Miniemulsion Polymerization Technology

Miniemulsion Polymerization Technology Vikas Mittal BASF SE, Polymer Research, Germany Scrivener )WILEY This Page...

Author: Vikas Mittal

56 downloads 1250 Views 19MB Size Report

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

Report copyright / DMCA form

DOWNLOAD PDF

Miniemulsion Polymerization Technology

Vikas Mittal

BASF SE, Polymer Research, Germany

Scrivener

)WILEY

This Page Intentionally Left Blank

Miniemulsion Polymerization Technology

Scrivener Publishing 3 Winter Street, Suite 3 Salem, MA 01970 Scrivener Publishing Collections Editors James E. R. Couper Richard Erdlac Pradip Khaladkar Norman Lieberman W. Kent Muhlbauer S. A. Sherif

Ken Dragoon Rafiq Islam Vitthal Kulkarni Peter Martin Andrew Y. C. Nee James G. Speight

Publishers at Scrivener Martin Scrivener ([email protected]) Phillip Carmical ([email protected])

Miniemulsion Polymerization Technology

Vikas Mittal

BASF SE, Polymer Research, Germany

Scrivener

)WILEY

Copyright © 2010 by Scrivener Publishing LLC. All rights reserved. Co-published by John Wiley & Sons, Inc. Hoboken, New Jersey, and Scrivener Publishing LLC, Salem, Massachusetts. Published simultaneously in Canada. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, (978) 750-8400, fax (978) 750-4470, or on the web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., Ill River Street, Hoboken, NJ 07030, (201) 748-6011, fax (201) 748-6008, or online at http://www.wiley.com/go/permission. Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages. For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at (800) 762-2974, outside the United States at (317) 572-3993 or fax (317) 572-4002. Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic formats. For more information about Wiley products, visit our web site at www.wiley.com. For more information about Scrivener products please visit www.scrivenerpublishing.com. Cover design by Russell Richardson. Library of Congress Cataloging-in-Publication ISBN 978-0-470-62596-5

Printed in the United States of America 10 9 8 7 6 5 4 3 2 1

Data:

Contents Preface

1

2

3

Miniemulsion Polymerization: An Overview y. Mittal 1.1 Introduction to Polymerization Techniques 1.2 Emulsion and Miniemulsion Polymerization 1.3 Properties of Miniemulsion Polymerization 1.4 Controlled Miniemulsion Polymerization References Multi-Functional Stabilizers in Miniemulsion Polymerization Alain Durand 2.1 Introduction 2.2 Stability of Initial Monomer Droplets 2.3 Stabilizers and Polymerization Processes 2.3.1 Mass-Transfer Processes 2.3.2 Reactive Stabilizers 2.4 Conclusion References Structured Copolymer Particles by Miniemulsion Polymerization V. Mittal 3.1 Introduction 3.2 Styrene-Dodecyl Methacrylate/Stearyl Methacrylate 3.3 n-Butyl Methacrylate-Crosslinking Monomers 3.4 Vinyl Acetate-Butyl Acrylate

xiii

l 2 3 10 19 22

25 25 27 30 30 31 39 39

43 43 46 49 51 v

vi

CONTENTS

3.5

Butyl Acrylate-(2-Methacryloxy)ethyl)trimethyl Ammonium Chloride 3.6 Butyl Acrylate-Methyl Methacrylate-Vinyl Acetate 3.7 Styrene-Acrylic Acid or 2-Aminoethyl Methacrylate Hydrochloride (AEMH) 3.8 Styrene-Butyl Acrylate 3.9 Styrene-Butadiene Rubber 3.10 Fluoroacrylate—LaurylMethylacrylate—Methyl Methacrylate 3.11 Polyurethane-Block-Polystyrene 3.12 Alkyd-Acrylic 3.13 Oil-Acrylate 3.14 Urethane-Acrylic References Encapsulation of Inorganic Nanoparticles by Miniemulsion Polymerization Jacqueline Forcada and Jose Ramos 4.1 Introduction 4.2 Miniemulsion Polymerization in the Presence of Inorganic Nanoparticles 4.2.1 Hydrophobization of Inorganic Nanoparticles 4.2.2 Dispersion of Hydrophobized Inorganic Nanoparticles in Monomer Phase 4.2.3 Miniemulsification of the Lipophilic Dispersion in Water 4.2.4 Polymerization of Droplets 4.3 Encapsulation of Silica Nanoparticles 4.3.1 Miniemulsion Polymerization with Hydrophilic Silica Nanoparticles 4.3.2 Miniemulsion Polymerization with Surface-Modified Silica Nanoparticles 4.3.3 Miniemulsion Polymerization with Locally Surface-Modified Silica Nanoparticles 4.4 Encapsulation of Magnetite Nanoparticles 4.4.1 Encapsulation of Magnetite by a Single Miniemulsion Polymerization Process 4.4.2 Encapsulation of Magnetite by a Double Miniemulsion Polymerization Process

53 54 55 57 57 61 62 63 65 67 68 71 71 73 73 75 75 76 76 77 78 83 85 86 89

CONTENTS

4.5 4.6

Conclusions and Future Perspectives Acknowledgements References

Polymeric Nanocapsules by Interfacial Miniemulsion Polymerization Guo-Rong Shan and Zhi-Hai Cao 5.1 Introduction 5.2 Organic Nanocapsules by Inter facial Miniemulsion Polymerization 5.2.1 Thermodynamic Prediction for the Morphology of Organic Nanocapsules 5.2.2 Particles Morphology of the System without Added NIPAM and DVB 5.2.3 Particles Morphology of the System with DVB 5.2.4 Particle Morphology of the System with Added NIPAM and DVB 5.2.5 Particle Size and Size Distribution in the Process of Polymerization 5.2.6 Mechanism for the Formation of Organic Nanocapsules through Interfacial Miniemulsion Polymerization 5.2.7 Influences on the Formation of Organic Nanocapsules through Interfacial Miniemulsion Polymerization 5.3 Organic-Inorganic Hybrid Nanocapsules by Interfacial Miniemulsion Polymerization 5.3.1 Thermodynamic Analysis and Morphological Prediction 5.3.2 Synthesis of Organic-Inorganic Hybrid Nanocapsules under Neutral Conditions 5.3.3 Synthesis of Organic-Inorganic Hybrid Nanocapsules under Acidic or Basic Conditions 5.3.4 Mechanism Analysis of Organic-Inorganic Hybrid Nanocapsules Formation 5.4 Conclusions References

91 92 92 97 97 99 99 101 103 105 109 112

113 117 117 119

124 134 136 137

viii

CONTENTS

Miniemulsion Polymerization of Vegetable Oil Macromonomers Shelby F. Thames, James W. Razvlins, and Sharathkumar K. Mendon 6.1 Introduction and Background 6.2 Emulsion Polymerization of Alkyds and Vegetable Oils 6.3 (Meth)acrylated Vegetable Oil Derivatives 6.4 Vegetable Oil Macromonomers 6.5 The Potential for Emulsion Polymerization of Model Saturated Monomers 6.6 Nucleation Mechanisms 6.7 Design of Thermosetting Latex Polymers 6.8 Classifying Monomer Solubility for Macro and Miniemulsion Polymerization 6.9 Soybean Acrylated Monomer Synthesis 6.10 Miniemulsion Polymerization 6.11 Conclusions References Controlled/Living Radical Polymerization in Aqueous Miniemulsion Catherine Lefay, and Julien Nicolas 7.1 Introduction 7.2 Controlled/Living Radical Polymerization in Bulk/Solution: General Considerations 7.2.1 CLRP Based on Reversible Termination 7.2.1.1 Nitroxide-Mediated Polymerization (NMP) 176 7.2.1.2 Atom Transfer Radical Polymerization (ATRP) 177 7.2.2 CLRP Based on Degenerative Transfer 7.2.2.1 Reversible Addition-Fragmentation Chain Transfer (RAFT) 7.2.2.2 Iodine Transfer Polymerization (ITP) 7.3 Nitroxide-Mediated Miniemulsion Polymerization 7.3.1 Oil-Soluble Bicomponent Initiating System 7.3.2 Water-Soluble Bicomponent Initiating System

139

139 143 145 146 150 152 154 158 160 160 168 169 173 174 174 175

178 179 180 182 183 185

CONTENTS

7.3.3 7.3.4

Oil-Soluble Monocomponent Initiating System Water-Soluble Monocomponent Initiating System 7.4 Atom Transfer Radical Miniemulsion Polymerization 7.4.1 Direct ATRP 7.4.2 Reverse ATRP 7.4.3 Simultaneous Reverse and Normal Initiation (SR&NI) ATRP 7.4.4 Activators Generated by Electron Transfer (AGET) ATRP 7.5 Reversible Addition-Fragmentation Chain Transfer Miniemulsion Polymerization 7.5.1 Key-Steps for the Success of RAFT Miniemulsion Polymerization 7.5.1.1 Inhibition and Retardation 7.5.1.2 Colloidal Instability 7.5.1.3 Livingness and Controlled Polymerization 7.5.2 RAFT Miniemulsion Polymerization of Vinyl Acetate 7.5.3 Nanocapsules Synthesized by RAFT Miniemulsion Polymerization 7.6 Iodine Transfer Polymerization in Miniemulsion 7.7 Conclusion References Inverse Miniemulsion Polymerization of Unsaturated Monomers Ignác Capek 8.1 Introduction 8.2 General 8.3 Kinetic Studies 8.4 Traditional and Nonconventional Inverse Latexes 8.4.1 Water Soluble Monomers 8.4.2 Hydrophobie Monomers 8.5 Controlled Radical Miniemulsion Polymerization 8.6 Amphiphilic and Associating Copolymers 8.7 Conclusion

ix

186 188 188 190 190 192 192 193 194 194 196 198 199 200 201 202 203 211 211 215 218 221 221 230 232 237 240

x

CONTENTS

8.8

9

Acknowledgements Abbreviations References

Double Miniemulsion Preparation for Hybrid Latexes R.Y. Hong, G. Liu, B. Feng, and H.Z. Li 9.1 Introduction 9.2 Hybrids via Mini-Emulsion Polymerization 9.3 Double-Miniemulsion Formation 9.4 Stability 9.5 Characterization 9.6 Applications 9.6.1 Effects of Reaction Conditions 9.6.1.1 Initiator Dosage 9.6.1.2 MMA Monomer Concentration 9.6.2 Rheological Properties of Magnetic Emulsion 9.6.2.1 Viscosity Versus Time 9.6.2.2 Viscosity with/without Magnetic Field 9.6.2.3 Applications of Magnetic Polymer Microspheres 9.7 Summary 9.8 Acknowledgments References

10 Surfactant Effect in Miniemulsion Polymerization for Biodegradable Latexes V. Soldi, B.G. Zanetti-Ramos, and E. Minatti 10.1 Introduction 10.2 Miniemulsion Polymerization of Biodegradable Latexes 10.3 Mechanisms of Surfactant Protection of Colloidal Dispersions 10.3.1 General Behavior of a Surfactant Molecule at the Interface 10.3.2 Mechanism 1: Lowering the Interfacial Tension 10.3.3 Mechanism 2: Electrostatic Stabilization 10.3.4 Mechanism 3: Steric Stabilization

244 244 246 251 252 253 255 255 257 261 266 266 267 268 268 269 270 271 272 272 277 277 278 282 282 284 286 288

CONTENTS

4

5

Effect of Surfactants on Miniemulsion Polymerization 10.4.1 Effect of Surfactant Type on the Particle Size and Latex Yield 10.4.2 Effect of Surfactant Concentration on Particle Size and Latex Yield 10.4.3 Effect of Surfactant on the Stability Final Remarks References

xi

291 291 294 297 298 299 303

This Page Intentionally Left Blank

Preface Miniemulsion polymerization provides many distinct advantages over the conventional emulsion polymerization because the monomer droplets are directly polymerized, whereas in the case of emulsion polymerization, the monomer is polymerized in the micelles and needs to travel through the aqueous phase. The polymerization of very hydrophobic monomers is thus difficult in the case of emulsion polymerization owing to limited diffusion through the aqueous phase. The miniemulsion polymerization does not suffer from these limitations and can lead to the polymerization of very hydrophobic as well as very hydrophilic monomers. The system parameters in miniemulsion polymerization include the choice of surfactant, co-surfactant, agitation speed, and by careful control on these parameters one can achieve efficient control on the polymerization reaction. The miniemulsion polymerization is also well suited for the various controlled living polymerization reactions like nitroxide mediated polymerization, atom transfer radical polymerization, as well as radical addition fragmentation chain transfer. Apart from that, the miniemulsion polymerization is also beneficial for the synthesis of encapsulated particles as well as core-shell particles to generate advanced polymer nanoparticles. There have been a large number of texts on emulsion and other forms of polymerization methods, but miniemulsion polymerization, though it provides unique routes for polymer particle synthesis, has been neglected. This book summarizes the recent advances in miniemulsion polymerization technology including the advances on the selection of surfactants and co-surfactants, the expansion of miniemulsion technology in various polymers and co-polymer systems, and the use of miniemulsion polymerization for the synthesis of advanced polymer particle morphologies. Chapter 1 provides the background information on the kinetics and mechanism of xm

xiv

PREFACE

miniemulsion polymerization technology as well as its comparison with the conventional emulsion polymerization. The importance of the selection of co-surfactants, initiators and surfactants is drawn out. Chapter 2 summarizes the advances in the multi-functional surfactants, i.e. the stabilizers which can be involved in other aspects of polymerization in addition to their function as stabilizers. Chapter 3 details the synthesis of structured copolymer particles by using miniemulsion polymerization. A large number of polymer and copolymer systems are demonstrated. Chapter 4 deals with the synthesis of the organic-inorganic hybrid particles by the encapsulation of silica and magnetite nanoparticles by polymers. Chapter 5 describes the synthesis of polymeric nanocapsules by interfacial miniemulsion polymerization. Miniemulsion polymerization is well suited for the polymerization of hydrophobic monomers, especially vegetable oil macromonomers that are too hydrophobic to traverse the aqueous phase during conventional emulsion polymerization. Such a system is described in Chapter 6. General considerations about four main CLRP techniques, namely nitroxide-mediated radical polymerization (NMP), atom transfer radical polymerization (ATRP), reversible addition-fragmentation chain transfer (RAFT), and iodine transfer polymerization (ITP) are given along with the review of their application in aqueous miniemulsion in Chapter 7. Chapter 8 focuses on the inverse miniemulsion polymerization of unsaturated monomers. Chapter 9 describes the double miniemulsion processes for the synthesis of organic-organic hybrid latexes. Chapter 10 focuses in general on the synthesis of the biodegradable latexes and specifically on the surfactant effects in these latexes. I would like to take this opportunity to express my heartfelt thanks to Scrivener Publishing for agreeing to publish the book. The contribution of my family especially my mother is to be mentioned as the book would not have been possible without it. Many thanks to my wife Preeti for co-editing the book as well as for many ideas for the better presentation of the information contained in the book. Vikas MITTAL Ludwigshafen.

1 Miniemulsion Polymerization: An Overview* V. Mittal BASF SE, Polymer Research, 67056 Ludwigshafen, Germany

Abstract

Miniemulsion polymerization exhibits distinct advantages over the conventional emulsion polymerization. The polymerization of monomer droplets is directly achieved when the diffusion of the monomer from the droplets to polymer particles is not required. This helps to polymerize water insoluble monomers. It also allows the presence of various system components like initiators, costabilizers, etc. directly at the site of the polymerization thus allowing better control. Conventionally, volatile hydrophobes or costabilizers like cetyl alcohol or hexadecane have been used. A number of advances have been reported in recent years on the use of costabilizers which are more compatible to the polymerization system. Use of polymers as costabilizers was reported to be very effective even though the polymer forms a poor costabilizer. Similarly, the comonomers and initiator have also been used as costabilizers. The use of chain transfer agents as costabilizers also opens the opportunities for molecular weight control in the polymer particles. These advances ensure that the particles are free from any low molecular weight impurity or volatile components. The living polymerization techniques like nitroxide mediated polymerization, atom transfer radical polymerization and reversible addition fragmentation chain transfer are also well suited for miniemulsion processes in order to generate specific morphologies in polymer particles and to control the molecular weight and its distribution in the particles. Keywords: diffusion, monomer droplet, costabilizer, surfactant, initiator, micelles, chain transfer, comonomer, miniemulsion, conversion, colloidal stability, controlled living polymerization.

•This review work was carried out at Institute of Chemical and Bioengineering, Department of Chemistry and Applied Biosciences, ETH Zurich, Zurich, Switzerland. V. Mittal (ed.) Miniemulsion Polymerization Technology, (1-23) © Scrivener Publishing LLC

1

2

1.1

MINIEMULSION POLYMERIZATION TECHNOLOGY

Introduction to Polymerization Techniques

Free radical polymerization can be carried out by using a number of different techniques. The simplest of these techniques is the bulk polymerization. In this technique, monomer is in the liquid form and the generated polymer is in solid form. Though the reaction mixture is free from any unwanted impurities or contamination leading to clean polymer, however, the viscosity of the system increases significantly due to the generation of polymer chains during the course of polymerization and thus the mixing of the system becomes extremely difficult leading to very broad molecular weight distributions in the polymer chains. Additionally, the polymer chains do no diffuse freely in the highly viscous medium leading to the accumulation of radicals at particular sites causing the polymerization rate to increase exponentially. Solution polymerization is an alternative method in which a solvent is used in which the monomer and polymer are soluble. The use of solvents eliminates the problems of higher viscosity and heat dissipation associated with bulk polymerization and allows one to stir the reaction medium easily. However, the choice of solvents must be proper; otherwise extensive chain transfer to solvent can take place resulting in only low molecular weight polymer chains. Precipitation polymerization is another form of polymerization in which the polymer is not soluble in the monomer or the reaction medium and precipitates out soon from the solution. Thus, precipitation polymerization starts as homogenous polymerization, but is soon turns into a heterogeneous polymerization. Dispersion polymerization is also similar to precipitation polymerization that the polymer formed is not soluble in monomer or organic solvent. After the formation of polymer particles, these particles are stabilized by added particle stabilizer and the polymerization proceeds in the polymer particles by the absorption of monomer into the polymer particles. Suspension or bead polymerization is a method in which monomer droplets are directly polymerized to generate high molecular weight polymers. In this method, water insoluble monomer is suspended in water with the aid of suspension stabilizers. The initiator used is also water insoluble or monomer soluble. The size of monomer droplets can be controlled depending on the ratio of monomer to the dispersion medium, i.e., water, the speed of agitation to generate droplets as well as by the amount of stabilizing agents.

MINIEMULSION POLYMERIZATION

3

During the polymerization, the monomer droplets polymerize independently and each droplet can be visualized as a bulk polymerization happening inside the droplet. Inverse suspension polymerization is also possible in which a water soluble monomer is used and its droplets are generated in an organic solvent. Initiators used are also water soluble and the monomer droplets are stabilized similarly by using suspension stabilizers. Emulsion polymerization is the one of the most versatile technique to generate small particles. With this technique, water insoluble monomers are polymerized by suspending them in water in the form of emulsion droplets stabilized by surfactants. The initiators used are water soluble in contrast to suspension polymerization where water insoluble initiators are used, the most common being potassium persulphate (KPS). Polymerization of extremely low water soluble monomers is very difficult with conventional emulsion polymerization. The low solubility of the monomer would not allow its diffusion to the polymer particles through the aqueous medium. Miniemulsion polymerization has been developed for such purposes, in which the monomer droplets generated by using high shear in the presence of an ionic surfactant and a co-surfactant or hydrophobe like hexadecane, are directly polymerized. The droplets and hence resulting polymer particles are generally in the size range of 50-500 nm. There are other forms of polymerization techniques like microemulsion, melt polycondensation and solution polycondensation etc. Figure 1.1 lists the large number of polymerization techniques used to synthesize a variety of polymers [1].

1.2

Emulsion and Miniemulsion Polymerization

The common mode of particle synthesis in emulsion polymerization is achieved by micellar nucleation method, though there is also the presence of homogenous nucleation especially in water soluble monomers. The surfactants like sodium dodecyl sulphate are added which at a concentration higher than the critical micelle concentration in the aqueous phase form micelles. These micelles owing to their hydrophobic nature inside the inner space are an ideal site for the radical entry as well as propagation of polymerization. The structure of surfactant is generally amphiphilic, with one part hydrophobic and the other part hydrophilic. These

4

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Solid & gas phase polymerization Bulk polymerization

Solution polymerization

Inverse suspension polymerization . M, , E m u i s i o n \ / Dispersion . Suspension \Upolymer,zat,onj L o l mer¡zat¡ ».polymerization/1 v / v ' /

Melt . polycondensationi

Polymerization techniques

Solution vpolycondensationy Interfacial . polymerization y

Inverse emulsion polymerization,

Miniemulsion polymerization ,

fMicro-suspension polymerization

Microemulsion .polymerization,

Precipitation polymerization^

Figure 1.1. Various polymerization techniques to generate a wide spectrum of polymers. Reproduced with permission from Nova Science Publishers [1].

molecules thus arrange themselves in a way that their hydrophilic parts are in interface with water. Every surfactant has a different critical micelle concentration value and it should be considered carefully while using different kinds of surfactants. The micelles generally have a size of 10 nm and generally 100-200 surfactant molecules form a micelle [2,3]· It is generally known that the surface tension of the solution decreases with the addition of surfactant at critical micelle concentration. However, it is not only the surface tension that is affected by the surfactant, rather a host of other properties of the solution are affected at critical micelle concentration. Once the monomer is added to the system, a small amount of monomer enters the micelles and some gets dissolved in the aqueous phase owing to the partial solubility in water. However, the majority of the monomer is generally present in the form of monomer droplets. These droplets are stabilized by the adsorption of surfactant molecules on the surface. The number of micelles is much

MINIEMULSION POLYMERIZATION

5

larger than the number of droplets and the droplet size may fall in the range of tens of micrometers [2,4]· When the polymerization is initiated by the addition of the initiator and after achieving the required polymerization temperature, the radicals are generated in the aqueous phase. The generated radicals have two possibilities to propagate further: to enter either the micelles or the monomer droplets. However, the experimental studies report that it is very rare that the radicals enter the monomer droplets. This is because of very large number of micelles present in the system as well as the architecture of the micelles provides ideal conditions for the monomer polymerization. When the radicals enter the micelles and start polymerizing the monomer contained in these micelles, the polymer particles form. These growing polymer particles are then supplied by the monomer molecules from the monomer droplets by diffusion through the aqueous phase. The termination of the radicals is quite slow as at a particular time during polymerization, there is rarely more than one radical per particle. The conventional emulsion polymerization is thus divided into three intervals as shown in Figure 1.2. On addition to the aqueous phase, the monomer enters the micelles as well as forms the monomer droplets apart from dissolution in water to some extent based on the solubility of the monomer as shown in Figure 1.2a [2]. The first interval, also termed as particles formation phase, is then initiated. The radicals re generated in the aqueous phase due to the decomposition of initiator. These radicals enter the micelles and initiate monomer polymerization leading to the generation of polymer particles. The number of particles keeps on increasing in this interval which also results in the continuous enhancement in the polymerization rate. The system, as shown in Figure 1.2b, thus consists of polymer particles, monomer droplets, and the inactive micelles. The particles keep on increasing in size, thus requiring more and more surfactant to stabilize the increasing surface area. This leads to the adsorption of the dissolved surfactant in the aqueous phase on the surface of the particles and the surfactant concentration thus falls much below the critical micelle concentration. This results into the destabilization of the remaining micelles and they provide their surfactant to stabilize the growing particles. The number of particles generated from total micelles in the beginning is generally in the range of 0.1%. At the end of first interval, all of the micelles either are polymer particles or are destabilized to lose the surfactant. In the second interval, the particles keep on growing in size and no new particles are nucleated

6

MINIEMULSION POLYMERIZATION TECHNOLOGY

Figure 1.2. Representation of various intervals of emulsion polymerization [1].

thus leading to the constant rate of polymerization. As the particles grow in size during the course of polymerization, they deplete the monomer content present in them. This depletion is continuously replenished by the absorption of more monomer from the water phase, which has been dissolved in it. The water phase in return absorbs more monomer from the monomer droplets resulting in the reduction of the size of the monomer droplets as shown in Figure 1.2c. After a certain conversion of the monomer is achieved, the monomer droplets also disappear which forms the transition period between the second and third interval. As shown in Figure 1.2d, the particles in this interval keep on polymerizing the monomer enriching them. Thus, concentration of the monomer in the particles decreases, and subsequently the polymerization rate also decreases in this interval. The number of particles thus also remains the same as the second interval and after the monomer has been completely depleted, the polymerization rate climbs down to zero.

MINIEMULSION POLYMERIZATION

7

Miniemulsion polymerization exhibits several advantages over conventional emulsion polymerization also called macroemulsion polymerization [5-8]. The most prominent advantage is the elimination of the need of the monomer to diffuse through the aqueous phase from the monomer droplets to the polymer particles during the course of polymerization. It is due to the reason that the monomer droplets are directly polymerized in this mode of polymerization. The monomer droplets are generated by shearing the system with high energy along with the addition of costabilizer (with the surfactant) which is needed to be hydrophobic in order to avoid the collapse of the monomer droplets by Ostwald ripening when the shearing of the system is stopped. Thus, in this mode of polymerization, it is of importance to avoid the micellar nucleation, therefore, the amount of surfactant is below the critical micelle concentration. Thus, miniemulsion polymerization differs from the macroemulsion polymerization significantly in the mechanism of particle nucleation. In fact, this difference also acts as an advantage of miniemulsion polymerization as micellar nucleation in conventional emulsion polymerization is extremely sensitive to a large number of factors such as amount of surfactant, amount of initiator, agitation speed, temperature of the polymerization reaction, mode of addition of the monomers, etc. The number of particles in miniemulsion polymerization is thus dependant on the shearing forces as well as the amount of surfactant and costabilizer, and is independent on the initiator amount. A significant advancement has been achieved in the living polymerization methods in miniemulsion polymerization. The colloidal stability is also much better in miniemulsion polymerization as compared to the conventional emulsion polymerization, which makes it a technique of choice. Figure 1.3 demonstrates the mechanism of miniemulsion polymerization [1]. The costabilizer and the surfactant are added along with monomer in the aqueous medium. The miniemulsion is then achieved by the action of shear. The application of shear breaks the bigger monomer droplets into the droplets of size range 10-500 nm which also forms the range of polymer particles generated by miniemulsion polymerization. However, as mentioned above, the size of the monomer droplets and hence polymer particles can be tuned by the amount of surfactant and costabilizer in combination with shearing forces. The surfactant is required in the system to eliminate the droplet coalescence by the action of Brownian motion or settling, whereas the costabilizer prevents the

8

MINIEMULSION POLYMERIZATION TECHNOLOGY

Figure 1.3. Process of miniemulsion polymerization. The molecules with open and filled circles represent the surfactant and costabilizer, respectively.

Ostwald ripening [5]. When the emulsion is subjected to shear by sonicator or mechanical homogenizer, the generation of small droplets is achieved in the liquid medium. As the droplets have a distribution is the sizes, the monomer tends to diffuse from the smaller droplets into the large ones if the monomer is even slightly soluble in the continuous phase. The surface area of the monomer droplets is quite high and most of the surfactant is adsorbed on the particle surface. As no micelles exist in the system therefore the particle nucleation takes place by radical entry into the droplets. The initiators used for the miniemulsion polymerization can be both water soluble as well as monomer soluble. Figure 1.4 is another representation of the comparison between the emulsion and miniemulsion polymerization processes.

MINIEMULSION POLYMERIZATION

9

Figure 1.4. Comparison of (a) emulsion and (b) miniemulsion polymerization.

Owing to the differences in the mechanism of particle nucleation as well as propagation, the rates of polymerization in both macroemulsion and miniemulsion polymerization are also different. As described above, the conventional emulsion polymerization has initiation of polymer particles in micelles followed by the diffusion of monomer from the monomer droplets to the polymer particles though the aqueous phase. The polymerization rate, therefore, first grows till the micelles are present in the system owing to the increase in the number of the particles. Subsequently, the rate of polymerization becomes constant as the polymer particles grow only in size and not in number owing to the use of all the surfactant to stabilize the polymer particles. The monomer in this interval is diffusing continuously to the polymer particles and is, therefore, getting depleted in the monomer droplets. When the droplets cease to exist, the rate starts to fall and becomes zero when all the monomer enriching the polymer particles is also consumed. However, in the case of miniemulsion polymerization, there is no diffusion of the monomer through the aqueous phase owing to direct polymerization of monomer droplets, therefore, there is no constant polymerization rate period in this mode of polymerization. This is, however, only true if the monomer is not diffusing from the small droplets to the large ones even in small extents. The rate initially grows owing to the increasing number of polymer particles by the entry of radicals in the droplets, and then drops down as the monomer in the droplets is depleted. One limitation of miniemulsion polymerization is

10

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

(A)

(B)

Figure 1.5. SEM micrographs of polystyrene latex generated by miniemulsion polymerization.

the use of hydrophobe which can be volatile in nature, the use of which thus the limits the applications of the polymer particles generated by miniemulsion. Figure 1.5 shows the representative scanning electron microscopy (SEM) micrographs of polystyrene particles synthesized with miniemulsion polymerization. The miniemulsion polymerization of hydrophobic monomers in the aqueous phase is achieved owing to no or insignificant dissolution of monomer in the aqueous phase which leads to the droplet stability in the system. However, if the monomers are water soluble, one can use inverse miniemulsion polymerization. Here, a hydrophobic reaction solvent or dispersion medium like cyclohexane is used instead of water, and the process is exactly similar to miniemulsion polymerization. Instead of a hydrophobe as a costabilizer, one must use a lipophobe such as sodium chloride, and the stabilizer is also different.

1.3

Properties of Miniemulsion Polymerization

If the Ostwald ripening is allowed to occur continuously, the monomer from the smaller particles would diffuse into the larger particles and extensive creaming would result. Costabilizers are therefore added as they help to stop the Ostwald ripening by stopping the diffusion of the monomers from monomer droplets. Therefore, they

MINIEMULSION POLYMERIZATION

11

should be very hydrophobic in nature and should be soluble in the monomers. As the costabilizers generally have a little water solubility therefore the Ostwald ripening still occurs but the time required for the destabilization of the miniemulsion runs in the range of months which thus allows enough time to achieve polymerization using stable miniemulsions. As mentioned earlier, the miniemulsions are generated by the combination of the high shear to break the bigger monomer droplets into the sub-micron monomer droplets in the presence of costabilizer to stop the diffusion of monomer from these particles. One must be clear that the addition of a costabilizer stops the conversion of a miniemulsion into a conventional emulsion; however, the addition of a costabilizer to conventional emulsion does not automatically convert it into a miniemulsion. It is only after addition of high shearing energy that it becomes a stable miniemulsion. The mechanical shear is generated by stirring, ultra-turrax or by ultrasonication. The mechanism of ultrasonication is primarily cavitation. Sonication is an attractive method for the laboratory scale miniemulsion generation, however, it is not suitable for the large scale processes and more efficient sharing devices are needed. In most of the reported studies over the use of miniemulsion polymerization, the use of anionic surfactant is most common. Sodium dodecyl sulphate is one the most commonly used anionic surfactant used for emulsion polymerizations. Therefore, it has also been automatically used for the majority of the miniemulsion polymerization reactions. These anionic surfactants are also attractive choices owing to their compatibility with the majority of the monomers and the initiators. However, the use of anionic surfactants is not suitable in controlled living polymerization achieved by the atom transfer radical polymerization and in such cases, non-ionic surfactants are mostly used. However, some studies have also reported the use of cationic surfactants like cetyltrimethylammonium bromide and dodecyltrimethylammonium chloride [9,10] and the resulting particles were reported to be similar to the particles achieved by using anionic surfactants. Some studies have also reported the use of nonionic surfactants. Wang et al. reported the use of poly (vinyl alcohol) as stabilizer with hexadecane as co-stabilizer [11]. The authors also reported that the use of hexadecane costabilizer was important as the use of poly(vinyl alcohol) was not sufficient to control the polymerization reaction. The costabilizer is conventionally required to be monomer soluble, water insoluble and with a low molecular weight. The insolubility

12

MINIEMULSION POLYMERIZATION TECHNOLOGY

in water leads to the elimination of the diffusion of monomer into the aqueous phase from the monomer droplets. The low molecular weight, on the other hand, allows to increase the weight ratio of costabilizer molecules as compared to monomer molecules in the monomer droplets. In the early studies cetyl alcohol and hexadecane have been used as costabilizers as they fit well to the characteristics required from an ideal costabilizer. However, these costabilizers are volatile in nature and their presence in the product may not be desirable for a number of applications. To circumvent these limitations, a number of studies have reported the use of various costabilizers which are not volatile in nature and help to achieve better acceptability of the polymer particles. Reimers et al. [12] reported the use of polymer as a costabilizer. The authors reported that using the polymer which is soluble in its own monomer would also fulfill the requirements of the costabilizer. The high molecular weight of the polymer was reported to make the polymer as poor costabilizer, but still the use of polymer as costabilizer was reported to reduce the diffusion of the monomer from the monomer droplets owing to high water insolubility of the polymer. The miniemulsion in such cases were reported to be thermodynamically unstable, but kinetically stable which still allowed the miniemulsions to be stable for the time scale suitable for the polymerization. The miniemulsions generated by using polymer as costabilizer were observed to be not true miniemulsion as they did get destabilized after a period of time, however, the polymer particles generated from these systems were similar to the systems where cetyl alcohol or hexadecane were used as costabilizers. This, therefore, completely eliminates the use of volatile hexadecane or other low molecular weight costabilizers in these miniemulsion polymerizations. The low molecular weight components are not desirable in the final latex, as these can easily migrate to other materials owing to their low molecular weight, thus causing health and safety concerns. Reimers et al. [12-13] reported on the effect of the amount of polymeric hydrophobe and its molecular weight on the generated size and size distribution of the monomer droplets. It was reported that the droplets diameters could range between 19.5 nm to 141.2 nm using polymeric costabilizer. These values were reported to be similar when the miniemulsion had hexadecane as co-stabilizer. The size of the droplets was generally observed to decrease on increasing the concentration of hexadecane, such phenomenon was also observed when polymer stabilizers were used. Though polymeric

MINIEMULSION POLYMERIZATION

13

costabilizer acted only as a poor costabilizer, the generated latexes with polymeric costabilizer were observed to have much lower polydispersity of 1.006 as compared to the 1.049 for the emulsion polymerization and 1.037 for the alkane stabilized miniemulsion. As the solubility parameter of PMMA is 19.3 MPa 1/2 , which is very similar to 18.9 MPa 1/2 value for the MMA monomer, therefore, indicating, using polymers as co-stabilizers in the polymerizations of its monomers ensures its solubility with monomer and elimination of diffusional degradation of monomer droplets owing to the water insolubility. Similarly, other monomer/polymer systems were reported to be effective. The polymeric costabilizers were also reported to be stable against the presence of small amount of inhibitors, retarders or other monomer impurities, which is generally not the case for low molecular weight co-stabilizers. Other similar studies have also been reported [14]. It was also reported that the used of a comonomer can also be employed to act as costabilizer. The comonomer can also subsequently get polymerized along with the monomer during the course of polymerization. [15]. Vinyl hexanoate, p-methyl styrene, vinyl stéarate etc. were used as comonomers with MMA and the reported system had stable miniemulsions with droplets diameters between 150 and 230 nm. The advantage of the comonomer as stabilizer is that it would reinforce the polymer chains, thus removing any concerns regarding the diffusion of low molecular weight costabilizer molecules out of the particles. Dodecyl methacrylate was also reported as a comonomer as well as costabilizer in the miniemulsion polymerization of styrene [16]. Figure 1.6 demonstrates the conversion as well as diameter evolution in the miniemulsion particles as a function of time as well as the concentration of costabilizer. Low water solubility of the costabilizer and better solubility with styrene monomer helped to achieve stable miniemulsions. Use of imitator has also been similarly reported to act also as costabilizer. Schork et al. [17] reported the use of lauroyl peroxide as initiator as well as costabilizer. The molecules first act as costabilizers helping to stabilize the monomer droplets and subsequently during the course of polymerization yield radicals by the thermal decomposition. The droplet sizes were also reported to be in the similar range as observed in miniemulsion polymerization and the ratio of polymer particles to the monomer droplets was also observed to be near unity thus confirming the efficiency of lauroyl peroxide molecules to act both as initiator as well as monomer costabilizer.

14

MINIEMULSION POLYMERIZATION TECHNOLOGY

Figure 1.6. Use of comonomer as a costabilizer [15]. Monomer conversion and (b) average particle size as a function of the reaction time: [DMA]. (Δ) 5; (O) 20; (*) 40 mM. Reproduced from reference 16 with permission from Elsevier.

In this process, similar to the use of comonomer as costabilizer, the initiator moieties are chemically incorporated in the polymer chains, thus leaving no low molecular weight residue in the polymer particles.

MINIEMULSION POLYMERIZATION

15

The use of chain transfer agent as costabilizer was reported to open opportunities for molecular weight control [5]. The chain transfer agent is generally difficult to transport to the polymer particles from the monomer droplets in the conventional emulsion polymerization, however, when the chain transfer agent is present at the site of polymerization as is possible in the case of miniemulsion polymerization, various possibilities of control of polymer reaction can be achieved. Dodecylmercaptan was reported to be used as a chain transfer agent as well as costabilizer for the miniemulsion polymerization of methyl methacrylate [18]. Sodium lauryl sulfate was used as surfactant and potassium persulphate as initiator. Stable monomer droplets were observed to form and the nucleation of the polymer particles proceeded by droplet polymerization owing to the presence of surfactant below the critical micelle concentration. When the surfactant concentration was held constant, the size of the droplets was observed to decrease on increasing the concentration of dodecylmercaptan, which is similar to the behavior as observed in the miniemulsion polymerization using hexadecane as costabilizer. The rate of polymerization was observed to be dependant upon factors such as the concentration of surfactant, co-surfactant as well as initiator. The authors reported that at low concentrations of initiator and surfactant (below CMC), the ratio of number of particles to number of droplets was 0.80 indicating that all the droplets were not nucleated. On the other hand, at higher concentrations of surfactant and initiator, the ratio was 13.56 indicting that nucleation of the polymer particles occurred not only by droplet nucleation but also by micellar nucleation. The surfactant concentration below CMC and initiator concentration at intermediate levels was observed to result a ratio near to unity indicating that the droplet nucleation resulted in the formation of polymer particles. It was observed by the authors that the value of chain transfer coefficient for the system of dodecylmercaptan as chain transfer agent and methyl methacrylate as monomer was in the range 0.6-0.8, which indicated that the chain transfer agent reacted only slightly more than the monomer thus ensuring its presence and effect throughout the course of polymerization. On the other hand, in the system with dodecylmercaptan as chain transfer agent and styrene as monomer, the chain transfer coefficient lies in the range of 15-20, indicating that it would be consumed very early in the polymerization reaction. Thus, its role as costabilizer would end very shortly in the polymerization

16

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

reaction leading to the Ostwald ripening of the smaller droplets into the larger ones subsequently resulting in the collapse of miniemulsion. Figure 1.7 lists various costabilizers reported in literature for miniemulsion polymerization [1]. Majority of the miniemulsion polymerizations have used the water soluble initiators. However, some studies also reported the use of oil soluble initiators for the polymerization of monomer droplets. Alduncin et al. [19,20] reported the use of lauroyl peroxide (LPO), benzoyl peroxide (BPO) and azobis(isobutyronitrile) (AIBN) initiators for the miniemulsion polymerization of styrene. Two different types of polymerization reactions were carried out. In the first case, o

o

CH 3 — (CH2)10 — C — O — O — C —(CH 2 ) 1 0 —CH 3 CH3 CH3 —(CH 2 ) 1 5 —OH

I

CH2=CH

Cl6H34

O

II

CH2 = C H — O ^ C — ( C H 2 ) 4 ^ C H 3

H 3 C—sr / o \

^s¡—CH 3 \ o /

H3C—Si^0/Si—CH3

-CH 2 -

OHo

Orlo

i ;

i

COOCHH3 COOCHH3

UH-i

i COOCHH3

— CH 2 — CH-[-CH 2 — CH j - C H 2 — CH —

) 6" ό

Figure 1.7. Various co-stabilizers used in the miniemulsion polymerization process. Reproduced with permission from Nova Science Publishers.

MINIEMULSION POLYMERIZATION

17

the initiator also acted as costabilizer so that no additional costabilizer was added to the polymerization medium. The second case was more conventional miniemulsion polymerization where hexadecane was used. Figure 1.8 demonstrates the time evolution of the droplet diameters. It was observed that the miniemulsions containing LPO had droplet sizes similar to those of the classical miniemulsion polymerizations and the size of these miniemulsions remained roughly constant for more than two hours after the sonication. The average size of the miniemulsions containing BPO and AIBN was, on the other hand, very large suggesting that these miniemulsions suffered a quick partial degradation after sonication. In the presence of hexadecane, however, the size distributions were almost similar irrespective of any initiator used. BPO and AIBN, on their own, are not water insoluble enough to avoid the Ostwald ripening of small particles. Figure 1.9 also represents the mechanism by which the three initiator systems work in the absence and presence of hexadecane. Figure 1.10 also represents the time evolution of conversion of the monomer when oil soluble initiator lauroyl peroxide was used [20]. As can be seen, the polymerization rate was not affected by the presence of hexadecane also confirmed by the particle size analysis. The conventional emulsion polymerization was also carried out for

Figure 1.8. Time evolution of the droplet diameter for the different miniemulsions. (O) LPO, (·) LPO+HD, (Δ) BPO, (A) BPO+HD, (D) AIBN, (■) AIBN+HD. Reproduced from reference 19 with permission from American Chemical Society.

18

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Figure 1.9. Schematic of various processes occurring during the miniemulsion polymerization with different oil soluble initiators. Reproduced from reference 19 with permission from American Chemical Society.

Figure 1.10. Time evolution of the conversion in the polymerizations initiated by LPO: (O) LPO, (·) LPO+HD and (Δ) conventional emulsion polymerization. Reproduced from reference 20 with permission from Elsevier.

MINIEMULSION POLYMERIZATION

19

comparison and it was reported that the polymerization could not reach full conversion owing to the extensive coagulation.

1.4

Controlled Miniemulsion Polymerization

Miniemulsion polymerization in the recent years has also been performed in the controlled living conditions in order to synthesize particles with specific morphologies or in order to control the molecular weight and its distribution [21-30]. Various living polymerization methods like nitroxide mediated polymerization, atom transfer radical polymerization as well as radical addition fragmentation chain transfer have been applied successfully to polymerization processes in miniemulsion. The schematic of nitroxide mediated polymerization is shown in Scheme 1.1 (Relation 1). The technique is based on the reversible termination of the radicals where the radicals are made inactive by reaction with other radicals but only temporarily. The species end capped by the nitroxide, as shown in Scheme 1.1, are termed as dormant species and they exhibit reversible dormant and active

—Pn-X

,——

-~Pn°

+ X°

(1)

Scheme 1.1. Representation of controlled polymerization. Reproduced from reference 31 with permission from American Chemical Society.

20

MINIEMULSION POLYMERIZATION TECHNOLOGY

behavior throughout the course of polymerization. Termination reactions do take place, but the extent of termination is greatly reduced. The emulsion polymerization required the diffusion of the nitroxodes from the aqueous phase to the polymer particles, which is not straight forward. However, in the case of miniemulsion polymerization, the diffusion of the control radicals is not required as these can be directly obtained in the monomer droplets, i.e. at the site of polymerization thus simplifying the process. Different nitroxodes have been reported in literature for different monomer systems like TEMPO and SGI as shown in equations 1 and 2 respectively. Apart from that, alkoxyamines and nitroxide terminated oligomers are also used as unimolecular systems, where the molecule acts both as nitroxide as well as initiator. Atom transfer radical polymerization (ATRP) is also based on reversible termination approach to achieve living polymerization.

M I N I E M U L S I O N POLYMERIZATION

21

The process involves an organic halide which reversibly terminates the polymer chains and generates reversibly active chains and the redox process is catalyzed by a transition metal compound such as cuprous chloride or bromide complexed with a ligand. Scheme 1.1 (Relation 2) shows the schematic of ATRP process. The limitation of ATRP process is the presence of transition metal compounds in the end product. Another limitation of this method is the possible interaction of copper compounds with the emulsifiers used in emulsion polymerization. The polymerization in emulsion phase can though work when no surfactant is used in the system or non ionic surfactants are used. Reversible addition fragmentation chain transfer (RAFT) is another form of controlled polymerization (Scheme 1.1, Relation 3)

Scheme 1.2. Systematic approached to synthesize hybrid polymers. Reproduced from reference 31 with permission from American Chemical Society.

22

MINIEMULSION POLYMERIZATION TECHNOLOGY

which operates by reversible transfer mode thus not similar to the reversible termination performed in ATRP or nitroxide mediated polymerization. The core of this process is a RAFT agent which contains dithioester groups. The living polymerization takes place as the transferred end group in the polymeric dithioester is as labile as the dithioester group in the starting RAFT agent. The disadvantage of the techniques operating by reversible termination is the partitioning of the deactivating species in the aqueous phase as well as organic phase. It complicates the concentration of active and dormant species in the polymer particles. However, the techniques based on reversible transfer do not suffer from these disadvantages, as the number of free propagating radicals in these polymerization methodologies practically remains unchanged. The initiator for the polymerization can be the conventional initiators like AIBN or benzoyl peroxide. Important advantage of this technique is the possibility of polymerization reaction to be carried out at lower temperatures. However it also suffers from the presence of excess or remainder RAFT agent which owing to the presence of sulphur also leads to color and odors to the product. Synthesis of hybrid polymers can also be achieved by using these methods as depicted in Scheme 1.2 [31].

References 1. V. Mittal, Advances in Polymer Latex Technology, New York, Nova Science Publishers, 2009. 2. G. Odian, Principles of Polymerization, Fourth Edition, New Jersey, John Wiley & Sons, Inc., 2004. 3. P.C. Hiemenz and R. Rajagopalan, Principles of Colloid and Surface Chemistry, New York, Marcel Dekker, Inc., 1997. 4. V.R. Gowariker, N. V. Viswanathan and J. Sreedhar Polymer Science, New Delhi, John Wiley & Sons, Wiley Eastern Limited, 1986. 5. F.J. Schork, Y. Luo, W Smulders, J.P. Russum, A. Butté, and K. Fontenot, Advances in Polymer Science, Vol. 175, p. 129,2005. 6. K. Landfester, Macromolecular Rapid Communications, Vol. 22, p. 896,2001. 7. P.A. Lovell and M.S. El-Aasser Emulsion Polymerization and Emulsion Polymers, Eds., England, John Wiley and Sons Limited, 1997. 8. K. Matyjaszewski and T.P. Davis, Eds., Handbook of Radical Polymerization, New Jersey, John Wiley & Sons, Inc., 2002. 9. K. Landfester, N. Bechtold, F. Tiarks, and M. Antonietti, Macromolecules, Vol. 32, p. 2679,1999. 10. M. Bradley, and F. Grieser, Journal of Colloid and Interface Science, Vol. 251, p. 1,2002.

MINIEMULSION POLYMERIZATION

23

11. S. Wang, and FJ. Schork, Journal of Applied Polymer Science, Vol. 54, p. 2157, 1994. 12. J.L. Reimers, and F.J. Schork, Journal of Applied Polymer Science, Vol. 60, p. 251, 1996. 13. J. Reimers, and F.J. Schork, Journal of Applied Polymer Science, Vol. 59, p. 1833, 1996. 14. H. Dong, J.W. Gooch, and F.J. Schork, Journal of Applied Polymer Science, Vol. 76, p. 105, 2000. 15. J.L. Reimers, and F,J. Schork, Polymer Reaction Engineering, Vol. 4, p. 135,1996. 16. C.S. Chern, and Y.C. Liou, Polymer, Vol. 40, p. 3763,1999. 17. J.L. Reimers, and F.J. Schork, Industrial & Engineering Chemistry Research, Vol. 36, p. 1085,1997. 18. D. Mouran, J. Reimers, and F. J. Schork (1996). Journal of Polymer Science, Part A: Polymer Chemistry, 34,1073-1081. 19. J. Asua, J. Alduncin, and J. Forcada, Macromolecules, Vol. 27, p. 2256,1994. 20. J.A. Alduncin, and J.A. Asua, Polymer, Vol. 35, p. 3758,1994. 21. G. Pan, E.D. Sudol, V.L. Dimonie, and M.S. El-Aasser, Macromolecules, Vol. 35, p. 6915,2002. 22. M.N. Alam, P.B. Zetterlund, and M. Okubo, Macromolecular Chemistry and Physics, Vol. 207, p. 1732,2006. 23. J.W. Ma, J.A. Smith, K.B. McAuley, M.F.Cunningham, B.Keoshkerian, and M.K. Georges, Chemical Engineering Science, Vol. 58, p. 1163, 2003. 24. J. Nicolas, B. Charleux, O. Guerret, and S. Magnet, Macromolecules, Vol. 37, p. 4453,2004. 25. M. Li, and K. Matyajaszewski, Macromolecules, Vol. 36, p. 6028,2003. 26. M. Li, K. Min, and K. Matyajaszewski, Macromolecules, Vol. 37, p. 2106,2004. 27. K. Min, H. Gao, and K. Matyajaszewski, Journal of American Chemical Society, Vol. 127, p. 3825,2005. 28. J.K. Oh, C. Tang, H. Gao, N.V. Tsarevsky, and K. Matyajaszewski, Journal of American Chemical Society, Vol. 128, p. 5578,2006. 29. G. Qi, C.W. Jones, and F.J. Schork, Macromolecular Rapid Communications, Vol. 28, p. 1010,2007. 30. M.F. Cunningham, Progress in Polymer Science, Vol. 33, p. 365, 2008. 31. J. Pyun, and K. Matyajaszewski, Chemistry of Materials, Vol. 13, p. 3436,2001.

This Page Intentionally Left Blank

2

Multi-Functional Stabilizers in Miniemulsion Polymerization Alain Durand Alain Durand Laboratoire de Chimie Physique Macromoléculaire, UMR 7568 CNRS - Nancy-University ENSIC, 1 rue Grandville, BP 20451, F-54001 Nancy cedex, France

Abstract Multi-functional surfactants in miniemulsion polymerization are defined as stabilizers which can be involved in other aspects of polymerization. For reactive surfactants, inisurfs, surfmers and transurfs have been reported in miniemulsion polymerization. Their applications in the preparation of latexes with well-controlled properties (chemical structure, particle morphology and colloidal stability) are overviewed. In addition, it is showed that some polymeric stabilizers can also suppress the need for ultrahydrophobes (acting against Ostwald ripening). Key words: Miniemulsion, polymerization, surfactants, inisurf, surfmers, transurfs

2.1

Introduction

In this chapter, we will define a stabilizer as a (macro)molecular species able to accumulate at liquid/liquid interfaces and which contributes to increase the kinetic stability of dispersions (like monomer emulsions or suspensions of polymeric particles) as well as to facilitate their formation. Stabilizers involved in miniemulsion polymerization processes have been the object of many studies because they are at the very heart of the process: they are involved in the stabilization of the droplets/particles, they are in direct contact with the reactants (monomers, initiators), they are present in

V. Mittal (ed.) Miniemulsion Polymerization Technology, (25-42) © Scrivener Publishing LLC

25

26

MINIEMULSION POLYMERIZATION TECHNOLOGY

rather high amounts when comparing to other polymerization processes in disperse media like emulsion or suspension and they are often retained in the final product. For all these reasons, stabilizers became rapidly an important topic in the literature on miniemulsion polymerization [1-4]. One major interest of miniemulsion process is the ability to produce a final dispersion of polymeric particles whose size distribution is an (almost) exact copy of that of the initial monomer emulsion. Another specificity of miniemulsion polymerization is the predominance of direct nucleation in previously stabilized droplets. This fact may promote some chemical processes taking place at interface (as we will see in the case of reactive surfactants). To approach that situation, several challenges need to be overcome in which the stabilizer is involved. First, each monomer droplet must be nucleated so that the whole population of the initial miniemulsion is converted into polymeric particles. Second, any homogeneous nucleation event should be avoided so that no supplementary particle could be produced. Third, droplet/particle identity should be preserved over the time needed for polymerization. A great deal of work has been carried out on how the structure of stabilizers should be selected according to the nature of continuous and disperse phase, the volume fraction of disperse phase and the conditions of polymerization, so as to conform the preceding requirements. A wide variety of stabilizers has been described for miniemulsion polymerization in scientific literature. A complete review of that topic would be above the scope of this chapter and the interested reader is referred to relevant monographs [3]. Briefly we could distinguish three different types of miniemulsion polymerization systems: oil-in-water (i.e. direct), water-in-oil (i.e. inverse) and non-aqueous. Molecular surfactants are particularly efficient for direct systems (especially ionic surfactants). Macromolecular surfactants are required for non-aqueous systems since steric repulsion is the main contribution to the colloidal stability of non-aqueous miniemulsions [5]. In the past 10 years, a certain number of publications about miniemulsion polymerization showed that the use of multi-functional stabilizers could be a promising approach. The general idea was to design molecular or macromolecular surfactants that could combine other functions to that of stabilizing the dispersed reaction medium. There were many interests in that strategy. First by the use of multi-functional surfactants it was possible to limit the number of components of a miniemulsion polymerization medium

MULTI-FUNCTIONAL STABILIZERS

27

and particularly to avoid the use of some labile compounds that could diffuse out of the final material. This latter process could be a problem by itself but could also modify end-use properties, particularly for coatings. In addition, a careful design of the stabilizer allows reducing the number of steps required for preparing surfacefunctionalized particles for targeted applications like drug delivery. The functional properties related to the surface of the particles are thus brought by the stabilizer used in the preparation of the particles. Finally, using reactive surfactants for confining polymerization processes at interface was a reported strategy for the control of particle morphology. Polymer chemistry was useful to design well-controlled macromolecular surfactants with reactive end-groups or to prepare polymeric stabilizers from more specific polymers. Controlled radical polymerization was particularly employed for the synthesis of specifically designed block copolymer surfactants. In most of cases, reactive groups were located at the end of the hydrophobic sequences. Chemical modification of structurally complex macromolecules, like polysaccharides, was another approach. Macromolecular surfactants were obtained following two strategies: either by the attachment of hydrocarbon groups on native polysaccharides or by growing graft chains from polysaccharide backbone using a controlled polymerization (grafting from strategy) [6-8]. In what follows, we will focus on the use of multi-functional stabilizers in miniemulsion polymerization, in other words of miniemulsion stabilizers which have another function (called the "supplementary function"). For the sake of clarity, we will distinguish the supplementary functions related to the stability of the initial monomer emulsion (mostly encountered with non-reactive surfactants) and those related to the polymerization reaction itself (essentially reactive surfactants).

2.2 Stability of Initial Monomer Droplets In any miniemulsion process, the initial emulsion of monomer may degrade either by coalescence or by Ostwald ripening (molecular diffusion) [9]. The generally followed strategy for reaching a shelf life consistent with the time needed for polymerization involves the combined uses of a surfactant acting against coalescence and a co-stabilizer preventing Ostwald ripening [3]. While the surfactant is expected to

28

MINIEMULSION POLYMERIZATION TECHNOLOGY

adsorb at oil/water interface for preventing coalescence events, the co-stabilizer is dissolved in the disperse phase and prevents molecular diffusion by an osmotic effect which has no relation with any interfacial activity. This is the reason why the word "cosurfactant" although widely employed is often misleading. The simplest co-stabilizer is a molecule which is readily soluble in the disperse phase while having an ultra low solubility in the continuous phase. For direct miniemulsions, a great deal of work has been carried out about the use of functional hydrophobes that could be involved either in the polymerization itself (monomer, chain transfer agent, initiator) or in the final properties of the latex (elastomers) [10-16]. Nevertheless, a limited number of papers examined the possibility of combining the two stabilizing functions (against coalescence and Ostwald ripening) into a single constituent of the miniemulsion recipe. This was achieved with some polymeric surfactants which were reported to stabilize direct miniemulsions without the need of adding any hydrophobe (like hexadecane) in the monomer phase (Table 2.1). To the best of our knowledge, no molecular surfactant has been reported to exhibit that property. The corresponding polymeric surfactants were anionic or cationic polyelectrolytes with long hydrophobic sequences or repeat units carrying long hydrocarbon tails (acrylic esters or quaternized amines). No direct comparison of the average diameter of the initial monomer emulsion and of the final latex is available. Nevertheless, with the cationic polymeric surfactants, no significant difference in the average particle diameter could be evidenced with or without the presence of hexadecane dissolved in the monomer [17]. With anionic polymers, latex particles with average diameters below 250 nm were obtained without adding any molecular hydrophobe in Table 2.1. Macromolecular surfactants suppressing the need for hydrophobe in miniemulsion polymerizations.

a

Stabilizer3

Monomer

(S)x-(VBC)y quaternized by C12 or C16 fatty amine

S

[17]

P(ODMA-co-MAA) andP(ODMA-co-AA)

S

[18]

C4H9S-CS-S-(BA)x-(AA)-C(CH3)-COOH C4H9S-CS-S-(S)x-(AA)y-à(CH3)-COOH

S,BA

[25]

For abbreviations, see footnote of Table 2.4.

Reference

MULTI-FUNCTIONAL STABILIZERS

29

the monomer [18]. Two alternative explanations can be proposed depending on the partition of the polymeric surfactant between the continuous and disperse phases. Oil-soluble polymeric surfactants containing octadecyl ester units were dissolved in styrene. Since these polymers were reported to act both as surfactants and as costabilizers, it can be assumed that the polymeric stabilizer dissolved in the droplets had an osmotic effect which slowed down Ostwald ripening. In addition, the presence of sodium carboxylate units led to a significant adsorption at the styrene/water interface preventing droplet coalescence by electrostatic repulsions. It must be noted that rather high amounts of polymeric surfactant were used (1.7-10 wt% relative to monomer) [18]. Similarly, carboxylated polyurethanes (PU) soluble in styrene were used as co-stabilizers in combination with SDS which was added in the aqueous phase as a surfactant (contrary to other studies in which the amphiphilic polymer was the sole surfactant) [19, 20]. The authors noticed that decreasing the molar mass of carboxylated PU was in favour of the shelf life of the monomer emulsion. Increasing the pH of the aqueous phase provided the same beneficial effect. These results were interpreted by considering both the adsorption of polymers at interface and the effect on Ostwald ripening. On the other hand, water-soluble comblike copolymers were also reported to suppress the need for a hydrophobic co-stabilizer in styrene miniemulsion polymerization. The authors showed that the kinetics of styrene conversion was unchanged by the presence or the absence of hexadecane dissolved in the monomer [17]. This strongly supports the idea that droplet ageing is very slow in both situations since droplet size has a significant effect on the overall rate of polymerization. Such results cannot be explained by an osmotic effect of the macromolecules. These copolymers carried long hydrocarbon chains (C12 or C16) ensuring a strong adsorption onto droplet surface. Their effect upon the rate of Ostwald ripening can be explained by a modification of interfacial elasticity. It was theoretically suggested by several authors that strongly adsorbed polymeric surfactants could slow down Ostwald ripening [8,21-23]. The almost irreversible adsorption of some sequences of polymeric surfactants could lead to an increase of the elastic modulus of surface layer as well as to a decrease of interfacial tension because of the reduction of droplet area upon shrinking. These combined effects could compensate the difference of chemical potential of the oil and consequently stop aging (or at least slow it down sufficiently for droplet nucleation to occur. This phenomenon was quantitatively treated and first

30

MINIEMULSION POLYMERIZATION TECHNOLOGY

experimental evidences were reported with oil-in-water emulsions stabilized with polymeric surfactants [8, 24]. Nevertheless, theoretical considerations stressed the need for minimal thickness of the covering layers in order to have a significant impact on the rate of Ostwald ripening. To the best of our knowledge, this aspect has not been experimentally investigated up to now. These considerations support the idea that a convenient design of polymeric stabilizers could suppress the need for hydrophobic co-stabilizers, taking the opportunity of elastic behaviour of the layer of polymeric surfactant at the surface of monomer droplets, which cannot be obtained with molecular surfactants. A direct comparison between cationic comblike copolymers and a cationic surfactant confirmed the strong difference in shelf-life of miniemulsions and the longer stability observed with the polymeric stabilizer [17]. Anionic diblock copolymers were also shown to act as co-stabilizers [25]. This result can be related to their strong adsorption at interfaces. Indeed these block copolymers contained long hydrophobic blocks and short anionic sequences (the ratio of the number of hydrophobic units to that of hydrophilic units is 4). This composition is very different from that of other gradient copolymers with the same repeat units but much more hydrophilic units (70 mole %) for which the presence of hydrophobe was essential for maintaining the colloidal stability during the reaction. At present, it is not possible to extract guidelines for a rational design of stabilizers that would also act as hydrophobes in miniemulsion polymerization. Nevertheless, several examples are available in the case of polymeric surfactants and give interesting tracks for further investigations.

2.3

Stabilizers and Polymerization Processes

2.3.1 Mass-Transfer Processes In miniemulsion polymerization, when chain polymerization is carried out, the initiator may be dissolved either in the disperse phase or in the continuous one. The high interfacial area allows an efficient droplet nucleation even if the initiator is initially dissolved in the continuous phase. The stabilizer forms a molecular (at least) layer covering the droplets and ensuring colloidal stability. When the initiator is dissolved in the continuous phase, one could wonder whether the stabilizing layer may interfere with the nucleation process and

MULTI-FUNCTIONAL STABILIZERS

31

particularly with the transfer processes. In case the answer is positive, this would imply that the stabilizer could have a direct influence on the reaction kinetics. Alkali-soluble resins were shown to limit the rate of BA/MMA miniemulsion polymerization when KPS was used as the initiator. This effect was attributed to the electrosteric barrier formed by the ASR. Such limitation was not observed when using a low-molar mass surfactant [26]. An example of steric repulsions was reported in the case of the ring-opening metathesis polymerization of norbornene carried out in miniemulsion [27, 28]. PEO-based ruthenium carbenes were designed and applied in the miniemulsion metathesis polymerization of norbornene. The simultaneous use of bulky initiator and neutral polymeric surfactant PS-b-EO at the surface of the droplets limited the diffusion of the initiator into monomers droplets by the presence of a steric barrier. This limitation had two consequences: the nucleation period increased with the molar mass of the initiator while the efficiency of initiation decreased (the experimental molar masses were higher than the expected ones).

2.3.2

Reactive Stabilizers

The stabilizer can be involved directly in some reactive steps of the polymerization reaction. One important consequence is that the stabilizers are covalently linked to the latex particles once miniemulsion polymerization has proceeded. Alternatively, the use of reactive stabilizer may be a way to confine reaction processes at the interface so as to obtain nanocapsules instead of bulk nanoparticles. Reactive stabilizers must comprise some functional groups in one or several parts of their chemical structure. To the best of our knowledge, in the large majority of reported examples, the reaction was carried out in direct miniemulsion polymerization. Consequently the functional groups were often located in the hydrophobic sequences of the molecular or macromolecular surfactants. The only exceptions were for molecular surfactants acting as initiators for ionic polymerizations through strong acid or basic functions. The chemical processes in which multi-functional stabilizers were involved belong to three categories: initiation, propagation and transfer (Table 2.2 to 2.4). The corresponding terms inisurf, surfmer and transurf, respectively, which were widely employed for emulsion polymerization will be used here for miniemulsion polymerization [29-31]. In a lot of examples, multi-functional stabilizers

S,DVB

TD

P(S-co-MA)-TEMPO

'For abbreviations, see footnote of Table 2.4.

BMA, EGDMA, DSDMA

HD

PE0114-PBMA49-C1

CRP

CRP

CRP

BA

HD

PEO120-PS15-Br, PEO120-PS29-Br, PEO^-PS^-Br, PEO^-Br

CRP

[43]

[42]

[41]

[39]

[40]

[32]

[37]

CRP RP

[35],[36]

CP

[38]

[33],[34]

AP

CRP

Reference

Polymerizationb

CRP

4-VP, DVB

S

MMA

S

p-MS

PGE, D4

Monomer

S, BA, BMA

-

PEO m -Br, PNIPAM89-Br HD

HD

[CH 2 =CH-Ph-CH 2 -0-(EO)-CO-CH 2 -CH 2 C(CH3)2-N=]2

CH3-0-(EO)nl-CO-C(CH3)2-(S)33-Br

HD

Oligostyrene

Br-C(CH 3 ) 2 -COO-(CH 2 ) ir N (CH 3 ) 3 Br

+

-

CH 3 -(CH 2 ) n -Ph-S0 3 H

CH 3 -(CH 2 ) n -Ph-S0 3 Na

-

Co-stabilizer

CH 3 -(CH 2 ) 11 (Ph-CH 2 )(CHy 2 NOH(CH3(CH2)11)2(CH3)2N+OH-

Inisurfa

Table 2.2. Molecular and macromolecular inisurfs used in miniemulsion polymerizations.

MULTI-FUNCTIONAL STABILIZERS

33

Table 2.3. Molecular and macromolecular surfmers used in miniemulsion chain polymerizations and step polymerization. Surfmera

Monomer

(CH 2 =CH-CH 2 -0-CH 2 -) 3 C-CH 2 -0(PO)x-(EO)v-R R = H or phosphate

S,BA

[49]

CH 2 =CH-CO-0-CH 2 -CH(CH 3 )(EO)x-(PO) -(EO)x-CH(CH3)-CH20-CO-CH=CH 2

S

[55]

CH3-(CH2)15-N+(CH3)2-(CH2)2-OOCC(CH3)=CH2, B r

S

[45]

CH2===CH-Ph-CH2-0-(DEA)x-(DMA)y partly quaternized

S

[46]

s

[56]

[CH 2 =CH-Ph-CH 2 -0-(EO) -CO-CH 2 CH 2 -C(CH 3 ) 2 -N=] 2

s

[32]

CF 3 -(CF 2 ) 7 -0-CO-CH=CH-COOH

S, BMA

[50]

S,MMA Na -0 3 SO-(CH 2 ) n -OOC-C(CH 3 )=CH 2 (CH3)3N+-(CH2)n-OOC-C(CH3)=CH2,Br

[44]

((DMA) 39 -0-CH 2 ) 2 -C(C 2 H.)-CH 2 -0CH 2 -CH=CH 2

+

a

Reference

Na + -OOC-(CH 2 ) 9 -OOC-CH=CH 2

AC8CBb, DACTPll b

[57]

CH 3 -(CH 2 ) 8 -Ph-CH=CH-CH 3 O-(EO)20-SO3NH4 CH 3 -(CH 2 ) n -CH-CH 2 -0-CH 2 -CH= CH 2 (n = 8,10) O-(EO)10-SO3NH4

S

[47]

Bu x -succinimide-(NH-CH 2 -CH 2 ) 4 -NH 2

HMDA, TDI

[48]

For abbreviations, see footnote of Table 2.4. AC8CB: polymerizable liquid crystal mesogen, DACTP11: diacrylate crosslinker.

b

were polymeric surfactants and more particularly amphiphilic block copolymers. The same trend has been evidenced for reactive surfactants used in emulsion polymerization. The use of controlled polymerizations allowed preparing amphiphilic copolymers with functional groups located either at the end of the chains or between different blocks (Figure 2.1 to 2.3). A few examples can be found where one stabilizer was involved in two auxiliary functions apart from that of providing colloidal stability to the emulsion. In one work, the transurf also replaced the hydrophobe to slow down

34

MINIEMULSION POLYMERIZATION TECHNOLOGY

Table 2.4. Molecular a n d macromolecular transurfs used in miniemulsion RAFT polymerizations Transurf 3

Monomer

Type of particles 0

Reference

S,MMA

BN

[44]

S,BA

BN

[25]

Ph-CS-S-C(CH3)(CN)-(CH2)2CO-NH-(CH 2 ) 2 -(EO) x -0-CH 3

S

BN

[58]

Ph-CH2-CS-S-(S)x-(MA)v-Ph

S

NC d

[59]

NC

[60]

Ph-CS-S-C(CH 3 )(CN)-(CH 2 ) 2 -COOH C4H9S-CS-S-(BA)x-(AA)yC(CH 3 )-COOH C4H9S-CS-S-(S)x-(AA) C(CH 3 )-COOH

C12H2.S-CS-S-(S)2-(AA)2CH(CH 3 )-COOH

s

HOOC-(CH 2 ) 2 -C(CH 3 )(CN)(DMA) -S-CS-CH.

MMA

BN

[61]

Ph-CH2-CS-S-(S)x-(MA)y-Ph

S

NC

[62]

NIPAM

NC

[63]

S,ND

NC

[64]

x

6

0

CH3-(EO)17-OOC-C(CH3)2-SSCS-C12H25 HOOC-(CH 2 ) 2 -C(CH 3 )(CN)(MAA)y-(S)y-SSC-S-CI2H25

"Abbreviations: AA, Acrylic acid; BA, Butyl acrylate ;BMA, Butyl methacrylate; Bu, Butylène; D4, Octamethylcyclotetrasiloxane; DEA, 2-(diethylamino)ethyl methacrylate; DMA, 2-(dimethylamino)ethyl methacrylate; DSDMA, bis(2-methacryloyloxyethyl) disulfide; DVB, Divinylbenzene; EGDMA, Ethylene glycol dimethacrylate; EO, Ethylene oxide; HD, Hexadecane; HMDA, 1,6-Diaminohexane; MA, Methyl acrylate; MAA, Methacrylic acid; MMA, Methyl methacrylate; ND, n-nonadecane; NIPAM, N-isopropylacrylamide; ODMA, Octadecyl methacrylate; PGE, Phenylglycidylether; p-MS, para-Methoxystyrene; PO, Propylene oxide; S, Styrene (with an index, otherwise sulfur atom); TD, tetradecane; TDI, Toluene 2,4-diisocyanate; TEMPO, 2,2,6,6-Tetramethyl-l-piperidinyloxyl; VBC, 4-Vinylbenzyl chloride; 4-VP, 4-Vinylpyridine. b Abbreviations: AP, Anionic polymerization; CP, Cationic polymerization; CRP, Controlled radical polymerization; RP, classical radical polymerization. C BN, bulk nanoparticles; NC, nanocapsules. d The final latex contained both nanocapsules and pure PS nanoparticles.

Ostwald ripening process [25]. This is, as far as we are aware, the only example of a reactive stabilizer acting as a hydrophobe. In another study, the macromolecular stabilizer was both inisurf and surfmer because it comprised both carbon double bonds at both

MULTI-FUNCTIONAL STABILIZERS

35

Figure 2.1. Schematic chemical structures of inisurfs used in miniemulsion. X is an halogen (bold line: hydrophilic block, circle: ionizable group).

Figure 2.2. Schematic chemical structures of surfmers used in miniemulsion. R : H or CH3 (bold line: hydrophilic block, circle: ionizable group).

Figure 2.3. Schematic chemical structures of transurfs used in miniemulsion. T is a functional group acting in chain transfer (dithioester...). (bold line: hydrophilic block, circle: ionizable group).

ends and azo bonds in the middle of the chain [32]. In all other studies, the stabilizer was involved in one supplementary function through one specific functional group. Inisurfs have been used in miniemulsion for various chain polymerization mechanisms: anionic, cationic as well as free-radical (Table 2.2). For ionic polymerizations, inisurfs were molecular surfactants in which the polar head was a strong acid or base [33-36]. Thus the

36

MINIEMULSION POLYMERIZATION TECHNOLOGY

counter ions (proton or hydroxide) were involved in the initiation of polymerization. During the anionic polymerization of phenyl glycidyl ether in miniemulsion, hydroxide ions (counter ions of the cationic surfactants) initiated the polymerization at the interface [34]. Then the anionic active centers formed complexes with ammonium groups with enough stability to allow a few propagation events to occur before termination with water molecules. The authors used surfactants mixtures combining cationic surfactant (with hydroxide counter ion) with nonionic ones. Increasing the nonionic surfactant content decreased the polymerization rate which was attributed to the reduction of the surface concentration of hydroxide ions. No hydrophobe was added into the monomer which, as noted by the authors, did not lead to emulsions with a correct stability. Consequently, the polymerization was carried out in conditions that may not be those of completely rigorous miniemulsion polymerization. Cationic polymerization of p-methoxystyrene was initiated by the use of dodecylbenzenesulfonic acid which acted as an inisurf [35,36]. The authors demonstrated that the addition of a hydrophobe could be avoided. Nevertheless, they observed that, in those conditions, the particle size was continuously decreasing while polymerization progressed. On the contrary, in the presence of hexadecane dissolved in monomer droplets, particle size was significantly lower and remained unchanged up to total monomer consumption. The proposed interpretation was that cationic polymerization started during the sonication step thus rapidly producing macromolecules in the droplets. These hydrophobic macromolecules were supposed to act as hydrophobe and prevent Ostwald ripening. As for anionic polymerization of PGE, the interfacial concentration of inisurf was shown to influence the rate of polymerization. Indeed this concentration directly impacts the amounts of protons at interface, which is the real initiator. In addition, a maximum number-average molar mass (around 1000 g / mol) was reached which is a common feature of ionic polymerizations carried out in miniemulsion. This fact is attributed to the low surface activity of oligomers having higher degrees of polymerization. Since these oligomers cannot maintain at interface, they are rapidly buried in the particles and a termination process occurs. Inisurfs have been employed for free radical miniemulsion polymerization mainly in the case of controlled radical polymerization (CRP). To the best of our knowledge, there is one example of study involving an inisurf for a miniemulsion polymerization following a classical mechanism [32]. The inisurf was used as the sole stabilizer and initiator for the preparation of polystyrene nanoparticles.

