University of Florida Thesis Or Dissertation Formatting

PRECISE POLYMER END GROUPS, AQUEOUS POLYMERIZATIONS, AND MACROMOLECULAR NANO-AGGREGATES

CHARLES ADRIAN FIGG

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2018

To my family and my friends, old and new.

ACKNOWLEDGMENTS

First, I would like to thank my family. My mother instilled scientific curiosity early in my life and has been my biggest supporter. I am constantly inspired by her resilience and ambition. I appreciate my sister as one of my best friends and for always pushing me to catch up with her successes, and my father and step-father for being beacons of reason and mindfulness. I am grateful to my grandpa, Charles Figg, and my grandmother, Juanita Figg, for teaching me the importance of civil duty and teaching.

They fought through adversity to desegregate the public school system and ensure every child in central New Jersey got the education they deserved. Their persistence and unyielding belief in the pursuit and realization of educational goals has guided me though my academic career.

I would like to thank Brent Sumerlin for being the mentor I desired and needed.

He gave me the freedom to discover my research voice, make mistakes, and learn from my mistakes, while always being insightful and guiding. Brent has taught me not only what it means to be a successful researcher, but a good community member and a patient leader. As I transition from a student to a colleague, I could not be more excited to see what our future collaborations will bring.

I also would like to thank my committee, Prof. Daniel A Savin, Prof. Ronald K.

Castellano, Prof. Stephen A. Miller, and Prof. Blanka Sharma, for their support during my graduate career.

Next, I would like to thank my past academic mentors. I began my science career taking readings on the radioactivity of a parking lot for Dr. Ernst Esch at Los Alamos

National Lab. He showed unyielding support of my interests, regardless of if they aligned with his. Prof. Bert Meijer for telling me “it does not matter if my group does

what you are interested in, pursue what you want, and I will support you,” and letting me research with Prof. Patricia Dankers because of my interest in biomaterials. Prof. Craig

Hawker for letting me join his lab, supporting me whenever I needed life advice, and having me work for one of the most influential mentors I have had, Prof. Brett Fors. Brett taught me the virtue in working hard to achieve and discover and the virtue of investing in the success of others. Prof. Cyrille Boyer for letting me have an Aussie winter holiday in his lab to learn PET-RAFT. Cyrille’s ambition and support for his students have further confirmed my career goals, while his support for me has endured far beyond the ten weeks I worked for him.

I could not have done this without the support of my friends. I thank my friends I have known my whole life, Brett, Michelle, and Ariana, for being there to commiserate about our educational goals, especially as we all pursue advanced degrees. I thank my college friends, specifically Michael, Trevor, John, and Ryan, for always planning a

Vegas or Napa trip when I needed it. My stammtisch group Nick, Kyle and Georg for the monthly escape of graduate life into the dungeon of Hogan’s.

I am grateful to the members of the Sumerlin Group, past and present, for the constant pillar of support and creativity. Specifically, I would like to thank the people I have collaborated with or who have provided immense assistance in my research: Nick,

Tomo, Georg, Megan, Jacob, Becky, Soma, Jimmy, John, and Bryan, I would also like to thank my many collaborators for their assistance in completing our research goals:

Prof. Daniel Savin, Prof. Ronald Castellano, Prof. Cyrille Boyer, Prof. Zesheng An, Prof.

Nathan Gianneschi, Prof. David Haddleton, Dr. Alexandre Simula, Kalkidan Gebre, Dr.

Kyle Bentz, Dr. Sivaprakash Shanmugam, Dr. Ashton Bartley, and Molly Touve.

Finally, I would like to thank Heather Drew. She inspires me every day, and I could not imagine anyone else to chase dreams with.

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ...... 4

LIST OF TABLES ...... 12

LIST OF FIGURES ...... 13

LIST OF ABBREVIATIONS ...... 17

ABSTRACT ...... 20

CHAPTER

1 BACKGROUND ...... 22

1.1 Precise Bond Formation ...... 23 1.2 Precise Polymer Synthesis ...... 23 1.2.1 Catalyst-Free RAFT Photopolymerizations ...... 24 1.2.2 Externally-Initiated RAFT Photopolymerizations ...... 26 1.2.3 Photoinduced Electron/Energy Transfer (PET) RAFT Polymerizations .... 28 1.3 Precise Nanoscale Interactions ...... 31

2 RESEARCH OBJECTIVE ...... 35

3 EFFICIENT AND CHEMOSELECTIVE SYNTHESIS OF OMEGA, OMEGA- HETERODIFUNCTIONAL POLYMERS...... 38

3.1 Introduction ...... 38 3.2 Results and Discussion ...... 40 3.3 Conclusion ...... 47 3.4 Materials and Methods ...... 47 3.4.1 Materials ...... 47 3.4.2 Characterization ...... 48 3.4.3 End-Group Functionalization of mPEG to Dichlorotriazine-mPEG ...... 48 3.4.4 2-Furylmethanethiol Conjugation to 2-mPEG-4,6-dichloro-1,3,5- triazine ...... 49 3.4.5 4-Benzylpiperidine Conjugation to 2-mPEG-4-(2-furylmethanethiol)-6- chloro-1,3,5-triazine ...... 49 3.4.6 Phenylmethanethiol Conjugation to 2-mPEG-4-(2-furylmethanethiol)- 6-chloro-1,3,5-triazine ...... 49 3.4.7 1-Phenylmethanamine Conjugation to 2-mPEG-4-(2- furylmethanethiol)-6-chloro-1,3,5-triazine ...... 50 3.4.8 Phenylmethanol Conjugation to 2-mPEG-4-(2-furylmethanethiol)-6- chloro-1,3,5-triazine ...... 50

3.4.9 1-(2-Furyl)methylamine Conjugation to 2-mPEG-4,6-dichloro-1,3,5- triazine ...... 50 3.4.10 4-Benzylpiperidine Conjugation to 2-mPEG-4-[1-(2- furyl)methylamine]-6-chloro-1,3,5-triazine ...... 51 3.4.11 Phenylmethanethiol Conjugation to 2-mPEG-4-[1-(2- furyl)methylamine]-6-chloro-1,3,5-triazine ...... 51 3.4.12 1-Phenylmethanamine Conjugation to 2-mPEG-4-[1-(2- furyl)methylamine]-6-chloro-1,3,5-triazine ...... 51 3.4.13 Phenylmethanol Conjugation to 2-mPEG-4-[1-(2-furyl)methylamine]- 6-chloro-1,3,5-triazine ...... 52 3.4.14 2-Furanmethanol Conjugation to 2-mPEG-4,6-dichloro-1,3,5-triazine.... 52

4 MILD AND EFFICIENT SYNTHESIS OF OMEGA, OMEGA- HETERODIFUNCTIONALIZED POLYMERS AND POLYMER BIOCONJUGATES ...... 53

4.1 Introduction ...... 53 4.2 Results and Discussion ...... 55 4.2.1 Conjugating BTF to mPEG-amine ...... 55 4.2.2 Synthesizing Omega, Omega-Homodifunctional Polymers ...... 55 4.2.3 Synthesizing Omega, Omega-Heterodifunctional Polymers ...... 57 4.2.4 Conjugating Omega, Omega-Heterodifunctional Polymers to Avidin ...... 60 4.3 Conclusions ...... 61 4.4 Materials and Methods ...... 62 4.4.1 Materials ...... 62 4.4.2 Characterization ...... 62 4.4.3 Synthesis of Benzotrifuranone (BTF) ...... 63 4.4.4 Synthesis of mPEG-Amine ...... 66 4.4.5 Synthesis of BTF Conjugates ...... 66 4.4.5.1 mPEG-amine conjugation to BTF ...... 66 4.4.5.2 One-pot m-PEG-amine conjugation to BTF followed by diallyl- amine addition conjugate ...... 67 4.4.5.3 Synthesis of 2-(4,6-Dihydroxy-2-oxo-5-(2-oxo-2-(prop-2-yn-1- ylamino)ethyl)-2,3-dihydrobenzofuran-7-yl)-N-heptylacetamide...... 67 4.4.5.4 mPEG-amine conjugation to BTF conjugate 4 ...... 68 4.4.5.5 Synthesis of N-Allyl-2-(5-(2-(heptylamino)-2-oxoethyl)-4,6- dihydroxy-2-oxo-2,3-dihydrobenzofuran-7-yl)acetamide (6) ...... 68 4.4.5.6 m-PEG amine conjugation to BTF conjugate 6 ...... 69 4.4.5.7 Synthesis of 2-(4,6-Dihydroxy-2-oxo-7-(2-oxo-2-(prop-2-yn-1- ylamino)ethyl)-2,3-dihydrobenzofuran-5-yl)-N-(oct-7-en-1- yl)acetamide (8) ...... 69 4.4.5.8 m-PEG amine conjugation to BTF conjugate 8 ...... 70 4.4.5.9 Copper-mediated azide-alkyne cycloaddition of mPEG-BTF conjugate 9 and azido coumarin ...... 70 4.4.5.10 Thiol-ene of polymer 10 with 2-mercaptoethylbiotin ...... 70 4.4.6 Avidin Conjugations ...... 71 4.4.6.1 Purification of 11 using avidin agarose gel beads ...... 71

4.4.6.2 Polymer conjugation to avidin ...... 71

5 COLOR-CODING VISIBLE LIGHT POLYMERIZATIONS TO ELUCIDATE THE ACTIVATION OF TRITHIOCARBONATES USING EOSIN Y...... 72

5.1 Introduction ...... 72 5.2 Results and Discussion ...... 73 5.2.1 Trapping Studies under Blue-Light Irradiation ...... 73 5.2.2 Trapping Studies under Green-Light Irradiation ...... 75 5.2.3 Polymerizations under Blue-Light Irradiation ...... 77 5.2.4 Polymerizations under Green-Light Irradiation ...... 81 5.3 Conclusions ...... 87 5.4 Materials and Methods ...... 87 5.4.1 Materials ...... 87 5.4.2 Characterization ...... 88 5.4.3 Procedures ...... 88 5.4.3.1 Solution pH preparation ...... 88 5.4.3.2 Example polymerization procedure (reductive PET-RAFT under blue-light irradiation) ...... 89 5.4.3.3 Example procedure for trapping studies (reductive PET-RAFT under blue-light irradiation) ...... 89 5.4.3.4 DMA chain extension (oxidative PET-RAFT under green-light irradiation) ...... 89

6 POLYMERIZATION-INDUCED THERMAL SELF-ASSEMBLY ...... 91

6.1 Introduction ...... 91 6.2 Results and Discussion ...... 93 6.3 Conclusions ...... 100 6.4 Materials and Methods ...... 101 6.4.1 Materials ...... 101 6.4.2 Characterization ...... 102 6.4.3 Procedures ...... 103 6.4.3.1 Synthesis of PDMA-AA macro chain transfer agent ...... 103 6.4.3.2 Synthesis of nanoparticles with a composition of PDMA34-b- P(DMA14-co-AA6)-b-PNIPAm26 ...... 104 6.4.3.3 Synthesis of nanoparticles with a composition of PDMA34-b- P(DMA14-co-AA6)-b-PNIPAm38 ...... 105 6.4.3.4 Synthesis of nanoparticles with a composition of PDMA34-b- P(DMA14-co-AA6)-b-PNIPAm49 ...... 106 6.4.3.5 Synthesis of nanoparticles with a composition of PDMA34-b- P(DMA14-co-AA6)-b-PNIPAm61 ...... 107 6.4.3.6 Synthesis of nanoparticles with a composition of PDMA34-b- P(DMA14-co-AA6)-b-PNIPAm73 ...... 108 6.4.3.7 Synthesis of nanoparticles with a composition of PDMA34-b- P(DMA14-co-AA6)-b-PNIPAm98 ...... 109

6.4.3.8 Synthesis of nanoparticles with a composition of PDMA34-b- P(DMA14-co-AA6)-b-PNIPAm121 ...... 110 6.4.3.9 Kinetics of PNIPAM chain-extension ...... 111

7 TUNING HYDROPHOBICITY TO PROGRAM BLOCK COPOLYMER ASSEMBLIES FROM THE INSIDE OUT ...... 113

7.1 Introduction ...... 113 7.2 Results and Discussion ...... 116 7.3 Conclusions ...... 127 7.4 Materials and Methods ...... 129 7.4.1 Materials ...... 129 7.4.2 Characterization ...... 129 7.4.3 Procedures for Kinetics ...... 132 7.4.3.1 Kinetics of poly(N,N-dimethylacrylamide) (PDMA) macro chain transfer agent synthesis (macro-CTA 1)...... 132 7.4.3.2 Kinetics of 90% DAAm, 10% DMA polymerization kinetics ...... 133 7.4.3.3 Kinetics of 85% DAAm, 15% DMA polymerization kinetics ...... 133 7.4.3.4 Kinetics of 80% DAAm, 20% DMA polymerization kinetics ...... 133 7.4.3.5 Kinetics of 75% DAAm, 25% DMA polymerization kinetics ...... 134 7.4.3.6 Reactivity ratios ...... 134 7.4.4 Procedures for Nanoparticle Synthesis ...... 135 7.4.4.1 PDMA macro chain transfer agent synthesis (macro-CTA 2) ...... 135 7.4.4.2 90% DAAm, 10% DMA DP2 = 54 polymerization ...... 135 7.4.4.3 90% DAAm, 10% DMA DP2 = 87 polymerization ...... 136 7.4.4.4 90% DAAm, 10% DMA DP2 = 141 polymerization ...... 137 7.4.4.5 90% DAAm, 10% DMA DP2 = 217 polymerization ...... 137 7.4.5 SEC-MALS discussion ...... 138 7.4.6 Worm length determination...... 142

8 FINAL REMARKS ...... 144

APPENDIX

A IDEAL OMEGA,OMEGA-HETERODIFUNCTIONALIZATION PATHWAY ...... 145

B FUNCTIONALIZATION OF COUMARIN-AZIDE AND BIOTIN TO PEG CHARACTERIZATION ...... 146

C ABSORPTION SPECTRA OF EOSIN Y AND TRITHIOCARBONATES ...... 147

D ADDITIONAL PET-RAFT POLYMERIZATION KINETIC DATA ...... 148

E POLYMERIZATION KINETICS OF N,N-DIMETHYLACRYLAMIDE AND DIACETONE ACRYLAMIDE ...... 151

F DYNAMIC LIGHT SCATTERING OF NANOPARTICLES DIRECTLY FOLLOWING POLYMERIZATION AND CROSSLINKING ...... 153

LIST OF REFERENCES ...... 155

BIOGRAPHICAL SKETCH ...... 175

LIST OF TABLES

Table page

3-1 Nucleophilic aromatic substitution conversions on monomethyl ether poly(ethylene glycol) triazine-based end groups using thiol, amine, and alcohol nucleophiles...... 42

5-1 Apparent rate of consumption of 2-(ethyl trithiocarbonate)propionic acid using different ratios of chain transfer agent:eosin Y:4-dimethylaminopyridine under blue light irradiation ...... 75

5-2 Results of photoinduced electron/energy-transfer reversible addition- fragmentation chain transfer polymerizations of N,N-dimethylacrylamide with a DMA to CTA ratio of 200:1 and a solution pH of 8.4-9.0...... 78

5-3 Results of Control Experiments of N,N-Dimethylacrylamide (5 M) with Different Irradiation Wavelengths and Reagents after 3 h of Irradiation...... 79

7-1 Characterization data of nanoparticle constituent unimers obtained using size exclusion chromatography equipped with a multi-angle laster light scattering detector...... 119

7-2 Complete nanoparticle profiles obtained using SEC-MALS...... 139

7-3 Light scattering studies of selected nanoparticles ...... 142

7-4 Worm length measurements from transmission electron microscopy images and calculated average worm length, worm length standard deviation, weighted worm length, and worm polydispersity (PDI)...... 143

LIST OF FIGURES

Figure page

1-1 Example of different length scales of polymer synthesis and where each chapter in this dissertation fits ...... 22

1-2 Initiator-free photopolymerizations where initiation occurs via carbon-sulfur bond photolysis and subsequent addition to monomer to achieve a degenerative chain-transfer equilibrium ...... 25

1-3 Externally-initiated RAFT photopolymerization mechanism analogous to a thermally-initiated RAFT polymerization ...... 27

1-4 Photo-induced electron transfer RAFT polymerization using a photocatalyst (PC) ...... 29

1-5 Proposed polymerization-induced self-assembly process proceeding through a dispersion-based mechanism...... 33

3-1 General approach to 2,4,6-trichloro-1,3,5-triazine (TCT) functionalization...... 39

3-2 Nucleophilic Substitutions to 2,4,6-trichloro-1,3,5-triazine using N,N- diisopropylethylamine (DIPEA) ...... 40

3-3 1H NMR spectra of monomethyl ether poly(ethylene glycol) and its derivatives ...... 41

3-4 Matrix assisted laser desorption-ionization time-of-flight mass spectra confirming the synthesis of omega, omega-heterodifunctionalization of monomethyl ether poly(ethylene glycol) ...... 44

3-5 Electrostatic potential maps from density functional theory calculations of 2,4,6-trichloro-1,3,5-triazine (TCT) substitution ...... 45

4-1 General Strategy to functionalize benzotrifuranone (BTF)...... 54

4-2 One-pot synthesis of omega, omega-homodifunctionalized polymer 3 beginning from benzotrifuranone (BTF) and monomethyl ether poly(ethylene glycol) amine (mPEG-amine) ...... 56

4-3 Monomethyl ether poly(ethylene glycol) amine (mPEG-amine) omega, omega-heterodifunctionalization using benzotrifuranone derivatives functionalized with various moieties ...... 58

4-4 Functionalization of monomethyl ether poly(ethylene glycol) omega, omega- heterodifunctionalized polymers (mPEG-BTF conjugate) and subsequent protein binding studies ...... 59

5-1 Possible visible-light induced RAFT polymerizations mechanisms ...... 74

5-2 Trapping studies of 2-(ethyl trithiocarbonate)propionic acid using different ratios of chain transfer agent:eosin Y:4-dimethylaminopyrdine ...... 76

5-3 Pseudo-first-order kinetics of photoinduced electron/energy-transfer reversible addition-fragmentation chain transfer polymerization ...... 79

5-4 Conversion, pseudo first-order kinetics, and experimental molecular weight versus theoretical molecular weight for photo electron-transfer reversible addition-fragmentation chain transfer polymerizations ...... 83

5-5 Chain-extension polymerizations of N,N-dimethylacrylamide under oxidative photo electron-transfer reversible addition-fragmentation chain transfer polymerization conditions ...... 86

6-1 Polymerization-induced thermal self-assembly scheme to achieve different polymer nanoparticle morphologies ...... 93

6-2 Synthesis of thermoresponsive block copolymers ...... 94

6-3 Molecular weight evolution during the polymerization-induced thermal self- assembly (PITSA)...... 95

6-4 Dynamic light scattering measurements of purified nanoparticles with varying poly(N-isopropylacrylamide) degrees of polymerization (PNIPAm DPn)...... 97

6-5 Transmission electron microscopy of the unpurified crosslinked polymer aggregates showing the progression of nanoparticle morphology ...... 98

6-6 Data from the PITSA chain extension of a N,N-dimethylacrylamide (DMA) and acrylic acid (AA) containing macro chain transfer agent ...... 100

7-2 Reversible addition-fragmentation chain transfer polymerization schemes for PISA ...... 115

7-3 Size exclusion chromatography characterization of self-assembled nanoparticles formed during polymerization-induced thermal self-assembly (after crosslinking) ...... 118

7-4 Transmission electron microscopy images of polymerization-induced thermal self-assembly synthesized nanoparticles with varying monomer feed ratios of diacetone acrylamide (DAAm) to N,N-dimethylacrylamide (DMA) ...... 121

7-5 Transmission electron microscopy images of crosslinked nanoparticles using a monomer feed ratio of 75:25 diacetone acrylamide to N,N- dimethylacrylamide (DAAm75) for the second block ...... 125

7-6 Morphology transitions according to core-block hydrophobicity ...... 126

7-7 Measured worm lengths and standard deviations from transmission electron microscopy (TEM) images with representative images of each worm at the measured second block degree of polymerization (DP2) ...... 128

A-1 Optimized reaction conditions limited unwanted mPEG byproducts, while taking full advantage of the reactivity of TCT...... 145

B-1 Matrix-assisted laser desorption-ionization time-of-flight mass spectrum of compounds 1, 9, and 10, confirming the presence of the dye-containing ,- heterodifunctionalized adduct...... 146

B-2 1H NMR spectrum of PEG functionalized with coumarin and biotin...... 146

C-1 Absorption spectra of eosin Y in water ...... 147

C-2 Absorption spectra of 2-(ethyl trithiocarbonate)propionic acid (ETPA) (ETPA, blue) and 2-(ethyl trithiocarbonate)-2-methylpropionic acid (CTA, red)...... 147

D-1 Molecular weight versus monomer conversion of photoinduced electron- transfer reversible addition-fragmentation chain transfer polymerizations ...... 148

D-2 Pseudo-first order kinetics plot for photo electron-transfer reversible addition- fragmentation chain transfer polymerizations varying the amount of 4- dimethylaminopyridine...... 148

D-3 Data for photoinduced electron-transfer reversible addition-fragmentation chain transfer polymerizations varying the amount of EY in basic conditions ....149

D-4 Data for photoinduced electron-transfer reversible addition-fragmentation chain transfer polymerizations varying the amount of EY in acidic conditions ...150

D-5 Electrospray ionization mass spectra of reductive PET-RAFT polymerization showing initiation from both the chain transfer agent and the reducing agent. ..150

E-1 Kinetics of a polymerization-induced thermal self-assembly polymerization using a monomer feed ratio of 90% diacetone acrylamide to 10% N,N- dimethylacrylamide ...... 151

E-2 Kinetics of a polymerization-induced thermal self-assembly polymerization using a monomer feed ratio of 85% diacetone acrylamide to 15% N,N- dimethylacrylamide ...... 151

E-3 Kinetics of a polymerization-induced thermal self-assembly polymerization using a monomer feed ratio of 80% diacetone acrylamide to 20% N,N- dimethylacrylamide ...... 151

E-4 Kinetics of a polymerization-induced thermal self-assembly polymerization using a monomer feed ratio of 75% diacetone acrylamide to 15% N,N- dimethylacrylamide ...... 152

F-1 Diacetone acrylamide:N,N-dimethylacrylamide monomer feed ratio of 90:10 with varying second block degrees of polymerization (DP2)...... 153

F-2 Diacetone acrylamide:N,N-dimethylacrylamide monomer feed ratio of 85:15 with varying second block degrees of polymerization (DP2)...... 153

F-3 Diacetone acrylamide:N,N-dimethylacrylamide monomer feed ratio of 80:20 with varying second block degrees of polymerization (DP2)...... 154

F-4 Diacetone acrylamide:N,N-dimethylacrylamide monomer feed ratio of 75:25 with varying second block degrees of polymerization (DP2)...... 154

LIST OF ABBREVIATIONS

AA Acrylic acid

ATRP Atom-transfer radical polymerization

BDF Benzodifuranone

BMF Benzomonofuranone

BTF Benzotrifuranone

C-S Carbon-sulfur

CTA Chain transfer agent

CuAAC Copper-mediated azide-alkyne cycloaddition

Ð or Ðm Molar mass dispersity

DAAm Diacetone acrylamide

DCM Dichloromethane

DCT Dichlorotriazine

DFT Density functional theory

Dh Hydrodynamic diameter

DIPEA Diisopropylethylamine

DLS Dynamic light scattering

DMA N,N-Dimethylacrylamide

DMAc N,N-Dimethylacetamide

DMAP 4-Dimethylaminopyridine

DP2 Second-block degree of polymerization

DPn Degree of polymerization

EDC 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide

EPHP N-ethylpiperidine hypophosphate

ETPA 2-(ethyl trithiocarbonate)propionic acid

LCST Lower critical solution temperature

Macro-CTA Macro chain transfer agent

MALDI-ToF MS Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry

MALS Multi-angle laser light scattering

MCT Monochlorotriazine

Mn Number-average molecular weight mPEG monomethyl ether poly(ethylene glycol) mPEG-amine monomethyl ether poly(ethylene glycol) amine

Mw Weight-average molecular weight

Mw, NP Weight-average molecular weight of nanoparticles

Mw, unimers Weight-average molecular weight of unimers

Nagg Aggregation number

NIPAM N-isopropylacrylamide

NMP Nitroxide-mediated polymerization

PBS Phosphate-buffered saline

PDMA Poly(N,N-dimethylacrylamide)

PEG Poly(ethylene glycol)

PEGA Poly(ethylene glycol) acrylate

PET Photoinduced electron/energy transfer

PET-RAFT Photoinduced electron/energy transfer reversible addition- fragmentation chain transfer

PC Photocatalyst

EY Eosin Y

PISA Polymerization-induced self-assembly

PITSA Polymerization-induced thermal self-assembly

PNIPAm Poly(N-isopropylacrylamide)

RAFT Reverisble addition-fragmentation chain transfer

RDRP Reversible-deactivation radical polymerization

Rg Radius of gyration

SEC Size-exclusion chromatography

TCT 2,4,6-Trichloro-1,3,5-triazine

TEM Transmission electron microscopy

THF Tetrahydrofuran

TPO (2,4,6-trimethylbenzoyl)diphenylphosphine oxide

TTC Trithiocarbonate

UCST Upper critical solution temperature

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

PRECISE POLYMER END GROUPS, AQUEOUS POLYMERIZATIONS, AND MACROMOLECULAR NANO-AGGREGATES

Charles Adrian Figg

May 2018

Chair: Brent Sumerlin Major: Chemistry

Polymer synthesis will be discussed on three length scales including precise bond formation, precise polymer synthesis, and precise nanoscale interactions.

Because of its commercial availability, well-defined nature, and ubiquity in biological applications, mPEG was chosen for omega, omega-heterodifunctional end-group modification strategies. First, two distinct functionalities at one chain end were achieved via consecutive and chemoselective nucleophilic aromatic substitution reactions on TCT with model thiols and amines. Next, omega, omega-heterodifunctionalized polymers and polymer bioconjugates under mild conditions using BTF are prepared. BTF enables introduction of differentially “clickable” functional groups (e.g., alkenes and alkynes) to mPEG-amine at ambient temperature using near-stoichiometric amounts of reagents.

Mechanistic investigations into aqueous visible-light RAFT polymerizations of acrylamides using eosin Y as a PET catalyst are then discussed. The photoinduced polymerization was found to be dependent upon the irradiation wavelength and reagents, where either reduction or oxidation of the PET catalyst leads to inherently different initiation and reversible-termination steps. Investigations into the role of PET catalyst, wavelength, and reducing agent demonstrated that precise polymers with

predictable molecular weights are best realized under an oxidative PET-RAFT mechanism.

PISA is a versatile technique to achieve a wide range of polymeric nanoparticle morphologies. First, the synthesis of block copolymers with a growing stimuli-responsive block to form various nanoparticle shapes is discussed and termed PITSA. A RAFT polymerization of NIPAm from a hydrophilic chain transfer agent was carried out in water above the cloud point of PNIPAm. Shown is that the resulting block copolymers form polymeric nanoparticles with a range of morphologies (e.g., micelles, worms, and vesicles) as a function of the PNIPAm block length.

Subsequently, how block copolymer hydrophobicity provides control over aggregate morphology in water and leads to control over the length of polymeric nanoparticle worms is discussed. Slight variations in a monomer feed ratio of DAAm and DMA dictated the block copolymer chain composition and were proposed to alter aggregate thermodynamics. Micelles, worms, and vesicles were synthesized and the highest level of control over worm elongation attained during a polymerization is reported, simply due to the polymer chain hydrophobicity.

CHAPTER 1 BACKGROUND

Polymer synthesis has broad applications, from modifying protein solution properties to lithography techniques decreasing the feature sizes of computer chips to making bullet proof materials. The eventual application relies on precise control over chemical transformations or polymer topologies over multiple length scales. For example, ultra-high molecular weight polyethylene used for bullet proof vests relies on chemical control over the catalyst activity for monomer insertion (precise bond formation) to result in linear chains (precise polymer synthesis) which can crystallize with other chains (precise nanoscale interactions) to yield the desired macroscopic properties. This synergistic interplay between synthesis and materials has elicited the pursuit of structurally complex polymeric materials with varying degrees of control over specific bond formation, polymer architecture, and nanoscale interactions. This dissertation will introduce synthesis techniques at each of these length scales to develop complex macromolecular materials (Figure 1-1).

Figure 1-1. Example of different length scales of polymer synthesis and where each chapter in this dissertation fits.

1.1 Precise Bond Formation

Modifications to specific bonds on polymer structures are generally used to impart functional handles or for conjugation to other materials. Polymer end-group modifications to synthesize telechelic polymers are particularly useful, as this route has shown effective in preparing polymer-protein conjugates,1 surface-grafted polymers,

(multi)block copolymers,2 star polymers,3 etc. Inefficient transformations on polymers can lead to functionally disperse mixtures of chains that are difficult to purify and demonstrate heterogeneous properties. Reliable synthetic approaches now exist to effect post-polymerization end-group modification,4-9 many involving “click” reactions

(e.g., cycloadditions,10-13 thiol-ene(-yne),14,15 and isocyanate ligation16) that offer excellent yields and chemoselectivity. Because these reactions display efficient functional group conjugation, the onerous purification processes often required during the synthesis of telechelic polymers is avoided.12,17-22 Additionally, the orthogonality and modularity of such transformations, in principle, makes them well-suited for the introduction of multiple different functionalities sequentially at single polymer chain ends. Even so, there are few examples23,24 of alpha, alpha- or omega, omega- heterodifunctional polymers prepared in this fashion, in part because required is pre- installation of two different reactive functional groups at a single polymer terminus.

Chapters 1 and 2 will focus on using symmetric trifunctional compounds to synthesize mPEG with two distinct functional groups at the omega-terminus.

1.2 Precise Polymer Synthesis

Visible light provides a low-energy irradiation source that has been exploited for a variety of reversible-deactivation radical polymerization (RDRP) techniques, including

reversible addition-fragmentation chain transfer (RAFT) polymerization and atom- transfer radical polymerization (ATRP).25,26 Photopolymerizations typically use either photochemically active initiators or electron/energy transfer catalysts to introduce the radicals needed for initiation. However, the majority of work in this area has involved polymerizations in organic solvents, with aqueous systems only recently having been considered. Aqueous systems allow synthesis in a green, benign solvent and provide significant advantages for polymerizations in biological contexts, which require low energy irradiation conditions to minimize or avoid damage to biomolecules (e.g., proteins, DNA, cells) that can occur during traditional UV-irradiated photopolymerizations. Moreover, aqueous polymerization conditions are also compatible with monomers/polymers that are not typically soluble in organic solvents

(e.g., ionic monomers and polyelectrolytes).27,28 The combination of low energy irradiation and benign aqueous conditions makes aqueous visible-light RAFT29 polymerizations ideal for many biological and materials applications.

1.2.1 Catalyst-Free RAFT Photopolymerizations

Most thiocarbonylthio (TCT) compounds absorb in the visible light region between 380-525 nm due to the n→* transition of the C=S group. This range of excitation wavelengths can be used to initiate polymerizations and mediate control via a photoiniferter mechanism.30 Initiation is proposed to occur via homolytic C-S bond cleavage to generate a carbon-centered radical capable of adding to monomer and a stable TCT radical that eventually deactivates growing chains via reversible termination

(Figure 1a). Polymerizations were initially reported using low irradiation intensity (4.8 W) and required longer reaction times to reach high monomer conversions (>12 h).31,32

Increasing light intensity (from 26 W to 208 W at  = 201 nm) has been shown to decrease reaction time without a significant loss in polymerization control that could

Figure 1-2. Initiator-free photopolymerizations where initiation occurs via carbon-sulfur bond photolysis and subsequent addition to monomer to achieve a degenerative chain-transfer equilibrium between propagating chains and reversible termination with the stable sulfur-centered radical (a). The polymerization rate can be increased using a tertiary amine (NR3) that is proposed to undergo a redox reaction with the excited-state thiocarbonylthio moiety (b). result from increased irreversible termination events.33 For example, when polymerizations were held at 17 C with 208 W of irradiation power, 95% monomer conversion was achieved in 20 min with good control over molecular weights and molecular weight distributions. However, maintaining the temperature at moderate levels was essential, as an analogous polymerization with the same irradiation intensity but without temperature control reached up to 80 C as a result of the polymerization exotherm and high irradiation intensity and resulted in a large increase in polymer molar

mass dispersity, most likely due to significant hydrolysis of the TCT at elevated temperatures.

Initiator-free photopolymerizations are an attractive route to well-defined polymers since every polymer chain is initiated from the R-group of the CTA and no additional reagents are needed. However, higher-energy violet or blue light is typically used to target the peak of the C=S n→* transition, which could be troublesome over long irradiation times or with high-intensity light sources. An alternative route to increase the rate of polymerization is via the addition of a tertiary amine.34 Tertiary amines are hypothesized to undergo a redox reaction with the excited state TCT to yield the TCT anion and a carbon-centered radical for initiation/propagation (Figure 1b). While adding tertiary amines can minimize reactions times to be more amenable for biological entities, high concentrations (i.e., stoichiometric amounts) of amine to CTA are often required.

1.2.2 Externally-Initiated RAFT Photopolymerizations

Analogous to conventional thermally-initiated RAFT polymerizations, visible-light aqueous polymerizations using an external initiating species that absorbs and fragments in the visible-light region are also possible (Figure 2). The water-soluble poly(ethylene glycol) acrylate (PEGA) was first investigated using (2,4,6- trimethylbenzoyl)diphenylphosphine oxide (TPO), a classic photoinitiator that has a visible-light absorbance up to  = 420 nm.35 PEGA was chosen as the monomer since it has been used extensively in polymer-protein engineering and shows a fast polymerization rate. An initialization period was observed for all polymerizations, an observation attributed to conversion of the CTA into oligomers prior to significant

propagation, but well-defined polymers were prepared with up to 80% conversion of

PEGA after 10 min. These results indicated that photolysis of the TPO initiator resulted in fast and well-controlled polymerizations at room temperature. The polymerization conditions were also amenable to other acrylic monomers at various pH ranges.36

However, the CTA was found to be more stable under acidic pH,37,38 while hydrolysis rates up to 3.3% were observed in solutions at pH 10.2. Interestingly, oxygen tolerance could be achieved for polymerizations initiated by TPO when the oxygen-scavenging enzyme glucose oxidase (GOx) and glucose were included.39

Figure 1-3. Externally-initiated RAFT photopolymerization mechanism analogous to a thermally-initiated RAFT polymerization.

