Citation
Synthesis and Improvement of Self-Healing Boronate Ester Hydrogels

Material Information

Title:
Synthesis and Improvement of Self-Healing Boronate Ester Hydrogels
Creator:
Deng, Christopher Chi-Long
Publisher:
University of Florida
Publication Date:
Language:
English

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Chemistry
Committee Chair:
SUMERLIN,BRENT S
Committee Co-Chair:
WAGENER,KENNETH B
Committee Members:
MILLER,STEPHEN ALBERT
VEIGE,ADAM S
BRENNAN,ANTHONY B

Subjects

Subjects / Keywords:
boronic
hydrogels

Notes

General Note:
Boronic acids have been a prevalent moiety in both organic chemistry and materials science where it has seen use as coupling agents, material crosslinks, and sensors. Much of the usefulness of boronic acid comes from its ability to bind with cis diols. The formation of these boronic and boronate esters is facile and reversible, creating a convenient dynamic covalent bond that can be used for a variety of purposes. Furthermore, the preponderance of diols in biological systems enhances the applicability of boronic acids in the biomaterials field. In this work, intramolecularly coordinated boronic acid monomers were used to create a variety of hydrogels that could heal at neutral and acidic pH. The ability to heal at these pH ranges is useful as boronic acids typically only bind at alkaline pH ranges, limiting its efficacy for biological applications. The first set of hydrogels were shown to heal in both qualitative cut and heal tests and in rheological measurements. The second set of healing hydrogels were synthesized from an improved boronic acid monomer. This new monomer addressed issues in copolymerization behavior in the first monomer that resulted in cloudiness in the hydrogel. The lack of transparency could possibly cause issues for using the hydrogel in optical applications. The copolymerization behavior of the new monomer was quantified with reactivity ratios and compared to the old monomer. The improved hydrogel remained clear and possessed better mechanical properties. Finally, a method was developed to selectively control the healing and stress relaxation behavior of the boronic acid hydrogel. In this method, a thermoresponsive copolymer was used to synthesize the hydrogel. Raising the temperature near the cloud point of the copolymer caused solubility induced aggregation, restricting polymer mobility, which in turn stopped healing and reduced stress relaxation behavior. The control of these behaviors could be of use in a cell culture scaffold and other applications. However, there are some drawbacks including reduction of modulus while restricting polymer mobility.

Record Information

Source Institution:
UFRGP
Rights Management:
All applicable rights reserved by the source institution and holding location.
Embargo Date:
12/31/2018

Downloads

This item has the following downloads:


Full Text

PAGE 1

SYNTHESIS AND IMPROVEMENT OF SELF HEALING BORONATE ESTER HYDROGELS By CHRISTOPHER CHI LONG DENG A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR T HE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2017

PAGE 2

2017 Christopher Chi long Deng

PAGE 3

To friends and family and loved ones departed

PAGE 4

4 ACKNOWLEDGMENTS Numerous people have supported me to this p oint. I would like to thank all of my previous teachers throughout my education for the foundation they gave to me and the valuable work that they do for all students. I thank Dr. Yi Zhang for allowing me to work in her lab and gain valuable experience. I would also like to thank Dr. Bill Dolbier for sharing his passion for organic chemistry and assisting me in entering graduate school. Of course, I am grateful towards Dr. Brent Sumerlin for accepting me in his group. His expectations and guidance have been valuable in my scientific growth. Also, I would like to thank my committee of Dr. Ken Wagener, Dr. Stephen Miller, Dr. Adam Veige, and Dr. Anthony Brennan. I appreciate the time taken from their busy schedules and their advice over the years. I would also like to acknowledge the University of Florida, the Department of Chemistry, and the Butler Polymer Laboratory for the opportunity to pursue my doctorate and providing a supportive environment for myself and all graduate students. I am grateful to the memb ers of the Sumerlin group. I have the most interaction with them on a daily basis and it has been a tremendous source of support both personally and professionally I acknowledge Dr. Will Brooks, my first lab mate for sharing in the agony of boronic acid s his general advice, and his friendship. I also thank Chris Kabb for being my second lab mate providing great advice, and putting up with me in general. I also extend my gratitude towards Dr. lin group members. You two were great friends, and together we occupied a unique position in the history of the Sumerlin lab. I thank Nick Carmean for being my on and off again workout partner and entertaining my automotive fantasies. Maybe the rally car d ream will come true one day. Finally, I thank my family. I know that I reach out as much as I should, but I appreciate that you are always there for me if I need it.

PAGE 5

5 T ABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 8 LIST OF FIGURES ................................ ................................ ................................ ......................... 9 LIST OF SCHEMES ................................ ................................ ................................ ...................... 13 LIST OF ABBREVIATIONS ................................ ................................ ................................ ........ 14 ABSTRACT ................................ ................................ ................................ ................................ ... 16 CHAPTER 1 SYNTHESIS AND APPLICATIONS OF BORON IC ACID HYDROGELS ....................... 18 1.1 Introduction ................................ ................................ ................................ ....................... 18 1.2 Boronic A cid H ydrogels ................................ ................................ ................................ ... 20 1.2.1 Early Examples of Boronic Acid H ydrogels ................................ .............................. 20 1.2.2 A B rief I ntroduction to R heology ................................ ................................ .............. 22 1.2.3 Self healing B oronic A cid H ydrogels ................................ ................................ ........ 2 6 1.2.4 Intramolecularly C oordinated B oronic A cids and H ydrogels D erived T he reof ........ 3 3 2 RESEARCH OBJECTIVE ................................ ................................ ................................ ..... 3 7 3 BORONATE ESTER HYDROGELS UNDERGO SELF HEALING AT NEUTRAL AND ACIDIC PH ................................ ................................ ................................ ................... 40 3.1 Introduction ................................ ................................ ................................ ....................... 40 3.2 Results a nd Discussion ................................ ................................ ................................ ..... 42 3.2.1 Synthesis of 2APBA Copolymers and Hydrogel F ormation ................................ ..... 42 3.2.2 Investigation of H ydrogel S elf healing ................................ ................................ ...... 4 5 3.2.3 Rheological S tudies ................................ ................................ ................................ .... 4 5 3.2.4 Examining the E ffect of C atechol O xidation on S elf healing ................................ ... 4 7 3.3 Conclusions ................................ ................................ ................................ ....................... 4 9 3.4 Experimental ................................ ................................ ................................ ..................... 50 3.4.1 Materials ................................ ................................ ................................ ..................... 50 3.4.2 Instrumentation and C haracterization ................................ ................................ ........ 51 3.4.3 Synthesis of 2 A crylamidopheynlboronic A cid P inacol E ster (2APBAE) ................ 52 3.4.4 Synthesis of Poly(2APBA co DMA) (P(2AP BA co DMA)) ................................ .... 52 3.4.5 Synthesis of Poly(PFPA co DMA) (P(PFPA co DMA)) ................................ .......... 5 3 3.4.6 Synthesis of Poly(DOPAAm co DMA) (P(DOPAAm co DMA) ............................. 5 3 3.4.7 Hydrogel S ynthes is ................................ ................................ ................................ .... 5 4

PAGE 6

6 4 C HANGE IN INTRAMOLECU LARLY COORDINATED BORONIC ACID MONOMER STRUCTURE RE SULTS IN IMPROVED SE LF HEALING HYDROGEL PROPERTIES ................................ ................................ ................................ .. 5 5 4.1 Introduction ................................ ................................ ................................ ....................... 5 5 4.2 Results and Discussion ................................ ................................ ................................ ..... 5 8 4.2.1 Synthesis of APAPBA C opolymers ................................ ................................ ........... 5 8 4.2.2 Reactivity R atio S tudies of APAPBAE and 2APBAE ................................ .............. 5 9 4.2.3 Hydrogel F ormation, T ransparency C omparison, and S elf healing S tudies .............. 61 4.2.4 Rheological S tudies ................................ ................................ ................................ .... 62 4.2.5 Effect of Buffer C oncentration on R heology ................................ ............................. 67 4 .3 Conclusions ................................ ................................ ................................ ....................... 6 8 4 .4 Experimental ................................ ................................ ................................ ..................... 6 8 4 .4.1 Materials ................................ ................................ ................................ ..................... 6 8 4 .4.2 Instrumentation and C haracterization ................................ ................................ ........ 6 9 4 .4.3 Synthesis of 3 A crylamido propanoic A cid ................................ ................................ 6 9 4 .4.4 Synt hesis of (2 (3 A cryla midopropanamido)phenyl)boronic A cid P inacol E ster (APAPBAE) ................................ ................................ ................................ ........................ 70 4 .4.5 Synthesis of P oly(APAPBA co DMA) (P(APAPBA co DMA)) ............................. 71 4 .4.6 Hydrogel S ynthesis ................................ ................................ ................................ .... 71 4 .4.7 Reactivity R atios ................................ ................................ ................................ ........ 71 5 TEMPERATURE INDUCED CONTROL OF SELF HEALING AND STRESS RELAXATION IN DYNAMIC COVALENT HYDROGELS ................................ ............. 7 3 5.1 Introduction ................................ ................................ ................................ ....................... 7 3 5.2 Re sults and Discussion ................................ ................................ ................................ ..... 7 5 5.2.1 Synthesis of APAPBA/NIPAM C opolymers and C loud P oint D etermination .......... 7 5 5.2.2 Examining the E ffect of T emperature on H ydrogel C reep and S elf healing ............. 7 6 5.2.3 Temperature D ependent R heological S tudies ................................ ............................ 7 8 5.3 Conclusions ................................ ................................ ................................ ....................... 8 4 5.4 Experimental ................................ ................................ ................................ ..................... 8 6 5.4.1 Materials ................................ ................................ ................................ ..................... 8 6 5.4.2 Instrumentation and C haracterization ................................ ................................ ........ 8 6 5.4.3 Synthesis of 3 A crylamidopropanoic A cid ................................ ................................ 8 7 5.4.4 Synthesis of (2 (3 A cryla midopropanamido)phenyl)boro nic A cid P inacol E ster (APAPBAE) ................................ ................................ ................................ ........................ 8 7 5 .4.5 Synthesis of P oly(APAPBA co NIPAM ) (P(APAPBA co NIPAM )) ...................... 8 8 5 .4.6 Hydrogel S ynthesis ................................ ................................ ................................ .... 8 9 6 CONCLUSIONS ................................ ................................ ................................ .................... 90 APPENDIX : SUPPLEMENTARY FIGURES A ND SCHEMES ................................ ................ 93 A.1 Gel Permeation Chromatography Traces ................................ ................................ ......... 93 A.2 Nuclear Magnetic Resonance Spectra ................................ ................................ ............. 9 6

PAGE 7

7 A.3 Rheometry ................................ ................................ ................................ ........................ 9 9 A.4 Supplemental Figur e s and Schemes ................................ ................................ .............. 10 6 LIST OF REFERENCES ................................ ................................ ................................ ............. 10 8 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 1 13

PAGE 8

8 LIST OF TABLES Table page 4 1 Parameter tables for the EKT reactivi ty ratio studies ................................ ........................ 5 9 5 1 Time required (s) to reach 50% of initial stress fo r 5 mol% and 10 mol% APAPBA P (APAPBA co NIPAM ) /PVOH hydrogels ................................ ................................ ....... 83 5 2 Time required (s) to reach 10% of initial stress for variou s 5 mol% and 10 mol% APAPBA P (APAPBA co NIPAM ) /PVOH hydrogels ................................ ...................... 83

PAGE 9

9 LIST OF FIGURES Figure page 1 1 Oxidation states of organoboron compounds ................................ ................................ ... 18 1 2 Equilibrium of boronic acid/boronate anion with cis diols ................................ ............... 19 1 3 Disintegration of insulin PVA/P(AAm 3APBA) film ................................ ..................... 22 1 4 Hydrogel displacement under shear between two rheometer plates ................................ .. 23 1 5 Relationship between strain deformation and stress response in a viscoelastic material ................................ ................................ ................................ .............................. 23 1 6 Typical strain sweep of a dynamic covalent hydrogel ................................ ...................... 24 1 7 Typical frequency sweep of a dynamic covalent hydrogel ................................ ............... 25 1 8 Scheme of PVOH cross linking with borate ions and with DMA NAAPBA copolymer ................................ ................................ ................................ .......................... 2 7 1 9 Gelation of HPMA/SHA and HPMA/PBA copolymers at pH 4.2 and pH 7.6 ................. 2 8 1 10 UV Vis spectra of gelation and de gelation of PEG stars and BDBA .............................. 2 9 1 11 11 B NMR of varying borax:DTT ratios ................................ ................................ .............. 30 1 12 Structure of phenylboronic acid and diol macromonomers ................................ ............... 3 1 1 13 Intermolecular and intramolecular coordination in Jeff 2AMPBA and Jeff 4AMPBA .... 3 5 3 1 Solution (10 wt%) of P(2 APBA co DMA) (10 mol% 2APBA) at pH 7.0 before and after addition of a solution (10 wt%) of PVOH in deionized water. ................................ 4 4 3 2 P(2APBA co DMA)/P(DOPAA m co DMA) hydrogels created in DI water and 1 wt% aqueous sodium bisulfite ................................ ................................ .......................... 4 4 3 3 Healing of hydrogels formed by crosslinking P(2APBA co DMA) (10 mol% 2APBA) with PVOH ................................ ................................ ................................ ......... 4 5 3 4 Frequency sweep of the hydrogels formed from P(2APBA co DMA) (10 mol% and 15 mol% 2AP BA) with PVOH at pH 4.0 ................................ ................................ ......... 4 7 3 5 Strain sweeps showing self healing of 2APBA hydrogels after fracture ......................... 4 7 3 6 Oxidized P(2APBA co DMA ) ( 10 mol% 2APBA ) / P(DOPAA m co DMA ) ( 10 mol% DOPAAm ) hydrogel s ................................ ................................ ............................. 4 8

PAGE 10

10 3 7 Unoxidized P(2APBA co DMA ) ( 10 mol% 2APBA )/ P(DOPAA m co DMA ) ( 10 mol% DOPAAm ) hydrogel s ................................ ................................ ............................. 4 9 4 1 Extended Kelen T d s plot of copolymerization of 2APBAE and DMA ....................... 60 4 2 Extended Kelen T d s plot of copolymerization of APAPBAE and DMA .................... 60 4 3 Hydrogels composed of PVOH and copolymers of DMA with 3APBA 2APBA, and APAPBA ................................ ................................ ................................ ........................... 61 4 4 T wo separate halves of a cut P (A PAPBA co DMA) (10 mol% APAPBA)/PVOH hydrogel ................................ ................................ ................................ ............................ 62 4 5 values from the linear viscoelastic region of strain sweeps conduct ed on hydrogels formed with P (APAPBA co DMA) ( 10 mol% APAPBA ) and hydrogels formed with P (APAPBA co DMA) (20 mol% APAPBA) ................................ ................................ ... 6 3 4 6 Strain sweep of P (APAPBA co DMA) (10 mol% APAPBA) hydrogel in pH 7 0.01 M acetate buffer ................................ ................................ ................................ ................ 6 4 4 7 values from frequency sweeps conduct ed on hydrogels formed with P (APAPBA co DMA) ( 10 mol% APAPBA ) and hydrogels formed with P (APAPBA co DMA) (2 0 mol% AP APBA) ................................ ................................ ................................ ......... 6 6 4 8 Frequency sweep of hydrogel s formed with P (APAPBA co DMA) (10 mol% APAPBA) in pH 3 0.1 M acetate buffer (1% strain) and pH 6 0.1 M acetate buffer (1% strain) ................................ ................................ ................................ ......................... 6 6 4 9 A) Strain sweeps and frequency sweeps of P (APAPBA co DMA) (10 mol% APAPBA) hydrogels in 0.1 M pH 4 and pH 7 acetate buffer ................................ ........... 6 7 5 1 Hydrogel self healing stems from the equilibrium of dynamic covalent crosslinks dissociating and reforming ................................ ................................ ................................ 7 3 5 2 P (APAPBA co NIPAM)/PVOH hydrogel under room temperature and at the cloud point ................................ ................................ ................................ ................................ ... 7 4 5 3 DLS measurements of 5 mg/mL solutions of P (APAPBA co NIPAM) (5 mol% APAPBA) and P (APAPBA co NIPAM) (10 mol% APAPBA) ................................ ....... 7 6 5 4 P (APAPBA co NIPAM) (10 mol% APAPBA)/PVOH hydrogel creep behavior at room temperature and after being heated on a hot plate at 40 C for 24 h.. ..................... 7 6 5 5 P (APAPBA co NIPAM) (10 mol% AP APBA)/PVOH hydrogel half fused with a P (APAPBA co DMA) (10 mol% APAPBA)/PVOH hydrogel half immers ed in 50 C water for 24 h ................................ ................................ ................................ ............... 7 7