MULTI-FUNCTIONAL STABILIZERS

37

In addition, the inisurf carried double bonds at both ends and, thus, acted also as a monomer. The length of the poly(ethylene oxide) sequences was shown to have a great influence on the colloidal stability of the particles during polymerization. The nanoparticles were expected to be covered by loops of PEO chains covalently attached at both ends, which is obviously consistent with the previous results concerning the influence of inisurf structure on colloidal stability. Sodium dodecylbenzenesulfonate (SDBS) was shown to greatly influence the rate of TEMPO-mediated styrene miniemulsion polymerization [37]. Other sulfonated surfactants did not exhibit such effect. The authors attributed those results to a participation of SDBS in reactions generating radicals. Nevertheless, the exact processes were not identified. Several studies involved molecular or macromolecular inisurfs in controlled radical polymerization [38-43]. To the best of our knowledge, only one work reported the use of a molecular inisurf for controlled radical polymerization [38]. Block copolymers prepared by ATRP were used in most cases. It was showed that the control of polymerization was obtained together with convenient colloidal stability. As underlined by the authors, with such systems, the molar masses obtained are closely linked to the average particle diameter. The only reported way to overcome this restriction is the addition of a hydrophobic co-initiator. It is noticeable that bromineterminated PEO were reported to be efficient reactive surfactants for ATRP and AGET ATRP [41]. Thus, the presence of a hydrophobic block in the initial macroinitiator was not required. Molecular and macromolecular surfmers (Table 2.3) were used in miniemulsion polymerization essentially with the aim of obtaining an irreversible anchoring of stabilizer at the surface of the final particles. According to the reactivity ratios, molecular surfmers can either be readily incorporated into the growing chains or homopolymerize. Thus, a transition may occur between molecular surfactants to macromolecular surfactant while polymerization proceeds [44]. Several studies compared the kinetics of polymerization either in the presence of surfmers or with non-reactive analogues. Results are available for both molecular and macromolecular surfmers. Comparing anionic and cationic molecular surfmers to non-reactive surfactants, Matahwa et al. concluded that the rate of polymerization, the molar mass distribution and the particle size were similar with cationic species with both styrene and methyl

38

MINIEMULSION POLYMERIZATION TECHNOLOGY

methacrylate [44]. With anionic surf mer, some differences were noticed. In another study about styrene miniemulsion polymerization stabilized with molecular cationic surfactants (surfmer or non-reactive), Cao et al. found similar particle sizes but faster polymerization kinetics with the surfmer [45]. With cationic polymeric surfactants used as stabilizers in styrene miniemulsion polymerization, Houillot et al. reported that particle sizes fell in the same range with the macromonomer and with its non-reactive analogue. Nevertheless, they also noticed that the kinetics of polymerization was slightly faster with the macromonomer despite similar droplet sizes [46]. Braganza et al. compared the incorporation of several surfmers in the final latex particles using either miniemulsion polymerization or emulsion polymerization [47]. They found that the overall incorporation was similar for the two processes but the fraction of surfmer buried into particle core was higher in the case of emulsion polymerization. To the best of our knowledge, there is only one example of surfmer used for step-polymerization in miniemulsion [48]. Polyurea nanocapsules were prepared using a surfmer carrying amine groups, in inverse minemulsion. The surfmer was involved in the formation of the polymeric shell and modified its permeability. Covalent anchoring of stabilizers onto particle surface provided excellent stability upon dialysis to the final latexes [46,49, 50]. Transurfs with controlled chemical structure were mainly applied for RAFT polymerization in miniemulsion (Table 2.4). One general idea was to use the transurf to confine radicals at the interface and thus prepare nanocapsules. The efficiency of such strategy was demonstrated in several examples. Poly(vinyl alcohol) (PVA) can be also considered as a transurf since it has been widely used as a polymeric stabilizer for emulsion and miniemulsion radical polymerization and it is involved in grafting reactions occurring through hydrogen abstraction followed by propagation. This latter process leads to a covalent attachment of PVA chains at the surface of the formed particles. Several studies have showed that miniemulsion process leads to a better accessibility of PVA chains as compared to emulsion polymerization. As a result, a higher fraction of attached PVA chains is obtained by the miniemulsion process [51]. Colloidal stability of the latexes was also reported to be improved by the miniemulsion process but this result was not directly linked to the amount of grafted PVA but to the chemical structure of the grafted chains [52-54].

MULTI-FUNCTIONAL STABILIZERS

39

2.4 Conclusion This chapter illustrated the concept of multifunctional stabilizer in miniemulsion polymerization, which has been particularly investigated over the last ten years. This concept is for a large part the application of that of reactive surfactant, which is well-known in emulsion polymerization. Nevertheless, the specificities of miniemulsion process allowed applications in controlling the morphology of the formed particles as well as their surface characteristics. In addition, stabilizers acting as hydrophobes were also reported which is specific to miniemulsion polymerization.

References 1. F.J. Schork, Y. Luo, W. Smulders, J.P. Russum, A. Butté, and K. Fontenot, Advances in Polymer Science, Vol. 175, p. 129,2005. 2. J.M. Asua, Progress in Polymer Science, Vol. 27. p. 1283,2002. 3. M. Antonietti, and K. Landfester, Progress in Polymer Science, Vol. 27, p. 689, 2002. 4. N. Bechthold, F. Tiarks, M. Willert, K. Landfester, and M. Antonietti, Macromolecular Symposia, Vol. 151, p. 549, 2000. 5. K. Landfester, M. Willert, and M. Antonietti, Macromolecules, Vol. 33, p. 2370, 2000. 6. A. Durand, and E. Marie, Advances in Colloid and Interface Science, Vol. 150, p. 90,2009. 7. H. Cramail, E. Cloutet, and K. Radhakrishnan, Macromolecular Engineering, Vol. 4, p. 2181, 2007. 8. T.F. Tadros TF, Advances in Colloid and Interface Science, Vol. 147-148, p. 281,2009. 9. I. Capek, Advances in Colloid and Interface Science, Vol. 107, p. 125,2004. 10. C.S. Chern, and C.-H. Lin, Journal of Applied Polymer Science, Vol. 92, p. 1961, 2004. 11. C.S Chern, Y.C. Liou, and T.J. Chen, Macromolecular Chemistry and Physics, Vol. 199, p. 1315,1998. 12. C.S. Chern, and J.-C. Sheu, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 38, p. 3188,2000. 13. C.S. Chern, and T.J. Chen, Colloids and Surfaces A, Vol. 138, p. 65,1998. 14. J.A. Alduncin, J. Forcada, and J.M. Asua, Macromolecules, Vol. 27,p. 2256, 1994. 15. G. Jia, N. Cai, Y. Xu, C. Liu, and X. Tan, European Polymer Journal, Vol. 43, p. 4453,2007. 16. Z.-Q. Yu, P.-H. Ni, J.-A. Li, and X.-L. Zhu, Colloids and Surfaces A,Vol. 242, p. 9,2004. 17. M. Manguian, M. Save, C. Chassenieux, and B. Charleux, Colloid and Polymer Science, Vol. 284, p. 142,2005.

40

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

18. G. Baskar, K. Landfester, and M. Antonietti, Macromolecules, Vol. 33, p. 9228, 2000. 19. Z.-Q. Yu, D.-Y. Lee, I.-W. Cheong, J.-S. Shin, Y-J. Park, and J.-H. Kim, Journal of Applied Polymer Science, Vol. 87, p. 1941, 2003. 20. Z.-Q. Yu, D.-Y. Lee, I.-W. Cheong, J.-S. Shin, Y-J. Park, and J.-H. Kim, Journal of Applied Polymer Science, Vol. 87, p. 1933,2003. 21. M.B.J. Meinders, W. Kloek, and T. van Vliet, Langmuir, Vol. 17, p. 3923, 2001. 22. M.B.J. Meinders, and T. van Vliet, Advances in Colloid and Interface Science, Vol. 108-109, p. 119,2004. 23. A.S. Kabal'nov, Journal of Dispersion Science and Technology, Vol. 22, p. 1, 2001. 24. S. Mun, and D.J. McClements, Langmuir, Vol. 22, p. 1551, 2006. 25. B.T.T. Pham, D. Nguyen, C.J. Ferguson, B.S. Hawkett, A.K. Serelis, and C.H. Such, Macromolecules, Vol. 36, p. 8907, 2003. 26. A.N.F. Peck, and J.M. Asua, Macromolecules, Vol. 41, p. 7928, 2008. 27. D. Quémener, V. Héroguez, and Y. Gnanou, Macromolecules, Vol. 38, p. 7977, 2005. 28. D. Quémener, V. Héroguez, and Y. Gnanou, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 44. p. 2784, 2006. 29. A. Guyot, Macromolecular Symposia, Vol. 179. p. 105, 2002. 30. A Guyot, Recent Research Development in Polymer Science, Vol. 8. p. 207,2004. 31. A Guyot, and K. Tauer, Advances in Polymer Science, Vol. I l l , p. 42,1994. 32. U. Yildiz, and K. Landfester, Polymer, Vol. 49, p. 4930, 2008. 33. C. Maitre, F. Ganachaud, O. Ferreira, J.-F. Lutz, Y. Paintoux, and P. Hémery, Macromolecules, Vol. 33, p. 7730,2000. 34. A. De Gunzbourg, S. Maisonnier, J.-C. Favier, C. Maitre, M. Masure, and P. Hémery, Macromolecular Symposia, Vol. 132, p. 359,1998. 35. S. Cauvin, R. Dos Santos, and F. Ganachaud, e-Polymers, 050, 2003 36. S. Cauvin, A. Sadoun, R. Dos Santos, J. Belleney, F. Ganachaud, and P. Hémery, Macromolecules, Vol. 35, p. 7919,2002. 37. M. Lin, J.C.C. Hsu, M.F. Cunningham, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 44, p. 5974,2006. 38. F. Stoffelbach, N. Griffete, C. Bui, and B. Charleux, Chemical Communications, p. 4807,2008. 39. F. Stoffelbach, B. Belardi, J.M.R.C.A. Santos, L. Tessier, K. Matyjaszewski, and B. Charleux, Macromolecules, Vol. 40, p. 8813,2007. 40. D. Xiong, Z. Li, R. Ma, Y. An, and L. Shi, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 47, p. 1651,2009. 41. W. Li, K. Min, K. Matyjaszewski, F. Stoffelbach, and B. Charleux,Macromolecules, Vol. 41, p. 6387, 2008. 42. W. Li, K. Matyjaszewski, K. Albrecht, and M. Möller, Macromolecules, Vol. 42, p. 8228,2009. 43. M.N. Alam, RB. Zetterlund, and M. Okubo, Polymer, Vol. 50, p. 1632,2009. 44. H. Matahwa, J.B. McLeary, R.D. Sanderson, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 44, p. 427,2006. 45. N. Cao, X. Wang, L. Song, and Z.C. Zhang, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 45, p. 5800,2007. 46. L. Houillot, J. Nicolas, M. Save, B. Charleux, Y Li, and S.P Armes, Langmuir, Vol. 21, p. 6726, 2005.

MULTI-FUNCTIONAL STABILIZERS

41

47. S.N. Braganza, E.S. Daniels, E.D. Sudol, V.L. Dimonie, A. Klein, and M.S. El-Aasser, PMSE Preprints, Vol. 99, p. 316,2008. 48. E.-M. Rosenbauer, K. Landfester, and A. Musyanovych, Langmuir, Vol. 25, p. 12084, 2009. 49. D. Crespy, A. Musyanovych, and K. Landfester, Colloid and Polymer Science, Vol. 284, p. 780,2006. 50. A. Pich, S. Datta, A. Musyanovych, H.-J.P. Adler, and L. Engelbrecht, Polymer, Vol. 46, p. 1323,2005. 51. S.J. Bohorquez, and J.M. Asua, Macromolecules, Vol. 41, p. 8597,2008. 52. N. Kim, E.D. Sudol, V.L. Dimonie, and M.S. El-Aasser, Macromolecules, Vol. 36, p. 5573,2003. 53. N. Kim, E.D. Sudol, V.L. Dimonie, and M.S. El-Aasser, Macromolecules, Vol. 37, p. 2427,2004. 54. N. Kim, E.D. Sudol, V.L. Dimonie, and M.S. El-Aasser, Macromolecules, Vol. 37, p. 3180,2004. 55. J.-W. Kim, J.-Y. Ko, J.-G. Park, J.-B. Jun, I.-S. Chang, and K.-D. Suh, Journal of Applied Polymer Science, Vol. 85, p. 328,2002. 56. P. Ni, M. Zhang, L. Ma, and S. Fu, Langmuir, Vol. 22, p. 6016,2006. 57. C M . Spillmann, J. Naciri, K.J. Wahl, YH. Garner III, M.-S. Chen, and B.R. Ratna, Langmuir, Vol. 25, p. 2419,2009. 58. A. Martins dos Santos, T. Le Bris, F. D'Agosto, and M. Lansalot, Macromolecules, Vol. 42, p. 946, 2009. 59. Y. Luo, and H. Gu, Macromolecular Rapid Communications, Vol. 27, p. 21,2006. 60. F. Lu, Y. Luo, and B. Li, Macromolecular Rapid Communications, Vol. 28, p. 868, 2007. 61. Q. Xiong, P. Ni, F. Zhang, and Z. Yu, Polymer Bulletin, Vol. 53, p. 1,2004. 62. Y. Luo, and H. Gu, Polymer, Vol. 48, p. 3262,2007. 63. F. Lu, Y Luo, B. Li, Q. Zhao, and F.J. Schork, Macromolecules, Vol. 43, p. 568, 2010. 64. F. Lu, Y Luo, and B. Li, Industrial and Engineering Chemistry Research, Industrial and Engineering Chemistry Research, Vol. 49, p. 2206,2010.

This Page Intentionally Left Blank

3

Structured Copolymer Particles by Miniemulsion Polymerization* V. Mittal BASF SE, Polymer Research, 67056 Ludwigshafen, Germany

Abstract

Miniemulsion polymerization has been used for the synthesis of a large number of copolymer particle systems. It has been successful in the incorporation of the very hydrophobic monomers in the polymer particles owing to the polymerization of monomer droplets and no requirement of the diffusion of the monomers from the aqueous phase. The hydrophobic monomer in such cases also acts as a costabilizer by granting stability to the monomer droplets against Ostwald ripening. Both batch as well as semicontinuous modes of copolymerization processes has been reported, which have significant impact on the monomer compositions of the copolymer chains. Various ionic and non-ionic surfactants have been employed as well as the use of various water soluble or oil soluble initiators. Redox initiator pairs have also been used in order to polymerize the hydrophobichydrophilic monomer pairs. The copolymerization reactions of not only liquid monomers, but the monomers in the gaseous form have also been copolymerized using miniemulsion processes. Keywords: miniemulsion, stability, Ostwald ripening, surfactant, costabilizer, hydrophobicity, monomer droplets, conversion, size distribution, redox initiation, initiator, hybrid.

3.1

Introduction

Copolymer particles are synthesized in order to combine the good properties of the individual polymer materials. Thus, it is more the copolymer or composite particles which find wider "This review work was carried out at Institute of Chemical and Bioengineering, Department of Chemistry and Applied Biosciences, ETH Zurich, Zurich, Switzerland. V. Mittal (ed.) Miniemulsion Polymerization Technology, (43-70) © Scrivener Publishing LLC

43

44

MINIEMULSION POLYMERIZATION TECHNOLOGY

commercial applications owing to the broader range of different properties, than the homopolymer particles. The kinetics of copolymerization is more complex in nature owing to the different reactivity ratios of the monomers. Scheme 3.1 represents the multitude of various copolymerization reaction steps when only two monomers are copolymerized. The polymerization involving three or four monomers leads to much more complex kinetics. The feeding methodology of monomers during the course of polymerization as well as reactivity ratios of the monomers are required to be considered in order to achieve copolymer with uniform compositions of the monomer in the particles. Batch polymerization of the monomers, which differ significantly in their reactivity ratios, would lead to the generation of gradient copolymer particles. In some cases, it is beneficial to achieve gradient in the concentration of the monomers to achieve special functionalities in the polymer particles, but in other cases, it is important to control the

Scheme 3.1. Representation of the copolymerization process. Reproduced from Nova Science Publishers with permission.

STRUCTURED COPOLYMER PARTICLES

45

composition of the polymer chains through out the particles. In the gradient copolymer particles, monomer concentration changes along the radius of the particles, i.e. the more reactive monomer has higher concentration at the core of the particles owing to its faster reaction, thus leaving the less reactive monomer to lag behind which subsequently is present mostly in the shell of the polymer particles. This phenomenon is also termed as compositional drift and it leads to compositions of the monomers in the polymer chains much different from the initial monomer feed ratios. Differential partitioning of the monomers in the aqueous and organic phases also leads to the compositional drift. Semibatch addition of the monomers is preferred over the batch addition, as this mode of monomer feeding allows better control over the chemical composition of the copolymer particles. It thus allows the composition of the polymer chains in the particles to have similar monomer compositions as in the monomer feeds. Semibatch mode of addition is also performed by two different methodologies: flooded addition and starved addition of monomers. Flooded addition of monomers corresponds to monomer addition protocol whereby the monomer addition rate is faster than the polymerization rate of the monomers. But the limitation of this method is the generation of significant extent of secondary nucleation. Secondary nucleation may be beneficial in some cases, but most of the times, it is not required. This limitation can be overcome by following the second methodology of monomer addition in starved addition mode, in which the monomer addition rate is slower than the monomer polymerization rate, and this allows the chemical composition of the polymer chains to be equal to the ratio of monomers in the feed. By the variation in the monomer feeds, the particles with specific characteristics of core and surface can be synthesized. The semibatch methods though allow to generate polymer particles with controlled composition, however, they also lead to the broadening of the molecular weight distribution of the polymer chains in the particles owing to the chain transfer to the polymer during the course of polymerization. Thus, apart from the method of polymerization i.e. emulsion or miniemulsion polymerization, the process by which the polymerization is carried out also determines the morphology as well as properties of the particles. The following paragraphs review the various copolymer particle systems in miniemulsion as reported in the literature.

46

MINIEMULSION POLYMERIZATION TECHNOLOGY

3.2 Styrene-Dodecyl Methacrylate/Stearyl Methacrylate Chern et al. [1-4] reported extensively on the copolymerization of dodecyl methacrylate and stearyl methacrylate with styrene in miniemulsion. Dodecyl methacrylate and stearyl methacrylate were also used as reactive costabilizers in these polymerizations, therefore, their amount was much lower as compared to styrene. Nonylphenol polyethoxylate with an average of 40 ethylene oxides per molecule (NP-40) and sodium dodecyl sulphate (SDS) were used as surfactants. Figure 3.1 demonstrates the cube of monomer droplet diameter as a function of time when the concentration of the comonomer is increased. In first case, SDS is used as surfactant whereas NP-40 is used in the second case. By the time derivative of the cube of droplet diameter, one can attain the quantitative information on the Ostwald ripening of the particles. On comparison it was observed that stearyl methacrylate was more effective in retarding the Ostwald ripening than dodecyl methacrylate. Figure 3.2 also demonstrates the conversion as well as size evolution of the polymer particles as a function of time. The behavior of two comonomers of dodecyl methacrylate and stearyl methacrylate was compared. As stearyl methacrylate was better in retarding the Ostwald ripening of the monomer droplets, the polymerization rate is faster in this case as the number of polymerization sites is also much higher than in the case of dodecyl methacrylate. The lower size of droplets in the case of stearyl methacrylate also leads to lower particles sizes after miniemulsion polymerization as represented in Figure 3.2b. Different amounts of hydrophilic monomers like 2-hydroxyalkyl methacrylates were also added to the styrene miniemulsion copolymerization with dodecyl methacrylate or stearyl methacrylate [4], which acted also as costabilizers, as mentioned above. The generation of hydroxyl groups on the surface of the particles owing to the incorporation of 2-hydroxyalkyl methacrylates leads to specific surface properties in the polymer particles which are of immense importance in defining their applications. The addition of such hydrophilic monomers leads to the generation of polymer nuclei in the aqueous phase thus leading to the homogenous mode of particle nucleation as compared to the droplet nucleation mode. As mentioned above that the use of dodecyl methacrylate comonomer also as costabilizer led to the generation of higher Ostwald ripening

STRUCTURED COPOLYMER PARTICLES

47

Figure 3.1. Cube of droplet diameter as a function of time by using two different surfactants and comonomers. Reproduced from reference 1 with permission from Elsevier.

48

MINIEMULSION POLYMERIZATION TECHNOLOGY

Figure 3.2. Monomer conversion as well as particle diameters as a function of time using two different comonomers or costabilizers. Reproduced from Elsevier with permission from reference 3.

than the stearyl methacrylate, therefore the particles generated in this system on the addition of 2.hydroxyalkyl methacrylates were predominately formed by homogenous nucleation. The use of stearyl methacrylate on the other hand does not lead to diffusional

STRUCTURED COPOLYMER PARTICLES

49

degradation of the monomer droplets thus helping to stabilize the monomer droplets better. The particles formed thus in this system after the incorporation of 2-hydroxyalkyl methacrylate were formed by droplet nucleation mode. Overall, in both the cases, increasing the amount of 2-hydroxyalkyl methacrylate in the system led to faster monomer conversion owing to the increase in the number of reaction nuclei. The overall particles size thus also dropped on increasing the extent of hydrophilic monomer. The size of the particles in the case of stearyl methacrylate comonomer and costabilizer was lower than the size of the particles in dodecyl methacrylate system owing to the lower extent of Ostwald ripening. The extent of monomer droplet nucleation was also increased on increasing the amount of 2-hydroxyalkyl methacrylate during the polymerization owing to the increased probability of capturing the waterborn particle nuclei by monomer droplets exhibiting a very large oil-water interfacial area. Two different kinds of 2-hydroxyalkyl methacrylate monomers (2-hydroxyethyl methacrylate (HEMA), and 2-hydroxypropyl methacrylate (HPMA)) were also added to the miniemulsion polymerization of styrene, but the effect of these monomers was similar.

3.3 n-Butyl Methacrylate-Crosslinking Monomers Ghazaly et al. [5] reported the miniemulsion copolymerization of n-butyl methacrylate with the various crosslinking monomers. These monomers included (a) macromonomer crosslinker (Mac) with a number average molecular weight of 3480 g / m o l and polydispersity index of 1-1 to 1.2, (b) ethylene glycol dimethacrylate (EDGMA) and (c) aliphatic urethane aery la te macromonomer (AU A). As the crosslinking monomers have very low water solubility thus limiting their diffusion through the aqueous phase, therefore miniemulsion polymerization was preferred over the emulsion polymerization. Hexadecane was used as a costabilizer and sodium dodecyl sulphate was used a surfactant in the copolymerization reactions. Both oil soluble and water soluble initiators were used. Potassium persulphate was the water soluble initiator while 2,2'-azobis(2-methy lbutyronitrile) (AMBN) was oil soluble initiator. In the case of oil soluble initiator, the initiator was added along with the monomers during the preparation of miniemulsions, whereas when water soluble initiator was used, it was added later to the miniemulsion.

50

MINIEMULSION POLYMERIZATION TECHNOLOGY

Stable miniemulsions were observed in all the three crosslinking co-monomers with n-butyl methacrylate and the resulting latexes were also stable. The development of gel content in the crosslinked latexes using oil soluble initiator as a function of n-butyl methacrylate was different for all the three crosslinking monomers. Faster rates of gel formation were reported for the copolymer with EDGMA and AUA, the fastest being EDGMA. The behavior was also overall similar when water soluble initiator was used. As the reactions in the system involved copolymerization, it was therefore also remarked that the differences in the reactivity ratios of the monomers leads to the compositional drift in the polymer chains, with more reactive monomers tending to polymerize first, followed by the polymerization of less reactive monomers. The reactivity ratio of the n-butyl methacrylate was much higher from Mac indicating that the propagating chains of poly(n-butyl methacrylate) would prefer to react with its own monomer than Mac. Mac monomer was also very hydrophobic, therefore, it was opined that it may be present mostly in the core of the particles and more of n-butyl methacrylate may then be found on the surface of the particles. The authors also postulated from the experimental data of the homo-polymerization of Mac using water and oil soluble initiators that the in the copolymer particles, the Mac chains may orient in a way that the less hydrophobic portion of these chains is oriented towards the surfaces of the particles that are more hydrophilic in nature. The use of water and oil soluble initiators also affected the kinetics of the system significantly. Owing to the different initiators, the radicals were generated at different sites. When a water soluble initiator was used, it produced radicals in the aqueous phase and reacted with the butyl methacrylate molecules dissolved in water. If the reactive end groups of Mac comonomer were present away in the core of the growing particles, it would then delay their polymerization and the n-butyl methacrylate polymerization would take place predominately. From the particle size distributions, it was concluded that when the copolymers of n-butyl methacrylate were synthesized with crosslinker of larger molecular weight, the polydispersity index in the size distribution of particles increased. It was attributed to the possibility of attainment of additional stability to the droplets against Ostwald ripening in the presence of long chain crosslinker molecules. The swelling behavior of the three copolymer particles was also analyzed and it was

STRUCTURED COPOLYMER PARTICLES

51

observed that the morphology of the copolymer particles was different from each other.

3.4

Vinyl Acetate-Butyl Acrylate

Copolymerization of vinyl acetate and butyl acrylate was reported by using the batch miniemulsion and macroemulsion polymerization methods, and the emulsion stability, polymerization kinetics, copolymer composition and particle properties were compared owing to these two different polymerization methods [6]. Sodium hexadecyl sulphate was used as surfactant and hexadecane was used as cosurfactant in the miniemulsion polymerization. Sodium hexadecyl sulphate was synthesized by the reaction of hexadecanol with chlorosulfonic acid followed by neutralization of alkyl sulfuric cid with sodium hydroxide in butanol. Miniemulsions were generated in two different ways. In the first case, hexadecane was mixed with the organic phase, i.e., monomer mixture, which was then mixed with the aqueous phase containing surfactant. In the second case, hexadecane was added to the aqueous phase containing surfactant first and the system was sonified to generate a miniemulsion. The monomer mixture was subsequently added to this miniemulsion. It was observed that the amount of sodium hexadecyl sulphate adsorption onto the mixed monomer droplets was significantly affected by hexadecane. Sodium hexadecyl sulphate concentrations of 5,10 and 20 mM were used and the presence of a small amount of hexadecane led to a drastic increase in the adsorption of surfactant in all the three concentrations. However, the adsorption leveled off at a hexadecane/sodium hexadecyl sulphate ratio of 3. As the amount of the surfactant adsorbed on the surface of droplets is relatable to the size of the droplets, therefore, at a given concentration of surfactant, the addition of hexadecane leads to the decrease in the monomer droplet size leading to the extension of surface area available for adsorption, which justifies the sharp increase in the amount of adsorption of the surfactant. Further increase in hexadecane amount leads to the droplet size to the minimum. It was thus observed that as hexadecane to sodium hexadecyl sulphate ratio and initial concentration of sodium hexadecyl sulphate was increased, the droplet size of monomer mixture was reduced and the emulsion stability was enhanced. Different recipes were produced by varying

52

MINIEMULSION POLYMERIZATION TECHNOLOGY

the amounts of surfactant and hexadecane. Induction period was observed to be present in all the miniemulsion polymerization trials. The presence of stable oxygen- containing air bubbles formed during the extensive emulsification process was reported to be a reason for such induction periods. It was also confirmed to the case as there was no induction period when no hexadecane was used for the emulsion polymerization. An increased surface resistivity to the entry of the radicals generated in the aqueous phase was also reported to be an alternate explanation, which occurs due to the presence of a thick layer of hexadecane-sodium hexadecyl sulphate at the surface of monomer droplets. The amount of coagulum generated in the polymerization reactions was also different, when hexadecane was used, only a small extent of coagulum was observed, on the other hand, the coagulum was significant in the trials without the use of hexadecane. The particle sizes were significantly different with and without the use of hexadecane. For latexes synthesized with 10 mM sodium hexadecyl sulphate, the particle diameter was observed to increase from 122 nm to 195 nm after the addition of hexadecane. The chemical compositions of the copolymer chains generated by conventional emulsion and miniemulsion polymerization were also significantly different. The polymer chains synthesized by miniemulsion process had lower contents of vinyl acetate units up to 70% conversion compared to the chains formed in emulsion polymerization process, thus confirming that the copolymerization achieved by the two methods have different kinetic behavior and leads to the generation of chemically different copolymer chains. Two different glass transition temperatures were observed in the latexes owing to the butyl acrylate and vinyl acetate rich phases. For the latex with miniemulsion polymerization, the glass transition temperatures were -29°C (butyl acrylate) and 52°C (vinyl acetate), whereas the temperatures were -20°C and 50°C respectively for the emulsion polymerization without hexadecane. The shift of 9°C in the miniemulsion latex was attributed to fewer amounts of vinyl acetate monomer units in the polymer chains during the first 70% conversion. It was concluded by the authors that hexadecane helped in the formation of stable emulsions with very small droplet size which became the locus of polymerization. The authors also opined that hexadecane may also act to hinder the rate of interparticle monomer transport during the polymerization and it was

STRUCTURED COPOLYMER PARTICLES

53

also observed to act as swelling promoter in the monomer-polymer system present in the particles.

3.5 Butyl Acrylate-(2-Methacryloxy)ethyl)trimethyl Ammonium Chloride Luo et al. reported the copolymerization of butyl acrylate with cationic monomer using the interfacial redox initiator system [7]. Butyl acrylate and cationic monomer 2-(methacryloyoxy)ethyl trimethyl ammonium chloride (MAETAC) form a hydrophobic and hydrophilic monomer pair and macroemulsion and miniemulsion polymerization reactions were carried out using cumene hydroperoxide and tetraethylenepentamine as initiators, hexadecane as cosurfactant and Triton X-405 as surfactant. The copolymerization of hydrophobic and hydrophilic monomers by emulsion polymerization is a challenge due to the residing of the hydrophilic monomers extensively in the aqueous phase. In this study, similarly, as butyl acrylate monomer is hydrophobic, it therefore stayed in the droplets, whereas owing to hydrophilicity, 2-(methacryloyoxy)ethyl trimethyl ammonium chloride significantly partitions in the aqueous phase. Gilbert et al. proposed that the section of initiator pair like cumene hydroperoxide and tetraethylenepentamine provides opportunities to graft hydrophilic monomer onto a hydrophobic polymer [8]. This redox initiator system forms a hydrophobic and hydrophilic pair system. Cumene hydroperoxide, owing to its hydrophobicity, remains significantly in the monomer droplets, whereas hydrophilic tetraethylenepentamine is significantly present in the aqueous phase. The authors underlined the ambiguity prevailing over the site of the radical generation from the oil soluble initiators like AIBN. Some studies have reported the generation of radicals in the aqueous phase even though AIBN is water insoluble, on the other hand, other studies reported the generation of the free radicals in the particles. The authors carried out the conventional emulsion polymerization of the system without hexadecane and using a water soluble initiator. It was observed that only MAETAC monomer was polymerized and butyl acrylate was not polymerized indicating that the free radicals could not enter the micelles to initiate the polymerization of hydrophobic monomer when a surfactant

54

MINIEMULSION POLYMERIZATION TECHNOLOGY

with a long hydrophobic tail was used thus requiring a modification in the components of the polymerization system. However, for the surfactant Triton X-405 used in the study, which has a long hydrophilic tail, it was suggested that the particles covered by this would have a viscous corona through which the radicals would penetrate to initiate the polymerization of the organic phase. Also, it was observed that the use of the redox initiator pair used in this study did not have the problem of initiation as mentioned above. Using this initiator pair, the radicals were generated at the aqueous organic interface and, thus, the radicals need not penetrate through the surfactant layer in order to initiate the polymer particles. It was also suggested that, owing to the water solubility of MAETAC monomer, it has the possibility to polymerize at two sites. The first is the polymerization in the aqueous phase and as the amount of butyl acrylate is not much in the aqueous phase owing to its water insolubility, therefore the generated chains would be either homopoly(MAETAC) or very rich in poly(MAETAC). The second possibility of the polymerization of MAETAC is at the interface of organic phase with aqueous phase. In such scenario, the butyl acrylate chains would also be included in the polymer chains. It was also reported that the more MAETAC was consumed at the very beginning of the macroemulsion polymerization than in the miniemulsion polymerization where no micelles exist. It was also observed that although more amount of MAETAC was consumed in the very beginning in the case of macroemulsion polymerization, the total amount of homopoly(MAETAC) was the same for both the macroemulsion and miniemulsion polymerizations. It was also observed that, in the case of miniemulsion polymerization, the homogenous nucleation occurred at low MAETAC concentration, but this was not the case at higher concentrations of MAETAC. Findings showed that although in both cases the kinetics of polymerization was significantly different, 18% of the MAETAC was polymerized in the aqueous phase in the final resulting latexes in both cases.

3.6 Butyl Acrylate-Methyl Methacrylate-Vinyl Acetate Unzue et al. [9] studied the terpolymerization of butyl acrylate, methyl methacrylate and vinyl acetate in the semicontinuous mode. The effect of operating conditions like solids content, initiator concentration, feed flow rate, amount of cosurfactant etc. on the kinetics

STRUCTURED COPOLYMER PARTICLES

55

of terpolymerization reaction was studied. Hexadecane was used as cosurfactant and potassium persulphate was used as initiator. The polymerization reactions were carried out following two set of experiments. In the first set, 10% of the miniemulsion was heated to 80°C and polymerized shortly for 15 minutes in batch mode. Subsequently, the remaining miniemulsion was fed to the reactor at constant feed rate. After the feeding operation was completed, the polymerization was run for further 1-2 hours. In the second set of experiments, the effect of operating conditions like partitioning of the monomer between the initial charge and the feed, type and concentration of surfactant, feed flow rate and sonication on the feasibility to obtain a 65 wt% solid content latex. The authors observed that in the beginning of the polymerization, the terpolymer was richer in methyl methacrylate and butyl acrylate and it contained almost no vinyl acetate. This effect was attributed to very different reactivity ratios of the monomers. Methyl methacrylate was the most reactive of the three followed by butyl acrylate and vinyl acetate. It was also observed that the homogeneity of the terpolymer improved when the instantaneous conversion increased as a result of the increase in the initiator concentration. The polymerization rate in the miniemulsion polymerization decreased as the concentration of the hexadecane cosurfactant increased. At the beginning of the process when the monomers accumulate in the reactor, it was observed that the instantaneous conversion was sensitive to the polymerization conditions, but the sensitivity was observed to significantly decrease as the polymerization proceeded further. It was concluded that the combination of anionic as well as nonionic emulsifiers was required to achieve stability in the system, and the anionic surfactant on its own was not sufficient to impart colloidal stability. The latex with high solid content of 65 wt% was also achieved and it was observed that the amount of coagulum increased when the anionic emulsified was used alone. Also, in the absence of sonication, the amount of coagulum was observed to increase.

3.7 Styrene-Acrylic Acid or 2-Aminoethyl Methacrylate Hydrochloride (AEMH) Musyanovych et al. [10] reported the synthesis of carboxyl and amino functionalized copolymer particles with styrene using the miniemulsion copolymerization reactions of styrene with acrylic acid or 2-aminoethyl methacrylate hydrochloride (AEMH). 2,2'-azobis

56

MINIEMULSION POLYMERIZATION TECHNOLOGY

(2-methylbutyronitrile) was used as oil soluble initiator. Lutensol AT-50 was used as non-ionic surfactant, whereas sodium dodecyl sulfate (SDS) and cetyltrimethylammonium chloride (CTMA-C1) were employed as ionic surfactants. The effect of the various comonomers and the surfactant types on the particles size as well as its distribution was investigated. The copolymer latexes polyStpolyAA and polySt-polyAEMH generated by using non-ionic surfactant were observed to show higher tendency to coagulate. The tendency to coagulate was more pronounced during the synthesis of polySt-polyAA latexes in comparison to the pure polystyrene or poly St-polyAEMH latexes. At 3 wt% amount of acrylic acid corresponding to total monomer content, the latex after 30 min of polymerization became viscous, and phase separation was observed in the end of polymerization. It was also observed that the diameter of the polymer particles sharply decreased with increasing acrylic acid content. The amino-functionalized copolymer particles were also reported to decreases sharply in size with the increase of the functional monomer till 3 wt%. The decrease was slow for the subsequent addition of AEMH until 20 wt% and an average particle size of about 120 nm was observed. For the generation of the latexes with ionic surfactants, polySt-polyAA latex copolymers particles was achieved by using anionic surfactant SDS, whereas polyStpolyAEMH copolymer latexes were synthesized by using cationic surfactant (CTMA-C1). The latexes generated with ionic surfactants were observed to have very low extent of coagulum generation. In contrast to the polymerization with non-ionic surfactant, polyStpolyAA latex particles generated by using the ionic surfactants were particularly stable even with 20 wt% of acrylic acid. The size of the particles was also observed to be constant until 10 wt% of added acrylic acid, which is again markedly different from the case of nonionic surfactants. At higher amounts of acrylic acid, the diameter was observed to sharply increase thus reaching an average value of 140 nm. The authors attributed the increase in particle size on increasing the content of acrylic acid to the formation of a "hairy" layer around the particles composed of the hydrophilic poly(acrylic acid) units. Interesting differences were also observed in the particle size distributions of the copolymer latexes with two different types of surfactants. The carboxyl and amino-functionalized particles generated by using non-ionic surfactants were observed to have high extent of polydispersity index and a bimodal size distribution of particles. The particles size distributions, on the other

STRUCTURED COPOLYMER PARTICLES

57

hand, in both the carboxyl and amino-functionalized particles synthesized by using ionic surfactant was monomodal, and the polydispersity index was not greater than 0.2. Figure 3.3 demonstrates carboxyl- and amino-functionalized polystyrene particles stabilized with non-ionic surfactant with 1 wt% of acrylic acid and with 3 wt% of AEMH. The results thus confirmed that it is not the functional comonomers themselves, but the combination of functional comonomer with non-ionic surfactant which leads to the generation of bimodality in the particle size. Figure 3.4 shows the mechanism of the bimodal particle generation in the presence of non-ionic surfactant suggested by the authors. By using the fluorescent dye molecules in miniemulsion polymerizations, they concluded that bimodal size distribution of the final particles may be a result of a budding-like effect, i.e., splitting of the miniemulsion droplets before or during the early stage of polymerization.

3.8

Styrene-Butyl Acrylate

Roberge et al. [11] reported the batch copolymerization of styrene and butyl acrylate for the synthesis of pressure sensitive adhesives. For the generation of miniemulsion system, organic components styrene, butyl acrylate and octadecyl acrylate were mixed separately from the Triton X-405 and SDS surfactants in water. Both solutions were then mixed followed by stirring and sonication to generate the miniemulsion. The reactions were carried out at 80°C and potassium persulphate was used as an initiator. Figure 3.5 demonstrates the conversion of the monomers as a function of reaction time for the various feed compositions. It was observed that the fast reaction rates were observed for all the different composition runs. Increasing the content of butyl acrylate in the system also enhanced the rate of polymerization as the conversion was enhanced as a function of time. Also, the polymerizations were observed to reach full conversions in all the different runs. Figure 3.6 also shows the impact of monomer feed composition on cumulative copolymer composition.

3.9

Styrene-Butadiene Rubber

Styrene butadiene copolymer latexes were synthesized by miniemulsion polymerization [12]. Various oil soluble initiators like

58

MINIEMULSION POLYMERIZATION TECHNOLOGY

Figure 3.3. TEM images (left) and particle size distribution analysis (right) of carboxyl- and amino-functionalized polystyrene particles stabilized with nonionic surfactant: (a) no functionality, (b) with 1 wt% of acrylic acid, (c) with 3 wt% of AEMH. Reproduced from reference 10 with permission from American Chemical Society.

STRUCTURED COPOLYMER PARTICLES

59

Figure 3.4. Mechanism of bimodal particle size distribution generation by using non-ionic surfactant in carboxyl and amino-functionalized particles. Reproduced from reference 10 with permission from American Chemical Society.

Figure 3.5. Conversion vs. reaction time for the miniemulsion copolymerization of styrene and butyl acrylate. Run 1: Styrene 5, butyl acrylate 95, Triton X-405 0.5, SDS 0.03 and octadecyl acrylate 0.5; run 4: Styrene 10, butyl acrylate 90, Triton X-405 0.5, SDS 0.03 and octadecyl acrylate 0.5; run 7: Styrene 15, butyl acrylate 85, Triton X-405 0.5, SDS 0.03 and octadecyl acrylate 0.5. Reproduced from reference 11 with permission from Elsevier.

60

MINIEMULSION POLYMERIZATION TECHNOLOGY

Figure 3.6. Cumulative feed composition for various feed compositions. Reproduced from reference 11 with permission from Elsevier.

2,2'-azobis(2-methylbutyronitrile) (V59), 2,2'-azobis(4-methoxy-2,4dimethyl valeronitrile) (V70), 2 / 2'-azobis(2,4-dimethyl valeronitrile) (V65) were used. N-dodecyl mercaptane (n-DM), and tert-dodecyl mercaptane (t-DM) were employed as chain transfer agents. Hexadecane was used as cosurfactant whereas SDS and potassium oléate were chosen as surfactants. The miniemulsion copolymerization of this system was special in nature owing to the gaseous form of butadiene, whereas most of the reported studied used the liquid monomers. To obtain miniemulsion, initiator, hexadecane and chain transfer agent were first dissolved in styrene. The solution was then dispersed in water by stirring at 2000 rpm for 1 hour. The suspension was then sonicated to achieve styrene based miniemulsion. The required amount of butadiene was condensed into a pre-cooled reaction vessel. The styrene miniemulsion was subsequently added to the frozen butadiene followed by homogenization of the reaction mixture by vigorous stirring. The polymerization of the monomers was carried out for 14 hours. The weight ratios of butadiene to styrene of 75:25 were used, which are also similar to the ratios commonly employed in the industrial polymerizations. The miniemulsions observed to be suitable for generating high solid content latexes up to 60 wt%. The particles also showed a narrow and monomodal size distribution and were stable and for several months. No coagulum was formed during this time. The polymerization reaction was performed at 72°C, when V59 was used as initiator. The average

STRUCTURED COPOLYMER PARTICLES

61

particle size of the samples was observed to be around 130 nm and a small polydispersity index values. The initiator V70 was used at 30°C for the miniemulsion copolymerization. The resulting particles were stable and narrowly dispersed, however, the conversion was too slow indicating that the initiator may not ne suitable for achieving high extent of monomer conversion in the time period of interest. The use of V65 initiator at 50°C also resulted in stable and narrowly dispersed latexes and the conversion was also high. Average particles size of 100 nm was observed. The authors also reported that by the use of chain transfer agents, the molecular weight distribution in the chains as well as amount of gel could be reduced. t-DM chain transfer agent was observed to be more efficient than n-DM agent. The gel content in the particles using the n-DM chain transfer agent was 28% whereas it was significantly reduced to 6% when t-DM was employed as chain transfer agent. The two surfactants of SDS and potassium oléate were not very different in behavior. The average particle size of the latexes using SDS as surfactant had somewhat smaller size as compared to the oléate stabilized latexes, however, the particles in both the latexes had low polydispersity.

3.10

Fluoroacrylate-Lauryl Methylacrylate-Methyl Methacrylate

Zhang et al. [13] reported the miniemulsion copolymerization of fluoroacrylate, lauryl methylacrylate, and methyl methacrylate monomers. It is difficult to achieve the polymerization of fluoroacrylate in the conventional emulsion polymerization owing to its hydrophobicity which retards its diffusion in the aqueous phase. Fluoroacrylate also acted as a cosurfactant apart from being a comonomer owing to its extremely hydrophobic nature. Oil soluble 2,2'-azobisisobutyronitrile (AIBN) was used as initiator and nonylphenol polyethoxylate with an average of 40 ethylene oxide units per molecule (NP-40) and cetyltrimethylammonium (CTAB), were used as surfactants. It was observed that the use of fluoroacrylate retarded the diffusional degradation of monomer droplets by Ostwald ripening and allowed the production of stable miniemulsions. The composition of the monomers in copolymer chains synthesized in miniemulsion polymerization was observed to be in good agreement with the feed ratios used. The chains, on the other hand, had significant differences in the calculated and theoretical

62

MINIEMULSION POLYMERIZATION TECHNOLOGY

values of monomer compositions, even at low feed ratio of fluoroacrylate, when the polymerization was carried out in emulsion. The size of the particles was similar to the size of the monomer droplets in the case of miniemulsion polymerization indicating efficient monomer droplet mechanism. The particle size distributions were broader than the particles synthesized conventional emulsion polymerization. The delayed nucleation of the monomer droplets is the result of the broad size distributions as the monomer droplets have much lower radical capture efficiency. On the other hand, in emulsion polymerization, the micelles are very quickly turned into growing polymer particles and thus the particles size distribution is also narrow. Reaction parameters were observed to greatly influence the morphology of the resulting copolymer latex. Polymerization rate and monomer conversion was reported to increase with increasing concentrations of surfactants and initiator. The particles generated in miniemulsion polymerization had a variety of internal structures, such as core—shell morphology or multiblob structures.

3.11

Polyurethane-Block-Polystyrene

Koenig et al. [14] reported the synthesis of water borne polyurethane-block-polystyrene latexes in miniemulsion polymerization. To obtain such functional latexes, miniemulsion copolymerization to obtain copolymer chains with specific blocks was performed by using macroazoinitiator. Styrene, isophorone diisocyanate (IPDI), 2,4-diethyl-l,5-pentanediol and a diol-functionalized azoinitiator were dispersed in water. As a first step, the polyaddition reaction was performed at room temperature to attain polyurethane block. Catalyst mixtures dibutyltindilaurate (DBTDL) and dimethyldodecylamine (DMDA) was used for this purpose. This led to the synthesis of a PU macroazoinitiator. Subsequently, the miniemulsion was heated to 72°C to initiate the free radical polymerization of styrene. Hexadecane was used as cosurfactant whereas sodium dodecyl sulphate was used as surfactant. It was observed that the copolymer could be successfully formed and it was not a mixture of two homopolymers. It was observed from the gel permeation chromatography, differential scanning calorimetry and nuclear magnetic resonance studies that 45% of a (linear) copolymer consisting of PU and PS was obtained. 28 wt% of homo-polystyrene was

STRUCTURED COPOLYMER PARTICLES

63

also observed owing to the presence of free azoinitiator molecules, which have not been incorporated in the PU macroazoinitiator. Transmission electron microscopy studies of the copolymer particles also exhibited homogeneous structure inside the particles.

3.12

Alkyd-Acrylic

Nabuurs et al. [15] reported the synthesis of alkyd-acrylic hybrid. Though the system was reported as an emulsion polymerization, but the authors subjected the emulsions to high shear before polymerization. Although no costabilizer was additionally added, but the alkyd is presumed to acts also a costabilizer. The polymerization reactions were initiated with potassium persulfate (KPS), and anionic or mixed anionic/non-ionic surfactants were used for stabilization. Nonylphenol-10 ethylene oxide sodium sulfate was used as a surfactant in most of the polymerization trials. To synthesize the hybrid, the alkyd was first dissolved in the acrylic monomers followed by the addition of surfactant and water. The emulsions after high shear homogenization were polymerized either in batch or in semi continuous mode. It was observed that the polymerization rates were slow and polymerization efficiency was also low owing to the presence unsaturated fatty acids in alkyd. The authors used various acrylic monomers, but methyl methacrylate was observed to generate the most finely dispersed pre-emulsion monomer droplets. When the polarity of the monomers was increased, i.e. when the monomer became too water soluble, the alkyd/acrylic ratio in the droplets was increased. This resulted in the more viscous system and in a less efficient particle size reduction. Figure 3.7 shows the conversion as well as polymerization rate of the hybrid polymerization reactions using the alkyds with varying concentrations of unsaturated aliphatic groups. It was observed that the increasing concentration of unsaturated fatty acids in the alkyd resulted in the enhanced extent of retardation of the polymerization rate. The conversion in the case of alkyds containing higher concentration of unsaturated groups did not reach to completion. The hybrid with fully saturated fatty acids was observed to reach a conversion of 100% after less than 1 h, whereas the hybrid with the highest concentration of unsaturated fatty acids had a monomer conversion of 93% after about 3 h. The enhancement of the properties of the hybrids over the individual components was observed as shown

64

MINIEMULSION POLYMERIZATION TECHNOLOGY

in Figure 3.8 for the hardness of the coatings generated from the latex particles. The coatings generated form the hybrid particles were much harder as compared to acrylic or alkyd coatings and the enhancement was also observed to retain as a function of time.

STRUCTURED COPOLYMER PARTICLES

3.13

65

Oil-Acrylate

Similar to the above reported, alkyd - acrylic system, Van Hamersveld et al. [16,17] also reported the use of oxidized triglycérides (e.g. sunflower oil) as initiator for the hybrid miniemulsion polymerization with methyl methacrylate. Unsaturated triglycérides were treated in the study with molecular oxygen to generate fatty-acid hydroperoxide groups. Fatty-acid hydroperoxides were then used as initiators for the hybrid miniemulsion polymerization reactions. Fatty-acid hydroperoxide Fe2 ethylenediaminetetraacetic acid sodium formaldehyde sulfoxylate (SFO-HP/Fe 2 /EDTA/SFS) formed the redox initiation system. The mini-emulsion system was stabilized with dodecyl sulphate as surfactant and hexadecane as hydrophobe. To generate miniemulsion, hydroperoxidized sunflower oil (SFO-HP) was added with monomer and hexadecane. The oil-monomer solution was then dispersed into a SDS-water solution. The miniemulsion was generated by sonication of the emulsion. Figure 3.9 shows the number average size distribution of the monomer droplets after homogenization. The miniemulsion was then transferred into a polymerization reactor and a concentrated solution of SFS was added. The polymerization was initiated by the adding of a catalytic amount of complexed Fe2 solution. The cryo transmission electron

66

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Figure 3.9. Number average size distribution of the sunflower oil - MMA monomer droplets after homogenization. Reproduced from reference 16 with permission from Elsevier.

microscopy analysis (Figure 3.10) revealed that the hybrid particles generated by initiation with the fatty-acid hydroperoxides did not show intraparticle heterogeneity, whereas the particles generated by initiation with t-butyl hydroperoxide has heterogeneous morphology. The particles contained light and dark colored phases, the lighter phases was present either on the side or in the centre of the particles. The light colored component of the particles was not observed to protrude out of the particles. Therefore, the morphology of these particles was described as core-shell, but the core was not always present at the centre of the particles. The presence of no such heterogeneity in the particles initiated by SFO-HP indicated that the use of fatty-acid hydroperoxides resulted in the formation of oilacrylate copolymer which acted as a compatibilizer for the copolymer particles. The effect of reaction parameters on the conversion was also recorded. When SFO-HP was used as hydroperoxide, the rate of polymerization did not depend on the concentration of SFS

STRUCTURED COPOLYMER PARTICLES

67

Figure 3.10. Cryo-TEM images of sunflower oil-poly(methyl methacrylate) hybrid latexes initiated by (a) t-butyl hydroperoxide and (b) SFO-HP. Reproduced from reference 17 with permission from Elsevier.

as shown in Figure 3.11a. The concentration of Fe2 also did not have a significant effect on the conversion or rate of polymerization as exhibited in Figure 3.11b. The increase in polymerization temperature as well as hydroperoxide concentration led to the enhancement of the polymerization rate as shown in Figure 3.11c and lid.

3.14

Urethane-Acrylic

Li et al. [18] reported the synthesis of urethane/acrylic hybrid miniemulsion latex nanoparticles. The miniemulsion polymerization of n-butyl methacrylate was performed both in the presence and absence of the urethane prepoylmers and the kinetics and rates of polymerization were compared. The behavior of conventional emulsion polymerization was also compared with the miniemulsion reaction trials. Hexadecane was used as costabilizer and sodium dodecyl sulphate was used as surfactant. Redox initiator pair of hydrogen peroxide and ascorbic acid was used for the initiation of miniemulsions. The homopolymerization of n-butyl methacrylate was observed to faster in miniemulsion as compared to conventional emulsion polymerization where no cosurfactant and sonication was used. It was further observed that the rate of

68

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Figure 3.11. Conversion vs. time curves for (a) various concentrations of SFS, (b) various concentrations of Fe, (c) various temperatures; and (d) various hydroperoxide concentrations. Reproduced from reference 16 with permission from Elsevier.

miniemulsion polymerization rate of n-butyl methacrylate was much faster when urethane prepolymer was added to the system. The authors attributed this behavior to the difference in the number of initial miniemulsion droplets. As the urethane prepolymer was more hydrophilic than n-butyl methacrylate, the initial size of the monomer droplets was much smaller than droplets containing pure n-butyl methacrylate. This led to the generation of the large number of monomer droplets in the system resulting in the creation of large number of polymerization loci.

References 1. 2. 3. 4.

C.S. Chern, and T.J. Chen, Colloids Surfaces A, Vol. 138, p. 65,1998. C.S. Chern, and T.J. Chen, Colloid and Polymer Science, Vol. 275, p. 1060,1997. C.S. Chern, and Y.C. Liou, Polymer, Vol. 40, p. 3763,1999. C.-S. Chern, and J.-C. Sheu, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 38, p. 3188, 2000.

STRUCTURED COPOLYMER PARTICLES

69

5. H.M. Ghazaly, E.S. Daniels, V.L. Dimonie, A. Klein, and M.S. El-Aasser, Journal of Applied Polymer Science, Vol. 81, p. 1721,2001. 6. J. Delgado, M.S. El-Aasser, and J. W. Vanderhoff, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 24, p. 861,1986. 7. Y. Luo, and F.J. Schork, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 39, p. 2696,2001. 8. R.G. Gilbert, J.F. Anstey, N. Subramaniam, and M.J. Monteiro, American Chemical Society, Division of Polymer Chemistry, Vol. 40, p. 102,1999. 9. M.J. Unzue, and J.M. Asua, Journal of Applied Polymer Science, Vol. 49, p. 81,1993. 10. A. Musyanovych, R. Rossmanith, C. Tontsch, and K. Landfester, Langmuir, Vol. 23, p. 5367,2007. 11. S. Roberge, and M.A. Dube, Polymer, Vol. 47, p. 799, 2006. 12. M.-V. Kohnle, U. Ziener, and K. Landfester, Colloid and Polymer Science, Vol. 287, p. 259, 2009. 13. Q. Zhang, X. Zhan, and F. Chen, Journal of Applied Polymer Science, Vol. 104, p. 641, 2007. 14. A. Koenig, U. Ziener, A. Schaz, and K. Landfester, Macromolecular Chemistn/ and Physics, Vol. 208, p. 155,2007. 15. T. Nabuurs, R.A. Baijards, and A.-L. German, Progress in Organic Coatings, Vol. 27, p. 163,1996. 16. E.M.S. van Hamersveld, J.J.G.S. van Es, and F.P. Cuperus, Colloids and Surfaces A: Physiochemical and Engineering Aspects, Vol. 153, p. 285,1999. 17. E.M.S. van Hamersveld, J.J.G.S. van Es, A.L. German, F.R Cuperus, R Weissenborn, and A.-C. Hellgren, Progress in Organic Coatings, Vol. 35, p. 235,1999. 18. M. Li, E.S. Daniels, V.L. Dimonie, E.D. Sudol, and M.S. El-Aasser, Polymeric Materials: Science & Engineering, Vol. 85, p. 258,2001.

This Page Intentionally Left Blank

4

Encapsulation of Inorganic Nanoparticles by Miniemulsion Polymerization Jacqueline Forcada2 and Jose Ramos1 1

Grupo de Física de Fluidos y Biocoloides, Departamento de Física Aplicada, Facultad de Ciencias, Universidad de Granada, 18071 Granada, Spain institute for Polymer Materials POLYMÄT and Grupo de Ingeniería Química, Facultad de Ciencias Químicas, Universidad del País Vasco/EHU, Apdo. 1072, Donostia-San Sebastián 20080, Spain

Abstract

Miniemulsion polymerization is a powerful technique to encapsulate inorganic nanoparticles into a polymer shell. However, prior to carrying out the miniemulsion process, the surface of the inorganic nanoparticles must be converted into a more hydrophobic one by using surface modifiers or coupling agents. The type of surface modifier used depends strongly on the type of inorganic nanoparticle to encapsulate. In this chapter, special attention will be paid to the encapsulation of silica and magnetite nanoparticles by miniemulsion polymerization using different surface modifiers. The use of the right combination of surface modifiers together with the adequate miniemulsion process (single or double) are the key factors for obtaining a high encapsulation degree of the inorganic nanoparticles together with a well-controlled morphology of the hybrid nanocolloids. Keywords: miniemulsion polymerization; inorganic nanoparticles; encapsulation

4.1

Introduction

Nanotechnology is one of the key technologies of the21 5t century, which is making great steps forward in the improvement of existing materials and the production of advanced and innovative materials in the V. Mittal (ed.) Miniemulsion Polymerization Technology, (71-96) © Scrivener Publishing LLC

71

72

MINIEMULSION POLYMERIZATION TECHNOLOGY

colloidal range based on both inorganic and polymeric materials as well as nanocomposites consisting of a mixture of them. In this way, considerable efforts have been devoted in the development of new polymer encapsulation techniques because polymer-encapsulated inorganic nanoparticles exhibit enhanced even novel properties (e.g., mechanical, chemical, electrical, rheological, magnetic and optical) and offer very interesting actual and potential applications in different fields, such as optics, catalysis, microelectronics, coating, cosmetics, inks, agriculture, drug release systems, diagnostics, and so on [1-5]. Among the different methods for the encapsulation of inorganic nanoparticles, heterophase polymerization is by far the most frequently used technique [6,7]. In this case, composite nanoparticles are prepared by carrying out the polymerization in the presence of inorganic nanoparticles via emulsion or miniemulsion polymerization. In the case of an emulsion polymerization, it is well known that homogeneous and micellar nucleations are the main mechanisms for particle formation. In the presence of inorganic particles dispersed in the aqueous phase, particles surface can be an additional site for particle nucleation. Thus, the control of the morphology of the composite nanoparticles can become complicated due to the competition among these nucleation mechanisms. However, in miniemulsion polymerization, particle nucleation occurs mainly inside submicrometer-sized monomer droplets (ranging from 50 to 500 nm in diameter). Because of their small size, the large overall surface area of the droplets can effectively compete for radical capture. If inorganic particles can be dispersed in the monomer phase followed by miniemulsification, then each miniemulsion droplet can indeed be treated as a small nanoreactor, which produces composite particles with adequate morphology control and a high encapsulation degree of inorganic particles. One of the most important issues during the preparation of inorganic/polymeric composite particles is the formation of specific interactions at the interface of the organic and inorganic phases. Therefore, surface modification of inorganic particles is a prerequisite step before performing the polymerization reaction. With this surface modification, two important goals can be achieved: introduction of different reactive groups useful in the subsequent polymerization reaction and enhancement of the surface hydrophobicity of the inorganic particles.

ENCAPSULATION OF INORGANIC NANOPARTICLES

73

The aim of this chapter is to present a review on the encapsulation of different kinds of inorganic nanoparticles by miniemulsion polymerization of hydrophobic monomers. Special attention will be paid to the encapsulation of silica and magnetite nanoparticles by miniemulsion polymerization using different surface modifiers or coupling agents, because the morphology of the hybrid nanocolloids is strongly affected by them.

4.2 Miniemulsion Polymerization in the Presence of Inorganic Nanoparticles In the last ten years, several reviews highlighted miniemulsion polymerization over other heterophase processes as a powerful technique to encapsulate inorganic solids successfully [6-15]. For the encapsulation of inorganic nanoparticles by miniemulsion, the inorganic/polymer interface as well as the polymer/water interface has to be carefully adjusted in order to obtain encapsulation as a thermodynamically favored state. The design of the interfaces is mainly dictated by the use of two surfactant systems, which govern the interfacial tensions, as well as by the use of appropriate functional comonomers, initiators, or co-stabilizers. The sum of all interfaces has to be minimized. In general, the most useful process appeared in literature to encapsulate hydrophilic inorganic nanoparticles by miniemulsion polymerization is shown in Figure 4.1 and consists of four steps: - S t e p 1: Hydrophobization of the hydrophilic inorganic nanoparticles. - Step 2: Dispersion and stabilization of hydrophobized inorganic nanoparticles in monomer phase. - Step 3: Miniemulsification of the lipophilic dispersion in water. - Step 4: Polymerization of droplets. 4.2.1

H y d r o p h o b i z a t i o n of I n o r g a n i c N a n o p a r t i c l e s

For the encapsulation of hydrophilic inorganic nanoparticles, the surfaces have to be converted into more hydrophobic by functionalization with a surfactant having a low HLB (hydrophilic/

74

MINIEMULSION POLYMERIZATION TECHNOLOGY

Figure 4.1. Encapsulation of inorganic nanoparticles by miniemulsion polymerization.

lipophilic balance) value, with coupling agents or surface modifiers. Apart from the enhancement of the surface hydrophobicity of the inorganic particles, in some cases different reactive groups are introduced for polymerization with the hydrophobic monomer. The type of surface modifier or coupling agent used depends on the type of inorganic nanoparticle to encapsulate. For example, calcium carbonate (CaC0 3 ) can be successfully encapsulated into polymer particles using stearic acid as hydrophobizing agent [16]. Erdem et al. [17-20] used OLOA 370, a polybutene succinimide pentamide, for a successful dispersion of titanium dioxide (Ti02) nanoparticles into styrene and cyclohexane. However, when the organic phase was a mixture of styrene and n-butylacrylate, Al-Ghamdi et al. [21] found that Solsperse 32000, a polyamine/polyester comb polymer with several anchor and several tails connected together in one molecule, was the best stabilizer for Ti0 2 . In the case of yttrium oxysulfide (Y202S) phosphorescent nanoparticles, the polymeric dispersant Solsperse 24000 improved Y 2 0 2 S dispersion and stability in styrene and methyl methacrylate monomers [22]. Fluorescent CdS/ZnS-coated CdSe or CdS quantum dots could be functionalized with a trialkylphosphine modified with an atom transfer radical polymerization (ATRP) initiator (i.e., 2-chloropropionate) [23], trioctylphosphine oxide (TOPO) [24,25], and 4-mercaptovinylbenzene [25] to be dispersed and stabilized in the organic phase previous to carry out the miniemulsion process. Zinc oxide (ZnO) nanoparticles were coated with 3-(trimethoxysilyl)propylmethacrylate (TPM) [26]

ENCAPSULATION OF INORGANIC NANOPARTICLES

75

and 3-aminopropyltriethoxysilane (APTS) [27] before dispersing them in styrene and styrene/n-butylacrylate, respectively. Silver (Ag) nanoparticles were made hydrophobic by dodecanethiol [28], and nano-alumina (A1203) with oleic acid (OA) [29]. On the other hand, in literature the surface modification of silica (Si02) and magnetite (Fe304) nanoparticles is studied in detail and several coupling agents are proposed for their hydrophobization. In the case of Si0 2 , silane coupling agents, such as TPM, 3-aminopropyltriethoxysilane (APTS), 3-glycidoxypropyltrimothoxysilane (GPM), or 3-mercaptopropyltrimethoxysilane (MPM) are commonly used [30-33], whereas for Fe 3 0 4 , fatty acids, such as oleic acid (OA), oleoyl sarkosine acid (OSA) or stearic acid are extensively used [34-37].

4.2.2 Dispersion of Hydrophobized Inorganic Nanoparticles in Monomer Phase In this step, a stable dispersion of inorganic nanoparticles in the organic phase is required. For that purpose, the previously hydrophobized inorganic nanoparticles are mixed with the monomer and the hydrophobe to polymerize and to suppress the Ostwald ripening effect, respectively. The stability of inorganic nanoparticles can be described by the well-known DLVO theory of electrostatic stabilization [38,39]. However, the primary minimum of the inorganic nanoparticle stabilization in nonaqueous media means steric stabilization. This arises from the adsoption of the anchoring groups of a stabilizer molecule onto the surface of each particle while the rest of the molecule, solvated by the medium, provides steric stabilization [40]. Thus, the key factor in this step is the nature and surface density of the surface modifier or coupling agent used.

4.2.3

Miniemulsification of the Lipophilic Dispersion in Water

The stable dispersion of inorganic nanoparticles in the monomer is miniemulsified in the water phase, employing a surfactant with high HLB which has a higher tendency to stabilize the monomer (polymer)/water interface. For that, the mixture is sonicated up to a stable miniemulsion is achieved. The stability of the miniemulsion droplets arises from the use of an ionic surfactant coupled with a low molecular weight, highly water-insoluble co-stabilizer (hydrophobe),

76

MINIEMULSION POLYMERIZATION TECHNOLOGY

which substantially retards diffusion of monomer out of the miniemulsion droplets (Otswald ripening effect) [9,10]. However, a successful encapsulation of inorganic nanoparticles via miniemulsion polymerization requires these nanoparticles to be substantially smaller than the monomer droplets generated [18,19].

4.2.4 Polymerization of Droplets The main difficulty in the process of miniemulsion polymerization is to avoid micellar nucleation and reduce homogeneous nucleation in order to ensure monomer droplet nucleation. In the case of encapsulation of inorganic nanoparticles, micellar and homogeneous nucleation will result in the formation of pure polymer particles decreasing encapsulation degree. Once a stable miniemulsion containing inorganic nanoparticles inside monomer droplets is obtained, adding the initiator and raising the temperature then start the polymerization. It is well known that in miniemulsion polymerization the nucleation occurs mainly inside monomer droplets. When inorganic nanoparticles are hydrophobic and dispersed directly in monomer droplets before polymerization, they will be encapsulated directly and do not need to diffuse into micelles; thus, the mass transfer of such large amount (in volume) of inorganic nanoparticles is avoided [9,10].

4.3

Encapsulation of Silica Nanoparticles

Monodisperse colloidal silica particles having diameters in the range 50-700 nm can be conveniently prepared by hydrolysis of tetra alkyl orthosilicates according to Stöber 's method [41]. Silica can be used in a wide range of colloidal products, ranging from paints and magnetic fluids to high-quality paper coatings because it is chemically inert and optically transparent. In addition, silica nanoparticles have also great potentials in biomédical applications due to good compatibility, resistance to decomposition in vivo and the presence of surface silanol groups. These groups can easily react with alcohols and silane coupling agents to produce dispersions that are not only stable in non-aqueous solvents but also provide ideal anchorage for covalent bonding of specific ligands [42-44]. Therefore, the encapsulation of silica nanoparticles by hydrophobic polymers allows the preparation of well-defined organic/inorganic

ENCAPSULATION OF INORGANIC NANOPARTICLES

77

hybrids nanomaterials with properties suitable for use in a variety of sensing or biomédical applications in addition to preparing materials suitable for reinforcing organic polymers.

4.3.1.

M i n i e m u l s i o n Polymerization with Hydrophilic Silica Nanoparticles

In a first approach, Tiarks et al. [45] carried out the polymerization of styrene (St) by a miniemulsion process in the presence of a hydrophobe (hexadecane), a coupling monomer (4-vinylpyridine) and bare silica nanoparticles of 22 nm, using 2,2'-azobis isobutironitrile (AIBN) as initiator. Depending on the reaction conditions and the surfactants employed (nonionic, anionic or cationic surfactant), different hybrid morphologies were obtained, comprising a raspberrylike morphology where the silica surrounds the latex particles and provides stabilization even without any low molecular weight surfactant. This morphology was obtained because under alkaline conditions 4-vinylpyridine provides a strong acid-base interaction with silica. At other compositions, non-coupled structures as well as completely encapsulation of silica nanoparticles in the polymer particle were obtained. The addition of the nonionic surfactant Lutensol AT50, due to the specific interaction of silica surfaces with poly(ethylene glycol), was counterproductive and worsens the particle size distribution and hybrid definition. The anionic surfactant sodium dodecyl sulfate (SDS), when applied under basic conditions, assisted the colloidal stability but did not really improve incorporation of silica nanoparticles, which was presumably due to the negative charge of both species. The cationic surfactant cetyltrimethylammonium chloride (CTMA) led to an improvement of the coupling between latex and silica nanoparticles. Using very large amounts of CTMA to hydrophobize the complete silica surface led to the encapsulation of silica within the latex particles. However, the non-homogeneity of both particles size and silica loading observed in this morphology, suggests that the colloidal stabilization of the silica nanoparticles within the monomer droplets is still to be improved. In more recent papers [46,47], raspberry-like composite microspheres with polystyrene (PSt) as cores and silica nanoparticles of 20 nm as shell were prepared through miniemulsion polymerization by using SDS as surfactant and 1-vinylimidazole [46] or 2-(methacryloyl)ethyltrimethylammonium chloride [47] as auxiliary cationic monomer. The strong acid-base interaction between acidic

78

MINIEMULSION POLYMERIZATION TECHNOLOGY

hydroxyl groups of the silica surface and basic amino groups of the cationic monomer promote the formation long-term stable PSt/Si0 2 nanocomposites with raspberry-like morphology, but no encapsulation was observed.

4.3.2

Miniemulsion Polymerization with Surface-Modified Silica Nanoparticles

It was found in literature that the most common silane coupling agent used to modify the surface of silica nanoparticles previous to encapsulate them by miniemulsion polymerization is TPM [48-54]. The surface silylation with TPM provides silica nanoparticles with an organic layer, thus achieving a considerable organophilation and compatibility with hydrophobic monomers. In addition, TPM bears a polymerizable double bound, acting as a comonomer in the polymerization of vinyl monomers. Zhang et al. [48] studied the encapsulation of TPM-modified silica nanoparticles by the miniemulsion polymerization of St using SDS as surfactant, hexadecane as hydrophobe and potasium persulfate (KPS) as initiator. They found that the size and morphology of the nanocomposite particles could be tuned by adjusting the silica particle size and surfactant concentration. For 45 nm silica nanoparticles, the size of the nanocomposite decreased from 200 to 80 nm with increasing surfactant concentration from 20 to 40 mM, and the number of silica nanoparticles encapsulated into each polymer particle gradually decreased and finally formed core-shell morphology (see TEM microphotographs in Figure 4.2). For 90 nm silica nanoparticles, the size of the nanocomposite particles also decreased from 180 to 130 nm with increasing surfactant concentration from 20 to 40 mM, but the core-shell morphology was kept unchanged. For 200 nm silica nanoparticles, some "raspberry-like" morphology was observed (see TEM microphotographs in Figure 4.3). In addition, they studied the effect of surface chemistry of the silica nanoparticles on the size and morphology of the nanocomposites particles by using two types of hydrofobized silica having 3.6 and 8.7 TPM molecules per square nanometer. The graft density of TPM on silica surface had almost no influence on the size and morphology of the nanocomposite nanoparticles. This behavior was completely different from that of an emulsion polymerization due to the different nucleation mechanism. In emulsion polymerization, the surface chemistry of the silica nanoparticles, especially the density

ENCAPSULATION OF INORGANIC NANOPARTICLES

79

Figure 4.2. TEM microphotographs of Si0 2 /polystyrene composite particles obtained using different SDS concentrations in the presence of 45 nm silica particles: (a) 20 mM; (b, c) 30 mM; (d) 40 mM. (Reprinted from Ref. 48 with permission from the American Chemical Society).

of TPM grafted on the silica surface, strongly influenced the final morphology of the nanocomposite particles [55]. Zhou et al. [49] found that the use of a soft monomer such as n-butyl acrylate (BA) also affected the encapsulation of silica nanoparticles by miniemulsion polymerization. The incorporation of BA together with St into the polymer was propitious for causing a multicore-shell morphology in comparison with only having a hard monomer (St) in the polymer, other parameters being equal. This multicore-shell structure was formed because BA and its polymer have much lower surface free energies than St and its polymer,

80

MINIEMULSION POLYMERIZATION TECHNOLOGY

Figure 4.3. TEM microphotographs of Si0 2 /polystyrene composite particles obtained using 20 mM SDS in the presence of silica particles with different sizes: (a) 45 nm; (b) 90 nm; (c) 200 nm. (Reprinted from Ref. 48 with permission from the American Chemical Society).

incorporating BA monomer should decrease the surface energy of the miniemulsion, increasing the droplet size of the miniemulsion. In addition, the TPM-modified silica nanoparticles had a surface structure similar to that of the BA molecules, which also helped in lodging more silica nanoparticles in each droplet. On the other hand, increasing the particle size of the silica or decreasing the surfactant content tended to form a normal core-shell or even raspberry-like structure. Furthermore, the encapsulation of TPM-modified silica nanoparticles by miniemulsion polymerization was also successful using a mixture of methyl methacrylate (MMA) and BA [51] and a mixture of St and 2-hydroxyethylmethacrylate (HEMA), or

ENCAPSULATION OF INORGANIC NANOPARTICLES

81

styrene sulfonic acid (SSA), or aminoethyl methacrylate hydrochloride (AEMH) [53]. Recently, Costoyas et al. [54] synthesized core-shell hybrid nanoparticles having narrow particle size distributions (PSDs) as well as a high degree of silica encapsulation by miniemulsion polymerization of St using SDS as surfactant, hexadecane as hydrophobe, and KPS as initiator. In this work, to enhance the hydrophobicity of the silica nanoparticles, apart from using TPM, another surface modifier was added simultaneously, i.e. oleic acid (OA). OA can be bonded to the silanol groups present at surface of silica nanoparticles by a simple hydrogen bond, and the double bonds are able to polymerize with a vinyl monomer. A synergistic effect was observed using TPM together with OA in the compatibilization step between the organic phase (monomer) and silica nanoparticles. In addition, to optimize the production of monodisperse hybrid particles, they studied the effect of the size of silica nanoparticles, the ratio styrene/silica, the surfactant concentration, and the presence of ethanol in the reaction. Increasing the size of the silica nanoparticles (66,107, and 202 nm) larger particle sizes of hybrid latexes (89,95, and 155 nm) were obtained. However, only the size of the latex with the smallest silica showed fully encapsulation. The decrease in St concentration a n d / or the increase in surfactant concentration caused smaller particle diameters and lower encapsulation degree. In addition, when the miniemulsion polymerization was carried out without ethanol a larger particle size and a higher conversion were achieved than those when ethanol was used. Figure 4.4 shows the TEM microphotographs of the latex obtained with ethanol (MP8), and without ethanol (MP11). As can be seen, latex particles prepared in the presence of alcohol (MP8, Figure 4.4.a) had a smaller particle diameter than those obtained without alcohol (MP11, Figure 4.4.b). The most of the hybrid particles of latex MP11 had one silica nucleus, finding some hybrid particles with 2 or 3 nuclei (Figure 4.4.c). In addition, the cycles of centrifugation and re-dispersion were effective and no pure polymer particles were observed in latex MP11H. A more challenging encapsulation of silica nanoparticles by miniemulsion polymerization used silane-coupling agents, which are able to initiate a controlled /living radical polymerization of hydrophobic monomer droplets. Alkoxamine initiators based on N-tert-butyl-1diethylphosphono-2,2-dimethylpropyl nitroxide (DEPN) and carrying a terminal functional group were synthesized in situ and grafted to silica surface by Bailly et al. [56]. The resulting grafted alkoxamines

82

MINIEMULSION POLYMERIZATION TECHNOLOGY

Figure 4.4. TEM microphotographs of latexes MP8 (a); MP11 (b); and hybrid particles of latex MP11 (after various cycles of centrifugation-redispersion) MP11H (c). (Reprinted from Ref. 54 with permission from Wiley).

were employed to initiate, through nitroxide-mediated polymerization (NMP), the growth of polystyrene chains from the silica surface. Then, the PSt-grafted silica nanoparticles were entrapped inside latex particles via miniemulsion polymerization forming silica-polystyrene core-shell particles. On the other hand, Bombalski et al. [57] reported the efficient synthesis of hybrid organic/inorganic nanoparticles using silica nanoparticles with surface tethered initiators an activators generated by electron transfer (AGET) ATRP miniemulsion process of BA. The surface-modified silica was prepared by reacting l-(chlorodimethylsilyl)propyl 2-bromoisobutyrate with hydroxyl groups on silica particle surface. In comparison to the bulk polymerization, using the same stoichiometry, miniemulsion allowed the preparation of hybrid materials with a higher yield, i.e., higher monomer conversion, and a higher polymerization rate without macroscopic gelation. Direct visualization by AFM provided additional evidence for the formation of wellcontrolled hybrids.