Compared to initiator-free systems where every chain is initiated from the CTA R- group, the overall polymer composition synthesized by RAFT in the presence of an external initiator contains more chain-end heterogeneity. Although only a small fraction of initiator relative to CTA is introduced into the system, that same fraction of chains are expected to either not contain the CTA-derived R-group at the  chain end (due to the

initiating fragments) or the thiocarbonylthio at the  chain end (due to irreversible radical-radical coupling or chain transfer events).

1.2.3 Photoinduced Electron/Energy Transfer (PET) RAFT Polymerizations

Photoinduced electron/energy-transfer (PET) catalysts are widely used in organic synthesis to conduct single-electron redox reactions.40 Many photocatalysts (PCs) have been reported to efficiently mediate both ATRP41,42 and RAFT polymerization43,44 through a redox reaction with the halide or TCT, respectively. However, the most

2 prevalent PCs (tris[2-phenylpyridinato-C , N)iridium(II) (Ir(ppy)3), phenothiazine, etc.) are poorly soluble in aqueous systems, precluding their use for polymerizations in water. Tris(2,2-bipyridyl)ruthenium(II) chloride (Ru(bpy)3Cl2) was the first PC to be employed under aqueous conditions by Boyer and co-workers.45 The oxidative quenching pathway of the catalyst was used to activate the thiocarbonylthio RAFT compound and initiate RAFT polymerization of DMA (Figure 3a). This pathway worked exceptionally well in water, even at low catalyst loadings relative to CTA ([CTA] :

-4 [Ru(bpy)3Cl2] = 1 : 210 ). Additionally, this polymerization approach worked for other acrylamides, acrylates, and methacrylates. When switching the polymerization medium to fetal bovine serum (FBS), the resultant polymers showed a broad molecular weight

-4 distribution (Ð > 1.5) when the [CTA] : [Ru(bpy)3Cl2] ratio was 1 : 210 . Upon

-3 increasing the [CTA] : [Ru(bpy)3Cl2] ratio tenfold to 1 : 210 , much lower molar mass dispersities were obtained. Therefore, the FBS was proposed to interact with the catalyst to encumber polymerization when catalyst loading was too low. Polymerizations using Ru(bpy)3Cl2 in the presence of ascorbic acid as a reducing agent, inducing a reductive quenching mechanism of the excited-state PC (Figure 3b), reach 90%

Figure 1-4. Photo-induced electron transfer RAFT polymerization using a photocatalyst (PC) via either an oxidative catalyst pathway (a) or a reductive catalyst pathway (b, NR3 can also be ascorbic acid). conversion in 30 min.46 These polymerization rates were especially effective to polymerize activated ester monomers in aqueous solutions while maintaining >85% of their hydrolytically sensitive ester moieties. An inexpensive and less-toxic alternative to

Ru(bpy)3Cl2 is the water soluble zinc porphyrin (Zn(II) meso-tetra(4- sulfonatophenyl)porphyrin), which also works with a variety of monomer classes and can be activated or deactivated according to the pH of the solution.47,48 Indeed, metal- based PCs work exceptionally well and facilitate fast polymerization rates in aqueous systems using low catalyst loadings and allow irradiation wavelengths from violet to red; however, the potential toxicity of Ru or Zn catalysts is a concern when using metal- based PET-RAFT polymerizations.

Metal-free approaches to PET-RAFT polymerization have also been reported that employ inexpensive organic dyes as PCs to make polymerization conditions more

amenable to biological applications. Eosin Y (EY) was the first organic dye used in aqueous polymerizations, as described by Hawker and coworkers.49 In this report, a tertiary amine co-catalyst reduced the excited triplet state of EY to the radical anion, which then readily reduced TCT compounds to liberate the carbon-centered radical that initiated polymerization and led to activation/propagation later in the polymerization.

Since a reductive quenching mechanism was employed, oxygen tolerance was realized.

Later mechanistic studies of EY under both blue and green irradiation wavelengths provided more insight into the use of this organocatalyst for RAFT polymerization.50

While the reductive catalyst pathway is more commonly reported, an oxidative pathway

(i.e., polymerization in the absence of a tertiary amine) yields polymer molecular weights much closer to predicted molecular weights. The authors hypothesized that this results from the high concentration of radicals present in reducing conditions leading to reduced control during polymerization. The pH of the polymerization solutions also affected the apparent rate of polymerization, as at a lower pH, a slower kp, app was observed, which was attributed to the partial protonation/deactivation of the catalyst and/or tertiary amine reducing agent.

PET-RAFT initiation relies on the PC activating the CTA to induce C-S bond cleavage, generally resulting in every polymer chain being initiated from the R-group.

Oxidative quenching mechanisms of the PC show exceptional characteristics of RDPD.

A reductive-quenching catalyst initiation mechanism also introduces oxygen tolerance to the polymerization via concurrent reduction of molecularly dissolved oxygen to yield peroxides, but other radical initiation species (e.g., peroxides, amine radical cations) may be introduced into the system causing some discrepancies in kinetics and

molecular weight control. Overall, these characteristics yield well-defined polymer distributions with homogeneous end groups and bode well for biological applications, especially those that require low catalyst loadings, mild irradiation wavelengths, and aerobic polymerizations conditions.

1.3 Precise Nanoscale Interactions

Self-assembled amphiphilic polymer nanoparticles have garnered substantial attention as potential delivery vehicles due to high stability at low solution concentrations, molecular encapsulation capabilities, and biocompatability.51-53 Micelles are perhaps the most exploited self-assembled polymer morphology because of their facile preparation and the ease with which they can be loaded.54,55 However, other types of self-assembled block copolymer morphologies, such as cylindrical micelles

(i.e., worms and nanorods) and vesicles, can have various advantages over the simple micellar structure. For example, in many cases polymeric worms and nanorods have higher loading capacities than micelles, though their solution properties, such as in vivo cell absorption, are not completely understood.56 Vesicles, or polymersomes, are able to encapsulate molecules in both their hydrophobic vesicle membrane and their hydrophilic interiors, providing the possibility of dual-component delivery.57,58

Although self-assembled polymer nanoparticles show increasing viability as delivery vehicles, inefficient and laborious preparation techniques may limit their broad application. Typically, the desired morphology that results from block copolymer assembly is developed through an often tedious bottom up approach at low concentrations (<5 w/w% solids).59,60 This strategy can entail multiple iterations of polymer synthesis, purification, and self-assembly via either solvent switching or film rehydration. The aggregated morphology is dictated primarily by the interfacial curvature

within the aggregate and is a function of the packing parameter (p) described in

Equation 1-1:

푣 푝 = 1-1 푎푙 where v is the volume of the hydrophobic chain, a is the optimal area of the hydrophilic head group, and l is the length of the hydrophobic chain.61 A lower packing parameter (p

≤ ⅓) is indicative of high interfacial curvature and leads to spherical micelles, while worms and vesicles are observed as the packing parameter increases (⅓ < p < ½ and p

≤ ½, respectively).62 However, another factor that controls whether micelles, worms, vesicles, or higher order nanoparticles are formed is the method and finesse in which they are synthesized, processed, and assembled.63 This often leads to arduous experimental optimization, opportunities for errors through the multi-step preparation process, and difficulty scaling up nanoparticle yields.

. The in situ synthesis of self-assembled block copolymer nanostructures by chain extending miscible polymers with an immiscible second block64-66 is redefining the field of block copolymer self-assembly. Dispersion-67-80 and emulsion-based81-84 polymerization-induced self-assembly (PISA)85-87 by reversible-deactivation radical polymerization has been used to achieve diverse polymer aggregate morphologies

(Figure 1-5). During PISA, various nanoparticle morphologies are obtained by the gradual in situ rearrangement of growing solvophobic chains and interparticle collisions in relatively concentrated solutions (up to 50 w/w% solids). These collisions allow the growing solvophobes to rearrange into lower energy morphologies as the hydrophobic segments grow to occupy larger volumes relative to the hydrophilic stabilizing block.

This rearrangement leads to an increasing chain packing parameter and a morphological transition from micelles to worms to vesicles during polymerization.

Reversible-deactivation radical polymerization (RDRP) methods provide the necessary control over polymer molecular weight needed to induce the gradual evolution of well-defined nanostructures during polymerization.88-90 Nitroxide-mediated polymerization (NMP) and atom transfer radical polymerization (ATRP) have been successfully employed for PISA;80,83,91,92 however, the majority of PISA reports rely on reversible addition-fragmentation chain transfer (RAFT) polymerization, due to the mild reaction conditions and functional group tolerance.93 Thus far, purely (and permanently) solvophobic polymers have been prepared by PISA via RAFT, including

Figure 1-5. Proposed polymerization-induced self-assembly process proceeding through a dispersion-based mechanism.

the synthesis of polystyrene in alcohols,67,72,74,94,95 poly(2-methoxyethyl acrylate) in water,96 poly(2-hydroxypropyl methacrylate) (PHPMA) in water,69 and nucleobase- containing polymers in chloroform and 1,4-dioxane.97

CHAPTER 2 RESEARCH OBJECTIVE

The purpose of this research was to develop new synthetic approaches to yield well-defined and precise macromolecular structures spanning discrete bond formation to the aggregation of thousands of well-defined polymer chains into nanoparticles.

The research discussed herein will begin in Chapter 3 with specific bond formation chemistry to synthesis omega, omega-heterodifunctional polymers using TCT.

TCT undergoes subsequent nucleophilic aromatic substitution reactions at increasing temperatures. This temperature specificity provides a scaffold to install three distinct functional groups according to the nucleophiles used. Hydroxyl, amine, and thiol nucleophiles proved to be excellent reactants for conjugation, proving the breadth and modularity of this approach. Importantly, analysis of the triazine ring electronics revealed that a certain order-of-addition is required to install three different heteroatom based nucleophiles (i.e., hydroxyl nucleophiles should be added first, then thiols, then amines).

Our initial interest in heterodifunctional polymers resulted from a desire to install a fluorescent tag onto a polymer-protein conjugate. However, the requisite heating used for TCT functionalization (up to 80 C) would disrupt the secondary and tertiary structures of proteins. Therefore, Chapter 4 reports another approach to achieve omega, omega-heterodifunctional polymers using BTF. BTF facilitates sequential lactone ring-opening reactions with amines at low (-41 C) to room temperatures.

Amines were conjugated to mPEG-amine containing “click” functional handles, and then subsequent reactions to install a coumarin dye and biotin for avidin binding were performed. Upon incubation of avidin with the omega, omega-heterodifunctional

polymers, polymer-protein conjugates with a fluorescent tag at the site of avidin conjugation were successfully synthesized.

Our interest in polymer-protein conjugates and RDRP techniques initiated the pursuit of identifying polymerization conditions that effectively polymerized acroyl monomers in water using mild visible light irradiation. Since our goal was to eventually use biomaterials (e.g., proteins, DNA), we began studying ways to limit the use of toxic metal PET catalysts previously reported. EY proved to be a very efficient PC for the polymerization of acrylamides in water. Chapter 5 details mechanistic investigations into polymerizations catalyzed by EY. Importantly, we found that limiting background initiation from the CTA and commonly employed reducing agents (e.g., tertiary amines, ascorbic acid) led to a significant improvement on polymer control, namely ways to improve experimental molecular weights to agree with theoretical molecular weights.

Chapters 6 and 7 describe new routes to synthesize nanoparticles derived from polymer aggregates. PISA is used to first introduce thermoresponsive polymers as amenable monomer for the in situ nanoparticle synthesis technique (Chapter 6), then the hydrophobicity of the self-assembled block is systematically altered to understand how inherent hydrophobicity affects aggregate morphology progression. Our group has been heavily involved in thermoresponsive polymers, and employing them as core- forming blocks for PISA further demonstrates how temperature can be used to affect macromolecular interactions. Chapter 7 experiments began as an evaluation of a comonomer system that displayed thermoresponsiveness when employed for PISA.

Our initial goal was to synthesize similar nanoparticle architectures that could be loaded with drugs, but that displayed markedly different temperatures of release according to

the hydrophobicity of the core-forming block. To our surprise, nanoparticle morphology was dramatically affected by the comonomer ratio (i.e., change in core-forming block hydrophobicity), with just a 5% change leading to different morphologies. Therefore,

Chapter 7 describes the evaluation of varying comonomer feed ratios on morphological outcome during PISA and how well-defined nanoparticle architectures can be achieved through minimal monomer manipulation.

CHAPTER 3 EFFICIENT AND CHEMOSELECTIVE SYNTHESIS OF OMEGA, OMEGA- HETERODIFUNCTIONAL POLYMERS

3.1 Introduction

*This chapter details our investigations in using cyanuric chloride (2,4,6-trichloro-

1,3,5-triazine (TCT)) as a convenient means to synthesize semi-telechelic ω,ω- heterodifunctionalized polymers through the installation of two functional groups at the hydroxyl terminus of the commodity polymer monomethyl ether poly(ethylene glycol)

(mPEG). TCT is frequently used as an organic synthesis building block to access complex molecular architectures due to its efficient reactivity with a variety of nucleophiles.98 As the electrophilicity of TCT and its partially functionalized dichlorotriazine (DCT), and monochlorotriazine (MCT) analogs are markedly different, nucleophilic aromatic substitution of the three chlorine atoms on TCT generally requires higher temperatures and/or stronger nucleophiles for each consecutive step (Figure 3-1, a). This differential reactivity is attributed to the increased electron delocalization in the aromatic ring after substitution of each electron-withdrawing substituent (i.e., chlorine) with electron-donating groups from the nucleophile residue. Consequently, we recognized that TCT’s ability to undergo sequential nucleophilic aromatic substitutions provides an ideal strategy for synthesizing complex macromolecular architectures.

Simanek and coworkers have extensively exploited the chemoselective nature of

TCT to minimize the number of functional group modifications and circumvent the protecting group chemistry typically required during dendrimer synthesis.99-108 TCT has

Figure 3-1. General approach to 2,4,6-trichloro-1,3,5-triazine (TCT) functionalization. (a) Varying temperatures allow chemoselective nucleophilic functionalization. (b) Model nucleophiles employed for sequential TCT functionalization. also been conjugated to polymer end groups, but relatively minimal characterization and/or incomplete consumption of its reactive sites has hindered its wide-spread applicability.109-112 However, other synthetic approaches towards the end-group functionalization of mPEG have received considerable attention, primarily due to mPEG’s important role in providing hydrophilicity and biocompatibility in a variety of biomedical applications.113-115 Given that mPEG contains no sites amenable to functionalization along its backbone, efficient end-group derivatization is critical for preparing the polymers needed for many of these applications. Traditional routes for mPEG end-group modification can require tedious multi-step syntheses and typically install only one reactive site per chain end.116-119 Additionally, there are few reports on the efficient ω,ω-heterodifunctionalization of mPEG.15,23,120

Therefore, we sought to establish a straightforward and methodical TCT conjugation approach to construct omega, omega-heterodifunctional mPEG, while generally investigating TCT functionalization more broadly as a powerful route to achieving specific polymeric end-group complexity. Herein, we report the efficient transformation of commercially available mPEG to omega, omega- heterodifunctionalized mPEG conjugates by chemoselective reactions with model thiol,

amine, and alcohol nucleophiles (Figure 3-1, b). The reaction conditions we identified resulted in the desired selective nucleophilic substitutions and avoided the variety of possible byproducts that could result from over/under consumption of reactive sites during each consecutive step (Figure A-1). In all cases, end-group functionalization was confirmed via 1H NMR spectroscopy and matrix assisted laser desorption-ionization time-of-flight mass spectrometry (MALDI-ToF MS). Moreover, using density functional theory (DFT) calculations, electrostatic potential maps were constructed to provide insight into the electronic structure and differential reactivity of the triazine ring of TCT and DCT.

Figure 3-2. Nucleophilic Substitutions to 2,4,6-trichloro-1,3,5-triazine using N,N- diisopropylethylamine (DIPEA), Yielding an omega, omega-heterodifunctional monomethyl ether poly(ethylene glycol) conjugate.

3.2 Results and Discussion

Amines are commonly used for TCT functionalization due to their high nucleophilicity and the resulting delocalization of the amine lone pair electrons after substitution. Previous model nucleophile/TCT reactivity studies have suggested that the reaction of a primary amine followed by a more reactive cyclic secondary amine can result in quantitative, consecutive nucleophilic substitutions.99 However, we sought to investigate the potential of using TCT as a facile strategy to orthogonally and efficiently

install two functional groups on a hydroxyl terminated polymer (i.e., the omega-terminus of 2000 g/mol mPEG). When mPEG (1) was treated with 2.5 equivalents of TCT at room temperature for 16 h, the hydroxyl group was quantitatively transformed to the

DCT derivative (i.e., 2-mPEG-4,6-dichloro-1,3,5-triazine (2)) (Figure 3-2). Product conversion of 1 to 2 was monitored using 1H NMR spectroscopy by the appearance of peaks attributed to the –CH2-O-triazine protons at delta = 4.65 ppm as shown in Figure

3-3. To investigate the nucleophilic scope of this approach for PEG functionalization, model thiol, amine, and alcohol nucleophiles were reacted with the DCT-mPEG derivative (2) and various MCT derivatives (Table 3-1).

Figure 3-3. 1H NMR spectra of monomethyl ether poly(ethylene glycol) (1, blue) and its derivatives after reaction with 2,4,6-trichloro-1,3,5-triazine (2, black), 2- furylmethanethiol (3, green), and 1-phenylmethanamine (4, red).

Table 3-1. Nucleophilic aromatic substitution conversions on monomethyl ether poly(ethylene glycol) triazine-based end groups using thiol, amine, and alcohol nucleophiles. Entry Electrophile Nucleophile Temp. (C) Conv.a (%) Cl HS N b 1 O O 0 – 25 >95 H3C 45O N N Cl Cl N HN 2 O 80 >95 H3C 45O N N S O

Cl HS N 3 O 80 16 H3C 45O N N S O

Cl H2N N 4 O 80 92 H3C 45O N N S O

Cl HO N 5 O 80 <5 H3C 45O N N S O

Cl H2N N b 6 O O 0 – 25 >95 H3C 45O N N Cl Cl N 7 O 80 >95 H3C 45O N N HN O

Cl HS N 8 O 80 24 H3C 45O N N HN O

Cl H2N N 9 O 80 >95 H3C 45O N N HN O

Cl HO N 10 O 80 <5 H3C 45O N N HN O

Cl HO N 11 O O 100 8 H3C 45O N N Cl aAfter 16 h using 1.0 equivalent of nuclophile to polymer, conversion determined by 1H NMR spectroscopy. bReactions held at 0 C for 3 h, then warmed to 25 C and stirred for 13 h.

The second nucleophilic substitution on the DCT ring of 2 with the model thiol 2- furylmethanethiol required a temperature where efficient conversion to 2-mPEG-4-(2- furylmethanethiol)-6-chloro-1,3,5-triazine (3) could be achieved without under or over consumption of the nucleophile. We observed that when a solution of 2 in dichloromethane (DCM) was treated with 1.00 equivalent of 2-furylmethanethiol at 0 °C

for 3 h and allowed to stir at room temperature for another 13 h, near-quantitative substitution of the second chlorine with the thiol occurred. Conversion was determined using 1H NMR spectroscopy by observing the appearance of peaks at delta = 4.45 ppm, attributed to the furfuyl methylene protons, and delta = 6.43 and 7.34 ppm, attributed to the aromatic furyl protons. For the final substitution on the MCT, 1.00 equivalent of 1- phenylmethanamine was added to 3 and allowed to react at 80 °C for 16 h to achieve 2- mPEG-4-(2-furylmethanethiol)-6-(1-phenylmethanamine)-1,3,5-triazine (4). 1H NMR spectroscopy of the products indicated a conversion of 92%, as evidenced by the appearance of new peaks attributed to the benzyl methylene protons at delta = 4.63–

4.67 ppm and the aromatic phenyl protons at delta = 7.29–7.38 ppm. Notably, the proton signals of the mPEG conjugate 4 suggested the presence of conformational isomers, which can likely be attributed to restricted nitrogen−triazine bond rotation due to partial double bond character.121 The presence of conformational isomers, which was also observed for other amine-TCT adducts, further supports the stabilizing effect of nitrogen’s considerable electron delocalization into the triazine ring.

Although 1H NMR spectroscopy showed the successful functionalization of the mPEG conjugate 2 with 2-furylmethanethiol and 1-phenylmethanamine, these data alone did not provide insight into the exact end-group composition. Therefore MALDI-

ToF MS was used to confirm that predominantly omega, omega-heterodifuntionalized mPEG was present (Figure 3-4). The MALDI-ToF MS spectrum of 2 showed a main distribution attributed to the 2 + Na+ adduct with an increased mass of 146.9 g/mol.

Minor distributions of 2 + K+ and an unidentifiable ionization degradation product were also visible. Single addition of 2-furylmethanethiol was confirmed via a mass increase of

77.99 g/mol, a major 3 + Na+ distribution, and a minor 3 + K+ distribution. Addition of 1- phenylmethanamine resulted in a mass increase of 71.10 g/mol with major 4 + Na+ and minor 4 + K+ distributions visible, confirming mPEG omega, omega- heterodifunctionalization. Importantly, no residual byproducts from inefficient nucleophilic substitution were observed throughout the MALDI-ToF MS characterization.

Figure 3-4. Matrix assisted laser desorption-ionization time-of-flight mass spectra confirming the synthesis of omega, omega-heterodifunctionalization of monomethyl ether poly(ethylene glycol) via incorporation 2,4,6-trichloro-1,3,5- triazine and subsequent reactions with model nucleophiles.

Throughout our studies, we realized the importance of considering the order in which different nucleophiles were added to the triazine ring. To elucidate the effect of triazine electronics during different nucleophile additions, DFT calculations using analogous amine, alcohol, and thiol substituents were performed on model DCT and

MCT compounds (Figure 3-5, B3LP 6-311+G** basis set). The electrostatic potential maps revealed that upon substitution of the first chlorine on TCT with an alcohol

nucleophile (i.e., mPEG, Figure 3-5, a), the resulting DCT remains relatively electron deficient because of the high electronegativity of oxygen. This observation is consistent with the low temperatures required to conjugate thiol (Table 3-1, entry 1) and amine

(Table 3-1, entry 6) nucleophiles to 2 (i.e., a DCT compound).

Figure 3-5. Electrostatic potential maps from density functional theory calculations of 2,4,6-trichloro-1,3,5-triazine (TCT) substituted with (a) a model alcohol (i.e., an analog of monomethyl ether poly(ethylene glycol) (mPEG)), followed by model (b) thiol, (c) amine, and (d) alcohol substituents (B3LYP 6-311+G** basis set).

Although a thiol nucleophile reacted near quantitatively with 2, subsequent addition of another thiol to the resulting MCT was inefficient (Table 3-1, entry 3), despite the relatively low electron density levels on the triazine ring observed in the electrostatic potential maps for the analogous model MCT with oxygen and sulfur substituents

(Figure 3-5, b). Therefore, even though sulfur has a relatively low electronegativity, its larger size likely limits orbital overlap and pi-orbital delocalization within the triazine ring, resulting in an electron-deficient MCT that was deactivated towards the addition of thiols as a third nucleophile.

Conjugating two amine nucleophiles to 2 required the same experimental reaction conditions as shown in Scheme 1, where each nucleophilic aromatic substitution resulted in >95% conversion, as determined by 1H NMR spectroscopy

(Table 3-1, entries 7 and 9). MALDI-ToF MS further confirmed the singular presence of the MCT-mPEG intermediate and omega, omega-heterodifunctionalized mPEG products. The electrostatic potential map of a model MCT compound containing oxygen and nitrogen substituents showed increased electron density within the ring, suggesting significant π-orbital delocalization of the nitrogen substituent (Figure 3-5, c). The stabilization that results from nitrogen substituents may deactivate the MCT-mPEG adducts towards addition of poorly stabilizing nucleophiles, such as thiols (e.g., phenylmethanethiol, Table 3-1, entry 8), while allowing amine nucleophiles to be good third nucleophiles regardless of the second nucleophile (Table 3-1, entries 2, 4, 7 and

9).

We were unable to conjugate additional alcohol nucleophiles to any triazine product that resulted after the initial TCT-mPEG conjugation (Table 3-1, entries 5, 10, and 11). The electrostatic potential map for a model MCT compound with two oxygen substituents suggests a highly electron deficient triazine ring would result in this case.

The electronic deficiency is believed to be from the high electronegativity of oxygen, which results in low π-orbital delocalization and high sigma–bond electron withdrawal.

Therefore, the electronic instability of the triazine ring that would result after substituting two chlorines with two alcohol nucleophiles appears too high to overcome during mPEG functionalizations.

These experimental and theoretical results provide insight into a preferred nucleophilic substitution order for TCT. Namely, to ensure efficient functionalization, the order of substitution should be (1) alcohol, (2) thiol, and (3) amine when these three different nucleophiles are required for polymer end-group functionalization. It should be emphasized that these experiments were conducted using a 1:1 ratio of nucleophile to the DCT-mPEG conjugate and that manipulating stoichiometry of less active nucleophiles would presumably lead to increased conversions of nucleophilic substitution reactions.

3.3 Conclusion

In summary, hydroxyl terminated polymers were shown to be successfully ω,ω- heterodifunctionalized employing the readily available compound TCT. Given the wide variety of available nucleophiles and established utilization of mPEG in numerous biological applications (e.g., protein conjugation), end-group modification using TCT provides unprecedented access to semi-telechelic heterodifunctional architectures.

Moreover, the facile ability of TCT to undergo sequential nucleophilic aromatic substitution reactions suggests it is a promising and versatile tool for polymer functionalization with applications well beyond the synthesis of (semi)telechelic polymers, specifically in the areas of macromolecular design and materials synthesis.

3.4 Materials and Methods

3.4.1 Materials

Poly(ethylene glycol) methyl ether (mPEG, 2000 g/mol, Sigma Aldrich), 2,4,6- trichloro-1,3,5-triazine (Sigma Aldrich, 99%), N,N-diisopropylethylamine (DIPEA, Alfa

Aesar, 98+%), 1-(2-furyl)-methylamine (Alfa Aesar, 99%), 4-benzylpiperidine (Sigma

Aldrich, 99%), 2-furanmethanol (Acros, 98%), 2-furylmethanethiol (Alfa Aesar, 99%), 1- phenylmethanamine (Sigma Aldrich, 99%), phenylmethanol (Alfa Aesar, 99%), phenylmethanethiol (Alfa Aesar, 99%), trans-2-[3-(4-tert-butylphenyl)-2-methyl-2- propenylidene]malononitrile (DCTB, Santa Cruz Biotechnology, ≥99%), and sodium trifluoroacetate (NaTFA, Aldrich, 98%) were used as received. All solvents were purchased from Fisher Scientific and used as received.

3.4.2 Characterization

1H NMR spectroscopy was conducted on an Inova 500 MHz, 2 RF channel instrument at 25 °C. Chloroform-d (Cambridge Isotopes Laboratories, Inc, 99.8%) solvent was used as received.

Matrix assisted laser desorption/ionization time-of-flight (MALDI-TOF-

TOF) was performed on a AB Sciex 5800 MALDI TOF/TOF (Framingham, MA) mass spectrometer operated in linear, positive ion mode with a 1 KHz N2 OptiBeam™ on-axis laser. Laser power was used at the threshold level required to generate signal. Spectra were analyzed using Polymerix Version 3 software (Sierra Analytics). Samples were prepared by mixing solutions of DCTB matrix (40.0 mg/mL in THF), NaTFA (1.00 mg/mL in THF), and polymer (2.00 mg/mL in THF) at a v:v:v ratio of 5:1:5 matrix:salt:polymer and 2.00 μL were spotted, then dried under N2, on a stainless steel AB Sciex Plate.

Density functional theory calculations were performed using Spartan

Student Edition version 5 (Wave Function, Inc.) using a B3LYP density functional method with 6-311G** scaling factors.

3.4.3 End-Group Functionalization of mPEG to Dichlorotriazine-mPEG

TCT (230 mg, 1.25 mmol) and mPEG (1.00 g, 0.500 mmol) were dissolved in

DCM (2.00 mL). DIPEA (161 mg, 1.25 mmol) was subsequently added and the reaction

stirred at room temperature for 16 h. The salts were removed by filtration and the polymer was precipitated 2× into cold diethyl ether and vacuum dried, yielding 2-mPEG-

4,6-dichloro-1,3,5-triazine.

3.4.4 2-Furylmethanethiol Conjugation to 2-mPEG-4,6-dichloro-1,3,5-triazine

2-mPEG-4,6-dichloro-1,3,5-triazine (1.00 g, 0.500 mmol) and DIPEA (65.1 mg,

0.505 mmol) were dissolved in DCM (5.00 mL) and cooled to 0 °C in an ice bath. 2-

Furylmethanethiol (57.0 mg, 0.500 mmol) was added and the reaction stirred at 0 °C for

3 h, warmed to room temperature (ca. 1 h), then stirred for an additional 12 h. The salts were removed by filtration, the polymer was precipitated into cold diethyl ether, and vacuum dried, yielding 2-mPEG-4-(2-furylmethanethiol)-6-chloro-1,3,5-triazine.

3.4.5 4-Benzylpiperidine Conjugation to 2-mPEG-4-(2-furylmethanethiol)-6-chloro- 1,3,5-triazine

2-mPEG-4-(2-furylmethanethiol)-6-chloro-1,3,5-triazine (150 mg, 7.50 × 10-2 mmol) and DIPEA (9.76 mg, 7.56 × 10-2 mmol) were dissolved in dioxane (0.750 mL). 4-

Benzylpiperidine (13.1 mg, 7.50 × 10-2 mmol) was added and the reaction was stirred at

80 °C for 16 h. The salts were removed by filtration, the polymer was precipitated into cold diethyl ether, and vacuum dried, yielding 2-mPEG-4-(2-furylmethanethiol)-6-(4- benzylpiperidine)-1,3,5-triazine.

3.4.6 Phenylmethanethiol Conjugation to 2-mPEG-4-(2-furylmethanethiol)-6- chloro-1,3,5-triazine

2-mPEG-4-(2-furylmethanethiol)-6-chloro-1,3,5-triazine (150 mg, 7.50 × 10-2 mmol) and DIPEA (9.76 mg, 7.56 × 10-2 mmol) were dissolved in dioxane (0.750 mL).

Phenylmethanethiol (9.30 mg, 7.50 × 10-2 mmol) was added and the reaction was stirred at 80 °C for 16 h. The salts were removed by filtration, the polymer was precipitated into cold diethyl ether, and vacuum dried yielding predominantly 2-mPEG-4-

(2-furylmethanethiol)-6-chloro-1,3,5-triazine with 16% 2-mPEG-4-(2-furylmethanethiol)-

6-phenylmethanethiol-1,3,5-triazine by 1H NMR spectroscopy conversion.

3.4.7 1-Phenylmethanamine Conjugation to 2-mPEG-4-(2-furylmethanethiol)-6- chloro-1,3,5-triazine

2-mPEG-4-(2-furylmethanethiol)-6-chloro-1,3,5-triazine (150 mg, 7.50 × 10-2 mmol) and DIPEA (9.76 mg, 7.56 × 10-2 mmol) were dissolved in dioxane (0.750 mL). 1-

Phenylmethanamine (8.03 mg, 7.50 × 10-2 mmol) was added and the reaction was stirred at 80 °C for 16 h. The salts were removed by filtration, the polymer was precipitated into cold diethyl ether, and vacuum dried, yielding 2-mPEG-4-(2- furylmethanethiol)-6-(1-phenylmethanamine)-1,3,5-triazine.

3.4.8 Phenylmethanol Conjugation to 2-mPEG-4-(2-furylmethanethiol)-6-chloro- 1,3,5-triazine

2-mPEG-4-(2-furylmethanethiol)-6-chloro-1,3,5-triazine (150 mg, 7.50 × 10-2 mmol) and DIPEA (9.76 mg, 7.56 × 10-2 mmol) were dissolved in dioxane (0.750 mL).

Phenylmethanol (8.10 mg, 7.50 × 10-2 mmol) was added and the reaction was stirred at

80 °C for 16 h. The salts were removed by filtration, the polymer was precipitated into cold diethyl ether, and vacuum dried, yielding mostly 2-mPEG-4-(2-furylmethanethiol)-6- chloro1,3,5-triazine.

3.4.9 1-(2-Furyl)methylamine Conjugation to 2-mPEG-4,6-dichloro-1,3,5-triazine

2-mPEG-4,6-dichloro-1,3,5-triazine (1.00 g, 0.500 mmol) and DIPEA (65.1 mg,

0.505 mmol) were dissolved in DCM (5.00 mL) and cooled to 0 °C in an ice bath. 1-(2-

Furyl)methylamine (48.5 mg, 0.500 mmol) was added and the reaction stirred at 0 °C for

3.4.10 4-Benzylpiperidine Conjugation to 2-mPEG-4-[1-(2-furyl)methylamine]-6- chloro-1,3,5-triazine

2-mPEG-4-[1-(2-furyl)methylamine]-6-chloro-1,3,5-triazine (150 mg, 7.50 × 10-2 mmol) and DIPEA (9.76 mg, 7.56 × 10-2 mmol) were dissolved in dioxane (0.750 mL). 4-

Benzylpiperidine (13.1 mg, 7.50 × 10-2 mmol) was added and the reaction was stirred at

80 °C for 16 h. The salts were removed by filtration, the polymer was precipitated into cold diethyl ether, and vacuum dried, yielding 2-mPEG-4-[1-(2-furyl)methylamine]-6-(4- benzylpiperidine)-1,3,5-triazine.

3.4.11 Phenylmethanethiol Conjugation to 2-mPEG-4-[1-(2-furyl)methylamine]-6- chloro-1,3,5-triazine

2-mPEG-4-[1-(2-furyl)methylamine]-6-chloro-1,3,5-triazine (150 mg, 7.50 × 10-2 mmol) and DIPEA (9.76 mg, 7.56 × 10-2 mmol) were dissolved in dioxane (0.750 mL).

