ORGANIZATION AND ORIENTATION OF ROD-TYPE POLYMER BRUSHES IN DIFFERENT BRUSH ARCHITECTURES UNDER SOLVENT TREATMENTS A Dissertation Presented to the Faculty of the Graduate School of Cornell University In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy by Hai Quang Tran May 2021 © 2021 Hai Quang Tran ORGANIZATION AND ORIENTATION OF ROD-TYPE POLYMER BRUSHES IN DIFFERENT BRUSH ARCHITECTURES UNDER SOLVENT TREATMENTS Hai Quang Tran, Ph. D. Cornell University 2021 Polymer brushes covalently attached to surfaces have been comprehensively studied for the last two decades since they can effectively transform surface properties for a variety of applications. However, the majority of previous studies have been focused on coil-type polymer brushes, so there is a lack of understanding in the behavior and molecular organization (e.g. conformation) of the equally important rod-type brushes when subjected to external factors such as solvent conditions or temperature. In this study, we found that for polypeptide rod-type brushes the polymer tilt angle from the surface normal changed as a function of brush thickness. Moreover, the number and nature of rod assemblies decreased while the size increased with increasing brush thickness. We showed that rod organization and orientation was strongly dependent on the conditions of a quenching process, depending on factors such as solvent miscibility and extraction rate. We also demonstrated polymer rigidity was crucial to the vertical orientation of the polypeptide. Additionally, polypeptide brushes can induce a change in liquid crystal (LC) orientation through a dispersion interaction, and the LC orientation was also proportional to the tilt of the polypeptide. Moreover, in mixed rod-coil brushes, before solvent quenching, coil brushes decreased the polypeptide tilt angle from the surface normal, resulting in a higher film thickness while after solvent quenching, coil brushes increased the number, but lowered the size of rod assemblies. We also studied nanopatterned polypeptide brushes prepared with electron beam lithography. Polypeptide brushes formed well-defined nanostructures inside pattern boundaries. Both the number of rod assemblies and brush thickness increased with pattern size. When the distance between adjacent patterns was comparable to polymer length, polymer bridges formed due to the interaction between polymer chains of adjacent patterns. The polymer bridges were alleviated by the addition of coil brushes into the otherwise empty areas. Mixed rod-coil brushes of mesogen-jacketed liquid crystalline polymer (MJLCP) and fluorinated polymer was studied. The high immiscibility between the polymers was crucial to the formation of rod assemblies whose size increased with increasing MJLCP thickness. The number of rod assemblies increased with thicker fluorinated brushes. Surface morphology can be tuned by controlling thickness of the two polymers. We also demonstrated coil brush synthesis without a typical deoxygenation step via surface initiated- single electron transfer-living radical polymerization (SI-SET-LRP). Various polymers and brush architectures could be produced with this method. BIOGRAPHICAL SKETCH Hai Tran was born and grew up in Ho Chi Minh city, Vietnam. After high school, he went to Iowa State University graduated in 2012 with a Bachelor of Science in Chemical Engineering. During his undergraduate, he had opportunities to work on undergraduate research projects in surface science and polymer synthesis. This research experience motivated him to pursue a PhD study. He then attended the University of Illinois for one year before transferring to Cornell to follow his fiancée. At Cornell, he joined the Ober’s group and started to work on projects about polymer brushes. During his time at Cornell, he studied the behaviors and organization of rod-type brushes in different brush architectures. v Dedicated to My beloved wife, daughter, parents, and siblings vi ACKNOWLEDGMENTS First, I would like to thank my advisor, Professor Christopher Ober for accepting me to his group. I really appreciated his patience, support, and guidance during my time at Cornell. Although I had a slow progress in my first two years, he was very patient and encouraged me to have a confidence in myself. While there were interruptions during my PhD study due to family issues, he gave me supports and encouragement so that I can complete my study. He also gave me brilliant ideas on polypeptide and liquid crystalline polymers which became main cores of my dissertation. Additionally, he introduced me to Prof. Abbott to start a fruitful collaboration. From him, I learned how to become an independent researcher. I would also like to thank my committee members: Professor Lara Estroff for her insight, and broad perspective about my project; and Professor Christopher Alabi for his time and letting me to use instruments in his lab. I want to thank my collaborators, Prof. Abbott and Michael Tsuei for their insights, and hard work on the project about the interaction between liquid crystal and polymer brushes. I also want to thank Dr. Ihsan Amin for guiding me on brush synthesis and characterization. I thank Yiren Zhang for his advices in chemical synthesis and Yuming Huang for help in brush patterning. I also enjoyed fruitful discussion with Dr. Florian Kafer, Dr. Ziwei Liu, Dr. Wei-Liang Chen and Dr. Roselynn Cordero. I will never forget the time we spent together in the Duffield office. Finally, I would like to thank my dear wife, Ngoc Phan, who gave me endless support and encouragement. Without her, I may not survive freezing winters in Ithaca and struggles in my research. I am also thankful to my mom and late dad for their sacrifice and support since I went to the U.S. vii TABLE OF CONTENTS BIOGRAPHICAL SKETCH .......................................................................................... v ACKNOWLEDGMENTS ............................................................................................ vii LIST OF FIGURES ...................................................................................................... xii LIST OF SCHEMES ................................................................................................... xiv LIST OF TABLES ....................................................................................................... xv LIST OF ABBREVIATIONS ..................................................................................... xvi CHAPTER 1 ................................................................................................................... 1 Synthesis, architectures and properties of rod-type and coil-type brushes .................... 1 1.1. Introduction .......................................................................................................... 1 1.2. Synthesis of polymer brushes using controlled radical polymerization from a surface ......................................................................................................................... 3 1.2.1. Atom transfer radical polymerization (ATRP) ............................................. 3 1.2.2. Reversible-addition fragmentation chain transfer (RAFT) ........................... 7 1.2.3. Nitroxide mediated polymerization (NMP) .................................................. 8 1.2.4. Ring-opening polymerization (ROP) ............................................................ 9 1.3. Advanced Brush Architecture ............................................................................ 12 1.3.1 Block Copolymer Brushes. .......................................................................... 12 1.3.2 Bottlebrush Architecture. ............................................................................. 14 1.3.3 Gradient Brushes. ......................................................................................... 16 1.3.4 Mixed Brushes. ............................................................................................ 17 1.3.5 Patterned Brushes. ........................................................................................ 19 1.3.6 Free Standing Brushes. ................................................................................. 22 1. 4. Rod-type polymers ............................................................................................ 24 1.4.1. Polypeptide brushes .................................................................................... 24 1.4.2 Mesogen Jacketed Liquid Crystalline Polymer (MJLCP) ........................... 29 1.5. Solvent annealing ............................................................................................... 34 1.6. Summary ............................................................................................................ 35 1.7. References .......................................................................................................... 38 CHAPTER 2 ................................................................................................................. 62 Synthesis, processing, and characterization of helical polypeptide rod-coil mixed brushes .......................................................................................................................... 62 viii 2.1 Abstract ............................................................................................................... 62 2.2 Introduction ......................................................................................................... 63 2.3 Results and Discussion ....................................................................................... 64 2.4 Conclusion .......................................................................................................... 73 2.5 Experimental Section .......................................................................................... 74 2.5.1 Materials ....................................................................................................... 74 2.5.2 Synthesis ...................................................................................................... 74 2.5.3 Characterization ........................................................................................... 76 2.6 Acknowledgement .............................................................................................. 77 2.7 Supporting Information ....................................................................................... 78 2.8 References ........................................................................................................... 84 CHAPTER 3 ................................................................................................................. 88 Organization and orientation of helical polypeptide brushes induced by solvent treatment ....................................................................................................................... 88 3.1 Abstract ............................................................................................................... 88 3.2 Introduction ......................................................................................................... 89 3.3 Experimental Section .......................................................................................... 92 3.3.1 Materials ....................................................................................................... 92 3.3.2 Synthesis of PBLG brushes .......................................................................... 92 3.3.3 Characterization of PBLG brushes ............................................................... 93 3.3.4 Solvent and Nonsolvent treatment of PBLG brushes .................................. 93 3.3.5 Preparation of thin LC films ........................................................................ 94 3.3.6 Microscopy Observation .............................................................................. 94 3.3.7 Changing PBLG chain persistence length via heating ................................. 94 3.4 Results and Discussions ...................................................................................... 94 3.4.2 LC response to PBLG brushes ................................................................... 101 3.4.3 Thermal Denaturation of PBLG Brushes ................................................... 104 3.4.4 Dependence of PBLG brush orientation on solvent quality ....................... 107 3.4.5 Reorganization of PBLG brushes with mixed solvents ............................. 108 3.5 Conclusion ........................................................................................................ 112 3.7 References ......................................................................................................... 126 CHAPTER 4 ............................................................................................................... 130 Nanopatterned helical polypeptide rod-type brushes via electron beam lithography 130 ix 4.1 Abstract ............................................................................................................. 130 4.2 Introduction ....................................................................................................... 131 4.3 Experimental section ......................................................................................... 134 4.3.1 Materials ..................................................................................................... 134 4.3.2 Synthesis .................................................................................................... 135 4.3.3 Characterization ......................................................................................... 137 4.4 Results and Discussion ..................................................................................... 139 4.4.1 Nanopatterned PBLG brushes .................................................................... 139 4.4.2 Effect of pattern size on brush morphology ............................................... 141 4.4.3 Binary Nanopatterned Rod-Coil Brushes ................................................... 143 4.5. Conclusion ....................................................................................................... 144 4.6 References ......................................................................................................... 145 CHAPTER 5 ............................................................................................................... 151 Tuning the surface properties of mixed Mesogen-Jacketed Liquid-Crystalline polymer rod-coil brushes .......................................................................................................... 151 5.1 Abstract ............................................................................................................. 151 5.2 Introduction ....................................................................................................... 152 5.3 Experimental Section ........................................................................................ 157 5.3.1. Materials .................................................................................................... 157 5.3.2 Synthesis of 2,5-bis[(4-butylbenzoyl)oxy]styrene (BBOS) ................. 157 5.3.3 Initiator immobilization ........................................................................ 158 5.3.4 Synthesis of PBBOS brushes ................................................................ 159 5.3.5. Solvent treatment .................................................................................. 160 5.3.6. Characterization ..................................................................................... 160 5.4 Results and Discussion ..................................................................................... 162 5.5 Conclusion ........................................................................................................ 172 5.6 References ......................................................................................................... 174 CHAPTER 6 ............................................................................................................... 182 Synthesis of polymer brushes via SI-SET-LRP in the presence of air using copper tape and hydrazine .............................................................................................................. 182 6.1 Abstract ............................................................................................................. 182 6.2 Introduction ....................................................................................................... 183 6.3 Experimental Section ........................................................................................ 186 6.3.1 Materials ..................................................................................................... 186 x 6.3.2 Synthesis .................................................................................................... 186 6.4 Results and Discussion ..................................................................................... 189 6.5 Conclusion ........................................................................................................ 199 6.6 References ......................................................................................................... 201 CHAPTER 7 ............................................................................................................... 205 Summary and Outlook ................................................................................................ 205 7.1 Mixed helical polypeptide rod-coil brushes ...................................................... 205 7.2 Organization and orientation of polypeptide brushes ....................................... 209 7.3 Fabrication of nanopatterned polypeptide brushes ........................................... 211 7.4 Mixed MJLCP rod- fluorinated coil brushes .................................................... 213 7.5 Surface-initiated Cu (0)- Control Radical Polymerization using Cu tape and reducing agents ....................................................................................................... 215 7.6 References ......................................................................................................... 216 xi LIST OF FIGURES Figure 1.1 Synthesis of polymer brushes on silicon substrate 4 Figure 1.2 Experimental setup for SI-ROP in vapor phase 11 Figure 1.3 Polymer brush architectures 13 Figure 1.4 Patterned polymer brushes 20 Figure 1.5 Conformation transition of polymer brushes to 24 chemical vapor and thermal treatment Figure 1.6 Solvent quenching treatment of PBLG brushes 27 Figure 1.7 Schematic illustration of MJLCP 29 Figure 1.8 Rod-coil diblock copolymers of MJLCP 31 Figure 1.9 Rod-rod diblock copolymers of MJLCP 33 Figure 2.1 FT-IR transmission spectra of PBLG homopolymer 66 brushes and mixed brushes Figure 2.2 The thickness of PMMA and mixed brushes, the D 68 ratio, and the tilt angle as a function of ATRP reaction time Figure 2.3 Water contact angle of mixed brush samples 70 Figure 2.4 AFM images of mixed polymer brushes 72 Figure 3.1 XPS analysis of PBLG brushes 95 Figure 3.2 AFM images of PBLG brushes of various thickness in 97 quenched state Figure 3.3 FT-IR spectra and the tilt angle of PBLG brushes of 99 various thickness Figure 3.4 AFM images of PBLG brushes of various thickness in 100 collapsed state Figure 3.5 Schematic illustrations of LC thin films on 50 nm 102 xii PBLG brushes Figure 3.6 AFM images, FT-IR spectra, and the tilt angle of 105 PBLG brushes as a function of heating time Figure 3.7 POM and AFM images of PBLG brushes treated with 110 10% chloroform in acetone Figure 4.1 Schematic illustration of the fabrication of 138 nanopatterned PBLG brushes Figure 4.2 SEM images of patterned PBLG and PS brushes 140 Figure 4.3 AFM images of patterned PBLG brushes 142 Figure 4.4 AFM images of patterned PBLG brushes, and 143 patterned binary brushes of PBLG and PNIPAM Figure 5.1 XPS spectra of polymer brushes 163 Figure 5.2 FT-IR spectra of PBBOS brushes and mixed brushes 164 Figure 5.3 Brush thickness, peak-to-valley height, number of rod 166 assemblies and WCA of mixed brushes as a function of PTFEMA thickness Figure 5.4 AFM images of mixed brushes as a function of 168 PTFEMA thickness Figure 5.5 Brush thickness, peak-to-valley height, number of rod 171 assemblies and WCA of mixed brushes as a function of PBBOS thickness Figure 5.6 AFM images of mixed brushes as a function of 172 PBBOS thickness Figure 6.1 PMMA thickness as a function of hydrazine 190 concentration and polymerization time Figure 6.2 PS thickness as a function of hydrazine concentration 193 xiii and polymerization time Figure 6.3 Brush thickness of P4VP, PHDFDMA, and 194 PPFBEMA as a function of hydrazine concentration Figure 6.4 AFM images of PMMA and PMMA-b-PS brushes 197 Figure 6.5 PMMA brush thickness as a function of the distance 199 between Cu surface and substate LIST OF SCHEMES Scheme 2.1 The synthesis procedure to make PBLG rod/PMMA 65 coil mixed brushes Scheme 3.1 Synthesis procedure of PBLG rod brushes 94 Scheme 4.1 Synthesis procedures for PBLG brushes, PS brushes, 139 and PNIPAM brushes Scheme 5.1 Synthesis of BBOS monomer, and mixed PBBOS- 161 PTFEMA rod-coil brushes Scheme 5.2 The schematic illustration of rod-coil mixed brushes 170 Scheme 6.1 Proposed mechanism of SI-SET-LRP in the presence 189 of air and hydrazine Scheme 7.1 Proposed hydrophilic and hydrophobic coil monomers 206 for mixed polypeptide- coil brushes Scheme 7.2 Side groups of polypeptides 207 Scheme 7.3 Proposed binary initiators 208 Scheme 7.4 Monomers of MJCLP 214 xiv LIST OF TABLES Table 2.1 Surface composition of mixed brush samples 69 measured with XPS Table 3.1 Surface roughness, teepee density and spacing of 96 PBLG brushes Table 3.2 Average PBLG tilt angles measured via FT-IR 107 following sequential solvent treatment. Table 3.3 Average orientations of 140 nm thick PBLG chains 108 following treatment with chloroform-acetone mixtures xv LIST OF ABBREVIATIONS Abbreviation Description 5CB 4-cyano’-pentylbiphenyl AFM Atomic force microscopy APDIPES 3-amino-propyldiisopropylethoxysilane APDMES 3-aminopropyldimethyl-ethoxysilane APTES 3-amino-propyltriethoxysilane ATRP Atom transfer radical polymerization BCP Block copolymr BiBB 2-bromo-2-methylpropionyl bromide CRP Control radical polymerization DMSO Dimethyl sulfoxide EBL Electron beam lithography FT-IR Fourier-transform infrared spectroscopy GPC Gel permeated chromatography LC Liquid crystal MJLCP Mesogen Jacketed Liquid Crystalline Polymer NMP Nitroxide mediated polymerization NMR Nuclear magnetic resonance NPB Nanopatterned polymer brushes P4VP Poly(4-vinylpyridine) PBBOS Poly{2,5-(4 butylbenzoyl)oxystyrene} PBLG Poly(γ-benzyl-L-glutamate) PHDFDMA Poly(heptadecafluorodecyl methacrylate) PMDETA 1,1,4,7,7-penta-methyldiethylenetriamine xvi PMMA Poly(methyl methacrylate) PMPCS Poly{2,5-bis[4-methoxyphenyl]oxycarbonyl)styrene} PNIPAM Poly(N-isopropyl acrylamide) POM Polarized optical microscope PPFBEMA Poly(perfluoro butyl ethyl methacrylate) PS Polystyrene PTFEMA Poly(2,2,2-trifluoroethyl methacrylate) RAFT Reversible-addition fragmentation chain transfer ROP Ring-opening polymerization SEM Scanning electron microscope SET-LRP Single electron transfer-living radical polymerization TEA Triethylamine TFE Trifluoroethanol WCA Water contact angle XPS X-ray photoelectron spectroscopy xvii CHAPTER 1 Synthesis, architectures and properties of rod-type and coil-type brushes 1.1. Introduction Polymer brushes may be defined as thin films of polymer chains covalently anchored to surfaces.1 Development of new theory to describe polymer brushes combined with new synthetic and characterization tools has led to a better understanding of the unique features of brushes and this, in turn, has triggered a significant growth of interest in the application of these important polymeric systems. Polymerization methods now permit the formation of complex polymer brush architectures with uniform molecular sizes, following growth from initiator sites bound to surfaces. The appropriate surface initiator coupled to controlled radical polymerization methods such as atom transfer radical polymerization (ATRP), nitroxide mediated polymerization (NMP) and reversible- addition fragmentation chain transfer (RAFT)1–3 also enables control of brush density and, thus, brush chain conformation from the substrate surface. In a typical dense polymer brush, the number of attachment or tethering points is sufficiently high that the crowded polymer chains are forced to stretch normal to a planar surface. This leads to residual tension in polymer brushes and may produce mechanochemical effects, influence location of brush chain ends in planar regions in the brush layer, and affects access by solvent and other molecules to the brush.4 1 Reproduced partially with permission from W-L. Chen, R. Cordero, H. Tran and C. K. Ober. 50th Anniversary Perspective: Polymer Brushes: Novel Surfaces for Future Materials. Macromolecules, 2017, 50 (11), pp 4089-4113. Copyright © 2017 American Chemical Society. 1 Covalently attached brushes may also be prepared by other methods in addition to “grafting from” surface initiation. "Grafting to" techniques have been extensively studied and may involve attachment of polymer chains to a substrate by exposing photoradical surface-bound initiators in the presence of a polymer film.5 The polymers become chemically linked to the surface and form the brush, enabling brush formation from polymers of extraordinarily high molecular weight. Alternatively, end functional chains have been tethered to surfaces using a simple thermal process to end link polymer chains to form a brush layer and create a neutral surface for the directed self-assembly of block copolymers for high resolution patterning.6 These methods cannot readily form the dense, extended brush possible when using initiation from uniform arrays of initiator groups. Polymer brushes are well known for transforming the nature of a surface with a layer just a few nanometers thick. Control of wetting properties, prevention of non-specific binding of biomolecules, colloidal stabilization and resistance to fouling are all examples of successful application of polymer brushes. Other uses include polymer compatibilizers7, new adhesive materials8,9, chromatographic devices10, and etch barriers.11 Responsive polymer brushes can be used to change wettability and binding properties, act as valves in microfluidics, control transport of ions and transduce chemical and biochemical signals.12 Synthetic polymer brushes are increasingly important as interfaces between materials and biological environments, as stimuli responsive surfaces, in drug delivery13, as surfaces for cell growth and for bioseparation.14 For example, charged brushes are used in lubrication and wetting and as antimicrobial coatings. Charge controls adsorption of molecules, enables attachment of specific molecules and living cells to surfaces and greatly influences such factors as non-specific binding on these surfaces. Nature uses brushes in interfaces to control surface wetting in the cartilage in joints15 or the surfaces of lung tissue, for lubrication16, 2 or to limit deposition of macromolecules (polysaccharides, proteins) onto surfaces. While there has been significant progress in both fundamental studies and applications of polymer brushes, prior studies were mostly focused on coil-type brushes. However, rod-type brushes with high persistence length may provide opportunities to develop new materials with unique nanostructures and interesting properties. This work examined the synthesis, processing, and characterization of rod-type brushes in various architectures to control surface topography, chain orientation, and organization with respect to the plane of the polymer film. In particular, this work demonstrated that rod- type brushes can be vertically aligned from a surface, driven by solvent treatment and phase separation from coil brushes. Moreover, rod-type brushes were self-assembled into periodic teepee-like nanostructures. Such a surface topography may allow precise control of surface functionalities and provide highly anisotropic materials with potential use in charge transport or biological interface systems. This chapter gave an overview of several excellent candidates for rod-type brushes with discussion of molecular design, polymer synthesis and chain orientation. Strategies for the preparation of complex brush structures and promotion of vertical alignment of rod brushes were also mentioned. 1.2. Synthesis of polymer brushes using controlled radical polymerization from a surface 1.2.1. Atom transfer radical polymerization (ATRP) ATRP is based on a reversible activation-deactivation process.17 An alkyl halide terminated initiator will undergo a homolytic and reversible cleavage with a transition metal complex to generate a reactive radical species to which monomers will be added. Then the new propagating radical will be deactivated by the oxidized form of the metal complex to produce the halogen capped dormant species. These repetitive cycles allow 3 for control of the radical polymerization for well-defined polymers. 18,19 An important advantage of this method is that it is well suited to the polymerization of (meth)acrylates with all the many functionalities that it provides. Ejaz et al. reported one of the earliest examples of the synthesis of polymer brushes via surface initiated (SI)-ATRP.17 The Figure 1.1. Synthesis of polymer brushes on silicon substrate using: a) SI-ATRP , b) SI-RAFT, c) SI-NMP, d) SI-ROP. Langmuir-Blodgett (LB) technique was used to lay down a well-organized layer of initiator on a silicon substrate followed by SI-ATRP of poly(methyl methacrylate)(PMMA) brushes. Free initiator p-toluenesulfonyl chloride (TsCl) was added to control the polymerization due to the extremely low concentration of the initiator on the surface.17 Matyjaszewski et al. reported the synthesis of homopolymer 4 and diblock copolymer brushes in the absence of free initiator (Figure 1.1a).1 Instead, CuBr2 was added at the beginning of the reaction to ensure a sufficient concentration of deactivating Cu(II) species. Jones et al. demonstrated the accelerating effect of water on SI-ATRP through a much faster formation of thick PMMA brushes compared to prior reports.20 The ability to grow poly(methyl methacrylate)-b-poly(2-hydroxyethyl methacrylate)(|-PMMA-b- PHEMA) block copolymer brushes proved the controlled nature of this method2. Huang et al. also confirmed the accelerating effect of water on SI-ATRP as a 700 nm thick PHEMA polymer brush layer was observed in aqueous media while the process formed a brush only 6 nm in the absence of water.21 The conventional ATRP method discussed above involves an immobilized ATRP initiator on a substrate from which polymer chains will grow. Sedjo et al. reported a modified method called reverse ATRP when a conventional radical initiator was used during the initial addition of the Cu(II)/ligand complex.22 Specifically, an immobilized azo initiator decomposed upon heating and reacted with Cu(II)/ligand complex to generate in situ ATRP initiator and Cu(I)/ligand complex. This simple method was successful in preparing poly(styrene)-b-poly(methyl methacrylate)(|-PS-b-PMMA) block copolymer brushes.22 In addition, conventional ATRP requires stringent deoxygenated conditions as oxygen, if present in the system, can quickly trap propagating radicals. Matyjaszewski and coworkers developed a simple technique called Activator Regenerated by Electron Transfer (ARGET) ATRP which involved addition of a minute amount of active copper catalyst in the presence of excess reducing agent while no deoxygenation step is needed.23 As proof, CuCl2 was used together with tin(II) 2-ethylhexanoate as a reducing agent to prepare poly(n-butyl acrylate)(PBA) homopolymer, and |-PBA-b-PS block copolymer brushes from silicon wafers.23 2 |- indicates the surface and the resulting block segment is shown grown from surface. 5 The major disadvantages of conventional SI-ATRP are its air free conditions, the use of metal catalysts which may be harmful for some applications and the inability to reuse the polymerization solution. As a result, Li et al. introduced electrochemically mediated ATRP (eATRP) to overcome some of these problems.24 A constant electrical potential was applied to generate Cu(I) catalyst through the reduction of Cu(II) in the vicinity of the ATRP-initiator modified electrode so that the polymerization can be initiated.24 This method was demonstrated by the growth of poly(3-sulfopropyl methacrylate, potassium salt) (PSPMA) brushes on a gold electrode in the presence of air. The same group then extended the eATRP technique to non-conducting substrates.25 In this case, an initiator modified silicon substrate was placed opposite a working electrode, and a potential was applied to reduce Cu(II) to Cu(I) which then diffused to the silicon substrate to initiate the polymerization. More interestingly, the smaller the gap between the electrode and the substrate was, the thicker the polymer brushes formed, so by tilting the electrode with respect to the substrate, a gradient of polymer brushes along the substrate can be produced.25 When a zinc plate was used as a sacrificial anode, it served to reduce Cu(II) to Cu(I) to initiate the polymerization.26 Only a microliter volume of polymerization solution was needed and tilting the Zn plate also allowed the formation of gradient polymer brushes. The recently developed techniques from the conventional SI-ATRP have overcome many of these issues and made this technique into an excellent choice for people at different synthetic levels and demonstrated industrial scale manufacture for real-life applications. The fabrication of gradient polymer brushes is also more straightforward with SI-ATRP than with SI-NMP or SI-RAFT (see below). 6 1.2.2. Reversible-addition fragmentation chain transfer (RAFT) RAFT polymerization is a versatile technique that involves a reversible regenerative chain transfer mechanism.27 Baum et al. reported the synthesis of polymer brushes from a silicon substrate via SI-RAFT. A surface immobilized azo initiator was used in the presence of a chain transfer agent (CTA) to polymerize styrene, MMA, and N,N- dimethylacrylamide (DMAEMA) with well controlled thickness (Figure 1.1b).3 Another strategy is to use a surface immobilized RAFT agent which can be attached to the surface through the leaving and reinitiating group (R group) or through a radical stabilizing group (Z group). The Z group approach was used to prepare homopolymer and diblock copolymer brushes on a silicon substrate.28,29 However, as this approach resembles a grafting-to method, polymer brushes with lower grafting density and reduced uniformity compared to other “grown from” brushes are more likely to form at high molecular weight.30 The R group approach was reported for polymer brush synthesis of methyl acrylate (MA), MMA, DMAEMA and their diblock copolymer brushes.31 This approach is similar to the grafting-from technique in ATRP, thus allowing a high density and uniform film, though the molecular weight distribution may be broadened by bimolecular termination.32,33 Advincula and coworkers reported the synthesis of polymer brushes via SI-RAFT on a conducting Au surface and on polythiophene films on electrode surfaces.34,35 SI-RAFT is a facile technique and is compatible with a wide range of monomers. Another advantage is that no metal catalyst is needed. However, this method generally does not produce polymer brushes as thick as other SI-controlled radical polymerization (CRP) techniques. In fact, polymer brushes with thickness less than 30nm are often reported using SI-RAFT. 28,29,31,36,37 The RAFT agent is also often expensive or not commercially available; therefore, multiple step syntheses may be needed. 7 1.2.3. Nitroxide mediated polymerization (NMP) NMP is based on the reversible reaction of growing radical chain ends with a stable nitroxide free radical.38 In general, a thermally unstable alkoxyamine goes under a reversible homolytic reaction to give a stable nitroxide radical and the reactive radical. Monomers are added to the radical to produce a polymeric radical which will be trapped by the stable nitroxide radical to yield dormant specie.39 The process is repeated to increase the degree of polymerization of the chain. Using a TEMPO-based initiator containing a chlorosilane anchoring group, Husseman et al. successfully synthesized PS brushes from a silicon substrate, showing the first example of SI-nitroxide mediated polymerization (SI-NMP) (Figure 1.1c).40 The addition of free alkoxyamine initiator is crucial for the success of SI-NMP as it resulted in a high concentration of nitroxide in the polymerization solution. It overcomes the issue of having extremely low concentration of surface bound initiator relative to the monomer concentration, leading to controlled polymer brush growth.40 The linear relationship between brush thickness and the molecular weight of a bulk polymer as well as conversion indicated that the growth of bulk polymer chains and polymer brushes were well controlled. The preparation of block and random copolymer brushes was also demonstrated. SI-NMP can also be done in a vapor phase to prepare homo and block copolymer of various monomers in more efficient way than in solution phase.41 The procedure of SI- NMP discussed so far is a unimolecular system.40 The second system for SI-NMP is bimolecular when an azo-based initiator is anchored on a surface following polymerization in the presence of a nitroxide to form in situ alkoxyamine.39,42 The bimolecular system resulted in a higher grafting density of the initiator due to the smaller size of the azo-initiator compared to the alkoxyamine-based initiator which is thought to be more stable than the azo-initiator.39 8 In addition, the TEMPO mediated system is mostly limited to styrene and its derivatives, but is not compatible with other types of monomers such as acrylates.2 Thus, other nitroxides have been developed. Parvole et al. reported the synthesis of PBA brushes from a silicon substrate using an immobilized azo initiator and the nitroxide, N- tert-butyl-N-1-diethylphosphono-2,2-dimenthylpropylnitroxyl (DEPN or SG1).42 Hawker and coworkers reported the formation of poly(tert-butyl acrylate) (PtBA) brushes on silicon substrate using an immobilized TEMPO-based alkoxyamine initiator in the presence of free 2,2,5-trimethyl-4-phenyl-3-azahexane-3-oxyl (TIPNO)-based alkoxyamine.43 A similar approach was used to synthesize copolymer brushes of 2- (dimethylamino ethyl)acrylate and styrene on a stainless steel surface.44 The advantage of SI-NMP is that no catalyst or metal is required, so it is an ideal technique to prepare polymer brushes for electronic or biological applications which are sensitive to impurities. Moreover, polymer brushes with a thickness from a few nm up to 200nm for various monomers were reported. 41,43 However, as the reaction temperature for SI-NMP is very high, it may not be suitable for a number of thermally sensitive monomers or substrates. Moreover, as TEMPO-based initiators do not work well for many monomers, a thoughtful selection of alkoxyamine is needed when using new types of monomers. 