PAGE 11

11 5 6 P(APAPBA co NIPAM) (10 mol% APAPBA)/PVOH hydrogel se lf healing at room temperature and 40 C.. ................................ ................................ ................................ .... 7 8 5 7 P(APAPBA co NIPAM) (10 mol% APAPBA)/PVOH hydrogel strain jumps at 25 C ................................ ................................ ................................ ................................ ...... 7 9 5 8 5 mol% and 10 mol% APAPBA P (APAPBA co NIPAM) PVOH hydrogel frequency sweeps from 20 C to 40 C ................................ ................................ ............................. 81 A 1 GPC trace s of 10% P(2APBAE co DMA) and 15% P(2APBAE co DMA) ................... 93 A 2 GPC trace of 10% P(DOPAAm co DMA) and 15% P(DOPAAm co DMA) ................. 94 A 3 GPC traces of 5 mol % P(APAPBAE co DMA ) 10 mol% P(APAPBAE co DMA) and 20 mol% P(APAPBAE co DMA) ................................ ................................ ............. 9 5 A 4 GPC traces of 5 mol % P(APAPBAE co NIPAM ) and 10 mol% P(APAPBAE co NIPAM) ................................ ................................ ................................ ............................ 9 5 A 5 19 F NMR spectra of P(PFPA co DMA) and P(PFPA c o DMA) after substitution with dopamine, showing cleavage of the pentafluorophenolate group ............................. 9 6 A 6 Typical 1 H NMR plot of DMA/2APBAE reactivity ratio study ................................ ....... 9 7 A 7 Typical 1 H NMR plot of DMA/APAPBA E reactivity ratio study ................................ .... 9 8 A 8 Frequency sweeps of hydrogels formed from P(2APBA co DMA) (10 mol% and 15 mol% 2APBA) with PV OH at pH 7.0 and P(2APBA co DMA) (10 mol%) with P(DOPAAm co DMA) hydrogels at pH 7.0 and 4.0 ................................ ........................ 9 9 A 9 Strain sweeps of 10 mol% P(2APBA co DMA) with P(DOPAAm co DMA) hydrogel and10 mol% P(2APBA co DMA) with PV OH hydrogel in DI water ............ 100 A 10 Strain sweep and frequency sweep of 10 mol% P(2APBA co DMA) with P(DOPAAm co DMA) hydrogel in DI water oxidized for 24 h ................................ ..... 101 A 11 Strain sweeps of 10 wt %, 10 mol % P(APAPBA co DMA) hydrogel at pH 3, 4, 5, 6, and 7 ................................ ................................ ................................ ................................ 102 A 12 Strain sweeps of 10 wt %, 20 mol % P(APAPBA co DMA) hydrogel at pH 3, 4, 5, 6, and 7 ................................ ................................ ................................ ................................ 102 A 13 Strain sweep and modulus recovery of 10 wt% 10 mol% P(APAPBA co DMA) hydrogel in pH 4 0.1 M acetate buffer and pH 6 0.1 M acetate buffer ........................... 103 A 14 Frequency sweeps of 10 wt%, 10 mol % P( APAPBA co DMA) hydrogels at pH 3, 4, 5, 6, and 7 ................................ ................................ ................................ ........................ 104

PAGE 12

12 A 15 Frequency sweeps of 10 wt %, 20 mol % P( APAPBA co DMA) hydrogels at pH 3, 4, 5, 6, and 7 ................................ ................................ ................................ .................... 104 A 16 Frequency sweeps of P (APAPBA co DMA)/PVOH hydrogel from 25 C to 40 C ..... 10 5 A 17 P (APAPBA co NIPAM) hydrogel stress relaxation graphs ................................ ........... 10 5 A 18 X ray structure of 2 acrylamido phenylboronic acid pinacol ester (2APBAE) showing the chelation of the carbonyl oxygen, O1, to the boron center B1 ................................ 10 6 A 19 Two halves of a cut 10 mol% P(APAPBA co DMA)/PVOH hydrogel after 4 hours of contact. No cut interface is seen indicating successful healing ................................ .. 10 6 A 20 hydrogels ................................ ................................ ................................ ......................... 10 7

PAGE 13

13 LIST OF SCHEMES Scheme page 3 1 Equilibrium of 2 acrylamidophenylboronic acid (2APBA) in water ................................ 42 3 2 Proposed mechanism of hydrogel formation between P(2APBA co DMA) (15 mol% 2APBA) with PVO H and P(DOPAAm co DMA) (15 mol% DOPAAm) at neutral pH ................................ ................................ ................................ ................................ ....... 4 3 3 3 Oxidation of catechols when exposed to oxygen ................................ ............................... 4 9 4 1 Equilibrium of 2 acrylamidophenylboronic acid (2APBA) in water ................................ 5 6 4 2 Possible delocalization pathway of a propagating radical in the APAPBA monomer ...... 5 7 4 3 Equilibrium of (2 (3 acrylamidopropanamido )phenyl)boronic acid (APAPBA) in water ................................ ................................ ................................ ................................ ... 5 8 4 4 General crosslinking scheme between P(APAPBA co DMA) and PVOH at neutral pH ................................ ................................ ................................ ................................ ....... 61

PAGE 14

14 LIST OF ABBREVIATIONS 2APBA 2 A crylamidophenylboronic acid 2APBAE 3APBA AIBN APAPBA APAPBAE BDBA CDCl 3 DCM DIPEA DMA DMAC DMF DMSO d 6 DOPAAm DTT EDC EKT 2 A crylamidophenylboron ic acid pinacol ester 3 Acrylamidophenylboronic acid Azobisisobutyronitrile (2 (3 A crylamidopropanamido)phenyl)boronic acid (2 (3 A crylamidopropanamido)phenyl)boronic acid pinacol ester 1,3 benzenediboronic acid Deuterated chloroform Dichloromethane D iisopropylethylamine N,N Dimethylacrylamide N,N Dimethylacetamide N,N Dimethylformamide Deuterated dimethyl sulfoxide Dopamine acrylamide Dithiothreitol 1 Ethyl 3 (3 dimethylaminopropyl )carbodiimide Extended Kelen T d s FITC FTIR GPC HCl HPMA LVE Fluor e scein isothiocyanate Fourier transform infrared Gel permeation chromatography Hydrochloric acid 2 H ydroxypropylmethacrylamide Linear viscoelastic

PAGE 15

15 MBA NAAP BA NaOH Methylene bisacrylamide N Acryloyl m aminophenylboronic acid Sodium hydroxide NIPAM NMR PBA P EG PFPA N Isopropylacrylamide Nuclear magnetic resonance Phenylboronic acid P oly(ethylene glycol) Pentafluorophenyl acrylate PVOH SHA TEA THF Poly(vinyl alcohol) S alicylhydroxamic aci d Triethylamine Tetrahydrofuran

PAGE 16

16 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 SYNTHESIS AND IMPROVEMENT OF SELF HEALING BORONATE ESTER HYDROGELS By Christopher Chi long Deng Decembe r 2017 Chair: Brent S. Sumerlin Major: Chemistry Boronic acids have been a prevalent moiety in both organic chemistry and materials science where it has seen use as coupling agents, material crosslinks, and sensors. Much of the usefulness of boronic aci d comes fr om its ability to bind with cis diols The formation of these boronic and boronate esters is facile and reversible, creating a convenient dynamic covalent bond that can be used for a variety of purposes. Furthermore, the preponderance of diols i n biological systems enhances the applicability of boronic acids in the biomaterials field. In this work, intramolecularly coordinated boronic acid monomers were used to create a variety of hydrogels that could heal at neutral and acidic pH. The ability to heal at these pH ranges is useful as boronic acids typically only bind at alkaline pH ranges, limiting its efficacy for biological applications. The first set of hydrogels were shown to heal in both qualitative cut / heal tests and rheological measuremen ts. The second set of healing hydrogels were synthesized from an improved boronic acid monomer. This new monomer addressed issues in copolymerization behavior in the first monomer that resulted in cloudiness in the hydrogel. The lack of transparency could cause issues for using the hydrogel in optical applications. The copolymerization behavior of the new

PAGE 17

17 monomer was quantified with reactivity ratios and compared to the old monomer. The improved hydrogel remained clear and possessed improved mechanical pro perties. Finally, a method was developed to selectively control the healing and s tress relaxation behavior of a dynamic covalent hydrogel. In this method, a thermoresponsive copolymer was used to synthesize the hydrogel. Raising the t emperature near the cloud point of the copolymer caused solubility induced aggregation restricting polymer mobility and limiting crosslink access which in turn stopped healing and reduce d stress relaxation The control of these traits could be of use in a cell culture scaf fold and other applications However, there are some drawbacks including reduction of modulus while restricting polymer mobility.

PAGE 18

18 CHAPTER 1 SYNTHESIS AND APPLICATIONS OF BORONIC ACID HYDROGELS 1.1 Introduction Boronic acids are arguably most known for the ir use in organic coupling reactions. However, there is a significant amount of boronic acid research conducted i n the area of polymer chemistry and material s science The ability of boronic acids to bind readily to cis diols can be utilized as crosslinks or for selective binding to biological molecules containing saccharides. As such, boronic acids are used in a variety of applications including structural crosslink ing biosens ing and drug delivery 1 3 Boronic acid complexation with diols is particularly intriguing since it exhibit s dynamic covalent behavior The ability for a bond to be reversi ble while maintaining covalent strength is a unique quality that can be leveraged for multiple applications in materials science Boronic a cids are boron compounds where two hydroxyls are attached to the boron and the third bond consists of a boron carbon bond. Oxidation of boranes produces borinic acid which can be oxidized to produce boronic acids. (Fig 1 1 ) Figure 1 1. Oxidation states of organoboron compounds Eventually, boronic acids will degrade into boric acid, a relatively benign compound found in many consumer products. As a solid boronic acids may for m dimeric diboroxanes or trimeric boroxines es pecially in anhydrous conditions Boronic acids typically exist in a trigonal planar form with a sp 2 hybridized boron center However, t he boron in boronic acids has an empty p orbital which imparts Lewis acidic character leading to facile coordination of basic molecules to the boronic acid. In doing so, the

PAGE 19

19 trigonal planar boronic acid is converted to a tetrahedral boronate anion with sp 3 hybridization In aqueous solution, boronic acids equilibrate between the trigonal planar boronic acid and the tetrahe dral boronate anion (Figure 1 2 ). Figure 1 2. Equilibriu m of boronic acid/boronate anion with cis diols The equilibrium is dependent on the boronic acid p K a where the boron centers have equal populations of trigonal planar and tetrahedral species. S ta bilization of the boronate anion is the primary driver of b oronic acid p K a and p K a values can range from 4 to 10.5 4 The boronic acid p K a plays a vital role in the formation of boronic esters and boronate esters. Boronic acids are capable of binding with cis 1,2 and cis 1, 3 diols to form cyclic esters. At a pH value b elow the boronic acid p K a the formation of boronic esters is dis favored while above the boronic acid p K a the formation of boro nate esters is favored The favorability of tetrahedral boronate ester formation compared to the trigonal planar boronic ester is attributed to the release of angle strain as the boron hybridization changes from sp 2 to sp 3 5 An initial study reported that the boronate anion was more reactive toward ester formation than the boronic acid by five orders of magnitude 6 However, subsequent reports have come out reporting superior reactivity of the trigonal species, although it is more susceptible to hydrolysis 7,8 Reaction mec hanics aside, it is evident that the boronic acid p K a is inst rumental in ester formation. Boronic e sterification is a

PAGE 20

20 significant reaction as it forms a dynamic covalent bond or a covalent bond that exhibits equilibrium control 9 A dynamic covalent bond maintains covalent strength but can al so be undone un der appropriate conditions. The combination of strength and malleability found in boronic/boronate esters makes boronic acids a desirable moiety in the synthesis of stimuli responsive and dynamic materials. Additionally, boronate ester forma tion occurs readily in aqueous solutions. Combined with the preponderance of cis diols in biological systems, boronic acids are extensively used in biomedical research 1.2 Boronic A cid H ydrogels 1.2.1 Early E xamples of Boronic Acid H ydrogels The first r eported formation of hydrogels crosslinked with a boron compound occurred over sixty years ago with the simple mixture of poly(vinyl alcohol) ( PVOH ) and borax 10 Kataoka et al 11 synthesized the first hydrogel created from interpolyme r complexation where a PVOH solution was mixed with an alkaline solution of poly( N vinylpyrrolidone co 3 acrylamidophenylboronic acid). Complexation occu r red at various wt% evidenced by in creasing solution viscosity. M acroscopic gelation was observed at 1 wt% Decomplexation between the polymers was obtained through the introduction of glucose solution. A s the glucose competed with PVOH to bind with the boronic acid, interpolymer complexation was interrupted, and a decrease in solution viscosity occurred. T he ability of boronic acids to bind with various saccharides has been used in many instances to produce boronic acid hydrogels with the purpose of sensing glucose. The impact of producing a blood glucose sensor that is fast and non invasive would be immens e There are a variety of detection methods for boronic acid hydrogel based glucose sensors As early as 1996, Kikuchi et al. 12 produced an electrochemical sensor utilizing boronic acid diol complexation. In this work, a boronic acid ter tiary amine

PAGE 21

21 copolymer and PVOH gel membrane was cast on a platinum electrode. When exposed to glucose in solution, the hydrogel swelled as the glucose displaced the PVOH. This swelling resulted in increased ion transport to the electrode and in creased elec trical current. The current changes were able to be observed in a glucose range of 0 300 mg/dL, a range comparable to physiological blood glucose levels. The swelling change in boronic hydrogels upon exposure to glucose is utilized in other sensor types as well. For example, Lowe et al 13 produced a series of hologra phic sensors starting in 2004. The sensor consisted of a thin layer of boronic acid hydrogel embedded with silver halide nanoparticles. As glucose bound to the boronic aci d and increased hydrogel swelling, the spacing between the silver scatterers increase d. Th e larger spacing in turn resulted in a red shift in the hologram. Improvements to the sensor w ere achieved by the use of 2 acrylamidophenylboronic acid ( 2APBA ) 14 The 2APBA monomer could more effectively bind with glucose at neutral pH due to the intramolecular coordination between the amide carbonyl and boron. Additionally, the 2APBA sensor d emonstrated improved selectivity towards glucose over lactate. The glucose responsiveness of boronic acid hydrogels can also be leveraged for drug delivery. In this application, the introduction of glucose elicits some response in the hydrogel that facilit ates drug release. In one of the earliest examples of drug delivery from a boronic hydrogel, Kataoka et al 15 synthesized a n N isopropylacrylamide/3 acrylamidophenylboronic acid ( NIPAM /3APBA ) copolymer crosslinked with N N methylenebisacrylamide ( MBA ) In release studies with fluorescein isothiocyanate ( FITC ) labeled insulin, the 3APBA hydrogel demonstrated significant insulin releas e upon exposure to glucose. The release was attributed to the increased hydrophilicity of the boronate i ons due to glucose binding producing a swelling increase which released mor e of the encapsulated insulin. C ompetitive binding of glucose

PAGE 22

22 is another mechanism for drug release In this scheme the drug or a polymer drug conj ugate is bound to the boronic acid polymer or hydrogel. Introduction of glucose results in cleavage of the drug or conjugate, resulting in drug release In a work by Zhang et al 16 FITC labeled insulin was conjugated to PVOH via EDC coupling. A solution containing this PVOH conjugate was combined with a poly(AAm co 3APBA) solution to form hydrogel films in a layer by layer fashion (Figure 1 3) Figure 1 3. Disintegration of insulin PVA/P(AAm 3 APBA) film when immersed in aqueous so lution as a result of the breakage of phenylboronate ester bonds. Reproduced with permission from ref 16. Copyright 2012 Royal Society of Chemistry. was studied in glucose solutions of various concentrations where increased rele ase was seen with increased glucose concentration The proportional release with respect to glucose concentration is a crucial trait for self regulating insulin delivery vehicles. The release was also sustained for a n extended period due to the layer by la yer architecture. 1.2.2 A Brief Introduction to R heology A slight detour will be taken now into the field of rheology as i t is an essential tool for the characterization of hydrogels and the forthcoming chapters will contain a non trivial amount of rheo logical studies Rheology is the study of how materials deform and flow. Rheological measurements of hydrogels are typically conducted on an oscillatory shear rheometer. In this setup, a circular hydrogel sample is placed between

PAGE 23

23 two circular plates one f ixed and the other moving, and an oscillatory shear is applied. T he stress ( ) applied is given by =F/A where F is the force applied to the moving plate and A is the area of the sample geometry (Figure 1 3 ) Figure 1 4 H ydrogel displacement under shear between two rheometer plates The strain ( ) in the material is given by the horizontal displacement of the samp le x/y. The shear modulus ( G ) is derived from these values where G = / When sinusoidal strain is applied to an ideal elastic solid, the stress response is in phase with the applied strai n. This type of material is analogous to a spring where any energy applied is stored. Like a spring, an elastic material will return to its original form once the strain is released. When sinusoidal strain is applied to an ideal viscous liquid, the stress response will be 90 out of phase. This type of material is anal o gous to a dashpot where applied energy is dissipated Any strain applied to an ideal viscous liquid will be permanent and increasing strain will cause flow. A viscoelastic material will have characteristics of both elastic and viscous materials where the stress response will have some phase angle ( ) between 0 and 90 17 (Figure 1 5 ) Figure 1 5 Relationship between strain deformation and stress response in a viscoelastic material The oscillatory stress response can be divided into two waves where one is in phase and the other is out of phase: = sin( ) + ) 18 From this equation, the storage

PAGE 24

24 modulus ( G G = G 0 sin( ) + G 0 cos( ) G and G G G like and if G G rial can be considered more fluid like. In one of the simplest models for viscoelastic liquids, the deformation can be modeled with a spring and dashpot in series. If a force acts upon this model, the spring will move immediately whereas the dashpot will h ave some delay. Upon cessation of force, the spring will return to its original shape but the dashpot displacement will remain. Therefore, a viscoelastic fluid will have some elastic character but will exhibit some permanent deformation when force is impr essed upon it. 19 While dynamic covalent hydrogel rheological behavior may roughly follow this model, it is unlikely that this simple approximation will completely describe the behavior There are, generally, three essential tests for characterizing dynamic covalent hydrogels. The first is the strain sweep, where a hydrogel is subjected to linearly increasing strain over time at constant angular frequency and constant temperature (Figure 1 6) Figure 1 6. Strain sweep showing typical dynamic covalent hydrogel behavior The strain sweep is conducted for two primary reasons. It establishes the linear viscoelastic (LVE) region where G R heological measurements

