SELF-ASSEMBLED POLYSTYRENE-BLOCK-POLY (ETHYLENE OXIDE) (PS-b- PEO) MICELLE MORPHOLOGIES IN SOLUTION. A Dissertation. Presented to

Transcription

1 SELF-ASSEMBLED POLYSTYRENE-BLOCK-POLY (ETHYLENE OXIDE) (PS-b- PEO) MICELLE MORPHOLOGIES IN SOLUTION A Dissertation Presented to The Graduate Faculty of The University of Akron In Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy Prachur Bhargava May 2007

2 SELF-ASSEMBLED POLYSTYRENE-BLOCK-POLY (ETHYLENE OXIDE) (PS-b- PEO) MICELLE MORPHOLOGIES IN SOLUTION Prachur Bhargava Dissertation Approved: Advisor Dr. Stephen Z. D. Cheng Accepted: Department Chair Dr. Mark D. Foster Committee Member Dr. Darrell H. Reneker Dean of the College Dr. George R. Newkome Committee Member Dr. W.L. Mattice Dean of the Graduate School Dr. George R. Newkome Committee Member Dr. Frank W. Harris Date Committee Member Dr. Alper Buldum ii

3 ABSTRACT In this research, we have investigated the self-assembly behavior of the amphiphilic diblock copolymer polystyrene-b-poly(ethylene oxide) PS 962 -b-peo 227 in DMF/water and DMF/acetonitrile mixtures. We have used solution conditions, namely copolymer concentration, solvent composition and temperature to control the morphology of aggregates. With increasing the water concentration in the DMF/water or the acetonitrile concentration in the DMF/acetonitrile system, the micelle morphology observed in transmission electron microscopy changed from spheres to worm-like cylinders and then, to vesicles. Critical micelle concentrations were determined by static light scattering. Morphological diagrams were constructed from the study of the micelle morphology changes in different copolymer concentrations. Based on the observations of morphological reversibility and annealing experiments, these two morphological diagrams were proven to be in thermodynamic equilibrium. Although the trend in morphological changes was identical in these two systems, there were remarkable differences in the morphological diagrams of PS 962 -b-peo 227. At higher copolymer concentrations a worm-network, which led to an order of magnitude increase in the inherent viscosity of the colloidal system was formed only in the DMF/water system. 4 The percentage of selective solvent required to induce morphological changes was much higher in case of DMF/acetonitrile system. The driving force for these morphological changes was understood to approach micelle free energy minimization. Approximate iii

4 micelle free energy calculations confirmed that the free energy decreases as the morphology changes from spheres to worm-like cylinders and then to vesicles with an increase in the selective solvent concentrations, and is dominated by the free energy of the interface. Further insight into the thermodynamic nature of morphological changes was achieved by inducing temperature driven morphological changes in DMF/water mixtures. With increasing the temperature, changes in the morphology from vesicles to worm-like cylinders and then to spheres were observed. Cooling the system back to room temperature regenerated the vesicle morphology indicating that the morphological changes are reversible. No hysteresis was observed in the morphological transitions during heating and cooling, indicating that the system is in thermodynamic equilibrium. The critical micellization temperatures and critical morphological transition temperatures were determined by turbidity measurements and were found to be dependent on the copolymer concentration and water content in the DMF/water mixture. The driving force for these morphological changes was understood to be reduction in the free energy of the corona and interfacial energy, which resulted in minimization of the micelle free energy. iv

5 DEDICATION This dissertation is dedicated to my grandparents, Dr. K.S. Bhargava and Mrs. Sheel Bhargava and my parents, Mr. Pradeep Bhargava and Dr. Keerti Bhargava. I am indebted to them for their support, guidance and encouragement. v

6 ACKNOWLEDGEMENTS I would like thank my advisor Dr. Stephen Z. D. Cheng for his invaluable guidance, believing in me and supporting me.i want to express my gratitude to my committee members Dr. Darrell H. Reneker, Dr. Frank W. Harris, Dr. Alper Buldum for their suggestions. I would like to thank my wife Dr. Sphurti Bhargava for her constant inspiration, helping with several experiments and useful discussions. I would like to thank my group members, especially Dr Kwang-Un-Jeong and Dr.Matthew J. Graham for their help on numerous occasions. Last, but not the least I want to thank all of my family back in India. vi

7 TABLE OF CONTENTS LIST OF TABLES..x Page LIST OF FIGURES....xi CHAPTER I. INTRODUCTION II. BACKGROUND Amphiphilic Molecules Amphiphilic block copolymers Amphiphilic block copolymers vs small amphiphilic molecules Applications of amphiphilic block copolymers and their self assembled micelle morphologies Micelle structure and morphology Methods to obtain micelles from amphiphilic block copolymers Single Solvent Process Two solvent or multi-solvent Solvent Process Controlling Micelle Morphology & Size Molecular parameters Solution parameters. 21 vii

8 2.8 Background on micellization and studies of Polystyrene-b-Poly(ethylene oxide)..27 III. OBJECTIVES..32 IV.EXPERIMENTAL Sample Synthesis Preparation of micellar solutions Transmission electron microscopy Static Light Scattering Experiments Turbidity measurements Viscosity measurements..36 V. MORPHOLOGICAL CHANGES WITH SOLVENT COMPOSITION Molecular mechanism of morphological Changes Thermodynamic and Kinetic Aspects of the Morphological Changes...49 VI. MICELLE FREE ENERGY CALCULATIONS...58 VII. EQUILBRIUM SIZES OF STRUCTURES Form for F total Exponents Coefficients 69 VIII. PHASE MORPHOLOGY OF SELF-ASSEMBLED WORMS...72 IX. TEMPERATURE INDUCED MORPHOLOGICAL TRANSITIONS Driving Force for the Morphological Changes Initial Copolymer Concentration Effects on the Morphological Changes Effect of Water Concentration on the Morphological Changes viii

9 9.4 Pathways of Morphological Changes in Isothermal Experiments...95 X. SUMMARY REFERENCES 104 ix

10 LIST OF TABLES Table Page 6.1 Micelle size and calculated parameters for a 0.5 (wt.) % initial copolymer concentration at three selective solvent concentrations Calculated free energy components and total micelle free energies at three selective solvent concentrations Calculated coefficients for the free energy dependence on size of structures at selective solvent concentrations given in Morphologies formed at different temperatures for different copolymer concentrations and 4.5 (wt.) % water concentration 91 x

11 LIST OF FIGURES e Figure Page 2.1 Structure of an amphiphilic molecule Self assembled Amphiphilic molecule Self- assembly of amphiphilic block copolymers Block copolymer micelles for drug encapsulation Drug loading in block copolymer micelles Multiple morphologies formed by amphiphilic block copolymers Formation of vesicles from amphiphilic block copolymers Multiple morphologies observed by transmission electron microscopy for Polystyrene-b-Poly(acrylic acid) Multiple morphologies (a) Helical for PS 40 -b-piaa 10 (b) pearl necklace for PS(193k)-b-PDMS(39k) observed by transmission electron microscopy Morphological Phase Diagram for polybutadiene-block-poly(ethylene oxide) (PBD-b-EO) Morphological Phase Diagram for polystyrene-block-poly(acrylic acid) PS 310 -b-paa Morphological Phase Diagram for of poly(acrylic acid)-(1,4)-polybutadiene diblock (PAA 75 -PBD 103 ) Effect of ion concentration on the morphology of PS-b-PEO Effect of ion PEO content on the morphology of PS-b-PEO Tubules(a) and other complex bilayer aggregates (b)& (c) of PS 240 -b- PEO xi

12 2.16 Vesicles with hollow rods for PS 100 -PEO Hexagonally packed hollow loops for PS-b-PEO Morphological transitions with adding water in the DMF/water system for initial copolymer concentration of 0.1 (wt.) %. (a) a pure sphere morphology at 2.75 (wt.) % water; (b) a mixed spheres and rods at 3.14 (wt.) % water; (c) a mixed spheres and long rods at 3.54 (wt.) % water; (d) a pure worm-like morphology at 4.17 (wt.) % water; (e) a mixed worm-like and vesicles at 4.24 (wt.) % water; and (f) a pure vesicle morphology at 4.52 (wt.) % water Morphological transitions with adding acetonitrile in the DMF/acetonitrile System for initial copolymer concentration of 0.5 (wt.) %. (a) a pure sphere morphology at 8 (wt.) % acetonitrile; (b) a mixed spheres and rods at 16 (wt.) % acetonitrile; (c) a mixed spheres and long rods at 30 (wt.) % acetonitrile; (d) a pure worm-like morphology at 37 (wt.) % acetonitrile; (e) a mixed worm-like and vesicles at 45 (wt.) % acetonitrile; and (f) a pure vesicle morphology at 70 (wt.) % acetonitrile Static Light Scattering measurements with adding water in the DMF/water system. The arrows point out the cmc at different initial copolymer concentrations Static Light Scattering measurements with adding acetonitrile in the DMF/ acetonitrile system. The arrows point out the cmc at different initial copolymer concentrations Turbidity measurements with adding water in the DMF/water system. The arrows point out the cmc at different initial copolymer concentrations Turbidity measurements with adding acetonitrile in the DMF/acetonitrile system. The arrows point out the cmc at different initial copolymer concentrations Morphological change diagram for PS 962 -b-peo 227 in the DMF/water system Morphological change diagram for PS962-b-PEO227 in the DMF/acetonitrile system Representation of morphologies (a) spherical observed at low selective solvent concentration (b) interacting spheres observed at low selective solvent concentration.(c) short rod (d) long rod /worm like with spherical end caps (e) lamella xii

13 5.10 Reversible morphological changes by adding DMF into the DMF/acetonitrile system with 0.6 (wt.) % initial copolymer concentration and 68 (wt.) % acetonitrile. (a) the original micelle morphology is pure vesicle; (b) mixed worm-like and vesicles at 50 (wt.)% of acetonitrile; (c) a pure worm-like morphology at 37 (wt.) % acetonitrile; (d) mixed spheres and rods at 25 (wt.) % acetonitrile, and (e) a pure sphere morphology at 7 (wt.) % acetonitrile Morphologies formed by a fast rate of acetonitrile addition to a 1 (wt.) % initial copolymer solution in DMF. Acetonitrile was added to 35 (wt.) % (a) after one day; (b) after two months Morphologies formed by a fast rate of acetonitrile addition to a 2 (wt.) % initial copolymer solution in DMF. Acetonitrile was added to 68(wt.) % (a) after1 day; (b) after two months Kinetically trapped morphologies formed by a fast rate of water addition to a 1 (wt.) % initial copolymer solution in DMF, water was added upto 3% wt.% Doughnut or toroid morphology formed by a fast rate of water addition to a 1 (wt.) % initial copolymer solution in DMF Complex bilayer aggregates formed by continuous addition of water to DMF TEM images after shadowing with Platinum (a) spheres (b) a mixture of spheres and vesicles (c) & (d) worm like micelles Porous spheres formed on adding 12wt% water to a 1.5wt% copolymer solution in THF Free energy changes with acetonitrile content for various morphologies in the copolymer-dmf/acetonitrile system Free energy changes with micelle size R for spheres in DMF/water system Free energy changes with micelle size R for cylinders in DMF/water system Free energy changes with micelle size R for lamellae in DMF/water system Worm network formed in 4 (wt.) % copolymer concentration and 3.9 (wt.) % water content in DMF/water mixture(a) Y and T shaped junctions (b) entangled network (c) microscopic size (d) multiple junctions Change in inherent viscosity on adding water in a 4 (wt.) % initial copolymer concentration solution in DMF xiii

14 8.3 Aligned cylinders at a 2 (wt.) % initial copolymer concentration and a 34 (wt.)% acetonitrile in the DMF/acetonitrile system Change in inherent viscosity on adding acetonitrile in a 4 (wt.) % initial copolymer concentration solution in DMF Representation of the Y junction Morphological changes on heating a system with 0.2 (wt.) % copolymer and 4.5 (wt.)% water concentration in DMF/water: (a) pure vesicles formed at room temperature; (b) a mixture of vesicles and worm-like cylinders at 45 ºC; (c) pure worm-like cylinders at 50 ºC; (d) a mixture of spheres and rod-like cylinders at 60 ºC; (e) pure spheres at 70 ºC Change in turbidity with temperature for a micelle system with 0.4 (wt.) % copolymer concentrations and 4.35 (wt.) % water concentration in DMF/water on heating and cooling Intermediate morphologies formed on heating a system with 0.2 (wt.) % copolymer and 4.5 (wt.) % water concentration in DMF/water: (a) a mixture of vesicles and rings at 40 ºC; (b) lamellae with protruding rods at 45 ºC; (c) circular lamellae with protruding rods; (d) vesicles with protruding rod Change in polymer solvent interaction parameter χp-s with temperature Individual polymer-solvent interaction parameters: χps-water, χps-dmf and χpeo-dmf Change in polymer solvent interaction parameter χp-s with temperature polymer- DMF/water interaction parameters: χps-solvent and χpeo-solvent Change in turbidity with temperature for micelle systems with 4.5 (wt.) % water concentration and different copolymer concentrations in DMF/water systems Change in intensity of scattered light with temperature for micelle systems with 0.1 (wt.) % copolymer and three different water concentrations in DMF/water: 6 (wt.) %, 8 (wt.) % and 10 (wt.) % Change in PS-solvent interaction parameter, χps-solvent, with temperature for different water concentrations in DMF/water systems Rod-like cylindrical morphology formed from a system with 0.4 (wt.) % copolymer concentration and 4.35 (wt.) % water concentration in DMF/water in which the original morphology was vesicles: (a) after rapidly heating to 50 ºC and equilibrating for 2 hrs; (b) after rapidly heating to 80 ºC, equilibrating for 2 hrs and then, quenching to 50 ºC and equilibrating for another 2 hrs 96 xiv

15 9.10 After quenching a system with 0.4 (wt.) % copolymer concentration and 4.35 (wt.) % water concentration in DMF/water in which the original morphology was vesicles to 50 ºC, the morphological changes with time: (a) large compound micelles after 30 min; (b) mixed large compound micelles and rod-like cylindrical micelles after 50 min; (c) rod-like cylindrical micelles after 1h; and (d) rod-like cylindrical micelles after 2h After rapidly heating a system with 0.4 (wt.) % copolymer concentration and 4.35 (wt.) % water concentration in DMF/water in which the original morphology was vesicles at room temperature to 65 ºC, equilibrating for 2h, then quenching to 50 ºC, the morphological changes with time: (a) large compound micelles after 30 min, (b) rod-like micelles after 1h, (c) rod-like micelles after 2h After rapidly heating a system with 0.4 (wt.) % copolymer concentration and 4.35 (wt.) % water concentration in DMF/water in which the original morphology was vesicles at room temperature to 70 ºC, the morphological changes with time: (a) long rods and spheres after 15 min, (b) short rods and spheres after 30 min xv

16 CHAPTER I INTRODUCTION Amphiphilic block copolymers have been known to self-assemble in solution to form a range of nanoscale morphologies 1, 2, 3, 4, 5, 6, 7, 8. These colloidal nanostructures have diverse applications such as carriers for drug delivery 9, 10, 11 and as nanotemplates 12. Key requirements for most applications include controlling the morphology and size of these self-assembled aggregates, which can be tuned via molecular 13 and solution parameters 14. Molecular parameters include the chemical repeat units, the relative block size, the molecular weight, and the polymer architecture. Solution parameters include solvent composition, polymer concentration, solution ph, temperature, and the presence of additives (ions, homopolymer, and surfactant). Most of the research work done in the last decade has focused on controlling the morphology of these aggregates by varying molecular parameters. Controlling the morphology by solution parameters offers the advantage of simplifying material synthesis. Only few research groups have used this approach and the method of preparation, and thermodynamic stability of these micelles has been a topic of discussion in the recent past 14,15. In this research, we have investigated the self-assembly behavior of an amphiphilic diblock copolymer, polystyrene-block-poly(ethylene oxide) (PS-b-PEO), in N,N-dimethylformamide (DMF)/water and DMF/acetonitrile 16. In both cases water and 1

17 acetonitrile are selective solvents for the PEO block. The degrees of polymerization of the PS and PEO blocks were 962 and 227 (PS 962 -b-peo 227 ), respectively. We have used solution conditions, namely copolymer concentration, solvent composition and temperature to control the morphology of aggregates. With increasing the water concentration in the DMF/water or the acetonitrile concentration in the DMF/acetonitrile system, the micelle morphology observed in transmission electron microscopy changed from spheres to worm-like cylinders and then, to vesicles. In between the concentration regions of two neighboring pure micelle morphologies, mixed morphologies such as spheres with short cylinders or worm-like cylinders with vesicles could be found. The morphological diagrams were constructed from the study of the micelle morphology changes in different copolymer concentrations and the critical micellization concentrations for both systems at different copolymer concentrations as determined by static light scattering experiments. Based on the observations of morphological reversibility and annealing experiments, these two morphological diagrams were proven to be in thermodynamic equilibrium. Although the trend in morphological changes was identical in these two systems, there were remarkable differences in the morphological diagrams of PS 962 -b-peo 227. At higher copolymer concentrations a worm-network, which led to an order of magnitude increase in the inherent viscosity of the colloidal system was formed only in the DMF/water system. The percentage of selective solvent required to induce morphological changes was much higher in case of DMF/acetonitrile system. These differences can be attributed to the large difference between the polymerselective solvent interaction parameters. The driving force for these morphological changes was understood to approach micelle free energy minimization. Approximate 2