1.2.4. Ring-opening polymerization (ROP) The ROP of N-carboxyanhydrides may be used to synthesize polypeptide brushes from immobilized primary-amine-initiators. In general, the amine will react with NCA to open the ring and produce the intermediate carbamic acid which then releases carbon dioxide and generates a new amine group at the chain end. The newly formed amine group then continues to react with NCA and propagate the growing chain 45 (Figure 1.1d). Using this approach, polymers from different types of amino acids such as 9 glutamate, aspartate, serine, cysteine, tryptophan or alanine have been reported 46. In addition, depending on the anchoring group, polypeptide can be grown on gold substrate from a thiol initiator or on silicon wafer, glass, or aluminum substrates using a silane initiator 46,47. Other substrates such as multi-wall carbon nanotubes (MWNTs) or polymer membranes were also reported, though multi-step treatment is usually required to introduce amino groups 48–50. The SI-ROP can be performed in solution, melt or the vapor phase. SI-ROP in solution: Schouten and co-workers reported the copolymerization of glutamate and aspartate on silicon wafers and glass slides in THF using 3- aminopropyltriethoxysilane (APTES) initiator. Wieringa showed the first order kinetics of the polymer brush growth of poly(benzyl-L-glutamate) (PBLG) and poly(methyl-L- glutamate) (PMLG) in DMF and the linear effect of monomer concentration on brush thickness 51. However, the brush growth leveled off after five hours due to the buildup of non-grafted polymer on top of the polymer brushes which prevents access of monomers to the reactive chain end. After washing off the free polymer, brush growth can restart again. The living characteristic of SI-ROP in solution was demonstrated through the synthesis of PBLG-b-PMLG block copolymer brushes 52. Nonetheless, a limitation of SI-ROP in solution is that impurities from solvents such as water can interfere with the polymerization resulting in early termination or formation of free polymer. For that reason, brush thickness over 40 nm is rarely reported with this method. 46 SI-ROP in the melt: Wieringa reported SI-ROP in the melt which requires less stringent conditions, but has a faster polymerization rate and is highly reproducible. In this method, monomer solution was spin-coated onto amine-functionalized silicon substrates to form a thin layer of monomer which was heating above the melting point of the monomer. This resulted in a 20nm polymer brush after 30 minutes 53. Although 10 this method can minimize impurities, non-grafted polymer caused by thermal polymerization may inhibit the brush growth. SI-ROP in the vapor phase: Change introduced a new method where monomers were vaporized under vacuum and condensed onto a substrate to start the polymerization 54. As illustrated in Figure 1.2, the monomers are spread at the bottom of a reactor while amine-functionalized substrate is held horizontally above it. The vacuum is applied and the reactor temperature is raised above the melting point of monomer to vaporize it. When vapor reactants deposit on the substrate, the surface amine will initiate the polymerization. This approach can eliminate all impurities and moisture as no solvent is needed. Brush thickness of up to 40 nm was synthesized and can be controlled by changing the pressure, temperature, distance between the monomer source and substrate, or reaction time. Lee demonstrated the versatility of this method by synthesizing polymer brushes from a wide range of amino acids such as benzyl-L- glutamate, methyl-L-glutamate, benzyl-L-aspartate, O-benzyl-L-serine, S-benzyl-L- cysteine, O-benzyl-L-tyrosine, L-tryptophan, L-phenylglycine, and L-phenylalanine 55. Figure 1.2. Experimental setup for SI-ROP in vapor phase. Reproduced with permission from ref. 54 11 Wang and Change further improved this method through controlling the temperatures of monomer source and substrate separately 56. Moreover, the enclosed space between the monomer and substrate was more confined to increase the effective concentration of vapor monomer. Under optimal conditions, PBLG brush thickness close to 200 nm can be achieved within 30 minutes. Thick brushes of poly(methyl L-glutamate) (PMLG), poly-L-phenylalanine (PLPA), poly(benzyl L-aspartate) (PBLA), poly(N- carbobenzyloxy L-lysine) (PCBL) as well as their block-copolymers such as PLPA-b- PBLG and PBLG-b-PCBL were successfully prepared 56. Zheng developed the vapor- deposition polymerization system with better control of temperature and pressure so that the process can occur at much higher pressure than those in previous studies. Besides, the composition of α-helix and random coil segments in PBLG brushes can be tuned by controlling the packing density 57. 1.3. Advanced Brush Architecture With the use of living polymerization methods, variations of polymer brush architecture such as block copolymer brushes, bottlebrush, mixed brushes, gradient brushes, free standing brushes and patterned brushes are possible and have been reported. Differences in architecture, for example, may lead to better control of surface formation and coverage, provide better access to specific functionality and enable structured surface formation. 1.3.1 Block Copolymer Brushes. Husseman et al. utilized nitroxide mediated polymerization (NMP) to prepare block copolymer brushes composed of a first block of polystyrene and a second block of a random copolymer of PS and PMMA on silicon substrates.40 Matyjaszewski et al. used SI-ATRP to synthesize |-PS-b-PMA and |-PS-b-PtBA diblock copolymer brushes.1 In addition, ARGET ATRP was reported for the preparation of |-PBA-b-PS copolymer 12 from silicon substrates.23 The formation of |-PS-b-PMMA diblock copolymer brushes can also be done by reverse-ATRP.22 Moreover, SI-RAFT was successfully used to Figure 1.3. a) Thermoresponsive polypeptide brushes with oligo(ethylene glycol) side chains. Reproduced with permission from ref. 71, b) stimuli responsive diblock copolymer brushes. Reproduced with permission from ref. 60, c) poly(acrylic acid) brushes with grafting density gradient. Reproduced with permission from ref. 84, d) molecular weight gradient polymer brushes via sacrificial-anode ATRP (sa ATRP). Reproduced with permission from ref. 26, e) nanoscale structural rearrangements of the Y-shaped brushes to form : (i) a corona with PS chains (blue) covered a core of poly(acrylic acid) chains (red) in toluene; (ii) swelling state of the brushes in non-selective solvent; (iii) PAA arms partially covered PS core in a good solvent for PAA. Reproduced with permission from ref. 103, f) reconstruction of mixed brushes in swollen state and dry state for both non-selective (Figure f1,2) and selective (Figure f3,4) solvents respectively. Reproduced with permission from ref. 97. grow |-PDMAEMA-b-PHEMA diblock brushes from silicon substrate with immobilized chain transfer agent.28 Similar approaches using SI-RAFT were 13 exmployed for the synthesis of |-PMMA-b-PDMAEMA, |-PMMA-b-PS and |-PS-b- PMA.31 Baum et al. reported the synthesis of polystyrene-b-poly(N,N- dimethylacrylamide) (|-PS-b-PDMA) and |-PDMA-b-PMMA via SI-RAFT from surface-immobilized azo initiators in the presence of 2-phenylprop-2-yl dithiobenzoate as chain transfer agent.3 Well defined block copolymer brushes can be produced by a combination of ATRP and RAFT techniques.58 Specifically, the bromine end groups of homopolymer brushes prepared via SI- ATRP were modified with a RAFT agent which was subsequently used to grow a second block in the brushes.58 The phase separation of |-PS-b-PMMA brushes upon solvent treatment was explored to create patterned nanostructures as an array of micelles with an ellipsoidal shape.59 Kumar et al. demonstrated the application of diblock copolymer brushes prepared with SI-ATRP as potential stimuli responsive materials for the control of dye release triggered by changes in temperature, light and pH (Figure 1.3b).60 Not only diblock but triblock copolymer brushes have been prepared from a silicon substrate61 or gold substrate62 via SI-ATRP. 1.3.2 Bottlebrush Architecture. Bottlebrush polymers differ significantly from more conventional single strand polymers. The majority of these materials are uncharged with reported side chains having a degree of polymerization (DP) ranging from 20 to 60 or more. They provide a different structural motif than the typical single strand brush and, as a result of the bulky side chains, may place the backbone under stress. As an example, Sheiko et al. reported untethered bottlebrush polymer chains coated on a surface with a PMMA backbone and DP 25 side chains of n-butyl acrylate and studied their main chain breakage kinetics of these polymers.63 Persistence length scales as side chain length: DP ~10 is flexible and 14 DP~100 is a rigid rod. These studies indicate that relatively short side chains will not place undue stress on a backbone chain.64 Nature uses polyelectrolyte bottlebrush polymers to create specific interfaces and solves the backbone stress problem by using spacing between the side chains and also spacing between the charges.65 For example, in contrast to most synthetic PEBs the charge is not placed directly on the backbone but is instead placed in the side chains offering some relief from coulombic stress.66,67 In addition, charge stress is also minimized by avoiding placement of anionic or cationic groups on each repeat of the side chain. As a result, tethered polyelectrolyte bottlebrush polymers offer a new and interesting architecture for the investigation of polyelectrolyte rushes provided that the side arms are kept relatively short (10 7 while surfactants and calcium ions can trigger the conformation change of PLGA from coil to helix 142. In a similar fashion, poly(L-lysine) (PLL) can transform between an α-helix, a β-sheet and a random coil depending on external factors 143. For instance, PLL is a random coil at pH < 7 and transforms to a helix at pH > 9, but becomes a β-sheet at pH > 11. Also, ClO -4 anions can induce a transition of PLL from a coil to a helix at pH = 7 while a coil to a β-sheet transition can be induced by surfactant 143 . Yang et al. studied chemoresponsive PMLG brushes and found that the polymer changed from α-helix to β-sheet after exposure to formic acid vapor, but reversed back to an α-helix after immersion in a mixture of DCA and chloroform. Moreover, the transition using formic acid was slower for PMLG brushes than its free polymer 144. 1.4.1.2. Orientation of helical polypeptide brushes α-Helix polypeptide brushes have reported persistence lengths between 100 and 200 nm, so they form rod-like structures 145. Thus, they can have an orientation with respect to given substrate. FT-IR spectroscopy can be used to determine the orientation of α- helical polypeptide brushes 139,146,147. In the case of PBLG, the transition dipole moment 25 directions of amide bonds with respect to the helix axis are fixed. To be specific, the transition moment of amide I is mostly parallel to the chain axis while that of the amide II is roughly perpendicular to its axis. Therefore, the ratio D of the absorbance of amide I over the absorbance of amide II can be used to estimate a tilt angle, though different equations are applied depending on FT-IR measuring modes: attenuated total reflection (ATR), transmission or reflectance. Using ATR-FTIR, Machida reported an angle of 33o with respect to the substrate 148. For reflectance FT-IR, the measurements are usually done on a metal substrate and only the transition dipole moment component perpendicular to a surface can be measured, so it is very useful in detecting out-of-plane orientation. Using this technique, Wang reported an angle of 41o on an aluminum substrate while Enriquez et al. obtained an angle of 37o on a gold substrate for PBLG brushes 146,149. For the FTIR transmission mode, brushes are commonly prepared on a double-polished silicon substrate, and the IR beam is perpendicular to the substrate. Thus, only the components of the transition dipole moments of the amides which are parallel to the substrate can be measured. The angle can be calculated using the following equation (1) 139,150 1 [sin2 𝜃 cos2 𝛼 + sin2 𝛼 (1+cos2𝐼 𝐼 𝜃)] D = A 2Amide I/AAmide II = K 1 (1) [sin2 𝜃 cos2 𝛼𝐼𝐼+ sin2 𝛼𝐼𝐼(1+cos2 𝜃)]2 Where D is the dichroic ratio of the peak area of the amide I over the amide II, αI and αII are transition dipole moment angles of the amide I and the amide II bands respectively, θ is the average helix tilt angle, and K is a proportionality constant involving the intrinsic oscillator strengths of the amide I and amide II vibrational modes. The proportionality constant K can be obtained from the transmission FT-IR spectrum of a PBLG Langmuir Blodgett film which is known to have the helices parallel to the substrate, so its tilt angle θ is 90o 151 . Thus, for LB film, the intensities of the 26 amide I and amide II are highest and lowest respectively. The D ratio and θ = 90o can be used to determine K. Helical polypeptides are naturally non-centrosymmetric when aligned152. If oriented in a magnetic or electric field it is possible to introduce piezoelectric and non-linear optical properties to such materials153. Jaworek and coworkers demonstrated that directionally aligned helical polypeptide brushes had large dipole moments and possessed piezoelectric properties 154. A change in thickness as a response to applied electric field was measured. Although the inverse piezoelectric coefficient was low, the voltage sensitivity was similar to that of commercial piezoelectric materials 154. Machida et al. showed that a spiral texture of nematic liquid crystal was induced by PBLG brushes while it was not observed on a cast film or LB film 148. The authors suggested that the oriented PBLG with the tilt angle of 57o from substrate normal played a role on the appearance of the spiral texture 148. Attempts have been made to control the orientation of polypeptide. Whitesell et al. reported the ability to directionally align helical polyalanine and polyphenylalanine on Figure 1.6. Solvent quenching treatment of PBLG brushes. Adapted with permission from ref 146 27 gold by the use of a designed aminotrithiol which has a spacing when attached on a surface similar to the helical diameter of polypeptide chains 155. Luijten et al. used a chemical cross-linking method to lock polyglutamate brushes in a perpendicular orientation permanently 150. Specifically, polymer brushes with polymerizable side groups were swelled in a polymerizable solvent and cross-linked via radical polymerization. After cross-linking, the helix tilt angle was 11o, indicating a nearly perpendicular orientation, which was significantly improved from the parallel orientation with the tilt angle of only 66o before swelling. It resulted in an increase in film thickness from 151 Å to 390 Å, and the perpendicular orientation was stable and could not be changed by any solvent treatments 150. Wang used a different approach called solvent quenching to promote unidirectional orientation of polymer brushes 146. Initially the PBLG rod brushes were swollen in a good solvent (chloroform in this case) and the rod segments extended. Subsequently the brush film was placed in a non-solvent such as acetone, extracting chloroform, and aggregating the PBLG rods which assembled together in stretched chain bundles to minimize contact with acetone. The rod bundles were retained after drying the solvent. The tilt angle was 49o and 3o from surface normal before and after quenching respectively. The rod bundles were stable in air or aqueous conditions for extended times, and other pairs of good and bad solvents gave similar results (Figure 1.6) 146. Yang et al. used this method to quench PMLG with a solvent pair of chloroform and ethanol, and observed significant changes in surface morphology, surface wettability, and film thickness 144. Wu et al. demonstrated the dependence of the electrical property of helical polypeptide brushes on rod orientation 156. Three different polymer films of spin-coated PBLG film, collapsed polymer brush, and quenched polymer brushes having tilt angles of 75o, 57o and 13o from surface normal respectively, were investigated in the study. The electrical properties varied from insulator for the spin-coated film to diode for quenched brushes. Moreover, the 28 vertically aligned brushes had a high rectification ratio comparable to that of an organic diode 156. 1.4.2 Mesogen Jacketed Liquid Crystalline Polymer (MJLCP) 1.4.2.1. Molecular design and synthesis The mesogen jacketed liquid crystalline rod-like polymers is a unique class of liquid crystalline (LC) polymer which was introduced by Q.-F. Zhou at Peking University in the late 1980s 157–159. Large pendent mesogenic groups are laterally attached to polymer Figure 1.7. Schematic illustration of mesogen-jacketed liquid crystalline polymer. The bulky mesogenic side groups force the backbone to stretch into rod-like conformation. Adapted with permission from ref 160 backbone via a C-C bond or very short linkage 160. More importantly, the linkages are strategically located at the center of gravity centers of the mesogenic side groups to minimize the torque exerted by the backbone on the side chain 160. Consequently, the mesogenic groups form a dense jacket around the main backbone which causes a steric effect to force the main chains into an extended conformation 161. Thus, the MJLCP chains can act as a supramolecular rod and have similar properties to main-chain liquid crystalline polymers (MCLCPs), though it is a class of side-on side-chain liquid crystalline polymers (SCLCPs) (Figure 1.7) 160,162. 29 The unique structure of MJLCP leads to an increase in persistence length and reported values for MJLCPs are between 11.5 to 13.5 nm163, significantly greater than what would normally be expected for a substituted vinyl polymer. Moreover, it is much easier to control the sizes of the MJLCP rods compared to other rod-type polymers. Specifically, the diameter of the rod can be controlled by changing the lengths of both rigid cores and flexible tails of the mesogenic side groups while the length of the rod can be tuned by varying the molecular weight (MW) of the polymers 164. MJLCPs have been synthesized mostly from 2,5-disubstitutied styrene with a C-C spacer 162. Monomers based on 2-vinylhydroquinone 165, 2-vinyl-1,4-phenylene-diamine 166, 2- vinylterephthalic acid 167, and vinyl terphenyl 168 have been reported. Alternative structures have been investigated with a methacrylate unit connected to the mesogenic group via a short linkage and polymers with similar properties have been produced 161,169. Considering the chemical structures of the monomers, controlled radical polymerization such as ATRP and NMP have been mainly used to synthesize MJLCPs and more complex architectures such as block-copolymers, bottlebrushes, superbranched or star polymers 162. Pragliola showed one of the first examples for the synthesis of MJLCPs via NMP 170. As an example, poly{2,5-(4 butylbenzoyl)oxystyrene} (PBBOS) was produced with a low PDI. In addition, the polymerization of PBBOS has a much higher reaction rate and greater conversion compared to polystyrene (PS) under identical reaction conditions. After that, Gopalan further examined the reaction mechanism and found that the liquid crystalline melt phase of BBOS can cause preordering effects of the monomer that speeds the polymerization 171. Zhang et al. demonstrated the ATRP of poly{2,5-bis[4- methoxyphenyl]oxycarbonyl)styrene} (PMPCS) 172. The living polymerization was shown by the linearity of the plots of ln([M]o/[M]) versus time and molecular weight versus conversion as well as the low PDI (< 1.3) throughout the polymerization. 30 1.4.2.2. Rod-coil diblock copolymers of MJLCP The advantages of highly controlled molecular size and persistence length of MJLCPs make them excellent candidates as rod-type segments for block copolymers. Nanostructures and properties of block copolymers containing MJLCP have been reported 164. Ober et al. showed a thorough study of rod-coil diblock, triblock and starblock copolymer with PBBOS as rod-block and polystyrene as coil block 173. Nanostructured phased separation of the block copolymers was confirmed with SAXS and TEM. The effects of molecular architecture on melt rheology were also reported. Li Figure 1.8. a) Chemical structure of PS-b-PMPCS; b) PS is the coil segment and PMPCS is the rod segment; c) two layers of rods between layers of coils. The rods are perpendicular to the layers; d) lamellar structure of PS-b-PMPCS. Adapted with permission from ref. 174 31 et al. investigated phase structures and morphologies of rod-coil diblock copolymers, PS−b-PMPCS 174. Lamellar microphase structure was observed for the diblock copolymers with an increase of d spacing at higher MW of both PS and PMPCS. Moreover, PMPCS rods form columns which are parallel to the lamellar normal, and there are two layers of columns in each MJLCP domain (Figure 1.8) 174. However, a decrease in the degree of polymerization of PMPCS from 34 to 32 changed the structure of the diblock copolymer to a tetragonal perforated layer 175. Furthermore, Shi et al. did a systematic study on PDMS-b-PMPCS diblock copolymers and observed various nanostructures ranging from a lamellar structure (LAM), and double gyroid structure (GYR) to a hexagonally packed cylinder (HEX) by simply changing the rod length of PMPCS 176. Moreover, increasing annealing temperature can cause phase transitions such as GYR to Fddd or HEX to a body centered cubic structure (BCC). 1.4.2.3. Rod-rod diblock copolymers of MJLCP A rod-rod is another interesting type of block copolymers. Zhou et al. reported the synthesis and self-assembly of rod-rod diblock copolymers of MJLCP and helical polypeptides 177. In particular, PMPCS and PBLG were synthesized via ATRP and ROP respectively before being connected via click chemistry. A lamellar nanostructure was observed when the volume fraction of PBLG (fPBLG) was 0.5. In this structure, PMPCS blocks were packed in a columnar nematic phase while PBLG blocks assembled into HEX structures (Figure 1.9) 177. When fPBLG increased to 0.69, a cylinder nanostructure was observed with the core of the cylinders made of PMPCS rods while the surrounding matrix was PBLG packed in HEX phase. After that, the effect of size disparity on the self-assembly of poly(octyl-4′-(octyloxy)-2-vinylbiphenyl-4-carboxylate)-b-poly(γ- benzyl-l-glutamate) (PVBP-b-PBLG) was examined when the rod diameter of PVBP is 32 significantly larger than that of PBLG 178. With an increase of fPVBP, the nanostructure of this rod-rod block copolymer changed from interdigitated lamellar to a bilayer lamellar structure. Also, increasing fPVBP disrupted the ordered structure of the HEX phase of PBLG while improving the order of the columnar nematic phase of PVBP. Despite the significant progress in the synthesis, properties and self-assembly behaviors of MJLCPs and their block copolymers, no study of MJLCP brushes has been Figure 1.9. Self-assembly of PMPCS-b-PBLG rod-rod diblock copolymer with an increase in volume fraction of PBLG a) poorly order packing; b) hexagon in lamella; d) hexagon in cylinder; c) and e) shearing geometry with x-axis as the shear direction. Adapted with permission from ref. 177. 33 ever reported. Given a great disparity in persistence length and molecular structure between rod and coil-type polymers as well as the reported differences in phase separation between coil-coil, rod-coil and rod-rod block copolymers, there are immense opportunities to explore MJLCP brushes and other architectures such as block copolymer brushes, gradient brushes, or mixed brushes. 1.5. Solvent annealing Solvent annealing has been demonstrated to play an important role to inducing long range order during the self-assembly of block copolymers 179. A solvent can swell polymers and enhancing the polymer mobility needed to form well-organized nanostructures 180. Various morphologies such as lamellae, gyroids, hexagonal or spherical structures may be tuned by controlling solvent conditions 181. Similarly, solvent treatment has been used to promote phase separation in mixed brushes. Lemieux et al. reported two distinct morphologies for mixed brushes of poly(methyl acrylate) (PMA) and poly(styrene-co-2,3,4,5,6-pentafluorostyrene) (PSF) when they were exposed to acetone, a selective solvent for PMA, and toluene, a selective solvent for PSF. A similar observation was reported when PS/PMMA mixed brushes were treated with selective solvents 100. Moreover, Zhao et al. demonstrated the formation of relatively ordered nanodomains when PS/PMMA mixed brushes were exposed to acetic acid. Minko observed a ripple phase for PS/PMMA mixed brushes when a non-selective solvent was used 97. Frederickson explored the effect of polymer compositions and solvent selectivity on the phase separation of PS/PMMA mixed brushes 98. After solvent annealing in THF, resembled 2D hexagonal phase was observed at low concentration of PS while at more equal concentration of two brushes, an interwoven continuous morphology was observed. Moreover, the phase morphology changed from cylindrical to ripple phases when solvent was switched from THF to benzene. Regarding rod-type 34 polymer brushes, as mentioned earlier, solvent quenching may be used to promote vertical alignment of helical polypeptide brushes. 1.6. Summary Rod-type polymers are vastly different from coil-type polymers in terms of molecular structure and persistence length. One unique characteristic of rod-type polymer brushes is the tilting order as demonstrated with helical polypeptide brushes; however, there is still a limited understanding on how to control such rod orientation. Moreover, as MJLCP brushes are a completely unexplored area, there is similar a lack of knowledge on tilting behavior of MJLCP brushes. Nonetheless, a recent theoretical study showed that main-chain liquid-crystalline polymer (MCLCP) brushes may have a tilting order. Since MJLCP chains can act as supramolecular rods, we can expect that MJLCP brushes have similar tilting behaviors as MCLP brushes. In addition, the tilting order plays a critical role in such properties of polymer brushes as their piezoelectric and dielectric properties, or even the alignment of liquid crystals which is crucial to LC display technology 46,182,183. Besides, as now understood from block copolymers 184,185, increased complexity provides more types of phases to discover. Thus, rod-coil mixed brushes which have not been previously studied, will offer new systems to explore. Thus the phase immiscibility and confinement provided by coil brushes may enable us to vertically align rod-type brushes and create periodic surface topography with specific surface chemical functions and highly anisotropic properties. In this thesis, the focus will be on the synthesis of rod-type polymer brushes, helical polypeptide and MJLCP, and their mixed brushes. Several fundamental questions will be addressed to understand important aspects of mixed brush and rod tilting behaviors: 1) What are the parameters which can affect the orientation of rod brushes and to what degree? 2) What is the mechanism behind the solvent quenching treatment? 3) How can 35 nanopatterning and confinement be used to aid in orienting rod brushes? 4) What role does the coil brush have in the alignment of rod brushes and surface morphology? With these questions in mind, the purpose of Chapter 2 is to provide fundamental insight into the effects of coil-type polymer brushes and solvent treatment on the orientation of helical polypeptide brushes. PBLG and PMMA brushes were prepared via SI-ROP and SI-ATRP respectively from mixed initiators. The molecular weight (MW) of PMMA brushes was varied to change its volume fraction. The helical orientation, rod self-assembly and surface morphology were characterized before and after solvent quenching. The presence of coil brushes clearly had a significant effect on the behavior of the helical peptides, and the effect was stronger with an increase of MW of PMMA. Moreover, this first ever example of mixed rod/coil brushes showed distinct differences compared to coil/coil mixed brushes and rod-coil diblock copolymers. In Chapter 3, we aimed to get a comprehensive picture of rod orientation and self- assembly of helical polypeptide brushes. Various parameters including polymer chain length, solvent quality, solvent mixture compositions, solvent quenching time, and persistence length were examined. These results clarified the mechanism underlying solvent quenching treatment. They can also explain discrepancies in the reported rod orientations from the literature. In previous studies, polymer brushes were prepared under different conditions which could lead to different MW and grafting density, and those can affect orientation. Moreover, the results may allow us to establish some design criteria which can be used to better control rod orientation and surface morphology. In Chapter 4, a bottom-up approach was used to pattern polypeptide brushes at the nanoscale. E-beam lithography was used to pattern a resist before the initiator was deposited in vapor phase into the exposed area. Through a systematic approach, we can grow polypeptide brushes into well-defined dots with diameter as small as 60 nm. Also, spacing between spiked patterns can be precisely controlled. After that, poly(NIPAM) 36 brushes, a thermoresponsive polymer, were successfully grown in the surrounding area to create patterned binary brushes which can be used as a platform for biological studies in the future. In Chapter 5, the major aim was to grow MJLCP brushes and its rod/coil mixed brushes. Compared to free polymer synthesis, a much greater amount of monomer is needed to completely cover a substrate so that homogenous and uniform brushes can be grown on a surface. Thus, we designed a simplified multi-gram scale monomer synthesis with significantly higher yield than reported in the literature. Then PBBOS brushes were prepared with well-controlled thickness via SI-NMP. A fluorinated polymer was prepared via SI-ATRP as coil-type brush to enhance the immiscibility and phase separation in mixed brushes. Solvent quenching treatment was applied for vertical alignment of the MJLCP rod to take place. The surface topography had teepee structures similar to one observed for polypeptide brushes. Volume fractions of the two polymers were varied by controlling the polymer thicknesses. The results suggested that the coil provides a pressure through phase incompatibility and confinement to promote vertical alignment of the MJLCP brush. Moreover, surface topography and wettability can be tuned by simply changing polymer thickness. In Chapter 6, we demonstrated the synthesis of polymer brushes via surface initiated- single electron transfer-living radical polymerization (SI-SET-LRP) using Cu surface and hydrazine in the presence of air. By removing the deoxygenation step, this method is more facile and effective than conventional control radical polymerization. The living character of this method was verified by the growth of block copolymer brushes. This method also allowed the preparation of gradient brushes by tilting the copper surface with respect to initiator-immobilized sample. Finally, in the summary chapter prospects for future studies of these rod-like brushes are given. 37 1.7. References 1. Matyjaszewski, K. et al. Polymers at Interfaces: Using Atom Transfer Radical Polymerization in the Controlled Growth of Homopolymers and Block Copolymers from Silicon Surfaces in the Absence of Untethered Sacrificial Initiator. Macromolecules 32, 8716–8724 (1999). 2. Hawker, C. J., Bosman, A. W. & Harth, E. New Polymer Synthesis by Nitroxide Mediated Living Radical Polymerizations. Chem. Rev. 101, 3661–3688 (2001). 3. Baum, M. & Brittain, W. J. Synthesis of Polymer Brushes on Silicate Substrates via Reversible Addition Fragmentation Chain Transfer Technique. Macromolecules 35, 610–615 (2002). 4. Zhou, F. & Huck, W. T. S. Surface grafted polymer brushes as ideal building blocks for “smart” surfaces. Phys Chem Chem Phys 8, 3815–3823 (2006). 5. Tovar, G., Paul, S., Knoll, W., Prucker, O. & Rühe, J. Patterning molecularly thin films of polymers—new methods for photolithographic structuring of surfaces. Supramol. Sci. 2, 89–98 (1995). 6. Ji, S. et al. Generalization of the Use of Random Copolymers To Control the Wetting Behavior of Block Copolymer Films. Macromolecules 41, 9098–9103 (2008). 7. Chen, X. & Armes, S. P. Surface Polymerization of Hydrophilic Methacrylates from Ultrafine Silica Sols in Protic Media at Ambient Temperature: A Novel Approach to Surface Functionalization Using a Polyelectrolytic Macroinitiator. Adv. Mater. 15, 1558–1562 (2003). 38 8. Husemann, M. et al. Surface-Initiated Polymerization for Amplification of Self- Assembled Monolayers Patterned by Microcontact Printing. Angew. Chem. Int. Ed. 38, 647–649 (1999). 9. Zhao, B. & Brittain, W. J. Polymer brushes: surface-immobilized macromolecules. Prog. Polym. Sci. 25, 677–710 (2000). 10. von Werne, T. A. et al. A Versatile Method for Tuning the Chemistry and Size of Nanoscopic Features by Living Free Radical Polymerization. J. Am. Chem. Soc. 125, 3831–3838 (2003). 11. Singhvi, R. et al. Engineering Cell Shape and Function. Science 264, 696–698 (1994). 12. Chen, C. S., Mrksich, M., Huang, S., Whitesides, G. M. & Ingber, D. E. Geometric Control of Cell Life and Death. Science 276, 1425–1428 (1997). 13. Aksay, I. A. et al. Biomimetic Pathways for Assembling Inorganic Thin Films. Science 273, 872–898. 14. Balazs, A. C., Singh, C., Zhulina, E., Gersappe, D. & Pickett, G. Patterned polymer films. MRS Bull. 22, 16–21 (1997). 15. Amiji, M. & Park, K. Surface modification of polymeric biomaterials with poly(ethylene oxide), albumin, and heparin for reduced thrombogenicity. J. Biomater. Sci. Polym. Ed. 4, 217–234 (1993). 16. Houbenov, N., Minko, S. & Stamm, M. Mixed polyelectrolyte brush from oppositely charged polymers for switching of surface charge and composition in aqueous environment. Macromolecules 36, 5897–5901 (2003). 39 17. Ejaz, M., Yamamoto, S., Ohno, K., Tsujii, Y. & Fukuda, T. Controlled graft polymerization of methyl methacrylate on silicon substrate by the combined use of the Langmuir-Blodgett and atom transfer radical polymerization techniques. Macromolecules 31, 5934–5936 (1998). 18. Barbey, R. et al. Polymer Brushes via Surface-Initiated Controlled Radical Polymerization: Synthesis, Characterization, Properties, and Applications. Chem. Rev. 109, 5437–5527 (2009). 19. Matyjaszewski, K., Göbelt, B., Paik, H. & Horwitz, C. P. Tridentate Nitrogen- Based Ligands in Cu-Based ATRP: A Structure−Activity Study. Macromolecules 34, 430–440 (2001). 20. Jones, D. M. & Huck, W. T. S. Controlled Surface-Initiated Polymerizations in Aqueous Media. Adv. Mater. 13, 1256–1259 (2001). 21. Huang, W., Kim, J.-B., Bruening, M. L. & Baker, G. L. Functionalization of Surfaces by Water-Accelerated Atom-Transfer Radical Polymerization of Hydroxyethyl Methacrylate and Subsequent Derivatization. Macromolecules 35, 1175–1179 (2002). 22. Sedjo, R. A., Mirous, B. K. & Brittain, W. J. Synthesis of Polystyrene- block - poly(methyl methacrylate) Brushes by Reverse Atom Transfer Radical Polymerization. Macromolecules 33, 1492–1493 (2000). 23. Matyjaszewski, K., Dong, H., Jakubowski, W., Pietrasik, J. & Kusumo, A. Grafting from Surfaces for “Everyone”: ARGET ATRP in the Presence of Air. Langmuir 23, 4528–4531 (2007). 40 24. Li, B., Yu, B., Huck, W. T. S., Zhou, F. & Liu, W. Electrochemically Induced Surface-Initiated Atom-Transfer Radical Polymerization. Angew. Chem. Int. Ed. 51, 5092–5095 (2012). 25. Li, B., Yu, B., Huck, W. T. S., Liu, W. & Zhou, F. Electrochemically Mediated Atom Transfer Radical Polymerization on Nonconducting Substrates: Controlled Brush Growth through Catalyst Diffusion. J. Am. Chem. Soc. 135, 1708–1710 (2013). 26. Yan, J. et al. Controlled Polymer-Brush Growth from Microliter Volumes using Sacrificial-Anode Atom-Transfer Radical Polymerization. Angew. Chem. Int. Ed. 52, 9125–9129 (2013). 27. Chiefari, J. et al. Living free-radical polymerization by reversible addition- fragmentation chain transfer: the RAFT process. Macromolecules 31, 5559–5562 (1998). 28. Peng, Q., Lai, D. M. Y., Kang, E. T. & Neoh, K. G. Preparation of Polymer−Silicon(100) Hybrids via Interface-Initiated Reversible Addition- Fragmentation Chain-Transfer (RAFT) Polymerization. Macromolecules 39, 5577–5582 (2006). 29. Stenzel, M. H., Zhang, L. & Huck, W. T. S. Temperature-Responsive Glycopolymer Brushes Synthesized via RAFT Polymerization Using the Z-group Approach. Macromol. Rapid Commun. 27, 1121–1126 (2006). 30. Zhao & Perrier, S. Synthesis of Well-Defined Homopolymer and Diblock Copolymer Grafted onto Silica Particles by Z-Supported RAFT Polymerization. Macromolecules 39, 8603–8608 (2006). 41 31. Rowe-Konopacki, M. D. & Boyes, S. G. Synthesis of Surface Initiated Diblock Copolymer Brushes from Flat Silicon Substrates Utilizing the RAFT Polymerization Technique. Macromolecules 40, 879–888 (2007). 32. Zhao, Y. & Perrier, S. Reversible Addition−Fragmentation Chain Transfer Graft Polymerization Mediated by Fumed Silica Supported Chain Transfer Agents. Macromolecules 40, 9116–9124 (2007). 33. Tsujii, Y., Ejaz, M., Sato, K., Goto, A. & Fukuda, T. Mechanism and Kinetics of RAFT-Mediated Graft Polymerization of Styrene on a Solid Surface. 1. Experimental Evidence of Surface Radical Migration. Macromolecules 34, 8872–8878 (2001). 34. Tria, M. C. R., Grande, C. D. T., Ponnapati, R. R. & Advincula, R. C. Electrochemical Deposition and Surface-Initiated RAFT Polymerization: Protein and Cell-Resistant PPEGMEMA Polymer Brushes. Biomacromolecules 11, 3422–3431 (2010). 35. Grande, C. D., Tria, M. C., Jiang, G., Ponnapati, R. & Advincula, R. Surface- Grafted Polymers from Electropolymerized Polythiophene RAFT Agent. Macromolecules 44, 966–975 (2011). 36. Zhai, G. et al. Functionalization of Hydrogen-Terminated Silicon with Polybetaine Brushes via Surface-Initiated Reversible Addition−Fragmentation Chain-Transfer (RAFT) Polymerization. Ind. Eng. Chem. Res. 43, 1673–1680 (2004). 37. Yu, W. H., Kang, E. T. & Neoh, K. G. Functionalization of Hydrogen- Terminated Si(100) Substrate by Surface-Initiated RAFT Polymerization of 4- 42 Vinylbenzyl Chloride and Subsequent Derivatization for Photoinduced Metallization. Ind. Eng. Chem. Res. 43, 5194–5202 (2004). 38. Hawker, C. J. “Living” free radical polymerization: a unique technique for the preparation of controlled macromolecular architectures. Acc. Chem. Res. 30, 373–382 (1997). 39. Brinks, M. K. & Studer, A. Polymer Brushes by Nitroxide-Mediated Polymerization. Macromol. Rapid Commun. 30, 1043–1057 (2009). 40. Husseman, M. et al. Controlled Synthesis of Polymer Brushes by “Living” Free Radical Polymerization Techniques. Macromolecules 32, 1424–1431 (1999). 41. Li, J., Chen, X. & Chang, Y.-C. Preparation of End-Grafted Polymer Brushes by Nitroxide-Mediated Free Radical Polymerization of Vaporized Vinyl Monomers. Langmuir 21, 9562–9567 (2005). 42. Parvole, J., Montfort, J.-P., Reiter, G., Borisov, O. & Billon, L. Elastomer polymer brushes on flat surface by bimolecular surface-initiated nitroxide mediated polymerization. Polymer 47, 972–981 (2006). 43. Husemann, M. et al. Manipulation of Surface Properties by Patterning of Covalently Bound Polymer Brushes. J. Am. Chem. Soc. 122, 1844–1845 (2000). 44. Ignatova, M. et al. Synthesis of Copolymer Brushes Endowed with Adhesion to Stainless Steel Surfaces and Antibacterial Properties by Controlled Nitroxide- Mediated Radical Polymerization. Langmuir 20, 10718–10726 (2004). 45. Hadjichristidis, N., Iatrou, H., Pitsikalis, M. & Sakellariou, G. Synthesis of Well- Defined Polypeptide-Based Materials via the Ring-Opening Polymerization of α- Amino Acid N-Carboxyanhydrides. Chem. Rev. 109, 5528–5578 (2009). 43 46. Shen, Y., Li, Z. & Klok, H.-A. Polypeptide brushes grown via surface-initiated ring-opening polymerization of α-amino acid N-carboxyanhydrides. Chin. J. Polym. Sci. 33, 931–946 (2015). 47. Wibowo, S. H., Sulistio, A., Wong, E. H. H., Blencowe, A. & Qiao, G. G. Polypeptide films via N-carboxyanhydride ring-opening polymerization (NCA- ROP): past, present and future. Chem. Commun. 50, 4971 (2014). 48. Li, J., He, W.-D., Yang, L.-P., Sun, X.-L. & Hua, Q. Preparation of multi-walled carbon nanotubes grafted with synthetic poly(l-lysine) through surface-initiated ring-opening polymerization. Polymer 48, 4352–4360 (2007). 