PAGE 25

25 are typically conducted at a strain level within the LVE region as to avoid another variable affecting the modulus. Additionally, at a high enough strain level, G G crossover which is indicative of separation of the hydrogel crosslinks and the hydr ogel achieving a more fluid nature. After this crossover, if the strain is returned to within the LVE region, G G exhibit some degree of recovery if the h ydrogel is dynamic covalent. The restoration of moduli is indicative of a self healing hy drogel and the rate and extent of moduli restoration can give an idea of healing speed and the extent of the healing respectively The second typical test for dynamic covalent hydrogels is the frequency sweep, where a hydrogel is subjected to i ncreasing a ngular frequency at constant strain and constant temperature (Figure 1 7) Figure 1 7. Frequency sweep showing typical dynamic covalent hydrogel behavior At the beginning of the test, the angular frequency is typically very low. In a dynamic covalent hy drogel, if the rate of crosslink exchange is faster than the frequency of applied strain, G G applied strain. In this so called terminal region, G 2 and G will increase with a slope of 1. The angular frequency at which G G gives the terminal relaxation frequency ( c ). From this frequency, the relaxation time ( ) can be calculated where c The value for is indicative of the ti me for the longest stress

PAGE 26

26 relaxation mode. 20 It can be loosely interpreted as the crosslink lifetime for a hydrogel system although other factors are involved for stress relaxation to occur besides crosslink lability The relaxation time gives a general idea of how qui ckly the hydrogel may heal or creep. As the angular frequency increases beyond c the crosslinks are unable to respon d to the strain in time and G G more solid The last test is the stress relaxation test. Here, the hydrogel is subjected to a set amount of initial strain at constant temperature and the stress response required to maintain the strain over time is measured. Over time, the stress required to maintain the strain will decrease due to polymer motion an d crosslink exchange 21 The stress relaxation experiment also gives an idea of crosslink exchange rate and polymer mobility as both will affect stress relaxation rate. The half life for the relaxation stress is typically used in papers discussing hydrogel cell scaffolds as hydrogel stress relaxation plays a role in cell differentiation 22 1.2.3 Self healing Boronic Acid H ydrogels The ability o f b oronic acids to readily form reversible bonds with diols allows for the synthesis of self healing hydrogels. As mentioned before, the first boronic acid hydrogel was formed from the complexation of borax and PVOH. The advantages of using polymeric boronic acid for hydrogel formation was first investigated by Ivanov et al 23 A hydrogel formed from a n N,N dimethylacrylamide (DMA)/ N acryloyl m aminophenylboronic acid ( NAAPBA) copolymer and PVOH was compared to a hydrogel formed from borax and PVOH (Figure 1 8)

PAGE 27

27 Figure 1 8. Scheme of PVOH cross linking with borate ions (a) and with DM A NAAPBA copolymer (b). Reproduced with permission from ref 18. Copyright 2004 Elsevier Ltd. The polymeric hydrogel required several times less boron concentration to form hyd rogels than the borax hydrogel due to t he ability of the polymer to form multiple crosslink points per polymer chain with P VOH Furthermore, the entropic penalty should be less with polymeric binding as opposed to small molec ule binding. The polymeric hydrogel exhibited a longer relaxation time than the borax gel with a comparable amount of boron. The longer relaxation time was ascribed to the complex interactions between the polymers in the hydrogel, such as chain entanglemen t and multiple crosslinking points. The first report to emphasize self healing in boronic acid hydrogels was done by Kiser et al 24 in 2007, where a hydrogel was formed using crosslinks between phenylboronic acid (PBA) and salicylhydroxamic acid (SHA) moieties. A vinyl PBA monomer and SHA were each copolymerized with either acrylic acid or 2 hydroxypropylmethacrylamide (HPMA). When the HPMA copolymer solutions were mixed at pH 4.2, a free flowing hydrogel formed. When th e gel was produced at pH 7.6, it was much more rigid without any flowing behavior

PAGE 28

28 Figure 1 9. Colored solutions of HPMA/SHA copolymers (a) and HPMA/PBA copolymers (b). A flowing hydrogel is formed at pH 4.2 (c) while a rigid hydrogel is formed at pH 7. 6 (d). Reproduced with permission from ref 19. Copyright 2007 WILEY VCH Verlag GmbH & Co. KGaA However, when the hydrogel was produced from the acrylic acid copolymers at pH 7.6, it became free flowing. This behavior was attributed to the ionic nature o f the polymers at that pH causing chain repulsion or a Donnan effect altering the crosslink equilibrium. The hydrogel behavior was also quantified via oscillatory shear rheometry. F requency sweeps of the HPMA cop olymer hydrogel revealed longer relaxation t imes at pH 7.6 than pH 4.2. The relaxation time difference correlated with the difference in flowing behavior seen at these pH values. Likewise, the acrylic acid copolymer hydrogel exhibited a comparable relaxation time at pH 7.6 to the relaxation time of the HPMA copolymer hydrogel at pH 4.2. The self healing effect was quantified by first fracturing the gel crosslinks with a high amount of stress. The stress was then lowered to within the LVE region to allow for modulus recovery. The HPMA copolymer hydrog el at pH 4.2 and the acrylic acid copolymer hydrogel at pH 7.6 exhibited near complete or complete recovery of their original storage modulus ( G Another example of a self healing boronic acid hydrogel was synthesized by Messersmith et al. 25 The hydrogel was formed from the coupling of a bifunctional crosslinker, 1,3 benzenediboronic acid (BDBA), a nd a 4 arm poly(ethylene glycol) (PEG) star functionalized with catechol on each arm. Mixing the component solutions resulted in hydrogel formation with in 30 min at pH 9 Lowering the

PAGE 29

29 pH to 3 caused the crosslinker to primarily exist in the less reactive b oronic acid form, resulting in a gel to sol transition. The gelation and de gelation processes w ere confirmed with UV Vis spectroscopy where formation of the boronate ester bond corresponded with a distinct peak at 490 nm. This peak disappeared upon acidif ication and de gelation of the hydrogel (Figure 1 10) Figure 1 10. UV Vis of gelation and de gelation. (A) UV vis spect ra of PEG crosslinker/BDBA at (a) pH 9, (b) pH 3 and (c) PEG star by itself at pH of 7.4. (B) Hydrogel formed from 7.5 wt% PEG star and BDBA at pH 9 and molar ratio of 1 : 8, and (C) sol ution produced after the hydrogel shown in B was acidified to pH 3 Reproduced with permission from ref 20. Copyright 2011 Royal Society of Chemistry. The hydrogel demonstrated self healing abilities in a cut/heal test. A hydrogel sample wa s cut in half and was then able to heal itself into a whole piece when the two halves were placed into contact with each other. Th e healing process only took 30 s after contact Self healing of the hydrogel was also characterized with strain sweeps. As str ain was ramped to 1000%, the crosslinks began to separate as seen with the crossover of G and G was lowered within the LVE region. Self healing boronic acid hydrogels have also been synthesized from multi arm polymers. In another work by Theato et al 26 a self healing hydrogel was created in a one pot reaction using borax, dithiothreitol (DTT), and 700 g/mol poly(ethylene glycol) diacrylate (PEGDA). Borax was added to aqueous solutions containing stoichiometric ratios of DTT and PEGDA. Borax was not only used to crosslink the diols in DTT but also served as a catalyst for Michael addition between the

PAGE 30

30 DTT thiol and PEGDA. 27 Gelation was fairly rapid, ranging from 40 s to 2 min depending on borax concentration. Completion of the thiol reaction was quantified via Fourier transform infrared spectroscopy (FTIR) where near complete disappearance of the thiol and acrylate peak s was observed Polymer formation was also seen in GPC analysis of the uncrosslinked hydrogel with a M n of 9100 g/mol. Furthermore, complexation between the borax and DTT was shown in 11 B NMR where borax complexing to both single and double DTT molecules o ccurred at a borax:DTT ratio of 1:2, whereas only double complexation occurred when the borax:DTT ratio was increased to 1:4 (Figure 1 11) Figure 1 11. 11 B NMR of borax/DTT = 1:2 at pH = 9 ( blue line) and pH = 3 ( green line) as well as borax/DTT = 1:4 at pH = 9 ( red line) Reproduced with permission from ref 22. Copyright 2015 American Chemical Society. Frequency sweeps were conducted for hydrogels with varying molar ratios of borax to DTT. As the molar ratio increased, the plateau modulus increased an d c decreased. The observed trend was attributed to the increased molar ratio of borax forming a greater number of crosslinks, thus strengthening the hydrogel and reducing the relaxation abilities of the polymer chains. These hydrogels were also shown to be self healing in a cut/heal test. Cut pieces of hydrogels were able to heal within 30 min, showing no signs of prior damage and being able to support their weight. Furthermore, the hydrogel was both pH and temperature responsive. At acidic pH, the crossli nks dissociated resulting in

PAGE 31

31 a gel to sol transition. Raising the pH back to 9 resulted in reformation of the hydrogel. As boronate ester formation was exothermic, increasing the temperature to 80 C resulted in bond dissociation and dissolution of the hyd rogel. Cooling the temperature down to 18 C allow the hydrogel to form again. These pH and temperature dependent cyclizations were reversible in three consecutive trials. In another work, by Anderson et al 28 tetra armed PEG stars were functionalized with various boronic acid moieties which were then complexed with diol terminate d tetra armed PEG stars. One of the advantages of crosslinking star polymers at their terminii is that a more homogen e ous crosslinked network is produced When random copolymers are crosslinked the distribution of crosslinks will be heterogeneous T hus th ere will be areas where the crosslink density is high er than average These areas will experience more significant stresses and fail earlier, ultimately leading to a weaker hydrogel. Additionally, using terminally bound stars largely eliminates loops and d angling chain ends that would also reduce hydrogel strength. In this work, 5k amino terminated PEG stars were functionalized with three separate boronic acid small molecules: 4 carboxy 3 fluorophenylboronic acid, 4 carboxyphenylboronic acid, and 2 formylph enylboronic acid, to form the conjugates PEG FPBA, PEG PBA, and PEG APBA respectively (Figure 1 12) Figure 1 12. Schematic depicting (a) hydrogel formation between four arm phenylboronic acid containing PEG and diol containing PEG and b ) c hemical structures of PEG phenylboronic acid macromonomers and PEG diol, and the representative mechanism of phenylboronic acid/ cis diol binding. Reproduced with permission from ref 23. Copyright 2015 WILEY VCH Verlag GmbH & Co. KGaA

PAGE 32

32 The indivi dual boronic acid molecules have p K a values of 7.8, 7.2, and 6.5 6.7 for 4 carboxyphenylboronic acid, 4 carboxy 3 fluorophenylboronic acid, and 2 formylphenylboronic acid respectively. T he amino PEG star was functionalized with D gluc on olactone to create the diol crosslinker A total of nine sets of hydrogels were created as each boronic acid star was coupled with the diol star at pH 6, 7, and 8. Frequency sweeps conducted on these hydrogels revealed viscoelastic behavior that was dependent up on the boron ic acid p K a values. At pH 6, PEG PBA did not exhibit any hydrogel formation, whereas PEG FPBA and PEG APBA did. As pH was increased to 7 and then 8, the plateau modulus increased for all hydrogels as more reactive boronate species were created, which resul ted in more crosslinks The higher crosslink density also resulted in a decrease in c values as relaxation ability was reduced These hydrogels were shown to be fully self healing in a cut/heal test where two cut pieces instantly healed when placed into c ontact. Additionally, the hydrogel demonstrated G G crossover when subjected to a strain of 500% indicating network dissolution. Once strain was lowered to 0.05%, G G values rapidly return ed to their original values, indicating self healing. The hy drogels were also shear thinning as the PEG FPBA hydrogel was able to be extruded through a 21 G syringe needle. The shear thinning ability was critical as the hydrogel was tested for a variety of biomedical applications. In particular, the PEG FPBA hydrog el was shown to have excellent release properties of various proteins, insulin, and immunoglobin when exposed to glucose solutions. Additionally, the hydrogel exhibited excellent cell viability via live/dead assays and a hydrogel injected subcutaneously i nto mice resulted in full clearance of the hydrogel at 21 days with no signs of inflammation or tissue damage.

PAGE 33

33 1.2.4 Intramolecularly coordinated Boronic A cids and Hydrogels Derived T hereof A few of t he previous work s highlight one of the main weaknesses of typical boronate ester crosslinked hydrogels. An alkaline pH is typically required for crosslinking with diols as this ensures the majority of the boron is in the more reactive boronate a nion form. Unable to gel at physiological pH the typical boronic acid hydrogel would be precluded from any biological applications. The most straightforward approach to this issue would be to simply lower the p K a of the boronic acid and thus lower the pH required to achieve binding. While previous molecular studies hav e shown that optimal binding pH is dependent on multiple factors including boronic acid p K a diol p K a buffer type, solvent, and steric hindrance 29 in hydrogel applications, a lower p K a is generally enough to ensure crosslinking at lower pH. Aryl boronic acids are more acidic than alkyl boronic acids and the aromatic group also provides a n excellent substrate for modif ication. Phenylboronic acid has a p K a of around 8.9 compared to methyl boronic acid which has a p K a of 10.4 4 The addition of electron withdrawing groups (EWGs) on the phenyl ring in proper po sitions further decreases the p K a as it stabilizes the negative charge of the boronate anion 30 The presence of EWGs on the phe nylboronic acid can lower the p K a to around 7. However, this approach adds significant synthetic complexity. An alternative approach to achieve binding at neutral and acidic pH would be to use intramolecularly coordinating boronic acids. In these mole cules a hetero atom forms a coordinate bond with the boron center The coordinat i on results in a tetrahedral boronate anion which more readily forms stable crosslinks with diols. Gnter Wulff synthesized t he first of these intramolecularly coordinated boronic a cids 31 In his monomer a dimethylaminomethyl group was inserted at the ortho position relative to the boronic acid. At this proximal position, the amine was able to form a coordinat e bond with the

PAGE 34

34 boron center resulting in a p K a of 5.2, the lowest value at the time. Wulff type boronic acids have been used in several instances in chemical sensors and small molecule capture devices 32 34 As expected, Wulff type boronic acids have also been employed in self healing hydrogels. Engbersen et al were able to create Wulff type boronic acid polymers through the functionalization of Jeffamine polymers 35 These polyethers were terminated by amino groups which were used to form imines with ortho formylphenyl boronic acid. Sodium borohydride was then used for reductive amination to form a coordinated boronic acid polymer (Jeff 2AMPBA) The same methodology was also applied to para formylphenylboronic acid to observe the behavior of a non coordinate d boronic ac id polymer (Jeff 4AMPBA) The p K a of Jeff 2AMPBA was determined via titration. The first p K a value of 6.4 corresponded to the protonated secondary amine. This value was appreciably lower than the 8.2 recorded for Jeff 4AMPBA. Deprotonation for the amine in Jeff 2AMPBA would be favored due to the amine forming a dative bond with the boronic acid. The coordinate bond also affected the second p K a of both polymers. In Jeff 4AMPBA, the second p K a was 9.5 whereas for Jeff 2AMPBA, it was high er than 11. The eleva tion in p K a was a direct result of the coordinate B N bond making it more difficult for a hydroxyl to form a boronate anion. Hydrogels were formed by combining the Jeffamine copolymers with PVOH from pH 3 to pH 9. Both Jeff 2AMPBA and Jeff 4AMPBA formed h ydrogels across the entire pH range. Frequency sweeps were conducted to determine the plateau modulus and relaxation times for both hydrogels. For Jeff 2AMPBA, both G increased as pH decreased These trends were attributed to the reduction in boron coordination from the amine resulting in more facile coupling with PVOH. Jeff 4AMPBA experienced the opposite trend where G increased as pH increased. Interestingly, both systems were able to form hydrogels at

PAGE 35

35 neutral and acidic pH. Th e broad range of gelation was attributed to intramolecular coordination in the case of Jeff 2AMPBA and intermolecular coordination for Jeff 4AMPBA (Figure 1 13) Figure 1 13. Intramolecular and intermolecular coordination in Jeff 2AMPBA and Jeff 4AMPBA respectively. R eproduced from ref 30. Copyright 2011 Royal Society of Chemistry. The Wulff type boronic acid is not the only type of intramolecular coordination possible for boronic acids. Liu et al synthesized a phenylboronic acid with a diisopropylamino carbonyl in t he ortho position 36 The phenylboronic acid was then bound to a catechol to form the boronate ester. X ray crystallography revealed c oordination of the car bonyl oxygen to the boron The B O dative bond was slightly longer than the B O ester bonds, which is expected for a coordinate bond compared to a covalent bond. Another study by Smith et al synthesized various ortho ureaphenylboronate esters to improve b inding with carboxylates 37 X ray crystallography also show ed carbonyl oxygen coordination to the boron in this system Additionally, 11 B NMR showed clear upfield shifts from non coordinated boronate esters, indicative of tetrahedral structure. Both the urea coordinated boronate esters and the non coordinated boro nate esters were subjected to titrations with tetrabutylammonium acetate. The coordinated boronate esters showed an order of magnitude larger association constants with the acetate compared to the non coordinated boronate esters. While this does not direct ly show the improved binding ability of a coordinate boronic acid with a diol, it does provide additional evidence for the formation

PAGE 36

36 of a tetrahedral boron via intramolecular coordination. Despite the demonstrated coordination between carbonyls and boronic acids, there has been limited use of this technique in material applications.