Phenylmethanethiol (9.32 mg, 7.50 × 10-2 mmol) was added and the reaction was stirred at 80 °C for 16 h. The salts were removed by filtration, the polymer was precipitated into cold diethyl ether, and vacuum dried, yielding predominately 2-mPEG-

4-[1-(2-furyl)methylamine]-6-chloro-1,3,5-triazine with 24% conversion to 2-mPEG-4-[1-

(2-furyl)methylamine]-6-phenylmethaneamine-1,3,5-triazine.

3.4.12 1-Phenylmethanamine Conjugation to 2-mPEG-4-[1-(2-furyl)methylamine]- 6-chloro-1,3,5-triazine

2-mPEG-4-[1-(2-furyl)methylamine]-6-chloro-1,3,5-triazine (150 mg, 7.50 × 10-2 mmol) and DIPEA (9.76 mg, 7.56 × 10-2 mmol) were dissolved in dioxane (0.750 mL). 1-

Phenylmethanamine (8.03 mg, 7.50 × 10-2 mmol) was added and the reaction was stirred at 80 °C for 16 h. The salts were removed by filtration, the polymer was precipitated into cold diethyl ether, and vacuum dried, yielding 2-mPEG-4-[1-(2- furyl)methylamine]-6-(1-phenylmethanamine)-1,3,5-triazine.

3.4.13 Phenylmethanol Conjugation to 2-mPEG-4-[1-(2-furyl)methylamine]-6- chloro-1,3,5-triazine

2-mPEG-4-[1-(2-furyl)methylamine]-6-chloro-1,3,5-triazine (150 mg, 7.50 × 10-2 mmol) and DIPEA (9.76 mg, 7.56 × 10-2 mmol) were dissolved in dioxane (0.750 mL).

Phenylmethanol (9.32 mg, 7.50 × 10-2 mmol) was added and the reaction was stirred at

80 °C for 16 h. The salts were removed by filtration, the polymer was precipitated into cold diethyl ether, and vacuum dried, yielding predominately 2-mPEG-4-[1-(2- furyl)methylamine]-6-chloro-1,3,5-triazine with negligible conversion to 2-mPEG-4-[1-(2- furyl)methylamine]-6-phenylmethanol-1,3,5-triazine.

3.4.14 2-Furanmethanol Conjugation to 2-mPEG-4,6-dichloro-1,3,5-triazine

2-mPEG-4,6-dichloro-1,3,5-triazine (200 mg, 0.100 mmol) and DIPEA (13.5 mg,

0.105 mmol) were dissolved in 1,4-dioxane (1.00 mL). 2-Furanmethanol (9.80 mg,

0.100 mmol) was added and the reaction was refluxed at 100 °C for 16 h. The salts were removed by filtration, the polymer was precipitated into cold diethyl ether, and vacuum dried, yielding mostly 2-mPEG-4,6,-dichloro-1,3,5-triazine with 8% 2-mPEG-4-

(2-furanmethanol)-6-chloro-1,3,5-triazine by 1H NMR spectroscopy conversion.

CHAPTER 4 MILD AND EFFICIENT SYNTHESIS OF OMEGA, OMEGA- HETERODIFUNCTIONALIZED POLYMERS AND POLYMER BIOCONJUGATES

4.1 Introduction

*To complement Chapter 3, reported here is a mild approach to introducing orthogonally “clickable” functional groups at single polymer chain ends using benzotrifuranone (BTF).122

Introduced recently,123-126 BTF is uniquely amenable to sequential aminolysis upon reaction with aliphatic amines to afford multifunctionalized molecules (Figure 4-1, a). The basis for both its stepwise ring-opening and a very fast initial aminolysis (k =

14.6 ± 0.2 M-1 s-1 for reaction with heptylamine at 24 °C in acetonitrile) is a synergism of electronic (i.e., inductive) effects and ring strain (the ring strain of BTF is ~ 28 kcal mol-1 based on DFT calculations).126 Compared to TCT, reaction of amines with BTF, while also requiring only near-stoichiometric amounts of reagents for complete conversion, occurs under milder conditions (TCT: 0–80 °C; BTF: –41 °C to room temperature) and generates no by-products (e.g., HCl). Given this, BTF might be a particularly useful reagent for achieving heterofunctionalization of polymers with, or in the presence of, acid- or thermally-sensitive (bio)molecules (e.g., proteins).

Reported here is the application of low-temperature and chemoselective heterofunctionalization of BTF to achieve polymer omega, omega-

*Adapted and reproduced from Ref. 122 with permission from the Royal Society of Chemistry. BTF and difunctional compounds were synthesized by Dr. Ashton Bartley under the supervision of Prof. Ronald K. Castellano. SDS-PAGE gels were conducted by Dr. Bryan S. Tucker under the supervision of Prof. Brent S. Sumerlin.

heterodifunctionalization at ambient temperatures. Shown specifically is how BTF enables introduction of differentially “clickable” functional groups (i.e., an alkyne for

Figure 4-1. General Strategy to functionalize benzotrifuranone (BTF) a) Efficient and chemoselective heterotrifunctionalization of BTF is achieved through reaction temperature, stoichiometry, and temporal control; b) monomethyl ether poly(ethylene glycol) amine (mPEG-amine) conjugation to BTF, yielding mPEG-benzodifuranone (mPEG-BDF); and, c) subsequent one-pot omega, omega-homodifunctionalization using allyl amine. copper-mediated azide-alkyne cycloaddition (CuAAC) and an alkene for thiol-ene reactions) to amine-terminated monomethyl ether poly(ethylene glycol) (mPEG-amine), allowing synthesis of a dye-functionalized polymeric bioconjugate. While polymers with alpha,alpha- or omega,omega-functionalization (prepared using multi-component reactions127,128 or thiolactone ring opening120) have previously been used for bioconjugate synthesis, the approach described here achieves such polymer constructs with unprecedented efficiency.

4.2 Results and Discussion

4.2.1 Conjugating BTF to mPEG-amine

The aminolysis of BTF has been studied extensively using relatively simple amine nucleophiles.123,126 To test this chemistry in the context of polymers, mPEG-amine

(1) was synthesized129 and reacted with a slight excess (1.25 equivalents) of BTF at room temperature (Figure 4-1, b). Consistent with recent kinetic studies,126 predominantly monoaminolysis of BTF was observed under these conditions (i.e., leaving two unreacted lactones). 1H NMR spectroscopy confirmed disappearance of the

–CH2-NH2 resonance at delta = 2.9 ppm (Figure 4-2, a) and the appearance of a singular amide -NH resonance at delta = 7.4 ppm after 16 h. MALDI-ToF MS revealed quantitative conversion of 1 to mPEG-benzodifuranone (2, mPEG-BDF), with the main distribution attributed to 2 + K+ corresponding to a mass increase of 246.1 Da (Figure 4-

2, b). With the utility of BTF for polymer end-group modification confirmed, we extended the chemistry to the synthesis of omega, omega-homodifunctional and omega, omega- heterodifunctional polymers.

4.2.2 Synthesizing Omega, Omega-Homodifunctional Polymers

First, the synthesis of an omega, omega-homodifunctional polymer in one-pot through the addition of a second nucleophile was tested. Allyl amine (5 equiv) was added to the solution of 1 and BTF after 16 h, and the reaction was left to stir for an additional 24 h at room temperature. Full conversion of 1 to the diallyl omega, omega- homodifunctional mPEG (3) was observed after 16 h by 1H NMR spectroscopy, based on the appearance of two vinyl proton signals (at delta = 5.1 and 5.8 ppm), a shift of the amide -NH peak (from delta = 7.4 to 7.0 ppm), and the appearance of a second amide

Figure 4-2. One-pot synthesis of omega, omega-homodifunctionalized polymer 3 beginning from benzotrifuranone (BTF) and monomethyl ether poly(ethylene glycol) amine (mPEG-amine); a) 1H NMR spectra of (top) mPEG-amine 1, (middle) mPEG-BDF 2, generated from addition of BTF to 1, and (bottom) omega, omega-homodifunctionalized polymer 3 formed from reaction of 2 with allyl amine; b) MALDI-ToF mass spectra corresponding to the reactions shown in part a; c) size-exclusion chromatography analysis of 1 and omega, omega-homodifunctionalized polymer 3 following a one-pot synthesis.

-NH resonance at delta = 6.3 ppm (integrating to two protons) resulting from the second and third lactone ring openings. SEC analysis confirmed that minimal polymer- polymercoupling occurred due to the absence of a high molecular weight shoulder

(Figure 4-2, c). MALDI-ToF MS further confirmed the quantitative conversion of 1 to 3 in a one-pot reaction via 2 (Figure 4-2, b). A major distribution attributed to 3 + Na+ and a minor distribution attributed to 3 + K+ were observed, accompanied by a mass increase of 98.26 Da compared to 2. These results confirmed that BTF was a viable option for executing a one-pot polymer end-group omega, omega-homodifunctionalization.

Unfortunately, any attempts to omega, omega-heterodifunctionalize 2 were unsuccessful due to the modest aminolysis rate discrimination (< 10:1 on a per lactone basis126) between the second and final lactone rings at room temperature and the insufficient solubility of mPEG-amine at lower temperatures that would have provided improved selectivity.126

4.2.3 Synthesizing Omega, Omega-Heterodifunctional Polymers

Fortunately, BTF provides an alternative approach to synthesize omega omega- heterodifunctional polymers through installation of the desired functional groups prior to polymer conjugation (Figure 4-3, a). To illustrate this approach, three BTF-derived monolactones (benzomonofuranones, or BMFs) were prepared using low-temperature

(-41 C) sequential amine addition to BTF (with heptylamine, allylamine, or propargylamine) using conditions reported previously.123 Subsequent reaction of 1 with

1.20-1.25 equivalents of the monolactones for 16 h at room temperature provided three distinct polymer conjugates: 5, derived from heptylamine and propargylamine addition;

7, derived from heptylamine and allylamine addition; 9, derived from octyl-7-ene-amine and propargylamine. Successful synthesis of each of the products was confirmed by 1H

NMR spectroscopy, with the appearance of the protons characteristic of each functional group and three unique amide -NH protons between delta = 6.1–7.1 ppm (Figure 4-3, b). Additionally, MALDI-ToF MS showed main distributions of 5 + Na+ with a shift of

400.6 Da, 7 + Na+ with a shift of 402.7 Da, and 9 + Na+ with a shift of 412.6 Da (Figure

4-3, c). Requiring at most ambient temperature and near stoichiometric quantities of reagents, BTF proved to be an excellent scaffold for the preparation of omega, omega- heterodifunctional polymers.

Figure 4-3. Monomethyl ether poly(ethylene glycol) amine (mPEG-amine) omega, omega-heterodifunctionalization using benzotrifuranone derivatives functionalized with various moieties; a) Reaction of various benzomonofuranones (BMFs) with mPEG-amine (1) to yield omega, omega- heterodifunctionalized polymers; 1H NMR (b) and MALDI-ToF MS (c) spectra confirming omega, omega-heterodifunctionalization.

Next, omega, omega-heterodifunctional polymer 9 was used to synthesize polymer bioconjugates through the installation of a coumarin dye and biotin. An azide derivative of coumarin 1 was conjugated to 9 using CuAAC to afford the dye- functionalized mPEG (10, Figure 4-4, a).12 By 1H NMR spectroscopy, disappearance of the terminal alkyne proton and appearance of the triazole proton at delta = 7.6 ppm confirmed successful reaction, while MALDI-ToF MS showed an appropriate shift in the molecular weight distribution (Figure B-1). SEC equipped with a UV-vis detector set to the lambdamax of coumarin in chloroform (~ 390 nm) further confirmed the dye was conjugated to the polymer (Figure 4-4, b). To synthesize the biotin-coumarin omega,

omega-heterodifunctional bioconjugate 11, a thiol-ene reaction between alkene 10 and

2-mercaptoethylbiotin was performed using 5 equivalents each of thiol and 2,2- dimethyoxy-2-phenylacetophenone.130 The product was obtained in 60% conversion following dialysis, comparable to reported thiol-ene reactions on isolated aliphatic alkenes,130 and the product structure was confirmed using 1H NMR by the appearance of the urea protons of biotin at delta = 6.3–6.4 ppm (Figure B-2).

Figure 4-4. Functionalization of monomethyl ether poly(ethylene glycol) omega, omega- heterodifunctionalized polymers (mPEG-BTF conjugate) and subsequent protein binding studies. a) Copper-mediated azide-alkyne cycloaddition of an azide derivative of coumarin 1 to 9, thiol-ene reaction of 10 with 2- mercaptoethylbiotin, and avidin binding to 11; b) size-exclusion chromatograms attained using a refractive-index (RI) detector coupled with a UV-vis detector set to 390 nm; c) avidin-functionalized agarose beads following incubation with 11 compared to native agarose beads; d) SDS- PAGE analysis confirming avidin association with 11. Lanes: (1) avidin, (2) avidin heated to 90 C for 10 min prior to SDS-PAGE to disrupt tetramers, (3) 11 incubated with avidin.

4.2.4 Conjugating Omega, Omega-Heterodifunctional Polymers to Avidin

For functional confirmation of the synthesis of 11, the polymer sample was incubated with avidin agarose gel beads, since avidin is known to have a strong non- covalent interaction with biotin.131 We reasoned that any polymer bioconjugates containing biotin (11) should bind to the beads while the polymer without biotin (10) could be removed via centrifugation. After 6 successive centrifugations followed by removal of the supernatant and washes of the remaining pellet with PBS, the beads displayed a yellow coloration suggesting successful polymer-protein binding between 11 and the bead-bound avidin (Figure 4-4, c). Subsequently, removal of 11 was carried out by incubating the polymer-conjugated resin in a solution of biotin in PBS for 48 h, after which the supernatant resulting from centrifugation was dialyzed to remove excess biotin and lyophilized.

The bioconjugate 11 was then incubated with free avidin for 16 h at 25 C to yield non-covalent polymer-protein conjugates in solution. Sodium dodecyl sulfate- polyacrylamide gel electrophoresis (SDS-PAGE) was used to characterize polymer- protein binding (Figure 4-4, d). Avidin exists predominately as a tetramer in solution, which has been shown to aggregate with SDS, limiting its gel mobility during electrophoresis.132 In lane 1, avidin aggregation is evident as no migration into the separating gel is observed. Heating avidin at 90 C for 10 min prior to SDS-PAGE effectively disrupted tetramer formation as evidenced by protein movement into the gel and resolution at the expected molecular weight (lane 2). Interestingly, when the solution containing 11 and avidin was run, a small band attributed to avidin aggregation was still observed, but a broad band attributed to the desired polymer-avidin conjugates

and a low molecular weight band attributed to avidin monomers were also observed

(lane 3). The prescence of avidin monomers suggested that the polymer end groups were disrupting tetramer formation, consequently limiting aggregation during SDS-

PAGE. Avidin tetramers are formed through strong van der Waals forces and hydrogen bonding,131 yet the end group of 11 seemed to effectively break up these interactions.

These SDS-PAGE results suggest that: (1) omega, omega-heterodifunctional polymer bioconjugates were synthesized using the facile nucleophilic end-group chemistry based on the sequential aminolysis of BTF; and (2) the functionalized polymer end groups broke up the strong interactions between avidin monomers and concurrently formed polymer-protein conjugates.

4.3 Conclusions

In conclusion, we have reported a new method for synthesizing omega, omega- heterodifunctionalized polymers and polymer bioconjugates. The approach employs the sequential aminolysis chemistry of benzotrifuranone (BTF) and allows polymer functionalization at ambient temperature in a limited number of synthetic steps using near-stoichiometric quantities of reagents. To illustrate the method and its versatility, orthogonally reactive click-functional handles were added to the chain end of mPEG- amine. Subsequent sequential reactions successfully afforded a polymer containing biotin and a coumarin dye; the polymer bioconjugates were able to bind avidin and were unexpectedly shown to break up the otherwise strong interactions between avidin monomers. This methodology holds promise for more elaborate polymer heterofunctionalization schemes given the compatibility of BTF with a broad scope of amine nucleophiles and thermally sensitive moieties, like proteins.

4.4 Materials and Methods

4.4.1 Materials

Monomethyl ether poly(ethylene glycol) (mPEG, 2000 g/mol, Sigma Aldrich, trans-2-[3-(4-tert-butylphenyl)-2-methyl-2-propenylidene]malononitrile (DCTB, Santa

Cruz Biotechnology, ≥99%), anthralin (AK Scientific, 98%), trifluoroacetic acid (TFA,

Fisher Scientific, 97%), sodium trifluoroacetate (NaTFA, Aldrich, 98%), and Pierce avidin agarose (Thermo Scientific) were used as received. All other reagents were purchased from VWR International and used as received. mPEG-amine was synthesized according to the procedure outline below. Oct-7-ene-amine,1 the azide derivative of coumarin 1,2 and 2-mercaptoethylbiotin3 were synthesized according to previous reports. All solvents were purchased from Fisher Scientific and used as received. DMF was purified using a Glass Contour solvent system (Glass Contour, Inc., now Pure Process Technology), degassed in 20 L drums, and passed through two columns of molecular sieves under an argon atmosphere. Thin Layer Chromatography

(TLC) was performed using aluminum-backed silica gel plates. The plates were developed using UV light and ninhydrin staining. Flash column chromatography was performed using SiO2-60 230-400 mesh silica gel.

4.4.2 Characterization

1H NMR spectroscopy was conducted on an Inova 500 MHz, 2 RF channel instrument at 25 °C. Chloroform-d (Cambridge Isotopes Laboratories, Inc., 99.8%) solvent was used as received.

Matrix assisted laser desorption/ionization time-of-flight (MALDI-TOF/TOF) was performed on a Bruker Microflex LRF MALDI TOF (Billerica, MA) mass spectrometer in reflectron, positive ion mode using an N2 on-axis laser. Spectra were collected in

flexControl (Bruker Daltronics Inc., Billerica, MA) and analyzed using flexAnalysis

(Bruker Daltronics Inc., Billerica, MA) and Polymerix Version 3 software (Sierra

Analytics, Modesto, CA). Analysis of mPEG-amine was performed by mixing dithranol matrix (20.0 mg/mL in DCM) and polymer solution (2.0 mg/mL in DCM with 2 drops trifluoroacetic acid) at a v:v ratio of 5:2 matrix:polymer and 3 L were spotted on a stainless steel Bruker MSP 96 target polished steel BC plate and air dried. Analysis of mPEG-BTF conjugates was performed by mixing solutions of DCTB matrix (10.0 mg/mL in THF) and polymer (2.00 mg/mL in THF) at a v:v ratio of 5:2 matrix:polymer and 2.00

μL were spotted, then dried under N2, on a stainless steel AB Sciex Plate.

Subsequently, 1.00 μL of a NaTFA solution (1.00 mg/mL in THF) was spotted on top of the polymer-matrix spot and dried under N2.

Size exclusion chromatography (SEC) was performed in N,N-dimethylacetamide

(DMAc) with 50 mM LiCl at 50 °C and a flow rate of 1.0 mL min-1 (Agilent isocratic pump, degasser, and autosampler, colums: Plgel 5 μm guard + two ViscoGel I-series

G3078 mixed bed columns: molecular weight range 0-20 × 103 and 1-100 × 104 g mol-

1). Detection consisted of a Wyatt Optilab T-rEX refractive index detector operating at

658 nm and a Wyatt miniDAWN Treos light scattering detector operating at 659 nm.

Absolute molecular weights and polydispersities were calculated using the Wyatt

ASTRA software and 100% mass recovery methods.

4.4.3 Synthesis of Benzotrifuranone (BTF)

To a 350 mL sealed pressure vessel was added 1,3,5-trimethoxybenzene (15 g,

89 mmol), paraformaldehyde (10 g, 33 mmol), and glacial acetic acid (35 mL). The solution was allowed to stir for one hour at room temperature. HBr (33% in AcOH, 75

mL) was added to the flask and the solution was stirred for 3 hours at 70 °C. The resulting orange solution was cooled to room temperature and allowed to stir for 16 hours. The reaction mixture was poured into water (700 mL), and CH2Cl2 was added until all solids were dissolved. The organic layer was separated and the aqueous layer was extracted with CH2Cl2. All organics were combined and washed successively with saturated NaHCO3, brine, and water. The organic layer was dried over Na2SO4, filtered, and concentrated under reduced pressure to afford a brown oil. The oil was purified via column chromatography (2:1 hexanes:CH2Cl2) to yield 1,3,5-tris(bromomethyl)-2,4,6-

1 trimethoxybenzene as a white solid (12 g, 29%). H NMR (CDCl3) δ 4.15 (s, 9H), 4.63

13 (s, 6H). C NMR (CDCl3) δ 22.6, 62.8, 123.4, 160.2. HRMS (DART) calculated for

+ C12H15O3Br3 [M + NH4] 463.8889, found 463.8903.

Presented here is a modified and improved synthetic procedure to the one we originally published in 2005.4 To a solution of 1,3,5-tris(bromomethyl)-2,4,6- trimethoxybenzene (17 g, 37 mmol) in acetonitrile (250 mL) and water (33 mL) was added KI (0.37 g, 1.5 mmol), KCN (12 g, 12 mmol) and 18-crown-6 (2.5 g, 6.2 mmol).

The resulting solution was allowed to stir overnight at room temperature. The solution was poured into ice water (700 mL) and suction filtered as soon as the ice melted. The filtered solid was dried under reduced pressure to afford 2,2ʹ,2ʺ-(2,4,6- trimethoxybenzene-1,3,5-triyl)triacetonitrile as a white solid (11 g, 100%). 1H NMR

13 (CDCl3) δ 3.72 (s, 6H), 4.02 (s, 9H). C NMR (CDCl3) δ 13.0, 62.9, 116.2, 117.7,

+ 158.6. HRMS (DART) calculated for C15H15N3O3 [M+NH4] 303.1452, found 303.1462.

Presented here is an updated synthetic procedure to the ones we originally published in 20054 and 2009.5 A solution of 2,2',2''-(2,4,6-trimethoxybenzene-1,3,5-

triyl)triacetonitrile (8.5 g, 28 mmol) in HBr (48%, 95 mL) was heated to reflux overnight.

The solution was cooled to room temperature at which a precipitate formed. Water (200 mL) was added to the flask and the mixture was extracted with ethyl acetate. The organic layers were combined, dried over MgSO4, and concentrated under reduced pressure. The solid was then hydrolyzed with NaOH (22 g NaOH in 100 mL water) at 60

°C for 3 hours. The resultant solution was cooled on an ice bath, acidified with concentrated HCl, and extracted with ethyl acetate. The organics were combined, dried over MgSO4, and concentrated under reduced pressure yielding 2,2ʹ,2ʺ-(2,4,6- trihydroxybenzene-1,3,5-triyl)triacetic acid (8.8 g, 95%). 1H NMR (DMSO-d6) δ 3.46 (s,

6H), 8.23 (s, 3H), 11.95 (s, 3H). 13C NMR (DMSO-d6) δ 29.7, 103.0, 153.0, 173.5.

+ HRMS (ESI) calculated for C12H12O9 [M=Na] 323.0374, found 323.0380.

Presented here is an updated synthetic procedure to the one we originally published in 2009.5 A solution of 2,2ʹ,2ʺ-(2,4,6-trihydroxybenzene-1,3,5-triyl)triacetic acid (4.4 g, 15 mmol) and polyphosphoric acid (PPA, 50 g) were heated to 110 °C overnight. The solution was cooled to 0 °C and 800 mL ice water was added with stirring. The aqueous solution was extracted with CH2Cl2 (emulsion). The organics were combined, washed with water, brine, and dried over Na2SO4. The solution was filtered and then clarified with activated carbon. The solution was concentrated under reduced pressure to yield BTF (1.1 g, 30%). The solid was purified by column chromatography

1 (98% CH2Cl2/acetone) and obtained was a light tan product. H NMR (CDCl3) δ 3.80 (s,

13 6H). C NMR (CDCl3) δ 30.3, 101.4, 150.3, 171.9. HRMS (GC-CI-MS) calculated for

+ C12H6O6 [M+H] 247.0243, found 247.0243.

4.4.4 Synthesis of mPEG-Amine

The synthesis was adapted from a previous report.6 First, mPEG (50 g, 25 mmol) was dissolved in toluene (300 mL) and dried during the azeotropic distillation of approximately 150 mL toluene. The flask was backfilled with N2, cooled to 0 °C, and

TEA (13 g, 13 mmol) was added. Subsequently, p-toluene sulfonyl chloride (24 g, 130 mmol) was dissolved in dry DCM and cannulated into the cooled flask. The reaction was held at 0 °C for 2 h then allowed to warm to room temperature and stir overnight. The salts were filtered off, the solution concentrated, and the polymer precipitated 2× into cold diethyl ether yielding 49 g of mPEG-TS. The amination was performed in two batches. One of the batches proceeded as follows: mPEG-TS (20 g, 10 mmol) was dissolved in 30% ammonium hydroxide solution (300 mL) in a plastic Nalgene bottle, wrapped in Parafilm, and left to stir for 7 days. The reaction was opened and placed in the back of the hood to allow the ammonia to evaporate over 7 days. The pH was then adjusted to 13 using 1.0 M NaOH and the aqueous solution was washed 4× with DCM.

The organic layers were combined, concentrated, and the polymer precipitated 2× into cold diethyl ether yielding the PEG-amine product 1 (38.8 g total yield across both batches).

4.4.5 Synthesis of BTF Conjugates

4.4.5.1 mPEG-amine conjugation to BTF

1 (40 mg, 2.0 × 10-2 mmol) and BTF (6.2 mg, 2.5 × 10-2 mmol) were dissolved in

THF (1.6 mL) and left to stir at room temperature for 16 h. The polymer was then precipitated from cold diethyl ether yielding the functionalized mPEG 2.

4.4.5.2 One-pot m-PEG-amine conjugation to BTF followed by diallyl-amine addition conjugate

mPEG-amine (100 mg, 5.0 × 10-2 mmol) and BTF (15 mg, 6.3 × 10-2 mmol) were dissolved in THF (2.0 mL) and left to stir at room temperature for 16 h. Subsequently, allyl amine (13 mg, 0.23 mmol) was added and the reaction left to stir for an additional

24 h. The polymer was then precipitated from cold diethyl ether yielding the functionalized mPEG 3.

4.4.5.3 Synthesis of 2-(4,6-Dihydroxy-2-oxo-5-(2-oxo-2-(prop-2-yn-1- ylamino)ethyl)-2,3-dihydrobenzofuran-7-yl)-N-heptylacetamide

To a solution of BTF (50 mg, 0.20 mmol) in dry DMF (2.0 mL) cooled to -41 °C was added propargylamine (11 mg, 0.20 mmol, 0.40 mL of 0.50 M solution in DMF).

The reaction stirred for 1 h and then heptylamine (23 mg, 0.20 mmol, 0.81 mL of 0.25 M solution in DMF) was added. The resulting reaction mixture stirred at -41 °C for 8 h and was then poured in to EtOAc (75 mL). The organic solution was washed with water (5 ×

25 mL) and water (1 × 25 mL), dried over Na2SO4, filtered, and concentrated. The residue was purified by column chromatography (40:60 EtOAc:Hex) to yield product 4

1 (0.42 g, 49%) as a mixture of regioisomers (43% and 57%). H NMR (DMSO-d6) δ 0.85

(t, 6H, J = 6.9 Hz), 1.24 (s, 6H), 1.40 (m, 4H), 3.04 (dq, 4H J = 5.5 Hz), 3.09 (t, 1H, J =

2.6 Hz), 3.11 (t, 1H, J = 2.5 Hz), 3.43 (s, 2H), 3.46 (s, 4H), 3.50 (s, 2H), 3.70 (s, 4H),

3.86 (ddd, 4H, J = 2.52 and 5.31 Hz), 8.26 (t, 1H, J = 5.5 Hz), 8.37 (t, 1H, J = 5.5 Hz),

8.50 (t, 1H, J = 5.5 Hz), 8.63 (t, 1H, J = 5.4 Hz), 9.71 (s, 1H), 9.92 (s, 1H), 10.7 (s, 1H)

13 10.9 (s, 1H). C NMR (DMSO-d6) δ 13.9, 22.0, 26.3, 28.1, 28.2, 28.3, 28.4, 28.7, 28.8,

30.4, 30.7, 30.8, 31.2, 31.2, 31.3, 31.6, 3.1.7, 39.0, 73.0, 73.2, 80.8, 81.1, 98.1, 98.5,

100.5, 100.7, 106.4, 106.6, 150.5, 150.7, 151.7, 152.1, 155.4, 155.5, 170.8, 171.4,

+ 171.9, 172.6, 172.6, 174.6. HRMS (ESI) calculated for C22H28N2O6 [M+H] 417.2020, found 417.2039.

4.4.5.4 mPEG-amine conjugation to BTF conjugate 4

1 (37 mg, 1.8 × 10-2 mmol) and BTF conjugate 4 (9.5 mg, 2.3 × 10-2 mmol) were dissolved in THF (0.73 mL) and left to stir at room temperature for 16 h. The polymer was then precipitated from cold diethyl ether yielding the functionalized mPEG 5.

4.4.5.5 Synthesis of N-Allyl-2-(5-(2-(heptylamino)-2-oxoethyl)-4,6-dihydroxy-2-oxo- 2,3-dihydrobenzofuran-7-yl)acetamide (6)

To a solution of BTF (40 mg, 0.16 mmol) in dry DMF (2.0 mL) cooled to -41 °C was added allylamine (9.1 mg, 0.16 mmol, 0.33 mL of 0.5 M solution in DMF). The resulting solution was allowed to stir for 1 h before the addition of heptylamine (18 mg,

0.16 mmol, 0.65 mL of 0.25 M solution in DMF). The solution then stirred for an additional 6 h at -41 °C before being poured in to EtOAc (75 mL). The organic solution was washed with water (5 × 25 mL) and brine (1 × 25 mL), dried over Na2SO4, filtered, and concentrated. The residue was purified by column chromatography (1:1

EtOAc:Hex) to yield product 6 (0.38 g, 56%) as a mixture of regioisomers (43% and

1 57%). H NMR (DMSO-d6) δ 0.85 (t, 6H, J = 6.5 Hz), 1.19 – 1.30 (m, 16H), 1.34 – 1.45

(m, 4H), 3.04 (dq, 4H J = 5.9 Hz), 3.45 (s, 2H), 3.46 (s, 2H), 3.49 (s, 2H), 3.50 (s, 2H),

3.65 – 3.73 (m, 8H), 5.01 – 5.08 (m, 2H), 5.09 – 5.12 (m, 1H), 5.15 – 5.18 (m, 1H), 5.70

– 5.85 (m, 2H), 8.20 (dt, 2H J = 5.0 Hz), 8.45 (dt, 2H, J = 5.0 Hz), 9.87 (s, 1H), 9.98 (s,

13 1H), 10.9 (s, 1H), 11.0 (s, 1H). C NMR (DMSO-d6) δ 13.9, 22.1, 26.3, 28.3, 28.4, 28.7,

28.8, 30.7, 30.8, 31.0, 31.2, 31.2, 31.3, 31.6, 31.7, 38.9, 39.0, 41.1, 41.2, 98.3, 98.5,

100.6, 100.7, 106.5, 106.7, 115.2, 115.4, 134.8, 134.9, 150.6, 150.7, 151.7, 151.9,

155.4, 155.5, 171.2, 171.7, 171.9, 172.4, 174.6, 174.6. HRMS (ESI) calculated for

+ C22H30N2O6 [M+H] 419.2177, found 419.2177.

4.4.5.6 m-PEG amine conjugation to BTF conjugate 6

1 (20 mg, 1.0 × 10-2 mmol) and BTF conjugate 6 (5.0 mg, 1.2 × 10-2 mmol) were dissolved in THF (0.20 mL) and left to stir at room temperature for 16 h. The polymer was then precipitated from cold diethyl ether yielding the functionalized mPEG 7.

4.4.5.7 Synthesis of 2-(4,6-Dihydroxy-2-oxo-7-(2-oxo-2-(prop-2-yn-1- ylamino)ethyl)-2,3-dihydrobenzofuran-5-yl)-N-(oct-7-en-1-yl)acetamide (8)

To a solution of BTF (50 mg, 0.20 mmol) in dry DMF (2.0 mL) at -41 °C was added propargylamine (11 mg, 0.20 mmol, 0.40 mL of 0.50 M solution in DMF). The resulting solution was allowed to stir for 1.5 h before the addition of oct-7-en-1-amine

(25 mg, 0.20 mmol, 0.81 mL of 0.25 M in DMF) was added and the solution stirred for 8 h at -41 °C. The reaction mixture was then poured in to EtOAc (75 mL). The organic solution was washed with water (5 × 25 mL) and brine (1 × 25 mL), dried over Na2SO4, filtered, and concentrated. The residue was purified by column chromatography (1:1

EtOAc:Hex) to yield product 8 (29 mg, 34%) as a pair of inseparable regioisomers (34% and 66%). 1H NMR (DMSO-d6) δ 1.25–1.42 (m, 16H), 2.00 (q, 4H, J = 6.7 Hz), 3.04

(dq, 4H J = 5.6 Hz), 3.09 (t, 1H, J = 2.5 Hz), 3.11 (t, 1H, J = 2.4 Hz), 3.43 (s, 4H), 3.46

(s, 2H), 3.49 (s, 2H), 3.70 (s, 4H), 3.86 (ddd, 4H, J = 2.49 and 5.30 Hz), 4.91–5.02 (m,

4H), 5.72 – 5.85 (m, 2H), 8.26 (t, 1H, J = 5.3 Hz), 8.37 (t, 1H, J = 5.5 Hz), 8.50 (t, 1H, J

= 5.5 Hz), 8.63 (t, 1H, J = 5.5 Hz), 9.71 (s, 1H), 9.92 (s, 1H), 10.7 (s, 1H), 10.9 (s, 1H).

13C NMR (DMSO-d6) δ 26.2, 28.2, 28.2, 28.6, 28.7, 30.4, 30.7, 30.9, 31.3, 31.7, 31.7,

33.1, 38.9, 38.9, 72.9, 73.2, 80.8, 81.1, 98.1, 98.5, 100.5, 100.7, 106.4, 106.6, 114.6,

138.8, 150.4, 150.7, 151.7, 152.1, 155.4, 155.5, 170.8, 171.5, 171.8, 172.6, 174.5,

174.6. HRMS (ESI) calculated for C23H28N2O6 [M+H]+ 429.2020, found 429.2033.

4.4.5.8 m-PEG amine conjugation to BTF conjugate 8

1 (35 mg, 1.8 × 10-2 mmol) and BTF conjugate 8 (9.4 mg, 2.2 × 10-2 mmol) were dissolved in THF (0.70 mL) and left to stir at room temperature for 16 h. The polymer was then precipitated from cold diethyl ether yielding the functionalized mPEG 9.