49. Yao, Y., Li, W., Wang, S., Yan, D. & Chen, X. Polypeptide Modification of Multiwalled Carbon Nanotubes by a Graft-From Approach. Macromol. Rapid Commun. 27, 2019–2025 (2006). 50. Liu, Z.-M. et al. Surface modification of microporous polypropylene membranes by the grafting of poly(γ-stearyl-l-glutamate). Eur. Polym. J. 39, 2291–2299 (2003). 51. Wieringa, R. H. et al. Surface grafting of poly (L-glutamates). 1. Synthesis and characterization. Langmuir 17, 6477–6484 (2001). 52. Wieringa, R. H., Siesling, E. A., Werkman, P. J., Vorenkamp, E. J. & Schouten, A. J. Surface Grafting of Poly( L -glutamates). 3. Block Copolymerization. Langmuir 17, 6491–6495 (2001). 53. Wieringa, R. H. & Schouten, A. J. Oriented thin film formation by surface graft polymerization of γ-methyl L-glutamate N-carboxyanhydride in the melt. Macromolecules 29, 3032–3034 (1996). 44 54. Chang, Y.-C. & Frank, C. W. Vapor deposition-polymerization of α-amino acid N-carboxy anhydride on the silicon (100) native oxide surface. Langmuir 14, 326–334 (1998). 55. Lee, N. H. & Frank, C. W. Surface-Initiated Vapor Polymerization of Various α- Amino Acids. Langmuir 19, 1295–1303 (2003). 56. Wang, Y. & Chang, Y.-C. Grafting of Homo- and Block Co-polypeptides on Solid Substrates by an Improved Surface-Initiated Vapor Deposition Polymerization. Langmuir 18, 9859–9866 (2002). 57. Zheng, W. & Frank, C. W. Surface-Initiated Vapor Deposition Polymerization of Poly(γ-benzyl- L -glutamate): Optimization and Mechanistic Studies. Langmuir 26, 3929–3941 (2010). 58. Rowe, M. D., Hammer, B. A. G. & Boyes, S. G. Synthesis of Surface-Initiated Stimuli-Responsive Diblock Copolymer Brushes Utilizing a Combination of ATRP and RAFT Polymerization Techniques. Macromolecules 41, 4147–4157 (2008). 59. Zhao, B., Brittain, W. J., Zhou, W. & Cheng, S. Z. Nanopattern formation from tethered PS-b-PMMA brushes upon treatment with selective solvents. J. Am. Chem. Soc. 122, 2407–2408 (2000). 60. Kumar, S., Dory, Y. L., Lepage, M. & Zhao, Y. Surface-Grafted Stimuli- Responsive Block Copolymer Brushes for the Thermo-, Photo- and pH-Sensitive Release of Dye Molecules. Macromolecules 44, 7385–7393 (2011). 45 61. Tomlinson, M. R. & Genzer, J. Formation and properties of multivariant assemblies of surface-tethered diblock and triblock copolymers. Polymer 49, 4837–4845 (2008). 62. Kim, J.-B., Huang, W., Bruening, M. L. & Baker, G. L. Synthesis of Triblock Copolymer Brushes by Surface-Initiated Atom Transfer Radical Polymerization. Macromolecules 35, 5410–5416 (2002). 63. Li, Y. et al. Sonication-induced scission of molecular bottlebrushes: Implications of the “hairy” architecture. Polymer 84, 178–184 (2016). 64. Sheiko, S. S. et al. Adsorption-induced scission of carbon–carbon bonds. Nature 440, 191–194 (2006). 65. Schaefer, L. & Schaefer, R. M. Proteoglycans: from structural compounds to signaling molecules. Cell Tissue Res. 339, 237–246 (2010). 66. Imberty, A., Lortat-Jacob, H. & Pérez, S. Structural view of glycosaminoglycan– protein interactions. Carbohydr. Res. 342, 430–439 (2007). 67. Bandtlow, C. E. & Zimmermann, D. R. Proteoglycans in the Developing Brain: New Conceptual Insights for Old Proteins. Physiol. Rev. 80, 1267–1290 (2000). 68. Wu, D., Zhao, C., Tian, J. & Zhao, H. Synthesis of PLLA-PEOMA comb- block -comb type molecular brushes based on AGET ATRP and ring-opening polymerization. Polym. Int. 58, 1335–1340 (2009). 69. Narupai, B. et al. Hierarchical comb brush architectures via sequential light- mediated controlled radical polymerizations. J. Polym. Sci. Part Polym. Chem. 54, 2276–2284 (2016). 46 70. Chang, Y.-C. & Frank, C. W. Vapor Deposition−Polymerization of α-Amino Acid N-Carboxy Anhydride on the Silicon(100) Native Oxide Surface. Langmuir 14, 326–334 (1998). 71. Shen, Y. et al. Shape-Persistent, Thermoresponsive Polypeptide Brushes Prepared by Vapor Deposition Surface-Initiated Ring-Opening Polymerization of α-Amino Acid N-Carboxyanhydrides. Macromolecules 48, 2399–2406 (2015). 72. Olanya, G., Thormann, E., Varga, I., Makuška, R. & Claesson, P. M. Protein interactions with bottle-brush polymer layers: Effect of side chain and charge density ratio probed by QCM-D and AFM. J. Colloid Interface Sci. 349, 265– 274 (2010). 73. Xu, Y. et al. pH and salt responsive poly(N,N-dimethylaminoethyl methacrylate) cylindrical brushes and their quaternized derivatives. Polymer 49, 3957–3964 (2008). 74. Lienkamp, K., Schnell, I., Groehn, F. & Wegner, G. Polymerization of Styrene Sulfonate Ethyl Ester by ATRP: Synthesis and Characterization of Macromonomers for Suzuki Polycondensation. Macromol. Chem. Phys. 207, 2066–2073 (2006). 75. Ruthard, C., Maskos, M., Kolb, U. & Gröhn, F. Finite-Size Networks from Cylindrical Polyelectrolyte Brushes and Porphyrins. Macromolecules 42, 830– 840 (2009). 76. Störkle, D., Duschner, S., Heimann, N., Maskos, M. & Schmidt, M. Complex Formation of DNA with Oppositely Charged Polyelectrolytes of Different Chain 47 Topology:  Cylindrical Brushes and Dendrimers. Macromolecules 40, 7998–8006 (2007). 77. Wang, M. et al. A Water-Soluble pH-Responsive Molecular Brush of Poly(N,N- dimethylaminoethyl methacrylate) Grafted Polythiophene. Macromolecules 41, 6993–7002 (2008). 78. Wintermantel, M. et al. Molecular Bottlebrushes. Macromolecules 29, 978–983 (1996). 79. Ballauff, M. & Borisov, O. Polyelectrolyte brushes. Curr. Opin. Colloid Interface Sci. 11, 316–323 (2006). 80. Pincus, P. Colloid stabilization with grafted polyelectrolytes. Macromolecules 24, 2912–2919 (1991). 81. Borisov, O. V. & Zhulina, E. B. Effects of ionic strength and charge annealing in star-branched polyelectrolytes. Eur. Phys. J. B - Condens. Matter Complex Syst. 4, 205–217 (1998). 82. Xu, Y. et al. Manipulating cylindrical polyelectrolyte brushes on the nanoscale by counterions: collapse transition to helical structures. Soft Matter 5, 379–384 (2009). 83. Wu, T., Efimenko, K., Vlček, P., Šubr, V. & Genzer, J. Formation and Properties of Anchored Polymers with a Gradual Variation of Grafting Densities on Flat Substrates. Macromolecules 36, 2448–2453 (2003). 84. Wu, T. et al. Behavior of surface-anchored poly (acrylic acid) brushes with grafting density gradients on solid substrates: 1. Experiment. Macromolecules 40, 8756–8764 (2007). 48 85. Wang, X., Tu, H., Braun, P. V. & Bohn, P. W. Length Scale Heterogeneity in Lateral Gradients of Poly( N -isopropylacrylamide) Polymer Brushes Prepared by Surface-Initiated Atom Transfer Radical Polymerization Coupled with In-Plane Electrochemical Potential Gradients. Langmuir 22, 817–823 (2006). 86. Tomlinson, M. R. & Genzer, J. Formation of Grafted Macromolecular Assemblies with a Gradual Variation of Molecular Weight on Solid Substrates. Macromolecules 36, 3449–3451 (2003). 87. Tomlinson, M. R. & Genzer, J. Formation of surface-grafted copolymer brushes with continuous composition gradients. Chem. Commun. 1350 (2003) doi:10.1039/b303823h. 88. Xu, C., Wu, T., Drain, C. M., Batteas, J. D. & Beers, K. L. Microchannel Confined Surface-Initiated Polymerization. Macromolecules 38, 6–8 (2005). 89. Poelma, J. E., Fors, B. P., Meyers, G. F., Kramer, J. W. & Hawker, C. J. Fabrication of Complex Three-Dimensional Polymer Brush Nanostructures through Light-Mediated Living Radical Polymerization. Angew. Chem. Int. Ed. 52, 6844–6848 (2013). 90. Bhat, R. R., Tomlinson, M. R. & Genzer, J. Orthogonal surface-grafted polymer gradients: A versatile combinatorial platform. J. Polym. Sci. Part B Polym. Phys. 43, 3384–3394 (2005). 91. Wu, T., Efimenko, K. & Genzer, J. Combinatorial Study of the Mushroom-to- Brush Crossover in Surface Anchored Polyacrylamide. J. Am. Chem. Soc. 124, 9394–9395 (2002). 49 92. Bhat, R. R., Chaney, B. N., Rowley, J., Liebmann-Vinson, A. & Genzer, J. Tailoring Cell Adhesion Using Surface-Grafted Polymer Gradient Assemblies. Adv. Mater. 17, 2802–2807 (2005). 93. Motornov, M., Sheparovych, R., Katz, E. & Minko, S. Chemical Gating with Nanostructured Responsive Polymer Brushes: Mixed Brush versus Homopolymer Brush. ACS Nano 2, 41–52 (2008). 94. Kumar Vyas, M., Schneider, K., Nandan, B. & Stamm, M. Switching of friction by binary polymer brushes. Soft Matter 4, 1024 (2008). 95. Ionov, L. et al. Gradient Mixed Brushes: “Grafting To” Approach. Macromolecules 37, 7421–7423 (2004). 96. Lemieux, M. et al. Reorganization of Binary Polymer Brushes: Reversible Switching of Surface Microstructures and Nanomechanical Properties. Macromolecules 36, 7244–7255 (2003). 97. Usov, D. et al. Three-Dimensional Analysis of Switching Mechanism of Mixed Polymer Brushes. Macromolecules 40, 8774–8783 (2007). 98. Price, A. D., Hur, S.-M., Fredrickson, G. H., Frischknecht, A. L. & Huber, D. L. Exploring Lateral Microphase Separation in Mixed Polymer Brushes by Experiment and Self-Consistent Field Theory Simulations. Macromolecules 45, 510–524 (2012). 99. Feng, J., Haasch, R. T. & Dyer, D. J. Photoinitiated Synthesis of Mixed Polymer Brushes of Polystyrene and Poly(methyl methacrylate). Macromolecules 37, 9525–9537 (2004). 50 100. Santer, S., Kopyshev, A., Yang, H.-K. & Rühe, J. Local Composition of Nanophase-Separated Mixed Polymer Brushes. Macromolecules 39, 3056–3064 (2006). 101. Zhao, B. Synthesis of binary mixed homopolymer brushes by combining atom transfer radical polymerization and nitroxide-mediated radical polymerization. Polymer 44, 4079–4083 (2003). 102. Zhao, B. A Combinatorial Approach to Study Solvent-Induced Self-Assembly of Mixed Poly(methyl methacrylate)/Polystyrene Brushes on Planar Silica Substrates: Effect of Relative Grafting Density. Langmuir 20, 11748–11755 (2004). 103. Julthongpiput, D., Lin, Y.-H., Teng, J., Zubarev, E. R. & Tsukruk, V. V. Y- Shaped Polymer Brushes: Nanoscale Switchable Surfaces. Langmuir 19, 7832– 7836 (2003). 104. Zhao, B. & He, T. Synthesis of Well-Defined Mixed Poly(methyl methacrylate)/Polystyrene Brushes from an Asymmetric Difunctional Initiator- Terminated Self-Assembled Monolayer. Macromolecules 36, 8599–8602 (2003). 105. Calabrese, D. R. et al. Design, Synthesis, and Use of Y-Shaped ATRP/NMP Surface Tethered Initiator. ACS Macro Lett. 4, 606–610 (2015). 106. Huang, X. et al. Synthesis of Hetero-Polymer Functionalized Nanocarriers by Combining Surface-Initiated ATRP and RAFT Polymerization. Small 8, 3579– 3583 (2012). 107. Patra, M. & Linse, P. Simulation of Grafted Polymers on Nanopatterned Surfaces. Nano Lett. 6, 133–137 (2006). 51 108. Lee, W.-K., Patra, M., Linse, P. & Zauscher, S. Scaling Behavior of Nanopatterned Polymer Brushes. Small 3, 63–66 (2007). 109. Jonas, A. M., Hu, Z., Glinel, K. & Huck, W. T. S. Chain Entropy and Wetting Energy Control the Shape of Nanopatterned Polymer Brushes. Macromolecules 41, 6859–6863 (2008). 110. Dong, R., Krishnan, S., Baird, B. A., Lindau, M. & Ober, C. K. Patterned Biofunctional Poly(acrylic acid) Brushes on Silicon Surfaces. Biomacromolecules 8, 3082–3092 (2007). 111. Zhou, Z., Yu, P., Geller, H. M. & Ober, C. K. Biomimetic Polymer Brushes Containing Tethered Acetylcholine Analogs for Protein and Hippocampal Neuronal Cell Patterning. Biomacromolecules 14, 529–537 (2013). 112. Pallandre, A., Glinel, K., Jonas, A. M. & Nysten, B. Binary Nanopatterned Surfaces Prepared from Silane Monolayers. Nano Lett. 4, 365–371 (2004). 113. Ahn, S. J. et al. Surface-Initiated Polymerization on Nanopatterns Fabricated by Electron-Beam Lithography. Adv. Mater. 16, 2141–2145 (2004). 114. Kaholek, M., Lee, W.-K., LaMattina, B., Caster, K. C. & Zauscher, S. Fabrication of Stimulus-Responsive Nanopatterned Polymer Brushes by Scanning-Probe Lithography. Nano Lett. 4, 373–376 (2004). 115. Liu, X., Li, Y. & Zheng, Z. Programming nanostructures of polymer brushes by dip-pen nanodisplacement lithography (DNL). Nanoscale 2, 2614 (2010). 116. Schmelmer, U. et al. Surface-Initiated Polymerization on Self-Assembled Monolayers: Amplification of Patterns on the Micrometer and Nanometer Scale. Angew. Chem. Int. Ed. 42, 559–563 (2003). 52 117. Lee, W.-K., Caster, K. C., Kim, J. & Zauscher, S. Nanopatterned Polymer Brushes by Combining AFM Anodization Lithography with Ring-Opening Metathesis Polymerization in the Liquid and Vapor Phase. Small 2, 848–853 (2006). 118. Schmelmer, U. et al. Nanostructured Polymer Brushes. Small 3, 459–465 (2007). 119. Steenackers, M., Jordan, R., Küller, A. & Grunze, M. Engineered Polymer Brushes by Carbon Templating. Adv. Mater. 21, 2921–2925 (2009). 120. Steenackers, M. et al. Structured Polymer Brushes on Silicon Carbide. Chem. Mater. 22, 272–278 (2010). 121. Jones, D. M., Smith, J. R., Huck, W. T., Alexander, C. & others. Variable adhesion of micropatterned thermoresponsive polymer brushes: AFM investigations of poly (N-isopropylacrylamide) brushes prepared by surface- initiated polymerizations. Adv. Mater. 14, 1130 (2002). 122. Discekici, E. H. et al. Simple Benchtop Approach to Polymer Brush Nanostructures Using Visible-Light-Mediated Metal-Free Atom Transfer Radical Polymerization. ACS Macro Lett. 5, 258–262 (2016). 123. Prucker, O., Schimmel, M., Tovar, G., Knoll, W. & Rühe, J. Microstructuring of molecularly thin polymer layers by photolithography. Adv. Mater. 10, 1073–1077 (1998). 124. Rastogi, A., Paik, M. Y., Tanaka, M. & Ober, C. K. Direct Patterning of Intrinsically Electron Beam Sensitive Polymer Brushes. ACS Nano 4, 771–780 (2010). 53 125. Rastogi, A., Paik, M. Y. & Ober, C. K. Development of a Directly Patterned Low-Surface-Energy Polymer Brush in Supercritical Carbon Dioxide. ACS Appl. Mater. Interfaces 1, 2013–2020 (2009). 126. Paik, M. Y. et al. Patterning of Polymer Brushes. A Direct Approach to Complex, Sub-Surface Structures. Nano Lett. 10, 3873–3879 (2010). 127. Hoshi, Y., Xu, Y. & Ober, C. K. Photo-cleavable anti-fouling polymer brushes: A simple and versatile platform for multicomponent protein patterning. Polymer 54, 1762–1767 (2013). 128. Kamada, T. et al. Patterning of photocleavable zwitterionic polymer brush fabricated on silicon wafer. Colloids Surf. B Biointerfaces 123, 878–886 (2014). 129. Xu, Y., Hoshi, Y. & Ober, C. K. Photo-switchable polyelectrolyte brush for dual protein patterning. J. Mater. Chem. 21, 13789–13792 (2011). 130. Hirtz, M. et al. Structured Polymer Brushes by AFM Lithography. Small 5, 919– 923 (2009). 131. Edmondson, S. & Huck, W. T. Quasi-2D Polymer Objects from Patterned, Crosslinked Polymer Brushes. Adv. Mater. 16, 1327–1331 (2004). 132. Edmondson, S., Frieda, K., Comrie, J. E., Onck, P. R. & Huck, W. T. S. Buckling in Quasi-2D Polymers. Adv. Mater. 18, 724–728 (2006). 133. Kelby, T. S. & Huck, W. T. S. Controlled Bending of Microscale Au−Polyelectrolyte Brush Bilayers. Macromolecules 43, 5382–5386 (2010). 134. Welch, M. E. & Ober, C. K. Characterization of Polymer Brush Membranes via HF Etch Liftoff Technique. ACS Macro Lett. 2, 241–245 (2013). 54 135. Ohm, C. & Ober, C. K. From surface coatings to polymer nanofilms: lifting off polymer brushes. RSC Adv. 3, 18482 (2013). 136. Estillore, N. C. & Advincula, R. C. Free-Standing Films of Semifluorinated Block Copolymer Brushes from Layer-by-Layer Polyelectrolyte Macroinitiators. Macromol. Chem. Phys. 212, 1552–1566 (2011). 137. Kohri, M. et al. Facile Synthesis of Free-Standing Polymer Brush Films Based on a Colorless Polydopamine Thin Layer. Macromol. Rapid Commun. 34, 1220– 1224 (2013). 138. Shen, Y., Li, Z. & Klok, H.-A. Polypeptide brushes grown via surface-initiated ring-opening polymerization of α-amino acid N-carboxyanhydrides. Chin. J. Polym. Sci. 33, 931–946 (2015). 139. Wieringa, R. H. et al. Surface Grafting of Poly( L -glutamates). 2. Helix Orientation. Langmuir 17, 6485–6490 (2001). 140. Yang, C.-T., Wang, Y. & Chang, Y.-C. Effect of Solvents and Temperature on the Conformation of Poly(β-benzyl- L -aspartate) Brushes. Biomacromolecules 11, 1308–1313 (2010). 141. Luijten, J., Vorenkamp, E. J. & Schouten, A. J. Reversible Helix Sense Inversion in Surface-Grafted Poly(β-phenethyl- L -aspartate) Films. Langmuir 23, 10772– 10778 (2007). 142. Wang, Y. & Chang, Y. C. Synthesis and Conformational Transition of Surface- Tethered Polypeptide:  Poly(l-glutamic acid). Macromolecules 36, 6503–6510 (2003). 55 143. Wang, Y. & Chang, Y. C. Synthesis and Conformational Transition of Surface- Tethered Polypeptide:  Poly(l-lysine). Macromolecules 36, 6511–6518 (2003). 144. Yang, C.-T., Wang, Y., Frank, C. W. & Chang, Y.-C. Chemoresponsive surface- tethered polypeptide brushes based on switchable secondary conformations. RSC Adv 5, 86113–86119 (2015). 145. Temyanko, E., Russo, P. S. & Ricks, H. Study of Rodlike Homopolypeptides by Gel Permeation Chromatography with Light Scattering Detection:  Validity of Universal Calibration and Stiffness Assessment. Macromolecules 34, 582–586 (2001). 146. Wang, Y. & Chang, Y. C. Preparation of Unidirectional End-Grafted α-Helical Polypeptides by Solvent Quenching. J. Am. Chem. Soc. 125, 6376–6377 (2003). 147. Duran, H., Yameen, B., Khan, H. U., Förch, R. & Knoll, W. Surface-initiated ring opening polymerization of N-carboxy anhydride of benzyl-l-glutamate monomers on soft flexible substrates. React. Funct. Polym. 73, 606–612 (2013). 148. Machida, S. et al. A Chiral Director Field in the Nematic Liquid Crystal Phase Induced by a Poly(.gamma.-benzyl glutamate) Chemical Reaction Alignment Film. Langmuir 11, 4838–4843 (1995). 149. Enriquez, E. P. & Samulski, E. T. Self-Assembled α-Helical Polypeptide Films. MRS Online Proc. Libr. Arch. 255, (1991). 150. Luijten, J., Groeneveld, D. Y., Nijboer, G. W., Vorenkamp, E. J. & Schouten, A. J. Cross-Linking-Induced Permanently Perpendicular Helix Orientation in Surface-Grafted Polyglutamate Films. Langmuir 23, 8163–8169 (2007). 56 151. Duda, G. et al. Preparation and characterization of monolayers and multilayers of preformed polymers. Thin Solid Films 159, 221–230 (1988). 152. Knoesen, A. et al. Sum-frequency spectroscopy and imaging of aligned helical polypeptides. IEEE J. Sel. Top. Quantum Electron. 10, 1154–1163 (2004). 153. Yen, C.-C., Tokita, M., Park, B., Takezoe, H. & Watanabe, J. Spontaneous Organization of Helical Polypeptide Molecules into Polar Packing Structure. Macromolecules 39, 1313–1315 (2006). 154. Jaworek, T., Neher, D., Wegner, G., Wieringa, R. H. & Schouten, A. J. Electromechanical Properties of an Ultrathin Layer of Directionally Aligned Helical Polypeptides. Science 279, 57–60 (1998). 155. Whitesell, J. K. & Chang, H. K. Directionally Aligned Helical Peptides on Surfaces. Science 261, 73–76 (1993). 156. Wu, J.-C., Chen, C.-C., Chen, K.-H. & Chang, Y.-C. Controlled growth of aligned α-helical-polypeptide brushes for tunable electrical conductivity. Appl. Phys. Lett. 98, 133304 (2011). 157. Zhou, Q., Zhu, X. & Wen, Z. Liquid-crystalline side-chain polymers without flexible spacer. Macromolecules 22, 491–493 (1989). 158. Zhang, D., Liu, Y.-X., Wan, X.-H. & Zhou, Q.-F. Synthesis and Characterization of A New Series of “Mesogen-Jacketed Liquid Crystal Polymers” Based on the Newly Synthesized Vinylterephthalic Acid. Macromolecules 32, 5183–5185 (1999). 57 159. Chai, C.-P. et al. Synthesis and Phase Structures of Mesogen-Jacketed Liquid Crystalline Polymers Containing 1,3,4-Oxadiazole Based Side Chains. Macromolecules 40, 9361–9370 (2007). 160. Chen, X.-F. et al. Mesogen-jacketed liquid crystalline polymers. Chem. Soc. Rev. 39, 3072 (2010). 161. Zhou, Q. F., Li, H. M. & Feng, X. D. Synthesis of liquid-crystalline polyacrylates with laterally substituted mesogens. Macromolecules 20, 233–234 (1987). 162. Gao, L.-C., Fan, X.-H., Shen, Z.-H., Chen, X. & Zhou, Q.-F. Jacketed polymers: Controlled synthesis of mesogen-jacketed polymers and block copolymers. J. Polym. Sci. Part Polym. Chem. 47, 319–330 (2009). 163. Wan, X. et al. Characterization of the chain stiffness for a mesogen-jacketed liquid crystal polymer: Poly2,5-BIS [(4-Methoxybenzoyl)Oxy] Styrene. Macromol. Symp. 96, 207–218 (1995). 164. Lyu, X.-L., Pan, H.-B., Shen, Z.-H. & Fan, X.-H. Self-assembly and Properties of Block Copolymers Containing Mesogen-Jacketed Liquid Crystalline Polymers as Rod Blocks. Chin. J. Polym. Sci. 36, 811–821 (2018). 165. Zhou, Q., Zhu, X. & Wen, Z. Liquid-crystalline side-chain polymers without flexible spacer. Macromolecules 22, 491–493 (1989). 166. Zhang, D., Zhou, Q.-F., Ma, Y.-G., Wan, X.-H. & Feng, X.-D. Mesogen-jacketed Liquid Crystal Polymers With Mesogens of Aromatic Amide Structure. Polym. Adv. Technol. 8, 227–233 (1997). 167. Zhang, D., Liu, Y.-X., Wan, X.-H. & Zhou, Q.-F. Synthesis and Characterization of A New Series of “Mesogen-Jacketed Liquid Crystal Polymers” Based on the 58 Newly Synthesized Vinylterephthalic Acid. Macromolecules 32, 5183–5185 (1999). 168. Zhu, Z. et al. Synthesis and characterization of a thermotropic liquid-crystalline poly[2,5-bis(4′-alkoxycarbonylphenyl) styrene]. J. Polym. Sci. Part Polym. Chem. 45, 830–847 (2007). 169. Zhou, Q.-F., Li, H.-M. & Feng, X.-D. Synthesis of a New Class of Side Chain Liquid Crystal Polymers—Polymers with Mesogens Laterally Attached Via Short Linkages to Polymer Backbones. Mol. Cryst. Liq. Cryst. Inc. Nonlinear Opt. 155, 73–82 (1988). 170. Pragliola, S., Ober, C. K., Mather, P. T. & Jeon, H. G. Mesogen-jacketed liquid crystalline polymers via stable free radical polymerization. Macromol. Chem. Phys. 200, 2338–2344 (1999). 171. Gopalan, P. & Ober, C. K. Highly Reactive 2,5-Disubstituted Styrene-Based Monomer Polymerized via Stable Free Radical Polymerization: Effect of Substitution and Liquid Crystallinity on Polymerization. Macromolecules 34, 5120–5124 (2001). 172. Zhang, H., Yu, Z., Wan, X., Zhou, Q.-F. & Woo, E. M. Effects of molecular weight on liquid-crystalline behavior of a mesogen-jacketed liquid crystal polymer synthesized by atom transfer radical polymerization. Polymer 43, 2357– 2361 (2002). 173. Gopalan, P., Zhang, Y., Li, X., Wiesner, U. & Ober, C. K. Liquid Crystalline Rod−Coil Block Copolymers by Stable Free Radical Polymerization:  Synthesis, Morphology, and Rheology. Macromolecules 36, 3357–3364 (2003). 59 174. Li, C. Y. et al. Hierarchical Assembly of a Series of Rod−Coil Block Copolymers:  Supramolecular LC Phase in Nanoenviroment. Macromolecules 37, 2854–2860 (2004). 175. Tenneti, K. K. et al. Perforated Layer Structures in Liquid Crystalline Rod−Coil Block Copolymers. J. Am. Chem. Soc. 127, 15481–15490 (2005). 176. Shi, L.-Y., Zhou, Y., Fan, X.-H. & Shen, Z. Remarkably Rich Variety of Nanostructures and Order–Order Transitions in a Rod–Coil Diblock Copolymer. Macromolecules 46, 5308–5316 (2013). 177. Zhou, Q.-H. et al. Synthesis and Hierarchical Self-Assembly of Rod−Rod Block Copolymers via Click Chemistry between Mesogen-Jacketed Liquid Crystalline Polymers and Helical Polypeptides. Macromolecules 43, 5637–5646 (2010). 178. Zhou, F. et al. Synthesis and Self-Assembly of Rod–Rod Block Copolymers with Different Rod Diameters. Macromolecules 46, 8253–8263 (2013). 179. Sinturel, C., Vayer, M., Morris, M. & Hillmyer, M. A. Solvent Vapor Annealing of Block Polymer Thin Films. Macromolecules 46, 5399–5415 (2013). 180. Paik, M. Y. et al. Reversible Morphology Control in Block Copolymer Films via Solvent Vapor Processing: An in Situ GISAXS Study. Macromolecules 43, 4253–4260 (2010). 181. Chavis, M. A., Smilgies, D.-M., Wiesner, U. B. & Ober, C. K. Widely Tunable Morphologies in Block Copolymer Thin Films Through Solvent Vapor Annealing Using Mixtures of Selective Solvents. Adv. Funct. Mater. 25, 3057– 3065 (2015). 60 182. Machida, S., Sano, K., Sunohara, K., Kawata, Y. & Mori, Y. The novel texture of a liquid crystal induced by poly-γ-benzyl-L-glutamate chemical reaction alignment (CRA) film. J. Chem. Soc. Chem. Commun. 1628–1629 (1992) doi:10.1039/C39920001628. 183. Li, X. et al. Engineering the anchoring behavior of nematic liquid crystals on a solid surface by varying the density of liquid crystalline polymer brushes. Soft Matter 14, 7569–7577 (2018). 184. Bohbot-Raviv, Y. & Wang, Z.-G. Discovering New Ordered Phases of Block Copolymers. Phys. Rev. Lett. 85, 3428–3431 (2000). 185. Krappe, U., Stadler, R. & Voigt-Martin, I. Chiral Assembly in Amorphous ABC Triblock Copolymers. Formation of a Helical Morphology in Polystyrene-block- polybutadiene-block-poly(methyl methacrylate) Block Copolymers. Macromolecules 28, 4558–4561 (1995). 61 CHAPTER 2 Synthesis, processing, and characterization of helical polypeptide rod-coil mixed brushes 2.1 Abstract Mixed polymer brushes of rod-type polypeptide and coil-type vinyl polymer brushes were synthesized via two sequential steps of vapor deposition surface-initiated ring- opening polymerization (SI-ROP) and surface-initiated atom transfer radical polymerization (SI-ATRP) respectively. The effect on polypeptide brushes by coil-type brushes of their surface morphology, film thickness and orientation were investigated before and after solvent quenching processes using chloroform and acetone. Before solvent quenching, the as-grown coil-type brushes forced the polypeptide brushes to stand up from the surface, resulting in higher film thickness, but the polypeptide brushes remained randomly oriented. After solvent quenching, polypeptide brushes tended to aggregate into conical bundles with an orientation perpendicular to the substrate, but coil-type brushes restricted the free arrangement of the polypeptide brushes and lessen their upward movement. Changes in film thickness, rod orientation, morphology and wettability were observed with increased molecular weight of the coil-type polymer in the mixed brushes *Reproduced with permission from Tran, H.; Zhang, Y.; Ober, C. K. Synthesis, Processing, and Characterization of Helical Polypeptide Rod–Coil Mixed Brushes. ACS Macro Lett. 2018, 7 (10), 1186–1191. Copyright 2018 American Chemical Society 62 2.2 Introduction Coil type polymer brushes are flexible but only form a planar surface morphology which restricts access by big molecules to surface functional groups on the flat, crowded surface in a variety of possible applications. In contrast, polypeptide rod brushes mixed with coil polymer brushes, by combining high persistence length rod segments in brush layers of coil chains, offer numerous possibilities in applications dealing with surface structure and control. The mismatch in rigidity combined with controlled chain length, gives rod brushes the potential to project a chain end with its specific chemical functionality from the coil-phase surface. Thus, specific functional groups can be distributed at an elevated height above a surface and at a reduced spacing to give the opportunity to explore conjugation with a target molecule such as an antibody or DNA. Moreover, coil brushes (in combination with rod brushes) may possess properties to prevent non-specific adsorption making such combined materials of interest for molecular recognition and studies of cell-surface interaction. However, without intervention rod brushes will not stand up from the surface. In this study, we demonstrated the use of both solvent treatment and coil brushes to control the orientation of rod brushes and the size and number of rod assemblies. Through this control, brush surface morphology and surface properties can be tuned. In this study, α-helical polypeptide brushes were used as models for rod polymers in general. Polypeptides are biocompatible and biodegrable, thus making them suited for several biological applications1,2 The poly(γ-benzyl-L-glutamate) (PBLG) rod was used in this study as it has an α-helical conformation which is the most thermodynamically-stable conformation.3 In contrast, poly(methyl methacrylate) (PMMA) used as the non-rod forming brush second brush is one of the most common vinyl coil forming polymers. The PBLG polypeptide and PMMA brushes were grown from mixed initiators using the grafting-from approach via SI-ROP of α-amino acid N- 63 carboxyanhydrides (NCAs) and SI-ATRP respectively. The grafting-from approaches were used to ensure a homogeneous and high grafting density film of polymer brushes. 4–7 In SI-ROP, the immobilized amine initiators on the surface will initiate a polymerization by reacting with NCA monomers to form the amide bond and release carbon dioxide while creating a new amine group at the end of the growing chain.5 In this study, the PBLG brushes were prepared via a vapor-phase approach mainly to produce homogeneous, well controlled thick films of polypeptides. 3,8–12. SI-ATRP was used to grow coil-type brushes because this versatile technique can produce well- controlled polymer brushes for a wide range of monomers under mild conditions. 4,13 2.3 Results and Discussion Our preliminary test indicated that the two polymerization reactions are orthogonal, so no PBLG brushes can be grown from ATRP initiator and vice versa. Moreover, the ATRP initiators remained reactive after exposure to SI-ROP conditions as PMMA brushes were successfully grown with similar thickness under same conditions compared to a control sample. In addition, previous studies of polypeptide rod-coil block copolymers and bottlebrush polymers indicated the high stability of polypeptide under ATRP polymerization conditions 14,15 which was also confirmed for PBLG brushes in our preliminary results. Therefore, the synthesis of mixed brushes was possible, but SI-ROP was carried out before SI-ATRP to avoid a diffusion problem of monomer molecules in the vapor phase toward reaction sites in the presence of PMMA brushes. The deposition process to form mixed initiators was also considered. 3- aminopropyl dimethylethoxysilane (APDMES) was used as initiator for SI-ROP to ensure a monolayer of initiator. A two-step deposition was used. ATRP initiator was first deposited on surface to form a uniform, homogenous layer of the initiator. 64 Subsequently APDMES was reacted with remaining active silanols on surface. (Scheme 2.1) As APDMES and ATRP initiators contain Nitrogen and Bromine respectively, in this study the ratio of N/Br which was determined as 1.65 using XPS should be close to the composition of the mixed initiators on surface. Scheme 2.1. The synthesis procedure to make PBLG rod/PMMA coil mixed brushes Using the approaches mentioned above for the initiator depositions and polymerizations, mixed PBLG/PMMA brushes were successfully synthesized. The presence of both types of brushes was confirmed with FT-IR. For PBLG brushes, α- helical conformation has characteristic peaks of amide I (1654 cm-1, backbone carbonyl stretching), amide II (1550 cm-1,C-N stretching), amide A (3290 cm-1, backbone N-H stretching), and the ester side chain (1734 cm-1). 8,16,17 The FT-IR spectra of mixed brushes contained those peaks, indicating the presence of PBLG brushes in the α-helical conformation. In addition, the significant increase in intensity of the ester side chain (1734 cm-1) and the presence of peaks of CH3 (2997 cm -1, C-H stretching) suggested the addition of PMMA brushes. (Figures 2.1 and S2.1) 65 Figure 2.1. FT-IR transmission spectra of PBLG homopolymer brushes (top) and mixed brushes (bottom) before and after solvent quenching 66 In order to understand the interaction between PBLG and PMMA brushes, a systematic study was carried out based on assumption that PMMA growth kinetics are not affected significantly by the presence of rod brushes. Various PMMA brush thicknesses were achieved via varying ATRP reaction time for substrates with the same rod brush thickness, which was defined as the measured thickness before the growth of PMMA brushes. In this work, a sample with the initial PBLG thickness of 62 nm were cut into four smaller samples (I-IV) from that PMMA brushes with different thicknesses ranging from 0, 31, 61 to 107nm for sample I to IV respectively were grown as determined from control samples of PMMA brushes. (Figure 2.2) Interestingly, the final thickness of mixed brush samples was significantly higher than either PBLG or PMMA control thickness. For example, the final thickness of sample IV were 142nm, larger than that of PBLG (62 nm) and PMMA control (107 nm), respectively. To gain more insight of final thicknesses of mixed brushes, surface compositions of mixed brushes were studied using XPS to know if the PMMA or PBLG layer was at the topmost surface (Figure S2.2). XPS showed that a significant percentage of nitrogen retained in the topmost layer of all the samples, indicating that PBLG brushes remained as major component (Table 1). Considering the significant change in mixed brush thickness, we hypothesized that the presence of PMMA brushes changed the orientation of PBLG brushes to more upright orientation, thus increasing the thickness. Initially, the PBLG brushes tend to orient parallel to the surface to minimize surface energy,16 but PMMA brushes occupied the space and forced PBLG brushes to be more perpendicular from the surface. Thus the final thickness was the new thickness of PBLG brushes at higher tilt angle. 67 Figure 2.2. The thickness of PMMA and mixed brushes, and the D ratio for A) before solvent quenching; B) after solvent quenching; C) tilt angle between the PBLG helices and the substrate before and after solvent quenching. The D value is defined as the ratio of amide I to amide II peak intensity. The samples have the same MW of PBLG but increasing thickness of PMMA from sample I to IV. The angle of quenched PBLG brushes (sample I) was assumed to be 87o as in Ref. 18 68 This hypothesis can also be supported by FT-IR measurements in transmission mode, since the intensity of the amide bonds changes with polypeptide orientation. The transition dipole moments with respect to an α-helix backbone are 39o and 75o for amide I and amide II respectively. 17 Thus, under the transmission mode, when the IR-beam went through the samples and interacted with PBLG brushes, only the active electric field components of the transition dipole moments of the amide bonds parallel to the substrate were measured.17. Then, the dichroic ratio D of amide I over amide II Table 2.1. Surface composition of mixed brush samples measured with XPS before solvent quenching. ATRP reaction C% N% O% time (hour) sample I 0 72.24 6.64 21.12 sample II 4 71.94 4.95 23.11 sample III 8 69.80 5.26 24.94 sample IV 20 69.58 4.69 25.73 adsorption can give a qualitative assessment of the average orientation of α-helical rod. The samples have the same MW of PBLG but increased thickness of PMMA 17–19 The decrease in D value is related to higher average tilt angle of the α-helical rods from sample I to IV by varying the ATRP reaction time. with respect to the substrate. The trend of decreasing D value from sample I to IV implied an increase in average tilt angle which was consistent with the increase in film thickness mentioned previously (Figure 2.1 and 2.2A). AFM images showed homogeneous and quite smooth surfaces for all samples (Figure S2.3) and indicated disordered orientations of PBLG brushes. Lastly, the water contact angle of the samples was also measured. No obvious change of water contact angle was observed from samples I-IV (Figure 2.3). 69 Subsequently a processing step called ‘solvent quenching’ was applied to enhance PBLG orientational order and aggregation.18 Mixed brush samples were first immersed in chloroform, a good solvent for PBLG to cause the chains to stand up before immediately transferring the samples into acetone, a bad solvent for PBLG so that the PBLG brushes would aggregate to minimize contact with acetone, thereby freezing any vertical orientation of PBLG brushes. 18,20 After solvent quenching, a dramatic change in thickness was observed for all the samples. (Figure 2.2B) However, the change in thickness was less significant for an increase in thickness of PMMA brushes, indicating a smaller change in the angle between the α-helix rod and the substrate. Our hypothesis was that the presence of PMMA brushes acted as a constraint on the rearrangement of PBLG brushes during solvent quenching. For sample I, PBLG brushes alone freely moved from parallel to perpendicular orientation while the entanglement of PMMA brushes made it harder for PBLG rods to move upright as seen in the samples II-IV. The dichroic ratio D again supported this hypothesis. The increase in D value from sample I-IV after solvent quenching indicated that the orientation of PBLG rods over the substrate was less perpendicular. Figure 2.3. Water contact angle of mixed brush samples before and after solvent quenching. 70 Previous studies reported an almost perpendicular orientation of PBLG brushes after solvent quenching. 18,20 Thus, we assume that the final thickness of PBLG homopolymer brushes (sample I, 235 nm) after solvent quenching should be nearly equal to the length of the PBLG brushes. As each monomer unit contributes 1.5 Å along the helical axis of the chain to the final length of PBLG brushes, the degree of polymerization was about 1500.17 Using the PBLG density of 1.32 g/cm3,21 the grafting density of PBLG helices was estimated as 0.143 chain/nm2 while the grafting density of PMMA brushes calculated from MW of free polymer was 0.56 chain/nm2.22,23 The tilt angle of PBLG rods in mixed brushes was ranged from 16o to 38o before quenching and from 87o to 51o after quenching (Figure 2.2C). AFM images also gave more information about mixed brush structure after solvent quenching. In particular, conical aggregates were observed for all the samples, though the height and sizes of aggregates varied from the samples I-IV (Figure 2.4). The average height of the aggregates and the spacing between neighboring aggregates grew smaller while the total number of aggregates per m2 increased from 42 to 98 aggregates going from sample I to IV. (Table S2.1) The presence of PMMA brushes disrupted the aggregation process, so only nearby PBLG rods could interact with each other to form aggregate bundles. The higher number of aggregates in mixed brushes may allow more chain-end functional groups of the PBLG rods to project from the surface. Moreover, the aggregates looked smaller and shorter in the mixed brushes because their lower parts were buried by the PMMA brush layer and became invisible to the AFM. Thus, with an increase in thickness of the PMMA layer from samples I to IV, the exposed part of the aggregates was smaller. 71 Figure 2.4. A) The schematic figure of rod-coil mixed brushes; B) The height images and cross-sectional profiles of mixed polymer brushes after solvent quenching. The white dash lines in the height images indicate the cross-sectional height measurement of the accompanying height profiles. All samples have the same molecular weight of polypeptide brushes but increased thickness of PMMA from sample I to IV. The scan size is 1µm x 1µm. 72 The water contact angle after quenching changed significantly. Sample I changed from 72.5o to 121o. This drastic change may be due to the formation of nanostructured rod aggregates. A similar effect was observed for lotus leaves when papillae structures of a leave can trap air when contacting with water drops and make the leaf become superhydrophobic.24 However, the addition of PMMA lowered the hydrophobicity as the angle of sample IV was 98o. At higher thickness of PMMA brushes, a higher volume of PBLG brushes was buried under a PMMA layer, thereby reducing the surface roughness.(Table S1) 2.4 Conclusion In conclusion, we have reported the study of mixed polymer brushes consisting of α-helical polypeptide rod and vinyl-type polymer coil. With constant starting rod PBLG thickness, the final thickness of mixed brushes prior to solvent quenching increase monotonically with the increase in MW of coil type PMMA brushes due to change in the orientation of PBLG rods. After solvent quenching, a significant increase in thickness was observed due to the formation of the conical aggregates of PBLG rods. However, monotonical decrease in the final thickness was seen because the entanglement of PMMA brushes restricted the movement of PBLG rods. As a result, less perpendicular rods were observed with an increase of PMMA brush thickness. Consequently, surface morphology and hydrophobicity also changed dramatically. Our studies showed that rod-coil mixed brushes behave differently from coil-coil mixed brushes, which have a subtle difference in height between two coil phases.22,25 This rod- coil system can lead to unique types of phase separation which can bring more chain- end functional groups of rod brushes to air interface. The versatility of ATRP polymerization and the high stability of polypeptides under ATRP conditions 73 demonstrated in our study showed the potential of stimuli-responsive coil-type brushes or block copolymer brushes in a rod-coil system. 