4, ee > 98 %)
a; R U R2 = Ph b; R^ = Me, R^ = Ph c; R^ = Ph, R2 = Me
Chiral Sulfoximines
311
the (25,S/?)-P-oxy sulfoximine intermediate shown below. Under kinetically controlled conditions, one would expect that the (2/?,S/?)-|3-oxy sulfoximine intermediate would predominate. It has been demonstrated that under these reaction conditions the 1,2-addition of 2a to ketones is reversible.^^ It is apparent that in this case the rate of formation of the (5)-oxirane product from the (25',S/?)-P-oxy sulfoximine intermediate is faster than that of the (/?)-oxirane product from the diastereomeric (2/?,S/?)-P-oxy sulfoximine intermediate. The reaction of the above sulfoximine anion with A^-benzylideneaniline gave (-)-l,2-diphenylaziridine ([a]D -12.9°).^^
A. Ph (-) The reaction of ketones with excess of the sulfoximine salt Na-2b gives oxetanes formed by ring opening of the initially formed oxirane and subsequent ring closure. These reactions are highly diastereoselective and generally the thermodynamically more stable oxetane, in which the C-O bond is "axial," is formed from cyclic ketones since the intermediate oxirane is formed under thermodynamically controlled conditions.^' In an attempt to enhance the optical purities of oxiranes via the above method the use of the ylide derived from (-)-Af-tosyl-5-methyl-5-neomenthylsulfoximine 86 has been investigated. The enantiomeric purities of the oxirane products 87 ranged from 56 to 86%.*^ A related study using 5-ej;o-2-bomylsulfoximines gave oxiranes in similar enantiomeric purities.*^
NTOS
Ph—S-CH2Na+ PhCOMe il O
DMSO RT
Na-(/?)-2b
_eN p O
c{^\ +PhSON(-)Tos Ph^""
Me Ph (2S.Sff)
(-)-(S) (yield 40%)
312
STEPHEN G. PYNE racemic Na-2b (3 equiv.) +i-butylcylcohexanone
DMSO 40 °C
Bu'
Bu'
DMSO, 30-32 °C
'YcH? Na®
87 yields 42-80% e.e. 56-86%
86
. . . .
^ ^
J? /f
N^\
II
O 88
89
1.K02CN=NC02K HOAc 2. aqueous base
90
/b-'-y p "N,
V'^^N^f^Me
O
Me
Chiral Sulfoximines
313
TMS
W^: H
MeaAl
NNO2
91a. R = -(CH2)6Me, R^ = H 91b. R = H. R^=-(CH2)6Me
\^'"(CH2)6Me \XT^(CH2)6Me H H 92a 92b
C. Other Reactions A new synthesis of diazenes (azoalkanes) has been developed using 4-(5,5-dimethylsulfoximino)-l,2,4-triazoline-3,5-dione 89.^ Treatment of fulvenes 88 with 89 followed by diimide reduction of the carbon-carbon double bond of the resulting cycloadduct and then mild base treatment gave the diazenes 90 in good overall yields. Trost"^^ has reported that A^-nitrosulfoximines undergo Lewis acid promoted cyclizations to tethered allyl silanes, enol silyl ethers, and activated aromatic rings. For example, the allyl silane 91 (3:1 mixture of 91a and 91b, respectively) underwent cyclization on exposure to trimethylaluminum in refluxing dichloromethane to give a 3:1 mixture of the diastereomeric methylenecyclohexanes 92a and 92b, respectively. This reaction proceeded with inversion of stereochemistry at the stereogenic carbon bearing the sulfoximine leaving group.
III. APPLICATIONS OF ^-HYDROXY SULFOXIMINES TO ASYMMETRIC SYNTHESIS A. Resolution of Racemic Chiral Cyclic Ketones p-Hydroxy sulfoximines are thermally labile and revert to their starting carbonyl compound and sulfoximine on mild thermolysis. This property has been exploited effectively as a method for the resolution of racemic chiral cyclic ketones.^^ For example, the addition of the lithium salt of (+)-(5)-2b (99% ee) under kinetically controlled conditions (-78 °C) to racemic menthone gave three of the four possible diastereomeric adducts. The major two adducts resulted from attack on the menthone from the equatorial direction. These diastereomeric adducts could be readily separated by column chromatography. Thermolysis of the individual two major diastereomeric carbinols at 140 °C gave d- and /-menthone, respectively, in high enantiomeric purities (90-93% ee). This methodology has been successfully applied to the resolution of other 2-substituted cyclohexanones as well as other chiral ketones that have served as advanced synthetic intermediates for the synthesis of natural products.^^"^^
314
STEPHEN G. PYNE racemic-menthone + lithiated (+)-(S)-2b H
»H
m ^ ^ / ^ •'\si,i:^x::z^ ^nod:::^ OH > ^ (50%)
OH (40%)
I 140 °C
dl/ (10%)
I 140°C H
[ICH3 + 0^/3^
[ICH3 + o^C^zZ-
(/•menthone (90% e.e.)
/-menthone (93% e.e.)
B. Synthesis of Alkenes via Reductive Elimination
Reductive elimination of P-hydroxy sulfoximines with aluminum amalgam in acetic acid gives alkenes in good yields^^In one study, the resolved carbinol adducts of the ketone 93 and (+)-(5)-2b were individually treated with aluminum amalgam in acetic acid to give natural (-)-|3-panasinsene and its antipode in high enantiomeric purity.^ ^ C. Asymmetric Synthesis of Alkenes with Axial Chirality
Axially chiral vinyl sulfoximines have been prepared with high diastereoselectivity (> 99:1) by asymmetric elimination of LiOSiMe3 from P-siloxy sulfoximines.^^ For example, addition of lithiated (+)-(5)-2a to ketone 94 gave the P-hydroxy sulfoximines 95 with a 99:1 diastereoselectivity. When the dianion of 95 was quenched at -78 °C with chlorotrimethylsilane, the vinyl sulfoximine 97 was
Q;.oH| CHr\'^^
:^^
NSiBu^Phg
60
Q R R = Me. Bu ^ (e.e. 97-100%)
Chiral Sulfoximines
315
1.lithiated(+)-(S)-2b 2. separation of diastereoisomers by chromatography
^ 93
AI(Hg) HOAC
NMe
(-)-p-panasinsene
isolated in 69% yield with 99:1 diastereoselectivity. Similarly the dianion of the a-methyl sulfoximine analogue of 95 (96) gave the a-methylvinyl sulfoximine 98 in 73% yield with a high diastereoselection (99:1). The geometry of the alkene is determined solely by the chirality at sulfur of the P'hydroxy sulfoximines; 99 and 101 were converted to the (Z)- and (£^-vinyl sulfoximines 100 and 102, respectively, in high yield and with a high product diastereoselection (99:1). These vinyl sulfoximines undergo nickel-catalyzed cross-coupling reactions with organometallic reagents to give optically active alkenes (see Section V.D for details).
H 94 H
H O ^ "HT ^
T
1.2n-BuLir 95R = H 2. Mel L 96 R = Me
2.TMSCI
O
ki H
fS
97R = H 98 R = Me J,.Ph NMe
316
STEPHEN G. PYNE
R2O
99
100
RgO-^^
101
102
1 [1= 4r s= -^-r NMe
^Ph
D. Directed Simmons-Smith Cyclopropanations
Resolved P-hydroxy sulfoximines derived from cyclic enones undergo diastereoselective Simmons-Smith cyclopropanation reactions to give, after thermolysis, cyclopropylketones in high enantiomeric purity (94-98%). Cyclopropanation occurs syn to the hydroxyl group of the P-hydroxy sulfoximine. This method is less diastereoselective for acyclic enones7^
(94% e.e.) E. Directed Osmylations
Osmylation of diastereomerically pure P-hydroxy sulfoximines, derived from 2a and cyclic enones, with a catalytic amount of osmium tetroxide (5 mol%) and trimethylamine A^-oxide (1.5 equiv) gives diastereomerically pure triols which on thermolysis yield 2,3-dihydroxy cyclic ketones in high enantiomeric purity ("100%" ee). Osmylation occurs syn to the sulfoximine group.^"^
HO OH F. Enantioselective Reactions
For the use of optically active P-hydroxy sulfoximines as ligands for enantioselective catalytic reactions, see Section VII.
Ch ira I Sulfoxim ines
317
IV. ALLYLIC SULFOXIMINES A. Synthesis Johnson disclosed the synthesis of the first reported allylic sulfoximine 104a in 1979 5 Treatment of racemic phenyl N-methylbenzenesulfonimidate 103 (X=OPh) with allyl lithium at 0-3 °C gave racemic 5-allyl-A^-methyl-5-phenylsulfoximine 104a in 71% yield. Harmata^^ has used a method related to that developed by Johnson^ to prepare the allylic sulfoximine 104b from the reaction of allyllithium with the sulfonimidoyl fluoride 103 (X=F). The yield, however, was low (20%). In 1991 Gais reported a useful method for preparing allylic sulfoximines via base-catalyzed (LiOMe, 3 equiv THF/toluene/n-hexane) isomerization of vinyl sulfoximines.^^ The combination of KOMe/THF was also found to be effective.^^ Vinyl sulfoximines can be readily obtained from the condensation of lithiated sulfoximines with aldehydes and ketones followed either by dehydration of the resulting P-hydroxy sulfoximines by treatment with methanesulfonyl chloride/triethylamine and then elimination of the resulting mesylate with DBU^^'^^ or triethylamine^^ or by trapping the intermediate lithium P-alkoxy sulfoximine with trimethylsilyl chloride^^ or methyl chloroformate^^ followed by elimination of the P-oxygen substituent with n-butyllithium^^ or potassium r^rr-butoxide,^^ respectively. The former method, using DBU, gives mixtures of the vinyl sulfoximine 105 and the allylic sulfoximine 106. Treatment of this mixture with KOMe/THF gives the allylic sulfoximines 1067^ The (E) isomer of 106 is usually the only or major isomer formed.^^ Gais^^ disclosed a useful method for preparing enantiomerically and diastereomerically pure (£)- or (Z)-allylic sulfoximines from (+)-(5)-5-(chloromethyl)/V-methyl-5-phenylsulfoximine 107. For example, treatment of 107 with (£)- or (Z)-l-propenyl cuprates 108a and 108b respectively gave the corresponding (£)and (Z) allylic sulfoximines 109a and 109b. Unfortunately, the yields were not high for these reactions. This method is also successful for the preparation of the corresponding 5-benzyl sulfoximine from the reaction of 107 with Ph2CuLiLiCN. Treatment of enantiomerically pure cyclic sulfonimidates 6 and 7 with allyllithium or allylmagnesium bromide gives optically active allylic sulfoximines 8 and 9 (R = allyl) as described in Section II.A.^^ The reactions of A^-phenyl-5-(methylphenyl)sulfoximidoyl chloride 109 with allyltrimethylsilane or allyltributylstanO II
NR 103 X = OPh, F
-104a 71% X = OPh, Ar = Ph, R = Me 104b 20% X = F, Ar = p-Tol, R = Ph
318
STEPHEN G. PYNE O II .S;^..,ph Me' NR
1.n-BuLi/THF •
2. R^CHgCOR^ 3. MsCi/EtaN 4. DBU
R = Ts, COgMe, Me
KOMe THF R2
O
11 NR 106
nane in the presence of aluminum chloride gave mixtures of the benzothiazine 110 and the allyl sulfoximine l l l 7 ^ The organostannane gave better yields of the ally lie sulfoximine 111. This method was successfully used to prepare the A^-<9-methylphenyl, A^-c?-bromophenyl, and A^-benzyl analogues of 111. The imination of racemic allyl phenyl sulfoxide 112 with O-mesitylsulfonylhydroxylamine (MSH)^^ using a modification of the procedure described by Johnson,^^ gave the allylic sulfoximines 113 in poor yield (29%).^^ This compound was readily converted to the A^-tosyl or A^-silyl derivatives 114.^^'^^ The optically active version of 114 (R = Ts) can be prepared from (5)-2b and acetaldehyde as shown below.^^ B. Lithiated Allylic Sulfoximines Structural
Studies
The crystal structures of three lithiated allylic sulfoximines have been reported."*^'^^ The X-ray crystal structure of lithiated 116/12-crown-4 complex showed solvent-separated contact ion-pairs of [Li(12-crown-4)2]"*' and the allylic sulfonimidoyl anion.^^ The anion adopts a conformation in which the p orbital at
1?
108a
•'/iDh -^ V'Ph •*
NMe
ff
108b ^1-
NMe
(35%)
(27%)
107
109a
Me
(''^^-^cuu.ucN ( f y .
CuLi.LiCN
108a
Me
108b
S;;"'Ph NMe 109b
Chiral Sulfoximines
p-Tol—S-CI II NPh
+
319
AlCU
y^ff'^^^^^
CH2CI2, -78 °C
109
-X+
111 [X = TMS, 2 7 % , X = SnBUg, 77%]
110a, X = T M S , 4 7 % 110b, X = SnBu3,4%
O II 112
MSH
O II
CH2CI2 (29%)
NH 113
O II NR 114(R = R3SI,Ts)
l.n-BuLi/THF (S)-2b
2. CH3CHO ^ 3. MsCI/EtgN 4. DBU KOMe THF O II .^^^^^^ NTs 114
320
STEPHEN G. PYNE
R
C^^ .NRi ^^^ >h H
.©
116, R = Me, R^=TMS 117, R = Ph, R^ = Me
Ca is gauche to both the oxygen and nitrogen substituents of the sulfur atom. This conformation suggests a stabilizing nc-a*-interaction between the nonbonding orbital on the a-carbon atom and the a* orbital of the S-Ph bond. A similar gauche conformation has been found in the solid state structure of racemic lithiated ll?."^ Lithiated 117 forms a dimer structure in which two allylic sulfonimidoyl anions with opposite chirality are linked by N-Li-O bridges to give an eight membered ring with the atom sequence (Li--N-S~0)2. Reactions with Alkylating Agents The alkylation of the lithiated allylic sulfoximine 118 (R=Ph or CH2Ph) is completely regioselective and gives only a-alkylation products 119.^^ The products were formed as mixtures of diastereoisomers, but the diastereomeric ratios were not reported. Lithiation and then methylation of the optically active allylic sulfoximine 120 gave the a-alkylation products 122 as a single diastereoisomer.^^ The stereochemistry of this compound was deduced by its transformation to a compound of known stereochemistry. The stereochemical outcome of this reaction was rationalized as arising from methylation of the lithiated species 121 in which the p orbital at Ca is gauche to both the oxygen and nitrogen substituents of the sulfur atom. Methylation of 121 would be expected to occur syn to lithium and anti to the S-Ph group.