PAGE 37

37 CHAPTER 2 RESEARCH OBJECTIVE The purpose of this research was to utilize carbonyl intramolecularly coordinated boronic acid monomers to create hydrogels that can sel f heal at neutral and acidic pH. While there have been various strategies to produce boronic acid hydrogels that self heal at neutral and acidic pH, there were no previous works that utilized intramolecular coordination with carbonyls. The use of carbonyls versus Wulff type boronic acids offer s several advantages. The first being that the effects of the amine p K a are dampened or eliminated. As seen in the Jeffamine boronic acid hydrogel, the amine p K a significan tly affected the strength of the hydrogel dep ending on the pH. Additionally, the use of carbonyl coordinating boronic acids allow s for monomers with a high rate of propagation, k p such as acrylamides and acrylates. The use of these monomers would be advantageous in copolymer s ynthesis due to general ease of polymerization along with compatibility with common hydrophilic monomers. Finally, the presence of cationic amines in a hydrogel could adversely affect cell viability 1 limiting potential use in biological applications. The research goal in chapt er 3 was to synthesize the first self healing hydrogel using carbonyl intramolecularly coordinated boronic acid monomers. In order to achieve this goal 2 acrylamidophenylboronic acid pinacol ester (2APBA E ) was copolymerized with N,N dimethylacrylamide (DM A) via conventional radical polymerization After the pinacol protecting group was removed the resulting poly(DMA co 2APBA) was dissolved in 10 wt% solution at pH 7 and pH 4 M ixing the polymer solution with PVOH solutions of the same pH produced self hea ling hydrogels. A catechol functionalized copolymer was also used to form a hydrogel, though issues with oxidation made it less effective for self healing.

PAGE 38

38 In chapter 4, the research focused on the synthesis and use of a new boronic acid monomer, (2 (3 ac rylamidopropanamido )phenyl)boronic acid (APAPBA). Th e APAPBA monomer was synthesized to address issues with the copolymerization behavior of 2APBA with other acrylamides. Specifically, 2APBA tended to prefer homopropagation versus cross propagation. The po lymerization behavior would result in the 2APBA copolymer containing blocky se gments of boronic acid monomer. The close proximity of bound PVOH chains would then lead to aggregation The aggregation manifested itself as cloudiness in the hydrogel after a p eriod of 24 h. Reactivity ratios were determin ed for both 2APBA and APAPBA. 2APBA preferentially homopropagated whereas APAPBA preferred cross propagation. Hydrogels formed from poly( APAPBA co DM A) and PVOH remained transparent while possessing faster heal ing times than the 2APBA hydrogel. The APAPBA hydrogel mechanical properties were also superior to those of the 2APBA hydrogel. All of these enhancements were attributed to improved co polymerization behavior stemming from the new monomer structure. Cha pter 5 focused on creating a self healing boronic acid hydro gel where the self healing and stress relaxation behavior could be controlled via thermal stimulus. Control was achieved by copolymer izing APAPBA with NIPAM. As before, solutions of this copolymer were mixed with PVOH solution to produce the hydrogel. Due to the temperature dependent solubility of NIPAM, the APAPBA copolymer would become increasingly hydrophobic near its cloud point. The hydrophobic APAPBA copolymer would aggregate, reducing polyme r mobility and limiting crosslink access Additionally, some water would be extruded from the hydrogel, increasing weight percent and further restricting polymer mobi lity. The constriction of movement prevented the typical self healing and creep behaviors seen in boronic acid hydrogels. Frequency sweeps revealed

PAGE 39

39 increas ed relaxation times for the hydrogel as temperature approached the copolymer cloud point. Also, stress relaxation experiments showed a decrease in long term stress relaxation at higher temper atures, correlating with the lack of creep and self healing behavior However, there was relatively little difference in short term stress relaxation at all temperatures, as it was still possible for some polymer motion even at the cloud point.

PAGE 40

40 CHAPTER 3 BORONATE ESTER HYDROGELS UNDERGO SELF HEALING AT NEUTRAL AND ACIDIC PH 3.1 Introduction Hydrogels consist of crosslinked networks of hydrophilic polymers swollen with an aqueous solution. 38 Hydrogels have been examined for a number of rol es, with much of the current literature focused on applications in the field of biomedical engineering, including use in drug delivery, tissue engineering, and wound dressings. 39 43 Hydrogels are well suited for biomedical applica tions due to their hydrophilic nature, porous structure, and adjustable stiffness. 44 Some hydrogels have been prepared with functional groups that permit reversible interactions or reversible crosslinking to allow for self healing character. 45 50 The ability of a hydrogel to repair damage autonomously is tremendously useful in a variety of areas, including in many of the applications mentioned above. 51,52 Reversibility in hydrogels can be accomplished by either relying on physical crosslinks or employing reversible covalent bonds. 53 The advantage of the latter approach is that the gels can maintain both the robust integrity of c ovalently crosslinked materials and the intrinsic reversibility of physically crosslinked systems. One example of a r eversible covalent bond that is used for hydrogel crosslinking is the complexation between boronic acids and 1,2 or 1,3 diols. 54 57 The boronate este r bond has been employed to create self healing networks, molecular sensors, and cancer cell sweepers. 58 The strength and reversibility of boronate ester crosslinks in aqueous media are governed by an equilibrium that is heavily dependent on solution pH and the p K a of the boronic acid. At pH values above the p K a of the boronic acid, boronate ester bond formation is favored while below the p K a the equilibrium favors the free boronic acid and diol. 59 61

PAGE 41

41 As this system is in a dynamic equilibrium between diol/boronic acids and boronate esters, bond rearrangement can occur, which allows for self healing i f the newly formed bonds span the damage interface between two boronate ester crosslinked materials. The p K a of most aryl boronic acids lies in the range of 8 9. As mentioned above, the p K a is essential for self healing applications of boronic acid based materials because the exchange between boronic acid/diol and boronate esters is most effective when the pH is above the p K a 29 Employing boronic acids that can heal at neutral and acidic pH would be advantageous, especially for biomedical applications. 62 One method to accomplish this is by using intramolecular coordination that can stabilize boronate ester formation at reduced pH. 4,32,35,63 In this work we employ a boronic acid containing acrylamide monomer to prepare water soluble polymers that can be crosslinked via boronate ester formation with diol containing polymers. The monomer, 2 acrylamidophenylboronic acid (2APBA), allows for intramolecular coor dination between the carbonyl oxygen of its a crylamide moiety and the boron of t he boronic acid group (Figure A 18 Scheme 3 1). 64 Compared to B N coordination present in Wulff type boronic acids, B O bonds are stronger and should therefore, be more capable of stabilizing the formed ester. 37,65 Given that ester formation between a boronic acid and diol is known to be favored by the boron existing in the tetrahedral form, polymers of 2APBA should be capable of forming st able boronate ester hydrogels when mixed with diol containing polymers. More importantly, we reasoned that intramolecular B O coordination in 2APBA would facilitate boronic acid diol complexation, even at neutral and acidic pH, while maintaining the dynamic exchange inherent to boronate esters in aqueous media. Therefore, hydrogels crosslinked via diol complexation with 2APBA units should demonstrate self healing behavior at pH ranges that were previous ly

PAGE 42

42 considered too low for boronic acid containing materials. Herein, we report the preparation and preliminary self healing properties of boronate ester hydrogels based on 2APBA in neutral and acidic pH ranges that would typically be considered too low for most boronic acid based polymers. Scheme 3 1. Equilibrium of 2 acrylamidophenylboronic acid (2APBA) in water. Intramolecular B O bond formation leads to the tetrahedral geometry of the boron atom that promotes boronate ester formation. 3.2 Results and Discussion 3.2.1 Synthesis of 2APBA C opolymers and H ydrogel F ormation We set out to prepare separate boronic acid containing copolymers and diol containing (co)polymers that would form hydrogels on mixing. N,N Dimethylacrylamide (DMA) was chosen as the c omonomer for polymerization with the pinacol protected ester of 2APBA (i.e., 2APBAE) because of its hydrophilicity and biocompatibility. 66 GPC analysis of the P(2APBAE co DMA) copolymers with 10 and 15 mol% 2APBA yielded M n values of approximately 3.4 10 4 g/mol ( = 3.5) and 3 8 10 4 g/mol ( = 4.2), respectively (Figure A 1 ). Two different diol containing copolymers were considered (Scheme 3 2). A dopamine acrylamide (DOPAAm) copolymer, P(DOPAAm co DMA), was prepared, as catechol g roups are known to bind strongly to boronic acids in aqueous media. 6 67 However, given that catechols are prone to oxidation at neutral pH when exposed to air, 68 which could potentially lead to loss of self healing ability over time, 69 poly(vinyl alcohol) (PVOH) was also used as an alternative polymeric diol, due to its high density of 1,3 diol groups along its backbone.

PAGE 43

43 Scheme 3 2. Proposed mechanism of hydrogel formation between P(2APBA co DMA) (15 mol% 2APBA) with PVOH and P(DOPAAm co DMA) (15 mol% DOPAAm) at pH 7 The dopamine polymers were synthesized via nucleophilic substitution of pentafluoro phenyl acrylate (PFPA) DMA copolymers with dopamine. 70 Activated ester chemistry allowed for facile synthesis of the desired polymer, with the functionalization being readily monitored by 19 F NMR spectroscopy via the cleavage of the penta fluorophenolate group (Figure A 5 ). 71 GPC analysis of the P(DOPAAm co DMA) copolymers with 10 and 15 mol% DOPAAm gave M n values of approximately 9.0 10 4 g/mol ( = 2.9) a nd 1.1 10 5 g/mol ( = 3.0), respectively (Figure A 2 ), though the insolubility of the polymer with 15 mol% DOPA prevented its subsequent use in hydrogels. The hydrogels of the P(2APBA co DMA) copolymers with P(DOPAAm co DMA) or PVOH were created in deion ized water and an acidic buffer at pH 4.0. Hydrogel formation, qualitatively confirmed through a vial inversion test, occurred rapidly after mixing the two boronic acid and diol containing polymer solutio ns and gently stirring (Figure 3 1 ).

PAGE 44

44 Figure 3 1. Solution (10 wt%) of P(2 APBA co DMA) (10 mol% 2APBA) at pH 7.0 A) before and B) after addition of a solution (10 wt%) of PVOH in deionized water. Reproduced with permission from ref 79. Copyright 2015 American Chemical Society. The rapid onset of gela tion is consistent with boronate ester formation under both acidic and neutral conditions. Interestingly, hydrogels created from P(DOPAAm co DMA) and P(2APBA co DMA) were initially colorless but darkened over time due to catechol oxidation. This process wa s noticeable within 24 h in deionized water and over several days in acidic buffer. However, when these hydrogels were created in a reducing solution of sodium bisulfite the hydrogel did not darken over time (Figure 3 2 ). Figure 3 2. P(2APBA co DMA)/P( DOPAA m co DMA) hydrogels created in DI water (left vial) and 1 wt% aqueous sodium bisulfite (right vial). Photos were taken at A) 0 h B) 24 h C) 108 h. The concentration of the original polymer solutions was 10 wt%. Reproduced with permission from ref 79. Copyright 2015 American Chemical Society.

PAGE 45

45 3.2.2 Investigation of H ydrogel S elf healing Given that all crosslinks within the hydrogel were expected to be dynamic covalent as a result of their rapid exchange due to boronate ester formation/hydrolysis, we dec ided to investigate the self healing characteristics of the gels. As a qualitative measurement of self healing ability, cut/heal tests were conducted on the hydrogels. Disks of each gel were created and subsequently sliced into two equal pieces. These halv es were then placed in contact to induce healing. In all cases, the gels appeared to heal within 60 min, as evidenced by the gradual disappearance of the scar at the damage site and by the gels regaining their mechanical integrity (Figure 3 3 ). This behavi or suggests that boronate ester formation across the damage interface allowed the gel to heal In aqueous media, boronate esters are under equilibrium with their boronic acid and diol constituents, which leads to a finite concentration of free boronic acid and diol groups available for covalent healing by esterification. The healed gels were strong enough to support their weight when suspended. Figure 3 3 Healing of hydrogels formed by crosslinking P(2APBA co DMA) (10 mol% 2APBA) with PVOH. A) Intact ge l B) Cut gel C) Gel halves placed in contact immediately after cutting D) Healed gel after 60 min E) Healed gel suspended under its weight. The concentration of both polymer solutions was 10 wt% prior to mixing. Reproduced with permission from ref 79. Copy right 2015 American Chemical Society. 3.2.3 Rheological Studies Rheometry was also employed to gain insight into the properties of the hydrogels and to quantify their self healing nature. The stiffness of the gels demonstrated a slight dependence on copo lymer composition for P(2APBA co DMA) copolymer, with

PAGE 46

46 increasing content of 2APBA leading to slightly higher storage moduli presumably due to the higher degree of crosslinking. Most importantly, the rheological behavior of the gels was consistent with tha t expected for systems involving dynamic (i.e., reversible) crosslinks. While permanently crosslinked gels demonstrate frequency independent moduli with storage modulus ( G G expected to demonstrate fre quency dependent moduli with G G For these boronate ester hydrogels, G G er 0.1 rad/s at pH 4.0 (Figure 3 4) and 7.0 (Figure A 8 (A)). These results indicate the time scale probed in these e xperiments is shorter than the lifetime of the reversible crosslinks, which gives rise to predominantly elastic behavior because the network is not allowed time to restructure when perturbed. Therefore, it can be deduced that the exchange rate of the boron ate esters is rapid for these materials, suggesting self healing behavior is possible at both neutral and acidic pH. Indeed, the gels crosslinked with the DOPAAm copolymers and the PVOH homopolymers exhibited self healing behavior in strain sweep e xperimen ts at acidic (Figure 3 5) and neutral pH (Figure A 9 ). The hydrogels were subjected to increasing strain ( 10 rad /s) until the crosslinks failed, as indicated by the crossover of G G this critical point, the hydrogel begins to exhibit more fluid c haracter than elastic character. After the strain was reduced to within the linear viscoelastic region, the crosslinks recoupled rapidly, restoring the original hydrogel moduli values. The PVOH hydrogels had an overall higher equilibrium modulus (~1000 Pa) than the DOPAAm hydrogels (~350 Pa) The modulus difference can likely be attributed to the increased crosslink density expected in the presence of the PVOH hydrogels, as the concentration of diols that can participate in crosslinking in this homopolymer is considerably higher

PAGE 47

47 than for the P(DOPAAm co DMA) copolymer. Interestingly, there is little difference in the strength of each hydrogel between neutral and acidic pH, which is directly attributable to the negligible difference in coordination strength a t these pH values as a result of the intramolecular complexation within 2APBA. Figure 3 4 Frequency sweep of the hydrogels formed from P(2APBA co DMA) (10 mol% and 15 mol% 2APBA) with PVOH at pH 4.0. Original polymer solutions were 10 wt%. Strain was c onstant at 5%. Figure 3 5 Strain sweeps showing s elf healing of 2APBA hydrogels after fracture. The top two plots demonstrate the change in modulus during the strain ramp described by the bottom plots. The hydro gel s fractured during the strain ramp up to 400% and rapidly healed after the strain was reduced A) hydrogels formed from P(2APBA co DMA) (10 mol% 2APBA) with P(DOPAAm co DMA) (10 mol% DOPAAm) at pH 4.0 and B) hydrogels formed from P(2APBA co DMA) (10 mol% 2APBA) with PVOH at pH 4.0. Original po lymer solutions were 10 wt%. Angular frequency was constant at 10 rad/s. 3.2.4 Examining the Effect of Catechol O xidation on S elf healing We were also interested in assessing the potential effect of catechol oxidation on the rheological and self healing p roperties of the gels containing the random

PAGE 48

48 P(DOPAAm co DMA) copolymers. Rheometry was conducted on a gel formed from P(2APBA co DMA) ( 10 mol% 2APBA) and P(DOPAAm co DMA) ( 10% DOPAAm) that was left exposed to air for 24 h. Despite the apparent oxidation ob served by the gradual change in color the gel retained the same moduli values and self healing abilities as a control gel that was kept in the absence of oxygen (Figure A 10 ). However, there was a noticeable increase in gel stiffness upon oxidation, whic h resulted in increased difficulty of setting the gap used in previous rheological testing. A visual creep test was conducted on another P(DOPAAm co DMA) hydrogel that was left to oxidize for a week in a closed container. Although the unoxidized boronate ester hydrogels demonstrated noticeable creep over time, a behavior attributable to the equilibrium between dynamic covalent crosslinks, the oxidized gel did not creep and reta i ned a definite shape (Figure 3 6 ). The same test was conducted on a P(DOPAAm c o DMA) hydrogel created and aged in a sodium bisulfite solution. This unoxidized hydrogel exhibited noticeable creep compared to the oxidized sample (Figure 3 7 ). Figure 3 6 Oxidized P(2APBA co DMA ) ( 10 mol% 2APBA )/ P(DOPAA m co DMA ) ( 10 mol% DOPAAm ) hy drogel at A) 0 h and B) 168 h. The hydrogel appears to retain its original dimensions after 168 h. Reproduced with permission from ref 79. Copyright 2015 American Chemical Society.