ENCAPSULATION OF INORGANIC NANOPARTICLES

83

4.3.3 Miniemulsion Polymerization with Locally Surface-Modified Silica Nanoparticles In the last years, some works presented the synthesis of asymmetric nanocomposites particles of polystyrene and silica via conventional (with a chemical initiator) [58,59] or radiation miniemulsion polymerization [60]. The key to obtain the asymmetric nanocomposite particle pairs was the combination of miniemulsion polymerization and the local surface modification of silica nanoparticles. Because of localized surface modification on the silica surface by using silane-coupling agents, such as n-octadecyltrimethoxysilane (ODMS) [58,59] or TPM [60], the nucleation and formation of the polymer nodule in miniemulsion polymerization took place only in the modified area on silica surface, thus ensuring the asymmetry morphology. In addition, controlling the monomer/silica weight ratio, different morphological anisotropic PSt/silica hybrid particles were prepared (see Figure 4.5). The morphology varied from mushroom-like, via hollow egg-like, to bowl-like structures with increasing monomer/silica weight ratio. This method offers an effective and feasible way to synthesize morphologically controllable polymer/inorganic anisotropic hybrid nanoparticles, which because to the existence of different surface functionalities, could have a wide range of potential applications in bioscience and molecular recognition. Polystyrene/silica hybrid asymmetric dimer particles were also synthesized by miniemulsion polymerization in one step [61]. In this case, a solution of St monomer with dissolved AIBN was added to tetraethyl orthosilicate (TEOS), which is the precursor of silica nanoparticles, TPM, and hexadecane to form an oil phase. On the other hand, cetyltrimethylammonium bromide (CTAB) dissolved in water was employed as the aqueous phase. The mixture of oil and water phase was sonicated and miniemulsion was carried out. After the polymerization of styrene, ammonia was added to the dispersion to form silica nanoparticles. Phase separation of the organic moiety and the inorganic moiety resulted from temperature raising and polymerization, and it was enhanced after the formation of polystyrene and silica due to the hydrophobicity of polystyrene and hydrophilicity of silica. In addition, the size of polystyrene particles in asymmetric dimer particles was adjusted easily either by changing the weight ratio of St/TEOS or by altering the sonication power during the miniemulsion preparation.

84

MINIEMULSION POLYMERIZATION TECHNOLOGY

Figure 4.5. TEM (a-e) and SEM (a'-e') images of anisotropic PS/silica hybrid particles synthesized in the presence of locally surface-modified silica nanoparticles. The weight ratio of monomer/silica was increased from a to e and a'to e'; (a) and (a'): 28 : 1, (b) and (b'): 60 : 1, (c) and (c'): 72 : 1 , (d) and (d'): 80 : 1 , and (e) and (e'): 100 :1.). (Reprinted from Ref. 60 with permission from the Royal Society of Chemistry).

ENCAPSULATION OF INORGANIC NANOPARTICLES

85

4.4 Encapsulation of Magnetite Nanoparticles Magnetic fluids or ferrofluids can be classified according to the hydrophilicity-hydrophobicity of the stabilizers used as hydrophilic and hydrophobic ferrofluids. They are stable dispersions of ultrafine magnetic particles in an organic or aqueous carrier medium. The stabilization of these ultrafine magnetic particles can be achieved by adsorption of stabilizers to hinder particles flocculation and sedimentation. Magnetite, Fe 3 0 4 , is the most common magnetic material used, and it is usually prepared by chemical co-precipitation of an aqueous of Fe 3+ /Fe 2+ solution (Fe3+: Fe2+ = 2:l(mol)) with a base in the presence of a stabilizer or surface modifier, such as oleic acid (OA), the most common used one. Magnetic polymer nanoparticles (MPNPs) have extensive applications in biomédical [62,63], bioengineering and biotechnology fields, such as cell separation [64], immunoassays [65], nucleic acid purification [66], DNA separation [67], enzyme immobilization [68], magnetic resonance imaging [69], and hyperthermia [70]. MPNPs exhibit high magnetic susceptibility to an external magnetic field and in addition, they are easily further functionalized and surface-modified by the attachment of various bioactive molecules [71]. In all the cases, MPNPs should fulfill certain criteria to fit further biotechnological applications [72], such as: no sedimentation, uniform size and size distribution, high and uniform magnetic content, superparamagnetic behavior, no toxicity, and no iron leaking. In general, however, the polymerization process for the encapsulation of magnetite may produce three possible types of particles in the resulting magnetic latex, i.e., magnetic polymeric nanoparticles (MPNPs; with magnetite encapsulated inside), pure polymer nanoparticles (PPNPs; without magnetite inside), and bare (free) magnetite nanoparticles (BMNPs; without polymer coating). The existence of PPNPs is not desirable because the magnetic properties of the composite particles will be decreased, and the existence of BMNPs is also not desired because they cannot be further functionalized due to their bare surfaces without polymer surrounding. In addition, particle size distribution (PSD) should be as narrow as possible so that MPNPs can respond to an external magnetic field as uniformly as possible.

86

MINIEMULSION POLYMERIZATION TECHNOLOGY

4.4.1 Encapsulation of Magnetite by a Single Miniemulsion Polymerization Process Most of the works found in literature devoted to encapsulate magnetite nanoparticles [73-98] proposed the single miniemulsion process consisting in four steps shown in Figure 4.1. As aforementioned, the successful incorporation of hydrophilic magnetite nanoparticles into hydrophobic polymer particles by miniemulsion polymerization relies on their surface modification to make magnetite/polymer compatible. Different surface modifiers have been used to obtain stable and hydrophobic magnetic nanoparticles, being oleic acid (OA) the most common one [75-84,87,92-97]. The carboxylate head of OA is able to anchor on the surface of iron oxide nanoparticles, while its hydrophobic tail ensures steric stabilization as well as compatibility with the solvent [34-37]. Oleoyl sarcosine acid (OSA) [73,95], bis(2-ethylhexyl) sulfosuccinate (AOT) [74], TPM [85,95], sorbitan oléate (Span-80) [86,95,96], alkylolammonium salts of low molecular weight polycarboxylic acid polymer (Disperbik-106, Disperbik-108, and Disperbik-111) [87,88], Sipomer PAM200, which is a phosphate-based poly(propylene glycol) methacrylate [89], lauric acid [90,91], stearic acid [95], and 12-hexanoyloxy-9-octadecenoic acid (HOA) [98] have also been used. However, most of them caused (i) inhomogeneous distribution of the magnetic nanoparticles inside and among the particles and / o r (ii) PPNPs a n d / o r (iii) BMNPs or magnetic aggregates in the aqueous phase a n d / o r (iv) broad PSD, a n d / o r (v) limited loading of the particles with magnetic material. With respect to the monomers, St has been the most common one used [73,74,76,78,80-98], but acrylates [75,77,79,8591,96] such as MMA and BA have also been successfully used. In addition, various surfactants, and hydrophobes have also been tested, being the most common ones SDS and hexadecane, respectively. The use of water-soluble or oil-soluble initiator affects greatly the morphology of the final composite. Mori et al. [84] observed that the size of monomer droplets/polymer particles increased from 60 to 300 nm during polymerization, keeping magnetite in core when potassium persulfate (KPS) or ammonium persulfate (APS) was used as the sole water-soluble initiator. In contrast, when 2,2'-azobis isobutyronitrile (AIBN) was used as the oil-soluble initiator, the size of the droplets/particles was retained to be 90 nm at the most and magnetite nanoparticles located at the surface of polystyrene particles.

ENCAPSULATION OF INORGANIC NANOPARTICLES

87

The effect of initiator on particle size in persulfate system was likely originated from the decrease of pH value and the increase of ionic strength, which induced the fusion of droplets/particles containing magnetite. When both AIBN and KPS were used, magnetite distributed randomly in polystyrene particles and the size of particles was around 200 nm. Mixed initiator systems employed to improve conversion and encapsulation of magnetite resulted in the formation with middle characteristics between those of sole initiator systems. On the other hand, recently ultrasound initiated miniemulsion polymerization was employed to prepare stable MPNPs [95-98] due to its many advantages, such as high polymerization rate, free chemical initiator, low surfactant concentration and low reaction temperature. It is well known that the challenges in almost all the strategies involving a single miniemulsion polymerization process to prepare MPNPs are to minimize, even eliminate, the formation of PPNPs and BMNPs during the preparation stage with an even greater challenge in obtaining MPNPs with very narrow PSD. However, to date these important issues have not been analyzed and studied in detail in literature. With the aim of optimizing the encapsulation of magnetite, Lu and Forcada [82] prepared MPNPs by miniemulsion polymerization in the presence of oil-based St ferrofluid with hexadecane as hydrophobe, AIBN as initiator and SDS as surfactant. Methacrylic acid (MAA) was used as comonomer, and hydroxyethyl cellulose (HEC) and polyvinylpyrrolidone (PVP) were used as aid stabilizers subsequently. Reducing the amount of SDS improved the Fe 3 0 4 encapsulation degree, narrowed the PSD, and reduced the number of BMNPs and PPNPs at the same time. The optimum percentage of SDS, based on the total amount of St and Fe 3 0 4 , was 2-3%. Increasing the amount of hexadecane improved the Fe 3 0 4 encapsulation degree, sharpened the PSD, and reduced the number of BMNPs and PPNPs. Increasing the amount of Fe 3 0 4 increased the difficulty of the encapsulation of Fe 3 0 4 . The number of BMNPs increased, and the PSD became much broader when the concentration of Fe 3 0 4 increased. The optimum concentration of Fe 3 0 4 was 10%. The use of MAA as a comonomer facilitated the encapsulation of Fe 3 0 4 , but at the cost of broadening the PSD. The particle size of the MPNPs was larger than that of the reactions carried out without MAA. HEC in the recipe assisted the encapsulation of Fe 3 0 4 and also improved the PSD of MPNPs. The most suitable concentration of HEC for use as a stabilizer aid was 2% together with 15% of

88

MINIEMULSION POLYMERIZATION TECHNOLOGY

hexadecane. PVP improved the encapsulation degree of Fe 3 0 4 and reduced the polydispersity of MPNPs. However, in the next work of Lu et al. [83] fairly monodisperse selfstabilized magnetic polymeric composite nanoparticles (SS-MPCPs) were prepared by surfactant-free miniemulsion polymerization using St as main monomer, sodium p-styrene sulfonate (NaSS) as ionic comonomer, hexadecane as hydrophobe, and AIBN as initiator in the presence of oleic acid-coated magnetite particles. TEM microphotographs of the SS-MPCPs synthesized with different amounts of NaSS are shown in Figure 4.6. As can be seen, the amount of NaSS had a notable effect on the magnetite encapsulation degree, particle size, and PSDs of the SS-MPCPs. From 5% to 15% of NaSS the encapsulation degree was increased, and some BMNPs but no PPNPs were observed. At 20% NaSS, the encapsulation of magnetite was successful, and neither BMNPs nor PPNPs were observed. The absence of PPNPs formed during the polymerization process indicated that monomer droplet nucleation was achieved entirely by using an emulsifier-free miniemulsion polymerization technique. Micellar nucleation was avoided completely in the absence of emulsifier, and homogeneous nucleation was also prevented. However, when the concentration of NaSS was 25% a significant number of PPNPs were observed. At this concentration, the amount of oligomers formed in the aqueous phase was enough to stabilize the particles generated by homogeneous nucleation and PPNPs were produced. On the other hand, the distribution of magnetite inside composite particles is an important factor, which can influence the magnetic characteristics of composite particles. In all the SS-MPCPs synthesized the magnetite particles are mainly in the core of the composite particles. Due to the ionic character of NaSS, the copolymerization of St with NaSS increases the hydrophihcity of the polymer formed and enhances the hydrophobicity of oleic acidcoated magnetite particles. Thus, magnetite particles are mainly in the core of SS-MPCPs. In addition, a few large composite particles in the magnetic latexes were observed. The distribution of magnetite particles inside the largest composite particles was completely different to that of the smallest ones. This means that the formation of this kind of particles was due to aggregation between some of the particles formed. Besides, it can be seen that the number of magnetite particles encapsulated inside each magnetic particle was not the same.

ENCAPSULATION OF INORGANIC NANOPARTICLES

89

Figure 4.6. TEM microphotographs of SS-MPCPs prepared with different amounts of NaSS (Reprinted from Ref. 83 with permission from the American Chemical Society).

4.4.2 Encapsulation of Magnetite by a Double Miniemulsion Polymerization Process Using the single miniemulsion polymerization process detailed above an inhomogeneous distribution of the magnetite in the polymer nanoparticles together with a limited magnetite content in the

90

MINIEMULSION POLYMERIZATION TECHNOLOGY

polymer matrix (15-20%) is obtained. In order to increase the magnetic content in the polymer particles another process was developed by Landfester et al. [99-101]. They encapsulated high amounts of magnetite particles into polystyrene particles by a three-step preparation route including two miniemulsion processes. In the first step, oleic acid coated magnetic nanoparticles in octane were prepared. In the second step, a dispersion of magnetite in octane was miniemulsified in water by using SDS as surfactant. After evaporation of the octane, the magnetite aggregates, which were covered by an OA/SDS bilayer, were mixed with a St miniemulsion and in the third step of the synthesis route; an ad-miniemulsification process was used to obtain final and full encapsulation. Here, a fusion/fission process induced by ultrasound was just effective for the monomer droplets, whereas the monomer coated magnetite aggregates stayed intact. That way, all monomer droplets were split and hetero-nucleated onto the magnetite aggregates to form a monomer film. After polymerization, polymer encapsulated magnetite aggregates were obtained, and they found that up to 40% magnetite could be encapsulated, resulting in particles with a high homogeneity of the magnetite content. On the other hand, magnetite/polystyrene latexes with narrow size distribution and high magnetite content were also prepared by a hybrid miniemulsion polymerization process containing binary droplets [102-104]. First of all, magnetite nanoparticles modified with oleic acid were synthesized and dispersed in octane. This ferrofluid was added to an aqueous solution with SDS as surfactant and treated ultrasonically to obtain miniemulsion droplets composed of magnetite nanoparticles aggregations with a diameter of 100-200 nm (Mag-droplets). Another miniemulsion made of St monomer droplets with a diameter of 3-4 mm (St-droplets) was prepared by membrane emulsification equipment and mixed with Mag-droplets to obtain a double-miniemulsion system, which contained microsized St droplets and nanosized magnetite aggregation droplets. With extremely low surfactants concentration, the nucleated loci were selectively controlled in the Mag-droplets, as the result of smaller droplet size and larger surface ratio. Both water soluble KPS and oil-soluble AIBN was adopted to initiate the polymerization. When AIBN was used, the obtained magnetic polystyrene latex (MPL) exhibited spherical shape. Nevertheless, microphase separation still occurred in the interior of MPL. However, in the presence of KPS, MPLs with particle size of 60-200 nm, narrow size distribution, and

ENCAPSULATION OF INORGANIC NANOPARTICLES

91

Figure 4.7 TEM images, at different magnifications, of MPLs obtained using KPS as initiator. (Reprinted from Ref. 103 with permission from Wiley).

high magnetite content (86 wt%) were attained successfully (see Figure 4.7). The magnetite content of as-synthesized MPLs depends on the volume ratio of Mag-droplets and St-droplets. In addition, without hexadecane in the St-droplets, the achieved MPLs have wide size distribution because of fast monomer diffusion during the polymerization process.

4.5

Conclusions and Future Perspectives

In this chapter, the incorporation of inorganic nanoparticles into miniemulsion polymerizations to give hybrid nanoparticles is revised. It can be concluded that the use of the right combination of surface modifiers or coupling agents together with the adequate miniemulsion process (single or double) are the key factors for attaining a high encapsulation degree of the inorganic nanoparticles together with a well-controlled morphology of the hybrid nanocolloids. Up to now, most of reports present the synthesis of polymerinorganic nanoparticles using only one type of inorganic nanoparticle. However, the use of two populations of inorganic nanoparticles opens new routes to the production of multilayer or multimodal hybrid nanoparticles [50,52,102,105]. Recently, van Berkel et al. [105] presented a convenient and highly modular co-encapsulation method by miniemulsion polymerization for the preparation of multimodal composite nanoparticles, consisting of a spherical,

92

MINIEMULSION POLYMERIZATION TECHNOLOGY

cross-linked poly(diviny lbenzene) matrix imbibed with two different types of inorganic core materials (MnFe 2 0 4 and Au nanoparticles). These particles exhibit a combination of properties characteristic of the incorporated inorganic material, in this case, magnetism due to MnFe 2 0 4 nanoparticles and UV-Vis absorption due to the surface plasmon resonance of Au nanoparticles. First, MnFe 2 0 4 and Au nanoparticles grafted with short polystyrene ligands were dispersed in the monomer, divinylbenzene. This monomer/nanoparticle solution was then emulsified with an aqueous solution of CTAB surfactant, via ultrasonication, to generate a miniemulsion of submicrometer monomer droplets containing the inorganic nanoparticles. The free radical polymerization of these droplets (using 2,2'-azobis(2-amidinopropane) dihydrochloride as initiator) yields the co-encapsulation of MnFe 2 0 4 and Au nanoparticles.

4.6

Acknowledgements

This work has been supported by the Spanish Programa Nacional de Materiales (MAT 2009-13155-C04-01). J. Ramos acknowledges the financial support by the Ministerio de Ciencia e Innovación: Subprograma Juan de la Cierva (JCI-2008-2217).

References 1. G. Schottner, Chemistry of Materials, Vol. 13, p. 3422, 2001. 2. C. Sanchez, G.J.deA.A. Soler-Illia, F. Ribot, T. Lalot, C.R. Mayer, and V. Cabuil, Chemistry of Materials, Vol. 13, p. 3061, 2001. 3. F Caruso, Advanced Materials, Vol. 13, p. 11,2001. 4. V.N. Manoharan, M.T. Elsesser, and D.J. Pine, Science, Vol. 301, p. 483, 2003. 5. Z. Yang, Z. Niu, Y. Lu, Z. Hu, and C.C. Han, Angewandte Chemie, International Edition, Vol. 42, p. 1943,2003. 6. E. Bourgeat-Lami, Journal of Nanoscience and Nanotechnology, Vol. 2, p. 1, 2002. 7. V. Castelvetro, and C. DeVita, Advances in Colloid and Interface Science, Vol. 108109, p. 167,2004. 8. K. Landfester, Advanced Materials, Vol. 13, p. 765,2001. 9. K. Landfester, Macromolecular Rapid Communications, Vol. 22, p. 896,2001. 10. M. Antonietti, and K. Landfester, Progress in Polymer Science, Vol. 27, p. 689, 2002. 11. J.M. Asua, Progress in Polymer Science, Vol. 27, p. 1283,2002. 12. K. Landfester, Topics in Current Chemistry, Vol. 227, p. 75, 2003.

ENCAPSULATION OF INORGANIC NANOPARTICLES

93

13. M.S. El-Aasser, and E.D. Sudol, Journal of Coatings Technology and Research, Vol. 1, p. 20, 2004. 14. K. Landfester, Annual Reviews in Materials Research, Vol. 36, p. 231,2006. 15. K. Landfester, Angewandte Chemie, International Edition, Vol. 48, p. 4488,2009. 16. N. Betchthold, F. Tiarks, M. Willert, K. Landfester, and M. Antonietti, Macromolecular Symposia, Vol. 151, p. 549, 2000. 17. B. Erdem, E.D. Sudol, V.L. Dimonie, and M.S. El-Aasser, Macromolecular Symposia, Vol. 155, p. 181,2000. 18. B. Erdem, E.D. Sudol, V.L. Dimonie, and M.S. El-Aasser, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 38, p. 4419, 2000. 19. B. Erdem, E.D. Sudol, V.L. Dimonie, and M.S. El-Aasser, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 38, p. 4431, 2000. 20. B. Erdem, E.D. Sudol, V.L. Dimonie, and M.S. El-Aasser, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 38, p. 4441,2000. 21. G.H. Al-Ghamdi, E.D. Sudol, V.L. Dimonie, and M.S. El-Aasser, Journal of Applied Polymer Science, Vol. 101, p. 3479,2006. 22. H. Kim, E.S. Daniels, S. Li, V.K. Mokkapati, and K. Kardos, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 45, p. 1038,2007. 23. A.C.C. Esteves, L. Bombalski, B. Cusick,A. Barros-Timmons, K. Matyjaszewski, and T. Trindade, Polymer Preprints, Vol. 46, p. 134,2005. 24. A.C.C. Esteves, A. Barros-Timmons, T. Monteiro, and T. Trindade, Journal of Nanoscience and Nanotechnology, Vol. 5, p. 766,2005. 25. N. Jourmaa, M. Lansalot, A. Théretz, and A. Elaissari, Langmuir, Vol. 22, p. 1810,2006. 26. J.J. Zhang, G. Gao, M. Zhang, D. Zhang, C.L. Wang, D.C. Zhao, and F.Q. Liu, Journal of Colloid Interface Science, Vol. 301, p. 78,2006. 27. E. Tang, and S. Dong, Colloid and Polymer Science, Vol. 287, p. 1025,2009. 28. H.D. Koh, M. Changez, J.P. Lee, and J.S. Lee, Macromolecular Rapid Communications, Vol. 30, p. 1583,2009. 29. A.R. Mahdavian, Y. Sarrafi, and M. Shabankareh, Polymer Bulletin, Vol. 63, p. 329, 2009. 30. A.R Philipse, and A. Vrij, Journal of Colloid and Interface Science, Vol. 128, p. 121.1989. 31. J.W. Goodwin, R.S. Harbron, and P.A. Reynols, Colloid and Polymer Science, Vol. 268, p. 766,1990. 32. R.D. Badley, W.T. Ford, FJ. McEnroe, and R.A. Assink, Langmuir, Vol. 6, p. 792.1990. 33. A. Van Blaaderen, and A. Vrij, Journal of Colloid and Interface Science, Vol. 156, p. 1,1993. 34. A. Wooding, M. Kilner, and D.B. Lambrick, Journal of Colloid and Interface Science, Vol. 144, p. 236,1991. 35. L. Shen, P.E. Laibinis, and T.A. Hatton, Langmuir, Vol. 15, p. 447,1999. 36. S. Sun, and H. Zeng, Journal of the American Chemical Society, Vol. 124, p. 8204, 2002. 37. M. Lattuada, and T.A. Hatton, Langmuir, Vol. 23, p. 2158,2007. 38. R.J. Hunter, Foundations of Colloid Science, Oxford, Clarendon Press, 1987. 39. S. Ross and I.D. Morrison, Colloidal Systems and Interfaces, New York, Wiley, 1988.

94

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

40. D.H. Napper, Polymeric Stabilization of Colloidal Dispersions, New York, Academic, 1983. 41. W. Stöber, A. Fink, and E. Bohn, Journal of Colloid and Interface Science, Vol. 26, p. 62,1968. 42. Z.F. Li, and E. Ruckenstein, Nano Letters, Vol. 4, p. 1463,2004. 43. S. Santra, H. Yang, D. Dutta, J.T. Stanley, P.H. Holloway, W. Tan, B.M. Moudgil, and R.A. Mericle, Chemical Communications, p. 2810,2004. 44. L.M. Rossi, L. Shi, EH. Quina, and Z. Rosenzweig, Langmuir, Vol. 21, p. 4277, 2005. 45. F. Tiarks, K. Landfester, and M. Antonietti, Langmuir, Vol. 17, p. 5775, 2001. 46. Y. Zhang, H. Chen, and Q. Zou, Colloid and Polymer Science, Vol. 287, p. 1221, 2009. 47. Y. Zhang, H. Chen, X. Shu, Q. Zou, and M. Chen, Colloid and Surfaces A: Physicochemical and Engineering Aspects, Vol. 350, p. 26,2009. 48. S.W. Zhang, S.X. Zhou, Y.M. Weng, and L.M. Wu, Langmuir, Vol. 21, p. 2124, 2005. 49. J. Zhou, S.W. Zhang, X.G. Qiao, X.Q. Li, and L.M. Wu, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 44, p. 3202, 2006. 50. J. Zhou, M. Chen, X.G. Qiao, and L.M. Wu, Langmuir, Vol. 22, pp. 1017510179,2006. 51. D.M. Qi, Y.Z. Bao, Z.X. Weng, and Z.M. Huang, Polymer, Vol. 47, p. 4622, 2006. 52. X.G. Qiao, M. Chen, J. Zhou, and L.M. Wu, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 45, p. 1028, 2007. 53. O. Töpfer, and G. Schmidt-Naake, Macromolecular Symposia, Vol. 248, p. 239, 2007. 54. A. Costoyas, J. Ramos, and J. Forcada, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 47, p. 935, 2009. 55. S. Reculusa, C. Mingotaud, E. Bourgeat-Lami, E. Duguet, and S. Ravaine, Nano Letters, Vol. 4, p. 1677,2004. 56. B. Bailly. A.C. Donnenwirth, C. Bartholome, E. Beyou, and E. Bourgeat-Lami, Journal ofNanomaterials, Vol. 2006, p. 1,2006. 57. L. Bombalski, K. Min, H. Dong, C. Tang, and K. Matyjaszewski,Macromolecules, Vol. 40, p. 7429, 2007. 58. W. Qiang, Y Wang, P. He, H. Xu, H. Gu, and D. Shi, Langmuir, Vol. 24, p. 606, 2008. 59. Y. Wang, H. Xu, W. Qiang, H. Gu, and D. Shi, Journal of Nanomaterials, Vol. 2009, p. 1,2009. 60. X. Ge, M. Wang, Q. Yuan, and X. Ge, Chemical Communications, p. 2765,2009. 61. W. Lu, M. Chen, and L. Wu, Journal of Colloid and Interface Science, Vol. 328, p. 98,2008. 62. J. Ugelstad, P. Stenstad, L. Kilaas, W.S. Prestvik, R. Herje, A. Bererge, and E. Homes, Blood Purification, Vol. 11, p. 349,1993. 63. A. Elaissari, R. Veyret, B. Mandrand, and J. Chatterjee, Surfactant Science Series, Colloidal Biomolecules, Biomaterials, and Biomédical Applications, Vol. 116, p. 1,2004. 64. M. Kuhara, H. Takeyama, T. Tanaka, and T. Matsunaga, Analytical Chemistry, Vol. 76, p. 6207, 2004. 65. Y Namba, M. Usami, and O. Suzuki, Analytical Sciences, Vol. 15, p. 1087,1999.

ENCAPSULATION OF INORGANIC NANOPARTICLES

95

66. A. Elaïssari, M. Rodrigue, F. Meunier, and C. Hervé, Journal of Magnetism and Magnetic Materials, Vol. 225, p. 127, 2001. 67. Oster, J. Parker, and L. A. Brassard, Journal of Magnetics and Magnetic Materials, Vol. 225, p. 145,2001. 68. S. Akgol, K. Yasemin, A. Denizli, and M.Y. Anca, Food Chemistry, Vol. 74, p. 281,2001. 69. F.Y. Cheng, C.H. Su, Y.S. Yang, C.S. Yeh, C.Y. Tsai, C.L. Wu, M.T. Wu, and D.B. Shieh, Biomaterials, Vol. 26, p. 729,2005. 70. J. Chatterjee, M. Bettge, Y. Haik, and CJ. Chen, Journal of Magnetics and Magnetic Materials, Vol. 293, p. 303, 2005. 71. X.Q. Liu, Y.P. Guan, Z.Y Ma, and H.Z. Liu, Langmuir, Vol. 20, p. 10278, 2004. 72. K. Landfester, and L.P. Ramirez, Journal of Physics: Condensed Matter, Vol. 15, p. S1345, 2003. 73. D. Hoffmann, K. Landfester, and M. Antonietti, Magnetohydrodynamics, Vol. 37, p. 217, 2001. 74. R. Betancourt-Galindo, R. Saldivar-Guerrero, O.S. Rodríguez-Fernández, L.A. García-Cerda, and J. Matutes-Aquino, Journal of Alloys and Compounds, Vol. 369, p. 87, 2004. 75. X. Liu, Y Guan, Z. Ma, and H. Liu, Langmuir, Vol. 20, p. 10278,2004. 76. I. Csetneli, M.K. Faix, A. Szilágyi, A.L. Kovács, Z. Németh, and M. Zrinyi, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 42, p. 4802,2004. 77. X. Liu, Y. Guan, H. Liu, Z. Ma, Y Yang, and X. Wu, Journal of Magnetics and Magnetic Materials, Vol. 293, p. I l l , 2005. 78. W. Zheng, F Gao, and H. Gu, Journal of Magnetics and Magnetic Materials, Vol. 293, p. 199,2005. 79. J.S. Nunes, C.L. de Vasconcelos, F.A.O. Cabrai, J.H. de Araújo, M.R. Pereira, and J.L.C. Fonseca, Polymer, Vol. 47, p. 7646, 2006. 80. R. Faridi-Majidi, N. Sharifi-Sanjani, and F. Agend, Thin Solid Films, Vol. 515, p. 368.2006. 81. R. Faridi-Majidi, N. Sharifi-Sanjani, and F Agend, Journal of Magnetics and Magnetic Materials, Vol. 311, p. 55,2007. 82. S. Lu, and J. Forcada, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 44, p. 4187,2006. 83. S. Lu, J. Ramos, and J. Forcada, Langmuir, Vol. 23, p. 12893, 2007. 84. Y Mori, and H. Kawaguchi, Colloids and Surfaces B - Biointerfaces, Vol. 56, p. 246.2007. 85. Z. Chen, K. Peng, and Y Mi, Journal of Applied Polymer Science, Vol. 103, p. 3660, 2007. 86. Y Sun, B. Wang, H. Wang, and J. Jiang, Journal of Colloid and Interface Science, Vol. 308, p. 332, 2007. 87. Q. Zhang, H. Zhang, G. Xie, and J. Zhang, Journal of Magnetics and Magnetic Materials, Vol. 311, p. 140, 2007. 88. Q. Zhang, G. Xie, H. Zhang, J. Zhang, and M. He, Journal of Applied Polymer Science, Vol. 105, p. 3525,2007. 89. N. Joumaa, P. Toussay, M. Lansalot, and A. Elaissari, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 46, p. 327,2008. 90. Y.D. Luo, C.A. Dai, and W.Y. Chiu, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 46, p. 1014,2008.

96

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

91. YD. Luo, C.A. Dai, and W.Y Chiu, Journal of Applied Polymer Science, Vol. 112, p. 975, 2009. 92. A.R. Mahdavian, Y. Sehri, and H. Salehi-Mobarakeh, European Polymer Journal, Vol. 44, p. 2482,2008. 93. S. Lu, J. Ramos, and J. Forcada, Macromolecular Symposia, Vol. 281,p. 89,2009. 94. S. Lu, R. Qu, and J. Forcada, Materials Letters, Vol. 63, p. 2539, 2009. 95. G. Qiu, Q. Wang, C. Wang, W. Lau, and Y Guo, Polymer International, Vol. 55, p. 265, 2006. 96. A.R. Mahdavian, M. Ashjari, and H.S. Mobarakeh, Journal of Applied Polymer Science, Vol. 110, p. 1242, 2008. 97. B.M. Teo, F. Chen, T.A. Hatton, F. Grieser, and M. Ashokkumar, Langmuir, Vol. 25, p. 2593, 2009. 98. Z. Qian, Z. Zhang, and Y Chen, Journal of Colloid and Interface Science, Vol. 327, p. 354,2008. 99. L.P Ramirez, and K. Landfester, Macromolecular Chemistry and Physics, Vol. 204, p. 22, 2003. 100. K. Landfester, and L.P. Ramirez, Journal of Physics: Condensed Matter, Vol. 15, p. SI 345, 2003. 101. V Holzapfel, M. Lorenz, C.K. Weiss, H. Schrezenmeier, K. Landfester, and V Mailander, Journal of Physics: Condensed Matter, Vol. 18, p. S2581, 2006. 102. H. Xu, L. Cui, N. Tong, and H. Gu, Journal of the American Chemical Society, Vol. 128, p. 15582, 2006. 103. L. Cui, H. Xu, P. He, K. Sumitomo, Y Yamaguchi, and H. Gu, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 45, p. 5285,2007. 104. T. Gong, D. Yang, J. Hu, W. Yang, C. Wang, and J.Q. Lu, Colloids and Surfaces A: Physicochemical and Engineering Aspects, Vol. 339, p. 232,2009. 105. K.Y. van Berkel, A.M. Piekarski, P.H. Kierstead, E.D. Pressly, PC. Ray, and C.J. Hawker, Macromolecules, Vol. 42, p. 1425,2009.

5

Polymeric Nanocapsules by Interfacial Miniemulsion Polymerization Guo-Rong Shan and Zhi-Hai Cao State Key Laboratory of Chemical Engineering, Institute of Polymerization and Polymer Engineering, Department of Chemical Engineering, Zhejiang University, Hangzhou 310027, China

Abstract A water-soluble co-monomer, N-isopropylacrylamide (NIPAM), and an oil-soluble crosslinker, divinylbenzene (DVB), and alkoxysilane monomer, γ-methacryloxypropyltrimethoxysilane (MPS), have been combined in a system for the synthesis of organic and organic-inorganic hybrid nanocapsules with crosslinked shells through interfacial miniemulsion polymerization by encapsulating a liquid non-sol vating hydrocarbon. Oligomers of poly(N-isopropylacrylamide) (PNIPAM) were dehydrated and separated from the aqueous phase and were adsorbed by the nanodroplets or latex particles and then anchored at their interfaces by means of a crosslinking reaction with DVB a n d / o r MPS, and the hydrolysis and condensation reactions of MPS. Nanocapsules were formed through encapsulation of the hydrocarbon by the newly produced polymers at the interfaces of the droplets. The crosslinked structure gradually grew to stabilize the shell morphology. Keywords: miniemulsion, nanocapsules, interfacial, nanodroplets, encapsulation, oligomer, morphology, crosslinker.

5.1

Introduction

Nanoparticles w i t h void h a v e s h o w n a great potential for the e n c a p sulation, e.g. catalysts, e n z y m e s , D N A s , medicines, or p i g m e n t s [1-3]. A large n u m b e r of innovative techniques h a v e been designed to elaborate hollow particles [4-10]. Recently, to directly encapsulate V. Mittal (ed.) Miniemulsion Polymerization Technology, (97-138) © Scrivener Publishing LLC

97

98

MINIEMULSION POLYMERIZATION TECHNOLOGY

liquid substances via a synthetic method in the emulsion, microemulsion, or miniemulsion has showed to be promising to synthesize nanocapsules [11-28], The formation of nanocapsules with an oily core in most literatures followed the mechanism of polymer phase separation in the oil droplets during polymerization. This process is controlled by the thermodynamic and kinetic factors. The thermodynamic aspect referring to the interfacial tensions of different phases can be influenced by the nature of oils, polymers, surfactants, and initiators. The kinetic factors included the viscosity in the dispersed phase and the mobility of polymer chains. The formulations and techniques for synthesizing the product with a high number fraction of nanocapsules need to be precisely designed. In addition, another limitation of the polymer phase separation in oil droplets is that the targeted nanocapsules cannot be obtained in the case that the morphology of nanocapsules is not the thermodynamically favorable state, because the diffusion of polymer chains to the interfaces of latex particles is no longer a spontaneous process. Interfacial polymerization has shown to be a robust alternative in the elaboration of nanocapsules through confining the processes of polymerization and polymer phase separation on the interfaces of latex particles. The polyurethane, polythiourea and polyurea nanocapsules with a hydrophilic core were synthesized in an inverse miniemulsion system [19-20], while a hydrophobic liquid core was encapsulated in a direct miniemulsion system [21] by interfacial polycondensation. Scott et al. reported an analogical method to the interfacial polycondensation for synthesizing nanocapsules but via alternating free radical copolymerization of hydrophobic maléate and hydrophilic vinyl ether monomers on the interfaces of the oil droplets in miniemulsion [22-23]. In addition, Luo et al. reported that the targeted nanocapsules could be synthesized via the interfacially confined controlled/living radical miniemulsion polymerization by using an amphiphilic RAFT agent [24-26]. Thermo-sensitive polymers such as poly(N-isopropylacrylamide) (PNIPAM) can undergo volume phase transition to separate out of the aqueous phase above their low critical solution temperature (LCST). The separated PNIPAM cannot dissolve into either the aqueous or the oil phases, and thus they are inclined to locate at the interfaces of the oil/water. Using the volume phase transition of PNIPAM, the thermo-sensitive capsules have been synthesized via the interfacial polymerization in an inverse emulsion system [27].

POLYMERIC NANOCAPSULES

99

The morphology of polystyrene nanocapsules was not the thermodynamically favorable state in the system containing surfactant, but the product with a high number fraction of the nanocapsules could be synthesized through the control of kinetic factors to perform interfacial miniemulsion polymerization by the synergic interaction of N-isopropylacrylamide (NIPAM) and divinylbenzene (DVB) [28]. The incorporation of functional groups into latex particles is of major interest in the field of colloidal science. Recently, the hydroxyl functionalized hybrid latex particles have been conveniently obtained via the incorporation of alkoxysilane units into the copolymer chains in a dispersed medium [29-31]. The synthesis of hybrid nanocapsules via miniemulsion (co)polymerization of γ-me thacryloxypropyltrimethoxysilane (MPS) and styrene (St) [18] was reported. In that case, the formation mechanism of nanocapsules mainly followed the polymer phase separation in oil droplets, and thus the products with high number fraction of nanocapsules only could be obtained in a limited range.

5.2

Organic Nanocapsules by Interfacial Miniemulsion Polymerization

5.2.1 Thermodynamic Prediction for the Morphology of Organic Nanocapsules Torza and Mason [32] discussed the equilibrium configurations of two immiscible liquid drops, designated as phase-1 and phase-3, suspended in a third immiscible liquid, phase-2, from the interfacial tensions and spreading coefficients. The relationship between spreading coefficients and interfacial tensions is given, s

.-=7,*-fy+7»)

where, y;. is the interfacial tension of i-phase and j-phase, S; is the spreading coefficient. If γ12 > γ23, the equilibrium configurations of two immiscible liquid drops can be judged by the spreading coefficient. Phase-1 is completely engulfed in phase-3 to form core-shell structure (Figure 5.1a) when S3 < 0, S2 < 0 and S3 > 0. Phase-1 is partially engulfed in phase-3

100

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

(A) Core-shell

(B) Hemispheres

(C) Individual particles

Figure 5.1. Schematic representation of the morphological prediction based on the the spreading coefficients, (a) S, < 0, S2 < 0 and S3 > 0 (b) Sx < 0, S2 < 0 and S3 < 0 (c) S, < 0, S2 > 0 and S3 < 0, O Phase-1 phase-2.

— Phase-3 Medium-

to form hemisphere structure (Figure 5.1b) when S2 < 0, S2 < 0 and S3 < 0. Phase-1 and phase-3 are completely separated (Figure 5.1c) when S1 < 0, S2 > 0 and S3 < 0. From Table 5.1, the interfacial tension of octane/water is greater than that of polystyrene/water in the system without any surfactant. Morphology predictions were conducted using the methods of Torza and Mason. Subscript assignment, 1, 2 and 3, accounts for the octane, water and polystyrene phase, respectively. According to the spreading coefficients result (S1 < 0, S2 < 0 and S3 > 0), core-shell structure is the thermodynamics stable state. And phase-1 is octane phase, so the octane core is engulfed in the polystyrene shell, which means capsule structure. The interfacial tensions of polystyrene/SDS aqueous solution and octane/SDS aqueous solution at different SDS concentrations are listed in Table 5.1. Upon the introduction of a surfactant, the interfacial tensions of both the polystyrene/SDS aqueous solution and octane/SDS aqueous solution decreased, but the latter showed a more pronounced response to the change in surfactant concentration. According to the result above, we could predict that the morphology of the system containing SDS is still core-shell structure, when phase-1 and phase-3 account for polystyrene and octane phase, respectively. Although the morphology of the final product is still a core-shell structure, the equilibrated morphology will be a polymer core with an oil shell in the system containing SDS based on the results of spreading coefficients in Table 5.1.

POLYMERIC NANOCAPSULES

101

Table 5.1. Interfacial tension between different phases, spreading coefficients, and morphology prediction. Reproduced from reference 28 with permission from Wiley. SDS 'Octane/W 'PS/W 'Octane/PS concentration (mN/m) (mN/m) (mN/m) (mmol/L) 0 1.25 2.5 5

30.1 25.8 21.5 16.8

49.7 11.3 4.6 4.8

7.2 7.2 7.2 7.2

[18] [18] [18] [18]

s, -27 -22 -24 -19

s 2 s3

Morphology Prediction

-73 12 Core-shell 3 -30 7 Core-shell b -19 10 Core-shell b -14 5 Core-shell b

"oil core and polymer shell; bpolymer core and oil shell.

5.2.2 Particles Morphology of the System without Added NIPAM and DVB The ratio of styrene/hydrocarbon was set at about 3:7. The surfactant concentration used throughout in this work was 5 mmol/L, and therefore the morphology of the nanocapsules should not be the thermodynamically favorable state according to the results of Table 5.1. Figure 5.2 shows that there is only little amount of 100 nm incomplete nanocapsules. Meanwhile, most of the particles are small solid balls. This result is in agreement with the thermodynamic prediction (Table 5.1). Landfester and coworkers [13] also found that polymerization initiated by KPS leaded to mostly solid spherical polystyrene particles, but not to the formation of nanocapsules. The evolution of the morphology and polymerization is shown in Figure 5.3. In this system, KPS was used as initiator, and therefore initiator radicals reacted with styrene dissolved in aqueous solution. Styrene oligomer propagated to the critical length (about 3 [33]), became insolubility and adsorbed on the surface of the droplets or latexes, while there are a lot of droplets and latexes in the miniemulsion system which provides huge surface area for adsorption. Because of the hydrophilic of the initiator fragment (S04~) and the interaction of electronegative charge of the initiator fragment (S04~) with SDS, the oligomer radicals are expected to remain at the oil/water interface temporarily and propagate with monomers. In the system, the polymer radicals will dissolve in oil droplets with the propagation of polymer chains (with the lipophilicity increasing).

102

MINIEMULSION POLYMERIZATION TECHNOLOGY

Figure 5.2. TEM image of latex particles of the system without added NIPAM and DVB (SDS 0.20 g, octane 8.33 g, HD 0.60 g, St 3.33 g, KPS 0.30 g, water 150 g, T = 75°C).

The polymers will separate from the oil phase as polymer concentration increases and the monomer is consumed. Moreover, as mentioned above, although the nanocapsules represent the thermodynamically favored state in the system without the addition of surfactant, the prevailing thermodynamic conditions in the system with surfactant are not expected to be favorable for the formation of nanocapsules because of the decrease in interfacial tension of polystyrene/SDS solution and octane/SDS solution when compared with pure water. Therefore, polymer chains separated at the interface can diffuse into the interior of the latex particles, and finally the droplets changed into solid particles. But along with polymerization, the viscosity of droplet increases, the diffusion of separated polymer chains became rather difficult, and some of the polymer radicals stayed at the oil/water interface, then a few irregular nanocapsules are formed. It is believed that these small solid particles were composed of a polymer core and an oil shell structure, and that their small size may have been due to the evaporation of octane during the process of TEM sample preparation and observation. Also, some visible octane droplets were found during the experiment.

POLYMERIC NANOCAPSULES

103

Figure 5.3. Schematic representation of the formation of latex particles in the system without added NIPAM and DVB.

5.2.3 Particles Morphology of the System with DVB Both of the thermodynamic prediction and experiment confirm that the styrene miniemulsion system, octane as template and SDS as emulsifier, will not form thermodynamic stable nanocapsules. Therefore, some kinetic aspects are presented to control the polymerization location and morphology. If polymer radical chains could be anchored and propagate at the interface of the latex particles, it should be possible to obtain nanocapsules. In order to confirm this hypothesis, a crosslinking agent, DVB, was introduced into the system. The result was shown in Figure 5.4. Although there were many solid particles, the number of regular nanocapsules increased remarkably. The formation and adsorption of radicals process was the same with previous ones. But DVB could limit the oligomer radicals at the droplet/water or latex/water interface. Therefore, oligomer radicals could react at the surface of droplet/water or latex/water. At the same time, monomer diffuses from the inside of droplet or latex to the interface and form crosslinking shell. Otherwise, the crosslinking shell could also slow down the oligomer radicals entering into the droplet or latex, and it made more oligomer radicals react at the interface to form nanocapsules. Actually, there were still many solid particles in the system initiated by KPS. This confirms that the hydrophilic of the initiator fragment (S04")

104

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Figure 5.4. TEM image of latex particles of the system with DVB but without NIPAM (SDS 0.20 g, octane 8.33 g, HD 0.60 g / St 3.33 g, DVB 0.40 g, KPS 0.30 g, water 150 g, T=75 °C).

and the interaction of electronegative charge of the initiator fragment (S04") with SDS were not enough to anchor most of the oligomer radicals at the oil/water interface to attack DVB. Most of the oligomer radicals would enter into the droplets or latex and react there. This would lead to polymer-core and oil-shell structure. The evolution of the reaction with DVB was shown in Figure 5.5. The locus of radical formation and the property of oligomer radicals will be decided by the type and property of the initiator. Both of these two aspects will influence the final morphology of particles. The oligomer radicals initiated by KPS have the initiator fragment (S04~), which could be adsorbed at the surface of droplets or latexes and drove down the interface tension between oil and water. Therefore, oligomer radicals could temporarily rest on the interface of droplets or latexes. AIBN, instead of KPS, was used to compare the influence of initiator type and property on the particle morphology, while other conditions kept changeless. AIBN is an oil-soluble initiator and dissolve in the oil droplet or latex. It could decompose into nonionic/oil-soluble radicals and initiate the monomer in the oil droplet or latex, which means the radicals

POLYMERIC NANOCAPSULES

105

Figure 5.5. Schematic representation of the formation of latex particles in the system with added DVB.

have no anchor effect at the surface of droplet or latex. Some of the nonionic/oil-soluble oligomer radicals can go out from or enter into the particles directly without any anchor effect, leading to solid particles (see Figure 5.6). van Zyl and coworkers[34] could not obtain nanocapsules with AIBN initiator system either. Comparing with Figure 5.4 and Figure 5.6, it can be found that there is more nanocapsules when using KPS as initiator rather than that of AIBN. That means KPS fragment oligomer radicals make it easier form nanocapsules.

5.2.4

Particle Morphology of the System with Added NIPAM and DVB

Previous analysis show that the precondition of interfacial miniemulsion polymerization is the diffusion rate of oligomer radicals and polymer chains being slow enough to be crosslinked by DVB on the interface of the latex particles. In other words, although the system with surfactant is thermodynamic unstable, nanocapsules could still be produced by controlling kinetic aspect. The above result shows that the anchor effect of oligomer radicals with KPS fragment should be enhanced. PNIPAM is the most familiar temperature-sensitive polymer. Its low critical solution temperature (LCST) is about 32 oç[35,36] Therefore, PNIPAM can undergo phase separation at the

106

MINIEMULSION POLYMERIZATION TECHNOLOGY

Figure 5.6. TEM image of latex particles of the system initiated by AIBN (SDS 0.20 g, octane 8.33 g, HD 0.60 g, St 3.33 g, DVB 0.40 g, AIBN 0.30 g, water 150 g, T=75 °C). Reproduced from reference 28 with permission from Wiley.

reaction temperature (75 °C), and the separated PNIPAM oligomer is expected to be locate at the droplet/water or latex/water interface, because it can not dissolve either aqueous or oil phase at the reaction temperature. One should be illuminated is that oligomer formed from ordinary water-soluble monomer (such as AA, MA) has so water-soluble that it may lead to aqueous phase nucleation. But PNIPAM oligomer radical could separate from aqueous phase and be adsorbed at the surface of droplet or latex, which reduce the probability of aqueous phase nucleation. In order to confirm the previous analysis, NIPAM was added into the system with DVB. In contrast with Figure 5.2 and Figure 5.4, the morphology of latex particles, shown in Figure 5.7, indicated that most of these latex were in the form of nanocapsules when some amount of NIPAM was introduced in the system. The numberaverage diameter of the nanocapsules and the average shell thickness estimated by TEM results was about 79 nm and 13.5 nm, respectively. The average void fraction of the capsule was about 28.4%, which is lower than the calculated value for the feed mixture composition. However, it should be borne in mind that this value is

POLYMERIC NANOCAPSULES

107

Figure 5.7. TEM image of nanocapsules of the system with added NIPAM and DVB (SDS 0.20 g, octane 8.33 g, HD 0.60 g, St 3.33 g, DVB 0.40 g, NIPAM 0.40 g, KPS 0.30 g, water 150 g, T = 75 °C). Reproduced from reference 28 with permission from Wiley.

only approximate and is lower than the exact value because of the existence of large and deformed capsules as well as a volume contraction because of the evaporation of octane in vacuum condition during the TEM sample preparation and observation. This estimation is also confirmed by the DLS analysis, because the number-average diameter of capsules obtained by DLS is 87.6 nm (Figure 5.8). To verify whether NIPAM participates in the copolymerization and into the shell, FTIR and solid-state 13C-NMR measurements were performed to characterize the compositions of the purified shellcopolymersofthenanocapsules.ComparingtheFTIRspectra shown in Figure 5.9, it can be seen that a sharp band at 1650 cm -1 , characteristic of the amide carbonyl groups of PNIPAM, as well as the specific bands of the aromatic ring of polystyrene at 1600, 1492, 906, 756 and 696 cm"1 are observed in the spectrum of the nanocapsules. The peak marked with an asterisk in the solid-state 13 C-NMR spectrum in Figure 5.10 is attributable to the carbonyl groups of PNIPAM. Thus, both the FTIR and solid-state 13 C-NMR spectra indicated that NIPAM molecules had been

108

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Figure 5.8. Particle size and size distribution of nanocapsules of the system with added NIPAM and DVB (SDS 0.20 g, octane 8.33 g, HD 0.60 g, St 3.33 g, DVB 0 .40 g, NIPAM 0.40 g, KPS 0.30 g, water 150 g, T=75 °C).

4000

3500

3000

2500

2000

1500

1000

500

Wavenumber (cm-1)

Figure 5.9. Fourier-transform infrared spectra of pure PSt and the copolymers of the nanocapsules. Reproduced from reference 28 with permission from Wiley.

copolymerized into the shell of the nanocapsules. In fact, that the morphology of latex particles relies on the NIPAM amounts (This will be shown in the following section) can also serve as an important evidence for the participation of NIPAM in the copolymerization.

POLYMERIC NANOCAPSULES LD

in

Tl-

CD

|CM -~

109

h-C*5 0 3 f ^ CO

LOO

ι-σ> CM

Ti-Tf CO CM CM

— i — ' — i — ' — i — ' — i — ■ — i —

180

160

140

120

100

80

60

40

20ppm

13

Figure 5.10. Solid-state C-NMR spectrum of the copolymers of the nanocapsules. Reproduced from reference 28 with permission from Wiley.

5.2.5

Particle Size and Size Distribution in the Process of Polymerization

The droplet and particle sizes during the polymerization were measured by DLS, the Z-average size and volume-average size of the particles was used to evaluate the size evolution and size distribution, respectively. The results in Figure 5.11(a) indicate a rapid decrease in the particle size during the initial stage of the polymerization, and thereafter (about 30min) the particle size remains constant. That the polydispersity index (PDI) of the particles decreasing distinctly elucidates the particles size distribution narrowing. This phenomenon could have been easily ascribed to secondary nucleation. However, the fact that nearly all of the latex particles were capsules did not support the occurrence of secondary nucleation. Droplet nucleation is a prerequisite for obtaining nanocapsules, whereas secondary nucleation leads to the formation of solid particles. As shown in Figure 5.11(b), the volume distribution of the initial droplets was broad, and some large droplets were presented in the initial miniemulsion. However, as the polymerization proceeded, the number of large droplets decreased. Generally, in miniemulsion polymerization, if the droplet size distribution is narrow and nucleation time is short, the monomer transportation

110

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY 170 160

0

N 150

'€ 140 to

a.

130 120

(A)

. \

1.0

V □

X

-a— Particle size -o-PDI

CO

300 min

- 0.6

120 min 0.4 .a 0.2

0.0 50 100 150 200 250 300 Polymerization time (min)

o

Q.

0 (B)

200 400 600 Volume particle size (nm)

800

Figure 5.11. Evolution of particle size (a) and volume distribution (b) during the process of interfacial miniemulsion polymerization with added NIPAM and DVB (SDS 0.20 g, octane 8.33 g, HD 0.60 g, St 3.33 g, DVB 0.40 g, NIPAM 0.40 g, KPS 0.30 g, water 150 g, T = 75°C). Reproduced from reference 28 with permission from Wiley.

between the droplets a n d / o r latex can be negligible. But in a broad size distribution system, small droplets have more opportunities to nucleate and capture radicals than larger ones because of their greater overall surface area. Once polymerization starts in small latex particles, monomers were consumed in these nucleated latex particles, resulting in the difference between the monomer concentration in small nucleated latex particles and large non-nucleated droplets. To maintain equilibrium, monomers would diffuse from the nonnucleated large droplets to the smaller nucleated latex particles. The droplets were a mixture of octane and monomer in this system, and therefore the concentration of octane in the large droplets would increase with the loss of monomer. Kim and Burgess [37] found that the interfacial tension of oil/ water interface is related to the oil phase constitute. The interfacial tension of the styrene/water interface is increased with decreasing styrene weight content in the oil phase, as can be seen in Figure 5.12. As a result, with the loss of monomers in large droplets, the interfacial tension of the oil/water interface was increased. It resulted in colloidal instability, and subsequently most of the non-nucleated large droplets coalesced to form macro-droplets. In addition, the coalescence of large droplets was consistent with the relatively low hydrocarbon encapsulation efficiency. The rapid decrease in the Z-average particle size during the

POLYMERIC NANOCAPSULES

111

60 ■

50

>

40 -

Ό-^.

CO

Ό co t

"~~~Ό-^

-/",

C

S Έ 30 ■

20

1

.

1

1

,

20 40 60 St weight content (wt%)

1

80

100

Figure 5.12. Influence of St weight content on the interfacial tension of oil mixture/water. Reproduced from reference 28 with permission from Wiley.

0 (A)

50 100 150 200 250 300 Polymerization time (min)

0

200 400 600 Volume particle size (nm) (B)

800

Figure 5.13. Evolution of particle size (a) and volume distribution (b) during the process of miniemulsion polymerization without added NIPAM (SDS 0.20 g, octane 8.33 g, HD 0.60 g, St 3.33 g, DVB 0.40 g, KPS 0.30 g, water 150 g, T=75°C). Reproduced from reference 28 with permission from Wiley.

initial phase of polymerization could also be reasonably ascribed to the simultaneous coalescence of large droplets. The particle size and size distribution of the system to which no NIPAM had been added, shown in Figure 5.13, demonstrate similar behavior to that of the system with NIPAM. As in the earlier analysis, the fast decrease in particle size during the initial phase of polymerization could again be ascribed to the disappearance of

112

MINIEMULSION POLYMERIZATION TECHNOLOGY

large droplets. Although TEM results showed that a large number of small solid particles were presented in the final product, the final sizes of the products with or without NIPAM were quite similar. It was further confirmed that particles with an inverted morphology, that is, with a solid polymer core and an oil shell, were produced.

5.2.6

Mechanism for the Formation of Organic Nanocapsules through Interfacial Miniemulsion Polymerization

The schematic representation in Figure 5.14 illustrates the mechanism for the formation of nanocapsules through interfacial polymerization. At first, the oil phase was dispersed into nanodroplets, and NIPAM molecules were mainly distributed in the aqueous phase and at the interface of the droplets, while only a small amount of styrene was dissolved in the aqueous phase because of its poor solubility in water. The water-soluble initiator, KPS, decomposed to form primary radicals, which mainly propagated with NIPAM to generate oligomer radicals of PNIPAM. The polymerization temperature used (75 °C) was much higher than the LCST of PNIPAM. As a result, PNIPAM oligomer radicals beyond the critical chain length separated from the aqueous phase.

Figure 5.14. Schematic representation of the process of the formation of nanocapsules through interfacial copolymerization in miniemulsion. Reproduced from reference 28 with permission from Wiley.

POLYMERIC NANOCAPSULES

113

Because of the tremendous interface areas of the droplets and latex particles, these were able to adsorb most of the phase separated PNIPAM oligomers. Chain ends of the oligomer radicals bearing S04~ groups were difficult to bring close to the interface of the latex particles because of the electrostatic repulsion between the initiator fragments and the surfactants. From the thermodynamic point of view, the adsorption of PNIPAM at the interface of droplets or latex particles was a favorable state, because PNIPAM, which is less hydrophobic than the hydrocarbon or polystyrene, lowered the interfacial Gibbs energy. In addition, the PNIPAM oligomers can not dissolve in both the aqueous and oil phase at the reaction temperature. The diffusion rate of oligomer radicals containing NIPAM units becomes slow enough to be cross linked by DVB, and then be anchored at the interfaces of the latex particles due to the decrease of mobility as the result of the formation of cross linked structure, leading to the interfaces to be the main loci of polymerization. Monomers diffused from the interiors of the latex particles as a supply for further polymerization. The crosslinked polymer shell could also slow down the diffusion rate of newly formed radicals and make it easier to crosslink. Consequently, nanocapsules with a crosslinked shell could be obtained through the encapsulation of the hydrocarbon by the accumulation of crosslinked polymers at the interfaces of latex particles.

5.2.7

Influences on the Formation of Organic Nanocapsules through Interfacial M i n i e m u l s i o n Polymerization

1) Amount of NIPAM Regular and high void fraction nanocapsules could be obtained by interfacial miniemulsion polymerization with both the contribution of NIPAM and DVB. NIPAM can slow down the oligomer radicals' ingoing rate, delay their residence time at the interface of the latex particles. The anchored radicals polymerize on the interface to ensure the interfacial polymerization. Once a crosslinked shell is formed at the interface of the latex particles, access by other radicals will be impeded. Under these circumstances, radicals without a NIPAM unit may also be crosslinked and anchored at the interface. Without adding NIPAM, only a small number of nanocapsules could be obtained in Figure 5.4. The number of nanocapsules increased markedly upon the addition of NIPAM, even a small

114

(A)

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

(B)

Figure 5.15. TEM images of nanocapsules and particles of the systems with different amounts of NIPAAM (SDS 0.20 g, water 150 g, octane 8.33 g, HD 0.60 g, St 3.33 g, DVB 0.40 g, KPS 0.30 g, T=75°C). (a) 0.2 g NIPAM (b) 1.0 g NIPAM [28].

amount (0.2g), as can be seen in Figure 5.15(a). Almost all of the droplets evolve into nanocapsules, as more NIPAM (0.4g) was introduced into the system (Figure 5.7). Although the nanocapsules still dominated in the final product, they were inclined to stick together in the larger amount (lg) NIPAM system and the contrast of the shell boundary is not clear. It is due to more NIPAM is incorporated into the shell. From the dependence of the number of capsules on the amount of NIPAM, it may be stated that for this system, 0.4g of NIPAM was enough to make nearly all droplets evolve to form nanocapsules. However, no evident aqueous phase nucleation could be observed. Therefore, PNIPAM is in the shell part and the rapid separation of PNIPAM oligomer could depress the aqueous phase nucleation. 2) Amount of DVB The mobility of (co)polymer chains can be conveniently decreased by adding crosslink agent to the system. However, addition of crosslink agent is not suitable for nanocapsules' formation in the system ruled by the inner phase separation mechanism. Unexpectedly, for the systems without or with a low amount of DVB, it is difficult to get nanocapsules (Figure 5.16(a) and Figure 5.16(b)). Although NIPAM was added in each runs and separated PNIPAM oligomer radicals were adsorbed by the droplets or latex, polymer chains would still propagate into hydrophobe and enter the oil phase

POLYMERIC NANOCAPSULES

(A)

115

(B)

Figure 5.16. TEM images of nanocapsules and particles of the systems with different amounts of DVB (SDS 0.20 g, water 150 g, octane 8.33 g, HD 0.60 g, St 3.33 g, NIPAM 0.40 g, KPS 0.30 g, T=75°C). Reproduced from reference 28 with permission from Wiley, (a) Og DVB (b) 0.20g DVB.

without the anchor effect of DVB. Another aspect is also adverse, in the system containing SDS, nanocapsule is not a thermodynamic stable state. The polymer chains separated on the interface will be driven into the interior of the latex by the thermodynamic force. The hypothesis was confirmed by Figure 5.16(a). However, addition of NIPAM in the system without DVB improved the fraction of regular nanocapsules (Figure 5.16(a) and Figure 4.2). For the systems with a low amount of DVB, the crosslinking reactions are so insufficient to anchor most of the oligomer radicals on the interface that many oligomer radicals could enter the droplets. The number fraction of nanocapsules slightly increased with the addition of 0.2 g DVB, but solid particles are primary. As the DVB amount increased to 0.4 g, the morphology of nanocapsules dominated in the latex particles. Lower chain mobility can be obtained by using a larger amount of DVB. Most of droplets undergo the interfacial polymerization. Newly formed radicals will probably be intercepted by the crosslinked shell and propagate, even if no crosslinking reactions happen on the interface. The different morphologies of the products obtained with different amounts of DVB provide strong evidence that the formation of nanocapsules follows the mechanism of interfacial polymerization in these systems with both NIPAM and DVB.

116

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

3) Weight content of monomer in the oil phase Higher weight content of monomer in the oil phase enhances the solubility of the polymer resulting in higher droplet viscosity, which is not favorable for polymer chain diffusion. Thus, formation of nanocapsules mainly followed an inner phase separation mechanism, which was strongly influenced by the internal viscosity of droplets, and this, in turn, was dependent on the solubility of the polymer in the oil mixture. In our previous research [18], nearly all droplets evolved to form solid particles when the weight content of monomer was 50%. The formation of nanocapsules follows an interfacial polymerization mechanism, and therefore the influence of the weight content of the monomer in the oil phase on the formation of nanocapsules can be analyzed from two aspects. Firstly, with increasing monomer content of the oil phase, the solubility of polymer chains in the oil droplets will increase. Once the radicals enter into the oil droplets, the crosslinked structure of the polymer chains will decrease the chain mobility, which is favorable for the formation of solid particles. Secondly, the higher viscosity also makes it hard for radicals transport to the central part of the droplet. It is favorable for nanocapsule structure. Figure 5.17 shows that the number of nanocapsules is still

Figure 5.17. TEM image of nanocapsules of the system with 50% monomer weight content in oil phase (SDS 0.29 g, water 150 g, octane 8.33 g, St 8.80 g, DVB 0.40 g, HD 0.87 g, NIPAM 0.58 g, KPS 0.74 g, T=75 °C).

MINIEMULSION POLYMERIZATION

117

more than that of solid particles at 50% weight content of monomer in the oil phase. However, a small increasing number of solid particles can also be observed in contrast to 30% weight content of the oil phase. It indicates that the solubility of oligomer radicals in the oil droplets will influence the morphology; however, the fact that most of the particles are nanocapsules supports the interfacial polymerization mechanism. Compared with low monomer weight content system, higher monomer weight content results in more regular nanocapsules.

5.3

Organic-Inorganic Hybrid Nanocapsules by Interfacial Miniemulsion Polymerization

5.3.1 Thermodynamic Analysis and Morphological Prediction The morphologies of ternary system of poly(MPS-co-St)/HD/ SDS solution were predicted again with the method of Torza and Mason [32]. According to the interfacial tensions of the polymer/ water, the polymer/SDS solutions, the oil/water and the oil/SDS solutions, and the spreading coefficients listed in Table 5.2, 5.3 and 5.4, and 5.5, the morphology of nanocapsules (oil core and polymer shell) is the most favorable state only in the case that water is used as the continuous phase. Upon the introduction of surfactant, both interfacial tensions of the poly(MPS-co-St)/SDS solutions and HD/SDS solutions decreased, but the latter showed a much more pronounced response to the change in surfactant concentration, resulting in γ /w > yo/w (seen in supporting information). Combined Table 5.2. Influences of monomer weight content on the oil/SDS solution interfacial tension. SDS concentration (mmol/L)

IHD/W

(mN/m)

γΟΛν (mN/m) (M% = 50 wt%) a

0

51.2

12.3

1.25

14.0

4.0

-5

7.5

1.8

5

6.2

1.8

2

a

M% represents the monomer weight content in the oil mixture of HD and monomers.

118

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Table 5.3. Influences of MPS weight contents on the interfacial tension of oil/SDS solution 3 . SDS concentration (mmol/L)

γΟΑν (mN/m) (MPS = 20.0 wt%)

y 0 / w (mN/m) (MPS = 33.3 wt%)

Ύο/w (mN/m) (MPS = 50.0 wt%)

0

12.26

11.12

11.43

1.25

4.00

4.11

3.51

2.5

1.83

1.67

1.68

5

1.83

1.65

1.63

The monomer weight content was 50%. Table 5.4. Surface tension, dispersion force and polar force of poly (MPS-co-St) and the interfacial tension of HD and poly(MPS-co-St)a. ys (mN/m)

■f (mN/m)

■f (mN/m)

Yo/p(mN/m)

poly(MPS-co-PS) (MPS=20.0 wt%)

28.43

24.38

4.05

4.67

poly(MPS-co-PS) (MPS=33.3 wt%)

32.53

28.03

4.50

8.22

poly(MPS-co-PS) (MPS=50.0 wt%)

27.45

22.67

7.34

9.17

HD

27.47

27.47

0

-

"γ., ~f, and ψ represent surface tension, dispersion force and polar force of poly(MPS-co-St) respectively; γο/ represents the interfacial tension of HD and poly(MPS-co-St). Table 5.5. Spreading coefficients of the ternary system of poly (MPS-co-St)-HD-water. p(MPS-co-St) p(MPS-co-St) SDS p(MPS-co-St) concentration (MPS = 20.0 wt%) (MPS = 33.3 wt%) (MPS = 50.0 wt%) (mmol/L) s, s 2 S3 s, s 2 S3 s, s 2 S3 0 1.25 5

-29.5 -72.9 20.2 -40.1 -62.4 23.6 -42.2 -22.7 23.9 -10.5 -29.1

1.2

-10.9 -25.9 -2.0 -14.0 -23.3 -4.37

-13.4 -16.4 4.04 -12.5 -11.9 -0.4 -13.7 -15.2 -4L63

POLYMERIC NANOCAPSULES

119

with the spreading coefficients listed in Table 5.5, the morphologies of final product may be an inverted core-shell (oil shell and polymer core) or a hemisphere in ternary system of poly(MPS-co-St)/ HD/SDS solution, depending on the SDS concentrations and the copolymer compositions.

5.3.2 Synthesis of Organic-Inorganic Hybrid Nanocapsules under Neutral Conditions 1) Role of cross-linking reaction in the formation of organic-inorganic hybrid nanocapsules Alkoxysilane, MPS, was introduced to synthesize nanocapsules with an organic-inorganic hybrid shell. Afore-mentioned thermodynamic analyses have shown that the morphology of nanocapsules is not the thermodynamically favorable state for a ternary system of poly(MPSco-St)/HD/SDS solutions. Therefore, it was necessary to synthesize the hybrid nanocapsules by performing the interfacial miniemulsion copolymerization through anchoring the oligomeric radicals on the interfaces of latex particles/water. It is well known that MPS can undergo the hydrolysis and condensation reactions in an aqueous system, e.g. emulsion or miniemulsion polymerization [29-31]. The hydrolysis and condensation rate of MPS depends greatly on the suspension pH. The FTIR spectrum (line a in Figure 5.18) of the hybrid copolymers synthesized under the neutral condition showed that a sharp absorbance peak at 821 and 1087 cm-1 belonging to Si-O-C supported that most of MPS did not participate in the hydrolysis. The spectrum of solid-state 29Si-NMR (Figure 5.19) showed that T° species dominated in the hybrid copolymers and only a small part of silicon species condensed (According to the conventional Tn nomenclature where n designates the number of Si-O-Si bonds, the signal at -43, -48, -58 and -68ppm corresponding to T° (CH30)3SiR), T1 (CH30)2Si(OSi) R), T2(CH30)Si(OSi)2R) and T3 species (Si(OSi)3R) respectively). Both FTIR and solid-state 29Si-NMR spectra definitely indicated that the hydrolysis and condensation of MPS remained at a low level under the neutral condition. Nevertheless, upon the introduction of MPS, the formation of cross-linking structure through condensation reactions taking place on the oil/water interfaces could help to anchor a part of oligomeric radicals on the interfaces of the latex particles /water. Therefore, some droplets could undergo interfacial copolymerization to form nanocapsules in a system without added DVB under the

120

MINIEMULSION POLYMERIZATION

TECHNOLOGY

Figure 5.18. FTIR spectra of shell copolymers synthesized via miniemulsion copolymerization by using HD as template under different pH conditions (SDS 0.216 g, water 150 g, HD 10.8 g, St 8.64 g, NIPAM 0.8 g, MPS 2.16 g, KPS0.15g,T = 70°C).

-42

T1

T2

h-48 -58

ΙΛ 66

100

50

0

_1_ -50

T3

_l_ -100

J_ -150

J -200

Chemical shift (ppm) 29

Figure 5.19. Solid-state Si-NMR spectrum of shell copolymers synthesized via miniemulsion copolymerization by using HD as template under the neutral condition (SDS 0.216 g, water 150 g, HD 10.8 g, St 8.64 g, NIPAM 0.8 g, MPS 2.16 g, KPS 0.15 g, T = 70°C).