ft
l.n-BuLiyTHF NR
118 (R = Ph, CHaPh)
, p1y 2 " ^ (54-81%)
r"Toi R^
NR
119 (R^ = Me, Et, allyl)
Chiral Sulfoximines
321 Mel
I
Li
.NTs 120
Ph
1. n-BuLi >2. Mel
121
P\.NTs Ph
.NTs
122
Reactions with Aldehydes and Ketones: a- versus
H Vh
y-Regioselectivity
The reaction of lithiated 123 with benzaldehyde gave a 5.3:1 mixture of the a-adduct 124 and the y-adduct 125, while a similar reaction with pivaldehyde produced only the a-adduct 124^^ These products were formed as mixtures of diastereoisomers, but the diastereomeric ratios were not reported. In related examples, a-adducts were exclusively obtained from the reaction of lithiated N-tertbutyldiphenylsilyl^^ and A^-methyl^^ allylic sulfoximines with aldehydes. Again, these products were mixtures of diastereoisomers. Lithiation of racemic N-tosyl allylic sulfoximine 126 followed by quenching the reaction at -78 °C with benzaldehyde or isobutyraldehyde gave the a-adducts 130a and 130b, respectively, as mixtures of diastereoisomers.^"* Interestingly, when these reactions were performed with an excess of the aldehyde (2 molar equiv) and
1. n-BuLi/THF ^N^Tol NPh 123
.\-Tol NPh
2. R^CHO
Sr^Tol NPh 125
124 R^ = Ph, 79%, 124:125 = 5.3:1 R^ = BuS 63%, 124:125 = 100:0
STEPHEN G. PYNE
322
warmed to room temperature, the novel 1,3-dioxanes 131 and 132 were formed in good yields [81 % (R = Ph) and 62% (R = Pr*)] as 1:1 mixtures of diastereoisomers. The 1,3-dioxanes 131 and 132 were proposed to arise from the anionic y-adduct 129 that must be formed from the kinetically favored anionic a-adduct 128 via a reversible aldol-type reaction.^"* In contrast, titanated allylic sulfoximines, which can be prepared from lithiated allylic sulfoximines by transmetallation with chlorotris(isopropoxy)titanium, react with aldehydes to give y-adducts in a highly
ff
rPh <;^^^^ NTs
126
o
LDA -78 "C 30min
?,
RCHO
NTs U* 127
-78 °C lOmin
ft 0°C NTs
R^ ^ 0 128
129
NH4CI -78 "C
f
OH •CHa
130a (R = Ph), d.r. = 56 :16 :16 :12 130b (R = PI*), d.r. = 39 : 39 : 22
H
NTs
H 131a (R = Ph) 131b (R = Pr')
O
H
r^ S;^"'NTs •CH2 O H
132a (R = Ph) 132b (R = Pr')
Chiral Sulfoximines
323
regioselective and diastereoselective manner.^^"^^ For example, lithiation of the enantiomerically pure allylic sulfoximines 133 and 134 with Az-butyllithium followed by the addition of chlorotris(isopropoxy)titanium and then an aldehyde gave exclusively the anr/-(Z)-Y-adducts 135 and 136, in a high diastereomeric purity (> 97%). The stereochemical outcomes of these reactions seem to be controlled predominantly by the configuration at the sulfur atom rather than that of the valine substituent. It has been suggested that transmetallation of lithiated 133 produces the configurationally stable titanium compound 137.^^'^^ These reactions are also highly regio- and diastereoselective with related cycloalkenyl sulfoximines.^^ The reactions of titanated 133 and 134 with (R) and (5) TBDMS-protected lactaldehyde [MeCH(OTBDMS)CHO] are also highly diastereoselective. Very high levels of diastereoselectivity (> 98%) were observed when the facial selectivity of the allylic sulfoximine anion matched that of the chiral aldehyde (the matched case).^^'^^ In the mismatched cases the diastereoselectivities were less but still
OTMS 133
135
^^:v^R OTMS 134
Ln-BuLi/THF •79 ^C ^ 2.CITi(OPrV0°C 3. R^CHO,-78°C 4. BU4NF
1. n-BuLi/THF Tol'"«.S^^^'^ -78^0 N R 2.CITi(OPrV0°C OH 3. R^CHO,-78°C 4. BU4NF 136
Ti(0PH)3 ''OTMS 137
STEPHEN G. PYNE
324
relatively high (d.s. 80-98%). In the case of the titanated allylic sulfoximine anion 138, the asymmetric induction from the chiral aldehyde was overridden by that of the sulfoximine anion and high diastereoselectivities were observed in both the anti-Cram and Cram products 139a and 139b, respectively. Optically active lithiated A^-methyl allylic sulfoximines 140 that have undergone transmetallation with chlorotris(isopropoxy)titanium also react with aldehydes to give anti-iZ)y-adducts 141 in a highly regioselective and diastereoselective manner (de > 95%).^^ The yields, however, are generally less than 50% suggesting that a diorganotitanium compound is formed that only transfers one allyl sulfoximine ligand to the aldehyde. In contrast, when the transmetallation is performed with chlorotris(diethylamino)titanium the resulting titanium species reacts with aldehydes to give {E)-syn'a-2idd\xcis 142 with very high regioselectivity (> 95%) and diastereoselectivity (> 95%).^^ The corresponding (Z)-isomer of 140 (R^ = Pr\ R2 = H) gives the corresponding (Z)-j)'n-a-adducts. In all cases the yields were good (70-76%).
Ti(0Pri)3
138
O
OTBDMS
anfrCram
Me ^OTMS 139a(cls>98%)
OTBDMS
Cram
OTBDMS
Me ^OTMS 139b(ds>98%)
OTBDMS
Chiral Sulfoximines
325 Ln-BuLi/THF.-78°C OH
H
^SC^^ Ph 140
3. R^CHO,-78 °C . •
1.A7-BuLiyTHF,-78°C 2. CITi(NEt3)3, -20 °C 3. RCHO, -78 °C
R2 = H
R2
« R2
Ri
^S^NMe
141
y P h .
PH 142 HO
Reactions with Enones: 1,2- versus 1,4-Addition and the Synthesis of Cyclopropanes
The reactions of lithiated sulfoximines 143 with cyclic enones give mixtures of regio- and diastereoisomers. The regioselectivity is dependent on the nature of the N-substituent on sulfur.^^'^^'^^ N-Tosyl derivatives give exclusively a-1,4-addition products (144), while this type of adduct is slightly favored in the case of A^-phenyl derivatives. N-r^rr-Butyl-diphenylsilyl (TBDPS) derivatives, however, favor y-1,4adducts (145). In all cases the adducts were formed as mixtures of diastereoisomers except in the case of the Y-l,4-adducts 145 (n= 1, Ar = Tol, R = Ph) which were obtained in low yield as a single diastereoisomer.^^ Treatment of a solution of racemic lithiated A^-tosy 1-5-ally 1-5-phenylsulfoximine 147 at -78 °C with the acyclic enones 146a-c (1.2 equiv) for 3 min gave, after quenching at -78 °C with acetic acid, the racemic 1,4-a adducts 149a-c in modest to excellent yields. The product diastereoselection ranged from 90:10 to 94:6. The relative (3/?*, 4/?*, S5*) stereochemistry of 149a was determined by X-ray structural analysis and has been rationalized as occurring via the transition state structure A in which the largest groups on the two reacting partners (R^ and the sulfonimidoyl group) are anti in order to minimize steric interactions.^^'^ Warming a solution of the anionic adducts 148a-c to room temperature for 1 h gave the racemic vinylcyclopropanes 150a-c in good yields (83-91% after column chromatography) and, in the case of the cyclopropyl phenyl ketones 150a and 150c, in lower diastereoselectivity than their respective Michael adducts 149a and 149c. In contrast, the diastereoselectivity observed for the cyclopropyl methyl ketone 150b was essentially identical to that found in its related Michael product 149b. Cyclopropane 150b was easily obtained diastereomerically pure by column chro-
STEPHEN G. PYNE
326 Ln-BuLiATHF -78 °C (CH2)
143
(CH2)i
144, S^ 0*^1 NR ^ Ar
145
Ar
R
n
144:145
reference
Tol
Ph
1
58:42
75
Tol
Ph
2
60:40
75
Ph
TBDPS
1
15:85
81
Ph
TBDPS
2
16:84
81
Ph
Ts
1
100:0
59
1 Ph
Ts
2
100:0
59
matography. Enantiomerically enriched (15, 2/?, 35)-150b was prepared from the reaction of enantiomerically enriched (5)-147 (ee 94%) and 146b under identical reaction conditions and procedures as described above. ^H NMR studies using the chiral shift agent europium tris[3-(heptafluoropropylhydroxymethylene)-(+)-camphorate] indicated that the enantiomeric purity of 150b was 95% after correction for the enantiomeric purity of (5)-147. The stereochemistry of 150b was established by NMR and NOE difference experiments. The stereochemistry of 150b was that expected for an intramolecular nucleophilic displacement reaction of the sulfoximidoyl group from the intermediate 148b, with inversion of stereochemistry at the carbon bearing the sulfonimidoyl group. The reaction of racemic 147 with (/?)-carvone, initially at -78 °C followed by warming to room temperature for 1 h, gave the vinylcyclopropane 151 in 72% yield and moderate diastereoselectivity (d.r. = 75:25). The stereochemistry of the major diastereoisomer shown in structure 151 from ^H NMR studies was that expected based on the stereochemical outcome of the reaction of racemic 147 with the achiral cyclic enones 146 and is consistent with our previously proposed chelated transition state^ for cyclic enones (compare with the transition state B). C. Rearrangements to Allylic Sulfinamides and Related Reactions
While the [2,3] sigmatropic rearrangement of allylic sulfilimines 152 to allylic sulfenamides 154 has been well documented,^^ the related thermal rearrangement
327
Chiral Sulfoximines
+ Li _
O
THF
R2
-78 °C
//
146
TsN
TsN 148
147
TsN' 149a (90%, d.r. = 93 : 7) 149b (66%, d.r. = 90:10) 149c (69%, d.r. = 94 : 6)
R2
R1
A
- ^
a; R U R2 = Ph b ; R U M e , R2 = Ph c; Ri = Ph, R2 = Me
O 150a(83%, d.r. = 79:21) 150b (91%, d.r. = 91 : 9) 150c (84%, d.r. = 84:16) of ally lie sulfoximines 153 to ally lie sulfinamides 155 is not generally a kinetieally favored process. For example, Tamura,^^ Harmata,^^ and Pyne^^ have reported that the simple allylie sulfoximines 156 were thermally stable in refluxing toluene solution. These results were in contrast to MNDO^^'^^ and ab initio^^ calculations which suggested that allylie sulfinamides should be thermodynamieally more stable than their isomeric allylie sulfoximines. These calculations indicated that this rearrangement process was unfavorable because of a high kinetic energy barrier. In 1994, Gais^"^ reported that enantiomerically pure y-phenyl-substituted allylie sulfoximines undergo partial rearrangement to their isomeric sulfinamides with retention of configuration of the S-atom. For example, thermolysis of the enantiomerically pure allylie sulfoximine 158 at 85 °C for 112 h gave, after chromatography, enantiomerically pure 158 and minor amounts of the two isomeric enantiomerically pure allylie sulfinamides 159 and 160 in almost equal yields. This rearrangement was thought to involve the ion-pair intermediate 161 which is stabilized by the phenyl substituent. In 1996, Pyne and Dong'^^ reported that the thermolysis of (5,5)-121 (ee 94%) in refluxing THF for 6 h gave a mixture of the rearranged allylie sulfinamides 163a
328
STEPHEN G. PYNE E- Li,
^--....,
1/
E*=146b Ph
Li"'"0
H
Ph^
^^
CO
^/^CH2
Me 1.2%N0E (1S, 2/=?, 3S)-150b (ee 95 %) and 164a. Exposure of this mixture to silica gel chromatography gave an inseparable 45:55 mixture of the sulfonamides 163b and 164b in 73% yield. These compounds were separated by HPLC and were determined to be 87 and 88% enantiomerically pure, respectively. The absolute stereochemistry of 164b was established by chemical correlation with a molecule of known absolute configuration. This thermal rearrangement was thought to occur via the intimate ion pair 162 with the anion being produced initially on the lower face of the cation for stereoelectronic reasons^^ or via a competing [2,3] sigmatropic rearrangement to give 163a. In 1995 Pyne and Dong^^ found that the ally lie sulfoximine 165 underwent a facile and completely regioselective and efficient rearrangement to the allylic sulfinamide 166 in the presence of tetrakis(triphenylphosphine)palladium(0) ((PPh3)4Pd) catalyst (5 mol%) at room temperature. Mild base hydrolysis of the reaction mixture (10% aqueous sodium hydroxide/methanol, 1:10, room temperature, 2 h) gave pure sulfonamide 167 after purification by column chromatography (silica gel) in 90% overall yield.
Chiral Sulfoximines
329 THF -78 °C to rt
NOE
NOE 151 72 %, d. r. = 75 : 25
B
R^Y^
R V ^
heat
^ "
X 154X = :
152X=: 153X = 0
155X = 0
p-r-.Y;,<^ Me 6 '
• NMe ciCHzCHzCI
158
85°C, 112h
Me
Me 161
O
156
157
(R^ = H or Ph R^ = aryl)
158 + (62%)
^N'
Pff Me
Me
159 ^ (11%)
Me
Me
160 (10%)
Ph
STEPHEN G. PYNE
330 1.THF, reflux, 6h '••
(S, S)-121
»-
2. silica gel SO(Ph)NTs (73%)
N(Ts)R 163a (R = SOPh) 163b (R = H)
162
© N(Ts)SOPh|
N(Ts)R 164a (R = SOPh) 164b (R = H)
1.(PPh3)4Pd(cat) H Ph
HQ,.^ H H
Ph
R
THF.RT, 10min TsN ••;& >>^Ph 165
r.u'^^^'^VX^ NTs ^ Ph ^^ ^^ ^^ 2. NaOH/H20/MeOH, 166 (R = SOPh) RT.2h 167 (R = H) (90%)
This mild rearrangement process was found to be general for both secondary 168^^ and primary ITO'^^ sulfoximines and in each case the reactions were completely regioselective and gave the primary allylic sulfonamide 172. The overall yields ranged from 79 to 95% and the reactions were found to be compatible with other functional groups (e.g., hydroxy and carbonyl groups). It was assumed that the above palladium-catalyzed rearrangements occur via the allyl-palladium cation complex intermediate 169 followed by attack of the ambident sulfmamide anion as a nitrogen nucleophile at the least hindered terminus of the allyl cation species to give 171. These palladium-catalyzed reactions also work well with cyclic allylic sulfoximines. For example, the cyclic sulfoximines 173 undergo palladium(0)-catalyzed rearrangement to their corresponding allylic sulfmamides which on mild base hydrolysis give exclusively the primary allylic sulfonamides 174 in excellent overall yields.^^ Optically active (5, 5)-121 gave the optically active sulfonamide 164b in 94% enantiomeric purity with overall retention of configuration at the allylic a-carbon.^^
331
Chiral Sulfoximines
TsNO'
1. Pd(0) THF.RT
1 . Pd(0) THF.RT "Ph 2.NaOH(aq) 168 MeOH, RT
R \ ; ^ ^ \ ^ ^ S ^ -r-Ph NTs 2. NaOH(aq) 170 MeOH, RT
e N(Ts)SOPh 169
H
S(0)Ph
OH-
Rv.^;;:;:>\^NTs 171
172(79-95%)
"
Oh
NTs
NHTs
11 • cat. Pd(0) 2. NaOH(aq) (CH2)n-' (89-90%)
174
n = 1-3
1. Pd(0)
LN
Pd
A ,Me H
2.0HSO(Ph)NTs (S, S)-121
© N(Ts)SOPh
164b (ee 94%)
N(Ts)H
332
STEPHEN G. PYNE
Treatment of the (E) a-sulfonimidoyl P,Y-unsaturated ketones 175a or 175b or the ester 175c with 10 mol% of freshly prepared (PPh3)4Pd in dry THF solution at room temperature for 1 h gave the unstable allylic sulfinamides 176a-c. Mild methanolysis of the reaction mixtures with triethylamine/methanol at room temperature gave pure (£)-sulfonamides 177a,b (y-amino a,P-unsaturated ketones) and the (£)-carbamate 177c (y-amino a,P-unsaturated ester) after purification of the crude reaction mixtures by column chromatography (silica gel) in overall yields of 32-68%.^'^ Optically active 177 (R^ = Ph, R^ = Cbz, ee = 20-62%) can be obtained from racemic 175 (R^ = Ph, R^ = Cbz) using a chiral ligand for the palladium catalyst.^^ Interestingly, in the presence of a bidentate ligand and palladium(O), 175 (R^ = Ph, R^ = Ts) gave not the expected allylic sulfmamide but the isomeric allylic sulfinimidic acid ester.^^ In principle, allylic sulfoximines can be used as substrates for the allylation of an external nucleophile (Nu) if that nucleophile can compete with the sulfmamide anion C for the palladium(O) complex B or if the formation of D is reversible. In 1997 Pyne and co-workers^^ reported that stabilized carbon and nitrogen nucleophiles can be efficiently allylated in a regioselective manner using allylic sulfoximines and palladium(O) catalysis (Eq. 5).