PAGE 49

49 Figure 3 7 Unoxidized P(2APBA co DMA ) ( 10 mol% 2APBA )/ P(DOPAA m co DMA ) ( 10 mol% DOPAAm ) hydrogel at A) 0 h and B) 168 h. Definite creep can be seen between the top and bottom photographs. Reproduced with permission from ref 79. Copyright 2015 American Chemical Society. The oxidized hydrogel also exhibited increased stiffnes s during rheometry. Unable to creep, the oxidized hydrogel transmitted more force on the rheometer at the same gap setting, exceeding the maximum allowed force of the instrument. Finally, the oxidized hydrogel was unable to self heal, even after 48 h, when subjected to the same cut/heal test used previously. These findings are attributed to oxidation of the catechols into their quinone form ( Scheme 3 3 ) simultaneously removing boronate ester crosslinking potential in the hydrogel and causing permanent cros slinks between quinone moieties. 72 Scheme 3 3 Oxidation of catechols when exposed to oxygen. A catechol undergoes an initial hydrogen abstraction and reaction with oxygen to produce the semiquinone radical. The semiquinone then reacts with oxygen aga in to produce the ortho quinone. 68 As these covalent crosslinks cannot rearrange like dynamic crosslinks, 21 they reduce the creep and stress relaxation behaviors exhibi ted in the boronate ester crosslinked hydrogel. 3.3 Conclusions In summary, we were able to create a boronic acid hydrogel that self heals at acidic and neutral pH by exploiting a boronic acid containing monomer capable of

PAGE 50

50 intramolecular coordination. Th is intramolecular coordination between the carbonyl oxygen and boron resulted in stabilized boronate ester crosslinks, which in turn allowed for hydrogels to be formed at neutral and acidic pH. Rheometry demonstrated that there was little dependence of pH on the strength of the hydrogels and that PVOH crosslinked materials exhibited greater strength than those crosslinked by P(DOPAAm co DMA) due to the higher crosslink density in the former. Additionally, we demonstrated that oxidation of the dopamine moie ties in P(DOPAAm co DMA) m ight lead to gradual changes in hydrogel properties including a reduction in self healing properties and minimalized creep. The broader pH range at which healing can occur increases the possible applications of boronate ester hyd rogels as biological materials, with the additional stability at lower pH values potentially being of benefit in acidic environments like the gastrointestinal tract. 3.4 Experimental 3.4.1 Materials N,N Dimethylacrylamide (DMA, Fluka, 98%) was passed th rough a column of basic alumina to remove inhibitor prior to polymerization Azobisisobutyronitrile (AIBN, Sigma, 98%) was recrystallized from ethanol. 2 Aminophenylboronic acid pinacol ester (Boron Molecular), poly(vinyl alcohol) (PVOH Sigma, 31K 50K MW, 87 89% hydrolyzed), acryloyl chloride (Alfa Aesar, 96%), catechol (Alfa Aesar, 99%), Alfa Aesar, 99%), dopamine hydrochloride (DOPA, Sigma), dichloromethane (DCM, Macro n), N,N dimethylacetamide (DMAC, Fisher), N,N dimethylformamide (DMF, BDH, 99.8%), tetrahydrofuran (THF, EMD, 99.5%), anhydrous magnesium sulfate (Fisher), hydroquinone (Alfa Aesar, 99%), sodium hydroxide (Macron), toluene (Sigma,

PAGE 51

51 99.5%), DMSO d 6 and CDCl 3 (Cambridge Isotopes) were used as received. Pentafluorophenyl acrylate (PFPA) was prepared as previously described. 71 3.4.2 Instrumentation and C haracterization 1 H and 19 F NMR spectroscopy were performed on Varian Inova 500 MHz NMR and Mercury 300 MHz NMR spectrometers, respectively. Gel permeation chromatography (GPC) was performed in DMAC (eluent = 0.05 M LiCl) at 50 C with a flow rate of 1.0 mL min 1 (Agilent 1260 Infinity Isocratic Pump G1310B, guard column + two ViscoGel I series G3078 mixed bed columns, molecular weight range: 0 20 10 3 and 0 100 10 4 g mol 1 ). Detection consisted of a Wyatt Optilab T rEX refractive index detector operating at 658 nm and a Wyatt miniDAWN TREOS laser light scattering detector (operating at 50 mW, 658 nm with detection angles of 49, 90, and 131). Molecular weights were determined using poly(methyl methacrylate) standards. UV Vis spectroscopy characterization was conducted on a Molecular Devices Spectra Max M2 spectrophotometer with Greiner Bio one 96 well clear bottom polypropylene reader plates. Rheometry was conducted at 25 C on a TA Instruments ARES LS1 rheometer using a 25 mm flat plate geometry with a 0.5 mm gap. X Ray Intensity data were collected at 100 K on a Bruker DUO diffractometer using MoK = 0.71073 ) and an APEXII CCD area detector. Raw data frames were read by program SAINT1 and integrated usin g 3D profiling algorithms. The resulting data were reduced to produce hkl reflections and their intensities and estimated standard deviations. The data were corrected for Lorentz and polarization effects and numerical absorption corrections were applied based on indexed and measured faces. T he structure was solved and refined in SHELXTL2013, using full matrix least squares refinement. The non H atoms were refined with anisotropic thermal parameters and all of the H atoms were calculated in

PAGE 52

52 idealized po sitions and refined riding on their parent atoms. The asymmetric unit consists of the molecule and a water molecule in general position. The water protons and the amino protons, along with those on C8 and C9 were obtained from a Difference Fourier map and refined freely. In the final cycle of refinement, 3512 reflections (of whi ch 3262 are observed (I)) were used to refine 218 parameters and the resulting R1, wR2 and S (goodness of fit) were 6.73%, 17.25% and 1.182, respectively. The refinement was carried out by minimizing the wR2 function using F2 rather than F values. R1 is calculated to provide a reference to the conventional R value but its function is not minimized 3.4.3 Synthesis of 2 A crylamidopheynlboronic A cid P inacol E ster (2APBAE) 2 Aminophenylboronic pinacol ester (5.00 g, 22.8 mmol) was dissolved in THF (150 mL) along with TEA (3.18 mL, 22.8 mmol), and the resulting solution was chilled to 0 C. A slight excess of acryloyl chloride (2.00 mL, 25.0 mmol) was diluted in THF (50 mL) and added drop wise over the course of 1 h via addition funnel. The solution was allowed to come to room temperature after the addition and was left to stir for 24 h. The mixture was filtered to remove the TEA salt, and the solvent was removed under reduced pressure to yield a slightly yellow solid. The monomer was recrystallized from toluene ( 2) to yield a white solid. Crystals of 2APBAE for X ray crystallography were prepared by solvent evaporation from wet DMAC. 3.4.4 Synthesis of Poly(2APBA co DMA ) (P(2APBA co DMA)) P(2APBA co DMA ) with 10 mol% 2APBA was synthesized via conventional radical poly merization Briefly, DMA (0.815 g, 8.22 mmol) and 2APBAE (0.250 g, 0.915 mmol) were added to a 20 mL vial, along with AIBN (1.50 mg, 9.1510 3 mmol) and DMAC (5.00 mL). The solution was purged with nitrogen for 30 min before being placed

PAGE 53

53 in a reaction bloc k preheated to 70 C. The reaction was allowed to proceed for 18 h to yield a polymer with 10 mol% 2APBAE content. Near quantitative monomer consumption was confirmed via 1 H NMR spectroscopy. The polymerization solution was dialyzed against 2% w/w NaOH in DI water, followed by deionized water, using 3500 MW cutoff dialysis tubing to purify the polymer and remove the pinacol protecting group simultaneously The polymer was recovered via lyophilization to yield a white powder of P(2APBA co DMA) ( M n = 3.4 10 4 g/mol; = 3.5). A polymer bearing 15 mol% 2APBAE was prepared in a similar manner ( M n = 3.8 10 4 g/mol; = 4.2). 3.4.5 Synthesis of Poly(PFPA co DMA) (P(PFPA co DMA)) P(PFPA co DMA) with 10 mol% PFPA was synthesized via conventional radical polymeriz ation DMA (1.00 g, 10.1 mmol) and PFPA (0.265g, 1.11 mmol) were added to a 20 mL vial along with AIBN (1.80 mg, 1.1010 2 mmol) and dry toluene (6.00 mL). The solution was purged with nitrogen for 30 min before being placed in a reaction block preheated t o 70 C. The reaction was allowed to proceed for 18 h. Near quantitative monomer consumption was confirmed via 1 H NMR spectroscopy. The polymer solution was precipitated into cold ether ( 2) and dried under vacuum to yield P(PFPA co DMA) ( M n = 9.0 10 4 g /mol; = 2.9). A polymer bearing 15 mol% PFPA was prepared in a similar manner ( M n = 1.1 10 5 g/mol; = 3.0). 3.4.6 Synthesis of Poly(DOPAAm co DMA) (P(DOPAAm co DMA)) P(PFPA co DMA) (1.29 g, 1.13 mmol PFPA groups) was dissolved in dry DMF (18 mL) with a slight excess of dopamine hydrochloride (0.235 g, 1.24 mmol). After the vial was purged with nitrogen for 30 min, two equivalents of TEA (0.315 mL, 2.26 mmol) were added. The vial was placed in a reaction heating block preheated to 70 C. The reaction wa s allowed to proceed for 18 h. Functionalization was confirmed via 19 F

PAGE 54

54 NMR spectroscopy by the complete disappearance of the pentafluorophenol ester peaks and appearance of the pentafluorophenol peaks. The dopamine functionalized polymer was purified via p recipitation into cold ether ( 2). The polymer was dried under vacuum and stored under nitrogen to prevent oxidation of the catechol. 3.4.7 Hydrogel S ynthesis Solutions of each polymer (10 wt%) were prepared in either deionized water or in a potassium bi phthal ate buffer at (pH 4.0). The P(2 APBA co DMA) solutions were then mixed with the P(DOPAAm co DMA) or PVOH solutions and lightly stirred to induce hydrogel formation.

PAGE 55

55 CHAPTER 4 C HANGE IN INTRAMOLECU LARLY COORDINATED BORONIC A CID MONOMER STRUCTURE RE SULTS IN IMPROVED SE LF HEALING HYDROGEL PROPERTIES 4.1 Introduction Hydrogels are ubiquitous in several fields of research, especially in the area of biomedical engineering where they are used in drug delivery, tissue engineering, and as wound dressings. 39 43 While hydrogels are inherently well suited to these applications, more advanced functionalities can be incorporated to expand their utility. One such functionality is reversible crosslinks. There has b een a wide variety of reversible crosslinks utilized in hydrogels, each with unique advantages and disadvantages. 73 The po tential benefits of reversibly crosslinked hydrogels in biomedical applications are significant and include self healing implants, improved mobility of cells in cell scaffolds, and injectable delivery. 74 76 In the field of reversibly crosslinked hydrogels, our group has focused on re versible covalent bonds. In one example, self healing hydrogels were created from copolymers of N,N dimethylacrylamide (DMA) and diacetone acrylamide. 77 Crosslinking was afforded via formation of oximes with bifunctional alkoxyamines. These hydrogels exhibited self healing behavior in both cut/heal tests and rheometry. A more relevant exa mple is the use of boronate esters as hydrogel crosslinks. The formation of boronate esters occurs when boronic acids are complexed with 1,2 or 1,3 diols. 54 57,78 The lability of t hese boronate esters in aqueous media is governed by an equilibrium that is heavily dependent on both the solution pH and the p K a of the boronic acid. At pH values above the p K a of the boronic acid, boronate ester bond formation is favored while below the p K a the equilibrium favors the free boronic acid and diol. 59 61 The system being in

PAGE 56

56 equilibrium is critical for self healing, as the rearrangement allows for cross linking moieties to break and reform to heal any defects. The main issue with using borona te esters as crosslinks is that exchange between boronic acid/diol and boronate esters is most effective when the pH is near the boronic acid p K a 29 As the p K a of most aryl boronic acids lies in the range of 8 9, materials containing boron ate esters are generally not suited for applications at physiological pH. The ability to form boronate cros s linked hydrogels at neutral and acidic pH was accomplished in a previous work by exploiting intramolecular coordination that stabilized ester formation at reduced pH. 4,32,35,63,79 The monomer 2 acrylamidophenylboronic acid (2APBA) was used to prepare water soluble polymers that were crosslinked via boronate ester formation with diol containing polymers. The carbonyl oxygen of the a crylamide moiety in the 2APBA monomer coordinates with t he boron of t he boronic acid group (Scheme 4 1 ). 64 Scheme 4 1. Equilibrium of 2 acrylamidophenylboronic acid (2APBA) in water. Intramolecular B O bond formation leads to the tetrahedral geometry of the boron atom that promotes boronate ester formation. This coordination shifts the equilibrium towards the boronate form that stabilizes formation of the boronate ester. As this occurs at neutral and even acidic pH, hydrogels crosslinked with variou s diol polymers were able to be formed in pH ranges well below those typical for boronic acid hydrogels. These hydrogels also demonstrated self healing in neutral and acidic media through cut/heal tests and rheometry.

PAGE 57

57 Despite the successful formation of th e hydrogel and demonstration of self healing, the 2APBA hydrogel exhibited one notable drawback. The hydrogel would become opaque over time, usually within 24 to 48 h. The opacity precludes its use in potential applications where optical transparency is ne eded, such as in contact lenses. If the hydrogel is used as a cell scaffold, the cloudiness could complicate or prevent photometric assays. The source of this opacity is likely attributed to the copolymer structure. Specifically, if 2APBA preferentially po lymerizes with itself over cross propagation with DMA, the result would be copolymers with high incorporation of 2APBA early in the polymerization When these homopolymer like chains are crosslinked with poly(vinyl alcohol) (PVOH), it could bring the PVOH chains in very close proximity leading to crystallization of PVOH, a phenomenon that has been previously reported 80 The unequal monomer incorporation could also create a more heterogeneous distribution of crosslinks, leading to unequal stress loading and weaker hydrogels. 81 The poor copolymerization behavior could be rationalized by delocalization of the 2APBA propagating radical. As seen in Scheme 4 2 the propagating vinyl radical has possible delocalization pathways throughout the boronic acid. Scheme 4 2 Possible delocalization pathway of a propagating radical in the APAPBA monomer. The creation of a more stable radical would explain the preferential polymerization of 2APBA. DMA would more likely add to 2APBA to create a stabilized p ropagating radical, and 2APBA would preferentially homopropagate to maintain that radical stability. W e reasoned that an ethylene spacer could be installed between the acrylamide

PAGE 58

58 and the amidoboronic acid moiety to retain the intramolecular coordination of the boronic acid while precluding the delocalization of the radical, (Scheme 4 3 ). Scheme 4 3 Equilibrium of (2 (3 acrylamidopropanamido )phenyl)boronic acid (APAPBA) in water The addition of the ethylene spacer prevent s potential radical delocalization as seen in Scheme 1. In this report we quantify the reactivity of APAPBA with DMA and discuss how this results in improved mechanical properties of self healing hydrogels created from APAPBA copolymers. 4.2 Results and Discussion 4.2.1 Synthesis of APAPBA C opolymers Varying ratios of pinacol ester protected APAPBA (APAPBAE) and DMA were copolymerized by conventional radical polymerization GPC analysis of poly(APAPBAE co DMA) (P(APAPBAE co DMA)) with 5 mol% and 10 mol% and 20 mol% APAPBAE yielded M n values of approximately 8.0 10 4 g/mol ( = 1.6), 7.9 10 4 g/mol ( = 1.6), and 3.3 10 4 g/mol ( = 2.1) respectively (Figure A 3 ). At lower feed ratios, APAPBAE appears to have improved copolymerization behavior over 2AP BAE, achieving higher molecular weights and lower dispersity. GPC analysis of poly(2APBAE co DMA) (P(2APBAE co DMA)) with 10 mol% and 15 mol% 2APBAE yielded M n values of approximately 3.4 10 4 g/mol ( = 3.5) and 3.8 10 4 g/mol ( = 4.2), respectively. P (APAPBA co DMA) with 20 mol% APAPBAE suffers a decrease in molecular weight and dispersity compared to the lower mol% copolymers. The large difference in molecular weight affected the resultant hydrogel properties and is discussed later.