18 micelle free energy calculations confirmed that the free energy decreases as the morphology changes from spheres to worm-like cylinders and then to vesicles with an increase in the selective solvent concentrations, and is dominated by the free energy of the interface. Further insight into the thermodynamic nature of morphological changes was achieved by inducing temperature driven morphological changes in DMF/water mixtures. With increasing the temperature, changes in the morphology from vesicles to worm-like cylinders and then to spheres were observed. Mixed morphologies like vesicles and rings, vesicles and worm-like cylinders and rod-like cylinders and spheres were also formed in the intermediate temperature regions. Cooling the system back to room temperature regenerated the vesicle morphology indicating that the morphological changes are reversible. No hysteresis was observed in the morphological transitions during heating and cooling, indicating that the system is in thermodynamic equilibrium. We also however, captured intermediate morphologies, which suggest the possible pathways for these changes. The critical micellization temperatures and critical morphological transition temperatures were determined by turbidity measurements and were found to be dependent on the copolymer concentration and water content in the DMF/water mixture. The morphological changes were only possible if the water content in the DMF/water mixture was low else the mobility of the polystyrene chains would be severely restricted. The driving force for these morphological changes was understood to be reduction in the interfacial energy and free energy of the corona, which resulted in minimization of the micelle free energy. Morphological observations at different time periods of isothermal experiments indicated that in the pathway of one equilibrium 3

19 morphology transfers to another, large compound micelles formed as an intermediate or metastable stage of this pathway. 4

20 CHAPTER II BACKGROUND 2.1 Amphiphilic Molecules Amphiphilic molecules are a special class of molecules, which have dual affinity or affinity for two or more different kind of environments. A simplest amphiphilic molecule is shown in Figure 1, which has a distinct hydrophilic head and a hydrophobic tail. Thus the dual affiliation is for hydrophilic and hydrophobic environments. Hydrophilic head group Hydrophobic hydrocarbon tail Figure 2.1 Structure of an amphiphilic molecule The most common example of amphiphilic molecules are surfactants which are present in the soaps and detergents which we use in everyday life. An example of a surfactant molecule is sodium dodecyl sulfate. Another class of amphiphilic molecules is lipids, 5

21 which are present in all living beings. An interesting property of these amphiphilic molecules is that they can self-assemble in solution to form a variety of structures. Selfassembly is defined as Spontaneous, reversible organization of pre-existing components by means of non-covalent interactions 17. These non-covalent interactions may include one or more of the following forces: hydrophobic forces, hydrogen bonding, van der Waals forces, and electrostatic forces. The self- assembly takes place at molecular concentrations above a critical concentration, which is known as the critical micelle concentration (CMC) Figure 2.2 shows some of the different self assembled structures small amphiphilic molecules can form. These include spheres, rods, monolayers, vesicles etc. Monolayer Vesicle Bilayer Sphere Figure 2.2 Self assembled Amphiphilic molecule In the human body, these molecules self-assemble to form a plethora of organized structures like the lipid bilayer membranes, which perform extremely sophisticated biological functions. These colloidal assemblies and their transformations are the very basis of many life processes. 6

22 2.2 Amphiphilic block copolymers This dual affiliation or amphiphilic nature can be incorporated into polymers by chemically linking together, blocks of two dissimilar repeating units to obtain amphiphilic block copolymers. Simplest case is when a hydrophobic block and a hydrophilic block are linked together. eg. polystyrene-block-poly(acrylic acid) PS-b-PAA where PS is hydrophobic and PAA is hydrophilic. Amphiphilic block copolymers can self-assemble to form micelles in aqueous media or when mixed with organic solvents. Micelles of block copolymers, which have a long corona and small core, are termed as star-like ; while, the ones with a large core and short corona are termed as crew cut. 18 Figure 2.3 shows a representation of a diblock copolymer and it s self-assembly. Graft block copolymers can also self assemble as shown in the figure. A plethora of morphologies have been recently reported, pioneered by Eisenberg and other research groups, some of which are analogous to those formed by small molecule amphiphiles, such as surfactants 13, 19, 20, 21, 22, 23, 24. Symmetric micelles, in which the core and corona forming blocks are similar in size, can also form multiple micelle morphologies. 24 We will review this in a later section. 2.3 Amphiphilic block copolymers vs small amphiphilic molecules Although small amphiphilic molecules and amphiphilic block copolymers share the same amphiphilic nature, there are several similarities and differences in their self-assembly behavior and properties of the self assembled aggregates. The similarity is that both these species can self-assemble under suitable conditions to form aggregates with multiple morphologies. The differences are as follows: 7

23 Figure 2.3 Self- assembly of amphiphilic block copolymers. 1. Size of molecules and the self assembled aggregates: Since polymer molecules are much larger than small amphiphiles, the size of block copolymer aggregates is usually larger as compared to the small amphiphiles. Also there is greater flexibility in tuning the size of these aggregates, which we will discuss in the later section. 2. Kinetic aspects: For small molecules the self-assembly process is instantaneous, while for amphiphilic block copolymers due to the large size kinetic aspects need to be considered and this adds another level of complexity to block copolymer self-assembly. 3. Degree of segregation of the core forming and corona block: In small molecules amphiphiles, the hydrocarbon tail and the hydrophobic head are strongly segregated. In case of amphiphilic block copolymer the degree of segregation can be tuned by choosing the chemical repeat units of the blocks. 8

24 4. Self-assembly polymorphism: Small molecule amphiphiles can form a variety of morphologies including spheres, rods, worm-like, bilayer vesicles etc. Biological amphiphiles like proteins can form complex helical morphologies. For block copolymers, work in many research groups has focused on obtaining multiple morphologies, which match the plethora of morphologies of small amphiphiles and biological amphiphiles. With the flexibility in designing block copolymers and their interesting self-assembly behavior in both aqueous and a variety of mixed solvents there is a potential of forming complex morphologies which may be not be possible for their small molecule counterparts. 5. Range of Applications: Self-assembled morphologies of amphiphilic block copolymers possess better stability and are more robust as compared with those formed by small molecule amphiphiles such as surfactants. This property coupled with the flexibility in designing them gives them potential for a range of applications, which are discussed, in the next section. The ratio of the hydrophobic part and hydrophilic can be easily tuned by choosing the molecular weight of the blocks. The flexibility of the aggregates can be tuned for a particular application by choosing the glass transition temperature of the blocks. Also the core forming block or the corona forming block can be selectively crosslinked to impart additional stability to these aggregates. 2.4 Applications of amphiphilic block copolymers and their self assembled micelle morphologies 1) Emulsifying agents and emulsion stabilizers: Amphiphilic block copolymers can be used as both emulsifying agents and emulsion stabilizers. For emulsifying agents, the 9

25 molecular weight is usually below 2000 while for post stabilization of emulsions higher molecular weight is required. Both steric effects and electrostatic effects can provide stabilization of emulsions. Block copolymers are being designed for emulsion polymerization such that the hydrophobic group matches with the latex polymers. This provides a maximum interaction between the hydrophobic block and the particle surface. Polymer (ethylene oxide)-block-poly(propylene oxide) copolymers have been investigated for the this application for different block lengths and ratios of the hydrophobic and hydrophilic block 25. Polystyrene-block-poly(ethylene oxide) (PS-b- PEO) has also been investigated as emulsion stabilizers 26. In related applications amphiphilic block copolymers can also be utilized as demulsifiers 27, defoamers 28 and foam stabilizers 29. 2) Drug delivery and Gene Delivery Self-assembled colloidal nanostructures of amphiphilic block copolymers have been investigated as vehicles for drug delivery for over a decade. Hydrophobic drugs can be incorporated in the hydrophobic core, while the hydrophilic shell provides steric stability Hydrophobic Drug Steric Stabilization Figure 2.4 Block copolymer micelles for drug encapsulation 10

26 and protects the drug from interacting with blood components 30 Poly(ethylene oxide) (PEO) based block copolymers with hydrophobic biodegradable polyesters (PLA, PGA etc.) and poly (amino acids) have been the highlight of last decade s research. Block copolymers with PEO in the corona are especially attractive due to its biocompatibility. 31 Small size of micelles in the range of nm make them attractive for this application. Enhanced permeation and retention effect displayed by these carriers can be used for passive delivery. Attaching ligands to the micelles can help to achieve active delivery and control the biodistribution. Drugs can be loaded in a variety of routes 32 as shown in Figure 2.5. Drugs have also been incorporated by covalently attaching them to the micelles. Drug release characteristics from micelles have also been studied and effort has been made in the literature to control the same and thus to obtain controlled release drug carriers. Figure 2.5 Drug loading in block copolymer micelles. 11

27 The cell uptake mechanism for the micelles is based on endocytosis 33. Recent developments include attaching ligands to micelles for receptor mediated endocytosis and developing ph sensitive micelles which can display the proton sponge effect. 34. Temperature driven drug delivery by thermo responsive micelles is also promising. This temperature dependent micellization and dimicellization is useful for temperature driven drug delivery. 35 Several block copolymers have been investigated in this respect including N-isopropylacrylamide based copolymers which is known to have a LCST behavior in water. 36, 37, 38 Block copolymers with a cationic block can be used to encapsulate DNA and serve as non-viral gene delivery vectors. Poly(ethylene imine) PEI and its copolymers, which display high cationic charge density and buffering capacity, have been researched extensively 39,40.The effect of morphology on the encapsulation efficiency and drug release has not been addressed in detail in literature. Most of the work done focuses on sphere morphology. Recently, worm like micelles and vesicles have also been investigated for encapsulating drugs and are anticipated to be an active research area in the future. 3) Applications in separation systems: The ability to form micelles by amphiphilic block copolymers can be used to solubilize hydrophobic compounds in the interior of the micelles. These molecules are an alternative to organic solvents for removal of organic molecules. Water based solutions of copolymers are also environmentally friendly. Temperature or ph induced association can be used as an external influence to trigger the separation process by inducing self- association or micellization by changing the stimuli. These stimuli responsive systems have potential use in the field of biotechnology in 12

28 which they can be used for extraction of sensitive bio-molecules like proteins. By external controls these species can be triggered to capture and release bio-molecules. The flexibility in design and possibility of attaching target ligands make them more attractive for these applications. PEO-PPO-PEO block copolymers have been utilized in these applications and investigated for structure property relationships. 41,42 Self- assembled micelles can also be used for separation of microorganisms from fluids and wastes 43 where microorganisms can be immobilized on the micelle surface. 4) Other Applications: Amphiphilic block copolymers can also be utilized in biosensors by immobilizing enzymes etc on their surface where their high specific surface area can be utilized. The self-assembly can also be used to obtain nanotemplates 12 and utilized for pattering etc. 2.5 Micelle structure and morphology Similar to small amphiphilic molecules amphiphilic block copolymers can also form a variety of morphologies. Figure 2.6 shows some of the different morphologies which can be obtained 44. In this schematic the grey part corresponds to the hydrophobic core and the individual black lines correspond to the hydrophilic chains. The morphologies in this figure are spherical, rod-like, interconnected rods, lamella, vesicles and large compound micelles (LCM). Eisenberg and co-workers investigated the internal structure of the LCMs. The bulk of the micelle consists of reverse micelles, which have the hydrophobic chains in the corona and hydrophilic chains in the core. The surface of these LCMs is stabilized by individual chains with hydrophilic chains on the surface. Let us look in detail at the vesicular morphology. As shown in Figure 2.7, one series of amphiphilic block copolymers can self-assemble to form one layer and another series of molecules 13

29 can form another layer and thus they can from bilayers which can be in the form of huge sheets and lamellae These lamella can close to form hollow structures; vesicles. The closure of the lamella is governed by the interaction of rim energy and bending energy 45. This will be discussed in detail in the section VI. Transmission electron microscopy (TEM) is a powerful tool to visualize directly these self-assembled morphologies. Figure 2.8 shows the multiple morphologies observed by transmission electron microscopy for polystyrene-b-poly(acrylic acid). In figures 2.8a-d spheres, rod, vesicles and large compound micelles can be observed respectively. 46 Other morphologies have also been reported. Sphere Cylinder/Rod Bicontinuous Rods Lamella Hydrophilic block hydrophilic in corona hydrophobic Vesicle Hydrophobic block Core of reverse micelles (hydrophilic) Large Compound Micelles Figure 2.6 Multiple morphologies formed by amphiphilic block copolymers. 14

30 Figure 2.7 Formation of vesicles from amphiphilic block copolymers. 100 nm 100 nm. 100 nm 1 µm Figure 2.8 Multiple morphologies observed by transmission electron microscopy for Polystyrene-b-Poly(acrylic acid). Helical morphology has been observed for a polystyrene-b-poly(isocyanopeptide) 47 as shown in Figure2.9a and pearl necklace morphology (Figure 2.9b) for Polystrene-bpolymer(dimethylsiloxane) PS(193k)-b-PDMS(39k) 48 15

31 2.6 Methods to obtain micelles from amphiphilic block copolymers Depending on the physiochemical nature of the block copolymer, micelles can be obtained by various routes. Below we discuss the various methods and give specific examples where the method has been used. A B 2000 Figure 2.9 Multiple morphologies (a) Helical for PS 40 -b-piaa 10 (b) pearl necklace for PS(193k)-b-PDMS(39k) observed by transmission electron microscopy Single Solvent Process In this method the block copolymer is dispersed in a selective solvent for one of the blocks and micelles are formed with the phobic block in the core and the philic block in the corona. This method is usually used for block copolymers in which the core forming block is flexible or its glass transition temperature is below the micellization temperature. This process can also be used if the core-forming block is glassy at the micellization temperature but is small as compared to the corona-forming block i.e for 16

32 star-like micelles with glassy cores. For example, Bates and co-workers 13 obtain micellar solutions of PBD-b-PEO by directly dispersing the block copolymer in water at room temperature which is a selective solvent for PEO. PBD is hydrophobic but is flexible at room temperature, and thus this process can be used. In this way, the polymer-solvent interaction parameter, χ P-S, does not vary. Another variation of this method is to dissolve both the blocks by heating a selective solvent and the cooling the system to induce micellization. Eisenberg and coworkers used this method to prepare micelles for polystyrene-b-poly(acrylic acid) in low alkanols (i.e. methanol to n-butanol) Two solvent or multi-solvent process In block copolymers where the core-forming block is glassy at the micellization temperature, it is not possible to obtain stable micelle morphologies by the single solvent process. Either the block copolymer will not disperse or one would obtain kinetically trapped morphologies if an attempt were made to use the single solvent process for such block copolymers. In these cases a two or multi-solvent process is used. In the first step, the block copolymer is dissolved in a good solvent for both the blocks. Then a solvent is added which is selective for one of the blocks, which is miscible with the common solvent. As the selective solvent is added the solvent composition continuously changes and thus the he χ P-S in this method changes. At a critical concentration of the selective solvent, micellization is induced and micelles are formed. This critical solvent concentration can be determined using static light scattering and turbidity measurements. One example is polystyrene-block-poly(acrylic acid) (PS 310 -b-paa 52 ) dissolved in dioxane as a common solvent, and a then selective solvent for PAA. Water is added 17

33 slowly to induce micellization 14,50. Other block polymers such as PS-b-PDMS and PS-b- PEO have also been studied for self- assembly using this method Both the above methods offer material and process parameters which can be used to control the size and morphology of the self-assembled aggregates which will be discussed in the next section. 2.7 Controlling Micelle Morphology & Size The size and morphology of these self-assembled aggregates can be controlled by molecular parameters, which are pertaining to the chemistry of the block copolymer, and solution parameters, which are related to the characteristics of the solution. The molecular parameters include the nature of chemical repeating units of the blocks, the sizes of each block, the overall molecular weight, and the architecture. It has been shown in the past decade that these parameters are key to tune the micelle morphologies. 12,22. In most cases when these parameters are varied to control the morphology and size, the single solvent process is used. The second category of the parameters, which have been investigated only recently, are the type of solvents and solvent quality, the solvent/nonsolvent ratio, the copolymer concentration, the ph value, additives such as salts, ions and homopolymer and the temperature. 48,50. When solvent quality and solvent/non-solvent ratio are used as controlling parameters, mostly two solvent method is used. The micelle morphology is also sometimes tuned by introducing specific interactions like hydrogen bonding or excess chirality in the monomer units. 47 Below we discuss the above parameters in detail and cite some examples from literature. The formation of different morphologies is explained by considering the micelle free energy, which consists of three components. 46 The free energy of the core, which relates 18

34 to the stretching of the core-forming block; the free energy of the interface, which relates to the surface tension between the core forming block and the solvent; and the free energy of the corona, to which the electrostatic and steric interactions of the coronaforming blocks contribute. Mathematically this can be expressed as F = F core + F interface + F corona (2.1) Many researchers have also theoretically treated the micelle free energy 51, 52, 53, 54. Below we discuss the above parameters in detail and cite some examples from literature and discuss how some of these parameters affect morphology of the block copolymer aggregates Controlling Micelle Morphology & Size: Molecular parameters (a)chemical repeat unit of the two blocks: The chemical nature of the repeat unit determines interaction parameter χ between the two blocks. This would determine the degree of segregation of the two blocks. Usually hydrophobic and hydrophilic blocks linked together are super segregated 45. The chemical repeat units would also decide specific interactions like hydrogen bonding. Ionic or protonable groups present in the backbone and/or side chains would result in additional electrostatic interactions which could affect the free energy of the corona. Many different chemical repeat units and combinations have been investigated, some examples will be discussed in the sections below. (b) Copolymer composition: This relates to the ratio of the hydrophobic block and the hydrophilic block. This one of the most important factor to control the morphology of these self-assembled aggregates. Eisenberg and coworkers studied the effect of poly(acrylic acid) content in polystyrene-block-poly(acrylic acid) on the morphology of 19