+ Pd(0)
Pd(0)
NTs
(5) Pd(0) +
Treatment of the racemic allylic sulfoximines 178,179, and 121 with 5 mol% of tetrakis(triphenylphosphine)palladium(0) (Pd(PPh3)4 in THF at room temperature for 10-30 min, in the presence of the nucleophiles dibenzylamine, sodium diethyl malonate, or the lithium salt of tert-buiyl A^-(diphenylmethylene)glycinate (BDMG) (1.2 molar equiv) gave the allylated products 180-184 in generally good yields with a good to high regioselectivity (Table 10). In general, the nucleophile added to the least substituted and/or sterically less demanding position of the allylic moiety. In the case of allylic sulfoximine 178, it was found that compound 180 could also be obtained in a similar yield by first converting 178 in situ to its isomeric allylic sulfmamide D (R = H, Eq. 5) by initial treatment of 178 with Pd(PPh3)4 in THF at room temperature for 15 min, followed by the addition of dibenzylamine. Thus, the allylic sulfmamide D (R = H, Eq. 5) is readily converted to its allylic cation A (R = H, Eq. 5) in the presence of Pd(0). The cyclic substrates 179 and 121 gave a mixture of regioisomers 181 and 182 and 183 and 184, respectively. The reaction of the secondary allylic sulfoximines 179 with dibenzylamine was completely regioselective and gave exclusively 181.
Chiral Su 1fox imines
333
Pd(PPh3)4 R i ^
RV
THF R^N^-f^O Ph 175a 175b 175c
Ph,F^ = Ts [a; b; R2 == Ph,R3 = COzMe c; R^ =•• OMe, R^= Ts R2 ==
MeOH
T
PhSO'
EtgN
•R3
H-^^R3
177a 177b 177c
176a 176b 176c
Starting compound 175a;R^=n-Bu 175b; R^ = n-Bu 175a; R^ =n-pent 175b; R^ = n-pent 175a; R^ =n-hexyl 175c; R* = Et
product 177a;R^=n-Bu 177b; R* = n-Bu 177a; R^ =n-pent 177b; R^ = n-pent 177a; R^ =n-hexyl 177c; R^ = Et
yield(%f 32 64 60 49 68 57
after chromatography
In the case of the reaction of the racemic ally lie sulfoximine rac-lll with the lithium salt of BDMG (Table 10, entry 5, Nu = CH(N=Ph2)C02Bu')), a 90 : 10 mixture of regioisomeric adducts 183 and 184 resulted. The major regioisomeric product 183 (Nu = CH(N=Ph2)C02Bu^) was a 74:26 mixture of diastereoisomers. The reactions of the secondary allylic sulfoximines 185-187 with dibenzylamine were completely regioselective and afforded the 1,4-amino alcohol 188 and the Y-amino enones 189 and 190, respectively (Table 11, entries 1-3). Treatment of enantiomerically pure (S, 5)-121 with dibenzylamine or sodium diethyl malonate in the presence of palladium(O) gave the essentially enantiomerically pure (ee > 98%) products (5J-191 (Nu = Bn2N) and (5)-191 (Nu = CH(C02Et)2), respectively (Table 12). Based on the sign of the specific rotation of 191 (Nu = Bn2N), the reaction of (5, 5)-121 with Pd(PPh3)4/dibenzylamine was shown to have occurred with overall retention of configuration at the stereogenic carbon, consistent with attack of the nucleophile on the palladium-allyl cation complex F, anti to the sterically demanding palladium(II) moiety (Eq. 6). The stereochemistry assigned to 191 (Nu = CH(C02Et)2) was made by analogy to that of 191 (Nu = Bn2N) and the known tendency of malonate nucleophiles to add anti to the palladium moiety in cations such as F (Eq. 6).
STEPHEN G. PYNE
334
Table 10. Allylation on Nucleophlles by Allylic Sulfoximines 178,179, and 121 Entry
Substrate^
Nucleophile^
fi^Ph
Products (YielcF (%))
^^;^-v^NBn2
NTs
180(67)
178
NTs
179
181 182 Nu = BngN (73); 181 : 182 = > 98 : < 2
3
179
B
Nu = CH( C02Et)2 ( 77); 181 :182 = 89 : 11
4
179
C
Nu = CH(N=CPh2)C02Bu^ (80); 181 .182 = 9 4 : 6
Mefi^Ph ^ ^
Me
^NTs
183 184 Nu = CH(N=CPh2)C02Bu^ (66); 183:184 = 90 : 10 (183; d.r .= 74 : 26)
rao121
Notes: ^Unless Indicated ail compounds are racemic. ^Nucleophiles: A, dibenzylamine, B, sodium diethyl malonate, C, lithium salt of te/t-butyl N(dlphenylmethylene)glycinate. ^After purification by column chromatography on silica gel.
^fd-^
Pd(0) -^:f^^\ inversion •n SO(Ph)NTs {S. S)-121
S
^H^H
Nu
(6)
inversion
0N(Ts)SOPh overall retention
Chiral
335
Sulfoximines
Table 11.
Allylation on Dibenzylamine by Allylic Sulfoximines 1 8 5 - 1 8 7
Entry
Substrate^
Nucleophile
Products (Yield^ (%))
OH
Ph
BugNH
BnaN,
Ph 188(66)
BUgNH
BnaN^^^x^Lp^
NTs
189(43)
186 CgH BUgNH
187
B'^'^
N^^
190(62)
Notes: 'Unless indicated, all compounds are racemic. ''After purification by column chromatography on silica gel.
Table 12.
Allylation on Nucleophiles by Optically Active Allylic Sulfoximine (S,S)-121
Entry
Substrate O Men ^
^
Products (Yielcfi (%))
Nucleophile'
.Ph
Me
^NTs
(S, S)-121
191
192
Nu = BnaN (60); 192 :192 = > 98:<2(191;e.e. >98%) (S. S)-121
B
Nu = CH( COgEOg (62); 191: 192 92:8(191;e.e. >98%)
Notes: ^Nucleophiles: A, dibenzylamine; B, sodium diethyl malonate. ''After purification by column chromatography on silica gel.
STEPHEN G. PYNE
336
Ph
TsN
193; R = Ph 194; R = Me 80-85° C 48 h
TsNH
195; R = Ph (49%) 196; R = Me (45%)
+ TsNHg Ph 197; R =Ph(10%) 198; R = Me (12%)
199 (35% from 193; 49% from 194)
Thermolysis of the racemic Y-sulfonimidoyl ketones 193 and 194 gave the 2,3-dihydrofurans 195 and 196, respectively.^^ When a sample of 193 was being dried for combustion analysis at 50-60 °C, it was noticed that the sample rapidly turned black. ^H NMR analysis of the black solid indicated the formation of a novel product. When this thermal process was repeated on a preparative scale at 80-85 °C for 48 h, the novel 2,3-dihydrofuran 195 could then be isolated, as a single diastereoisomer, in 49% yield after purification by column chromatography. In addition, the known sulfonamide 197 and 4-methylbenzenesulfonamide 199 were also isolated in yields of 10 and 35%, respectively. The structure of compound 195 was unequivocally determined by a single-crystal X-ray structural analysis that showed that the allylic carbon (C(4)) in 193 had undergone inversion of stereochemistry on cyclization to 195. When a solution of 193 was heated to reflux in toluene for 2 h, the same three products 195,196, and 199 were isolated in yields of 42, 18, and 32%, respectively. Heating racemic y-sulfonimidoyl ketone 194 in the solid state under similar conditions afforded the analogous dihydrofuran 196 in 45% yield plus the sulfonamides 198 (12%) and 199 (49%) after purification by column chromatography. The dihydrofuran 196, however, was an 87:13 mixture of trans and cis isomers, respectively, that could not be separated. A tentative mechanistic scheme, involving cyclization of the enol form of the ketone and nucleophilic displacement of the sulfoximine group, was proposed to account for this chemistry. ^^
Chiral Sulfoximines
337
R2CuLi MeN
R2CuLi 200
BFsEtaO
HH
HH 202
D. Allylic Substitution Reactions with Organometallic Reagents
Gais has reported on the substitution reactions of optically active endocyclic allylic sulfoximines with organocopper reagents. For example, the allylic sulfoximine 200 on reaction with homocuprates (RjCuLi, where R = alkyl) gave exclusively the endocyclic alkenes 201 via an "Sj^2-like" displacement reaction. When these reactions were conducted in the presence of BF30Et2 (1 equiv), exocyclic alkenes 202 were formed (> 98% regioselectivity) via an "Sj^2'-like" displacement of sulfinamide.^^ Since their initial communication, a full account of this study has been disclosed.^^^ In general, the reactions of primary endocyclic allylic A^-methyl5-phenylsulfoximines with organocuprate-LiI reagents give products of a-substitution while the addition of boron trifluoride results in the formation of Y-substitution products.
H
n-BuCu. Lil ,Ph p2 O 203
NMe
H
n-Bu ,CHp
THForEtgO
n-Bu R2
205
204
O^ N(H)Me 206
338
STEPHEN G. PYNE
^
n-BuCu.LiI ,1^2 o 207
NMe THForEtgO
^'
^'
(«) R2 208
J!^2 209
O^ ''N(H)Me 206
Enantiomerically pure (£)-acyclic sulfoximines 203 react with n-BuCuLiI in the presence of boron trifluoride to give almost exclusively y-substitution products 204.^°^ The enantiomeric purities of the y-substitution products ranged from 2 to 72% depending on the nature of the solvent and the substituents R^ and R^ in 203. (5)-N-methyl-5-phenylsulfmamide 206 was isolated in high yields (> 80%) and enantiomeric purity (> 95%). The corresponding (Z)-acyclic sulfoximines 207 react with n-BuCuLiI in the presence of boron trifluoride to give almost exclusively y-substitution products 208.^^^ The enantiomeric purities of the y-substitution products ranged from 12 to 92% depending on the nature of the solvent and the substituents R^ and R^ in 207. (5)-N-methyl-5-phenylsulfmamide 206 was again isolated in high yields (> 80%) and enantiomeric purity (> 95%). Further studies revealed that the nature of the yV-substituent had little influence on the regio- or enantioselectivities of these reactions.
V. VINYL SULFOXIMINES A. Synthesis The synthesis of vinyl sulfoximines via the elimination reactions of P-hydroxy sulfoximines^^'^^'^^'^^'^^ has been discussed in Section IV.A. The method of Craig-^"^ is particularly useful for the preparation of A^-unsubstituted vinyl sulfoximines 210 which can be readily substituted at nitrogen by reactions with a number of reagents (see Section II.A) including trifluoromethane sulfonic anhydride (triflic anhydride).^^ A/^-Tosyl vinyl sulfoximines 211 can be prepared in a one-pot reaction via an in situ Wadsworth-Emmons procedure from 5-methyl-5-phenyl-A^-tosylsulfoximine by sequential treatment at -78 °C with n-BuLi, potassium r^rr-butoxide, and diethyl chlorophosphate, followed by addition of an aldehyde and warming to 0 °C.^^^ The resulting A^-tosyl vinyl sulfoximines 211 are formed almost exclusively as the (£)-isomer in good overall yields (60-91%).