4.4.5.9 Copper-mediated azide-alkyne cycloaddition of mPEG-BTF conjugate 9 and azido coumarin

mPEG-BTF conjugate 9 (50 mg, 2.1 × 10-2 mmol), azido coumarin (6.2 mg, 2.3 ×

10-2 mmol), and N ,N ,N ,N, N-pentamethyldiethylene triamine (2.6 mg, 1.5 × 10-2 mmol) were dissolved in DMF (0.75 mL) in a Schlenk flask and the solution was purged with argon for 20 min. The reaction was placed in liquid N2 and Cu(I)Br (1.5 mg, 1.0 ×

10-2 mmol) was added on top the frozen solution. The head-space was purged with argon for an additional 20 min, the solution thawed, and the reaction set to stir for 24 h at room temperature. The reaction was quenched by exposure to air, filtered through neutral alumina, and polymer 10 was precipitated from cold diethyl ether.

4.4.5.10 Thiol-ene of polymer 10 with 2-mercaptoethylbiotin

10 (30 mg, 1.1 × 10-2 mmol), 2-mercaptoethylbiotin (17 mg, 5.5 × 10-2 mmol), and

2,2-dimethoxy-2-phenylacetophenone (14 mg, × 10-2 mmol) were dissolved in DMF

(0.50 mL) in a Schlenk flask, wrapped in foil, and purged with argon for 20 min. The flask was then placed under UV irradiation for 2 h. The reaction was quenched by exposure to air and polymer 11 was purified via dialysis against DI water (1000 MCWO tubing) and lyophilization.

4.4.6 Avidin Conjugations

4.4.6.1 Purification of 11 using avidin agarose gel beads

11 (1.5 mg, 0.5 mL of a 3.0 mg/mL PBS solution) was added to a suspension of avidin agarose gel beads (0.50 mL) in a centrifuge tube. The mixture was agitated for 2 h at room temperature. Subsequently, the solution was centrifuged, the supernatant was removed, PBS was added (0.50 mL), and the solution was agitated for 1 h at room temperature. This cycle was repeated 5× until the supernatant appeared colorless, while the gel retained a light yellow coloration. The avidin-agarose gel beads were added to a biotin solution in PBS (1.0 mL, ca. 1.00 × 10-2 M) and stirred for 48 h at room temperature. Isolation of the polymer was conducted via 4 cycles of centrifugation, supernatant collection, and addition of PBS. The collected supernatant was combined, dialyzed against DI water (1000 MWCO tubing), and lyophilized yielding the polymer product 11.

4.4.6.2 Polymer conjugation to avidin

Biotin-containing polymer conjugate 11 (1.5 mg, 4.97 × 10-4 mmol) and avidin

(0.60 mg, 9.95 × 10-6 mmol) were dissolved in PBS (1.2 mL) and stirred at room temperature for 16 h. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis was performed using aliquots taken directly from this solution.

CHAPTER 5 COLOR-CODING VISIBLE LIGHT POLYMERIZATIONS TO ELUCIDATE THE ACTIVATION OF TRITHIOCARBONATES USING EOSIN Y

5.1 Introduction

*Eosin Y (EY) presents an interesting opportunity to catalyze polymerizations in aqueous medium, as the organodye is water-soluble and has been shown to effectively activate PET-RAFT polymerizations of methacrylates and vinyl ketones in organic solvent.133,134 Indeed, EY has recently been reported to catalyze aqueous polymerizations in the presence of cells,49 proteins,135 or oxygen (when ascorbic acid is added).136 To further expand the applicability of this photocatalyst, we sought to explore the polymerization mechanism and seek opportunities to optimize the initiation and reversible-termination steps.

EY absorbs both blue and green light (Figure C-1), with these absorptions being attributed to the dimer and monomer in solution, respectively.137 Therefore, we hypothesized that polymerizations may occur under irradiation at both wavelengths. The mechanism of initiation can be controlled through the addition of a sacrificial reducing agent, either through reduction of the excited-state EY (reductive PET-RAFT, Figure 5-

1, a) or oxidation of the excited-state EY (oxidative PET-RAFT, Figure 5-1, b).

Additionally, since trithiocarbonate (TTC) compounds absorb in blue wavelengths

(Figure C-2) and may undergo photolysis of the carbon-sulfur (C-S) bond,32,138 it was necessary to also consider a photoiniferter30 polymerization mechanism where the C-S bond homolytic cleavage occurs independent of catalyst (Figure 5-1, c). Herein, we

report our investigations into how polymerization conditions may affect the mechanism of visible-light PET-RAFT polymerizations and the resultant polymer properties.

Specifically, the power of dictating mechanism via wavelength was found to greatly improve polymerization control. Therefore, this report shows that an understanding of redox processes during polymerization may lead to optimized aqueous polymerization conditions and access to a variety of well-defined polymer compositions using biologically-relevant polymerization conditions.

5.2 Results and Discussion

5.2.1 Trapping Studies under Blue-Light Irradiation

To study the mechanism for generation of carbon-centered radicals during photopolymerization with TTCs (ZSC(=S)SR), model trapping studies with 2-(ethyl trithiocarbonate)propionic acid (ETPA) and 15 equiv. of the hydrogen source N- ethylpiperidine hypophosphate (EPHP)139 were performed (Figure 5-2), in a manner analogous to an end-group removal technique we recently reported.140 The monomethyl

R-group was chosen to best emulate the sterics of a growing acryloyl chain end. The disappearance of the methine proton adjacent to the TTC (at delta = 4.55 ppm) was monitored using 1H NMR spectroscopy and was attributed to homolytic cleavage of the

C-S bond and subsequent radical quenching by EPHP. The slope of the pseudo-first order kinetic plot of ETPA consumption was then evaluated to determine the rate of radical generation from the C-S bond cleavage. Low equivalents of EY (Table 5-1, Entry

3) lead to a faster disappearance of the methine proton than photolysis (Table 5-1,

Entry 5), and higher concentrations of EY (Table 5-1, Entry 2) led to faster disappearance of the methine proton. This increased rate in disappearance in methine

Figure 5-1. Possible visible-light induced RAFT polymerizations mechanisms a) proposed photopolymerization mechanism where the excited state eosin Y is reduced by a tertiary amine, leading to a reductive photo electron-transfer reversible addition-fragmentation chain transfer polymerization (PET-RAFT) mechanism; b) proposed photopolymerization mechanism where the excited state eosin Y is oxidized by a trithiocarbonate, leading to an oxidative PET- RAFT mechanism; c) proposed photoiniferter mechanism where the trithiocarbonate species undergoes photolysis upon excitation. proton was consistent with EY-inducing radical formation through an oxidative PET pathway under blue light (Figure 5-2, a). When the tertiary amine (4- dimethylaminopyridine (DMAP)) was added in the absence of EY (Table 5-1, Entry 4), consumption of the methine proton of ETPA was faster than in the absence of EY (when

the ratio of ETPA to EY was 1:0, Table 5-1, Entry 5) indicating that the tertiary amine can also undergo a redox reaction with the excited state TTC to induce radical formation. This observation is in agreement with previous studies by Qiao34,141-143 and

Konkolewicz144 where a faster rate of polymerization was observed when only tertiary amine was added to a TTC. As expected, in the presence of both EY and DMAP (Table

5-1, Entry 1), the rate of consumption of ETPA was faster than the other trials. Since the excited-state EY is gaining an electron from the tertiary amine, the difference between the redox potential of the radical anion EY (-1.06 V)145 and the ground state redox potential of the TTC (-0.6 V)43 is large, yielding a favorable electron transfer from the reduced EY to the TTC. Overall, these trapping studies indicate that multiple mechanisms of initiation, having similar rates of radical generation, may occur simultaneously when both EY and a tertiary amine are used as PET catalysts under blue-light irradiation.

Table 5-1. Apparent rate of consumption of 2-(ethyl trithiocarbonate)propionic acid using different ratios of chain transfer agent:eosin Y:4-dimethylaminopyridine under blue light irradiation Entry ETPA:EY:DMAP Apparent rate of ETPA disappearance (min-1) 1 1:0.0004:1 5.58  10-3 2 1:0.004:0 2.31  10-3 3 1:0.0004:0 1.81  10-3 4 1:0:1 5.49  10-3 5 1:0:0 1.52  10-3

5.2.2 Trapping Studies under Green-Light Irradiation

Next, the trapping studies were conducted under green light to further understand the impact of irradiation wavelength on C-S bond cleavage (Figure 5-2, b).

Unfortunately, due to the rapid degradation of EY under green-light irradiation (as evidenced by the decrease in rate of ETPA consumption after 3 h and loss of the

Figure 5-2. Trapping studies of 2-(ethyl trithiocarbonate)propionic acid using different ratios of chain transfer agent:eosin Y:4-dimethylaminopyrdine under a) blue light irradiation, and b) green light irradiation using 15 equiv. of N- ethylpiperidine hypophosphate (EPHP) as a hydrogen source. characteristic pink color of EY), the rates of ETPA degradation could not be fit using a linear regression. However, we observed important qualitative differences between radical initiation under green light versus blue light. The experiments containing only

ETPA (ETPA:EY:DMAP ratio of 1:0:0) or ETPA and DMAP (ETPA:EY:DMAP ratio of

1:0:1) led to minimal disappearance of the methine proton adjacent to the TTC of ETPA.

Low conversion is expected since TTCs show such a low absorbance at green

wavelengths (Figure C-2). Importantly, these results indicate the photoiniferter and tertiary amine reduction mechanism play minor roles during green-light irradiation.

Trapping experiments that included EY and ETPA (ETPA:EY:DMAP ratio of 1:0.0004:0) showed moderate consumption of the methine proton after 20 h (43%), confirming the excited-state EY can efficiently undergo oxidation to induce C-S bond cleavage of TTCs under green light (Figure 5-1, b), which could be used to initiate radicals and lead to reversible activation during polymerization. When EY, ETPA, and DMAP were present

(ETPA:EY:DMAP ratio of 1:0.0004:1), faster consumption of the methine proton was observed, which confirms the excited-state EY can also induce TTC cleavage via the reductive PET pathway (Figure 5-1, a) under green-light. Furthermore, these results indicate that a sole initiation mechanism (oxidative PET-RAFT) can occur under green- light irradiation, while a photoiniferter mechanism is unavoidable under blue-light irradiation, where both the ETPA and EY can absorb.

5.2.3 Polymerizations under Blue-Light Irradiation

To test the effect of wavelength and reagents on aqueous photopolymerizations,

N,N-dimethylacrylamide (DMA) was subject to different polymerization conditions at basic pH (8.4-9.0). First, blue-light irradiation was evaluated using 2-(ethyl trithiocarbonate)-2-methylpropionic acid (CTA) and a DMA to CTA ratio of 200:1 (Table

5-2, Figure 5-3, Figure D-1). Since TTCs were observed to undergo photolysis under blue light during the trapping studies and in previous reports,31,32 an iniferter mechanism was first tested in which no EY or DMAP were added to the polymerizations. Monomer conversion reached 40% after 3 h, and experimental molecular weights were consistently ~15% higher than the theoretical values (Table 5-2, Entry 1). To test

whether an amine would significantly affect the rate and control during polymerization, a reaction using 1 eq. of DMAP relative to CTA was performed (Table 5-2, Entry 2).

Slightly slower kinetics compared to the reaction undergoing an iniferter process were observed, and the polymer dispersity remained low (Ð = 1.01-1.11) across all the polymerization conditions tested. During control experiments (Table 5-3), DMAP was observed to inhibit the auto-polymerization of DMA under blue-light irradiation, and a similar inhibition could be occurring during the PET-RAFT polymerizations as well.

Overall, these results indicate that well-defined polymers can be attained in water using blue light to induce C-S bond cleavage and initiate polymerizations, corroborating other reports of using C-S photolysis under blue light to synthesize acrolyl polymers.31,34,141,144

Table 5-2. Results of photoinduced electron/energy-transfer reversible addition- fragmentation chain transfer polymerizations of N,N-dimethylacrylamide with a DMA to CTA ratio of 200:1 and a solution pH of 8.4-9.0. a Entry CTA:EY:DMAP Irradiation Monomer Mn,theo Mn,MALLS Dispersity, color conversionb (g/mol)c (g/mol)d Ð 1 1:0:0 bluee 0.40 8320 9540 1.01 2 1:0:1 bluee 0.25 5170 7440 1.11 3 1:0.0004:0 bluee 0.87 17500 18500 1.01 4 1:0.0004:1 bluee 0.82 16500 19000 1.01 5 1:0.0004:2 bluee 0.95 19000 21500 1.02 6 1:0.001:2 bluee 0.93 18600 20300 1.02 7 1:0.001:0 greenf 0.92 18500 19400 1.02 8 1:0.001:2 greenf 0.94 18800 20500 1.05 achain transfer agent:eosin Y:4-dimethylaminopyridine, bdetermined using 1H NMR c spectroscopy, theoretical number average molecular weights (Mn,theo) calculated from monomer conversion, dnumber average molecular weights obtained using gel permeation chromatography equipped with a multi-angle light scattering detector with 0.05 M LiCl in N,N-dimethylacetamide as the eluent e11 mW/cm2, f6.2 mW/cm2.

Figure 5-3. Pseudo-first-order kinetics of photoinduced electron/energy-transfer reversible addition-fragmentation chain transfer polymerization with different ratios of [chain transfer agent]:[eosin Y]:[4-dimethylaminopyridine] performed under blue-light irradiation (11 mW/cm2).

Table 5-3. Results of Control Experiments of N,N-Dimethylacrylamide (5 M) with Different Irradiation Wavelengths and Reagents after 3 h of Irradiation. Wavelength Reagents DMA DMA:EY:DMAP conversiona Blue (450 nm) 200:0:0 0.16 200:0:1 <0.05 200:0.0004:0 0.91 200:0.0004:1 0.88 Green (520 nm) 200:0:0 <0.05 200:0:1 <0.05 200:0.0004:0 0.89 200:0.0004:1 0.98 aConversion determined by 1H NMR spectroscopy

Next, we investigated oxidative PET-RAFT polymerization under blue-light irradiation. Initial polymerizations using a CTA:EY ratio of 1:0.0004 led to high monomer conversion after 3 h (Table 5-2, Entry 3). Actual molecular weights were slightly higher

than theoretical molecular weights, although the molecular weight dispersity remained low (Ð = 1.01). DMAP was then added to favor the reductive PET-RAFT polymerization mechanism. Using a CTA:EY:DMAP ratio of 1:0.0004:1 resulted in similar rates of polymerization, with 82% monomer conversion being achieved after 3 h (Table 5-2,

Entry 4). However, molecular weights were consistently ~15% higher than theoretical molecular weights (Figure D-1), possibly due to degradation of the CTA (vide infra). The concentration of amine was found to affect the apparent rate of polymerization (Figure

D-2), and using 2 equivalents relative to CTA yielded full monomer conversion in 3 h with good control over dispersity, comparable to a previous report using organic solvents.146 Interestingly, keeping the equivalents of DMAP constant and increasing the equivalents of EY to yield ratios of CTA:EY:DMAP=1:0.001:2 (Table 5-2, Entry 5) and

1:0.002:2 did not significantly affect the rates (Figure D-3). These results indicate that, at these low concentrations of catalyst, EY is not significantly affecting the apparent rate of polymerization, but the amine can slightly affect the rate of polymerization.

In conclusion, for PET-RAFT polymerizations under blue-light irradiation, at least four different mechanisms of trithiocarbonate activation occur when CTA, tertiary amine, and EY are present: 1) TTC photolysis, 2) tertiary amine-accelerated TTC photolysis, 3) oxidative PET-RAFT, and 4) reductive PET-RAFT using the tertiary amine catalyst.

Consequently, we hypothesize the variety of mechanisms of radical initiation contributes to higher-than-predicted experimental molecular weights, although well-defined polymers with low dispersities are achieved (since degenerative chain-transfer occurs during all four).

5.2.4 Polymerizations under Green-Light Irradiation

We then tested polymerization conditions using green-light irradiation at two different pH values (4.4 and 8.4) to compare the mechanism of reductive versus oxidative PET-RAFT (Figure 5-4). In addition to involving lower energy light that may pose less problems during polymerization in the presence of sensitive systems (e.g., proteins, cells), the main advantage of using green light over blue light for RAFT polymerizations is that the contribution of homolytic photolysis to initiation should be dramatically reduced. Indeed, polymerization solutions containing only CTA or CTA and

DMAP showed negligible conversion after 3 h. Under basic conditions, when a CTA:EY ratio of 1:0.001 was used (Table 5-2, Entry 7), high conversions (92%) were reached after 3 h, indicating the PET catalyst effectively activated the TTC without reducing agent present (i.e., oxidation of the excited state EY). In contrast to all other polymerization conditions discussed up to this point, experimental molecular weights closely agreed with theoretical values until high monomer conversion, where an increase in MW can potentially be attributed to chain-chain coupling commonly observed in RAFT polymerizations in monomer-starved conditions. Additionally, these results indicate that tertiary amine is not necessary to achieve fast rates of polymerization under these conditions. To test whether the agreement between theoretical and experimental molecular weight values were due to wavelength or initiation mechanism, DMAP was added for a CTA:EY:DMAP ratio of 1:0.001:2 to access a reductive PET-RAFT mechanism (Table 5-2, Entry 8) at basic pH. Full monomer conversion was reached after 3 h, and a linear pseudo first-order kinetics plot, linear increase in molecular weights with monomer conversion, and low molar mass dispersities were observed (Ð = 1.05-1.15). However, unlike the oxidative mechanism

without DMAP, the actual molecular weights were ~10-15% higher than predicted values (analogous to polymerizations under blue-light irradiation using reductive PET-

RAFT conditions).

To determine whether any molecular weight discrepancy was occurring due to

CTA hydrolysis under basic conditions,38 polymerizations were conducted at acidic pH

(4.4, Figure 5-4). The oxidative EY conditions (with CTA:EY = 1:0.001) yielded polymers with good agreement between theoretical and experimental molecular weights.

However, the reductive EY conditions (CTA:EY:DMAP = 1:0.001:2) at a pH of 4.4 still yielded a ~15% discrepancy between theoretical and actual molecular weights, suggesting that CTA hydrolysis is not the main cause of these differences. Although the rates of polymerization were generally slower at a pH of 4.4, increasing the EY concentration (i.e., CTA:EY = 1:0.003) led to a linear pseudo-first order kinetic plot, slightly faster kinetics, and excellent agreement between theoretical and experimental molecular weights (Figure D-4). The slower rates of polymerization were attributed to protonation of the excited state catalyst147 and a higher rate of bimolecular triplet decay148, lowering the amount of active state catalyst and reducing the probability of electron transfer from the triplet state of EY to the CTA. From these results, we concluded that molecular weight control was best when the polymerization occurred exclusively via the oxidative PET-RAFT mechanism with green light.

Overall, the polymerization and trapping results provide fundamental insight into the application of PET-RAFT polymerization in water using EY as the photocatalyst.

Previous reports of aqueous polymerizations using EY involved tertiary amines as a

Figure 5-4. Conversion, pseudo first-order kinetics, and experimental molecular weight versus theoretical molecular weight for photo electron-transfer reversible addition-fragmentation chain transfer polymerizations containing different ratios of chain transfer agent:eosin Y:4-dimethylaminopyridine performed under green-light irradiation (6.2 mW/cm2). reducing agent for the excited-state EY, effectively favoring the reductive PET-RAFT mechanism (Figure 5-1, a). While this addition results in fast and linear polymerization kinetics and oxygen tolerance, the concurrence of photolysis (Figure 5-1, c) and

reduction (Figure 5-1, a) of the excited-state trithiocarbonate (in blue wavelengths) and the oxidative PET-RAFT mechanism (Figure 5-1, b) in both blue and green wavelengths can lead to multiple mechanisms of initiation. It should also be noted that while electron transfer mechanisms are discussed, proposed energy transfer mechanisms of initiation could be occurring but are difficult to distinguish from PET pathways.

Rapid consumption of the CTA is essential to attain polymers of controlled molecular weights and low dispersities. Since all the polymerization conditions described here yielded well-defined polymers, we reasoned that a considerable amount of the CTA during polymerization is consumed from degenerative chain transfer events early in the polymerization.149 Even if TTC cleavage rates differ between the mechanisms, we reasoned that degenerative chain transfer favors rapid consumption of the CTA prior to significant monomer propagation. Importantly, each initiation mechanism yields the same carbon centered propagating radical but results in different trithiocarbonyl species. Moreover, DMA was found to undergo initiation from EY, regardless of the wavelength or when tertiary amine was present, so this introduces an unavoidable route of radical generation (Table 5-3). When a kinetic time point was qualitatively evaluated by electrospray ionization mass spectrometry, the chains synthesized under reductive PET-RAFT conditions showed a population of chains initiated from the DMAP reducing agent instead of the desired CTA, but a population deriving from DMA initiation was not observed (Figure D-5). The background initiation route resulting from reducing agent intermediates150 demonstrates that reductive PET-

RAFT conditions can cause a decrease in polymer chain-end homogeneity and reduced molecular weight control. While a lower than expected molecular weight should occur if

chains are initiated from sources other than CTA-derived radicals, we hypothesize that a high concentration of radicals from the multiple routes of initiation very early in the polymerization may induce degradation of a non-negligible amount of the CTA, ultimately leading to higher-than-expected polymer molecular weights. Additionally, while we expect that only a small fraction of chains could be deleteriously affected by the differing mechanisms of initiation (e.g., initiation from DMAP), an overall beneficial increase in the characteristics attributed to controlled radical polymerizations was observed when the initiation mechanism was limited to oxidative PET-RAFT under green-light irradiation.

Another key difference in the control over polymerization in each mechanism could result from the differences in reaction intermediates and deactivation steps. Since multiple mechanisms can occur simultaneously, multiple radical termination reactions

(either reversible or irreversible) between the reaction intermediates could occur. These different modes of deactivation could lead to differences in polymerization control and discrepancies between theoretical and experimental molecular weights. Finally, additional termination events involving the reducing agent amine radical intermediates150 during reductive PET-RAFT conditions may lead to further reductions in polymerization control. Regardless of the irradiation wavelength, additional unavoidable termination events involving radicals that result from DMA initiation by EY could also be occurring.

Our investigations of the kinetics of these polymerizations also revealed loss of

EY during irradiation (most likely due to the formation of the leuco form of EY commonly formed under anaerobic conditions),151,152 as the polymerizations lost color and a

Figure 5-5. Chain-extension polymerizations of N,N-dimethylacrylamide under oxidative photo electron-transfer reversible addition-fragmentation chain transfer polymerization conditions under green-light irradiation (6.2 mW/cm2) after sequential addition of EY photocatalyst and monomer at a solution pH of 4.4. noticeable decrease in apparent rate was observed over time. Previous reports attribute

EY degradation to reaction with oxygen species153,154 or photobleaching in the presence of hydrogen sources or electron acceptors.155,156 To ensure EY degradation did not affect chain-end retention of the TTC CTA, after 3 h additional EY was added to an oxidative PET-RAFT polymerization under green light irradiation at a pH of 4.4 (Figure

5-5). Linear pseudo first-order kinetics were observed, theoretical molecular weights closely matched experimental values, and good blocking efficiency was observed. EY and DMA were added a third time to the same flask, and good chain-end retention of the CTA was again observed by SEC with a symmetric shift to lower retention times, confirming catalyst degradation did affect the resultant polymers.

5.3 Conclusions

In conclusion, we have investigated the initiation mechanisms available during aqueous PET-RAFT polymerizations. Due to the propensity of the trithiocarbonate species to undergo both catalytic reduction and independent photolysis, the only approach that is proposed to induce one mechanism of polymerization, with typical characteristics of controlled radical polymerization, are under oxidative PET-RAFT conditions during green-light irradiation with EY. The reliable control of molecular weights is paramount in controlled radical polymerizations, and elucidation that only oxidative PET-RAFT provided polymers with targeted molecular weights and fast polymerization kinetics suggests increased understanding of the role of the possible mechanisms of PET can lead to more well-defined polymeric materials.

5.4 Materials and Methods

5.4.1 Materials

4-dimethylaminopyridine (DMAP, Acros Organics, 99%), and 1-ethylpiperidine hypophosphite (EPHP, Aldrich, 95%) were used as received. 2-(ethyl trithiocarbonate)-

2-methylpropionic acid (CTA) and 2-(ethyl trithiocarbonate)propionic acid (ETPA) were synthesized according to a previous report.1 N,N-Dimethylacrylamide (DMA, 99%,

Sigma-Aldrich) was filtered through basic alumina prior to use. Eosin Y (EY, Santa Cruz

Biotechnology, 99%) was prepared as a 1 mg/mL solution in deionized (DI) water prior to use. All solvents were used as received.

For a light source, a CO-Z 3W E27 RGB LED Light Blub was placed in a desk lamp and changed to the desired polymerization color.

5.4.2 Characterization

1H NMR spectroscopy was conducted on either an Inova 500 MHz, 2 RF channel instrument or a Mercury 300 MHz 4nuc instrument at 25 C. D2O (Cambridge Isotopes

Laboratories, Inc., 99.9%) and DMSO-d6 (Cambridge Isotopes Laboratories, Inc.,

99.9%) solvents were used as received.

Size exclusion chromatography (SEC) was performed in N,N-dimethylacetamide

(DMAc) with 50 mM LiCl at 50 °C and a flow rate of 1.0 mL min-1 (Agilent isocratic pump, degasser, and autosampler, colums: Plgel 5 μm guard + two ViscoGel I-series

G3078 mixed bed columns: molecular weight range 0-20 × 103 and 1-100 × 104 g mol-

1). Detection consisted of a Wyatt Optilab T-rEX refractive index detector operating at

658 nm and a Wyatt miniDAWN Treos light scattering detector operating at 659 nm.

Absolute molecular weights and polydispersities were calculated using the Wyatt

ASTRA software and 100% mass recovery methods.

Electrospray ionization mass spectrometry was performed on a Thermo Scientific

LCQ Deca Ion Trap. Samples were prepared by taking a kinetic aliquot of a reductive

PET-RAFT polymerization after 10 min (conversion by 1H NMR spectroscopy = 9.5%,

Mn, theo = 2100 g/mol), diluting with a solution of 1% formic acid in acetonitrile, and directly injected.

5.4.3 Procedures

5.4.3.1 Solution pH preparation

For basic solution polymerizations (pH = 8.4-9.0): polymerizations containing DMAP were found to have a pH between 8.4-9.0, while polymerizations without DMAP were adjusted to pH 8.4 using 1 M NaOH.

For acidic solution polymerizations (pH = 4.4): polymerizations without DMAP were found to have a pH between 4.4-4.5, while polymerizations containing DMAP were adjusted to pH 4.4 using aqueous HCl.

5.4.3.2 Example polymerization procedure (reductive PET-RAFT under blue-light irradiation)

DMA (1.04 mL, 1.00 g, 10.1 mmol), CTA (11.3 mg, 5.05 × 10-2 mmol), EY (13.0

L, 1.30 × 10-2 mg, 2.00 × 10-5 mmol), and DMAP (12.3 mg, 0.101 mmol) were dissolved in water (2.02 mL) in a Schlenk flask. N,N-dimethylformamide (DMF, 0.100 mL) was added as an internal standard, and the solution was purged with Argon (Ar) for

15 min. Subsequently, the flask was placed over the light source approximately 5 cm away from the bulb. Time point aliquots were taken periodically using a purged needle and quenched by exposure to air. After 3 h, the flask was removed from the light source and quenched via exposure to air.

5.4.3.3 Example procedure for trapping studies (reductive PET-RAFT under blue- light irradiation)

ETPA (10.6 mg, 5.05 × 10-2 mmol), EPHP (135.7 mg, 75.7 × 10-2 mmol), DMAP

(12.3 mg, 10.1 × 10-2 mmol), and EY (13.0 L, 1.30 × 10-2 mg, 2.00 × 10-5 mmol) were dissolved in dimethylsulfoxide (2.02 mL) with DMF as an internal standard in a 10 mL

Schlenk flask. After purging the solution for 15 min with Ar, the flask was placed over the light source approximately 5 cm away from the bulb. Time point aliquots were taken after 3, 8 and 20 hours with a purged needle and quenched via exposure to air.

5.4.3.4 DMA chain extension (oxidative PET-RAFT under green-light irradiation)

DMA (1.04 mL, 1.00 g, 10.1 mmol), CTA (11.3 mg, 5.05 × 10-2 mmol), and EY

(32.4 L, 32.4 × 10-2 mg, 5.00 × 10-5 mmol) were dissolved in water (2.02 mL) in a

Schlenk flask. DMF (0.100 mL) was added as an internal standard, and the solution was purged with Ar for 15 min. Subsequently, the flask was placed over the light source approximately 5 cm away from the bulb. After 3 h, the flask was removed from the light sourced and quenched via exposure to air. A solution aliquot was removed for analysis and additional EY (32.4 L, 32.4 × 10-2 mg, 5.00 × 10-5 mmol) was added. The solution was purged with Ar for 15 min and returned to the light source. After 3 h, the flask was removed from the light sourced and quenched via exposure to air. A solution aliquot was removed for analysis and additional DMA (0.500 mL, 0.481 mg, 4.86 mmol) and EY

(32.4 L, 32.4 × 10-2 mg, 5.00 × 10-5 mmol) were added. The solution was purged with

Ar for 15 min and returned to the light source. After 3 h, the flask was removed from the light source and quenched via exposure to air.

CHAPTER 6 POLYMERIZATION-INDUCED THERMAL SELF-ASSEMBLY

6.1 Introduction

*Imparting stimuli-sensitivity introduces an interesting characteristic to amphiphilic nanoparticles, bestowing “smart” behavior that can induce a phase transition from hydrophilic to hydrophobic under specific stimulus.157 Thermoresponsive polymers,158 for example, poly(N-isopropylacrylamide) (PNIPAm), are particularly attractive as their lower critical solution temperature (LCST) can be tailored by varying molecular weight,159 incorporating comonomers,160,161 or functionalizing end-group moieties.55,162-

165 Recently, Zhong and co-workers showed that temperature responsive vesicles can be prepared by dissolving a poly(PEG-b-acrylic acid-b-NIPAm) copolymer in water above the LCST of PNIPAm with subsequent crosslinking of the acid groups.166

Monteiro and co-workers demonstrated a temperature directed morphological transformation by synthesizing a poly(NIPAm-b-styrene) copolymer with a RAFT nanoreactor technique with induced phase separations via a variation of the solubility of the styrene core.167 Stucky and co-workers reported temperature responsive polymer nanoparticles from a precipitation polymerization of NIPAm and subsequent diacrylamide core crosslinking.168 An and co-workers prepared thermoresponsive

PNIPAm nanogels via a diacrylate core crosslinking comonomer,169 while Zhang and co-workers reported multiple PNIPAm copolymers by dispersion polymerization.170,171

Armes and co-workers have considered thermoresponsive polymers for morphological changes, using upper critical solution temperature (UCST) behavior to tune the

*Adapted and reproduced from Ref. 125 with permission from the Royal Society of Chemistry.

hydrophobicity of PHPMA–containing block copolymer nanoparticles synthesized by

PISA.172,173

The versatility of PISA, especially in aqueous media,85 offers substantial promise for facile soft nanoparticle synthesis.174 However, as far as we are aware, taking advantage of the reversible phase transition of stimuli-responsive polymers for self- assembly during PISA has not been reported. Herein, we report the thermally-triggered aggregation and reorganization of a growing NIPAm polymer to form diverse self- assembled polymer morphologies that depend on the PNIPAm degree of polymerization. We refer to this approach as polymerization-induced thermal self- assembly (PITSA) and believe that it offers considerable promise for the facile synthesis of responsive polymer nanoparticles. To do this, a hydrophilic RAFT macro chain transfer agent (macro-CTA) composed of primarily N,N-dimethylacrylamide (DMA) was chain extended with NIPAm in water at 70 °C (Figure 6-1, a). By performing the chain extension above the LCST of PNIPAm, the growing PNIPAm blocks self-assembled as they became increasingly insoluble. As the growth of the PNIPAm continued, the observed morphology of the self-assembled aggregates evolved from micelles, to fused micelles/worm-like micelles, to branched worms, and eventually to vesicles (Figure 6-1, b). As opposed to previous reports of PISA that lead to aggregates that are readily characterized due to the stability provided by the solvophobic block, the assemblies prepared by PITSA are susceptible to dissociation on cooling. Therefore, to aid with characterization of the nanoparticles, we included a small fraction of acrylic acid (AA) units within the hydrophilic PDMA block to allow subsequent crosslinking with

ethylenediamine.166 The resulting polymer self-assemblies were thus stable and more readily characterized.

Figure 6-1. Polymerization-induced thermal self-assembly scheme to achieve different polymer nanoparticle morphologies a) reversible addition-fragmentation chain transfer polymerization chain-extension of a hydrophilic macro chain transfer agent at 70 °C resulted in different self-assembled polymeric nanoparticle morphologies that were subsequently crosslinked. b) Progression of polymer nanoparticle morphology with increasing hydrophobic polymer degree of polymerization.

6.2 Results and Discussion

The hydrophilic macro-CTA was synthesized by the homogeneous polymerization of DMA in DMAc at 70 °C with sequential addition of AA. DMA conversion was monitored by 1H NMR spectroscopy, and when 32% of monomer was consumed, AA was added to the reaction vessel (Figure 6-2, a). The polymerization was quenched at an overall DMA conversion of 52%, and end-group analysis of the purified polymer revealed a final composition of polyDMA34-b-poly(DMA14-co-AA6). Size exclusion chromatography (SEC) equipped with a multi-angle light scattering (MALS) detector showed a unimodal and symmetric molecular weight distribution with an absolute Mn of 5690 g/mol (ĐM = 1.05). Chain extension polymerizations with varying

feed ratios of NIPAm were carried out to investigate how the degree of polymerization of

PNIPAm affected the self-assembly of the nanoparticles. For brevity, only four of the block copolymers with varying degrees of polymerization of the PNIPAm block will be discussed in detail. Each polymerization was conducted at 15 w/w% solids in water at

70 °C for 3 h, at which point monomer conversions had reached between 82-99% by 1H

NMR spectroscopy. The variance in monomer conversions could be attributed to the varying [monomer]:[CTA]:[initiator] ratios. By conducting the chain extension above the

LCST of the growing PNIPAm block, the developing block copolymers self-assembled during polymerization, which was apparent from the reaction mixture gradually transitioning from homogeneous to heterogeneous as conversion increased. The viscosity of the solutions appeared to reach a maximum at the intermediate PNIPAm

172 DPn,theo presumably from worm entanglements. The polymerizations were quenched while holding the temperature above 70 °C, ensuring that the newly formed nanoparticle morphologies remained intact (Figure 6-2, b).