2.5 Experimental Section 2.5.1 Materials Allyl-2-bromo-2-methylpropionate, ethyl α-bromoisobutyrate, chlorodimethylsilane, Pt on activated carbon (10 wt %), 2,2′- bipyridine, pyridine, copper (1) chloride, inhibitor remover (for removing hydroquinone and monomethyl ether hydroquinone), anhydrous toluene, methyl methacrylate, γ-benzyl L-glutamate, triphosgene were purchased from Sigma Aldrich. 3-aminopropyldimethyl-ethoxysilane (APDMES) was purchased from Gelest. Copper tape (882-L COPPER) with 88.9 μm copperfilm was purchased from Lamart Co. Deionized water with a resistivity of 18.2 MΩ•cm at 25 °C was obtained from Millipore’s Milli-Q Synthesis A10 system. Tetrahydrofuran (THF) was distilled over sodium. All other solvents were purchased from Fisher Scientific. Si(100) double- sided polished wafers were from WRS Materials. Cantilevers were purchased from Applied NanoStructures, Inc. (ACCESS-NC) 2.5.2 Synthesis y-Benzyl L-Glutamate N-Carboxyanhydride The NCA monomer was synthesized using a literature procedure.26 ATRP silane initiator Hydrosilylation of allyl 2-bromo-2-methylpropionate with chlorodimethylsilane was carried out using a literature procedure to obtain monofunctional ATRP initiator, 3- (chlorodimethylsilyl)propyl 2-bromo-2-methylpropionate. 27 Immobilization of silane initiators 74 Double-polished wafer pieces were oxidized using a Harrick Plasma Cleaner for 10 minutes, rinsed with ethanol, and heated to further remove moisture. In a glove box, the wafer pieces were immersed in a toluene solution of the ATRP initiator (25 mM) and anhydrous pyridine (0.5 mM) for 16 hours at room temperature. The substrates were then removed from the solution and washed with methanol and toluene sequentially. Wafer pieces were then blown dry under nitrogen gas. After that, in a glove box, the ATRP initiator-immobilized wafer samples were immersed in a toluene solution of APDMES (120 mM) at room temperature for 24 hours. The substrates were then removed from the solution and washed with toluene and blown dry under nitrogen gas. PBLG brush polymerization 10 mg of NCA monomer was added and spread at the bottom of a 10-mL glass beaker. An initiator-immobilized wafer sample was placed on top of the beaker and was fixed to the beaker with copper tape. The beaker and sample was then put into a glass vessel which was subsequently evacuated to 500 mtorr and heated up in an oil bath at 105oC for a controlled time period. After the reaction, the substrate was soaked in chloroform overnight, subsequently sonicated in chloroform, and dried under nitrogen gas. PMMA brush polymerization Wafer samples with tethered PBLG brushes and ATRP initiators were placed into a Schlenk flask which was then evacuated and back-filled with argon for 5 times. Methyl methacrylate (94 mmol), methanol (8 mL) and water (2 mL) were purged with argon in for about 40 minutes. CuCl (173 mg, 1.74 mmol), bipyridine (544 mg, 3.48 mmol) were added in another Schlenk flask equipped with a magnetic stir bar. The flask was evacuated and purged with argon for 5 times. The monomer solution was then cannulated into the ligand and copper salt. The mixture was stirred for 10 minutes before cannulated into the flask containing the wafer substrates. The polymerization was kept 75 at 30oC for different reaction time. After the reaction, the samples were washed with chloroform and blown dry under nitrogen gas. Quenching process The polymer brush samples were immersed into chloroform for 30 minutes. Then they were quickly transferred into acetone for 20 minutes. After that, they were blown dry under nitrogen gas. Free PMMA polymer polymerization A similar procedure to PMMA brush polymerization was used with ethyl α- bromoisobutyrate as free ATRP initiator (2000:1 molar ratio of monomer to initiator). The polymerization was stopped by opening the reaction to air. PMMA was precipitated in methanol, dissolved back in THF, and ran through a column of aluminum oxide to remove the catalyst. The polymer was re-precipitated in methanol and dried in vacuum oven. The purified polymer was dissolved in THF at 1mg/mL for GPC analysis. 2.5.3 Characterization Optical Ellipsometry Thicknesses of initiator film, and PMMA homopolymer brushes were measured with an Imaging Ellipsometer - Nanofilm EP3 at 532 nm laser Fourier Transform Infrared Spectroscopy (ATR-FTIR) A Bruker Optics – Vertex 80v was used to determine the presence and conformation of polymer brushes. The spectrums were taken in transmission mode. Spectra were recorded at 4 cm-1 resolution, and 256 scans were taken. Atomic Force Microscopy (AFM) 76 Asylum MFP-3D atomic force microscopy was used to characterize the topography and thickness of mixed polymer brushes. The thickness of the polymer brushes was determined by measuring the height difference between a brush area and a scratched (polymer removed) area. The dry topography was characterized with AC tapping mode using silicon cantilevers (model: ACCESS-NC) Etcher Oxford PlasmaLab 80+ was used to remove PMMA homopolymer brushes on backside of samples. X-ray Photoelectron Spectroscopy (XPS) XPS measurements were done on a Surface Science Instruments SSX-100 with operating pressure ~2x10-9 Torr. Monochromatic Al Kα x rays (1486.6 eV) with 1 mm diameter beam size were used. Photoelectrons were collected at a 55° emission angle. A hemispherical analyzer determined electron kinetic energy, using a pass energy of 150 V for wide/survey scans. Contact angle Contact angle measurements were performed using VCA Optima Contact Angle. The measurements were taken on 4 different locations on the sample. Gel permeation chromatography (GPC) Waters ambient temperature GPC equipped with a Waters 410 differential refractive index detector was used to determine Molecular weight and dispersity of free PMMA polymer. GPC columns were eluted at a rate of 1.0mL/min with THF at 40oC. 2.6 Acknowledgement This works is supported by the National Science Foundation (DMR-1506542). The 77 authors would like to acknowledge fruitful discussions with Prof. Su-mi Hur. This work made use of the Cornell Center for Materials Research Shared Facilities which are supported through the NSF MRSEC program (DMR-1719875) and Cornell NanoScale Facility, an NNCI member supported by NSF Grant ECCS-1542081. 2.7 Supporting Information Figure S2.1. FT-IR spectra of mixed PBLG and PMMA polymer brushes before quenching. The labeled peaks are amide I (1654 cm-1, backbone carbonyl stretching), amide II (1550 cm-1,C-N stretching), amide A (3290 cm-1, backbone N-H stretching), the ester side chain (1734 cm-1), and CH3 (2997 cm -1, C-H stretching). 78 Figure S2.2. XPS spectra of PBLG brushes before quenching. 79 Figure S2.3. The height images of mixed polymer brush samples before quenching. All the samples have the same molecular weight of PBLG but with increasing thickness of PMMA from sample I to IV. 80 Figure 1 Figure S2.4. The phase images of mixed polymer brush samples after quenching. All the samples have the same molecular weight of PBLG but with increasing thickness of PMMA from sample I to IV. 81 Figure S2.5. XPS spectra of mixed initiators (APDMES and ATRP initiators) on silicon substrate. The ratio of Nitrogen (1s) and Bromine (3d) is 1.65 82 Table S2.1. Roughness, average height of conical aggregates above PMMA layer, and estimated number of aggregates per micron square. All the samples have the same molecular weight of PBLG but with increasing thickness of PMMA from sample I to IV. Sample I II III IV before quenching 1.8 2.5 2.0 2.1 roughness (nm) after quenching 56.6 46.8 38.1 15.8 Conical aggregate height above PMMA 232 183 128 55 layer (nm) Number of aggregates per micrometer 42 57 69 98 square Grafting density calculation a) PBLG brushes 1.32𝑔 ℎ𝜌𝑁 62𝑛𝑚 ∗ ∗ 6.02 ∗ 10 23𝑐ℎ𝑎𝑖𝑛𝑠/𝑚𝑜𝑙 𝐴 𝜎 = = 𝑐𝑚 3 = 0.143 𝑐ℎ𝑎𝑖𝑛/𝑛𝑚2 𝑀𝑛 342954𝑔/𝑚𝑜𝑙 b) PMMA brushes 1.19𝑔 ℎ𝜌𝑁 61𝑛𝑚 ∗ 3 ∗ 6.02 ∗ 10 23𝑐ℎ𝑎𝑖𝑛𝑠/𝑚𝑜𝑙 𝐴 𝜎 = = 𝑐𝑚 = 0.56 𝑐ℎ𝑎𝑖𝑛/𝑛𝑚2 𝑀𝑛 78034𝑔/𝑚𝑜𝑙 Note: for PMMA brushes, 61 nm and Mn of 78,034 g/mol are the height of PMMA brushes and molecular weight of free PMMA polymer synthesized in the same ATRP conditions for 8 hours. 83 2.8 References 1. Kar, M., Pauline, M., Sharma, K., Kumaraswamy, G. & Sen Gupta, S. Synthesis of Poly- L -glutamic Acid Grafted Silica Nanoparticles and Their Assembly into Macroporous Structures. Langmuir 27, 12124–12133 (2011). 2. Ma, Y., Shen, Y. & Li, Z. Different cell behaviors induced by stereochemistry on polypeptide brush grafted surfaces. Mater Chem Front 1, 846–851 (2017). 3. Lee, N. H. & Frank, C. W. Surface-Initiated Vapor Polymerization of Various α- Amino Acids. Langmuir 19, 1295–1303 (2003). 4. Barbey, R. et al. Polymer Brushes via Surface-Initiated Controlled Radical Polymerization: Synthesis, Characterization, Properties, and Applications. Chem. Rev. 109, 5437–5527 (2009). 5. Wibowo, S. H., Sulistio, A., Wong, E. H. H., Blencowe, A. & Qiao, G. G. Polypeptide films via N-carboxyanhydride ring-opening polymerization (NCA- ROP): past, present and future. Chem. Commun. 50, 4971 (2014). 6. Shen, Y., Li, Z. & Klok, H.-A. Polypeptide brushes grown via surface-initiated ring-opening polymerization of α-amino acid N-carboxyanhydrides. Chin. J. Polym. Sci. 33, 931–946 (2015). 7. Chen, W.-L., Cordero, R., Tran, H. & Ober, C. K. 50th Anniversary Perspective : Polymer Brushes: Novel Surfaces for Future Materials. Macromolecules 50, 4089–4113 (2017). 8. Chang, Y.-C. & Frank, C. W. Vapor deposition-polymerization of α-amino acid N-carboxy anhydride on the silicon (100) native oxide surface. Langmuir 14, 326– 334 (1998). 84 9. Zheng, W. & Frank, C. W. Surface-Initiated Vapor Deposition Polymerization of Poly(γ-benzyl- L -glutamate): Optimization and Mechanistic Studies. Langmuir 26, 3929–3941 (2010). 10. Shen, Y. et al. Shape-Persistent, Thermoresponsive Polypeptide Brushes Prepared by Vapor Deposition Surface-Initiated Ring-Opening Polymerization of α-Amino Acid N -Carboxyanhydrides. Macromolecules 48, 2399–2406 (2015). 11. Wang, Y. & Chang, Y.-C. Grafting of Homo- and Block Co-polypeptides on Solid Substrates by an Improved Surface-Initiated Vapor Deposition Polymerization. Langmuir 18, 9859–9866 (2002). 12. Yang, C.-T., Wang, Y. & Chang, Y.-C. Effect of Solvents and Temperature on the Conformation of Poly(β-benzyl- L -aspartate) Brushes. Biomacromolecules 11, 1308–1313 (2010). 13. Khabibullin, A., Mastan, E., Matyjaszewski, K. & Zhu, S. Surface-Initiated Atom Transfer Radical Polymerization. in Controlled Radical Polymerization at and from Solid Surfaces (ed. Vana, P.) vol. 270 29–76 (Springer International Publishing, 2015). 14. Huang, C.-J. & Chang, F.-C. Polypeptide Diblock Copolymers: Syntheses and Properties of Poly(N-isopropylacrylamide)-b-Polylysine. Macromolecules 41, 7041–7052 (2008). 15. Ding, J., Xiao, C., Tang, Z., Zhuang, X. & Chen, X. Highly Efficient “Grafting From” an α-Helical Polypeptide Backbone by Atom Transfer Radical Polymerization. Macromol. Biosci. 11, 192–198 (2011). 85 16. Wieringa, R. H. et al. Surface grafting of poly (L-glutamates). 1. Synthesis and characterization. Langmuir 17, 6477–6484 (2001). 17. Wieringa, R. H. et al. Surface Grafting of Poly( L -glutamates). 2. Helix Orientation. Langmuir 17, 6485–6490 (2001). 18. Wang, Y. & Chang, Y. C. Preparation of Unidirectional End-Grafted α-Helical Polypeptides by Solvent Quenching. J. Am. Chem. Soc. 125, 6376–6377 (2003). 19. Vlasova, E., Volchek, B., Tarasenko, I. & Vlasov, G. Spectroscopic Investigation of Polypeptide Plane Brushes. Macromol. Symp. 305, 116–121 (2011). 20. Wu, J.-C., Chen, C.-C., Chen, K.-H. & Chang, Y.-C. Controlled growth of aligned α-helical-polypeptide brushes for tunable electrical conductivity. Appl. Phys. Lett. 98, 133304 (2011). 21. Machida, S. et al. A Chiral Director Field in the Nematic Liquid Crystal Phase Induced by a Poly(.gamma.-benzyl glutamate) Chemical Reaction Alignment Film. Langmuir 11, 4838–4843 (1995). 22. Zhao, B. Synthesis of binary mixed homopolymer brushes by combining atom transfer radical polymerization and nitroxide-mediated radical polymerization. Polymer 44, 4079–4083 (2003). 23. Ionov, L. & Minko, S. Mixed Polymer Brushes with Locking Switching. ACS Appl. Mater. Interfaces 4, 483–489 (2012). 24. Sun, T., Feng, L., Gao, X. & Jiang, L. Bioinspired Surfaces with Special Wettability. Acc. Chem. Res. 38, 644–652 (2005). 25. Price, A. D., Hur, S.-M., Fredrickson, G. H., Frischknecht, A. L. & Huber, D. L. Exploring Lateral Microphase Separation in Mixed Polymer Brushes by 86 Experiment and Self-Consistent Field Theory Simulations. Macromolecules 45, 510–524 (2012). 26. Daly, W. H. & Poché, D. The preparation of N-carboxyanhydrides of α-amino acids using bis(trichloromethyl)carbonate. Tetrahedron Lett. 29, 5859–5862 (1988). 27. Ramakrishnan, A., Dhamodharan, R. & Rühe, J. Controlled Growth of PMMA Brushes on Silicon Surfaces at Room Temperature. Macromol. Rapid Commun. 23, 612–616 (2002). 87 CHAPTER 3 Organization and orientation of helical polypeptide brushes induced by solvent treatment 3.1 Abstract Polymer brushes can transform surface properties via tailoring of chemical functional groups, architecture, or processing conditions. Previous studies have reported that helical poly(γ-benzyl-L-glutamate) (PBLG) brushes, when treated with either good (chloroform) or poor (acetone) solvents form either collapsed or teepee-like assemblies, respectively but there is a lack of understanding in the mechanism and key parameters involved in the process. Here we showed that the organization and orientation were strongly dependent on polymer thickness. In the collapsed state, the polymer tilt angle from the surface normal was proportional to PBLG brush thickness while in the quenched state, the polymer tilt angle decreased with an increase of brush thickness. Moreover, teepee-like assemblies were bigger for thicker brushes, resulting in higher surface roughness, though the number of teepee-like assemblies decreased. Additionally, PBLG brushes can induce liquid crystal (LCs) response through a dispersion interaction, and there was a proportional correlation between LC orientation and PBLG tilt angle. We also demonstrated that the success of the quenching process was reliant on solvent miscibility, extraction rate and polymer rigidity. When mixed solvents of chloroform and acetone were used for quenching instead of pure acetone, micron-scale patterned regions of PBLG brushes were observed, resulting from the formation of solvent droplets on the brush surface during drying process. Finally, LCs can be utilized for imaging PBLG brush organization to complement AFM and FT-IR to characterize heterogenous patterned surfaces. 88 Note: This work was done in collaboration with the Abbott lab. Hai Tran did the synthesis of monomer and polymer brushes, and characterization of brushes using XPS, FT-IR and AFM. Michael Tsuei characterized the Liquid Crystal responses on brushes using POM. 3.2 Introduction Polymer brushes are thin films of polymer chains covalently attached to surfaces.3 Polymer brushes have been widely used to modify surface properties and morphology through tuning chemical structures, molecular weight, grafting density or functional groups of the polymers.1–3 Given the versatility of polymerization methods, polymer brushes have demonstrated potential applications as adhesive materials 4,5, chromatographic devices 6, controlled transport of ions7, and platforms for molecular recognition. 8 Polymer brushes may be categorized according to their stiffness as either coil-type or rod-type brushes. α-Helical polypeptide brushes with reported persistence lengths between 100 and 200 nm,9 have been studied as a model for rod-type polymers. Due to the wide array of amino acids, polypeptide brushes have been demonstrated as promising stimuli-responsive materials to external factors such as pH, temperature, and ionic strength. 10 Moreover, a distinct aspect of high persistence length polypeptide brushes compared to coil-type brushes is their orientation or tilt angle with respect to a surface. Among many polypeptide brushes, poly(γ-benzyl-L-glutamate) (PBLG) is noteworthy due to its stable helical structure and possession of both non- centrosymmetric piezoelectric and dielectric properties.11 The orientation of PBLG brushes is critical to their properties as Jaworek et al. demonstrated that directionally aligned PBLG brushes had large dipole moments and piezoelectric properties.12 In 89 addition, Wu et al. showed that the electrical properties of PBLG brushes changed from insulator to diode when the tilt angle from the surface normal decreased from 75° to 13°.13 For this reason, several approaches have been reported to control the orientational states and properties of grafted polypeptide brushes. Whitesell et al. designed an aminotrithiol initiator to control spacings between polymer chains and directionally align polypeptide brushes.14 Alternatively, Luijten et al. used a chemical cross-linking approach to lock polymer brushes into a vertical orientation.15 Of relevance to our study, Wang et al. used solvent quenching to switch the orientation of polymer brushes.16 In this process, the PBLG brushes were first swollen with a good solvent such as chloroform to extend them from the surface before transfer into a non-solvent such as acetone, and dried. Exposure to poor solvent resulted in the formation of teepee-like assemblies of polymer brushes. This treatment enabled a change in the PBLG tilt angle from 49° in the collapsed state to 3° in the quenched state. However, there are only limited studies following that initial work. In this report, we aimed to develop a comprehensive picture of the underlying mechanism and any parameters related to the quenching process. A limit of the quenching process was that it only revealed two states: collapsed and quenched. Thus, it motivated us to study intermediate states and more selective control in tuning PBLG tilt angles. In addition, Machida et al. showed a spiral texture of nematic liquid crystal induced by PBLG brushes,17 and this observation inspired us to explore how the tilt angle and organization of PBLG brushes may influence the ordering of LCs supported on the polymer brushes. Moreover, the long range orientational order of LCs can amplify any changes in nanoscale structure at interfaces to be visualized in optical textures.18–20 For instance, Xiao et al. showed how the conformation of poly(6-(4- methoxyazobenzene-4'-oxy)hexyl methacrylate) (PMMAZO) brushes changed the orientation of LCs supported on the brushes. At low grafting density, the mesogenic 90 side groups of the polymer brushes aligned perpendicular to the substrate, resulting in homeotropic anchoring of the LCs. In contrast, at high grafting density, polymer brushes extended away from the surface and caused the mesogenic side group to align tangentially to the substrate, resulting in planar anchoring of the LCs.21 We envisioned that if a relationship between PBLG orientation and LC ordering existed, LCs would be a new effective method to visualize any tiny changes of PBLG orientation and organization. Thus LCs can complement FT-IR and AFM which are the two common and useful techniques to characterize PBLG brushes.22–24 While FT-IR can determine the orientation of polypeptide brushes, the resulting value is the average tilt angle of all polymer chains across mm2 surface areas, so it is not effective to measure a heterogenous surface. On the other hand, AFM is a good method for mesoscale surface structure, but not useful for molecular-scale features or any changes below the polymer surface.25 Thus, LCs may provide some insight beyond the capabilities of FT-IR and AFM. In this work, we reported the effect of PBLG brush thickness on polymer organization and orientation in both collapsed and quenched states. Second, we explored LC response to changes in polymer orientation and polymer assemblies. We also established relationship between LC ordering and PBLG orientation. In addition, we demonstrated how polymer chain rigidity, solvent miscibility, and extraction time impacted the quenching process and affected polymer orientation, organization, and consequent LC ordering. Lastly, we showed how mixed solvent treatment induced complex surface patterning of PBLG brushes, and how low molar mass LCs were utilized to probe the polymer organization. 91 3.3 Experimental Section 3.3.1 Materials Anhydrous toluene, γ-benzyl L-glutamate, and triphosgene were purchased from Sigma Aldrich and used as received. 3-aminopropyldimethyl-ethoxysilane (APDMES) was purchased from Gelest. Copper tape (882-L COPPER) was purchased from Lamart Co. Deionized water with a resistivity of 18.2 MΩ•cm at 25 °C was obtained from Millipore’s Milli-Q Synthesis A10 system. Tetrahydrofuran (THF) was distilled over sodium. All other solvents were used as received from Fisher Scientific. Si(100) double- sided polished wafers were from WRS Materials. Cantilevers were from Applied NanoStructures, Inc. (ACCESS-NC). The nematic LC 4-cyano’-pentylbiphenyl (5CB) was purchased from HCCH (Jiangsu Hecheng Display Technology Co., Ltd). Chloroform and hexane were purchased from VWR Chemicals BDH. 3.3.2 Synthesis of PBLG brushes y-Benzyl L-Glutamate N-Carboxyanhydride (NCA monomer) was synthesized by a published procedure.26 Glass slides and silicon wafers were cut into 12 mm x 25 mm pieces and cleaned in a solution of 70% (v/v) H2O, 15% (v/v) HCl (ACS reagent, 37%) and 15% H2O2 (ACS reagent, 30%) at 50°C for one hour. The slides and wafers were then washed with methanol, dried with nitrogen gas and placed in an oven at 110 oC for 10 minutes to remove any moisture. Subsequently, the slides and wafers were oxidized using a Harrick Plasma Cleaner for 10 minutes. Glass slide and silicon wafer pieces were immersed in a toluene solution of APDMES (120 mM) at room temperature for 16 hours. The substrates were then washed with toluene and methanol before drying with nitrogen gas. 10 mg of NCA monomer was added and spread at the bottom of a 10-mL glass beaker. An initiator-immobilized substrate was fixed on top of the beaker with copper tape. The beaker and the substrate were then put into a glass chamber connected to a 92 Schlenk line. The glass chamber was subsequently evacuated to 500 mtorr and heated up in an oil bath at 105 oC for different amounts of time to control polymer brush thickness. After the reaction, the substrate was sonicated in chloroform, and blow-dried with nitrogen gas. 3.3.3 Characterization of PBLG brushes Polymer brushes on doubled-polished wafers were characterized by Bruker Optics – Vertex 80v Fourier Transform Infrared Spectroscopy (FT-IR). The spectrums were taken in transmission mode, and recorded with 4 cm-1 resolution and 256 scans. X-ray photoelectron spectroscopy (XPS) measurements were done with a Surface Science Instruments SSX-100 using a monochromatic Al Kα source (1486.6 eV). Photoelectrons were collected at a 55° emission angle. Asylum MFP-3D atomic force microscopy (AFM) was used to characterize surface topography of polymer brushes with AC tapping mode using silicon cantilevers (model: ACCESS-NC). 3.3.4 Solvent and Nonsolvent treatment of PBLG brushes Treatment of all PBLG brushes with single component or mixtures of solvents and nonsolvents required initial solvation of the brushes in 3 mL chloroform for 2 min. Solvent quenching in acetone required direct transfer of the chloroform solvated brush into 3 mL acetone for 1 min and subsequent drying via compressed air. Nonsolvent- nonsolvent mixture treatment required direct transfer of the chloroform solvated brush into mixtures of acetone and water for 1 min and subsequent drying via compressed air. Solvent-nonsolvent mixture treatment required direct transfer of solvated brushes into mixtures of chloroform and acetone for 1 min and subsequent drying via compressed air. 93 3.3.5 Preparation of thin LC films 20 μm thick LC films were prepared by pipetting 0.5 μL of LC (i.e. 5CB or MBBA) into the pores of 75 mesh (thickness 20 μm; lateral pore size 285 μm) transmission electron microscopy (TEM) grids that were supported on the PBLG brushes. Excess LC was then removed to produce a LC film with thickness of 20 μm. 3.3.6 Microscopy Observation An Olympus BX41 microscope with 4x, 20x and 50x objectives, two rotating polarizers and a Moticam 10.0 MP camera was used for optical microscopy. 3.3.7 Changing PBLG chain persistence length via heating 130 nm thick PBLG brushes were prepared and heated to 180 °C for either 0 min, 4 min, 15 min, or 300 min. After heat treatment, the samples were cooled to 25 °C and surface topography of the brushes was measured via AFM, average brush tilt angles were measured via FT-IR, and the lateral organization of the brushes was measured via LC films. 3.4 Results and Discussions 3.4.1 Dependence of PBLG brush orientation on brush thickness Scheme 3.1. Synthesis procedure detailing surface initiation and SI-ROP used to make PBLG rod brushes. 94 In our initial experiments, we examined the effect of polymer brush thickness on polymer orientation and self-assembly. Surface-initiated vapor-phase ring opening polymerization was used to control polymer brush thickness by varying reaction time as reported in previous studies (Scheme 3.1).27 The presence of polymer brushes on the substrate was verified via X-ray photoelectron spectroscopy (XPS). The measured surface composition (17% oxygen, 76% carbon, 7% nitrogen) was close to the theoretical composition of PBLG (18.75% oxygen, 75% carbon, 6.25% nitrogen) (Figure 3.1). The polymer brushes were solvated in chloroform before either drying with compressed air or quenching in acetone to induce either a collapsed state or quenched state respectively as described by Wang et al. 28 AFM was used to characterize surface morphology and film thickness. Figure 3.1: XPS analysis of PBLG brushes shows elemental composition of 17% oxygen, 76% carbon, 7% nitrogen that is consistent with the theoretical composition of PBLG. 95 In this work, four sets of PBLG brushes with different thicknesses measured in the quenched state (12 nm, 25 nm, 50 nm and 140 nm) were studied. AFM showed that quenched PBLG brushes formed nanoscale teepee-like structures for all four thicknesses, though the magnitude of polymer assemblies changed significantly for 12 nm to 140 nm thick samples (see Figure 3.2). Moreover, the size and shape of teepee- like assemblies were more pronounced for thicker brushes. For example, with much bigger teepees, the 140nm thick sample had an average spacing of 110 nm and roughness of 32.6 nm while the 12 nm thick sample had a spacing of 51 nm and roughness of 3.49 nm (Table 3.1). However, the number of teepee-like assemblies was reduced for thicker brushes as the AFM cross section profile revealed that the 12 nm thick and 140 nm thick samples comprised 18 and 9 teepees per 1 μm line, respectively. The clear difference in teepee height and diameter suggested a change of polymer orientation with brush thickness. Table 3.1: Surface roughness, teepee density and spacing of PBLG brushes 96 Figure 3.2: 3D AFM images and 2D height profiles of (a) 12 nm, (b) 25 nm, (c) 50 nm, and (d) 140 nm thick PBLG brushes in quenched state after treated with acetone. Dotted white lines show cross-section of 3D images to generate 2D height profiles. 97 To gain more insight into brush organization in the quenched state, FT-IR in transmission mode was utilized. The characteristic peaks of amide I (1654 cm-1, backbone carbonyl stretching), amide II (1550 cm-1, C-N stretching), and amide A (3290 cm-1, backbone N-H stretching) were observed for all four different brush thicknesses, indicating an α-helical conformation for the PBLG brushes (Figure 3.3a-d). Moreover, the dichroic ratio of amide I over amide II peaks can be used to estimate the average tilt angles with respect to the surface normal of quenched PBLG brushes (sections S2). We found that there was an inverse correlation between PBLG tilt angle and brush thickness in the quenched state. Specifically, a 12 nm thick brush had an average tilt angle of 25.6° from the surface normal while a brush of 140 nm has a tilt angle of 15.1° (Figure 3.3e). This observation suggests that thicker brushes were more vertically oriented, and the results were consistent with the shape of teepees observed by AFM. It can be explained that long polymer chains only need a slight tilt to aggregate with neighboring polymer chains and form teepee-like assemblies while a shorter chain requires a greater relative movement to meet other chains. We also studied the PBLG brushes in the collapsed state using AFM and FT-IR. In contrast to the quenched state, surface morphology of polymer brushes in the collapsed state was quite flat and mostly featureless for brush thickness over 25 nm (see Figure 3.4). Such behavior suggests that the polymer chains were randomly oriented, consistent to what has been reported in the literature.28 Interestingly, 12 nm thick brushes exhibited teepee-like structures, though less pronounced than the quenched state. An AFM cross-section profile showed that there were about 7 teepees per 1 μm line for brushes in the collapsed state. This observation implies that 12 nm brushes were somewhat vertical, but less perpendicular than that of the quenched state. 98 Figure 3.3: (a-d) FTIR spectra of a) 12 nm, b) 25 nm, c) 50 nm, d) 140 nm-thick PBLG brushes treated with chloroform (red) and acetone (black (e) Average PBLG chain tilt angles from the surface normal as a function of PBLG brush thickness for chloroform (green triangles) and acetone-treated (red circles) PBLG brushes. (f) Average LC tilt angles from the surface normal as a function of PBLG brush thickness for chloroform (green triangles) and acetone-treated (red circles) PBLG brushes. 99 Figure 3.4: 3D AFM images and 2D height profiles of (a) 12 nm, (b) 25 nm, (c) 50 nm, and (d) 140 nm thick PBLG brushes in collapsed state after treated with chloroform. Dotted white lines show location of cross-section of 3D images used to generate 2D height profiles. FT-IR confirmed that 12 nm thick brushes had the smallest tilt angle (35.4°) from the 100 surface normal among the four samples (12 nm, 25 nm, 50 nm, and 140 nm) in the collapsed state, though the angle was still higher than that for 12 nm thick brushes in quenched state. Moreover, an increasing trend in average tilt angle (35 ± 1° to 45 ± 1°) with increasing brush thickness was observed for PBLG in the collapsed state (Figure 3.3e). Overall, we found that brush thickness had significant effects on PBLG tilt angle, but the effects showed opposite trends for the collapsed and quenched states. 3.4.2 LC response to PBLG brushes We learnt that PBLG brushes in the collapsed and quenched states were vastly different in terms of surface morphology and average tilt angle. In addition, Machida et al. observed a spiral texture of nematic liquid crystals (LC) on PBLG brushes, but not on a cast film or LB film 17. Therefore, it was of interest to further investigate the effect of surface morphology and polymer tilt angle on the ordering of a nematic LC. First, we attempted to put LC films of 5CB, hosted within 20 μm-thick metal grids, on either collapsed or quenched 50 nm PBLG brushes. The LC film in contact with collapsed brushes exhibited a bright optical texture, when viewed between crossed polarizers, indicative of planar alignment of LCs with average LC tilt angles of 87 ± 3° (Figure 3.5b,d), while the LC film supported on quenched brushes exhibited a dark optical appearance, indicating homeotropic (perpendicular) alignment of the LC (Figure 3.5c,e). We also confirmed that 5CB was not a good solvent for PBLG and cannot change the organization of PBLG brushes. Specifically, no significant change in the average tilt angle of both collapsed and quenched brushes was observed after contact with 5CB (section S3). Moreover, we verified that the 5CB film in contact with either initiator-functionalized substrates or a cast film of PBLG exhibited a bright optical texture, indicative of planar alignment of the LCs (section S4). Therefore, the significant differences in LC response between PBLG brushes in collapsed and quenched states 101 must be from changes in brush organization. From the AFM and FT-IR results discussed above, two possible factors are surface roughness and polymer orientation. Figure 3.5: (a) Molecular structure of 5CB. (b,c) optical micrographs (crossed polarizers) and corresponding schematic illustrations of LC thin films on 50 nm PBLG brushes in a (b,d) collapsed and (c,e) quenched conformation. 102 Subsequently we explored how the PBLG brushes regulated the orientations of the LCs. We wanted to find out whether the change in LC orientation reported above was a result of the PBLG brushes directly changing the lowest free energy orientation of the LCs (i.e. easy axis) or a result of a change in the energy cost to reorient the LCs at the PBLG brush interface (surface anchoring energy; S5). We confined a 4 μm thick LC film between two quenched 140 nm thick PBLG brushes (Figure S3a) and observed a dark optical texture indicative of homeotropic alignment (Figure S3b). Confining 4 μm thick LC films between collapsed 140 nm thick brushes resulted in a uniform bright optical texture, when viewed between crossed polarizers, indicating planar alignment of the LCs (Figure S3c). These observations implied that a change in the orientation of PBLG within brushes induces a change in the lowest free energy orientation of the LCs (i.e. easy axis). Next, we examined LC response to PBLG brushes of different thicknesses from 12 nm to 140 nm. We observed that LCs supported on quenched PBLG brushes exhibited a dark optical texture indicative of homeotropic alignment regardless of polymer brush thickness with a LC tilt angle of 0° from the surface normal. Since the roughness changed significantly from 32.6 nm for 140 nm thick brushes to 3.49 nm for 12 nm thick brushes while there was no change in LC tilt angle with brush thickness, it suggested that the teepee-like structures and roughness only had a subtle effect on LC alignment. In contrast, we more readily observed changes in optical texture of LCs supported on collapsed PBLG brushes. Specifically, LCs supported on collapsed 140 nm thick brushes exhibited planar alignment while LCs on collapsed 12 nm thick brushes showed homeotropic alignment. In addition, collapsed PBLG brushes with intermediate thickness (~20 – 50 nm) exhibited a range of interference colors (Figure S4). As determined by the Michel-Levy birefringence chart, the interference colors were used to obtain the LC tilt angle (Fig. 3f). Interestingly, for collapsed brushes, LC tilt 103 angles were proportional to PBLG brush thickness. Although AFM showed subtle teepee-like features for 12 nm thick brush, its roughness was only 2.7 nm while samples of other thicknesses had a roughness of less than 1.7 nm. It indicated that surface roughness was not a main factor in changes in LC behavior. Therefore, we believe that PBLG tilt angles play a major role on LC response through intermolecular interactions. For that reason, we wanted to determine whether the interactions were driven by dipoles or dispersion forces. Since 5CB possesses positive dielectric anisotropy and positive anisotropic polarizability (Δε = + 13, Δn = + 0.18),29 we decided to test the response of N-(4-methoxybenzylidene)-4-butylaniline (MBBA) which has negative dielectric anisotropy and positive anisotropic polarizability (Δε = -0.7, Δn = + 0.21).29 We observed that MBBA LC films in contact with collapsed 50 nm brushes exhibited a bright optical texture while the LC film supported on quenched 50 nm brushes exhibited a dark optical appearance (Figure S5). Since 5CB and MBBA gave similar results, we concluded that the dispersion interactions between PBLG and LC were the primary cause of LC responses. 3.4.3 Thermal Denaturation of PBLG Brushes As discussed above, there was a strong correlation between PBLG tilt angles and brush thickness (Figures 3.3). However, the effect of brush thickness on PBLG tilt angle diminished for longer brushes. Thus, it was of interest to explore the importance of polymer stiffness on the tilt angle of PBLG brushes and their assemblies. Since the intramolecular hydrogen bonds between backbone N-H and C=O groups dictate the helical secondary structure of PBLG brushes, we attempted to reduce polymer stiffness by disrupting the hydrogen bonds via thermal treatment (Figure 3.6a).30 Specifically, samples of acetone-quenched 130 nm thick PBLG brushes were heated at 180°C for 0 min, 4 min, 15 min and 300 min before re-quenching them with chloroform and acetone. 104 Figure 3.6: (a) Schematic illustration of PBLG chain; (b-e) AFM 3D images and (f-i) corresponding micrographs of 20 μm-thick LC films supported on PBLG brushes heated for b,f) 0 min, c,g) 4 min, d,h) 15 min, e,i) 300 min. j) FTIR spectra of acetone- quenched 130 nm-thick PBLG brushes following 0 min, 4 min, 15 min, and 300 min of heating at 180 °C, and re-quenching in acetonek) Average PBLG tilt angles estimated via FTIR as a function of heating time. Scale bar in the micrograph is 200 μm. 105 AFM was used to probe polymer organization. A continuous transition from teepee-like assemblies to flat morphology was observed with an increase in heating time from 0 min to 300 min (Figure 3.6b-e), suggesting an increase in PBLG tilt angle from the surface normal. We further tested LC response of 5CB on heat-treated samples and observed a continuous transition of optical textures of LC films from dark to bright on 0 min and 300 min heated samples respectively (Figure 3.6f-i). Using a Michel-Levy chart, LC tilt angles from the surface normal were estimated as 0° at 0 min, 18 ± 9° at 4 min, 47 ± 12° at 15 min, and 89 ± 1° at 300 min of heating at 180°C. Additionally, the FT-IR spectra showed an increase in the intensity of random coil peaks (amide I- 1680- 1691 cm-1) as a function of heating time (Figure 3.6j).31,32 The presence of random coil peaks was a qualitative indicator of helix to coil conversion and implied a decrease in polymer persistence length. Moreover, the PBLG tilt angle from the surface normal increased with heating time, thereby indicating the critical role of polymer stiffness in polymer tilt angle (Figure 3.6k). Interestingly, both PBLG and LC tilt angles increased with longer heating time. This correlation was consistent with the results obtained from the case of variation in PBLG thickness discussed above. It confirmed again the relationship between PBLG orientation and LC tilt angle. As indicated by FT-IR data, longer heating time resulted in a higher amount of coil segments in polymer chains, thereby decreasing polymer rigidity. Consequently, polymer brushes were more likely to flop over and could not stand up in a vertical orientation to form teepee-like assemblies. The results emphasized the critical role of polymer stiffness on the polymer tilt angle and organization. It also explained why teepee-like structures have never been reported for any coil-type polymer brushes. 