35 the block copolymer 22. They used a two solvent method.in the first step the block copolymer was dissolved in N,N-dimethylformamide (DMF) which is a common solvent for both the blocks, then water was added drop wise until 25wt% water was reached. The resulting colloidal solution were placed in dialysis bags and dialyzed against water to remove the DMF. The initial copolymer concentration in DMF was kept fixed to 2wt%. They found that as the PAA content in the block copolymer decreased, the morphology of the aggregates changed from spheres to cylinders, to bilayer (lamella and vesicles) and eventually to compound micelles. The aggregates were analyzed by transmission electron microscopy. The multiple morphologies are shown in Figure 2.8. A systematic study, on the effect of block copolymer composition and molecular weight on the morphology of polybutadiene-block-poly(ethylene oxide) (PBD-b-PEO) has been carried out by Bates and coworkers 13. They varied the PEO content in the block copolymer and the molecular weight of the PBD block. The micelles were obtained by dispersing the block copolymer in water at room temperature, which is a selective solvent for the PEO block. This method could be used, as PBD is flexible at room temperature. They obtained a morphological phase diagram, which maps equilibrium micelle morphologies as a function of the varying parameters. The morphologies obtained and the morphological phase diagram is shown in Figure They found that on decreasing the PEO content the morphology changed from spheres to worm like micelles and then to bilayers. They observed that the worm like micelles formed a network for a degree of polymerization of the PBD of 170 and a weight fraction of PEO of

36 2.7.2 Controlling Micelle Morphology & Size: Solution parameters (a) Solvent and concentration: The effect of solvent is usually studied by using two solvent approaches. The effect of solvent can be divided into three parts. First, is the effect of solvent/non solvent ratio or the solvent composition, second the nature of the common solvent, and third the effect of selective solvent. Eisenberg and workers have studied the effect of solvent composition and common solvent on the block copolymer morphologies of PS-b-PAA 14,50, however the effect of different selective solvents has not been studied so far on any block copolymer system. The effect of solvent composition has been studied in terms of different water contents in dioxane/ water mixture. A two solvent process was used and the morphology was visualized by TEM. They found that on increasing the water content in the dioxane/water mixture the morphology changed from spheres to a mixture of spheres and cylinders, to pure cylinders, then to a mixture of cylinders and vesicles and finally to pure vesicles. They also studied the effect of copolymer concentration and obtained a morphological phase diagram which is shown in Fig Varying the solvent composition changes the polymer solvent interaction parameter χ which has and effect on the interfacial free energy and thus affects the morphology. This is the only polymer and solvent system for which a study had been carried out. Very recently, Bang and workers studied the morphology for polystyrene-bpoly(isoprene) PS(13000)-b-PI(71000)in different solvent mixtures of di-n-butyl phthalate(dbp), diethyl phthalate (DEP), and dimethyl phthalate (DMP) 55. They obtained sphere, rods and vesicle morphology by varying the solvent composition. 21

37 N PB w PEO Figure 2.10 Morphological Phase Diagram for polybutadiene-block-poly(ethylene oxide) (PBD-b-PEO) However they obtained mostly mixed morphology in the compositions they investigated. The morphology was studied by SAXS( small angle X-ray scattering ) and cryo TEM. Eisenberg and coworkers studied the effect of common solvent on the morphology of PS-b-PAA. 46 The morphology depends also on the nature of the common solvent in which the copolymers are initially dissolved For 2% PS(500)-b-PAA(60) copolymers, when pure DMF is used as the common solvent, spherical micelles were formed; however, if the initial common solvent is THF or dioxane, vesicles are produced. 22

38 Spheres + Rods Spheres Rods Rods+Vesicles Vesicles Copolymer Concentration (wt%) Water content (wt%) Figure 2.11 Morphological Phase Diagram for polystyrene-block-poly(acrylic acid) PS 310 -b-paa 52. The choice of common solvent determines the dielectric constant of the medium and would effect the inter-corona interactions, in addition to affecting the polymer solvent interaction parameter χ. This would also effect the swelling of the core and thus the free energy of the core. Putuax and coworkers studied the effect of the copolymer concentration and architecture on the morphology of micelles of poly(styrene-b-isoprene) 23.For linear chains spherical 23

39 micelles were obtained, while for similar cyclic copolymer chains giant worm like micelles and vesicles were obtained. On increasing the copolymer concentration the morphology changed from planar particles to worm like micelles and then to vesicles at higher concentrations. (b) Effect of additives: Additives, which can be added, include salts such as sodium chloride (NaCl), calcium chloride (CaCl 2 ), surfactants such as sodium dodecyl sulfate and homopolymer. Also alkali such as sodium hydroxide (NaOH) and acid such as HCl can also be added to tune the morphology. Eisenberg and coworkers studied the morphogenic effect of the addition of salts such as CaCl 2 and NaCl on the micelles 46. As the concentration of added salt increases, the morphologies change from spheres to vesicles, which is similar to the effect of adding HCl, increasing the copolymer concentration or decreasing the length of soluble block. However, changes in the aggregate morphology can be achieved with much lower concentrations of CaCl 2 than of NaCl. An example is the morphology of the aggregates made from 1 wt% PS(410)-b- PAA(25)copolymers in DMF. Without any additives, the aggregates are small spheres. When NaCl is added to a final concentration of 10mM, vesicles are formed. However, if CaCl 2 is used as the additive, vesicles are formed when the added final concentration of CaCl 2 is as low as 90 mm. Discher and coworkers studied the effect of ph, NaCl and CaCl 2 concentration on the morphology of poly(acrylic acid)-(1,4)-polybutadiene diblock (PAA 75 -PBD 103 ) in detail 24. They obtained a morphological phase diagram which is shown in Figure They observed spheres, worm like, a worm network above an additive concentration and higher order vesicles. Decreasing the ph or increasing the salt 24

40 content the morphology changed from spheres, to a mixture of spheres and cylinders, then to cylindrical network and finally to vesicles. Figure 2.12 Morphological Phase Diagram for of poly(acrylic acid)-(1,4)-polybutadiene diblock (PAA 75 -PBD 103 )Eisenberg and coworkers studied the effect of homopolymer styrene content in the PS-b-PAA 46. For a block copolymer which formed vesicles, adding homopolymer styrene changed the morphology from vesicles to rods and then to spheres.c) Effect of temperature Temperature has been used as a tool to control the block copolymer self-assembly for several years. 56, 57, 58, 59, 60, 61, 62, 63 Depending on the nature of the polymer-solvent interactions, if the system displays an upper critical solution temperature (UCST) behavior, micellization takes place below a critical temperature; while if the polymer displays a lower critical solution temperature (LCST) behavior micellization takes place 25

41 above a critical temperature. In either case this temperature is known as the critical micellization temperature (CMT). This temperature dependent micellization and dimicellization is specifically useful for temperature driven drug delivery. As mentioned earlier, N-isopropylacrylamide based copolymers have been investigated in this respect which is known to have a LCST behavior in water. 36,37,38 Poly(ethylene oxide)-block-poly(propylene oxide)-block-poly(ethylene oxide) (PEO-b- PPO-b-PEO) triblock copolymer has also been exhaustively investigated due its interesting LCST behavior in water. 64, 65, 66, 67, 68, 69 At higher concentrations it also exhibits ordered phases such as cubic, hexagonal and lamellar phases. 65 Relatively less literature is available on using temperature to control micelle morphologies of block copolymers in dilute solution. 48,58 Pispas and Hadjichristidis investigated the effect of temperature on polybutadiene-block-poly(ethylene oxide) (PBD-b-PEO) micelle solutions in water by static and dynamic light scattering (SLS and DLS). 70 They observed a slight increase in the intensity of scattered light and a decrease in the hydrodynamic radius with increasing temperature. But they could not conclude any changes in the micelle morphology. A temperature-concentration phase diagram was mapped for polystyrene-blockpolydimethlysilocsine) (PS-b-PDMS) in which PS formed the core and PDMS formed the corona by Iyama et al. 58 Using SLS and DLS they estimated distinct phases of spheres and cylinders on changing the temperature and polymer concentration. The effect of temperature on micelle morphology has so far mostly been studied by indirect techniques such as SANS, SLS and DLS. Very recently, temperature-induced morphological changes were studied by atomic force microscopy (AFM) for polystyrene-blockpolyisoprene (PS-b-PI) in a selective solvent for PI blocks. 71 The copolymer composition 26

42 was specifically chosen to be on the phase boundary of a calculated morphological phase diagram. For two different block copolymer compositions they observed rods to sphere change and vesicle to cylindrical change, respectively. The changes were shown to be reversible although required over fifteen hrs for the forward change to take place, almost a month to change the morphology back completely. 2.8 Background on micellization and studies of Polystyrene-b-Poly(ethylene oxide) Since we are interested in the micellization of PS-b-PEO, this subsection will be devoted to relevant work in literature for this specific block copolymer. Micellization of polystyrene-block-poly(ethylene oxide) (PS-b-PEO) diblock copolymers was first investigated almost four decades ago in a ethyl benzene which is selective solvent for the PS block. 72 In 1982, spherical micelles in water, which is a selective solvent for the PEO block, were studied via static light scattering experiments. 73 Recently, the micellization of PS-b-PEO in the presence of various additives was also reported. 74,75 Eisenberg and his coworkers obtained multiple micelle morphologies of PS-b-PEO by two approaches. First, by adding milli-molar amounts of salts or acid. 21 They obtained a mixture of spheres and rods and lamella with protruding rods. These are shown in figure 2.13 Secondly; they varied the molecular weight of the PEO block. i.e the PEO composition in the block copolymer. 76 In this case they obtained rods, vesicles and large compound micelles as shown in figure However in most cases, different mixed micelle morphologies were observed. A pure single morphology was rarely found. The method they used to prepare the micelles was to first dissolve the block copolymer in DMF which is a common solvent for both PS and PEO and then add water which is a selective solvent for the PEO block up to 25wt %. Using a similar method, complex bilayer aggregates 27

43 including tubules were obtained, which were dependent on the relative block length, polymer concentration, and annealing time. 77,78 These are shown in Figure nm 200 nm PS(240)-b-PEO(180) No salt PS(240)-b-PEO(180) 1.3mM LiCl Figure 2.13 Effect of ion concentration on the morphology of PS-b-PEO They were explained to be kinetically quenched morphologies. More recently, by changing the common solvent they observed, vesicle with hollow rods, which was identified to be a trapped intermediate morphology 79,80 and is shown in Figure They also observed hexagonally packed hollow loops when THF was used as a common solvent (Figure 2.17). The effect of polymer architecture and common solvent on micelle morphologies of PEO-b-PS-b-PEO triblock copolymers was also reported. 81 To summarize, amphiphilic block copolymers have been researched extensively for micellization behavior. Most of the work done focuses on the spherical morphology. However, multiple morphologies for amphiphilic block copolymers were observed almost a decade back. Multiple morphologies have been obtained by primarily varying the molecular parameters and more specifically the block copolymer composition. 28

44 PS(240)-b-PEO(15) PS(240)-b-PEO(45) PS(240)-b-PEO(180) Figure 2.14 Effect of ion PEO content on the morphology of PS-b-PEO. Only recently solution parameters have been investigated to control the morphology and has been limited only to a few block copolymers. Controlling the morphology by solution parameters has some advantages over using the molecular parameters. These include simplifying material synthesis where one can use only one block copolymer composition and obtain different morphologies. Also this method enables more control to obtain exclusively one morphology at a time and provides greater insight into morphological transitions. However, controlling the morphology by solution parameters requires more stringent preparation conditions and care has to be taken to differentiate between kinetically trapped and thermodynamically stable morphologies, which we will discuss in detail in Chapter V. For PS-b-PEO, the spherical morphology has been characterized extensively. Multiple morphologies have been obtained by varying mostly the molecular parameters resultingin mixed morphologies. The effect of solution conditions on crew-cut micelle morphologies and their changes in PS-b-PEO, has not been carried out systematically and no morphological phase diagram in dilute solution exists for this system. 29

45 500 nm Figure 2.15 Tubules(a) and other complex bilayer aggregates (b)& (c) of PS 240 -b-peo 15 Figure 2.16 Vesicles with hollow rods for PS 100 -PEO 30 30

46 Figure 2.17 Hexagonally packed hollow loops for PS-b-PEO 31

47 CHAPTER III OBJECTIVES 1) To obtain multiple morphologies for amphiphilic block copolymer Polystyrene-b- Poly(ethylene oxide) by varying only the solution conditions and keeping molecular parameters fixed :PS(100k)-b-PEO(10k), PS962-b-PEO227, (f V PS ) ) To control the morphology by observing the morphogenic effect of: A. Common solvent/selective solvent ratio B. Copolymer concentration C. Selective solvent D. Temperature 3) To obtain morphological phase diagrams. 4) To conduct free energy calculations, design and conduct insightful experiments for understanding thermodynamic origins of morphological changes. 32

48 CHAPTER IV EXPERIMENTAL 4.1 Sample Synthesis The diblock copolymer PS 962 -b-peo 227 was synthesized using living anionic polymerization based on a standard route which is published elsewhere. 82 The synthesis was carried out by Dr. Joseph X. Zheng.In brief, the PS precursor was characterized by size exclusion chromatography (SEC) using polystyrene standards and had a number average molecular weight of 100k g/mol (the degree of polymerization of ~ 962) and a polydispersity of The number average molecular weight of PEO blocks was determined by proton nuclear magnetic resonance to be 10k g/mol (the degree of polymerization of ~ 227). A polydispersity of 1.04 in the overall copolymer was determined by SEC using the universal calibration. The volume fraction of PS blocks (f PS V ) was Preparation of micellar solutions We have used the approach of dissolving the copolymer in N,N-dimethylformamide (DMF) and then, adding water or acetonitrile, which are selective solvents for the PEO block. Therefore, two solvent systems (DMF/water and DMF/acetonitrile) for this diblock copolymer have been investigated. The PS 962 -b-peo 227 was first dissolved in 33

49 anhydrous DMF by stirring at room temperature for a few days to obtain stock solutions of different copolymer concentrations [0.1 to 8 (wt.) %], and they were then sealed with Teflon tape and stored at room temperature. Anhydrous DMF was filtered through a filter of 0.02 µm pore size before making the stock solutions. The method for preparing solutions was similar in both systems. The difference was only in how fast the selective solvent was added and how long the samples were equilibrated. In the DMF/water system, each ~ 10 ml of stock solution was taken in a vial, and deionized water was added drop-wise to reach a predetermined DMF/water ratio. The contents of the vial were stirred continuously to allow rapid mixing of DMF and water. Each drop added was ~ 0.1 (wt) % water of the total solution weight, and at least a 30-min gap was kept between adding consecutive drops. After the predetermined DMF/water ratio was reached, solutions were sealed and left to equilibrate for a period of time between 6-8 hrs and up to several days with mild stirring. This process was repeated to obtain micellar solutions with different DMF and water ratios. After equilibrating the solutions, they were left standing without stirring for at least 30 min before taking the samples for micelle morphological observations. In the DMF/acetonitrile system, 1 (wt.) % acetonitrile of the total solution weight was added every 10 mins. The samples were then sealed and equilibrated at least overnight before preparing the samples for morphological observations. For studying the effect of temperature, the micellar solutions prepared at room temperature, in vials were sealed with Teflon tape and kept at the desired temperature in a temperature-controlled oil bath with a digital temperature controller having an accuracy of ± 0.1 C. During heating and cooling, the micelle samples were equilibrated at the desired temperature for two to four hrs dependent upon the 34

50 equilibrating time needed. In order to study the morphological evolution in an isothermal condition where the morphological changes occur, the micelle samples were taken out at different periods of time from the isothermal temperature oil bath, and their morphologies were monitored under TEM. 4.3 Transmission electron microscopy: Micelle morphological observations were performed on a TEM (Philips TECNAI) with an accelerating voltage of 120 kv. In order to observe micelle morphologies under TEM yet retain the original morphological sizes and geometries as in the solution, a small amount of solution was quenched in excess water ( 100 times) to quickly vitrify the PS blocks into its glassy state and then, a drop from the quenched solution was placed on a carbon-coated grid. After a few min, the excess solution was blotted away with filter paper. The grids were dried at room temperature and atmospheric pressure for several hrs before examination in the TEM. Since the PS blocks were in their glassy state, the quenched morphologies were kept in TEM. 14 Alternatively, the quenched samples were also placed in dialysis tubes and dialyzed against distilled water for 4 days to remove DMF. Distilled water was changed twice a day. The dialyzed aqueous solutions were then used to prepare the sample for TEM observations. Using both methods to prepare the samples, identical results were obtained. 4.4 Static Light Scattering Experiments: In order to determine the critical micellization concentration (cmc), static light scattering experiments were conducted using a Brookhaven Instrument coupled with a BI-200SM goniometer, BI-9000AT correlator, and an EMI-9863 photomultiplier tube for photon counting. A Meller Griot 35 mw He-Ne laser was used as light source (632.8 nm). A 35