Figure 6-2. Synthesis of thermoresponsive block copolymers a) macro chain transfer agent synthesis of polyDMA34-b-poly(DMA14-co-AA6) by RAFT (co)polymerization of N,N-dimethylacrylamide (DMA) and acrylic acid (AA) b) RAFT chain-extension using N-isopropylacrylamide and subsequent crosslinking.

The molecular weights of the block copolymers were determined by taking an aliquot from the polymerization mixture, purifying via dialysis, and conducting SEC-

MALS. A linear increase in the absolute molecular weight was observed with increasing

NIPAm feed ratio (Figure 6-3, a). While traditional PISA often leads to polymers with somewhat broad molecular weight distributions and/or multimodal SEC traces with high molecular weight shoulders, all of the block copolymers formed by this new approach were unimodal and of low molar mass dispersity (ĐM < 1.12) (Figure 6-3, b).

Figure 6-3. Molecular weight evolution during the polymerization-induced thermal self- assembly (PITSA) of polyDMA34-b-poly(DMA14-co-AA6)-b-polyNIPAmn by the chain extension of a macro chain transfer agent containing N,N- dimethylacrylamide (DMA) and acrylic acid (AA), polyDMA34-b-poly(DMA14- co-AA6), with N-isopropylacrylamide (NIPAm). a) Absolute number-average molecular weight (Mn) as a function of NIPAm feed ratio b) SEC chromatograms as a function of PNIPAm degree of polymerization (PNIPAm DPn).

These results suggest that the block copolymers prepared by PITSA were well-defined and indicative of excellent blocking efficiency, even at high monomer conversions

(Figure 6-3, b). The enhanced control observed here is potentially the result of a low concentration of radicals during polymerization and/or high mobility of the propagating chains within the nanoparticle cores due to the presence of water in the core, which can serve to plasticize the growing PNIPAm block.175 The typical methods used to analyze block copolymer assemblies require either sample preparation/characterization at ambient temperature. Therefore, it was necessary to crosslink the nanoparticles formed during PITSA immediately after polymerization and prior to cooling. The pendant acid groups of the acrylic acid units in the hydrophilic shell of the nanoparticles were crosslinked by addition of ethylenediamine in the presence of 1-ethyl-3-(3- dimethylaminopropyl)carbodiimide (EDC) and catalytic 4-dimethylaminopyridine

(DMAP) (Figure 6-2, b). These reactions were conducted at polymerization temperatures and immediately after the polymerizations were quenched by opening the reaction mixture to air. The polymerization solutions remained viscous after crosslinking, but subsequently changed from yellow to colorless due to aminolysis of the trithiocarbonate RAFT end group. Upon cooling to ambient temperature, the viscosity reduced, implying dissolution of some of the nanoparticles, likely due to the inherently low efficiency of EDC coupling in neutral water.176 Dynamic light scattering (DLS) analysis of the reaction mixtures showed distributions that could be attributed to the presence of both unimers and nanoparticles with larger hydrodynamic diameters (Dh).

To remove the unimer impurities, the nanoparticles were purified via dialysis and lyophilization. DLS measurements of the purified nanoparticles showed predominantly

distributions of higher Dh, suggesting primarily the larger nanoparticles remained (Figure

6-4).

Figure 6-4. Dynamic light scattering measurements of purified nanoparticles with varying poly(N-isopropylacrylamide) degrees of polymerization (PNIPAm DPn).

The nanoparticles with PNIPAm blocks of DPn = 52, 101, and 137 resulted in large size distributions that might be expected of higher order assemblies, such as worms and vesicles. The resulting DLS histogram for the nanoparticles from the block copolymer with PNIPAm with DPn of 137 showed a distribution that also contained a population of small particles, which might result from residual unimers or coupled chains. It should be noted that while these results indicate the presence of particles of a size range that might be expected for self-assembled block copolymers, the nonspherical and multimodal nature of these distributions suggests these sizes are approximate.

To gain further insight into the morphology of the block copolymer aggregates formed during the polymerization, transmission electron microscopy (TEM) was conducted after crosslinking of the nanoparticles formed in situ by reacting with ethylenediamine. As the PNIPAm DPn increased, the predominant nanoparticle morphology changed. TEM images of the cooled reaction solution containing the

crosslinked particles revealed multiple nanoparticle morphologies (Figure 6-5).

However, smaller particles (<10 nm) in the background were also present. These small features were attributed to residual polymeric impurities that arise from incomplete crosslinking and dead polymeric chains.72 Nevertheless, the majority of these impurities were successfully removed by dialysis, as evidenced by DLS and TEM.

Figure 6-5. Transmission electron microscopy of the unpurified crosslinked polymer aggregates showing the progression of nanoparticle morphology as the poly(N-isopropylacrylamide) degree of polymerization (PNIPAm DPn) increased. (2% uranyl acetate aqueous solution negative stain).

With a PNIPAm DPn = 11, uniform micelles of approximately 20 nm diameter were obtained. The PNIPAm DPn = 52 resulted in a mixture of micelles and fused micelles/worms. The observed micelles ranged in size from 20 to 40 nm, whereas multiple sizes of fused micelles and worms were observed. As the PNIPAm chains reached a DPn = 101, predominantly worms and branched worms were observed, with minimal residual micelles being present. By increasing the DPn of the PNIPAm to 137,

structures that appeared to be vesicle-like were visible. Combinations of the discussed polymeric morphologies were also observed in other examples of PITSA.

Kinetic studies of the aqueous NIPAm chain extensions were conducted to gain further insight into the nature of the polymerization. Five separate polymerizations with identical stoichiometry ([NIPAm]:[polyDMA34-b-poly(DMA14-co-AA6) macro-CTA]:[I] =

121:1:0.04) were conducted at 70 °C. A separate polymerization was quenched every

30 min, and NIPAm conversion was monitored by 1H NMR spectroscopy. The first-order kinetic plots suggested a constant rate of polymerization for the first ~120 min, after which the rate of polymerization appeared to increase (Figure 6-6, a). The observed increase in polymerization rate was also observed in previous PISA systems and is attributed to the self-assembly of growing solvophobic chains, which leads to an environment conducive to monomer solubilization.174 Once this occurs, the monomer is thought to swell the assembled regions of solvophobic chains. The high local monomer concentration that results from this leads to an increased rate of polymerization.

Therefore, the point at which the change in rate was observed provides insight into the critical solution DPn of the PNIPAm block at which self-assembly occurs at 70 °C. The evolution of Mn versus NIPAm conversion showed an approximately linear increase in molecular weight at low conversions, with slightly higher than expected molecular weights being present at elevated conversion (Figure 6-6, a, inset). The corresponding

SEC traces shifted to higher molecular weights during the polymerization, with unimodal distributions and good macro-CTA blocking efficiency being apparent (Figure 6-6, b).

Therefore, despite the induced nucleation of the PNIPAm into hydrophobic regimes, the polymerization still proceeded with good control over the molecular weight and

dispersity while maintaining the mediating trithiocarbonate RAFT groups up to high monomer conversions.

Figure 6-6. Data from the PITSA chain extension of a N,N-dimethylacrylamide (DMA) and acrylic acid (AA) containing macro chain transfer agent, [polyDMA34-b- poly(DMA14-co-AA6)], with N-isopropylacrylamide (NIPAm) at 70 °C. a) First- order kinetics plot showing an increase in rate during the polymerization. Inset: Mn vs. conversion. b) SEC traces showing unimodal distributions shifting to higher molecular weights as the polymerization proceeds.

6.3 Conclusions

This report demonstrated for the first time polymerization-induced thermal self- assembly (PITSA), using the triggered hydrophobic self-assembly of PNIPAm moieties to induce nanoparticle reorganization during polymerization in an aqueous medium.

PISA is typically limited to polymer/solvent combinations that allow monomer solubility and polymer insolubility in a permanently selective solvent. The PITSA approach reported here expands this highly useful process to allow for the preparation of

100

responsive polymer nanoparticles by exploiting the temperature-sensitivity of PNIPAm in water. Given the variety of stimuli-responsive polymers that have been reported, this approach dramatically expands the library of possible PISA systems and introduces possible new approaches to synthesizing “smart” polymeric nanoparticle delivery vehicles using PITSA (“PITSA delivery”). Although our studies have focused exclusively on a thermoresponsive polymer with LCST-type behavior, this same general strategy should be applicable to temperature-sensitive polymers that demonstrate UCST behavior as well. Moreover, polymers that demonstrate responsive behavior to other stimuli, including changes in pH or salt concentration, irradiation with light, etc. may also be amenable slices to the stimuli-responsive PISA pie.

6.4 Materials and Methods

6.4.1 Materials

4-Dimethylaminopyridine (DMAP, > 99%, Sigma-Aldrich), 4,4´- azobiscyanovaleric acid (ACVA, 98%, Alfa-Aesar), ethylenediamine (99%, Alfa Aesar),

N,N-dimethylacetamide (DMAc, > 99%, Fisher Scientific), trioxane (> 99.5%, Acros

Organics), 1-ethyl-3-(3-dimethylaminopropyl)carbodiimideHCl (96%, Combi-Blocks) were used as received. S-Ethyl-S´-(α,α´-dimethyl-α˝-acetic acid)-trithiocarbonate chain- transfer agent (CTA) was synthesized according to a previously published report.1 2,2´-

Azobisisobutyrlnitrile (AIBN, 98%, Sigma-Aldrich) was recrystallized 3× from ethanol prior to use. N-Isopropylacrylamide (NIPAm, > 98%, TCI Chemicals) was recrystallized

3× from hexanes prior to use. N,N-Dimethylacrylamide (DMA, 99%, Sigma-Aldrich) and acrylic acid (AA, 99.5%, Alfa Aesar) were filtered through basic alumina prior to use.

101

6.4.2 Characterization

1H NMR spectroscopy was conducted on an Inova 500 MHz, 2 RF channel instrument at 25 °C. Chloroform-d (Cambridge Isotopes Laboratories, Inc, 99.8%),

DMSO-d6 (Cambridge Isotopes Laboratories, Inc, 99.9%), and D2O (Cambridge

Isotopes Laboratories, Inc, 99.9%) solvents were used as received.

Size exclusion chromatography (SEC) was performed in dimethylacetamide

(DMAc) with 50 mM LiCl at 50 °C and a flow rate of 1.0 mL min-1 (Agilent isocratic pump, degasser, and autosampler, colums: Plgel 5 μm guard + two ViscoGel I-series

G3078 mixed bed columns: molecular weight range 0-20 × 103 and 1-100 × 104 g mol-

1). Detection consisted of a Wyatt Optilab T-rEX refractive index detector operating at

658 nm and a Wyatt miniDAWN Treos light scattering detector operating at 659 nm.

Absolute molecular weights and polydispersities were calculated using the Wyatt

ASTRA software and 100% mass recovery methods. Prior to absolute molecular weight determination, samples with NIPAm DPn,theo = 26, 49, 73, and 121 were purified via dialysis against DI water for 4 days (Spectra/Por 3 Dialysis Membranes (3500 MWCO) from Spectrum Laboratories) before being isolated by then freeze-drying.

Dynamic light scattering analysis was performed with a Malvern Zetasizer Nano

ZS (Model No. ZEN 3600, Malvern Instruments Ltd., Worcestershire UK) at 25 °C.

Purified samples were prepared by dialyzing against DI water for 7 days using

Spectra/Por Float-A-Lyzer G2 100k MWCO 5 mL tubes from Spectrum Labs and subsequently freeze-dried. All samples were diluted to 5 mg/mL solutions and placed in a quartz low-volume cuvette for analysis.

Transmission electron microscopy was conducted on an H-700 from Hitachi High

Technologies America, Inc., Schaumber, IL USA. Digital images were acquired with a

102

Veleta 2k × 2k camera and iTEM software (Olympus Soft-Imaging Solutions Corp.,

Lakewood, CO). Electron Microscopy Sciences Formvar Carbon Film on 400 mesh nickel grids (FCF400-Ni) were used for all measurements. For unpurified samples, 10

μL polymerization solution sample diluted to 2.5 mg/mL was placed on a grid for 1 min.

The excess solvent was wicked off and the grid air-dried. For staining, a drop of 2% uranyl acetate in water was placed on the grid and allowed to sit for 45 s. The excess solvent was wicked off and the grid air-dried. Purified samples from dialysis were diluted to 0.25 mg/mL, and TEM grids were prepared analogous to unpurified samples. For osmium tetroxide staining, a purified 0.125 mg/mL nanoparticle solution was placed on a grid for 1 min. The excess solvent was wicked off and the grid air-dried. The grid was then stained in an osmium tetroxide vapor chamber for 2 h.

6.4.3 Procedures

6.4.3.1 Synthesis of PDMA-AA macro chain transfer agent

DMA (8.00 g, 80.8 mmol), CTA (181 mg, 0.808 mmol), and AIBN (6.60 mg, 0.404 mmol) were placed in a Schlenk flask to yield a [DMA]:[CTA]:[I] ratio = [100]:[1]:[0.05].

DMAc (20 mL) was added to make the monomer concentration 5 M, and 1,3,5-trioxane

(120 mg) was added as an internal standard. The flask was sealed with a glass stopper, and a rubber septum was placed over the arm joint. The reaction was subject to 3 freeze-pump-thaw cycles, and left to stir at 70 °C. DMA conversion was monitored by 1H

NMR spectroscopy (chloroform-d, 500 MHz). Meanwhile, acrylic acid (1.75 g, 24.2 mmol) was purged with N2 in a separate flask for 30 min. When the DMA polymerization had been stirring for 60 min (32% monomer conversion by 1H NMR spectroscopy), the acrylic acid was cannulated into the reaction vessel. After 90 min (53% monomer conversion by 1H NMR spectroscopy), the flask was removed from heat, opened to the

103

atmosphere, and placed in a cold-water bath to stir. The solution was added drop-wise to cold diethylether to precipitate a yellow polymer. The polymer was subsequently re- precipitated into cold diethylether from THF and vacuum dried, yielding 4.07 g yellow

1 powder. H NMR spectroscopy end group analysis (DMSO-d6, 500 MHz) allowed a comparison of the methylene peak next to the trithiocarbonate group (t, 1H, 3.61 ppm) to the acrylic acid proton (s, 11.5 – 12.5 ppm) and indicated a polymer composition of

PDMA48AA6 (Mn,NMR = 5410 g/mol) while GPCMALS absolute molecular weight analysis found an Mn = 5690 g/mol and ĐM = 1.05. Final macro-CTA composition from conversion and end-group analysis: polyDMA34-b-poly(DMA14-co-AA6).

6.4.3.2 Synthesis of nanoparticles with a composition of PDMA34-b-P(DMA14-co- AA6)-b-PNIPAm26

The macro-CTA polyDMA34-b-poly(DMA14-co-AA6) (70.0 mg, 0.0123 mmol),

NIPAm (45.2 mg, 0.400 mmol), ACVA (0.140 mg, 5.00 × 10-4 mmol, 1.00 mg/mL solution in DI water) were placed in a vial with a rubber septum to give a

[NIPAm]:[CTA]:[I] ratio of [32]:[1]:[0.04]. DI water (0.654 g) was added to achieve a 15 w/w% solids concentration. The solution was deoxygenated by purging with N2 for 30 min, then placed in a heating block at 70 °C. After 3 h, the vial was opened to the atmosphere, while still in the heating block, and subsequently stirred for 30 min to ensure reaction quenching. An aliquot (200. μL) of solution was taken for 1H NMR spectroscopy (D2O, 500 MHz) and SEC analysis. A conversion of 84% was calculated by integrating the methyne (m, 1H, 3.90 ppm) and vinyl peaks (d, 1H, 5.65 ppm) on

NIPAm to the methyne peak on PNIPAm (s, 1H, 3.80 ppm). This gave a PNIPAm

DPn,theo = 26. The aliquot was dialyzed and freeze-dried, and an absolute number

104

average molecular weight of Mn = 7020 g/mol, ĐM = 1.12, and PNIPAm DPn = 11 was found by MALS.

While the quenched reaction vial was still in the heating block at 70 °C, a solution of EDC (28.7 mg, 0.150 mmol) and DMAP (2.44 mg, 0.0200 mmol, 10.0 mg/mL solution in DI water) was prepared in a separate vial. The solution was placed on the heating block for 10 min to reach 70 °C and subsequently added to the reaction vessel. After 30 min, ethylenediamine (2.16 mg, 0.0360 mmol) was added, and the solution was left to stir overnight at 70 °C. The amount of EDC was calculated by using 2× excess theoretical moles of acrylic acid (0.0738 mmol), calculated from the polymer composition (6 AA units) and amount of macro-CTA added (0.0123 mmol), while the amount ethylenediamine added was calculated as 0.5 equivalents of theoretical moles of AA.

6.4.3.3 Synthesis of nanoparticles with a composition of PDMA34-b-P(DMA14-co- AA6)-b-PNIPAm38

The macro-CTA polyDMA34-b-poly(DMA14-co-AA6) (70.0 mg, 0.0123 mmol),

NIPAm (56.5 mg, 0.500 mmol), ACVA (0.140 mg, 5.00 × 10-4 mmol, 1.00 mg/mL solution in DI water) were placed in a vial with a rubber septum to give a

[NIPAm]:[CTA]:[I] ratio of [38]:[1]:[0.04]. DI water (0.718 g) was added to achieve a 15 w/w% solids concentration. The solution was deoxygenated by purging with N2 for 30 min, then placed in a heating block at 70 °C. After 3 h, the vial was opened to the atmosphere, while still in the heating block, and subsequently stirred for 30 min to ensure reaction quenching. An aliquot (200. μL) of solution was taken for 1H NMR spectroscopy (D2O, 500 MHz) and SEC analysis. A conversion of 92% was calculated by integrating the methyne (m, 1H, 3.90 ppm) and vinyl peaks (d, 1H, 5.65 ppm) on

105

NIPAm to the methyne peak on PNIPAm (s, 1H, 3.80 ppm). This gave a PNIPAm

DPn,theo = 38.

6.4.3.4 Synthesis of nanoparticles with a composition of PDMA34-b-P(DMA14-co- AA6)-b-PNIPAm49

The macro-CTA polyDMA34-b-poly(DMA14-co-AA6) (70.0 mg, 0.0123 mmol),

NIPAm (67.8 mg, 0.600 mmol), ACVA (0.140 mg, 5.00 × 10-4 mmol, 1.00 mg/mL solution in DI water) were placed in a vial with a rubber septum to give a

[NIPAm]:[CTA]:[I] ratio of [49]:[1]:[0.04]. DI water (0.781 g) was added to achieve a 15 w/w% solids concentration. The solution was deoxygenated by purging with N2 for 30 min, then placed in a heating block at 70 °C. After 3 h, the vial was opened to the atmosphere, while still in the heating block, and subsequently stirred for 30 min to ensure reaction quenching. An aliquot (200. μL) of solution was taken for 1H NMR spectroscopy (D2O, 500 MHz) and SEC analysis. A conversion of 99% was calculated by integrating the methyne (m, 1H, 3.90 ppm) and vinyl peaks (d, 1H, 5.65 ppm) on

106

NIPAm to the methyne peak on PNIPAm (s, 1H, 3.80 ppm). This gave a PNIPAm

DPn,theo = 49. The aliquot was dialyzed and freeze-dried, and an absolute number average molecular weight of Mn = 11600 g/mol, ĐM = 1.12, and PNIPAm DPn = 52 was found by MALS.

6.4.3.5 Synthesis of nanoparticles with a composition of PDMA34-b-P(DMA14-co- AA6)-b-PNIPAm61

The macro-CTA polyDMA34-b-poly(DMA14-co-AA6) (70.0 mg, 0.0123 mmol),

NIPAm (90.4 mg, 0.800 mmol), ACVA (0.140 mg, 5.00 × 10-4 mmol, 1.00 mg/mL solution in DI water) were placed in a vial with a rubber septum to give a

[NIPAm]:[CTA]:[I] ratio of [65]:[1]:[0.04]. DI water (0.910 g) was added to achieve a 15 w/w% solids concentration. The solution was deoxygenated by purging with N2 for 30 min, then placed in a heating block at 70 °C. After 3 h, the vial was opened to the atmosphere, while still in the heating block, and subsequently stirred for 30 min to ensure reaction quenching. An aliquot (200. μL) of solution was taken for 1H NMR

107

spectroscopy (D2O, 500 MHz) and SEC analysis. A conversion of 94% was calculated by integrating the methyne (m, 1H, 3.90 ppm) and vinyl peaks (d, 1H, 5.65 ppm) on

NIPAm to the methyne peak on PNIPAm (s, 1H, 3.80 ppm). This gave a PNIPAm

DPn,theo = 61.

6.4.3.6 Synthesis of nanoparticles with a composition of PDMA34-b-P(DMA14-co- AA6)-b-PNIPAm73

The macro-CTA polyDMA34-b-poly(DMA14-co-AA6) (70.0 mg, 0.0123 mmol),

NIPAm (102 mg, 0.900 mmol), ACVA (0.140 mg, 5.00 × 10-4 mmol, 1.00 mg/mL solution in DI water) were placed in a vial with a rubber septum to give a [NIPAm]:[CTA]:[I] ratio of [73]:[1]:[0.04]. DI water (0.975 g) was added to achieve a 15 w/w% solids concentration. The solution was deoxygenated by purging with N2 for 30 min, then placed in a heating block at 70 °C. After 3 h, the vial was opened to the atmosphere, while still in the heating block, and subsequently stirred for 30 min to ensure reaction

1 quenching. An aliquot (200. μL) of solution was taken for H NMR spectroscopy (D2O,

108

500 MHz) and SEC analysis. A conversion of >99% was calculated by integrating the methyne (m, 1H, 3.90 ppm) and vinyl peaks (d, 1H, 5.65 ppm) on NIPAm to the methyne peak on PNIPAm (s, 1H, 3.80 ppm). This gives a PNIPAm DPn,theo = 73. The aliquot was dialyzed and freeze-dried, and an absolute number average molecular weight of Mn = 17100 g/mol, ĐM = 1.07, and PNIPAm DPn = 101 was found by MALS.

6.4.3.7 Synthesis of nanoparticles with a composition of PDMA34-b-P(DMA14-co- AA6)-b-PNIPAm98

The macro-CTA polyDMA34-b-poly(DMA14-co-AA6) (70.0 mg, 0.0123 mmol),

NIPAm (136 mg, 1.20 mmol), ACVA (0.140 mg, 5.00 × 10-4 mmol, 1.00 mg/mL solution in DI water) were placed in a vial with a rubber septum to give a [NIPAm]:[CTA]:[I] ratio of [98]:[1]:[0.04]. DI water (1.17 g) was added to achieve a 15 w/w% solids concentration. The solution was deoxygenated by purging with N2 for 30 min, then placed in a heating block at 70 °C. After 3 h, the vial was opened to the atmosphere, while still in the heating block, and subsequently stirred for 30 min to ensure reaction

109

1 quenching. An aliquot (200. μL) of solution was taken for H NMR spectroscopy (D2O,

6.4.3.8 Synthesis of nanoparticles with a composition of PDMA34-b-P(DMA14-co- AA6)-b-PNIPAm121

The macro-CTA polyDMA34-b-poly(DMA14-co-AA6) (70.0 mg, 0.0123 mmol),

NIPAm (168 mg, 1.49 mmol), ACVA (0.140 mg, 5.00 × 10-4 mmol, 1.00 mg/mL solution in DI water) were placed in a vial with a rubber septum to give a [NIPAm]:[CTA]:[I] ratio of [121]:[1]:[0.04]. DI water (1.35 g) was added to achieve a 15 w/w% solids concentration. The solution was deoxygenated by purging with N2 for 30 min, then placed in a heating block at 70 °C. After 3 h, the vial was opened to the atmosphere, while still in the heating block, and subsequently stirred for 30 min to ensure reaction

1 quenching. An aliquot (200. μL) of solution was taken for H NMR spectroscopy (D2O,

110

500 MHz) and SEC analysis. A conversion of >99% was calculated by integrating the methyne (m, 1H, 3.90 ppm) and vinyl peaks (d, 1H, 5.65 ppm) on NIPAm to the methyne peak on PNIPAm (s, 1H, 3.80 ppm). This gives a PNIPAm DPn,theo = 121. The aliquot was dialyzed and freeze-dried, and an absolute number average molecular weight of Mn = 21200 g/mol, ĐM = 1.06, and PNIPAm DPn = 137 was found by MALS.

6.4.3.9 Kinetics of PNIPAM chain-extension

The macro-CTA polyDMA34-b-poly(DMA14-co-AA6) (350. mg, 0.0615 mmol),

NIPAm (841 mg, 7.45 mmol), and ACVA (0.700 mg, 2.50 × 10-3 mmol, 1.00 mg/mL solution in DI water) ([NIPAm]:[CTA]:[I] ratio of [121]:[1]:[0.04]) were placed in a vial. DI water (6.76 g) was added to achieve a 15 w/w% solids concentration with 1,3,5-trioxane

(215 mg) as an internal standard. The solution was divided into six vials containing 1.20 mL each. The vials were deoxygenated by purging with N2 for 30 min then placed in a heating block at 70 °C. Every 30 min a vial was removed from the heat, quenched by

111

opening to the atmosphere in a cold water bath, and analyzed by 1H NMR spectroscopy

(CDCl3, 500 MHz) and SEC for conversion and molecular weight.

112

CHAPTER 7 TUNING HYDROPHOBICITY TO PROGRAM BLOCK COPOLYMER ASSEMBLIES FROM THE INSIDE OUT

7.1 Introduction

*Initially, we sought to expand the route of synthesizing thermoresponsive systems to access block copolymers that self-assemble and dissociate at predetermined temperatures dictated by the composition of the responsive block. N,N-

Dimethylacrylamide (DMA) is a hydrophilic monomer which yields a hydrophilic polymer, while diacetone acrylamide (DAAm) is a hydrophilic monomer that leads to a hydrophobic polymer and has recently been utilized by An and others for PISA. 177-179

Additionally, the hydrophobicity of copolymers containing DMA and DAAm can be discretely tuned by varying the comonomer ratio during polymerization. 160,180,181 We hypothesized that if these copolymers composed the core-forming block of polymer aggregates, the resulting nanoparticles would show composition-dependent thermoresponsive behaviors. Indeed, we observed that the thermal stability of the assemblies formed during PISA was directly dependent on the ratio of DMA to DAAm.

More interestingly, minor changes in the monomer feed ratio and the resulting variation in hydrophobicity also led to dramatic differences in aggregate morphology. The effect of comonomer ratio within the core-forming block during PISA has only been considered relatively recently.182-185

The PISA mechanism in water (Figure 1-5) begins as chain extension from a water-soluble homopolymer with a water-soluble monomer that yields a hydrophobic

*Adapted and reproduced with permission from Macromolecules 2017, 50, 935-943. Copyright 2017 American Chemical Society. Data in Table 7-3 was collected by Dr. Kyle C. Bentz under the supervision of Prof. Daniel A. Savin. Data in Table 7-4 was collected with the help of Dr. R. Nicholas Carmean under the supervision of Prof. Brent S. Sumerlin.

113

block.174 When the second block reaches a critical degree of polymerization, the growing amphiphilic chains assemble in situ to form micelle-like morphologies, thus transitioning the homogeneous system into a dispersion polymerization. The increasing chain length of the hydrophobic block causes the copolymers to rearrange into thermodynamically favored morphologies dictated by the packing parameter, the degree of polymerizations of the hydrophobic block, and the interfacial curvature and surface tension between the hydrophilic and hydrophobic blocks.186 As we changed the composition of the core-forming block by varying the monomer feed ratio, we observed differences in the morphological progression during the course of the polymerization.

These results suggested aggregate morphology was being dictated by the inherent hydrophobicity of the block copolymers, a phenomenon that affects chain mobility and unimer exchange between micelles.187-190 Therefore, we realized that our system provided an opportunity to explore another fundamental aspect of block copolymer self- assembly and reorganization: how hydrophobicity affects nanoparticle morphology during PISA.

In the work presented here, a hydrophilic poly(N,N-dimethylacrylamide) (PDMA) macro chain transfer agent (macro-CTA) was first synthesized by RAFT polymerization to high monomer conversion (Figure 7-2, a). In a one-pot approach,191,192 this macro-

CTA was chain extended with varying molar ratios of DAAm to DMA to result in statistical copolymer segments with compositions dependent on the monomer feed ratio

(Figure 7-2, b). Various nanoparticle morphologies were observed (e.g., micelles, worms, branched worms, or vesicles) with a remarkable dependence on polymer composition.

114

Figure 7-2. Reversible addition-fragmentation chain transfer polymerization schemes for PISA a) N,N-dimethylacrylamide (DMA) polymerization to yield the hydrophilic macro chain transfer agent (macro-CTA). b) Polymerization-induced thermal self-assembly aqueous chain extension of the macro-CTA with DMA and diacetone acrylamide (DAAm), and subsequent crosslinking of the DAAm groups to provide stable nanostructures.

In addition to control over nanoparticle morphology, the proposed changes in aggregate thermodynamics through comonomer modulation provided unprecedented evidence of controlled worm growth during PISA—a phenomenon unrealized in most self-assembly techniques in selective solvents,193 except seeded crystallization-driven self-assembly.194 Worm-worm transitions of polymeric nanoparticles have been reported following self-assembly, 195-204 but as far as we are aware, there have been no reports of controlling worm length to the extent described here for amphiphilic block copolymer in solution. Here, we present our results on the dramatic morphology changes that result from small variations in block copolymer composition, and more broadly, provide insight into how tuning block copolymer solvophilicity can lead to extraordinary control over polymer self-assembly and reorganization.

115

7.2 Results and Discussion

A PDMA macro-CTA (Mw,MALS = 8040 g/mol, ĐM = 1.17) was synthesized in water, reaching full monomer conversion (>95% by 1H NMR spectroscopy) after 3 h.

Through a one-pot polymerization approach, the macro-CTA solution was diluted with an aqueous monomer solution to 15 w/w% concentration. Polymerization kinetics using monomer feed ratios of 90% DAAm to 10% DMA (DAAm90, Figure E-1), 85% DAAm to

15% DMA (DAAm85, Figure E-2), 80% DAAm to 20% DMA (DAAm80, Figure E-3), and

75% DAAm to 25% DMA (DAAm75, Figure E-4) were monitored at 70 °C to determine the copolymerization behavior of the comonomers. Similar reactivity ratios (rDAAm = 0.86 and rDMA = 0.80) of the monomers were found that suggested relatively statistical monomer incorporation during PISA.

The compositional effects of the block copolymers on the resultant morphologies were studied by varying both comonomer feed ratio and the degree of polymerization of the second block (DP2). Copolymerizations targeting a specific DP2 and containing monomer feed ratios of DAAm90, DAAm85, DAAm80, or DAAm75 were left for 16 h so that each polymerization achieved >95% monomer conversion. Aliquots from the polymerizations were taken periodically, diluted with DMAc (i.e., a non-selective solvent), and characterized by 1H NMR spectroscopy for monomer conversion and size- exclusion chromatography with multi-angle light scattering detection (SEC-MALS) for molecular weights of the unimers. Because the block copolymer aggregates that formed during chain extension of the macro-CTA with DMA and DAAm were thermoresponsive (i.e., susceptible to dissociation on cooling), the aggregates that resulted during PISA were crosslinked immediately after polymerization at 70 C to

116

facilitate characterization of morphology and size. Crosslinking of the hydrophobic block was achieved by reacting a difunctional alkoxyamine with the ketone functional groups of the DAAm units to result in hydrolytically stable oxime linkages.205-207 These “locked” nanostructures could be readily characterized by dynamic light scattering (DLS), SEC-

MALS, and transmission electron microscopy (TEM). DAAm75 was the lowest incorporation of DAAm considered, as polymerizations with lower ratios of the hydrophobic monomer gelled during crosslinking, presumably from blocks not being sufficiently hydrophobic to induce phase separation.

The unimer weight-average molecular weights (Mw,unimer) of the second block were determined by SEC-MALS. Purified aliquots of the polymerization solution prior to crosslinking offered evidence of excellent blocking efficiency and low polymer molar mass dispersities by SEC-MALS for all compositions at DP2 = 50, 80, 130, and 200

(Figure 7-3, a-d, solid lines). Importantly, the unimers targeting the same DP2, but with different compositions, possessed similar molecular weights regardless of the resultant morphology (Table 7-1). For example, unimers with DP2 of 141 (Table 7-1, entries 9-12) all had absolute molecular weights of approximately 22,000-24,000 g/mol, facilitating evaluation of aggregate morphology based solely on inherent chain hydrophobicity.