PAGE 59

59 4.2.2 Reactivity R atio S tudies of APAPBAE and 2APBAE R eactivity ratio studies were conducted to define the copolymerization behavior of 2APBAE and APAPBAE via copolymerization at different feed ratios of DMA:boronic acid comonomer. In this study, the reactivity ratio of DM A is denoted as r 1 where: Here, k 11 is the rate constant of the propagating DMA radical reacting with DMA monomer and k 12 is the rate constant of the propagating DMA radical reacting with the boronic acid monomer. Similarly, r 2 is the reactivity ratio of t he boronic acid where r 2 = k 22 / k 21 ; k 22 is the rate constant of the propagating boronic acid radical reacting with boronic acid monomer and k 21 is the rate constant of the propagating boronic acid radical reacting with DMA monomer. Due to previous observa tions of 2APBA being consumed early in the polymerization it was expected that its reactivity ratio would exceed unity. In other words, the rate of homopropagation would be larger than cross propagation. On the other hand, APAPBA should demonstrate improv ed incorporation with DMA and have a reactivity ratio of less than unity. The extended Kelen T d s (EKT) method (Section 4.4.7) was used in this study at DMA:boronic acid comonomer ratios of 90:10, 70:30, 50 :50 or 40:60, and 30:70 (Table 4 1). Conversions were held to 15% or less in all copolymerizations. Table 4 1. Parameter tables for the EKT reactivity ratio studies DMA/2APBAE ratio x y 1 2 z F G 90/10 9.00 6.06 0.060 0.057 0.085 0.663 13.8 7.63 0.392 0.708 70/30 2.33 0.76 0.125 0.077 0.236 0.298 8.58 0.80 0.056 0.601 50/50 1.00 0.30 0.389 0.181 0.596 0.220 6.26 3.16 0.265 0.524 30/70 0.43 0.09 0.122 0.033 0.159 0.195 2.35 4.66 0.580 0.292 DMA/APAPBAE ratio x y 1 2 z F G 90/10 9.00 8.40 0.071 0.070 0.075 0.931 9.69 7.95 0.744 0.907 70/30 2.33 2.85 0.050 0.053 0.043 1.228 1.89 1.50 0.522 0.655 40/60 0.67 1.28 0.085 0.118 0.062 1.976 0.33 0.14 0.109 0.249 30/70 0.43 1.58 0.080 0.163 0.044 3.937 0.10 0.15 0.135 0.093

PAGE 60

60 In the EKT plot, r 1 r 2 m t he DMA:2APBAE plot (Figure 4 1 ) were r 1 = 0.91, r 2 = 7.4, and r 1 r 2 = 6.7. Figure 4 1. Extended Kel en T d s plot of copolymerization of 2APBAE and DMA. 1.30 and R 2 = 0.91. The reactivity ratio data suggests that 2APBAE is more likely to polymerize with itself rather than DMA and copolymers with blocks of 2APBAE could form duri ng the copolymerization, which is consistent with our prediction. In contrast, APAPBAE was expected to propagate with DMA to a greater degree. From the DMA:APAPBAE study, values of r 1 = 0.80, r 2 = 0.009, and r 1 r 2 of 0.007 were deri ved from the EKT plot ( Figure 4 2 ). Figure 4 2. Extended Kelen T d s plot of copolymerization of APAPBAE and DMA. 0.009 and R 2 = 0.96.

PAGE 61

61 Interestingly, r 2 is very close to 0, indicating that APAPBAE vastly preferred cross propagation to DMA, unlike 2APBAE. Th e r 2 value combined with the r 1 value o f 0.80 suggests that P (APAPBA co DMA) would have a more uniform distribution of APAPBA without blocky segments of boronic acid. 4.2.3 Hydrogel F ormation, Transparency Comparison, and S elf healing S tudies After evidence of improved monomer incorporation was obtained, the next step was to synthesize hydrogels and ascertain improvements in material properties. Aqueous solutions of P( APAPBA co DMA) and PVOH were created at 10 wt% at pH 3, 4, 5, 6, and 7 in 0.1 M acetate buffer. Mixing these two solutions resul ted in rapid gelation with mild sti rring at all pH values (Scheme 4 4 ). Scheme 4 4 General crosslinking scheme between P(APAPBA co DMA) and PVOH at neutral pH. The optical clarity of the APAPBA hydrogel was then compared to that of a 2APBA hydrogel an d a 3 acrylamidophenylboronic acid (3APBA) hydrogel (Figure 4 3 ). Figure 4 3. Hydrogels composed of PVOH and copolymers of DMA with A) 3APBA B) 2APBA C) APAPBA. The boronic acid polymers contain 10 mol % boronic acid and are crosslinked with PVOH soluti ons at 10 wt %. Photo courtesy of author.

PAGE 62

62 The APAPBA hydrogel and 2APBA hydrogel were 72 hours old while the 3APBA hydrogel was over a year old. As shown, only the 2APBA hydrogel is cloudy. As 2APBA is the only monomer of the three that can have propagatin g radical delocalization it lends support to the notion that the cloudiness is caused by aggregation of PVOH chains due to the poor copolymerization of 2APBA. The self healing ability of the APAPBA hydrogel was qualitatively measured via a cut/heal test. In this test, two separate halves of 10 mol% APAPBA hydrogel, with one contrasted with cresol red dye, were placed in contact. A timeline of this test is shown in Figure 4 4 where the hydrogel appears to be entire ly healed within 15 min. Figure 4 4 Hydrogel s elf healing test. A) Two separate halves of a cut P( APAPBA co DMA) (10 mol% APAPBA)/PVOH hydrogel. B) The two halves are placed into contact. C) After 15 min, the cut interface is diminished as self healing takes place. D) After 30 min, the gel can be pick ed up and support its weight. Photo courtesy of author. At 30 min the cut interface appeared to be even more faded, and the hydrogel was able to be suspended under its weight. After 4 h, there was no indication, apart from the dye, that the hydrogel was e ve r split in to two pieces (Figure A 17 ). The time taken to heal the APAPBA hydrogel (15 min) is considerably less than the healing time of the 2APBA hydrogel (60 min). 4.2.4 Rheological S tudies Oscillatory shear rheometry was used to characterize the visc oelastic properties and self healing behavior of the APAPBA hydrogel. Rheometry was conducted at 10 mol% and 20 mol% APAPBA concentration. Rheometry was also attempted with an

PAGE 63

63 APAPBA homopolymer, but the homopolymer was not soluble at pH 3 or pH 4. Additio nally, when homopolymer solutions at pH 5 and pH 6 were crosslinked with PVOH, the result was extrusion of water and formation of a dehydrated polymer disk rather than a hydrogel. A dynamic strain sweep was first conducted to determine the linear viscoelas tic region of the hydrogel along with the storage modulus ( ) and loss modulus ( ). As pH increases, increases, with a slight deviation at pH 7, as shown i n Figure 4 5 Figure 4 5 values from the linear viscoelastic region of strain sweeps cond ucted on hydrogels formed with P( APAPBA co DMA) ( 10 mol% APAPBA) (blue circles) and hydrogels formed with P( APAPBA co DMA) (20 mol% APAPBA) (orange diamonds). The increase in modulus with increasing pH is possibly due to the protonation of the boronic a cid amide drawing electron density from the carbonyl. Th e reduced carbonyl ele ctron density would weaken the coordinate bond to the boron center decreasing the stabilization of the boronate ester. This observation could also be due to the improved solubil ity of the polymer at higher pH values. The APAPBA hydrogels proved to have higher values than the 2APBA hydrogels, which were around 1 k Pa, even though the 2APBA hydrogels had a higher boronic acid content (15 mol% vs 10 mol%). The increase in modulus could be due to a variety of factors. As discussed earlier, APAPBA

PAGE 64

64 incorporates more evenly in the polymer as compared to 2APBA. The improvement in monomer distribution would result in more homogeneous crosslinking which in turn would result in a more homo geneous allocation of stress on the crosslinks and increased strength. Additionally, the ethylene spacer may provide increased accessibility of the boronic acids for crosslinking. Interestingly, the hydrogels formed with P( APAPBA co DMA) (10 mol% APAPBA) exhibited higher moduli values than hydrogels formed with P( APAPBA co DMA) (20 mol% APAPBA) (Figures A 11 /A 12 ). The reduction in strength is likely due to the lower molecular weights of the P( APAPBA co DMA) (20 mol% APAPBA) precursors. The 20 mol% polymer had less than half the molecular weight of the 10 mol% polymer, which would result in less entanglement of polymers within the hydrogel resulting in a lower modulus, a phenomen on that has been observed in other boronic acid hydrogels. 35 The self healing ability of these hydrogels was investigated by first conducting a ramping strain sweep The increasing strain eventually results in crosslink separation which is shown by G an d G 6 Figure 4 6 Strain sweep of P( APAPBA co DMA) (10 mol% APAPBA) hydrogel in pH 7 0.01 M acetate buffer. The initial strain is set at 1% (10 rad/s). When strain is increased to 300% (10 rad/s), a G G are broken When the strain is returned to 1%, moduli values recover towards their original values as crosslinks re form.

PAGE 65

65 At this point the hydrogel demo nstrates more fluid character than elastic character. The strain is then reduced to within the linear viscoelastic region immediately after the strain sweep to allow the crosslinks to reform. During this recovery period, the moduli values return to their o riginal values as the crosslinks reconnect. There is a discernible difference in the amount of time required for modulus recovery between acidic and neutral pH, where the recovery time is shorter at acidic pH (Figure A 13 ). The difference is attributed to the dependence of relaxation time ( ) with respect to pH. As the hydrogels with lower are more labile and quicker to reform, there is less time required to recover the initial modulus. Frequency sweeps were conducted to determine values. For covalently crosslinked hydrogels, the modulus is typically frequency independent with > In contrast, hydrogels with dynamic crosslinks will demonstrate frequency dependent moduli values. At lower angular frequencies, the hydrogel crosslinks have enough time t o rearrange and allow for stress relaxation of the hydrogel. In this regime, > and the hydrogel can be considered to be in a more fluid like state. At higher angular frequencies, the crosslinks cannot rearrange quickly enough to dissipate stress, and this results in a crossover of and G ( c ). This point is then used to determine where c As pH increases, the / G crossover decreases, resulting in larger values (Figure 4 7 ). Once again, this trend could be due to the decreased electron donation at lower pH due to amide protonation. The difference in crosslink lability could easily be seen during handling. At lower pH, the hydrogels flowed much more rapidly when perturbed, demonstrating spinnbarkeit behavior whereas the hydrogels near n eutral pH would retain their shape for a l onger timeframe (Figure 4 8 ).

PAGE 66

66 Figure 4 7 values from frequency sweeps conducted on hydrogels formed with P( APAPBA co DMA) ( 10 mol% APAPBA) (blue circles) and hydrogels formed with P( APAPBA co DMA) (2 0 mol% APAPBA) (orange diamonds). With increasing pH, increases. Figure 4 8 Frequency swee ps at pH 4 and pH 7. A) Frequency sweep of hydrogel formed with P( APAPBA co DMA) ( 10 mol% APAPBA) in pH 3 0.1 M acetate buffer (1% strain) B) Frequency sweep of hydrogel formed with P( APAPBA co DMA) (1 0 mol% APAPBA) in pH 6 0.1 M acetate buffer (1% strain) Photo courtesy of author. The 10 mol% hydrogels exhibit larger values than the 20 mol% hydrogels, probably due to greater chain entanglement of the higher molecular weight 10 mol% copolymers (Figures A 14 /A 15 ). Compared to the 2APBA hydrogels, the APAPBA hydrogels

PAGE 67

67 universally have lower values, which explains the faster healing times of the APAPBA hydrogel. 4.2.5 Examining Buffer Effects on R heo logy Considering that the boronic acid monomer exhibits anionic character due to intramolecular coordination, the effect of buffer concentration was explored Specifically if the boronic acid monomer is indeed negatively charged, a lower buffer concentration should reduce charge screening. The reduction in charge screening would result in a more extended polymer chain structu re due to anionic repulsion. The extended struct ure could result in different hydrogel properties as the polymer chain is more accessible for crosslinking., R heometry was conducted t o test this hypothesis on hydrogels formed with P( APAPBA co DMA) ( 10 mol% APAPBA) in 0.01 M and 0.1 M acetate buffer at pH 4 and pH 7. As shown in Figure 4 9 the reduced buffer concentration resulted in both higher modulus and larger values. Figure 4 9 Rheology at different buffer concentrations. A) Strain sweeps of P( APAPBA co DMA) ( 10 mol% APAPBA) hydrogels in 0.1 M pH 4 and pH 7 acetate buffer (10 rad/s) B) Strain sweeps of P( APAPBA co DMA) ( 10 mol% APAPBA) hydrogel in 0.01 M pH 4 and pH 7 acetate buffer (10 rad/s) C) Frequency sweeps of P( APAPBA co DMA) ( 10 mol% APAPBA) hydrogels in 0.1 M pH 4 and pH 7 acetate buffer (1% strain) D) Frequency sweeps of P( APAPBA co DMA) ( 10 mol% APAPBA) hydrogels in 0.01 M pH 4 and pH 7 acetate buffer (1% strain)

PAGE 68

68 4.3 Conclusions In summary, we were able to create a hydrogel that exhibited self healing ability over a wide pH range and demonstrated improved optical clarity and mechanical properties over previously reported materials. An ethylen e spacer used to separate the polymerizable vinyl group from the phenyl boronic acid moiety in the monomer seemed to improve copolymerization behavior with DMA, leading to more homogeneous monomer incorporation. The improved monomer incorp o ration resulted i n reduced hydrogel cloudiness and increased gel modulus. The increased spacing of the boronic acid from the polymer backbone also seemed to enhance crosslinking, resulting in faster hydrogel healing times. Furthermore, the hydrogel exhibited polyelectrolyt e behavior as reduced buffer concentration resulted in improved modulus and stiffness. However, a drawback was encountered increased APAPBA comonomer feed ratio resulted in lower molecular weight cop olymers as seen in P( APAPBAE co DMA) (20 mol% APAPBA). The lower molecular weight of the copolymer precursors produced lower moduli values and shorter values in the resultant hydrogel. Nevertheless, the results demonstrated how a change in monomer design could improve polymerization behavior and in turn improve hydrogel mechanical properties. 4.4 Experimental 4.4.1 Materials N,N dimethylacrylamide (DM A, Fluka, 98%) was passed through a column of basic alumina to remove inhibitor prior to polymerization Azobisisobutyronitrile (AIBN, Sigma, 98%) was recrystallized from ethanol. 2 Aminophenylboronic acid pinacol ester (2APBAE, Boron Molecular), poly (vinyl alcohol) (PVOH Sigma, 31K 50K

PAGE 69

69 MW, 87 89% hydrolyzed), acryloyl chloride (Alfa Aesar, 96%), beta alanine (Alfa Aesar, 98%), dichloromethane (DCM, Macron), 1 [ bis (dimethylamino)methylene] 1 H 1,2,3 triazolo[4,5 b ]pyridinium 3 oxid hexafluorophosphate ( HATU, TCI, >98%), diisopropylethylamine (DIPEA, Alfa Aesar, 99%), N, N dimethylacetamide (DMAc, Fisher), N,N dimethylformamide (DMF, BDH, 99.8%), ethyl acetate (Fisher, 99.9%), hydrochloric acid (Macron), tetrahydrofuran (THF, EMD, 99.5%), anhydrous sodium sulfate (Fisher), sodium chloride (Fisher), sodium hydroxide (NaOH, Fisher), DMSO d 6 MeOD, and CDCl 3 (Cambridge Isotopes) were used as received. 4.4.2 Instrumentation and C haracterization 1 H NMR spectroscopy was performed on a Varian Inova 500 MHz NMR sp ectrometer. Molecular weights and molecular weight distributions were determined via gel permeation chromatography (GP C) in DMAc with 50 mM LiCl at 50 C and a flow rate of 1.0 mL/min (Agilent isocratic pump, degasser, and autosampler; ViscoGel I series 1 series G3078 mixed bed columns, with Molecular weights and dispersities were calculated using Wyatt ASTRA software. Molecular weights were determined u sing poly(methyl methacrylate) (PMMA) standards with molecular weights ranging from 2.210 3 to 9.8810 5 g/mol. Detection consisted of a Wyatt Optilab T rEX refractive index detector operating at 658 nm. Rheometry was conducted at 25 C on a TA Instruments ARES LS1 rheometer using a 25 mm flat plate geometry. Rheometer gap ranged from 0.6 mm to 0.7 mm. 4.4.3 Synthesis of 3 A crylamidopropanoi c A cid Sodium hydroxide (10.0 g, 0.250 mol ) was dissolved in deionized water (100. mL) and the solution was chilled i n an ice bath. To this, beta alanine (10.0 g, 0.112 mol )

PAGE 70

70 was added After 30 min in the ice bath a solution of acryloyl chloride (12.0 mL, 0.148 mol ) in THF (100 mL) was added dropwise. After addition, the solution was allowed to come to room temperature and stir for 18 hours The pH of the solution was slowly lower ed to ~2 with conc entrated hydrochloric acid while maintaining the reaction cold in an ice bath. The acidic solution was saturated with sodium chloride before it was repeatedly extract ed with et hyl acetate. The organic layers were dried over anhydrous sodium sulfate before removing the solvent under reduced pressure. The resulting product was used without further purification. (5 5% yield) 4.4.4 Synthesis of (2 (3 A crylamidopropanamido)phenyl)boro nic A cid P inacol E ster (APAPBAE) 3 A crylamidopropanoic acid (0.561 g, 3.92 mmol) was added to DCM (10.0 mL) along with DIPEA (1.02 g, 7.89 mmol). The solution was stirred until 3 acrylamidopropanoic acid dissolved, at which point HATU (1.49 g, 3.92 mmol) was added The resulting yellow solution was again stirred until all of the HATU was dissolved In a separate solution, 2 aminophenylboronic acid pinacol ester (0.853 g, 3.92 mmol) was dissolved in DCM (5.00 mL). Both solutions were chilled in a salt ice b ath for 30 min before adding the 2 aminophenylboronic acid pinacol ester solution into the HATU solution dropwise over 5 10 minutes. The mixture was allowed to stir in the salt ice bath, and after some time, the desired product began to crystallize from t he reaction solution. The crystals were agitated to ensure continued stirring of the solution. After 3 h ours the solution was placed into the freeze r to finish crystallization The product was recovered by filtration and washed with a minimal amount of ve ry cold DCM. The product was essentially pure at this point but was recrystallized again from toluene to ensure no DCM remained. It is important not to use an excess of HATU, as it is insoluble in DCM until it reacts and would potentially contaminate the f inal product. (70% yield)