Chiral Sulfoximines
^^
339
O
1.n-BuLi/THF
Q
Tf20
Q
y
2. RCHo
y
pz!^
y
\TMS3.MeOCOCI R ^ \ H 4. KOBu\ THF 210
^ ^ ^
^Tf
5. 2M HCI (aq)
A^-Arylsulfonyl-5-ethenyl-5-phenylsulfoximines have been prepared from A^arylsulfonyl-5-methyl-5-phenylsulfoximines by deprotonation and then treatment with Eschenmoser's salt. Treatment of the resulting tertiary amine with an excess of methyl iodide followed by base treatment gave 212a,b in 26-42% overall yields.^^'^^^ Bromination of 212a,b and then elimination with triethylamine gave the a-bromo derivatives 213a,b in 26-32% yields.^^ Paley^^ has reported a method for preparing A^-r^At-butyldimethylsilyl (E)-vinyl-, dienyl-, and enynyl-sulfoximines from the reactions of the /V-r^rr-butyldimethylsilyl p-0-tosyl vinyl sulfoximine 214. Compound 214 was prepared in 60-65% yield in a one-pot reactionfromits corresponding 5-methylsulfoximine via a deprotonation (with lithium tetramethylpiperidine), formylation, and tosylation sequence. Treatment of 214 with "higher-order" cuprates (R2CuCNLi2, R = Me, Et, Bu, Pr^, Ph) in diethyl ether solution gave (£0-vinyl sulfoximines 217 in fair to good yields (50-74%). The yield of 217 (R = Me) could be enhanced from 50% using organocopper chemistry to 87% by employing trimethylaluminum and palladium(O) catalysis. Treatment of 214 with 1-hexenyldimethylalane or divinylethylalane using palladium(O) catalysis gave the corresponding (£)-enynyl- and dienyl-sulfoximines 215 and 216, respectively. Both Gais and Jackson have reported the preparation of a-alkyl and a-trimethylsilyl vinyl sulfoximines 219 by a-lithiation of vinyl sulfoximines 218 with butyllithium or methyllithium followed by treatment with alkyl halides, chlorotrimethylsilane,'^^'^^'^^'^'^^^ or diphenyl disulfide.^^
O yII ^^
Q Tl
1. n-BuLi/THF 2. KOBu^ THF
'^NTos^(^*0)2P(0)CI R 4. RCHO
1.n-BuLi/THF 2.CH2=NMe2l NSOgAr 3. Mel 4. NaHCOa
p. 11
Q M
LBrg
NSOgAr 2- BgN ^ 212a,Ar=Ph, 212b, Ar = 2,4,6-Pr'3C6H2
1 Br
NSOgAr
^^ ^^'^
340
STEPHEN G. PYNE B. Michael Reactions
Organometallic Reagents Enantiomerically pure chiral vinyl sulfoximines having a chiral auxiliary at nitrogen undergo conjugate addition of alkyllithium and organocopper reagents with high asymmetric induction at the P-position (Tables 13 and 14). 11 The stereochemical outcomes of these reactions seemed to be chiefly governed by the chirality at sulfur of the sulfoximine group rather than the chiral norephedrine derived auxiliary. For example, vinyl sulfoximines 220 and 223 underwent conjugate addition of alkyllithium with opposite 7i-face diastereoselectivity (cf. entry 1, Table 13, and entries 1 and 2, Table 14). The stereochemical outcomes of these reactions were rationalized by invoking the initial formation of the complex 226 between the organometallic reagent and the sulfoximine via coordination at the sulfoximine nitrogen. The organometallic reagent may then be directed preferentially to one of the diastereotopic faces of the vinyl group. It was discovered that alkylcopper reagents (RCu) underwent conjugate additions in the opposite diastereofacial sense to alkyllithium and dialkylorganocopper reagents (R2CuLi) and with a very high diastereofacial selectivity (Table 13, entries 3,5-7). By the nature of their preparation [RLi + Cul (0.5 or 1.0 equiv)], both RjCuLi and RCu contain 1 equiv of soluble Lil which can compete with the organometallic reagents for chelation at the sulfoximine nitrogen. The reversal of 7C-facial diastereoselectivity with RCu was explained by the attack on RCu on the Li"*"-chelated species 227 from the least sterically demanding 7i-face. Consistent with this proposal was the reduction in diastereoselectivity in favor of 222 when Lil "free" n-BuCu was employed (Table 13, entry 6). 1.LiTMP/THF 2. DMF/THF
O
O il NTBDMS
NTBDMS 3.TsCI 214 M e g A I — = - -Bu Pd(0) (CH2=CH)2AIEty Pd(0) O y NR 215 (R = TBDMS)
NR 216 (R = TBDMS)
RgCuCNLJg
O II f^
^ ^NR 217 (R = TBDMS)
Chiral Sulfoximines
341 1. n-BuLi or
O
2. R2X or R^ PhSSPh 219 (R' = Me, Ts; l=P = alkyl, TMS, SPh)
218
Further experiments, in which the stereochemical outcomes could be rationalized as arising from analogous coordinated intermediates to 226 and 227, were performed on the enantiomerically pure vinyl sulfoximines 228 and 231.'^ Vinyl sulfoximine 228 underwent conjugate addition of RjCuLi in the expected stereochemical sense, presumably via a coordinated intermediate analogous to 226 (Table
Table 13. Addition of Organometallics ( R ' M ) to Vinyl Sulfoximine 220 0
J..^N
Me
f J .. - ^ MeO
-i,^ Ph
^_p,^^^„p_ -25°CtoO°C 2. NH4CI
220 O
J;.R1 '
H 221
Entry
R
1 2 3 4 5 6 7 8
Ph Ph Ph Me Me Me n-Bu PhCH2CH2
222 R}M n-BuLi n-Bu2CuLi n-BuCu n-BujCuLi n-BuCu n-BuCu(Lir'free") MeCu MeCu
Yield (%) 69 76 71 77 81 68 72 75
Diastereoselection ^221:222; 73:27 86:14 5:95 81:19 5:95 33:67 4:96 5:95
342
STEPHEN G. PYNE Table 14. Addition of Organometallics (RM) to Vinyl Sulfoximine 223 O '"vr^^Me
^^
M e O - ^ ^"'Ph 223
1.RM.THF.
-25 °C to 0 °C 2.NH4CI
O
O
II Q
iph
.,
Q«"»"Ph
224 Entry 1 2
225 RM
Yield (%)
D'\asiereose\eci\on (224:225;
MeLI n-BuLJ
85 82
96:4 95:5
15, entry 1). In the absence of Lil, which could also complex to the sulfoximine nitrogen, the diastereoselection, in favor of diastereoisomer 229, was enhanced (Table 15, entry 2). When 228 was precomplexed with ZnBr2 prior to the addition of Me2CuLi, the reaction proceeded in the opposite stereochemical sense and favored the diastereoisomer 230 (Table 15, entry 3). This result was clearly consistent with attack of Me2CuLi on a zinc-coordinated intermediate analogous to 227. The reactions of 228 with MeCu (Table 15, entries 4 and 5) occurred in the same stereochemical sense as the reaction of 220 and RCu described above. The stereochemical outcome of the reaction of the vinyl sulfoximine 231 with Me2CuLi (Table 16, entry 1) appears anomalous, whereas that from the reaction of
O
O
ic'.P^ ,.Me MeO" 226
'"Ph
RM
J'-Ph MeC^ 227
Me ">'/.''ph
Chiral Sulfoximines
343
this substrate with the other organometallic reagents reported in Table 16 (entries 2-5) is consistent with those in Tables 13-15. In 1996, Jackson^^^ reported the stereoselective addition of organometallic reagents to N-tosyl a-trimethylsilyl vinyl sulfoximines. Treatment of these compounds with either alkyl or phenyllithium, dialkylcopperlithium, or alkyl Grignard reagents (R^M), followed by quenching with mild acid and then desilylation with tetrabutylammonium fluoride gave P-substituted sulfoximines in moderate to good overall yields. The diastereoselection varied from 1:1 to 25:1 and was dependent on the nature of the groups R and R^ and the metal M. Organolithium reagents gave the best overall yields and levels of diastereoselectivity. Two examples that worked well include the A^-tosyl a-trimethylsilyl vinyl sulfoximines 234a,b. In both cases the diastereomeric ratio of the products 235a,b was 25:1 and the overall yields were greater than 65%. The relative stereochemistry of 235a and 235b was determined by X-ray crystal structure analysis. The a-unsubstituted vinyl sulfoximines 236 underwent a-deprotonation with organolithium reagents in contrast to the vinyl sulfoximines 220 and 223 which undergo conjugate addition reaction with alkyl-
Table 15, Addition of Organometallics (RM) to Vinyl Sulfoximine 228
O ^S'C'Ph
^
1.RM.THF. -25 °C to 0 "C 2. NH4CI
228 O
O
LH
LR
T"'H
n-Bu'^P,
Ph 229
Entry 1 2 3 4 5
RM MejCuLi Me2CuLi (Lil "free") M e j C u U Z n B r j d . l equiv) MeCu MeCu (Lil "free")
T"'H
n-Bu^>^
Ph 230
Yield (%)
Diastereoselection (229.-230J
60 72 64 83 74
88:12 94:6 12:88 15:85 20:80
STEPHEN G. PYNE
344
Table 16. Addition of Organometallics (RM) to Vinyl Sulfoximine 231
RM, THF. Ph n-Bu^
-25°C to 0°C 2. NH4CI
231
f
Ph
'
N^^M«
"^Ph T""H Ph
JUiR 232 Entry 1 2 3 4 5
RM Me2CuLi Me2CuLi (Lil "free") Me2CuLi ZnBr2 (^-^ equiv) MeCu MeCu (Lil "free")
233 Yield (%)
Diastereoselection ^32.233;
60 69 65 79 80
23:77 90:10 12:88 21:79 13:87
lithiums, perhaps related to the formation of a strong chelate between the sulfoximine nitrogen and the methoxy oxygen atom.^^'^^ The stereochemical outcome of the reactions of 234 with organolithium reagents was rationalized as occurring via attack on the conformation 237 with possible assistance from the sulfoximine oxygen atom.^^^ X-ray structural analysis on 236 (R = Ph) suggested that attack would occur on the conformation 238 in which the S=0 and C=C bonds are approximately syn coplanar and approach of the nucleophile would occur anti to the large 5-phenyl substituent.^^^ A later and more comprehensive study of the solid-state structures of vinyl sulfoximines has been reported by Jackson.^^^ In the latter study, vinyl sulfoximines 236 (R = H, Me) were found to have a conformation in which the S=0 and C=C bonds are approximately syn coplanar while vinyl sulfoximines 236 (R = c-C^Hu, PhCH2CH2, PrO had a conformation in which the S=0 and C=C bonds are approximately syn coplanar.^^^
Chiral Sulfoximines
345
9,
._i..
I NTs TMS
R'
^ „ ^,^ ^ 2. BU4NF ( d r . = 25:1)
234a, R = PH 234b, R =f>C6Hi 1
Q Ph N NTs
. , 235a, R = Pi', R^ = Me 235b, R = o-CeH 11, R^ = Ph
s "
NTs 236
R^^NTs
>4^s^NTs
TMS 237
238
\ Ph
Resonance-Stabilized Carbanions Base-catalyzed addition of nitroethane and cyclic P-keto esters to racemic A^-phthalimido-5-/7-tolyl-5-vinylsulfoximine with either an enantiomerically pure chiral amine (quinine) or under phase-transfer conditions in the presence of an enantiomerically pure phase transfer catalyst (A^-benzylquininium chloride or A^-dodecyl-A^-methylephedrinium bromide), proceeded with little or no asymmetric induction at the newly created stereogenic carbon center.^^^ It is not clear from this report whether these reactions were terminated at 50% conversion or less, a condition necessary to observe kinetic resolution of the vinyl sulfoximine. Only in the case of the reaction of the vinyl sulfoximine with nitroethane was the product obtained with a measurable optical rotation ([a]D25 + 5.4° (CHCI3)), ^^^ ^^^ enantiomeric purity was not determined. The unreacted vinyl sulfoximine was recovered and found to have an enantiomeric purity of 7%. The reaction of the a-phenylthio vinyl sulfoximine 239 with lithiated phenyl phenylthiomethyl sulfone 240 gave a 3:1 mixture of the cyclopropanes 241a and 241b, respectively.^^ In contrast, the a-unsubstituted vinyl sulfoximine 242 gave a mixture of the cyclopropyl sulfone 243, isolated as a single diastereoisomer in 49% yield, and the cyclopropylsulfoximine 244, which was difficult to characterize.^^
STEPHIEN G. PYNE
346
KF Tol—S-CH=CH2
+
-\A/2 RCH-W
quinine
I
or chiral R^R2R3R^NX
9 II
R *l
,
Tol—S-CH2-CH2—C—W^
P RCH-W^
=EtN02
o
COgEt
Nitrogen Nucleophiles
The reaction of racemic A^-phthalimido-5-p-tolyl-5-vinylsulfoximine with a deficiency (0.5 molar equiv) of enantiomerically pure (-)-ephedrine resulted in a kinetic resolution of the vinyl sulfoximine.^^^ When the reaction was conducted at -30 °C the unreacted vinyl sulfoximine could be recovered with an enantiomeric purity of 46%. (-)-Amphetamine and (+)-l-phenylethylamine were not effective for kinetic resolution. The analogous (Z)-propenyl sulfoximine also underwent kinetic resolution with (-)-ephedrine, but the extent of kinetic resolution was not determined. The enantiomerically pure vinyl sulfoximines 245a and 245b, on treatment with hydroxide, undergo cyclization to give chiral isoquinolines with a modest diastereoselectivity. Reductive desulfurization of the major diastereomeric products from these cyclization reactions (247a and 246b) with Raney nickel gave the chiral isoquinoline alkaloids, (-)-(5)-carnegine and (+)-(/?)-carnegine, respectively, in high enantiomeric purity (95% ee).^^"^ The stereochemical outcome of these cyclizations seems largely governed by the chirality at sulfur in 245 and not by the chiral auxiliary ligand (Table 17). Changing
Chiral Sulfoximines
347 M%
Tol—S-CH=:CH 2
NHMe OH (0.5 equiv.)
O
II
Me
Tol—S-CHa-CHa-NMe—1^ N Osr-N-^O
Tol—S-CH=CH 2
II
HO—l^ Ph
N Oci^N^o (-) (46% ee)
the reaction solvent from methylene chloride (CHjClj) to methanol (MeOH) in the reaction of 245a with benzyltrimethylammonium hydroxide ([PhCH2NMe3]*[OH]") dramatically affects the diastereoselectivity (from 48% to 16%). Surprisingly, the reaction temperature had little effect on the diastereoselectivity. It was proposed that in a nonpolar aprotic solvent (CHjClj) the reaction proceeds via the intermediate 248 in which there is H-bonding between the NH of the amino group and the nitrogen of the sulfoximine moiety. O .. II
Li^SPh
L . ^"^^ SPh 239
THF
SOoPh -78°CtoRT '^ 240 (75%) O II
P h ( C H 2 ) 2 ^ * ^ ^ N ^ ^ g ^ + 240
PhS pjo'* *, '^'^ SPh f^"^ 241a, X = Ph, Y = O 241b, X = 0 , Y = Ph
A^SOaPh
THF ; -78 °C to RT
ph(CH2)2-'
^sPh 243 (49%)
242
PhS Sv Ph(CH2)j
244
NTs
348
STEPHEN G. PYNE MeO
MeO
^ks^^A.