51 cylindrical glass scattering cell with a diameter of 12 mm was placed at the center of a thermostated bath (± 0.01 C) with decahydronaphthalene used for refractive index matching. The measurements were carried out at 90 scattering angle at 25 C. The glass scattering cells were extensively cleaned by ultrasonication in THF and ethanol to eliminate any dust and impurity. The stock solutions were filtered into the scattering cells through filters of 0.45 µm pore size. Water or acetonitrile was added drop-wise through filters of 0.02 µm pore size. For studying the effect of temperature, the samples were equilibrated at the desired temperature for 1 hr before taking measurements. 4.5 Turbidity measurements: Turbidity was measured by a Hewlett Packard HP 8453 UV-Visible Spectrophotometer with UV-Visible Chem Station software. The measurements were carried out at a wavelength of 700 nm where the absorption of the aggregates was minimized. The sample vials were sealed with Teflon tape and placed in a temperature cell in which the temperature was controlled by a temperature controller. DMF was used as the reference for all the measurements. The preparation of the solution was identical with those for light scattering. 4.6 Viscosity measurements: Viscometric experiments were performed on a Schott-Gerate viscometry system using an Ubbelohde capillary viscometer. The time taken for the solutions to flow through the capillary was recorded and used to calculate inherent viscosity of the solutions by comparing with the time of pure solvent to flow through the capillary. The viscometer was placed in a temperature-controlled bath at 25 ºC. 36

52 CHAPTER V MORPHOLOGICAL CHANGES WITH SOLVENT COMPOSITION In this chapter we have investigated the self-assembly behavior of amphiphilic diblock copolymer, polystyrene-block-poly(ethylene oxide) (PS-b-PEO), in N,Ndimethylformamide (DMF)/water and DMF/acetonitrile. In both cases water and acetonitrile are selective solvents for the PEO block. Micelle morphologies of the block copolymer in both systems could be controlled by varying copolymer and selective solvent concentrations. Figure 5.1 shows a set of TEM images of the micelle morphologies for 0.1 (wt.) % initial PS 962 -b-peo 227 concentration in DMF/water system. At a low water concentration of 2.75 (wt.) %, the micelles are spherical (Figure 5.1a). They change to mixed spheres and short cylinders (the length < 400 nm) at a water concentration of 3.14 (wt.) % (Figure 5.1b). At a water concentration of 3.54 (wt.) % as shown in Figure 5.1c, the mixed spheres and cylinders can still be observed. However, the length of the cylindrical micelles becomes increasingly long (several µms) with a decreased population of spheres. Sometimes, the pearl necklace morphology is also observed to coexist. At a water concentration of 4.17 (wt.) %, a morphological change occurs to form exclusively long worm-like micelles (Figure 5.1d). On further increasing the water concentration to 4.24 (wt.) %, mixed worm-like micelles and vesicles appear (Figure 5.1e). Finally, the morphology changes to exclusively vesicles at a water 37

53 concentration of 4.52 (wt.) % (Figure 5.1f). If water is added further the morphology does not change and we observed vesicles even when water is added up to 50 (wt.) %. We believe that polystyrene blocks are frozen at much lower water contents [~ 10 (wt.) %] due to its hydrophobic nature and thus the morphology does not change further. B 50 nm 100 nm 100 nm 100 nm 100 nm 100 nm Figure 5.1 Morphological transitions with adding water in the DMF/water system for initial copolymer concentration of 0.1 (wt.) %. (a) a pure sphere morphology at 2.75 (wt.) % water; (b) a mixed spheres and rods at 3.14 (wt.) % water; (c) a mixed spheres and long rods at 3.54 (wt.) % water; (d) a pure worm-like morphology at 4.17 (wt.) % water; (e) a mixed worm-like and vesicles at 4.24 (wt.) % water; and (f) a pure vesicle morphology at 4.52 (wt.) % water. From TEM observations, the distribution of wall thickness of the vesicles seems to become narrow at higher water contents; therefore, by varying the water concentration in a region between 2.75 and 4.52 (wt.) %, the micelle morphologies of PS 962 -b-peo 227 undergo controlled changes which are very sensitive to the small increment in water 38

54 concentration. In the system of PS 962 -b-peo 227 in DMF/acetonitrile for an initial copolymer concentration of 0.5 (wt.) %, Figure 5.2 shows a set of TEM images of the micelle morphologies. A B C 0.2 µm 0.2 µm 0.2 µm D E F 0.2 µm 0.2 µm 0.2 µm Figure 5.2 Morphological transitions with adding acetonitrile in the DMF/acetonitrile system for initial copolymer concentration of 0.5 (wt.) %. (a) a pure sphere morphology at 8 (wt.) % acetonitrile; (b) a mixed spheres and rods at 16 (wt.) % acetonitrile; (c) a mixed spheres and long rods at 30 (wt.) % acetonitrile; (d) a pure worm-like morphology at 37 (wt.) % acetonitrile; (e) a mixed worm-like and vesicles at 45 (wt.) % acetonitrile; and (f) a pure vesicle morphology at 70 (wt.) % acetonitrile. The PS 962 -b-peo 227 block copolymer first forms spheres at a low acetonitrile concentration of 8 (wt.) % (Figure 5.2a). The micelle morphology changes to mixed spheres and cylinders in the acetonitrile concentrations between 16 and 30 (wt.) % (Figures 5.2b and 5.2c), to worm-like cylinders at 37 (wt.) % (Figure 5.2d), then to mixed cylinders and vesicles at 45 (wt.) % (Figure 5.2e), and finally to vesicles at 70 (wt.) % 39

55 (Figure 5.2f). Therefore, the micelle morphology changes take place in a much broader region of the acetonitrile concentration between 8 and 70 (wt.) %On adding acetonitrile further, the morphology does not change and remains as vesicles. At these higher acetonitrile contents [>70 (wt.) %], we expect the mobility of PS blocks would be severely restricted, as acetonitrile is bad solvent for PS, and thus, the morphology may not change. In order to construct micelle morphological diagrams of these two systems, we need systematic investigations of these changes with respect to different initial copolymer and selective solvent concentrations. Light scattering experiments were utilized to determine the cmc for the diagrams. In the system of PS 962 -b-peo 227 in DMF/water, Figure 5.3 shows that the cmc for 0.1 (wt.) % initial copolymer concentration is at 2.5 (wt.) % of water, and for 0.5 (wt.) % initial copolymer concentration, the cmc is at 1.9 (wt.) % of water (the arrows in Figure 3). On the other hand, in the system of PS 962 -b- PEO 227 in DMF/acetonitrile, the light scattering experiments in Figure 5.4 show that for 0.1 (wt.) % initial copolymer concentration the cmc is at 13 (wt.) % of acetonitrile, and for 0.5 (wt.) % initial copolymer concentration, the cmc is at 2 (wt.) % of acetonitrile (the arrows in Figure 4). The arrows point out the cmc at different initial copolymer concentrations. On adding acetonitrile further to this solution, we observe that the scattered intensity shows further jumps at certain acetonitrile concentrations. The position of these jumps closely coincides with the acetonitrile concentrations where we observe morphological changes in TEM. Thus these jumps in the scattered intensity could be related to the morphological changes. 40

56 1x10 5 8x wt% 0.5 wt% 6x10 4 I (a.u.) 4x10 4 2x Water Content (wt.%) Figure 5.3 Static Light Scattering measurements with adding water in the DMF/water system. The arrows point out the cmc at different initial copolymer concentrations. Turbidity experiments revealed an overall range of selective solvent concentrations in which morphological changes take place. For example in the system of PS 962 -b-peo 227 in DMF/water, Figure 5.5 shows that turbidity increases sharply and then becomes constant in a short range of water concentration implying that the morphological changes occur in a very narrow range with water content. On the other hand, in the system of PS 962 -b-peo 227 in DMF/ acetonitrile, turbidity experiments in Figure 5.6 illustrate that the turbidity increases over a wide range of acetonitrile content implying that the morphological changes occur in a broad range of acetonitrile content. 41

57 (wt.) % 0.5 (wt.) % I (a.u.) Acetonitrile content (wt.%) 3500 Figure 5.4 Static Light Scattering measurements with adding acetonitrile in the DMF/acetonitrile system. The arrows point out the cmc at different initial copolymer concentrations. Figures 5.7 and 5.8 show two morphological diagrams for the DMF/water and DMF/acetonitrile systems, respectively, in a region of initial PS 962 -b-peo 227 concentrations from 0.1 to 8 (wt.) %. In this study, TEM is the primary technique utilized to construct these diagrams. The morphological boundaries in these two figures are drawn using lines between the data points of different micelle morphologies. In the case of the DMF/water system, the successive points are taken with small increments of the water concentration, since the morphology is very sensitive to the amount of water added 42

58 into this system. For the DMF/acetonitrile system, the increment of successive data points is relatively large in terms of the acetonitrile concentration. 3 1 wt.% 0.1 wt.% 2 Turbidity Water Content (wt.%) Figure 5.5 Turbidity measurements with adding water in the DMF/water system. Smaller increments are used only near the morphological change regions in order to obtain more accurate observations for determining the morphological boundaries. Note that in the DMF/water system the selective solvent concentration changes within several percentages to induce the micelle morphological changes from spheres to worm-like cylinders and then to vesicles; while, in the DMF/acetonitrile system, the selective 43

59 solvent changes within several ten percentages to complete the micelle morphological changes wt.% 1.6 Turbidity Acetonitrile Content (wt.%) Figure 5.6 Turbidity measurements with adding acetonitrile in the DMF/acetonitrile system. Yet both systems possess the identical pathway of morphological changes with increasing the selective solvent concentrations (Figures 5.7 and 5.8). The morphological changes in both systems also shift to lower selective solvent concentrations as the initial PS 962 -b- PEO 227 concentration increases. The question is: what is the reason for this vast difference in selective solvent concentrations between these two systems where the morphological changes occur. 44

60 Actual Polymer Concentration (wt.%) spheres spheres + cylinders cylinders cylinders + vesicles vesicles Water Content (wt.%) 10 1 Figure 5.7 Morphological change diagram for PS 962 -b-peo 227 in the DMF/water system. As the selective solvent is added, the solvent becomes progressively poorer for the micelle core formed by the PS blocks. Quantitatively, the poorness of the selective solvent with respect to the PS block is determined by the PS-solvent interaction parameter, if we neglect the entropic contribution. These parameters can be estimated using the van Laar-Hildebrand equation of 83 χ P-S = [V S /(RT)] (δ P - δ S ) 2 (1) where V S is the molar volume of the solvent and δ P and δ S are solubility parameters of the polymer and selective solvent, respectively. Since the solubility parameter of PS is 9.04 (cal/cm 3 ) 0.5, while that of water is 23.4 (cal/cm 3 ) 0.5, and acetonitrile is 11.9 (cal/cm 3 ) 0.5,37 45

61 Actual Polymer Concentration (wt.%) spheres spheres + cylinders cylinders cylinders + vesicles vesicles Acetonitrile Content (wt.%) Figure 5.8 Morphological change diagram for PS 962 -b-peo 227 in the DMF/acetonitrile system. the χ P-S values for both systems can be calculated as χ PS-water = 6.27 and χ PS-acetonitrile = The χ PS-water value is close to nine times higher than the χ PS-acetonitrile value, indicating that water is a much poorer solvent for PS as compared to acetonitrile. Adding a small amount of water can result in a substantial increase in the interfacial free energy to initiate the micellization. This is equivalent to adding a large amount of acetonitrile. The increase in the interfacial free energy leads to an increase in the micelle size and thus, an increase of the stretching of the PS blocks in the core. This ultimately gives rise 46

62 to the morphological changes. The breadths of morphological changes and the regions of stabilized morphologies can thus be tuned by choosing appropriate selective solvents. An attempt was made to make a reduced plot for both morphological diagrams without success. The reason may be due to the solvent interaction difference between DMF/water and DMF/acetonitrile which has not been taken into consideration. 5.1 Molecular mechanism of morphological Changes Based on the TEM observations we can hypothesize the molecular mechanism of morphological changes and structure of the various micelle morphologies. Figure 5.9a shows a representation of the sphere morphology formed at low selective solvent concentrations. The grey region corresponds to the PS core while the individual chains correspond to the PEO corona. Note that in these representations the core and corona are shown of similar size to simplify representation, while actually the PS core is much bigger than the PEO corona. The spheres can be present as separated spheres or interconnected in which the corona chains interact as shown in Figure 5.9 b. Note that at this stage even though the corona of the spheres interact, the cores are distinct for the spheres. PEO chains (Corona) Polystyrene Core Figure 5.9a Representation of spherical morphology observed at low selective solvent concentration. 47

63 Figure 5.9b Representation of interacting spheres morphology observed at low selective solvent concentrations. On increasing selective solvent concentration, short rods are formed. The interacting spheres can lead to the formation short rods such and there is no distinction between their cores (Figure 5.9c). Figure 5.9c Representation of short rod morphology. As more selective solvent is added more and more spheres can add to these small rods, thus increasing the length of the rods and decreasing the relative population of the spheres. At a certain selective solvent concentration exclusively worm like micelles are formed which have spherical end caps as shown in Figure 5.9d. (Actual length would be much bigger) 48

64 Figure 5.9d Representation of long rod /worm like morphology with spherical end caps. In case of DMF/water, adding water further leads to the formation of junctions which we will discuss in detail in Chapter VIII. A representation of these junctions will be discussed in that chapter. Adding more selective solvent leads to formation of lamella and vesicles, a representation of lamella is shown in Figure 5.9e.The lamella are bilayers with two layers of the copolymer chains. These lamellae close to form vesicles Figure 5.9e Representation of lamella. 5.2 Thermodynamic and Kinetic Aspects of the Morphological Changes. The criteria for achieving stable morphologies are that the obtained morphologies must be stable over a long period of time and the morphological changes due to varying the selective solvent concentration must be reversible. 50 The question is whether the 49

65 morphological changes in Figures 5.7 and 5.8 are the thermodynamic phase diagrams. In other words, does the rate of adding selective solvent affect the morphological changes? It is known that the evolution towards equilibrium micelle morphologies depends upon mobility of the core forming blocks, which are associated with the solubility of the amphiphilic diblock copolymer in the solvent. 15 In general, our experience is that the rate of adding selective solvents must be slow enough and amount of selective solvent added each time must be small enough in order to obtain the stable micelle morphologies. The micelle morphologies observed in Figures 5.7 and 5.8 are stable over months as long as the systems are kept closed. In order to further prove that the micelle morphologies obtained are truly in thermodynamic equilibrium, we designed two sets of experiments. First, we chose to add DMF into a DMF/acetonitrile system with an initial PS 962 -b-peo 227 concentration of 0.6 (wt.) % and an acetonitrile concentration of 68 (wt.) %. Figure 5.10 shows a set of TEM images of the micelle morphological changes after slowly adding different amounts of DMF. Figure 5.10a is the initial vesicle morphology of the system. Adding DMF to this system leads to the formation of mixed worm-like cylinders and vesicles at 50 (wt.) % acetonitrile (Figure 5.10b), worm-like cylinders at 37 (wt.) % acetonitrile (Figure 5.10c), mixed spheres and cylinders at 25 (wt.) % acetonitrile (Figure 5.10d), and spheres at 13 (wt.) % of acetonitrile (Figure 5.10e). In this experiment, the rate of adding DMF is critical. When the DMF concentration is initially low (the acetonitrile concentration is high), the mobility of PS block is severely restricted and thus, DMF has to be added slowly [1 (wt.) % increment of the total solution weight per 10 min]. When the DMF concentration becomes gradually high, DMF can be added 50

66 at a faster rate [1 (wt.) % increment of the total solution weight per 1 min]. Figure 5.10 thus reveals that the micelle morphological changes are reversible. A B C 0.2 µm 0.2 µm D E 0.2 µm 100 nm 100 nm Figure 5.10 Reversible morphological changes by adding DMF into the DMF/acetonitrile system with 0.6 (wt.) % initial copolymer concentration and 68 (wt.) % acetonitrile. (a) the original micelle morphology is pure vesicle; (b) mixed worm-like and vesicles at 50 (wt.)% of acetonitrile; (c) a pure worm-like morphology at 37 (wt.) % acetonitrile; (d) mixed spheres and rods at 25 (wt.) % acetonitrile, and (e) a pure sphere morphology at 7 (wt.) % acetonitrile. Second, we examine the thermodynamic stability of the morphologies by adding the selective solvent quickly to generate kinetically quenched morphologies. We then carried out annealing experiments to monitor whether these kinetically generated morphologies return to the thermodynamically stable morphology. We started with a system which had 1 (wt.) % initial PS 962 -b-peo 227 concentration in pure DMF, the 51

67 selective solvent acetonitrile was added at a fast rate [1 (wt.) % increment of the total solution weight per 1 min] to reach an acetonitrile concentration of 35 (wt.) %. The quenched morphology formed is mixed spheres and cylinders as shown in Figure 5.11a. Note that based on Figure 5.8, this system should display a stable morphology of worm-like cylinders. The system was then annealed at room temperature for 1 month; the mixed morphology gradually changes to worm-like cylinders as shown in Figure 5.11b. Therefore, the worm-like cylinders should be the thermodynamically stable morphology. A B 0.2 µm 0.2 µm Figure 5.11 Morphologies formed by a fast rate of acetonitrile addition to a 1 (wt.) % initial copolymer solution in DMF. Acetonitrile was added to 35 (wt.) % (a) after one day; (b) after two months. In another experiment for a system with 2 (wt.) % of the initial PS 962 -b-peo 227 concentration in pure DMF, the selective solvent of acetonitrile was added at a fast rate [1 (wt.) % increment of the total solution weight per 1 min] to an acetonitrile concentration of 68 (wt.) %. Quenched morphology of mixed spheres and vesicles is observed as shown in Figure 5.12a. Note that based on Figure 5.8, this system should display a morphology of pure vesicles. We annealed the system for two months at room temperature; the mixed morphologies can be observed to be in the process of returning to 52