106 3.4.4 Dependence of PBLG brush orientation on solvent quality As we observed a large effect of the quenching process on PBLG tilt angle, we wanted to explore other solvents beyond the solvent pair of chloroform and acetone. First, we tested methanol which is a nonsolvent for PBLG but miscible with chloroform. A tilt angle of 15.7° was observed for 140 nm PBLG brushes after quenching process with chloroform and methanol (Table 3.2). However, when water, which is a nonsolvent for PBLG but immiscible with chloroform, was used to replace acetone, a tilt angle of 45.5° was obtained for 140 nm PBLG brushes after quenching process. It indicated that two solvents used in the quenching process must be miscible with each other to obtain a vertical orientation of PBLG brushes. It also implied that the extraction of a good solvent from swollen PBLG brushes into a nonsolvent played an important role in the quenching process. Hence, we further investigated the effect of extraction kinetics on PBLG orientation. Specifically, we incubated chloroform-solvated PBLG brush samples in acetone for either 20 seconds or 20 minutes before transferring the samples into water for 1 minute and drying. Since water is miscible with acetone, but immiscible with chloroform, we expected that water can stop the extraction process, and the extraction time would be either 20 seconds or 20 minutes. Interestingly, FT-IR revealed that PBLG tilt angle was 41.3 ± 3.4° for 20-second incubation time and 15.9 ± 0.93° for Table 3.2. Average PBLG tilt angles measured via FT-IR following sequential solvent treatment. 107 20-minute incubation time (Table 3.2). Moreover, LCs on PBLG brushes with 20 min incubation time had homeotropic alignment while in case of 20 second incubation time, LCs showed large patches of planar anchoring and large patches of homeotropic anchoring. It suggested that the extraction time was critical to the success of quenching. 3.4.5 Reorganization of PBLG brushes with mixed solvents Table 3.3: Average Orientations of 140 nm Thick PBLG Chains Measured via FT-IR Following Treatment with Chloroform-Acetone Mixtures Up to this point, we gained insights into the quenching process, and the importance of solvent miscibility and extraction time to PBLG orientation. Thus, we applied those understandings to obtain a greater diversity of PBLG brush organization. Instead of pure acetone, a solvent mixture of chloroform and acetone was used to quench chloroform-solvated PBLG brushes. We expected that a solvent mixture would have different extraction rate than pure acetone and may cause different types of polymer assemblies. PBLG brushes (140 nm thick) were swollen in chloroform before transferring into solvent mixtures of either 10% or 25% v/v chloroform in acetone. FT- IR revealed PBLG tilt angles of 37° and 39° for 10% and 25% v/v chloroform in acetone, respectively. The angles were between those treated with pure chloroform (45°) and pure acetone (15°) (Table 3.3 and Figure S7). The results suggested that PBLG may 108 adopt intermediate states between the collapsed and teepee-like states. This observation is reasonable because mixed solvents should extract chloroform from solvated brushes less effectively than pure acetone does. We further characterized the samples quenched with mixed solvent by using 20 μm- thick LC films and observed complex optical textures. For the samples treated with 10% v/v chloroform in acetone, heterogenous textures showed small circular domains with tilted alignment and diameters of 6 ± 4 μm (Figure 3.7a; red bordered box). Retardance measurements showed that the region inside the small circular domains had LC tilt angles of 47 ± 9° while the region outside the circular domains showed LC tilt of 49 ± 9°. There were also large circular homeotropic domains with diameters of 26 ± 3 μm (Figure 3.7a; yellow bordered box) From the relationship between PBLG and LC tilt angles established above (Figures 3.3 and S6), we estimated that the PBLG in the inner and outer regions of small circular tilted domains had a tilt angle of ~38° and ~39° from the surface normal, respectively while the tilt angle of PBLG in large circular homeotropic domain was between 15° and 35° . We also estimated that the large homeotropic domains occupied ~ 6.5% of the LC film while the tilted domains and its surrounding area occupied 93% of the film. Therefore, the average PBLG tilt angle across the entire sample was calculated to be between 37° and 38.3° from the surface normal, and it was in good agreement with that average tilt angle measured via FT-IR (37°± 1.1°). The results suggested that we can use LCs as an alternative approach to estimate PBLG tilt angle. To evaluate the effectiveness of this approach, we used it to estimate PBLG tilt angle for 140 nm thick brushes treated with 25% v/v chloroform in acetone and compare it with the angle obtained via FT-IR. Using LCs, an average PBLG tilt of 109 33.9° ≤ θ ≤ 39.4°, was obtained and it matched well with that measured via FTIR (θ = Figure 3.7: (a,b) 20 μm thick LC films supported on (a) 140 nm-thick and (b) 50 nm- thick PBLG brush quenched with 10% chloroform in acetone showing (a) large circular homeotropic domains surrounded by continuous planar domain containing smaller circular domains with tilted LC anchoring (red inset), and (b) micrometer-scale circular domains with tilted LC anchoring (blue inset). (c,d) corresponding 2D AFM cross sections and (e,f) 2D height profiles (blue dashed lines) of (c,e) 140 nm and (d,f) 50 nm thick PBLG brushes, respectively. 110 39.0 ± 1°; Table 3.3). It proved that LCs approach can be used to estimate PBLG tilt angle for micrometer scale regions. As LC texture suggested there were differences in organization of PBLG across small circular tilted domains, AFM was used to further characterize the circular domains, and it showed that the circular domains had ring-like shape (Figure 3.7c). Height profiles of the rings showed that the ring-like region was roughest with distinct teepee-like features (91 ± 17 nm- height PBLG) while the inside (45 ± 7 nm) and outside (25 ± 17 nm) of the domains showed non-distinct surface topography (Figure 3.7e and Section S10). These differences in surface topography across the ring indicate different PBLG tilt angles, such that taller assemblies on the ring represent clusters of PBLG with smaller tilt with respect to the surface normal. Overall, the AFM and LC results suggest that LCs can probe the presence of micrometer-scale heterogeneity in PBLG brush topography. As PBLG tilt angles and surface morphology are dependent on brush thickness, it was of interest to investigate how 50 nm PBLG brushes would respond to a solvent treatment of 10% v/v chloroform in acetone. LC films supported on 50 nm PBLG brushes after the treatment exhibited aperiodic circular tilted domains with a continuous gradient of interference colors and diameter in the range of 2 - 20 μm (blue box, Fig. 7b). The center of the domain had LC tilt angles of θ = 58 ± 5° from the surface normal which correspond to PBLG tilt angles of θ = 40 ± 0.5° from the surface normal. AFM showed that the circular domains had ring-like shape (Figure 3.7d). However, a height profile across the ring looked quite different from that of 140 nm brushes. AFM also revealed that the ring had a quite flat topography while the inner and outer regions had the teepee-like structures (Figures 3.7f and S9). 111 Overall, quenching PBLG brushes with solvent mixtures resulted in circular/ring-like structures which were distinct from uniform teepee-like structures if treated with pure acetone. We hypothesized that the formation of ring-like structures was from the nucleation of either chloroform or acetone droplets at the brush interface due to phase separation. While we do not fully understand the mechanism behind the localized phase separation of chloroform and acetone, we observed a similar phenomenon in an analogous system. In that system LC films supported on acetone- quenched 140 nm brushes were immersed in a chloroform saturated aqueous solution (Cchloroform ~ 1% v/v). Initially, the LC films became isotropic phase indicating the penetration of chloroform into the LC film. After a two-hour incubation time, the LC film on brushes was removed from the chloroform saturated aqueous solution and heated to 60 °C for 30 min to completely remove chloroform from the LC film. After that, it was cooled to room temperature before characterizing with a polarized microscope. Micron-scale circular tilted LC domains which had similar shape to those seen in the case of mixed solvent treatment were observed (Figure S10). In addition, a formation of circular tilted domains did not occur if pure water was used instead of chloroform saturated aqueous solution. Thus, those observations supported the idea that chloroform diffused from the aqueous phase through the LC film and nucleated into droplets on the polymer brush interface, resulting in circular tilted domains. 3.5 Conclusion Through this study, we carefully characterized the effect of solvent treatments and polymer thickness on the organization and orientation on PBLG brushes. We also investigated how the changes of PBLG impacted the ordering of LC film supported on the brush. In the teepee-like state, the size and shape of the teepee-like assemblies were dependent on polymer thickness (or polymer chain length). The teepee-like assemblies 112 of 140 nm thick brushes were bigger and more pronounced than those of 12 nm brushes, but the number of brush assemblies was smaller for 140 nm PBLG. The 140 nm brushes also resulted in higher surface roughness. Moreover, for quenched brushes, there was a decrease in average PBLG tilt angle from the surface normal with increasing brush thickness. In contrast, for chloroform-treated collapsed brushes, 12 nm thick PBLG brushes exhibited teepee-like structures while brushes thicker than 25 nm looked quite flat. In addition, chloroform-treated brushes showed an increase in average PBLG tilt angle with increasing brush thickness. We also observed that LCs exhibited homeotropic alignment for all brush thicknesses in the teepee-like state. Conversely, LC tilt angles from the surface normal increased from 0° for 12 nm brushes to 90° for 140 nm brushes in the collapsed state. We concluded that there was a proportional correlation between LC and PBLG tilt angles for collapsed-state brushes. We also found that the dispersion interactions between PBLG brushes and the LC probe was the main driving force for the observed LC responses. Through thermal denaturation experiments, we showed the dependence of PBLG tilt angle and organization on polymer stiffness. Acetone-quenched PBLG brushes which had more random coil content formed much smaller and less uniform teepee-like assemblies and resulted in higher LC tilt angles from the surface normal. In addition to polymer stiffness, solvent quality and extraction time also contributed to the success of quenching and brush organization. A good solvent and nonsolvent of PBLG must be miscible with each other so that the nonsolvent can extract the good solvent from PBLG and force the brushes into assemblies. Moreover, when a mixture of chloroform and acetone was used, circular domains with locally distinct organizations of PBLG were observed. 113 While we currently do not fully understand the factors behind the differences in the domains and how to control the size and periodicity of the domains, we predict that it will strongly depend on factors such as interaction parameters between the solvents and the polymer, interaction between solvent and non-solvent, polymer thickness, grafting density and surface roughness of the PBLG brush. However, our work demonstrated that LC can be utilized for imaging PBLG so that it can complement FT- IR and AFM to characterize PBLG brushes in terms of PBLG tilt angles and local polymer organization, especially for heterogenous and micron-sized patterns. 3.6 Supporting Information S1. Chain Length and Grafting Density Calculation For 50 nm-thick brushes, the average tilt angles with respect to the surface normal of the collapsed and quenched PBLG brushes are 43.0 ± 0.3° and 20.2 ± 0.8°, respectively. Thus, the polymer chain length is calculated as For PBLG brush, each monomer unit contributes 1.5 Å along the helical axis of the chain, so the degree of polymerization is The molecular weight of polymer is Mn = 353 × 219 g/mol = 77307 g/mol (3) Polymer brush thickness in collapsed state is determined as hcol = l × sin(90° − 43°) = 38.7 nm (4) 114 S2. Estimation of Average PBLG Tilt Angle via FTIR We estimated the average tilt angles of the PBLG in the collapsed and quenched brushes using the dichroic ratio from two characteristic amide peaks (amide I - 1654 cm-1 and amide II - 1550 cm-1) via FTIR. Using Equation 6 below, we estimated the average tilt angles with respect to the surface normal of the collapsed and quenched PBLG brushes to be 43.0 ± 0.3° and 20.2 ± 0.8°, respectively Where, R is the dichroic ratio of the amide I and amide II bands, AAmide I is the area under the amide I band, AAmide II is the area under the amide II band, K (3.6) is a proportionality constant calculated from the transmission FTIR spectrum of a polyglutamate Langmuir Blodgett film, αI and αII are transition dipole moment angles of the amide I (39° from helical axis) and amide II (75° from helical axis) bands respectively, and θ is the average PBLG helix tilt angle. S3. Effect of PBLG Brush Pretreatment with 5CB on PBLG Tilt Angle Table S1: Average PBLG tilt angles of brushes treated with either chloroform or acetone before and after a 15 min incubation of brushes in 5CB. 50 nm PBLG Collapse Quench (Dichroic Ratio) (Dichroic Ratio) NO LC 43 ± 0.3° 20 ± 0.8° LC Treated 43 ± 0.4° 19 ± 0.9° 115 Figure S1: FTIR spectra of 50 nm-thick PBLG brushes treated with chloroform (red) and acetone (black), and subsequently incubated in the 5CB prior to rinsing off the LC with acetone and drying. The labeled peaks are amide I (1654 cm-1, backbone carbonyl stretching) and amide II (1550 cm-1, C-N stretching) of the α-helices, and the ester side chain (1734 cm-1). 116 S4. LC Response to Initiator-functionalized Substrates and PBLG Thin Films Figure S2: Optical micrographs between crossed polarizers of 20 μm-thick LC film supported on a) 3-Aminopropyldimethylethoxysilane (APDMES; initiator)- functionalized substrate, and b) PBLG thin film in the absence of initiator. Scale bar is 200 μm. We performed two controls to determine the LC response to both the initiator that tethers PBLG chains to the solid substrate and free PBLG in the absence of initiator. In our first control, a glass substrate was functionalized with initiator (3- Aminopropyldimethylethoxysilane (APDMES)) and subsequently contacted with a 20 μm-thick LC film. The LC films in contact with initiator-functionalized substrates exhibited a bright optical texture, when viewed between crossed polarizers, indicative of planar alignment of LCs (see Figure S2a). In our second control, we observed that LCs in contact with a thin film of PBLG free of initiator, that was formed by depositing chloroform-solubilized PBLG chains from solution onto a glass substrate via evaporation, exhibited planar alignment (see Figure S2b). The controls described above indicate that differences in LC response between collapsed and quenched PBLG brushes is a consequence of a change in the organization of the brush. The controls described above indicate that differences in LC response between collapsed and quenched PBLG brushes is a consequence of a change in the organization of the brush 117 S5. PBLG Brushes Change the Easy Axis of LCs Figure S3: a) Schematic illustration of 4 μm-thick LC film confined between two 50 nm-thick PBLG brushes treated with either b) acetone or c) chloroform. When viewed between crossed polarizers, b) homeotropic alignment of LCs was observed for the LC films confined between acetone-quenched brushes, and c) planar alignment of LCs was observed for LC films confined between chloroform-collapsed brushes. As described in the main text, 20 μm-thick LC films with boundary conditions of air and PBLG brushes that were pretreated with either chloroform or acetone exhibited different optical textures. We determined if the observed difference in LC alignment for chloroform and acetone-treated brushes was a result of a decrease in the surface anchoring energy of the LC or a change in the easy axis of the LCs by confining thin LC films between 2 glass slides decorated with 50 nm-thick PBLG brushes that were pretreated with either chloroform or acetone and separated by 4 μm diameter glass spacers. For LC films confined between acetone-treated brushes, a dark optical texture was observed when viewed between crossed polarizers, indicating homeotropic alignment of the LC. In contrast, LC films confined between chloroform treated-brushes exhibited a bright optical texture indicating non-homeotropic alignment of the LC. This result indicates that PBLG brushes change the easy axis of the LC. 118 S6. LC Response to PBLG Brushes of Variable Thickness Figure S4: (a) Schematic illustration of LC tilt at PBLG brush interface and (b-e) micrographs (crossed polarizers) of LC thin films on chloroform-treated PBLG brush with (b,c) thickness (z) gradient of (b) z= 16-20 nm (increasing left to right), (c) z= 25- 32 nm (increasing left to right), (d) z= 32 nm, (e) z= 40 nm, (f) z= 50 nm. 119 S7. Probing Intermolecular Interactions Between LCs and PBLG Figure S5: Optical micrographs of 20 μm-thick a,b) 5CB and c,d) MBBA film supported on a,c) chloroform-collapsed and b,d) acetone-quenched 50 nm thick PBLG brushes between crossed polarizers. Scale bar is 200 μm. 120 S8. Relationship Between LC and PBLG Tilt Angles Figure S6: Average LC tilt angles from the surface normal as a function of PBLG chain tilt angle for solvent-treated PBLG brushes. Data points show mean values and the error bars represent 1 standard deviation (n=3). 121 S9. FT-IR Spectra of Mixed Solvent Treated 140 nm-thick PBLG Brushes Figure S7: FT-IR spectra of 140 nm thick PBLG polymer brushes treated with chloroform and acetone mixtures. Four different solvent compositions (0%, 10%, 25% and 100% chloroform in acetone) were used. The labeled peaks are amide I (1654 cm- 1, backbone carbonyl stretching) and amide II (1550 cm-1, C-N stretching) of α-helical secondary structure, the ester side chain (1734 cm-1). 122 S10. 2D and 3D Height Profiles of 140 nm-thick PBLG Brushes via AFM Figure S8: (a,c,d) 3D and (b,d,f) 2D height profiles measured via AFM of a,b) inner domain, c,d) ring-like region, e,f) outer continuous domain of 140 nm-thick PBLG brushes treated with 10% chloroform in acetone. Dotted white lines show cross- section of 3D profile used to generate 2D height profiles. 123 S11. 2D and 3D Height Profiles of 50 nm-thick PBLG Brushes via AFM Figure S9: (a,c,d) 3D and (b,d,f) 2D height profiles measured via AFM of a,b) inner domain, c,d) ring-like region, e,f) outer continuous domain of 50 nm-thick PBLG brushes treated with 10% chloroform in acetone. Dotted white lines show cross- section of 3D profile to generate 2D height profiles. 124 S12. In Situ Reorganization of PBLG Brushes Figure S10: Micrographs of 20 μm thick 5CB films, a,d) under bright field microscopy and b,c,e,f) between crossed polarizers, supported on acetone-quenched 140 nm PBLG brushes in air after 2 hour incubation in a-c) water and d-f) water saturated with chloroform (Cchloroform ≤ 1% v/v). 125 3.7 References 1. Zoppe, J. O. et al. Surface-Initiated Controlled Radical Polymerization: State-of- the-Art, Opportunities, and Challenges in Surface and Interface Engineering with Polymer Brushes. Chem. Rev. 117, 1105–1318 (2017). 2. Barbey, R. et al. Polymer Brushes via Surface-Initiated Controlled Radical Polymerization: Synthesis, Characterization, Properties, and Applications. Chem. Rev. 109, 5437–5527 (2009). 3. Chen, W.-L., Cordero, R., Tran, H. & Ober, C. K. 50th Anniversary Perspective : Polymer Brushes: Novel Surfaces for Future Materials. Macromolecules 50, 4089– 4113 (2017). 4. Husemann, M. et al. Surface-Initiated Polymerization for Amplification of Self- Assembled Monolayers Patterned by Microcontact Printing. Angew. Chem. Int. Ed. 38, 647–649 (1999). 5. Zhao, B. & Brittain, W. J. Polymer brushes: surface-immobilized macromolecules. Prog. Polym. Sci. 25, 677–710 (2000). 6. von Werne, T. A. et al. A Versatile Method for Tuning the Chemistry and Size of Nanoscopic Features by Living Free Radical Polymerization. J. Am. Chem. Soc. 125, 3831–3838 (2003). 7. Chen, C. S., Mrksich, M., Huang, S., Whitesides, G. M. & Ingber, D. E. Geometric Control of Cell Life and Death. Science 276, 1425–1428 (1997). 8. Burkert, S. et al. Protein Resistance of PNIPAAm Brushes: Application to Switchable Protein Adsorption. Langmuir 26, 1786–1795 (2010). 126 9. Temyanko, E., Russo, P. S. & Ricks, H. Study of Rodlike Homopolypeptides by Gel Permeation Chromatography with Light Scattering Detection:  Validity of Universal Calibration and Stiffness Assessment. Macromolecules 34, 582–586 (2001). 10. Wang, Y. & Chang, Y. C. Synthesis and Conformational Transition of Surface- Tethered Polypeptide:  Poly(l-glutamic acid). Macromolecules 36, 6503–6510 (2003). 11. Hwang, Y. et al. Piezoelectric properties of polypeptide-PMMA molecular composites fabricated by contact charging. Polymer 52, 2723–2728 (2011). 12. Jaworek, T., Neher, D., Wegner, G., Wieringa, R. H. & Schouten, A. J. Electromechanical Properties of an Ultrathin Layer of Directionally Aligned Helical Polypeptides. Science 279, 57–60 (1998). 13. Wu, J.-C., Chen, C.-C., Chen, K.-H. & Chang, Y.-C. Controlled growth of aligned α-helical-polypeptide brushes for tunable electrical conductivity. Appl. Phys. Lett. 98, 133304 (2011). 14. Whitesell, J. K. & Chang, H. K. Directionally Aligned Helical Peptides on Surfaces. Science 261, 73–76 (1993). 15. Luijten, J., Groeneveld, D. Y., Nijboer, G. W., Vorenkamp, E. J. & Schouten, A. J. Cross-Linking-Induced Permanently Perpendicular Helix Orientation in Surface- Grafted Polyglutamate Films. Langmuir 23, 8163–8169 (2007). 16. Wang, Y. & Chang, Y. C. Preparation of Unidirectional End-Grafted α-Helical Polypeptides by Solvent Quenching. J. Am. Chem. Soc. 125, 6376–6377 (2003). 127 17. Machida, S. et al. A Chiral Director Field in the Nematic Liquid Crystal Phase Induced by a Poly(.gamma.-benzyl glutamate) Chemical Reaction Alignment Film. Langmuir 11, 4838–4843 (1995). 18. Popov, P., Mann, E. K. & Jákli, A. Thermotropic liquid crystal films for biosensors and beyond. J. Mater. Chem. B 5, 5061–5078 (2017). 19. Bukusoglu, E., Bedolla Pantoja, M., Mushenheim, P. C., Wang, X. & Abbott, N. L. Design of Responsive and Active (Soft) Materials Using Liquid Crystals. Annu. Rev. Chem. Biomol. Eng. 7, 163–196 (2016). 20. Miller, D. S., Wang, X. & Abbott, N. L. Design of Functional Materials Based on Liquid Crystalline Droplets. Chem. Mater. 26, 496–506 (2014). 21. Li, X. et al. Engineering the anchoring behavior of nematic liquid crystals on a solid surface by varying the density of liquid crystalline polymer brushes. Soft Matter 14, 7569–7577 (2018). 22. Tran, H., Zhang, Y. & Ober, C. K. Synthesis, Processing, and Characterization of Helical Polypeptide Rod–Coil Mixed Brushes. ACS Macro Lett. 7, 1186–1191 (2018). 23. Wieringa, R. H. et al. Surface Grafting of Poly(l-glutamates). 2. Helix Orientation. Langmuir 17, 6485–6490 (2001). 24. Vlasova, E., Volchek, B., Tarasenko, I. & Vlasov, G. Spectroscopic Investigation of Polypeptide Plane Brushes. Macromol. Symp. 305, 116–121 (2011). 25. Dazzi, A. et al. AFM–IR: Combining Atomic Force Microscopy and Infrared Spectroscopy for Nanoscale Chemical Characterization. Appl. Spectrosc. 66, 1365– 1384 (2012). 128 26. Daly, W. H. & Poché, D. The preparation of N-carboxyanhydrides of α-amino acids using bis(trichloromethyl)carbonate. Tetrahedron Lett. 29, 5859–5862 (1988). 27. Chang, Y.-C. & Frank, C. W. Vapor deposition-polymerization of α-amino acid N- carboxy anhydride on the silicon (100) native oxide surface. Langmuir 14, 326–334 (1998). 28. Wang, Y. & Chang, Y. C. Preparation of Unidirectional End-Grafted α-Helical Polypeptides by Solvent Quenching. J. Am. Chem. Soc. 125, 6376–6377 (2003). 29. Mushenheim, P. C. & Abbott, N. L. Hierarchical organization in liquid crystal-in- liquid crystal emulsions. Soft Matter 10, 8627–8634 (2014). 30. Ushiki, H., Tanaka, F. & Mita, I. Study on the fluorescence depolarization of a chromophore attached to polymer chain end. Aggregation behaviours of poly(γ- benzyl l-glutamate) in semi-concentrated solution up to liquid crystal region. Eur. Polym. J. 22, 827–829 (1986). 31. F. M. EL-Mahdy, A. & Kuo, S.-W. Diphenylpyrenylamine-functionalized polypeptides: secondary structures, aggregation-induced emission, and carbon nanotube dispersibility. RSC Adv. 8, 15266–15281 (2018). 32. Kotharangannagari, V. K., Sánchez-Ferrer, A., Ruokolainen, J. & Mezzenga, R. Thermoreversible Gel–Sol Behavior of Rod–Coil–Rod Peptide-Based Triblock Copolymers. Macromolecules 45, 1982–1990 (2012). 129 CHAPTER 4 Nanopatterned helical polypeptide rod-type brushes via electron beam lithography 4.1 Abstract In what we believe the first work of nanopatterned polypeptide rod-type brushes, we combined electron-beam lithography (EBL) and surface-initiated vapor-deposition polymerization (SI-VDP) to prepare nanopatterned helical poly(benzyl-L-glutamate) (PBLG) brushes with feature size as small as 65 nm. After a solvent treatment with chloroform and acetone, PBLG brushes formed well-defined teepee-like nanostructures inside patterned features, indicating vertical orientations of PBLG brushes. In comparison, polystyrene (PS) brushes grown from same-sized patterned ATRP initiators were collapsed, thereby broadening the feature size. Thus, this work demonstrated the effect of polymer stiffness on polymer conformation. Moreover, the number of teepees per feature increased with the patterned feature size. When the edge- to-edge spacing between adjacent features was comparable to polymer length, PBLG brushes from adjacent features started to interact with each other. PBLG brush thickness also increased with patterned feature size, and that was probably due to greater exposure to vapor-phase initiators and monomers during the synthesis. Lastly, the additional presence of poly(N-isopropyl acrylamide) PNIPAM brushes in unpatterned areas can prevent the interaction of PBLG from adjacent features, resulting in more well-defined nanostructures of PBLG brushes. Note: Hai Tran did the synthesis of monomer, the initiators and polymer brushes, and characterization of brushes using AFM and ellipsometry while Yuming Huang did EBL and characterization via SEM. 130 4.2 Introduction A polymer brush system which is composed of polymer chains covalent attached to a surface enabled versatile modifications of surface chemistry and morphology through a thin film of polymer coating.1,2 Among various polymer brush architectures such as block copolymer brushes3, mixed brushes4, or detached brushes5, patterned polymer brushes have gained a great interest because they offer possibilities to precisely control targeted chemical functional groups in a specific surface location. The advances in lithography and polymerization methods have contributed to the dramatic progress in the research of patterned polymer brushes.6,7 In general, there are two common approaches to fabricate patterned polymer brushes. The first approach is the “top-down” in which pre-synthesized homogenous polymer brushes are selectively removed by degradation under radiation exposure or by mechanical forces.7 A pattern resolution of 50 nm can be achieved through the “top-down” via electron beam lithography (EBL) 8. While the “top-down” approach is quite straightforward, it has some drawbacks, such as post-processing residue, or over-etching.6 The second approach is the “bottom-up” in which patterns of surface-immobilized initiators serve as a template from which polymer brushes are grown via surface-initiated polymerization. 6,7 There are two ways to produce patterned initiators. In the first method, uniform layers of initiators were immobilized to a substrate before deep UV7,9 or EBL10 were utilized for ablative patterning of the initiators. In the second method, patterned resist layers served as a mask for the deposition of initiators into exposed area.11–13 Patterned brushes have been used for cell growth direction14, targeted binding 15. biosensors6,16, patterned functionalities6,17,18, platform to study cell-surface interactions19, thermoresponsive surfaces17 , antifouling surfaces 20, actuator 21, or controlling surface wettability 22. 131 Moreover, it is worth noting that fundamental properties of nanopatterned polymer brushes (NPB) are different from those of homogenous brushes and micropatterned brushes because for NPB, polymer contour length is comparable to pattern size. For example, there is a scaling behavior for NPB when the height of NPB is proportional to the pattern size. 23 Specifically, Ahn et al. showed that for pattern size of 1.8 μm, the polymer brush thickness was 300 nm while for pattern size of 600 nm, the thickness was only 170 nm even under same polymerization conditions via SI-ATRP due to less chain crowding. 24 Furthermore, Rastogi et al. reported that line broadening of nanopatterned polymer brushes was observed due to polymer chain relaxation in the voided regions.8 Chen et al. used the “bottom-up” approach to fabricate patterns of polyelectrolyte brushes with the smallest feature dimension of 200 nm via deep-UV lithography and observed a reduction in degrafting for smaller patterns due to an increase of stress relaxation at such dimensions.25 Additionally, Yu et al. found that when NPB was swollen in a good solvent and pitch distance was similar to the polymer size, polymer chains from adjacent patterned features can interact with each other.20 Although patterned polymer brushes have been studied extensively, most polymer brushes in the reported literature were coil-type polymers. In contrast, there is very limited work on patterned rod-type polymer brushes, especially for polypeptide brushes. 26,27 Due to the intramolecular hydrogen bonding, α-helix polypeptide brushes with reported persistence lengths between 100 and 200 nm,28 have a much higher stiffness than any coil-type brushes. For that reason, they have been commonly used as a role-model for studies of rod-type brushes. Owing to their persistence length, helical polypeptide brushes have a unique property which is their orientation (i.e. tilt angles). Moreover, depending on the form, density and composition of the amino acid, polypeptide brushes can respond to external stimuli including such factors as pH, temperature, and ionic strength. 29 Given those interesting properties of polypeptide 132 brushes, their nanopatterned brushes are likely to behave differently from coil-type NPB. However, to the best of our knowledge, there is a complete lack of understandings of helical polypeptide NPB while there are only a few studies on micropatterned polypeptide brushes. For instance, Kratzmuller et al. used microcontact printing (μCP) to produce patterned poly(benzyl-L-glutamate) (PBLG) brushes on gold surface. 30 In another approach, Wang et al. used “bottom-up” approach through photolithography to pattern the initiators from which PBLG brushes were grown via surface-initiated vapor- deposition polymerization (SI-VDP). 31 However, those studies were mainly on the polymer synthesis without much further characterization on polymer behaviors, and more importantly, the patterns were in micron size. Therefore, it was our interest to investigate helical polypeptide NPB and gain insights into behaviors of nanostructured rod-type polymer brushes. Particularly, in this work, we used PBLG brushes due to their stable helical structure and synthesized them via “the bottom-up” approach. EBL was utilized to prepare nanopatterned mask of ZEP-520A ebeam resist for the subsequent vapor-phase deposition of the initiators from which PBLG brushes were grown via SI- VDP. We also examined the scaling behavior of NPB of PBLG by varying the pattern size. Furthermore, we investigated the effect of “solvent quenching” treatment on nanopatterned PBLG. In this treatment, as described by Wang et al.32 for homopolymer brushes, the PBLG rod brushes are swollen in a good solvent such as chloroform and the rod segment extended. Subsequently the brush film is placed in a non-solvent such as acetone, extracting the chloroform, aggregating the PBLG rods which assemble together in stretched chain bundles. Consequently, the randomly oriented rod brushes became vertically aligned. The average tilt angle of PBLG also changed from 49° to 3° from surface normal accordingly. Moreover, an amine function on each rod chain end results in rod bundles with an amine functional tip. Thus, the teepee-like nanostructures of aggregated rod bundles resemble the protein spikes consisting of intertwined protein 133 strands with a functional outermost tip. The protein spikes on virus surfaces are known for host cell recognition, cell invasion and avoiding identification by antibodies.33 However, with homopolymer PBLG brushes, the size and spacings of teepee-like assemblies were poorly controlled with spacing greater than 100 nm. EBL and SI-VDP would allow precise control over the location and spacings of patterned teepees with a resolution limit of well below 100 nm. Therefore, NPB of PBLG may offer interesting possibilities in the study of molecular recognition and cell-surface interactions. Specifically, the functional tips of teepee-like rod bundles may serve as attachment points for cell membrane mimics or accessible transmembrane protein. On the other hand, the spacing of rod bundles can be significantly reduced to less than 50 nm and they can be subsequently modified with ligands for antibody binding. We also attempted to prepare binary rod/coil NPB by the growth of PNIPAM brushes via SI-ATRP in void area so that we can have a surface with complex architecture and dual stimuli-responsive properties. 4.3 Experimental section 4.3.1 Materials Allyl-2-bromo-2-methylpropionate, 2-bromo-2-methylpropionyl bromide (BiBB), anhydrous triethylamine (TEA), hydrazine, copper(I) bromide, pyridine, hydrogen hexachloroplatinate hexahydrate catalyst, basic alumina, anhydrous toluene, N,N- dimethylacetamide (DMAc), styrene, N-isopropylacrylamide (NIPAM), 1,1,4,7,7- penta-methyldiethylenetriamine (PMDETA), triphosgene and γ-benzyl L-glutamate were purchased from Sigma Aldrich. Dimethylethoxysilane and 3-amino- propyldiisopropylethoxysilane (APDIPES) were purchased from Gelest. NIPAM monomer was recrystallized from hexane. Styrene was passed through basic alumina to remove the inhibitor before use. Copper tape (882-L COPPER) with 88.9 μm 134 copperfilm was purchased from Lamart Co. Deionized water with a resistivity of 18.2 MΩ • cm at 25 °C was obtained from Millipore’s Milli-Q Synthesis A10 system. All other solvents were purchased from Fisher Scientific. Si (100) polished wafers were from WRS Materials. Cantilevers were purchased from Applied NanoStructures, Inc. (ACCESS-NC). ZEP-520A positive resist (57k) and ZEDN50 developer were provided by Zeon Chemicals LP. 4.3.2 Synthesis 1. Benzyl L-Glutamate N-Carboxyanhydride34 and (3-(2-bromoisobutyryl)propyl) dimethylethoxysilane (BIDS)35 were synthesized following the reported procedure. 2. Patterned PBLG Brushes Synthesis: Preparation of the E-Beam Mask: Silicon wafers were thoroughly cleaned using a Harrick Plasma Cleaner for 10 minutes, then rinsed with ethanol and acetone, and heated at 110 ℃ to remove moisture. ZEP-520A was spin-coated onto the wafer and baked at 170 ℃ for 2 min. The resist film thickness was around 150 nm (FilMetrics F50-EXR). The resist was then exposed using the JEOL 9500 EBL system with a beam current of 2 nA and a dose of 350 μC/cm2. The exposed film was immersed into ZEDN50 developer for 3 min, transferred into methyl isobutyl ketone for 30 seconds, and rinsed with Propan-2-ol for 30 s. The sample was dried with a stream of nitrogen gas. Residual resist in the exposed area was then cleaned by the Oxford PlasmaLab 80Plus RIE System at 200 mTorr and 40 W for 30 s. Immobilization of Silane Initiators: APDIPES (100 μL) was added into a small glass vial and placed into a closed chamber. The wafer with patterned resist was then placed into the same chamber beside the glass vial. The chamber was then brought to vacuum 135 (1 torr) and placed in an oil bath at 70 ℃ for a controlled time period. The ZEP-520A resist was then removed by sonicating in N, N-dimethylacetamide and isopropanol for 5 min and dried under nitrogen gas. PBLG Brushes Synthesis via SI-ROP: the NCA monomer (10 mg) was added to a 10 mL beaker. The initiator-immobilized wafer was placed on top of the beaker with the pattern at the center. The beaker and the substrate were then put into a closed chamber and brought under vacuum (0.5 torr) and elevated temperature (105℃) for two hours. After the reaction, the substrate was soaked in chloroform overnight, rinsed for 30 s and dried with nitrogen gas. 3. Synthesis of patterned PS brushes The preparation of E-beam mask was the same as that for patterned PBLG brushes. Preparation of patterned immobilized ATRP Initiators: the substrate with immobilized APDIPES was placed in a 20 ml vial with 5 ml anhydrous dichloromethane (DCM). TEA (100 ml) and BiBB (100 ml) were then added to the vial and the reaction was carried out for 90 min. The substrate was then cleaned with DCM and methanol and dried with a nitrogen gas stream. PS Brush Polymerization: PS brushes were synthesized via surface-initiated Cu(0) mediated controlled radical polymerization (SI-CuCRP) using a similar procedure reported elsewhere.36 Glass slide was cut into small pieces with a similar size to initiator- immobilized substrates. Cu tape was stuck onto glass pieces. A piece of Cu-taped glass was clamped together with the substrate with patterned immobilized ATRP initiator substrate by a copper clamp. A Teflon space with thickness of 1 mm was used to separate the pieces. The setup was put into a solution of 2 mL styrene, 1 mL dimethyl sulfoxide (DMSO), 36 μL PMDETA, and 1.5 μL hydrazine, and polymerization was conducted at 136 105oC for a given time. Afterward, the glass slides were removed from the solution and rinsed with dichloromethane, ethanol, and dried with nitrogen gas. 4. Synthesis of PNIPAM Brushes as Secondary Brushes Deposition of ATRP initiator: Samples with patterned PBLG brushes was immersed into a solution of BIDS in anhydrous toluene (50 mM) overnight at room temperature. After that, the samples were washed with toluene, methanol and dried with nitrogen gas. PNIPAM brush polymerization: Samples with patterned PBLG brushes and immobilized ATRP initiator were placed into a 25 mL Schlenk flask which was then evacuated and backfilled with Argon for five times. CuBr (14.5 mg, 0.1 mmol) was put into another 50 mL Schlenk flask with a magnetic stir bar, and it was evacuated and backfilled with Argon for five time. NIPAM (0.56 g, 5 mmol) and PMDETA 54 mg, 65 μL, 0.3 mmol) were dissolved in 10 mL of a 7:3 methanol: water mixture and purged with Argon for 20 minutes before being cannulated into the flask containing CuBr. The mixture was stirred for 20 minutes before being cannulated into the flask containing the PBLG samples. The polymerization was run at room temperature. After that, the samples were sonicated in methanol, ethanol and water and dried with nitrogen. 