N(C(XF3)Me
245
MeO
tAeoYX^^^*^ 246
NMe
s r"'H Ph
Table 17. Base-Induced Cyclization of 245a,b Sulfoximine 245a 245a 245a 245a 245b 245b 245b 245b
Base [PhCH2NMe3]*[OH][PhCHjNMeal^IOH][PhCH2NMe3l+[OH]LI^OH[PhCHjNMeal^IOH][PhCHjNMcaJ-'IOH][PhCHjNMeaJ-'IOH]Li+OH-
Solvent CH2CI2 CH2CI2 MeOH MeOH-H20(2:1) CH2CI2 CH2CI2 MeOH MeOH-H20(2:1)
rm 0 -40 0 0 0 -40 0 0
Diastereoisomeric Ratio a46):a47) 26:74 28:72 58:42 65:35 71:29 68:32 54:46 65:35
349
Chiral Sulfoximines
Oxygen Nucleophiles Racemic A^-tosyl vinyl sulfoximines 236 subjected to nucleophilic epoxidation with lithium f^rf-butylperoxide in THF at -50 °C for 5 min gave the sulfoximinooxiranes 249 as single diastereoisomers in excellent yields (72-97%)^^ The relative stereochemistry of 249 (R = Pr*) was established by X-ray crystal structure analysis. The epoxidation process occurred from the same diastereoface of the C=C of 236 as the addition of organolithium reagents to 236 (see structures 235a,b above) In contrast, the reaction of 236 (R = Pr*) with alkaline hydrogen peroxide gave a 1.7:1 mixture of diastereomeric sulfoximinooxiranes. The nucleophilic epoxidation of 236 (R = W) was also highly diastereoselective using lithium triphenylmethylperoxide whereas the analogues potassium reagents were poorly diastereoselective.^^^ It was suggested that coordination of the lithium cation to the sulfoximine oxygen was essential to obtain a high level of diastereoselectivity in these reactions.^ ^^ When the R substituent in 236 contains other stereogenic centers, these epoxidation reactions gave variable ratios of diastereoisomers. Matched situations usually, depending on the nature of the epoxidation reagent, result in a very high diastereoselectivity (25:1) (e.g., 250 to 251) whereas in the unmatched situations the diastereoselectivity was generally poorer. Treatment of enantiomerically pure sulfoximinooxiranes 249 with magnesium bromide in the presence of tetrabutylammonium borohydride gave the enantiomerically enriched bromohydrins 252 in good yields and with enantiomeric purities ranging from 70 to 91%.^*^ Treatment of the rerr-butyldimethylsilyloxy vinyl sulfoximine 139b with tetrabutylammonium fluoride in THF at 0 ^'C afforded the 2,3,4,5-tetrasubstituted dihydrofuran 253 in greater than 97% diastereoselectivity.*^*^^ This method works equally well with the other diastereoisomers of 139b. QH
350
STEPHEN G. PYNE
O
Q O
, ^ % ^ F ^
.Bu-OOLi
^ L ,
p^l^^*~Ph
236
249 O
5 3-^0
O
"NTS
-50 °C
250
o^T Me-V^
Me
'NTs 251
llle (d.r.=25:1)
Cyanide Ion The reaction of A^-tosyl vinyl sulfoximines 236 with lithium cyanide in DMF at room temperature for 1 h gave the vinyl nitriles 254 in good yields. *^^ Treatment of 236 with lithium dimethylphosphonate in THF at -78 °C to room temperature gave moderate yields of the vinyl phosphonates 255.^^^ These yields could be improved to 54-64% by isolation of the initial Michael adducts by quenching these reactions at -20 °C and then treatment of these products with sodium methoxide in methanol at reflux. These reactions proceed via the intermediates 256 and 257. C. Cross-Coupling Reactions The nickel-catalyzed cross-coupling reaction of the vinyl sulfoximine 258 with diarylzinc reagents in the presence of a salt (MgBr2, LiBr, or ZnCl2) gave exclusively the (£)-exocyclic alkenes 259 in high enantiomeric purity (> 98%).^^^ Unfortunately, this method could not be extended to the cross-coupling reaction with dialkylzinc reagents. Nickel- and magnesium-catalyzed coupling of the optically active vinylsulfoximine 260 and the organozinc reagents 261a,b gave the optically active allyl silanes 262a,b (ee > 95%) in excellent yields (91-95%).^^^ This method also worked efficiently on an optically active 3-oxa-carbacyclin intermediate. ^^^ ^ O O
MgBrg, BU4NBH4 NTs
249
_. B^
(67-85%) 252
(e.e. 70-91%)
Chiral Sulfoximines
351 )H
T01-5S 1^^^^^^
Me
/
OTMS
\
139b
O II ^
Me
BU4NF. THF
OTBDMS
0°C (61%)
Tni-Q/
\_J
\ ^ ^ ^ e #
OH
OTMS 253
X
LiCN, DMF 1h, RT
NTs
236
254
(63-81%) (MeO)2P(0)Li, THF -78 °C to RT Nu
^P(0Me)2
^
R ^
NU
O ©
NTs
256
255
O
257
MeN^ Ar2Zn ^ NiCl2(cippp) salt ,« R^O 258
R^O (dppp = Ph2P(CH2)3PPh2)
*, R^O
259
R^O
STEPHEN G. PYNE
352 TMS
MeN
Zn(CH2SiMe2X)2 2 6 1 a , X = Me 261b, X = OPr' ^ NiCl2.clppp MgBr2
a-Metallated (metal = Li or BrMg) vinyl sulfoximines undergo nickel-catalyzed substitution with organometallic reagents to give vinyl organometallic compounds.^^'^^^ For example, the a-lithiated (Z)-vinyl sulfoximine 264, which is stable to isomerization to (E)-266 at -78 °C, when treated with phenyllithium in the presence of 5 mol% NiCl2(PPh3)2 gave the (Z)-vinylsilane 265 as a single diastereoisomer in 72% yield. The same (Z)-isomer of 265 was obtained starting from (£)-266, which is formed at -30 °C from (Z)-264. It was assumed that 265 arises from a l,5-0,C-silyl migration from the vinyl lithium intermediate 267.^^ D. Cycloaddition Reactions
The Diels-Alder reactions of racemic ^-(/7-tolyl)-5-/?-tolyl-5-vinylsulfoximine with dienes gave mixtures of diastereomeric cycloadducts in good yield (Table 18). When cyclopentadiene and 1,3-cyclohexadiene were employed as dienophiles, the endo diastereomeric products 268c and 268d (n = 1,2) predominated. ^^^
MeLi, ^ ^ ^ ' O EtgO -70 ^C
.^^^^Li
PhLi, NiCl2(PPh3)2
piA^x^s^Ph
'^®Du/^\^^'^^-70toO°C 263
Me EtgSi
(72%)
^
265
-30 °C 2h Ov NMe
PhLi,
S>Xp,^NiCl2(PPh3); Mini /DPI
J2 Et20, 2h. -70 to 0 °C
(59%)
265
353
S(0)(NTos)Tol
2 6 9 a , R U M e , R^ = H 269b, R U H, R^ = Me
The intramolecular Diels-Alder reactions of vinyl sulfoximines 270 (AZ = 1, 2) have been studied by Craig.^^^'^^^ In all cases mixtures of four diastereomeric cycloadducts were formed. When n = 1 the major diastereoisomer was the transfused compound 271 (n = 1) while when n = 2 the major diastereoisomer was the cw-fused compound 273 {n = 2) The diastereoselectivity of these reactions when n = 1 were essentially independent of the nature of the N-substituent in 270, while when 71 = 2 the yV-2,4,6-triisopropylphenyIsulfonyl (Tris) derivative gave the highest selectivity for 273.
Table 18. Diels-Alder Reactions of Racemic N-(p-Tolyl)-S-p-tolyl-S-vinylsulfoximine with Dienes Diene
Yield (%)
Cyclopentadiene
81
1,3-Cyclohexadiene
95
2,3-Dimetliyl-1,3-butadiene
95
Cycloadducts (Diastereoselection) 268a+268b+268c +268 (1:1:4:5) [268a+268b]+268c +268d (7:41:52) 269a+269b (4:1)
354
STEPHEN G. PYNE
II
(CM (CH2)n
r
(CH2)n
PhMe heat
XN 270
r
n
Ts Ts
1 2
yield 271 :272: 273:274 (%) 72 39 : 31 : 25 : 5 70 2 0 : 1 4 : 4 0 26
Tris Tris
1 2
79 75
35 : 30 : 30 : 5 15:5:65:15
Tf
1 2
89 68
38 :32 :22 : 8 24 : 21 : 35 : 20
(CH2)n
273
274
|
PhMe heat (53%)
TfN
TfN 276(90%)
277(10%)
275
278
279 (X = Ts, 14% X = Tf, 20%)
280 (X = Ts, 86% X = Tf, 80%)
Chiral Sulfoximines
355
Ph I H R' H Pll f ^ 75-80 °C Ph,. ^N >=< + >=N©
Ph I
be
282
R2N
281
r^ H Me Ph Ph
1
R2
Yield (%)
Ts 46 Ts 55 Ts 67 Tris 66 Ph Me 43
283 :284 65 64 67 75 47
35 36 33 25 53
The trienes 275 and 278 underwent cyclization to give only two cycloadducts. The major adduct from 275 was the trans-fused adduct 276, while that from 278 was the cw-fused product 280.^^^ The reactions of the vinyl sulfoximines 281 with C,A^-diphenylnitrone 282 are highly regioselective and give only 4-sulfonimidoyl-isoxazolidine cycloadducts 283 and 284.^^ These reactions proceed with modest 7i-facial selectivity with respect to the dipolarophiles 281. The stereochemical outcomes of these reactions are consistent with attack on the ground-state conformation 238 of the sulfoximine through an "^n
VI.
a-SULFONIMIDOYL KETONES AND ESTERS A. Synthesis
a-Sulfonimidoyl ketones 285 have been prepared from 5-methylsulfoximines via three different methods: (1) by deprotonation with n-butyllithium followed by reaction with aldehydes and then oxidation of the resulting P-hydroxy sulfoximines,^^ (2) by deprotonation with n-butyllithium followed by reaction with nitriles and then acid hydrolysis,^^^'^^"^ and (3) by deprotonation with lithiumdiisopropylamide (LDA) followed by reaction with esters.^^^ The titanium tetrachloridecatalyzed reaction of silyl enol ethers with A^-methylbenzenesulfonimidoyl fluoride also gives a-sulfonimidoyl ketones.^ a-Sulfonimidoyl esters 285 (R^ = OR) can be
356
STEPHEN G. PYNE
readily prepared from 5-methylsulfoximines by deprotonation with ^-butyllithium followed by reaction with dimethyl carbonate. ^^"^ B. Reactions Reactions with AHcylating Reagents Methylation of the lithium salt of (5)-(+)-A^-ethoxycarbonylmethyl-5-aryl-5-/7tolyl sulfoximines (286, 287) is 100% diastereoselective when the 5-aryl group is capable of coordination to the lithium cation."^ The addition of the potassium salts of the sulfoximinyl esters 289 and 292 to diene-molybdenum (288) and dienyliron complexes (291) gave adducts which, after desulfonylation, yielded the enantiomerically enriched organometallic complexes 290 and 293. The potassium salts gave the highest diastereoselectivities and the more sterically demanding A^-(dimethylthexysilyl) derivative 289 gave products with the highest enantiomeric purity. ^^^ Reactions with Aldehydes The Knoevenagel-type condensation of the a-sulfonimidoyl ketones 294a or 294b or the a-sulfonimidoyl ester 294c with aldehydes proceeded in modest to good yields (46-87%) and gave the {E) a-sulfonimidoyl P,Y-enones 295a and 295b and the {E) a-sulfonimidoyl (3,Y-unsaturated ester 295c as mixtures of two diastereomeric compounds.^^ Treatment of P-carbonyl sulfoximines 296 (R=Bu^ COuMe, N(Pr^)2) with diethylzinc gave the corresponding ethyl zinc derivatives.^ The X-ray crystal structure of the ethyl zinc derivative (R=Bu') showed a dimeric O-metallated
NR2n V'' CH3" ^R^'
1.n-BuLi,THF 2.R1CH0 3. oxidation
1. n-BuLi, THF 2. MeOCOgMe CHg-^^RS
l.n-BuLi.THF ^R^ Q 2. R^CN V orRiCOgR CHa'^'^RS 3 H"^
I %^ R I ^ ^ ^ ' ' ^ R3
285
Chiral Sulfoximines
357
OR O
OR O ^Tol
l^e'
N. \ CH(-)CO 2Et
i^^/CO^Et
286 R = Me 287 R = CH2CH20Me
diastereoselection R = Me, 76:22 R = CH2CH20Me, 100:0
enolate form (297) in which two enolates of opposite chirality were connected to form an eight-membered (Zn--0~S-N)2 ring. In contrast, the organometallic compound formedfromthe reaction of 296 (R = N(Pr*)2) formed a dimeric C-metallated carbonyl species (298) that involved an eight-membered (Zn-C-C-0)2 ring structure. Solution NMR studies suggested that 298 (R = N(Pr')2) had a much higher electron density at the a-sulfur carbon than 297 (R = Bu') consistent with the observation that 298 (R = N(Pr^)2) was more reactive toward electrophiles. For example, while 297 (R = Bu') did not react with methyl iodide or benzaldehyde, compound 298 (R = N(Pr*)2) reacted with benzaldehyde in the presence of chlorotrimethylsilane to give 299 in greater than 90% diastereoselectivity.
O
Mo(CO)2Cp PF.
(CHg)
11 11 NSi-thexylMe2
1. Ph—S—CH{-)C02Me 289
288
Mo(CO)2Cp
K*
290
2. Al{Hg)
(n = 1,80%ee) (n = 2, 85% ee)
Fe(CO)3
Fe(C0)3
a 291
(CH2)nl^^C02Me
1. Ph*-S—CH(-)C02Me
:l 292
.G02Me
NTos ^-
2. AI(Hg)
293 (30% ee)
358
STEPHEN G. PYNE
R2
R^CHgCHO, MeCN
^ S V. R^N ^ I ; ^ O
^^^^
piperidine, HOAc, 3 A molecular sieves, rt
^^
" Y " ^ R^
^ s \ R^N ^ I ^ o
(46-870/0)
^^
294a; R^ = Ph. R^ = Ts 294b; R2 = Ph, R^ = COgMe 294c; R2 = OMe, R^ = Ts
295a; R^ = Ph, R^ = Ts 295b; R^ = Ph, R^ = COgMe 295c; R^ = OMe. R^ = Ts
[R^ = Bu, pent, hexyl, Et]
Reductions
P-Keto sulfoximines 300 undergo diastereoselective reductions at -78 °C with sodium borohydride or diborane to give mixtures of diastereomeric P-hydroxy sulfoximines. The product diastereoselection increases as the steric demand of the substituent R of 300 increases (Table 19). Reductive removal of the P-sulfoximine group of the diastereomeric mixture of P-hydroxy sulfoximines gives secondary alcohols with the (5)-configuration.^^^
Et I Me>.
Me^
N
O
EtgZn
^
rj
N
O R
296
°
°
297 (R = Bu*)
Me^
Me^ 'A
II
299
298
(R = N(PH)2)
( R = N(P02)
Cbiral Sulfoximines
359
rri .c^ 9. :l
II xC>
Ph—S—CH,
NaBH4or
'R
11
r
BHgTHF
(major)
HO.
H
NMe 300 Raney Ni
H
PH
CH3
^R
.H
^C\, CH2
(minor) O
11 Ph—S—CH2-
(S)
•I NR
(43-69% ee)* '[corrected for enant. purity of 300 ] Table 19.
Diastereoselective Reductions of 300
^-Keto-sulfoximine
aoo; (R)
Reducing Agent
Me Et Pr' Ph Bu^ n-hexyl Ph Bu' Bu*
NaBH4 NaBH4 NaBH4 NaBH4 NaBH4 BH3 BH3 BH3 BH.
Yield (%) 100 100 100 100 100 75 90 83 91
Diastereoselection 50:50 56:44 60:40 70:30 75:25 75:25 80:20 83:17 90:10
VIL SULFOXIMINES AS LIGANDS FOR ASYMMETRIC SYNTHESIS The X-ray crystal structures of P-hydroxy sulfoximines coordinated to metals are known, including complexes to ethylzinc^^^ and vanadium.^^^ These complexes involve coordination of the metal to the hydroxyl oxygen and the sulfoximine nitrogen. A palladium(II) bidentate pyridine-sulfoximine complex^^^ and simple sulfoximine-copper(II) and zinc complexes are also known. ^^^ These latter complexes involve coordination of the metal to the sulfoximine nitrogen.
360
STEPHEN G. PYNE
In 1979, Johnson reported the enantioselective reduction of ketones with stoichiometric amounts of optically active P-hydroxy sulfoximine-borane complexes. ^^^ Prochiral alkyl phenyl ketones (RCOPh) undergo enantioselective reduction with enantiomerically pure p-hydroxy sulfoximine borane complexes (301 and 302). These complexes are prepared by reaction of the corresponding P-hydroxy sulfoximine with borane at -78 °C. The structures 301 and 302 have been suggested for these complexes. In the case of the borane complex 301, the enantioselectivity increased as the steric bulk of the R substituent of the ketone (RCOPh) was decreased from Pr^ to Me. The analogous reductions of methyl alkyl ketones (MeCOR) with these borane complexes were less enantioselective (3-27% ee).^^^ In 1993, Bolm reported that these reactions could be performed using catalytic quantities (10 mol%) of the chiral P-hydroxy sulfoximine. ^^^ The enantiomeric purities of the product alcohols ranged from 52% (1-indanone) to 93% (PhCOCH20SiR3). In many cases the enantiomeric purities were enhanced using sodium borohydride as reductant in the presence of chlorotrimethylsilane.^^^ These methods have been extended to the asymmetric reductions of imines.^^"^ A^-SPhsubstituted imines gave the highest enantioselectivities and these reductions proceeded in the same stereochemical sense as the reductions of ketones. NH \j
OH
O^
\ .