Following crosslinking, the reactions were cooled to room temperature before

DLS size measurements were performed (Figures F-1, F-2, F-3, F-4). Variation in observed nanoparticle sizes and size distributions (i.e., unimodal versus multimodal) across similar unimer molecular weights suggested that increasing the amount of DMA in the copolymer significantly affected morphology. The nanoparticle SEC traces (Figure

7-3, a-d, dotted lines) were consistent with the size distributions observed during DLS

117

Figure 7-3. Size exclusion chromatography characterization of self-assembled nanoparticles formed during polymerization-induced thermal self-assembly (after crosslinking) a) diacetone acrylamide (DAAm) to N,N- dimethylacrylamide (DMA) monomer feed ratio of 90:10 (DAAm90); b) DAAm to DMA monomer feed ratio of 85:15 (DAAm85); c) DAAm to DMA monomer feed ratio of 80:20 (DAAm85); d) DAAm to DMA monomer feed ratio of 75:25 (DAAm75). characterization and further suggested that monomer feed composition had a large effect on the resultant nanoparticles. Inevitably, some SEC results may be affected by the variety of anisotropic structures that could be present, but qualitative comparisons can still be made. When an intermediate value of DP2 for DAAm80 and DAAm75 was characterized, the nanoparticles eluted at shorter retention times with broad and multimodal distributions. Subsequently, for polymerizations with higher DP2,

118

Table 7-1. Characterization data of nanoparticle constituent unimers obtained using size exclusion chromatography equipped with a multi-angle laster light scattering detector. Entry Second block %DAAm in Unimer weight- degree of monomer feed average molecular polymerization, weight, Mw, unimer ( a DP2 103 g/mol) 1 54 90 11.6 2 85 11.7 3 80 11.3 4 75 14.3 5 87 90 15.6 6 85 15.6 7 80 18.0 8 75 18.7 9 141 90 21.7 10 85 22.2 11 80 23.8 12 75 22.2 13 217 90 35.2 14 85 33.1 15 80 30.6 16 75 39.5 aAccording to initial monomer feed ratio and >95% monomer conversion by 1H NMR spectroscopy. the nanoparticles eluted at longer retention times, and symmetric traces were observed.

This change in elution time and distribution shape could be indicative of an anisotropic- isotropic morphology change that would accompany a worm-to-sphere transition.

Nanoparticle traces for DAAm90 and DAAm85 contrasted these results where a steady decrease in retention time was observed according to increasing DP2. Additional discussion on the effect of unimer composition on nanoparticle weight-average molecular weight (Mw,NP), aggregation number (Nagg), and radius of gyration (Rg) attained during nanoparticle MALS is in the supplementary information. One conclusion drawn from the constant increases in Mw,NP, Nagg, and Rg during MALS analysis is that, regardless of composition or DP2, the nanostructures may not be kinetically trapped and

119

instead favor reorganization. This reorganization would cause the chains to adopt more thermodynamically favorable conformations, agreeing with recent SAXS studies by

Armes.208 However, only correlational conclusions can be drawn using MALS, because the apparent variety of nanoparticle shapes resulted in difficult characterization and data analysis. Fortunately, this characterization limitation between nanoparticles possessing a similar unimer DP2 provides compelling evidence that chain composition drastically affected the morphology.

TEM imaging is perhaps the most useful characterization technique to understand the morphological transitions of polymer chains during aggregation and reassembly (although the technique is limited by sample size). As with the DLS and

SEC-MALS results, we observed dramatic effects of copolymer composition on nanoparticle morphology. DAAm90 exhibited only spherical micelle structures at all DP2 values analyzed (Figure 7-4, a-d). Though vesicles have been reported for copolymers

177 containing DAAm self-assembled blocks with a high DP2, we did not observe any bilayer structures over the ranges considered. With increasing amounts of DMA within the hydrophobic block (DAAm85), worms were observed at a DP2 of 87 (Figure 7-4, f), while a DP2 of 141 (Figure 7-4, g) appeared to result in a worm-to-micelle transition and a smaller population of vesicular structures which remained constant at DP2 = 217

(Figure 7-4, h).

The transitions exhibited similarity to other worm-micelle transitions observed in post-polymerization self-assembly.195,201,209 Presumably, the core-forming block is becoming increasingly hydrophobic as the polymerization ensues; therefore, the interfacial energy would be expected to increase between the poly(DAAm-co-DMA)

120

Figure 7-4. Transmission electron microscopy images of polymerization-induced thermal self-assembly synthesized nanoparticles with varying monomer feed ratios of diacetone acrylamide (DAAm) to N,N-dimethylacrylamide (DMA) and second block degrees of polymerization (DP2) of 54, 87, 141, and 217; a-d) 90% DAAm:10% DMA (DAAm90); e-h) 85% DAAm:10% DMA (DAAm85); i-l) 80% DAAm:20% DMA (DAAm80); m-p) 75% DAAm:25% DMA (DAAm75). PTA = sodium phosphotungstate negative stain, UA = uranyl acetate negative stain block (DP2) and water. To quell the unfavorable interaction, the core-forming chains could be stretching to lower the core-solvent surface interactions. Chain stretching would lower the interfacial area between the hydrophilic and hydrophobic blocks; however, bringing the corona chains closer together increases steric repulsion. To relieve the coronal steric congestion, the length of each worm could shorten to provide a

121

higher number of spherical end-caps, which provide thermodynamic relief through a small hydrophilic-hydrophobic interfacial area and more volume for the corona chains.196 The constant decrease in solvent quality for the core chains most likely explains why only micelle-like morphologies were observed for DAAm90. The high interfacial tension between the hydrophobic block and water may inhibit any morphological transition, as the chains could favor sacrificing the entropic penalty of elongation over the enthalpic surface tension penalty.

Further increasing the DMA content to 20% in the monomer feed ratio (DAAm80) yielded morphological transitions more customary of PISA. Micelles were first observed with a DP2 = 54 (Figure 7-4, i), worms at DP2 = 87 (Figure 7-4, j), branched worms and other intermediate structures such as “jellyfish” and “octopus” at a DP2 = 141 (Figure 7-

4, k), and vesicles at a DP2 = 217 (Figure 7-4, l). However, a worm-to-micelle transition was still observed at this composition. Following the worms at DP2 = 87 and 141, a minor population appeared to undergo reversion to micelles as the degree of polymerization was increased. Therefore, in the final morphological composition (DP2 =

217, Figure 7-4, l), mostly vesicles with a small population of micelles persisted. Though a micelle-to-worm-to-fused worm-to-vesicle transition for DAAm80 seemed to have occurred, the presence of a minor worm-to-micelle transition inhibits the block copolymers formed at that monomer feed ratio from truly encompassing a hierarchical morphological transformation, showing multiple possible morphologies at high DP2.

DAAm75 yielded the most diverse and distinct morphological transitions. A DP2 of 54 resulted in micelles, comparable to the micelles achieved with the other compositions (Figure 7-4, m). At a DP2 = 87, worms were observed (Figure 7-4, n);

122

however, these worms appeared shorter in length than those of the other monomer compositions previously discussed, and many micelle-like structures were also present.

Upon further increasing the second block degree of polymerization to 141, highly extended worm structures were achieved (Figure 7-, 4o). Worm entanglements proved to encumber dispersion of the worms in solution needed to image isolated worms, but the lengths were observed to be up to microns in length. Additionally, few branching points can be seen over the length of the worms. At the highest DP2 of 217, predominately vesicles were observed (Figure 7-4, p). Importantly, all the spherical morphologies at a DP2 value of 217 displayed a bilayer-like structure, and their propensity to hexagonally pack on the TEM grids provides evidence of their narrow size dispersity.

The morphological transformations of DAAm75 proved to be particularly interesting. Micelle-to-worm transitions appeared to occur at a higher DP2 than in the other monomer feed ratios, which could be attributed to the DMA imparting more hydrophilicity to the chains. The enhanced hydrophilicity may have led to a later onset of aggregation and subsequent reorganization of unimers. Worm phases were also observed over a large range of DP2, and the worms appeared to predominantly favor elongation over branching at DP2 values of 111 and 127 (Figure 7-5, a, b). Elongation continued to result in worms up to microns in length with minimal branching points and a few lamellar structures at a DP2 of 141 (Figure 7-5, c). These results are in stark contrast to previous PISA reports where the linear worm phase was abrupt and branched worms were commonly observed. The worm to vesicle transition observed at

DP2 = 174 resulted in morphologies previously observed during PISA morphological

123

transitions, including lamellar bilayers, “octopi,” and branched structures (Figure 7-5, d).174 However, the majority of the worm-to-vesicle transitions appeared to occur from vesicles budding off from the worms (black arrows, Figure 7-5, d). A similar transformation was reported by Zhu and Hayward for blends of poly(ethylene oxide)-b- polystyrene copolymers during post-polymerization self-assembly.210 The bilayers were attributed to local phase separation due to the introduction of copolymers containing different block lengths. The nanoparticles reported here may be undergoing similar transitions, where chains possessing lower molecular weight PDMA blocks are partitioning into the core of the worms to form a bilayer. However, further mechanistic insight is required to fully understand this newly observed PISA transformation.

The morphological phase diagram (Figure 7-6, a) and transitions observed

(Figure 7-6, b) emphasize the significant impact of the hydrophobicity of the self- assembling polymer block on aggregate morphology. Vesicles were observed only when higher amounts of hydrophilic DMA were incorporated into the feed ratio, while worms were observed over a larger compositional range. Therefore, nanoparticle morphology was heavily dependent upon the hydrophobicity of the core-forming polymer block, where the hydrophobicity dictated progression of a polymer aggregate via a hierarchical morphological transition or an alternative reversion to micelles.

The average worm length for each of the compositions was determined by TEM to observe the differences in worm-worm transitions during the polymerization. With low concentrations of DMA (DAAm85), the worm length stayed between 76-94 nm across a range of DP2 values (Figure 7-7, a). Upon increasing the DMA content (DAAm80), the worm length peaked around a DP2 of 84, after which, a broadening of the length

124

Figure 7-5. Transmission electron microscopy images of crosslinked nanoparticles using a monomer feed ratio of 75:25 diacetone acrylamide to N,N- dimethylacrylamide (DAAm75) for the second block and varying degrees of polymerization (DP2); a) DP2 of 111 showing worm and micelle morphologies; b) DP2 of 127 showing predominately worm morphologies; c) DP2 of 141 showing elongated and branched worm morphologies (black arrows denote branching or points of lamellar formation); d) DP2 of 176 showing a worm-to- vesicle transition through branch points and budding from worms (black arrows denote points of vesicles budding from worms).

125

Figure 7-6. Morphology transitions according to core-block hydrophobicity a) phase diagram depicting the effect of varying the ratio of diacetone acrylamide (DAAm) to N,N-dimethylacrylamide in the monomer feed, resulting in a variety of morphologies; b) observed transition of morphology according to monomer feed ratio and the degree of polymerization of the block copolymer. distribution was observed as branching and multiple junction points developed (Figure

7,7 b). The polymerizations with the highest DMA content (DAAm75) yielded worms that grew at least an order of magnitude longer (from 76 nm to 770 nm) over the range of

DP2 values considered, until the worm-to-vesicle transition was approached and worm length plummeted (Figure 7-7, c, only isolated worms were measured—the actual average worm length is expected to be higher).

126

We reasoned that worm elongation proceeded through a step-wise process where the growing polymer chains induced an increase in packing parameter and necessitated a lower interfacial curvature. As the volume of the hydrophobic blocks increases, the worm end-caps become the most thermodynamically unfavorable section of the nanoparticles.186 The difference in energy between the cylindrical worm and spherical end-cap could cause the ends to undergo inelastic collisions with each other, while most collisions that do not include two end-caps would be elastic (Figure 7-7, d).

Consequently, worm length would grow exponentially, while branching via fusion of worms would be much slower. However, if the growing polymer chains become too hydrophobic, a high interfacial energy favors spheres over cylinders.196,211 The balance between the increasing volume of the hydrophobic block and the inherent hydrophobicity of the block (i.e., the number of worm end-caps present) would then dictate worm length. Therefore, the worm elongation (and reversion back to micelles) observed here is controlled by the hydrophilic-hydrophobic balance of the core chains, where the interfacial energy between the core and corona acts to inhibit or induce worm elongation during polymerization.

7.3 Conclusions

New examples of aqueous PISA systems have been difficult to identify and effectively modulate due to the stringent monomer/polymer requirements. We have demonstrated that nanoparticle morphology is defined by the hydrophobic nature of the growing amphiphilic chains. Moreover, predetermined morphologies can be targeted by carefully controlling incorporation of hydrophilic comonomers during PISA. Insight into the contribution of chain solvophilicity during polymer self-assembly and reorganization

127

Figure 7-7. Measured worm lengths and standard deviations from transmission electron microscopy (TEM) images with representative images of each worm at the measured second block degree of polymerization (DP2); a) diacetone acrylamide (DAAm):N,N-dimethylacrylamide (DMA) monomer feed ratio of 85:15 (DAAm85); b) DAAm:DMA monomer feed ratio of 80:20 (DAAm80); c) DAAm:DMA monomer feed ratio of 75:25 (DAAm75); d) proposed step-wise worm elongation controlled through the hydrophobicity of the constituent unimers. provided an understanding of the delicate balance between entropic and enthalpic penalties during polymer self-assembly and reorganization. This concept led to the strategic synthesis of a variety of nanoparticles including micelles, worms, micron-length worms, branched worms, and vesicles, and the elucidation of intermediate morphologies not observed previously during PISA (e.g., vesicles budding off from

128

worms). Regulation of block copolymer amphiphilicity also yielded the highest level of control over worm elongation reported during or after polymerization without a seeded initiator. Therefore, this elucidation of the contributions of block copolymer hydrophobicity during self-assembly may provide powerful new insight and strategies for programming the morphology and dimensions of block copolymer solution aggregates.

7.4 Materials and Methods

7.4.1 Materials

4,4´-Azobiscyanovaleric acid (ACVA, 98%, Alfa-Aesar), trioxane (> 99.5%, Acros

Organics), deionized water ASTM Type II (DI water, Aqua Solutions, Inc.) were used as received. S-Ethyl-S´-(α,α´-dimethyl-α˝-acetic acid)-trithiocarbonate chain transfer agent

(CTA) was synthesized according to a previously published report.212 Diacetone acrylamide (DAAm, > 98%, TCI Chemicals) was recrystallized 2× from ethyl acetate and once from hexanes prior to use. N,N-Dimethylacrylamide (DMA, 99%, Sigma-Aldrich) was filtered through basic alumina prior to use. O,O´-1,3-

Propanediylbishydroxyamine2HCl (crosslinker, > 99%, Sigma-Aldrich) was prepared as a 250 mg/mL deionized (DI) water solution immediately prior to use.

7.4.2 Characterization

1H NMR spectroscopy was conducted on an Inova 500 MHz, 2 RF channel instrument at 25 °C. D2O (Cambridge Isotopes Laboratories, Inc, 99.9%) solvent was used as received.

Size exclusion chromatography (SEC) was performed in N,N-dimethylacetamide

(DMAc) with 50 mM LiCl at 50 °C and a flow rate of 1.0 mL min-1 (Agilent isocratic pump, degasser, and autosampler, colums: Plgel 5 μm guard + two ViscoGel I-series

G3078 mixed bed columns: molecular weight range 0-20 × 103 and 1-100 × 104 g mol-

129

1). Detection consisted of a Wyatt Optilab T-rEX refractive index detector operating at

658 nm and a Wyatt miniDAWN Treos light scattering detector operating at 659 nm.

Absolute molecular weights and polydispersities were calculated using the Wyatt

ASTRA software and 100% mass recovery methods. Prior to absolute molecular weight determination, unimer samples were purified via dialysis against DI water for 3 days

(Spectra/Por 3 Dialysis Membranes (3500 Molecular weight cut-off (MWCO) from

Spectrum Laboratories), and nanoparticle samples were purified via dialysis against DI water for 3 days using Spectra/Por Float-A-Lyzer G2 100k MWCO 5 mL tubes from

Spectrum Labs before being isolated by freeze-drying.

Dynamic light scattering (DLS) analysis was performed with a Malvern Zetasizer

Nano ZS (Model No. ZEN 3600, Malvern Instruments Ltd., Worcestershire UK) and multi-angle dynamic light scattering measurements were performed on an ALV/CGS-3 four-angle, compact goniometer system (Langen, Germany), which consisted of a 22 mW HeNe linear polarized laser operating at a wavelength of 휆 = 632.8 푛푚 and scattering angles from 휃 = 42 − 150°. Fluctuations in the scattering intensity were measured via a ALV/LSE-5004 multiple tau digital correlator, and analyzed via the intensity autocorrelation function (g(2)(τ)). Decay rates, Γ, were obtained from single- exponential fits using a second-order cumulant analysis, and the mutual diffusion coefficient, Dm, was calculated through the relation according to equation 7-1:

2 Γ = 푞 퐷푚 7-1

2 where q is the scalar magnitude of the scattering vector. The hydrodynamic radius (Rh) was calculated through the Stokes-Einstein equation 7-2:

130

푘퐵푇 7-2 퐷푚 ≈ 퐷표 = 6휋휂푠푅ℎ where Dm is approximately equal to the tracer diffusion coefficient, Do, kB is the

Boltzmann constant, T is the absolute temperature, and ηs is the solvent viscosity.

Light scattering samples were performed at 25 °C following polymerization and crosslinking. Polymerization samples were diluted to 5 mg/mL solutions in water or

DMAC, passed through a 0.45 um Nylon syringe filter, and placed in a plastic cuvette for analysis (Malvern) and to 1.25 mg/mL, passed through 0.45 um Nylon syringe filter, and placed into a borosilicate, pre-cleaned cuvette for analysis (ALV). The appearance

(and disappearance) of multimodal distributions suggested that the varying hydrophilicity of the copolymers had an effect on the final morphology attained during

PISA.

Transmission electron microscopy (TEM) was conducted on an H-700 from

Hitachi High Technologies America, Inc., Schaumber, IL USA. Digital images were acquired with a Veleta 2k × 2k camera and iTEM software (Olympus Soft-Imaging

Solutions Corp., Lakewood, CO). Electron Microscopy Sciences Formvar Carbon Film on 400 mesh nickel grids (FCF400-Ni) were used for all measurements. For unstained samples, 10 μL purified nanoparticle solution (0.1 mg/mL) was spotted on the grid for 15 s. The excess solvent was wicked off and the grid air-dried. For sodium phosphotungstinate (PTA) staining, a drop of 0.5% PTA in water was placed on the grid and allowed to sit for 15 s. The excess solvent was wicked off and the grid was air- dried. For uranyl acetate (UA) staining, a drop of 2% UA in water was placed on the grid and allowed to sit for 45 s. The excess solvent was wicked off and the grid air-dried.

131

Worm length was measuring using ImageJ software with measured TEMs in the “Worm

Characterization” section.

7.4.3 Procedures for Kinetics

7.4.3.1 Kinetics of poly(N,N-dimethylacrylamide) (PDMA) macro chain transfer agent synthesis (macro-CTA 1)

DMA (3.85 g, 38.9 mmol), CTA (194 mg, 0.864 mmol), and ACVA (12.1 mg,

0.0432 mmol) were placed in a Schlenk flask to yield a [DMA]:[CTA]:[I] ratio =

[45]:[1]:[0.05]. DI water (15 mL) was added to make the monomer concentration 2.6 M.

The flask was sealed with a glass stopper, and a rubber septum was placed over the arm joint. The reaction was purged with nitrogen for 45 min, and left to stir at 70 °C for 3 h. Time points were taken every 30 min, and DMA conversion was monitored by 1H

NMR spectroscopy by comparing the vinyl monomer peaks to the 6 protons on the amide group (D2O, 500 MHz).

Weight-average molecular weight by SEC-MALS yielded an Mw = 7830 g/mol and a Ð = 1.13. The remaining initiator in solution was calculated using the half-life equation 7-3 and assuming a 10 h half-life time of AVCA at 70 °C:

푡/푡1/2 [퐼] = [퐼]0(0.5) 7-3

[퐼] = (0.727 푚푔/푚퐿)(0.5)3/10

[퐼] = 0.655 푚푔/푚퐿

The concentration of polymer (assuming > 95% monomer conversion as observed by

1H NMR) was found by dividing the total solids added by the volume of water added, resulting in a macro-CTA 1 concentration of 270 mg/mL. These concentrations were then used for all subsequent polymerizations.

132

7.4.3.2 Kinetics of 90% DAAm, 10% DMA polymerization kinetics

Macro-CTA 1 polymerization solution (635 mg, 135 mg macro-CTA 1, 0.0199 mmol macro-CTA 1, 0.328 mg ACVA, 1.17  10-3 mmol ACVA), DMA (40.9 uL, 39.4 mg,

0.398 mmol), and DAAm (605 mg, 3.58 mmol) were added to give a [macro-CTA

1]:[DMA]:[DAAm]:[initiator] ratio of [1]:[20]:[180]:[0.06]. The solution was diluted with DI water (4.70 g) to give a final solids concentration of 15 w/w%. The solution was then split into 6 vials containing 0.860 mL each and left to stir at 70 °C. Monomer conversion was monitored by 1H NMR spectroscopy by comparing the vinyl peaks of DAAm and

DMA to the backbone polymer peaks. SEC showed the evolution of molecular weight and efficient blocking efficiency, confirming the efficacy of the RAFT polymerization.

7.4.3.3 Kinetics of 85% DAAm, 15% DMA polymerization kinetics

Macro-CTA 1 polymerization solution (635 mg, 135 mg macro-CTA 1, 0.0199 mmol macro-CTA 1, 0.328 mg ACVA, 1.17  10-3 mmol ACVA), DMA (61.5 uL, 59.1 mg,

0.596 mmol), and DAAm (572 mg, 3.38 mmol) were added to give a [macro-CTA

1]:[DMA]:[DAAm]:[initiator] ratio of [1]:[30]:[170]:[0.06]. The solution was diluted with DI water (4.61 g) to give a final solids concentration of 15 w/w%. The solution was then split into 6 vials containing 0.850 mL each and left to stir at 70 °C. Monomer conversion was monitored by 1H NMR spectroscopy by comparing the vinyl peaks of DAAm and

DMA to polymer backbone peaks. SEC showed the evolution of molecular weight and efficient blocking efficiency, confirming the efficacy of the RAFT polymerization.

7.4.3.4 Kinetics of 80% DAAm, 20% DMA polymerization kinetics

Macro-CTA 1 polymerization solution (635 mg, 135 mg macro-CTA 1, 0.0199 mmol macro-CTA 1, 0.328 mg ACVA, 1.17  10-3 mmol ACVA), DMA (82.0 uL, 78.8 mg,

133

0.795 mmol), and DAAm (538 mg, 3.18 mmol) were added to give a [macro-CTA

1]:[DMA]:[DAAm]:[initiator] ratio of [1]:[40]:[160]:[0.06]. The solution was diluted with DI water (4.52 g) to give a final solids concentration of 15 w/w%. The solution was then split into 6 vials containing 0.830 mL each and left to stir at 70 °C. Monomer conversion was monitored by 1H NMR spectroscopy by comparing the vinyl peaks of DAAm and

DMA to the polymer backbone peaks. SEC showed the evolution of molecular weight and efficient blocking efficiency, confirming the efficacy of the RAFT polymerization.

7.4.3.5 Kinetics of 75% DAAm, 25% DMA polymerization kinetics

Macro-CTA 1 polymerization solution (635 mg, 135 mg macro-CTA 1, 0.0199 mmol macro-CTA 1, 0.328 mg ACVA, 1.17  10-3 mmol ACVA), DMA (82.0 uL, 78.8 mg,

0.795 mmol), and DAAm (538 mg, 3.18 mmol) were added to give a [macro-CTA

DMA to the polymer backbone peaks. SEC showed the evolution of molecular weight and efficient blocking efficiency, confirming the efficacy of the RAFT polymerization.

7.4.3.6 Reactivity ratios

Reactivity ratios were found to be rDAAm = 0.860 and rDMA = 0.804 using

Finneman-Ross data fitting, f = moles of DAAm in the monomer feed, F = molar composition of DAAm in the polymer, and limiting monomer conversions between 8-

10% by 1H NMR spectroscopy.

134

7.4.4 Procedures for Nanoparticle Synthesis

7.4.4.1 PDMA macro chain transfer agent synthesis (macro-CTA 2)

DMA (3.85 g, 38.9 mmol), CTA (174 mg, 0.777 mmol), and ACVA (10.9 mg,

0.0389 mmol) were placed in a Schlenk flask to yield a [DMA]:[CTA]:[I] ratio =

[50]:[1]:[0.05]. DI water (15 mL) was added to make the monomer concentration 2.6 M.

The flask was sealed with a glass stopper, and a rubber septum was placed over the arm joint. The reaction was purged with nitrogen for 45 min, and left to stir at 70 °C for 3 h. Weight-average molecular weight by SEC-MALS of a purified sample yielded an Mw =

8040 g/mol and a Ð = 1.17. The remaining initiator in solution was calculated to be

0.595 mg/mL using the half-life equation (7-3) and the 10 h half-life time of AVCA at 70

°C. The concentration of polymer (assuming > 95% monomer conversion as observed by 1H NMR) was found by dividing the total solids added by the volume of water added, resulting in a macro-CTA 2 concentration of 269 mg/mL. These concentrations were then used for all subsequent polymerizations and number-average molecular weight (Mn

= 6870 g/mol) was used to calculate monomer feed ratios.

7.4.4.2 90% DAAm, 10% DMA DP2 = 54 polymerization

Macro-CTA 2 polymerization solution (317 mg, 67.2 mg macro-CTA 2, 9.78  10-

3 mmol macro-CTA 2, 0.149 mg ACVA, 5.32  10-4 mmol ACVA), DMA (5.45 uL, 5.25 mg, 0.0530 mmol), and DAAm (80.6 mg, 0.477 mmol) were added to give a [macro-

CTA 2]:[DMA]:[DAAm]:[initiator] ratio of [1]:[5.4]:[48.6][0.05]. The solution was diluted with DI water (0.771 g) to give a final solids concentration of 15 w/w% and purged with

N2 for 15 min. The solution was then left to stir at 70 °C for 16 h. Following polymerization, a 0.2 mL aliquot was taken for unimer characterization. The remaining monomer unit concentration of DAAm was then calculated as follows:

135

• Total DAAm before polymerization: 80.6 mg

• Concentration of DAAm monomer units in solution: 78.9 mg/mL

• DAAm removed with 0.2 mL aliquot: 15.8 mg

• Remaining DAAm monomer units in solution: 64.8 mg (0.383 mmol)

Crosslinker solution (27.4 uL, 6.86 mg, 0.0383 mmol) was added via micro syringe. The solution was stirred at 70 °C for approximately 1 min, then removed from the oil bath and allowed to cool to room temperature.

7.4.4.3 90% DAAm, 10% DMA DP2 = 87 polymerization

Macro-CTA 2 polymerization solution (317 mg, 67.2 mg macro-CTA 2, 9.78  10-

3 mmol macro-CTA 2, 0.149 mg ACVA, 5.32  10-4 mmol ACVA), DMA (8.70 uL, 8.40 mg, 0.0848 mmol), and DAAm (129 mg, 0.7632 mmol) were added to give a [macro-

CTA 2]:[DMA]:[DAAm]:[initiator] ratio of [1]:[9]:[78][0.05]. The solution was diluted with

DI water (1.12 g) to give a final solids concentration of 15 w/w% and purged with N2 for

15 min. The solution was then left to stir at 70 °C for 16 h. Following polymerization, a

0.2 mL aliquot was taken for unimer characterization. The remaining monomer unit concentration of DAAm was then calculated as follows:

• Total DAAm before polymerization: 129 mg

• Concentration of DAAm monomer units in solution: 94.5 mg/mL

• DAAm removed with 0.2 mL aliquot: 18.9 mg

• Remaining DAAm monomer units in solution: 110 mg (0.651 mmol)

Crosslinker solution (46.6 uL, 11.7 mg, 0.0654 mmol) was added via micro syringe. The solution was stirred at 70 °C for approximately 1 min then, removed from the oil bath and allowed to cool to room temperature.

136

7.4.4.4 90% DAAm, 10% DMA DP2 = 141 polymerization

Macro-CTA 2 polymerization solution (317 mg, 67.2 mg macro-CTA 2, 9.78  10-

3 mmol macro-CTA 2, 0.149 mg ACVA, 5.32  10-4 mmol ACVA), DMA (14.2 uL, 13.6 mg, 0.138 mmol), and DAAm (210 mg, 1.24 mmol) were added to give a [macro-CTA

2]:[DMA]:[DAAm]:[initiator] ratio of [1]:[14]:[127][0.05]. The solution was diluted with DI water (1.69 g) to give a final solids concentration of 15 w/w% and purged with N2 for 15 min. The solution was then left to stir at 70 °C for 16 h. Following polymerization, a 0.2 mL aliquot was taken for unimer characterization. The remaining monomer unit concentration of DAAm was then calculated as follows:

• Total DAAm before polymerization: 210 mg

• Concentration of DAAm monomer units in solution: 108 mg/mL

• DAAm removed with 0.2 mL aliquot: 21.7 mg

• Remaining DAAm monomer units in solution: 188 mg (1.11 mmol)

Crosslinker solution (79.6 uL, 19.9 mg, 0.111 mmol) was added via micro syringe. The solution was stirred at 70 °C for approximately 1 min, then removed from the oil bath and allowed to cool to room temperature.

7.4.4.5 90% DAAm, 10% DMA DP2 = 217 polymerization

Macro-CTA 2 polymerization solution (317 mg, 67.2 mg macro-CTA 2, 9.78  10-

3 mmol macro-CTA 2, 0.149 mg ACVA, 5.32  10-4 mmol ACVA), DMA (21.7 uL, 20.9 mg, 0.211 mmol), and DAAm (323 mg, 1.91 mmol) were added to give a [macro-CTA

2]:[DMA]:[DAAm]:[initiator] ratio of [1]:[22]:[195][0.05]. The solution was diluted with DI water (2.49 g) to give a final solids concentration of 15 w/w% and purged with N2 for 15 min. The solution was then left to stir at 70 °C for 16 h. Following polymerization, a 0.2

137

mL aliquot was taken for unimer characterization. The remaining monomer unit concentration of DAAm was then calculated as follows:

• Total DAAm before polymerization: 323 mg

• Concentration of DAAm monomer units in solution: 118 mg/mL

• DAAm removed with 0.2 mL aliquot: 23.6 mg

• Remaining DAAm monomer units in solution: 299 mg (1.77 mmol)

Crosslinker solution (127 uL, 31.7 mg, 0.177 mmol) was added via micro syringe. The solution was stirred at 70 °C for approximately 1 min, then removed from the oil bath and allowed to cool to room temperature.

7.4.5 SEC-MALS discussion

Nanoparticle weight-average molecular weight (Mw,NP), nanoparticle aggregation number (Nagg), and nanoparticle radius of gyration (Rg) (Table 1) were compared to further study how unimer composition affected PISA where Nagg was calculated by dividing the Mw,NP of the aggregates by Mw,unimer. The Mw,NP increased for all monomer feed ratios with increasing unimer molecular weight. Assuming a constant or increasing

Nagg, an increase in nanoparticle molecular weight is expected during PISA, as the unimer chains are constantly growing. Interestingly, DAAm75 showed a nanoparticle molecular weight at DP2 = 217 almost an order of magnitude larger than the other three compositions, even though elution times were similar, suggesting the presence of a different morphology.

Nagg of all the nanoparticles increased with increasing Mw,unimer. This result suggests the nanoparticles formed are not kinetically trapped and are able to overcome

138

Table 7-2. Complete nanoparticle profiles obtained using SEC-MALS %DAA Second Unimer Unim Nanoparti Nanoparti Radius Observed m in Block Molecul er cle cle of Morpholo Monom Degree of ar dn/dc Molecular Aggregati Gyratio gyc er Polymerizati Weight, value Weight, on n, Rg a b Feed on, DP2 Mw,unime s Mw,NP Number, (nm) 3 6 r (10 (mL/g (10 Nagg g/mol)b )b g/mol)b 90 54 11.6 0.070 1.11 96 16 M 8 90 87 15.6 0.072 3.93 252 20 M 2 90 141 21.7 0.072 11.6 535 23 M 6 90 217 35.2 0.077 45.2 1280 30 M 0 85 54 11.7 0.074 1.49 127 11 M 6 85 87 15.6 0.073 17.1 1100 27 W 9 85 111 -- -- 0.934 -- 30 W + M 85 127 -- -- 10.5 -- 29 W + M 85 141 22.2 0.072 36.7 1650 30 W + M 5 85 174 -- -- 13.9 -- 27 M 85 217 33.1 0.070 38.1 1150 30 M 4 80 54 11.3 0.071 0.699 62 12 M 6 80 87 18.0 0.063 18.7 1040 79 W 2 80 111 26.7 0.058 34.2 1280 75 W + M 8 80 127 27.7 0.067 24.2 873 63 W + M 1 80 141 23.8 0.071 28.2 1180 63 W + M 4 80 174 28.1 0.076 25.8 918 35 V + W + 7 M 80 217 30.6 0.072 31.2 1020 31 V + M 5 75 54 14.3 0.085 0.622 43 13 M 4 75 87 18.7 0.079 2.11 113 23 M + W 1

139

Table 7-2. Continued %DAA Second Unimer Unim Nanoparti Nanoparti Radius Observed m in Block Molecul er cle cle of Morpholo Monom Degree of ar dn/dc Molecular Aggregati Gyratio gyc er Polymerizati Weight, value Weight, on n, Rg a b Feed on, DP2 Mw,unime s Mw,NP Number, (nm) 3 6 r (10 (mL/g (10 Nagg g/mol)b )b g/mol)b

75 111 19.3 0.0745 9.77 506 68 M + W 75 127 22.0 0.0701 50.0 2270 87 W 75 141 22.2 0.0812 65.1 2930 89 W 75 174 32.6 0.0729 119 3650 61 V + W 75 217 39.5 0.0638 148 3750 39 V aAccording to initial monomer feed ratio and >95% monomer conversion, bdetermined using a multi-angle laser light scattering, cobserved during transmission electron microscopy imaging where M = micelle, W = worm, and V = vesicle. unfavorable entropic and steric interactions to combine into more thermodynamically favorable morphologies. Armes and coworkers recently reported similar results for spherical morphologies,208 therefore our results provide further evidence of the combinatorial nature of PISA throughout the assembly and reorganization process. The difference in relative hydrophobicity of aggregated blocks is expected to affect Nagg, as more hydrophobic chains should be more collapsed and therefore require more chains per nanoparticle to lower the interfacial area per chain.187 When comparing the nanoparticles formed when DP2 = 217, the aggregation number decreases from

DAAm90 (Table 1, entry 4) to DAAm85 (Table 1, entry 8) to DAAm80 (Table 1, entry

12). However, DAAm75 showed a significantly higher Nagg when DP2 = 217 (Table 1, entry 16) than the other three monomer feed ratios suggesting this monomer ratio contains a different morphology. Other conclusions from Nagg are difficult to obtain due to the apparent distribution of nanoparticle morphologies present between different compositions.