PAGE 71

71 4.4.5 Synthesis of P oly(APAPBA co DMA) (P(APAPBA co DMA)) P(APAPBA co DMA ) with 10 mol% APAPBA was synthesized via conventional radical polymerization Briefly, DMA (1.000 g, 10.08 mmol) and APAPBAE (0.400 g, 1.16 mmol) were added to a 20 mL borosilicate glass vial, along with AIBN (1.85 mg, 1.1310 2 mmol) and DMAC (7.00 mL). The solution was purged with nitrogen for 30 min before being placed in a metal reaction block preheated to 70 C. The reaction was allowed to proceed for 18 h to yield a polymer with 10 mol% APAPBAE content. Near quantitative monomer consumption was confirmed via 1 H NMR spectroscopy. The polymerization solution was dialyzed against 0.5 % w/w NaOH in deionized water for 24 hours, followed by deionized water fo r 48 hours, using 3500 MW cutoff dialysis tubing to purify the polymer and remove the pinacol protecting group simultaneously The polymer was recovered via lyophilization to yield a white powder of P(APAPBA co DMA) ( M n = 7.9 10 4 g/mol; = 1.6). Polymer s containing 5 mol% ( 8.0 10 4 g/mol; = 1.6) and 20 mol% ( M n = 3.8 10 4 g/mol; = 4.2) APAPBA were prepared in a similar manner. 4.4.6 Hydrogel S ynthesis Solutions of each polymer (10 wt%) were prepared in acetic acid/sodium acetate buffers at pH 3, 4, 5, 6, and 7. The P(APAPBA co DMA) solutions were then mixed with PVOH solutions and lightly stirred to induce hydrogel formation. 4.4.7 Reactivity R atios Reactivity ratios were conducted via the extended Kelen T d s method. Monomer ratios of 90:10, 70:30, 50:50, and 30:70 were used for DMA/2APBAE. Monomer ratios of 90:10, 70:30, 40:60, and 30:70 were used for DMA/APAPBAE. The polymerization was conducted in DMAc at 0.5 M in a 10 mL Schlenk flask. The

PAGE 72

72 monomer to initiator ratio was 500:1. 2,6 Dimethylanilin e was used as an internal standard for proton NMR determination of monomer conversion (Figures S1/S2). The polymerization solution was subjected to three freeze pump thaw cycles before being backfilled with nitrogen. The studies were run at 70 C in a pre heated oil bath. Total monomer conversion typically ranged from 5 % to 15 %. In the extended Kelen T d s 1 2 are defined as where molar conversion of the monomers, x is the feed ratio ( f 1 /f 2 ) and y is the copolymer composition ratio (F 1 / F 2 ). 1 2 are used to calculate the factor z where The factor z is used along with y to calculate F and G w here The smallest ( F m ) and largest ( F M ) F The factors F G T linearized EKT equation Thus, r 1 r 2

PAGE 73

73 C HAPTER 5 TEMPERATURE INDUCED CONTROL OF SELF HEALING AND STRESS RELAXATION IN DYNA MIC COVALENT HYDROGELS 5.1 Introduction Dynamic covalent bonds exhibit reversibility under equilibrium control. 9 Under certain conditions or stimuli, dynamic covalent bonds can break and reform while under other conditions the bonds remain quite stable. The ability to break and reform while possessing the strength of a traditional covalent bond makes dynamic covalent bonds amenable for use in self healing materials such as gels 77,79,82 The lability of a dynamic covalent bond allows for rearrangement of crosslinks near a defect or damaged area which leads to he aling of the material (Figure 5 1 ). Figure 5 1. Hydrogel self healing behavior stems fr om the equilibrium of dynamic c ovalent crosslinks dissociating and reforming. This equilibrium causes the presence of uncrosslinked moieties at the cut interface, which allows for self healing to occur. Adapted with permission from ref 73. Copyright 2014 R oyal Society of Chemistry. S elf healing hydrogels have potential in important applications such as soft actuators and cell scaffolds 83 85 However, the ability of the crosslinks to rearrange can also be detrimental. The crosslink lability that gives rise to self healing also cau ses stress relaxation Th is fluid like trait is an issue for any application that requires structural rigidity. For hydrogel cell scaffold s the mobility of the encapsulated cells is crucial for cell differentiation and viability 22,86,87 The advantage o f using dynamic covalent hydrogels is that the reversible nature of the crosslinks allows for more facile cell

PAGE 74

74 movement without compromising hydrogel structural integrity or requiring an external stimulus. However, it may be difficult to control relaxation behavior of the hydrogel while maintaining mild conditions that are amenable for cell culture. In both examples, the ability to selectively control the hydrogel self healing and stress relaxation behavior would elevate the usefulness of the hydrogel. In a soft actuator, a dynamic covalent hydrogel could be rigid when needed to perform certain functions but then be able to self heal at rest. The ability to control stress relaxation combined with precise hydrogel architecture could result in the creation of complex tiss ue structures. In this work, thermal control of self healing and stress relaxation is achieved by crosslinking poly(vinyl alcohol) (PVOH) with a copolymer of 2 (3 acrylamidopropanamido) phenyl boronic acid (APAPBA) and N isopropylacrylamide (NIPA M). The dynamic nature of the boronate ester crosslink provides the behavior necessary for self healing and stress relaxation while NIPAM provides a mechanism to arrest the lability of the hydrogel. When polymerized NIPAM exhibits temperature dependent so lubility in aqueous solutions. Above its cloud point, a NIPAM polymer will be insoluble in water. For the poly( APAPBA co NIPAM ) (P(APAPBA co NIPAM)) copolymer the increased hydrophobicity induces aggregation of the copolymers along with some extrusion of w ater from the hydrogel (Figure 5 2 ). Figure 5 2 Thermoreponsive hydrogel scheme. On the left, a P( APAPBAE co NIPAM) (dark blue/green) /PVOH (light blue) hydrogel at room temperature. After it is heated near the cloud point of P( APAPBAE co NIPAM), the co polymer aggregates as the NIPAM becomes hydrophobic (orange) and water is extruded

PAGE 75

75 The combination of polymer aggregation and water extrusion creates a more constrained environment where polymer mobility is restricted, which in turn increases hydroge l rel axation time and prevents self healing. The thermally induced change in polymer mobility was characterized via qualitative healing and creep tests as well as rheo logy. These tests showed that polymer mobility could be reduc ed u sing a mild thermal stimulus causing slower stress relaxation and prevention of self healing The change in mobility was fully reversible as a return normal properties. 5.2 Results and Discussion 5 .2.1 Synthesis of APAPBA/NIPAM Copolymers a nd Cloud Point D etermination Copolymers of APAPBA and NIPAM were synthesized via conventional radical polymerization with 5 mol% and 10 mol% of APAPBA. G PC analysis of P( APAPBAE co NIPAM) with 5 mol% and 10 mol% APAPBAE yielded M n values of approximately 7.9 10 4 g/mol ( = 1.6) and 3.3 10 4 g/mol ( = 2.1) respectively (Figure A 4 ). Copolymers were purified and the pinacol protecting group was removed by dialysis against basic water followed by neutral water. The cloud points of each copolymer were dete rmined via dynamic light scattering (DLS). Due to APAPBA bei ng hydrophilic, the P( APAPBA co NIPAM) copolymers were expected to have a higher cloud point than the typical value of 32 C found in NIPAM homopolymers. Derived count rate, representative of solu tion turbidity, was measured versus temperature. As the solution temperature exceeded the cloud point temperature, the count rate rose sharply from increasing turbidity resulting in an infle ct ion point. The cloud point was determined by equating linear fit lines before and after the inflection point. The corresponding DLS data is shown in Figure 5 3

PAGE 76

76 Figure 5 3 DLS measurements of 5 mg/mL solutions of A) P( APAPBA co NIPAM) (5 mol% APAPBA) and B) P( APAPBA co NIPAM) (10 mol% APAPBA) As expected both c opolymers exhibited a cloud point over 32 C with cloud points of 34.6 C and 40.3 C for the 5 mol% and 10 mol% APAPBA copolymers respectively. Cloud point increased with increased APAPBA mol% makes it less favorable for wate r molecules to dissociate from NIPAM. 5.2.2 Examining the Effect of T emperature on Hydrogel Creep and S elf healing Hydrogels were synthesized from the mixing of 10 wt% aqueous solutions of P( APAPBA co NIPAM) and PVOH Gelation seen through the vial invers ion test, occurred within 10 seconds of stirring the two solutions at all conditions. Simple creep tests were conducted to observe the effect of temperature on hydrogel viscoelasticity. A 10 wt% hydrogel synthesized from 10 mol% P( APAPBA co NIPAM) and PVOH was heated to 40 C for 18 h on a hot plate. At this temperature, the copolymer chains within the hydrogel are aggregated due to hydrophobicity, reducing polym er mobility. As seen in Figure 5 4 the hydrogel retains its original shape and shows no signs o f significant dimensional change Figure 5 4 Temperature effects on hydrogel flow. A) P( APAPBA co NIPAM) (10 mol% APAPBA)/PVOH hydrogel immediately after formation at room temperature B) The same hydrogel after being heated on a hot plate at 40 C for 24 h. The hydrogel shows no significant signs of flow C) Upon return to room temperature, the hydrogel creeps across the vial bottom after 45 min. Photo courtesy of author.

PAGE 77

77 Upon return to 25 C, polymer mobility is restored and the hydrogel can creep and cover the bottom of th e vial. If immersed in an aqueous solution a dynamic covalent hydrogel will eventually dissociate as swelling forces push the crosslinks away from each other. A P( APAPBA co NIPAM) hydrogel immersed in water at room temperature would display the same behavior However, above the cloud point, the restriction of polymer mobility should prevent hydrogel dissolution. A P( APAPBA co NIPAM) hydrogel was attached to a non thermoresponsive hydrogel made from a n APAPBA / N,N dimethylacrylamide (D MA) copolymer and PVOH (Figure 5 5) to test this hypothesis The combined hydrogel was then immersed in 50 C DI water for 24 h. During this time, the hydrogel half made from the DMA copolymer dissolved while the NIPAM copolymer hydrogel half remained inta ct. The P( APAPBA co NIPAM) hydrogel returned to its regular appearance shortly after being removed from the vial and sitting at room temperature. Figure 5 5 Temperature effects on gel dissolution. A) P( APAPBA co NIPAM) (10 mol% APAPBA)/PVOH hydrogel h alf fused with a P( APAPBA co DMA) (10 mol% APAPBA)/PVOH hydrogel half B) Only the P( APAPBA co NIPAM) hydrogel half remains after immersion in 50 C water for 24 h C) The P( APAPBA co NIPAM) gel half returns to normal appearance upon cooling to room temperat ure Photo courtesy of author. The reduction in polymer mobility near the copolymer cloud point was also seen in self healing experiments. At 25 C, a 10 mol% P( APAPBA co NIPAM) hydrogel exhibits normal self healing behav io r when split in half (Figure 5 6 ).

PAGE 78

78 Figure 5 6 Self healing test of A) P( APAPBA co NIPAM) (10 mol% APAPBA)/PVOH hydrogel at room temperature. The two halves took approximately 45 min to heal. B) P( APAPBA co NIPAM) (10 mol% APAPBA)/PVOH hydrogel split at room temperature. I mmediately after putting the halves together, the hydrogel is immersed in 40 C phosphate buffer. C) The hydrogel turned opaque at 40 C. After 1 h of heating, there is no sign of healing in the hydrogel. Photo courtesy of author. However, at 40 C, the s plit halves of the hydrogel are unable to self heal. There were some signs of hydrogel shrinkage as the hydrogel halves pulled away from each other. However the area in contact did not self heal, nor did any self healing occur when the edges were pushed b ack into contact. Upon return to room temperature, self healing ability returned to normal 5.2.3 Temperature dependent R heological S tudies R heological measurements were also conducted to quantify t he arrest of self healing For these experiments, hydroge ls were created from 1 mol%, 5 mol%, 10 mol%, and 15 mol% P( APAPBA co NIPAM). However, the 15 mol% P( APAPBA co NIPAM) hydrogel proved to be too brittle to conduct rheological measurements reliably. On the other hand, the 1 mol% P( APAPBA co NIPAM) hydrogel dissolved into polymer solution when heated to its cloud point. As such, data was only obtained from the 5 mol% and 10 mol% hydrogels. Strain sweeps were conducted at different temperatures to quantify the effect of heating on s el f healing. As seen in Figu re 5 7 upon exposure to high strain, the storage modulus ( G G crossover indicating breakage of crosslinks as the hydrogel takes on a more fluid character. When strain is returned to the linear viscoelastic (LVE) region, G G turn to their original values. This process is reversible over multiple cycles.

PAGE 79

79 Figure 5 7 Temperature effects on strain sweeps. A) P( APAPBA co NIPAM) (10 mol% APAPBA)/PVOH hydrogel strain jumps at 25 C. Within the LVE region at 1% strain G G ), G G separated and the hydrogel takes on a fluid character. Upon return to 1% strain, the moduli values recover to their original values. B) The same hydrogel strain jumps at 40 C. The initial G is reduced an d the hydrogel is unable to recover the initial G value after the high strain cycle. C) Strain jumps conducted after the hydrogel has rested at 25 C for 45 min. The moduli values have largely returned to their original values at 25 C, and the self heali ng ability is mostly recovered as well. However, there is a reduction in storage and loss modulus when the hydrogel is heated to 40 C The moduli decrease may be attributed to several factors. Naturally, some decrease in modulus is expected due to increas ed polymer motion and a decreased population of bound crosslinks This effect can be seen in Figure A 16, where frequency sweeps are conducted on a non thermoresponsive hydrogel at increasing temperatures. While G APAPBA/NIPAM hydrogel. Additionally, G thermoresponsive gel whereas G The APAPBA/NIPAM copolymer is the likely cause for the precipitous drop of G The thermally induced collapse of these polymer chains would not only cause some crosslink separation but would also reduce availability of boronic acid moieties due to steric hindrance. A fter the first G G crossover, G did not return to its original value, though it did equilibrate at a lower value. This behavior is likely due to the strain separating crosslinks which are then unable to reform due to steric constraints. It is worthwhile to note that G returned to above G r the first high strain cycle, indicating that some population of crosslinks was able to reform This result is somewhat in contrast to the cut/heal test where no healing took

PAGE 80

80 place at 40 C. However, the two observa tions are a result of the two different experimental methods. With the cut/heal test, there is mechanical separation of the hydrogel and the healing mechanism requires polymer motion and crosslink exchange across the cut interface. In contrast, the strain sweep maintains the structure of the hy drogel and does not separate the crosslinks to an unrecoverable degree. The reduction of G in the high temperature strain sweep can be attributed to both the extrusion of water and the aggregated copolymer structures reducing mobility. When the temperature is returned to 25 C, normal self healing behavior is recovered as G G are able to return to their original values, demonstrating that the thermally induced restriction of polymer mobility is reversible. F requency sweeps also revealed decreased polymer mobility with increased temperature In this experiment, the hydrogel was subjected t o increasing angular frequency starting from low frequency at a constant strain. A typical dynamic covalent hydrogel will demonstrate frequency dependent behavior where G larger tha n G In this range, t he imposed strain is slow enough that the dynamic crosslinks have enough time to rearrange in response As angular frequency increase s, there is not enough time for the crosslinks to respond and G wi ll rise above G G G c ), which is representative of the slowest mode of relaxation. The terminal relaxation frequency can then be used to determine the relaxation time where c Th e relaxation time can be seen as a metric for crosslink and polymer lability and the ability for the hydrogel to dissipate stress. In Figure 5 8 frequency sweeps for P( APAPBA co NIPAM) with 5 mol% and 10 mol% APAPBA are depicted in 5 degree intervals from 20 C to 40 C

PAGE 81

81 Figure 5 8 Frequency s weeps of A) P( APAPBA co NIPAM) (10 mol% APAPBA)/PVOH hydrogel from 20 C to 40 C and B) P( APAPBA co NIPAM) (5 mol% APAPBA)/PVOH hydrogel from 20 C to 40 C. At 20 C and 25 C, the frequency sweep s demonstrate frequency dependent behavior expected of dynamic covalent hydrogels. Also as the temperature increased, c increased as well. Th e change in terminal relaxation frequency is attributed to decreased crosslink population The increased temperatu re would disfavor crosslink formation as boronate esterification is typically exothermic. 88,89 A dditionally, h igher temperatures provide greater internal energy to overcome the positive activation energy required to break boronate ester crosslinks. 90,91 Thus, there would be less bound crosslinks at a given moment at higher temperature than at a lower temperature. There wa s a transition point at 30 C w here modulu s started to decrease, presumably due to the onset of the cloud point transitio n. As this occu r red the copolymer chains b eca me more hydrophobic and began to aggregate, resulting in lower modulus as discussed before. Progressing even closer to t he cloud point, at 35 C, 37 C, and 40 C, G G d to drop as aggregation increased. Stress relaxation wa s affected as c decreased and eventually disappear ed off the measured frequenc y range at 40 C. At this temperature range restrictio n of polymer mobility counter ed the elevated polymer motion from increased energy In contrast, frequency sweeps conducted on a non thermoresponsive P( APAPBA co DMA) hydrogel showed no increase in at higher temperatures (Figure A 16). Instead, modulus