POLYMERIC NANOCAPSULES

121

Figure 5.20. Influence of DVB weight contents on the morphology of latxe particles synthesized by miniemulsion copolymerization by using HD as template with 5 mmol/L SDS (SDS 0.216 g, water 150 g, HD 10.8 g, St 8.64 g, NIPAM 0.8 g, MPS 2.16 g, KPS 0.15 g, T =70°C) (aland bl are the typical TEM images of sample; a2 and b2 are the TEM images at the boundary of particle aggregates), (al) DVB = 0 g (a2) DVB = 0 g (bl) DVB = 0.4 g (b2) DVB = 0.4 g.

neutral condition. The typical TEM image shown in Figure 5.20(al) indicated that the number fraction of nanocapsules was close to half of the overall latex particles, while the TEM image at the boundary of particle aggregates seen in Figure 5.20(a2) showed that the solid particles dominated in this area. Without DVB, the radicals were produced in the aqueous phase, and the primary radicals would mainly propagate with NIPAM and

122

MINIEMULSION POLYMERIZATION TECHNOLOGY

MPS because their relatively larger water solubility than St. It was believed that the oligomeric radicals were inclined to locate at the interfaces of droplet/water at first because of their relative hydrophilicity and poor solubility in the oil phase. However, they would propagate with oil monomers (St and unhydrolyzed MPS) at the interfaces of the latex particles/water, and thus the hydrophobicity of the polymer chains increased. In this system, most oligomeric radicals and the polymer chains would diffuse into the interior of droplets under the drive of the thermodynamic force due to the insufficient formation of the crosslinked structure. Consequently, more than half of latex particles with the morphology of solid particles appeared in the final product. The cross-linking reaction can be conveniently improved by adding some cross-linkers (e.g. DVB). The formation of crosslinked structure as the introduction of DVB can decrease the mobility of polymer chains, and thus the crosslinked polymer chains were anchored at the interfaces of latex particles/water. The number fraction of hybrid nanocapsules increased obviously, seen in Figure 5.20(bl) and 5.20(b2) as the introduction of DVB. This clearly indicated that the formation of solid particles without DVB mainly relied on the insufficient cross-linking reactions to anchor oligomeric radicals and polymer chains on the interfaces. These results could be regarded as evidence for that the formation of nanocapsules followed the mechanism of interfacial copolymerization. 2) SDS concentration A series of experiments varied DVB weight content with 10 mmol/L SDS have been performed to investigate the SDS concentration on the morphology of the particles. Only very few nanocapsules could be found in the product of the system absent of DVB seen in Figure 5.21(a). As the introduction of DVB, the number fraction of nanocapsules increased, but more than half of particles were solid particles in the system with 0.4 g DVB shown in Figure 5.21(b). The number fraction of nanocapsules increased as the increase of the weight content of DVB, and the nanocapsules dominated in the product of the system with 0.8 g DVB seen in Figure 5.21(c). Through comparing the morphologies of products with different SDS concentrations (Figure 5.20 and 5.21), it could be found that the amount of solid particles in the system with 10 mmol/1 SDS was obviously more than that of the system with 5 mmol/L SDS, especially for systems with a low level of DVB. The higher the SDS concentration,

POLYMERIC NANOCAPSULES

123

Figure 5.21. Influence of DVB weight contents on the morphology of latxe particles synthesized by miniemulsion copolymerization by using HD as template with 10 mmol/L SDS (SDS 0.432 g, water 150 g, HD 10.8 g, St 8.64 g, NIPAM 0.8 g, MPS 2.16 g, KPS 0.15 g, T = 70°C). (a) DVB = 0 g (b) DVB = 0.4 g (c) DVB = 0.8 g.

the larger is the thermodynamic driving force for the formation of non-nanocapsules. A similar dependency of the number fraction of nanocapsules on the SDS concentration has been reported by several other literatures in which the formation of nanocapsules followed the mechanism of polymer phase separation in oil droplets. Landfester et al. [13] found that the nanocapsules could be obtained in the system with 0.5 wt% SDS (with respect to the organic phase), but not in the system with 4 wt% SDS. Luo et al. [15] reported that most of particles had a half-moon structure in the case with a high level of surfactant.

124

MINIEMULSION POLYMERIZATION TECHNOLOGY

Ni et al. [18] also found that the solid particles dominated in the final product with a high level of surfactant concentration. Under the neutral condition, only a small part of MPS molecules participate in condensation reactions, and thus most of oligomeric radicals could not be anchored at the interfaces of oil/water in the system absent of DVB. The formation of the final morphology in the system with a low level of cross-linking reactions is thermodynamically controlled, similar to that of the system following the mechanism of polymer phase separation in oil droplets. However, the formation of nanocapsules can still be obtained controlling the kinetic factors by introducing some amount of DVB. This has been attested by the fact that most of droplets evolved to form nanocapsules in the system with 0.8 g DVB and 10 m m o l / L SDS.

5.3.3

Synthesis of Organic-Inorganic Hybrid Nanocapsules under Acidic or Basic Conditions

For the miniemulsion copolymerization of MPS and St, the hydrolysis and condensation of MPS are significantly influenced by the suspension pH, and both of them will be accelerated under the basic or acidic condition. In the previous section, since the hydrolysis and condensation of MPS remained at a low level under the neutral condition, an additional amount of DVB was required to improve the anchoring effect of oligomeric radicals for synthesizing products with a high number fraction of nanocapsules. In this section, it is tried to elaborate products with a high number fraction of nanocapsules to improve the anchoring effect of oligomeric radicals via the increase of the condensation degree through performing reactions under the acidic or basic conditions. 1) Suspension pH values In order to synthesize products with good colloidal and storage stability, the reactions under weak acidic or basic conditions are conducted. The pH values of the initial miniemulsions and the final latexes were summarized in Table 5.6. According to the FTTR spectra shown in Figure 5.18, all the characteristic absorbance peaks belonging to the units of St, NIPAM and MPS could be found in the resulting copolymers of hybrid shells. It indicated that all monomers participated in the shell formation of nanocapsules.

POLYMERIC NANOCAPSULES

125

Table 5.6. p H values of the initial miniemulsion and the final latexes under different pH conditions. Types of buffers

None

ADMP and DNHP

Sodium bicarbonate

Sodium tetraborate decahydrate

Initial pH

7.1

6.8

8.5

9.3

Final pH

2.6

6.5

8.9

9.1

The number fraction of nanocapsules was less than 50% of the overall number of latex particles in the product synthesized under the buffered neutral system in the absence of DVB (Figure 5.20(al) and Figure 5.20(a2)). The decomposition of KPS will decrease the pH value of the system, but for a buffered system, the suspension pH only decreased from 6.8 to 6.5. The hydrolysis and condensation rate remains at a low level attested by the results of solid-state 29 Si-NMR and FTIR seen in Figure 5.18 and Figure 5.19. Without added DVB, the anchoring effect only depending on the condensation reactions was insufficient to perform the interfacial copolymerization in the most of droplets, and thus the product with a high number fraction of nanocapsules could not be obtained under the buffered neutral system. The pH value of the unbuffered neutral system decreased since the polymerization started as a result of the decomposition of KPS, and the pH value of the product was 2.6. The decrease of pH value from 7 dramatically promoted the hydrolysis and condensation of MPS. The spectrum of solid-state 29Si-NMR of the hybrid copolymers shown in Figure 5.22(a) indicated that all MPS underwent the condensation reactions to form condensed silicone species. This was supported by the fact that the characteristic absorbance peak at 821 cm"1 disappeared in the spectrum of FTIR (Line b in Figure 5.18). Although the main silicone species was the incomplete condensed species (T2), the cross-linking degree under this condition was significantly higher than that of the buffered neutral system. Consequently, the number fraction of nanocapsules obviously increased compared to the buffered neutral system. The TEM image shown in Figure 5.23(al) indicated that nearly all latex particles were nanocapsules in the typical area of the sample. However, it was clearly seen that some solid particles still could be found at the boundary of particle aggregates (Figure 5.23(a2)). The solid

126

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

-58 T2

-66

? T°

1

T -48

-66

f , t

100 (A)

i

50

r

i

>

T3

-57 T2

V

V

-A,\

i

T3

i

i

0 -50 -100 -150 -200 Chemical s hift (ppm)

/

_ YfV"Vf 100 (B)

50

1

,

lV

,

I

.

A

0 -50 -100 -150 -200 Chemical shift (ppm)

Figure 5.22. Solid-state 29Si-NMR spectra of shell copolymers synthesized via miniemulsion copolymerization by using HD as template under different pH conditions (SDS 0.216 g, water 150 g, HD 10.8 g, St 8.64 g, NIPAM 0.8 g, MPS 2.16 ¡ KPS 0.15 g, T = 70°C).(a) pH = 7.1, unbuffered (b) pH=8.5.

particles may be produced by homogenous nucleation or the diffusion of copolymer chains into the interior of particles under the thermodynamic force. Further detailed discussion about the formation of these solid particles will be found in the following sections. The FTIR spectrum of Line c in Figure 5.18 showed that the absorbance peak of Si-O-C at 821 cm-1 decreased obviously for the product synthesized at pH 8.5. This indicated that the hydrolysis of MPS has been promoted at pH 8.5. Moreover, most resultants of hydrolysis have condensed to form the complete condensed species (T3) seen in Figure 5.22(b). Compared to that of the buffered neutral system, the degree of hydrolysis and condensation of MPS has been significantly promoted. As the increase in the level of the cross-linking reactions at pH 8.5, the oligomeric radicals were more inclined to be crosslinked and anchored on the latex particles/water interfaces to perform interfacial copolymerization. As a result, the product with the high number fraction of nanocapsules was obtained at pH 8.5, as shown in Figure 5.23(bl). But some solid particles still appeared at the boundary of the particle aggregates (Figure 5.23(b2)). First effort to decrease the amount of solid particles at the boundary of particle aggregates was increasing the pH value to 9.3. However, the number fraction of nanocapsules did not improve obviously. This indicated that the degree of hydrolysis and condensation of MPS still could not fulfill the requirement of the formation of nanocapsules under this condition.

POLYMERIC NANOCAPSULES

127

Figure 5.23. TEM images of nanocapsules synthesized via miniemulsion copolymerization by using HD as template under different pH conditions (SDS 0.216 g, water 150 g, HD 10.8 g, St 8.64 g, NIPAM 0.8 g, MPS 2.16 g, KPS 0.15 g, T = 70°C). (a) pH = 7.1, unbuffered; (b) pH = 8.5; (c) pH = 9.3. (al, bl and cl are the typical TEM images of sample; a2, b2 and c2 are the TEM images at the boundary of particle aggregates).

128

MINIEMULSION POLYMERIZATION TECHNOLOGY

2) MPS weight content at pH 8.5 The formation of solid particles in the product synthesized with 20 wt% MPS at pH 8.5 may be induced by the insufficient anchoring effect of oligomeric radicals due to the small amount of MPS. Thus, the second effort to decrease the amount of solid particles has been made by increasing the MPS weight content, because the level of hydrolysis and condensation reactions is anticipated to be improved through the increase of MPS weight content which in turn results in the increase of the anchoring effect of oligomeric radicals at the interfaces. The contributions to the increase of the anchoring effect of oligomeric radicals at the interfaces of oil/water as the increase in the level of hydrolysis and condensation reactions originate from two aspects. First, the increase of the hydrophilicity of the oligomeric radicals due to the incorporation of more hydrolyzed MPS units can decrease the diffusion rate of oligomeric radicals into the interior of droplets or latex particles, and thus the possibility to be crosslinked through condensation of oligomeric radicals at the interfaces increases. Secondly, as the increase of MPS amount participating into the hydrolysis and condensation reaction, more oligomeric radicals undergo condensation reactions and are anchored at the interfaces of droplets/water and latex particles/ water to perform interfacial copolymerization. The solid-state 29Si-NMR spectrum of the hybrid copolymer with 50 wt% MPS shown in Figure 5.24 was similar to that of the system with 20% MPS (Figure 5.22(b)). Considering the larger MPS amount used in the present system than that of the system with 20 wt% MPS and the similar portion of uncondensed and condensed species as shown in the 29Si-NMR spectrum, the amount of condensation resultants were expected to be more in the present system with 50 wt% MPS. In contrast to the product with 20 wt% MPS, the number fraction of nanocapsules obviously increased in the product synthesized with 33.3 wt% MPS, seen in Figure 5.25(al) and Figure 5.25(a2). Only very few solid particles could be found at the boundary of particle aggregates. As the MPS weight content increased to 50 wt%, the latex particles with the nanocapsules morphology dominated in both the typical area and the boundary of particle aggregates too (Figure 5.25(bl) and Figure 5.25(b2)). These results definitely showed that more droplets underwent interfacial copolymerization due to the increase of the anchoring effect of oligomeric radicals as a result of the increase in the level of hydrolysis and condensation reactions.

POLYMERIC NANOCAPSULES

129

57 T2

V,SV/>A/%>WWV/ÍVV'

100

50

0 -50 -100 Chemical shift (ppm)

-150

-200

Figure 5.24. Solid-state 29Si-NMR spectrum of shell copolymers synthesized by miniemulsion copolymerization by using HD as template with 50.0 wt% MPS relative to overall monomers at pH = 8.5 (SDS 0.216 g, water 150 g, HD 10.8 g, St 5.40 g, NIPAM 0.8 g, MPS 5.40 g, KPS 0.15 g, T=70 °C).

3) DVB weight contents at pH 8.5 The number of solid particles at the boundary of particle aggregates was reduced through increasing the MPS weight content as presented in the previous section. The anchoring effect of oligomeric radicals on the interfaces of latex particles was improved due to the increase of the cross-linking reaction and the hydrophilicity of oligomeric radicals with the increase of MPS weight content. But it is still impossible to clearly distinguish the respective role of cross-linking reactions or the hydrophilicity of oligomeric radicals in the formation of nanocapsules. The hydrophilicity of oligomeric radicals is independent of the introduction of DVB, while the crosslinking reactions greatly rely on the amount of DVB. Therefore, the third effort to decrease the amount of solid content was made by changing the amount of DVB used in the system. The differences in the number fraction of nanocapsules of the products with different DVB weight contents can be completely ascribed to the different levels of the cross-linking reaction. The TEM images of the products with different DVB weight contents synthesized under pH 8.5 were shown in Figure 5.26. The results indicated that in the product with 0.2 g DVB nearly all the

130

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

(A1)

(A2)

(B1)

(B2)

Figure 5.25. TEM images of nanocapsules synthesized via miniemulsion copolymerization by using HD as template with different MPS weight contents at pH = 8.5CSDS 0.216 g, water 150 g, HD 10.8 g, NIPAM 0.8 g, St+MPS 10.80 g, KPS 0.15 g, T = 70°C). (a) MPS = 33.3 wt%, (b) MPS = 50.0 wt%. (al and bl are the typical TEM images of sample; a2 and b2 are the TEM images at the boundary of particle aggregates).

latex particles in the typical areas were nanocapsules (Figure 5.26(31)), and the number of solid particles at the boundary of particle aggregates obviously decreased, seen in Figure 5.26(a2). Only very few solid particles could be found in the final product with 0.4 g DVB. These results clearly indicated that the formation of most solid particles in the product without DVB was due to the insufficient crosslinking reactions to anchor the oligomeric radicals at the interfaces of oil/water, but not the hydrophilicity of the oligomeric radicals.

POLYMERIC NANOCAPSULES

(A1)

(A2)

(B1)

(B2)

131

Figure 5.26. TEM images of nanocapsules synthesized via miniemulsion copolymerization by using HD as template with different DVB weight contents at pH 8.5 (SDS 0.216 g, water 150 g, HD 10.8 g, St 8.64 g, NIPAM 0.8 g, MPS 2.16 g, KPS 0.15 g, T = 70°C). (a) DVB= 0.2 g, (b) DVB = 0.4 g. (al and bl are the typical TEM images of sample; a2 and b2 are the TEM images at the boundary of particle aggregates).

4) Monomer weight content at pH 8.5 The last effort to decrease the amount of solid particles was made by increasing the monomer weight content used in the system. The TEM images shown in Figure 5.27 for the product with 66.7 wt% monomer weight content synthesized at pH 8.5 indicated that the nanocapsules dominated in the typical area of sample (Figure 5.27(al)), but the latex particles were mainly composed of solid particles at the boundary of particle aggregates (Figure 5.27(a2)). Compared to the product with

132

(A1)

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

(A2)

Figure 5.27. TEM images of nanocapsules synthesized via miniemulsion copolymerization by using HD as template with 66.7% monomer weight content synthesized at 8.5 (SDS 0.216 g, water 150 g, HD 7.2 g, St 11.52 g, NIPAM 0.8 g, MPS 2.88 g, KPS 0.15 g, T=70 °C). (al is the typical TEM images of sample; a2 is the TEM images at the boundary of particle aggregates).

50 wt% monomer weight content (Figure 5.23(al ) and Figure 5.23(a2)), the number of solid particles slightly increased in this system. As the increase of overall amount of monomers, the amount of MPS increases too, leading to the increase in the level of hydrolysis and condensation reactions under the basic condition. This is believed to be favorable to perform interfacial copolymerization in more droplets by enhancing the anchoring effect of oligomeric radicals. However the solubility of oligomeric radicals in the oil phase will increase as the increase of the monomer weight content too, resulting in the dissolution of more oligomeric radicals in the oil phase which is unfavorable to form nanocapsules through interracial copolymerization. Consequently, under the control of these two factors, the number fraction of nanocapsules did not increase for the system with 66.7 wt% monomer weight content, compared to the system with 50 wt%. It is worth noting that the nanocapsules synthesized with high monomer weight content were well-defined and with a more uniform size. 5) Morphological evolution in the process of miniemulsion copolymerization The TEM images of the particle morphology at different conversions were shown in the Figure 5.28. At the stage of low conversion

POLYMERIC NANOCAPSULES

(A)

(B)

(C)

(D)

133

(E)

Figure 5.28. Morphological evolution in the process of miniemulsion copolymerization of MPS and St by using HD as template at pH 8.5 (SDS 0.216 g, water 150 g, HD 10.8 g, St 8.64 g, NIPAM 0.8 g, DVB 0.4 g / MPS 2.16 g, KPS 0.15 g, T = 70°C).

134

MINIEMULSION POLYMERIZATION TECHNOLOGY

(conversion=12.7 %), the latex particles with a low contrast could be found in the product (Figure 5.28(a)). The morphology of latex particles is not stable enough to resist deformation induced by the oil evaporation in the process of the preparation and the observation of TEM sample. It is believed these particles are nanocapsules but with a thin polymer shell because of the low conversion. As the polymerization proceeded, more polymers were produced in the system (conversion=36.5 %), and the contrast of the latex particles increased correspondingly (Figure 5.28(b)). At this stage, the shell of nanocapsules was still thin but clearly distinguishable. The shell stability of nanocapsules further increased as the increase of conversion to 56.8% (Figure 5.28(c)). As the conversion increased to 75.5%, the resultant nanocapsules are elliptical (Figure 5.28(d)). For the product with the high conversion, the well-defined spherical nanocapsules dominated in the final product (Figure 5.28(e)). It indicated that with the increase of the amount and the cross-linking degree of copolymers in the shell, the morphology of nanocapsules became stable enough to keep spherical even with the loss of HD. The morphological evolution of latex particles definitely indicated that the morphology of nanocapsules has been formed at the initial stage of polymerization. The polymerization in the rest of reaction time contributed to increase the shell thickness and the cross-linking degree of copolymers, resulting in the increase of shell stability. These results again indicated that the formation of nanocapsules mainly follows the mechanism of interfacial copolymerization in present system. 5.3.4

M e c h a n i s m A n a l y s i s of O r g a n i c - I n o r g a n i c Hybrid Nanocapsules Formation

1) Anchoring effects The prerequisite to perform interfacial miniemulsion copolymerization is that the separated oligomeric radicals are adsorbed by the nanodroplets and then anchored at the oil/water interfaces. The anchoring effect of oligomeric radicals is governed by thermodynamic (the hydrophilicity and the oil solubility of oligomeric radicals, the thermodynamically favorable state) and kinetic (the formation of crosslinked structures) aspects concurrently. For the present systems, from the thermodynamic point of view, the hydrophilicity of oligomeric radicals can be increased via the incorporation of hydrolyzed MPS units in the oligomeric radicals;

POLYMERIC NANOCAPSULES

135

In addition, at the reaction temperature (70 °C), the PNIPAM oligomeric radicals cannot dissolve into either the aqueous phase or the oil phase. Both of them are favorable to locate the oligomeric radicals on the interfaces of oil/water to perform the interfacial copolymerization. From the kinetic point of view, the anchoring effects of oligomeric radicals can be reinforced by the introduction of crosslinker, and the condensation of MPS. 2) Process analysis In the pre-treating process, the oil phase was homogenized to form nanodroplets, resulting in the drastic increase of the interfacial area of oil/water. Under the neutral conditions, a small part of MPS participates in the hydrolysis at the interfaces of oil /water. The monomers in the aqueous phase were consisted of NIPAM and hydrolyzed MPS, while the un-hydrolyzed MPS, St and DVB were in the oil phase. For the basic or acidic system, MPS molecules not only took part in the hydrolysis reactions, but also in the condensation reactions. The monomers in the aqueous phase were consisted of NIPAM, hydrolyzed MPS and the polysiloxane oligomers below the critical chain length, while the unhydrolyzed MPS and St were in the oil phase. Under the neutral condition, upon the introduction of KPS, the oligomeric radicals containing the NIPAM and hydrolyzed MPS units were produced in the aqueous phase. At the reaction temperature, these oligomeric radicals could separate out of the aqueous phase, and then be adsorbed by oil droplets. The adsorbed oligomeric radicals were inclined to locate at the interfaces of oil/ water due to their hydrophilicity and the oil insolubility. The mobility of polymer chains was reduced through the reactions with DVB as a result of the formation of the crosslinked structures. Therefore, the oligomeric radicals were anchored, and then polymerized at the interfaces of oil/water. For the acidic or basic system, the propagation of oligomeric radicals to the critical chain length could be through the free radical copolymerization, the condensation reaction, or the combination of them. In another word, the separation of oligomeric radicals out of the aqueous phase could be promoted by the condensation. The separated oligomeric radicals could be adsorbed by the nanodroplets or latex particles. These oligomeric radicals containing silanol groups were anticipated to reduce the interfacial tension of droplets/water or latex particles /water, and therefore were inclined to locate at the interface of droplets or

136

MINIEMULSION POLYMERIZATION TECHNOLOGY

latex particles. The oligomeric radicals could be immobilized on the interfaces of latex particles/water via the condensation reactions and then propagate there. The monomer diffuses from the interiors of latex particles to the interfaces as a supply for the monomer consumption, and the newly-produced copolymers separated at the interfaces. Although the morphology of nanocapsules is not the thermodynamically favorable state for the system containing surfactant, the crosslinked copolymer chains cannot diffuse into the interiors of latex particles due to their low mobility. Consequently, the nanocapsules are obtained by the encapsulation of HD template via the accumulation of the copolymers at the interfaces.

5.4

Conclusions

The thermody namic prediction for the ternary system of PS /octane / SDS aqueous solution and poly(MPS-co-PS)/HD/SDS aqueous solution based on the interfacial tensions showed that the morphology of nanocapsules was not the most favorable state. However, the product with a high number fraction of well-defined organic and organic-inorganic hybrid nanocapsules still could be synthesized via interfacial miniemulsion polymerization of St, NIPAM, DVB, a n d / o r MPS. NIPAM and DVB played key roles in inducing the interfaces of the nanodroplets to be the loci of polymerization. In a well-designed system, although the droplet size distribution was broad and a rapid decrease in particle size was observed during the initial stage of the polymerization, the fact that nearly all droplets evolved to form organic nanocapsules indicated the exclusion of secondary nucleation. An investigation of the influence of the amount of DVB on the formation of nanocapsules in the presence of NIPAM has further confirmed that the organic nanocapsules are formed by an interfacial miniemulsion polymerization mechanism. Under the neutral condition, the results showed that the hydrolysis and condensation reactions of MPS remained at a low level. Therefore, an additional amount of DVB was required to improve the anchoring effects of oligomeric radicals on the interfaces of oil/water to promote more nanodroplets to perform the interfacial copolymerization. An investigation of the influence of

POLYMERIC NANOCAPSULES

137

the amount of DVB on the formation of organic-inorganic hybrid nanocapsules in the presence of NIPAM has further confirmed that the nanocapsules are formed by an interfacial miniemulsion polymerization mechanism. The number of solid particles increased as the increase of the SDS concentration, especially in the system with the insufficient cross-linking reactions. The morphology of latex particles could be controlled by the hydrolysis and condensation reactions of MPS via changing the pH value of suspension. Increasing or decreasing the pH value from 7, the number fraction of organic-inorganic hybrid nanocapsules increased as the result of the increase of the condensation reactions attested by the results of 29Si-NMR spectra and TEM. However, some solid particles still could be found at the boundary of particle aggregates of the product with 20 wt% MPS synthesized under the acidic and basic conditions. The number fraction of organic-inorganic hybrid nanocapsules could be further improved through increasing the MPS weight content or introducing a small amount of DVB at pH 8.5. The morphological evolution in the process of copolymerization indicated that the formation of organic-inorganic hybrid nanocapsules following the interfacial miniemulsion polymerization.

References 1. A.V. Korobko, W. Jesse, and J.R.C. van der Maarel, Langmuir, Vol. 21, p. 34, 2005. 2. K. Bouchemal, S. Briançon, E. Perrier, H. Fessi, I. Bonnet, and N. Zydowicz, International journal of Pharmaceutics, Vol. 269, p. 89, 2004. 3. A.V. Jovanovic, J.A. Flint, M. Varshney, T.E. Morey, D.M. Dennis, and R.S. Duran, Biomacromolecules, Vol. 7, p. 945,2006. 4. F. Caruso, R.A. Caruso, and H. Möhwald, Science, Vol. 282, p. 1111,1998 5. F. Caruso, M. Spasova, A. Susha, M. Giersig, and R.A. Caruso, Chemistry of Materials, Vol. 13, p. 109,2001. 6. Q. Zhang, E.E. Remsen, and K.L. Wooley, Journal of the American Chemical Society, Vol. 122, p. 3642, 2000. 7. X.Y. Liu, M. Jiang, S.L. Yang, M.Q. Chen, D.Y. Chen, C. Yang, and K. Wu, Angewandte Chemie International Edition, Vol. 41, p. 2950,2002. 8. L. Sun, R.M. Crooks, and V. Chechik, Chemical Communications, p. 359, 2001 9. T.K. Mandai, M.S. Fleming, and D.R. Walt, Chemistry of Materials, Vol. 12, p. 3481, 2000. 10. M. Chen, L. Wu, S. Zhou, and B. You, Advanced Materials, Vol. 18, p. 801,2006. 11. C.J. McDonald, K.J. Bouck, A.B. Chaput, and C.J. Stevens, Macromolecules, Vol. 33, p. 1593, 2000.

138

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

12. J. Jang, and K. Lee, Chemical Communications, p. 1098,2002. 13. F. Tiarks, K. Landfester, and M. Antonietti, Langmuir, Vol. 17, p. 908,2001. 14. A.J.P. van Zyl, R.D. Sanderson, D. de Wet-Roos, and B. Klumperman, Macromolecules, Vol. 36, p. 8621,2003. 15. Y.W. Luo, and X.D. Zhou, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 42, p. 2145,2004. 16. H.D. Koh, and J.S. Lee, Macromolecular Rapid Communications, Vol. 28, p. 315, 2007. 17. Y. Chen, H.R. Liu, Z.C. Zhang, and S.J. Wang, European Polymer Journal, Vol. 43, p. 2848,2007. 18. K.F. Ni, G.R. Shan, and Z.X. Weng, Macromolecules, Vol. 39, p. 2529,2006 19. D. Crespy, M. Stark, C. Hoffmann-Richter, U. Ziener, and K. Landfester, Macromolecules, Vol. 40, p. 3122,2007. 20. N. Jagielski, S. Sharma, V. Hombach, V. Mailander, V. Rasche, and K. Landfester, Macromolecular Chemistry and Physics, Vol. 208, p. 2229,2007. 21. L. Torini, J.F. Argillier, and N. Zydowicz, Macromolecules, Vol. 38, p. 3225, 2005. 22. C. Scott, D. Wu, C.C. Ho, and C.C. Co, Journal of the American Chemical Society, Vol. 127, p. 4160,2005. 23. D. Wu, C. Scott, C.C. Ho, and C.C. Co, Macromolecules, Vol. 39, p. 5848,2006. 24. Y.W. Luo, and H.Y. Gu, Macromolecular Rapid Communications, Vol. 27, p. 21, 2006. 25. Y.W. Luo, and H.Y. Gu, Polymer, Vol. 48, p. 3262,2007 26. F.J. Lu, Y.W. Luo, and B.G. Li, Macromolecular Rapid Communications, Vol. 28, p. 2868, 2007. 27. Q.H. Sun, and YL. Deng, Journal of the American Chemical Society, Vol. 127, p. 8274, 2005. 28. Z.H. Cao, and G.R. Shan, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 47, p. 1522, 2009. 29. S.W. Zhang, S.X. Zhou, Y.M. Weng, and L.M. Wu, Langmuir, Vol. 22, p. 4674, 2006. 30. E. Bourgeat-Lami, I. Tissot, and F. Lefebvre, Macromolecules, Vol. 35, p. 6185, 2002. 31. Z.H. Cao, G.R. Shan, G. Fevotte, N. Sheibat-Othman, and E. Bourgeat-Lami, Macromolecules, Vol. 41, p. 5166,2008. 32. S. Torza, and S.G. Mason, Journal of Colloid and Interface Science, Vol. 33, p. 67, 1970. 33. F.K. Hansen, and J. Ugelstad, Journal of Polymer Science, Polymer Chemistry, Vol. 16, p. 1953,1978. 34. A.J.P. van Zyl, R.F.P. Bosch, J.B. McLeary, R.D. Sanderson, and B. Klumperman, Polymer, Vol. 46, p. 3607,2005 35. M. Heskins, and J.E. Guillet, Journal of Macromolecular Science Chemistry, Vol. 2, p. 1441,1969. 36. X.C. Xiao, L.Y. Chu, WM. Chen, S. Wang, and R. Xie, Langmuir, Vol. 20, p. 5247, 2004. 37. H. Kim, and D.J. Burgess, Journal of Colloid and Interface Science, Vol. 241, p. 509, 2001.

6 Miniemulsion Polymerization of Vegetable Oil Macromonomers Shelby F. Thames*, James W. Rawlins, and Sharathkumar K. Mendon S F Thames, Polymer Science Research Center, The University of Southern Mississippi, Thames-Rawlins Research Group, Hattiesburg, USA

Abstract

Miniemulsion polymerization is well suited for the polymerization of hydrophobic monomers, especially VOMMs that are too hydrophobic to traverse the aqueous phase during conventional emulsion polymerization. Since VOMMs possess some inherent hydrophobicity due to their fatty acid content, they are intrinsically suited for miniemulsion polymerizations, and the resulting polymers are crosslinkable at ambient conditions. While other miniemulsions can be made to crosslink only via in situ grafting, crosslinking in VOMM miniemulsions occurs after film formation is complete, thereby enabling improved flow and leveling during film formation with subsequent oxygen uptake for ambient crosslinking and polymer characteristic changes. Unlike many other vegetable oil derivatives, VOMMs effectively stabilize monomer droplets for miniemulsion polymerization. Novel soybean oil-based VOMMs were synthesized and copolymerized with conventional acrylic and (meth)acrylate monomers at various weight percents ranging from 5 to 70%. Gel content studies confirm retention of the VOMM unsaturation during the polymerization and consequently they cure to solid, high performance polymers via ambient auto-oxidative polymerization, after they are applied to a chosen substrate. Keywords: vegetable oil, acrylic acid, auto-oxidative polymerization, soybean oil, in-situ, gel content, unsaturation, ambient crosslinking, vegetable oil macromonomers, VOMMs, miniemulsions, and emulsion polymerization.

6.1 Introduction and Background Vegetable Oil M a c r o m o n o m e r s (VOMMs) are derived from vegetable oils, i.e., triglycéride esters of fatty acids (Figure 6.1). The chemical V. Mittal (ed.) Miniemulsion Polymerization Technology, (139-172) © Scrivener Publishing LLC

139

140

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY O

II

CH2—O—C

R

CH

R2

I

°

0—C

CH2—0

n

R3

O Ri,R2, and R3 are fatty acid residues

Figure 6.1. Vegetable Oil.

Table 6.1. Common vegetable oil fatty acids. Stearic acid

CH3-(CH2)16-COOH

Oleic acid

CH3-(CH2)7-CH=CH-(CH2)7-COOH

Linoleic acid

CH 3 -(CH 2 ) 4 -CH=CH-CH 2 -CH=CH-(CH 2 ) 7 -COOH

Linolenic acid

CH 3 -CH 2 -CH=CH-CH 2 -CH=CH-CH 2 -CH= CH-(CH2)7-COOH

Eleostearic acid

CH 3 -(CH 2 ) 3 -CH=CH-CH=CH-CH=CH(CH2)7-COOH

Ricinoleic acid

CH3-(CH2)5-CH(OH)-CH2-CH=CH-(CH2)7-COOH

composition of the fatty acids and the proportions in which they are present determine the general properties of the vegetable oils. Furthermore, the fatty acid composition of vegetable oils is affected by environmental conditions such as soil type, growing location, and climatic changes such as humidity and temperature [1]. Most fatty acids contain 18 carbon atoms and are characterized by varying degrees of unsaturation (conjugated or unconjugated) (Table 6.1). Fatty acid chains inherently impart flexibility and hydrophobicity to vegetable oils and their derivatives. The propensity of unsaturated fatty acid chains to react with atmospheric oxygen and polymerize is reflected in the "drying" characteristics of vegetable oils. Drying, in terms of vegetable oil chemistry, defines the oxygen initiated polymerization of various fatty acid chains with subsequent conversion to high molecular weight semi-solid or solid materials. In this respect, conjugated double bonds are more reactive than unconjugated double bonds.

POLYMERIZATION OF VEGETABLE O I L MACROMONOMERS

141

All vegetable oils contain a proportion of saturated acids such as palmitic acid and stearic acid, and the monounsaturated acid, i.e., oleic acid, that make little or no contribution to the drying process. Consequently, vegetable oils are distinguished from one another by their "drying" ability and are segregated via their iodine value (IV) into non-drying oils with IVs of 120 (coconut, cottonseed), semidrying oils with IVs of 120-150 (safflower, sunflower, soybean), and as drying oils with IVs > 150 (linseed oil, tung oil). The fatty acid compositions of various vegetable oils are listed in Table 6.2 [2]. Vegetable oils are a key component of oil-modified polyesters (alkyd resins) synthesized via esterification of vegetable oils (or vegetable oil derived acids and esters) with a variety of acids or anhydrides such as phthalic anhydride, isophthalic acid, or trimellitic anhydride, and polyols such as glycerol, trimethylol propane, or pentaerythritol. Despite their excellent gloss, flexibility, and favorable cost-performance ratio, alkyd resins possess poor hydrolytic resistance due to their ester functionality and tend to yellow during exterior exposure as further auto-oxidation increases visible light absorption through chromophore formation. On the other hand, acrylic derived polymers are easily synthesized via a number of polymerization techniques, including emulsion polymerization, and offer excellent color and color retention on exterior weathering, but coatings based on

Table 6.2. Fatty acid composition of common vegetable oils. Saturated

Oleic

Linoleic

Linolenic

Iodine value

Cottonseed

26

20

54

-

108

Linseed

11

21

18

50

180

Safflower

10

12

78

-

145

Soybean

15

22

55

8

130

Sunflower

12

26

62

-

135

a

5

8

4

3

165

3

3

4

-

85

Oil

Tung

Castor*

Contains 80% eleostearic acid. "Contains 90% ricinoleic acid.

142

MINIEMULSION POLYMERIZATION TECHNOLOGY

petroleum-based monomers sometimes possess limited flexibility. The complementary features of these two major resin types have resulted in significant efforts to produce high quality, waterborne alkyd-acrylic hybrids combining the advantages of both polymer types. Monomer hydrophobicity plays an important and controlling role in emulsion polymerizations. For instance, monomers must possess sufficient water solubility to diffuse through the water phase into the surfactant stabilized micelle where polymerization occurs. Monomers unable to diffuse into micelles will reside in monomer droplets, often unpolymerized, and over time will attempt to coalesce in order to minimize their surface energy. Thus, the hydrophobic nature of each monomer is vastly important both singularly and in relation to each other for efficient and random copolymerization, not only in particle nucleation but also in particle growth throughout the feeding stage. Monomers such as butyl acrylate (BA) and methyl methacrylate (MMA) are used extensively in emulsion polymerization chemistry and despite their general hydrophobicity, possess sufficient water solubility for efficient emulsion polymerization. Table 6.3 lists the water solubility of a number of monomers commonly used in emulsion polymerizations [3]. It can be seen that the low water solubility (<0.001 wt%) of monomers with >12 carbons limits their use in emulsion polymerizations. This Table 6.3. Monomer solubility in water. Monomer

Solubility in water (wt%), 20°C

Methyl acrylate

5.2

Vinyl acetate

2.5

Methyl methacrylate

1.5

Butyl acrylate

0.16

Styrene

0.03

2-Ethylhexyl acrylate

0.01

Vinyl neo-pentanoate

0.08

Vinyl 2-ethylhexanoate

<0.01

Vinyl neo-nonanoate

<0.001

Vinyl neo-undecanoate

< 0.001

POLYMERIZATION OF VEGETABLE OIL MACROMONOMERS

143

is especially true when the polymerization recipe includes monomers with widely varying water solubilities wherein some monomers readily diffuse into micelles and diffusion of others are inhibited by their higher hydrophobicity.

6.2 Emulsion Polymerization of Alkyds and Vegetable Oils Emulsions containing vegetable oil derivatives were designed to combine the advantages of solvent-soluble polyester polymers with the environmentally responsible, human health friendly, "Green" features of waterborne polymers. Unlike traditional latexes, vegetable oil latexes contain vegetable oil/alkyd particles stabilized as oil-in-water emulsions. Choosing the appropriate surfactant and emulsification protocol are critical to ensure colloidal stability of alkyd/vegetable oil derived emulsions. Nabuurs et al. investigated emulsion polymerization of acrylate monomers in the presence of alkyds and noted that acrylate monomer polymerization was severely retarded by alkyd unsaturation reacted in their chosen conditions [5]. However, the alkyd-acrylic hybrid systems were much better than the alkyd-acrylic blends in terms of film formation. For instance, the minimum filming temperature (MFT) of the hybrids was 5°C whereas the MFT of the alkyd-acrylic blend with the same overall composition was 21 °C, which coincided with the glass transition temperature (T ) of the acrylic resin. Studying the same system via mmiemulsion polymerization, Wang et al. were able to achieve copolymerization of the alkyd and acrylate [6]. Only 20-30% of the alkyd double bonds were consumed or lost during the polymerization process under their chosen conditions. The fraction of polyacrylate grafted onto the alkyd increased with alkyd content and ranged from 22-70%. Wu et al. investigated alkyds as a co-stabilizer in miniemulsion polymerization of butyl acrylate and methyl methacrylate and verified the formation of poly(acrylate-graftalkyd) as the predominant polymeric structure [7]. At least 20% alkyd was necessary to ensure stabilization. Only 17-28% of the alkyd double bonds were consumed during the polymerization, van Hamersveld et al. conducted miniemulsion copolymerization of an alkyd (oil length 83%) with MMA, and claimed lack of intraparticle heterogeneity via cryogenic transmission electron microscopy (cryo-TEM) analysis. The authors contended that the use of partly hydroperoxidized sunflower

144

MINIEMULSION POLYMERIZATION TECHNOLOGY

oil as the oil soluble initiator resulted in the formation of triglycéride modified polyacrylate molecules acting as a compatibilizer between the alkyd and PMMA phases [8]. However, in a subsequent study, van Hamersveld et al. reported that an AFM study of the film formation process of the same oil/alky d-acrylic hybrid latexes showed phase separation between the oil and the acrylic phases upon drying. The films consisted of deformed acrylic particles embedded in a continous matrix of oil [9]. Tsavalas et al. noted that hybrid miniemulsion polymerization of an MMA-alkyd system gave incomplete conversion while full conversion was achieved for an identical recipe without the alkyd [10]. The limiting conversion was attributed to the presence of relatively inactive alkyl allyl macroradicals formed due to chain transfer (retardive chain transfer mechanism). A physical explanation was also postulated proposing the partitioning of the monomer-alkydpolymer mixture and the formation of a hard polyacrylate shell that prevented initiator penetration. In a follow-up study, Hudda et al. developed mathematical models to describe the kinetics of an alkydacrylate system, i.e., a homogenous model based on retardive chain transfer kinetics and a heterogeneous model based on an alkyd core and acrylic shell [11]. The authors concluded that retardive chain transfer alone could not explain the limited conversion but the coreshell model in which polymerization occurred in an acrylic-rich shell while the alkyd-rich core served as a reservoir for the acrylic monomer and alkyd did result in correct simulation of the limiting conversion. To circumvent the immiscibility between the alkyds and the growing acrylate chains, Guo and Schork investigated hybrid miniemulsion polymerization of butyl acrylate and methyl methacrylate in the presence of a lower molecular weight analog of alkyds, i.e., linoleic acid and safflower seed oil [12]. The authors reported that hybrid particles from the acrylate/linoleic acid system exhibited homogeneous morphology with degrees of grafting near 100% and a single T while hybrid particles from acrylate/safflower seed oil exhibited lower degree of grafting and a core-shell structure with two T s. Tsavalas et al. studied the morphology of latexes synthesized via miniemulsion copolymerization of a medium oil length alkyd with MMA and BA by transmission electron microscopy (TEM), scanning electron microscopy (SEM), atomic force microscopy (AFM), and nuclear magnetic resonance (NMR) spectroscopy [13]. Hybrid alkyd-MMA systems were shown as a raspberry morphology with alkyd core and PMMA shell. BA/MMA copolymer-alkyd polymers exhibited typical core-shell morphology with alkyd core

POLYMERIZATION OF VEGETABLE O I L MACROMONOMERS

145

and smooth acrylic shell. In a BA-alkyd system, however, a continuous particle phase of PBA with small internally dispersed island domains of alkyd was observed.

6.3 (Meth)acrylated Vegetable Oil Derivatives Bunker and Wool synthesized acrylated methyl oléate (AMO) by epoxidizing methyl oléate with a peroxy acid and reacting the product with acrylic acid [14]. Despite using 15 wt% sodium dioctyl sulfosuccinate as the surfactant for emulsion polymerization, the low water solubility of this macromonomer (approximately 10~7 M) gave only 91% conversion after 18 hours of reaction. Branching was attributed to chain transfer to polymer via hydrogen abstraction from a tertiary backbone C-H bond. In a second study, Bunker et al. adopted miniemulsion polymerization to synthesize pressure sensitive adhesives (PSAs) by copolymerizing MMA with AMO and reported complete conversion within 1 hour using only 2% surfactant by weight [15]. While the AMO and petroleum-based PSAs had comparable tack, shear strength and elasticity, the AMO-based polymers had lower peel strength. The authors proposed that the molecular weight of the AMO polymer needed to be increased via appropriate crosslinking to achieve the desired balance of properties. Spagnola et al. pursued the synthesis of functionalized fatty acid derivatives by adding a free radical polymerizable double bond to a fatty alcohol [16]. Specifically, oleyl alcohol and a mixture of oleyl and linoleyl alcohols were esterified with methacrylic anhydride to yield the respective methacrylated esters, i.e., oleyl methacrylate (OM) and a mixture of linoleyl and oleyl methacrylates (LOM). The comonomer, 2-ethylhexyl methacrylate (EHMA), was chosen for its low T (-10.15°C) and lack of hydrogen abstraction, which can lead to a grafting or crosslinking reactions during polymerization. The OM and LOM were each copolymerized with EHMA via miniemulsion batch copolymerization at 0, 5, 20, and 40 wt% based on monomer phase using a thermal initiator. The authors noted a slight conversion decrease for the EHMA with increasing amounts of OM incorporation and a large drop in conversion for EHMA with increasing concentrations of LOM in the system. These data are consistent with Tsavalas et al.'s results discussed above. Oleyl hydrocarbon chains possess fewer unsaturated sites than linoleyl chains and are therefore less susceptible to hydrogen abstraction. The improved conversion noted with increased reaction times for the LOM system

146

MINIEMULSION POLYMERIZATION TECHNOLOGY

was attributed to retardation effects from hydrogen abstraction being overcome by generation of new radicals from the thermal initiator. Although the radicals formed by chain transfer were sufficiently less reactive and did not reduce the reaction rate, they participated in polymerization reactions and formed a high degree of insoluble polymer (75±2% and 77.6+1% for 3% and 28.2% OM, respectively, and 88.3±1.1% and 86±3.3% for 3% and 28.2% LOM, respectively, after 24 hours). To reduce the amount of hydrogen abstraction and the reactivity of the radicals formed, the polymerizations were repeated at ambient with a redox initiator. The redox-initiated copolymerizations proceeded to completion much faster than the thermally initiated systems and offer significantly lower fractions of insoluble polymer (4.7±3.3% and 12.2±2.4% for 3% and 28.2% OM, respectively, and 39.8+1.6% and 51.1±1.2% for 3% and 28.2% LOM, respectively, after 3 hours). The authors attributed these results to reductions in hydrogen abstraction a n d / or that the fatty acid radical was less reactive in the ambient temperature redox-initiated system.

6.4

Vegetable Oil Macromonomers

In the early 90s, Thames et al. developed a series of vegetable oil derivatives, termed vegetable oil macromonomers (VOMMs), functionalized for efficient incorporation into emulsions [17]. VOMMs are a viable biobased alternative to obviate the need for coalescing solvents in waterborne coatings. Although waterborne coatings are regarded as the closest and environmentally favored alternative to solvent-based coatings, they are often formulated with significant amounts of solvents, i.e. coalescing aids, to facilitate good film formation of high T polymers. The disadvantages of this technology are clear; the volatile organic compounds (VOCs) formulated into the coatings recipe enter the atmosphere during the drying process as pollutants and constitute real and confirmed adverse health effects. VOCs are an environmental concern as they react with atmospheric nitrogen oxides to form ozone. Environmental Protection Agency (EPA) findings have linked ground level ozone to increased asthmatic and respiratory conditions in humans [18]. Even short-term exposure to very low levels of ozone can cause chest pain, coughing, nausea, throat irritation, congestion, and reduced lung capacity. In addition, ozone can exacerbate cardiac and lung conditions such as bronchitis, asthma, pneumonia, emphysema, and heart disease.

POLYMERIZATION OF VEGETABLE O I L MACROMONOMERS

147

In view of the detrimental effect of ozone, the EPA imposes restrictions on the maximum VOC content permissible in coatings. VOMMs have three distinct structural and design characteristics favoring their use for high performance environmentally-responsible emulsions: a. VOMMS, by virtue of their large monomer size, are excellent coalescing aids (plasticizing monomers) and facilitate smooth film formation at zero VOC, b. VOMMs copolymerize with other monomers via their acrylate and (meth)acrylate functionality, and consequently are retained in the final film where they become an integral part of the applied coating, and c. VOMMs, when forming part of a polymer system, crosslink to higher molecular weight polymers during ambient cure. Thus, VOMM-derived polymers function as thermoplastic polymers during application and thermosetting polymers as auto-oxidative polymerization progresses during ambient cure. The overall objective of the polymer design process is to achieve and optimize control of the MFT and final T (after auto-oxidation) of VOMM emulsion polymers. This can be accomplished by choosing appropriate amounts and type of VOMMs for incorporation into the latex polymer composition. Emulsion polymers synthesized in this way represent the synergism of two technologies, i.e., alkyds and acrylics, where the advantages of each are retained and the disadvantages minimized. As evident from Bunker et al.'s work discussed above, achieving the appropriate degree of auto-oxidative crosslinking is critical if one is to control and utilize the inherent plasticization efficacy of the VOMM fatty acid chains. In general, unmodified vegetable oils lack reactive functional groups with the exception of a few oils such as castor and lesquerella oils (which are hydroxyl functional), while vemonia oil offers epoxide functionality. The hydroxyl moiety is a convenient site to incorporate polymerizable acrylate and methacrylate functionalities. Consequently, acrylated castor oil methyl esters (Figure 6.2) and acrylated castor oil (Figure 6.3) were among the first VOMMs to be synthesized and studied. The synthesis process has since been expanded to include other vegetable oils and reactants to produce a variety of VOMMs. Triglycérides are fatty acid esters, and compositionally, the presence of ester groups, either from the glyceryl backbone or introduced

148

M I N I E M U L S I O N POLYMERIZATION T E C H N O L O G Y CH 3 —0—C—(CH 2 ) 7 —CH = CH —CH 2 —CH —(CH 2 ) 5 —CH 3 O

O

Figure 6.2. Acrylated castor oil methyl ester. O O

II

¡I

H 2 C=CH — C — O

I

-O — C — (CH2)7 — CH = CH—CH 2 — CH — (CH 2 ) 5 —CH 3 O O

lí

II

H 2 C=CH — C — O

I

HC—O —C—(CH 2 ) 7 —CH = CH —CH 2 —CH —(CH 2 ) 5 —CH 3 O O

II

H2C=CH— C— O

I

H 2 C—O —C—(CH 2 ) 7 —CH = CH—CH 2 —CH —(CH 2 ) 5 —CH 3

Figure 6.3. Idealized acrylated castor oil (drawn with all three fatty acids as ricinoleic acid derivatives, typically the oil is limited to about 80% ricinoleic acid).

during derivatization steps, reduce long-term stability. Furthermore, the likeliness of chain transfer to allylic double bonds during free radical polymerization posed significant challenges, especially with VOMMs based on vegetable oils containing higher amounts of unsaturation. To overcome these hurdles, several VOMMs were designed and synthesized by modifying the composition, connectivity and length of polymerizable acrylate or methacrylate moieties linked with the fatty acid chain. A few such VOMMs are listed in Figure 6.4 [19,20,21,22]. The novel VOMMs were studied to determine the positive attributes for incorporation in the polymer backbone via emulsion polymerization, and achieving the desired final film properties. The areas of study include: 1. 2. 3. 4. 5.

Solubility parameter, Water solubility/ water partitioning, Functionality, Molecular weight, Polymer design for retention of allylic unsaturation (triene content minimization), 6. Color, 7. Propensity for degradative chain transfer,

POLYMERIZATION OF VEGETABLE O I L MACROMONOMERS

149

8. Emulsion polymer and particle stability, 9. Pre-cured and post-cured polymer T changes as a function of polymer composition, and 10. VOMM plasticization efficacy for film formation.

O

O

I!

H

il

H

I

O

! ü

H 2 C—CH —C—OCH 2 CH 2 —O —C — N — R — N — C — O R ' — O H OCN —R — NCO ¡s a diisocyanate (A)

HO — R'—OH is a castor oil ora vegetable oil monoglyceride (diol) O

H

¡I

H

I

O

HO—R" -O — C — N — R - -N — C- OR'—X OCN — R —NCO is a diisocyanate HO — R'—X is hydroxy vinyl ether or hydroxy (meth)acrylate (B)

HO — R"—OH is a castor oil or a vegetable oil monoglyceride (diol) O

II

-R —N —C—OR'—OH - C — OR'H2C = C H — R — NCO is an acrylate isocyanate (C)

HO — R"—OH is a castor oil or a vegetable oil monoglyceride (diol) O

I!

O H

II

I

H O

I

II

O H

II

I

H O

I II

H,C=HC—C—OCHpCH, - O - C —N —R —N —C—OR'—OC—N—R — N — C - O X X —OH is a fluorinated alcohol OCN — R — NCO is a diisocyanate (D)

HO — R'—OH is a castor oil or a vegetable oil monoglyceride (diol)

(E)

XisHorCH,

Figure 6.4. VOMM Range. (Continued)

150

MINIEMULSION POLYMERIZATION TECHNOLOGY

CH, OH

CHo—O — C — R , (F)

N-CH?-CHp-0-C-C=CH? II C(CH3 O

o CH 2 ~- 0 - - C — \ ^ - v

v / ~ \ ^

o

CH-- 0 - - C - R ,

í

\

r

0=J

o il

OH

CH 2 - - 0 - - C — R 2 (G)

Ri

V=0

v

- ^ ' ^ ^

OH

0 X II I 0 - C H 2 - C H - C H 2 -- 0 - C - C = C H 2

XisHorCH3 and R2 are fatty acid chains

OH

CH 2 —O—C—R.

^ \ / ^

N-CH2-CH=CH2

I

H

(H)

R, and R2 are fatty acid chains

CH—0—C—R,

I

o

CH 2 — O — C — R (I)

R-0-CH=CHP

R, and R2 are fatty acid chains, HO-R-0-CH=CH2 is a vinyl ether

Figure 6.4. V O M M Range.

6.5 The Potential for Emulsion Polymerization of Model Saturated Monomers VOMMs are based on unsaturated fatty acids containing eighteen carbon atoms and intrinsically possess very low solubility in water. Moreover, when the fatty acid or vegetable oil derivative, the loss

POLYMERIZATION OF VEGETABLE O I L MACROMONOMERS

151

of a functional group capable of hydrogen bonding further reduces its aqueous affinity. Figure 6.5 demonstrates that saturated carbons are the source of unfavorable interactions with the water phase. Ruelle and Kesselring quantified that addition of two méthylènes to a parent functional molecule (alcohols, acids, alkanes, methyl esters, or amines) reduced the molecules water solubility by an order of magnitude [23]. Ostwald ripening rates, i.e., the diffusion rate of monomer from smaller to larger droplets brought about by a difference in droplet pressures, are determined by measuring the droplet particle size over time. Under these conditions, the diffusion rate of the most hydrophobic monomer is proportional to the rate of droplet growth, allowing their diffusion rate to be determined using the Lifshitz-Slyozov-Wagner relationship. Alkanes bearing more than ten carbons have the capability to kinetically stabilize emulsions by retarding monomer flux from smaller to larger particles. After homogenizing the dispersion with a microfluidizer or sonicator, each monomer droplet must maintain an equal concentration of hydrophobic material. Ripening readily occurs when alkanes with less than ten carbons are employed, leading to phase separation within hours. On the contrary, hexadecane and octadecane imparts stability for days or months. Gilbert concluded

Figure 6.5. Water solubility versus number of carbons for acids, acrylates, alcohols, alkanes, and methyl esters.

152

MINIEMULSION POLYMERIZATION TECHNOLOGY

that homogeneous nucleation does not occur when a C12 acrylate is emulsion polymerized due to lack of monomer in the aqueous phase. Furthermore, Chern and Lin investigated lauryl methacrylate, a C12 methacrylate, as a stabilizer in miniemulsions and initiated ca. 30% of the monomer droplets [24]. Matsumoto et al. observed a 4.2 hour and 2.3 hour induction period of a C12 and C10 methacrylate homopolymerization, respectively, using conventional macroemulsion techniques [25]. On the contrary, an induction period was not noted when MMA and hexyl methacrylate were polymerized under similar conditions, and they concluded the induction periods resulted from the hydrophobic monomers inability to form oligomeric radicals (z-mers) that are generally thought to initiate polymerization. Therefore, these findings indicate diffusional limitations are encountered in macroemulsion when monomers have less than 10"4 wt% water solubility. The diffusion rates of lauric (C12), palmitic (C16), and stearic (C18) fatty acids have been measured to be ca. 10 7 cm 2 /sec indicating that the acid group does not supplement the polarity sufficiently to offset the hydrophobic effect [26]. Moreover, extremely nonpolar hexadecane yields diffusivity ca. 10®cm2/sec. In an emulsion polymerization, the monomer's water solubility chiefly controls its diffusion rates and surfactant has been shown to have minor effects on these rates. Anionic surfactants were observed to slightly increase Ostwald ripening rates upon increasing its concentration thirty times [27]. Nonionic surfactants enhance diffusion to a similar degree except when added to a pre-dispersed emulsion. When homogenizing oil droplets in the presence of a nonionic surfactant, the emulsion degrades by molecular diffusion of oil through the continuous phase, and micellar diffusion is enhanced when a large concentration of nonionic surfactant is added to a pre-formed emulsion. In the latter scenario, the mass transport mechanism has been shown to be a function of particle fusion-fission phenomenon and molecular diffusion [28]. Therefore, only monomers having moderately different water solubilities can be influenced by the choice of surfactant and processing conditions.

6.6 Nucleation Mechanisms Several mechanisms could be active in the particle formation stage of conventional, macro- and miniemulsion polymerization including micellar nucleation [29,30], homogeneous nucleation [31], and droplet initiation [32]. The dominant mechanism is chiefly

POLYMERIZATION OF VEGETABLE OIL MACROMONOMERS

153

determined by the monomer(s) hydrophilicity, surfactant concentration, and droplet's particle size. For example, micellar nucleation dominates when slightly water soluble monomers are polymerized in the presence of copious surfactant and large monomer droplets (>1μπ\). Homogeneous nucleation dominates when very hydrophilic monomers are copolymerized with surfactant levels below the CMC (or surfactant-free). Finally, monomer droplet nucleation can be produced using very hydrophobic monomers, a [S] < CMC, and intense homogenization to reduce the droplets size. Hansen and Uglestead first recognized droplet nucleation to be a dominant mechanism in miniemulsions, and they demonstrated that a very hydrophobic "cosurfactant" or hydrophobe sufficiently retards diffusional degradation to allow radical penetration into the droplets [32]. Typically, cosurfactants are fugitive, lipophilic substances, e.g., hexadecane; however, reactive cosurfactants are also effective in reducing Ostwald ripening. Chern and Sheu utilized stearyl methacrylate (SMA) at low concentrations (24mM) to demonstrate the effect of methacrylic acid on particle nucleation [33], They reported slightly enhanced droplet nucleation upon addition of acrylic acid or methacrylic acid to a miniemulsion polymerization. While miniemulsion has the advantage of copolymerizing diffusionally limited monomers, secondary nucleated particles can be formed and compete with the droplets as polymerization sites. This phenomenon is favored when the [S] > CMC and slightly water soluble monomers are employed, and miniemulsion polymerizations employ water insoluble monomers with the surfactant concentrations at or below the CMC to avoid micellar nucleation. However, commercial latexes are usually manufactured with surfactant levels higher than the CMC to impart shear stability and reduce coagulum. Higher concentrations of a reactive hydrophobe, SMA or stearyl acrylate (SA), were than what is routinely employed were investigated, and a slightly water soluble monomer, MMA, was copolymerized with the reactive hydrophobe at [S] » CMC. The crystalline character of SA and SMA emulsion copolymers allowed probing its distribution amongst the polymer chains. Jordan et al. explored the reactivity and thermal transitions of SA and methyl methacrylate (MMA) solution copolymers, and found a minimum of forty monomer wt% SA was required to produce crystalline domains in SA/MMA solution copolymer [34]. This research formulated SA and SMA at lower concentrations, ten monomer wt%, in macro- and "pseudo" miniemulsion conditions. The broadness and

154

MINIEMULSION POLYMERIZATION TECHNOLOGY

area of the melting point was correlated to the extent of copolymerization occurring inside monomer droplets. As previously mentioned, SA was emulsion copolymerized at 30 wt% with BA using a phase transfer catalyst, ß-cyclodextrin, to carry SA into the micelles [35]. This research showed random incorporation of SA in emulsion copolymers decreases its T and produced amorphous domains at these levels. However, without the catalyst, no T decrease was observed and crystalline melting peaks were observed, indicating SA was locally concentrated in distinct domains formed by two polymerization sites.

6.7 Design of Thermosetting Latex Polymers While there are a number of variables that must be considered in the design of thermosetting latexes, secondary particle nucleation or homopolymerization of crosslinker in the aqueous phase or droplet is detrimental to network formation. The difficulties encountered in forming effective network formation result from kinetic or thermodynamic implications or a combination of the two, and the major issues are listed below: 1. Crosslinking before complete coalescence 2. Low crosslinker concentration near the particle interface 3. Homopolymerization of very water soluble crosslinking comonomer during synthesis 4. Homopolymerization of water insoluble crosslinking comonomer during synthesis 5. Competitive particle nucleation mechanisms Bufkin and Grawe first reported crosslinked films by introducing functional monomers in a two-stage monomer addition method that increases the probability of having functional groups near the surface [36]. Complete network formation is dependent on competitive crosslinking reaction rates and polymer diffusion coefficients [37-39]. Maintaining a balance between these two competing phenomena is vital to the design of ambient curable latexes. This issue is not as critical when auto-oxidative crosslinking mechanisms are employed due to their considerably reduced reaction rates as compared to other ambient cured polymers. During macroemulsion copolymerization, monomers exhibiting large differences in both

POLYMERIZATION OF VEGETABLE O I L MACROMONOMERS

155

aqueous phase partitioning and reactivity ratios have increased liveliness of creating non-uniform monomer-particle distributions. Comonomers exhibiting similar reactivity ratios and water solubilities produce polymers and films with more evenly distributed crosslinking sites throughout the particles. Semi-batch feeding is capable of altering the comonomer distribution of latexes synthesized with comonomers possessing moderate differences in reactivities or water solubilities. However, this processing method cannot improve the copolymerization when two significantly different solubility (water solubility/hydrophobicity) comonomers are macroemulsion copolymerized [40]. Distinctly different copolymer compositions are generated when copolymerizing water soluble functional monomers, e.g., acrylic acid, with other slightly water soluble comonomers, e.g., glycidyl methacrylate, to produce a crosslinked film [38,41]. A significant fraction of the water-soluble functional comonomer will homopolymerize in the aqueous phase yielding a broad comonomer composition distribution or a blocky like copolymer [42]. For example, when BA is copolymerized with sodium acrylate, homopolymerization primarily occurs in both organic and aqueous phases. Very little copolymerization occurs between BA and sodium acrylate. On the other hand, when MMA and sodium acrylate are copolymerized, more homogeneous copolymers are formed because MMA's water solubility is an order of magnitude greater than that of BA [43]. Random copolymerization of insoluble and slightly soluble monomers requires reduced monomer feed rates to create monomer starve feed conditions. When a functional monomer homopolymerizes, complete network formation cannot occur, leading to ineffective network formation [44]. For instance, acrylic acid and methacrylic acid (MAA) are thermodynamically driven to the particle/water interface due to their increased polarity and hydrogen bonding. However, at pH > 6, MAA fully partitions into the aqueous phase [45]. Utilizing semibatch methods at a reduced pH permit acid functional groups to be present on the particle's surface for interparticle crosslinking. In a batch polymerization, the comonomers' reactivity ratios and water solubilities chiefly determine the copolymers microstructure and distribution throughout the particles using the macroemulsion technique. Comonomers with similar polymerizable moieties, e.g., acrylic esters, display similar reactivity ratios and the resulting polymers chemical and particle compositional distribution is chiefly controlled by the comonomers partioning behavior between the

156

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

aqueous phase, polymeric particles, and monomer droplets. Very water soluble or water insoluble comonomers (Figure 6.6) that do not diffuse into the polymerization locus before free radical polymerization occurs, broadens the chemical composition distribution and in extreme circumstances, generates two distinct polymers in separate particle populations. Figure 6.6 displays water solubility values of monomers commonly used in latexes, along with two fatty acids for comparison [46,47,48,49]. Fatty acids have negligible water solubility due to the presence of eighteen nonpolar carbons incapable of hydrogen bonding. Since the fatty structures are the fundamental building blocks for VOMMs, their hydrophobicity will be carried over, and the miniemulsion technique must be used to copolymerize the macromonomers with typical acrylic monomers. Furthermore, GPC analysis (Figure 6.7) did not indicate the presence of residual fatty acid, and the product's reduced retention

Figure 6.6. Monomer water solubility chart where DMF= Dimethylformamide, MAA = Methacrylic Acid, HEA = Hydroxyethyl Acrylate, HEMA = Hydroxyethyl Methacrylate, 2-EHMA = 2-Ethylhexyl Methacrylate, f-Bu Styrene = ferf-Butyl Styrene, and Ricin. Acid = Ricinoleic Acid.

POLYMERIZATION OF VEGETABLE O I L MACROMONOMERS 240.00

J i PEGMM /< I J

LLPEGMM

220.00

157

' ill

200.00

' i!« I 'ί'

180.00 160.00

i

140.00 120.00

1

100.00

i

80.00

/

i

V ,1 y III

60.00

77 ü

40.00

1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00 Minutes

Figure 6.7. GPC overlay of LLPEGMM and PEGMM.

times indicate the ethoxylated methacrylate was esterified. The number of peaks in the PEGMM and LLPEGMM chromatograms reveals the distribution of ethoxylated repeat units. Both analyses have more than six distinct n values most likely ranging from 3 to 9. Literature data on similar ethoxylated molecules correlated well with the measured distribution. As the average number of ethoxylated units increases, broader distributions are obtained. Our analysis indicates thirteen distinct n values of PEGMM having an average of ten repeat units. Ferguson et al. also reported larger distributions of repeat units as the number of ethoxylated repeat units with increasing n values. RP-HPLC chromatography is a powerful tool for analyzing reaction conversion while providing estimates for each of the analyte's partition coefficient. Figure 6.8 shows RP-HPLC chromatograms comparing stearyl acrylate (very hydrophobic), LLPEGMM, and common monomers used in emulsions. The four monomers with the shortest retention times do not have diffusional limitations, and analyte water solubility decreases with increasing retention time. Comparing the retention times of stearyl acrylate to the ethoxylated VOMM, shows ethylene oxide's hydrophilic contributions. Combining the six possible ethoxylated methacrylates having different n values with the four fatty acids in soybean oil produces 24 different ethoxylated VOMMs and explains the broad spectrum

158

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY -20% Non-reacted Ethoxylate Methacrylate 250

Λ

200

> CD O)

c

o

Q. C/> CU

Œ

150 50 H

Ethoxylated VOMM

!.>"""

0

% -100 ω ω -150-1 £

2-EHA

100

-50 H

Stearyl Acrylate

MMAI

! BA

-200-250-3006

8

10 Retention Time (min)

12

14

Figure 6.8. RP-HPLC of typical emulsion monomers, LLPEGMM, and SA.

observed in Figure 6.5. Most of the monomer is more hydrophobic than an eight-carbon acrylate, i.e., 2-EHA, and the mean fraction of LLPEGMM has water solubility comparable to a ten-carbon acrylate. Most of the monomer fractions are more hydrophobic than a twelve-carbon acrylate; while larger n values and unsaturated fatty acids produce more water soluble VOMMs. The hydrophobicity of a twelve-carbon acrylate was shown to stabilize monomer droplets; however, using the semi-continuous feed method can minimize these effects.

6.8

Classifying Monomer Solubility for Macro and Miniemulsion Polymerization

The type of emulsion polymerization employed is determined by the comonomer water solubilities, concentration of surfactant(s), ([S]), and desired polymer microstructure. As stated previously, semibatch processing renders slightly more homogeneous copolymerization of comonomers varying in water solubilities by less than one or two orders of magnitude. Warson and Fitch divided latex monomers into four categories based on aqueous affinity.

POLYMERIZATION OF VEGETABLE O I L MACROMONOMERS

159

1. Very water insoluble monomers, e.g., styrene and vinyl stéarate 2. Appreciably water soluble monomers producing water insoluble polymers, e.g., methyl methacrylate and acrylonitrile 3. Appreciably water soluble monomers producing slightly hydrophilic polymers, e.g., vinyl acetate 4. Highly water soluble monomers producing very hydrophilic polymers, e.g., MAA and acrylic acid Under special conditions, comonomers of group 4 can be macroemulsion copolymerized with comonomers of groups 2 and 3. Group 1 monomers are not readily incorporated using traditional macroemulsion polymerization and when they are copolymerized with group 4 monomers homopolymerization will occur in separate phases [50,51]. Only monomers possessing water solubility greater than 10"7wt% are capable of diffusing through the aqueous phase on a reasonable timescale. Other monomers cannot diffuse out of the monomer droplet sufficiently quick to match the reactions timescale and ultimately become concentrated in hydrophobic monomer droplets as the more water soluble monomers are moving across the aqueous phase to growing particles, creating two distinct particle populations. On the contrary, miniemulsion polymerization methods evenly distribute hydrophobic monomers into micelles and minimize or eliminate the necessity of macroemulsion phase transfer from larger monomer droplets. Monomer flux between droplets and particles is inhibited to maintain equal hydrophobe concentration throughout the particles. A reduction in monomer diffusion combined with a low free surfactant concentration promotes monomer droplet nucleation while decreasing the probability of homogeneous and micellar nucleation. Moreover, homogeneous nucleation is further suppressed by polymerizing water insoluble monomers. The inherent hydrophobicity of oils and oil-based derivatives limit the ability of VOMMs to migrate from the monomer droplets to the site of polymerization through the aqueous phase. Therefore, it was necessary to functionalize the VOMM to increase its hydrophilic character and enhance its ability to be used as a copolymerizable hydrophobe for miniemulsion polymerization. Unlike other hydrophobes, the acrylate and methacrylate functionalities improve covalent bonding to form polymers while the VOMM sites of allylic unsaturation retain their potential for crosslinking during and after film formation.

160

6.9

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Soybean Acrylated Monomer Synthesis

Soybean oil (500 g, 0.56 mole) and 83.5 mg of 2-mercaptobenzothiazole were heated to 100°C under a nitrogen purge of 1 hour to remove the oxygen before adding 111.8 g (1.12 moles) maleic anhydride (MA) to the blend. The reaction temperature was increased to 215°C and maintained for 2.5 hours to yield maleinized soybean oil. Next, hydroxyethyl acrylate (26.95 g), phenothiazine (0.153 g), and 1-methyl imidazole (0.29 g) were added to the reactor. The reaction was heated to 115°C and maintained at that temperature for 2.5 hours to yield soybean acrylated monomer (SAM) (Figure 6.9). The FTIR spectrum of SAM shows absorptions at 1785 and 1853 cm"1 that arise from the stretch vibrations of the residual maleic anhydride (Figure 6.10). The absorption at 3010 cm'1 and 1650 cm"1 result from the C-H and C=C stretch vibrations, respectively, of the fatty acid double bond. Acrylate (-CH=CH 2 ) absorptions are noted at 919 cm'1 and 984 cm 1 -while maleic acid carboxylic group is seen at 1408 cm"1 and 1750 cm 1 , although the latter is partially overlapped by the ester carbonyl group. The wide and low absorptions at 2500 cm'1 and 3400 cm 1 are attributed to -OH vibration.

6.10

Miniemulsion Polymerization

Unlike conventional emulsion polymerization, miniemulsion polymerization occurs inside monomer droplets. Energetic homogenization is essential to achieve the desired submicron droplet size, and the droplets are protected against diffusional degradation and

'^/wwvvvvvvvvwvv

- fatty acid chain

Figure 6.9. Soybean acrylate monomer (SAM).

POLYMERIZATION OF VEGETABLE O I L MACROMONOMERS

161

8-

6-

1

4000

1

3500

1

1

3000

Figure 6.10. FTIR spectra of S A M .

1

1

2500

1

1

2000

1

1

1500

Wavenumber (cm-1)

1

1

1000

1

1—

500

droplet coagulation by using a hydrophobe/costabilizer and an appropriate surfactant. Homogenization is typically achieved via the use of rotor-stator systems, sonifiers, or high-pressure homogenizers. Hydrophobie materials employed in miniemulsions include low molecular weight molecules [52], pigments [53], and oligomers or polymers [54-56]. The low water solubility of the hydrophobes prevents their diffusion through the water phase and keeps all other comonomers inside the droplet, thus minimizing the system's free energy. Since most hydrophobes are not covalently bonded to the polymer, they are prone to leaching from the coating or phase separation, prompting research efforts directed at developing reactive hydrophobes [57-59]. Typically, the surfactant is dissolved in water while the costabilizer is dissolved in the monomers, and the mixture is subject to high energy homogenization. For instance, when cetyl alcohol is used as a costabilizer, it can be dissolved in the surfactant solution by heating beyond its melting point, at which time the mixture is cooled and sonicated to break the gel phase before adding monomer and proceeding with homogenization to form droplets. With SAM, monomer droplets were generated by inverting a water-in-oil (w/o) emulsion, a process referred to as catastrophic phase inversion. SAM, ammonium bicarbonate, anionic surfactant

162

MINIEMULSION POLYMERIZATION TECHNOLOGY

Rhodapex® CO-436 [ammonium salt of nonylphenol-4-ethoxy sulfate (58 wt%) in water (27 wt%) and ethanol (15 wt%)], and deionized water were added to a reaction kettle, and placed in a 65°C water bath (Table 6.4). The contents were stirred at 100 rpm for 10 minutes to form a water-in-oil emulsion at which time phase inversion to an oil-in-water emulsion was affected with the continued addition of water. Samples were subjected to particle size analysis and stability evaluations. A temperature of 80°C was chosen for the polymerization. BA and MMA were added continuously over 20 minutes to swell the already formed VOMM droplets, and the system was allowed to equilibrate for an additional 30 minutes at which time a sample was collected for particle size analysis to verify droplet swelling. The miniemulsion was polymerized in a semi-continuous mode by adding ammonium persulfate over a two-hour period. For comparison, acrylic controls [control 1 (BA/ MMA), and control 2 (BA/MMA/Acrylic acid)] were synthesized via conventional emulsion polymerization (Table 6.5). Monomers and initiator were added over a two-hour period. Particle size distribution analysis was used to follow the nucleation mechanism and particle growth of the polymerization process. Figure 6.11 (0 min, immediately before initiator addition) confirms the presence of large monomer droplets (200-2200 nm). Figure 6.12 (6 min) shows generation of small particles (50 nm) by micellar nucleation as predicted for emulsion polymerization when the surfactant concentration is above the critical micelle

Table 6.4. Miniemulsion formulation. Materials

Weight (g)

SAM

16.95

Ammonium bicarbonate

0.42

Rhodapex CO-436

1.30

Deionized water

112.24

Butyl acrylate

16.50

Methyl methacrylate

12.65

Ammonium persulfate

0.83

POLYMERIZATION OF VEGETABLE O I L MACROMONOMERS

163

Table 6.5. Conventional emulsion formulation. Materials

Weight (g)

Rhodapex CO-436

13.10

Ammonium bicarbonate

1.65

Water

307.70

Butyl acrylate

165.00

Methyl methacrylate

126.50

Acrylic acid Ammonium persulfate

0.00/10.00 2.42

Figure 6.11. Particle size analysis of monomer droplets. concentration (CMC). The figure also shows disappearance of monomer droplets early in the reaction; since they act as monomer reservoirs, as monomer diffuses from the droplets to the formed particles; the droplets shrink in size and eventually disappear. Yet another reason for droplet disappearance is that particles outnumber droplets after the nucleation stage, and since the simple particle size analysis system cannot detect distributions below 1 %, the droplets may not be identified even if they are present. Figures 6.12 (25 min) and 6.12 (90 min) show particle growth throughout the feeding stage (from 60 nm at 25 min to 110 nm at 90 min), which is another characteristic of conventional emulsion polymerization and contrasts with the miniemulsion results discussed below.