140

Fortunately, the average Rg may provide some insight into distribution of nanoparticle morphologies present. The continuous, monotonic increase in Rg exhibited by DAAm90 further suggests there are few anisotropic structures during PISA for this composition. When the DMA ratio is increased by only 5% in the monomer feed

(DAAm85), the Rg increases significantly at a low DP2 (Table 1, entry 5) then remains approximately the same for higher DP2 values. A more dramatic peak in Rg was observed for DAAm80 (Table 1, entry 10) and DAAm75 (Table 1, entry 15) at an intermediate DP2 accompanied by broad and multimodal SEC traces. This large increase then decrease in Rg may be indicative of a sphere-to-worm-to-sphere transition, as worms are known to exhibit large and disperse Rg values due to the morphology’s high aspect ratio.

To confirm some of the results we observed during DLS analysis using the

Zetasizer and SEC-MALS were accurate, we performed subsequent DLS measurements in DMAc using an ALV/CGS-3 four-angle, compact goniometer system.

This analysis allows for more accurate determination of hydrodynamic radii because angular dependence of the diffusion coefficient can be accounted for, which is particularly important for disperse samples. The  value obtained by dividing the Rg

(obtained by SEC-MALS) by the Rh (obtained by diffusion coefficients given by the slope in Γ vs q2 plots) agreed with the morphologies observed during TEM imaging and helped to check the accuracy of the data obtained by SEC-MALS (Table S2).

Theoretically,  values for micelles and vesicles are 0.775 and 1.0, respectively. As Rg begins to exceed the value of Rh the  value begins to exceed unity, which indicates the presence of elongated structures. For the samples analyzed using the multi-angle

141

analysis technique, the  values observed are in agreement with the theoretical values for the morphologies determined by TEM analysis. Specifically, samples with micelle morphology DAAm90/DP2 87 and DAAm90/DP2 217 were determined to have  values of 0.87 and 0.77, respectively, which are in excellent agreement with the theoretical value of 0.775. Sample DAAm85/DP2 141, which exhibited a mixed morphology of micelles and worms, gave a  value of 1.0, which may be expected due to the blend of morphologies. Finally, sample DAAm80/DP2 87 which exhibited worm morphology, gave a  value of 2.2, which agrees with the extended, anisotropic morphology shown in TEM analysis. Detailed analysis of the remaining samples is outside the scope of the current manuscript, but will be pursued in future publications, specifically the interesting anisotropic, worm morphology systems.

Table 7-3. Light scattering studies of selected nanoparticles %DAA Second Hydrodynamic Hydrodyna Radiu , Observed b b m in Block Radius, Rh (water) mic Radius, s of DMA Morpholo d Monom Degree of Rh (DMAc) Gyrati c gy c er Polymerizati on, (Rg/R a Feed on, DP2 Rg h) Ratio (DMAc ) 90 87 - 23 20 0.87 M 90 217 34 39 30 0.77 M 85 141 24 30 30 1.0 W + M 80 87 45 36 79 2.2 W aAccording to initial monomer feed ratio and >95% monomer conversion, bdetermined using multi-angle DLS, cdetermined suing SEC-MALS, dobserved during transmission electron microscopy imaging where M = micelle and W = worm.

7.4.6 Worm length determination

Worm lengths were measured end-to-end of randomly selected worms across different TEM images from different regions of the TEM grids using ImageJ software.

Measured worm length (L) and number of worms measured (n) were used to calculate

142

average worm length (Ln, Equation 7-4), weighted worm length (Lw Equation 7-5), and worm length polydispersity (PDI, Equation 7-6) with the table represented in Table S3.

Σ 퐿 7-4 퐿 = 푛 푛

Σ 퐿2 7-5 퐿푤 = 퐿푛푛

퐿 7-6 푃퐷퐼 = 푤 퐿푛

Table 7-4 Worm length measurements from transmission electron microscopy images and calculated average worm length, worm length standard deviation, weighted worm length, and worm polydispersity (PDI). %DAAm Second Block Sample Average Worm Weighted Worm in Degree of Size Worm Length Worm PDI Monomer Polymerization Length Standard Length Feed (Ln, nm) Deviation (Lw, nm) 85 87 210 94 57 128 1.4 85 111 150 98 57 130 1.3 85 127 150 86 43 100 1.2 85 141 100 76 26 85 1.1 80 87 148 343 253 528 1.5 80 111 114 208 126 284 1.4 80 127 150 184 159 321 1.7 80 141 77 468 452 900 1.9 80 176 119 133 64 163 1.2 75 87 200 76 35 93 1.2 75 111 200 318 266 540 1.7 75 127 100 469 290 647 1.4 75 141 58 774 389 966 1.3 75 176 68 292 172 392 1.3

143

CHAPTER 8 FINAL REMARKS

Macromolecular chemistry relies on the interplay of interactions and reactions on multiple length scales to elicit macroscopic properties. To achieve these goals, new chemistry must be developed at each scale to provide specific and desired reaction outcomes. This dissertation describes two new approaches using TCT and BTF that enable omega, omega- heterodifunctional polymers to be synthesized. Aqueous RAFT polymerizations using visible-light and a photocatalytic organic dye are described in mechanistic detail to provide well-defined polymers in biologically-relevant conditions.

New approaches to amphiphilic block copolymer nanoparticles using thermoresponsive polymers and an understanding on the interplay between hydrophobicity and aggregate morphology outcome are described. Therefore, this dissertation introduces synthetic approaches to specific, complex, and well-defined materials, while also providing fundamental understandings of the processes described.

144

APPENDIX A IDEAL OMEGA,OMEGA-HETERODIFUNCTIONALIZATION PATHWAY

Figure A-1. Optimized reaction conditions limited unwanted mPEG byproducts, while taking full advantage of the reactivity of TCT.

145

APPENDIX B FUNCTIONALIZATION OF COUMARIN-AZIDE AND BIOTIN TO PEG CHARACTERIZATION

Figure B-1. Matrix-assisted laser desorption-ionization time-of-flight mass spectrum of compounds 1, 9, and 10, confirming the presence of the dye-containing ,-heterodifunctionalized adduct.

Figure B-2. 1H NMR spectrum of PEG functionalized with coumarin and biotin.

146

APPENDIX C ABSORPTION SPECTRA OF EOSIN Y AND TRITHIOCARBONATES

Figure C-1. Absorption spectra of eosin Y in water

Figure C-2. Absorption spectra of 2-(ethyl trithiocarbonate)propionic acid (ETPA) (ETPA, blue) and 2-(ethyl trithiocarbonate)-2-methylpropionic acid (CTA, red).

147

APPENDIX D ADDITIONAL PET-RAFT POLYMERIZATION KINETIC DATA

Figure D-1. Molecular weight versus monomer conversion of photoinduced electron- transfer reversible addition-fragmentation chain transfer polymerizations containing different ratios of [chain transfer agent]:[eosin Y]:[4- dimethylaminopyridine] performed under blue-light irradiation.

Figure D-2. Pseudo-first order kinetics plot for photo electron-transfer reversible addition-fragmentation chain transfer polymerizations varying the amount of 4-dimethylaminopyridine in ratios of [chain transfer agent]:[eosin Y]:[4- dimethylaminopyridine] performed under blue-light irradiation.

148

Figure D-3. Data for photoinduced electron-transfer reversible addition-fragmentation chain transfer polymerizations varying the amount of EY in ratios of [chain transfer agent]:[eosin Y]:[4-dimethylaminopyridine] performed under blue-light irradiation and a pH 8.4-9.0.

149

Figure D-4. Data for photoinduced electron-transfer reversible addition-fragmentation chain transfer polymerizations varying the amount of EY in ratios of [chain transfer agent]:[eosin Y] performed under green-light irradiation, in the absence of DMAP, and a pH of 4.4.

Figure D-5. Electrospray ionization mass spectra of reductive PET-RAFT polymerization showing initiation from both the chain transfer agent and the reducing agent.

150

APPENDIX E POLYMERIZATION KINETICS OF N,N-DIMETHYLACRYLAMIDE AND DIACETONE ACRYLAMIDE

Figure E-1. Kinetics of a polymerization-induced thermal self-assembly polymerization using a monomer feed ratio of 90% diacetone acrylamide to 10% N,N- dimethylacrylamide, as monitored by 1H NMR spectroscopy and size exclusion chromatography.

Figure E-2. Kinetics of a polymerization-induced thermal self-assembly polymerization using a monomer feed ratio of 85% diacetone acrylamide to 15% N,N- dimethylacrylamide, as monitored by 1H NMR spectroscopy and size exclusion chromatography.

Figure E-3. Kinetics of a polymerization-induced thermal self-assembly polymerization using a monomer feed ratio of 80% diacetone acrylamide to 20% N,N- dimethylacrylamide, as monitored by 1H NMR spectroscopy and size exclusion chromatography.

151

Figure E-4. Kinetics of a polymerization-induced thermal self-assembly polymerization using a monomer feed ratio of 75% diacetone acrylamide to 15% N,N- dimethylacrylamide, as monitored by 1H NMR spectroscopy and size exclusion chromatography.

152

APPENDIX F DYNAMIC LIGHT SCATTERING OF NANOPARTICLES DIRECTLY FOLLOWING POLYMERIZATION AND CROSSLINKING

Figure F-1. Diacetone acrylamide:N,N-dimethylacrylamide monomer feed ratio of 90:10 with varying second block degrees of polymerization (DP2).

Figure F-2. Diacetone acrylamide:N,N-dimethylacrylamide monomer feed ratio of 85:15 with varying second block degrees of polymerization (DP2).

153

Figure F-3. Diacetone acrylamide:N,N-dimethylacrylamide monomer feed ratio of 80:20 with varying second block degrees of polymerization (DP2).

Figure F-4. Diacetone acrylamide:N,N-dimethylacrylamide monomer feed ratio of 75:25 with varying second block degrees of polymerization (DP2).

154

LIST OF REFERENCES

(1) Cobo, I.; Li, M.; Sumerlin, B. S.; Perrier, S. Smart Hybrid Materials by Conjugation of Responsive Polymers to Biomacromolecules. Nat. Mater. 2015, 14, 143–159.

(2) Bates, F. S.; Hillmyer, M. A.; Lodge, T. P.; Bates, C. M.; Delaney, K. T.; Fredrickson, G. H. Multiblock Polymers: Panacea or Pandora's Box? Science 2012, 336, 434–440.

(3) Golas, P. L.; Matyjaszewski, K. Marrying Click Chemistry with Polymerization: Expanding the Scope of Polymeric Materials. Chem. Soc. Rev. 2010, 39, 1338–1354.

(4) Goldmann, A. S.; Glassner, M.; Inglis, A. J.; Barner-Kowollik, C. Post‐ Functionalization of Polymers via Orthogonal Ligation Chemistry. Macromol. Rapid Commun. 2013, 34, 810–849.

(5) Lowe, A. B. Thiol–Ene “Click” Reactions and Recent Applications in Polymer and Materials Synthesis: a First Update. Polym. Chem. 2014, 5, 4820–4870.

(6) Das, A.; Theato, P. Activated Ester Containing Polymers: Opportunities and Challenges for the Design of Functional Macromolecules. Chem. Rev. 2015, 116, 1434–1495.

(7) De Bruycker, K.; Billiet, S.; Houck, H. A.; Chattopadhyay, S.; Winne, J. M.; Prez, Du, F. E. Triazolinediones as Highly Enabling Synthetic Tools. Chem. Rev. 2016, 116, 3919–3974.

(8) Durmaz, H.; Sanyal, A.; Hizal, G.; Tunca, U. Double Click Reaction Strategies for Polymer Conjugation and Post-Functionalization of Polymers. Polym. Chem. 2012, 3, 825–835.

(9) Iha, R. K.; Wooley, K. L.; Nyström, A. M.; Burke, D. J.; Kade, M. J.; Hawker, C. J. Applications of Orthogonal “Click” Chemistries in the Synthesis of Functional Soft Materials. Chem. Rev. 2009, 109, 5620–5686.

(10) Hiltebrandt, K.; Elies, K.; D’hooge, D. R.; Blinco, J. P.; Barner-Kowollik, C. A Light-Activated Reaction Manifold. J. Am. Chem. Soc. 2016, 138, 7048–7054.

(11) Billiet, S.; De Bruycker, K.; Driessen, F.; Goossens, H.; Van Speybroeck, V.; Winne, J. M.; Prez, Du, F. E. Triazolinediones Enable Ultrafast and Reversible Click Chemistry for the Design of Dynamic Polymer Systems. Nat. Chem. 2014, 6, 815–821.

155

(12) Gondi, S. R.; Vogt, A. P.; Sumerlin, B. S. Versatile Pathway to Functional Telechelics via RAFT Polymerization and Click Chemistry. Macromolecules 2007, 40, 474–481.

(13) Mueller, J. O.; Schmidt, F. G.; Blinco, J. P.; Barner-Kowollik, C. Visible‐Light‐ Induced Click Chemistry. Angew. Chem. Int. Ed. 2015, 54, 10284–10288.

(14) Roth, P. J.; Boyer, C.; Lowe, A. B.; Davis, T. P. RAFT Polymerization and Thiol Chemistry: a Complementary Pairing for Implementing Modern Macromolecular Design. Macromol. Rapid Commun. 2011, 32, 1123–1143.

(15) De, S.; Khan, A. Efficient Synthesis of Multifunctional Polymersviathiol–Epoxy “Click” Chemistry. Chem. Commun. 2012, 48, 3130–3132.

(16) Gody, G.; Roberts, D. A.; Maschmeyer, T.; Perrier, S. A New Methodology for Assessing Macromolecular Click Reactions and Its Application to Amine– Tertiary Isocyanate Coupling for Polymer Ligation. J. Am. Chem. Soc. 2016, 138, 4061–4068.

(17) Quek, J. Y.; Liu, X.; Davis, T. P.; Roth, P. J.; Lowe, A. B. RAFT-Prepared Α- Difunctional Poly(2-Vinyl-4,4-Dimethylazlactone)S and Their Derivatives: Synthesis and Effect of End-Groups on Aqueous Inverse Temperature Solubility. Polym. Chem. 2015, 6, 118–127.

(18) An, Z.; Tang, W.; Wu, M.; Jiao, Z.; Stucky, G. D. Heterofunctional Polymers and Core–Shell Nanoparticlesvia Cascade Aminolysis/Michael Addition and Alkyne–Azide Click Reaction of RAFT Polymers. Chem. Commun. 2008, 6501–6503.

(19) Samoshin, A. V.; Hawker, C. J.; de Alaniz, J. R. Nitrosocarbonyl Hetero-Diels– Alder Cycloaddition: a New Tool for Conjugation. ACS Macro Lett. 2014, 3, 753–757.

(20) Yu, B.; Chan, J. W.; Hoyle, C. E.; Lowe, A. B. Sequential Thiol‐Ene/Thiol‐Ene and Thiol‐Ene/Thiol‐Yne Reactions as a Route to Well‐Defined Mono and Bis End‐Functionalized Poly(N‐Isopropylacrylamide). J. Polym. Sci. A Polym. Chem. 2009, 47, 3544–3557.

(21) Simula, A.; Nurumbetov, G.; Anastasaki, A.; Wilson, P.; Haddleton, D. M. Synthesis and Reactivity of Α,Ω-Homotelechelic Polymers by Cu(0)-Mediated Living Radical Polymerization. Eur. Polym. J. 2015, 62, 294–303.

(22) Roth, P. J.; Jochum, F. D.; Zentel, R.; Theato, P. Synthesis of Hetero- Telechelic Α,Ω Bio-Functionalized Polymers. Biomacromolecules 2009, 11, 238–244.

156

(23) Zhang, B.; Zhang, H.; Elupula, R.; Alb, A. M.; Grayson, S. M. Efficient Synthesis of High Purity Homo-Arm and Mikto-Arm Poly(Ethylene Glycol) Stars Using Epoxide and Azide-Alkyne Coupling Chemistry. Macromol. Rapid Commun. 2014, 35, 146–151.

(24) Driessen, F.; Martens, S.; Meyer, B. D.; Prez, Du, F. E.; Espeel, P. Double Modification of Polymer End Groups Through Thiolactone Chemistry. Macromol. Rapid Commun. 2016, 37, 947–951.

(25) McKenzie, T. G.; Fu, Q.; Uchiyama, M.; Satoh, K.; Xu, J.; Boyer, C.; Kamigaito, M.; Qiao, G. G. Beyond Traditional RAFT: Alternative Activation of Thiocarbonylthio Compounds for Controlled Polymerization. Adv. Sci. 2016, 3, 1500394.

(26) Pan, X.; Malhotra, N.; Simakova, A.; Wang, Z.; Konkolewicz, D.; Matyjaszewski, K. Photoinduced Atom Transfer Radical Polymerization with Ppm-Level Cu Catalyst by Visible Light in Aqueous Media. J. Am. Chem. Soc. 2015, 137, 15430–15433.

(27) Huang, T.; Cui, Z.; Ding, Y.; Lu, X.; Cai, Y. The Use of Electrostatic Association for Rapid RAFT Synthesis of Histamine Polyelectrolyte in Aqueous Solutions at and Below 25 °C. Polym. Chem. 2015, 7, 176–183.

(28) Tong, J.; Shi, Y.; Liu, G.; Huang, T.; Xu, N.; Zhu, Z.; Cai, Y. Visible Light Mediated Fast Iterative RAFT Synthesis of Amino‐Based Reactive Copolymers in Water at 20 °C. Macromol. Rapid Commun. 2013, 34, 1827–1832.

(29) Hill, M. R.; Carmean, R. N.; Sumerlin, B. S. Expanding the Scope of RAFT Polymerization: Recent Advances and New Horizons. Macromolecules 2015, 48, 5459–5469.

(30) Otsu, T. Iniferter Concept and Living Radical Polymerization. J. Polym. Sci. A Polym. Chem. 2000, 38, 2121–2136.

(31) McKenzie, T. G.; Fu, Q.; Wong, E. H. H.; Dunstan, D. E.; Qiao, G. G. Visible Light Mediated Controlled Radical Polymerization in the Absence of Exogenous Radical Sources or Catalysts. Macromolecules 2015, 48, 3864– 3872.

(32) Xu, J.; Shanmugam, S.; Corrigan, N.; Boyer, C. Catalyst-Free Visible Light- Induced RAFT Photopolymerization. In Controlled Radical Polymerization: Mechanisms; ACS Symposium Series; Vol. 1187, pp. 247–267.

(33) Lewis, R. W.; Evans, R. A.; Malic, N.; Saito, K.; Cameron, N. R. Ultra-Fast Aqueous Polymerisation of Acrylamides by High Power Visible Light Direct Photoactivation RAFT Polymerisation. Polym. Chem. 2018, 9, 60–68.

157

(34) Fu, Q.; McKenzie, T. G.; Tan, S.; Nam, E.; Qiao, G. G. Tertiary Amine Catalyzed Photo-Induced Controlled Radical Polymerization of Methacrylates. Polym. Chem. 2015, 6, 5362–5368.

(35) Shi, Y.; Gao, H.; Lu, L.; Cai, Y. Ultra-Fast RAFT Polymerisation of Poly(Ethylene Glycol) Acrylate in Aqueous Media Under Mild Visible Light Radiation at 25 °C. Chem. Commun. 2009, 1368–1370.

(36) Shi, Y.; Liu, G.; Gao, H.; Lu, L.; Cai, Y. Effect of Mild Visible Light on Rapid Aqueous RAFT Polymerization of Water-Soluble Acrylic Monomers at Ambient Temperature: Initiation and Activation. Macromolecules 2009, 42, 3917–3926.

(37) David B Thomas; Sumerlin, B. S.; Andrew B Lowe, A.; Charles L McCormick. Conditions for Facile, Controlled RAFT Polymerization of Acrylamide in Water†. Macromolecules 2003, 36, 1436–1439.

(38) David B Thomas; Anthony J Convertine; Roger D Hester; Andrew B Lowe, A.; Charles L McCormick. Hydrolytic Susceptibility of Dithioester Chain Transfer Agents and Implications in Aqueous RAFT Polymerizations†. Macromolecules 2004, 37, 1735–1741.

(39) Tan, J.; Liu, D.; Bai, Y.; Huang, C.; Li, X.; He, J.; Xu, Q.; Zhang, L. Enzyme- Assisted Photoinitiated Polymerization-Induced Self-Assembly: an Oxygen- Tolerant Method for Preparing Block Copolymer Nano-Objects in Open Vessels and Multiwell Plates. Macromolecules 2017, 50, 5798–5806.

(40) Prier, C. K.; Rankic, D. A.; MacMillan, D. W. C. Visible Light Photoredox Catalysis with Transition Metal Complexes: Applications in Organic Synthesis. Chem. Rev. 2013, 113, 5322–5363.

(41) Fors, B. P.; Hawker, C. J. Control of a Living Radical Polymerization of Methacrylates by Light. Angew. Chem. Int. Ed. 2012, 51, 8850–8853.

(42) Treat, N. J.; Sprafke, H.; Kramer, J. W.; Clark, P. G.; Barton, B. E.; Read de Alaniz, J.; Fors, B. P.; Hawker, C. J. Metal-Free Atom Transfer Radical Polymerization. J. Am. Chem. Soc. 2014, 136, 16096–16101.

(43) Xu, J.; Jung, K.; Atme, A.; Shanmugam, S.; Boyer, C. A Robust and Versatile Photoinduced Living Polymerization of Conjugated and Unconjugated Monomers and Its Oxygen Tolerance. J. Am. Chem. Soc. 2014, 136, 5508– 5519.

(44) Zhou, H.; Johnson, J. A. Photo‐Controlled Growth of Telechelic Polymers and End‐Linked Polymer Gels. Angew. Chem. Int. Ed. 2013, 52, 2235–2238.

158

(45) Xu, J.; Jung, K.; Corrigan, N. A.; Boyer, C. Aqueous Photoinduced Living/Controlled Polymerization: Tailoring for Bioconjugation. Chem. Sci. 2014, 5, 3568–3575.

(46) Niu, J.; Page, Z. A.; Dolinski, N. D.; Anastasaki, A.; Hsueh, A. T.; Soh, H. T.; Hawker, C. J. Rapid Visible Light-Mediated Controlled Aqueous Polymerization with in Situ Monitoring. ACS Macro Lett. 2017, 6, 1109–1113.

(47) Shanmugam, S.; Xu, J.; Boyer, C. Aqueous RAFT Photopolymerization with Oxygen Tolerance. Macromolecules 2016, 49, 9345–9357.

(48) Shanmugam, S.; Xu, J.; Boyer, C. A Logic Gate for External Regulation of Photopolymerization. Polym. Chem. 2016, 7, 6437–6449.

(49) Niu, J.; Lunn, D. J.; Pusuluri, A.; Yoo, J. I.; O'Malley, M. A.; Mitragotri, S.; Soh, H. T.; Hawker, C. J. Engineering Live Cell Surfaces with Functional Polymers via Cytocompatible Controlled Radical Polymerization. Nat. Chem. 2017, 9, 537–545.

(50) Figg, C. A.; Hickman, J. D.; Scheutz, G. M.; Shanmugam, S.; Carmean, R. N.; Tucker, B. S.; Boyer, C.; Sumerlin, B. S. Color-Coding Visible Light Polymerizations to Elucidate the Activation of Trithiocarbonates Using Eosin Y. 2018, 51, 1370–1376.

(51) Rösler, A.; Vandermeulen, G. W. M.; Klok, H.-A. Advanced Drug Delivery Reviews. Adv. Drug Deliv. Rev. 2012, 64, 270–279.

(52) Elsabahy, M.; Wooley, K. L. Design of Polymeric Nanoparticles for Biomedical Delivery Applications. Chem. Soc. Rev. 2012, 41, 2545–2561.

(53) Smith, A. E.; Xu, X.; McCormick, C. L. Stimuli-Responsive Amphiphilic (Co)Polymers via RAFT Polymerization. Prog. Polym. Sci. 2010, 35, 45–93.

(54) Allen, C.; Maysinger, D.; Eisenberg, A. Nano-Engineering Block Copolymer Aggregates for Drug Delivery. Colloids Surf., B 1999, 16, 3–27.

(55) Roy, D.; Cambre, J. N.; Sumerlin, B. S. Future Perspectives and Recent Advances in Stimuli-Responsive Materials. Prog. Polym. Sci. 2010, 35, 278– 301.

(56) Tao, L.; Hu, W.; Liu, Y.; Huang, G.; Sumer, B. D.; Gao, J. Shape-Specific Polymeric Nanomedicine: Emerging Opportunities and Challenges. Exp. Biol. Med. 2011, 236, 20–29.

159

(57) Tanner, P.; Baumann, P.; Enea, R.; Onaca, O.; Palivan, C.; Meier, W. Polymeric Vesicles: From Drug Carriers to Nanoreactors and Artificial Organelles. Acc. Chem. Res. 2011, 44, 1039–1049.

(58) Discher, B. M.; Won, Y. Y.; Ege, D. S.; Lee, J. C.; Bates, F. S.; Discher, D. E.; Hammer, D. A. Polymersomes: Tough Vesicles Made From Diblock Copolymers. Science 1999, 284, 1143–1146.

(59) O'Reilly, R. K.; Hawker, C. J.; Wooley, K. L. Cross-Linked Block Copolymer Micelles: Functional Nanostructures of Great Potential and Versatility. Chem. Soc. Rev. 2006, 35, 1068–1083.

(60) Rodriguez-Hernandez, J.; Checot, F.; Gnanou, Y.; Lecommandoux, S. Toward “Smart” Nano-Objects by Self-Assembly of Block Copolymers in Solution. Prog. Polym. Sci. 2005, 30, 691–724.

(61) Sijbesma, R. P.; Meijer, E. W. Quadruple Hydrogen Bonded Systems. Chem. Commun. 2003.

(62) Blanazs, A.; Armes, S. P.; Ryan, A. J. Self-Assembled Block Copolymer Aggregates: From Micelles to Vesicles and Their Biological Applications. Macromol. Rapid Commun. 2009, 30, 267–277.

(63) Hayward, R. C.; Pochan, D. J. Tailored Assemblies of Block Copolymers in Solution: It Is All About the Process. Macromolecules 2010, 43, 3577–3584.

(64) Ferguson, C. J.; Hughes, R. J.; Pham, B. T. T.; Hawkett, B. S.; Gilbert, R. G.; Algirdas K Serelis, A.; Such, C. H. Effective Ab Initio Emulsion Polymerization Under RAFT Control. Macromolecules 2002, 35, 9243–9245.

(65) Ferguson, C. J.; Hughes, R. J.; Nguyen, D.; Pham, B. T. T.; Gilbert, R. G.; Serelis, A. K.; Such, C. H.; Hawkett, B. S. Ab Initio Emulsion Polymerization by RAFT-Controlled Self-Assembly. Macromolecules 2005, 38, 2191–2204.

(66) Yamauchi, K.; Hasegawa, H.; Hashimoto, T.; Tanaka, H.; Motokawa, R.; Koizumi, S. Direct Observation of Polymerization-Reaction-Induced Molecular Self-Assembling Process: in-Situ and Real-Time SANS Measurements During Living Anionic Polymerization of Polyisoprene-Block-Polystyrene. Macromolecules 2006, 39, 4531–4539.

(67) Wan, W.-M.; Pan, C.-Y. One-Pot Synthesis of Polymeric Nanomaterials via RAFT Dispersion Polymerization Induced Self-Assembly and Re-Organization. Polym. Chem. 2010, 1, 1475–1484.

160

(68) Derry, M. J.; Fielding, L. A.; Warren, N. J.; Mable, C. J.; Smith, A. J.; Mykhaylyk, O. O.; Armes, S. P. In Situ Small-Angle X-Ray Scattering Studies of Sterically-Stabilized Diblock Copolymer Nanoparticles Formed During Polymerization-Induced Self-Assembly in Non-Polar Media. Chem. Sci. 2016, 7, 5078–5090.

(69) Blanazs, A.; Ryan, A. J.; Armes, S. P. Predictive Phase Diagrams for RAFT Aqueous Dispersion Polymerization: Effect of Block Copolymer Composition, Molecular Weight, and Copolymer Concentration. Macromolecules 2012, 45, 5099–5107.

(70) Warren, N. J.; Mykhaylyk, O. O.; Ryan, A. J.; Williams, M.; Doussineau, T.; Dugourd, P.; Antoine, R.; Portale, G.; Armes, S. P. Testing the Vesicular Morphology to Destruction: Birth and Death of Diblock Copolymer Vesicles Prepared via Polymerization-Induced Self-Assembly. J. Am. Chem. Soc. 2015, 137, 1929–1937.

(71) Pei, Y.; Lowe, A. B. Polymerization-Induced Self-Assembly: Ethanolic RAFT Dispersion Polymerization of 2-Phenylethyl Methacrylate. Polym. Chem. 2014, 5, 2342–2351.

(72) Zhao, W.; Gody, G.; Dong, S.; Zetterlund, P. B.; Perrier, S. Optimization of the RAFT Polymerization Conditions for the in Situ Formation of Nano-Objects via Dispersion Polymerization in Alcoholic Medium. Polym. Chem. 2014, 5, 6990– 7003.

(73) Kang, Y.; Pitto-Barry, A.; Willcock, H.; Quan, W.-D.; Kirby, N.; Sanchez, A. M.; O'Reilly, R. K. Exploiting Nucleobase-Containing Materials – From Monomers to Complex Morphologies Using RAFT Dispersion Polymerization. Polym. Chem. 2014, 6, 106–117.

(74) Karagoz, B.; Esser, L.; Duong, H. T.; Basuki, J. S.; Boyer, C.; Davis, T. P. Polymerization-Induced Self-Assembly (PISA) – Control Over the Morphology of Nanoparticles for Drug Delivery Applications. Polym. Chem. 2014, 5, 350– 355.

(75) Truong, N. P.; Whittaker, M. R.; Anastasaki, A.; Haddleton, D. M.; Quinn, J. F.; Davis, T. P. Facile Production of Nanoaggregates with Tuneable Morphologies From Thermoresponsive P(DEGMA-Co-HPMA). Polym. Chem. 2016, 7, 430– 440.

(76) Tan, J.; Sun, H.; Yu, M.; Sumerlin, B. S.; Zhang, L. Photo-PISA: Shedding Light on Polymerization-Induced Self-Assembly. ACS Macro Lett. 2015, 1249– 1253.

161

(77) Yeow, J.; Xu, J.; Boyer, C. Polymerization-Induced Self-Assembly Using Visible Light Mediated Photoinduced Electron Transfer–Reversible Addition– Fragmentation Chain Transfer Polymerization. ACS Macro Lett. 2015, 4, 984– 990.

(78) Yeow, J.; Sugita, O. R.; Boyer, C. Visible Light-Mediated Polymerization- Induced Self-Assembly in the Absence of External Catalyst or Initiator. ACS Macro Lett. 2016, 558–564.

(79) Yeow, J.; Shanmugam, S.; Corrigan, N.; Kuchel, R. P.; Xu, J.; Boyer, C. A Polymerization-Induced Self-Assembly Approach to Nanoparticles Loaded with Singlet Oxygen Generators. Macromolecules 2016, acs.macromol.6b01581.

(80) Delaittre, G.; Save, M.; Gaborieau, M.; Castignolles, P.; Rieger, J.; Charleux, B. Synthesis by Nitroxide-Mediated Aqueous Dispersion Polymerization, Characterization, and Physical Core-Crosslinking of pH- and Thermoresponsive Dynamic Diblock Copolymer Micelles. Polym. Chem. 2012, 3, 1526–1538.

(81) Delaittre, G.; Dire, C.; Rieger, J.; Putaux, J.-L.; Charleux, B. Formation of Polymer Vesicles by Simultaneous Chain Growth and Self-Assembly of Amphiphilic Block Copolymers. Chem. Commun. 2009, 2887–2889.

(82) Zhang, W.; D’Agosto, F.; Boyron, O.; Rieger, J.; Charleux, B. Toward a Better Understanding of the Parameters That Lead to the Formation of Nonspherical Polystyrene Particles via RAFT-Mediated One-Pot Aqueous Emulsion Polymerization. Macromolecules 2012, 45, 4075–4084.

(83) Qiao, X. G.; Lansalot, M.; Bourgeat-Lami, E.; Charleux, B. Nitroxide-Mediated Polymerization-Induced Self-Assembly of Poly(Poly(Ethylene Oxide) Methyl Ether Methacrylate-Co-Styrene)-B-Poly(N-Butyl Methacrylate-Co-Styrene) Amphiphilic Block Copolymers. Macromolecules 2013, 46, 4285–4295.

(84) Zhang, X.; Rieger, J.; Charleux, B. Effect of the Solvent Composition on the Morphology of Nano-Objects Synthesized via RAFT Polymerization of Benzyl Methacrylate in Dispersed Systems. Polym. Chem. 2012, 3, 1502–1509.

(85) Warren, N. J.; Armes, S. P. Polymerization-Induced Self-Assembly of Block Copolymer Nano-Objects via RAFT Aqueous Dispersion Polymerization. J. Am. Chem. Soc. 2014, 136, 10174–10185.

(86) Canning, S. L.; Smith, G. N.; Armes, S. P. A Critical Appraisal of RAFT- Mediated Polymerization-Induced Self-Assembly. Macromolecules 2016, 49, 1985–2001.

162

(87) Rieger, J. Guidelines for the Synthesis of Block Copolymer Particles of Various Morphologies by RAFT Dispersion Polymerization. Macromol. Rapid Commun. 2015, 36, 1458–1471.

(88) Moad, G.; Rizzardo, E.; Thang, S. H. Radical Addition–Fragmentation Chemistry in Polymer Synthesis. Polymer 2008, 49, 1079–1131.

(89) Matyjaszewski, K. Atom Transfer Radical Polymerization (ATRP): Current Status and Future Perspectives. Macromolecules 2012, 45, 4015–4039.

(90) Nicolas, J.; Guillaneuf, Y.; Lefay, C.; Bertin, D.; Gigmes, D.; Charleux, B. Nitroxide-Mediated Polymerization. Prog. Polym. Sci. 2013, 38, 63–235.

(91) Charleux, B.; Delaittre, G.; Rieger, J.; D’Agosto, F. Polymerization-Induced Self-Assembly: From Soluble Macromolecules to Block Copolymer Nano- Objects in One Step. Macromolecules 2012, 45, 6753–6765.

(92) Sugihara, S.; Armes, S. P.; Lewis, A. L. One‐Pot Synthesis of Biomimetic Shell Cross‐Linked Micelles and Nanocages by ATRP in Alcohol/Water Mixtures. Angew. Chem. 2010, 122, 3578–3581.