68 vesicles as shown in Figure 5.12b. The annealing time of two months is apparently not long enough to completely reach the stable vesicles. A B 0.2 µm 0.2 µm Figure 5.12 Morphologies formed by a fast rate of acetonitrile addition to a 2 (wt.) % initial copolymer solution in DMF. Acetonitrile was added to 68(wt.)% (a) after1 day; (b) after two months. We can thus conclude that although a fast rate of adding selective solvent may form kinetically quenched micelle morphologies (in most cases, they are mixed morphologies), the thermodynamically stable morphology can always be recovered as long as the PS blocks are not in their glassy state and the system is annealed for sufficient time. The appropriate equilibrium time required to recover the thermodynamically stable morphologies is dependent on the chain dynamics, which is in turn dependent on the molecular weight of the chains and the solvent composition 50. At high acetonitrile content the chain dynamics would be slow as compared to low acetonitrile content, as acetonitrile is a bad solvent for polystyrene, and the friction forces would also increase as the micelle core would be less swollen 84. Thus, the annealing time required to recover the thermodynamically stable morphologies is dependent on the concentration of acetonitrile in our system. 53

69 In case of DMF/water fast rate of water addition also produced kinetically trapped morphologies and no morphological transitions can be induced if the rate of water addition is fast. Figure 5.13 shows mixture of spheres and vesicles which is formed when rate of water addition is almost 10 times faster. Also, toroid or doughnut shaped morphology is observed sometimes as seen if Figure 5.14 In an extreme case, if water is added almost continuously to a DMF solution, complex bilayer aggregates are obtained as kinetically trapped morphologies. These were found not to change even after several months, and settled in the solution. These are shown in Figure Figure 5.13 Kinetically trapped morphologies formed by a fast rate of water addition to a 1 (wt.) % initial copolymer solution in DMF, water was added upto 3% wt.%. The self-assembled micelles spheres, rods and vesicles are expected to be three dimensional as opposed to flat discs or ribbons. In order to confirm this we shadowed the micelles with Platinum and observed in TEM. Figure 5.16 shows the TEM images of 54

70 shadowed spheres, worm-like micelles and vesicles and clearly indicates that these aggregates are three-dimensional. Figure 5.14 Doughnut or toroid morphology formed by a fast rate of water addition to a 1 (wt.) % initial copolymer solution in DMF. We also changed the common solvent to THF and we observed that porous spheres were formed in case of THF/water (Figure 5.17). However in case of THF/acetonitrile spheres, rods and vesicles were formed similar to the earlier observations. This indicates complex interactions between the solvents which seem to affect the morphology would be interesting to study the morphological changes for mixed common solvent and mixed selective solvent systems. For example use a mixture of DMF and THF as a common solvent or use of a mixture of water and acetonitrile as a mixed selective solvent. Using the later system can lead to observation of other morphologies which cannot be observed in case of either water or acetonitrile. This is because, in case of water the χ polymer-solvent and other morphologies may not be thermodynamically accessible While in case of acetonitrile χ polymer-solvent increases slowly so that, a lot of selective solvent needs to be added leading to kinetic barriers for accessing other morphology. 55

71 Figure 5.15 Complex bilayer aggregates formed by continuous addition of water to DMF. 56

72 A B C D Figure 5.16 TEM images after shadowing with Platinum (a) spheres (b) a mixture of spheres and vesicles (c) & (d) worm like micelles. Figure 5.17 Porous spheres formed on adding 12wt% water to a 1.5wt% copolymer solution in THF. 57

73 CHAPTER VI MICELLE FREE ENERGY CALCULATIONS In order to further understand the reasons for the appearance of micelle morphological changes from spheres to worm-like cylinders, to vesicles in different selective solvent and initial PS 962 -b-peo 227 concentrations, free energies in each of these micelle morphologies must be estimated. In these micelles, the PS blocks form the core and the PEO blocks form the corona. 85 We have utilized the size and geometry data which is obtained from the TEM images as the first approximation (the precise calculation needs the size and geometry of micelles in solution). The first parameter which needs to be calculated is the degree of stretching of PS blocks in the core (S c ) based on the equation of 46 S = R / (6.1) c R o where R is the radius of the PS core in the spheres or in the worm-like cylinders. In the case of vesicles, R is half of the wall thickness. The quantity R o is the unperturbed end-toend distance of a PS chain which can be calculated from the following equation 78 R o =0.067M 0.5 (6.2) where M is the number average molecular weight of the PS block. We approximate the observed micelle radius to be the radius of the PS cores, since in our case the PS block composition is much higher as compared to PEO composition ( f PS V is 0.914). 58

74 Furthermore, the PEO coronas are collapsed during the preparation of the samples for the TEM observations. The second parameter is the interfacial area per chain (s). It can be calculated based on the following equations 46 s sphere = 3V S N PS /(f R core ) s cylinder = 2V S N PS /(f R core ) (6.3a) (6.3b) and s lamella = V S N PS /(f R lamella ) (6.3c) where f is the volume fraction of PS blocks in the core, V S is the volume of PS monomer, and N PS is the degree of polymerization of the PS block. Note that we use R lamella (half the wall thickness) to approximate the calculation in the case of vesicles. If we assume sufficiently dense PS blocks in the core, f approaches to unity. The value of f would be closer to unity when the selective solvent concentrations are high as compared to when selective solvent concentrations are low. The observed dimensions and the calculated parameters based on eqs for both DMF/water and DMF/acetonitrile systems for 0.5 (wt.) % initial copolymer concentrations are listed in Table 6.1 at three representative selective solvent concentrations (where the pure micelle morphologies appear). High magnification TEM images and ImageJ software were used to determine the sizes of the aggregates. As the morphology changes from spheres to worm-like cylinders, to vesicles, both the S c and s values decrease. Since the corona can be considered as tethered chains on a convex or quasi-planar substrate, the strength of corona repulsion can also be evaluated by the reduced tethering density, ~ 86, 87,88 σ, which is defined as 59

75 ~ σ = π R 2 g /s (6.4) where s is the interfacial area per chain and R g is the radius of gyration of the tethered PEO chains in the specific solution. The R g value in this study is calculated using a scaling law reported in water at room temperature by Devanand and Selser. 89 This scaling law also predicted a correct R g value of PEO for a particular molecular weight in DMF. 90 We thus used this scaling law to approximate the R g values for PEO in good solvents. The ~ σ values listed in Table 6.1 increase from 3 to 7.2 as the morphology changes. This implies that the coronas in this study start approaching the boundary between the non-interaction and crossover regimes in spheres, and pass the onset of the chain overcrowding to enter the crossover regime to generate some repulsion in vesicles. They are far away from the highly stretched brush regime. 86,87,88 The total free energy of one chain in a micelle is F = F core + F interface + F corona (6.5) Let us consider each of these components individually. The term F core is the elastic free energy of the core and can be estimated by the degree of stretching S c in the core 51 F core /kt = k j S c 2 (6.6a) Table 6.1: Micelle size and calculated parameters for a 0.5 (wt.) % initial copolymer concentration at three selective solvent concentrations Morphology Solvent concentration (wt.) % Diameter or wall thickness (nm) S c (stretching) s (area per chain, nm 2 ) Reduced Tethering density Water AN* Water AN* Water AN* Water AN* Water AN* Spheres ±4.0 Cylinder ±2.2 Lamella ±3.3 *AN stands for acetonitrile 42.5 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±0.4

76 The coefficient k j has been calculated for the case of dense cores: k sphere = 3π 2 /80, k cylinder = π 2 /16, and for the case of lamellae k lamella = π 2 /8. 52 As shown in Table 1, S < 1 in some cases. This may be due to two reasons. First, during the preparation of TEM samples using excess water the micelles could deswell, and reduce the degree of stretching of the PS blocks; and/or secondly, there might be a possibility that the chains could actually be compressed, especially when the solvent quality becomes increasingly poor for the cores. In a recent study by SAXS and cryo-tem for PS-b-PI block copolymer in solution, the value of S c was found to be less than one in the case of vesicles. 55 Since Eq. 6.6a would predict a decrease in free energy even if the compression increases, we need to modify this equation so that it predicts an increase in the free energy when compression increases. If we define the degree of compression as 1/S c, and assume that the coefficient k j is identical in compression as in stretching, we can modify eq. 6.6a as c F core /kt = k j (1/S c ) 2 (6.6b) Therefore, eq. 6.5a is applicable when the core is stretched, and eq. 6.6b is applicable when core is compressed in our calculations. The term F interface is related to the interfacial energy between the core (PS) blocks at the interface and the solvent. Therefore, 51 F interface = γs (6.7a) where s is the interfacial are per chain, which we have already determined, and γ is the surface tension, which is related to χ core-solvent by the following expression 3,52,53 γ = (kt/a 2 )(χ core-solvent /6) 0.5 (6.7b) 61

77 where a is the PS monomer length. This expression originally derived for two homopolymers, 52 has been regularly utilized in micellization theories. 3,53 The term χ coresolvent for different solvent concentrations can be estimated using the solubility parameters 83 and respective volume fractions. The term F corona is based on the expression recently proposed by Zhulina et al. in their theory of diblock copolymer micelles. 51 When the corona is treated as chains tethered on a quasi-planar substrate, the upper-limit value of the F corona values for these micelle morphologies can be provided, since all of these morphologies possess convex surface for the corona chains to be tethered onto and thus, the chain stretching must be reduced in the real cases. This free energy expression is expressed by F corona / kt ^ ^ = C C F N ( sa H PEO ) 2 1 / 2ν (6.8a) where N PEO is the degree of polymerization of the PEO block, a is the PEO monomer length, and ν is the scaling exponent, which is equal to 3/5 for a good solvent. The ^ C and ^ H C F values are numerical pre-factors, and in the case of a good solvent, they are C ^ H = C 1/ 3 1/ 3 H ( l / a) υ ^ C F = C F (6.8b) (6.8c) where C H and C F are model parameters which were calculated to be 0.68 and 1.38 respectively, 51 l is the kuhn length, a is the PEO monomer length, and υ is the excluded volume parameter of PEO in the specific solvent. The values of l and a are 1.1 nm and 0.35 nm, respectively. 91,92 The excluded volume parameter υ is related to the second 91 virial coefficient A 2 62

78 υ = 2A 2 M 0 2 / (a 3 N A ) (9d) where M 0 is the molecular weight of the PEO monomer, and N A is the Avogadro s number. The values of A 2 in literature for our specific molecular weight and/or scaling laws have enabled us to determine the values of A 2 for PEO in DMF, in water and in 89,,,95,, acetonitrile To calculate the values at different concentrations of the systems, we have further used the individual A 2 values and the volume fractions of the respective solvents. Using eqs. 6.8a 6.8d, the free energy of the corona can therefore be calculated. The calculated individual free energy components and the total free energy for a 0.5 (wt.) % initial copolymer concentration at representative solvent concentrations (given in Table 6.1) are listed in Table 6.2. The total free energy decreases as the morphology changes from spheres to worm-like cylinders to vesicles. The F core /kt increases from spheres to worm-like cylinders and then to vesicles as the chains are compressed. When χ core-solvent increases, the F interface /kt decreases as the micelle size becomes increasingly large. The F corona /kt increases as the reduced tethering density increases leading to increased chain repulsions. However, both of the F core and the F corona values are much smaller compared to the F interface, which is dominant. Table 6.2: Calculated free energy components and total micelle free energies at three selective solvent concentrations Morphology F core /kt F interface /kt F corona /kt F total /kt Water AN* Water AN* Water AN* Water AN* Spheres ±4 61.6±3.82 ±0.05 ±0.02 ±3.8 ±3.5 ±0.15 ±0.12 Cylinder ±3 47.3±3.79 ±0.1 ±0.1 ±2.8 ±3.5 ±0.1 ±0.25 Lamella 3.1 ± ± ± ± ± ±0.3 40±4 32±1.5 *AN stands for acetonitrile 63

79 In the above calculations, we have not considered the rim energy of the lamella as compared to closed vesicles. This rim energy is E rim = 2πRγ, where R is the radius of the circular lamella and γ is the surface tension. For our system in the concentration region of forming vesicles, γ 2kT based on eq. 6.7b; therefore, for a circular lamella with a radius of 200 nm, E rim 800πkT. This huge free energy increase can be avoided if the lamella closes to form vesicles. However, in bending the lamella to form a vesicle, a bending energy E bend needs to be considered. In spherical vesicles the bending energy is independent of the size of the vesicle and is equal to E bend = 8πk where is the bending c k c modulus. 97,98 A typical value for block copolymer bilayers is k c 40kT so that E bend 320πkT, a total energy required to bend a lamella to form a spherical vesicle. Quantitatively, for a circular lamella having a radius of 200 nm, the number of chains [the aggregation number (N agg )] can be calculated by 46 N agg = fπr 2 t/n PS V PS (6.9) where R is the radius of the lamella (200 nm), t is the wall thickness (~ 30 nm in our vesicles),v S is the volume of the PS repeating unit, and N PS is the degree of polymerization of the PS block. If we assume a sufficiently dense PS core and f is close to unity, the N agg is calculated to be Therefore, for each chain E bend 320πkT/23000 = 0.04kT. Similarly, the rim energy E rim per chain can also be calculated to be 0.1kT. The difference between these two values would decide the equilibrium appearance of vesicles. In block copolymer vesicles, however, another possibility is that significant corona chain segregation may occur: the long chains tend to be on the outside and the 64

80 short ones on the inside. Since steric repulsion between long chains is stronger than between the short ones, a natural curvature may form to stabilize the vesicle. Thus, the maintenance of closure may not require a bending modulus to be part of the thermodynamics. 99, 100, 101 Based on the free energy calculations, in principle, we can estimate the change in free energy for different micelle morphologies with increasing the selective solvent concentration. Figure 6.1 shows the change in total free energy (F/kT) versus acetonitrile concentration for spheres, worm-like cylinders, and vesicles in the DMF/acetonitrile system, which is only a first approximation. There are two broad concentration regions exhibiting the mixed micelle morphologies. Our calculations only partially enter these two concentration regions, since in these two regions, the micelle size and geometry cannot be precisely defined. This is particularly in the case of short cylinder micelles. It is speculated that if the free energies are going to cross over, they must be at around 21 (wt.) % between spheres and worm-like cylinders and at around 50 (wt.) % between cylinders and vesicles. According to the free energy changes with acetonitrile concentrations in Figure 6.1, we should observe morphological changes as first-order-like transitions from one type of micelles to another. Namely, only one morphology in its own concentration region, which possesses the lowest free energy, can be observed. Any other morphology with respect to this one is metastable. 102 Note that between two single morphologies there must be mixed morphologies in a narrow concentration region. 103 However, we have observed broad concentration regions exhibiting mixed morphologies. Several possibilities may exist. One is that the free energy levels of neighboring micelle morphologies are close to each other, although they do crossover; another is that the 65

81 regions of mixed morphology may represent thermodynamic equilibrium. 103 Further experiments on micellization and their morphological changes at different temperatures have been designed and are conducted to judge the explanations. They will be discussed in Chapter IX F/kT spheres cylinders vesicles Acetonitrile Content (wt)% Figure 6.1 Free energy changes with acetonitrile content for various morphologies in the copolymer-dmf/acetonitrile system. 66

82 CHAPTER VII EQUILBRIUM SIZES OF STRUCTURES We saw in Chapter VI that the total free energy of the micelles (F total ) is dominated by the free energy of the interface (F interface ).This dominance was found to apply in all the three micelle morphologies of spheres, rods and vesicles and in both solvent systems. The dominance of F total by F interface makes it important to perform an appropriate quantitative analysis of the equilibrium size of the structures. This is because changes in the size of the structures would strongly affect the F interface and thus the F total. Thus it is necessary to determine the equilibrium micelle size (R) or the value of R which would minimize the total free energy. 7.1 The form for F total For a particular set of conditions, a specified solvent, block copolymer and structure (morphology), the dependence of free energy on R can be written as F kt total F kt F F kt core int erface corona b1 b 2 b 3 = + + = a 1 R + a 2 R + a 3 R (7.1) kt The coefficients b 1,b 2, b 3 can be determined for the theoretical expressions described earlier in Chapter VI. Then the coefficients a 1,a 2 and a 3 can be determined for conditions in Table 6.1 by using the observed R values and the calculate free energies in Table

83 7.2 Exponents For F cor e, from Eq. 6.6a and Eq. 6.1 F kt R core 2 = k j S c = k j = R o 2 a 1 R 2 (7.2) F core is the proportional to R 2. The system dependent proportionality constant is a 1 and b 1 =2. For F interface, from Eq. 6.7a and 6.7b and Eq 6.3a, 6.3b and 6.3c F 0.5 int erface 2 kt 1 χ core solent xvs N PS a = γ s = = 2 (7.3) a 6 fr R For sphere: x= 3, R=R core For cylinder: x = 2, R= R core For lamella: x=1, R =R lamella F interface is proportional to R -1 and the system dependent proportionality constant is a 2 and b 2 =-1 For F corona, from Eq. 6.8a, 6.8b and 6.8c and Eqs 9d, and Eqs. 6.3a, 6.3b and 6.3c F corona kt ^ = C H ^ C F N PEO s a 2 1/ 2ν = C H l a 1/ 3 υ 1/ 3 C F N PEO 2 a fr xvs N PS 1/ 2ν = a R 3 1/ 2ν (7.4) F corona is proportional to R 1/2ν where ν=3/5 in good solvent. The system dependent proportionality constant is a 3 and b 3 = 5/6. The complete form of the free energy expression is F kt = a / 6 1R + a2r + a3r (7.5) 68