164

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Figure 6.13. Monomer droplet distribution of miniemulsion containing SAM.

Figure 6.13 illustrates the stability of monomer droplets containing SAM. No change in monomer droplet size distribution was detected even after a month under ambient conditions. It is believed that SAM's hydrophobicity prevents the Ostwald ripening effect, and translates to monomer droplet stability in the experimental conditions. After confirming monomer droplet stability, additional comonomers were added to swell the droplets, and the miniemulsion was polymerized by adding the initiator as a continuous feed. Particle size analysis was performed throughout the polymerization process. Figure 6.14 indicates that the final particle size distribution

POLYMERIZATION OF VEGETABLE O I L MACROMONOMERS

165

Figure 6.14. Particle size distribution throughout miniemulsion polymerization.

replicates the initial monomer droplet distribution exhibited at the start of the polymerization, suggesting that droplet nucleation was the predominant mode as dynamic laser light scattering is unable to detect particles at <1% concentration. Above its CMC, Rhodapex CO-436 displays surface tension values of 32.2 m N / m . The fact that the miniemulsion surface tension values before and after polymerization are higher than 32.2 m N / m adds additional confirmation that the miniemulsion polymerization proceeded with minimal micellar nucleation (Table 6.6) as desired. The minimal change in surface tension before and after miniemulsion polymerization further validates that droplet nucleation was minimized. On the other hand, with conventional emulsion polymerization, the surface tension increased from 32.4 m N / m before polymerization to 38-40 m N / m after polymerization that is attributed to adsorption of free surfactant required to stabilize the high number of growing particles as they increased in size and typically shrink in quantity. An additional nucleation mechanism is suggested as a small particle size distribution (50 nm) was noted during conventional emulsion polymerization. The ability of the particle size analyzer to detect micellar nucleation was confirmed by adding additional surfactant to a miniemulsion before polymerization. Figure 6.15 shows the particle size distribution of a miniemulsion conducted

166

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Table 6.6. Surface tension of miniemulsion and conventional emulsion polymerization. Before polymerization (mN/m)

After polymerization (mN/m)

Control 1

32.4

38.9

Control 2

32.6

40.3

Miniemulsion

36.5

37.7

Figure 6.15. Droplet and particle size analysis before and after polymerization with surfactant concentration above CMC.

by increasing the surfactant concentration by 1 g. The smaller size particles (60 nm) correspond to micellar nucleation while the larger particles correspond to droplet nucleation. In this case, the number of particles nucleated by micellar nucleation was high enough to be detected by the dynamic light scattering analyzer. To ensure that the final analysis corresponds to particles and not monomer droplets, monomer content was measured throughout the polymerization (Figure 6.16). Residual monomer after polymerization was below 1000 ppm reaffirming that the distribution corresponds to polymer particles and not monomer droplets. The miniemulsion gel content was compared with that of two conventional emulsion polymers (Table 6.7). The 0% gel content of

POLYMERIZATION OF VEGETABLE O I L MACROMONOMERS

167

Figure 6.16. Monomer content during miniemulsion polymerization.

Table 6.7. Gel content of conventional emulsion polymers and miniemulsions. VOMM content

% Gel Liquid

Film, air

Film, N 2

Control 1

0%

0.30

0.34

0.32

Control 2

0%

4.70

25.08

24.90

Miniemulsion

35%

0.00

80.50

36.50

the miniemulsion in its liquid state indicated minimal, if any, highly branched or crosslinked chains were produced and support the preservation of VOMM unsaturation during miniemulsion polymerization. A high gel content (80.5%) was quantified after the miniemulsion was applied as a thin film and dried at ambient conditions. These data confirm crosslinking during the air-drying process. On the other hand, the same miniemulsion placed in a nitrogen atmosphere where oxidative cure was unlikely due to the absence of oxygen, showed only moderate gel content (36.5%). The detectable gel content was attributed to and consistent with a combination of residual hydroperoxide formation prior to application (during monomer

168

MINIEMULSION POLYMERIZATION TECHNOLOGY

and polymer synthesis) and radical propagation of already activated allylic double bonds. The lesser degree of crosslinking is typical of systems starved of oxygen during the curing process.

6.11

Conclusions

Vegetable oils have been used in many polymer types designed principally for the coatings industry. They provide plasticization and organic solubility to polymer systems, aid in application, improve adhesion, and penetrate porous substrates very effectively, yet will readily cure at ambient via auto-oxidative polymerization in air to yield crosslinked polymers with excellent performance properties. With the advent of synthetic resins, acrylics achieved popularity due to their excellent color and color retention, especially upon exterior exposure. However, neither technology met all industry needs in that vegetable oil derived polymers were soluble in organic solvents and thus were high in VOCs, and consequently are significant contributors to smog formation and the attendant health issues. While the acrylic and vinyl-acrylic systems form excellent polymers via emulsion polymerization, they require the addition of coalescing aids classified as VOCs and lack the flow and leveling attributes of the vegetable oil derived polymers. While many researchers have attempted to combine, or copolymerize as it were, alkyd and acrylic polymer types to achieve the positive attributes of both, success has been sporadic and questionable in many cases until now. VOMMs offer a viable solution to this challenge by incorporating fatty acid chains into an acrylic backbone via emulsion polymerization. Furthermore, VOMM derived polymers possess no VOCs and the internal plasticizing effects of the VOMMs allow formulation of zero to low VOC and essentially "green" products. Miniemulsion polymerization is well suited for the polymerization of hydrophobic monomers, especially VOMMs that are too hydrophobic to traverse the aqueous phase during conventional emulsion polymerization. Since VOMMs possess some inherent hydrophobicity due to their fatty acid content, they are intrinsically suited for miniemulsion polymerizations, and the resulting polymers are crosslinkable at ambient conditions. While other miniemulsions can be made to crosslink only via in situ grafting, crosslinking in VOMM miniemulsions occurs after film formation

POLYMERIZATION OF VEGETABLE O I L MACROMONOMERS

169

is complete, thereby enabling improved flow and leveling during film formation with subsequent oxygen uptake for ambient crosslinking and polymer characteristic changes. Unlike many other vegetable oil derivatives, VOMMs effectively stabilize monomer droplets for miniemulsion polymerization. Novel soybean oil-based VOMM were synthesized and copolymerized with conventional acrylic and (meth)acrylate monomers at various weight percents ranging from 5 to 70%. Gel content studies confirm retention of the VOMM unsaturation during the polymerization and consequently they cure to solid, high performance polymers via ambient auto-oxidative polymerization, after they are applied to a chosen substrate. VOMMs are ideal for industrial/commercial use as they will facilitate increased commercialization of alternative crops in environmentally responsible coatings, reduce the consumption of non-renewable resources, eliminate the need for organic solvents, coalescing aids or plasticizers, and provide environmentally friendly polymers and "Green" products for use in coatings and adhesives.

References 1. D. Swern, Bailey's Industrial Oil and Fat Products, 4th ed., New York, John Wiley & Sons, 1979. 2. Surface Coatings, 3rd ed., Journal of Oil and Colour Chemists Association, Australia. 3. C. Quintero, Vegetable Oil Macromonomer Kinetics and Loci during Emulsion Polymerization, Ph. D. Dissertation, Dec 2005. 4. G. Östberg, and B. Bergenstáhl, Journal of Coatings Technology, Vol. 68, p. 39, 1996. 5. T. Nabuurs, R.A. Baijards, and A.L. German, Progress in Organic Coatings, Vol. 27, p. 163,1996. 6. S.T. Wang, FJ. Schork, G.W. Poehlein, and J.W. Gooch, Journal of Applied Polymer Science, Vol. 60, p. 2069,1996. 7. X.Q. Wu, F.J. Schork, and J.W. Gooch, Journal of Applied Polymer Science (Polymer Chemistry), Vol. 37, p. 4159,1999. 8. E.M.S. van Hamersveld, J.J.G.S. van Es, and F.P. Cuperus, Colloids & Surfaces A, Vol. 153, p. 285,1999. 9. E.M.S. van Hamersveld, J.J.G.S. van Es, AL. German, F.P. Cuperus, P. Weissenborn, and AC. Hellgren, Progress in Organic Coatings, Vol. 35, p. 235, 1999. 10. J.G. Tsavalas, Y. Luo, and F.J. Schork, Journal of Applied Polymer Science, Vol. 87, p. 1825,2003. 11. L. Hudda, J.G. Tsavalas, and F.J. Schork, Polymer, Vol. 46, p. 993, 2005.

170

M I N I E M U L S I O N POLYMERIZATION T E C H N O L O G Y

12. J. Guo, and F.J. Schork, Macromolecular Reaction Engineering, Vol. 2, p. 265, 2008. 13. J.G. Tsavalas, F.J. Schork, and K. Landfester, JCT Research, Vol. l(l),p. 53, 2004. 14. S.P. Bunker, and R.P. Wool, Journal of Polymer Science Part A; Polymer Chemistry, Vol. 40(4), p. 451, 2002. 15. S. Bunker, C. Staller, N. Willenbacher, and R.P. Wool, International Journal of Adhesion & Adhesives, Vol. 23, p. 29, 2003. 16. L. Spagnola, E.S. Daniels, V.L. Dimonie, M.S. El-Aasser, and A. Klein, Polymeric Materials: Science and Engineering, Vol. 99, p. 753, 2008. 17. S.F. Thames, K.G. Panjnani, and O.S. Fruchey, U.S. Patent 6,001,913. 18. http://airnow.gov, accessed Sep 2009. 19. G.E. Booth, The Design, Synthesis, and Study of Novel Vegetable Oil Macromonomers and Their Structure-Property Relationships in Emulsion Polymers, Ph. D. Dissertation, 2002. 20. C.C. Blackwell, The Evaluation of Vegetable Oil Based Macromonomers in Emulsion Polymerization, Ph. D. Dissertation, 2004. 21. K.L. Diamond, Kinetic and Thermodynamic Investigation of Vegetable Oil Derived Macromonomers for Use in Low-VOC Architectural Coatings, Ph.D. Dissertation, 2004. 22. C.T. Williams, Design, Synthesis, and Characterization of Vegetable Oil Macromonomers in Auto-Oxidative Crosslinking Latexes, Ph.D. Dissertation, 2005. 23. P. Ruelle, and U.W. Kesselring, Journal of Pharmaceutical Sciences, Vol. 87, p. 1015,1998. 24. C.S. Chern, and T.J. Chen, Colloids and Surfaces, Vol. 138, p. 65,1998. 25. A. Matsumoto, M. Noriyasu, A. Hiroyuki, I. Jun-ichi, and I. Capek, Polymer, Vol. 40, p. 5687,1999. 26. J.S. Patton, B. Stone, C. Papa, R. Abramowitz, and S.H. Yalkowsky, Journal of Lipid Research, Vol. 25, p. 189,1984. 27. A.S. Kabalnov, Langmuir, Vol. 10, p. 680,1994. 28. T.K.N. Hoang, V.B. La, L. Deriemaeker, and R. Finsy, Physical Chemistry Chemical Physics, Vol. 6, p. 1413,2004. 29. W.D. Harkins, Journal of American Chemical Society, Vol. 69, p. 1428,1947. 30. W.V. Smith, and R. H. Ewart, Journal of Chemical Physics, Vol. 16, p. 592,1948 31. R.M. Fitch, and C.H. Tsai, Polymer Colloids I, New York, Plenum Press, 1971. 32. J. Uglestad, M.S. El-Aasser, and J.W. Vanderhoff, Journal of Polymer Science, Polymer Letters Edition, Vol. 11, p. 503,1973. 33. C.S. Chern, and J.C. Sheu, Polymer, Vol. 42, p. 2349, 2001. 34. E.F. Jordan, G.R. Riser, B. Artymyshyn, J.W. Pensabene, and A.N. Wrigley, Journal of Polymer Science, Part A-l, Vol. 10, p. 1657,1972. 35. R.J. Leyrer, and W Machtle, Macromolecular Chemistry and Physics,Vo\. 201, p. 1235,2000. 36. B.G. Bufkin, and J.R. Grawe, Journal of Coatings Technology, Vol. 50, p. 41, 1978. 37. F. Tronc, W. Chen, M.A. Winnik, S.T. Eckersley, and G.D. Rose, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 40, p. 4098,2002. 38. M.S. El-Aasser, S. Mohammed, E.S. Daniels, L.H. Sperling, and A. Klein, Journal of Applied Polymer Science, Vol. 66, p. 1869,1997.

POLYMERIZATION OF VEGETABLE O I L MACROMONOMERS

171

39. M.A. Winnik, Macromolecules, Vol. 32, p. 7692,1999. 40. A. Zosel, Mechanical Properties of Films from Polymer Latices, Polymers for Advanced Technologies, Vol. 263,1995. 41. M. Hidalgo, J.Y. Cavaill, J. Guillot, A. Guyot, J. Perez, and R. Vassoille, Journal of Polymer Science, Part B: Polymer Physics, Vol. 33, p. 1559,1995. 42. M.S. El-Aasser, S. Krishnan, A. Klein, and E.D. Sudol, Macromolecules, Vol. 36, p. 3511,2003. 43. H. Warson, and C. A. Finch, Applications of Synthetic Resin Latices, West Sussex, John Wiley & Sons, 2001, Vol. 1. 44. J. Guillot, and D. Charmot, Polymer, Vol. 33, p. 352,1991. 45. J. Guillot, A.M. Santos, and T.F. McKenna, Chemical Engineering Science, Vol. 53, p. 2143,1998. 46. D.R. Bassett, Journal of Coatings Technology, Vol. 73, p. 43, 2001. 47. A.P. Freidig, H.J.M. Verhaar, and J.L.M. Hermens, Vol. 33, p. 3038,1999. 48. M. Earll, Syracuse Research Association, http://www.syrres.com/esc/datalog.htm. 49. H. Tanii, and K. Hashimoto, Toxicology Letters, Vol. 11, p. 125,1982. 50. U.S. Satpathy, and A.S. Dunn, British Polymer Journal, Vol. 20, p. 521,1988. 51. J.M. Asua, Progress in Polymer Science, Vol. 27, p. 1283,2002. 52. F. Tiars, K. Landfester, and M. Antonietti, Langmuir, Vol. 17, p. 908,2001. 53. S. Lelu, C. Novat, C. Graillât, A. Guyot, and E. Bourgeat-Lami, Polymer International, Vol. 52, p. 542,2003. 54. G. Baskar, K. Landfester, and M. Antonietti, Macromolecules, Vol. 33, p. 9228, 2000. 55. S. Wang, and F.J. Schork, Journal of Applied Polymer Science, Vol. 54, p. 2157, 1994. 56. C M . Miller, E.D. Sudos, and C.A. Silebi, Macromolecules, Vol. 28, p. 2772, 1995. 57. J.A. Alducin, J. Forcada, and J.M. Asua, Macromolecules, Vol. 27, p. 2256, 1994. 58. C.S. Chern, and T.J. Chen, Colloid & Polymer Science, Vol. 275, p. 546,1997. 59. C.S. Chern, and T.J. Chen, Colloids & Surfaces, Vol. 138, p. 65,1998.

This Page Intentionally Left Blank

7

Controlled/Living Radical Polymerization in Aqueous Miniemulsion Catherine Lefay,1 and Julien Nicolas 2 laboratoire Chimie Provence, UMR CNRS 6264, Universités d'Aix-Marseille I, II et III, avenue Escadrille Normandie Niemen, 13397 Marseille cedex 20, France laboratoire de Physico-Chimie, Pharmacotechnie et Biopharmacie, UMR CNRS 8612, Université Paris-Sud XI, Faculté de Pharmacie, 5 rue Jean-Baptiste Clément, 92296 Châtenay-Malabry, France

Abstract

In the last 15 years, the field of free-radical polymerization has undergone an important renewal with the discovery of controlled/living radical polymerization (CLRP). Even though CLRP techniques are now well understood in homogeneous media such as bulk or solution, their transposition in aqueous dispersed media is far less straightforward. Because initial attempts to conduct CLRP in emulsion generally failed, research has been mainly focused on miniemulsion polymerization, which represents a simplified model of emulsion polymerization. In this chapter, recent advances in the achievements of CLRP in aqueous miniemulsion will be covered. General considerations about four main CLRP techniques, namely nitroxide-mediated radical polymerization (NMP), atom transfer radical polymerization (ATRP), reversible addition-fragmentation chain transfer (RAFT) and iodine transfer polymerization (ITP), will be given before reviewing their application in aqueous miniemulsion. Keywords: Miniemulsion, controlled /living radical polymerization, block copolymer, NMP, ATRP, RAFT

V. Mittal (ed.) Miniemulsion Polymerization Technology, (173-210) © Scrivener Publishing LLC

173

174

7.1

MINIEMULSION POLYMERIZATION TECHNOLOGY

Introduction

Controlled/Living Radical Polymerizations (CLRP) techniques have emerged in the nineties and have provided simple routes to prepare (co)polymers of predetermined molecular weight (MW) and narrow molecular weight distribution (MWD), as well as block copolymers and various complex macromolecular architectures (e.g. star, gradient or comb-shaped polymers). As a consequence, the advent of CLRP has brought a revolution in the field of macromolecular synthesis and has led to the preparation of a wide range of new polymeric materials which can be used for several applications such as surfactants, adhesives, thermoplastic elastomers, biomédical and electronic devices [1-5]. Radical polymerization in aqueous dispersed media (emulsion, dispersion, etc.) represents one of the most important industrial processes for the production of synthetic polymers. Therefore, the obvious challenge was to apply these CLRP techniques in water-borne systems. Even though CLRP techniques are well understood in bulk or solution, their transposition in those environmentally friendly media is far less straightforward [6-13]. Indeed, due to the difficulties and limitations previously encountered in ab initio emulsion polymerization under controlled/living conditions, the miniemulsion process has been then selected as it can be seen as a simplified model of ab initio emulsion polymerization regarding the complex nucleation step [13-16]. Indeed, only monomer droplet nucleation exists in miniemulsion, and those droplets behave as individual reactors in which the different polymerization steps take place.

7.2

Controlled/Living Radical Polymerization in Bulk/Solution: General Considerations

Conventional free-radical polymerization is a widely used industrial process to prepare a broad range of familiar polymers in various reaction media [17,18]. Even though this technique is easy to carry out, a limitation arises from a lack of control over the macromolecular architecture which avoids the synthesis of well-defined polymers with different levels of control such as composition, molecular shape, chain length and a, co-functionality. In order to circumvent these drawbacks, CLRP techniques have been developed.

CONTROLLED /LIVING RADICAL POLYMERIZATION

175

Prior to the development of CLRP, all these polymer characteristics were hardly achievable. Only living ionic polymerization [19] was able to furnish such (co)polymers but under very drastic polymerization conditions and only with a limited choice of monomers. Contrary to conventional free-radical polymerization, CLRP techniques are based on a reversible activation-deactivation equilibrium between dormant (or capped) chains and active chains (propagating radicals) that enables all polymer chains to grow simultaneously at the same rate. Indeed, the reduction of irreversible termination events between propagating radicals allows a good control of both molecular weights and molecular weights distribution to be achieved. Polymer chains are consequently well defined in terms of architecture and composition. A successful controlled/living radical polymerization system should exhibit the following features: (i) a linear evolution of the logarithmic conversion (ln[l /(1-conversion)]) with time, accounting for a constant concentration of propagating radicals; (ii) a linear increase in the number-average molar mass (Mn) with monomer conversion (the degree of polymerization (DPn) is predetermined by the consumed monomer to initially introduced initiator molar ratio); (iii) low polydispersity indexes (PDIs) close to a Poisson distribution (Mw/Mn = 1 + l/DP n ); (iv) a quantitative a- and co-functionalization and (v) the possibility for the chains, after the initial monomer consumption, to further grow when additional monomer is introduced which allows block copolymers to be synthesized [17]. Among the different CLRP methods, nitroxide-mediated polymerization (NMP) [20], atom transfer radical polymerization (ATRP) [21,22], reversible addition-fragmentation chain transfer (RAFT/ MADIX) [23] and iodine transfer polymerization (ITP) [24] are the most widely studied. They are commonly divided into two categories: CLRP based on reversible termination (NMP and ATRP) or based on reversible (or degenerative) transfer (RAFT and ITP). 7.2.1

CLRP B a s e d o n R e v e r s i b l e T e r m i n a t i o n

CLRP based on reversible termination are governed by the persistent radical effect (PRE) and does not require the addition of a radical initiator because the activation step generates a propagating radical [5]. The PRE can be explained in general by considering "dormant" species (noted P-Y in Figure 7.1 and P-X in Figure 7.3) that decompose into active species (or propagating radicals noted P') and persistent species (the nitroxide in the case of NMP and the Mt n+1 X/L m catalyst

176

M I N I E M U L S I O N POLYMERIZATION T E C H N O L O G Y dead polymer

P-Y

«

kd

kc

/

P*

\

+

Y*

M Figure 7.1. Activation-deactivation equilibrium in nitroxide-mediated polymerization (NMP).

in the case of ATRP, see Figures 7.1 and 7.3). Persistent species can not undergo self reaction but can only react with carbon-centered radicals. Therefore, propagating radicals are consumed by both self termination and by reaction with persistent species. After a short initiating period when both active and persistent species concentration increases similarly with time, the persistent species accumulate in the medium. The large excess of persistent species compared to active ones then enhances the probability of recombination between active and persistent species instead of irreversible termination between two active species. The key step is then the establishment of an equilibrium between active and dormant species, strongly shifted towards the latter. 7.2.1.1

Nitroxide-Mediated

Polymerization

(NMP)

NMP [20,25,26] is based on a reversible termination reaction between a growing radical P* and a free nitroxide Y" (also called persistent radical or control agent) to form a largely predominant (macro) alkoxyamine PY (Figure 7.1). The equilibrium between active and dormant species is a thermal process; the (macro)alkoxyamine regenerates the propagating radical and the nitroxide by homolytic cleavage at high temperature (generally in the 70-130°C range). A typical NMP can be set up following two different methods: (i) by using a bicomponent initiating system (i.e. a conventional radical initiator and a free nitroxide) or (ii) by using a monocomponent initiating system via a preformed alkoxyamine (see Figure 7.2 for the structures of nitroxides usually employed in NMP). Based on the work of Rizzardo and Solomon [25,27,28], Georges and co-workers reported first the controlled radical polymerization of styrene with (2,2,6,6-tetramethylpiperidinyl-l-oxy) (TEMPO) as control agent [26]. However, as TEMPO was almost exclusively limited to styrenic monomers (only a sterically hindered TEMPO

CONTROLLED /LIVING RADICAL POLYMERIZATION

Y.

y-

Y-

v n ¿< /

\

(A)

Ό-Ν

(B)

O /—

Ό-Ν

177

/

(C)

(D)

Figure 7.2. Structure of nitroxides used as mediators in NMP. (a) (2,2,6,6tetramethylpiperidinyl-1-oxy) (TEMPO); (b) N-ferf-butyl-N[l-diethylphosphono-(2,2-dimethylpropyl)] nitroxide (SGI or DEPN); (c) N-tertbutyl-N-[l-phenyl-2-(methylpropyl)]nitroxide (TIPNO) and (d) (2,2-diphenyl-3phenylimino-2,3-dihydroindol-l-yloxyl nitroxide (DPAIO).

derivative allowed the control of the n-butyl acrylate polymerization) [29], new acyclic nitroxides (see Figure 7.2b and 7.2c) have been designed to enhance the range of controllable monomers. More precisely, N-íerí-butyl-N-[l-diethylphosphono-(2,2-dimethylpropyl)] nitroxide (SGI or DEPN) [30,31] and N-ferf-butyl-IV-[l-phenyl-2(methylpropyl)]nitroxide (TIPNO) [32,33] are now able to control the polymerization of styrenics [31,33], alkyl aery lates [31,33], acrylic acid [34,35], acrylamides [36,37] and dienes [20,38]. More recently, a particular nitroxide (2,2-diphenyl-3-phenylimino-2,3-dihydroindol1-yloxyl nitroxide, DPAIO, see Figure 7.2d) has been designed to control methacrylic esters [39], which represents an alternative to the SGl-mediated copolymerization approach of methacrylates with a very small amount of a comonomer such as styrene [40-43] or acrylonitrile [44]. 7.2.1.2

Atom Transfer Radical Polymerization

(ATRP)

The mechanism of ATRP [5,21,22,45-47] is based on a rapid exchange of a halide atom (especially Cl or Br) between a growing radical and a dormant species, via a redox process involving a transition metal complex (Figure 7.3). Various transition metals can be employed in ATRP (Cu, Ru, Fe, Ni, etc.) but investigations of ATRP in aqueous dispersed media were exclusively based on copper complexes. This equilibrium is in fact largely shifted towards the dormant species that ensures the success of the control of the polymerization. In the first ATRP process, called direct ATRP, the transition metal complex in a lower oxidation state (Mt n /L m with Mtn the metal and Lm a ligand) is directly added to the reaction as an activator and reacts reversibly

178

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY dead polymer kt

k *act

P-X + Mtn/Lm « - — ' Kdeact

X: CI, Br, I Ml·. Cu, Ru, Fe, Ni, Pd, Rb

''

P* /

+

Mt n+1 X/L m

>

M

Figure 7.3. Activation-deactivation equilibrium in atom transfer radical polymerization (ATRP).

with the dormant species (PX, with X a halogen atom) to form a deactivator (Mtn+1X/Lm) and the active specie P \ In contrast, when the polymerization is initiated by a conventional initiator and a metal complex at the higher oxidation state, the process is called reverse ATRP. Simultaneous Reverse and Normal Initiation (SR&NI) process takes advantage of both normal and reverse ARTP as Cu(II), an alkyl halide and a radical initiator are initially present in the reaction medium [48]. It provides a way to reduce the amount of copper complex and to prepare more complex macromolecular architectures. Recently, new ATRP processes have been developed namely activators generated by electron transfer (AGET) [49] and activators regenerated by electron transfer (ARGET) [50]. The AGET process employs a reducing agent (e.g. ascorbic acid or tin(II) 2-ethylhexanoate) which reacts with Mt n+1 X/L m to generate the active catalyst (Mt n /L m ). The process then follows a direct ATRP process. AGET ATRP allows the preparation of pure block copolymers with no homopolymer of the second monomer. The ARGET process uses an excess of reducing agent that allows a significant reduction of the amount of metal in the media. However, if ATRP can be used with a large range of monomers, functional monomers bearing acid or amine function remain today a barrier to overcome.

7.2.2 CLRP Based on Degenerative Transfer CLRP systems based on degenerative transfer operate by the exchange of activity between active and dormant species via a reversible chain transfer mechanism. Contrary to ATRP and NMP (PRE-based systems), such CLRP processes require the addition of a radical initiator because the activation-deactivation equilibrium is not associated with a change in the concentration of (macro)radicals.

CONTROLLED/LIVING RADICAL POLYMERIZATION

7.2.2.1

Reversible Addition-Fragmentation Transfer (RAFT)

179

Chain

The RAFT process using thiocarbonylthio compounds, including dithioesters and trithiocarbonates, was reported by the CSIRO group in early 1998 [51]. French researchers reported a similar process but with xanthate RAFT agents (MADIX) in late 1998 (see Table 7.1) [52]. The historical development of this process has been recently reviewed [53]. RAFT agents noted Z-C(=S)SR act as transfer agents by a twostep addition-fragmentation mechanism (see Figure 7.4) [23,53-58]. Such a transfer agent bears a C=S double bond reactive toward radical addition and a R and Z groups that are generally chosen according to the monomer to be polymerized. Indeed, the Z substituent should convey the transfer agent an appropriate reactivity toward propagating radicals and give the intermediate radicals 2 and 5 an appropriate stability (Figure 7.4). In addition, the R* radical resulting from fragmentation (or homolytic cleavage) of the S-R bond of 2 or 5 has to efficiently reinitiate the polymerization. In conclusion, reversible chain transfer requires that: (i) 1 and 3 are active transfer agents under polymerization conditions; (ii) R is a good living group and (iii) propagating radicals and R" efficiently reinitiate the polymerization (that means that the reinitiation rate constant, k, is higher than the propagation rate constant, k ). RAFT agents are then characterized by their transfer constant C. with C = k /k where k and k are J

tr

tr

tr

p

tr

the rate constants of transfer and propagation, respectively. Table 7.1. Various RAFT agents noted Z-C(=S) SR depending on the nature of the Z group. RAFT agent

Z

Dithioester

R, (alkyl, aryl)

Dithiocarbonate (xanthate)

OR,

Dithiocarbamate

NRjR.,

Trithiocarbonate

SR,

Phosphoryl dithioester

P ( 0 or S)(ORj)2

Dithiocarbazate

NCR^-NR^

p

180

M I N I E M U L S I O N POLYMERIZATION T E C H N O L O G Y Pre-equilibrium: reversible chain tranfer dead polymer kt / Pn\ + V S _ k ) 1 k py z M 1

R

kadd " - * Pn-S^S-R k add " I z 2

^

kß p

dead polymer v kt ' P n - S ^ S + . R* I V kn z V 3 M 3

Main equilibrium: chain equilibration dead

dead polymer

Polymer \k<

kt / P

*

+

kP ; M

V

S _ P m

I

z

^="

P

n-VS~Pm ^

T

^

P

z

n-

S

^

S

r

z

+

/ P™

v{ k, ^ M

M

4 5 6 Figure 7.4. Mechanism of reversible addition-fragmentation chain transfer (RAFT).

Until recently the major drawback of RAFT was the lack of universal control agent. In particular, dithioesters or trithiocarbonates were suitable for controlling the polymerization of more activated monomers such as styrene (S), methyl methacrylate (MMA), methacrylic acid (MAA), methacrylamide (MAM), acrylic acid (AA), acrylamide (AM) or acrylonitrile (AN). However, they inhibit or delay the polymerization of less activated monomers such as vinyl acetate (VAc), N-vinylpyrrolidone (NVP) or N-vinylcarbazole (NVC) for which xanthates or dithiocarbamates (RAFT agents of low transfer constant Ctr) are required. The choice of R and Z groups is thus crucial to achieve a good control of the polymerization [57,58]. A solution to this problem was recently proposed by Moad, Rizzardo and co-workers who succeeded to elaborate a universal switchable RAFT agent (Figure 7.5) [59]. Unfortunately, RAFT polymerization in bulk/solution is often accompanied by both inhibition and retardation [23,53-56]. If inhibition is generally attributed to the pre-equilibrium and a slow re-initiation by R* (Figure 7.4), the retardation seems to be related to the intermediate radical of the main equilibrium. 7.2.2.2

Iodine Transfer Polymerization (ITP)

Iodine transfer polymerization (ITP) is a degenerative polymerization which requires an alkyl iodide (RI) as a transfer agent. ITP was first

CONTROLLED /LIVING RADICAL POLYMERIZATION \ +

v//

S-R

\ ^

cj

H+

SR base

RAFT controls VAc, NVP, NVC

*-<

SR

ö"

'

181

S-R

RAFT controls MMA, MA, Sty

Figure 7.5. Switchable RAFT agent.

dead polymer p·

dead polymer

Kpx

+ p -| —

M M Figure 7.6. Mechanism of Iodine Transfer Polymerization (ITP).

developed in the late 1970s by Tatemoto and co-workers [24,60]. In this CLRP process, the initiating radical A* generated by thermal decomposition of a conventional initiator (e.g. AIBN) adds onto monomer. The resulting radical propagates before exchanging iodine with the transfer agent RI (e.g. C6F13I). The main equilibrium between dormant and propagating species is described on Figure 7.6 [61]. The main drawback of ITP is the instability of the iodoalkyl compound. To overcome this drawback, Lacroix-Desmazes and co-workers have proposed a new process based on the direct reaction between radicals and molecular diiodine (I2) which plays the role of a radical trap [62,63]. The chain transfer agents (A-I and Pn-I) are then generated in situ. This process is called reverse iodine transfer polymerization (RITP) by analogy with reverse ATRP. Several patents based on the RITP process have been recently filled by Solvay [64,65] and Akzo [66]. Nevertheless, RITP still have some drawbacks that need to be overcome. In particular, an inhibition period is usually observed at the beginning of the polymerization and the evolution of M n and PDI versus conversion depends on the exchange constant C = k Ik Γ

σ

ex

ex

p

with kex and k the rate constant respectively of exchange and propagation [24]. In addition, due to low Cex values, the control over the polydispersity is usually poor.

182

MINIEMULSION POLYMERIZATION TECHNOLOGY

7.3 Nitroxide-Mediated Miniemulsion Polymerization Only TEMPO, TEMPO derivatives, SGI nitroxide as well as their corresponding alkoxyamines have been used in miniemulsion polymerization so far. The different initiating systems are presented in Table 7.2. Basically, nitroxide-mediated miniemulsion polymerization can be divided into two main categories referring to the nature of the initiating system; either bicomponent or monocomponent, each of them could be either oil- or water-soluble (i.e. oil or aqueous phase initiation). Table 7.2. Various initiating systems in nitroxide-mediated miniemulsion polymerization. Type of the initiating system

Solubility of the radical initiator

Initiating system

Monomers

Ref.

Radical Nitroxide initiator Bicomponent Oil-soluble

BPO

TEMPO

Sty

[67-70]

AIBN

SGI

Sty

[71]

Bicomponent Water-soluble K 2 S 2 O 8

Sty TEMPO and derivatives

[69,70, 72-74]

SGI

Sty

[71,75]

PS-TEMPO, PS-TEMPO-OH

Sty, nBA

[70,74, 76-86]

PS-TIPNO-OH

MBA

[87]

BST-TEMPO

Sty, nBA

[83,87, 88]

MONAMS

Sty, nBA

[89-93]

Sty, nBA

[94]

K2S2O8/

Na 2 S 2 0 5 Alkoxyamine Oil-soluble

Alkoxyamine Water-soluble BlocBuilder™

CONTROLLED/LIVING RADICAL POLYMERIZATION

183

7.3.1 Oil-Soluble Bicomponent Initiating System In a miniemulsion process, the monomer droplets behave as independent microreactors, which allow the use of the same reagents as in bulk. In this view, a bicomponent initiating system is very simple to achieve in miniemulsion as very few parameters have to be changed with respect to a classical polymerization system. Prodpran et al. were the first to report a successful nitroxidemediated miniemulsion polymerization of styrene at 125°C [67]. They used benzoyl peroxide (BPO, see Figure 7.7), TEMPO and Dowfax 8390 (see Figure 7.8) as the surfactant together with hexadecane in order to prevent Ostwald ripening. They also studied the miniemulsion polymerization owing to the thermal autoinitiation of

o

9 o _ κ+ o-s-o-o-s-o κ'

<ïs.

" * * ■

ö

o (A)

0

°

NC- -N=N-

0 - S - S - 0 Na+

Na

Ö 0 (C)

2HCI

HN

y.

NH NH,

"ΌΟΟΗ

HOOC

-N=NCN CN

(F)

(E)

O

-N=N-N=N-

HO-

-OH (G)

-CN

(D)

-N=N-

H,N

ö

(B)

O

(H)

Figure 7.7. Radical initiators used in CLRP in miniemulsion: (a) benzoyl peroxide (BPO); (b) potassium persulfate (KPS); (c) sodium metabisulfite (SMBS); (d) 2,2'-azobis-isobutyronitrile (AIBN); (e) 2,2'-azobis(2-methylpropionamidine) dihydrochloride (V-50); (f) 4,4'-azobis(4-cyanopentanoic acid) (ACPA); (g) 2,2'-azobis[2-[1 -(2-hydroxyethyl)-2-imidazolin-2-yl]propane]dihydrochloride (VA-060) and (h) 2,2'-azobis[2-(2-imidazolin-2-yl)propane]dihydrochloride (VA-044).

184

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

(E) (F) Figure 7.8. Surfactants used in CLRP in miniemulsion: (a) sodium dodecylbenzylsulfate (SDBS); (b) cetyltrimethylammonium bromide (CTAB); (c) polysorbate 80 (Tween 80); (d) Dowfax 8390; (e) sodium dodecylsulfate (SDS) and (f) Brij 98.

styrene. Latexes at 20 wt.% solids were targeted and a good control was achieved with polydispersity indexes ranging from 1.14 to 1.6. However, large particles were obtained with a broad particle size distribution. Cunningham et al. investigated the rate-accelerating effects of camphorsulfonic acid (CSA) on nitroxide-mediated styrene miniemulsion polymerization at 135°C [68]. In this case, TEMPO and TEMPO-OH were used in conjunction with sodium dodecylbenzylsulfate (SDBS) as a surfactant. For all conditions, molecular weights were higher in the presence of CSA, although the degree of rate enhancement due to its addition varied considerably. A careful attention was also paid to the effect of the water solubility of the nitroxide over the quality of the control [69]. TEMPO offered a better control than TEMPO-OH whereas the kinetics was not significantly affected whatever the nitroxide.

CONTROLLED/LIVING RADICAL POLYMERIZATION

185

When the acyclic SGI nitroxide was selected instead of TEMPO, the polymerization of styrene could be performed at 90°C using AIBN as a monomer-soluble radical initiator [71]. However, even though fair PDIs were obtained, a limited monomer conversion (-60%) after 24 h was obtained probably due to the persistent radical effect together with the almost negligible rate of thermal autoinitiation of styrene at such temperature, resulting in an accumulation of free SGI in the polymerization medium. 7.3.2

Water-Soluble Bicomponent Initiating System

In this case, the polymerization starts in the aqueous phase and leads to the formation of oligoradicals or oligomeric alkoxyamines which then enter the monomer droplets, thus becoming the primary locus of polymerization. This change with respect to an oil-soluble initiating system might affect the polymerization kinetics, the initiator efficiency and hence the control over the molecular weights and the molecular weight distribution. The TEMPO/potassium persulfate (KPS) system was employed by MacLeod et al. for the miniemulsion polymerization of styrene at 135°C [72]. With an optimized [TEMPO]0/[KPS]0 initial ratio of 2.9, a very fast polymerization was achieved (87% monomer conversion in 6 h) together with narrow MWD. This result was explained by partition of the TEMPO between the organic and the aqueous phase, which would lead to a significant decrease of the nitroxide concentration in the polymerization locus. While TEMPO- or TEMPO-OH-terminated polystyrene were used as macroinitiators for the miniemulsion polymerization of n-butyl acrylate (nBA) to prepare the corresponding block copolymers by simple chain extension [70,74], when TEMPO was replaced by the more hydrophilic TEMPO-OH during the synthesis of PS homopolymers, significant differences appeared [69,73]. With TEMPO, the initiator efficiency was higher with KPS than with BPO whereas with TEMPO-OH, no such difference was observed. Aqueous phase kinetics together with partition coefficient of the nitroxides and of oligomeric alkoxyamines were shown to be of great importance regarding the outcome of the polymerization. With SGI, the potassium persulfate/sodium metabisulfite redox initiating system allowed the styrene polymerization rate to be strongly enhanced compared to the similar polymerization initiated by AIBN at 90°C [71,75]. However, pH has to be adjusted in

186

MINIEMULSION POLYMERIZATION TECHNOLOGY

order to avoid side reactions with the nitroxide. An optimal [SG1] 0 / [KPS]0 ratio of 1.2 was found to be the best compromise regarding induction period necessary to the in situ formation of the SG1based alkoxyamine. Styrene conversion reached 90% in 8 h with molecular weights in good agreement with the theoretical values and PDIs in the 1.5-2.0 range.

7.3.3

Oil-Soluble Monocomponent Initiating System

The use of a monocomponent initiating system instead of a bicomponent one allows a high initiating efficiency to be obtained (usually close to 100%) which is of crucial importance in CLRP as the number of chains is in that case well-defined. In miniemulsion, one can distinguish between use of molecular alkoxyamine and macromolecular alkoxyamine (i.e. nitroxide-terminated oligomers). In the latter case, the macroalkoxyamine is generally a short PS-TEMPO, either synthesized and isolated prior to its use in miniemulsion [70,76,77,79,81,82,95], or synthesized after a first step in bulk stopped at low monomer conversion followed by emulsification of the medium without further purification [74,80]. Pan et al. used a PS-TEMPO macroinitiator (Mn = 7050 g.mol·1) at 125°C with hexadecane as a costabilizer and Dowfax 8390 as a surfactant [76,77,95]. However, thermal autoinitiation of styrene was noticed leading to molecular weights lower than expected ones together with PDI increasing up to -1.8 with monomer conversion (~70% in 12 h). They observed that the polymerization rate was independent of both the macroinitiator and the surfactant concentration (and thus independent of the number of nanoparticles). Surprisingly, Keoshkerian et al. who prepared a similar macroinitiator in situ where able to obtain very high styrene conversion (>99%) within 6 h together with PDI as low as 1.15 [74]. Chain extensions with ft-butyl acrylate from PS-TEMPO macroinitiators in miniemulsion turned to be an efficient route to poly(n-butyl acrylate)-frpolystyrene (PnBA-b-FS) block copolymer synthesis [70,74]. When a polystyrene oligomer terminated by the TIPNO-OH nitroxide (well-suited for acrylates) was chosen, 86% nBA conversion was reached within 3 h, still yielding satisfying PDI. A modified TEMPO-mediated miniemulsion process which did not require a volatile costabilizer due to the use of a PS-TEMPO macroinitiator, and based on the continuous addition of ascorbic acid was reported [78]. It yielded nearly complete styrene conversions

CONTROLLED /LIVING RADICAL POLYMERIZATION

187

within 2-3 h while preserving narrow molecular weight distribution and a high degree of livingness. Besides, addition of ascorbic acid or a free-radical initiator allowed the TEMPO-mediated miniemulsion styrene polymerization to be conducted at low temperature such as 100°C while exhibiting reasonable polymerization rates and preserving the living character of the polymers [82]. However, PDI were slightly higher (~1.4-1.6) than those obtained under classical experimental conditions. Preformed molecular oil-soluble alkoxyamine, either based on TEMPO and derivatives [83,87,88] or SGI (MONAMS) [89-92], have been used for the polymerization of styrene and n-butyl acrylate in aqueous miniemulsion. A mathematical model developed by Ma et al. predicted that under TEMPO-based molecular alkoxyamine initiation, the living fraction of the polymer should decrease with monomer conversion and the PDI should increase when reaching high monomer conversion [88]. However, for the polymerization of styrene initiated by the BST-TEMPO alkoxyamine (Figure 7.9a), adequate experimental conditions led to narrow molar mass distributions (PDI ~1.3) and high monomer conversions [83]. In contrast, with nBA in conjunction with ascorbic acid at 135°C [87], a higher PDI was obtained (-1.6 at 60% monomer conversion) compared to previous results involving the TIPNO-OH nitroxide. The more versatile SGl-based oil-soluble alkoxyamine (the socalled MONAMS, Figure 7.9b) allowed better results to be obtained [92]. Different experimental conditions were investigated for the homopolymerization of nBA [91 ] and its copolymerization with styrene to yield gradient copolymers at 112°C [89]. For instance, molecular weights masses up to 50,000 g.mol·1 together with PDI as low as 1.2-1.4 could be obtained. Besides, a narrow distribution composition was noticed and copolymers chains exhibited a gradient composition. The 20—45%

V-O-N O—', o~L y-p-o—/

(Β)

Y-O-N OHO^ V- p-o-

(C)

Figure 7.9. Molecular alkoxyamines used in miniemulsion NMP: (a) BST-TEMP; (b) MONAMS and (c) BlocBuilder™.

188

MINIEMULSION POLYMERIZATION TECHNOLOGY

solids latexes were very stable, even though average diameters were rather broad. Besides this, if an appropriate sequential monomer addition is set up, poly[(n-butyl acrylate)-b-poly(n-butyl acrylateco-styrene)] block copolymers within stable 20% solids latexes can be synthesized [90]. The lack of remaining PnBA homopolymer after the second step also demonstrated the high living fraction of the PrcBA-SGl first block. 7.3.4

Water-Soluble M o n o c o m p o n e n t Initiating System

The use of the BlocBuilder™ SGl-based water-soluble alkoxyamine (Figure 7.9c) represents the only example to date of a water-soluble monocomponent initiating system in NMP in miniemulsion [94]. For nBA at 112°C at 20% solids, all the features of a controlled system were obtained. However, a poor initiating efficiency have been noticed in the case of styrene at 120°C due to highly pronounced persistent radical effect leading to a very slow chain growth in the aqueous phase. This was circumvented by the addition of a very small amount of a well-chosen comonomer (methyl acrylate) acting here as an entry rate enhancer due to appropriate copolymerization conditions. Concerning the colloidal characteristics, average diameters were significantly lower (260-310 nm) with a narrower particle size distribution than those obtained with the MONAMS under identical conditions, probably due to the negative charge of the BlocBuilder™ alkoxyamine. It's worth mentioning that this study further led to the first successful system of nitroxide-mediated emulsion polymerization under controlled /living conditions [13,96-99].

7.4

Atom Transfer Radical Miniemulsion Polymerization

In contrast to other CLRP methods, ATRP may not be the most straightforward technique to apply in aqueous dispersed media due to the presence of both an activator and a deactivator to control the polymerization kinetics and the chain growth. However, numerous ATRP studies carried out in miniemulsion, mainly using catalysts based on copper complexes, have been reported so far. Due to the unique nature of the ATRP process, a successful ATRP

CONTROLLED /LIVING RADICAL POLYMERIZATION

(E)

189

(F)

Figure 7.10. ATRP ligands used in miniemulsion polymerization: (a) bipyridine (bpy); (b) dinonylbipyridine (dNbpy); (c) pentamethyldiethylenetriamine (PMDETA); (d) nitrogen-based tetradentate ligands [R = lauryl (LAJREN), 2-ethylhexyl (EHAJREN) or butyl (BAJREN)]; (e) 4/4'/4"-Tris(5-nonyl)-2,2':6'/ 2"-terpyridine (tNtpy) and (f) N/N-bis(2-pyridylmethyl)octadecylamine (BPMODA).

in miniemulsion relies on the high solubility of both the Cu(I) and the Cu(II) complexes in the monomer phase, which is directly connected to the nature of the ligand (see Figure 7.10). Indeed, only hydrophobic enough ligands, such as substituted bipyridines, would be efficient. In addition, the surfactant should be carefully selected in order to avoid interaction with the catalyst. In this view, only cationic and non-ionic surfactants are suitable for achieving living miniemulsion polymerizations, even though a good stability

190

MINIEMULSION POLYMERIZATION TECHNOLOGY

is only obtained with non-ionic ones [100]. Various side reactions may also occur, mainly due to the nature of the solvent. Direct ATRP was the first process which has been investigated. However, due to the technical difficulties encountered during the emulsification step due to the sensitivity of Cu(I) complexes to oxidation when exposed to air, less oxygen sensitive approaches have been successfully employed such as reverse ATRP, SN&NI ATRP and AGET ATRP. 7.4.1

D i r e c t ATRP

Homopolymerization of n-butylmethacrylate (nBMA) was successfully reported in miniemulsion at 70°C with ethyl 2-bromoisobutyrate as the initiator, CuBr/dNbpy as the catalyst and 13.5 wt.% (with respect to the monomer) of a non-ionic surfactant (Brij 98) [101]. In 2 h, monomer conversion reached more than 70% with a good control and livingness (PDI < 1.5), together with an average diameter around 300 nm. The same catalytic system was used by Kagawa et al. who reported a two-step direct ATRP miniemulsion process at 70°C for the preparation of P(z'BMA)-b-PS diblock copolymer latex stabilized by the Tween 80 surfactant (6-10 wt.% with respect to ¿BMA) [102]. Due to good control and high degree of livingness, well-defined block copolymers were synthesized and nanostructured nanoparticles adopting an "onion-like" morphology were observed. 7.4.2

R e v e r s e ATRP

Direct ATRP can be advantageously replaced by reverse ATRP, which starts with a conventional radical initiator and a Cu(II) complex (tolerant to oxygen), thus allowing Cu(I) oxidation problem to be efficiently overcome. Besides, a broad variety of water-soluble initiators are commercially available which confer a great flexibility to the process. However, reverse ATRP exhibits an induction period (which depends on the polymerization temperature and on the amount of deactivator with respect to the initiator), until Cu(II) concentration becomes low enough (due to the deactivation of radicals) for the ATRP equilibrium to take place. Reverse ATRP in miniemulsion has been successfully reported for nBMA at 90°C with Cu(II)/dNbpy complex in the presence of

CONTROLLED/LIVING RADICAL POLYMERIZATION

191

hexadecane and 13.5 wt.% (with respect to the monomer) of Brij 98 as the surfactant [101]. The nature of the radical initiator was varied and it was shown that the water-soluble V-50 initiator led to a better control over molecular weights with a higher initiating efficiency than with AIBN. No significant change concerning the colloidal characteristics was however noticed when compared to the direct ATRP counterpart [101]. More recently, Li and Matyjaszewski used a highly active ligand (Figure 7.10d) together with the VA-060 radical initiator and Brij 98 for the miniemulsion polymerization of nBMA [103]. These optimized conditions allowed the reaction temperature to be decreased down to 70°C and the amount of surfactant to be strongly reduced (2.3 wt.% with respect to the monomer) while maintaining good control and livingness together with satisfying colloidal stability at 20% solids. The hydrophobicity of the catalyst can be easily changed by varying the nature of the ester moiety of the ligand. When the surfactant concentration was too high, a too hydrophobic ligand resulted in a loss of control/livingness due to both micellar and droplets nucleation with an insufficient concentration of deactivator in the polymerization loci (due to its inability to transport via the aqueous phase). If the hydrophobicity of the ligand was tuned appropriately, surfactant concentration had no effect on the system (only the particle size distribution was bimodal) and molecular weights ranged from 30,000 to 100,000 g.mol·1 together with PDIs between 1.5 and 1.7. The high degree of livingness of such latexes was demonstrated by performing chain extension after 98% monomer conversion with a second load of «BMA and surfactant, thus opening the door to block copolymer synthesis in miniemulsion ATRP [103]. The cationic CTAB surfactant was chosen by Simms and Cunningham in conjunction with EH6TREN as the ligand and the VA-044 initiator for reverse ATRP in miniemulsion of nBMA at 90°C [104]. Both a linear increase of molecular weights and a decrease of PDIs values with monomer conversion were shown, together with a good colloidal stability provided by only 1 wt.% of CTAB with respect to the monomer. By using ascorbic acid/hydrogen peroxide as the redox initiating system, the same authors reported the successful synthesis of high molecular weight (~106 g.mol·1) P(nBMA) or PMMA with low PDI values in reverse miniemulsion ATRP at 60°C [105,106]. The nature of the initiating system and compartmentalization effects have been put forward to explain this achievement [107].

192

MINIEMULSION POLYMERIZATION TECHNOLOGY

7.4.3 Simultaneous Reverse and Normal Initiation (SR&NI) ATRP The SR&NI ATRP represents a significant improvement in polymerization conducted in aqueous dispersed media as it allows the problems of both direct and reverse ATRP to be overcome [108]. This process has been applied to the miniemulsion polymerization of nBMA at 80°C under AIBN initiation, using BPMODA or tNtpy as the ligand and Brij 89 to ensure the colloidal stabilization [109]. The SR&NI ATRP approach led to good control/livingness and colloidal properties with 5-8 times less copper complex in the medium compared to reverse ATRP. Homopolymers of nBA and styrene were also successfully synthesized by miniemulsion SR&NI ATRP [110,111]. The advantage of this technique compared to reverse ATRP is the possibility to prepare complex macromolecular architectures via the use of multifunctional alkyl halide initiators. For instance at 60°C with tNtpy or BPMODA, the synthesis of diblock, triblock and 3-arm star polystyrene and poly(«-butyl acrylate) has been reported in miniemulsion SR&NI ATRP using a trifunctional ATRP initiator [109]. Besides, poly(methyl acrylate)-b-polystyrene star-block copolymers were obtained by sequential monomer polymerization from a multifunctional alkyl halide initiator [110]. A good control of the polymerization was shown with low PDIs in the 1.18-1.37 range. However, a small amount of homopolymer deriving from the second monomer was detected and highly supposed to be generated by the radical initiator, which is a drawback inherent to the SR&NI ATRP process.

7.4.4 Activators Generated by Electron Transfer (AGET) ATRP AGET ATRP, which represents one of the most advanced ATRP method to date, was carried out in miniemulsion. nBA, MA and styrene were polymerized at 80°C in the presence of ascorbic acid as the reducing agent together with CuBr 2 /BPMODA as the catalyst [112]. PMA-fr-PS diblock and 3-arm star block as well as various gradient copolymers of high purity were obtained [113,114]. Interestingly, if the reducing agent is added in excess, it consumes oxygen as well as it reduces the Cu(II) complex and thus allows AGET ATRP to be performed in the presence of air (i.e. the deoxygenation step could be skipped) [115]. Stoffelbach et al. used an amphiphilic poly(ethylene glycol)-k-PS block copolymer as both the stabilizer and the initiator

CONTROLLED/LIVING RADICAL POLYMERIZATION

193

for the miniemulsion AGET ATRP of rcBMA, styrene or nBA at 80°C without any further surfactant addition [116]. A good control of the polymerization was obtained with rather low PDIs and relatively high initiation efficiency. Stable latexes were recovered with average diameters in the 130-250 nm range. The great flexibility of AGET ATRP also allowed well-defined hybrid materials to be prepared in miniemulsion via the use of functionalized silica acting as a macroinitiator, resulting in coreshell silica/PnBA hybrid particles [117]. In comparison to bulk system, the miniemulsion approach permitted to prepare hybrid materials with a higher yield and without macroscopic gelation. Besides, a similar miniemulsion AGET ATRP was recently employed for the high yield synthesis of well-defined molecular brushes by polymerization of nBA from poly(2-(2-bromopropionyloxy) ethyl methacrylate) (PBPEM) macroinitiator, where macroscopic gelation was not observed [118]. Water-in-oil (W/O) miniemulsion polymerization, also called inverse miniemulsion polymerization, is a method particularly welladapted for the synthesis of water-soluble polymers. This process has been carried out under AGET ATRP conditions by Matyjaszewski and co-workers for the polymerization of poly(ethylene glycol) methyl ether methacrylate (MePEGMA) [119,120]. Well-controlled polymers (PDI < 1.3) and stable nanoparticles of -200 nm in diameter were prepared. More complex architectures were then proposed such as PMePEGMA-fr-PHEMA diblock and PHEMA-b-PMePEGMA-bPHEMA triblock copolymers [121] as well as hydrophilic polymer gels [120] and cross-linked hydrophilic MePEGMA-based polymer particles for drug delivery purposes [122,123].

7.5

Reversible Addition-Fragmentation Chain Transfer Miniemulsion Polymerization

RAFT is one of the most versatile CLRP techniques regarding the choice of monomers and the polymer architecture. As the control agent is attached to the polymer chain end (except in the early polymerization stages with a low MW RAFT agent), no partition of the RAFT agent is expected between the aqueous and organic phases. To ensure an efficient control of the polymerization, the RAFT agent has to be placed in the locus of polymerization, (i.e. in the monomer droplets).

194

MINIEMULSION POLYMERIZATION TECHNOLOGY

Depending on the pH and the temperature, the rate of hydrolysis of the dithiocarbonyl moiety of the RAFT agent can be significant [124]. This hydrolysis phenomenon is in particular more crucial for reverse RAFT miniemulsion (of acrylamide monomer for instance) [125] than for direct RAFT miniemulsion polymerization as in this latter case, the control agent is mainly compartmentalized in the particles (i.e. in the organic phase). RAFT miniemulsion polymerization has been recently reviewed by Zetterlund and co-workers [10] and by Cunningham [7]. In addition, numerous papers have appeared reporting on the RAFT miniemulsion of a large range of monomers such as: styrene[126-134] [135-144],4-acetoxystyrene[145],MMA[135,138,142,146-148],EHMA [129,136], nBMA [129,136,147,149-151], nBA [127,128,141,152,153], MAA [153], fluorinated alkyl methacrylates [149,151], acrylamide (inverse RAFT miniemulsion) [125], VAc [154,155] and vinyl saccharide monomers based on glucose and fructose [147]. Various diblock or triblock copolymers based on these monomers have been synthesized as well.

7.5.1

Key-Steps for the Success of RAFT Miniemulsion Polymerization

RAFT is certainly the CRP technique that encountered the most difficulties to be applied to aqueous dispersed media. The major drawbacks and challenges that had to be overcome are described in the following parts. 7.5.1.1

Inhibition and Retardation

Inhibition in RAFT polymerization in bulk/solution is usually related to the pre-equilibrium [129,156]. Inhibition in RAFT miniemulsion polymerization using low MW RAFT agents is often more important than in bulk due to exit of the RAFT agent leaving group to the aqueous phase. The extent of inhibition will therefore be linked to the hydrophobicity of this leaving group [152,157,158]. Similarly, retardation in RAFT miniemulsion polymerization is generally more severe than in bulk/solution counterparts. A possible explanation could be the exit to the aqueous phase of the radical released from fragmentation of a low MW RAFT agent intermediate radical followed by termination in the aqueous phase or termination after re-entry in a particle, in the case of a "zero-one" system [158]. In

CONTROLLED /LIVING RADICAL POLYMERIZATION

195

addition, the degree of retardation seems to decrease with increasing the hydrophobicity of the released radical [132,135]. This was supported by Lansalot and co-workers who showed that the polymerization rates in miniemulsion polymerizations mediated by 1-phenylethyl phenyldithioacetate (PEPDTA, see Figure 7.11) were higher than those of miniemulsion polymerizations mediated by cumyl dithiobenzoate (CDB) and 1-phenylethyl dithiobenzoate (PEDB). It suggested that the main parameter for controlling the polymerization rate in RAFT miniemulsion polymerization is the possible escape of R* to the aqueous phase. Moreover, retardation in miniemulsion RAFT polymerization compared to the conventional miniemulsion polymerization can be nearly suppressed by using a PS-PEPDTA macroRAFT agent instead of the corresponding low MW RAFT agent. These results have been confirmed by Butté and co-workers who used a PS oligomer obtained by bulk polymerization in the presence of ferf-butyl dithiobenzoate [135]. Since RAFT is a reversible chain transfer process, it was believed that the (mini)emulsion polymerization kinetics will not be affected by the presence of a RAFT agent. Nevertheless, Monte Carlo simulations performed by Prescott and co-workers [158,159] have proven that the presence of a RAFT agent may modify the "zero-one" kinetics model of emulsion polymerization because of the chain length dependence of kt. Luo and co-workers [133] have then theoretically investigated the intrinsic effect of RAFT reactions on the (mini)

K

O r0 (A)

αχ)

χ)

aV (B)

(C)

o

s s

(D)

CN (E)

Figure 7.11. Structure of several RAFT agents used in miniemulsion polymerization, (a) cumyl dithiobenzoate (CDB); (b) 2-cyanoprop-2-yl dithiobenzoate (CPDB); (c) 1-phenylethyl phenyldithioacetate (PEPDTA); (d) methyl (ethoxycarbonothioyl)sulfanyl acetate (MESA) and (e) 4-cyano-4[(thiobenzoyl) pentanoic acid.

196

MINIEMULSION POLYMERIZATION TECHNOLOGY

emulsion polymerization system and confirmed their conclusions by the miniemulsion polymerization of styrene with PS-PEPDTA and PS-CPDB (with CPDB for 2-cyanoprop-2-yl dithiobenzoate) as macroRAFT agents. By modifying the Smith-Ewart equation for RAFT (mini)emulsion and by taking into account the different kinetic properties of propagating and intermediate radicals and considering a zero-one system, they concluded that retardation effect was an inherent feature of RAFT (mini)emulsion [156]. Consequently, in the presence of a RAFT agent, retardation will always occur contrary to a system with no control agent. 7.5.2.2

Colloidal Instability

If no appropriate polymerization conditions are employed [136,160] miniemulsion RAFT polymerization can be accompanied by a characteristically colored organic layer, the presence of a large coagulum fraction a n d / o r a slow polymerization rate [129]. The origins of such colloidal instability are still not well understood. Huang and coworkers have studied the instability of styrene miniemulsion using CDB as a RAFT agent and stabilized by SDS and hexadecane [130]. The observed instability was assigned to the initial Ostwald ripening process leading to large micron-size droplets with a relatively low concentration of RAFT agent and hexadecane, which phase separated as an oil layer after centrifugation. The Ostwald ripening also created small droplets containing high concentrations of RAFT agent, that coagulate to form droplets of large diameters, which further separate as a colored layer at the top of the dispersed media. As the polymerization takes place in different loci, bimodal MMD and particles with broad or bimodal particle size distribution were obtained. On the other hand, explanations of Luo and co-workers were in contradiction with Huang and co-workers conclusions as in their study, the colloidal instability was believed to originate from the "superswelling" effect that slows down droplet nucleation [131,134,160]. According to this theory, the high concentration of low MW oligomers in CLRP and thus in RAFT polymerization at low conversion gives rise to superswelling where large amounts of monomer diffuse from non-nucleated monomer droplets to particles that contain oligomers. In addition, Ugelstadt and co-workers have proven, theoretically and experimentally, that oligomers are very effective swelling agents and can significantly increase the

CONTROLLED /LIVING RADICAL POLYMERIZATION

197

monomer swelling properties of polymer particle [161,162]. Luo and co-workers thoroughly investigated the deleterious effect of superswelling on RAFT miniemulsion polymerization and their theoretical modeling [160] predicted that the interfacial tension of the miniemulsion could play a key role in the superswelling. A decrease of the interfacial tension would reduce the chemical potential of monomer within minidroplets and then decrease the superswelling degree. Consequently, it would enhance the colloidal stability together with the PSD and the MWD. These theoretical works were experimentally confirmed in the case of a styrene batch miniemulsion polymerization with PEPDTA RAFT agent under KPS initiation, in conjunction with hexadecane as a costabilizer and SDS as a surfactant [131]. The post addition of SDS increased the surfactant coverage of the minidroplets and suppressed the superswelling of the first particles. These results were confirmed by another study [134] where the authors showed that, in the case of styrene RAFT miniemulsion polymerization, a higher level of SDS and hexadecane than those employed in conventional mimiemulsion polymerization could narrow the MWD to 1.3 and yield narrow PSD similar to conventional miniemulsion polymerization. On the other hand, it seems that superswelling can be decreased, or even suppressed, by using appropriate experimental conditions such as the use of oligomeric RAFT agents instead of low molecular weight counterparts. For instance, Vooslo and co-workers [163] prepared dithiobenzoateterminated polystyrene oligomers in bulk and then emulsified the mixture in the presence of hydrophobe to create a stable miniemulsion. No colored surface layer was then observed. The type of surfactant (anionic, cationic or non-ionic) can also have a strong impact on the colloidal stability. In particular, De Brouwer and co-workers [129,136] reported stability problems with styrene, BMA, MMA and HEMA during RAFT miniemulsion polymerization with dithiobenzoates as RAFT agents under ionic surfactant stabilization (SDS or CTAB). In contrast, this instability was not observed with non-ionic surfactant (Igelpal890 or Brij98) and the polymerization was well controlled even though retardation was still observed. To enhance colloidal stability, surfmers (i.e. polymerizable surfactants) [142,164] or functionalized RAFT agent can be used. In particular, carboxylic acid functionalized PS [137] or PMMA [146] particles exhibit a better stability due to the presence of ionizable carboxylic acid groups (coming from the RAFT agent) at the surface of particles, hence increasing the electrostatic stabilization.

198

MINIEMULSION POLYMERIZATION TECHNOLOGY

7.5.2.3

Livingness and Controlled

Polymerization

In some cases, miniemulsion RAFT polymerization can be less controlled and/or living compared to bulk/solution counterparts. Superswelling phenomenon seems to be the main explanation because the redistribution of monomer and low MW RAFT agents between non-nucleated monomer droplets and polymer particles can create multimodal molecular weight distributions [160]. The dramatic influence of superswelling was confirmed by latter studies dealing with fundamental aspects of RAFT miniemulsion polymerization [131,133,134,140,165]. In particular, they showed that in the case of styrene RAFT miniemulsion polymerization mediated by PEPDTA, when the surfactant coverage is lower than 40%, two kinds of particles grew in parallel: polymer and oligomer particles (that are assumed to be induced by the superswelling effect). Both the PSD and the MWD of the final latex are then consequently bimodal whereas a post addition of surfactant prevented oligomer particles formation [131]. Similar results were obtained by Tonge and co-workers [166] as well as by McLeary and co-workers [138] who showed that styrene and MMA miniemulsion polymerization with 4-cyano-4-(thiobenzoyl) pentanoic acid as a RAFT agent yielded good control and good living character but only for high ionic surfactant concentrations (10 wt.% of CTAB or SDS with respect to monomer) and high hexadecane concentration (4 wt.% with respect to monomer). Very recently, Tobita [167] investigated the theoretical MWD obtained for RAFT miniemulsion polymerization depending on the particle diameter. They showed that the MWD decreased with a decreasing of the particle diameter (PDIs were 1.01; 1.09; 1.27 and 1.61 respectively for 50; 100; 150 and 300 nm). These results were in good agreement with previous studies dealing with the surfactant level and the RAFT miniemulsion stability. Luo and co-workers [140] studied the synthesis of high molecular weight polystyrene by miniemulsion RAFT polymerization mediated by PEPDTA. They showed that contrary to a batch polymerization, a two-step RAFT semi-batch polymerization provided narrow molecular weight distribution. The first step consisted in the batch miniemulsion polymerization of a fraction of styrene in presence of the RAFT agent. Once 80% of monomer conversion were reached, the remaining amount of styrene and additional water were continuously added. A 80,000 g.mol"1 polystyrene was obtained with a PDI of 1.35, significantly narrower than for the equivalent bulk system.

CONTROLLED/LIVING RADICAL POLYMERIZATION

199

Smulders and co-workers [168] reported a similar result while using a xanthate-mediated nBA seeded emulsion polymerization. Finally, an interesting study of Zhang and co-workers [150] showed that during the CPDB-mediated miniemulsion polymerization of nBMA, ß-cyclodextrin acted as both a stabilizer and a solubilizer, thus helping the transport of the hydrophobic low MW RAFT agent to the droplets and particles. In particular, latexes obtained with ß-cyclodextrin exhibited a narrower MWD (-1.3).

7.5.2

RAFT Miniemulsion Polymerization of Vinyl Acetate

Vinyl acetate (VAc) is a highly important monomer as its polymer is used in several applications such as adhesives, paints or biomédical materials. In particular, poly(vinyl acetate) is the precursor to poly(vinyl alcohol), which is a polymer of high importance in the pharmaceutical area. Poly(vinyl alcohol) presents the advantages of being water-soluble, non-toxic, together with interesting bioadhesive properties. VAc belongs to the "less activated monomers" class as the radical derived from VAc is highly reactive (i.e. poorly stabilized) and has the tendency to extensively undergo chain transfer and chain termination reactions. In addition, because of its reactivity, the VAcderived radical will quickly add to activated double bonds such as the C=S double bond of RAFT agents. Consequently, in the case of the RAFT process, the intermediate radical will be relatively stable with a low tendency to fragment as VAc propagating radical is a poor living group. Attempts to produce PVAc homopolymer by NMP or ATRP have so far met with very limited success [169-172]. The controlled polymerization of VAc by RAFT has been reported recently with both dithiocarbamates and xanthates [173-178], whereas dithioesters RAFT agents generally inhibit the polymerization [179]. According to Stenzel and co-workers who studied several xanthates [177], methyl (ethoxycarbonothioyl)sulfanyl acetate (MESA) provided the best results with the higher polymerization rate and the lower MWD. Simms and co-workers [154] as well as Russum and co-workers [155] respectively used watersoluble (VA-060 and VA-044) and oil-soluble (AIBN) initiators for the VAc miniemulsion polymerization mediated by the MESA RAFT agent. Compared to water-soluble initiators, the use of AIBN

200

MINIEMULSION POLYMERIZATION TECHNOLOGY

led to a better agreement between experimental and theoretical molecular weights as well as lower MWDs. Nevertheless, Russum and co-workers observed more irreversible termination events in miniemulsion than in the equivalent bulk system leading to a larger proportion of dead polymer chains and a higher PDI [155].

7.5.3

Nanocapsules Synthesized by RAFT Miniemulsion Polymerization

Extensive research has been devoted to synthesize nanoparticles for the encapsulation of several classes of molecules such as drugs, dyes or perfumes. According to the targeted applications and experimental conditions, several particles morphologies can be designed such as core-shell and hollow particles. RAFT miniemulsion polymerization has recently proven to be an efficient technique to synthesize nanocapsules, as reported by Luo and co-workers [180-182] who developed amphiphilic macroRAFT agents to both control and stabilize interfacially confined miniemulsion polymerization. They first used styrene and maleic anhydride oligomers (SMA-RAFT) synthesized by PEPDTA-mediated RAFT copolymerization [180,182]. The hydrophilicity of SMA-RAFT agent was tuned by the ammonolyze of anhydride groups into carboxyl and amide groups via ammonia treatment. KPS was used as initiator, nonadecane (ND) as core material and styrene as monomer without further addition of stabilizer. Due to their amphiphilic properties, ammonolyzed SMARAFT molecules self-assembled at the water/droplets interface once the miniemulsion is formed via ultrasonication. In a similar way, after addition of several monomer units, radicals derived from the water-soluble initiator also become surface active and anchored at the minidroplets/water interface and then transfer to the RAFT agent. As the radical is located at the interface, the polymerization of styrene is thus confined at the interface and polymer chains grow inwards leading to a polymer shell. The addition of 0.5 wt.% SDS and lowering of the ammonia /anhydride group ratio improved nonadecane encapsulation by narrowing particle size distribution and decreasing the amount of matrix-type polystyrene particles probably coming from homogeneous nucleation (Figure 7.12). Further improvements in terms of narrow particle size distribution and uniform polystyrene shell thickness were obtained by using the same technique but with a poly(acrylic acid)-b-poly(styrene) RAFT agent, together with hexadecane as core material and a postaddition of SDS [181].

CONTROLLED /LIVING RADICAL POLYMERIZATION

201

Figure 7.12. RAFT interfacial miniemulsion polymerization.

7.6 Iodine Transfer Polymerization in Miniemulsion Few studies were actually devoted to ITP in miniemulsion [183-186]. In particular, Charleux and co-workers [183] have revealed that a continuous addition of styrene in a ITP miniemulsion system with C6F]3I as a transfer agent and ACPA as an initiator allowed a linear increase of molecular weights values versus conversion but could not hamper the increase of PDI values. Charleux and co-workers [185] also showed that polystyrene-fr-poly(n-butyl acrylate) copolymers could be obtained in a two steps process. The first step consisted in the batch styrene miniemulsion polymerization with C6F13I as a transfer agent and the second step in the addition of n-butyl acrylate to this seed latex, in one shot or in a continuous way. The polydispersity of the PS block was close to 1.5 but increased to 2 when the poly(n-butyl acrylate) block was formed. More recently, Lacroix-Desmazes and co-workers [186] synthesized poly(styrene)-fopoly(dimethylsiloxane)-fr-poly(styrene) (PS-b-PDMS-fr-PS) triblock

202

MINIEMULSION POLYMERIZATION TECHNOLOGY

copolymers by ITP of styrene in miniemulsion with a preformed oc,o>diiodopoly(dimethylsiloxane) macrotransfer agent that also plays the role of hydrophobe as well. The polymerization revealed a linear increase of experimental MW values versus monomer conversion but they were always higher than the theoretical ones. Moreover, as expected with ITP method, the polydispersity indexes were high, around 1.7-1.8 at the end of the reaction. In a recent article, Lacroix-Desmazes and co-workers [187] reported for the first time the RITP of styrene in miniemulsion with SDS as a surfactant and hexadecane as a hydrophobe. To counterbalance the hydrolysis of iodine into HOI, I3", Γ and I03", hydrogen peroxide was continuously added into the reaction medium under acidic conditions. I2 was then continuously regenerated by oxidation of iodide Γ that overcomes the problem of deviation of the molecular weight observed without H 2 0 2 addition. They then showed that RITP in miniemulsion, upon H 2 0 2 addition, allowed to accurately control the molecular weight of PS chains by finely tuning the initial concentration of I2. They indeed obtained experimental MW values in good agreement with the theoretical ones and PDI values comprised between 1.46 and 1.75.

7.7

Conclusion

The four main CLRP techniques (NMP, ATRP, RAFT and (R)ITP) have been shown to be efficient in the synthesis of well-controlled and living (co)polymers in miniemulsion as well as complex macromolecular architectures. In the case of NMP, the use of the acyclic SGI nitroxide instead of the originally studied cyclic nitroxide TEMPO permitted to decrease the polymerization temperature and to broaden the range of polymerizable monomers in aqueous environment. In particular, the oilsoluble SGl-based alkoxyamine MONAMS enabled the synthesis of well-controlled gradient copolymers whereas the commercially available SGl-based alkoxyamine Blocbuilder™, represents the only example to date of water-soluble monocomponent initiating system. Concerning ATRP, due to the oxygen sensitivity of Cu(I) complexes, reverse, SN&NI and AGET ATRP were preferentially employed. With generally Cu-based catalysts, non-ionic surfactant and hydrophobic ligands, well-controlled and living miniemulsion polymerizations have been performed, yielding block, star and hybrid (co)polymers particles.