(93) Lowe, A. B.; McCormick, C. L. Reversible Addition–Fragmentation Chain Transfer (RAFT) Radical Polymerization and the Synthesis of Water-Soluble (Co)Polymers Under Homogeneous Conditions in Organic and Aqueous Media. Prog. Polym. Sci. 2007, 32, 283–351.

(94) Karagoz, B.; Boyer, C.; Davis, T. P. Simultaneous Polymerization-Induced Self-Assembly (PISA) and Guest Molecule Encapsulation. Macromol. Rapid Commun. 2014, 35, 417–421.

(95) Karagoz, B.; Yeow, J.; Esser, L.; Prakash, S. M.; Kuchel, R. P.; Davis, T. P.; Boyer, C. An Efficient and Highly Versatile Synthetic Route to Prepare Iron Oxide Nanoparticles/Nanocomposites with Tunable Morphologies. Langmuir 2014, 30, 10493–10502.

(96) Liu, G.; Qiu, Q.; Shen, W.; An, Z. Aqueous Dispersion Polymerization of 2- Methoxyethyl Acrylate for the Synthesis of Biocompatible Nanoparticles Using a Hydrophilic RAFT Polymer and a Redox Initiator. Macromolecules 2011, 44, 5237–5245.

(97) Kang, Y.; Pitto-Barry, A.; Willcock, H.; Quan, W.-D.; Kirby, N.; Sanchez, A. M.; O'Reilly, R. K. Exploiting Nucleobase-Containing Materials – From Monomers to Complex Morphologies Using RAFT Dispersion Polymerization. Polym. Chem. 2014.

163

(98) Blotny, G. Recent Applications of 2,4,6-Trichloro-1,3,5-Triazine and Its Derivatives in Organic Synthesis. Tetrahedron 2006, 62, 9507–9522.

(99) Simanek, E. E.; Abdou, H.; Lalwani, S.; Lim, J.; Mintzer, M.; Venditto, V. J.; Vittur, B. The 8 Year Thicket of Triazine Dendrimers: Strategies, Targets and Applications. Proc. R. Soc. A 2010, 466, 1445–1468.

(100) Lim, J.; Simanek, E. E. Triazine Dendrimers as Drug Delivery Systems: From Synthesis to Therapy. Adv. Drug Deliv. Rev. 2012, 64, 826–835.

(101) Steffensen, M. B.; Hollink, E.; Kuschel, F.; Bauer, M.; Simanek, E. E. Dendrimers Based on [1,3,5]-Triazines. J. Polym. Sci. A Polym. Chem. 2006, 44, 3411–3433.

(102) Zhang, W.; Nowlan, D. T., III; Thomson, L. M.; Lackowski, W. M.; Simanek, E. E. Orthogonal, Convergent Syntheses of Dendrimers Based on Melamine with One or Two Unique Surface Sites for Manipulation. J. Am. Chem. Soc. 2001, 123, 8914–8922.

(103) Hollink, E.; Simanek, E. E. A Divergent Route to Diversity in Macromolecules. Org. Lett. 2006, 8, 2293–2295.

(104) Patra, S.; Kozura, B.; Huang, A. Y. T.; Enciso, A. E.; Sun, X.; Hsieh, J.-T.; Kao, C.-L.; Chen, H.-T.; Simanek, E. E. Dendrimers Terminated with Dichlorotriazine Groups Provide a Route to Compositional Diversity. Org. Lett. 2013, 15, 3808–3811.

(105) Lim, J.; Kostiainen, M.; Maly, J.; da Costa, V. C. P.; Annunziata, O.; Pavan, G. M.; Simanek, E. E. Synthesis of Large Dendrimers with the Dimensions of Small Viruses. J. Am. Chem. Soc. 2013, 135, 4660–4663.

(106) Lim, J.; Simanek, E. E. Synthesis of Water-Soluble Dendrimers Based on Melamine Bearing 16 Paclitaxel Groups. Org. Lett. 2008, 10, 201–204.

(107) Enciso, A. E.; Garzoni, M.; Pavan, G. M.; Simanek, E. E. Influence of Linker Groups on the Solubility of Triazine Dendrimers. New J. Chem. 2015, 1247– 1252.

(108) Enciso, A. E.; Abid, Z. M.; Simanek, E. E. Rapid, Semi-Automated Convergent Synthesis of Low Generation Triazine Dendrimers Using Microwave Assisted Reactions. Polym. Chem. 2014, 5, 4635–4640.

(109) Abuchowski, A.; van Es, T.; Palczuk, N. C.; Davis, F. F. Alteration of Immunological Properties of Bovine Serum Albumin by Covalent Attachment of Polyethylene Glycol. J. Biol. Chem. 1977, 252, 3578–3581.

164

(110) Gotoh, Y.; Tsukada, M.; Minoura, N. Chemical Modification of Silk Fibroin with Cyanuric Chloride-Activated Poly(Ethylene Glycol): Analyses of Reaction Site by Proton NMR Spectroscopy and Conformation of the Conjugates. Bioconjugate Chem. 1993, 4, 554–559.

(111) Shafer, S. G.; Harris, J. M. Preparation of Cyanuric‐Chloride Activated Poly(Ethylene Glycol). J. Polym. Sci. A Polym. Chem. 1986, 24, 375–378.

(112) Falchi, A.; Taddei, M. PEG-Dichlorotriazine (PEG-DCT): a New Soluble Polymer-Supported Scavenger for Alcohols, Thiols, Phosphines, and Phosphine Oxides. Org. Lett. 2000, 2, 3429–3431.

(113) Pelegri-O’Day, E. M.; Lin, E.-W.; Maynard, H. D. Therapeutic Protein–Polymer Conjugates: Advancing Beyond PEGylation. J. Am. Chem. Soc. 2014, 136, 14323–14332.

(114) Treetharnmathurot, B.; Ovartlarnporn, C.; Wungsintaweekul, J.; Duncan, R.; Wiwattanapatapee, R. Effect of PEG Molecular Weight and Linking Chemistry on the Biological Activity and Thermal Stability of PEGylated Trypsin. Int. J. Pharm. 2008, 357, 252–259.

(115) Pasut, G.; Veronese, F. M. State of the Art in PEGylation: the Great Versatility Achieved After Forty Years of Research. J. Control. Release 2012, 161, 461– 472.

(116) Zhang, B.; Zhang, H.; Myers, B. K.; Elupula, R.; Jayawickramarajah, J.; Grayson, S. M. Analytica Chimica Acta. Analytica Chimica Acta 2014, 816, 28– 40.

(117) Opsteen, J. A.; van Hest, J. C. M. Modular Synthesis of Block Copolymers via Cycloaddition of Terminal Azide and Alkyne Functionalized Polymers - Chemical Communications (RSC Publishing). Chem. Commun. 2005, 57–59.

(118) Harris, J. M.; Struck, E. C.; Case, M. G.; Paley, M. S.; Yalpani, M.; Van Alstine, J. M.; Brooks, D. E. Synthesis and Characterization of Poly(Ethylene Glycol) Derivatives. J. Polym. Sci. Polym. Chem. Ed. 1984, 22, 341–352.

(119) Zalipsky, S. Functionalized Poly(Ethylene Glycols) for Preparation of Biologically Relevant Conjugates. Bioconjugate Chem. 1995, 6, 150–165.

(120) Zhao, Y.; Yang, B.; Zhang, Y.; Wang, S.; Fu, C.; Wei, Y.; Tao, L. Fluorescent PEGylation Agent by a Thiolactone-Based One-Pot Reaction: a New Strategy for Theranostic Combinations. Polym. Chem. 2014, 5, 6656–6661.

165

(121) Katritzky, A. R.; Ghiviriga, I.; Steel, P. J.; Oniciu, D. C. Restricted Rotations in 4,6-Bis- and 2,4,6-Tris-(N,N-Dialkylamino)-S-Triazines. J. Chem. Soc., Perkin Trans. 2 1995, 443–447.

(122) Figg, C. A.; Bartley, A. N.; Kubo, T.; Tucker, B. S.; Castellano, R. K.; Sumerlin, B. S. Mild and Efficient Synthesis of Omega, Omega-Heterodifunctionalized Polymers and Polymer Bioconjugates. Polym. Chem. 2017, 8, 2457–2461.

(123) Baker, M. B.; Ghiviriga, I.; Castellano, R. K. Molecular Multifunctionalization via Electronically Coupled Lactones. Chem. Sci. 2012, 3, 1095–1099.

(124) Li, Y.; Lampkins, A. J.; Baker, M. B.; Sumpter, B. G.; Huang, J.; Abboud, K. A.; Castellano, R. K. Benzotrifuranone: Synthesis, Structure, and Access to Polycyclic Heteroaromatics. Org. Lett. 2009, 11, 4314–4317.

(125) Figg, C. A.; Simula, A.; Gebre, K. A.; Tucker, B. S.; Haddleton, D. M.; Sumerlin, B. S. Polymerization-Induced Thermal Self-Assembly (PITSA). Chem. Sci. 2015, 6, 1230–1236.

(126) Baker, M. B.; Ferreira, R. B.; Tasseroul, J.; Lampkins, A. J.; Abbas, Al, A.; Abboud, K. A.; Castellano, R. K. Selective and Sequential Aminolysis of Benzotrifuranone: Synergism of Electronic Effects and Ring Strain Gradient. J. Org. Chem. 2016, 81, 9279–9288.

(127) Yang, B.; Zhao, Y.; Wang, S.; Zhang, Y.; Fu, C.; Wei, Y.; Tao, L. Synthesis of Multifunctional Polymers Through the Ugi Reaction for Protein Conjugation. Macromolecules 2014, 47, 5607–5612.

(128) Zhang, Q.; Zhao, Y.; Yang, B.; Fu, C.; Zhao, L.; Wang, X.; Wei, Y.; Tao, L. Lighting Up the PEGylation Agents via the Hantzsch Reaction. Polym. Chem. 2016, 7, 523–528.

(129) Elbert, D. L.; Hubbell, J. A. Conjugate Addition Reactions Combined with Free- Radical Cross-Linking for the Design of Materials for Tissue Engineering. Biomacromolecules 2001, 2, 430–441.

(130) Koo, S. P. S.; Stamenović, M. M.; Prasath, R. A.; Inglis, A. J.; Prez, Du, F. E.; Barner-Kowollik, C.; Van Camp, W.; Junkers, T. Limitations of Radical Thiol‐ Ene Reactions for Polymer–Polymer Conjugation. J. Polym. Sci. A Polym. Chem. 2010, 48, 1699–1713.

(131) Livnah, O.; Bayer, E. A.; Wilchek, M.; Sussman, J. L. Three-Dimensional Structures of Avidin and the Avidin-Biotin Complex. PNAS 1993, 90, 5076– 5080.

166

(132) Bayer, E. A.; Ehrlich-Rogozinski, S.; Wilchek, M. Sodium Dodecyl Sulfate‐ Polyacrylamide Gel Electrophoretic Method for Assessing the Quaternary State and Comparative Thermostability of Avidin and Streptavidin. Electrophoresis 1996, 17, 1319–1324.

(133) Xu, J.; Shanmugam, S.; Duong, H. T.; Boyer, C. Organo-Photocatalysts for Photoinduced Electron Transfer-Reversible Addition–Fragmentation Chain Transfer (PET-RAFT) Polymerization. Polym. Chem. 2015, 6, 5615–5624.

(134) Lee, I.-H.; Discekici, E. H.; Anastasaki, A.; de Alaniz, J. R.; Hawker, C. J. Controlled Radical Polymerization of Vinyl Ketones Using Visible Light. Polym. Chem. 2017, 8, 3351–3356.

(135) Tucker, B. S.; Coughlin, M. L.; Figg, C. A.; Sumerlin, B. S. Grafting-From Proteins Using Metal-Free PET–RAFT Polymerizations Under Mild Visible- Light Irradiation. ACS Macro Lett. 2017, 6, 452–457.

(136) Yeow, J.; Chapman, R.; Xu, J.; Boyer, C. Oxygen Tolerant Photopolymerization for Ultralow Volumes. Polym. Chem. 2017, 8, 5012–5022.

(137) Chakraborty, M.; Panda, A. K. Spectral Behaviour of Eosin Y in Different Solvents and Aqueous Surfactant Media. Spectrochim. Acta, Part A 2011, 81, 458–465.

(138) McKenzie, T. G.; da M Costa, L. P.; Fu, Q.; Dunstan, D. E.; Qiao, G. G. Investigation Into the Photolytic Stability of RAFT Agents and the Implications for Photopolymerization Reactions. Polym. Chem. 2016, 7, 4246–4253.

(139) Shastri, L. V.; Huie, R. E.; Neta, P. Formation and Reactivity of Hypophosphite and Phosphite Radicals and Their Peroxyl Derivatives. J. Phys. Chem. 1990, 94, 1895–1899.

(140) Carmean, R. N.; Figg, C. A.; Scheutz, G. M.; Kubo, T.; Sumerlin, B. S. Catalyst-Free Photoinduced End-Group Removal of Thiocarbonylthio Functionality. ACS Macro Lett. 2017, 6, 185–189.

(141) Fu, Q.; Xie, K.; McKenzie, T. G.; Qiao, G. G. Trithiocarbonates as Intrinsic Photoredox Catalysts and RAFT Agents for Oxygen Tolerant Controlled Radical Polymerization. Polym. Chem. 2017, 8, 1519–1526.

(142) Fu, Q.; Ruan, Q.; McKenzie, T. G.; Reyhani, A.; Tang, J.; Qiao, G. G. Development of a Robust PET-RAFT Polymerization Using Graphitic Carbon Nitride (G-C3N4). Macromolecules 2017, 50, 7509–7516.

167

(143) Fu, Q.; McKenzie, T. G.; Ren, J. M.; Tan, S.; Nam, E.; Qiao, G. G. A Novel Solid State Photocatalyst for Living Radical Polymerization Under UV Irradiation. Sci. Rep. 2016, 6, 5559.

(144) Allegrezza, M. L.; DeMartini, Z. M.; Kloster, A. J.; Digby, Z. A.; Konkolewicz, D. Visible and Sunlight Driven RAFT Photopolymerization Accelerated by Amines: Kinetics and Mechanism. Polym. Chem. 2016, 7, 6626–6636.

(145) Lazarides, T.; McCormick, T.; Du, P.; Luo, G.; Lindley, B.; Eisenberg, R. Making Hydrogen From Water Using a Homogeneous System Without Noble Metals. J. Am. Chem. Soc. 2009, 131, 9192–9194.

(146) Xu, J.; Shanmugam, S.; Duong, H.; Boyer, C. Organo-Photocatalysts for Photoinduced Electron Transfer – Reversible Addition-Fragmentation Chain Transfer (PET-RAFT) Polymerization. Polym. Chem. 2014.

(147) Grossweiner, L. I.; Zwicker, E. F. Transient Measurements of Photochemical Processes in Dyes. I. the Photosensitized Oxidation of Phenol by Eosin and Related Dyes. J. Chem. Phys. 1961, 34, 1411–1417.

(148) Zwicker, E. F.; Grossweiner, L. I. Transient Measurements of Photochemical Processes in Dyes. Ii. the Mechanism of the Photosensitized Oxidation of Aqueous Phenol by Eosin 1. J. Phys. Chem. 1963, 67, 549–555.

(149) Carmean, R. N.; Becker, T. E.; Sims, M. B.; Sumerlin, B. S. Ultra-High Molecular Weights via Aqueous Reversible-Deactivation Radical Polymerization. Chem 2017, 2, 93–101.

(150) Hu, J.; Wang, J.; Nguyen, T. H.; Zheng, N. The Chemistry of Amine Radical Cations Produced by Visible Light Photoredox Catalysis. Beilstein J. Org. Chem. 2013, 9, 1977–2001.

(151) Phillips, K.; Read, G. Novel Radical Couplings in the Photoreduction of Xanthenoid Dyes with Tribenzylamine. J. Chem. Soc., Perkin Trans. 1 1986, 0, 671–673.

(152) Zakrzewski, A.; Neckers, D. C. Bleaching Products of Rose Bengal Under Reducing Conditions. Tetrahedron 1987, 43, 4507–4512.

(153) Amat-Guerri, F.; López-González, M. M. C.; Martínez-Utrilla, R.; Sastre, R. Singlet Oxygen Photogeneration by Ionized and Un-Ionized Derivatives of Rose Bengal and Eosin Y in Diluted Solutions. J. Photochem. Photobiol. 1990, 53, 199–210.

168

(154) Herculano, L. S.; Malacarne, L. C.; Zanuto, V. S.; Lukasievicz, G. V. B.; Capeloto, O. A.; Astrath, N. G. C. Investigation of the Photobleaching Process of Eosin Y in Aqueous Solution by Thermal Lens Spectroscopy. J. Phys. Chem. B 2013, 117, 1932–1937.

(155) Padon, K. S.; Scranton, A. B. A Mechanistic Investigation of the Three‐ Component Radical Photoinitiator System Eosin Y Spirit Soluble, N‐ Methyldiethanolamine, and Diphenyliodonium Chloride. J. Polym. Sci. A Polym. Chem. 2001, 39, 715–723.

(156) Avens, H. J.; Bowman, C. N. Mechanism of Cyclic Dye Regeneration During Eosin‐Sensitized Photoinitiation in the Presence of Polymerization Inhibitors. J. Polym. Sci. A Polym. Chem. 2009, 47, 6083–6094.

(157) Stuart, M. A. C.; Huck, W. T. S.; Genzer, J.; Müller, M.; Ober, C.; Stamm, M.; Sukhorukov, G. B.; Szleifer, I.; Tsukruk, V. V.; Urban, M.; et al. Emerging Applications of Stimuli-Responsive Polymer Materials. Nat. Mater. 2010, 9, 101–113.

(158) Roy, D.; Brooks, W. L. A.; Sumerlin, B. S. New Directions in Thermoresponsive Polymers. Chem. Soc. Rev. 2013, 42, 7214–7243.

(159) Lutz, J.-F.; Akdemir, Ö.; Hoth, A. Point by Point Comparison of Two Thermosensitive Polymers Exhibiting a Similar LCST: Is the Age of Poly(NIPAM) Over? J. Am. Chem. Soc. 2006, 128, 13046–13047.

(160) De, P.; Sumerlin, B. S. Precision Control of Temperature Response by Copolymerization of Di(Ethylene Glycol) Acrylate and an Acrylamide Comonomer. Macromol. Chem. Phys. 2013, 214, 272–279.

(161) Barker, I. C.; Cowie, J. M. G.; Huckerby, T. N.; Shaw, D. A.; Soutar, I.; Swanson, L. Studies of the “Smart” Thermoresponsive Behavior of Copolymers of N-Isopropylacrylamide and N,N-Dimethylacrylamide in Dilute Aqueous Solution. Macromolecules 2003, 36, 7765–7770.

(162) Gibson, M. I.; O'Reilly, R. K. To Aggregate, or Not to Aggregate? Considerations in the Design and Application of Polymeric Thermally- Responsive Nanoparticles. Chem. Soc. Rev. 2013, 42, 7204–7213.

(163) De, P.; Li, M.; Gondi, S. R.; Sumerlin, B. S. Temperature-Regulated Activity of Responsive Polymer−Protein Conjugates Prepared by Grafting-From via RAFT Polymerization. J. Am. Chem. Soc. 2008, 130, 11288–11289.

(164) Li, H.; Bapat, A. P.; Li, M.; Sumerlin, B. S. Protein Conjugation of Thermoresponsive Amine-Reactive Polymers Prepared by RAFT. Polym. Chem. 2011, 2, 323–327.

169

(165) Jones, M. W.; Gibson, M. I.; Mantovani, G.; Haddleton, D. M. Tunable Thermo- Responsive Polymer–Protein Conjugates via a Combination of Nucleophilic Thiol–Ene “Click” and SET-LRP. Polym. Chem. 2011, 2, 572–574.

(166) Xu, H.; Meng, F.; Zhong, Z. Reversibly Crosslinked Temperature-Responsive Nano-Sized Polymersomes: Synthesis and Triggered Drug Release. J. Mater. Chem. 2009, 19, 4183–4190.

(167) Kessel, S.; Truong, N. P.; Jia, Z.; Monteiro, M. J. Aqueous Reversible Addition‐ Fragmentation Chain Transfer Dispersion Polymerization of Thermoresponsive Diblock Copolymer Assemblies: Temperature Directed Morphology Transformations. J. Polym. Sci. A Polym. Chem. 2012, 50, 4879–4887.

(168) An, Z.; Shi, Q.; Tang, W.; Tsung, C.-K.; Hawker, C. J.; Stucky, G. D. Facile RAFT Precipitation Polymerization for the Microwave-Assisted Synthesis of Well-Defined, Double Hydrophilic Block Copolymers and Nanostructured Hydrogels. J. Am. Chem. Soc. 2007, 129, 14493–14499.

(169) Xu, Y.; Li, Y.; Cao, X.; Chen, Q.; An, Z. Versatile RAFT Dispersion Polymerization in Cononsolvents for the Synthesis of Thermoresponsive Nanogels with Controlled Composition, Functionality and Architecture. Polym. Chem. 2014, 5, 6244–6255.

(170) Li, Q.; Huo, F.; Cui, Y.; Gao, C.; Li, S.; Zhang, W. Doubly Thermoresponsive Brush‐Linear‐Linear ABC Triblock Copolymer Nanoparticles Prepared Through Dispersion RAFT Polymerization. J. Polym. Sci. A Polym. Chem. 2014, 52, 2266–2278.

(171) Li, Q.; Gao, C.; Li, S.; Huo, F.; Zhang, W. Doubly Thermo-Responsive ABC Triblock Copolymer Nanoparticles Prepared Through Dispersion RAFT Polymerization. Polym. Chem. 2014, 5, 2961–2972.

(172) Blanazs, A.; Verber, R.; Mykhaylyk, O. O.; Ryan, A. J.; Heath, J. Z.; Douglas, C. W. I.; Armes, S. P. Sterilizable Gels From Thermoresponsive Block Copolymer Worms. J. Am. Chem. Soc. 2012, 134, 9741–9748.

(173) Cunningham, V.; Ratcliffe, L.; Blanazs, A.; Warren, N. J. Tuning the Critical Gelation Temperature of Thermo-Responsive Diblock Copolymer Worm Gels. Polymer 2014, 5, 6307–6317.

(174) Blanazs, A.; Madsen, J.; Battaglia, G.; Ryan, A. J.; Armes, S. P. Mechanistic Insights for Block Copolymer Morphologies: How Do Worms Form Vesicles? J. Am. Chem. Soc. 2011, 133, 16581–16587.

170

(175) Wu, C.; Zhou, S. Laser Light Scattering Study of the Phase Transition of Poly(N-Isopropylacrylamide) in Water. 1. Single Chain. Macromolecules 1995, 28, 8381–8387.

(176) Nakajima, N.; Ikada, Y. Mechanism of Amide Formation by Carbodiimide for Bioconjugation in Aqueous Media. Bioconjugate Chem. 1995, 6, 123–130.

(177) Zhou, W.; Qu, Q.; Xu, Y.; An, Z. Aqueous Polymerization-Induced Self- Assembly for the Synthesis of Ketone-Functionalized Nano-Objects with Low Polydispersity. ACS Macro Lett. 2015, 495–499.

(178) Qu, Q.; Liu, G.; Lv, X.; Zhang, B.; An, Z. In Situ Cross-Linking of Vesicles in Polymerization-Induced Self-Assembly. ACS Macro Lett. 2016, 316–320.

(179) Jiang, Y.; Xu, N.; Han, J.; Yu, Q.; Guo, L.; Gao, P.; Lu, X.; Cai, Y. The Direct Synthesis of Interface-Decorated Reactive Block Copolymer Nanoparticles via Polymerisation-Induced Self-Assembly. Polym. Chem. 2015, 6, 4955–4965.

(180) Tang, X.; Han, J.; Zhu, Z.; Lu, X.; Chen, H.; Cai, Y. Facile Synthesis, Sequence-Tuned Thermoresponsive Behaviours and Reaction-Induced Reorganization of Water-Soluble Keto-Polymers. Polym. Chem. 2014, 5, 4115–4123.

(181) Mukherjee, S.; Hill, M. R.; Sumerlin, B. S. Self-Healing Hydrogels Containing Reversible Oxime Crosslinks. Soft Matter 2015, 11, 6152–6161.

(182) Yang, P.; Ratcliffe, L. P. D.; Armes, S. P. Efficient Synthesis of Poly(Methacrylic Acid)-Block-Poly(Styrene-Alt-N-Phenylmaleimide) Diblock Copolymer Lamellae Using RAFT Dispersion Polymerization. Macromolecules 2013, 46, 8545–8556.

(183) Shi, P.; Zhou, H.; Gao, C.; Wang, S.; Sun, P.; Zhang, W. Macro-RAFT Agent Mediated Dispersion Copolymerization: a Small Amount of Solvophilic Co- Monomer Leads to a Great Change. Polym. Chem. 2015, 6, 4911–4920.

(184) Zhou, J.; Zhang, W.; Hong, C.; Pan, C. Promotion of Morphology Transition of Di-Block Copolymer Nano-Objects via RAFT Dispersion Copolymerization. Polym. Chem. 2016, 7, 3259–3267.

(185) Ratcliffe, L. P. D.; Blanazs, A.; Williams, C. N.; Brown, S. L.; Armes, S. P. RAFT Polymerization of Hydroxy-Functional Methacrylic Monomers Under Heterogeneous Conditions: Effect of Varying the Core-Forming Block. Polym. Chem. 2014, 5, 3643–3655.

171

(186) Figg, C. A.; Kubo, T.; Sumerlin, B. S. Efficient and Chemoselective Synthesis of Omega, Omega-Heterodifunctional Polymers. ACS Macro Lett. 2015, 1114– 1118.

(187) Jacquin, M.; Muller, P.; Cottet, H.; Théodoly, O. Self-Assembly of Charged Amphiphilic Diblock Copolymers with Insoluble Blocks of Decreasing Hydrophobicity: From Kinetically Frozen Colloids to Macrosurfactants. Langmuir 2010, 26, 18681–18693.

(188) Lejeune, E.; Drechsler, M.; Jestin, J.; Müller, A. H. E.; Chassenieux, C.; Colombani, O. Amphiphilic Diblock Copolymers with a Moderately Hydrophobic Block: Toward Dynamic Micelles. Macromolecules 2010, 43, 2667–2671.

(189) Laruelle, G.; François, J.; Billon, L. Self-Assembly in Water of Poly(Acrylic Acid)-Based Diblock Copolymers Synthesized by Nitroxide-Mediated Polymerization. Macromol. Rapid Commun. 2004, 25, 1839–1844.

(190) Bendejacq, D. D.; Ponsinet, V.; Joanicot, M. Chemically Tuned Amphiphilic Diblock Copolymers Dispersed in Water: From Colloids to Soluble Macromolecules. Langmuir 2005, 21, 1712–1718.

(191) Ratcliffe, L. P. D.; Ryan, A. J.; Armes, S. P. From a Water-Immiscible Monomer to Block Copolymer Nano-Objects via a One-Pot RAFT Aqueous Dispersion Polymerization Formulation. Macromolecules 2013, 46, 769–777.

(192) Derry, M. J.; Fielding, L. A.; Armes, S. P. Industrially-Relevant Polymerization- Induced Self-Assembly Formulations in Non-Polar Solvents: RAFT Dispersion Polymerization of Benzyl Methacrylate. Polym. Chem. 2015, 6, 3054–3062.

(193) Truong, N. P.; Quinn, J. F.; Whittaker, M. R.; Davis, T. P. Polymeric Filomicelles and Nanoworms: Two Decades of Synthesis and Application. Polym. Chem. 2016, 7, 4295–4312.

(194) Gilroy, J. B.; Gädt, T.; Whittell, G. R.; Chabanne, L.; Mitchels, J. M.; Richardson, R. M.; Winnik, M. A.; Manners, I. Monodisperse Cylindrical Micelles by Crystallization-Driven Living Self-Assembly. Nat. Chem. 2010, 2, 566–570.

(195) Burke, S. E.; Eisenberg, A. Kinetics and Mechanisms of the Sphere-to-Rod and Rod-to-Sphere Transitions in the Ternary System PS310-B- PAA52/Dioxane/Water. Langmuir 2001, 17, 6705–6714.

(196) Zhang, L.; Eisenberg, A. Thermodynamic vs Kinetic Aspects in the Formation and Morphological Transitions of Crew-Cut Aggregates Produced by Self- Assembly of Polystyrene-B-Poly(Acrylic Acid) Block Copolymers in Dilute Solution. Macromolecules 1999, 32, 2239–2249.

172

(197) Chen, L.; Shen, H.; Eisenberg, A. Kinetics and Mechanism of the Rod-to- Vesicle Transition of Block Copolymer Aggregates in Dilute Solution. J. Phys. Chem. B 1999, 103, 9488–9497.

(198) Zhang, L.; Eisenberg, A. Formation of Crew-Cut Aggregates of Various Morphologies From Amphiphilic Block Copolymers in Solution. Polym. Adv. Technol. 1998, 9, 677–699.

(199) Truong, N. P.; Quinn, J. F.; Dussert, M. V.; Sousa, N. B. T.; Whittaker, M. R.; Davis, T. P. Reproducible Access to Tunable Morphologies via the Self- Assembly of an Amphiphilic Diblock Copolymer in Water. ACS Macro Lett. 2015, 381–386.

(200) Jain, S.; Bates, F. S. On the Origins of Morphological Complexity in Block Copolymer Surfactants. Science 2003, 300, 460–464.

(201) Fernyhough, C.; Ryan, A. J.; Battaglia, G. pH Controlled Assembly of a Polybutadiene–Poly(Methacrylic Acid) Copolymer in Water: Packing Considerations and Kinetic Limitations. Soft Matter 2009, 5, 1674–1682.

(202) Lovett, J. R.; Warren, N. J.; Ratcliffe, L. P. D.; Kocik, M. K.; Armes, S. P. pH- Responsive Non-Ionic Diblock Copolymers: Ionization of Carboxylic Acid End- Groups Induces an Order-Order Morphological Transition. Angew. Chem. Int. Ed. 2015, 54, 1279–1283.

(203) Penfold, N. J. W.; Lovett, J. R.; Warren, N. J.; Verstraete, P.; Smets, J.; Armes, S. P. pH-Responsive Non-Ionic Diblock Copolymers: Protonation of a Morpholine End-Group Induces an Order–Order Transition. Polym. Chem. 2016, 7, 79–88.

(204) Zhu, J.; Hayward, R. C. Wormlike Micelles with Microphase-Separated Cores From Blends of Amphiphilic AB and Hydrophobic BC Diblock Copolymers. Macromolecules 2008, 41, 7794–7797.

(205) Mukherjee, S.; Bapat, A. P.; Hill, M. R.; Sumerlin, B. S. Oximes as Reversible Links in Polymer Chemistry: Dynamic Macromolecular Stars. Polym. Chem. 2014, 5, 6923–6931.

(206) Kalia, J.; Raines, R. T. Hydrolytic Stability of Hydrazones and Oximes. Angew. Chem. 2008, 120, 7633–7636.

(207) Hill, M. R.; Mukherjee, S.; Costanzo, P. J.; Sumerlin, B. S. Modular Oxime Functionalization of Well-Defined Alkoxyamine-Containing Polymers. Polym. Chem. 2012, 3, 1758–1762.

173

(208) Jones, E. R.; Mykhaylyk, O. O.; Semsarilar, M.; Boerakker, M.; Wyman, P.; Armes, S. P. How Do Spherical Diblock Copolymer Nanoparticles Grow During RAFT Alcoholic Dispersion Polymerization? Macromolecules 2016, 49, 172– 181.

(209) Lund, R.; Willner, L.; Richter, D.; Lindner, P.; Narayanan, T. Kinetic Pathway of the Cylinder-to-Sphere Transition in Block Copolymer Micelles Observed in Situ by Time-Resolved Neutron and Synchrotron Scattering. ACS Macro Lett. 2013, 2, 1082–1087.

(210) Zhu, J.; Hayward, R. C. Spontaneous Generation of Amphiphilic Block Copolymer Micelles with Multiple Morphologies Through Interfacial Instabilities. J. Am. Chem. Soc. 2008, 130, 7496–7502.

(211) Lei, L.; Gohy, J.-F.; Willet, N.; Zhang, J.-X.; Varshney, S.; Jérôme, R. Tuning of the Morphology of Core−Shell−Corona Micelles in Water. I. Transition From Sphere to Cylinder. Macromolecules 2003, 37, 1089–1094.

(212) Li, H.; Bapat, A. P.; Li, M.; Sumerlin, B. S. Protein Conjugation of Thermoresponsive Amine-Reactive Polymers Prepared by RAFT. Polym. Chem. 2011, 2, 323–327.

174

BIOGRAPHICAL SKETCH

Adrian was born and raised in Los Alamos, NM with the expectation that the last thing he would do in life is chemistry. Having chemist parents that worked at Los

Alamos National Lab, he grew bored with hearing incessant chemistry dialogue after 18 years. Therefore, he enrolled at the University of California, Santa Barbara (UCSB) for college, far away from the reaches of parental interests. Hoping the beach, weather, and social activities would distract him from the central science, Adrian thought he was safe. Much to his dismay, however, the brainwashing kicked in during organic chemistry. Adrian was enthralled with the conceptual fundamentals of organic reactions.

Driven by his desire to travel, Adrian started his chemical research through an exchange program with Bert Meijer’s polymer research group in Eindhoven,

Netherlands. Upon returning to UCSB, Adrian promptly began work in Craig Hawker’s lab under the guidance of Brett Fors—so began Adrian’s love affair with polymers. With a healthy work-life balance, Adrian was able to graduate with a degree in organic chemistry, almost 2 years of research experience, and a commitment to attend the

University of Florida for his Ph.D.

Adrian began his graduate career with the notion of getting a good government or industrial job, not pursing academia. In parallel with the rest of his chemistry career,

Adrian soon realized that what he thought he wanted the least, he wanted the most.

Diligent hours in the literature and hood, coupled with his desire to understand conceptual fundamentals, led to success in the lab. His desire to travel led to a summer conducting research in Australia and conferences in the UK, California, Washington,

D.C., and Massachusetts. Now on the cusp of post-graduate life, Adrian cannot wait to see what chemistry, and the world, have in store for him.

175