84 7.3 Coefficients The values of a 1, a 2, a 3 for specific system studied can be determined from the sizes (R values) in Table 6.1 and the values of individual free energy components in Table 6.2 using Eqs. 7.2, 7.3 and 7.4. For R (nm) and ignoring the uncertainties, the results are given in Table 7.1. Now we can calculate the dependence of F/kT on the size of the structures, for different morphologies for different solvent systems. The equilibrium structures specified by the free energy expression occur at minima of F/kT vs R. We can plot this dependence for the different morphologies. Figure 7.1, Figure 7.2 and Figure 7.3 shows the total free energy change of spheres, cylinders and lamella in DMF/water with Table 7.1: Calculated coefficients for the free energy dependence on size of structures at selective solvent concentrations given in Table 6.1. Morpohlogy Solvent a1 a2 a3 Water Spheres AN Water Cylinder AN Water Lamella AN changes in micelle radius R. The arrows in the Figures indicate the size of micelles R observed in TEM. The minimum free energy for spheres in DMF/water is at R~75nm while the observed micelle radius for the spheres is R~24.5nm. Accordingly the minimum free energy at R~75nm is lower than the free energy calculated for the observed spherical micelles in Table 6.2. These differences in the R min and minimum free energy are present for all three morphologies in both DMF/water and DMF/acetonitrile system. However these differences are biggest in case of spheres and less for cylinders and least for the lamella morphologies. 69

85 F/kT R(nm) Figure 7.1 Free energy changes with micelle size R for spheres in DMF/water system. F/kT R(nm) Figure 7.2 Free energy changes with micelle size R for cylinders in DMF/water system F/kT R(nm) Figure 7.3 Free energy changes with micelle size R for lamellae in DMF/water system 70

86 There could be several reasons for these differences in the observed micelle size and the micelle size (R) where the free energy is minimum:- 1) The PS core of the micelles present in the solution is swollen by the solvent. As the water content in the solvent mixture increases the solvent quality becomes bad for PS core and it would be less swollen. Thus the micelles would be most swollen in case of spheres and least swollen in case of vesicles/lamella. During the preparation of samples for TEM, the micelles would deswell. Thus the size of the micelle core observed in TEM would be less than actual size of the core in the solution and this could contribute to the differences in the observed micelle size in TEM and the R min ( R for minimum free energy). 2) The free energy expression used in Chapter VI and the individual free energy components might be an overestimation for the free energy of the interface and/or underestimation of the free energy of the core and corona. 3) Assuming that the free energy expression is accurate, there is a third possibility, that the morphologies observed in this research are long lived metastable morphologies. We have performed extensive experiments to check the thermodynamic stability of these including annealing and reversibility experiments. From an experimental point of view the morphologies are the micelles are stable for over several months, but they could still be metastable morphologies which are kinetically trapped and cannot evolve to the equilibrium size in the time frame of this research (even several years). All the above three factors could contribute in part or fully to the disparity in the observed micelle sizes and the calculate Rmin. 71

87 CHAPTER VIII PHASE MORPHOLOGY OF SELF-ASSEMBLED WORMS In chapter V, we discussed the self-assembled micelle morphologies of PS 962 -b-peo 227 in DMF/water and DMF/acetonitrile mixtures. With increasing the concentration of the selective solvent (water in the DMF/water or the acetonitrile in the DMF/acetonitrile system), the micelle morphology observed in TEM changed from spheres to worm-like cylinders and then, to vesicles. In this chapter, let us focus on the concentrations where the worm-like cylinders are formed in PS 962 -b-peo 227 in DMF/water system. We observed that on increasing the copolymer concentration a worm-network was formed 104. Figure 8.1 shows the worm network formed for a 4 (wt.)% initial copolymer concentration and 3.9 (wt.)% water in a DMF/water system. The network consists of self-assembled worm-like cylinders interconnected by the Y- and T- shaped junctions (Figure 8.1a). The worm-like cylinders are highly entangled as shown in Figure 8.1b. The network can form in a macroscopic size (Figure 8.1c) and we also observe multiple junctions (Figure 8.1d). Such a network structure can be considered as counterpart of the bicontinuous structures formed in the bulk. 105 This network is stable, and does not change for several months if no additional water is allowed to add into the system. It has been found that in the DMF/water system the formation of network of PS 962 -b-peo 227 as shown in Figure

88 requires a critical initial copolymer concentration, and this concentration is determined to be 0.4 (wt.)%. Below this critical concentration, no network appears although some Figure 8.1 Worm network formed in 4 (wt.) % copolymer concentration and 3.9 (wt.) % water content in DMF/water mixture(a) Y and T shaped junctions (b) entangled network (c) microscopic size (d) multiple junctions. branching is visible in the worm-like cylinders. This concentration dependence is in accordance with some theoretical predictions 106,107 and experimental observation 108 for network formation in small molecule surfactants. It is observed that the network formation is accompanied by an abrupt increase in the macroscopic viscosity of the 73

89 system.to quantify the magnitude of this increase, capillary viscometer measurements were conducted. The inherent viscosity (η ih ) is calculated based on an equation of η ih = [ln (t / t 0 )] / C (1) where t is the time taken by the copolymer solution to flow through the capillary, t 0 is the time taken by pure DMF to flow through the capillary, and C represents the copolymer solution concentration in g/dl. 109 Figure 8.2 shows the change of η ih values with adding water to a 4 (wt.)% initial copolymer concentration in DMF. As we saw in Chapter V adding water induces micellization and morphological changes from spheres to worm-like cylinders and then to vesicles. The viscosity in this figure does not show a substantial increase within the phase morphology of mixed spheres and worm-like cylinders at relatively low water concentrations. However, when the water (which is the selective solvent of the PEO block) concentration reaches 3.9 (wt.) %, it undergoes the micelle morphological change to the worm-like cylinder network, the η ih value exhibits a sudden increase of over four times. Further increasing the water concentration leads to enter the vesicle morphology, where no network is formed. The η ih value sharply reduces and returns to almost the original value due to this morphological change. However, disregarding an extensive experimental effort, such a worm-like cylinder network is not found in the DMF/acetonitrile system even at high copolymer concentrations [up to 8 (wt.) %]. On increasing the copolymer concentration, the worm-like cylinders tend to align parallel to each other rather to be interconnected as shown in Figure 8.3. Only a mild and gradual increase in η ih value is observed at the onset of the micelle morphological changes from the mixed sphere and rod-like cylinders to the worm-like cylinders. Quantitative 74

90 measurements of the viscosity values in different micelle morphologies are shown in Figure 8.4. Figure 8.2 Change in inherent viscosity on adding water in a 4 (wt.) % initial copolymer concentration solution in DMF. If the worm-like cylinder micelles are sufficiently long and flexible, they can behave like polymers. They can get entangled and show viscoelasticity 68. A network of worm like cylinders which is both entangled and interconnected, can increase the viscosity by two mechanisms. The mechanism of increase in viscosity due to entanglements can be compared to the increase in polymer viscosity due to entanglements, with some limitations 68. The interconnections (the Y- and T- junctions) increase the viscosity by introducing crosslinks in the worm-like cylinders and this can be compared to increase 75

91 Figure 8.3 Aligned cylinders at a 2 (wt.) % initial copolymer concentration and a 34 (wt.)% acetonitrile in the DMF/acetonitrile system. in polymer viscosity due to chemical crosslinks. However, since the crosslinks in the worm network are not chemical it would be interesting to study its viscoelastic behavior. Also, the viscosity of worm-like cylindrical micelle solution increases with increasing the worm length 110. A worm network can be considered as worm-like cylinder with an infinite length as opposed to individual worm-like cylinder micelles, which would have a finite length. Thus, for an entangled and interconnected worm network (Figure8.1) in case of DMF/water we observe much more increase in the viscosity than in case of worm- like cylinders in DMF/acetonitrile (Figure 8.3). Such a network has been observed for small molecule surfactants, and it is found to be dependant upon concentrations and temperatures. 108 The network formation has also been treated theoretically by Safran and coworkers 111. In small molecule surfactants and micro-emulsions, an important concept is the spontaneous curvature c 0 which is the preferred curvature of the amphipihlic monolayer towards water or oil 111 The c 0 value decreases with increasing temperature 76

92 and causes the morphological changes from spheres to cylinders that interconnect via junctions. 108 An average curvature (H) can be defined as 112 Figure 8.4 Change in inherent viscosity on adding acetonitrile in a 4 (wt.) % initial copolymer concentration solution in DMF. H = 1/ R 1 + 1/ R ) / 2 (2) ( 2 where R 1 and R 2 are the radii of curvatures in two perpendicular directions in a symmetric geometric object. If we assign positive and negative signs to R 1 and R 2, H can be referred as the net curvature. In a case of a sphere, R 1 = R 2 = R, and H = 1/R. A cylinder possesses R 1 = R and R 2 =, and H = 1/(2R). When H = 0 due to R 1 = -R 2, it illustrates a planer bilayer or a saddle shaped surface. 77

93 In the case of block copolymers, the formation of Y junctions results in saddle points with negative curvature and thus the net curvature of the copolymer network is reduced. 14 A representation of the Y junction is shown in Figure 8.5 where R is the negative curvature. Therefore, as the Y- junctions populate, the net curvature would decrease and finally, lead to formation of a network. The network is also considered to have favorable configurational entropy due to the existence of several possible PEO chains R PS core Figure 8.5 Representation of the Y junction. configurations of the network. 113 These different configurations can be in terms of different branch lengths and branch length distributions which can exist in a network. However, there is also a loss in entropy as the free ends of the worm-like cylinders are constrained to meet at a junction. In the worm-like cylinder network, free ends of the worm-like cylinders (also referred as spherical end caps 13,108 ) are largely eliminated due 78

94 to the formation of interconnections. It has been known that both the interfacial area per chain and degree of stretching of the core blocks are higher for the spherical geometry than those in the cylindrical geometry. 46 Therefore, existence of the spherical end caps are thermodynamically unfavorable. The end capping energy has been found to be dependent upon the curvature, and it decreases as the spontaneous curvature decreases when junctions exist. 108 In order to reduce the end caps, the worm-like cylinders may also form toroids or rings which have been observed recently in an amphipihlic triblock copolymer system. 114 Yet the ring formation requires that the worm-like cylinders should be flexible enough in order to bend (with a small bending energy). 115 In our DMF/water system, the worm-like cylinders with PS core may not be sufficiently flexible at room temperature due to the high vitrification temperature of the PS blocks. One of the driving forces to form the network can be hypothesized to reduce the free energy by decreasing the number of end caps to form the interconnections if the length of the worm-like cylinders is not very long. Therefore, formation of the network is a result of interplay among the bending energy (curvature), the end capping energy, and the network configurational entropy. One factor which influences the bending energy and the network entropy is the number density of the junctions. All these factors contribute to the free energy of the network which can be expected to attain a configuration which would minimize the free energy. This would decide the relative appearance of end caps and junctions. In the case of block copolymers the core chain stretching and steric interactions at the junctions and saddle shaped surfaces also need to be considered. However, why this is not the case in the DMF/acetonitrile system of PS 962 -b- PEO 227? Does this imply that the end capping energy is not much unfavorable in this 79

95 system? We expect that this must be associated with the PS-solvent interaction parameters. The χ PS-water value is close to nine times higher than the χ PS-acetonitrile value, 16 indicating that water is a much poorer solvent for PS compared to acetonitrile. In the DMF/water system, the cylinders should tend to reduce the net curvature by forming a network while in DMF/acetonitrile they should not. Namely, the diameter of the wormlike cylinders in DMF/water system should be smaller than that in DMF/acetonitrile system. However, this is not the case as indicated by our TEM observations: the average diameter of the worm-like cylinders in DMF/water system is 35 nm while that in DMF/acetonitrile is 38 nm. Another possibility is that the worm-like cylinders formed in the DMF/acetonitrile system of PS 962 -b-peo 227 are long enough to down play the density of end caps. Further investigation is currently undertaking. The branched worm-like cylinders and network structures are also considered sometimes as intermediaries between the worm-like and lamellar morphologies. 24 It has been observed that the morphological change from worm-like cylinders to vesicles in PBD 103 -b-paa 75 occur via branched cylinders and a worm-like cylinder network, 24 while in the case of crew cut morphology of PS 310 -b-paa 52 in DMF/water, the change from cylinders to vesicles occurs without the formation of a network. 50 In our case of the block copolymer PS 962 -b-peo 227, when the solvent mixture is DMF/water, the change from cylinders to vesicles occurs via branched cylinders and the network; while in the DMF/acetonitrile solvent mixture, the change occurs without the formation of the network. Again, the solvent plays a crucial role in deciding the morphological pathway of the cylinder to vesicle changes. 80

96 CHAPTER IX TEMPERATURE INDUCED MORPHOLOGICAL TRANSITIONS In CHAPTER V, we had seen that the morphology of PS 962 -b- can be precisely controlled by varying the solvent composition and copolymer concentration. In this chapter we discuss the reversible morphological changes of PS 962 -b-peo 227 which can induced by solely changing the temperature for fixed copolymer concentration and solvent composition in a DMF/water system 116. The changes are reversible and require only a few hours of equilibrating for the forward and backward changes. To best of our knowledge this is the first direct observation of accessing solely by changing the temperature, all the three classical micelle phase morphologies in dilute solution i.e., spheres, cylinders and vesicles for a block copolymer with fixed block and solvent composition and at fixed copolymer concentration. The morphological changes are also followed for different copolymer and water concentrations in the DMF/water system. Intermediate or metastable micelle morphology is also found in the pathway from one thermodynamically stable morphology to another during isothermal experiments. These observations are useful to achieve an insight in the mechanism of these morphological changes. Figure 9.1 shows the morphological changes induced by increasing temperature for a system with 0.2 (wt). % copolymer concentration and 4.5 (wt.) % water concentration in DMF/water. At room temperature the micelle morphology is vesicles as 81

97 shown in Figure 9.1a. At 45 C the morphology changes to a mixture of worm-like cylinders and vesicles (Figure 9.1b). When the temperature is increased to 50 C exclusively worm-like cylinders are formed as shown in Figure 9.1c. At 60 C a mixture of spheres and rod-like cylinders is obtained (Figure 9.1d) and finally at 70 C, the morphology changes to aggregated spheres (Figure 9.1e). A B C D E Figure 9.1 Morphological changes on heating a system with 0.2 (wt.) % copolymer and 4.5 (wt.)% water concentration in DMF/water: (a) pure vesicles formed at room temperature; (b) a mixture of vesicles and worm-like cylinders at 45 ºC; (c) pure wormlike cylinders at 50 ºC; (d) a mixture of spheres and rod-like cylinders at 60 ºC; (e) pure spheres at 70 ºC. The spheres are aggregated because at higher temperatures, the solvent quality of water for PEO is not as good as at room temperature, and it might be less effective to provide 82

98 colloidal stabilization to the micelles. It is demonstrated that by changing only the temperature we can access spherical, worm-like cylindrical and vesicular morphologies without changing compositions of the system. Note that all these observations on the morphological changes are taken after the samples were equilibrated four hrs, and thus, they are assumed in equilibrium condition (see below for experimental evidence). When each of these samples is cooled back to room temperature the vesicular morphology is regenerated. Detailed morphological studies show that the morphological changes during cooling for the system with 0.2 (wt.) % copolymer concentration and 4.5 (wt.) % water concentration in DMF/water are completely reversible. Namely, the TEM morphological observations are from Figure 9.1e back to Figure 9.1a during cooling. Furthermore, no hysteresis has been observed during cooling compared with those observations during heating. This indicates that the times required conducting the morphological change during heating and cooling are close to identical. We also follow the change in turbidity when a system with 0.4 (wt.) % copolymer concentrations and 4.35 (wt.) % water concentration in DMF/water is heated from room temperature and cooled from high temperatures as shown in Figure 9.2. It is evident that the turbidity changes during heating and cooling are identical, again, indicating the absence of hysteresis in the morphological changes. Micelle rings are sometimes formed and observed, when the vesicles changes towards the cylinders Figure 9.3a shows the rings coexisting with the vesicles obtained on heating the system with 0.2 (wt). % copolymer concentration and 4.5 (wt.) % water concentration in DMF/water from room 83

99 Turbidity Heating Cooling Temperature ( 0 C) Figure 9.2 Change in turbidity with temperature for a micelle system with 0.4 (wt.) % copolymer concentrations and 4.35 (wt.) % water concentration in DMF/water on heating and cooling. temperature to 40 C. Initially, the cylinders formed would be short and rod-like and thus, the end-capping energy would be important as compared to the overall free energy of the micelle. In order to decrease the overall free energy by eliminating the ends, the rings form if the bending energy is less significant as compared to the end-capping energy. We have also captured some intermediate morphologies for the vesicle to cylinder change. Figures 9.3b and 9.3c show lamellae with protruding rods, suggesting that rods are formed from lamellae which are originated from vesicles which provide some evidence 84