5. Quenching process The polymer brush samples were immersed into chloroform for 30 minutes. Then they were quickly transferred into acetone for 20 minutes. After that, they were blown dry under nitrogen gas. 4.3.3 Characterization Thickness of PNIPAM brushes outside of patterned PBLG brushes was measured with 137 an Imaging Ellipsometer - Nanofilm EP3 at 532 nm laser at 50-60° angle of incidence. A Cauchy model was used to fit the data, in which the Cauchy layer was representative of the polymer brush. An Asylum MFP-3D atomic force microscope was used to characterize the topography and thickness of patterned brushes. The dry topography was characterized with AC tapping mode using silicon cantilevers (model: ACCESS-NC). Scanning electron microscope images were taken using a Zeiss Ultra 55 scanning electron microscope (SEM) with an accelerating voltage ranging from 0.5 to 1kV. No coating was done on any of the examined surfaces. Figure 4.1. Schematic illustration of the fabrication of nanopatterned PBLG brushes. Electron-beam lithography was used to pattern a resist film into nanohole arrays. APDIPES was vapor-deposited into the nanoholes. After the resist lift-off, PBLG brushes were grown from the patterned APDIPES via SI-VDP 138 4.4 Results and Discussion 4.4.1 Nanopatterned PBLG brushes EBL was used to pattern a ZEP-520 resist film (~150 nm) into nanohole arrays with a center-to-center distance of 200 nm. After a brief descumming process via reactive ion etching (RIE), vapor deposition of APDIPES was carried out using a similar procedure described by Fetterly et al. 37 APDIPES was selected due to its bulky size to minimize the diffusion of APDIPES through the resist film. The vapor-deposition conditions were optimized to prevent the background contamination of the silane in unexposed areas. After the resist lift-off, NPB of PBLG was grown from the patterned silane via SI-VDP (Figure 1 and Scheme 4.1a). Scheme 4.1. Synthesis procedures for a) PBLG brushes via SI-VDP; b) PS brushes via SI- CuCRP; c) PNIPAM brushes via ATRP After the solvent treatment with chloroform and acetone, the quenched NPB of PBLG was characterized with SEM and AFM. Dot patterns of PBLG brushes with the diameter as small as 65 nm were fabricated (Figure 4.2A). A clear contrast between the dots and 139 surrounding area implied that all the polymer chains stayed inside the boundaries of the patterns. Further characterization with AFM showed teepee-like structure with a height of ~74 nm, and there was one teepee per dot (Figure 4.3A). It implied that nanopatterned PBLG brushes can stand up on their own without any secondary support, probably due to their high persistence and strong intermolecular interaction. To examine the effect of polymer stiffness on the polymer organization, we also prepared PS brushes using the same pattern size of the silane. Specifically, APDIPES was reacted with BiBB to form ATRP initiators which were then used for the growth of PS brushes (Scheme 4.1b). It ensured no change in pattern size after the conjugation of BiBB. The SEM image of patterned PS brushes in Figure 4.2C showed that the edge of the patterned PS brushes became blurry. The dark regions were the original pattern of dots while the lighter region was from brush collapse. At the center of the dots, high steric repulsion forced polymer chains stretch away from the surface, resulting in higher thickness and darker color in SEM image. In contrast, at the edge of the dots, less chain crowding reduced chain stretching, resulting in lower thickness and brighter color. This observation was Figure 4.2. SEM micrographs of A) PBLG brushes with dot diameter of ~65 nm; B) PBLG brushes with dot diameter of ~110 nm and multi-teepees per dot; C) PS brushes with a diameter of ~65 nm for the inner circle and ~160 nm for the outer circle. All scale bars are 200 nm. 140 consistent with the literature. 8,24 The distinct difference between patterned PBLG and PS brushes emphasized the crucial effect of polymer stiffness on chain conformation. 4.4.2 Effect of pattern size on brush morphology As the pattern size increased, several interesting phenomena were observed. First, instead of one big teepee per dot, the PBLG brushes formed separate and smaller teepees in a dot (Figure 4.2B). Based on what we learnt from homogenous PBLG brushes, teepee size was proportional to polymer thickness (i.e. chain length). Therefore, when pattern size was relatively big compared to brush thickness, several teepees would form in a dot. To overcome this issue, we can simply increase the polymer brush thickness by tuning the polymerization conditions. Figure 4.2B also showed “bridges” between adjacent dots. As the center-to-center distance was kept at 200 nm, increasing the pattern size reduced the edge-to-edge distance. For this reason, PBLG brushes at the edge of a pattern started to interact with brushes of the adjacent patterns. Furthermore, “bridges” only formed between edge-adjacent dots, but not if the dots were diagonally adjacent. A simple explanation is that a diagonal is always ~1.41 times longer than an edge. Hence, to prevent the formation of “bridge”, we can increase the distance between patterns. However, for some applications such as molecular recognition, which requires small spacing between spikes, an increase of the distance between patterns is not feasible. Thus, we explore an alternative solution for this problem later in this chapter. Next, we investigated the scaling behavior of PBLG brushes by varying pattern size. AFM showed that with an increase of pattern size from 65 nm to 115 nm, the number of teepees per dot increased (Figure 4.3), and that was consistent with the SEM results. Moreover, the brush thickness also increased from 74 nm to 105 nm. The correlation between pattern size and brush thickness was similar to the scaling behavior of coil-type brushes as reported in the literature.23,24 The underlying cause is probably 141 from a lower grafting density or lower polymerization rate. Since the initiator deposition and the polymerization were all in vapor phase, bigger patterns can expose more to the initiator and monomer, resulting faster adsorption rates. Figure 4.3. AFM images (top) and the corresponding cross-sectional profile (bottom) of A) patterned PBLG brushes with dot diameter of ~65 nm; B) patterned PBLG brushes with dot diameter of ~115 nm. The cross-section areas are marked by the white lines. 142 4.4.3 Binary Nanopatterned Rod-Coil Brushes While patterned PBLG brushes exhibited interesting properties and potentials in biological applications, the addition of secondary brushes can further modify the surface with unique dual properties. For this purpose, SI-ATRP was conducted to grow the secondary brushes following the deposition of BIDS, an ATRP initiator of ethoxysilane, to void area between patterned PBLG brushes. PNIPAM was chosen in this case due to its stimuli-responsive properties.38 As PNIPAM established a change in solubility in water at its lower critical solution temperature (LCST) of 32℃,17 one can control the Figure 4.4. AFM images (top) and the corresponding cross-sectional profile (bottom) of A) patterned PBLG brushes; B) patterned binary brushes of PBLG and PNIPAM. The cross-section areas are marked by the white lines. 143 swelling of the PNIPAM film in solution to expose or envelop the teepee structures by controlling the surrounding temperature. PNIPAM brushes with a thickness of 25 nm as measured by ellipsometry were grown to form binary nanopatterned rod/coil brushes. AFM was used to characterize surface morphology of the binary brushes. While the teepee nanostructures of PBLG brushes retained after the polymerization of PNIPAM, bottom part of the teepees were covered under PNIPAM layer, causing the teepee to look smaller under AFM (Figure 4.4). Similarly, the brush thickness changed from 64 nm for patterned PBLG brushes to 32 nm for binary brushes, indicating the presence of PNIPAM brushes. In addition, as PNIPAM brushes existed between PBLG patterns, they can block the interactions of PBLG brushes from adjacent dots. As a result, there was a decrease in “bridges” for the binary brushes. The features of binary brushes had more rounded and distinct shape than that of patterned PBLG brushes, as shown in Figure 4.4. 4.5. Conclusion Herein, we report the preparation of nanopatterned helical PBLG brushes via SI-VDP following EBL patterned initiator formation. This “bottom-up” approach allowed a precise control of pattern size and spacings at the nanoscale. NPB with feature size as small as 65 nm were achieved, and the polymer brushes were assembled into teepee- like nanostructures within pattern boundaries after the solvent quenching with chloroform and acetone. In contrast, PS brushes grown from same pattern size were collapsed, resulting in broadening of pattern size. This distinct difference between PBLG and PS brushes indicated the strong dependence of polymer morphology on its stiffness. Moreover, a patterned feature of PBLG brushes may contain more than one teepee if the pattern size was relatively big. It implied that polymer chain length dictated the teepee size. When adjacent patterned features were closed enough, polymer brushes 144 from adjacent features can start to interact with each other. Moreover, polymer thickness became higher for bigger pattern size, and we attributed that scaling behavior to a greater chance of exposure to the initiators and monomers in vapor phase for bigger pattern. We also introduced PNIPAM brushes into unpatterned areas via SI-ATRP to produce binary patterned rod/coil brushes. PNIPAM brushes brought thermoresponsive property to the brush system while blocking the interaction of PBLG brushes from adjacent patterned features, thereby resulting more well-defined patterned features. In the future, we want to exploit the thermoresponsive property of PNIPAM brushes to either shrink or swell the brushes. Hence, they can expose or envelop teepees of PBLG brushes. We are also interested in functionalizing the outermost tip of the teepees so that this brush system can be used as a platform for cell-surface interaction, molecular recognitions, sensors, or chemical devices. 4.6 References 1. Chen, W.-L., Cordero, R., Tran, H. & Ober, C. K. 50th Anniversary Perspective: Polymer Brushes: Novel Surfaces for Future Materials. Macromolecules 50, 4089– 4113 (2017). 2. Zoppe, J. O. et al. Surface-Initiated Controlled Radical Polymerization: State-of- the-Art, Opportunities, and Challenges in Surface and Interface Engineering with Polymer Brushes. Chem. Rev. 117, 1105–1318 (2017). 3. Kumar, S., Dory, Y. L., Lepage, M. & Zhao, Y. Surface-Grafted Stimuli- Responsive Block Copolymer Brushes for the Thermo-, Photo- and pH-Sensitive Release of Dye Molecules. Macromolecules 44, 7385–7393 (2011). 145 4. Price, A. D., Hur, S.-M., Fredrickson, G. H., Frischknecht, A. L. & Huber, D. L. Exploring Lateral Microphase Separation in Mixed Polymer Brushes by Experiment and Self-Consistent Field Theory Simulations. Macromolecules 45, 510–524 (2012). 5. Welch, M. E. & Ober, C. K. Characterization of Polymer Brush Membranes via HF Etch Liftoff Technique. ACS Macro Lett. 2, 241–245 (2013). 6. Yu, Q., Ista, L. K., Gu, R., Zauscher, S. & López, G. P. Nanopatterned polymer brushes: conformation, fabrication and applications. Nanoscale 8, 680–700 (2015). 7. Welch, M. E. & Ober, C. K. Responsive and patterned polymer brushes. J. Polym. Sci. Part B Polym. Phys. 51, 1457–1472 (2013). 8. Rastogi, A., Paik, M. Y., Tanaka, M. & Ober, C. K. Direct Patterning of Intrinsically Electron Beam Sensitive Polymer Brushes. ACS Nano 4, 771–780 (2010). 9. Ahmad, S. A., Leggett, G. J., Hucknall, A. & Chilkoti, A. Micro- and Nanostructured Poly[oligo(ethylene glycol)methacrylate] Brushes Grown From Photopatterned Halogen Initiators by Atom Transfer Radical Polymerization. Biointerphases 6, 8–15 (2011). 10. Tsujii, Y. et al. Fabrication of patterned high-density polymer graft surfaces. II. Amplification of EB-patterned initiator monolayer by surface-initiated atom transfer radical polymerization. Polymer 43, 3837–3841 (2002). 11. Pallandre, A., Glinel, K., Jonas, A. M. & Nysten, B. Binary Nanopatterned Surfaces Prepared from Silane Monolayers. Nano Lett. 4, 365–371 (2004). 12. Dong, R., Krishnan, S., Baird, B. A., Lindau, M. & Ober, C. K. Patterned Biofunctional Poly(acrylic acid) Brushes on Silicon Surfaces. Biomacromolecules 8, 3082–3092 (2007). 146 13. Chen, W.-L. et al. Morphology of Nanostructured Polymer Brushes Dependent on Production and Treatment. Macromolecules 50, 4715–4724 (2017). 14. Zhou, Z., Yu, P., Geller, H. M. & Ober, C. K. Biomimetic Polymer Brushes Containing Tethered Acetylcholine Analogs for Protein and Hippocampal Neuronal Cell Patterning. Biomacromolecules 14, 529–537 (2013). 15. Chiang, E. N., Veatch, S. L., Holowka, D., Ober, C. & Baird, B. A. Investing Early Signaling Events in IgE-FcɛRI Activation Using SEM. Biophys. J. 98, 274a (2010). 16. Welch, M. E. et al. Generalized Platform for Antibody Detection using the Antibody Catalyzed Water Oxidation Pathway. J. Am. Chem. Soc. 136, 1879–1883 (2014). 17. Yu, Q., Johnson, L. M. & López, G. P. Nanopatterned Polymer Brushes for Triggered Detachment of Anchorage-Dependent Cells. Adv. Funct. Mater. 24, 3751– 3759 (2014). 18. Gevrek, T. N., Bilgic, T., Klok, H.-A. & Sanyal, A. Maleimide-Functionalized Thiol Reactive Copolymer Brushes: Fabrication and Post-Polymerization Modification. Macromolecules 47, 7842–7851 (2014). 19. Senaratne, W., Andruzzi, L. & Ober, C. K. Self-Assembled Monolayers and Polymer Brushes in Biotechnology:  Current Applications and Future Perspectives. Biomacromolecules 6, 2427–2448 (2005). 20. Yu, Q. et al. Nanopatterned polymer brushes as switchable bioactive interfaces. Nanoscale 5, 3632–3637 (2013). 147 21. Huck, W. T. S. Responsive polymers for nanoscale actuation. Mater. Today 11, 24–32 (2008). 22. Tian, Y., Su, B. & Jiang, L. Interfacial Material System Exhibiting Superwettability. Adv. Mater. 26, 6872–6897 (2014). 23. Lee, W.-K., Patra, M., Linse, P. & Zauscher, S. Scaling Behavior of Nanopatterned Polymer Brushes. Small 3, 63–66 (2007). 24. Ahn, S. J. et al. Surface-Initiated Polymerization on Nanopatterns Fabricated by Electron-Beam Lithography. Adv. Mater. 16, 2141–2145 (2004). 25. Chen, W.-L. et al. Reduced Lateral Confinement and Its Effect on Stability in Patterned Strong Polyelectrolyte Brushes. Langmuir 33, 3296–3303 (2017). 26. Wibowo, S. H., Sulistio, A., Wong, E. H. H., Blencowe, A. & Qiao, G. G. Polypeptide films via N-carboxyanhydride ring-opening polymerization (NCA-ROP): past, present and future. Chem. Commun. 50, 4971–4988 (2014). 27. Shen, Y., Li, Z. & Klok, H.-A. Polypeptide brushes grown via surface-initiated ring-opening polymerization of α-amino acid N-carboxyanhydrides. Chin. J. Polym. Sci. 33, 931–946 (2015). 28. Temyanko, E., Russo, P. S. & Ricks, H. Study of Rodlike Homopolypeptides by Gel Permeation Chromatography with Light Scattering Detection:  Validity of Universal Calibration and Stiffness Assessment. Macromolecules 34, 582–586 (2001). 29. Wang, Y. & Chang, Y. C. Synthesis and Conformational Transition of Surface- Tethered Polypeptide:  Poly(l-glutamic acid). Macromolecules 36, 6503–6510 (2003). 148 30. Kratzmüller, T., Appelhans, D. & Braun, H.-G. Ultrathin Microstructured Polypeptide Layers by Surface-initiated Polymerization on Microprinted Surfaces. Adv. Mater. 11, 555–558 (1999). 31. Wang, Y. & Chang, Y. C. Patterning of Polypeptide Thin Films by the Combination of Surface-Initiated Vapor-Deposition Polymerization and Photolithography. Adv. Mater. 15, 290–293 (2003). 32. Wang, Y. & Chang, Y. C. Preparation of Unidirectional End-Grafted α-Helical Polypeptides by Solvent Quenching. J. Am. Chem. Soc. 125, 6376–6377 (2003). 33. Haywood, A. M. Virus receptors: binding, adhesion strengthening, and changes in viral structure. J. Virol. 68, 1–5 (1994). 34. Daly, W. H. & Poché, D. The preparation of N-carboxyanhydrides of α-amino acids using bis(trichloromethyl)carbonate. Tetrahedron Lett. 29, 5859–5862 (1988). 35. Wang, T.-L., Ou, C.-C. & Yang, C.-H. Synthesis and properties of organic/inorganic hybrid nanoparticles prepared using atom transfer radical polymerization. J. Appl. Polym. Sci. 109, 3421–3430 (2008). 36. Zhang, T., Du, Y., Müller, F., Amin, I. & Jordan, R. Surface-initiated Cu(0) mediated controlled radical polymerization (SI-CuCRP) using a copper plate. Polym. Chem. 6, 2726–2733 (2015). 37. Fetterly, C. R., Olsen, B. C., Luber, E. J. & Buriak, J. M. Vapor-Phase Nanopatterning of Aminosilanes with Electron Beam Lithography: Understanding and Minimizing Background Functionalization. Langmuir 34, 4780–4792 (2018). 149 38. Choi, B.-C., Choi, S. & Leckband, D. E. Poly(N-isopropyl acrylamide) Brush Topography: Dependence on Grafting Conditions and Temperature. Langmuir 29, 5841–5850 (2013). 150 CHAPTER 5 Tuning the surface properties of mixed Mesogen-Jacketed Liquid-Crystalline polymer rod-coil brushes 5.1 Abstract Mixed brushes, driven by polymer-polymer immiscibility, can form microphase- separated structures. In prior studies, mixed brushes have been focused on coil-type polymers. Given a disparity in stiffness between rod and coil polymers, mixed rod-coil brushes may possess unique properties. In this work, a mesogen-jacketed liquid crystalline polymer (MJLCP) was explored as a rod-type brush because such polymers have high persistence length due to the crowding effect of large mesogenic side groups which jacket the polymer backbone. Poly(2,2,2-trifluoroethyl methacrylate) (PTFEMA) was chosen as coil-type brushes to enhance polymer immiscibility with MJLCP. Poly[2,5-(4 butylbenzoyl)oxystyrene] (PBBOS), a typical MJLCP, was synthesized via SI-NMP while PTFEMA brushes were grown via SI-ATRP from mixed ATRP/NMP initiators to produce mixed rod-coil brushes. When PBBOS brush thickness remained constant and PTFEMA brush thickness was increased, a transition in surface topography from teepee-like structures to nanoholes were observed. The teepee-like structures were constructed from PBBOS rod assemblies. With thin PTFEMA brushes, exposed portions of the rod assemblies above the coil film were seen as teepee-like structures under AFM while for thick PTFEMA brushes, the coil film was thicker than the rod assemblies, resulting in nanoholes. Moreover, increasing the relative number of PTFEMA brushes reduced the size of teepee structures, but also increased the number of brush teepees. Surface hydrophobicity also decreased accordingly. On the other hand, by keeping PTFEMA brush thickness the same while increasing PBBOS thickness, teepees became larger, but the number of teepees became smaller. As a result, the 151 surface roughness and hydrophobicity increased. Overall, surface properties can be tuned by controlling the relative thickness of the two polymers. 5.2 Introduction Polymer brushes are polymer chains with one end tethered to a substrate.1,2 As a result, polymer thin films with submicron thickness can dramatically change the surface properties of a substrate. In the last twenty years, the great advances in surface-initiated controlled radical polymerization (SI-CRP) such as surface-initiated atom transfer radical polymerization (SI-ATRP)3,4, nitroxide mediated polymerization (SI-NMP)5 or reversible addition-fragmentation chain transfer (SI-RAFT) polymerization6 have enabled the growth of polymer brushes with control of monomer chemistry, molecular weight (MW), dispersity, composition and architecture.2 In addition to homopolymer brushes, complex architectures have been reported such as mixed brushes, gradient brushes, block copolymer brushes, or patterned brushes.1,7 Particularly, mixed brushes are composed of two or more types of polymer chains which are randomly bound together to a common substrate. The mixed brushes can exhibit the combined properties of the two polymer components. 8,9 Mixed brushes have shown potential application in electrochemical gating,10 control of surface friction, 11 and control of protein adsorption.12 Mixed brushes may be prepared via “grafting-to” or “grafting-from” methods. With “grafting-to”, a mixture of two different pre-synthesized polymers with a functional chain-end were chemically linked to a surface to form mixed brushes. For example, carboxy-terminated poly(tert-butyl acrylate) and poly(2-vinyl pyridine) were linked to a surface from a polymer solution via a coupling reaction between the carboxy groups and epoxy groups on the substrate.12 Using the same approach, other mixed brushes have also been reported. 13,14 However, the “grafting-to” method resulted in a low grafting density and thickness due to reported steric hindrance.15 With a “grafting- 152 from” method, surface-immobilized initiators were used to grow polymer brushes from the surface with high grafting density. Lemieux et al. used surface-linked azo initiators to grow poly(methyl acrylate) (PMA), and poly(styrene-co-2,3,4,5,6- pentafluorostyrene) (PSF) brushes were subsequently grown from unreacted initiators.16 To have better control of polymer molecular size and dispersity, Zhao et al. used mixed ATRP and NMP initiators for the growth of mixed brushes.17 In particular, poly(methyl methacrylate) PMMA brushes were synthesized from ATRP initiator while polystyrene (PS) brushes were grown from NMP initiators. Variations in polymer MW and composition were reported using this method. 18,19 Alternatively, a binary Y-initiator which contains both ATRP and NMP sites can be used to produce uniformly dispersed mixed brushes.20,21 When the two polymers in mixed brushes are immiscible, they can undergo lateral microphase separation.19,20,22 Price et al. investigated the phase separation of PS and PMMA mixed brushes under non-selective solvents and observed ripple, cylinder and hemisphere structures as a function of polymer fraction.22 On the other hand, Feng et al. observed dimple-type structure for PS-PMMA mixed brushes after exposure to a selective solvent.23 Reversible microstructures were also reported for mixed brushes of PMA and PSF.16 Although there is significant progress in the study of mixed brushes in terms of synthesis or phase separation, all polymers which have been used to date are coil-type brushes with a persistence length of 1 nm or less.24,25 Recently, we reported the first study of helical polypeptide rod and coil mixed brushes.26 In that study, the presence of coil-type PMMA showed an effect on the orientation and organization of polypeptide rod-type brushes, resulting from a significant change in surface morphology. Thus, mixed rod-coil polymer brushes with mismatches in polymer stiffness may create unique microstructures that cannot be seen in mixed coil-coil 153 brushes. In this work, we attempted to use the mesogen-jacketed liquid crystalline rod- type polymer (MJLCP) for a mixed rod-coil brush system. The MJLCP was first introduced by Zhou in the late 1980s.27–29 MJLCP have large pendent mesogenic groups connected to polymer backbones via a very short linkage,30 thereby inducing a crowding effect to force the polymer backbone into an extended conformation.31 The resulting ordered structure results in an increase in persistence length for MJCLP with reported values between 11.5 to 13.5 nm32 when the MJCLP chains act as a supramolecular rod. Therefore, a MJLCP has a much higher stiffness than coil-type polymers. Moreover, the sizes of MJLCP can be easily tuned by varying the length of the mesogenic group and the polymer MW.33 A typical mesogenic side chain of MJLCP has three p-substituted ester-linked phenylene rings with the central ring connecting to the polymer backbone via a C-C spacer. 34 MJLCPs with the polymethacrylate backbone have also been reported. 31,35 CRP such as ATRP or NMP has been employed to synthesize MJLCPs. 34 Ober and coworkers reported the synthesis of poly[2,5-(4 butylbenzoyl)oxystyrene] (PBBOS) via NMP. 36 The polymerization of PBBOS resembles that of PS, but at a higher reaction rate due to a chain ordering effect. On the other hand, Zhang et al. also synthesized poly(2,5-bis[4- methoxyphenyl]oxycarbonyl)styrene) (PMPCS) via ATRP.37 Owing to their high persistence length, MJCLPs have been used as rod segments for block copolymers. Gopalan et al. reported the synthesis of rod-coil diblock, triblock copolymers of PBBOS and PS which induced nanostructured phase separation. 38 Moreover, Hsiao studied rod-coil diblock copolymers of PMPCS and PS, and observed lamellar structure and tetragonal perforated layer structures at high and low MW of PMPCS respectively.39,40 Similarly, Shi et al. showed different nanostructures such as lamellar (LAM) phases, double gyroid (GYR) structures, and hexagonally (HEX) packed cylinders for polydimethylsiloxane-b-PMPCS diblock copolymers by varying 154 MW of PMPCS. 41 While there is significant progress in the synthesis, processing, and characterization of MJLCP and their block copolymers, to the best of our knowledge, no study of MJLCP brushes has been ever reported. From what has been learned from rod-coil block copolymers of MJCLP, it is anticipated that MJCLP rod-coil mixed brushes may enable a rich array of phase separated structures. Moreover, our previous study of helical polypeptide rod and coil mixed brushes showed vertical alignment of rod brushes and their organization into teepee-like structures. 26 However, the study generated a number of unaddressed questions. First, the role of brush persistence length on the ability to vertically align rod brushes was not well understood. Second, the effect of rod size and the interaction between the rod and coil brushes on teepee size and spacing was also unclear. Therefore, a MJLCP rod brush with tunable size and a lower rigidity than polypeptide can give insights into those questions. Specifically, in this work, we used PBBOS for rod-coil mixed brushes since it can be synthesized via SI- NMP with chain ends that can be modified to a specific chemical function while coil- type polymer brushes can be grown using SI-ATRP. As mixed coil-coil brushes and rod-coil block copolymers have demonstrated the crucial role of polymer immiscibility in phase separation, we decided to use a fluorinated polymer for the coil brush due to the known highly immiscibility with other polymers.42 Moreover, fluorinated polymers are hydrophobic, and thermally stable.43,44 Poly(2,2,2-trifluoroethyl methacrylate) (PTFEMA) was chosen because its block copolymers were reported to have a long-range order nanostructured morphology. 43,45 PTFEMA is also an electron-beam and deep-UV sensitive material so that the fluorinated segments can be selectively removed with lithography to create patterned brushes for further studies.46,47 More importantly, SI-ATRP is well-suited for the highly controlled synthesis of PTFEMA brushes.48 In this work, we demonstrated for the first time the synthesis of PBBOS brushes via SI-NMP. PTFEMA brushes were added via 155 SI-ATRP to prepare mixed rod-coil brushes. To promote phase separation, a solvent treatment called “solvent quenching” was applied to the mixed brushes. Wang et al. used this method to change the organization of polypeptide brushes from randomly oriented to vertically aggregated teepee structures.49 The polypeptide brushes were swollen in a good solvent (i.e. chloroform), and then transferred into a non-solvent (i.e. acetone). A sudden change in solvent conditions forced the swollen polypeptide brushes to aggregate and minimize exposure to acetone. Accordingly, the tilt angle of the polypeptide brushes changed from 49° to 3° from the surface normal. Since the solvent quenching worked well for polypeptide brushes, we attempted to apply it for the PBBOS-PTFEMA mixed brushes. Moreover, we conducted a systematic study to investigate the effect of phase separation in terms of polymer thickness on surface topography and rod brush organization. Either rod or coil brush thickness was kept constant while varying the thickness of the other polymer. The mixed brush thickness, surface topography, roughness, and hydrophobicity were characterized by ellipsometry, AFM, and water contact angle measurement to explore the effect of polymer thickness on surface properties. Overall, the results showed no vertical alignment of PBBOS homopolymer brushes while the PBBOS rod- PTFEMA coil system exhibited teepee- like structures similar to that of the reported polypeptide rods. The results indicated the role of the coil brush on providing a pressure through phase incompatibility and confinement to promote vertical alignment of the rod brush. Moreover, when phase separation was enhanced by increasing the volume fraction of PTFEMA coil, the number of teepees composed of PBBOS rods increased while the size and spacing of teepees decreased. This observation was similar to that of the helical rod-coil mixed brushes. On the other hand, by increasing PBBOS length, the size and spacing of teepees increased while the teepee density decreased. 156 5.3 Experimental Section 5.3.1. Materials Allylamine, anhydrous toluene, α-bromoisobutyryl bromide (BIBB), chlorodimethylsi- lane, Pt on activated carbon (10 wt %), 4,4′-dinonyl-2,2′-dipyridyl (dNnbpy), α,α,α- trifluorotoluene, anhydrous tetrahydrofuran (THF), anhydrous pyridine, 4-n- butylbenzoylchloride, benzoyl peroxide (BPO, >97%), styrene (St, 99%), 2,2,6,6- tetramethylpiperidinooxy (TEMPO, 98%), triethylamine (TEA), sodium hydride (60% dispersion in oil, allyl bromide, ethylbenzene, di-tert-butyl peroxide, copper (I) bromide, inhibitor remover (for removing hydroquinone and monomethyl ether hydroquinone) were purchased from Sigma Aldrich. 2,5-dihydroxybenzaldehyde, potassium tert-butoxide, methyl triphenylphosphonium bromide were purchased from AK Scientific Inc. The monomer 2,2,2-trifluoroethyl methacrylate [TFEMA; stabilized with methylhydroquinone (MEHQ)] was purchased from TCI America. TFEMA was passed through an inhibitor remover to remove the inhibitor before use. Deionized water with a resistivity of 18.2 MΩ•cm at 25 °C was obtained from Millipore’s Milli-Q Synthesis A10 system. All other solvents were purchased from Fisher Scientific. Si(100) double-sided polished wafers were purchased from WRS Materials. Cantilevers were purchased from Applied NanoStructures, Inc. (ACCESS-NC) 5.3.2 Synthesis of 2,5-bis[(4-butylbenzoyl)oxy]styrene (BBOS) 2,5-dihydroxystyrene. To a solution of methyltriphenylphosphonium bromide (13.2 g, 37 mmol) and potassium tert-butoxide (6.72 g, 60 mmol) in THF (80 mL), a solution of 2,5-dihydroxybenzaldehyde (2 g, 14 mmol) in THF (20 mL) was added dropwise, and the reaction mixture was stirred at room temperature for 48 hours. After that, saturated ammonium chloride solution (80 mL) was added to quench the reaction. The resulting solution was concentrated via rotary evaporation and extracted with ethyl acetate (5 times) and dried over magnesium sulfate. The organic extract was evaporated to dryness 157 and the crude product was purified by flash chromatography (ethyl acetate/ hexane = 1:10) to get 2,5-dihydroxystyrene as a slightly yellow solid (1.18 g, 60%) 1H-NMR (400 MHz, DMSO-d6): δ = 5.00-5.73 ppm, 2H for =CH2; 6.50-6.53 ppm, 1 H for -CH=; 6.53- 7.00 ppm, 4H for benzene rings; and 8.68 (s), 8.88 (s) for -OH BBOS. To a solution of 2,5-dihydroxystyrene (1 g, 7 mmol) and TEA (3.35 g, 33 mmol) in anhydrous THF, 4-n-butylbenzoylchloride (4.34 g, 22 mmol) was added dropwise, and the reaction mixture was stirred in an ice bath for 24 hours. Then water (5 mL) was added to quench the reaction. The solution was concentrated and extracted with dichloromethane (3 times), water ( 2 times) and saturated sodium bicarbonate (2 times). The combined organic fractions were dried over magnesium sulfate and evaporated to dryness. The crude product was purified by flash chromatography (ethyl acetate/hexane = 1:25) to get BBOS as a white solid (2.5 g, 75%) 1H NMR (400 MHz, CDCl3): δ = 0.95 ppm, 6H for CH3 ; 1.35 ppm, 4H for -CH2CH2CH2; 1.65 ppm, 4H for -CH2CH2; 2.70 ppm, 4H for -CH2; 5.10–5.90 ppm, 2H for =CH2 ; 6.60–6.95 ppm, 1H for -CH= ; and 7.00–8.30 ppm, 11H for benzene rings 5.3.3 Initiator immobilization ATRP silane initiator (2-bromo-2-methyl-N-{3-[chloro(dimethyl)silyl]- propyl}propenamide) was synthesized following a published procedure. 50 1-(2- (allyloxy)-1-phenylethoxy)-2,2,6,6-tetramethylpiperidine was prepared following a previous procedure.51 Then it underwent hydrosilylation with chlorodimethylsilane using Pt/C (10% Pt) as a catalyst to give NMP silane initiator, (1-(2-(3- (chlorodimethylsilyl)propoxy)-1-phenylethoxy)-2,2,6,6-tetramethylpiperidine). Wafer pieces were cleaned by sonication in acetone, followed by isopropanol for 10 minutes each and dried with nitrogen. The moisture was further removed by baking in an oven 158 at 110°C for 10 min. Then, the wafers were oxidized using a Harrick Plasma Cleaner for 10 min. In a glovebox, the wafer pieces were immersed overnight in a toluene solution of the initiators (15 mM ATRP silane initiator and 15 mL NMP silane initiator) and pyridine (0.5 mM) for 16 hours at room temperature. The substrates were then removed from the solution and cleaned with toluene then methanol and dried under nitrogen gas. 5.3.4 Synthesis of PBBOS brushes The free NMP initiator (1-Phenyl-1-(2,2,6,6-tetramethyl-1-piperidinyloxy)ethane) was synthesized using previous procedures.52–54 Initiator-immobilized silicon substrates were placed into a 20-mL glass vial and ~ 500 mg of a mixture of BBOS monomer and the free NMP initiator at 1000:1 molar ratio was added to the vial to cover the substrates. The vial was put into a custom-made glass vessel which was subsequently evacuated and backfilled with argon (three times) and kept under argon atmosphere. After that, the vessel was heated to 115oC in an oil bath to start the polymerization for various amounts of time. After the reaction, the substrates were rinsed and sonicated with DCM and THF to remove any physiosorbed free polymer and dried under nitrogen gas. Synthesis of PTFEMA brushes Wafer samples with tethered PBBOS brushes and immobilized-ATRP initiators were placed into a Schlenk flask which was then evacuated and back-filled with argon for 5 times. In another flask, the monomer 2,2,2-trifluoroethyl methacrylate (94 mmol), and the solvent α,α,α-trifluorotoluene (2 mL) were purged with argon for at least 1 hour. In another Schlenk flask equipped with a magnetic stir bar, CuBr (173 mg, 1.74 mmol), dNnbpy (544 mg, 3.48 mmol) were added. The flask was evacuated and backfilled with argon for 5 times. The monomer solution was then cannulated into the ligand and copper 159 salt. The mixture was stirred for 10 minutes before cannulated into the flask containing the wafer substrates. The polymerization was kept at 80oC for different reaction time. After the polymerization, the samples were washed and sonicated in THF for 5 minutes and dried under nitrogen gas. 5.3.5. Solvent treatment The polymer brush samples were immersed into chloroform for 30 minutes. Then they were quickly transferred into hexane for 20 minutes. After that, they were blown dry under nitrogen gas. 5.3.6. Characterization NMR spectrum were measured using a Varian INOVA-400 NMR spectrometer using CDCl3 as the solvent, unless stated otherwise. Thicknesses of initiator film, PBBOS and PTFEMA homopolymer brushes were measured with an Imaging Ellipsometer - Nanofilm EP3 at 532 nm laser at 50-60° angle of incidence. A Cauchy model (Cauchy layer/silicon substrate) was used to fit the data. Fourier Transform Infrared (FT-IR) measurements were performed in transmission mode with a Bruker Optics – Vertex 80v. Spectra were recorded at 4 cm-1 resolution, and 256 scans were taken for each measurement with bare silicon wafer as reference. Surface topography, thickness and roughness of mixed polymer brushes were measured using an Asylum MFP-3D atomic force microscopy (AFM). The thickness of the polymer brushes was determined by measuring the height difference between a brush area and a scratched (polymer removed) area. The dry topography was characterized with AC tapping mode using silicon cantilevers (model: ACCESS-NC) in air at room temperature. The surface composition of the mixed initiators and the polymer brushes was determined by X-ray 160 Photoelectron Spectroscopy (XPS) using a Surface Science Instruments SSX-100 with operating pressure of ~2x10-9 Torr. Monochromatic Al Kα x rays (1486.6 eV) with 1 mm diameter beam size was used. Photoelectrons were collected at a 55° emission angle. A hemispherical analyzer determined electron kinetic energy, using a pass energy of 150 V acquired at 1 eV/ step for wide/survey scans. VCA Optima Contact Angle was used to measure the static water contact angle on a modified sample by dispending 2.0 μL droplets of deionized water. Scheme 5.1. a) Synthesis of BBOS monomer, and b) synthesis of mixed PBBOS- PTFEMA rod-coil brushes via SI-NMP and SI-ATRP, respectively. 161 5.4 Results and Discussion The quantity of monomer needed for brush synthesis is usually much greater than that of free polymer synthesis because a substrate must be fully immersed into the reaction mixture. However, prior synthesis of BBOS monomer is five-step reaction with low yield.55 Therefore, it is not feasible to go through that route for large scale preparation of BBOS. Instead, we designed a simplified multi-gram scale monomer synthesis with significantly higher yield (Scheme 5.1). Specifically, 2,5- dihydroxybenzaldehyde was converted via Wittig reaction to 2-vinyl-1,4- dihydroxybenzene which was subsequently reacted with 4-butylbenzoyl chloride to give BBOS at in a total yield of 45%. The reaction setup and purification are simple enough to produce large amounts of BBOS in a short time. Since the controlled synthesis of free PBBOS was reported by Pragliola et al.36, we adapted similar polymerization conditions for the brush growth from immobilized-NMP initiator. The presence of PBBOS brushes on a silicon substrate was verified by ellipsometry, FT-IR and XPS. For ~50 nm brush thickness as measured by ellipsometry, the measured elemental composition was 85.9% carbon and 14.1% oxygen (Figure 5.1a), similar to its theoretical composition (88.2% carbon and 11.8% oxygen). The peak for Si2p (99 eV) was not observed in the survey spectra due to a thick polymer film on top of the silicon substrate. The FT-IR spectra of the PBBOS brushes provided a characteristic stretching peak of C=O group from the mesogenic side chains at 1735 cm-1 (Figure 5.2). In addition, the peaks at 1490 cm-1 and 1610 cm-1 corresponded to the stretching vibration of the carbon-carbon bonds in the benzene rings of mesogenic groups. For the preparation of mixed polymer brushes, a similar approach to the work of Zhao17 using mixed ATRP and NMP initiators was utilized. The initiators with monochlorosilane anchoring group were used to ensure a monolayer of uniformly 162 Figure 5.1. XPS survey scan spectra of (a) PBBOS homopolymer brushes; (b) mixed ATRP and NMP initiators; (c) PTFEMA homopolymer brushes; (d) mixed PBBOS- PTFEMA brushes. distributed initiators. Using mixed initiators allowed us to control the composition of the two initiators which could consequently influence the final composition of the rod and coil polymers. However, in this study, the initiator composition was kept constant so that we can clearly see the effect of other parameters such as polymer molecular weight on phase separation. Therefore, a solution of 15 mM ATRP initiator and 15 mM NMP initiator was used to prepare all the samples in this study. The initiator thickness as measured by ellipsometry was 0.9 ± 0.1 nm, indicating the formation of a monolayer 163 film. XPS also showed the peaks for Br3d (69 eV), N1s (399 eV), C1s (284 eV) and Si2p (99 eV) in the surveyed spectrum (Figure 5.1b), confirming the presence of the initiators. Moreover, ATRP initiator has 1 Bromine atom and 1 Nitrogen atom while NMP initiator contains 1 Nitrogen atom. Thus, the ratio of Br to N can be used to determine the relative composition of ATRP and NMP initiators on the substrate. From XPS data, the ratio of NMP initiator to ATRP initiator was 0.70:1. The discrepancy between the ratios of the two initiators in the solution and on the surface was probably from differences in molecular size or adsorption rate of the initiators. Although SI-ATRP and SI-NMP is known to be orthogonal18, the order of polymerization was carefully considered due to the significant difference in monomer size between TFEMA and BBOS. In fact, the size of TFEMA and BBOS were 0.58 nm and 1.9 nm respectively, as calculated using MacSpartan Plus program. Moreover, the polymerization of PBBOS was in bulk while that of PTFEMA was in solution. Figure 5.2. a) FT-IR spectra of PBBOS brushes (blue), and mixed brushes (black) 164 Therefore, it would be extremely hard for BBOS to diffuse through PTFEMA brushes before reaching NMP reactive sites. Our preliminary results showed failed attempts to grow PBBOS brushes in the presence of PTFEMA brushes. For that reason, SI-NMP of PBBOS was carried out before SI-ATRP of PTFEMA. Additionally, SI-ATRP of PTFEMA was run at 80 oC which is low enough to not activate NMP reaction sites.20 XPS showed the new peak for F1s (686 eV) which came from PTFEMA brushes. The measured elemental composition of mixed brushes was 63.4% carbon, 15.6% oxygen and 21.0% fluorine (Figure 5.1d). The percentage of fluorine in mixed brushes was significantly lower than that of PTFEMA homopolymer brushes (33.9%) (Figure 5.1c) due to the presence of PBBOS brushes. In addition, FT-IR measurements on mixed brushes showed the peaks at 1284 cm-1 and 657 cm-1 which can be assigned to stretching and bending vibrations of the C-F bond in the CF3 group of PTFEMA. 