CONTROLLED/LIVING RADICAL POLYMERIZATION

203

RAFT miniemulsion polymerization has successfully enabled the synthesis of various homo and block copolymers with a large range of monomers and in particular with VAc. Nevertheless, some major drawbacks had to be overcome to achieve stable and controlled systems. Inhibition and retardation phenomena, probably linked to the exit from particles of radical released from fragmentation of (low MW) RAFT agent, were reduced by using macro-RAFT agent. The presence of an upper colored layer assigned to the superswellig effect and related to colloidal instability together with broad MWD could be avoided by using oligo-RAFT agent, surfmers or a high amount of non-ionic surfactant and hydrophobe. If (R)ITP has been less extensively described in the literature so far than the three previously described CLRP techniques in miniemulsion, the available results are nevertheless very promising in terms of control and latex stability as well as from an industrial point of view. If CLRP in miniemulsion is nowadays a well-understood polymerization system, it's worth mentioning that it has also been an important step towards the understanding and the implementation of true (i.e. ab initio) emulsion processes under controlled/living conditions, which are industrially more appealing for the production of (co)polymers colloids and nanoparticles.

References 1. K. Matyjaszewski, Controlled/Living Radical Polymerizations: From Synthesis to Materials, Vol. 944, Washington, DC, American Chemical Society, 2006. 2. K. Matyjaszewski, Advances in Controlled/Living Radical Polymerization Vol. 854, Washington, DC, American Chemical Society, 2003. 3. Κ.Α. Davis, and K. Matyjaszewski, Advances in Polymer Science, Vol. 159, p. 2, 2002. 4. K. Matyjaszewski, Progress in Polymer Science, Vol. 30, p. 858,2004. 5. W.A. Braunecker, and K. Matyjaszewski, Progress in Polymer Science, Vol. 32, p. 93, 2007. 6. M.F. Cunningham, Progress in Polymer Science, Vol. 27, p. 1039, 2002. 7. M.F. Cunningham, Progress in Polymer Science, Vol. 33, p. 365,2008. 8. J. Qiu, B. Charleux, and K. Matyjaszewski, Progress in Polymer Science, Vol. 26, p. 2083, 2001. 9. M. Save, Y. Guillaneuf, and R.G. Gilbert, Australian Journal of Chemistry, Vol. 59, p. 693,2006. 10. P.B. Zetterlund, Y. Kagawa, and M. Okubo, Chemical Reviews, Vol. 108, p. 3747,2008. 11. J.B. McLeary, a n d B. Klumperman, Soft Matter, Vol. 2, p. 45,2006.

204

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

12. J.K. Oh, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 46, p. 6983, 2008. 13. B. Charleux, and J. Nicolas, Polymer, Vol. 48, p. 5813, 2007. 14. M.F. Cunningham, Comptes Rendus Chimie, Vol. 6, p. 1351, 2003. 15. J.M. Asua, Progress in Polymer Science, Vol. 27, p. 1283,2002. 16. F.J. Schork, Y.W. Luo, W. Smulders, J.P. Russum, A. Butte, and K. Fontenot, Advances in Polymer Science, Vol. 175, p. 129,2005. 17. K. Matyjaszewski and T.P Davis, Handbook of Radical Polymerization, Hoboken, John Wiley & Sons, Inc., 2002. 18. G. Odian, Principles of Polymerization, 4th Edition, John Wiley & Sons, Inc., 2004. 19. M. Szwarc, Nature, Vol. 178, p. 1168,1956. 20. C.J. Hawker, A.W. Bosman, and E. Harth, Chemical Reviews, Vol., p. 3661, 2001. 21. M. Kamigaito, T. Ando, and M. Sawamoto, Chemical Reviews, Vol. 101, p. 3689, 2001. 22. K. Matyjaszewski, and J.H. Xia, Chemical Reviews, Vol. 101, p. 2921, 2001. 23. S. Perrier, and P. Takolpuckdee, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 43, p. 5347,2005. 24. G. David, C. Boyer, J. Tonnar, B. Ameduri, P. Lacroix-Desmazes, and B. Boutevin, Chemical Reviews, Vol. 106, p. 3936, 2006. 25. D.H. Solomon, E. Rizzardo, and P. Cacioli, Polymerization process and polymers produced thereby, U.S. Patent 4,581,429, assigned to CSIRO, 8 April, 1986. 26. M.K. Georges, R.P.N. Veregin,P.M. Kazmaier,and G.K. Hamer,Macromolecules, Vol. 26, p. 2987,1993. 27. E. Rizzardo, and D.H. Solomon, Polymer Bulletin, Vol. 1, p. 529,1979. 28. G. Moad, E. Rizzardo, and D.H. Solomon, Macromolecules, Vol. 15, p. 909, 1982. 29. C.A. Knoop, and A. Studer, Journal of the American Chemical Society, Vol. 125, p. 16327, 2003. 30. D. Benoit, S. Grimaldi, J.-P. Finet, P. Tordo, M. Fontanille, and Y. Gnanou, Controlled/living free-radical polymerization of styrene and n-butyl acrylate in the presence of a novel asymmetric nitroxyl radical, in K. Matyjaszewski, ed., Controlled radical polymerization, ACS symposium series, Vol. 685, Washington, DC, pp. 225, 1998. 31. D. Benoit, S. Grimaldi, S. Robin, J.-P. Finet, P. Tordo, and Y Gnanou, Journal of the American Chemical Society, Vol. 122, p. 5929,2000. 32. D. Benoit, S. Grimaldi, J.-P. Finet, P. Tordo, M. Fontanille, and Y Gnanou, Polymer Preprints (American Chemical Society, Division of Polymer Chemistry), Vol. 38, p. 729,1997. 33. D. Benoit, V. Chaplinski, R. Braslau, and C.J. Hawker, Journal of the American Chemical Society, Vol. 121, p. 3904,1999. 34. L. Couvreur, C. Lefay, J. Belleney, B. Charleux, O. Guerret, and S. Magnet, Macromolecules, Vol. 36, p. 8260,2003. 35. C. Lefay, J. Belleney, B. Charleux, O. Guerret, and S. Magnet, Macromolecular Rapid Communications, Vol. 25, p. 1215,2004. 36. T. Diaz, A. Fischer, A. Jonquieres, A. Brembilla, and P. Lochon, Macromolecules, Vol. 36, p. 2235, 2003.

CONTROLLED/LIVING RADICAL POLYMERIZATION

205

37. K. Schierholz, M. Givehchi, P. Fabre, F. Nallet, E. Papon, O. Guerret, and Y. Gnanou, Macromolecules, Vol. 36, p. 5995,2003. 38. D. Benoit, E. Harth, P. Fox, R.M. Waymouth, and C.J. Hawker, Macromolecules, Vol. 33, p. 363,2000. 39. Y. Guillaneuf, D. Gigmes, S.R.A. Marque, P. Astolfi, L. Greci, P. Tordo, and D. Bertin, Macromolecules, Vol. 40, p. 3108,2007. 40. B. Charleux, J. Nicolas, and O. Guerret, Macromolecules, Vol. 38, p. 5485,2005. 41. J. Nicolas, P. Couvreur, and B. Charleux, Macromolecules, Vol. 41, p. 3758, 2008. 42. J. Nicolas, C. Dire, L. Mueller, J. Belleney, B. Charleux, S.R.A. Marque, D. Bertin, S. Magnet, and L. Couvreur, Macromolecules, Vol. 39, p. 8274,2006. 43. J. Nicolas, L. Mueller, C. Dire, K. Matyjaszewski, and B. Charleux, Macromolecules, Vol. 42, p. 4470,2009. 44. J. Nicolas, S. Brusseau, and B. Charleux, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 48, p. 34,2010. 45. J.S. Wang, and K. Matyjaszewski, Journal of the American Chemical Society, Vol. 117, p. 5614,1995. 46. M. Kato, M. Kamigaito, M. Sawamoto, and T. Higashimura, Macromolecules, Vol. 28, p. 1721,1995. 47. T. Pintauer, and K. Matyjaszewski, Chemical Society Reviews, Vol. 37, p. 1087, 2008. 48. J. Gromada, and K. Matyjaszewski, Macromolecules, Vol. 34, p. 7664,2001. 49. K. Min, H. Gao, and K. Matyjaszewski, Journal of the American Chemical Society, Vol. 127, p. 3825,2005. 50. W. Jakubowski, and K. Matyjaszewski, Angewandte Chemie, International Edition, Vol. 45, p. 4482,2006. 51. T.P. Le, G. Moad, E. Rizzardo, and S.H. Thang, Polymerization with living characteristics, WO9801478, assigned to Du Pont, 15 January, 1998. 52. P. Corpart, D. Charmot, T. Biadatti, S. Zard, and D. Michelet, Method for block polymer synthesis by controlled radical polymerization, W09858974, assigned to Rhodia Chimie, 30 December, 1998. 53. G. Moad, E. Rizzardo, and S.H. Thang, Polymer, Vol. 49, p. 1079, 2008. 54. J. Chiefari, Y.K. Chong, F. Ercole, J. Krstina, J. Jeffery, T.P.T Le, R.T.A. Mayadunne, G.F. Meijs, C.L. Moad, G. Moad, E. Rizzardo, and S.H. Thang, Macromolecules, Vol. 31, p. 5559,1998. 55. G. Moad, E. Rizzardo, and S.H. Thang, Australian Journal of Chemistry, Vol. 58, p. 379,2005. 56. G. Moad, E. Rizzardo, and S.H. Thang, Australian Journal of Chemistry, Vol. 59, p. 669,2006. 57. A. Favier, and M.T. Charreyre, Macromolecular Rapid Communications, Vol. 27, p. 653,2006. 58. G. Moad, E. Rizzardo, and S.H. Thang, Accounts of Chemical Research, Vol. 41, p. 1133, 2008. 59. M. Benaglia, J. Chiefari, Y.K. Chong, G. Moad, E. Rizzardo, and S.H. Thang, Journal of the American Chemical Society, Vol. 131, p. 6914,2009. 60. M. Tatemoto, Y. Yutani, and K. Fujiwara, Process for producing perfluorovinyl ether sulfonic acid derivatives and block copolymers therefrom, W09843952, assigned to Daikin Industries Ltd., 8 October, 1988. 61. M.C. Iovu, and K. Matyjaszewski, Macromolecules, Vol. 36, p. 9346,2003.

206

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

62. P. Lacroix-Desmazes, R. Severac, and B. Boutevin, Macromolecules, Vol. 38, p. 6299, 2005. 63. C. Boyer, P. Lacroix-Desmazes, J.-J. Robin, and B. Boutevin, Macromolecules, Vol. 39, p. 4044, 2006. 64. Boutevin, B. Otazaghine, P. Lacroix-Desmazes, M. Dubreuil, and V. Bodart, a-Fluoroacrylic polymers with low polydispersity index, WO 2003097705, assigned to Solvay, 27 November, 2003. 65. P. Lacroix-Desmazes, R. Severac, B. Boutevin, V. Bodart, and V. Kurowsky, Iodinated organic substances of low molecular mass and process for preparing them, WO 2004094356, assigned to Solvay, 4 November, 2004. 66. D.E.P. Mestach, R.H.G. Brinkhuis, and P.J.M.D. Elfrink, Method for making methacrylate unit-containing low-polydispersity polymers, WO 2004009644, assigned to Akzo Nobel, 29 January, 2004. 67. T. Prodpran, V.L. Dimonie, E.D. Sudol, and M.S. El-Aasser, Macromolecular Symposia, Vol. 155, p. 1,2000. 68. M.F. Cunningham, K. Tortosa, M. Lin, B. Keoshkerian, and M.K. Georges, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 40, p. 2828, 2002. 69. M.F. Cunningham, K. Tortosa, J.W. Ma, K.B. McAuley, B. Keoshkerian and M.K. Georges, Macromolecular Symposia, Vol. 182, p. 273, 2002. 70. K. Tortosa, J.-A. Smith and M.F. Cunningham, Macromolecular Rapid Communications, Vol. 22, p. 957, 2001. 71. M. Lansalot, C. Farcet, B. Charleux, J.P. Vairon, R. Pirri and P. Tordo, ACS Symposium Series, Vol. 768, p. 138,2000. 72. P.J. MacLeod, R. Barber, P.G. Odell, B. Keoshkerian and M.K. Georges, Macromolecular Symposia, Vol. 155, p. 31,2000. 73. M.F. Cunningham, M. Xie, K.B. McAuley, B. Keoshkerian and M.K. Georges, Macromolecules, Vol. 35, p. 59,2002. 74. B. Keoshkerian, P.J. MacLeod and M.K. Georges, Macromolecules, Vol. 34, p. 3594,2001. 75. C. Farcet, M. Lansalot, B. Charleux, R. Pirri and J.P. Vairon, Macromolecules, Vol. 33, p. 8559, 2000. 76. G. Pan, E.D. Sudol, V.L. Dimonie and M.S. El-Aasser, Macromolecules, Vol. 34, p. 481,2001. 77. G. Pan, E.D. Sudol, V.L. Dimonie and M.S. El-Aasser, Macromolecules, Vol. 35, p. 6915,2002. 78. M. Lin, M.F. Cunningham and B. Keoshkerian, Macromolecular Symposia, Vol. 206, p. 263, 2004. 79. T. Nakamura, PB. Zetterlund and M. Okubo, Macromolecular Rapid Communications, Vol. 27, p. 2014, 2006. 80. M. Lin, J.C.C. Hsu and M.F. Cunningham, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 44, p. 5974, 2006. 81. PB. Zetterlund, T. Nakamura and M. Okubo, Macromolecules, Vol. 40, p. 8663, 2007. 82. M.F. Cunningham, D.C.T. Ng, S.G. Milton and B. Keoshkerian, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 44, p. 232,2005. 83. M.F. Cunningham, M. Lin, and B. Keoshkerian, ]CT Research, Vol. 1, p. 33,2004. 84. P.B. Zetterlund, M.N. Alam, H. Minami, and M. Okubo, Macromolecular Rapid Communications, Vol. 26, p. 955,2005.

CONTROLLED/LIVING RADICAL POLYMERIZATION

207

85. M.N. Alam, P.B. Zetterlund, and M. Okubo, Macromolecular Chemistry and Physics, Vol. 207, p. 1732,2006. 86. Y. Saka, P.B. Zetterlund, and M. Okubo, Polymer, Vol. 48, p. 1229,2007. 87. B. Keoshkerian, A.R. Szkurhan, and M.K. Georges, Macromolecules, Vol. 34, p. 6531,2001. 88. J.W. Ma, J.A. Smith, K.B. McAuley, M.F. Cunningham, B. Keoshkerian, and M.K. Georges, Chemical Engineering Science, Vol. 58, p. 1163,2003. 89. C. Farcet, B. Charleux, and R. Pirri, Macromolecular Symposia, Vol. 182, p. 249, 2002. 90. C. Farcet, B. Charleux, and R. Pirri, Macromolecules, Vol. 34, p. 3823,2001. 91. C. Farcet, J. Nicolas, and B. Charleux, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 40, p. 4410, 2002. 92. C. Farcet, C. Burguiere, and B. Charleux, Surfactant Science Series, Vol. 115, p. 23, 2003. 93. C. Farcet, J. Belleney, B. Charleux, and R. Pirri, Macromolecules, Vol. 35, p. 4912, 2002. 94. J. Nicolas, B. Charleux, O. Guerret, and S. Magnet, Macromolecules, Vol. 37, p. 4453, 2004. 95. G. Pan, E.D. Sudol, V.L. Dimonie, and M.S. El-Aasser, ACS Symposium Series, Vol. 801, p. 139,2002. 96. J. Nicolas, B. Charleux, O. Guerret, and S. Magnet, Angewandte Chemie, International Edition, Vol. 43, p. 6186, 2004. 97. J. Nicolas, B. Charleux, O. Guerret, and S. Magnet, Macromolecules, Vol. 38, p. 9963, 2005. 98. J. Nicolas, B. Charleux, and S. Magnet, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 44, p. 4142,2006. 99. J. Nicolas, A.-V Ruzette, C. Farcet, P. Gerard, S. Magnet, and B. Charleux, Polymer, Vol. 48, p. 7029,2007. 100. S. Jousset, J. Qiu, K. Matyjaszewski, and C. Granel, Macromolecules, Vol. 34, p. 6641,2001. 101. K. Matyjaszewski, J. Qiu, N.V Tsarevsky, and B. Charleux, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 38, p. 4724,2000. 102. Y. Kagawa, H. Minami, M. Okubo, and J. Zhou, Polymer, Vol. 46, p. 1045, 2005. 103. M. Li, and K. Matyjaszewski, Macromolecules, Vol. 36, p. 6028,2003. 104. R.W. Simms, and M.F. Cunningham, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 44, p. 1628,2006. 105. R.W. Simms, and M.F. Cunningham, Macromolecules, Vol. 40, p. 860,2007. 106. R.W. Simms, and M.F. Cunningham, Macromolecular Symposia, Vol. 261, p. 32, 2008. 107. R.W. Simms, and M.F. Cunningham, Macromolecules, Vol. 41, p. 5148,2008. 108. M. Li, and K. Matyjaszewski, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 41, p. 3606,2003. 109. M. Li, K. Min, and K. Matyjaszewski, Macromolecules, Vol. 37, p. 2106,2004. 110. M. Li, N.M. Jahed, K. Min, and K. Matyjaszewski, Macromolecules, Vol. 37, p. 2434, 2004. 111. K.E. Min, M. Li, and K. Matyjaszewski, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 43, p. 3616, 2005.

208

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

112. K. Min, H. Gao, and K. Matyjaszewski, Journal of the American Chemical Society, Vol. 127, p. 3825,2005. 113. H. Gao, K. Min, and K. Matyjaszewski, Macromolecular Chemistry and Physics, Vol. 207, p. 1709, 2006. 114. K. Min, J.K. Oh, and K. Matyjaszewski, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 45, p. 1413, 2007. 115. K. Min, W. Jakubowski, and K. Matyjaszewski, Macromolecular Rapid Communications, Vol. 27, p. 594, 2006. 116. F. Stoffelbach, B. Belardi, J.M.R.C.A. Santos, L. Tessier, K. Matyjaszewski, and B. Charleux, Macromolecules, Vol. 40, p. 8813, 2007. 117. L. Bombalski, K. Min, H. Dong, C. Tang, and K. Matyjaszewski, Macromolecules, Vol. 40, p. 7429, 2007. 118. K. Min, S. Yu, H.-i. Lee, L. Mueller, S.S. Sheiko, and K. Matyjaszewski, Macromolecules, Vol. 40, p. 6557,2007. 119. J.K. Oh, F. Perineau, and K. Matyjaszewski, Macromolecules, Vol. 39, p. 8003, 2006. 120. K. Oh Jung, C. Tang, H. Gao, V. Tsarevsky Nicolay, and K. Matyjaszewski, Journal of the American Chemical Society, Vol. 128, p. 5578,2006. 121. J.K. Oh, H. Dong, R. Zhang, K. Matyjaszewski, and H. Schlaad, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 45, p. 4764, 2007. 122. K. Oh Jung, J. Siegwart Daniel, H.-i. Lee, G. Sherwood, L. Peteanu, O. Hollinger Jeffrey, K. Kataoka, and K. Matyjaszewski, Journal of the American Chemical Society, Vol. 129, p. 5939,2007. 123. J.K. Oh, DJ. Siegwart, and K. Matyjaszewski, Biomacromolecules, Vol. 8, p. 3326, 2007. 124. S.W. Prescott, M.J. Ballard, E. Rizzardo, and R.G. Gilbert, Macromolecules, Vol. 35, p. 5417, 2002. 125. G.G. Qi, C.W. Jones, and F.J. Schork, Macromolecular Rapid Communications, Vol. 28, p. 1010,2007. 126. J.P. Russum, C.W. Jones, and F.J. Schork, Macromolecular Rapid Communications, Vol. 25, p. 1064, 2004. 127. J.P. Russum, C.W. Jones, and F.J. Schork, Industrial & Engineering Chemistry Research, Vol. 44, p. 2484,2005. 128. W.W. Smulders, C.W. Jones, and F.J. Schork, Macromolecules, Vol. 37, p. 9345, 2004. 129. J.G. Tsavalas, F.J. Schork, H. de Brouwer, and M.J. Monteiro, Macromolecules, Vol. 34, p. 3938, 2001. 130. X.Y. Huang, E.D. Sudol, V.L. Dimonie, C D . Anderson, and M.S. El-Aasser, Macromolecules, Vol. 39, p. 6944,2006. 131. L. Yang, Y.W. Luo, and B.G. Li, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 44, p. 2293,2006. 132. M. Lansalot, T.P. Davis, and J.P.A. Heuts, Macromolecules, Vol. 35, p. 7582, 2002. 133. Y.W. Luo, R. Wang, L. Yang, B. Yu, B.G. Li, and S.P. Zhu, Macromolecules, Vol. 39, p. 1328, 2006. 134. L. Yang, Y.W. Luo, and B.G. Li, Polymer, Vol. 47, p. 751,2006. 135. A. Butte, G. Storti, and M. Morbidelli, Macromolecules, Vol. 34, p. 5885, 2001. 136. H. de Brouwer, J.G. Tsavalas, F.J. Schork, and M.J. Monteiro, Macromolecules, Vol. 33, p. 9239, 2000.

CONTROLLED/LIVING RADICAL POLYMERIZATION

209

137. H.J. Lee, J.M. Lee, S.E. Shim, B.H. Lee, and S. Choe, Polymer, Vol. 46, p. 3661, 2005. 138. J.B. McLeary, M.P. Tonge, D.D. Roos, R.D. Sanderson, and B. Klumperman, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 42, p. 960, 2004. 139. I. Uzulina, N. Gaillard, A. Guyot, and K. Claverie, Comptes Rendus Chimie, Vol. 6, p. 1375,2003. 140. L. Yang, Y.W. Luo, and B.G. Li, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 43, p. 4972, 2005. 141. A. Bowes, J.B. McLeary, and R.D. Sanderson, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 45, p. 588,2007. 142. H. Matahwa, J.B. McLeary, and R.D. Sanderson, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 44, p. 427, 2006. 143. G.G. Qi, C.W Jones, and J.F. Schork, Industrial & Engineering Chemistry Research, Vol. 45, p. 7084, 2006. 144. Z.Q. Yu, X.L. Ji, and P.H. Ni, Colloid and Polymer Science, Vol. 285, p. 211,2006. 145. S. Kanagasabapathy, A. Sudalai, and B.C. Benicewicz, Macromolecular Rapid Communications, Vol. 22, p. 1076,2001. 146. S.E. Shim, H. Lee, and S. Choe, Macromolecules, Vol. 37, p. 5565,2004. 147. M. Al-Bagoury, K. Buchholz, and E.J. Yaacoub, Polymers for Advanced Technologies, Vol. 18, p. 313,2007. 148. L. Yang, Y. Luo, X.Z. Liu, and B. Li, Polymer, Vol. 50, p. 4334,2009. 149. Z. Qinghua, Z. Xiaoli, C. Fengqiu, S. Ying, and W. Qiongyan, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 45, p. 1585, 2005. 150. F. Zhang, P.H. Ni, Q.F. Xiong, and Z.Q. Yu, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 43, p. 2931, 2005. 151. X.D. Zhou, P.H. Ni, Z.Q. Yu, and F. Zhang, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 45, p. 471,2007. 152. Y.W. Luo, B. Liu, Z.H. Wang, J. Gao, and B.G. Li, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 45, p. 2304, 2007. 153. R. Bussels, C. Bergman-Gottgens, J. Meuldijk, and C. Koning, Polymer, Vol. 46, p. 8546,2005. 154. R.W Simms, T.P. Davis, and M.F. Cunningham, Macromolecular Rapid Communications, Vol. 26, p. 592,2005. 155. J.P. Russum, N.D. Barbre, C.W. Jones, and F.J. Schork, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 43, p. 2188, 2005. 156. C. Barner-Kowollik, M. Buback, B. Charleux, M.L. Coote, M. Drache, T. Fukuda, A. Goto, B. Klumperman, A.B. Lowe, J.B. Mcleary, G. Moad, M.J. Monteiro, R.D. Sanderson, M.P. Tonge, and P. Vana, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 44, p. 5809, 2006. 157. A.D. Peklak, and A. Butte, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 44, p. 6114,2006. 158. S.W. Prescott, M.J. Ballard, E. Rizzardo, and R.G. Gilbert, Macromolecular Theory and Simulations, Vol. 15, p. 70,2006. 159. S.W. Prescott, Macromolecules, Vol. 36, p. 9608,2003. 160. Y.W. Luo, J. Tsavalas, and F.J. Schork, Macromolecules, Vol. 34, p. 5501,2001. 161. J. Ugelstad, K.H. Kaggerud, F.K. Hansen, and A. Berge, Makromolekulare Chemie, Vol. 180, p. 737,1979. 162. J. Ugelstad, P.C. Mork, K.H. Kaggerud, T. Ellingsen, and A. Berge, Adv. Colloid. Interface Sei., Vol. 13, p. 101,1980.

210

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

163. J.J. Vosloo, D. De Wet-Roos, M.P. Tonge, and R.D. Sanderson, Macromolecules, Vol. 35, p. 4894, 2002. 164. A. Guyot, Advances in Colloid and Interface Science, Vol. 108-109, p. 3, 2004. 165. Y.W. Luo, and X.Z. Liu, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 42, p. 6248, 2004. 166. M.P. Tonge, J.B. McLeary, J.J. Vosloo, and R.D. Sanderson, Macromolecular Symposia, Vol. 193, p. 289, 2003. 167. H. Tobita, Macromolecular Theory and Simulations, Vol. 18, p. 120,2009. 168. W. Smulders, and M.J. Monteiro, Macromolecules, Vol. 37, p. 4474,2004. 169. F. Simal, S. Delfosse, A. Demonceau, A.F. Noels, K. Denk, F.I. Kohl, T. Weskamp, and W.A. Herrmann, Chemistry-a European Journal, Vol. 8, p. 3047, 2002. 170. M. Wakioka, K.Y. Baek,T. Ando, M. Kamigaito, and M. Sawamoto,Macromolecules, Vol. 35, p. 330,2002. 171. J.H. Xia, H.J. Paik, and K. Matyjaszewski, Macromolecules, Vol. 32, p. 8310, 1999. 172. A. Debuigne, J.R. Caille, N. Willet, and R. Jerome, Macromolecules, Vol. 38, p. 9488, 2005. 173. D. Charmot, P. Corpart, H. Adam, S.Z. Zard, T. Biadatti, and G. Bouhadir, Macromolecular Symposia, Vol. 150, p. 23,2000. 174. M.L. Coote, and L. Radom, Macromolecules, Vol. 37, p. 590, 2004. 175. M. Destarac, D. Charmot, X. Franck, and S.Z. Zard, Macromolecular Rapid Communications, Vol. 21, p. 1035, 2000. 176. 1A. Favier, C. Barner-Kowollik, T.P. Davis, and M.H. Stenzel, Macromolecular Chemistry and Physics, Vol. 205, p. 925,2004. 177. M.H. Stenzel, L. Cummins, G.E. Roberts, T.P. Davis, P. Vana, and C. BarnerKowollik, Macromolecular Chemistry and Physics, Vol. 204, p. 1160,2003. 178. M.H. Stenzel, T.P. Davis, and C. Barner-Kowollik, Chemical Communications, Vol., p. 1546,2004. 179. E. Rizzardo, J. Chiefari, R.T.A. Mayadunne, G. Moad and S.H. Thang, In Controlled/Living Radical Polymerization: Progress in ATRP, NMP and RAFT, Matyjaszewski, K., Ed.; ACS Symposium Series 768; American Chem Society: Washington, DC, 2000. 180. Y.W. Luo, and H.Y. Gu, Macromolecular Rapid Communications, Vol. 27, p. 21,2006. 181. F.J. Lu, Y.W. Luo, and B.G. Li, Macromolecular Rapid Communications, Vol. 28, p. 868, 2007. 182. Y.W. Luo, and H.Y. Gu, Polymer, Vol. 48, p. 3262, 2007. 183. M. Lansalot, C. Farcet, B. Charleux, J.R Vairon, and R. Pirri, Macromolecules, Vol. 32, p. 7354,1999. 184. A. Butte, G. Storti, and M. Morbidelli, Macromolecules, Vol. 33, p. 3485,2000. 185. C. Farcet, M. Lansalot, R. Pirri, J.R Vairon, and B. Charleux, Macromolecular Rapid Communications, Vol. 21, p. 921,2000. 186. E. Pouget,J. Tonnar, C. Eloy, P. Lacroix-Desmazes,and B. Boutevin,Macromo/ecu/es, Vol. 39, p. 6009, 2006. 187. J. Tonnar, P. Lacroix-Desmazes, and B. Boutevin, Macromolecules, Vol. 40, p. 186, 2007.

8

Inverse Miniemulsion Polymerization of Unsaturated Monomers Ignác Capek Slovak Academy of Sciences, Polymer Institute, Dúbravská cesta, Bratislava, Slovakia and Trenctn University, Faculty of Industrial Technologies, Púchov, Slovakia

Abstract

Inverse miniemulsions can be generated by sonification of the aqueous solution of polar monomers in the unpolar surfactant saturated phase. For the formulation of miniemulsions, a wide variation of nonionic surfactants and amphiphilic block copolymers could be used. The inverse monomer miniemulsion can be easily polymerized to latexes by using water- and oil-soluble initiators, interfacial active initiators and UV and γ irradiation. The traditional and nontraditional inverse miniemulsion polymerizations and copolymerizations are ideal processes for the preparation of hydrophilic polymers and amphiphilic copolymers, polymer nanoparticles and composite nanoparticles. Controlled/living radical polymerization in the inverse miniemulsions provides a versatile nontraditional route for synthesis of hydrophilic and amphiphilic (co)polymers with narrow molecular weight distribution, designed architectures, and useful end functionalites. These procedures can be used to develop novel thermally responsive polymer microspheres, the composite polymer/ inorganic (magnetic) nanoparticles, polymer covered nanoparticles, core/ shell nanoparticles and hollow nanoparticles. Keywords: Inverse miniemulsion, lypophobe, classical and controlled/ living radical polymerization polymerization, kinetics, nanoparticles, nanocomposites.

8.1

Introduction

Inverse miniemulsions are metastable colloids m a d e out of t w o immiscible fluids, o n e being dispersed in the other, in the presence of V. Mittal (ed.) Miniemulsion Polymerization Technology, (211-250) © Scrivener Publishing LLC

211

212

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

surfactants, lypophobes, cosurfactants and some additives. Surfactant molecules are usually added to an inverse emulsion polymerization recipe to provide the system with colloidal stability and they selfassemble to micelles above a critical micellar concentration (CMC) (Scheme 8.1). The hydrophobic surfactant prevents water droplets from coalescence by steric stabilization. The role of a cosurfactant (lypophobe) is to suppress molecular diffusion. Most of the miniemulsions are kinetically stable, but under an ideal condition thermodynamically stable miniemulsions can also be prepared. The traditional radical miniemulsion polymerization is usually much faster than the solution or bulk radical polymerization, the resulting polymer is in a form of latex and it has much higher average molecular weight but broader molecular weight distribution. The miniemulsion polymerization can easily be carried up to a relatively high conversion of monomer to polymer. Most monomer molecules dwell in the monomer-saturated aqueous droplets. The mechanism of inverse miniemulsion polymerization primarily involves monomer droplet nucleation. In some cases when miniemulsion is not enough stable then micellar and homogeneous nucleations can contribute to the overall polymerization. The minidroplets

Scheme 8.1. The schematic representation of the possible solubilization loci for hydrophilic monomer in the inverse micelles and the type of inverse micelles ((1) below CMC and (2) (spheres), (3) (oblates) (4) (rods), (5) (bilayers), and (6) vesicles above CMC).

INVERSE MINIEMULSION POLYMERIZATION

213

may have a very large total droplet surface area and, therefore, may compete effectively with the monomer-swollen micelles for the oligomeric radicals generated in oil phase (with the oil-soluble initiator). After capture of the oligomeric radicals from the continuous phase or by the generation of primary and later oligomeric radicals within the droplets (with the water soluble initiator), the polymerizing minidroplets are then transformed into the monomer-swollen polymer particles. These nucleated droplets or highly monomer-swollen polymer particles might have a larger number of radicals per particle ñ than the monomer-swollen polymer nanoparticles generated by the micellar mechanism. This is the reason why the polymerization kinetics can be governed by pseudo-bulk kinetics rather than by the Smith-Ewart zero-one kinetics. The preferable formation of polymer inside the minidroplets would depress the interaction of radicals with emulsifier or chain transfer of radicals to emulsifier, which generally leads to incorporation of emulsifier, coemulsifier or additives units into the polymer matrix and changes in the physical properties of the latex product. The chemically incorporated emulsifier on the particle surface can increase the colloidal stability of latex particles. The miniemulsion polymerization, if other nucleation mechanisms are eliminated, might produce a latex product with a more uniform particle size distribution (PSD). In the case of appropriately formulated miniemulsions where polymerization is initiated in each droplet and the solubility of the monomer in the continuous phase is low, the ideal, limiting case of a 1:1 copy of the droplets to the particles can be obtained [1]. Inverse miniemulsion polymerization is a process in which a liquid monomer (or a solution of the monomer) is dispersed into a continuous phase using: (1) a high energy input process (ultrasounds, high shear rate...), (2) a surfactant and (3) a compound that is solubilised in the droplets and exhibits extra-low solubility in the continuous phase. Under these conditions, the initial droplet size could be in the range 100-500 nm (Scheme 8.2). For the traditional radical polymerization in inverse miniemulsions, a large variety of hydrophilic monomer can be used, such as hydroxyethyl methacrylate, (meth)acrylamide, or (meth)acrylic acid [2] to prepare polymer and copolymer nanoparticles (latexes). The nonconventional controlled/living radical polymerization (CRP) provides a versatile route for synthesis of (co)polymers with narrow molecular weight distribution, designed architectures, and useful end functionalites [3]. Various methods for CRP include

214

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Scheme 8.2. Schematic representation of direct and inverse droplets.

atom transfer radical polymerization (ATRP) [4], stable free radical polymerization (SFRP) [5], degenerative transfer polymerization (DT) with alkyl iodides [6], reversible addition-fragmentation transfer polymerization (RAFT) [7], and cobalt-mediated radical polymerization (CMRP) [8]. The presence of polymeric chains at the surface of nanoparticles has a significant contribution to the colloidal stability of the dispersions. In most cases, synthetic polymeric surfactants were used as stabilizers for the initial miniemulsion: block, random and graft amphiphilic copolymers [9, 10]. These amphiphilic copolymers can also act as costabilizer, hydrophobe and/or lypophobe. In a few examples, native polymers or hydrophobically modified native and synthetic polymers can be used as polymeric stabilizers of the initial emulsion [11]. Polymeric stabilizers based on dextran, a neutral bacterial polysaccharide, were used for the preparation of both direct and inverse emulsions. Several classes of amphiphilic and associative polymers have been developed: depending on the method of synthesis, the hydrophobes can be end-attached (telechelic polymers) or distributed either statistically or as small blocks in the hydrophilic polymeric backbone

INVERSE MINIEMULSION POLYMERIZATION

215

(multisticker polymers). They are usually obtained either by chemical modification of a precursor polymer or by free-radical copolymerization of the appropriate monomers. The inverse miniemulsion process has [12] shown to lead to multiblock copolymers in which the hydrophobic blocks are of tunable length and number, according to the experimental conditions [13]. The design of such materials with controllable amphiphilic and rheological properties is quite attractive. One of the best methods to prepare composite particles is the inverse miniemulsion polymerization. Owning to the character of the droplet nucleation, inorganic or organic particles or other hydrophilic or hydrophobic additives could be encapsulated inside or on the surface layer of the latex particles depending on the location of additives after miniemulsification [14]. We summarize the most recent research development in the field of inverse miniemulsion polymerization, including the kinetics and mechanism of this process and the preparation of classical polymer, composite and nonconventional nanoparticles (dispersions).

8.2

General

The inverse miniemulsion obtains its stability by using a combination of an effective surfactant and an osmotic pressure agent which is practically insoluble in the continuous phase and prevents the minidroplets from Ostwald ripening. Inverse-emulsions are typically sterically stabilized with a hydrophobic nonionic surfactant (with a low HLB). The process of miniemulsion formation can be best described by applying extreme shear forces to a system consisting of a continuous phase, a dispersed phase (water a n d / o r monomer), a surfactant, and an osmotic pressure agent (Scheme 8.3). The mechanism of miniemulsion polymerization strongly deviates from those for microemulsion and emulsion polymerization. The monomer droplets act as a monomer reservoir in the emulsion polymerization. Microdroplets act as a reaction loci and monomer reservoir in the microemulsion polymerization. For the miniemulsion polymerization, the loci of reaction are the monomer droplets and the mechanism is called droplet nucleation (Scheme 8.4) [15, 16]. As a result, monomer droplets in miniemulsion act as nanoreactors for in situ polymerization, resulting in a one-to-one copy of droplets to latex particles synthesized [17,18].

216

MINIEMULSION POLYMERIZATION TECHNOLOGY

Scheme 8.3. Schematic representation of homogenization of water in the oil continuous phase and the formation o / w miniemulsion droplets (OPA - osmotic pressure agent, M - monomer, W - water).

Scheme 8.4. Schematic representation of inverse miniemulsion polymerization process.

INVERSE MINIEMULSION POLYMERIZATION

217

The inversion miniemulsion polymerization of hydrophilic acrylic and methacrylic monomers (acrylamide (AAm), acrylic acid (AA), methacrylic acid (MA), N-isopropylacrylamide (NIPAM) and others) can be initiated by water-soluble initiators (e.g., potassium persulfate, ammonium persulfate), oil-soluble initiator (e.g., benzoyl peroxide, 2,2'- azobisisobutyronitrile (AIBN),...), surface-active initiators and radiation sources (γ-ray or UV). The basic mechanistic events proceeding in the inverse miniemulsion polymerization are as follows: • generation of the free radicals either in the continuous or disperse phase • entry of the free radicals into the latex particles • propagation of macromolecules within the particles, accompanied by transfers and intra-particle terminations that may occur • exit (desorption) of the free radicals from the small particles • termination, propagation and other fates of various free radicals, which are present in the disperse and continuous phases. The initiation of inverse miniemulsion polymerization starts by primary radicals derived from initiator dissloved in the monomer droplets or entering of radicals from the continuous phase into the monomer saturated aqueous droplet. When we use an oil-soluble initiator (AIBN) then the initiation might proceed by a two-step process. When a water soluble initiator as ammonium peroxodisulfate (APS) is used then the initiation starts directly by reaction of radicals derived from APS with monomer. The water-soluble initiator would greatly suppress the chance of homogeneous (organic phase) nucleation while the use of oil-soluble one could result in an increase in this occurrence. The cage effect approach might be operative in the APS initiated polymerization with the particle size much below 100 nm. Furthermore, the desorption/entry events of monomeric radicals could be operative in small particles. In both cases the rate of polymerization is lowered. The use of an interfacial active initiator provides the start of polymerization at the interface. The probability of starting of a chain growth depends on the location of monomer. In the case of a water soluble monomer the reaction loci are transferred to the

218

MINIEMULSION POLYMERIZATION TECHNOLOGY

particle core. When a hydrophobic monomer is used then the reaction loci are transferred to the continuous oil phase. When the inverse miniemulsion is irradiated by γ-ray, many very active intermediates such as e" a , H* and "OH are then generated owing to the radiolysis of water. These radicals are very reactive and therefore they can react not only with monomer but also with stabilizer or costabilizer located at the particle surface. This enables to form a different structures of polymers and polymer particles (graft copolymers, core/shell, hollow nanoparticles,..). The partitioning of water-soluble monomers (acrylic acid ( AA) or acrylamide (AAm)) in a two-phase water-in-oil (w/o) system may strongly influence the kinetics of inverse miniemulsion polymerization. When the monomer partitioning is operative then: 1) the polymerization of monomer proceeds in the continuous phase and the homogeneous nucleation appears, a n d / o r 2) the adsorption of precipitated oligomers or primary particles by the particles might decrease the colloidal stability of polymer particles. The partitioning of some monomers (acrylic acid) can be depressed by the deprotonation of AA by NaOH. The increased hydrophilicity of monomer decreases its solubility in the continuous oil phase and enhances the stability of inverse miniemulsion.

8.3

Kinetic Studies

There is no true Interval II for the miniemulsion polymerization system. The occasionally observed constant polymerization rate in the conversion range of 20-50% may result from the equilibrium between two opposing effects, that is, the growing population of latex particles and the decreased monomer concentration at the reaction loci. Besides, the gel effect (the Smith-Ewart case 3 or pseudo-bulk polymerization system) may counterbalance the continuously decreased monomer concentration in the polymer particles. The kinetics of inverse miniemulsion polymerization and copolymerization of acryalmide was reported by Capek et al. [19-24]. The conversion-time data for the inverse miniemulsion polymerization of acrylamide initiated by ammonium peroxodisulfate (APS) at different emulsifier (Tw 85) concentrations are shown in Figure 8.1 [21]. The conversion time curves are concave downward and the total conversion is reached during a few minutes after the start of polymerization. A fast polymerization can be attributed to the very high propagation rate constant for acryalamide, the high concentration

INVERSE MINIEMULSION POLYMERIZATION

219

of polymer particles (reaction loci) and compartmentalization of reaction loci. The dependence of the polymerization rate on the conversion (Figure 8.1) can be described by a curve with two rate intervals. The rate of polymerization (R ) first strongly increases up to a maxim u m (R max) and then slightly decreases to the final conversion. The initial increase in the polymerization rate can be attributed to the robust particle nucleation and the gel effect. The maximum polymerization rate lies in the conversion range 20 - 50 %. The decrease in the polymerization rate with increasing conversion is mainly attributed to a decrease of monomer concentration at the reaction loci. The maximal rate of both APS- and AIBN-initiated polymerizations increases with increasing the emulsifier concentration and appears at [Tw 85] = 0.11 mol.dnv 3 and at [Tw 85] = 0.08 mol.dnr 3 in the APS-initiated polymerization and in the AIBN-initiated polymerization, respectively (Table 8.1). This may be discussed in terms of different coverge of particle surfaces for different surfactant concentrations. The decrease in the polymerization rate with increasing [Tw 85] at a high emulsifier concentration range is attributed to the decreased monomer and radical concentrations

Conversion / % 40

60

ε "δ

ε

o

Ü

Time / min

Figure 8.1. Variation of monomer conversion and the rate of polymerization in the inverse miniemulsion polymerization of AAm with Tw 85 concentration, reaction time and conversion. Recipe: 100 g cyclohexane (CH), 35 g water, 5 g AAm, 0.104 g APS, 60°C. [Tw 85] x 107(mol.dm-3CH): (bottom curves) 2.72, (top curves) 13.6 [21].

220

MINIEMULSION POLYMERIZATION TECHNOLOGY

in the polymer particle, the chain transfer events and the exit of monomeric radicals. The diameter (D ) of final polymer particles decreases and the number of final polymer particles (N ) increases with increasing the emulsifier concentration (Table 8.1). The increase in the particle concentration is much more pronounced at higher emulsifier concentration range. The strong increase in the number of polymer particles can be attributed to the robust particle nucleation caused by the increased radical capture efficiency of droplets and homogeneous nucleation as well. The dependence of the rate of copolymerization of AAm and crosslinker (N,N'-methylenebis(acrylamide) (MBA) vs. conversion is described by a curve with the two rate intervals - similar to that observed for the emulsion polymerization of AAm [23,24]. The maximal rate of polymerization slightly increases with increasing the crosslinker (MBA) concentration; R max °c [MBA] X = 0 1 6 (APS). This was attributed to the crosslinked density of polymer network (particles). Crosslinking reactions decrease the amount of soluble monomer(s) by increasing both the amount of polymer and the crosslinking density of the network [25]. After complete conversion of monomers to polymer, only the network and the diluent remain in the reaction system [26-28]. Table 8.1. Variation of kinetic and colloidal parameters in the inverse miniemulsion polymerization of AAm initiated by AIBN (1) or APS (2) with the Tw 85 concentration. 3 [Tw 85] x 102 (mol · dm 3 )

R

p,max

xlO 4

(mol · dnr3 · s"1)

a

D p (nm)

N xl0 1 6 /dm 3 P

ΠΧ10 2 /

particle

(1) (2)

(1)

(2)

(1) (2)

2.72

(1) 15.0 (2) 9.0

200 265

2.0 0.9

3.713.8

5.44

13.7 20.6

162 193

3.8 2.2

3.0 7.7

8.16

14.4 26.2

152 130

4.6 7.2

2.6 3.0

10.88

11.7 29.7

134 110

6.8 12.2

1.4 2.0

13.6

10.6 26.6

124 90

8.5 31.7

1.10.7

100 g cyclohexane, 35 g water, 5 g AAm, 0.2108 g AIBN or 0.104 g APS, 60°C, Tw 85, R max and N are related to the continuous phase (cyclohexane).

INVERSE MINIEMULSION POLYMERIZATION

221

Ge et al. have studied the kinetics of inverse miniemulsion copolymerization of (2-methacryloyloxyethyl) trimethyl ammonium chloride with acrylamide [29]. The rate of polymerization (R ) could be represented by R « D087[M]137[E]°53. Furthermore, it was evidenced that polymerization mostly occurred in the aqueous monomer droplets, regardless of the various conditions. Chen et al. studied kinetic behavior in inverse emulsion polymerization of AAm [30]. In order to avoid coagulation in the toluene system, the amount of emulsifier greater than 10 wt% was needed.

8.4 Traditional and Nonconventional Inverse Latexes 8.4.1 Water Soluble Monomers Colloidal particles of poly(2-hydroxyethyl methacrylate) (PHEMA) homo- and copolymers are often prepared by dispersion polymerization [31]. The inverse miniemulsion polymerization, with the aid of oil soluble surfactants, has also been utilized to synthesize nanometer-sized PHEMA particles dispersed in organic media [16, 32]. Hydroxyethyl methacrylate (HEMA) is a moderately hydrophilic monomer where the inverse miniemulsion polymerization process can be used to obtain PHEMA particles [32]. For most of the experiments cyclohexane as oil phase and PEGA200 (PEGA200, a poly(ethylene oxide) azo initiator (poly(ethylene glycol)isobutyrate with an ethylene oxide molecular weight of 200 g.mol·1) and AIBN as initiators were chosen. It was reported that the addition of a small amounts of water increases the colloidal stability of minidroplets, thus, it acts as a weak lipophobe. Rather small inverse latex particles in the size range between 80 and 160 nm and narrow size distributions are obtained. Independent of the initiator type, it was found that with increasing amount of surfactant, the particle size decreases as expected. The use of AIBN as initiator results in a smaller particle size than the use of PEGA200. The robust droplet nucleation via entry of oligomeric radicals leads to the formation of high number of small nanoparticles. The incorporation of PEG tails might somewhat influence the surfactant decoration of the particles as well as the area per surfactant molecule, Asurf. In the case of AIBN, stable latexes with as low as 1.6 wt% of surfactant KLE3729

222

MINIEMULSION POLYMERIZATION TECHNOLOGY

(a block copolymer consisting of poly(ethylene-co-butylene) and poly(ethylene oxide)) can be successfully synthesized: D(nm)/KLE3729(wt%)/A surf : 149/1.6/25.3,129/4.0/11.5,107/4.8/11.7, 82/9.6/7.7 and 78/16/4.8 The area per surfactant molecule, Asurf, decreases with decreasing particle size and follows the same trend as in the case of direct miniemulsions [33]. This means that the surface coverage is a function of the particle size: for smaller particles more surfactant is required in order to obtain stable latexes. Also organic-inorganic hybrid PHEMA particles embedded with magnetic iron oxide nanoparticles [34, 35] and clays [36] were synthesized. However, these procedures have resulted in preparation of PHEMA with broad molecular weight distribution (i.e. M w /M n >2.0) because of the use of an uncontrolled free-radical polymerization process. The inverse miniemulsion was prepared by mixing a solution of HEMA, crosslinker, and VA-086 (2,2'- azobis(methyl-N-(2hydroxyethyDpropionamide) in water to a dispersion of MMT20 (montmorilonit) in cyclohexane at room temperature [36]. The droplet size of the MMT20-stabilized HEMA inverse miniemulsion was in the size of about 500 - 550 nm. HEMA could be polymerized to form latex particles of about 720 nm; however, the formed latex was less stable than the PAAm latexes. Furthermore, the content of fouling in the reactor formed during the reaction was much higher than the PAAm-based particles. Without the addition of a strong lipophobe, the AAm inverse miniemulsions (in CH) after sonication show only a low stability (less than 1 h) [32]. A polymerization of such an unstable emulsion leads to a large distribution in particle size, coagulum and a large fraction of large particles. The addition of a strong lipophobe NaCl increased the stability of the miniemulsions up to several days.The particle size decreases with increasing amounts of initiator (AIBN) the surfactant (KLE3729): D(nm)/KLE3729(g)/Asurf (nm2): 193/0.12/18,142/0.25/12 and 142/0.5/7 Depending on the particle size, the surface demand per surfactant molecule A , is between 18 nm 2 at low surfactant amounts (0.12)

INVERSE MINIEMULSION POLYMERIZATION

223

and 7 nm 2 for higher surfactant amounts (0.5). Acrylamide particles were also synthesized in a miniemulsion polymerization stabilized by a low molecular weigh surfactant Span 80. In this case, a much lower surfactant concentration is used to obtain particles with a particle size around 80 nm. The inorganic-based composite nanoparticles were prepared by inverse miniemulsion (co)polymerization of AAm and crosslinker initiated by VA-086 in the presence of MMT20 [36]. It is well established that solid particles with special features will selfassembly at the liquid-liquid interface to reduce the interfacial energy between the two immiscible liquids (see Scheme 8.5). Systems with a higher weight fraction of water (ca. 10 wt.%) produced unstable inverse emulsions in the presence of the MMT20 clay platelets. The average diameter of monomer droplets in the AAm/cyclohexane (CH)/water/MMT20 miniemulsions containing around 4 wt% of both water and AAm decreased with increasing the concentration of MMT20 as follows: D d (nm)/ MMT20(wt %): 680/0.27, 600/0.49 and 510/0.69 A schematic illustration of the surfactant-free inverse miniemulsion polymerization in the presence of hydrophobic MMT is given in Scheme 8.6.

Scheme 8.5. Schematic representation of the stabilization mechanism of inverse emulsions. Key: (1) low HLB surfactants around the water droplets; (2,3) solid particles around the droplet; (4) an inverse emulsion stabilized by clay platelets; (5) polymer chains around the droplets.

224

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Scheme 8.6. Schematic representation of the stabilizing function of the hydrophobic clay platelets in the inverse miniemulsion.

The final particles appeared to be about 200 nm larger than the initial emulsion droplets (ca. 700 nm), while there was no significant change for the PDI. DLS measurements also revealed that the particle size increased gradually during the polymerization: Dp (nm)/time(min): 500/0, 740/45 and 890/90 The clay platelets are located at the surface of the latex particle, similar to what has been reported by Bon et al. [37]. The rugged surface morphology obtained by SEM approach clearly indicates that the PAAm latex particles are covered by a layer of clay platelets. A "fluffy" structure followed by the cryo-TEM approach was clearly formed around the particles. A series of magnetic composite nanoparticles were prepared by miniemulsion (co)polymerization of AAm and MBA in the presence of iron oxide nanoparticles [35]. Hydrophilic magnetic nanoparticles with excellent stability were achieved by adding polymethacrylic acid during the preparation of magnetic iron oxide. Such particles dispersed readily in dilute ammonia and formed a stable magnetic fluid. On increasing the amount of Span 80 (sorbitan monooleate), the particle size of the magnetic composite polymer/Fe 3 0 4 nanoparticles decreased to ca. 80 nm. The drop size strongly decreased with ultrasonicating time. An increase in the ferrofluid content in the polymerization system reduced the particle size of the magnetic polymer microspheres. On increasing the amount of Fe 3 0 4 dispersion from 0.25 to 1.4 g, the particle size of the magnetic nanospheres decreased from 160 to 67 nm. This is because during the fusion/ fission process in ultrasonication, the droplets favored to fission than fusion with the increase of the amount of Fe 3 0 4 . The magnetic nanoparticles are supposed to act as lypopbobe. DLS analysis showed

INVERSE MINIEMULSION POLYMERIZATION

225

that the particle size of magnetic polymeric latex fluctuated around 80 nm by variation of MBA concentration. The formation of composite PMMA/KGM (polysaccharide, konjac glucomannan) nanoparticles resulted from the grafting reactions in the inverse miniemulsion system [38]. With increasing amount of AIBN, the graft efficiency (GE) exhibited a maximum value at 1.5 mmol/L. On the one hand, with the increase in initiator concentration, the number of free radicals generated in the system increased greatly, which favored the graft copolymerization and led to an increasing GE. On the other hand, when the initiator concentration was too high, the large amount of AIBN in the system produced excess radicals, and the homopolymerization of AAm occurred, which resulted in a decrease in GE: GE(%)/AIBN(mmol.dm-3): 40/5, 86/10, 95/15, 85/20, 70/25 The decrease in GE at [AIBN] >95 mmol.dm 3 can be discussed in terms of the primary radical termination of less reactive graft radicals. The partitioning of AIBN between the oil and water phases increases the amount of AIBN in the aqueous droplets and its participation in the termination reactions (AIBN is partially soluble in water). Futhermore, the decreased radical entry efficiency of radicals into the particles decreases the rate of grafting polymerization. Under such conditions accumulation of surface active AAm [39] at the particle surface favours the formation of PAAm. As the amount of Span 80 increased, the number of reaction loci (droplets and micelles) increased, which greatly increased GE. However, at Span 80 >6 wt%, the number of reaction loci did not increase and instead became constant. Moreover, the chance of occurring chain-transfer reactions to the surfactant increased accordingly because Span 80 contained many active types of hydrogen. Thus, the GE values reached a maximum and then decreased slightly. The most widely studied thermally responsive hydrogels are based on poly(N-isopropylacrylamide) (PNIPAM), which undergoes a volume transition in water upon heating above 32°C, during which the extended hydrophilic coils collapse into hydrophobic globules [40], this causes the gels shrink and expel most of the absorbed water. The reversible process allows PNIPAM hydrogels to adsorb proteins through hydrophobic interactions at high temperatures (above 32°C), and this is followed by release at low temperatures (below 32°C) [41]. Similarly, PNIPAM hydrogels or nanoparticles can be used in temperature-controlled adsorption/

226

MINIEMULSION POLYMERIZATION TECHNOLOGY

isolation of proteins [42], peptides [43], and other biomolecules [44]. The separation process can be operated without large changes in the environment. Mild conditions for biomolecules can be maintained during the entire separation process. The transition of poly(N-isopropylacrylamide) from hydrophilic to hydrophobic above its lower critical solution temperature (LCST) was adopted in a designed unique interfacial polymerization process by which the temperature-sensitive hollow microspheres could be synthesized in situ [45]. The NIPAM monomer was first dissolved in an aqueous phase and then emulsified with toluene to form a water-in-oil (w/o) emulsion in the presence of Span 80. To conduct an interfacial polymerization at the oil/water interface, a redox initiation system containing benzoyl peroxide (BPO) in oil phase and tetraethylenepentamine (TEPA) in water phase was used as the interfacial initiator. During the reaction process, the reductant TEPA and the oxidant BPO will diffuse to the oil/water interface first to generate free radicals. After that, the polymerization of NIPAM will start spontaneously at the interface. If the polymerization is carried out at the temperature above the LCST of PNIPAM, the PNIPAM will be neither water-soluble nor oil-soluble (toluene as oil phase). For this reason, the formed PNIPAM layer will be restricted at the oil/water interfacial area at the temperature above its LCST. At the same time, the crosslinking agent (divinylbenzene, DVB) in oil phase will also diffuse to the interface and participate in the polymerization. As the polymerization continues, all monomers and cross-linkers are reacted and an insoluble crosslinked PNIPAM network is formed at the interface with a hollow microspheric structure. There are several approaches to prepare sensitive composite polymer/inorganic magnetic microspheres [46]. They involve formation of the inverse miniemulsion by dispersing a hydrophilic polymer, magnetite nanoparticles, a surfactant, and a crosslinking agent in oil [47]. In a second approach, microspheres are obtained by polymerization of an inverse (mini)emulsion prepared from a monomer, crosslinker, magnetite, water, initiator, and surfactant. Another polymerization technique includes inverse suspension polymerization [48]. A major advance would be to develop composite nanoparticles and microparticles that show two or more sensitive stimulies. For example, the temperature sensitivity of PNIPAM and the magnetic properties of superparamagnetic particles can be quickly separated from even complex solutions in magnetic field. Magnetic separation advantageously replaces classical techniques, such as

INVERSE MINIEMULSION POLYMERIZATION

227

centrifugation, filtration, or chromatography. Use of magnetic particles reduces time-consuming manipulation steps, and purification and concentration of captured biopolymers in small volumes. In magnetic separations, the isolated cells or enzymes are subject to very low mechanical stress compared with centrifugation or filtration. Suspensions of small particles containing immobilized enzymes enable reproducible pipetting of small volumes, which is important from the practical point of view. Mackova et al. have investigated the effect of various reaction parameters on the inverse (mini)emulsion polymerization of NIPAAm yielding excellent monodisperse, thermally responsive nano- and microspheres [49]. This process involves aqueous ferrofluid droplets containing water-soluble monomer dispersed with the aid of oil-soluble surfactants in continuous organic phase by sonication. NIPAM and MBA were copolymerized in the presence of iron oxide nanoparticles under Span 80 emulsification and oil-soluble initiators such as AIBN and 2,2'-azobis(2-methyloctanenitrile), (AMON). The magnetic P(NIPAM-co-20%MBA) microspheres obtained with 10 wt% y-Fe203 in the monomers had a spherical shape with a smooth surface and substantially smaller size. They were rather polydisperse and contained about 8 wt% Fe suggesting that all the iron oxide present in the feed was incorporated in the particles. It was observed that a narrow particle size distribution can be obtained only at a certain concentration of iron oxide in the dispersed phase, from which magnetic microspheres are formed by polymerization. This concentration corresponded to 5-10 wt% of iron oxide in monomers, when individual spherical particles were formed and the system seemed to be well stabilized (with D ca 5 μιη). At 15-20 wt% y-Fe203 in monomers, partly irregular particles with D ca 2 μτη were observed and at more than 20 wt% of iron oxide the product completely aggregated. As expected, the particle size decreased with increasing concentration of emulsifier and initiator: D ^m)/Span 80(wt.%): 4.5/1.5, 3.6/2.5,3.8/3.5,2.8/4.2, 2.2/5.0 D (um)/AMON(wt.%): 2.7/0.1, 2.0/0.2,1.6/0.22,1.4/0.33,1.2/0.43 P(NIPAAm-co-MBAAm) spheres were prepared by APS-initiated suspension copolymerization in cotton seed oil [48]. If water-soluble APS was the initiator, maghemite colloid partly aggregated, and, as a consequence, rather large, about 40 μηι in diameter, magnetic crosslinked PNIPAAm particles were formed.

228

MINIEMULSION POLYMERIZATION TECHNOLOGY

Poly(acrylic acid) nanoparticles were prepared by the inverse miniemulsion polymerization of acrylic acid (AA) in the presence and absence of a crosslinker (diethylene glycol diacrylate) [32]. To decrease the solubility of acrylic acid in cyclohexane a weak lypophobe (water) was added to the system. A further significant decrease in the oil-solubility of acrylic acid in cyclohexane was observed by the addition of the NaCl solution. Using the NaOH solution, the partitioning of acrylic acid was shifted more into the water phase, and at a 2:1 ratio NaOH: acrylic acid only 7% of the acrylic acid was found to be soluble in cyclohexane. With increasing the surfactant concentration decreased the particle size: D(nm)/KLE3729/Asurf (nm2): 101/2.5/25,80/5.0/16, 76/6.0/13.2 In the case of low surfactant concentration, Asurf is 24.9 nm 2 and decreases to 13.2 nm 2 in the case of higher amount. The used block copolymer turned out to be a very efficient stabilizer. Since the interfacial tensions of miniemulsions before and after polymerization were very similar, it was concluded that droplet nucleation was the main initiation mechanism. Acrylic acid was terpolymerized with sodium acrylate (SA) and MBA the crosslinking agent MBA in the inverse miniemulsions [50]. The miniemulsion was stabilized by Span 80 and initiated by a redox system sodium metabisulfite (SMBS) and APS. The polymerization proceeded rapidly within several minutes on mixing under ultrasonication. This can be attributed to the high rate of initiation and robust nucleation of monomer droplets. TEM photographs of latex particles without costabilizer demonstrated a formation of core-shell structure, in which poly(acrylic acid) (PAA) was formed in the shell layer and water was present in the core. The morphology was resulted from the phase separation of PAA in water and an enhanced tendency of PAA diffusing out to the solvent phase. When NaCl was used as the lypophobe, the poor stability of miniemulsion resulted in large amount of latex particles produced from homogeneous nucleation during the synthesis. Using sodium hydroxide with the molar ratio of N a O H / A A (neutralization degree) in the formulation varied from 0.36, 0.54, to 0.72 no phase separation was observed before and after the polymerization. Furthermore the conversion was almost complete if the molar ratio of costabilizer/monomer was higher than 0.36.

INVERSE MINIEMULSION POLYMERIZATION

229

The results indicated that NaOH not only introduced an osmotic pressure but also increased the hydrophilicity of the acrylic acid monomer by deprotonation of carboxylic acid. The particle nucleation and growth loci were mainly in original droplets and only a small amount of particles was produced from homogeneous nucleation when NaOH was used as the costabilizer and the surfactant concentration was less than its CMC value (Span 80, CMC -3.2 g in 80 ml of cyclohexane) [51]. NaOH not only enhanced the hydrophilicity of monomer droplet but also favored the synthesis of spherical latex particles due to the increase of interfacial tension. By comparing the size distributions of monomer droplets and latex particles, two distribution curves were nearly identical when the surfactant concentration was below its CMC in SI ([NaOH]/[AA] = 0.54,1 g Span 80), but a shrinking feature from droplets to latex particles was observed in S2 ([NaOH]/[AA] = 0.54, 3.5 g Span 80). This result revealed that a micellar/homogeneous nucleation mechanism was significant when the surfactant concentration was higher than the CMC (S2). The average size of droplets was in the order of Sl(ca. 200 nm) >S2 (ca. 100 nm) as expected. The higher concentration of surfactant provided more surface coverage and reduced the size of droplets. Luo et al. have combined the hydrophilic poly(acrylic acid-cosodium acrylate)/P(AA-SA) latex particles with hydrophobic oleic acid (OA) -ZnO nanoparticles into composite latex particles by inverse miniemulsion polymerization [52]. The polymerization was initiated by a redox initiators (APS and SMBS). This condition was used to decrease the solubility of monomer in cyclohexane, and to enhance the droplet nucleation. The surfactant (Span 80) concentration was controlled below the CMC value to avoid the micellar nucleation. The results showed that sodium hydroxide was more effiecient than sodium chloride (see above). With increasing the concentration of costabilizer from 0.36, 0.54, to 0.72 ((the molar ratio of N a O H / A A ) , sample A l o to A3o, the hydrophilic nature of monomer (AA-SA) was enhanced as well as the size of monomer droplets (without OA-ZnO NPs, respectively). According to the formulation in preparing P(AA-SA) latex particles, OA-ZnO was introduced into the continuous phase and the same procedure was adopted to synthesize the composite latex particles. The results showed that droplet nucleation was the dominant route when OA-ZnO was

230

MINIEMULSION POLYMERIZATION TECHNOLOGY

present. The particle sizes in Alo-A3o mostly ranged from 100 nm to 300 nm. On the other hand, for the corresponding composite latex particles, two differences were revealed. First, small nanoparticles, ZnO, with size less than 25 nm appeared in the TEM photographs and located on or around the surface of P(AA-SA) latex particles. Second, the morphology of ZnO/P(AA-SA) composite latex particles was more spherical than pure latex particles. The growth mechanism of composite latex particles in the presence of OA-ZnO was proposed in Scheme 8.7.

8.4.2

Hydrophobie Monomers

Polymer materials with hydrophobic polymer domains exhibit a range of supramolecular structures and functionalities, which potentially allow for chemical tailoring of the materials properties for targetspecific applications [53]. Kopelman and co-workers have developed the (mini)emulsion polymerization method for producing nanosized polymer hydrogels with fluorescence molecules entrapped in the polymer matrices, i.e., "probes encapsulated by biologically localized embedding" (PEBBLE) [54]. Well-structured and functionalized polymer capsules, "shell cross-linked Knedel" (SCK) [55], have also been produced by Wooley and co-workers from chemical modifications of functional block copolymer micelles [56]. Yang et al. have reported on the development of a facial method for the preparation of hollow superparamagnetic nanocomposite microspheres via inverse miniemulsion polymerization of styrene initiated by γ-ray [57, 58]. The polymerization of styrene is initiated by the free radicals at the surface of magnetic nanoparticles (MNPs) located at oil/water interface. Aqueous monomer saturated minidroplets

Scheme 8.7. Growth mechanism of P(AA-SA)/ZnO composite latex particles.

INVERSE MINIEMULSION POLYMERIZATION

231

dispersed in kerosene are stabilized by Span 80 and a strong lypophobe NaCl. Furthermore, the weak costabilizer (lypophobe) water affects greatly the size distribution of hollow superparamagnetic nanocompostie microspheres. When the content of Span 80 is too little, the stability of the inverse miniemulsion is so bad that the size distribution of the final products is wide. The size of nanoparticles (NPs) (0.27 g Span 80) is in the range of 184-522 nm. Then, with the increase of Span 80 amount to 0.46 g, the well-shape spheres can be obtained and the size is in the range of 246-510 nm. Further, when the amount of Span 80 increases to 0.92 g, the well-shape spheres can also be obtained and the size is in the range of 215-400 nm, indicating that the size distribution becomes better. It is reasonable that more amount of surfactant leads to the smaller size and the better size distribution for emulsion polymerization system. The particle size and the shell thickness (Sthick) increased with increasing the dose rate (DR) (1 g MPs): D(nm)/Sthick(nm)/DR(Gy/min): 1510/210/100, 520/63/81, low/low/63 When the inverse miniemulsion is irradiated by γ-ray, many active intermediates such as e*a, H* and Ό Η are then generated owing to the radiolysis of water. As these radicals try to enter into the oil phase, they first should react with styrene soluble in water and intitiate the formation of oligomeric radicals. Then they also interact with the active hydroxyl groups at the surface of metal nanoparticles located at water/oil interface and abstract hydrogens from them, leading to the formation of free radicals at the surface of MNPs and polymerization at the interface zone. The formation of hollow nanoparticles favours the polymerization of styrene at the surface of MPs. When the initiating radicals ( · Η and ·ΟΗ) produced by γ-rays are located in water phase then they can initiate the production of composite Fe 3 0 4 /PSt particles [59]. When the emulsion is irradiated by γ-rays, the radicals such as · Ο Η and · Η are produced owing to the radiolysis of water. The polymerization of styrene is initiated by the radicals at the organic layer of ammonium oléate as well as in the aqueous phase. A small amount of styrene is dissolved in water. Polymerization advances by continuous diffusion of styrene into these particles from oil phase (cyclohexane) through interface to the aqueous droplets. After polymerization, the magnetic nanoparticles are encapsulating in polymer core. The composite particles are spherical in shape and the magnetite particles are

232

MINIEMULSION POLYMERIZATION TECHNOLOGY

encapsulated by PSt. The average size of three samples is approximately 150 nm, 200 nm and 250 nm, thus, the average size of composite particles increases with the amount of styrene. However, when 15% silicon oil is added into continuous phase, a majority of particles are very small and some pure PSt particles exist. This can be attributed to the micellar or homogeneous nucleation of particle in the continuous phase. Sarkar et al. have reported a new interfacial radical polymerization method for synthesizing novel polymer capsules with a hydrophobic poly(tert-butyl acrylate) shell and a hydrophilic poly(allylamine) interior [60]. This method allowed the productions of narrowly dispersed polymer capsules with the nominal diameters of the capsules variable from 50 to 1000 nm by controlling the polymerization conditions. This approach is based on the generation of the initiating polyradical on a water-soluble amine polymer in the aqueous phase (of the inverse (mini)emulsion) by hydrogen peroxide and subsequent grafting of a hydrophobic vinyl monomer to the polyradical from the water/oil interface [61]. The macroradicals generated in situ act as emulsifier molecules and stay at the water/oil interface of the inverse emulsion which is further stabilized by Span 80 [62].

8.5 Controlled Radical Miniemulsion Polymerization When polymerization occurs within the aqueous droplets conventional stable colloidal particles can be prepared upon the addition of radical initiators. However, due to the use of uncontrolled freeradical polymerization processes (RP), all of the inverse miniemulsion polymerizations resulted in the preparation of polymers with broad molecular weight distribution (i.e., M w /M n >2.0) and without chain-end functionality. Controlled/living radical polymerization (CRP), however, provides a versatile route for synthesis of (co) polymers with narrow molecular weight distribution, designed architectures, and useful end functionalites [3]. CRP reactions have been examined in heterogeneous aqueous dispersions primarily due to environmental benefits, better process control, and the feasibility of commercial production of latex particles [63]. Reversible addition fragmentation transfer (RAFT) polymerizations and atom transfer radical polymerization (ATRP), have

INVERSE MINIEMULSION POLYMERIZATION

233

been demonstrated to be potentially useful, and are important new approaches to the production of polymers with well-defined or special architectures in a controlled manner [4,64,65]. By applying RAFT polymerization to inverse miniemulsion, one can in principle combine the advantages of both of these techniques and offer a convenient way to synthesize unique or well-defined structured polymer colloids such as hydrophilic polymers and nanogels or amphiphilic block copolymers. Matyjaszewski et al. have reported the first CRP of water-soluble monomers in an inverse miniemulsion, a process that allows for the controlled synthesis of nanometer-sized colloidal particles of water-soluble polymers [66]. This process involves aqueous droplets (including water-soluble monomers), stably dispersed with the aid of oil-soluble surfactants in a continuous organic media by sonification [16]. McCormick and co-workers have used RAFT polymerization for the preparation of water soluble polymers and they also followed the kinetics of polymerization as well. The deviation from pseudo first order kinetics at higher conversion was accompanied with a decreased polymerization rate already at medium conversions where RAFT agent hydrolysis and aminolysis influenced the polymerization process [67]. Albertin et al. also observed the RAFT agent hydrolysis in the RAFT polymerization of methacrylic glycomonomer in the presence of added base. They claimed that hydrolysis of the free RAFT agent and the end-of-chain dithiobenzoyl groups was promoted by the high pH of the solution and the hydrolysis could lead to the deviation from a well-controlled RAFT process [68]. A new strategy was developed for the synthesis and functionalization of nanometer-sized colloidal particles of well-controlled, watersoluble polymers [66, 69]. This approach involves the utilization of a new initiation process for ATRP-named activators generated by electron transfer (i.e., AGET ATRP) for the polymerization of watersoluble monomers in the inverse miniemulsion. The additional introduction of a functional crosslinker, such as a disulfide [70], allows the synthesis of crosslinked degradable nanoparticles. Application of this methodology results in the preparation of materials with the following features: (1) The colloidal particles preserve a high degree of halide end functionality. This halide functionality enables further chain extension and formation of block copolymers and/or functionalization with biorelated molecules. (2) Due to the utilization of a

234

MINIEMULSION POLYMERIZATION TECHNOLOGY

controlled radical polymerization process, namely ATRP, nanogels with a uniformly crosslinked network can be prepared [71]. (3) In a reducing environment the crosslinked nanogels can degrade to individual polymeric chains with relatively narrow molecular weight distribution (M / M < 1.5). w

n

Variations of kinetic, colloidal and molecular weight parameters with initiator, costabilizer and water contents were investigated in the inverse miniemulsion copolymerization of oligo(ethylene glycol) monomethyl ether methacrylates (OEOMA) with different molecular weights, OEOMA300 (M = 300 g/mol, pendent EO units DP = 7), a OEOMA475 (M = 475 g/mol, pendent EO units DP ~9)), and OEOMA1100 (M = 1100 g/mol, pendent EO units DP = 23), was chosen as the water-soluble monomer [66, 72] (Scheme 8.8). Span 80 formed a stable inverse miniemulsion of OEOMA with water in cyclohexane, allowing for the synthesis of stable P(OEOMA) particles. A water-soluble, poly(ethylene oxide) (PEO) - functionalized bromoisobutyrate (PEO5000-Br) was used as the initiator for the preparation of PEO-block-P(OEOMA) copolymers. Tris[(2-pyridyl)methyl]amine (TPMA) as the ligand forming a complex with CuBr2 was used to stabilize the ATRP activator and deactivator. The AGET ATRP process involves the use of an oxidatively stable Cu(II) precursor that can generate the active Cu(I) catalyst by reaction with nonradical-forming reducing agents. Water-soluble ascorbic acid (AscA) was used as the reducing agent. The successful synthesis of uncrosslinked colloidal particles consisting of PEO-block-P(OEOMA) copolymers with M w /M n < 1.3 was demonstrated using OEOMA (OEOMA300 and OEOMA1100) macromonomers (Scheme 8.8). DLS measurements show the size of the P(OEOMA300) and P(OEOMAllOO) colloidal particles to be around 120-150 nm and 188-207 nm in diameter with narrow size distribution and high stability. Polymerization was first order, indicating a constant

Scheme 8.8. Illustration of inverse miniemulsion AGET ATRP.