100 that the rods may also be occasionally formed directly from the vesicles, although this morphology is only seen rarely. A B C D Figure 9.3 Intermediate morphologies formed on heating a system with 0.2 (wt.) % copolymer and 4.5 (wt.) % water concentration in DMF/water: (a) a mixture of vesicles and rings at 40 ºC; (b) lamellae with protruding rods at 45 ºC; (c) circular lamellae with protruding rods; (d) vesicles with protruding rod. 9.1 Driving Force for the Morphological Changes. The morphological changes are based on the free energy of the micelle which consists of the free energy of the core, free energy of the corona and the free energy of the 85

101 interface. 46 To better understand these changes, we should obtain an insight into the PS and PEO interactions with the solvents and how they are affected by changing the temperature. The polymer-solvent interactions can be estimated by the polymer-solvent interaction parameters if we neglect the entropic contribution. As we saw earlier, these parameters can be estimated using the van Laar-Hildebrand equation of 83 χ P-S = V S /(RT) (δ P - δ S ) 2 (9.1) where V S is the molar volume of solvent, δ P and δ S are solubility parameters of polymer and solvent, respectively. For sake of simplicity, we neglect the temperature dependence of δp, δ S (they are slightly temperature dependent 83 ) and V S in our calculations. As is evident, equation 9.1 predicts a decrease in the χ P-S value as the temperature increases. To obtain a better idea we plot this equation in Figure 9.4a for PS-water, PS-DMF and PEO- DMF interaction parameters, χ PS-water, χ PS-DMF and χ PEO-DMF, respectively. It can be seen that the relative decrease in χ PS-water is much more significant than the χ PS-DMF and χ PEO- DMF. It is known that PEO has a LCST behavior in water with a LCST temperature around 100 C, and the χ PEO-water increases as temperature increases. 93 Therefore, for a 4.5 (wt.) % water concentration in the DMF/water system with an assumption that the system obeys the ideal solution behavior (the solubility parameters in the mixture obey a linear addition scheme), we can plot the change in χ PS-solvent and χ PEO-solvent with temperature as shown in Figure 9.5. It is evident that at room temperature the value of χ PS-solvent is much higher than χ PEO-solvent, indicating that the solvent mixture (DMF/water) is poor for PS blocks and good for PEO blocks. As the temperature increases the χ PS-solvent decreases (Figure 9.5), the mixed solvent thus becomes less poor for PS blocks. On the other hand, the χ PEO-solvent does not change significantly with temperature, since the χ PEO-water increases 86

102 and χ PEO-DMF decreases with increasing temperature. The overall solvent mixture χ PEOsolvent thus remains almost constant. We can also further exam the solvent quality for PEO by analyzing the change in the excluded volume parameter υ with temperature. The excluded volume parameter is related to the second virial coefficient A 2 by the following expression 91 υ = 2A M a (9.2) 2 0 / 3 N av where M 0 is the molecular weight of the PEO monomer, and N av is the Avogadro s number. The second virial coefficient A 2 is related to the χ P-S by the expression of 117 A 2 2 = V01 M ( 1/ 2 χ P S ) / M 0 ( V01 V02 ) /( M n 0 ) (9.3) where V 01 is the molar volume of the solvent, χ P-S is the interaction parameter, V 02 is the molar volume of the PEO monomer and M n is the number average molecular weight of the PEO polymer. We have already estimated the χ PEO-solvent in Figure 9.5, and using eqs2 and 3, we can calculate the excluded volume parameter υ at different temperatures, assuming all other values are independent of temperature. After calculating and plotting, we find that the change in υ with temperature is insignificant. Therefore, the interaction between PEO and mixed solvent is kept almost unchanged with changing temperature. Let us now calculate the micelle free energy of the system at different temperatures based on the size and geometry information obtained from the TEM observations and equations we have described in Chapter V. It is surprising that even though there is a reduction in the χ PS-solvent and a corresponding reduction in the surface tension (γ), the calculated free energy of the interface (F interface ) remains almost constant for micelles formed at different temperatures. This is because the interfacial area per chain s increases from vesicles to 87

103 cylinders and then to spheres. Therefore, as temperature is increased, γ decreases but the area each chain occupies, s, increases, and F interface = γ s possesses little changes. It can be concluded that the free energy of the interface should not be the major driving force for the micelle morphological changes. A χ PS-DMF PS-Water PEO-DMF Temperature ( 0 C) Figure 9.4 Change in polymer solvent interaction parameter χ P-S with temperature Individual polymer-solvent interaction parameters: χ PS-water, χ PS-DMF and χ PEO-DMF. However, we found that there is a decrease in the free energy of the corona (F corona ) which is dependent mainly on the excluded volume parameter υ and the tethering density (1/s) as we saw in Chapter V. Calculations conducted earlier here using equations 9.2 and 9.3 revealed that the excluded volume parameter does not change significantly with temperature, but the tethering density (1/s) decreases from vesicles to cylinders and then, 88

104 to spheres. Therefore, it is speculated that the change of F corona acts as the major driving force for the morphological changes with temperature as compared to the case of changing the selective solvent concentrations where the change in F interface is the major driving force for the morphological changes as we saw in Chapter V. B χ PS-Solvent PEO-Solvent Temperature ( 0 C) Figure 9.5 Change in polymer solvent interaction parameter χ P-S with temperature polymer-dmf/water interaction parameters: χ PS-solvent and χ PEO-solvent. 9.2 Initial Copolymer Concentration Effects on the Morphological Changes. We also follow the morphological changes with temperature for different initial copolymer concentrations while keeping the water concentration fixed at 4.5 (wt.) %. Figure 9.6 shows the turbidity measurements for 0.2 (wt.) %, 1 (wt.) % and 2 (wt.) % copolymer micelle systems with change in temperature. We can define two characteristic temperatures from these three sets of data. First temperature is the critical transition 89

105 temperature (CTT), which is the minimum temperature above which the morphological changes can occur; and second is the critical micellization temperature (CMT), which is the temperature above which no micelles exist. It is evident from these three sets of data that both CTT and CMT are dependent on the copolymer concentration. As the copolymer concentration is increased, both CTT and CMT shift to higher temperatures. The CTT values for 0.2 (wt.) %, 1 (wt.) % and 2 (wt.) % are 40 C, 45 C and 50 C, respectively; while the CMT values are 70 C, 85 C and 95 C respectively (the vertical arrows in Figure 9.6). Turbidity (a.u.) Temperature 0.2 (wt.) % 1.0 (wt.) % 2.0 (wt.) % 0 C Figure 9.6 Change in turbidity with temperature for micelle systems with 4.5 (wt.) % water concentration and different copolymer concentrations in DMF/water systems. 90

106 The results of the micelle morphological changes in the systems having DMF/water mixture with 4.5 (wt.) % water concentration and different copolymer concentrations are summarized in Table 9.1. It is evident that the individual morphological changes are shifted to higher temperatures as the copolymer concentration is increased. Table 9.1: Morphologies formed at different temperatures for different copolymer concentrations and 4.5 (wt.) % water concentration. Polymer Temperature (ºC) Concentration (wt.) % V V+C C C+S S 0.4 V V V+C C C+S C+S 2 V V V+C V+C C C+S V: Vesicles, C: Cylinders, and S: Spheres For example, for 0.2 (wt.) %, 0.4 (wt.) % and 2 (wt.) % copolymer concentrations the rod-like cylindrical morphology is formed at 50 C, 60 C and 70 C, respectively. At 80 C, for the 0.2 (wt.) % copolymer concentration no micelles exist; while for 0.4 (wt.) % and 2 (wt.) % the micelle morphology is still a mixed spheres and rod-like cylinders. The effect of concentration can be understood on the basis of aggregation number and the polymer-solvent interaction parameter χ P-S. The aggregation number N agg is related to the concentration from the surfactant theory in small molecules 46 N agg = 2( C / CMC) 1/ 2 (9.4) where C is the surfactant concentration, and here, we use as the copolymer concentration, and CMC is the critical micellization concentration. This equation is valid for simple micelle morphologies of spheres, rods and vesicles, although N agg is less sensitive to C in case of spheres 118 In our case the CMC is actually a function of the water concentration, which decreases as the concentration water increases as we saw in Chapter IV, Therefore, 91

107 as the copolymer concentration increases the N agg increases. At higher aggregation numbers the micelle size becomes large, resulting in higher core chain stretching and thus, a greater reduction in the free energy of the interface is required to stabilize the micelle morphology. This can now be achieved by increasing the temperature. Moreover, it is known that the polymer-solvent interaction parameter χ P-S is related to the polymer concentration φ by the following equation 83 χ P S = χ χ1φ + χ 2φ +... (9.5) where χ 0, χ 1, χ 2 are coefficients. At low copolymer concentrations the higher order terms are negligible. But at high concentrations, they become important and increase the effective χ P-S. Thus, when the copolymer concentration increases, the χ P-S increases and the micelle systems need to be heated to higher temperatures to cause a sufficient decrease in χ P-S, which will lead to the morphological changes. 9.3 Effect of Water Concentration on the Morphological Changes. Figure 9.7 shows the change in static SLS intensity for a 0.1 (wt.) % copolymer concentration with different water concentrations. As can be seen in this figure, the water concentration in the DMF/water mixture has a strong influence on the intensity changes. For the water concentrations of 6 (wt.) % and 8 (wt.) %, the scattered intensity qualitatively decreases with increasing temperature; while for the system with a 10 (wt.) % water concentration the intensity increases with increasing temperature. The question is why the scattering intensity-temperature dependence is so different for these systems with respect to the water concentration? We investigated the morphological changes using TEM and found that the morphology changes from vesicles to worm-like cylinders 92

108 and then to a mixture of spheres and rod-like cylinders, when the water concentrations are 6 (wt.) % and 8 (wt.) %. I (a.u.) (wt.)% 8 (wt.)% 10 (wt.)% Temperature 0 C Figure 9.7 Change in intensity of scattered light with temperature for micelle systems with 0.1 (wt.) % copolymer and three different water concentrations in DMF/water: 6 (wt.) %, 8 (wt.) % and 10 (wt.) %. However, the vesicular micelle morphology does not change in the system with the water concentration of 10 (wt.) % up to 90 C. We can use eq 9.1 to predict the change in χ PSsolvent with temperature for different water concentrations in DMF/water mixtures. The results are plotted in Figure 9.8 The χ PS-solvent decreases significantly for all these three water concentrations on increasing the temperature. Therefore, the reduction in the free energy of the interface must be 93

109 (wt.)% 8 (wt.)% 10 (wt.)% χ Temperature ( 0 C) Figure 9.8 Change in PS-solvent interaction parameter, χ PS-solvent, with temperature for different water concentrations in DMF/water systems. similar in all the three systems. The reason for the fact that no morphological changes are seen in the case of 10 (wt.) % water cannot be attributed to the thermodynamic considerations. In another aspect, we can consider that the mobility of PS blocks could be severely restricted due to its strong hydrophobic nature even though the temperature is close to its T g. The good solvent DMF greatly enhances the mobility of the PS blocks, which is the necessary condition for the observation of the morphological changes, it can, however, only take place at low water concentrations. With increasing the water concentration, the micelle morphological change may be very slow, and no morphological changes could be observed with changing temperatures in the time period 94

110 of our experiments. Pispas and Hadjichristidis reported a similar increase in scattered intensity for the PBD-b-PEO micelle system in aqueous solutions. 70 In that case; a possible explanation could also be that even though PBD is in a rubbery state, its mobility is still restricted at a high water concentration due to its hydrophobic nature. 9.4 Pathways of Morphological Changes in Isothermal Experiments. In order to investigate whether the morphological change pathways are identical from different initial micelle phases, we monitor the morphological changes at an isothermal temperature after both rapid heating from the room temperature or rapid cooling from a high temperature. A system with 0.4 (wt.) % PS 962 -b-peo 227 copolymer concentration and 4.35 (wt.) % water concentration in DMF/water is used. The initial morphology is vesicles at room temperature. First, this system is rapidly heated from room temperature to 50 C and equilibrated there for two hrs, the morphology observed is mainly wormlike cylinders as shown in Figure 9.9a. The second experiment is that the same system is first heated to 80 ºC and equilibrated for two hrs. The copolymer forms spherical micelles. This system is then quenched to 50 ºC and equilibrated for another two hours. The morphology observed is also mainly worm-like cylinders as shown Figure 9.9b. Different from the first experiment, the cylinders are formed from the spherical micelles and the system is being quenched from a high temperature of 80 C. Therefore, irrespective of the initial morphology (vesicles or spheres) and the initial temperature (room temperature or 80 ºC) of the system from which the morphology at 50 ºC is accessed, the identical worm-like cylinder morphology is formed. The morphology at a particular temperature is independent of the initial phases and the thermal histories of the systems. Therefore, the micelle morphologies obtained are thermodynamically stable. 95

111 A B Figure 9.9 Rod-like cylindrical morphology formed from a system with 0.4 (wt.) % copolymer concentration and 4.35 (wt.) % water concentration in DMF/water in which the original morphology was vesicles: (a) after rapidly heating to 50 ºC and equilibrating for 2 hrs; (b) after rapidly heating to 80 ºC, equilibrating for 2 hrs, and then, quenching to 50 ºC and equilibrating for another 2 hrs. We still have a question in our mind: what are the pathways of these two morphological changes when the micelle system is brought from the room temperature or from the higher temperature to 50 C? We first try to determine the minimum time required at 50 C to achieve the thermodynamically stable micelle. Within this minimum time, we have then opportunities to catch the intermediate or metastable morphologies during the morphological change to the worm-like cylinders. Namely, we want to observe the morphological changes with time after the system is quickly brought to 50 C. Again, a system with 0.4 (wt.) % copolymer concentration and 4.35 (wt.) % water concentration in DMF/water at room temperature is quickly heated to 50 ºC. The morphological changes are observed in TEM after time periods of 30 mins, 50 mins, 1h 96

112 and 2h, and the respective morphologies are shown in Figures 9.10a 9.10d. At 30 mins in Figure 9.10a, we surprisingly observe large compound micelles, which have not been found in any of our previous equilibrium micelle morphologies. When the time is at 50 mins, the morphology becomes mostly worm-like cylinders mixed with some large compound micelles (Figure 9.10b). After 1h the morphology observed is exclusively worm-like cylinders (Figure 9.10c), and this morphology does not change with longer period of time such as 2h as shown in Figure 9.10d. Therefore, a period of 1h is the minimum time required to obtain the thermodynamically stable morphology. In the next experiment, the system is first rapidly heated to 65 ºC to obtain mixed micelles of spheres and rods-like cylinders, and equilibrated for two hrs. It is then quenched to 50 ºC, and we start to monitor the morphology formed after 30 mins, 1h and 2h. These morphological changes are shown in Figures 9.11a c, respectively. At a period of time of 30 mins, we also observe large compound micelles (Figure 9.11a), which are similar to those we observed in Figure 9.10a where the system was heated from room temperature. After 1h the rod-like cylindrical morphology is formed (Figure 9.11b), and retain the same morphology after 2h as shown in Figure 9.11c. Therefore, the time required to equilibrate the micelle morphology for the system cooled from a higher temperature is identical to that required for the system heated from room temperature. We can conclude that there is no hysteresis in the morphological changes in these isothermal experiments, disregarding whether the system is brought to 50 C from rapid heating or cooling. Large compound micelles have been previously observed in several block copolymers including PS 240 -b-peo diblock copolymer, polystyrene-block- 97

113 poly(acrylic acid) PS 200 -b-paa 4 in which the hydrophilic block is very short 22 and polystyrene-block-poly(4-vinylpyridine) PS 306 -b-p4vp A B C D D Figure 9.10 After quenching a system with 0.4 (wt.) % copolymer concentration and 4.35 (wt.) % water concentration in DMF/water in which the original morphology was vesicles to 50 ºC, the morphological changes with time: (a) large compound micelles after 30 min; (b) mixed large compound micelles and rod-like cylindrical micelles after 50 min; (c) rod-like cylindrical micelles after 1h; and (d) rod-like cylindrical micelles after 2h. Eisenberg et al. have shown that these large compound micelles are reverse micelles in an almost continuous PS phase in the bulk and hydrophilic chains on their surface which 98

114 provide colloidal stabilization. 22 We observe that the large compound micelles are formed during the isothermal experiments after heating a vesicle morphology system from room temperature to 50 ºC or cooling a system with the mixed sphere and worm-like cylinder morphology from 65 C to 50 ºC These large compound micelles are not thermodynamically stable, and gradually change to rod-like cylindrical micelles. Figure 9.11 After rapidly heating a system with 0.4 (wt.) % copolymer concentration and 4.35 (wt.) % water concentration in DMF/water in which the original morphology was vesicles at room temperature to 65 ºC, equilibrating for 2h, then quenching to 50 ºC, the morphological changes with time: (a) large compound micelles after 30 min, (b) rod-like micelles after 1h, (c) rod-like micelles after 2h. 99