56 Moreover, the characteristic peak of the carbonyl group was shifted from 1735 cm-1 to 1742 cm-1 with higher intensity due to the addition of C=O groups at 1751 cm-1 from PTFEMA brushes (Figure 5.2). AFM revealed a lamellar structure of mixed brushes after solvent-vapor annealing in THF. These results implied that the two polymers were highly immiscible, thereby inducing phase separation. We attempted to promote vertical alignment of PBBOS brushes by using solvent quenching. Specifically, a sample of mixed brushes was solvated in chloroform, a good solvent of PBBOS to extend the rod segments. After that, the sample was transferred into hexane, a non-solvent of PBBOS. By doing that, we expected hexane to extract chloroform from the swollen polymers, thereby aggregating PBBOS chains together in stretched chain bundles. Interesting surface topography was observed after solvent quenching and we would further explore it later in this work. 165 It is known from block copolymer studies that phase separation is largely governed by χN where χ is the Flory-Huggins interaction parameter and N is the degree of polymerization.57 Thus, for mixed rod-coil brushes, we investigated how the thickness of each polymer (i.e. polymer molecular weight) could dictate the formation of microphase-separated structures. In the first system, the thickness of PBBOS brushes was held constant while varying PTFEMA brush thickness through a control of polymerization time. Specifically, a PBBOS brush sample with 72 nm thickness was Figure 5.3. (a) Brush thickness; (b) Peak to valley height; (c) number of rod assemblies; and (d) water contact angle for sample A to D which have the same PBBOS brush thickness but increased PTFEMA thickness 166 cut into four smaller samples from which PTFEMA brushes with different thicknesses ranging from 0, 26, 105 to 161 nm for sample A to D respectively were synthesized (Figure 5.3). It ensured that the four samples had the same grafting density and thickness of PBBOS brushes so that the main difference among them was PTFEMA brush thickness. The thickness of PTFEMA in mixed brushed was determined from PTFEMA homopolymer brushes grown from mixed initiators using the same polymerization conditions. We understand that the presence of PBBOS brushes may affect the polymerization kinetics of PTFEMA formation, resulting in a small discrepancy in PTFEMA thickness between homopolymer and mixed brushes. However, the general trend of the effect of PTFEMA brush thickness on phase separation and surface morphology of mixed brushes if present still holds true. After solvent quenching with chloroform and hexane, the thicknesses of mixed brushes which were measured from a scratched (polymer removed) area using AFM ranged from 72 nm to 185 nm for sample A to D respectively (Figure 5.3a). For the samples B to D, the thicknesses of mixed brushes were higher than those of PBBOS or PTFEMA brushes, suggesting a significant change in polymer conformation or organization of mixed brushes compared to homopolymer brushes. For this reason, surface morphology was characterized via AFM to gain insight into polymer organization. Interestingly, a distinct transition was observed for samples A to D (Figure 5.4). While sample A with no PTFEMA (i.e. PBBOS homopolymer brush) looked very flat, sample B with ~ 26 nm PTFEMA brushes had teepee-like nanostructures which were nearly identical to that shown for polypeptide brushes as reported elsewhere.49 The teepee height, measured as peak-to-valley height, was ~ 140 nm for sample B (Figure 5.4B) When PTFEMA thickness was increased to 105 nm in sample C, the teepees looked much shorter with a height of ~ 37 nm, though the mixed brush thickness (158 nm) was almost the same to that of sample B (164 nm). These 167 Figure 5.4. 3D AFM projections of mixed brushes with 72 nm thickness of PBBOS and A) 0 nm; B) 26 nm; C) 105 nm; D) 161 nm of PTFEMA brushes. The samples were characterized after solvent quenching. The scan size is 1µm x 1µm. observations led us to hypothesize that teepee structures were formed from the assemblies of PBBOS chains driven by the solvent treatment. Thus, the heights of rod assemblies were approximately closed to the thickness of mixed brushes. However, films of PTFEMA brushes partly enveloped the bottom of the rod assemblies so that only the exposed parts above coil film were characterized using AFM (Scheme 5.2). Consequently, increasing PTFEMA brush thickness resulted in smaller teepees. When PTFEMA thickness increased to ~ 161 nm in sample D, the formation of nanoholes with a depth of ~ 50 nm occurred instead of teepees, and the mixed brush thickness increased to ~ 185 nm. The drastic change in surface topography happened when PTFEMA 168 brushes became comparable or thicker than the height of the rod assemblies. The topmost surface was PTFEMA while the rod assemblies were below it, causing nanoholes. The high immiscibility between the two polymers prevented PTFEMA from completely covering the rod assemblies of PBBOS. The mixed brush thickness (~ 185 nm) was similar to the true thickness of PTFEMA brush in mixed brushes, and it was slightly higher than that of PTFEMA homopolymer brushes (~ 161 nm). It implied that bulky PBBOS caused PTFEMA chains to stretch more in mixed brushes than that in homopolymer brushes. So far, the results suggested that the role of the coil brushes provided a pressure through phase incompatibility and confinement to vertically align the rod brushes into teepee-like assemblies. While a thin film of PTFEMA (~ 26 nm) was enough to induce the nanostructured, an increase in PTFEMA thickness may further enhance the phase separation. In fact, the number of teepees increased from ~ 70 for sample B to ~ 100 for sample C (Figure 5.3c). In addition, for the sample D, the number of nanoholes was ~ 121, confirming the correlation between PTFEMA thickness and the number of nanostructures. It can be explained that PTFEMA brushes promoted the local confinement of PBBOS, thereby increasing the number of rod assemblies. The transition in surface topography with PTFEMA thickness caused a change in surface roughness which in turn affected the surface hydrophobicity (Figure 5.3d). The lowest static water contact angle (WCA) was 93° for sample A (roughness 2.1 nm) while the highest WCA was 139° for sample B (roughness 35.7 nm). Such nanostructure-induced hydrophobicity was similar to the phenomena observed in lotus leaves 58, or gecko skin. 59 After gaining insight into the effect of PTFEMA thickness on surface topography, we proceeded to investigate how PBBOS brush thickness may influence the phase-separated nanostructures. Samples of different PBBOS thicknesses ranging 169 from 0 nm in sample E, 37 nm in sample F, 52 nm in sample G, and 72 nm in sample B (Figure 5.5) were used to grow 26 nm PTFEMA brushes. A significant increase in mixed brush thickness compared to that of homopolymer brushes was observed. Moreover, thickness of mixed brushes became larger with increasing PBBOS thickness (Figure 5.5a). Scheme 5.2. The schematic illustration of rod-coil mixed brushes AFM revealed that all the mixed brush samples (F,G and B) displayed teepee structures while PTFEMA homopolymer brush looked featureless (sample E) (Figure 5.6). Surface roughness changed from 0.74 nm for sample E to 33.5 nm for sample B. The increasing trend in surface roughness was the result of the formation of teepee structures in mixed brushes (samples F, G and B) (Figures 5.4 and 5.6). Moreover, size of teepees became bigger, as indicated by peak-to-valley height, with an increase of PBBOS thickness (Figure 5.5). The results supported the hypothesis above that the 170 Figure 5.5. (a) Brush thickness; (b) Peak to valley height; (c) number of rod assemblies; and (d) water contact angle for sample E, F, G and B which have the same PTFEMA brush thickness but increased PBBOS thickness teepees were constructed from rod assemblies, and the height difference between mixed brush thickness and the peak-to-valley height came from the presence of PTFEMA film. Although thicker PBBOS brush resulted in bigger teepee-like nanostructures, the number of teepees was reduced. The number of teepees changed from ~108, 94 to 70 for samples F, G and B, respectively (Figure 5.5c). It can be explained that thicker PBBOS brushes (i.e. longer polymer chains) can interact and aggregate with other chains from farther away, thereby forming bigger teepee-like rod assemblies. Since a larger teepee contained more rods, the number of teepees became fewer. Additionally, surface hydrophobicity increased with surface roughness and teepee size. WCA increased from ~96° for sample E to ~139° for sample B (Figure 5.5d). 171 Figure 5.6. AFM 3D images of mixed brushes with 26 nm PTFEMA brush and E) 0 nm; F) 37 nm; G) 52 nm of PBBOS brushes. The scan size is 1µm x 1µm. 5.5 Conclusion In this study, we developed a new, more efficient synthesis of BBOS monomer to improve the yield and scalability. Then we successfully polymerized PBBOS brushes via SI-NMP. Although PBBOS brushes showed a featureless surface, by adding PTFEMA brushes grown via SI-ATRP into the system to produce rod-coil mixed brushes which were then treated with a solvent quench step, teepee-like nanostructures of rod assemblies were observed. This observation indicated the critical role of coil 172 brushes through phase separation in the assembly of rod brushes. Furthermore, a systematic investigation was carried out to examine the effect of brush thickness on surface topography. With the same PBBOS brush thickness and an increase in PTFEMA brush thickness, surface topography changed from teepee structures to nanoholes. The explanation came from the relative height of the rod assemblies and coil film. When the rod assemblies were taller than the coil film, exposed parts above the coil film of the rod assemblies were seen as teepees under AFM. In contrast, when the rod assemblies were shorter and below the coil film, it resulted in nanoholes. Accordingly, surface roughness and hydrophobicity decreased with an increase of coil brush thickness. In addition, thicker coil brushes enhanced phase separation and confined the aggregation of rod brushes, thereby causing a decrease in teepee size but an increase in the number of teepees (or holes). On the other hand, when coil brush thickness remained constant and rod brush thickness increased, teepees became bigger, but the number of teepees was reduced. This observation can be explained by longer rod brushes interacting and forming assemblies with other rod chains from a greater distance. As a result, the surface roughness and hydrophobicity increased with thicker rod brushes. Overall, we demonstrated that the relative thickness of the rod and coil brushes can greatly impact the surface topography and other properties. Moreover, the nanostructures qualitatively resemble a surface of lotus leaves or gecko skins, with the mixed brushes having potential for low adhesion, self-cleaning, antibacterial surfaces.60 However, since this is the first reported MJLCP rod-coil brushes, many questions still remain to be addressed. The organization, conformation, or orientation of rod brushes in the assemblies needs to be verified using methods such as grazing-incidence wide-angle X-ray scattering (GI- WAXS). Moreover, size and spacing of the nanostructures were not uniform across the sample, and this issue may be due to uneven distribution of the initiators and the 173 polymers on surface. Thus, a Y-shaped initiator might be used to have better control of nanostructure. Given the effect of coil brush thickness, gradient coil brushes can also be used to produce a continuous transition of surface properties along a sample. Additionally, since the size of a rod polymer can be adjusted by changing the chemical structure of the mesogenic side groups, there will be an interest in exploring the effect of rod size on surface topography. With the large library of MJLCPs, mixed brushes of other MJLCPs are worth investigating in the future. 5.6 References 1. Chen, W.-L., Cordero, R., Tran, H. & Ober, C. K. 50th Anniversary Perspective : Polymer Brushes: Novel Surfaces for Future Materials. Macromolecules 50, 4089–4113 (2017). 2. Zoppe, J. O. et al. Surface-Initiated Controlled Radical Polymerization: State-of- the-Art, Opportunities, and Challenges in Surface and Interface Engineering with Polymer Brushes. Chem. Rev. 117, 1105–1318 (2017). 3. Barbey, R. et al. Polymer Brushes via Surface-Initiated Controlled Radical Polymerization: Synthesis, Characterization, Properties, and Applications. Chem. Rev. 109, 5437–5527 (2009). 4. Kang, C., Crockett, R. M. & Spencer, N. D. Molecular-Weight Determination of Polymer Brushes Generated by SI-ATRP on Flat Surfaces. Macromolecules 47, 269–275 (2014). 5. Brinks, M. K. & Studer, A. Polymer Brushes by Nitroxide-Mediated Polymerization. Macromol. Rapid Commun. 30, 1043–1057 (2009). 174 6. Boyer, C., Stenzel, M. H. & Davis, T. P. Building nanostructures using RAFT polymerization. J. Polym. Sci. Part Polym. Chem. 49, 551–595 (2011). 7. Welch, M. E. & Ober, C. K. Responsive and patterned polymer brushes. J. Polym. Sci. Part B Polym. Phys. 51, 1457–1472 (2013). 8. Drechsler, A., Elmahdy, M. M., Uhlmann, P. & Stamm, M. pH and Salt Response of Mixed Brushes Made of Oppositely Charged Polyelectrolytes Studied by in Situ AFM Force Measurements and Imaging. Langmuir 34, 4739–4749 (2018). 9. Estillore, N. C. & Advincula, R. C. Stimuli-Responsive Binary Mixed Polymer Brushes and Free-Standing Films by LbL-SIP. Langmuir 27, 5997–6008 (2011). 10. Motornov, M., Sheparovych, R., Katz, E. & Minko, S. Chemical Gating with Nanostructured Responsive Polymer Brushes: Mixed Brush versus Homopolymer Brush. ACS Nano 2, 41–52 (2008). 11. Kumar Vyas, M., Schneider, K., Nandan, B. & Stamm, M. Switching of friction by binary polymer brushes. Soft Matter 4, 1024 (2008). 12. Uhlmann, P. et al. In-Situ Investigation of the Adsorption of Globular Model Proteins on Stimuli-Responsive Binary Polyelectrolyte Brushes. Langmuir 23, 57– 64 (2007). 13. Ionov, L. et al. Gradient Mixed Brushes: “Grafting To” Approach. Macromolecules 37, 7421–7423 (2004). 14. Minko, S. et al. Synthesis of Adaptive Polymer Brushes via “Grafting To” Approach from Melt. Langmuir 18, 289–296 (2002). 15. Polymer brushes: synthesis, characterization, applications. (Wiley-VCH, 2004). 175 16. Lemieux, M. et al. Reorganization of Binary Polymer Brushes: Reversible Switching of Surface Microstructures and Nanomechanical Properties. Macromolecules 36, 7244–7255 (2003). 17. Zhao, B. Synthesis of binary mixed homopolymer brushes by combining atom transfer radical polymerization and nitroxide-mediated radical polymerization. Polymer 44, 4079–4083 (2003). 18. Zhao, B. A Combinatorial Approach to Study Solvent-Induced Self-Assembly of Mixed Poly(methyl methacrylate)/Polystyrene Brushes on Planar Silica Substrates: Effect of Relative Grafting Density. Langmuir 20, 11748–11755 (2004). 19. Zhao, B., Haasch, R. T. & MacLaren, S. Solvent-Induced Self-Assembly of Mixed Poly(methyl methacrylate)/Polystyrene Brushes on Planar Silica Substrates:  Molecular Weight Effect. J. Am. Chem. Soc. 126, 6124–6134 (2004). 20. Zhao, B. & He, T. Synthesis of Well-Defined Mixed Poly(methyl methacrylate)/Polystyrene Brushes from an Asymmetric Difunctional Initiator- Terminated Self-Assembled Monolayer. Macromolecules 36, 8599–8602 (2003). 21. Calabrese, D. R. et al. Design, Synthesis, and Use of Y-Shaped ATRP/NMP Surface Tethered Initiator. ACS Macro Lett. 4, 606–610 (2015). 22. Price, A. D., Hur, S.-M., Fredrickson, G. H., Frischknecht, A. L. & Huber, D. L. Exploring Lateral Microphase Separation in Mixed Polymer Brushes by Experiment and Self-Consistent Field Theory Simulations. Macromolecules 45, 510–524 (2012). 176 23. Feng, J., Haasch, R. T. & Dyer, D. J. Photoinitiated Synthesis of Mixed Polymer Brushes of Polystyrene and Poly(methyl methacrylate). Macromolecules 37, 9525–9537 (2004). 24. Wignall, G. D., Ballard, D. G. H. & Schelten, J. Measurements of persistence length and temperature dependence of the radius of gyration in bulk atactic polystyrene. Eur. Polym. J. 10, 861–865 (1974). 25. Rubinstein, M. & Colby, R. H. Polymer Physics. (OUP Oxford, 2003). 26. Tran, H., Zhang, Y. & Ober, C. K. Synthesis, Processing, and Characterization of Helical Polypeptide Rod–Coil Mixed Brushes. ACS Macro Lett. 7, 1186–1191 (2018). 27. Zhou, Q., Zhu, X. & Wen, Z. Liquid-crystalline side-chain polymers without flexible spacer. Macromolecules 22, 491–493 (1989). 28. Zhang, D., Liu, Y.-X., Wan, X.-H. & Zhou, Q.-F. Synthesis and Characterization of A New Series of “Mesogen-Jacketed Liquid Crystal Polymers” Based on the Newly Synthesized Vinylterephthalic Acid. Macromolecules 32, 5183–5185 (1999). 29. Chai, C.-P. et al. Synthesis and Phase Structures of Mesogen-Jacketed Liquid Crystalline Polymers Containing 1,3,4-Oxadiazole Based Side Chains. Macromolecules 40, 9361–9370 (2007). 30. Chen, X.-F. et al. Mesogen-jacketed liquid crystalline polymers. Chem. Soc. Rev. 39, 3072 (2010). 31. Zhou, Q. F., Li, H. M. & Feng, X. D. Synthesis of liquid-crystalline polyacrylates with laterally substituted mesogens. Macromolecules 20, 233–234 (1987). 177 32. Wan, X. et al. Characterization of the chain stiffness for a mesogen-jacketed liquid crystal polymer: Poly2,5-BIS [(4-Methoxybenzoyl)Oxy] Styrene. Macromol. Symp. 96, 207–218 (1995). 33. Lyu, X.-L., Pan, H.-B., Shen, Z.-H. & Fan, X.-H. Self-assembly and Properties of Block Copolymers Containing Mesogen-Jacketed Liquid Crystalline Polymers as Rod Blocks. Chin. J. Polym. Sci. 36, 811–821 (2018). 34. Gao, L.-C., Fan, X.-H., Shen, Z.-H., Chen, X. & Zhou, Q.-F. Jacketed polymers: Controlled synthesis of mesogen-jacketed polymers and block copolymers. J. Polym. Sci. Part Polym. Chem. 47, 319–330 (2009). 35. Zhou, Q.-F., Li, H.-M. & Feng, X.-D. Synthesis of a New Class of Side Chain Liquid Crystal Polymers—Polymers with Mesogens Laterally Attached Via Short Linkages to Polymer Backbones. Mol. Cryst. Liq. Cryst. Inc. Nonlinear Opt. 155, 73–82 (1988). 36. Pragliola, S., Ober, C. K., Mather, P. T. & Jeon, H. G. Mesogen-jacketed liquid crystalline polymers via stable free radical polymerization. Macromol. Chem. Phys. 200, 2338–2344 (1999). 37. Zhang, H., Yu, Z., Wan, X., Zhou, Q.-F. & Woo, E. M. Effects of molecular weight on liquid-crystalline behavior of a mesogen-jacketed liquid crystal polymer synthesized by atom transfer radical polymerization. Polymer 43, 2357–2361 (2002). 38. Gopalan, P., Zhang, Y., Li, X., Wiesner, U. & Ober, C. K. Liquid Crystalline Rod−Coil Block Copolymers by Stable Free Radical Polymerization:  Synthesis, Morphology, and Rheology. Macromolecules 36, 3357–3364 (2003). 178 39. Li, C. Y. et al. Hierarchical Assembly of a Series of Rod−Coil Block Copolymers:  Supramolecular LC Phase in Nanoenviroment. Macromolecules 37, 2854–2860 (2004). 40. Tenneti, K. K. et al. Perforated Layer Structures in Liquid Crystalline Rod−Coil Block Copolymers. J. Am. Chem. Soc. 127, 15481–15490 (2005). 41. Shi, L.-Y., Zhou, Y., Fan, X.-H. & Shen, Z. Remarkably Rich Variety of Nanostructures and Order–Order Transitions in a Rod–Coil Diblock Copolymer. Macromolecules 46, 5308–5316 (2013). 42. Hillmyer, M. A. & Lodge, T. P. Synthesis and self-assembly of fluorinated block copolymers. J. Polym. Sci. Part Polym. Chem. 40, 1–8 (2002). 43. Cintora, A. et al. Block copolymers containing stable radical and fluorinated blocks with long-range ordered morphologies prepared by anionic polymerization. Polym. Chem. 10, 5094–5102 (2019). 44. Wang, Z. & Zuilhof, H. Self-Healing Superhydrophobic Fluoropolymer Brushes as Highly Protein-Repellent Coatings. Langmuir 32, 6310–6318 (2016). 45. Nakatani, R. et al. Perpendicular Orientation Control without Interfacial Treatment of RAFT-Synthesized High-χ Block Copolymer Thin Films with Sub- 10 nm Features Prepared via Thermal Annealing. ACS Appl. Mater. Interfaces 9, 31266–31278 (2017). 46. Maeda, R., Hayakawa, T. & Ober, C. K. Dual Mode Patterning of Fluorine- Containing Block Copolymers through Combined Top-down and Bottom-up Lithography. Chem. Mater. 24, 1454–1461 (2012). 179 47. Rastogi, A., Paik, M. Y., Tanaka, M. & Ober, C. K. Direct Patterning of Intrinsically Electron Beam Sensitive Polymer Brushes. ACS Nano 4, 771–780 (2010). 48. Bhairamadgi, N. S., Pujari, S. P., van Rijn, C. J. M. & Zuilhof, H. Adhesion and Friction Properties of Fluoropolymer Brushes: On the Tribological Inertness of Fluorine. Langmuir 30, 12532–12540 (2014). 49. Wang, Y. & Chang, Y. C. Preparation of Unidirectional End-Grafted α-Helical Polypeptides by Solvent Quenching. J. Am. Chem. Soc. 125, 6376–6377 (2003). 50. Menzel, M. et al. Entropic death of nonpatterned and nanopatterned polyelectrolyte brushes. J. Polym. Sci. Part Polym. Chem. 57, 1283–1295 (2019). 51. Lu, C.-H., Wang, J.-H., Chang, F.-C. & Kuo, S.-W. Star Block Copolymers Through Nitroxide-Mediated Radical Polymerization From Polyhedral Oligomeric Silsesquioxane (POSS) Core. Macromol. Chem. Phys. 211, 1339–1347 (2010). 52. Husseman, M. et al. Controlled Synthesis of Polymer Brushes by “Living” Free Radical Polymerization Techniques. Macromolecules 32, 1424–1431 (1999). 53. Hawker, C. J., Barclay, G. G., Orellana, A., Dao, J. & Devonport, W. Initiating systems for nitroxide-mediated “living” free radical polymerizations: Synthesis and evaluation. Macromolecules 29, 5245–5254 (1996). 54. Hawker, C. J., Barclay, G. G. & Dao, J. Radical Crossover in Nitroxide Mediated “Living” Free Radical Polymerizations. J. Am. Chem. Soc. 118, 11467–11471 (1996). 180 55. Ezrin, M., Updegraff, I. H. & Cassidy, H. G. Electron Exchange—Polymers. III. Polymers and Copolymers of Vinylhydroquinone. J. Am. Chem. Soc. 75, 1610– 1614 (1953). 56. Heo, H. J., Han, D. J. & Sohn, E.-H. Impact of trifluoromethyl groups on the control of surface and optical properties of poly(methyl methacrylate). J. Fluor. Chem. 219, 92–97 (2019). 57. Sinturel, C., Bates, F. S. & Hillmyer, M. A. High χ–Low N Block Polymers: How Far Can We Go? ACS Macro Lett. 4, 1044–1050 (2015). 58. Sun, T., Feng, L., Gao, X. & Jiang, L. Bioinspired Surfaces with Special Wettability. Acc. Chem. Res. 38, 644–652 (2005). 59. Watson, G. S. et al. A gecko skin micro/nano structure – A low adhesion, superhydrophobic, anti-wetting, self-cleaning, biocompatible, antibacterial surface. Acta Biomater. 21, 109–122 (2015). 60. Jaggessar, A., Shahali, H., Mathew, A. & Yarlagadda, P. K. D. V. Bio-mimicking nano and micro-structured surface fabrication for antibacterial properties in medical implants. J. Nanobiotechnology 15, (2017). 181 CHAPTER 6 Synthesis of polymer brushes via SI-SET-LRP in the presence of air using copper tape and hydrazine 6.1 Abstract In this work, we employed surface initiated- single electron transfer-living radical polymerization (SI-SET-LRP) using Cu tape and hydrazine to prepare polymer brushes in the presence of air. In this method, hydrazine reduced a Cu2O layer on Cu tape to pristine Cu(0) which was quickly reacted with oxygen and oxidized back to Cu2O. This redox cycle continued until all oxygen was consumed. By eliminating the deoxygenation step, this method is more facile and effective than conventional control radical polymerization to prepare thick polymer brushes with excellent control of molecular weight and chain-end fidelity. Various monomers from styrene to fluorinated methacrylates were successfully polymerized using this method. The effect of hydrazine concentration on brush thickness was also investigated. At low concentration, hydrazine can definitely accelerate the polymerization and brush thickness increased with greater hydrazine concentration. However, when hydrazine concentrations were excessive, surface Cu(0) activator was overproduced while Cu(II) deactivator was decreased. This situation led to a high concentration of propagating radicals and bimolecular termination occurred, thus decreasing brush thickness. We also optimized the hydrazine concentration for each monomer. The effect of temperature on SI-SET-LRP was also studied with PMMA brushes. A higher temperature increased polymerization rate, resulting thicker brushes at short reaction time. However, the polymerization had a higher degree of living character at lower temperature, resulting in higher brush thickness for long reaction time. Due to the facile setup, polymer brushes on wafer-sized substrates were also prepared. The livingness of the system was demonstrated by the formation of diblock and triblock copolymer brushes. The effect of the distance between 182 a Cu surface and the initiator-bearing surface was investigated to determine the optimal separation distance. Moreover, by tilting a Cu surface with respect to the initiator- bearing surface, gradient polymer brushes were achieved. 6.2 Introduction Polymer brushes have drawn substantial interest in the last twenty years for their ability to transform the properties of a surface such as hydrophobicity, adhesion, friction, and anti-fouling berhavior.1,2 Applications of polymer brushes have been reported for biosensors3,4, thermoresponsive surfaces5, chromatographic devices6, drug delivery7, directing of cell growth,8 or control of surface wettability 9. Since their first introduction by Matyjaszewski10 in 1999, surface-initiated atom transfer radical polymerization (SI-ATRP) has been the predominant method for polymer brush synthesis due to its versatility for a wide range of monomers and precise control over polymer composition and architecture.2,11 A major issue of conventional SI-ATRP is a requirement for stringent deoxygenation conditions which prevent it from being used in the industry. Thus, modified SI-ATRP has been developed to overcome this issue. Matyjaszewski reported an improved method called Activator Regenerated by Electron Transfer (ARGET) ATRP which eliminated deoxygenation by the introduction of excess reducing agent.12 However, this method has been largely used for polymerization of methacrylate monomers in water or polar solvents.13,14 The results were also not reproducible unless protocol was strictly followed.14 Huck and coworkers developed another method called electrochemically mediated ATRP (eATRP) when a negative potential was applied to generate Cu(I) catalyst from Cu(II) at a working electrode.15 As Cu(I) diffused from the electrode to an initiator-immobilized substrate, there was a gradient of Cu(I) concentration along the path from the electrode to the substrate. Thus, by tilting the substrate at an angle to the substrate, gradient polymer brushes were 183 produced.15 The method was furthered improved by replacing the electrode with a sacrificial anode such as zinc metal which reduced Cu(II) to Cu(I) .16 However, they only reported the brush synthesis of hydrophilic monomers in aqueous solution. Percec introduced a method called the single electron transfer-living radical polymerization (SET-LRP) as an alternative to ATRP.17 In SET-LRP, a halide terminated initiator is activated by Cu(0) via an outer-sphere single-electron transfer process to generate Cu(I) and reactive radical species to which monomers are added.18– 20 The Cu(I) species disproportionate into Cu(0) activator and Cu(II) deactivator. The latter deactivates the propagating radicals to dormant chains. By this mechanism, the disproportionation of Cu(I) plays a key role in the activation-deactivation process.19 Moreover, the disproportionation of Cu(I) was dependent on solvent polarity and ligands, so solvents such as DMSO, alcohol and water in combination with N- containing ligands such as PMDETA or tris(2-(dimethylamino) ethyl)amine (Me6- TREN) are commonly used.17,21 SET-LRP can be catalyzed by Cu(0) powders or wires to polymerize methacrylate, acrylate, acrylamide or zwitterionic monomers with excellent control of polymer molecular weight and dispersity under mild conditions. 21,22 SET-LRP was used to prepare high-order multiblock copolymers with high chain- end fidelity, indicating the livingness of the approach. Interestingly, no purification was required after each block formation step because the reaction was carried out to full monomer conversion.23 Other polymer architectures such as star polymers or dendritic macromolecules have been reported.24,25 Originally, SET-LRP required stringent deoxygenated procedures such as freeze-pump-thaw or inert-gas bubbling, but Percec overcame this limitation by the addition of reducing agent such as hydrazine hydrate into the system.19 He demonstrated the synthesis of polymethacrylate in the presence of air with high conversion and 184 excellent chain-end fidelity. The role of hydrazine was to reduce Cu2O from the copper wire surface to Cu(0). Although the reaction conditions and mechanism of SET-LRP for the synthesis of free polymer in solution were well studied a decade ago, the synthesis of polymer brushes via SET-LRP only gained attention recently. Zhang et al. reported the first study of surface-initiated SET-LRP (i.e. SI-CuCRP) using a copper plate.26 In this method, a copper plate and a silicon substrate with immobilized ATRP initiators were sandwiched face to face by a clamp and the distance between the two surfaces was 0.5 mm. Numerous vinyl monomers were successfully polymerized and the fabrication of block copolymer brushes, gradient brushes and patterned brushes were also reported, though anaerobic conditions were still required. After that, Zhang further utilized Cu plate to grow polymer brushes under ambient conditions.27 However, the method only worked well for hydrophilic polymers in aqueous media. Thus, it was in our interest to improve the method so that it could be used for the polymerization of styrenic and fluorinated monomers in the presence of air. Inspired by Percec’s work, we attempted to introduce hydrazine into the system so that the deoxygenation step was eliminated. Moreover, our preliminary results showed that after being used for a while, especially at high reaction temperature, the surface of the Cu plate became contaminated with the deposit of organic residues. This coating can definitely affect the performance and reproducibility of SI- SET-LRP using old Cu plate. For that reason, we used a commercially available, and cheaper material which is Cu tape as an alternative and disposable Cu source. In this work, we report the preparation of hydrophobic polymer brushes via SI-SET-LRP in air using Cu tape and hydrazine. The effects of hydrazine concentration, temperature and the distance between the sample and Cu surface on the brush growth were investigated for various monomers. Moreover, the facile setup allowed us to grow brushes on wafer- 185 sized sample. The living character of this method was demonstrated by the preparation of diblock and triblock copolymer brushes. Other architectures such as gradient brushes were also possible to produce. 6.3 Experimental Section 6.3.1 Materials 2-Bromo-2-methylpropionyl bromide (BiBB), anhydrous triethylamine (TEA), CuBr, hydrazine, basic alumina, inhibitor remover (for removing hydroquinone and monomethyl ether hydroquinone), anhydrous toluene, styrene (S), methyl methacrylate (MMA), 2-(dimethylamino)ethyl methacrylate (DMAEMA), 4-vinylpyridine (4VP), heptadecafluorodecyl methacrylate (HDFDMA), trifluoroethanol (TFE),1,1,4,7,7- penta-methyldiethylenetriamine (PMDETA), anhydrous acetone, anhydrous dichloromethane (DCM) were purchased from Sigma Aldrich. 3-amino- propyltriethoxysilane (APTES) was purchased from Gelest. Perfluoro butyl ethyl methacrylate (PFBEMA) was purchased from Fluoryx Inc. Copper wire (20 gauge) was from Fisher. Styrene was passed through basic alumina to remove the inhibitor before use. Other monomers were passed through inhibitor remover to remove the inhibitor before use. Copper tape (882-L COPPER) with 88.9 μm copperfilm was purchased from Lamart Co. Deionized water with a resistivity of 18.2 MΩ • cm at 25 °C was obtained from Millipore’s Milli-Q Synthesis A10 system. All other solvents were purchased from Fisher Scientific. 6.3.2 Synthesis Immobilization of Silane Initiators: Silicon wafer was cleaned with piranha solution (H2O2 : H2SO4, 1:3 v/v) at 90°C for 60 minutes. Warning: Piranha solution reacts violently with organic materials. Then the silicon wafer was cleaned with D.I water and dried under nitrogen gas. The clean substrate was immersed into a 5% (v/v) APTES 186 solution in dry acetone and was sonicated for 45 minutes. After the formation of self- assembled monolayer (SAM) of APTES on surface, the sample was rinsed with acetone and dried under nitrogen gas. The sample was cured in an oven at 110°C for 30 minutes. The sample was placed into a solution of 2% (v/v) TEA in dry DCM in an ice bath. Then BiBB was added dropwise into the solution to get a final concentration of 2% (v/v) BiBB. The reaction was run for 12 hours before the sample was removed and rinsed with DCM, ethanol, and acetone, and dried under nitrogen gas. The APTES-BiBB layer was 2 nm as measured by ellipsometry. Synthesis of polymer brush via SI-SET-LRP: Cu tape was attached on glass pieces. A piece of Cu-taped glass was clamped together with a ATRP initiator-modified silicon substate by a copper clamp. Two spacers with 1 mm thickness were put at the two ends of the sample to separate it from Cu-taped glass. The setup was put into a monomer solution and polymerization was conducted for a given time. Afterward, the sample was removed from the solution and rinsed with dichloromethane, ethanol, water and dried with nitrogen gas. The recipe for each monomer was as following: MMA ( 2 mL monomer, 1 mL DMSO, 36 μL PMDETA); S (2 mL monomer, 1 mL DMSO, 36 μL PMDETA); 4VP (1.2 mL monomer, 1.8 mL DMSO, 22 μL PMDETA); DMAEMA ( 1.2 mL monomer, 1.2 mL H2O, 0.6 mL MeOH); HDFDMA (1.2 mL monomer, 1.8 mL TFE); PFBEMA (1.2 mL monomer, 1.8 mL TFE). The concentration of reducing agent was varied from 0 to 63.6 mM. The reaction temperature was 25°C or 60°C for MMA, 105°C for S, and 60°C for the other monomers. Block copolymer brushes: PMMA brushes were prepared via SI-SET-LRP as described above (3.18 mM hydrazine at 25°C). After that, the brush sample was subjected to a second SI-SET-LRP to grow the second block of PS brushes (15.9 mM hydrazine at 105°C). 187 Gradient brushes: an ATRP-initiator modified substrate was sandwiched with a Cu- tape glass. The two pieces had one end in direct contact and the other end was separated by a spacer with 1 mm thickness. The setup was subjected to SI-SET-LRP of MMA (3.18 mM hydrazine at 25°C). Synthesis of polymer brush via SI-SET-LRP using copper wire: copper wire wrapped around a stir bar was placed into a reaction mixture of MMA monomer at 3.18 mM of hydrazine. An initiator-bearing substrate was also put into the solution. The reaction was carried out at 25°C for two hours. Synthesis of polymer brush via SI-ATRP: styrene (7 mL, 61.1 mmol), anisole (2.5 mL), and PMDETA (67 μL, 0.32 mmol) were added together into a Schlenk flask and subsequently subjected to freeze-pump-thaw (3 cycles). In another Schlenk flask equipped with a stir bar, CuBr (23 mg, 0.16 mmol) was added, and the flask was evacuated and backfilled with Argon for five times. The reaction mixture was then cannulated to the flask containing CuBr. Polymerization was run at 105°C for two hours. 6.3.3 Characterization Thicknesses of polymer brushes were measured using Spectroscopic Imaging Auto- Nulling Ellipsometer EP³-SE (Nanofilm Technologies GmbH, Germany). All measurements were performed with 532 nm laser starting at an angle of incidence of 50o and ending at 60o with an angle increment of 1o. The data were fitted with a Cauchy model/silicon oxide/silicon stack model using the TFCompanion software (SemiconSoft, US). Veeco Icon AFM was used to characterize surface morphology and thickness of block-copolymer brushes. It was done with AC tapping mode using Olympus AC160TS probes. VCA Optima Contact Angle was used to measure the static water contact angle. 