HO..H (ee 52-93%)
Optically active P-hydroxy N-methyl sulfoximines have been used as catalysts for the enantioselective transfer of an ethyl group from diethylzinc to aldehydes to give secondary alcohols in enantiomeric excesses of 61-88%.^^^'^^^ Related chiral ligands have been used with nickel acetylacetone to promote the enantioselective Michael addition of diethylzinc to chalcones.^^^ MeN
^
OH
V
HO^.H
R ^ H R^ = Me, -(CH2)n- (n = 3 or 9) (ee 61-88%)
Chiral titanium reagents derived from optically active P-hydroxy sulfoximines and Ti(0-Pr^)4 promote the asymmetric addition of trimethylsilyl cyanide to alde;hydes.^^^ The resulting cyanohydrins are generally formed in 74-91% enan-
Chiral Sulfoximines OH
0
1
II
361
+
9
— •
Ef"i%HrV'Ph Ph
' -^NM^ 1 301 'BF3
(S) (e.e. 31-60%)
OH
0
1
II
+
1
Hq
ph^^"f^cH3-V;ph £^
H
— •
Phf^
^R
'^ +NMe 302 •BF3 (e.e. 57-74%)
O H
(fl)-303
+ TMSCN
HO
pN
R-^H Ti(0PH)4 CH2CI2 HN OH
(e.e. 74-91%)
Me>»yS
(ff)-303
Li
THF
Cu n-Bu yMegCeHgCn/® 304
0
0
+
k^
-78 °C
A 1 1 (e.e. = 99%)
362
STEPHEN G. PYNE (S)-307 (5 mol %)
OAC
MeOgC^^COgMe
+ CHgCCOgMe); Ph 305
[Pd(allyl)CI]2
^ ^
^^
2mo\% g s A{(3 (3 equiv.) (S)-306 (77%. e.e. 73%) eqt AcOK
^ . N
11 Q//
CH2CH2-0(H0)C6H4
307
tiomeric excess. The (/?)-P-hydroxy sulfoximine 303 was found to be particularly effective. When substoichiometric amounts of 303 and Ti(0-Pr^)4 were used (20 mol%), the enantioselectivities and yields dropped dramatically. For example, with benzaldehyde the stoichiometric reaction gave the corresponding cyanohydrin in 91% ee (yield 72%) while when catalytic conditions were employed the ee was 43% (yield 29%). Gais has used chiral cyclic a-sulfonimidoyl carbanions as nontransferable ligands in copper-mediated enantioselective Michael additions to cyclic enones."*^ For example, the organocopper reagent 304 underwent conjugate addition to cyclohexenone to give (5)-3-butylcyclohexanone in 99% enantiomeric excess. Bolm has discovered that chiral sulfoximine/palladium complexes formed between pyridine-sulfoximine ligands and palladium(O) catalyze the enantioselective allylation of dimethyl malonate with the racemic allyl acetate 305.*^^ The best enantiomeric purity of the product 306 was obtained using the chiral ligand 307.
VIM. CONCLUSIONS Sulfoximines are versatile reagents for diastereoselective and asymmetric synthesis. They continue to find many synthetic applications as both nucleophilic and electrophilic reagents. While the nucleophilic character of sulfoximine reagents has been well exploited,^ the use of the sulfoximine group as a nucleofuge is more recent and adds to the synthetic use of these compounds. The palladium(0)-catalyzed chemistry of allylic sulfoximines and the use of chiral sulfoximines as ligands in catalytic asymmetric synthesis are areas of recent development that have potentially useful applications. Further work is required to understand the factors that determine the diastereoselection and the stereochemical outcomes of these reactions. These studies will result in enhanced product diastereo- and enantioselectivities and make these reagents even more attractive to the wider synthetic chemistry community.
Chiral Sulfoximines
363
ACKNOWLEDGMENTS I would like to thank the Australian Research Council for supporting our chiral sulfur chemistry program over 14 years. I sincerely thank Sandra Chapman, Dorothy David, Branko Dikic, Chris Dixon, Zemin Dong, Renate Griffith, Kylie Hellmund, and Gareth O'Meara for their fine contributions to sulfur chemistry in my laboratories. Collaborations in this area of research with Allan White, Gemot Boche, Guy Solladie, and Andreas Pfalz are gratefully acknowledged.
REFERENCES 1. For earlier reviews see: Johnson, C. R. Aldrichimica Acta 1985,18,3; Johnson, C. R.; Barbachyn, M. R.; Meanwell, N. A.; Stark, C. J., Jr.; Zeller, J. R. Phosphorus Sulfur 1985, 24, 151; Pyne, S. G. Sulfur Rep. 1992, 72, 57; Pyne, S. G. Sulfur Rep. 1999, 27, 281. 2. Johnson, C. R.; Schroeck, C. W. J. Am. Chem. Soc. 1973, 95,7418. 3. Shiner, C. S.; Berks, A. H. J. Org. Chem. 1988, 55, 5543. 4. Brandt, J.;Gais, H.-J. Tetrahedron: Asymmetry 1997, 5, 909. 5. Johnson, C. R.; Jonsson, E. U.; Wambsgans, A. J. Org. Chem. 1979,44, 2061. 6. Johnson, C. R.; Bis, K. G.; Cantillo, J. H.; Meanwell, N. A.;Reinhard, M. F. D.;Zeller, J. R.;Vonk, G. R J. Org. Chem. 1983,48,1. 7. Johnson, C. R.; Jonsson, E. U.; Bacon, C. C. J. Org. Chem. 1979,44, 2055. 8. Harmata, M. Tetrahedron Lett. 1989,30,437. 9. Johnson, C. R.; Wambsgans, A. J. Org. Chem. 1979,44, 2278. 10. Rees, C. W; Storr, R. C. J. Chem. Soc. C1969, 1474 . 11. Pyne, S. G. J. Org. Chem. 1986,57, 81. 12. Pyne, S. G. Tetrahedron Utt. 1986,27, 1691. 13. Reggelin, M.; Weinberger, H. Tetrahedron Lett. 1992, 33, 6959. 14. Reggelin, M.; Welcker, R. Tetrahedron Lett. 1995,36, 5885. 15. Johnson, C. R.; Janiga, E. R. J. Am. Chem. Soc. 1973, 95, 7692. 16. Johnson, C. R.; Kirchhoff, R. A.; Corkins, H. G. J. Org. Chem. 1974,39, 2458. 17. (a) Tumura, Y; Sumoto, K.; Minamikawa, J.; Ikeda, M. Tetrahedron Lett. 1972,4137. (b) Tumura, Y.; Minamikawa, J.; Sumoto, K.; Fujii, S.; Ikeda, M. J. Org. Chem. 1973,38,1239. 18. Johnson, C. R.; Kirchhoff, R. A.; Reischer, R. J.; Katekar, G. E J. Am. Chem. Soc. 1973,95,4287. 19. Kwart, H.; Kahn, A. A. J. Am. Chem. Soc. 1967,89,1950. 20. Moriyama, M.; Numata, T.; Gae, S. Org. Prep. Prvced. Int. 1974,6, 207. 21. Cram, D. J.; Day, J.; Rayner, D. R.; von Schriltz, D. M.; Duchamp, D. J.; Garwood, D. C. J. Am. Chem. Soc. 1970, 92,7369. 22. Abramovitch, R. A.; Takaya, T. J. Chem. Soc, Perkin Trans. I 1975, 1806. 23. Greenwald, R. B.; Evans, D. H. Synthesis 1977, 650. 24. Johnson, C. R.; Uvergne, O. J. Org. Chem. 1989,54,986. 25. (a) Hwang, K.-J. J. Org. Chem. 1986,57,99. (b) Hwang, K.-J.; Logusch, E. W; Brannigan, L. H.; Thompson. M. R. J. Org. Chem. 1987,52, 3435. 26. Pearson, A. J.; Blystone, S. L.; Nar, H.; Pinkerton, A. A.; Roden, B. A.; Yoon, J. / Am. Chem. Soc. 1989, 777,134. 27. Pyne, S. G.; Dikic, ^.Tetrahedron Lett. 1991, 31, 5231. 28. Williams, T. R.; Booms, R. E.; Cram, D. J. J. Am. Chem. Soc. 1971, 93, 7338. 29. Johnson, C. R.; Schroeck, C. W ; Shanklin, J. R. J. Am. Chem. Soc. 1973, 95,7424. 30. Johnson, C. R.; Schroeck, C. W. /. Am. Chem. Soc. 1973, 95, 6190. 31. Hambley, T. W; Raguse, B.; Ridley, D. D. Aust. J. Chem. 1987,40, 61. 32. Schaffner-Sabba, K.; Henrici, H.; Renfroe, H. B. / Org. Chem. 1977, 42, 952.
364
STEPHEN G. PYNE
33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48.
Craig, D.; Geach, N. J. Synlett 1993, 481. Mutti, R.; Wintemitz, P. Synthesis 1986, 426. Bach, T.; Korber, C. Tetrahedron Lett. 1998,39, 5015. Veal, H. S.; Levin, J.; Swem, D. Tetrahedron Lett. 1978, 503. Johnson, C. R.; Kirchhoff, R. A. J. Org. Chem. 1979,44,2280. Huang, S.-L.; Swem, D. J. Org. Chem. 1979,44, 2510. Gaggero, N.; D'Accolti, L.; Colonna, S.; Curci, R. Tetrahedron Lett. 1997,38,5559. Bosshamer, S.; Gais, H.-J. Synthesis 1998,919. Gais, H.-J.; Erdelmeier, I.; Lindner, H.; Vollhardt, J. Angew. Chem. Int. Ed. Engl. 1986, 25, 938. Gais, H.-J.; Dingerdissen, U.; Kruger, C ; Angermund, K. / Am. Chem. Soc. 1987,109, 3773. Chassaing, G.; Marquet, A. Tetrahedron 1978, 34, 1399. Muller, J. F. K.; Batra, R.; Spingler, B.; Zehnder, M. Helv. Chim. Acta 1996, 79, 820. Muller, J. F. K.; Neuburger, M.; Zehnder, M. Helv. Chim. Acta 1997, 80, 2182. Hambley, T. W.; Raguse, B.; Ridley, D. D. Aust. J. Chem. 1986,39, 1833. Trost; B. M.; Matsuoka, R. T. Synlett 1992, 27. (a) Pyne, S. G.; Dikic, B. Tetrahedron Lett. 1990,31, 5231. (b) Dikic, B. Ph.D. Thesis, University of WoUongong, 1991. Pyne, S. G.; Dikic, B.; Skelton, B. W.; White, A. H. Aust. J. Chem. 1992,45, 807. Pyne, S. G.; Dikic, B.; Skelton, B. W.; White. A.H. J. Chem. Soc, Chem. Commun. 1990, 1376. Johnson, C. R.; Stark, C. J., Jr. /. Org. Chem. 1982, 47, 1193. Pyne, S. G.; Boche, G. J. Org. Chem. 1989,54, 2663. Boche, G. Angew. Chem. Int. Ed. Engl. 1989, 28, 111. Pyne, S. G.; Dong, Z.; Skelton, B. W; White, A. H. / Chem. Soc, Perkin Trans. 11994, 2607. Pyne, S. G.; Dikic, B.; Skelton, B. W; White, A. H. J. Chem. Soc, Chem. Commun. 1990, 1376. Entry 7 in Table 1 was incorrectly reported in the initial communication as was the stereochemistry at C-2 in structure 7. The figure caption therein also requires amendment, thus: 0(3)-S(3)C(31,4), 109.9 (1);C(4)-S(3)-C(31), 102.3(2)°. Toda, R; Imai, N. J. Chem. Soc, Perkin Trans. 11994, 2673.
49. 50. 51. 52. 53. 54. 55. 56.
57.
59
58. In the original publication lithiated sulfoximines should have been referred to as lithiated (S)-sulfoximines and not (/?)-lithiated sulfoximines. 59. Pyne, S. G.; Dong, Z.; Skelton, B. W; White, A. H. / Org. Chem. 1997,62, 2337. 60. Corey, E. J.; Chaykovsky, M. J. Am. Chem. Soc 1965, 87,1353. 61. Welch, S. C ; Rao, A. S. C. P; Lyon, J. T; Assercq, J.-M. J. Am. Chem. Soc, 1983,105, 252. 62. Tij, S. S.; Soman, R. Tetrahedron: Asymmetry 1994,5, 1513. 63. Tij, S. S.; Shah, A. C ; Lee, D.; Newton, G.; Soman, R. Tetrahedron: Asymmetry 1995, 6, 1731. 64. Meehan, S.; Little, R. D. J. Org. Chem. 1997, 62, 3779. 65. Johnson, C. R.; Zeller, J. R. Tetrahedron 1984,40,1225. 66. Salomon, R. G.; Sachinvala, N. S.; Roy, S.; Basu, B.; Raychaudhuri, S. R.; Miller, D. B.; Sharma, R. B. J. Am. Chem. Soc 1991,113, 3085. 67. Johnson, C. R.; Penning, T. D. J. Am. Chem. Soc 1988,110, 4726. 68. Paquette, L. A.; Heidelbaugh, T. M. Synthesis 1998, (Special Issue SI), 495. 69. Davis, F. A.; Zhou, P; Murphy, C. K.; Sundarababu, G.; Qi, H. Y; Han, W; Przeslawski, R. M.; Chen, B. C ; Carroll, P J.J. Org. Chem. 1998, 63, 2273. 70. Johnson, C. R.; Kirchhoff, R. A. / Am. Chem. Soc 1979,101, 3602. 71. Johnson, C. R.; Meanwell, N. A. J. Am. Chem. Soc 1981,103,7667. 72. Erdelmeier, I.; Gais, H.-J.; Linder, H. J. Angew. Chem. Int. Ed. Engl. 1986,25,935. 73. Johnson, C. R.; Barbachyn, M. R. J. Am. Chem. Soc 1982,104, 4290. 74. Johnson, C. R.; Barbachyn, M. R. J. Am. Chem. Soc 1984,106, 2459. 75. Harmata, M.; Claassen, R. J., II. Tetrahedron Lett. 1991, 32, 6497. 76. Bund, J.; Gais, H.-J.; Erdelmeier, I. J. Am. Chem. Soc 1991,113, 1442.