INVERSE MINIEMULSION POLYMERIZATION

235

concentration of active centers during the reaction. Molecular weight increased monotonically with conversion, and polydispersity remained low, M w / M n < 1.3 up to 90% conversion. The particle size of P(OEOMA300) was smaller than that of P(OEOMAllOO). This difference can be attributed to the more hydrophobic character ofOEOMA300. In the presence of disulfide-functionalized dimethacrylate (DMAPEOSS) the obtained crosslinked P(OEOMA300) particles did not dissolve in any solvent, including THF and water. DLS measurements indicate that the size of the crosslinked particles dispersed in cyclohexane was 260 nm in diameter with a relatively broad size distribution. In general, the size of crosslinked particles was larger than that of uncrosslinked particles dispersed in cyclohexane. These particles showed excellent colloidal stability in dispersion media. The size determined by atomic force microscopy (AFM) analysis was 1.6-2.3 times larger (ca. 425 nm), compared to the values determined by DLS measurements. The ATRP approach was also applied on the inverse miniemulsion ATRP of water-soluble HEMA [73]. AGET ATRP) was utilized [69] and used to prepare nanoparticles of well-controlled double-hydrophilic PEO-b-PHEMA diblock and PHEMA-b-PEOb-PHEMA triblock copolymers. These block copolymers self-assembled in water into stable micellar nanoparticles with PHEMA core and PEO shell (Scheme 8.9). AGET ATRP of hydrophilic HEMA was successfully conducted in cyclohexane inverse miniemulsion, resulting in the formation of nanoparticles of well-controlled PHEMA-containing block copolymers [73]. A stable inverse miniemulsion of HEMA formed in the presence relatively small amount of KLE (0.12 g in 24 g cyclohexane) (KLE is a block copolymer consisting of poly(ethylene-co-butylene) (59 wt %) and poly(ethylene oxide) (41 wt %); molecular weight is Mn = 8100 g/mol). A KLE polymer with similar composition (44 wt% EO, Mn = 6600 g/mol) has earlier been used to prepare stable HEMA inverse miniemulsions [34]. Increasing the amount of block copolymer surfactant resulted in faster polymerization and smaller particle size. The rate of polymerization with 1.4 wt% KLE was slightly faster than that observed in the presence of 0.6 wt% KLE. DLS measurement shows the size of the particles to be 146 nm in diameter with relatively narrow size distribution. Molecular weight increased linearly with conversion (1.4 wt% KLE). Molecular weight distribution was narrow, with M w /M n -1.2-1.3 up to over 70% conversion.

236

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Scheme 8.9. (a) Preparation of nanoparticles of PHEMA-containing block copolymers by AGET ATRP of HEMA in inverse miniemulsion. (b) Core-shell micellar nanoparticles of PHEMA-b-PEO-b-PHEMA triblock copolymers through self-assembly in water.

RAFT inverse miniemulsion polymerization also provides a potential route to facilitate process control of the homopolymerization and copolymerization of water soluble monomers (acrylamide) [74]. This polymerization system contains except of basic components as a hydrophobic surfactant (B246SF), cyclohexane, water, initiator (a water soluble 4,4'-azobis(4-cyanovaleric acid) (ABCP) and an oil'soluble AIBN), acrylamide, a costablizer (MgS04) also a RAFT agent (2-(2 carboxyethylsulfanyl thiocarbonylsulfanyl)propionic acid). When the water-soluble initiator ABCP was used, the polymerization proceeded in a controlled/living manner at low conversion and pseudo-living manner at high conversion both in inverse miniemulsion and solution. When AIBN was utilized, however, a significant loss of control was observed with prolonged reaction time. The poly(acrylamide) latexes produced in both experiments had good colloidal stability. The polymerization rates, both decreased with reaction time. As in the solution polymerization, the molecular weights in both experiments showed a deviation from the predicted value after the conversions reached -50%. However, despite these similarities, there were some significant differences between the two inverse miniemulsion polymerizations. The molecular weight

INVERSE MINIEMULSION POLYMERIZATION

237

of poly(acrylamide) in Expt. 2 (AIBN) was much higher than that of Expt. 3 (ABCP) at the same conversion. The PDIs in Expt. 2, as high as 1.7, were much broader than that of Expt. 3, which is indicative of less control in Expt. 2. It should be noted that the degree of livingness/control at low conversions is comparable to the only other report of inverse miniemulsion polymerization with a control agent present [72]. In those works, good control was achieved up to a conversion of 65-80%. The RAFT polymerizations here appear well controlled at conversions up to 50% and only deviate at higher conversions. The use of a buffer solution in the aqueous phase to suppress the suspected RAFT agent hydrolysis or aminolysis may improve the control/ livingness in the inverse miniemulsions [75].

8.6 Amphiphilic and Associating Copolymers Water-soluble polymers and amphiphilic copolymers are presently the subject of extensive research because of their important applications as stabilizers, flocculants, compatibilizers and absorbants [76]. Hydrophobically modified polyacrylamides form an important class of amphiphilic and associating polymers. They are usually obtained by polymerization in solution or by a micellar polymerization technique [77]. Polymerization reactions in inverse (mini)emulsions are choice methods for the synthesis of high molecular weight, water-soluble polymers in the form of latexes, that is, water-swollen polymer particles dispersed in an organic, continuous phase [78]. The advantages of the techniques are the high solid contents (25-35 wt%) with low viscosities of the dispersions, good handling, and storage facilities. Pabon et al. [79] showed that the use of an inverse (mini)emulsion polymerization route allows one to increase considerably the solid contents in the final products (up to 25 wt %). The associative copolymers were formed of an acrylamide/sodium acrylate backbone hydrophobically modified with small amounts (~ 0.5 mol %) of a series of amphiphilic comonomers, isooctylphenoxy-poly(oxyethylene)(n) methacrylates (with 1 < n < 12.5). This process yields high molecular weight polymers encapsulated within water droplets dispersed in an organic medium, which facilitates their subsequent use in many potential applications. The presence of polymeric chains at the surface of nanoparticles has a significant contribution to the colloidal stability of the dispersions especially at high ionic strength because of the polymeric chains

238

MINIEMULSION POLYMERIZATION TECHNOLOGY

acting as steric barriers against aggregation [9, 10]. In most cases, synthetic polymeric surfactants were used as stabilizers for the initial miniemulsion: block, random and graft amphiphilic copolymers. These amphiphilic copolymers can also act as costabilizer, hydrophobe and/or lypophobe. The polymeric surfactant can be a macromonomer which leads to an irreversible anchoring of the polymeric chains at the surface of the particles. In a few examples, a native polysaccharide or hydrophobically modified polysaccharides were used as polymeric stabilizers of the initial emulsion. This process led to polysaccharide-covered nanoparticles [11]. Polymeric stabilizers based on dextran, a neutral bacterial polysaccharide, were used for the preparation of both direct and inverse emulsions. Durand et al. have examined the use of these amphiphilic polysaccharides as stabilizers in miniemulsion polymerization [80]. Rotureau et al. have reported the preparation of polysaccharide-covered nanoparticles by free-radical inverse miniemulsion polymerization process, involving the use of amphiphilic polysaccharides [81]. Depending on the chemical structure of the amphiphilic polysaccharides, inverse miniemulsions were used as reaction media, leading to different kinds of nanoparticles. Hydrogel nanoparticles were prepared by the inverse miniemulsion polymerization of acrylamide and MBA initiated by AIBN in the presence of sodium sulfate as a lypophobe. That kind of nanoparticles could encapsulate hydrophilic molecules provided that they have a sufficient hydrodynamic volume to be retained within the network. In that case, highly hydrophobic dextran derivatives with a very low water-solubility were used as stabilizers during the polymerization. Amphiphilic polysaccharides based on dextran, a neutral bacterial polysaccharide have been used in the inverse miniemulsion polymerization. Dextran is composed of glucose units linked in a-1,6 and consists in slightly branched chains [82]. Randomly modified polysaccharides with varying amount of hydrocarbon groups attached to glucose units were prepared. Two polymer series, Dex26KPy and Dex26KPy org were obtained with attached hydroxylated octyl groups and phenoxy groups. Emulsifying the epoxide into an alkaline solution of dextran (dex) led to a third series of polymers noted Dex26KPy aq [83]. Anionic dextran derivatives were prepared by modification of Dex26KPy or Dex26KC6y, replacing epoxide by with 1,3-propane sulfone. Various amphiphilic polymers were obtained differing by the charge density of the polysaccharide backbone. Finally, polylactide-grafted dextran copolymers

INVERSE MINIEMULSION POLYMERIZATION

239

were prepared and named as Dex33K-g-yPLAm, where m is the number average molar mass of polylactide (PLA) grafts. Inverse miniemulsions were obtained with Dex26KPy and Dex33K-g-yPLAm polymers provided that the polymers exhibited a significant solubility in glycerol triacetate. This solubility was obtained either by increasing the degree of substitution (for Dex26KPy polymers) or by increasing the length of the side chains (for Dex33K-g-yPLAm polymers). Droplet diameter of water-inglycerol triacetate emulsions stabilized by Dex33K-g-36PLA2.3K as a function of the weight ratio of polymeric surfactant to dispersed phase (°=D) varies as follows: D(nm)/ « D : 280/0.01,260/0.02, 215/0.042,170/0.06 For water-in-glycerol triacetate emulsions, sodium sulfate appeared to be the most efficient lypophobe for lowering the rate of ageing at a given concentration of 0.5 mol/L, for all polymeric surfactants (either Dex26KP180 or Dex33K-g-36PLA2.3K). This fact should be a result of the difference of osmotic pressure of the dispersed phase when changing the salt from sodium chloride to sodium sulfate. The effect of salt concentration had been detailed for water-in-dichloromethane emulsions stabilized by Dex26KP180 with sodium chloride dissolved in the dispersed aqueous phase [83]. For high enough amounts of polymeric surfactant, the size of the final particles is close to that of the initial droplets. With neutral polymeric surfactants, the size of the final particles becomes significantly higher than that of the initial droplets when the relative amount of polymeric surfactant becomes too low. The value of <*D = 0.05 seems to be a correct minimum value for a good control of nanoparticle size. The use of anionic dextran derivatives limits this phenomenon because of the electrostatic repulsions which contribute to colloidal stability. Since droplet kinetic stability during minimemulsion polymerization is controlled by the rate of Ostwald ripening, another way to obtain smaller drops would be to vary the nature and amount of lipophobe. Pabon et al. have extended the inverse miniemulsion polymerization technique to the synthesis of hydrophobically modified water-soluble polymers (HMWSPs) [79]. The hydrophilic backbone of the copolymers was formed from an acrylamide/sodium acrylate (AAm/NaA) mixture. The incorporation of the hydrophobic groups (~0.5 mol %) was achieved by free-radical copolymerization of these

240

MINIEMULSION POLYMERIZATION TECHNOLOGY

water-soluble monomers with a series of amphiphilic comonomers, the isooctylphenoxy- poly(oxyethylene)(n) methacrylates, whose number of EO units was varied from 1 to 12. The process yielded high molecular weight polymers (Mw = 6 x 106) encapsulated within water-swollen droplets dispersed in an organic medium (solid content ~25 wt%). High yields of conversion were reached (99%) with, however, a partial incorporation of the minor hydrophobic component in the copolymer. When a hydrophilic redox initiator was used, the incorporation level increased with the HLB of the amphiphilic comonomer, that is, with the number of EO units. The free-radical copolymerization of a water-soluble monomer, acrylamide (AAm), with a hydrophobic comonomer, either N,Ndihexylacrylamide (diC6) or Ν,Ν-diphenylacrylamide (diPh) in inverse emulsions was used with the aim of designing high solid content materials with controllable properties [84]. The choice of these comonomers was dictated by the fact that the use of N,Ndialkylacrylamides instead of N-monoalkylacrylamides leads to copolymers that not only are homogeneous in composition but also have a more efficient thickening abilities. The best formulation conditions required for the formation of stable emulsions were formulated. In particular, in the case of diPh, which is not soluble in either the organic phase or the aqueous phase, the use of cosolvents is required to facilitate its solubilization in either phase. The process leads to stable inverse latices consisting of water droplets swollen with the hydrophobically modified polyacrylamides and dispersed in the organic phase (polymer solids content -25 wt%). The nature of the redox initiator (hydrophilic or lipophilic) was shown to play an important role on the level of incorporation and on the distribution of the hydrophobic units in the copolymers, which in turn affects their rheological behavior in aqueous solution. The incorporation of diPh in the copolymer was maximized when all of the reaction components are located in the aqueous dispersed phase. The copolymer thus obtained has a statistical microstructure. When diPh was solubilized in the organic phase, the use of a lipophilic initiator enabled increased incorporation into the final copolymer by a factor of 4.

8.7

Conclusion

The miniemulsion technique involves dispersion of a large number of homogenized monomer droplets in oil with the aid of emulsifier,

INVERSE MINIEMULSION POLYMERIZATION

241

coemulsifier, lypophobe, different additives and (co)solvents. The inverse miniemulsion polymerization of water soluble and oil soluble monomers can be initiated by the water-soluble, oil-soluble and surface active initiators and by γ-rays and UV light. In the ideal case, the particle nucleation is primarily controlled by the polymerization in the minidroplets. Under such conditions, the large area of minidroplets promotes the capture of radicals from the continuous phase. When we use a water soluble initiator as APS then the initiation starts directly by reaction of radicals derived from APS with monomer in the monomer minidroplet. When we use the oilsoluble initiator (for example AIBN), initiator derived primary radicals ((CH 3 ) 2 (CN)C· = R·) propagate in an oil phase to form oligomeric radicals (M.·) that can either terminate or enter a particle. Alternatively, the transfer can also occur. The chain - transfer derived monomeric radicals (M·) can start to propagate or exit the particle, leading to a large variety of possible fates, among which propagation, re-entry and termination are the most prominent. The use of an interfacial active initiator provides the start of polymerization at the interface. The probability of starting of a chain growth depends on the location of monomer. In the case of a water soluble monomer the reaction loci are transferred to the particle core. When a hydrophobic monomer is used then the reaction loci are transferred to the continuous oil phase. These reactions enable to form different structured polymers and nanoparticles as graft copolymers, hollow nanoparticles and/or core-shell nanoparticles. When the inverse miniemulsion is irradiated by γ-ray, many very active intermediates such as e* , H* and *OH are then generated owing to the radiolysis of water. These radicals are very reactive and therefore they can react not only with monomer but also with stabilizer or costabilizer located at the particle surface. These reactions enable to form hollow nanoparticles and core-shell nanoparticles as well. For the dispersion of the water soluble monomers in nonpolar dispersion media, surfactants with low HLB values are required. To obtain stable miniemulsions, the strong "lipophobe" should be mixed to the water phase in order to build up an osmotic pressure in the droplets. In the case of appropriately formulated miniemulsions where polymerization is initiated in each droplet and the solubility of the monomer in the continuous phase is low, the ideal, limiting case of a 1:1 copy of the droplets to the particles can be obtained. The final product is usually a colloid dispersion of hydrophilic polymer particles in continuous oil phase. Rather small and narrow

242

MINIEMULSION POLYMERIZATION TECHNOLOGY

distributed latexes in a size range between 50 nm < d < 400 nm were made of hydroxyethyl acrylate, (meth)acrylamide, and (meth) acrylic acid where the low molecular weight hydrophobic stabilizer and nonionic amphiphilic block copolymers. The rate of miniemulsion polymerization practically decreases throughout the reaction. This is attributed to the continuous decrease of monomer concentration at the reaction loci due to polymer chain growth, dilution of the monomer phase by hydrophilic chains of emulsifier, polymer chains, lypophobe, partitioning of monomer between minidroplets and the continuous phase, etc. The dependence of the polymerization rate on the conversion can be described by a curve with two rate intervals. The rate of polymerization first strongly increases up to a maximum and then slightly decreases to the final conversion. The initial increase in the polymerization rate can be attributed to the robust particle nucleation and the gel effect. The maximum polymerization rate lies in the conversion range 20 - 50 %. There is no Interval II with the constant monomer concentration at the reaction loci and constant number of the polymer particles. The decrease in the polymerization rate with increasing conversion is mainly attributed to a decrease of monomer concentration at the reaction loci. On increasing the amount of stabilizer (Span 80, Tween 85, nonionic amphiphilic block copolymers...), the particle size of the polymer and composite nanoparticles decreases. On increasing in the ultrasonicating time, the particle size of the polymer nanoparticles decreases. An increase in the some additive (ferrofluid) content in the polymerization system reduces the particle size of the polymer microspheres. This is because during the fusion /fission process in ultrasonication, the droplets favored to fission than fusion with the increase of the amount of surfactant or cosufactant. The crosslinker favours the formation of smaller and crosslinked polymer particles with decreased swelling ability. Mixing the clay platelets that are dispersed in organic solvents with the monomer aqueous solution results in a slight decrease in the particle size. One of the best methods to prepare composite particles is the inverse miniemulsion polymerization. Owning to the character of the droplet nucleation, inorganic or organic particles or other hydrophilic or hydrophobic additives could be encapsulated inside or on the surface layer of the latex particles depending on the location of additives after miniemulsification. The typical method based on miniemulsion polymerization is to suspend inorganic particles

INVERSE MINIEMULSION POLYMERIZATION

243

and nanosubstances in the dispersed phase and then polymerize the monomer to form composite nanoparticles. In miniemulsion polymerization, the monomer droplets with such additives act as "nanoreactors", the composite polymer/inorganic nanoparticles can be prepared in situ. The copolymerization of hydrophilic with hydrophobic monomer can initiate the formation of randoms or block copolymers at the particle interfaces. The formation of copolymers can be varied by the choice of reaction conditions and the initiator type. Furthermore, the presence of hydrophobic oligomers in the continuous phase can decrease the solubility of acrylamid or acrylic acid monomers in the continuous oil phase. The final result could be the increased colloidal stability of miniemulsion droplets. A major advance would be to develop composite nanoparticles and microparticles that show two or more sensitive stimulies. For example, the temperature sensitivity of poly(N-isopropylacrylamide) and the magnetic properties of superparamagnetic particles can be quickly separated from even complex solutions in magnetic field. Magnetic separation advantageously replaces classical techniques, such as centrifugation, filtration, or chromatography. In magnetic separations, the isolated cells or enzymes are subject to very low mechanical stress compared with centrifugation or filtration. When the inverse miniemulsion is irradiated by γ-ray, many active intermediates such as e\ , H* and *OH are then generated owing to the radiolysis of water. As these radicals try to enter into the oil phase, they first should react with styrene soluble in water and initiate the formation of oligomeric radicals. Then they also interact with the active hydroxyl groups at the surface of magnetic nanoparticles located at water/oil interface and abstract hydrogens from them, leading to the formation of free radicals at the surface of larger nanoparticles and polymerization at the interface zone with the formation of hollow nanoparticles. When the initiating radicals (•H and ·ΟΗ) produced by γ-rays are located in water phase then they can initiated the production of classical polymer particles. This procedure was used to prepare Fe 3 0 4 /PSt magnetic composite nanoparticles. All of the classical inverse miniemulsion polymerizations resulted in the preparation of polymers with broad molecular weight distribution (i.e.,M w /M n >2.0). Controlled/living radical polymerization of water souble monomers in the inverse miniemulsion, however, provides a versatile route for synthesis of (co)polymers with narrow

244

MINIEMULSION POLYMERIZATION TECHNOLOGY

molecular weight distribution, designed architectures, and useful end functionalites. This strategy was also developed for the synthesis and functionalization of nanometer-sized colloidal particles of well-controlled, water-soluble polymers. It involves the utilization of a new initiation process for atom transfer radical polymerization (ATRP) - named activators generated by electron transfer (i.e., AGET ATRP) for the polymerization of water-soluble monomers. The additional introduction of a functional crosslinker, such as a disulfide, allows the synthesis of crosslinked degradable nanoparticles. By applying reversible addition fragmentation transfer polymerization to inverse miniemulsion of water-soluble monomers, one can in principle synthesizes unique or well-defined structured polymer colloids such as hydrophilic polymers and nanogels or amphiphilic block copolymers.

8.8

Acknowledgements

This research is supported by the SAV-FM-EHP-2008-01-01 project, the APVV projects No's 0362-07 and 0030-07, and NFP No. 26240220011.

Abbreviations AA AAm ABCP AFM AGET AIBN AMON APS AscA ATRP BPO CH CMC CMRP CRP diC6

acrylic acid acrylamide 4,4'-azobis(4-cyanovaleric acid) atomic force microscopy activators generated by electron transfer 2,2'- azobisisobutyronitrile 2,2'-azobis(2-methyloctanenitrile) ammonium peroxodisulfate ascorbic acid atom transfer radical polymerization benzoyl peroxide cyclohexane critical micellar concentration cobalt-mediated radical polymerization controlled/living radical polymerization N,N-dihexylacrylamide

INVERSE MINIEMULSION POLYMERIZATION

diPh DLS DMA-PEOSS Dp

DR DT DVB GE HEMA HLB HMWSPs KGM KLE3729 LCST M M MA MBA Mi MMT20 MNPs M /M w'

n

NIPAM NPs OA OEOMA OPA P(AA-SA) PAA PAAm PEBBLE PEGA200 PEO

245

Ν,Ν-diphenylacrylamide dynamic light scattering disulfide-functionalized dimethacrylate diameter of final polymer particles dose rate degenerative transfer polymerization divinylbenzene the graft efficiency hydroxyethyl methacrylate hydrophilic-lipophilic balance hydrophobically modified water-soluble polymers polysaccharide, konjac glucomannan block copolymer consisting of poly(ethyleneco- butylène) (59 wt %) and poly(ethylene oxide) (41 wt %) lower critical solution temperature monomer monomeric radicals methacrylic acid (Ν,Ν' -methylenebis(acry lamide) oligomeric radicals montmorilonit magnetic nanoparticles ratio of average weight molecular weight and the average number molecular weight N-isopropylacrylamide nanoparticles oleic acid oligo(ethylene glycol) monomethyl ether methacrylates osmotic pressure agent poly(acrylic acid-co-sodium acrylate) poly(acrylic acid) polyacrylamide probes encapsulated by biologically localized embedding a poly(ethylene oxide) azo initiator (poly(ethylene glycol)isobutyrate with an ethylene oxide molecular weight of 200 g.mol-1) poly(ethylene oxide)

246

M I N I E M U L S I O N POLYMERIZATION T E C H N O L O G Y

PHEMA PNIPAM PSD PSt RAFT RP R

p ^p,max

SCK SFRP SMBS Span 80 ^thick

TEM TEPA THF TPMA VA-086 W w/o «D

poly(2-hydroxyethyl methacrylate) poly(N-isopropylacrylamide) particle size distribution polystyrene reversible addition-fragmentation transfer polymerization free-radical polymerization processes the rate of polymerization maximum shell cross-linked Knedel stable free radical polymerization sodium metabisulfite sorbitan monooleate shell thickness transmission electron microscopy tetraethylenepentamine tetrahydrofuran tris [ (2-pyridyl)methyl]amine 2,2'-azobis(methyl-N-(2-hydroxyethyl) propionamide water water-in-oil the weight ratio of polymeric surfactant to dispersed phase

References 1. K. Landfester, N. Bechthold, S. Förster, and M. Antonietti, Macromolecular Rapid Communications, Vol. 20, p. 81,1999. 2. J. Barton and I. Capek, in E. Horwood, ed., Radical Polymerization in Disperse Systems", Bratislava, Chichester and Veda, 1994. 3. K.A. Davis, and K. Matyjaszewski, Advances in Polymer Science, Vol. 159, p. 2, 2002. 4. K. Matyjaszewski, and J. Xia, Chemical Reviews, Vol. 101, p. 2921, 2001. 5. C.J. Hawker, A.W. Bosman, and E. Harth, Chemical Reviews, Vol. 101, p. 3661, 2001. 6. M.C. Iovu, and K. Matyjaszewski, Macromolecules, Vol. 36, p. 9346,2003. 7. S. Perrier, and P. Takolpuckdee, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 43, p. 5347,2005. 8. A. Debuigne, J.-R. Caille, and R. Jerome, Angewandte Chemie, International Edition, Vol. 44, p. 1101,2005. 9. M. do Amaral, H. de Brouwer, S. Van Es, and J.M. Asua, Macromolecular Symposia, Vol. 226, p. 167, 2005.

INVERSE M I N I E M U L S I O N POLYMERIZATION

247

10. P. Ni, M. Zhang, L. Ma, and S. Fu, Langmuir, Vol. 22, p. 6016,2006. 11. E. Rotureau, E. Marie, M. Leonard, E. Dellacherie, T.A. Camesano, and A. Durand, Colloids and Surfaces A, Vol. 288, p. 62,2006. 12. E. Volpert, J. Selb, and F. Candau, Macromolecules, Vol. 29, p. 1452,1996. 13. K.D. Branham, G.S. Shafer, C.E. Hoyle, and C.L. McCormick, Macromolecules, Vol. 28, p. 6175,1995. 14. YD. Luo, C.A. Dai, and W.Y. Chiu, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 46, p. 1014, 2008. 15. J. Ugelstad, M.S. El-Aasser, and J.W. Vanderhoff, Journal of Polymer Science, Polymer Letters, Vol. 11, p. 503,1973. 16. [M. Antonietti, and K. Landfester, Progress in Polymer Science, Vol. 27, p. 689, 2002. 17. K. Ouzineb, C. Graillât, and T.F. McKenna, Journal of Applied Polymer Science, Vol. 97, p. 745,2005. 18. K. Landfester, Advanced Materials, Vol. 13, p. 765,2001. 19. I. Capek, Designed Monomers and Polymers, Vol. 6, p. 399,2003. 20. I. Capek, Central European Journal of Chemistry, Vol. 3, p. 291,2003. 21. I. Capek, Polymer Journal, Vol. 36, p. 1,2004. 22. I. Capek, Polymer-Plastics Technology Engineering, Vol. 44, p. 539,2005. 23. U.Yildiz, I. Capek, D. Berek, Y. Sarov, and I.W. Rangelow, Polymer International, Vol. 56, p. 364,2007. 24. I. Capek, L. Fialová, and D.Berek, Designed Monomers and Polymers, Vol. 11, p. 123,2008. 25. [A. Matsumoto, Advances in Polymer Science, Vol. 123, p. 41,1995. 26. H. Matsumoto, H. Hayashi and H. Aota, "Emulsion allyl methacrylate/alkyl methacrylate copolymerizations based on mechanistic discussion of freeradical crosslinking multivinyl polymerizations", in M. Nomura, H. Tobita, and K. Suzuki, eds., Polymeric Microspheres, Kyoto University Press, Chapter 12, pp. 147-162,2007. 27. A. Matsumoto, K. Kodama, H. Aota, and I. Capek, European Polymer Journal, Vol. 35, p. 1509,1999. 28. A. Matsumoto, N. Murakami, H. Aota, J. Ikeda, and I. Capek, Polymer, Vol. 40, p. 5687,1999. 29. X. Ge, Q. Ye, X. Xu, Z. Zhang, and Q. Sun, Radiation Physics and Chemistry, Vol. 50, p. 253,1997. 30. L.W. Chen, B.Z. Yang, and M.L. Wu, Progress in Organic Coatings, Vol. 31, p. 393,1997. 31. S.-H. Cho, J.-H. Ryu, J.-G. Park, and K.-D. Suh, European Polymwe Journal, Vol. 41, p. 2209,2005. 32. K. Landfester, M. Willert, and M. Antonietti, Macromolecules, Vol. 33, p. 2370, 2000. 33. K. Landfester, N. Bechthold, F. Tiarks, and M. Antonietti, Macromolecules, Vol. 32, p. 5222,1999. 34. K. Wormuth, Journal of Colloid and Interface Science, Vol. 241, p. 366,2001. 35. Z.Z. Xu, C.C. Wang, W.L. Yang, YH. Deng, and S.K. Fu, Journal of Magnetism and Magnetic Materials, Vol. 277, p. 136,2004. 36. D.J. Voorn, W. Ming, and A.M. Van Herk, Macromolecules, Vol. 39, p. 2137, 2006.

248

M I N I E M U L S I O N POLYMERIZATION T E C H N O L O G Y

37. S. Cauvin, P.J. Colver, and S.A.F. Bon, Macromolecules, Vol. 38, p. 7887, 2005. 38. D. Tian, and H.-Q. Xie, Journal of Applied Polymer Science, Vol. 108, p. 3122, 2008. 39. M.T. Carver, E. Hirsch, J.C. Wittmann, R.M. Fitch, and F. Candau, Journal of Physical Chemistry, Vol. 93, p. 4867,1989. 40. C. Wu, and S. Zhou, Macromolecules, Vol. 28, p. 8381,1995. 41. Q. Duan, Y. Miura, A. Narumi, X. Shen, S.-I. Sato, T. Satoh, and T. Kakuchi, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 44, p. 1117,2006. 42. Y.K. Gong, L. Luo, A. Petit, D.J. Zukor, O.L. Huk, J. Antoniou, FM. Winnik, and F. Mwale, Journal of Biomédical Materials Research A, Vol. 72, p. 1,2005. 43. H. Kanazawa, K. Yamamoto, Y. Kashiwase, Y. Matsushima, N. Takai, A. Kikuchi, Y. Sakurai, and T. Okano, Journal of Pharmaceutical and Biomédical Analysis, Vol. 15, p. 1545,1997. 44. E. Uguzdogan, E.B. Denkbas, and A. Tuncel, Macromolecular Bioscience, Vol. 2, p. 214,2002. 45. Q. Sun, and Y. Deng, Journal of the American Chemical Society, Vol. 127, p. 8274, 2005. 46. S. Lu, and J. Forcada, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 44, p. 4187,2006. 47. G. Liu, H. Yang, and J. Zhou, Biomacromolecules, Vol. 6, p. 1280,2005. 48. D. Müller-Schulte, and T. Schmitz-Rode, Journal of Magnetism and Magnetic Materials, Vol. 302, p. 267, 2006. 49. H. Macková, D. Králová, and D. Horák, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 45, p. 5884,2007. 50. Y.-D. Luo, C.-A. Dai, and W.-Y Chiu, Journal of Colloid and Interface Science, Vol. 330, p. 170,2009. 51. C.L. Lin, W.Y. Chiu, and T.M. Don, Journal of Applied Polymer Science, Vol. 100, p. 3987, 2006. 52. Y.-D. Luo, C.-A. Dai, and W.-Y. Chiu, Journal of Polymer Science: Part A: Polymer Chemistry, Vol. 46, p. 8081, 2008. 53. K.L. Wooley, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 38, p. 1397,2000. 54. H.A. Clark, M. Hoyer, M.A. Philbert, and R. Kopelman, Analytical Chemistry, Vol. 71, p. 4831,1999. 55. K.B. Thurmond, T. Kowalewski, and K.L. Wooley, Journal of the American Chemical Society, Vol.118, p. 7239,1996. 56. Q. Zhang, E.E. Remsen, and K.L. Wooley, Journal of the American Chemical Society, Vol. 122, p. 3642,2000. 57. S. Yang, H. Liu, and Z. Zhang, Journal of Polymer Science: Part A: Polymer Chemistry, Vol. 46, p. 3900, 2008. 58. S. Yang, and H.R. Liu, Journal of Materials Chemistry, Vol. 16, p. 4480,2006. 59. Y. Chen, Z. Qian, and Z. Zhang, Colloids and Surfaces A: Physicochemical and Engineering Aspects, Vol. 312, p. 209,2008. 60. D. Sarkar, J. El-Khoury, S.T. Lopina, and J. Hu, Macromolecules, Vol.38, p. 8603, 2005. 61. C A . Uraneck, and R.J. Sonnenfeld, US 2697700,19541221,1954. 62. P.-Y. Jiang, Z.-C. Zhang, and M.-W. Zhang, Journal of Polymer Science: Part A: Polymer Chemistry, Vol. 34, p. 695,1996.

INVERSE MINIEMULSION POLYMERIZATION

249

63. J.M. Asua, Progress in Polymer Science, Vol. 27,1283, 2002. 64. B. Sumerlin, A. Lowe, D. Thomas, and C. McCormick, Macromolecules, Vol. 36, p. 5982, 2003. 65. M. Kamigaito, T. Ando, and M. Sawamoto, Chemical Reviews, Vol. 101, p. 3689,2001. 66. J.K. Oh, C. Tang, H. Gao, N.V. Tsarevsky, and K. Matyjaszewski, Journal of the American Chemical Society, Vol. 128, p. 5578,2006. 67. D.B. Thomas, A.J. Convertine, L.J. Myrick, C.W. Scales, A.E. Smith, A.B. Lowe, Y.A. Vasilieva, N. Ayres, and C.L. McCormick, Macromolecules, Vol. 37, p. 8941, 2004. 68. L. Albertin, M.H. Stenzel, C. Barner-Kowollik, and T.P. Davis, Polymer, Vol. 47, p. 1011,2006. 69. W. Jakubowski, and K. Matyjaszewski, Macromolecules, Vol. 38, p. 4139, 2005. 70. Y. Li, and S.P. Armes, Macromolecules, Vol. 38, p. 8155, 2005. 71. F. Isaure, P.A.G. Cormack, S. Graham, D.C. Sherrington, S.P. Armes, and V. Buetuen, Chemical Communications, p. 1138,2004. 72. J.K. Oh, F. Perineau, and K. Matyjaszewski, Macromolecules, Vol. 39, p. 8003, 2006. 73. J.K. Oh, H. Dong, R. Zhang, K. Matyjaszewski, and H. Schlaad, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 45, p. 4764,2007. 74. G. Qi, C.W. Jones, and F.J. Schork, Macromolecular Rapid Communications, Vol. 28, p. 1010, 2007. 75. D.B. Thomas, B.S. Sumerlin, A.B. Lowe, and C.L. McCormick, Macromolecules, Vol. 36, p. 1436, 2003. 76. C.L. McCormick, J. Bock, and D.N. Schulz, in H.F. Mark, N.M. Bikales, C.G. Overberger, G. Menges, eds., Encyclopedia of Polymer Science and Engineering, 2nd ed., New York, Wiley, Vol. 17, p. 730,1989. 77. J.J. Effing, I.J. McLennan, and J.C.T. Kwak, Journal of Physical Chemistry, Vol. 98, p. 2499,1994. 78. F. Candau, in K.L. Mittal, P. Kumar, eds., Handbook of Microemulsion Science and Technology, New York, Marcel Dekker, Chapter 22, p. 679,1999. 79. M. Pabon, J.-M. Corpart, J. Selb, and F. Candau, Journal of Applied Polymer Science, Vol. 84, p. 1418,2002. 80. A. Durand, E. Marie, E. Rotureau, M. Léonard, and E. Dellacherie, Langmuir, Vol. 20, p. 6956,2004. 81. E. Rotureau, J. Raynaud, B. Choquenet, E. Marie, C. Nouvel, J.-L. Six, E. Dellacherie, and A. Durand, Colloids and Surfaces A: Physicochemical and Engineering Aspects, Vol. 331, p. 84,2008. 82. E. Nordmeier, Journal of Physical Chemistry, Vol. 97, p. 5770,1993. 83. A. Durand, and E. Dellacherie, Colloid and Polymer Science, Vol. 284, p. 536,2006. 84. M. Pabon, J.-M. Corpart, J. Selb, and F. Candau, Journal of Applied Polymer Science, Vol. 91, p. 916,2004.

This Page Intentionally Left Blank

9 Double Miniemulsion Preparation for Hybrid Latexes R.Y. Hong 12 , G. Liu1, B. Feng1, and H.Z. Li2 College of Chemistry, Chemical Engineering & Materials Science, and Key Laboratory of Organic Synthesis offiangsu Province, Soochow University, SIP, Suzhou 215123, China 2 State Key Laboratory of Multi-phase Complex Systems, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100080, China

Abstract

There is no doubt that the properties of hybrid organic-organic latexes show significant improvements as compared with the properties of simple blends of polymer latexes. The improvements have been observed for properties such as the solvent resistance, the tensile strength, the hardness, the gloss and the adhesive strength on the metals. In other cases, the weatherability has been the object of such improvement. It appears that the main difference between simple blends and true hybrid organic-organic polymer colloids relies on a better homogeneity of the two polymeric materials inside a latex particle. Another drawback in the production of true hybrid materials comes from the required synthetic methods. Up to now, mini-emulsion polymerization has been restricted mostly to the laboratory, due to the difficulty to obtain mini-emulsification of polymer solutions in suitable monomers. It is clear that the usual method of ultrasonication cannot be applied on an industrial scale. However, there is some hope that efficient methods of emulsification may be developed, such as the use of Manton-Gaulin machines or the so-called static mixer, in which the emulsification can be obtain upon pressure forcing the solution in suitable device. Keywords: hybrid, homogeneity, emulsion, seed, sonication, static mixer, adhesive strength, weatherability.

V. Mittal (ed.) Miniemulsion Polymerization Technology, (251-276) © Scrivener Publishing LLC

251

252

9.1

MINIEMULSION POLYMERIZATION TECHNOLOGY

Introduction

There is an obvious interest to combine the properties of the polymer and inorganic nanomaterials in one material. It is expected that the preparation of such hybrid materials should lead to improvements versus blends of two materials. The term hybrid polymer is usually applied to polymer compounds containing an inorganic material covalently bonded to the polymer. Recently, efforts have been made to develop these materials as nanocomposites and, of course, hybrid composite latexes are a big part of this effort. Recent reviews have been devoted to this topic [1,2]. Another class of hybrid polymers is composed of at least two kinds of organic polymers, normally incompatible, and again covalently bonded to each other. This review is limited to the second class of hybrid materials, specifically as dispersed latexes of small particle size. The main interest of these materials is to intimately combine the properties of the different kinds of polymers. The main applications of these materials lie in the so-called waterborne coatings, such as paints, paper coating products, adhesives, textile sizing, and printing inks. The simplest way to prepare such hybrid latexes is the blend of two different polymer latexes. However, many polymers cannot be prepared directly as polymer latexes; in the domain of waterborne coatings, the acrylic (or styrene-acrylic) latexes, which are the more advanced class of coatings, can be prepared directly through a radical emulsion copolymerization processes. Important classes of products such as alkyd resins or polyurethane (PU) are normally obtained through polycondensation or polyaddition and must be put in the solution before emulsification. The main drawback of these processes is the possible limitation of their solubility in volatile solvents, and the fact that after emulsification of these solutions, the solvents must be evaporated. This is not a particularly environmentally friendly process when compared with waterborne coatings. For the most part, such hybrids are simple blends of acrylic and polycondensate latexes, as mentioned in a recent review [3]. True hybrid organic-organic products were reported in the open literature only recently, i.e., in the late 1990s. Two main routes are known for their synthesis. The first is the mini-emulsion polymerization [4] in which a polycondensate, such as polyester, PU, alkyd resin, or epoxy resin, is dissolved in radically polymerizable monomer(s). In the second route, the polycondensate, suitably modified, directly

DOUBLE MINIEMULSION FOR HYBRID LATEXES

253

forms latex in the presence of water, and this latex is used as a seed in an emulsion polymerization of suitable monomers. It is interesting to note that it is possible now to combine these two routes as some polycondensates can now be prepared via mini-emulsion and then can be used as seeds for an emulsion polymerization.

9.2

Hybrids via Mini-Emulsion Polymerization

The mechanisms of conventional emulsion and mini-emulsion polymerizations are in some ways significantly different. A conventional unseeded (i.e., no small particles added at the beginning) batch emulsion polymerization reaction may be divided into three intervals. Particle nucleation occurs during Interval I and is usually completed at low monomer conversion (2-10%) when most of the monomer is located in relatively large (1-10 mm) droplets. Particle nucleation takes place when radicals formed in the aqueous phase grow via propagation and then enter into micelles or become large enough in the continuous phase to precipitate and form primary particles which may undergo limited flocculation until a stable particle population is obtained. Significant nucleation of particles from monomer droplets is discounted because of the small total surface area of the large droplets. Interval II involves polymerization within the monomer-swollen polymer particles with the monomer supplied by diffusion from the droplets. Interval III begins when the droplets disappear or at least reach a polymer fraction similar to that of the particles and continues to the end of the reaction. Mini-emulsion polymerization [4] involves the use of an effective surfactant/co-stabilizer system to produce very small (0.01-0.5 mm) monomer droplets. The droplet surface area in these systems is very large, and most of the surfactant is adsorbed at the droplet surfaces. Particle nucleation is primarily via radical (primary or oligomeric) entry into monomer droplets, since little surfactant is present in the form of micelles, or as free surfactant available to stabilize particles formed in the continuous phase. The reaction then proceeds by polymerization of the monomer in these small droplets; hence, there is no true Interval II. Mini-emulsions are produced by the combination of a high shear to break up the emulsion into submicron monomer droplets, and

254

MINIEMULSION POLYMERIZATION TECHNOLOGY

a surfactant/costabilizer system to retard monomer diffusion from the submicron monomer droplets. Both are necessary to affect predominant droplet nucleation (nucleation in which a preponderance of the particles originate from droplets rather than from micelles or from homogeneous nucleation). High shear is provided by a sonicator or a mechanical homogenizer. The surfactant is necessary to retard droplet coalescence caused by Brownian motion, settling or Stokes law creaming of settling. The costabilizer (also referred to in earlier works as a cosurfactant or hydrophobe) prevents Ostwald ripening [5]. When a liquid emulsion is subjected to high shear, small droplets will result. There will still be a statistical distribution of droplet sizes. If the monomer is even slightly soluble in the continuous aqueous phase (and most are as evidenced by the fact that Interval II of macroemulsion polymerization takes place), monomer will, over time, diffuse from the smaller monomer droplets into the larger ones. This results in a lower interfacial area (and interfacial energy), since the loss of interfacial area of the smaller droplets is larger than the gain in interfacial area of the larger ones. The reduction in interfacial energy is the driving force for degradation of the small droplets. If Ostwald ripening is allowed to continue unchecked, creaming of the monomer will occur as the droplet sizes become large enough for Stokes law creaming to occur. A co-stabilizer function is to prevent Ostwald ripening by retarding monomer diffusion from the smaller droplets to the larger. Co-stabilizers should be highly insoluble in the aqueous phase(so that they will not diffuse out of the droplets) and highly soluble in the monomer droplets. Under these conditions, diffusion of monomer out of the smaller droplets results in an increase in the concentration of the co-stabilizer in those particles (since, by definition co-stabilizers are too insoluble in the aqueous phase to leave the droplet). The increase in free energy associated with the concentration of the costabilizer balances the decease due to reduced interfacial area cause by Ostwald ripening, and, at some point, ripening stops. In the original discovery of mini-emulsion polymerization, Ugelstad and co-workers [6] used either cetyl alcohol (CA: water solubility (vol/vol) estimated at 6 x 108 [7]) or hexadecane (HD: water solubility estimated at 1 x 109 [7]) to retard monomer diffusion from submicron monomer droplets. Both CA and HD are volatile organic components and therefore not entirely desirable in the final product. Other researchers used polymer, chain transfer agents and comonomers as

DOUBLE MINIEMULSION FOR HYBRID LATEXES

255

stabilizers. Because transport of material across the aqueous phase is minimized in mini-emulsion polymerization, such systems are natural for creating hybrid polymers by free radical polymerization of one monomer in the presence of another polymer. In addition, other chemists engaged in polymerization can take advantage of the colloidal stability of a miniemulsion [8]. In this technology, a preformed polymer (or oligomer) is dissolved into a monomer (or monomer solution). A mini-emulsion is then created from the monomer-polymer solution, and this mini-emulsion is polymerized via standard techniques. Care must be taken in creating the miniemulsion, because the polymer solution will likely have a high viscosity (higher than that of a simple monomer mini-emulsion), and droplet breakup by the shear device may be more difficult. On the other hand, the preformed polymer may act as the co-stabilizer, eliminating the need for HD or other co-stabilizers. For some chemistries, grafting will take place during the polymerization of the monomer, and for some polymers phase separation will occur. In any case, the product is a submicron dispersion of one polymer in another and may well have a practical value.

9.3

Double-Miniemulsion Formation

Ultrasonic emulsification is also very efficient in reducing droplet size but it is only appropriate for small batches [9,10]. A recent study [11] on the preparation of polymerizable nano-emulsions has shown that the efficiency of the dispersion process is strongly dependent on the ultrasonication time at different amplitudes and that the more hydrophobic the monomer is, the longer the sonication time required. The principle of the encapsulation of magnetite in polymer particles by the miniemulsion process is shown in Figure 9.1 [12].

9.4

Stability

The small droplet size of nano-emulsions confers stability against sedimentation (or creaming) because the Brownian motion and consequently the diffusion rate are higher than the sedimentation (or creaming) rate induced by the gravity force. Ostwald ripening or molecular diffusion, which arises from emulsion polydispersity and the difference in solubility between small and large droplets,

256

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Figure 9.1. The principle of the encapsulation of magnetite in polymer particles by the miniemulsion process. Reproduced from reference [12] with permission from IOP publishing.

is the main mechanism for nano-emulsion destabilization [13]. The Lifshitz-Slezov [14] and Wagner [15] (LSW) theory predicts a linear relationship between the cube of the radius, r3, and time, t, with the slope being the Ostwald ripening rate. The LSW theory assumes that the droplets of the dispersed phase are spherical, the distance between them is higher than the droplet diameter and the kinetics is controlled by molecular diffusion of the dispersed phase in the continuous phase. According to this theory, the Ostwald ripening rate in o / w emulsions is directly proportional to the solubility of the oil in the aqueous phase. In fact, Taylor [16] has suggested that Ostwald ripening might be used as a tool to estimate the thermodynamics of solution of oils in water. Izquierdo et al. [17] found that the Ostwald ripening rate obtained experimentally was higher than calculated theoretically according to the LSW theory. The discrepancy was attributed to factors not taken account of in this theory such as oil transport due to the presence of micelles and /or microemulsion droplets in the aqueous phase, increase in droplet Brownian motion and lowering of the interfacial Gibbs elasticity. Experimental Ostwald ripening rate about three orders of magnitude higher than the calculated have been also reported in w / o nano-emulsions stabilized with fluorinated surfactants by Courrier et al. [18].

DOUBLE MINIEMULSION FOR HYBRID LATEXES

257

The reduction of the Ostwald ripening process by the addition to the system of a small amount of a second oil with low solubility in the aqueous phase[19], is well known [13]. It has been recently reported that, for an ethoxylated nonionic surfactant system, the Ostwald ripening rate can also be decreased by adding a second surfactant with the same alkyl chain length and higher degree of ethoxylation than the primary surfactant [20].

9.5

Characterization

DSD is by far the most important characteristic of a monomer miniemulsion because if affects directly both miniemulsion stability and droplet nucleation. Therefore, the understanding of the mechanisms ruling miniemulsion polymerization strongly depends on an accurate determination of the DSD. Azad et al. [21] stained/hardened styrene miniemulsion droplets with Os0 4 . Samples were withdraw after 5-10 min and placed on a formvar, carbon coated and allow to dry. The grids were examined by transmission electron microscopy (TEM). Inferior TEM pictures were obtained when the mole ratio of Os0 4 /styrene was less than 1/1 or higher than 2/1 as well as when the reaction time was longer than 30 min. The spherical shape of the stained particles was assessed by shadowing the stained droplets with a 80/20 Pt/Pd alloy at an angle of 30°. This method was unsuccessfully tried by Choi et al. [22] who pointed out that the monomer droplet sizes found by Azad et al. [21] were in the 0.4-1.5 pm range, which is much larger than that expected in the miniemulsion system. Choi et al. [22] used a freezefracturing technique to obtain information on the DSD in the miniemulsion. In this technique, a small emulsion droplet was frozen by immersion in liquid nitrogen, fractured with a cold knife, shadowed and replicated with platinum/carbon. The sample was then placed in water to dissolve away the emulsion and the remaining carbon film was examined by TEM (Figure 9.2). In this method, the miniemulsion does not need to be diluted, but the path of fracture does not necessarily cross the center of the droplets, and therefore the sizes observed in the micrograghs are less or equal to the actual diameters. Van Hamersveld et al. [23] used cryo-TEM (Figure 9.3) to study the morphology of oil-acrylate miniemulsions. A thin aqueous film was thermally fixated by vitrification in liquid ethane and imaged at -170°C. A quantitative analysis of the average droplet

258

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Figure 9.2. The fracture plane passed around the droplets.

size was not made because of the possible occurrence of size segregation in the sample. Figures 9.4 and 9.5 respectively shows the SEM and TEM devices generally used for the characterization. Ugelstad et al. [24] estimated the total surface of the monomer droplets, which is related to the droplet size and a good indication of the relative size of different systems, by measuring the amount of surfactant adsorbed on the droplets. Erdem et al. [25] extended the soap titration method [26] to the measurement of the average droplet size in miniemulsion. In this method, the miniemulsion is titrated with the surfactant to detect at what concentration plus the saturation adsorption area of the surfactant, the total surface area of the droplets (At) can be estimated. Then the volumer-surface

D O U B L E M I N I E M U L S I O N FOR H Y B R I D LATEXES

259

Figure 9.3. Cryo-TEM image of monomer droplets. Reproduced from reference [23] with permission from Elsevier.

Figure 9.4. Hitachi SEM S-4700

average droplet diameters is calculated by means of the following equation: d dvs =6V t /A t

(1)

where Vt is the total volume of the monomer droplets. The results obtained with this method were consistent with those obtained through capillary hydrodynamic fractionation (CHDF) [27].

260

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Figure 9.5 TEM TecnaiG220.

Miller et al. [27] used CHDF to measure the DSD of monomer miniemulsions. In this technique, the diluted miniemulsion is carried through an open capillary by a carrier fluid. A parabolic velocity profile develops in the capillary. Small droplets are able to flow closer to the capillary wall than large droplets, and hence they are carried by slower streamlines and will elute later than the large droplets. The interpretation of the fractograms is based on a calibration obtained my measuring the elution times of polystyrene latex particle standards, and it is open to discussion whether calibrations made using hard polymer particles are directly applicable to soft monomer droplets. In principle CHDF provides the DSD, but the fractograms showed by Miller et al. [27] presented small droplet size tails that might be artifacts. Nevertheless, the technique yields values of the droplet size that are in good agreement with the sizes of the polymer particles obtained after polymerization of these miniemulsions. Light scattering has been frequently used to determine the droplets size of monomer [28]. It is very easy to implement by this technique and the measuring is very rapid, but by no means trouble-free. In this technique, vast dilution is needed to avoid multiple scattering. Dilution might affect the DSD, e.g. by desorption of surfactant and by dissolution of the monomer in the aqueous phase. In order to minimize this effect, the dilution is made using the aqueous phase of the same preparation, light scattering is not well suited to determine the size distribution of polydisperse systems [29] because the inversion of the autocorrelation function is illconditioned. Small differences in the autocorrelation function may

DOUBLE MINIEMULSION FOR HYBRID LATEXES

261

result in quite different size distributions after inversion namely, quite different types of size distributions correspond to almost the same autocorrelation function. Landfester et al. [30] used small-angle neutron scattering (SANS) in an attempt to determine the droplet size of styrene miniemulsion. In addition to the fact that this is not a technique readily available, SANS shares with light scattering the same drawback in data analysis, namely, that the raw data are fitted to a given model. Thus, Landfester et al. [30] assumed that the DSD was well represented by a SchulzZimm distribution, but there is no guarantee that this would be the case. For example, the DSD determined using CHDF by Miller et al. [27] cannot be well represented by a Schulz-Zimm distribution. In conclusion, there is no direct method that guarantees the accurate determination of the DSD. This handicaps the study of the mechanisms involved in miniemulsion polymerization. The droplet size of the miniemulsion is often inferred from the PSD of the latex obtained after polymerization of the monomer miniemulsion. However, this may lead to substantial errors.

9.6

Applications

In recent years, magnetic latexes have attracted much attention. The magnetic latexes have the advantage of rapid and easy separation of particles upon the application of an external magnetic field [31 ] and demonstrate potential applications in biomédical and diagnostics fields, including cellular therapy in cell labeling [32], drug delivery [33], cell separation [34], biosensors [35], immobilization of biomolecules such as oligonucleotides [36], proteins [37] and antibodies [38]. In order to successfully apply magnetic latexes to biomédical fields, it is necessary to obtain magnetic polymer particles with properties of no sedimentation [39], near-nanosized distribution [40], high and uniform superparamagnetic content [41], no iron leaking and non-toxicity [42]. The pioneering work in synthesis of magnetic latexes has been reported by Guesdon et al. [43] in 1977, who synthesized magnetic particles by the polymerization of acrylamide and agarose in the presence of iron oxide nanoparticles. Charmot et al. [44] investigated the emulsion polymerization of styrene in the presence of organic magnetic fluids. Several schemes for preparation of magnetic latexes have been reported based on the research of Charmot et al.

262

MINIEMULSION POLYMERIZATION TECHNOLOGY

The common route to synthesize magnetic polymer particles is monomer polymerization by dispersing surface modified magnetite particles directly into the liquid phase of monomer and then initiating polymerization of the monomer in the presence of the magnetic particles. However, the magnetic polymer microspheres obtained from the conventional monomer polymerization are often incompletely and non-uniformly encapsulated, leading to the nonuniform size of resultant particles. Due to limited polymerization methodology in the encapsulation process, new approaches have been explored and developed for the synthesis of nano-sized magnetic polymer spheres, including emulsion polymerization [45,46], miniemulsion polymerization [47,48], emulsifier-free emulsion polymerization [49], dispersion polymerization [50], suspension polymerization [51-53], and microemulsion polymerization [44]. Among these polymerization processes, miniemulsion polymerization [54,55] is considered as one of the novel polymerization methods. In the presence of co-emulsifier, miniemulsion polymerization takes place inside the stable monomer droplets as reaction place in which the average diameter of monomer droplets is about 50-500 nm. Recently, a novel double miniemulsion has been developed based on the miniemulsion, and has been used to prepare polymer microspheres or nanospheres encapsulating magnetic particles. The novel method of double emulsion polymerization process is to combine the two miniemulsions to synthesize magnetite-encapsulated polymer particles. In addition, by controlling the reaction conditions, it is easier to produce monodispersed, nanoscale and superparamagnetic polymer spheres by double miniemulsion than other methods. Gu et al. [56,57] had proposed a modified miniemulsion/emulsion polymerization to obtain magnetite-polystyrene (Fe304-PS) microspheres. The obtained microspheres have the particle diameter of several micrometers along with the narrow size distribution and high magnetite content. The objective of the present investigation is to apply this double miniemulsification process for the encapsulation of magnetic particles in the poly (methyl methacrylate) (PMMA) to obtain monodispersed magnetic polymer microspheres [58] (Figure 9.6). First, oleic acid coated magnetite particles were synthesized by co-precipitation and dispersed into octane to obtain a magnetic fluid. Then the magnetic fluid was miniemulsified into water using sodium dodecylsulfate (SDS) as emulsifier and hydrophobic reagent hexadecane as an osmotic agent. Furthermore, another prepared MMA miniemulsion

D O U B L E M I N I E M U L S I O N FOR H Y B R I D LATEXES

263

Figure 9.6. The reaction sketch for preparing magnetic PMMA microspheres through double miniemulsion polymerization [58].

was added into the former emulsion dropwise to carry on polymerization at 80°C. The brown magnetic emulsion was prepared. The morphology and magnetic properties of the prepared PMMA microspheres were characterized. The effects of initiator dosage and MMA concentration on PMMA conversion were investigated. Stable (oil-in-water) magnetic emulsion was prepared by polymerization involving two miniemulsions (A and B). A typical experiment was described as follows [58]: Miniemulsion (A) was prepared using 5 ml of magnetic fluid (oleic acid coated Fe 3 0 4 nanoparticles dispersed in octane with 10% (w/v) magnetite content) and 0.35 g of hydrophobic reagent hexadecane mixed in surfactant solution consisting of sodium dodecyl sulfate (SDS: 0.35 g) and distilled water (25 g). The above mixture was stirred for 0.5 h at 1300 rpm, and then adding AIBN followed by ultrasonic irradiation for 10 min in an ice-cooled bath. Miniemulsion (B) consisting of certain amount of MMA, 0.12 g of hexadecane, 0.12 g of SDS, 0.15 g of Tween-80 and 20 g of distilled water was prepared by magnetic stirring at ambient temperature for 1 h followed by ultrasonic irradiation for 15 min. Then Miniemulsion (A) was transferred to a 250 mL, threenecked glass reactor equipped with an Ar inlet and an ordinary polytetrofluoroethyl agitator was used throughout the experiments. Miniemulsion (B) was added in portions, with intense stirring, to the Miniemulsion (A). Emulsion polymerization was carried out at 80°C with a thermal bath for 90 min. Finally, the brown magnetic emulsion was obtained. The PMMA/Fe 3 0 4 magnetic microspheres can be separated from the emulsion under an external magnetic field, and can be redispersed into the emulsion with agitation.

264

MINIEMULSION POLYMERIZATION TECHNOLOGY

In the present investigation, magnetic PMMA microspheres were obtained by double-miniemulsion polymerization and magnetic separation. The surface morphology and particle size of these microspheres were studied by TEM and SEM, as shown in Figure 9.7. It can be seen from the TEM images that magnetic PMMA microspheres demonstrate excellent monodispersity and all the composite particles are spherical in shape. The average diameters of the magnetic PMMA microspheres of Samples 1, 2 and 3 are 100,120 and 150 nm, respectively. Comparing the samples of different weight ratios of MMA to OMP (Table 9.1), it is obvious that the diameter of the microspheres increased slightly with the increasing weight ratio of MMA to OMP. Therefore, the weight ratio of MMA to OMP was an important factor in controlling the particle size of magnetic PMMA microspheres. In addition, as observed in Figure 9.7a (Sample 1), the distribution of the magnetite nanoparticles encapsulated inside the PMMA microspheres is comparatively uniform, however, there is some OMP leaking outside the polymer spheres, while the morphology of magnetic microspheres (Samples 2 and 3) become uniform and magnetite particles seem to disperse homogeneously in the polymer spheres with the increasing weight ratio of MMA to OMP. In addition, the morphology of magnetic PMMA microspheres was studied by SEM. The SEM images of Samples 1 and 3 were shown in Figure 9.8a and Figure 9.8b. From Figure 9.8a, it is apparent that OMP congregated inside microspheres, and the content of Fe 3 0 4 nanoparticles in each microsphere was not consistent. Thus, the morphology of Sample 1 was not uniform and there existed flaw apparently which indicated the incomplete encapsulation of OMP. This phenomenon may be explained that the low weight ratio of MMA to OMP leads to more magnetic particles existed in the reaction system. On the other hand,

(A)

(B)

(C)

Figure 9.7. TEM of magnetic PMMA microspheres (a) sample 1, (b) sample 2 and (c) sample 3 [58].

D O U B L E M I N I E M U L S I O N FOR H Y B R I D LATEXES

265

Table 9.1. Synthesis magnetic PMMA microspheres. AIBN (mg)

Magnetic fluid (ml)

Monomer: OMP (weight ratio)

600

6

5

6

2

900

9

5

9

3

1200

12

5

12

4

600

12

5

6

5

600

24

5

6

Sample No.

Methyl methacrylate (mg)

1

All experiments were carried out at 80°C for 90min [58].

(A)

(B)

Figure 9.8. SEM of magnetic PMMA microspheres (a) sample 1; (b) sample 3 [58].

the encapsulation ability of microsphere is limited, so excessive magnetic nanoparticles could not be well encapsulated by PMMA microsphere. In Figure 9.8b, while weight ratio of MMA to OMP is high, it is found that the morphology of microspheres is spherical and the OMP were dispersed homogeneously in the polymer microspheres, which reasonably matched with TEM observation (Figure 9.7c). The magnetic properties of oleic-acid-coated magnetite particles and magnetic PMMA microspheres were characterized by a vibrating sample magnetometer (VSM). Figure 9.9 shows the typical room-temperature magnetization curves of bare Fe304, OMP and magnetic PMMA microspheres of Sample 3, respectively. As shown in this figure, the saturation magnetization of bare Fe304 nanoparticles is 65.6 emu/g and oleic acid coated magnetite particles is 47.7 emu/g. For the magnetic PMMA microspheres, the saturation magnetization is 23.7 emu/g. All the samples demonstrate a

266

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Figure 9.9 Magnetization curves obtained by VSM at room temperature of (a) Fe 3 0 4 nanoparticles; (b) oleic-acid-coated magnetite and (c) magnetic PMMA microspheres of sample 3 [58].

typical superparamagnetic behavior. However, the saturation magnetization of the nanoparticles was significantly smaller than that of bulk magnetite, which is 84 emu/g [59]. The lower saturation magnetization can be ascribed to the rather small size of magnetite nanoparticles and the mass of the added oleic acid on the particles [60]. In addition, low saturation magnetization of magnetic PMMA microspheres may be attributed large part to the oxidation during the polymerization, which leads to the formation of some non-magnetic iron oxide (Fe 2 0 3 ). 9.6.1

Effects of R e a c t i o n C o n d i t i o n s

The effect of initiator dosage and MMA concentration on PMMA conversion were investigated respectively as follows. 9.6.1.1

Initiator Dosage

The effect of the initiator dosage on PMMA conversion with fixed amount of MMA was shown in Figure 9.10. Comparing the plots of different initiator dosages, the reaction rate is in the following order: 24 mg AIBN > 12 mg AIBN > 6 mg AIBN. It is apparent

D O U B L E M I N I E M U L S I O N FOR H Y B R I D LATEXES

267

Figure 9.10. Conversion versus time of emulsion polymerization at different initiator dosages [58].

that the polymerization rate was higher under the condition of a higher initiator concentration. Higher amount of the initiator could increase the polymerization of MMA in the emulsion. However, when the concentration of initiator exceeded 50 mg, the excessive amount of the initiator could initiate the polymerization of MMA in the solution which leads to the production of nonmagnetic PMMA. In addition, the reason for the conversion of less than 100% was due to the glassy effect [61]. When the glass-transition temperature of the polymer particles (T of magnetic PMMA particles of Sample 3 is 108.16°C, as shown in the supplementary material) was higher than the reaction temperature (80 °C), the monomer propagation was difficult to continue in the polymer particles; thus, the final conversion could not reach 100%. 9.6.2.2

MMA Monomer

Concentration

The effect of MMA monomer concentration on MMA conversion was investigated, and the result was shown in Figure 9.10. Comparing the plots of different MMA monomer concentrations, the reaction rate is in the following order: 6g MMA > 9g MMA > 12g MMA. As Wang [62] reported, with a lower concentration of MMA in the polymerization system, the magnetic fluid acted as the seeds in the polymerization and the reaction rate was very fast. When the concentration of

268

MINIEMULSION POLYMERIZATION TECHNOLOGY

MMA increased, the reaction rate slowed down. It can be explained that most of the layer structure of the surfactant of the magnetic fluid was destroyed by the excess monomer and then the polymerization was mostly initiated by self-nucleation [63].

9.6.2 9.6.2.2

Rheological Properties of Magnetic Emulsion Viscosity Versus Time

Figure 9.11 shows the viscosities of magnetic emulsions of different MMA concentrations at a fixed shear rate (140 s_1). We can see from the figure that the viscosity of magnetic emulsions do not change with the shearing time at the constant shear rate and the viscosity basically keeps constant which indicated the magnetic emulsions were stable. When the magnetic emulsions were moved with the rotor, the PMMA coated solid particles also rotated with the rotor, leading to the destruction of magnetic latex structures at steady state. Because of the solid content of magnetic emulsions is below 20% and the concentration of magnetic PMMA particles was low which will not cause the entanglement of polymer chains. This can be explained why the viscosity of magnetic emulsion do not changed with the shear time.

Figure 9.11. Time evolution of magnetic emulsion viscosity with different samples: (a) sample 1; (b) sample 2; (c) sample 3 [58].

DOUBLE MINIEMULSION FOR HYBRID LATEXES

9.6.2.2

269

Viscosity with/without Magnetic Field

In this section, the magnetic field with the intensity of H = 17 mT was applied to the magnetic emulsions between the spindle and cup of the rotating rheometer. The intensity of magnetic field was measured using a Hall-effect sensor. The viscosities of magnetic emulsions with and without magnetic field were illustrated in Figure 9.12. By comparing the curves, it is found that the applied magnetic field had obvious effects the viscosity of magnetic emulsion. Since the viscosity of water cannot response to the applied magnetic field, the viscosity of the magnetic emulsion is determined by the properties and contents of the magnetic PMMA particles. Under magnetic field, the magnetic PMMA particles were polarized and arranged their orientation along the direction of the magnetic field, also the interaction among the magnetic PMMA particles was enhanced. Therefore, the flow resistance increased. Finally, it gave higher viscosity of the emulsion. In addition, it is observed that there is an abrupt change of viscosity at the shear rate of 125 and 90 s"1 for magnetic emulsions with and without magnetic field, respectively. This phenomenon may be due to the destruction of microstuctures of magnetic emulsion at high hear stress [49]. With the increasing shear rate, these microstructures of magnetic emulsions would be destroyed gradually under the shear stress which may cause an abrupt change

Figure 9.12. Viscosity vs. shear rate of magnetic emulsion (sample 3) with and without magnetic field [58].

270

MINIEMULSION POLYMERIZATION TECHNOLOGY

of viscosity when shear stress reaches some extent. However, the applied magnetic field can also rearrange the magnetic nanoparticles, leading to the formation of orderly microstructures [50]. So the viscosity of magnetic emulsion under magnetic field had an abrupt change at higher shear rate than that without magnetic field. 9.6.2.3

Applications of Magnetic Polymer Microspheres

Magnetic polymer nano- and micro-spheres are an area of great interest because the combination of easily modifiable polymer support and its responsiveness to a magnetic field greatly enhances possibilities of its easy manipulation, separation, and targeting. Recently developed magnetic nanoparticles can be used in data storage applications (if they are not affected by temperature fluctuations), catalysis, and biomédical applications [58,64]. Xu et al. [65,66] utilized superparamagnetic polymer beads as building blocks to fabricate photonic crystals configurable with a magnetic field. They applied a strong magnetic field to the beads, which induced strong interparticle interactions and eventually led to the formation of a 3D crystal with variable lattice constants. Helseth and co-workers [67,68] investigated crystallization of superparamagnetic beads under a magnetic field generated by ID nanomagnets. The magnetic-potential well was created using a bismuth-substituted ferrite garnet film. The large, in-plane magnetized domains of the garnet film were separated by a domain wall. The domain wall served as a nanomagnet, creating a potential well with a net magnetization moment in the direction perpendicular to the film surface and attracting the microbeads. Under a slow flow of the suspension, the beads started to assemble on one side of the potential well and finally made a well-ordered monolayer of spherical beads. When the beads were crystallized between two domain walls, a hexagonal structure between the walls or a chain connecting the two walls was observed. The same group also studied the effect of an external magnetic field on the crystal formation at the domain wall. Interestingly, when the external magnetic field was weak, chains of beads along the magnetic field were obtained as stable structures, whereas they fluctuated under a strong magnetic field. Doyle et al. [69] have capitalized on the chain formation of superparamagnetic beads to fabricate biochips for DNA separation. They confined a suspension of superparamagnetic beads in a thin gap between two glass slides and applied a magnetic field normal to the

DOUBLE MINIEMULSION FOR HYBRID LATEXES

271

layer to align the beads along the field. The assembled chains in the thin layer provided microchannels of which the dimensions were controlled by the size and concentration of the magnetic beads. Switching off the magnetic field destroyed the channels by turning the particles back into a disordered suspension.

9.7

Summary

There has been much less effort in the synthesis of hybrid organicorganic latexes and their characterization than for the corresponding hybrids of inorganic-organic materials. It seems that the reason for this situation is that industrial researchers consider the properties of such hybrid materials to be not that much different from the properties of blends of latexes, which are more easily produced, while the compatibility between inorganic sol-gel materials and polymer latexes is much more difficult to achieve. However, there is no doubt that the properties of hybrid organic-organic latexes show significant improvements as compared with the properties of simple blends of polymer latexes. The improvements have been observed for properties such as the solvent resistance, the tensile strength, the hardness, the gloss and the adhesive strength on the metals. In other cases, the weatherability has been the object of such improvement. It appears that the main difference between simple blends and true hybrid organic-organic polymer colloids relies on a better homogeneity of the two polymeric materials inside a latex particle. An example of this improved homogeneity is shown on Figure 9.7. Another drawback in the production of true hybrid materials comes from the required synthetic methods. Up to now, mini-emulsion polymerization has been restricted mostly to the laboratory, due to the difficulty to obtain mini-emulsification of polymer solutions in suitable monomers. It is clear that the usual method of ultrasonication cannot be applied on an industrial scale. However, there is some hope that efficient methods of emulsification may be developed, such as the use of Manton-Gaulin machines or the so-called static mixer, in which the emulsification can be obtain upon pressureforcing the solution in suitable device. The second method of seeded emulsion polymerization from a modified polycondensate latex is closer to the classical technology, although it needs a proper modification, which may be costly. The last point of interest may be the comparison of the properties of the

272

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

materials, which may be prepared by the two methods. It seems that better results can be obtained from the mini-emulsion route. The reason for that is probably due to the better homogeneity of the corresponding products, prepared in one step instead of two steps for the seeded emulsion route.

9.8

Acknowledgments

The project was supported by the National Natural Science Foundation of China (NSFC, Nos. 20876100 and 20736004), the National Basic Research Program of China (973 Program, No. 2009CB219904), the State Key Lab. of Multiphase Complex Systems of the Chinese Academy of Science (No. 2006-5), the Key Lab. of Organic Synthesis of Jiangsu Prov., R&D Foundation of Nanjing MedicalUniv. (NY0586), Post-doctoral Science Foundation of Jiangsu Prov., National Post-doctoral Science Foundation (20090451176) and the Commission of Science and Technology of Suzhou Municipality (YJS0917),China.

References 1. V. Castelvetro, and C D . Vita, Advances in Colloid and Interface Science, Vol. 108-109, p. 167,2004. 2. E. Bourgeat-Lami, 'Organic/inorganicnanocomposite colloids', in H.S. Nalwa, ed., Encyclopedia of Nanoscience and Nanotechnology, Los Angeles, American Scientific Publishers, pp. 305-32, 2004. 3. M. Jowkar-Deriss, and O. Karlsson, Colloids and Surfaces A, Vol. 245, p. 115, 2004. 4. J.M. Asua, Progress in Polymer Science, Vol. 27, p. 1283,2002. 5. W. Ostwald, Z. Phys. Chem., Vol. 34, p. 495,1900. 6. J. Ugelstad, M. Elaasser, and J.W. Vanderho, Journal of Polymer Science, Polymer Letters, Vol. 11, p. 503,1973. 7. C. Chern, and T. Chen, Colloids and Surfaces A, Vol. 138, p. 65,1998. 8. K. Landfester, Macromolecular Rapid Communications, Vol. 22, p. 896, 2001. 9. P. Walstra, 'Emulsion stability', in P. Becher, ed., Encyclopedia of Emulsion Technology, New York, Marcel Dekker, pp. 1-62,1996. 10. J. Floury, A. Desrumaux, M.A.V. Axelos, and J. Legrand, journal of Food Engineering, Vol. 58, p. 227,2003. 11. K. Landfester, J. Eisenblätter, and R. Rothe, JCT Research, Vol. 1, p. 65, 2004. 12. L. Katharina, and P.R. Liliana, Journal of Physics: Condensed Matter, Vol. 15, p. S1345,2003. 13. T.F. Tadros, P. Izquierdo, J. Esquena, and C. Solans, Advances in Colloid and Interface Science, Vol. 108-109, p. 303, 2004.