188 Scheme 6.1. Proposed mechanism of SI-SET-LRP in the presence of air and hydrazine 6.4 Results and Discussion It is reasonable to expect that SI-SET-LRP and SET-LRP have similar mechanisms, because they use Cu(0) to catalyze the reaction. The only difference is that SI-SET-LRP is surface-confined as initiators were covalently attached to a planar substrate and in close distance to the planar copper surface. Based on our understanding of SET-LRP, we propose a mechanism for SI-SET-LRP in the presence of reducing agent and oxygen (Scheme 6.1). As Cu(0) is easily oxidized to Cu2O in air at ambient temperature 28, it is expected that commercial Cu(0) (i.e. Cu tape) has a thin film of Cu2O on surface. Although Cu2O is also a catalyst for SET-LRP, it is much less reactive than Cu(0).17 The presence of hydrazine in the reaction mixture would result in reduction of a Cu2O layer to Cu(0) and subsequent activation of the copper surface. Both a freshly cleaned Cu(0) surface and pristine Cu(0) act as oxygen scavenger to react with O2 in the mixture and become oxidized back to Cu2O. The redox cycle continues 189 until all oxygen is depleted. In addition, the atomic Cu(0) can diffuse through the reaction medium to the surface-immobilized initiators. In the presence of N-ligands (i.e. PMDETA), Cu(0) acts as an SET activator for the initiator through heterolytic cleavage of the C-Br bond so that propagating radicals and Cu(I)X/L complexes are formed.19 Dipolar aprotic solvent (i.e. DMSO) promotes the spontaneous disproportionation of Cu(I)X/L complex to atomic Cu(0) and Cu(II)X2/L which subsequently deactivates the propagating radical.19 Figure 6.1. a) PMMA brush thickness as a function of hydrazine concentration for 2 hours of reaction; b) PMMA brush thickness as a function of reaction time with 3.18 mM of hydrazine. Since hydrazine can clearly affect the consumption of oxygen in the reaction mixture and the concentrations of activator and deactivator, we initially investigated the effect of hydrazine on polymer brush growth. In general, Cu-tape on glass was sandwiched with an initiator-bearing substrate at a distance of 1 mm by putting spacers at the two ends of the sample. For the preparation of PMMA brushes, the substrate setup 190 was placed in a reaction mixture of the monomer, hydrazine, DMSO as solvent, and PMDETA as ligand. The concentration of hydrazine was varied from 0 to 63.6 mM. The reaction was run at either 25°C or 60°C for two hours before the sample was removed from the solution and cleaned. At 25°C and without hydrazine, no formation of PMMA brushes was observed. By adding a minute amount of hydrazine (1.59 mM), homogeneous PMMA brushes with a thickness of 69 nm were produced. Increasing hydrazine concentration to 3.18 mM resulted in a brush thickness of 119 nm. However, the brush thickness started decreasing when hydrazine concentration increased beyond 3.18 mM. Indeed, the thickness reduced to 66 nm at a concentration of 63.6 mM hydrazine. The dependence of brush thickness as a function of hydrazine concentration indicated the crucial role of hydrazine on brush growth (Figure 6.1a). Without hydrazine, activation of the initiators came from bulk Cu2O which is much less reactive than atomic Cu(0), and any initiator-derived radicals would be quickly terminated by oxygen trapping, thereby causing unsuccessful brush growth. In the presence of hydrazine (1.59 mM), pristine atomic Cu(0), an efficient catalyst for SET-LRP, was generated. Moreover, Cu(0) efficiently removed oxygen, thereby reducing early termination of the radicals. Consequently, the polymerization could proceed to produce PMMA brushes. Higher concentration of hydrazine (3.18 mM) resulted in faster generation of Cu(0) and depletion of oxygen. In addition, it was likely that the propagation rate was higher with increasing hydrazine concentration, similar to what was reported in the literature for free polymers.19 Higher concentration led to much thicker brushes of 119 nm. However, excess hydrazine had a detrimental effect on brush growth. After all oxygen was consumed, it was expected that the production of pristine Cu(0) was proportional to the remaining quantity of hydrazine, so at a high concentration of hydrazine, a rapid initiation could increase the propagating radical concentration. Moreover, hydrazine may reduce some Cu(II) to Cu(0), thereby 191 decreasing the concentration of deactivator. As a result, it was more likely to increase the chance of irreversible bimolecular termination, thus lowering the brush thickness. A similar phenomenon was reported for SET-LRP of free poly(methyl acrylate) in excess hydrazine when the conversion reached a plateau of 50% with an increase in polydispersity.19 Fleischmann et. al reported that the polymerization rate of free PMMA in DMSO via SET-LRP was accelerated with an increase of the reaction temperature while maintaining a low polydispersity and first order kinetics.18 Thus, we attempted to carry out SI-SET-LRP of PMMA brushes at 60°C. The relationship between PMMA brush thickness and hydrazine concentration was similar to that at 25°C (Figure 6.1a). However, a thin brush of 15 nm was observed without hydrazine indicating the rate of propagation was significantly higher for 60°C than that of 25°C, though an early termination still occurred by oxygen trapping. An addition of hydrazine improved the polymerization and brush thickness increased to 180 nm at 1.59 mM of hydrazine. The difference in brush thickness between the two reaction temperatures was significant at a low concentration of hydrazine, but quickly diminished with an increase of hydrazine content. The polymerization kinetics study was carried out to examine the livingness of the system (Figure 6.1b). At 25°C using 3.18 mM of hydrazine, the linear correlation between brush thickness and reaction time was observed for the first three hours before it started to level off. It is noteworthy that the only source of the halide group (Br) in the system came from the surface-immobilized initiators which was a tiny amount. For that reason, the amount of Cu(I)X/L and Cu(II)X2/L complexes present was extremely small. Initially, the complexes were located in the small space between Cu surface and initiator-bearing surface. However, over time, the complexes could escape from within the sandwiched region, thereby reducing the effective Cu(I)X/L and Cu(II)X2/L 192 complex concentration participating in the polymerization. This reduction increased the likelihood of bimolecular termination between neighboring polymer chains, resulting in a plateau in brush growth rate. At 60°C and using 3.18 mM of hydrazine, brush growth was linear for the first two hours and subsequently leveled off. It suggested that although at the beginning, the polymer growth rate at 25°C was lower than that of 60°C, the ability of maintain the livingness of the reaction resulted in a higher brush thickness at 25°C than that at 60°C after six hours of reaction. We also attempted to grow PS brushes via SI-SET-LRP at 105°C. Without hydrazine, there was no formation of brushes. At 1.59 mM hydrazine, the polymerization was dramatically accelerated to produce PS brushes with a thickness of Figure 6.2. a) PS brush thickness as a function of hydrazine concentration at 105°C for 2 hours of reaction time; b) PS brush thickness as a function of reaction time at 105°C with 15.9 mM of hydrazine 85 nm after two hours (Figure 6.2a). The brushes displayed a static water contact angle (WCA) of 94.5°, confirming the presence of PS. The relationship between brush thickness and hydrazine concentration was similar to that of PMMA brushes. 193 Specifically, higher amounts of hydrazine produced more pristine Cu(0) and removes oxygen more quickly, thus increasing the initiation and propagation rates. The optimal hydrazine concentration was 15.9 mM for PS brushes of 133 nm. However, excess hydrazine above this level decreased the brush thickness due to bimolecular termination. For comparison, PS brushes were also prepared with SI-ATRP using similar conditions (105°C, two hour reaction, PMDETA as ligand, CuBr as catalyst and anisole as solvent), Figure 6.3. Brush thickness as a function of hydrazine concentration for a) P4VP at 60°C for 2 hours; b) PHDFDMA at 60°C for 5 hours; c) PPFBEMA at 60°C for 5 hours 194 and even after a degassing step of freeze-pump-thaw cycles, only 25 nm of PS brushes was produced. It indicated that SI-SET-LRP was more effective than SI-ATRP in preparation of PS brushes. Polymerization kinetics studies was also performed at 15.9 mM hydrazine. Brush thickness increased linearly with time within the first two hours before reaching a plateau (Figure 6.2b). In addition, P4VP was also successfully prepared via SI-SET-LRP. The trend of P4VP thickness as a function of hydrazine was similar to that of PMMA and PS. Brush thickness increased from 0 to 81 nm after two hours of reaction time at 60°C when hydrazine concentration is changed from 0 to 15.9 mM (Figure 6.3a). Then the thickness gradually decreased to 35 nm when hydrazine concentration increased to 254 mM. P4VP brushes of 81 nm had a WCA of 56.5°. The versatility of SI-SET-LRP was further demonstrated by the synthesis of fluorinated polymer brushes. Although SET-LRP was has been used to polymerize fluorinated acrylates and methacrylates,29 to our knowledge, our work was the first report of the synthesis of fluorinated polymer brushes via SI-SET-LRP. Specifically, HDFDMA and PFBEMA were the two monomers in the study. Instead of DMSO, trifluoroethanol (TFE) was used as solvent because it contains both fluorinated and hydroxyl sites. While the fluorinated site can enhance the solubility of the monomers, the hydroxyl group can mediate the disproportionation of Cu(I)X/L which is the key step of SET-LRP.29 The polymerization was carried at 60°C in air, much lower than the elevated temperature (110°C) for SI-ATRP of fluorinated brushes which also required a deoxygenation step.30 Interestingly, a higher concentration of hydrazine was needed compared to that of the previous monomers. The brushes started to form using 3.18 mM hydrazine and the thickness reached to its highest value (54 nm for PHDFDMA and 37 nm for PPFBEMA brushes) at 31.8 mM hydrazine (Figures 6.3b,c). It can be explained that the solubility of hydrazine in fluorinated mixtures was lower than in DMSO. PHDFDMA and PPFBEMA brushes displayed WCAs of 127° 195 and 117° respectively. PHDFDMA had a higher WCA due to longer fluoroalkyl side chains. Overall, the relationship between brush thickness and hydrazine concentration was similar for all the polymers, though the optimal conditions were different due to the nature of monomers and solvents. As a deoxygenation step was eliminated in SI-SET-LRP by adding hydrazine, it became possible to prepare brushes on a large sample. Specifically, a wafer-size initiator-bearing substrate was sandwiched with a Cu-tape surface and the setup was placed in a petri dish before a reaction mixture with MMA as monomer was introduced. After 1.5 hours at room temperature, PMMA brushes with a thickness of 91 nm were obtained. The uniformity in thickness across the sample was confirmed with ellipsometry. Moreover, the living character of SI-SET-LRP in air was demonstrated by the preparation of block copolymer brushes. In a first example, PMMA homopolymer brushes of 79 nm were prepared (Figures 6.4a-c). AFM revealed that the surface was smooth with a roughness of 0.4 nm. WCA for PMMA brushes was 68°. After thorough cleaning, the brush sample was subjected to another SI-SET-LRP to grow PS brushes. AFM showed an increase in polymer film thickness to 157 nm (Figure 6.4d-f), indicating the addition of PS block. The height of PS block was estimated as 78 nm from the difference between the former and latter thicknesses. AFM also showed some phase separation in |-PMMA-b-PS brushes. The surface roughness for the diblock copolymer brushes was 1.4 nm and WCA was 83°. Other diblock and triblock copolymer brushes were also prepared including |-PMMA-b-PDMAEMA, |- PDMAEMA-b-PS, and |-PDMAEMA-b-PS-b-PMMA brushes. These examples indicated the high livingness of SI-SET-LRP using Cu tape and hydrazine. 196 Figure 6.4. a-c) AFM image of PMMA brush thickness, height profile and surface morphology; d-f) AFM image of |-PMMA-b-PS brush thickness, height profile and surface morphology. The scale bars are 2 μm for (a) and (d), and 200 nm for (c) and (f) After being generated from the reduction of Cu2O by hydrazine at the Cu surface, pristine atomic Cu(0) had to diffuse from the Cu metal surface to the opposite initiator-bearing surface so that the polymerization could start. Moreover, the volume of confined space between the two surfaces can change with the distance D between the surfaces. Higher D should result in a larger volume. Thus, assuming that all the Cu (Cu(0) and ligated Cu complexes) was trapped in the zone between the two surfaces, 197 the concentration of Cu in the space was inversely proportional to D. Consequently, separation could affect the polymerization rate. For that reason, we investigated the effect of D on the polymerization in terms of brush thickness. Specifically, D was varied by changing the thickness of spacers at the two ends of a sample while keeping all other reaction conditions (reaction time, temperature, hydrazine concentration) constant. PMMA brush thickness as a function of D was plotted (Figure 6.5a). When the two surfaces were in contact (D = 0), only thin PMMA brushes were observed (h = 6 nm). Interestingly, brush thickness increased almost linearly with D and reached a maximum value of 70 nm at D = 0.84 mm. However, when D was increased beyond 0.84 mm, brush thickness started to decrease. At D = 3 mm, h was 37 nm. For D less than 0.85 mm, the concentration of Cu decreased with an increase of D. At D = 0, a high concentration of Cu(0) could cause fast activation of dormant chains, thus increasing the concentration of propagating radicals. Consequently, bimolecular termination between neighboring radicals was more likely to occur. It led to lower molecular weight, and even lower grafting density, thereby decreasing brush thickness. At a higher D, the concentration of Cu(0) was lower, so the polymerization had a higher degree of livingness, and the polymer brush became thicker. However, when D was too large (> 0.85 mm), Cu(I)X/L and Cu(II)X2/L complexes were more likely to escape from the confined space. Without the complexes, the equilibrium of activation/deactivation was disrupted, and the polymerization became less controlled. As a result, polymer brushes became thinner. For comparison, Cu tape was replaced by Cu wire wrapped around a stir bar, so there was no confined space for Cu pieces. Consequently, PMMA brushes with a thickness of ~ 10 nm was observed. As the correlation between brush thickness and D was established, we attempted to produce gradient brushes by tilting Cu surface. Specifically, a 1 mm spacer was placed at one end of an initiator-bearing substate while Cu surface was in contact with 198 the substrate at the other end. By doing that, the distance between Cu surface and the sample was monotonically varied along the sample. We expected that the local concentration of Cu changed accordingly along the sample. Using MMA as monomer, a linear increase in brush thickness along the sample was observed (Figure 6.5b). Figure 6.5. a) PMMA brush thickness as a function of the distance D between Cu surface and initiator-bearing surface; b) thickness of gradient PMMA brushes with a tilt Cu plate 6.5 Conclusion We reported SI-SET-LRP in which a planar Cu surface was kept parallel in a short distance to an initiator-bearing substate so that both the activator and deactivator were in the confined space. Hydrazine was used as a reducing agent to facilitate the deoxygenation of the reaction mixture. Cu(0) generated from the reduction of a Cu2O layer by hydrazine was oxidized by oxygen and all oxygen was quickly consumed through this redox process. The dependence of brush thickness as a function of hydrazine concentration was examined. A minute amount of hydrazine can enhance the polymerization, but excess hydrazine can increase the concentration of Cu(0) activator 199 and reduce the concentration of Cu(II) deactivator, thus causing bimolecular termination. Optimal amount of hydrazine for controlled polymerization was determine for several monomers. The effect of temperature on SI-SET-LRP was also discussed. While an elevated temperature can speed up the polymerization, low temperature led to a higher control of polymerization. The living character of SI-SET-LRP in air was demonstrated by the preparation of block copolymer brushes. Polymer brushes can be produced over a large area, indicating the potential of this method for industrial applications. The concentrations of activator and deactivator in the confined space was varied by simply adjusting the distance between a Cu surface and the initiator-bearing surface. Brush thickness was changed accordingly. By tilting a Cu surface with respect to the initiator-bearing sample, the local concentration of Cu was varied along the sample so that gradient brushes can be prepared. In the current system, the source of Br atoms was only from the initiators attached to the surface. By introducing a halide source such as sodium bromide to the system, there is a higher degree of freedom to tune the concentrations of Cu(I)X/L and Cu(II)X2/L. Such a change will give us another parameter to have better control of the polymerization. While hydrazine is an effective reducing agent, it is highly toxic. Thus, there is a need to explore more environmentally friendly reducing agents such ascorbic acid, thiourea dioxide. In addition, as solvent and ligand are known to play an important role on the disproportionation of Cu(I), future work should also explore other choices of solvent and ligand to further improve SI-SET-LRP. Moreover, an Fe surface is an interesting choice to replace Cu surface due to its low environmental toxicity and its lower cost. Acknowledgement We would like to acknowledge the excellent support on polymer synthesis and AFM characterization from Ihsan Amin and Clemens Liedel, respectively . 200 6.6 References 1. Chen, W.-L., Cordero, R., Tran, H. & Ober, C. K. 50th Anniversary Perspective: Polymer Brushes: Novel Surfaces for Future Materials. Macromolecules 50, 4089–4113 (2017). 2. Zoppe, J. O. et al. Surface-Initiated Controlled Radical Polymerization: State-of- the-Art, Opportunities, and Challenges in Surface and Interface Engineering with Polymer Brushes. Chem. Rev. 117, 1105–1318 (2017). 3. Welch, M. E. et al. Generalized Platform for Antibody Detection using the Antibody Catalyzed Water Oxidation Pathway. J. Am. Chem. Soc. 136, 1879– 1883 (2014). 4. Yu, Q., Ista, L. K., Gu, R., Zauscher, S. & López, G. P. Nanopatterned polymer brushes: conformation, fabrication and applications. Nanoscale 8, 680–700 (2015). 5. Yu, Q., Johnson, L. M. & López, G. P. Nanopatterned Polymer Brushes for Triggered Detachment of Anchorage-Dependent Cells. Adv. Funct. Mater. 24, 3751–3759 (2014). 6. von Werne, T. A. et al. A Versatile Method for Tuning the Chemistry and Size of Nanoscopic Features by Living Free Radical Polymerization. J. Am. Chem. Soc. 125, 3831–3838 (2003). 7. Aksay, I. A. et al. Biomimetic Pathways for Assembling Inorganic Thin Films. Science 273, 872–898. 201 8. Zhou, Z., Yu, P., Geller, H. M. & Ober, C. K. Biomimetic Polymer Brushes Containing Tethered Acetylcholine Analogs for Protein and Hippocampal Neuronal Cell Patterning. Biomacromolecules 14, 529–537 (2013). 9. Tian, Y., Su, B. & Jiang, L. Interfacial Material System Exhibiting Superwettability. Adv. Mater. 26, 6872–6897 (2014). 10. Matyjaszewski, K. et al. Polymers at Interfaces: Using Atom Transfer Radical Polymerization in the Controlled Growth of Homopolymers and Block Copolymers from Silicon Surfaces in the Absence of Untethered Sacrificial Initiator. Macromolecules 32, 8716–8724 (1999). 11. Matyjaszewski, K. Advanced Materials by Atom Transfer Radical Polymerization. Adv. Mater. 30, 1706441 (2018). 12. Matyjaszewski, K., Dong, H., Jakubowski, W., Pietrasik, J. & Kusumo, A. Grafting from Surfaces for “Everyone”: ARGET ATRP in the Presence of Air. Langmuir 23, 4528–4531 (2007). 13. Dunderdale, G. J., Urata, C., Miranda, D. F. & Hozumi, A. Large-Scale and Environmentally Friendly Synthesis of pH-Responsive Oil-Repellent Polymer Brush Surfaces under Ambient Conditions. ACS Appl. Mater. Interfaces 6, 11864– 11868 (2014). 14. Dunderdale, G. J., England, M. W., Urata, C. & Hozumi, A. Polymer Brush Surfaces Showing Superhydrophobicity and Air-Bubble Repellency in a Variety of Organic Liquids. ACS Appl. Mater. Interfaces 7, 12220–12229 (2015). 15. Li, B., Yu, B., Huck, W. T. S., Liu, W. & Zhou, F. Electrochemically Mediated Atom Transfer Radical Polymerization on Nonconducting Substrates: Controlled 202 Brush Growth through Catalyst Diffusion. J. Am. Chem. Soc. 135, 1708–1710 (2013). 16. Yan, J. et al. Controlled Polymer-Brush Growth from Microliter Volumes using Sacrificial-Anode Atom-Transfer Radical Polymerization. Angew. Chem. Int. Ed. 52, 9125–9129 (2013). 17. Percec, V. et al. Ultrafast Synthesis of Ultrahigh Molar Mass Polymers by Metal- Catalyzed Living Radical Polymerization of Acrylates, Methacrylates, and Vinyl Chloride Mediated by SET at 25 °C. J. Am. Chem. Soc. 128, 14156–14165 (2006). 18. Fleischmann, S. & Percec, V. SET-LRP of methyl methacrylate initiated with CCl4 in the presence and absence of air. J. Polym. Sci. Part Polym. Chem. 48, 2243–2250 (2010). 19. Fleischmann, S., Rosen, B. M. & Percec, V. SET-LRP of acrylates in air. J. Polym. Sci. Part Polym. Chem. 48, 1190–1196 (2010). 20. Jiang, X., Rosen, B. M. & Percec, V. Immortal SET–LRP mediated by Cu(0) wire. J. Polym. Sci. Part Polym. Chem. 48, 2716–2721 (2010). 21. Nguyen, N. H. & Percec, V. Dramatic acceleration of SET-LRP of methyl acrylate during catalysis with activated Cu(0) wire. J. Polym. Sci. Part Polym. Chem. 48, 5109–5119 (2010). 22. Ding, W. et al. Synthesis of zwitterionic polymer by SET-LRP at room temperature in aqueous. J. Polym. Sci. Part Polym. Chem. 49, 432–440 (2011). 23. Soeriyadi, A. H., Boyer, C., Nyström, F., Zetterlund, P. B. & Whittaker, M. R. High-Order Multiblock Copolymers via Iterative Cu(0)-Mediated Radical 203 Polymerizations (SET-LRP): Toward Biological Precision. J. Am. Chem. Soc. 133, 11128–11131 (2011). 24. Whittaker, M. R., Urbani, C. N. & Monteiro, M. J. Synthesis of linear and 4-arm star block copolymers of poly(methyl acrylate-b-solketal acrylate) by SET-LRP at 25 °C. J. Polym. Sci. Part Polym. Chem. 46, 6346–6357 (2008). 25. Rosen, B. M., Lligadas, G., Hahn, C. & Percec, V. Synthesis of dendritic macromolecules through divergent iterative thio-bromo “Click” chemistry and SET-LRP. J. Polym. Sci. Part Polym. Chem. 47, 3940–3948 (2009). 26. Zhang, T., Du, Y., Müller, F., Amin, I. & Jordan, R. Surface-initiated Cu(0) mediated controlled radical polymerization (SI-CuCRP) using a copper plate. Polym. Chem. 6, 2726–2733 (2015). 27. Zhang, T. et al. Wafer-scale synthesis of defined polymer brushes under ambient conditions. Polym. Chem. 6, 8176–8183 (2015). 28. Grouchko, M., Kamyshny, A. & Magdassi, S. Formation of air-stable copper– silver core–shell nanoparticles for inkjet printing. J. Mater. Chem. 19, 3057–3062 (2009). 29. Samanta, S. R., Cai, R. & Percec, V. SET-LRP of semifluorinated acrylates and methacrylates. Polym. Chem. 5, 5479–5491 (2014). 30. Wang, Z. & Zuilhof, H. Antifouling Properties of Fluoropolymer Brushes toward Organic Polymers: The Influence of Composition, Thickness, Brush Architecture, and Annealing. Langmuir 32, 6571–6581 (2016). 204 CHAPTER 7 Summary and Outlook 7.1 Mixed helical polypeptide rod-coil brushes Through this comprehensive study, we have a better understanding of rod-type homopolymer brushes and mixed rod-coil brushes. In chapter 2, which to the best of our knowledge is the first published work of rod-coil mixed brushes, we studied how the presence of coil brushes can affect the organization and orientation of polypeptide brushes. PBLG polypeptide and PMMA brushes were grown from mixed initiators via SI-ROP and SI-ATRP respectively. While keeping the same molecular weight (MW) of PBLG, the MW of PMMA was varied by controlling the polymerization time. A trend was observed that longer PMMA brushes tended to push PBLG rods up from surface, resulting in a higher tilt angle of PBLG which was verified by FT-IR measurements. Consequently, the film thickness of mixed brushes became higher than that of homopolymer brushes. However, PBLG rods were still randomly oriented, making the topmost surface look flat and homogenous. After solvent quenching with chloroform and acetone, PBLG rods assembled into rod bundles, resulting in teepee-like nanostructures. With the entanglement of PMMA in mixed brushes, PBLG rods can not freely move as in the homopolymer system. Therefore, after quenching, the tilt angle of PBLG in mixed brushes was less than that of the homopolymer. Moreover, the number of teepee-like nanostructures per μm2 increased with the addition of coil brushes. However, as only the topmost part of the teepees above the PMMA layer can be seen by AFM, the size and height of the teepees looked smaller at a higher MW of PMMA brushes. In addition, the change in surface morphology and chemical groups exposed on the surface changed surface wettability. An increase in volume fraction of PMMA in mixed brushes made the surface more hydrophilic. With mixed polypeptide rod-coil 205 brushes, surface topography and chemical functionalities can be controlled. Therefore, they offer opportunities for molecular recognition and studies of cell-surface interaction. As an example, a preliminary study of cell behavior on mixed polypeptide rod- coil brushes was done in collaboration with Daniel group. For homopolymer brushes of PBLG, cell adhesion was much stronger on a polymer in the collapsed state that than in quenched state. It suggested that the high surface roughness of a quenched polymer brush reduced cell adhesion significantly. In addition, the morphology of cells on quenched polymers had lower circularity, indicating that polymer nanostructured teepees can stretch cells directionally. For mixed brushes in the collapsed state, an increase in MW of PMMA lowered cell adhesion. As XPS results showed a higher content of PMMA on the topmost surface for longer PMMA brushes, it implied that PMMA is less cell-compatible compared to PBLG. A decrease in cell adhesion was also observed for quenched mixed brushes, but the effect was weaker as the presence of PMMA decreased the surface roughness. For the same reason, cell circularity increased with longer PMMA brushes. Overall, the use of mixed polypeptide/coil brushes gave us an opportunity to reversibly control surface topography and chemical compositions which can induce different cell behaviors. Moreover, it has been reported that when helical polypeptides were aligned in a Scheme 7.1. Proposed hydrophilic (blue) and hydrophobic (red) coil monomers for mixed polypeptide- coil brushes magnetic or electric field, it can possess piezoelectric and non-linear optical properties.1 206 For example, the piezoelectric coefficient of a poled composite of 30% PBLG in PMMA is an order of magnitude less than that of poly(vinylidene fluoride), a typical piezoelectric material.2 Since a PMMA brush and solvent quenching can induce non- centrosymmetric alignment of PBLG, we will measure the piezoelectric properties of the mixed PBLG-coil brushes in the future. If the result is promising, mixed brushes will be a potential material for integrated actuators. Since our study implied that the interaction of the two polymers dictated the surface morphology, we are interested in exploring other polypeptides and coil type polymers, given the versatility of SI-ROP and SI-ATRP. We suspect that nanostructure size and spacing are dependent on the immiscibility of the rod and coil brushes. Thus, we will choose coil-type polymers with different hydrophilicity and functional groups to tailor polymer miscibility (Scheme 7.1). In addition to PBLG, other helical polypeptides can be synthesized from a number of amino acids such as substituted Scheme 7.2. Side groups of polypeptides glutamates, aspartates, tryptophan, and lysine (Scheme 7.2). Polypeptide brushes will 207 be selected based on the rod diameter, functional side groups and the solubility so that we will investigate the effect of those parameters to the formation of nanostructures in mixed brushes. In addition to XPS, FT-IR and AFM, we will use other characterization techniques such as grazing incidence small angle X-ray scattering (GISAXS), and neutron reflectivity to gain more valuable information about rod and coil structures. GISAXS can provide information about lateral microstructure while neutron reflectivity will reveal the internal organization of the brushes. In reported work, we prepared mixed initiators using the separate deposition of two initiators. Although that approach allows us to control the ratio of the two initiators, we can not ensure a uniform distribution of the initiators at the molecular level. A fluctuation in local grafting density of the initiators can affect the uniformity in size and spacings of the nanostructures. Alternatively, we will use a binary initiator containing two different reactive sites, each of which is used to grow a distinct type of polymer. Herein, we propose a new binary initiator which has an ATRP site and an amine site to Scheme 7.3. Proposed binary initiators 208 grow coil polymer and polypeptide, respectively (Scheme 7.3). We expect that the initiator can significantly improve the uniformity in mixed brushes. 7.2 Organization and orientation of polypeptide brushes Although we explored how coil brushes and solvent quenching influenced the orientation and organization of PBLG brushes, other factors had not been examined. Thus, in Chapter 3, the effect of polymer thickness (i.e. MW), solvent quality, and quenching conditions on PBLG homopolymer brushes was investigated. We found that rod orientation (or tilt angle) was strongly dependent on PBLG thickness measured in the quenched state. PBLG tilt angle from the surface normal decreased with an increase of polymer thickness, as verified by FT-IR. Moreover, AFM revealed that the size of nanostructures was proportional to polymer thickness while the number of nanostructures was decreased. We suspect that longer PBLG chains can reach other chains from farther location to aggregate together into nanostructures, thereby increasing the nanostructure size. Consequently, a larger nanostructure contains a higher number of rods, so the number of nanostructures became smaller. As a result, both the surface roughness and the spacing between adjacent nanostructures increased. On the other hand, PBLG brushes in the collapsed state after treatment with chloroform had different behaviors. AFM showed that brushes thicker than 25 nm looked very flat while 12 nm PBLG brushes had subtle features. Moreover, for chloroform-treated brushes, the rod tilt angle from the surface normal increased for thicker brushes. The changes in the rod tilt angle and surface topography with polymer thickness inspired us to examine liquid crystal (LC) response upon contact with brushes. Interestingly, 5CB LCs exhibited homeotropic alignment for all brush thicknesses in the quenched state, indicating an LC tilt angle of 0° from surface normal. On the other hand, for collapsed brushes, LC tilt angle changed from 0° for 12 nm brushes to 90° for 140 nm brushes. 209 After further investigation, we deduced that PBLG tilt angle of collapsed brushes was proportional to LC tilt angle, and it was caused by the dispersion interaction between PBLG and LCs. We suspected that the dramatic change in polymer organization and orientation after solvent quenching was partly due to the high persistence length of PBLG. Since the polymer stiffness was from intramolecular hydrogen bonding along the polymer backbone, we decided to break hydrogen bonds through thermal treatment, thereby reducing polymer persistence length. FT-IR showed an increased amount of random coil with thermal treatment time, indicating qualitatively the decrease in persistence length. PBLG tilt angle from the surface normal also increased accordingly. However, the formation of nanostructures was much less pronounced for brushes with longer thermal treatment time. It suggested that the persistence length played a critical role on PBLG tilt angle and organization. In the past studies, only pure chloroform and acetone were used as good solvent and non-solvent for PBLG respectively during the quenching process. Thus, we attempted to explore the importance of solvent quality on PBLG behaviors. The solvent quenching was not successful when a pair of chloroform and water was used while it worked well for a pair of chloroform and methanol. It implied that a good solvent and non-solvent of PBLG must be miscible and the extraction of a good solvent from solvated PBLG to a non-solvent was the basic mechanism of the quenching process. Moreover, the extraction time should be long enough to remove most of the good solvent from PBLG brushes. In addition, we tried to manipulate the extraction efficiency by using mixed solvents instead of pure non-solvent. As a result, the extraction rate of mixed solvents became weaker and the formation of heterogenous ring-like micro-sized domains were observed. Although we do not have a clear explanation, we attributed the domains to nucleated droplets of chloroform on a brush surface. Since FT-IR only 210 provides the average tilt angle of PBLG across mm2 surface areas, the method can not give information of micro-sized patterns. Since we established a correlation between the tilt angles of LCs and PBLG, LCs was demonstrated for the estimation of PBLG tilt angle on patterned, heterogenous surface. It suggested that LC can complement other techniques in characterization of PBLG brushes. Overall, we established criteria to control both the rod orientation and brush organization through polymer thickness, stiffness, and solvent conditions of the quenching process. However, we do not know whether they only apply to PBLG. Thus, it is a great interest to explore other helical polypeptides mentioned above (Scheme 7.2), and even other rod-type polymers. Moreover, the conformation transition from helix to coil through thermal treatment is permanent. Thus, stimuli-responsive polypeptides which can change conformation reversibly upon exposure to external conditions such as pH, surfactant, or ions will give us more freedom in changing surface topography and properties. 3–5 Given the great difference in thickness upon solvent treatment, polypeptide brushes can be used to open or close nanochannel gates. 7.3 Fabrication of nanopatterned polypeptide brushes In chapter 4, we fabricated nanopatterned PBLG brushes using the “bottom-up” approach. Electron-beam lithography (EBL) was used to pattern a film of ebeam resist into arrays of nanoholes to which initiators was deposited. Following the resist lift-off, nanopatterned PBLG brushes with feature size as small as 65 nm were grown via SI- ROP. We found that PBLG brushes can form teepee-like structures within the pattern area while polystyrene brushes were collapsed due to chain relaxation, thereby enlarging the pattern size. Moreover, the number of teepees in a patterned area depends on both polymer length and the size of the patterns. Thick brushes (i.e. long polymer chain) in a small, patterned area formed only one teepee while short brushes in a big, patterned 211 area assembled into several teepees. In addition, when the distance between two adjacent patterned features were comparable to polymer thickness, polymer brushes of adjacent features can interact with each other, thereby forming a bridge of polymer connecting the two adjacent features. We also observed a scaling behavior of patterned brushes where bigger patterned features contained thicker brushes. Furthermore, PNIPAM brushes were grown from unpatterned areas via SI-ATRP to form binary patterned rod-coil brushes. The presence of PNIPAM brushes can obstruct the interaction of PBLG brushes of adjacent features, hence reducing polymer bridges. In general, by using nanopatterned PBLG brushes, the size and spacing of teepee-like rod bundles can be precisely controlled at nanoscale. Since there were primary amine functional groups at the rod chain ends and the rods are vertically aligned in the rod bundles, we expect that all the amine groups are located at the tip of the rod bundles. Therefore, we should functionalize the tips with targeted molecules. To evaluate the possibility of tip modification, we should label the tips with heavy elements such as phosphorus, bromine or iodine and subsequently use secondary ion mass spectrometry (SIMS) for depth profile information. With the ability of tip modification, we should further use the tips as attachment points for cell membrane lipid mimics. Then the system can serve as a platform to study cell membranes with potential applications such as biosensing or bioseparation. Moreover, although in our work, the center-to-center spacing is 200 nm and the pattern size is 65 nm, we should definitely improve the fabrication conditions to further reduce the spacing and the pattern size while coil brush can block unwanted interactions of rod brushes of adjacent features. Coil brush such as PNIPAM can also prevent non-specific adsorption on surface. Then the rich surface topography of nanopatterning brushes can be used for molecular recognition or antibody bindings. 212 7.4 Mixed MJLCP rod- fluorinated coil brushes In Chapter 5, we investigated Mesogen-Jacketed Liquid Crystalline polymer (MJLCP) as a new type of rod brushes. Although this class of polymer has a lower persistence length than that of polypeptides, it is still significantly higher than that of coil brushes. A five-step synthesis of monomer with low yield was probably a roadblock for any study of MJCLP brush. In this work, we redesigned the monomer synthesis to two-step reaction to solve this bottle-neck issue. Then mixed MJLCP rod-coil brushes were prepared by growing PBBOS and PTFEMA brushes via SI-NMP and SI-ATRP, respectively. After the solvent quenching, while PBBOS homopolymer brushes looked flat, the mixed brushes displayed teepee nanostructures similar to that of PBLG brushes. It suggested that coil brushes through phase separation due to high immiscibility promoted the vertical alignment and aggregation of rod brushes. We also demonstrated that the relative thickness of the two polymers determined surface topography. By keeping PBBOS brush thickness constant and varying PTFEMA thickness, surface topography changed from teepee structures to nanoholes. We suspect that the transition occurred when coil brush surface changed from below rod-peak level to above it. Increasing the coil brush content also increased the number of rod nanostructures while decreasing the size. On the other hand, using the same coil brush thickness while increasing PBBOS thickness, the size of rod bundles became larger while reducing the number of rod bundles. Accordingly, all the changes in brush organization impacted the surface topography and properties. 213 Scheme 7.4. Monomers of MJCLP In this work, we only varied the polymer thicknesses while the grafting density remained the same. Thus, an interesting direction is to vary the compositions of the two polymers on a surface. We will simply change the concentrations of the two initiators during the deposition step. GISAXS, GIWAXS and neutron reflectivity can also reveal more valuable information about the orientation and organization of brushes in the system. However, as discussed above, mixed initiators may cause inhomogeneous distribution of the two polymers, hence disrupting a long-range order. Thus, a binary initiator which contains ATRP and NMP sites (Scheme 7.3) is worth trying to obtain a higher uniformity in size and spacing of rod bundles. Given the number of MJLCPs that have been reported (Scheme 7.4), future work should consider new MJCLP brushes beyond PBBOS so that we will have other rod brushes with different structure and functional groups. Another advantage of the MJLCP is that the rod diameter can be tuned by tailoring the mesogenic structures such as aliphatic tail length or the number of phenyl rings in the rigid core. We suspect that the rod diameter will have a great 214 impact on the organization and conformation of rod brushes in mixed brushes. Another direction is mixed rod-rod brushes which are a completely new system. Rod-rod block copolymer was reported with interesting structures such as hexagons in lamella or hexagons in cylinders.6 As SI-ATRP, SI-NMP and SI-ROP are compatible methods, it is feasible to prepare mixed MJLCP- polypeptide brushes which may result in new nanostructures. 7.5 Surface-initiated Cu (0)- Control Radical Polymerization using Cu tape and reducing agents In Chapter 6, we introduce a new polymerization method developed from Cu (0)- Control Radical Polymerization. By using Cu tape as Cu (0) source and a reducing agent, a deoxygenation step is not needed. While both hydrazine and ascorbic acid can be used as reducing agent, hydrazine was more effective for higher growth rate. Moreover, the simple reaction setup in this method allowed the preparation of uniform polymer brushes on wafer-size sample. Thus, this method provides the opportunity for polymer brushes to be used in industrial applications. The method is versatile enough to grow hydrophilic, styrenic, or fluorinated polymer brushes. The concentration of reducing agent was also optimized to get thick and homogenous brushes. The living characteristic of this method was demonstrated by the preparation of diblock and triblock copolymer brushes. By simply tilting the Cu substate with respect to initiator- immobilized sample, gradient brushes were produced. To improve the feasibility and scalability of this method, aqueous solution should be explored to replace organic solvents for the polymerization. Other parameters such as ligand concentration, types of ligands, concentration of halide ions are also worth investigating to further improve the polymerization method. 215 7.6 References 1. Yen, C.-C., Tokita, M., Park, B., Takezoe, H. & Watanabe, J. Spontaneous Organization of Helical Polypeptide Molecules into Polar Packing Structure. Macromolecules 39, 1313–1315 (2006). 2. Hwang, Y. et al. Piezoelectric properties of polypeptide-PMMA molecular composites fabricated by contact charging. Polymer 52, 2723–2728 (2011). 3. Luijten, J., Vorenkamp, E. J. & Schouten, A. J. Reversible Helix Sense Inversion in Surface-Grafted Poly(β-phenethyl- L -aspartate) Films. Langmuir 23, 10772–10778 (2007). 4. Yang, C.-T., Wang, Y., Frank, C. W. & Chang, Y.-C. Chemoresponsive surface- tethered polypeptide brushes based on switchable secondary conformations. RSC Adv 5, 86113–86119 (2015). 5. Wang, Y. & Chang, Y. C. Synthesis and Conformational Transition of Surface- Tethered Polypeptide:  Poly(l-glutamic acid). Macromolecules 36, 6503–6510 (2003). 6. Zhou, Q.-H. et al. Synthesis and Hierarchical Self-Assembly of Rod−Rod Block Copolymers via Click Chemistry between Mesogen-Jacketed Liquid Crystalline Polymers and Helical Polypeptides. Macromolecules 43, 5637–5646 (2010). 216