Chiral Sulfoximines
365
77. Gais, H.-J.; Muller, H.; Bund, J.; Scommoda, M.; Brandt, J.; Raabe, G. J. Am, Chem. Soc. 1995, 777,2453. 78. Bailey, P. L.; Clegg, W.; Jackson, R. F. W.; Meth-Cohn, O. J. Chem. Soc, Perkin Trans. 1 1993, 343. 79. Pyne, S. G.; Dong, Z. J. Org. Chem. 1996,67, 5517. 80. Bosshammer, S.; Gais, H.-J. Synlett 1998, 99. 81. Pyne, S. G.; Boche, G. Tetrahedron 1993,49, 8449. 82. Gais, H.-J.; Lenz, D.; Raabe, G. Tetrahedron Lett. 1995, 36,1431. 83. Hainz, R.; Gais, H.-J.; Raabe, G. Tetrahedron: Asymmetry 1996, 7, 2505. 84. Pyne, S. G.; Dong, Z.; Skelton, B. W.; White, A. H. Sulfur Lett. 1997, 20, 255. 85. Reggelin, M.; Weinberger, H. Angew. Chem. Int. Ed. Engl. 1994, 55, 444. 86. Reggelin, M.; Weinberger, H.; Gerlach, M.; Welker, R. J. Am. Chem. Soc. 1996, 775, 4765. 87. Reggelin, M.; Weinberger. H.; Heinrich, T. LiebigsAnn. 1997, 1881. 88. Gais, H.-J.; Muller, H.; Decker, J.; Hainz, R. Tetrahedron Lett. 1995,5(5, 7433. 89. Hainz, R.; Gais, H.-J.; Raabe, G. Tetrahedron: Asymmetry 1996, 7, 2505. 90. Pyne, S. G.; Dong, Z.; Skelton, B. W; White, A. H. J. Chem. Soc, Chem. Commun. 1994,751. 91. Gilchrist, T. L.; Moody, C. J. Chem. Rev. 1977, 77,409. 92. Tamura, Y; Matsushima, H.; Minamikawa, J.; Ikeda, M.; Sumoto, K. Tetrahedron 1975,57,3035. 93. Harmata, M.; Glaser, R.; Chen, G. S. Tetrahedron Lett. 1995, 36, 9145 94. Gais, H.-J.; Sconunoda, M.; Lenz, D. Tetrahedron Lett. 1994, 35, 7361. 95. Pyne, S. G.; Dong, Z.; Skelton, B. W; White, A. H. J. Chem. Soc, Chem. Commun. 1995, 445. 96. Pyne, S. G.; Dong, Z. Tetrahedron Lett. 1995, 36, 3029. 97. David, D. M.; O'Meara, G. W; Pyne, S. G. Tetrahedron Utt. 1996, 37, 5417. 98. Pyne, S. G.; David, D. M.; Dong, Z. Tetrahedron Lett. 1998, 37, 8499. 99. Pyne, S. G.; O'Meara, G. W; David, D. M. Tetrahedron Lett. 1997, 38, 3623. 100. Pyne, S. G.; Dong, Z.; Skelton, B. W; White, A. H. Aust. J. Chem. 1998, 57, 209. 101. Gais, H.-J.; Muller, H.; Bund, J.; Sconunoda, M.; Brandt, J.; Raabe, G. J. Am. Chem. Soc 1995, 777, 2453. 102. Scommoda, M.; Gais, H.-J.; Bosshammer, S.; Raabe, G. J. Org. Chem. 1996, 61,4379. 103. Craig, D.; Geach, N. J. Synlett 1992,299. 104. David, D. M.; Bakavoli, M.; Pyne, S. G.; Skelton, B. W.; White, A. H. Tetrahedron 1995. 57, 12393. 105. This procedure was originally developed by Dr. S. E. Gibson (nee Thomas) and Dr. A. Weirzchleyski, Imperial College. London, unpublished work (see Acknowledgment in Ref. 104). 106. Paley, R. S.; Snow, S. R. Tetrahedron Lett. 1990, 57. 5853. 107. Erdelmeier, I.; Gais, H.-J. / Am. Chem. Soc 1989, 777, 1125. 108. Jackson, R. F. W; Briggs, A. D.; Brown, P A.; Clegg, W; Elsegood, M. R. J.; Frampton. C. / Chem. Soc, Perkin Trans. I 1996.1673. 109. Bailey. P L.; Hewkin. C. T; Clegg. W; Jackson, R. F W. J. Chem. Soc, Perkin Trans. 11993, 577. 110. Dong, Z.; Pyne, S. G.; Skelton, B. W; White, A. H. Aust. J. Chem. 1993,46,143. H I . Briggs, A. D.; Clegg, W ; Elsegood. M. R. J,; Framton. C. S.; Jackson. R. F. W. Acta Crystallogr A 1998. C54,1335. 112. Annunziata. R.; Cinquini. M.; Colonna. S. J. Chem., Soc Perkin Trans. 11980. 2422. 113. Annunziata. R.; Cinquini. M. /. Chem. Soc, Perkin Trans. 11979.1684. 114. Pyne. S. G. / Chem. Soc, Chem. Commun. 1986.1686. 115. Briggs. A. D.; Jackson. R. F W; Clegg, W; Elsegood, M. R. J.; Kelly, J.; Brown, P A. Tetrahedron Lett. 1994,35, 6945. 116. Bailey, P L.; Briggs, A. D.; Jackson, R. F W; Pietruszka, J. Tetrahedron Lett. 1993. 34, 6611. 117. Bailey. P L.; Jackson. R. F W. Tetrahedron Utt. 1991. 32, 3119. 118. Gais. H.-J.; Bulow. G. Tetrahedron Utt. 1992. 33,461. 119. Gais. H.-J.; Bulow. G. Tetrahedron Utt. 1992. 33, 465.
366
STEPHEN G. PYNE
120. Glass, R. S.; Reineke, K.; Shanklin, M. J. Org. Chem. 1984,49,1527. 121. Craig, D.; Geach, N. J. Tetrahedron: Asymmetry 1991, 2, 1177. 122. Craig, D.;Geach, N. J.; Pearson, C. J.; Slawin, A. M. Z.; White, A. J. R; Williams, D.J. Tetrahedron 1995,57,6071. 123. Johnson, C. R.; Stark. C. J., Jr.; J. Org. Chem. 1982, 47, 1196. 124. Bolm, C ; Muller, J.; Zehnder, M.; Neuburger, M. A. Chem. Eur. J. 1995, 7, 312. 125. Annunziata, R.; Cinquini, M.; Cozzi, F. J. Chem. Soc, Perkin Trans. 11981, 1109. 126 Pearson, A. J.; BIystone, S. L.; Nar, J.; Pinkerton, A. A.; Roden, B. A.; Yoon, J. / Am. Chem. Soc. 1989, 777, 134. 127. Bolm, C ; Muller, J.; Schlingloff, G.; Zehnder, M.; Neuburger, M. J. Chem. Soc, Chem. Commun.1993, 182. 128. Bolm, C ; Bienewald, P.; Harms, K. Synlett 1996, 775. 129. Bolm, C ; Kaufmann, D.; Zehnder, M.; Neuburger, M. Tetrahedron Lett. 1996, 37, 3985. 130. Zehnder, M.; Bolm, C ; Schaffner, S.; Kaufmann, D.; Muller, J. Liebigs Ann. 1995, 125. 131. Johnson, C. R.; Stark, C. J., Jr. Tetrahedron Lett. 1979, 4713. 132. Bolm, C ; Felder, M. Tetrahedron Utt. 1993, 34, 6041. 133. Bolm, C ; Seger, A.; Felder, M. Tetrahedron Lett. 1993, 34, 8079. 134. Bolm, C ; Felder, M. Synlett 1994, 655. 135. Bolm, C ; Muller, J. Tetrahedron 1994, 50, 4355. 136. Bolm, C ; Felder, M.; Muller, J. Synlett 1992, 439. 137. (a) Bolm, C ; Muller, P Tetrahedron Lett. 1995,36,1625. (b) Bolm, C ; Muller, P; Harms, K. Acta Chem. Scand. 1986, 50, 305.
INDEX
Acinetobacter calcoaceticus (CYMO), 62-63 2-Acyl-l, 3-dithianes, asymmetric sulfoxidation of, 141-143 2-Acyl-2-alkyl-1,3-dithiane 1 -oxides conjugate addition of lithium organocuprate, 125-127 "one pot" preparation of, 139-140 stereoselective enolate alkylation, 127-130 systems preparation, 119-120 Acylation, of DiTOX, 139-140 Acylsilanes H2S/HCI reactions, 3-4 Lawesson's Reagent, 4-5 (Me3Si)2S reactions, 4 Alcohol in DAG methodology, 91-96 tautomycin synthesis, 160 Aldehydes 1,3-dithiane dioxide reactions, 147-148 1,3-dithiolane-l-oxide reactions, 148 lithiated sulfoximines reactions, 293-300 "one pot" method for deriving sulfinimines, 255 a-sulfonimidoyl ketones and esters, reactions, 356-358
Alkoxides, intramolecular addition, 172-173 2-Alkyl-2-acyl-1,3-diothiolane 1-oxides, 148-150 Alkyl sulfoximines lithiated sulfoximines, 288-312 synthesis, 284-288 Alkylating reagents, 290-293 Allylic sulfmamides, rearrangements from allylic sulfoximines, 326-337 Allylic sulfoximines lithiated allylic sulfoximines, 318-326 organometallic reagents and substitution reactions, 337-338 rearrangement to allylic sulfmamides, 326-337 synthesis of, 317-318 Allylsulfones in [3+2]cycloadditions sulfmimine reactions, 275-277 Aminoalkylating agents, 130-132 a-Aminoketones, 132-137 a-Aminosulfenimides, used to prepare sulfinimines, 253 Aminosulfites, sulfoxide synthesis, 84-85 Andersen method, 77-109 a-Arylpropanoic acids, 144-145, 146
367
368
Arylsilythiones, 7-8 Asymmetric oxidation of sulfides, 60-77 diastereoselective oxidations, 60-61 enatioselective oxidations, 61-62, 66-77 Asymmetric synthesis, sulfoximines, 283-362 a-Azidosulfoxides, used to prepare sulfinimines, 259 Aziridines, synthesis, 310-312 Azoalkanes, synthesis, 313 Azomethine ylides in [3+2]cycloadditions sulfinimine reactions, 275-277 Biological sulfoxidation, 62-66 Brassinolide, 165 a-Bromo vinyl sulfones, 171 Cancer research, 99, 156 Carbohydrate chemistry, iodine in, 37-54 Carbon bonds cleavage forfluorideions, 29 electroauxiliaries for, 54 formed via Pummerer reactions, 234-236 in silicon induced reorganizations, 221-222 Carbon-carbon bonds electroauxiliaries for, 54 formed via Pummerer reactions, 234-236 in silicon induced reorganizations, 221-222, 233 Carbon nucleophiles conjugate addition to a,P-unsaturated sulfoxides and sulfones, 157-169 sulfinimine reactions, 265-273 Chelation-mediated facially selective cycloaddition reactions, 137-13 8
INDEX
Chiral reagents a,P-unsaturated sulfoxides, 156 Chirality alkenes, axial chirality, 314-316 base effect, 91-94 carbon or silicon and transfer, 19-20 chiral controllers, 104, 204 criteria for ideal auxiliary, 118-119 cyclic sulfides, 117-118 cyclic sulfoxides as chiral auxiliaries, 117-150 Diels-Alder reaction, 185 ethoxy vinyl esters (EVE) and transfer, 237-239 selenoxides and transfer, 244-246 sugars as source, 160 of sulfoxides, 58-59 sulfoxides used to direct chiral centers, 156 thioacylsilanes, 19-20 Claisen rearrangements, 200-201 Cross-coupling reactions, 350-352 Cyanide ions, reactions in vinyl sulfoximines, 350 Cyanides, sulfinimine reactions, 266-267 Cyclic sulfoxides as chiral auxiliaries, 118-147 [l+2]Cycloadditions, sulfinimine reactions, 273-275 [3+2]Cycloadditions, sulfinimine reactions, 275-277 Cycloalkanone synthesis, one pot method, 140-141 Cyclopropanation reactions lithiated sulfoximines, 306-310 Simmons-Smith, 316 a,(3-unsaturated sulfoxides and sulfones, 201-206 CYMO (Acinetobacter calcoaceticus), 62-63
Index
DAG methodology (See diacetone-D-glucose (DAG) methodology) Danishefsky's diene, 137-138 Diacetone-D-glucose (DAG) methodology, 89-109 applications, 99-109 base effect, 91-94 origins of diastereoselectivity, 96-98 scope and limitations, 98-99 Dialkyl sulfoxides, 74 Diastereoselective oxidation of sulfides, 60-61 Diastereoselectivity in development of DiTOX asymmetrical building blocks, 120-125 Diasteroselective synthesis, sulfoximines, 283-362 Diazenes (azoalkanes), synthesis, 313 Diazomethane, 184-185 Diels-Alder reactions, 137-138, 150 creation of chiral centers, 185 intramolecular (IMDA), 194-195 ketene equivalents in, 192-193 sulfmimines, 277-278 vinyl sulfoxides and vinyl sulfones, 185-195 vinyl sulfoximines, 352-355 yielding deltacyclane, 180-181 Dienes and cycloaddition thioacylsilanes, 10-11 thioaldehyde S-oxides, 19 Dienolates, sulfmimine reactions, 270-272 (/?)-(-)-2,6-dimethylheptanoic acid, 143-144 1,3 Dipoles and cycloaddition thioacylsilanes, 9-10 "Disarmed" thioglycosides, activation of, 44-49 1,3-Dithiane 1-oxides, 118-147 1,3-Dithiane dioxides, 147-148 1,3-Dithiolane-1-oxides, 148-150
369 1,3-Dithiolane dioxides, 150 ^^m-Dithiols, 3 DiTOX, 118-147 acylation of, 139-140 asynmietric building blocks, 120-147 Perkin ring synthesis using, 140 preparation of optically pure substrates, 141-143 DiTOX asynmietric building blocks applications of, 143-144 development of, 120-147 Grignard reagents and ketones, 120-122 Electron spin resonance (ESR) spectroscopy, 31-32 Electrophiles reaction with Z-a-silyl vinyl sulfides, 25-27 a-vinyl anion reactions, 177-179 Electrophilic additions to a,P-unsaturated sulfoxides and sulfones, 176-179 Enantiomeric sulfides, synthesized with DAG, 102 Enantioselective sulfoxidation, 61-62 by chiral oxaziridines, 75-76 by chiral peroxides, 68-69 metallo-(salen)-catalyzed oxidation, 72-75 Orsay system, 67-68 Padova system, 68-69 Pasini system, 72 Sharpless method, 66-67, 69-72 Enantiospecificity, in Punmierer reactions, 230-231 Enethiolizable thioacylsilanes, 20-25 Enolates amination and a-aminoketones, 134-137 bromination, 132-134
370
stereoselective enolate alkylation, 127-130 Enones, reactions with lithiated sulfoximines, 306-310 Enzyme-catalyzed sulfoxidation, 62-64 Ephedrine sulfonamides, 84-85 Epoxidation of a,P-unsaturated sulfoxides and sulfones, 201-204 Epoxy vinyl sulfoxides, 163-164 Eschenmoser's salt, 130-131 ESR (electron spin resonance) spectroscopy, 31-32 Esters sulfmate esters, synthesis, 81-83 a-sulfonimidoyls, 355-356 Ethoxy vinyl esters (EVE), 236-239 Ruoride ions, reaction with Z-a-silyl vinyl sulfides, 29-31 Fluoro-desilylation, stereochemistry of, 18-19 Free radicals, 31-32 Fungi and sulfoxides, 64-65 Gallili antigen, 43 D-Glucose derivatives, synthesis of, 160 Glyco-amino acid synthesis, 43-44 Glycobiology, 38 Glycopeptides, synthesis, 43-44 Glycoside synthesis, 38 (see also thioglycosides) Glycoside synthesis, tuning donor reactivity, 50-51 Glycosyl donors, 38 Glycosyl sulfoxides, 51-53 Glycosylation, iodine-promoted, 41-42 Grignard reagents [3,3]-sigmatropic rearrangements, 200 sulfinimine reactions, 267-269
INDEX