115 We wanted to determine whether the large compound micelles can also appear when the system is heated to even higher temperatures. We rapidly heat a vesicle system to 70 ºC and monitor the morphological change of micelles with time. The result is shown in Figures 9.12a and 9.12b. The morphology changes from original vesicles to cylinders and spheres after 15 mins (Figure 9.12a). After 30 mins as shown in Figure 9.12b, these long cylinders become rod-like short cylinders and spheres. No large compound micelles at 70 ºC are observed. Therefore, the large compound micelle morphology for PS 962 -b-peo 227 is at the best a thermodynamically metastable morphology if it is not unstable. 102;120 This also indicates that the pathways of the morphological changes may be temperature dependent. Further research needs to be carried out for understanding the pathways of these morphological changes at different temperature and time scales. A B Figure 9.12 After rapidly heating a system with 0.4 (wt.) % copolymer concentration and 4.35 (wt.) % water concentration in DMF/water in which the original morphology was vesicles at room temperature to 70 ºC, the morphological changes with time: (a) long rods and spheres after 15 min, (b) short rods and spheres after 30 min. 100

116 CHAPTER X SUMMARY In this research, micelle morphologies of PS 962 -b-peo 227 have been studied in both DMF/water and DMF/acetonitrile systems at room temperature. TEM observations illustrate that by adding the selective solvent (water or acetonitrile) to a copolymer solution in DMF, the micelle morphologies change from spheres to worm-like and finally, to vesicles. In between two micelle morphologies, mixed morphology also exists. Morphological diagrams for PS 962 -b-peo 227 in these two systems have been completed by determining the cmc s using light scattering experiments. Although the trend in morphological changes is identical in both these systems, a remarkable difference in the selective solvent concentrations in these two diagrams remains. The morphological changes in DMF/water system occur in a very narrow region of water concentration; while, in DMF/acetonitrile they occur in a broad region of acetonitrile concentration. This difference is due to very different values of PS-solvent interaction parameters. Reversibility and annealing experiments reveal that we have obtained thermodynamic equilibrium morphologies. The micelle free energies have been quantitatively estimated as a first approximation. It has been shown that the energies decrease from spheres to worm-like cylinders and then, to vesicles. Among three free energy components, the F interface dominates as compared to the terms of F core and F corona. If the free energies of 101

117 two neighboring morphologies crossover, the micelle morphology change should take place as a first-order-like transition. However, in the broad selective solvent concentration regions, mixed micelle morphologies exist. We have also observed the formation of a self-assembled worm-like cylinder network for PS 962 -b-peo 227 in DMF/water. The network consists of entangled self assembled worms which are interconnected by mostly Y- and T- shaped junctions. The formation of this network is accompanied by a more than four fold increase in the inherent viscosity of the colloidal system. A similar network is not formed in the DMF/acetonitrile system, where the worm-like cylinders tend to align leading to only a marginal increase in the inherent viscosity of the system. The concentration dependence of the network is similar to that found in small molecule surfactants. The formation of the network is governed by interplay between end capping energy, bending energy (curvature) and the network configurational entropy. The final configuration of the network should be one which minimizes the free energy. Several reasons have been suggested for the differences in the morphological behavior in these two solvent systems, and more detailed work is required to get a better insight in this respect. However, it is certain that the solvent plays an important role in determining the phase morphology of worm-like cylinder micelles and the morphological pathway for the micelle changes. Further we have also observed reversible morphological changes of PS b-peo 227 in DMF/water systems solely by changing the temperature. On increasing the temperature from room temperature, the morphology changes from vesicles to worm-like cylinders and then, to spheres with mixed morphologies in between. During cooling the system from high temperature, the micelle morphologies return from spheres to worm- 102

118 like cylinders and finally reach vesicles at room temperature. The change in morphology with temperature was supported by static and dynamic light scattering measurements. The critical transition temperatures, critical micellization temperatures and individual morphological change temperatures are found to increase on increasing the copolymer concentration. The water concentration in the DMF/water system is also crucial in these changes as it affects the mobility of the PS chains. No morphological changes can be observed when the system contains a water concentration which is higher than or equal to 10 (wt.) %. The main driving force for these changes is analyzed to be reduction of the free energy of the corona. No hysteresis has been found in the morphological changes achieved by heating and cooling the system. However, large compound micelles are observed as an intermediate or metastable morphology during the initial stage of the morphological change in the system when the isothermal temperature is at 50 C after the system is rapidly heated from room temperature or cooled from high temperatures. However, the appearance of these intermediate or metastable micelles is isothermalcondition dependent. 103

119 REFERENCES 1 Whitmore, M. D.; Noolandi, J. Macromolecules 1985, 18, Halperin A. Macromolecules 1987, 20, Nagarajan, R. Ganesh, K. J. Chem. Phys. 1989, 90, Hilfiker, R.; Wu, D. Q.; Chu, B. J. Colloid Interface Sci. 1990, 135, Gast, A. P.; Vinson, P. K.; Cogan-Farinas, K. A. Macromolecules 1993, 26, Qin, A.; Tian, M.; Ramireddy, C.; Webber, S. E.; Munk, P. Macromolecules 1994, 27, Antonietti, M.; Heinz, S. Schmidt, M.; Rosenauer, C. Macromolecules 1994, 27, Glatter, O.; Scherf, G.; Schillén, K. Brown, W. Macromolecules 1994, 27, Alexandridis, P.; Lindman, B. Amphiphilic Block copolymers: Self Assembly and applications; Elsevier: New York, NY, 2000, p Kim, Y.; Dalhaimer, P.; Christian, D.V.; Discher D. E. Nanotechnology 2005, 16, S Allen, C.; Han, J.; Yu, K.; Maysinger, D.; Eisenberg, A. J. Controlled Release 2000, 63, Jenekhe, S. A.; Chen, X. L. Science 1999, 283, Jain, S.; Bates, F. S. Science 2003, 300, Choucair, A.; Eisenberg, A. Eur. Phys. J. E 2003, 10, Won, Y. Y.; Brannan; A. K.; Davis, H.T.; Bates F.S. J. Phys. Chem. B 2002,106,

120 16 Bhargava, P.; Zheng, J. X.; Li, P.; Quirk, R. P.; Harris, F. W.; Cheng, S. Z. D. Macromolecules 2006, 39, Whitesides, G. M.; Grzybowski, B. Science 2002, 295, Halperin, A. Tirrel, M.; Lodge, T. P. Adv. Polym. Sci. 1992, 100, Astafieva, I.; Zhong, X. F.; Eisenburg, A. Macromolecules 1993, 26, Gao, Z.; Varshney, S. K.; Wong, S.; Eisenberg, A. Macromolecules 1994, 27, Zhang, L.; Yu, K.; Eisenberg, A. Science 1996, 272, Zhang, L.; Eisenberg, A. J. Am. Chem. Soc. 1996, 118, Putaux, J. L.; Minatti, E.; Lefebvre, C.; Borsali, R.; Schappacher, M.; Deffieux, A. Faraday Discuss. 2004, 28, Geng, Y.; Ahmed, F.; Bhasin, N.; Disher, D. E. J. Phys. Chem. B 2005, 109, Faers, M. A.; Luckham, P.F.;Colloids Surfaces A 1994, 86, Prirma, I.; Makromol. Chem. Macromol. Symp., 1990,35/ Taylor,S.E. Chem. Ind., 1992, 20, Schmolka,I.R.,Nonionic surfactants (M.J. Schick ed.) Surfactant Science Series 1,Marcer Dekker, New York, 1966, p Schwarz, E.G. in Silicone Technology,Vol. 14 (P.F. Bruins, ed.) Interscience, New York, 1970, p Jones, M.C.; Leroux,J.C. European J. of Pharm. Biopharm. 1999,148, p1. 31 Dunn, S. E.; Brindley, A.; Davis, S. S.; Davies, M. C.; Illum, L. Pharm. Res. 1994, 11, Gaucher, G.; Dufresne M.H.; Sant, V. P.; Kang, N.; Maysinger, D.; Leroux, J.C. Journal of Controlled Release 2005,109, Kakizawa, Y.; Kataoka, K.; Adv. Drug Deliv. Rev. 2002, Behr J.P., CHIMIA 1997 Nr. 1/2 Jan./Feb

121 35 Liu, S. Q.; Tong, Y. W.; Yang, Y. Biomaterials 2005, 26, Liu, X. M; Wang, L.; Huang, J.; He, C. Biomaterials 2004, 25, Andre, X.; Zhang, M.; Mueller, A. H. E. Macromol. Rapid Commun. 2005, 26, Wei, H.; Zhang, X. Z.; Zhou, Y.; Cheng, S. X.; Zhuo, R. X. Biomaterials 2006, 27, Park, J. Bull. Korean Chem. Soc. 2001, Vol. 22, No.10, Middaugh R. et al. Euro. J. Pharm. Sc Tontisakis,A Hilfeker, Chu,B. J. Colloid Interface Sc,1990., McFann G.J.;Johnston,P.N. Hurter and A.T. Hatton, Ind. Eng. Chem. Res., 1993, Cohen,Y. Bioresource Technology (Accessed Aug. 2005). 45 Antonietti, M.; Förster, S. Adv. Mater. 2003, 15, Zhang, L.; Eisenberg, A. Polym. Adv. Technol. 1998, 9, Cornelissen J. J. J. M.; Fisher, M.; Sommerdijk, A. J. M. N.; Nolte, R. J. M. Science 1998, 280, Chen, E.; Xia, Y.; Graham, M. J.; Foster, M. D.; Mi, Y.; Wu, W.; Cheng, S. Z. D. Chem. Mater. 2003, 15, Desbaumes, L.,Eisenberg,A. Langmuir 1999, 15, Shen, H.; Eisenberg, A. J. Phys. Chem. B 1999, 103, Zhulina, E. B.; Adam, M.; LaRue, I.; Sheiko, S. S.; Rubinstein, M. Macromolecules 2005, 38, Semenov, A. N. Sov. Phys. JETP 1985, 15, Helfand, E.; Tagami, Y. J. Polym. Sci. Part B 1971, 9, Munch, M. R. Gast, A. P. Macromolecules 1988, 21,

122 55 Bang, J.; Jain, S.; Li, Z.; Lodge, T P.; Pedersen, J S.; Kesselman, E.; Talmon, Y. Macromolecules 2006, 39, Zhou, Z.; Chu, B.; Peiffer, D. G. Macromolecules 1993, 26, Fukumine, Y.; Inomata, K.; Takano, A.; Nose, T. Polymer 2000, 41, Iyama, K.; Nose, T. Polymer 1998, 39, Solomatin, S. V.; Bronich, T. K.; Eisenberg, A.; Kabanov, V. A.; Kabanov, A. V. Langmuir 2004, 20, Aathimanikandan, S. V.; Savariar, E. N.; Thayumanavan, S. J. Am. Chem. Soc. 2005, 127, Bang, J.; Viswanathan, K.; Lodge, T. P.; Park, M. J.; Char, K. J. Chem. Phys. 2004, 121, Liu, X. M.; Yang, Y. Y.; Leong, K. W. Journal of Colloid and Interface Science 2003, 266, Wei, H.; Zhang, X. Z.; Zhou, Y.; Cheng, S. X.; Zhuo, R. X. Biomaterials 2006, 27, Linse, P. Macromolecules 1993, 26, Glatter, O.; Scherf, G.; Schillén, K. Brown, W. Macromolecules 1994, 27, Michels, B.; Waton, G.; Zana, R. Colloids Surf. A 2001, , Su, Y.-L.; Wang, J.; Liu, H.-Z. Langmuir 2002, 18, Waton, G.; Michels, B.; Steyer, A.; Schosseler, F. Macromolecules 2004, 37, Dwyer, C.; Viebke, C.; Meadows, J. Colloids and Surfaces A: Physicochem. Eng. Aspects 2005, 254, Pispas, S.; Hadjichristidis, N. Langmuir 2003, 19, LaRue, I.; Adam, M.; Pitsikalis, M.; Hadjichristidis, N.; Rubinstein, M.; Sheiko, S. S. Macromolecules 2006, 39, Franta E. Journal de Chimie Physique 1966, 63,

123 73 Riess G.; Rogez, D. Polymer Preprints 1982, 23, Bronstein, L. M.; Chernyshov, D. M.; Timofeeva, G. I.; Dubrovina, L. V.; Valetsky, P. M. Langmuir 1999, 15, Bronstein, L. M.; Chernyshov, D. M.; Timofeeva, G. I.; Dubrovina, L. V.; Valetsky, P. M.; Obolonkova, E. S.; Khokhlov, A. R. Langmuir 2000, 16, Yu, K.; Eisenberg, A. Macromolecules 1996, 29, Yu, K.; Zhang, L.; Eisenberg, A. Langmuir 1996, 12, Yu, K.; Eisenberg, A. Macromolecules 1998, 31, Yu, K.; Bartels, C.; Eisenberg, A. Macromolecules 1998, 31, Yu, K.; Bartels, C.; Eisenberg, A. Langmuir 1999, 15, Yuan, J.; Li, Y.; Li, X.; Cheng, S.; Jiang, L.; Feng, L.; Fan, Z. Eur. Poly. J. 2003, 39, Quirk, R. P.; Kim, J.; Kausch, C.; Chun, M. S. Polym. Int. 1996, 39, Barton, A. F. M. Handbook of Polymer-Liquid Interaction Parameters and solubility parameters; CRC Press:Boston, MA, 1990, p Tian, M.; Qin, A.;Ramireddy, C.;Webber, S. E.; Munk, P.; Tuzar, Z.; Prochazka, K. Langmuir 1993, 9, Based on our surface enhanced Raman scattering experiments conducted on the micelles, the PEO blocks are on the surface as corona, and no PS blocks appear on the micelle surface. 86 Kent, M. S. Macromol. Rapid Commum. 2000, 21, Chen, W. Y.; Zheng, J. X.;Cheng, S. Z. D.; Li, C. Y.; Huang, P.; Zhu, L.; Xiong, H.; Ge, Q.; Guo, Y.; Quirk,R. P.; Lotz, B.; Deng, L.; Wu, C.; Thomas, E. L. Physical Review Letters 2004, 93, Zheng, J. X.; Xiong, H.; Chen, W. Y.; Lee, K.; Van Horn, R. M.; Quirk, R. P.; Lotz, B.; Thomas, E. L.; Shi, A.; Cheng, S. Z. D. Macromolecules 2006, 39, Devanand, K.; Selser, J. C. Macromolecules 1991, 24,

124 90 Lund, R.; Willner, L.; Stellbrink, J.; Radulescu, A.; Richter, D. Macromolecules 2004, 37, Rubinstein, M.; Colby, R. H. Polymer Physics; Oxford University Press:Oxford, UK, Hansen, P. L.; Cohen, J. A.; Podgornik, R.; Parsegian, V. A. Biophysical Journal 2003, 84, Venohr, H.; Fraaije, V.; Strunk, H.; Borchard, W. Europ. Polym. J. 1998, 34, Duval, M.; Sarazin, D. Polymer 2000, 41, Brandrup, J.; Immergut, E. H. Polymer Handbook, 2nd ed.; Wiley-Interscience:New York, NY, 1975; pp. VII 38, VII Elias, H. G.; Lys, H. Makromolekulare Chemie 1966, 92, Antonietti, M.; Förster, S. Adv. Mater. 2003, 15, Oberdissea, J. Eur. Phys. J. B 1998, 3, Luo, L.; Eisenburg, A. J. Am. Chem. Soc. 2001, 123, Luo, L.; Eisenburg, A. Angew. Chem. Int. Ed. 2002, 41, Liu, F.; Eisenberg, A. J. Am. Chem. Soc. 2003, 125, Keller, A. Cheng, S. Z. D.; Polymer 1998, 39, Koningsveld, R.; Stockmayer, W. H.; Nies, E. Polymer Phase Diagrams, Oxford Universiy Press:New York, 2001, pp Bhargava, P; Zheng, J. X.; Li, Quirk, R. P.; Cheng, S. Z. D.. J. Polym. Sci. Part B. 2006, 44, Hasegawa, H.; Tanaka, H.; Yamasaki, K.; Hashimoto, T. Macromolecules 1987, 20, Bohbot, Y.; Ben-Shaul, A.; Granek, R.; Gelbart, W. M. J. Chem. Phys. 1995, 103, Drye, T. J.; Cates, M. E. J. Chem. Phys. 1992, 96,

125 108 Bernheim-Groswasser, A.; Wachtel, E.; Talmon, Y. Langmuir 2000, 16, Rudin, A. The elements of Polymer Science and Engineering: an introductory text and reference for engineers and chemists; Academic Press: San Diego, CA, Raghavan, S.R.; Fritz, G.; Kaler, E.W. Langmuir Zilman, A.; Tlusty, T.; Safran, S. A. Journal of Phys.: Condens. Matter 2003, 15, S Evans D. F.; Wennerstrom, H. The Colloidal Domain: Where Physics, Chemistry, Biology, and Technology Meet, Wiley-Vch, New York, Tlusty, T.; Safran, S. A. Journal of Phys.: Condens. Matter 2000, 12, A Jintao, Z.; Liao,Y.; Jiang,W. Langmuir 2004, 20, Vab dee Schoot, P.; Wittmer J. P. Macromol. Theor. Simul. 1999, 8, Bhargava, P; Tu, Y.; Zheng, J. X.; Li, Quirk, R. P.; Cheng, S. Z. D. J. Am. Chem. Soc.; 2007; 129(5) p Adames, W.; Breuer, W.; Michalczyk, A.; Borchnard, W. Eur. Polym. J., 1989, 25, J. N. Israelachvili, Intermolecular and Surface Forces, 2nd edn, Academic Press, London 1992 p Liu, L.; Yong, G.; Cong, Y.; Li, B.; Han, Y. Marcomol. Rapid Commun. 2006, 27, Cheng, S. Z. D.; Keller, A. Ann. Rev. Mater. Sci. 1998, 28,