DOUBLE M I N I E M U L S I O N FOR HYBRID LATEXES

273

14. I.M. Lifshitz, and V.V. Slyozov, Journal of Physics and Chemistry of Solids, Vol. 19, p. 35,1961. 15. C. Wagner, Z Elektrochem, Vol. 65, p. 581,1961. 16. P. Taylor, Advances in Colloid and Interface Science, Vol. 106, p. 261, 2003. 17. P. Izquierdo, J. Esquena, T.F. Tadros, J.C. Dederen, J. Feng, and M.J. Garcia, et al., Langmuir, Vol. 20, p. 6594, 2004. 18. H.M. Courrier, T.F. Vandamme, and M.P. Krafft, Colloids and Surfaces A: Physicochemical and Engineering Aspects, Vol. 244, p. 141,2004. 19. P. Izquierdo, Studies on nano-emulsion formation and stability. Ph.D. thesis, Barcelona: University of Barcelona, 2004. 20. P. Izquierdo, J. Feng, J. Esquena, T.F. Tadros, J.C. Dederen, M.J. Garcia, N. Azemara and C. Solansa, Journal of Colloid and Interface Science, Vol. 285, p. 388, 2005. 21. A.R.M. Azad, J. Ugelstad, R.M. Fitch, F.K. Hansen, 'Emulsification and emulsion polymerization of styrene using mixtures of cationic surfactant and long chain fatty alcohols or alkanes as emulsifiers', in I. Piirma, J.L. Gordon, eds., Emulsion Polymerization, ACS Symposium Series, Vol. 24, Washington, DC, ACS, pp. 1-23,1976. 22. Y.T. Choi, M.S. EI-Aasser, E.D. Sudol, and J.W. Vanderhoff, Journal of Polymer Science, Polymer Chemistry Edition, Vol. 23, p. 2973,1985. 23. E.M.S. van Hamersveld, J.J.G.S. van Es, and F.P Cuperus, Colloids and Surfaces A,Vo\. 153, p. 285,1999. 24. J. Ugelstad, PC. Mork, K.H. Kaggerud, T. Ellingsen, and A. Berge, Advances in Colloid and Interface Science, Vol. 13, p. 101,1980. 25. B. Erdem, Y. Sully, E.D. Sudol, V.L. Dimonie, and M.S. EI-Aasser, Langmuir, Vol. 16, p. 4890, 2000 26. S.H. Maron, M.E. Elder, and I.N. Ulevich, Journal of Colloids Science, Vol. 9, p. 89,1954. 27. C M . Miller, J. Venkatesan, C.A. Silebi, E.D. Sudol, and M.S. EI-Aasser, Journal of Colloid and Interface Science, Vol. 162, p. 11,1994. 28. VS. Rodriguez, Interparticle monomer transport in miniemulsion copolymerization, PhD Dissertation, Lehigh University, 1988. 29. O. Elizalde, P. Leal, and J.R. Leiza, Particle Size Characterization, Vol. 17, p. 236,2000. 30. K. Landfester, N. Bechthold, S. Foster, and M. Antinietti, Macromolecular Rapid Communications, Vol. 20, p. 81,1999. 31. O. Olsvik, T. Popovic, E. Skjerve, K.S. Cudjoe, E. Homes, J. Ugelstad, and M. Uhlen, Clinical Microbiology Reviews, Vol. 7, p. 43,1994. 32. A. Ito, H. Honda, and T. Kobayashi, Cancer Immunology Immunotherapy, Vol. 55, p. 320,2005. 33. S. Goodwin, C. Peterson, C. Hoh, and C. Bittner, Journal of Magnetism and Magnetic Materials, Vol. 194, p. 132,1999. 34. M. P. Morales, O.B. Miguel, R.P. Alejo, J.R. Cabello, S.V. Verdaguer, and K. O'Grady, Journal of Magnetism and Magnetic Materials, Vol. 266, p. 102, 2003. 35. R. Arshady, Biomaterials, Vol.14, p. 5,1993. 36. R. Hiergeist, W. Andrä, N. Buske, R. Hergt, I. Hilger, U.Richter, and W. Kaiser, Journal of Magnetism and Magnetic Materials, Vol. 201, p. 420,1999. 37. A.P. Astalan, F. A h r e n t o r p , C. Johansson, K. Larsson, a n d A. Krozer, Biosensors and Bioelectronics, Vol. 19, p . 945,2004.

274

M I N I E M U L S I O N POLYMERIZATION T E C H N O L O G Y

38. F. Ganachaud, A. Elaissari, C. Pichot, A. Laayoun, and P. Cros, Langmuir, Vol. 13, p. 701,1997. 39. J. Revilla, A. Elaissari, P. Carrière, and C. Pichot, Journal of Colloid and Interface Science, Vol. 180, p. 405,1996. 40. T. Delair, C. Pichot, and B. Mandrand, Colloid and Polymer Science, Vol. 272, p. 72,1994. 41. P. Wang, C. Lee, and T. Young, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 43, p. 1342, 2005. 42. D.G. Shchukin, I.L. Radtchenko, and G.B. Sukhorukov, Materials Letters, Vol. 57, p. 1743,2003. 43. C. Wilhelm, C. Billotey, J. Roger, J.N. Pons, J.C. Bacri, and F. Gazeau, Biomaterials, Vol. 24, p. 1001, 2003. 44. J. Panyam, and V. Labhasetwar, Advanced Drug Delivery Reviews, Vol. 55, p. 329,2003. 45. J. L. Guesdon, Immunochemie, Vol. 14, p. 443,1977. 46. D. Charmot, P. Corpart, H. Adam, S.Z. Zard, T. Biadatti, and G. Bouhadir, Macromolecular Symposia, Vol. 150, p. 23,2000. 47. F. Montagne, O.M. Monval, C. Pichot, and A. Elaissari, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 44, p. 2642,2006. 48. D. Horák, and N. Chekina, Journal of Applied Polymer Science, Vol. 102, p. 4348, 2006. 49. G. H. Qiu, Q. Wang,C. Wang, W.L. Lau,and YL. Guo, Ultrasonics Sonochemistry, Vol. 14, p. 55,2007. 50. Q. Y Zhang, G. Xie, H.P. Zhang, J.P. Zhang, and M. He, Journal of Applied Polymer Science, Vol. 105, p. 3525,2007. 51. P.H. Wang, and C.Y Pan, Journal of Applied Polymer Science, Vol. 75, p. 1693, 2000. 52. D. Horák, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 39, p. 3707, 2001. 53. L.C.S. Maria, M.C.A.M. Leite, M.A.S. Costa, J.M.S. Ribeiro, L.F. Senna, and M.R. Silva, Materials Letters, Vol. 58, p. 3001, 2004. 54. Z.Y. Ma, YP. Guan, X.Q. Liu, and H.Z. Liu, Journal of Applied Polymer Science, Vol. 96, p. 2174,2005. 55. YH. Lee, J.H. Rho, and B.S. Jung, Journal of Applied Polymer Science, Vol. 89, p. 2058,2003. 56. VS. Zaitsev, D.S. Filimonow, I.A. Presnyakov, R.J. Gambino, and B.J. Chu, Journal of Colloid and Interface Science, Vol. 212, p. 49,1999. 57. M. Antonoett, and K. Landfester, Progress in Polymer Science, Vol. 27, p. 689, 2002. 58. R.Y Hong, B. Feng, G. Liu, H.Z. Li, J.M. Ding, Y Zheng, and D.G. Wei, Journal of Applied Polymer Science, Vol. 112, p. 89,2009. 59. J. M. Asua, Progress in Polymer Science, Vol. 27, p. 1283,2002. 60. F. Tiarks, K. Landfester, and M. Antonietti, Macromolecular Chemistry and Physics, Vol. 202, p. 51,2001. 61. H. Xu, L.L. Cui, N.H. Tong, and H.C. Gu, Journal of the American Chemical Society, Vol. 128, p. 15582,2006. 62. P.C. Wang, W.Y Chiu, CF. Lee, and T.H. Young, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 42, p. 5695, 2004.

D O U B L E M I N I E M U L S I O N FOR H Y B R I D LATEXES

275

63. L.L. Cui, H.C. Gu, H. Xu, and D.L. Shi, Materials Letters, Vol. 60,p. 2929,2006. 64. R.Y. Hong, B. Feng, G. Liu, S. Wang, H.Z. Li, J.M. Ding, Y Zheng, and D.G. Journal of Alloys and Compounds, Vol. 476, p. 612,2009. 65. X. Xu, G. Friedman, K.D. Humfeld, S.A. Majetich, and S.A. Asher, Chemistry of Materials, Vol. 14, p. 1249, 2002. 66. X. Xu, G. Friedman, K.D. Humfeld, S.A. Majetich, and S.A. Asher, Advanced Materials, Vol. 13, p. 1681, 2001. 67. L.E. Helseth, H.Z. Wen, R.W. Hansen, T.H. Johansen, P. Heinig, and T.M. Fischer, Langmuir, Vol. 20, p. 7323,2004. 68. H.Z. Wen, L.E. Helseth, and T.M. Fischer, Journal of Physical Chemistry B, Vol. 108, p. 16261,2004. 69. P.S. Doyle, J. Bibette, A. Bancaud, and J.-L. Viovy, Chips Science, 1. 295, p. 2237, 2002.

This Page Intentionally Left Blank

10 Surfactant Effect in Miniemulsion Polymerization for Biodegradable Latexes V. Soldi*, B.G. Zanetti-Ramos, and E. Minatti Departmento de Química, Universidade Federal de Santa Catarina, Florianopolis, Brazil

Abstract

This chapter summarizes the main contributions from the literature concerning biodegradable latexes nanoparticles. Most of the studies in the literature describe the effects of different surfactants used in the polymerization techniques, reaction mechanisms, and surfactant effects on the size and morphology of the nanoparticles. We have described herein some such studies, focusing mainly on the presence of biodegradable components in nanoparticles obtained by miniemulsion polymerization, the main mechanisms associated with the presence of surfactants and, in particular, we discuss the effect of surfactants on the size, morphology and polymerization processes of biodegradable latexes. The presence of surfactants can significantly affect the formation and stability of biodegradable latexes (nanoparticles) obtained through miniemulsion processes.

Keywords: biodegradable latexes, miniemulsion, surfactant effect

10.1

Introduction

This chapter addresses three topics which summarize the main contributions from the literature concerning biodegradable latexes. Firstly, the most recent findings associated with the miniemulsion processes used to obtain biodegradable latexes are discussed. Along with some concepts, a selection of the contributions is reported in which either biodegradable or partially biodegradable contents V. Mittal (ed.) Miniemulsion Polymerization Technology, (277-302) © Scrivener Publishing LLC

277

278

MINIEMULSION POLYMERIZATION TECHNOLOGY

were used to produce the miniemulsion nanoparticles. The second topic is the stabilization mechanisms promoted by surfactants in colloidal dispersions, such as in miniemulsion polymerization. In the main topic, the effects of surfactants on the miniemulsion polymerization of biodegradable latexes are described, including their influence on the size, yield, concentration and stability of the biodegradable latexes obtained.

10.2 Miniemulsion Polymerization of Biodegradable Latexes Polymer latexes are defined as polymer nanoparticles dispersed in a non-solvent, or more specifically, they can be described as relatively stable nanodroplets (size range 30-500 nm) of one phase dispersed in a continuous phase [1-3]. These polymer nanoparticles can be efficiently synthesized by miniemulsion polymerization by shearing a system containing oil, water, surfactant and a hydrophobe. Basically, the stability of a miniemulsion system is controlled by the hydrophobe, which acts as an osmotic agent stabilizing the system against Ostwald ripening, and by the surfactants which suppress the coalescence of the droplets. The stabilization of the nanoparticles formed is dependent on a combination of the effects due to the presence of surfactants and an osmotic pressure agent [1-16]. The homogenization of miniemulsion nanoparticles is strongly dependent on the energy transferred to the reaction medium according to the preparation technique used. In general, ultrasonication is used for the homogenization of small quantities, favoring the formation of more mono-disperse and smaller nanoparticles. On the other hand, mechanical systems (p. ex. Ultra-Turrax®) have been used for the emulsification of larger quantities, favoring in most cases, the formation of polydisperse nanoparticles [3,4,8,13,14,17]. Most of the published papers related to the synthesis of colloidal nanoparticles by miniemulsion polymerization focus on monomers, such as styrene, methyl methacrylate, acrylic and methacrylic acids, acrylamide and ethylene, which are not specifically biodegradable systems [6,18-20]. The authors describe different polymerization techniques, reaction mechanisms, surfactant effects and other aspects associated with the miniemulsion process. Considering that our objective is to focus on biodegradable latexes, we summarize in this topic the main contributions from the literature concerning

SURFACTANT EFFECT FOR BIODEGRADABLE LATEXES

279

the miniemulsion polymerization of these systems. The biocompatibility and biodegradability of miniemulsion nanoparticles can be improved using polysaccharides and proteins to coat or to stabilize nanoparticles for applications such as drug delivery systems. For example, using amphiphilic polysaccharides based on dextran, polysaccharide-covered nanoparticles were prepared using a miniemulsion polymerization process [21,22]. Through a direct miniemulsion process the authors obtained nanoparticles combining a highly hydrophobic core (polystyrene or poly(lauryl methacrylate)) with a hydrophilic surface coating (dextran). On the other hand, following an inverse miniemulsion process, crosslinked poly(acrylamide) nanoparticles were obtained. The size and surface coverage of the nanoparticles were controlled by the polymerization conditions. The authors also observed that the particle size decreased with an increase in dextran concentration and molecular weight. At the same time, through scanning electron microscopy (SEM) micrographs the authors showed that highly monodisperse polystyrene nanoparticles were synthesized at high dextran concentrations. In another study, amphiphilic glycopolymers, polylactide-grafted dextran copolymers, were also synthesized [23]. Depending on the proportion of polylactide/ dextran, these copolymers exhibited solubility either in water or in organic solvents and they were able to stabilize either direct or inverse emulsions. The droplet size was dependent on the amount of amphiphilic copolymer in the continuous phase. By inverse miniemulsion copolymerization of acrylamide and Ν,Ν'-methylenebisacrylamide in the presence of a polylactidegrafted dextran copolymer as a stabilizer, polyacrylamide hydrogel nanoparticles were prepared. Marie et al. (2002) [24] prepared nanocapsules of styrene using chitosan as a stabilizer, Jeffamine D2000 or Gluadin as biocompatible co-stabilizers and diepoxide as a crosslinker in the presence of an inert oil. The presence of the co-stabilizers allows the preparation of biocompatible and biodegradable nanocapsules with diameters in the range of 100-300 nm. The size and morphology can be seen in Figure 10.1. In a similar way, biocompatible and biodegradable hollow microspheres, using carboxyl-functionalized polystyrene particles as a core template and the chitosan crosslinked with glutaraldehyde as the shell, were prepared [25]. Particles with average size close to 300 nm and polydispersity equal to unity were prepared independently of the acrylic acid fraction present at the microsphere surface.

280

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Figure 10.1. TEM pictures micrographs of polystyrene latexes stabilized by chitosan: left, sample with low molecular weight chitosan; right, sample with high molecular weight chitosan. [From Ref. 24 with permission from American Chemical Society.

Monodisperse nanoparticles of biocompatible and biodegradable polymers such as poly(L-lactide), poly[(D,L-lactide)-co-glycolide] (50:50), and poly(epsilon-caprolactone) were prepared by combining the emulsion/solvent evaporation method and miniemulsion technique [26]. The authors observed differences between the various polymers in terms of the particle size (range of 80-200 nm), size distribution and degradation time. Ishimoto et al. (2009) [27] obtained nanoparticles through the miniemulsion copolymerization of poly(alkyl methacrylate-gra/i-lactic acid)s. The new bio-based polymer system allowed miniemulsions stable in water to be produced, which can be used, for example, in coating processes as a biomass-polymer material. The size distribution of the monomer droplets was measured by dynamic light scattering showing diameters in the range of 220-340 nm, depending on the surfactant used. Recently, we synthesized biocompatible nanoparticles using a natural triol (castor oil) as a monomer and isophorone diisocyanate (IPDI) in aqueous medium using the miniemulsion polymerization technique [28,29]. The stabilization of the nanoparticles was achieved by adding nonionic surfactants (Tween 80 or Pluronic F68) and olive oil as a co-stabilizer. Under these conditions nanoparticles with diameters in the range of 250-300 nm were obtained.

SURFACTANT EFFECT FOR BIODEGRADABLE LATEXES

281

In a similar way, Gaudin and Sintes-Zydowicz (2008) [5] synthesized polyurethane nanocapsules based on a biocompatible inert oil by interfacial polycondensation of IPDI and 1,6-hexanediol in direct miniemulsion process with Miglyol 812, which is a triglycéride oil currently used for drug delivery applications, as the organic dispersed phase. Stable nanocapsules with average diameter of 170-190 nm were obtained depending on the surfactants used. The ultrasonication technique used in the miniemulsion polymerization with Miglyol 812, appeared to favored the formation of polyurethane nanoparticles with a small size. The miniemulsion polymerization of styrene and lauryl methacrylate in the presence of the biodegradable surfactant N-stearoyldi(sodium)glutamate was very efficient compared to the latexes produced with sodium dodecyl sulfate [30]. In the case of styrene, the size of the nanoparticles obtained with N-stearoyl-di(sodium) glutamate (range of 45-80 nm) was similar to those produced in the presence of sodium dodecyl sulfate. Analysis by dynamic light scattering and transmission electron microscopy (TEM) suggested that the nanoparticles were quite monodisperse. The miniemulsion polymerization of pure lauryl methacrylate latexes produced very large particles. However, for this monomer, the combination of N-stearoyl-di(sodium)glutamate with an appropriate co-stabilizer allows the production of very small latexes. Biodegradable nanogels based on poly(oligo (ethylene oxide) monomethyl ether methacrylate) were prepared by inverse miniemulsion polymerization [31]. The authors reported the utility of the nanogels as carriers for controlled drug delivery scaffolds to target specific cells. The biodégradation process was evaluated for soluble polymers in the presence of water-soluble biocompatible glutathione tripeptide as a reducing agent. A miniemulsion process was also used to synthesized glutaraldehyde crosslinked gelatin nanoparticles [32]. As observed in Figure 10.2, the particle size, determined by dynamic light scattering, increases considerably with the percentage of gelatin in the system. For example, the size changed from ca. 210 nm to 420 nm increasing the amount of gelatin from 5 to 30%, independently of the thermal cycle temperature. The nanoparticles were stable even with a very small amount of surfactant used in the miniemulsion process. The authors also observed that the particle size does not change with crosslinker concentration above 3 wt % in relation to gelatin concentration.

282

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Figure 10.2. Particle size of the samples prepared with varying gelatin concentration and cross-linker concentration 4.7% for GelA (type-A, acidic porcine) in the thermal cycle. [From Ref. 32 with permission from American Chemical Society].

Biodegradable polymer nanoparticles were synthesized by direct enzymatic polymerization of miniemulsions consisting of lactone nanodroplets [2]. As described by the authors, for the enzymatic reaction in bulk or solution approximately 50 wt% of enzyme is needed in relation to the monomer. Also, the conversion is approximately 80%, even after a long reaction time (five days). In the following sections the action mechanisms and effects of surfactants in miniemulsion polymerization will be described in greater detail.

10.3

Mechanisms of Surfactant Protection of Colloidal Dispersions

10.3.1 General Behavior of a Surfactant Molecule at the Interface The thermodynamics of a given system with more than one phase is widely influenced by the number of atoms/molecules that are closest to the boundaries. To be at the interface is energetically unfavorable for a given molecule: the Gibbs energy of a system increases with the interfacial area σ (Equation 1). As a result, the thermodynamic equilibrium state for a disperse system is complete phase segregation, leading to the minimum interfacial area.

SURFACTANT EFFECT FOR BIODEGRADABLE LATEXES

UL/ 7

283

ω

In colloid dispersions, the size of the phase boundary area relative to the volume of the system is such that a substantial fraction of the total mass of the system is present at the boundaries. In these cases, surfactants can play a major role in the system. If we take a simple oil/water emulsion as an example of colloid dispersion, we see that the phenomena occurring at the phase boundaries are so different to the expected bulk phase interactions that the entire behavior of the system is determined by interfacial processes. The characteristic molecular structure of surfactants consists of a structural group that has very weak intermolecular interaction with the solvent, known as a lyophobic group, covalently bound to another group that has strong intermolecular interaction with the solvent, called the lyophilic group. This is known as an amphipathic or amphiphilic structure and these molecules have a whole set of unique behaviors as a consequence of the amphiphilic structure. When a molecule with an amphiphilic structure is dissolved in a solvent, the lyophobic group may promote a local ordering of the solvent molecules in the attempt to cage this group; this leads to a decrease in the entropy of the solvent and, consequently, an increase of the free energy of the system. When this occurs, the system responds in such a way as to minimize contact between the lyophobic group and the solvent. As a result, some of the surfactant molecules are expelled to the interfaces of the system, with their lyophobic groups oriented so as to minimize contact with the solvent molecules. Let us consider the case of an oil phase dispersed in water ( o / w emulsion): the hydrophobic parts of the surfactant molecules are oriented predominantly toward the oil. Since oil molecules are essentially nonpolar, as are the hydrophobic groups, this decrease in the difference between the two phases in contact with each other at the interface results in a decrease in the interfacial tension. On the other hand, the presence of the hydrophilic group prevents the surfactant from being expelled completely from the solvent as a separate phase, since that would require dehydration of the hydrophilic group. The amphiphilic structure of the surfactant therefore causes not only concentration of the surfactant at the interface and reduction of the interfacial tension, but also orientation of the molecule at the interface with its hydrophilic group in the aqueous phase and

284

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

its hydrophobic group oriented away from it. We will see that the mechanisms of protection of emulsions with surfactants are based on these behaviors. 10.3.2

M e c h a n i s m 1: L o w e r i n g t h e Interfacial T e n s i o n

A general prerequisite for the existence of a stable interface between two phases is that the free energy of formation of the interface is positive; when it is negative or zero, the system will undergo complete mixing of the two phases, becoming homogeneous. As predicted by Equation 1, the emulsification of a biphasic mixture of immiscible liquids will increase the total interfacial area and increase the Gibbs energy of the system. The spontaneous path will be the complete segregation of the dispersion into two separate phases with minimum interface area. Hence, here lies the first problem associated with the stability of emulsions: they are thermodynamically unstable. One way to reduce this instability is to decrease the interfacial tension between the liquids, for instance, through the addition of a surfactant as an emulsifier [33]. In the experiment referred to in Figure 10.3, the surface tension of water is lessened by the addition of an amphiphilic biosurfactant, and the surface tension becomes stable when the concentration of NaC reaches the critical micellar concentration (CMC). Above this concentration, surfactant molecules will no longer be excluded at the interface and will form micelles dispersed in the bulk. The presence of 64 i 62-i 60 i 58 \

E 56 i E 54-j 'S) 52-i 50 i 48 H 46-í

-13.5 -13.0 -12.5 -12.0 -11.5 -11.0 -10.5 -10.0 ln[NaC/mol.crrr3]

-9.5

Figure 10.3. Effect of sodium cholate (NaC) concentration on the surface tension of water.

SURFACTANT EFFECT FOR BIODEGRADABLE LATEXES

285

free micelles is mostly undesirable in emulsions, due to the bridge flocculation process that micelles may promote among the droplets. Therefore, if the surfactant is amphiphilic, the concentration must be wisely chosen, taking into account the partial volume of each phase and the total interfacial area. At concentrations below the CMC, the surfactant molecules will tend to adsorb at the interface and build a monolayer film at the liquid surface; the concentration of surfactant molecules at the interface is called Γ2 and can be determined by fitting the first segment of a curve such as that in Figure 10.3 according to the Gibbs adsorption isotherm, expressed in Equation 2.

snaincj The monolayer film of surfactant molecules produces a surface pressure π at the interface, which increases with surfactant concentration, as shown in Figure 10.4. The pressure is dependent on the physical state of the monolayer film. Lower pressures favor the gaseous state where the molecules are very dispersed and Γ2 is very low, while high pressures indicate that the film is in a more compact state, such as the liquid-condensed (LC) or solid (S) state [34]. The interfacial tension is reduced when the surface pressure is increased. The chemical structure of the surfactant plays a very important role in the stability of emulsions [35]. Its amphiphilic characteristic can be adjusted according to the chemical groups in the hydrophobic and hydrophilic regions. In general, ionic groups are more hydrophilic than nonionic polar groups and the hydrophobicity increases with the number and size of alkylic chains of the hydro phobic region of the surfactant molecule. The relationship between the hydrophilic

Figure 10.4. Illustration of different states of monolayer films of amphiphilic surfactant molecules adsorbed at the w / o or w/air interface.

286

MINIEMULSION POLYMERIZATION TECHNOLOGY

and hydrophobic characteristic of a surfactant molecule is represented by the HLB (hydrophilic/lipophilic balance) number. Surfactants with a low HLB number (3 < HLB < 6), i.e., more lipophilic than hydrophilic, are more appropriate for systems where the continuous phase is nonpolar, as is the case of w / o emulsions [36]. However, if the HLB is too low (HLB < 3), the hydration of the hydrophilic group will be weak and will not allow the nesting of the surfactant molecule at the interface and it will probably migrate to the oil phase. Surfactants with higher HLB numbers (8 < HLB < 18) will favor o / w emulsions and since these have a stronger hydrophilic characteristic they are well suited for systems where the continuous phase is polar. If the HLB is too high (HLB>18) the system can no longer be considered an o / w emulsion, as the surfactant will tend to form micelles in the aqueous phase. However, some oil can be dissolved in the micelle core driving the system to a microemulsion, i.e., an emulsion comprised of swollen micelles. In this case, the dispersion is thermodynamically stable. To demonstrate that Equation 1 still holds for microemulsions, it can be assumed that a swollen micelle is a very small droplet of oil where a Gibbs monolayer of surfactant is strongly adsorbed, leading to a very compact film (LC or S) and producing a very high surface pressure π. Assuming that the interfacial tension y is given by Equation 3, it can become negative if π > γθ. 7 = 7ο~π

(3)

Nevertheless, surfactants with a high HLB number, especially ionic surfactants, can be mixed with surfactants with a lower HLB number and used in emulsions. As we will see in the following section, this strategy can be used to charge the droplet interface of an o / w emulsion, even when using nonionic surfactants.

10.3.3

Mechanism 2: Electrostatic Stabilization

A droplet can become electrically charge in several ways, including the selective adsorption of small ions and the adsorption of larger molecules at the interface, such as ionic surfactants, biopolymers and polyelectrolytes. In this context, an ionic surfactant can be used as an additive to surfactants with lower HLB values in order to develop an electrical double layer in the droplet surroundings and, consequently, increase the repulsion between two neighboring droplets (Figure 10.5). The adsorption of anionic surfactants will

SURFACTANT EFFECT FOR BIODEGRADABLE LATEXES

287

Charged particle

Ionosphere

Electrostatic repulsion

Figure 10.5. The charge at the particle surface produces a diffuse layer of surrounding ions and an energy barrier when the particles approach each other.

give the particle a negative charge, while the adsorption of cationic surfactants will give it a positive charge at the interface. When the droplets are stabilized by addition of proteins, the ionization of the amino acid residues plays a very important role, and the charge of the droplet, and consequently the stability of the emulsion, will be strongly dependent on the pH of the aqueous phase. The mechanism of electrical repulsion can be described considering the DLVO model, which dictates that the total potential of interaction between two colloidal particles is the sum of the attraction potential (VA), which is due to van der Waals attractive forces, and of the repulsion potential (VR), which is related to the presence of electrical charge at the surface of the particles (Equation 4). The maximum potential represents the energy barrier against aggregation. If particles approach each other with sufficient kinetic energy to overcome this energy potential, aggregation occurs and the suspension is destabilized. The value of VR plays a significant role in the stability of a dispersed medium. This potential is a function of the particle radius a and of the surface potential Ψ0, and also decreases exponentially with the distance H from the particle surface (Equation 5). The exponential coefficient κ is the reciprocal of the Debye length κ_1, which can be taken as the width of the electrical double layer of the ionic atmosphere surrounding the particle. The electrical repulsion between particles inhibits the coagulation process and increases with κ_1. The effect of ionic strength must be taken into account when using ionic surfactants or polyelectrolytes due to the self-salt effect. The Debye length decreases with an increase in ionic strength, as

288

M I N I E M U L S I O N POLYMERIZATION T E C H N O L O G Y

shown in Equation 6. This means that the increase in ionic strength promotes the contraction of the electrical double layer and decreases the value of VR. Thus, the coagulation of particles is more favorable since VA becomes more dominant [37]. VT = VA+VR V

VR

_bru

-~^T6

T

0

(4)

-KH

K \2nF2I)

(5) (6)

One way to quantify the contribution of electrical repulsions to the stability of emulsions is by measuring the so-called Zeta potential. The Huckel equation (Equation 7) shows that the electrokinetic Zeta potential is inversely proportional to the Debye length and to the particle radius a. In most cases, when there is no other stabilization mechanism, it can be assumed that an emulsion is kinetically stable if the Zeta potential is greater than the threshold of 130mV I. 4πεα(1 + κα)

ÍJ\

As predicted by Equations 5 and 7, an increase in the ionic strength leads to a reduction in the Debye length and thus a decrease in the Zeta potential. Therefore, the ionic strength of the aqueous phase has to be low in order for this mechanism of protection to be effective. When using polyelectrolytes as stabilizers, the protection is not only electrical but the longer hydrophilic chain has an important conformational entropy, which can also provide steric stabilization, as we will see in the following section.

10.3.4

Mechanism 3: Steric Stabilization

When amphiphilic surfactants with large lyophilic groups (such as block copolymers) are adsorbed in the interfacial region of a colloidal dispersion, another mechanism of protection of the colloid stability takes place. The surfactant acts not only by reducing the interfacial tension but also creating the so-called steric stabilization. The mechanism of this protection is based on two different thermodynamic contributions: osmotic pressure and configurational entropy.

SURFACTANT EFFECT FOR BIODEGRADABLE LATEXES

289

For colloidal particles with a monolayer of polymeric chains adsorbed at the interface with a thickness of d, the osmotic effect will appear when two particles are closer than the distance 2d (see Figure 10.6). The origin of this effect lies in the differential osmotic pressure which appears when the polymeric chains of the two particles start to overlap. Thus, the local concentration of the polymer is higher than that of the neighborhood, i.e. the space between the particles has a lower osmotic pressure, leading to a spontaneous osmotic flow of solvent molecules into the overlap zone. This gives rise to a new repulsion potential, Vosm. However, when the distance between the particles is less than the value of d, some molecules will be forced to undergo elastic compression. The compression has a thermodynamic effect, decreasing the configurational entropy of the chains, as shown in Figure 10.7. Thus, the approximation of two particles has an entropy penalty and will increase the Gibbs energy of the system. This effect will create a different repulsion potential, Vvr. In order to regain the lost entropy, the particles must move apart allowing them more freedom of movement. The surfactant used for steric stabilization does not necessarily need to be amphiphilic, since long chain polymers are able to adsorb at the interface. If this is the case, the polymeric protective agent must be strongly anchored to the particle surface. If the molar weight of the polymer is too high, there is the possibility that the

Figure 10.6. Illustration of repulsion potential Vosm. If the particles become closer an osmotic flow of solvent is spontaneously formed, driving the particles away from each other.

290

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Loss of configurational entropy

Figure 10.7. Illustration of repulsion potential Vvr. If the particles become closer the number of possible configurations of the adsorbed polymeric chains is reduced, leading to a decrease in entropy.

various possible points of attachment will encounter two different particles rather than attach to the same particle. This is particularly the case when there is a large excess of particles relative to the concentration of polymer. Attachment of the same polymer chain to two particles essentially ties them together and brings them closer, in effect leading to particle flocculation. On the other hand, if the polymer molar weight is too low or it is poorly adsorbed on the particle surface, another phenomenon may occur: depletion flocculation. As the particles approach each other, the non-adsorbed polymer will be squeezed out from the overlapping region; as a result, the local concentration of polymer decreases and becomes less than that in the bulk, leading to a change in the local osmotic balance. Thus, there will be an osmotic flow of solvent molecules away from the overlapping region, creating a hydrodynamic suction effect between the particles, resulting in loss of stability and flocculation. The temperature also plays a very important role in the action of the polymeric protective agents. The protection only exists if the polymeric chains are under appropriate solvent conditions, i.e., the temperature must be above temperature-θ for a LCST polymer or below temperature-θ for a UCST polymer. Under poor solvent conditions, the protection vanishes and the opposite effect is observed: the adsorbed layers may add an additional attractive force between the particles due to van der Waals attraction between the two polymeric layers [38]. If the polymer is a polyelectrolyte or it is used mixed with ionic surfactants, the protection also extends to electrostatic repulsion,

SURFACTANT EFFECT FOR BIODEGRADABLE LATEXES

291

as described above. In most cases, all of the protection mechanisms can be used together to improve the dispersion stability.

10.4

Effect of Surfactants on Miniemulsion Polymerization

As previously discussed, the colloidal stability is usually controlled by the type and amount of surfactant. In the miniemulsion formulations a wide range of anionic, cationic and nonionic surfactants can be used, resulting in negatively or positively charged particles (electrostatic stabilization) or long polymeric chains at the particle interfaces (entropie and osmotic stabilization). Anionic and cationic surfactants have been employed for the formation of monodisperse droplets with sizes between 30 and 200 nm, while nonionic oligomeric or polymeric surfactants are suitable for the formation of droplets between 100 and 800 nm [13]. The obtainment of biodegradable latexes by miniemulsion polymerization is a new and interesting field of research since these latexes have many important applications. Some examples are given in Table 10.1, including polyurethane, ε-caprolactam, gelatin, poly(L-lactide) and others. The effect of surfactants on the latex particle size, yield and stability, will now be discussed.

10.4.1 Effect of Surfactant Type on the Particle Size and Latex Yield In miniemulsion formulations a wide range of anionic, cationic and nonionic surfactants can be used, resulting in polymeric dispersions with different charges and levels of stability [13]. As summarized in the second topic of this chapter, there are two main forms of stabilization to prevent premature coagulation of latex particles: (a) electrostatic repulsion between the polymer particles (provided by means of anionic surfactants and negatively charged functional groups located at the polymer/water interface); and (b) entropic/osmotic stabilization by hydrophilic macromolecules located at the surface of the polymer particles (the hydrophilic groups originate from nonionic surfactants or protective colloids). The hydrophilic groups attract water, thus creating a so-called protective waterbarrier between the particles that prevents coagulation. If protective colloids are used, normally in combination with surfactants, the latex

292

M I N I E M U L S I O N POLYMERIZATION TECHNOLOGY

Table 10.1. Surfactant type and concentration used in the synthesis of biodegradable latex by miniemulsion polymerization. Particle size (nm)

Ref.

Latexes

Surfactant (% vs monomer)

Polystyrene or lauryl methacrylate coated with dextran

Dextran 0.1 %a

170

[22]

Polystyrene

Chitosan 1 %

227

[24]

Polyurethane

SDS 3% SDS 2% SDS1%

196 208 183

[9]

Polyurethane

SDS 3% Pluronic F68 2.5% Pluronic F68 10.0%

250 480 240

[5]

Polyurethane

SDS 10.0% Tween 80 5 - 20% Pluronic F68 10.0%

463-2475b [28,29] 261-292 285

ε-caprolactam

PE-co-Bu-b-EO 4-10%

154-922

[39]

Lauryl methacrylate

N-stearoyl-di(sodium) glutamate 10%

42-177

[30]

Poly (L- lactide)

No show

100-200

[40]

340-240 295-400

[41]

740 200 185

[23]

Poly(butylanoacrylate) Span 80 1, 3 and 5% Tween 80 1, 3 and 5% Polylactide-grafteddextran

Dextranl.25% Dextran 2.5% Dextran 5.0%

Gelatin

Poly[(butyleneco-ethylene)-b-ethylene oxide] P(B/E-bEO) 0.3%

140-420

[32]

Poly(oligo(ethylene oxide) monomethyl ether methacrylate (POEOMA)

Span 80

151-225

[31]

n-Butylcyanoacrylate (BCA)

Dodecylbenzenesulfonic acid

200

[42]

Poly(alkyl methacrylategraft-lactic acid)s

SDS 1.0%-Sodium dialkyl sulfosuccinate (PEREX)1.0%

333

[27]

(Continued)

SURFACTANT EFFECT FOR BIODEGRADABLE LATEXES

293

Table 10.1. Surfactant type and concentration used in the synthesis of biodegradable latex by miniemulsion polymerization. (Continued) Latexes

Surfactant (% vs monomer)

Styrene coated by lipid Lutensol AT50 Methyl methacrylate

Lecithin

PEG-coated poly Brij®78 2-32 g/L (n-butyl cyanoacrylate) Brij®700 2-32 g/L Tween®80 2-32 g/L

Particle size (nm)

Ref.

176-210

[43]

223

[44]

180-530 180-250 90-260

[45]

'Versus dispersant phase; bBimodal distribution.

is referred to as a colloid-stabilized latex. If protective colloids are absent, the latex is called a colloid-free latex [46]. As can be observed in Table 10.1, different types of surfactant can be used in biodegradable latex synthesized by miniemulsion polymerization. Also, it can be observed that even with different surfactant types, it is possible to obtain particles with similar sizes. A comparison between sodium dodecyl sulfate (SDS), polyethylene oxide)77-poly(propylene oxide)29-poly(ethylene oxide)77 (Pluronic F68) and poly(ethyleneoxide)20-sorbitane monooleate (Tween 80) used in biodegradable polyurethane latex synthesized by miniemulsion polymerization was carried out by Zanetti-Ramos et al. (2008) [29]. In this study the authors verified that SDS is not efficient in this system, resulting in a bimodal size distribution, with the latex comprising 463 nm and 2475 nm particles . The surfactant SDS provides stabilization by inferring negative charge to the particle interface and thus producing an electrostatic repulsion, and it is the surfactant most commonly used in miniemulsion polymerization. However, its use in polyurethane latex synthesis resulted in low latex yields (around 50%). This is attributed to the poor affinity of the anionic surfactant for the polyol used. Conversely, the nonionic surfactants gave a monomodal particle size distribution and good yields. The latexes synthesized with Tween 80 or Pluronic F68 were similar in terms of yields (85%) and particle size distribution. Tween and Pluronic can be considered as polymeric surfactants that stabilize the miniemulsion by both entropie and osmotic mechanisms. This behavior seems to be more appropriate for the stabilization of the natural polyol used in the above-mentioned study. Riess (1999) [47] observed that the conventional surfactants with low molecular weight, such as sodium lauryl sulfate, often have

294

MINIEMULSION POLYMERIZATION TECHNOLOGY

a negative effect on the properties of the final polymer latexes. The HLB value of these surfactants is usually too high to stabilize o / w emulsions and they are not able to provide steric or osmotic protection. Polymeric surfactants provide some unique advantages in emulsion polymerization, such as low foaming, and good chemical and mechanical stability of the latex. The polydispersity is generally broader for systems stabilized by nonionic surfactants because electrostatic stabilization is more efficient than steric stabilization. Furthermore, for the nonionic systems, the surface coverage is dependent on particle size, the coverage of the particles increasing with decreasing particle size [13]. The particle size measurements are sensitive to any variation in pH and/or acid surfactant concentration leading to chain depolymerization/ repolymerization, affecting, consequently, the particle stability [42]. In general, samples are characterized straight after dilution minimizing the effect of changes in the particle size occurring over time. However, as can be seen in Figure 10.8, the particle size and size distribution change over time, as was observed for the n-butyl cyanoacrylate/dodecylbenzenesulfonic acid miniemulsions [42]. 10.4.2

Effect of Surfactant C o n c e n t r a t i o n o n Particle S i z e a n d Latex Yield

Several papers discuss the effect of surfactant concentration on the latex synthesized by miniemulsion processes (see Table 10.1).

Figure 10.8. Particle size (dz) and polydispersity index (Poly) evolutions as a function of time in the miniemulsion model experiment. [From Ref. 42 with permission from American Chemical Society].

SURFACTANT EFFECT FOR BIODEGRADABLE LATEXES

295

For example, Zanetti-Ramos et al. (2008) studied the effect of surfactant concentration with Tween 80 [29]. Tween 80 is used in several hundred pharmaceutical and cosmetic products, owing to its attractive cost and low toxicity [48], as compared to SDS [49]. The concentration of the surfactant (Tween 80) was varied from 5 to 20 wt.% (with respect to the monomer). Generally, the surfactant concentration should be below the CMC (critical micellar concentration) upon formation of the miniemulsion, in order to prevent any side micellar polymerization: the free micelle can dissolve some of the oil phase and become swelled micelles (microemulsion) [50]. However, a monomodal size distribution was also observed when the formulations were prepared with different surfactant types in concentrations above or below the CMC value [29]. The concentration of surfactant used in miniemulsion polymerization affects the particle size. Generally, on increasing the surfactant concentration smaller particle sizes are obtained. This effect can be observed in Table 10.1 for different polymer systems and in Figure 10.9 for a styrene miniemulsion [51]. Besides the four surfactants included in Figure 10.9, a similar behavior was observed with the C12 sulfonium surfactant and lecithin. In another study, it was observed that the particle diameter decreased from 292 nm to 261 nm when the amount of Tween

Figure 10.9. Variation of the particle size by changing the relative amount and type of surfactant in a styrene miniemulsion at a volume fraction of the dispersed phase of 0.2. [From Ref. 1 with permission from Elsevier].

296

MINIEMULSION POLYMERIZATION TECHNOLOGY

80 surfactant increased from 5 to 20 wt.% vs monomer (see Table 10.1) [29], in agreement with the literature data (13,50,52-57). The effects of concentration and surfactant type were well characterized in a study on poly(butylcyanoacrylate) nanocapsules obtained by interfacial polymerization in miniemulsions for the delivery of DNA molecules [41]. In general, the particle size decreased with increasing surfactant concentration. However, effects due to the surfactant type were not significant in this case. The effect of surfactant concentration (Tween 80) on the average particle size and yield of the formulation was analyzed for polyurethane systems (28,29). As previously discussed, the average particle diameter decreased with the surfactant concentration (Figure 10.10a). The yield of the formulation increased when a higher surfactant concentration was used [2] (Figure 10.10b). E 300 S 290 Φ

ε

■§ 280 270 g, 260-1 2 250 (A)

10 wt % (vs monomer)

20

10 wt % (vs monomer)

20

100 95· 90

> 85 80 (B)

Figure 10.10. Effect of Tween 80 concentration on: (A) average PUR particle diameter (nm) and (B) PUR particle formulation yield (%). [From Ref. 29 with permission from Elsevier].

SURFACTANT EFFECT FOR BIODEGRADABLE LATEXES

297

Moreover, it was observed that a concentration of at least 5 wt.% of Tween 80 is necessary to favor particle stability and therefore to avoid aggregation. It is to be expected that an insufficient amount of the emulsifier would fail to stabilize all the nanoparticles and thus some of them would tend to aggregate. As a result, nanospheres with large size would be produced [47]. Romio et al. (2009) [58] observed that the surfactant concentration also affects the morphology and encapsulation efficiency (EE) of nanocapsules synthesized by miniemulsion polymerization. The author observed that using 0.02g of surfactant (lecithin) the EE was 70%, and using 0.06g this value was above 80% and at higher concentrations the EE decreased [22].

10.4.3 Effect of Surfactant on the Stability Zeta potential can be taken as an index for the stability of the nanoparticles as it is an indirect measurement of particle surface potential [20]. In most cases, the higher the absolute value of the zeta potential of the nanoparticles, the larger the amount of charge on their surface, which increases the VR (see Equation 6) between the nanoparticles dispersed in the medium and thus results in a higher energy barrier against coalescence [55]. However, if the particles have low zeta potential values then there is no force to prevent the particles coming together and flocculating. Flocculation may contribute to the instability of the nanoemulsions during storage, especially for samples with low surfactant concentration. Generally, the influence of flocculation can be reduced or eliminated by steric stabilization or by electrostatic stabilization [57]. Freitas and Müller (1998) [59] demonstrated that a zeta potential value of around -25mV ensures a high-energy barrier to stabilize the nanosuspensions such as solid lipid nanoparticles. Polyurethane nanoparticles are found to be negatively charged at pH around 6.8 [29]. The zeta potential is strongly dependent on the pH of the system. Generally, the absolute value of the zeta potential is reduced on decreasing the pH [23]. The influence of the surfactant on the surface charge of PUR nanoparticles can be observed in Table 10.2. The higher negative electrical charge was provided by the presence of the SDS on the particle surface, as would be expected due to its anionic characteristic [29].

298

MINIEMULSION POLYMERIZATION TECHNOLOGY

Table 10.2. Zeta potential as a function of surfactant type in polyurethane latex synthesized by a miniemulsion process. Zeta Potential3 (mV)

pH

Tween 80 - 20%

-22.1 ± 0.12

6.9

Tween80-10%

-23.0 ± 0.20

6.8

Tween 80 - 5%

-31.5 ±0.17

7.1

PluronicF68-10%

-25.0 ± 0.95

6.9

SDS-10%

-52.3 ± 0.31

6.8

Surfactant

"Values are means ± S.D.; n=3. -20 |

,

,

> § -25 .2 Έ S Í

ñ

-30 -

_35 I

0

,

,

1

1

»

· 1

5

1

10 15 20 Tween 80 concentration (wt%)

25

Figure 10.11. Zeta potential as a function of surfactant concentration at 25°C. [From Ref. 28 with permission from Elsevier] . In order to investigate the influence of the nonionic surfactants on the origin of the surface charges, the zeta potential of the droplets as a function of the surfactant concentration was measured [29]. A slight reduction in the surface electrical charge can be observed when the Tween 80 concentration increases from 5% to 10% (Figure 10.11). This result is in agreement with Marinova et al. (1996) [60]. The nanoemulsion stabilization during the polymerization stage is probably not promoted by the surface charge effect, but is associated with the capacity to reduce the interfacial tension of the system a n d / o r the formation of a barrier that would hinder coalescence.

10.5 Final Remarks In this chapter we have described the main topics associated with miniemulsion processes used to obtain biodegradable latexes.

SURFACTANT EFFECT FOR BIODEGRADABLE LATEXES

299

In general, the literature is very limited in relation to studies on this theme, and most of them describe the effects of different surfactants used in the polymerization techniques, reaction mechanisms, and surfactant effects on the size and morphology of the nanoparticles. We have described herein some such studies, focusing mainly on the presence of biodegradable components in nanoparticles obtained by miniemulsion polymerization, the main mechanisms associated with the presence of surfactants and, in particular, we discuss the effect of surfactants on the size, morphology and polymerization processes of biodegradable latexes. In summary, the presence of surfactants can significantly affect the formation and stability of biodegradable latexes (nanoparticles) obtained through miniemulsion processes.

References 1. M. Antonietti, and K. Landfester, Progress in Polymer Science, Vol. 27, p. 689,2002. 2. K. Landfester, Annual Reviews in Materials Research, Vol. 36, p. 231,2006. 3. K. Landfester, M. Willert, and M. Antonietti, Macromolecules, Vol. 33, p. 2370,2000. 4. E.D. Sudol and M.S. El-Aasser, in P.A. Lovell, M.S. El-Aasser, eds., Emulsion Polymerization and Emulsion Polymers, Chichester, p. 699,1997. 5. F. Gaudin, and N. Sintes-Zydowicz, Colloids and Surfaces A, Vol. 331, p. 133,2008. 6. Z. Qian, J. Chen, Y. Chen, Z. Zhang, and H. Liu, Colloids and Surfaces A, Vol. 295, p. 7, 2007. 7. M.S. El-Asser and C M . Miller, J.M. Asua, ed., in Polymeric Dispersions: Principles and Applications, Dordrecht, The Netherlands, p. 109,1997. 8. J.M. Asua, Progress in Polymer Science, Vol. 27, p. 1283, 2002 9. L. Torini, J.F. Argillier, and N. Zydowicz, Macromolecules, Vol. 38, p. 3225,2005. 10. G. Baskar, K. Landfester, and M. Antonietti, Macromolecules, Vol. 33, p. 9228, 2000. 11. K. Landfester, N. Bechthold, F. Tiarks, and M. Antonietti, Macromolecules, Vol. 32, p. 5222,1999. 12. U. Yildiz, K. Landfester, and M. Antonietti, Macromolecular Chemistry and Physics, Vol. 204, p. 1966, 2003. 13. K. Landfester, Advanced Materials, Vol. 10, p. 765,2001. 14. M. Antonietti, and K. Landfester, Progress in Polymer Science, Vol. 27, p. 689, 2002. 15. M. Takasu, T. Shiroya, K. Takeshita, M. Sakamoto, and H. Kawaguchi, Colloid and Polymer Science, Vol. 282, p. 740,2004. 16. A. Durand, and E. Marie, Advances in Colloid and Interface Science, Vol. 150, p. 90, 2009. 17. K. Landfester, J. Eisenblätter, and R. Rothe, Journal of Coatings Technology, Vol. 1, p. 65, 2004.

300

M I N I E M U L S I O N POLYMERIZATION T E C H N O L O G Y

18. F. Tiarks, K. Landfester, and M. Antonietti, Langmuir, Vol. 17, p. 5775, 2001. 19. U. Yildiz, and K. Landfester, Polymer, Vol. 49, p. 4930,2008. 20. G. Baskar, and K. Landfester, M. Antonietti, Macromolecules, Vol. 33, p. 9228, 2000. 21. C. Ladaviere, M.-C. Averlant-Petit, O. Fabre, A. Durand, E. Dellacherie, and E. Marie, Colloid and Polymer Science, Vol. 285, p. 621,2007. 22. E. Rotureau, J. Raynaud, B. Choquenet, E. Marie, C. Nouvel, J.-L. Six, E. Dellacherie, and A. Durand, Colloids and Surfaces A, Vol. 331, p. 84,2008. 23. J. Raynaud, B. Choquenet, E. Marie, E. Dellacherie, C. Nouvel, J.-L. Six, and A. Durand, Biomacromolecules, Vol. 9, p. 1014, 2008. 24. E. Marie, K. Landfester, and M. Antonietti, Biomacromolecules, Vol. 3, p. 475, 2002. 25. Z. Qian, Z. Zhang, H. Li, H. Liu, and Z. Hu, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 46, p. 228, 2008. 26. A. Musyanovych, J. Schmitz-Wienke, V. Mailänder, P. Walther, and K. Landfester, Macromolecular Bioscience, Vol. 8, p. 127,2008. 27. K. Ishimoto, M. Arimoto, H. Ohara, S. Kobayashi, M. Ishii, K. Morita, H. Yamashita, and N. Yabuuchi, Biomacromolecules, Vol. 10, p. 2719,2009. 28. B.G. Zanetti-Ramos, E. Lemos-Senna, R. Borsali, V Soldi, E. Cloutet, and H. Cramail, Polymer, Vol. 47, p. 8080, 2006. 29. B.G. Zanetti-Ramos, E. Lemos-Senna, H. Cramail, E. Cloutet, R. Borsali, and V. Soldi, Materials Science and Engineering C, Vol. 28, p. 526,2008. 30. U. Yildiz, K. Landfester, and M. Antonietti, Macromolecular Chemistry and Physics, Vol. 204, p. 1966, 2003. 31. J.K. Oh, DJ. Siegwart, H. Lee, G. Sherwood, L. Peteanu, J.O. Hollinger, K. Kataoka, and K. Matyjaszewski, Journal of the American Chemical Society, Vol. 129, p. 5939,2007. 32. A. Ethirajan, K. Schoeller, A. Musyanovych, U. Ziener, and K. Landfester, Biomacromolecules, Vol. 9, p. 2383,2008. 33. K. P. Velikov, O. D. Velev, K. G. Marinova, and G. N. Constantinides, Journal of Chemical Society, Faraday Transactions, Vol. 93, p. 2069,1997. 34. S. Ni, W. Lee, B. Li, and A.R. Esker, Langmuir, Vol. 22, p. 3672, 2006. 35. A. Wiacek, E. Chibowski, Colloids and Surfaces A, Vol. 159, p. 253,1999. 36. S. Slomkowski, in Colloidal Biomolecules, Biomaterials, and Biomédical Applications, New York, Marcel Dekker, 2005. 37. M. Sauer and W. Meier, in Colloids and Colloids Assemblies, Weinhein, WileyVCH, 2004. 38. L.H. Sperling, in Introduction to Physical Polymer Science, New Jersey, John Wiley and Sons, 2006. 39. D. Crespy, and K. Landfester, Macromolecules, Vol. 38, p. 6882, 2005. 40. M. Urban, A. Musyanovych, and K. Landfester, Macromolecular Chemistry and Physics, Vol. 210, p. 961,2009. 41. A. Musyanovych, and K. Landfester, Progress in Colloid and Polymer Science, Vol. 134, p. 120,2008. 42. C. Limouzin, A. Caviggia, E Ganachaud, and Patrick Hemery, Macromolecules, Vol. 36, p. 667, 2003. 43. S.K.E. Messerschmidt, A. Musyanovych, M. Altvater, P. Scheurich, K. Pfizenmaier, K. Landfester, and R.E. Kontermann. Journal of Controlled Release, Vol. 137, p. 69,2009.

SURFACTANT EFFECT FOR BIODEGRADABLE LATEXES

301

44. Α.Ρ. Romio, N. Bernardy, E.L. Senna, P.H.H. Araujo, and C. Sayer, Materials Science and Engineering C, Vol. 29, p. 514,2009. 45. M. Wu, E. Dellacherie, A. Durand, and E. Marie, Colloids and Surfaces B, Vol. 69, p. 147,2009. 46. Product Bulletin. General Principles of Emulsion Polymerization with VeovaO Monomers, http://www.resins.com/resins/eu/pdf/vv-2-0.pdf. 47. G. Riess, Colloids and Surfaces A, Vol. 153, p. 99,1999. 48. S.I. Simöes, J.M. Tapadas, C M . Marques, M.E.M. Cruz, M.B.F. Martins, and G. Cevc, European Journal of Pharmaceutical Sciences, Vol. 26, p. 307,2005. 49. S. Sirisattha, Y. Momose, E. Kitagawa, and H. Iwahashi, Water Research, Vol. 38, p. 61,2004. 50. H. Huang, H. Zhang, J. Li, S. Cheng, F. Hu, and B. Tan, Journal of Applied Polymer Science, Vol. 68, p. 2029,1998. 51. K. Landfester, Topics in Current Chemistry, Vol. 227, p. 75, 2003. 52. J. Delgado, M.S. El-Aasser, and J.W. Vanderhoff, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 24, p. 861,1986. 53. K. Landfester, N. Bechthold, S. Förster, and M. Antonietti, Macromolecular Rapid Communications, Vol. 20, p. 81,1999. 54. Y. Luo, and X. Zhou, Journal of Polymer Science, Part A: Polymer Chemistry, Vol. 42, p. 2145, 2004. 55. S. Feng, and G. Huang, Journal of Controlled Release, Vol. 71, p. 53,2001. 56. G. Chansiri, R.T. Lyons, M.V Patel, and S.L. Hem, Journal of Pharmaceutical Sciences, Vol. 88 p. 454,1999. 57. I. Capek, Advances in Colloid and Interface Science, Vol. 107, p. 125,2004. 58. A.P. Romio, C. Sayer, P.H.H. Araújo, M. Al-Haydari, L. Wu, and S.R.P. Rocha, Macromolecular Chemistry and Physics, Vol. 210, p. 747,2009. 59. C. Freitas, and R.H. Müller, International Journal of Pharmaceutics, Vol. 168, p. 221,1998. 60. K.G. Marinova, R.G. Alargova, N.D. Denkov, O.D. Velev, D.N. Petsev, I.B. Ivanov, and R.P. Borwankar, Langmuir, Vol. 12, p. 2045,1996.

This Page Intentionally Left Blank

Index Page numbers in italics indicate figures or tables. acrylated methyl oléate (AMO), 145 activators generated by electron transfer ATRP, 192-3 AIBN (oil soluble initiator), 16, 61, 86,104,217,227 aliphatic urethane acrylate macromonomer (AUA), 49 alkoxysilane monomer, 99,119 alkyd resins, 252 alkyd-acrylic hybrid synthesis, 63-4 AMBN (oil soluble initiator), 49, 56,60 AMON (oil soluble initiator), 227 amphiphilic copolymers, 214-15, 237-40 amphiphilic glycopolymers, 279 amphiphilic polysaccharides, 238 anionic diblock copolymers as co-stabilizers, 30 anionic surfactants, 11 associating copolymers, 214-15, 237-40 atom transfer radical miniemulsion polymerization, 188-90,189 atom transfer radical polymerization (ATRP), 20-1,22,177-8,275 limitation, 21 nonionic surfactants, 11 auto-oxidative polymerization, 147

azobis(2,4-dimethyl valeronitrile) (oil soluble initiator), 60 azobis(4-methoxy-2,4-dimethyl valeronitrile) (oil soluble initiator), 60 bare magnetite nanoparticles (BMNPs), 85 batch polymerization, 44,155 benzoyl peroxide (oil soluble initiator), 16, 217 biodegradable latexes, miniemulsion polymerization, 278-82 Brownian motion/settling, 7 bulk polymerization, 2 butyl acrylate-(2-methacryloxy) ethyl)- trimethyl ammonium chloride, copolymerization, 53-4 polymerization kinetics, 54 butyl acrylate-methyl methacrylate—vinyl acetate, terpolymerization, 54-5 emulsion stability, 55 carboxyl and amino functionalized copolymer particles synthesis, 55-7 bimodal particle size distribution generation, 57,59 particle size distribution, 57,58

303

304

INDEX

carboxylated polyurethanes, 29 catastrophic phase inversion, 161 cationic surfactants, 11 cetyl alcohol (costabilizers), 12, 161,254 cetyltrimethyl ammonium bromide, 11 cetyltrimethylammonium chloride (CTMA-C1), 56, 77 chain transfer agent as costabilizer, 15 CLRP based on degenerative transfer, 178 iodine transfer polymerization, 180-1,181 reversible additionfragmentation chain transfer, 179-80,179-81 CLRP based on reversible termination, 175-6 atom transfer radical polymerization, 177-8,178 nitroxide-mediated polymerization, 176-7, 276-7 colloid-free latex, 293 colloid-stabilized latex, 293 colloidal stability, 7 comonomers, 13,14 advantage, 13 as costabilizer, 13, 24 composite polymer/ magnetic nanoparticles, 224,226,243 compositional drift, 45 controlled miniemulsion polymerization, 19-22 controlled radical miniemulsion polymerization, 27,173 bulk/solution, 174-5 features, 175 in inverse miniemulsions, 232-7 copolymer particles, 43 copolymerization process, 44,44 costabilizers, 7,10-17, 26 preventing Ostwald ripening, 7-8

cosurfactants, 153 crosslinking in VOMM miniemulsions, 168-9 diffusion, 3, 5, 9 direct ATRP, 190 direct miniemulsions, polymeric surfactants, 28 dispersion polymerization, 2 divinylbenzene (DVB), 99,103-8, 114-15 DLVO theory of electrostatic stabilization, 75 D(nm)/KLE3729,222 dodecyl methacrylate (comonomer), 13, 46 dodecylmercaptan (chain transfer agent and costabilizer), 15 dodecyltrimethylammonium chloride, 11 double-miniemulsion preparation for hybrid latexes applications, 261-6,264-6 characterization, 257-61, 258-60 drawbacks, 252 effects of reaction conditions initiator dosage, 266-7 MMA monomer concentration, 267,267-8 formation, 255,256 stability, 255-7 droplet nucleation, 215, 226 emulsion polymerization, 3-6 homogeneous and micellar nucleations, 72 intervals, 5, 6 vs miniemulsion polymerization, 8, 9 particles formation phase, 5-6 emulsion polymerization of alkyds and vegetable oils, 143-5

INDEX

emulsion polymerization of model saturated monomers, 150-2, 252 encapsulation of inorganic nanoparticles, 72, 73-6, 74 dispersion of hydrophobized, 75 hydrophobization, 73-5 miniemulsification of lipophilic dispersion, 75-6 polymerization of droplets, 76 encapsulation of magnetite nanoparticles, 85-91 double miniemulsion polymerization, 88-91 magnetic polystyrene latex synthesis, 90-1, 92 single miniemulsion polymerization magnetic polymer nanoparticles synthesis, 87-8 single miniemulsion polymerization, 86-8 SS-MPCPs synthesis, 88, 89 surface modifiers, 86 encapsulation of silica nanoparticles, 76-84 hydrophilic, 77-8 locally surface-modified, 83-4 Si0 2 /polystyrene composite particles, 79-80 surface-modified, 78-82 ethylene glycol dimethacrylate (EDGMA), 49 flooded addition of monomers, 45 fluoroacrylate-lauryl methylacrylate-methyl methacrylate, miniemulsion copolymerization, 61-2 free radical polymerization, 2,174 functionalized fatty acid derivatives, 145

305

heterophase polymerization, 72 hexadecane (costabilizers), 11-12, 15,17,49 homogeneous nucleation, 153 homogenization, 161 homogenous nucleation, 3, 48, 54,126 hybrid latexes preparation, 252 hybrid organic-organic latexes, 251,252-3 mini-emulsion polymerization, 252-3 seeded emulsion polymerization, 252-3 hybrid polymer, 252 hybrids via mini-emulsion polymerization, 253-5 mechanisms of mini-emulsions, 253-4 unseeded batch emulsion polymerization reaction, 253 hydrophobes, volatile, limiting polymer particles generation, 10 hydrophobically modified polyacrylamides, 237, 240 hydrophobically modified watersoluble polymers, 239 hydrophobicity, 53, 61 hydroxyethyl methacrylate (HEMA), 221 in situ grafting, 168 inisurfs, 35,36 anionic polymerization, 35-6 cationic polymerization, 35-6 chemical structures, 35 free radical polymerization, 36-7 ionic polymerizations, 35-6 molecular and macromolecular, 32,37 initial monomer emulsion, stability of, 27-30

306

INDEX

initiators, 2-3 interfacial polymerization, 98 inverse miniemulsion polymerization, 211-12,212, 213,224 controlled radical miniemulsion polymerization, 232-7,234, 236 hydrophobic monomers, 230, 230-2 initiation, 217 interfacial active initiator, 217-18 kinetic studies, 218-21,219-20 mechanism of, 212-13 mechanistic events, 217 poly(acrylic acid) nanoparticles, 228 traditional radical polymerization, 212,213 UV and γ irradiation, 217 water- and oil-soluble initiators, 217 water soluble monomers, 221-30,223^t inverse suspension polymerization, 3,10 iodine transfer polymerization, 180-1,181 in miniemulsion, 201-2 isophorone diisocyanate (IPDI), 62, 280 lauroyl peroxide (oil soluble initiator), 13 Lifshitz-Slyozov-Wagner relationship, 151 Lifshitz-Slezov and Wagner (LSW) theory, 256 lipophobe (sodium chloride), 10 Lutensol AT-50, non-ionic surfactant, 56, 77,293 lypophobes, 212,214, 228,231, 241-2

macroemulsion polymerization, 7 polymerization rate, 9 macromolecular surfactants, 26,28 preparation, 27 suppressing need for hydrophobe, 28 macromonomer crosslinker, 49 MAETAC, 53-4 magnetic emulsion, rheological properties applications of magnetic polymer microspheres, 270-1 viscosity vs time, 268,268 viscosity with/without magnetic field, 269,269-70 magnetic fluids, 85 magnetic nanoparticles, 224,230, 231 magnetic PMMA microspheres, 264-5 magnetic polymer nanoparticles (MPNPs), applications, 85 magnetic separation, 227 magnetite, 85 magnetite coupling agents, 75 mass-transfer processes, 30-1 melt polycondensation, 3 (meth)acrylated vegetable oil derivatives, 145-6 methacrylic acid, 217 γ-methacryloxypropyltrimethoxysi lane (MPS), 99,128 micellar nucleation method, 3, 7, 72,153 micelles, 4 miniemulsion polymerization, 3, 6-10 advantages, 7,153 avoiding micellar nucleation, 7 colloidal stability, 7 vs emulsion polymerization, 8, 9 encapsulation of inorganic nanoparticles, 73-6

INDEX

limitation, 9-10 mechanism, 7, 8 particle nucleation, mechanism of, 8-9, 72 polymerization rate, 9 polystyrene particles synthesis, 10,20 properties, 10-10 specificity, 26 see also macroemulsion polymerization miniemulsion stabilizers, 27 molecular surfactants, 26 monomer conversion, 13,14,18, 46,48,57,59,62-3, 64 monomer droplet nucleation, 153 monomer droplets, 5-6, 7 monomer hydrophobicity, 142 monomer polymerization, 5 monomer water solubility for macro and miniemulsion polymerization, 158-9 monomeric radicals, 241 multi-functional surfactants, 25 n-butyl methacrylate-crosslinking monomers, miniemulsion copolymerization, 49-51 N-isopropylacrylamide (NIPAM), 99,105-8,113-14,217 nanocapsules of styrene, 279 nanocapsules synthesis, RAFT miniemulsion polymerization, 200 nanotechnology, 71 nitroxide-mediated miniemulsion polymerization, 182,182 nitroxide mediated polymerization, 19,19, 176-7,177 (N,N'-methylenebis(acrylamide) (MBA), 220 non-aqueous miniemulsion polymerization, 26

307

non-reactive surfactants, 27,37 nonionic surfactants, 11 nonylphenol-10 ethylene oxide sodium sulfate, surfactant, 63 conversion and polymerization rate, 64 hardness of coatings, 65 NP-40, surfactant, 46, 61 nucleation mechanisms, 152-4 octane/SDS aqueous solution, interfacial tension, 100 oil-acrylate synthesis, 65-7, 66-8 oil-in-water (direct) miniemulsion polymerization, 26 oil-soluble bicomponent initiating system, 183-4,183-5 oil soluble initiators, 49,57-8, 217 polymerization of monomer droplets, 17-18 oil-soluble monocomponent initiating system, 186-8,187 oil-soluble polymeric surfactants, 29 oleic acid, 229 oligo(ethylene glycol) monomethyl ether methacrylates (OEOMA), 234 organic-inorganic hybrid nanocapsules by interfacial miniemulsion polymerization mechanism analysis of formation anchoring effects, 134-5 process analysis, 135-6 synthesis under acidic/basic conditions DVB weight contents at pH 8.5,129-30,131 monomer weight content at pH 8.5,131-2, 232 morphological evolution, 132-4, 233

308

INDEX

MPS weight content at pH 8.5, 128, 229-30 suspension pH values, 124-7, 125-7 synthesis under neutral conditions role of cross-linking reaction, 119-22, 220-2 SDS concentration, 122-4, 223 thermodynamic analysis and morphological prediction, 117-18,117-19 organic nanocapsules by interfacial miniemulsion polymerization influences on formation amount of DVB, 114-15, 225 amount of NIPAM, 113-14, 224 weight content of monomer in oil phase, 226,116-17 mechanism for formation, 222, 112-13 particle size and size distribution, 109-12, 220-22 particles morphology of system with added NIPAM and DVB, 105-8 with DVB, 103-5, 204-5 without added NIPAM and DVB, 105-9, 206-9 thermodynamic morphology prediction, 99-101, 200, 202 osmotic pressure agent, 215,226 Ostwald ripening, 7-8,10-11,16 rates, 151 ozone, adverse effects, 146 p-methyl styrene (comonomer), 13 P(AA-SA)/ZnO composite, 230, 230 particle size distribution, 213,227 particle size distributions, 50, 56, 58-9, 62 PEBBLE, 230

PEGA200,221 persistent radical effect (PRE), 175 poly(2-hydroxyethyl methacrylate) (PHEMA), 221 poly(acrylamide) nanoparticles, 279 poly(acrylic acid) (PAA), 228 poly(alkyl methacrylate-gra/f-lactic acid) copolymerization, 280 polydispersity index (PDI), 109 poly[(D,L-lactide)-co-glycolide], 280 poly(epsilon-caprolactone), 280 polyethylene oxide) (PEO), 222, 234 poly(L-lactide), 280 polylactide-grafted dextran copolymers, 279 polymer chemistry, 27 polymer-encapsulated inorganic nanoparticles, properties and applications, 72 polymer encapsulation techniques, 72 polymer latexes, 278 polymeric costabilizer, 12 polymeric stabilizers, 38, 214,238 suppress need for hydrophobes, 30 polymeric surfactants, 28,30 polymerization rate, 9 fa tors affecting, 15 polymerization techniques, 4 poly(N-isopropylacrylamide) (PNIPAM), 98,105-6,225, 226 polysaccharide, konjac glucomannan, 225 polysaccharide-covered nanoparticles, 279 polystyrene/octane/SDS aqueous solution, interfacial tension, 100

INDEX

polystyrene particles synthesis, 10,10 polystyrene/silica hybrid asymmetric particles synthesis, 83, 84 polyurethane, 252 polyurethane-block-polystyrene synthesis, miniemulsion polymerization, 62-3 poly(vinyl alcohol) advantages, 199 and hexadecane costabilizer, 11 pharmaceutical applications, 199 transurf, 38 potassium persulphate, initiator, 3, 15, 49, 55, 57, 63 precipitation polymerization, 2 premature coagulation of latex particles prevention, 291 pressure sensitive adhesives (PSAs), 145 properties of miniemulsion polymerization, 10-19 pure polymer nanoparticles (PPNPs), 85 radical miniemulsion polymerization, 212 RAFT miniemulsion polymerization, 193-4,214 colloidal instability, 196-7 inhibition and retardation, 194-6,295 livingness and controlled polymerization, 198-9 nanoparticles synthesis, 200,201 vinyl acetate, 199-200 reactive cosurfactants, reducing Ostwald ripening, 153 reactive stabilizers, 31-8 control of particle morphology, 27 initiation, propagation and transfer, 31, il-4 reactive surfactants, 27, 33

309

redox initiator pairs, 53—4, 67 reverse ATRP, 190-1 reversible addition fragmentation chain transfer (RAFT), 21-2, 179-80,179-80 advantages and disadvantages, 22 RP-HPLC chromatography, 157,158 semibatch addition of monomers, 45 shell cross-linked Knedel (SCK), 230 silane coupling agents, 75, 81, 83 silica applications, 76 simultaneous reverse and normal initiation ATRP, 192 sodium dodecyl sulphate, surfactant, 3,11,46,49,56, 62,67 sodium dodecylbenzenesulfonate, influence rate of TEMPOmediated styrene miniemulsion, 37 sodium lauryl sulfate, surfactants, 15,293 sodium metabisulfite (SMBS), 228 solution polycondensation, 3 solution polymerization, 2 soybean acrylated monomer synthesis, 160, 260-2 soybean oil, 160 Span 80 (sorbitan monooleate), 224 stable free radical polymerization (SFRP), 214 starved addition of monomers, 45 static mixer, 271 stearyl methacrylate, retarding Ostwald ripening, 46 Stöber 's method, 76 styrene-acrylic acid /2-aminoethyl methacrylate hydrochloride, copolymerization, 55-7

310

INDEX

styrene-dodecyl methacrylate/ stearyl methacrylate, copolymerization, 46-9, 47-8 styrene-butadiene rubber synthesis, miniemulsion polymerization, 57-61 styrene-butyl acrylate, batch copolymerization, 57 conversion vs. reaction time, 57,59 cumulative feed composition, 57,60 surface modification, inorganic particles, 72 surfactant effect in miniemulsion, 291,292-3 effect of surfactant concentration on particle size and latex yield, 294-6,294-7 particle size and latex yield, 291-4 stability, 297-8,297-8 surfactant protection of colloidal dispersions, mechanisms electrostatic stabilization, 286-8, 287 general behavior of surfactant molecule, 282-4 lowering interfacial tension, 284-5,284-6 steric stabilization, 288-91, 289-90 surfactants, 3-4 critical micelle concentration, 4 preventing coalescence, 7 surfmers, 37 chemical structures, 35 molecular and macromolecular, 33,37 polymerization kinetics, 37-8 step-polymerization, 38 suspension/bead polymerization, 2

tetraethylenepentamine, 226 thermally responsive nano- and microspheres, 227 thermosetting latex polymers design, 154-8,156-8 kinetic / thermodynamic implications, 154 TPM-modified silica nanoparticles, encapsulation of, 78 transurfs, 38 chemical structures, 35 molecular and macromolecular, 34 triglycérides, 147-8 tris[(2-pyridyl)methyl]amine, 234 Triton X-405, surfactant, 53-4, 57 Ultra-Turrax®, 278 ultrasonic emulsification, 255 ultrasonication, 11, 278 urethane/acrylic hybrid synthesis, 67-8 VA-086 (2,2'- azobis(methylN-(2- hydroxyethyl) propionamide), 222 vegetable oil, 240,141 vegetable oil fatty acids, 140 composition, 141 vegetable oil macromonomers (VOMMs), 139,146-50 acrylated castor oil, 247 acrylated castor oil methyl esters, 247 disadvantages, 146 miniemulsion polymerization, 160-8, 262-7 gel content, 166-8, 267 range, 149-50 structural and design characteristics, 147 study areas, 148-9 vinyl acetate, RAFT miniemulsion polymerization, 199-200

INDEX

vinyl acetate-butyl acrylate, copolymerization, 51-3 copolymer composition, 52 emulsion stability, 51 particle properties, 52 polymerization kinetics, 52 vinyl hexanoate (comonomer), 13 vinyl stéarate (comonomer), 13 volatile organic compounds (VOCs), 146 VOMM plasticization efficacy, 149

311

water-in-oil (inverse) miniemulsion polymerization, 26,193 water-soluble bicomponent initiating system, 185-6 water-soluble comblike copolymers, 29 water soluble initiator, 50, 217, 236, 241 water-soluble monocomponent initiating system, 188 water soluble monomers, 242 waterborne coatings, 146, 252

This Page Intentionally Left Blank

Also of Interest Check out these published and forthcoming related titles from Scrivener Publishing Introduction to Industrial Polyethylene: Properties, Catalysts, Processes by Dennis P. Malpass. Published 2010. ISBN 978-0470-62598-9. Demystifies the largest volume manmade synthetic polymer by distilling the fundamentals of what polyethylene is, how it's made and processed, and what happens to it after its useful life is over. A Concise Introduction to Additives for Thermoplastic Polymers by Johannes Karl Fink. Published 2010. ISBN 978-0-470-60955-2. Written in an accessible and practical style, the book focuses on additives for thermoplastic polymers and describes 21 of the most important and commonly used additives from Plasticizers and Fillers to Optical Brighteners and Anti-Microbial additives. It also includes chapters on safety and hazards, and prediction of service time models. Handbook of Engineering and Specialty Thermoplastics Volume One: Polyolefins and Styrenics by Johannes Karl Fink. Published 2010. ISBN 978-0-470-62483-5. Volume Two: Polyethers and Polyesters edited by Sabu Thomas and Visakh P.M. Forthcoming late 2010. Volume Three: Nylons edited by Sabu Thomas and Visakh P.M. Forthcoming late 2010. Volume Four: Water Soluble Polymers edited by Johannes Karl Fink. Forthcoming 2011.

Polymer Nanoutube Nanocomposites edited by Vikas Mittal. Published 2010. ISBN 978-0-470- 62592-7 Provides a one-stop source for the information on synthesis, properties, and potential applications of nanotube reinforced polymer nanocomposites. A Guide to Safe Material and Chemical Handling by Nicholas P. Cheremisinoff and Anton Davletshin. Published 2010. ISBN 978-0470-62582-8 The volume provides an assembly of useful engineering and properties data on materials of selection for process equipment, and the chemical properties, including toxicity of industrial solvents and chemicals.