a-Halo carbolic acids, 132-134 a-Haloenolates, [1+2] cycloaddition and sulfinimines, 274-275 a-Haloketones, 132-134 a-Halophosphonate carbanions, [1+2] cycloaddition and sulfinimines, 275 Halosulfoximides, used to prepare sulfinimines, 258-259 Helminthosporium, 65 Hetero Diels-Alder reactions, 277-278 Heteroatom nucleophiles, conjugate addition to a,P-unsaturated sulfoxides & sulfones, 170-176 Heterodienes and cycloaddition, thioacylsilanes, 12 H2S/HCI, reactions with acylsilanes, 3-4 Hydrides, reduction reaction of sulfinimines, 263-264 a-Hydroxy acids, synthesis, 147-148 2-Hydroxy groups in organic synthesis, 160 P-Hydroxy sulfoximines alkene synthesis, 314-315 in asymmetric synthesis, 313-316 enantioselective reactions, 316 osmylations, directed, 316 reductive elimination of, 314 resolution of racemic chiral cyclic ketones, 313-314 Simmons-Smith cyclopropanations, 316 P-Hydroxy sulfoximines, synthesis, 293-302 a-Hydroxyketones, 145-147 Imines reactions with lithiated sulfoximines, 302-306 sulfur bonding, 249-278 Interhalogens as thioglycoside activator, 46-49
Index
Iodine in carbohydrate chemistry, 37-54 chemoselective activation, 42-43 chemospecific activation of thioglycosides, 42-43 glycosylation promotion by, 41-42 interhalogens as source, 46-49 Lewis acids and, 51-54 as thioglycoside activator, 45-46 as thiophilic reagent, 39-41 IPNS (isopenicillin synthetone), 244 Iron-mediated reactions of a,P-unsaturated sulfoxides and sulfones, 208 Isopenicillin synthetone (IPNS), 244 Ketene equivalents, 192-193 Ketones development of DiTOX asymmetrical building blocks, 120-125 P-hydroxy sulfoximines and, 313-314 reactions with lithiated sulfoximines, 300-302 resolution of racemic chiral cyclic, 313-314 a-sulfonimidoyls, 355-356 Lawesson's Reagent and acylsilanes, 4-5 Lewis acids, 137-138 as chelating agents, 191 iodine as, 51-54 promoting cyclization of A^-nitrosulfoximines, 313 Ligands for asymmetric synthesis, sulfoximines, 359-362 Lithiated allylic sulfoximines, 318-326 Lithiated sulfoximines aldehyde reactions, 293-300
371
alkylating reagents, reactions, 290-293 cyclopropanes, synthesis, 306-310 enone reactions, 306-310 P-hydroxy sulfoximines, synthesis, 293-302 imine reactions, 302-306 ketone reactions, 300-302 oxiranes, oxetanes, and aziridines, synthesis, 310-312 structural studies, 288-290 Lithium organocuprate reagents, 125-127, 165 Mannich reactions P-aminoketones, 134 asynmietric, 130-132 (Me3Si)2S, reactions with acylsilanes, 4 Metal-catalyzed reactions, 206-208 Methanesulfmates, 98-100 Methanesulfmyl chloride, reactions, 95 /ran^-2-Methylene-1,3-dithiolane dioxides, 150 Methylthioketones, 7 Michael-initiated ring closure (MIRC) process, 172-173,206 Michael reactions oxygen nucleophiles, 349-350 sulfmimines, 250, 265-273 synthesis of a-silyl vinyl sulfides, 24 vinyl sulfoximines, 340-349 Microbiological sulfoxidation, 64-66 Mortierella isabellina, 65 Nitrile oxides, 181-182 Nitrogen nucleophiles Michael reactions in sulfoximines, 346-349 sulfmimine reactions, 265 Nitrones, 182-184 ^-Nitrosulfoximines, cyclizations of, 313
372
nucleophiles carbon, 157-169, 265-273 in displacement reactions, 134 heteroatom, 170-176 nitrogen, 346-349 oxygen, 49, 265, 349-350 phosphorus, 265 reaction with Z-a-silyl vinyl sulfides, 27-29 sulfur, 265 Nucleophilic substitution of chiral sulfur derivatives, 77-109 Olefins, photoinduced cycloaddition of thioacylsilanes, 12-13 Oligosaccharide structures, synthesis, 38 Organocuprates 2-acyl-2-alkyl-l,3-dithiane 1 oxide conjugate addition, 125-127 8^2'reactions, 163-164 Organometallic reagents DiTOX asymmetric building blocks, 125-127 and DiTOX asynunetric building blocks, 120-122 substitution reactions of allylic sulfoximines, 337-338 sulfite reactions, 79-81 thioacylsilanes, 8-9 vinyl sulfoximines, 340-344 Organosulfiir chemistry, a,p-unsaturated sulfides and sulfones, 155-208 Orsay system, 67-68 Osmylations, directed, 316 Oxaziridines, chiral, 75-76 Oxazolidinones, chiral, 85-89 Oxetanes, synthesis, 310-312 Oxidation of thioacylsilanes, 13-14 Oxime carbanions, sulfinimine reactions, 272 Oxiranes, synthesis, 310-312
INDEX
Oxisuran, 100-101 Oxygen nucleophiles Michael reactions in sulfoximines, 349-350 reactions of glycosyl halides, 49 sulfinimine reactions, 265 Padova system, 68-69 Palladium-mediated reactions of a,P-unsaturated sulfoxides and sulfones, 206-208 Pasini system of enantioselective sulfoxidation, 72 Penicillin biosynthesis, 241, 244 Pericyclic reactions of a,(3unsaturated sulfoxides and sulfones [2+2] cycloadditions, 179-181 [3+2] cycloadditions, 181-185 [4+2] cycloadditions, 185-196 Perkin ring synthesis, 140, 141 Peroxidases and sulfoxidation, 62-64 Peroxides, chiral, 68-69 Pharmaceutical applications, 69-71, 100-101, 156 rran5-2-Phenylcyclohexanol, 81-83 a-Phosphate carbanions, sulfinimine reactions, 272-273 Phosphines, reduction reaction of sulfinimines, 264 Phosphorus nucleophiles, sulfinimine reactions, 265 a-Phosphoryl vinyl p-tolyl sulfoxides, 205 Piperylene, vinyl sulfoxides and Diels-Alder reaction, 188-189 Polonovski-type rearrangements, 222-224 Prochiral sulfides catalytic asymmetric oxidation, 143 enantioselective oxidation of, 61-62
Index
Pummerer reactions/rearrangements carbon-carbon bond formation, 234-236 described, 216 effect of substituent groups on silyl function, 233-234 enantiospecificity, 230-231 ethoxy vinyl esters (EVE) used to induce, 236 intermolecular cyclization and SKA, 239-244 mechanistic interpretations, 216-217 planning for asymmetric, 219-224 radical P-additions to sulfoxides, 196-200 selenoxides, 244-246 sila-Pummerer or silyl Pummerer rearrangements, 220 silicon-induced, 219-222 O-silylated ketene acetals (SKA), 197, 220-230 stereoselective examples, 217-219 vinyl sulfoxides, 224 vinylogous, 198-199 Ramberg/Backlund rearrangement, 171 Resonance stabilized carbanions, 345 Sj^2' additions to a,P-unsaturated sulfoxides and sulfones, 157-169 S-oxides (silyl sulfmes), 13-14, 16-20 Selenoxides, 244-246 Sharpless method, 66-67, 142 Silicon-induced Pummerer and Polonovski rearrangements, 219-224 Z-a-Silyl enethiols, 20-21 a-Silyl enethiols, 21-22 Silyl functions, 233-234 Silyl sulfines (S-oxides), 13-14, 16-20
373
Z-a-Silyl vinyl sulfides, 25-27 a-Silyl vinyl sulfides, 22-30, 25-30 O-Silylated ketene acetals (SKA) additive Pummerer reactions, 224, 231-233 induced Pummerer rearrangements, 224-227,230-231 Polonovski reaction, 222-224 Pummerer reactions, 197 reaction mechanism of Pummerer rearrangements, 227-230 SKAs and silyl function, 233-234 sufoxides and Pummerer rearrangements, 220-222 [3,3]-Sigmatropic rearrangements, 200-201 Simmons-Smith cyclopropanations, 316 SKA (see O-silylated ketene acetals (SKA)) Solid-phase organic synthesis (SPOS), 170-171 Sparsomycin, 99, 156 Spectroscopy electron spin resonance (ESR), 31-32 thioacylsilanes, 6-7 Spin trapping agents, thioacylsilanes, 31-32 SPOS (solid-phase organic synthesis), 170-171 Sugars and chirality, 160 Sulfenic acids (RSOH), 250 Sulfenimines, used to prepare sulfinimines, 250-253 Sulfides asymmetric oxidation of, 60-77 methods to prepare chiral sulfoxides, 142-143 Sulfinamides, used to prepare sulfinimines, 256-258 Sulfinate esters, synthesis of, 81-83
374
Sulfinates, used to prepare sulfmimines, 254-256 Sulflnimines from a-aminosulfenimides, 253 from a-azidosulfoxides, 259 elimination reactions, 262 halogenation reactions, 263 Michael addition reactions, 265-273 from iV-sulfmylamides, 259-260 oxidation reactions, 262-263 preparation of, 250-260 reduction reactions, 263-264 from sulfenimines, 250-253 from sulfinamides, 256-258 from sulfinates, 254-256 from sulfinyl halides, 253-254 from sulfonylazides, 260-261 synthesis and reactions, 249-278 from thiosulfinates, 256 Sulfinyl group, effectiveness as chiral controllers, 58 Sulfinyl halides, used to prepare sulfinimines, 253-254 iV-sulfinyl oxazolidinones, 85-89 yV-sulfinylamides, used to prepare sulfinimines, 259-260 Sulfite methodology, synthesis of chiral sulfoxides, 78-81 Sulfones, a,P-unsaturated, 155-208 a-Sulfonimidoyl ketones and esters aldehyde reactions, 356-358 alkylating reagent reactions, 356 reduction reactions, 358-359 synthesis, 355-356 Sulfonium intermediates, 219 a-Sulfonyl carbanions, sulfinimine reactions, 272 Sulfonylazides, used to prepare sulfinimines, 260-261 Sulfoxidation biological, 62-66 predicting absolute configuration at sulfur, 143
INDEX
Sulfoxides, 57-110 biologically active, structure of, 59 chiral sulfoxide synthesis, 57-110 cyclic, 117-150 methods to prepare chiral sulfoxides from sulfides, 142-143 nucleophilic substitution on, 77-109 optically pure, 60, 77-109 silicon-induced rearrangements, 219-224 a,P-unsaturated sulfoxides, 155-208 Sulfoximides, used to prepare sulfinimines, 258-259 Sulfoximines alkyl sulfoximines, 284-313 allylic sulfoximines, 317-338 diastereoselective and asymmetric synthesis, 283-362 P-hydroxy sulfoximines and asymmetric synthesis, 313-316 ligands for asymmetric synthesis, 359-362 a-sulfonimidoyl ketones and esters, 355-358 vinyl sulfoximines, 338-355 Sulfur bonding imines, 249-278 Sulfur heterocycles five-membered, 138, 148-150 six-membered, 118-148 Sulfur nucleophiles, sulfinimine reactions, 265 Sulfur stereogenic center, 261-262 Sulftirane intermediates, 219 Thietane formation, 12-13 Thio-Claisen rearrangements, 199-201 Thioacetyl trimethylsilane, decomposition, 7 Thioacylsilanes, 1-32 chirality, 19-20 dienes and cycloaddition, 10-11 1,3 dipoles and cycloaddition, 9-10
Index
enethiolizable thioacylsilanes, 20-25 heterodienes and cycloaddition, 12 organometallic reagents and, 8-9 oxidation, 13-14 photoinduced cycloaddition with olefins, 12-13 reactivity of aromatic, 8-14 a-silyl vinyl sulfides, 25-30 spectroscopy of, 6-7 spin trapping agents, 31-32 synthesis of, 3-5 thermal stability of, 7-8 thioaldehyde equivalents, 14-19 Thioaldehydes S-oxides, 16-20 thioacylsilanes as equivalents, 14-19 Thioalkyltrimethylsilanes, alternatives to, 53-54 Thioglycoside donors, 42-43 tuning reactivity of, 50-51 Thioglycosides, 39 activation of "disarmed," 44-49 "disarmed" activation, 44-49 donors, 42-43, 50-51 synthesis of, 53-54 Thiolactones, 24-25 Thionation deviating behavior during, 7-8 methods, 3-5 Thiones, cycloadditions of, 12 Thiophilic reagents, iodine, 39-41 Unsaturated silylated thiolactones, 24-25 a,p-Unsaturated sulfones (See a,P-Unsaturated sulfoxides and sulfones) a,P-Unsaturated sulfoxides and sulfones carbon nucleophiles, 157-169 cyclopropanation reactions and, 204-206
375
electrophilic additions to, 176-179 epoxidation of, 201-204 heteroatom nucleophiles, 170-176 iron-mediated reactions of, 208 metal-catalyzed reactions of, 206-208 nucleophilic additions, 157-176 palladium-mediated reactions of, 206-208 pericyclic reactions of, 179-186 Dsiltoxins, 156 a-Vinyl anions of vinyl sulfone, 178-179 Vinyl sulfones anticancer agents, 156 Diels-Alder reactions, 185-195 electrophilic additions, 176-179 intermolecular conjugate additions to, 158-159 intermolecular cycloadditions to, 193-194 intramolecular cycloadditions to, 194-196 iron-mediated reactions, 208 palladium-mediated reactions, 206-208 Vinyl sulfoxides chirality, 205 Diels-Alder reactions, 185-195 electrophilic additions, 176-179 epoxy vinyl sulfoxides, 163-164 intermolecular conjugate additions to, 157-158 intermolecular cycloadditions to, 185-193 a-phosphcM-yl vinyl p-tolyl sulfoxide, 205 Pummerer reactions, 224 Vinyl sulfoximines cross-coupling reactions, 350-352 cyanide ions in reactions, 350 cycloaddition reactions, 352-35
INDEX
376
Michael reactions, 340-349 nitrogen nucleophiles, 346-349 organometallic reagents, 340-344 oxygen nucleophiles, 349-350 resonance-stabilized carbanions, 345 synthesis, 338-339
Vinylogous Pummerer reactions, 198-199 Xenotransplantation, 43 Yeast and sulfoxides, 65-66