PAGE 82

82 decrea sed and c inc reased with rising temperature. A comparison of values for all the studied hydrogels can be seen in Figure A 20 Since increased at higher temperatures, there should be a corresponding decrease in stress relaxation rate as the reduced ab ility of the crosslinks to exchange studies were conducted to ascertain the effect of temperature on hydrogel stress relaxation. In this experiment, the hydrogel is sub jected to a constant strain and the stress required to maintain that strain is measured In viscoelastic materials, stress relaxation occurs as the initial stress is dissipated through various mechanisms. For dynamic covalent hydrogels, the primary means of stress relaxation is through the unbinding of the crosslinks and subsequent polymer motion. 21 Given that polymer mobility and crosslink accessibility is reduced at temperatures closer to the cloud point, there should be a decrease in stress relaxation rates for hydrogels at higher temperatures. In addition to the 10 mol% P( APAPBA co NIPAM), stress relaxation experiments were conducted on various hydrogel systems with increased polymer mobility. These various hydrogels were chosen to determine how various methods of enhanc ing polymer mobility would affect stress relaxation. A 5 mol% P( APAPBA co NIPAM) hydrogel was tested since the lower crosslink percentage should allow for greater polymer freedom Hydrogels synthesized in 10 mM pH 4 acetate buffer were also tested T he bor onate ester of the APAPBA monomer has shown increased lability at lower pH values, likely attributed to amide protonation reducing the electron density being donated to the boron center from the carbonyl. Finally, a hydrogel made with low molecular weight (LMW) PVOH (13000 23000 g/mol ) was tested to determine if reduced polymer entanglement would increase stress relaxation The s tress relaxation graphs for a ll hydrogels are seen in Figure A 17

PAGE 83

83 where normalized stress is plotted against time at temperatures ranging from 25 C to 40 C. In general, the short term stress relaxation rate remained relatively constant while the long term stress relaxation rate decreased as the temperature was increased The short term and long term stress relaxation were quantifi ed with 1/2 and 1/10 ; the se factors represent the time taken for stress to be reduced to 50% and 10% of the original stress value s respectively. Overall, both 1/2 and 1/10 decrease d from 25 C to 30 C as polymer motion was increased without substantia l copolymer aggregation However, once t emperatures approach ed the cloud point, the 1/2 values increa sed as polymer mobility decreased The 1/2 values do not vary much between each temperature for all system s, although, in general, the systems with incre ased polymer mobility do show slightly lower 1/2 values than those of the 10 mol% P( APAPBA co NIPAM) hydrogel (Table 5 1 ). Table 5 1. Time required (s) to reach 50% of initial stress for 5 mol% and 10 mol% APAPBA P( APAPBA co NIPAM ) /PVOH hydrogels Tempera ture 10 mol% APAPBA pH 7 1/2 10 mol% APAPBA pH 4 1/2 10 mol% APAPBA/ LMW PVOH pH 7 1/2 5 mol% APAPBA pH 7 1/2 25 C 7.1 0.9 2.8 0.1 5.3 0.4 3.3 0.4 30 C 5.7 0.2 2.1 0.1 5.0 0.3 2.8 0.3 35 C 4.9 0.6 2.1 0.1 4.0 0.3 3.4 0.5 37 C 5.0 0.7 2.4 0.1 3.8 0.1 6.5 1.2 40 C 6.0 1.9 3.3 0.2 4.0 0.1 14.7 1.1 Table 5 2. Time required (s) to reach 10% of initial stress for various 5 mol% and 10 mol% APAPBA P( APAPBA co NIPAM ) /PVOH hydrogels Temperature 10 mol% APAPBA p H 7 1/10 10 mol% APAPBA pH 4 1/10 10 mol% APAPBA/ LMW PVOH pH 7 1/10 5 mol% APAPBA pH 7 1/10 25 C 42.3 3.3 11.9 0.2 27.5 2.2 11.9 0.3 30 C 35.9 0.7 9.5 0.4 25.3 1.0 10.1 0.3 35 C 45.5 11.4 12.0 1.6 24.6 0.1 30.5 1.5 37 C 69. 2 35.5 28.9 5.7 26.4 0.9 665.0 223.8 40 C 114.4 67.3 100.5 20.8 35.4 0.1 N/A While polymer mobility wa s reduced at higher temperatures, it was not entir ely locked resulting in the similar 1/2 values. However, polymer mobility and crossl ink access w ere limited enough to result in an increase in 1/10 values with increased temperature (Table 5

PAGE 84

84 2) T he 10 mol% P( APAPBA co NIPAM) hydrogels synthesized in pH 4 buffer and with LMW PVOH also experienced increased 1/10 values with increased tem peratur e, though the greater polymer mobility of these systems diminished the trend. T he 5 mol% P( APAPBA co NIPAM) hydrogel displayed a somewhat different 1/10 trend compared to the 10 mol% hydrogels. The reduced number of crosslinkable moieties in the 5 mol% APAPBA hydrogel s hould result in lower 1/10 values than those seen in the 10 mol% APAPBA hydrogel s Indeed, this was seen at 25 C and 30 C where t he lesser percentage of crosslinks appeared to compensate for the lower cloud point of the 5 mol% APAP BA copolymer However, there is a sharp upturn in 1/10 at 35 C At 37 C, there is a massive increase in 1/10 and at 40 C ; stress does not even decay to 10% of the original value. The smaller mol% of APAPBA that helped maintain polymer mobility at lowe r temperatures was likely more entrapped once copolymer aggregation began in earnest. With fewer crosslinks able to dissipate stress, the 5 mol% APAPBA hydrogel was rendered less capable of stress relaxation. 5.3 Conclusions In summary, self healing and st ress relaxation behavior of dynamic covalent hydrogels was thermally controlled through the synthesis of NIPAM copolymers. Although polymer motion was increased with elevated temperature, the restriction of and subsequent aggregation prevented any large scale polymer motion needed for creep and self healing. The P( APAPBA co NIPAM ) /PVOH hydrogel demonstrated normal self healing and creep behavior at room temperature. However, at temperatures near the polymer cloud point, both creep and self healing were arrested. In contrast, a dynamic covalent hydrogel with no NIPAM incorporation retained its dynamic nature at elevated temperatu res. The thermally

PAGE 85

85 controlled behavior was also confirmed with various rheological expe riments. Strain sweeps of the P( APAPBA co NIPAM ) /PVOH hydrogel showed normal self healing behavior at 25 C where G G their original values when strain is lowered within the LVE region. The recovery of moduli was not seen when the strain sweep was conducted at 40 C. Upon return to 25 C, the hydrogel was able to recover its original self healing abilities, demonstrating the reversibility of the polymer mobility transition. Frequency sweeps were cond ucted to assess the relaxation time of the hydrogels with respect to temperature. While relaxation time initially increased, it started to decrease once the temperature approached the cloud point. Near or above the cloud point, the relaxation time increase d to a degree where it could not be measured on the given frequency scale. Stress relaxation studies were used to ascertain the temporal profile of stress dissipation and in turn, the extent of polymer mobility. Additional hydrogel systems with enhanced po lymer mobility were examined in this study as well. Across all hydrogels, there was relatively little change in 1/2 as temperature increased since polymer mobility was not restricted enough to prevent short range motion. However, there was a distinct increase in 1/10 as temperature increased because the polymer mobility was restricted enough to prevent long range motion and subsequent stress relaxation. Th e simple incorporation of a the r moresponsive monomer provides a simple method of controlling the stress relaxation and healing behavior of dynamic covalent hydrogels. Future work will be considered for improving t he change in relaxation time with temper ature and with eliminating the drop in modulus with increasing temperature.

PAGE 86

86 5.4 Experimental 5.4.1 Materials N,N D imethylacrylamide (DMA, Fluka, 98%) was passed through a column of basic alumina to remove inhibito r prior to polymerization N Isopropylacrylamide was recrystallized thrice from hexane. 2 Azobisisobutyronitrile (AIBN, Sigma, 98%) was recrystallized from ethanol. 2 Aminophenylboronic acid pinacol ester (2APBAE, Boron Molecular), poly(vinyl alcohol) ( PVOH Sigma, 31K 50K MW, 87 89% hydrolyzed), acryloyl chloride (Alfa Aesar, 96%), beta alanine (Alfa Aesar, 98%), dichloromethane (DCM, Macron), 1 [ bis (dimethylamino)methylene] 1 H 1,2,3 triazolo[4,5 b ]pyridinium 3 oxid hexafluorophosphate (HATU, TCI, >98%), diisopropylethylamine (DIPEA, Alfa Aesar, 99%), N, N dimethylacetamide (DMAc, Fisher), N,N dimethylformamide (DMF, BDH, 99.8%), ethyl acetate (Fisher, 99.9%), hydrochloric acid (Macron), tetrahydrofuran (THF, EMD, 99.5%), anhydrous sodium sulfate (Fisher), sodium chloride (Fisher), sodium hydroxide (NaOH, Fisher), DMSO d 6 MeOD, and CDCl 3 (Cambridge Isotopes) were used as received. 5.4.2 Instrumentation and C haracterization 1 H NMR spectroscopy was performed on a Varian Inova 500 MHz NMR spectrometer. Mole cular weights and molecular weight distributions were determined via size exclusion chromatography (SEC) in DMAc with 50 mM LiCl at 50 C and a flow rate of 1.0 mL/min (Agilent isocratic pump, degasser, and autosampler; ViscoGel I series series G 3078 mixed bed columns, with Molecular weights and dispersities were calculated using Wyatt ASTRA software. Molecular weights were determined using poly(methyl methacrylate) (PMMA) standar ds with

PAGE 87

87 molecular weights ranging from 2.210 3 to 9.8810 5 g/mol. Detection consisted of a Wyatt Optilab T rEX refractive index detector operating at 658 nm. Rheometry was conducted at on a TA Instruments DHR 2 rheometer using a cross hatched 40 mm flat ge ometry. Rheometer gap ranged from 0.8 mm to 1.0 mm. 5.4.3 Synthesis of 3 A crylamidopropanoic A cid Sodium hydroxide (10.0 g, 0.250 mol ) was dissolved in deionized water (100. mL) and the solution was chilled in an ice bath. To this, beta alanine (10.0 g, 0 .112 mol ) was added After 30 min in the ice bath a solution of acryloyl chloride (12.0 mL, 0.148 mol ) in THF (100 mL) was added dropwise. After addition, the solution was allowed to come to room temperature and stir for 18 hours The pH of the solution w as slowly lower ed to ~2 with conc entrated hydrochloric acid while maintaining the reaction cold in an ice bath. The acidic solution was saturated with sodium chloride before it was repeatedly extract ed with ethyl acetate. The organic layers were dried over anhydrous sodium sulfate before removing the solvent under reduced pressure. The resulting product was used without further purification. (5 5% yield) 5.4.4 Synthesis of (2 (3 A cryla midopropanamido)phenyl)boronic A cid P inacol E ster (APAPBAE) 3 A crylamidop ropanoic acid (0.561 g, 3.92 mmol) was added to DCM (10.0 mL) along with DIPEA (1.02 g, 7.89 mmol). The solution was stirred until 3 acrylamidopropanoic acid dissolved, at which point HATU (1.49 g, 3.92 mmol) was added The resulting yellow solution was ag ain stirred until all of the HATU was dissolved In a separate solution, 2 aminophenylboronic acid pinacol ester (0.853 g, 3.92 mmol) was dissolved in DCM (5.00 mL). Both solutions were chilled in a salt ice bath for 30 min before adding the 2 aminophenylb oronic acid pinacol ester solution into the HATU solution dropwise over 5 10 minutes. The mixture was allowed to stir in the salt

PAGE 88

88 ice bath, and after some time, the desired product began to crystallize from the reaction solution. The crystals were agitate d to ensure continued stirring of the solution. After 3 h ours the solution was placed into the freeze r to finish crystallization The product was recovered by filtration and washed with a minimal amount of very cold DCM. The product was essentially pure a t this point but was recrystallized again from toluene to ensure no DCM remained. It is important not to use an excess of HATU, as it is insoluble in DCM until it reacts and would potentially contaminate the final product. (70% yield) 5.4.5 Synthesis of P o ly(APAPBA co NIPAM) (P(APAPBA co NIPAM)) P(APAPBA co NIPAM) with 10 mol% APAPBA was synthesized via conventional radical polymerization Briefly, NIPAM (0.75 g, 6.63 mmol) and APAPBAE (0.255 g, 0.74 mmol) were added to a 20 mL borosilicate glass vial, alon g with AIBN (1.21 mg, 7.3710 3 mmol) and DMAC (3.75 mL). The solution was purged with argon for 30 min before being placed in a metal reaction block preheated to 70 C. The reaction was allowed to proceed for 18 h to yield a polymer with 10 mol% APAPBAE c ontent. Near quantitative monomer consumption was confirmed via 1 H NMR spectroscopy. The polymerization solution was dialyzed against 0.5 % w/w NaOH in deionized water for 24 hours, followed by deionized water for 48 hours, using 3500 MW cutoff dialysis tu bing to purify the polymer and remove the pinacol protecting group simultaneously The polymer was recovered via lyophilization to yield a white powder of P(APAPBA co NIPAM) ( 3.3 10 4 g/mol, = 2.1). Polymers containing 5 mol% APAPBA ( 7.9 10 4 g/mol, = 1.6) were prepared in a similar manner.

PAGE 89

89 5.4.6 Hydrogel F ormation Solutions of each polymer (10 wt%) were prepared in acetic acid/sodium acetate buffers at pH 4 and in phosphate buffers at pH 7. The P(APAPBA co NIPAM) solutions were then mixed with PV OH solutions and lightly stirred to induce hydrogel formation.

PAGE 90

90 CHAPTER 6 CONCLUSIONS AND FUTURE DIRECTIONS The work presented in this dissertation focused on the synthesis and characterization of self healing intram olecularly coordinated boronate ester hydrogels. The use of carbonyl coordinated boronic acid polymers in hydrogels was unprecedented and offered several advantages over Wu lff type boronic acid polymers. The first work demonstrated that 2APBA could be read ily polymerized with other hydrophilic monomers to form a water soluble copolymer. Combining aqueous solutions of the synthesized copolymers with aqueous solutions of PVOH or P( D OPAAm co D MA ) rendered hydrogels almost instantaneously. The resultant hydroge ls were able to self heal at both neutral and acidic pH as seen with qualitative cut/heal tests and rheo logy The ability to self heal at these lower pH ranges is atypical for most boronic acids and is a powerful tool for use in biological applications. Th e use of the dopamine modified copolymer also revealed an interesting insight. Catechols have been shown to have a high binding constant with boronic acids and are often used as the diol in crosslinks. However, the dopamine copolymer in the study exhibited quite rapid oxidation in standard atmosphere, evidenced by the dark coloration typical of oxidized catechols. Th e oxidation resulted in decreased self healing behavior and creep and precluded the use of catechols as the diol crosslinker in future studies. While the first intramolecularly coordinated boronic acid hydrogel was a good start, it did have some disadvantages. Primarily, the hydrogel turned cloudy and opaque over time. The lack of transparency could lead to issues with either optical techniques for characterization or use in optical applications. The cloudiness was attributed to the aggregation of PVOH polymers as this had been previously documented in the creation of PVOH hydrogels. The aggregation could be explained by the copolymerization

PAGE 91

91 beha vior of 2APBAE. As 2APBAE preferred homopropagation to cross propagation, blocky segments of boronic acid would result in the copolymers. The proximity of the PVOH chains bound to these copolymers would result in cloudiness. A new intramolecularly coordina ted boronic acid monomer APAPBAE, was synthesized to remedy this aggregation Reactivity ratio studies conducted on APAPBAE and 2APBAE homopropagation tendency for cross propagation. Thus, the APAP BA hydrogel remained transparent after synthesis. Furthermore, the improved copolymer structure and increased accessibility of the boronic acid moiety resulted in larger modulus values and faster self healing times. This work demonstrated the importance of crosslinker structure and that a molecular change in the crosslinker could evoke change on the macroscopic scale. Finally the third work was conducted to install control over the self healing and stress relaxation behaviors of the self healing hydrogel The ability to control these important characteristics of the hydrogel would be instrumental in synthesizing higher o rder hydrogel constructs, such as precise cell scaffolds. To achieve this control, APAPBA was copolymerized with NIPAM. The temperature d ependent solubility of NIPAM ensured that the APAPBA NIPAM copolymer would aggregate near its cloud point. The subsequent reduction in polymer mobility would limit both self healing and stress relaxation. This approach was chosen since temperature is a fac ile stimulus to control, and no complex synthesis steps are required. The thermal control of self healing and stress relaxation was seen in cut/heal tests, creep tests, and in rheo logical studies Frequency sweeps revealed increasing relaxation times for t he hydrogel as temperature approached the copolymer cloud point. Also, stress relaxation experiments showed a decrease in long term stress relaxation at higher temperatures, correlating with the lack of

PAGE 92

92 creep and self healing behavior However, there is re latively little difference in short term stress relaxation at all temperatures, as it is still possible for some polymer motion even at the cloud point. The drawback to this approach was a reduction in modulus near the cloud point of the APAPBA copolymer. The reduction in modulus was attributed to a reduction in the effective number of crosslinks due to decreased accessibility within the collapsed NIPAM copolymers Nevertheless, the work demonstrated a straightforward method to selectively tune stress rela xation at a temperature range relevant to biological applications. Future work on this project would search for a method to control self healing and stress relaxation without the reduction in hydrogel modulus. In summation, these three research projects proved the viability and tunability of carbonyl intramolecularly coordinated boronic acids in self healing hydrogels. The field of self healing hydrogels remains quite relevant, especially with the advances in biomedical research. The 2APBA hydrogel was in volved in a collaboration with the University of Delaware to explore its viability as a cell culture substrate. This type of collaborative study would also be valuable for the other two hydrogel systems. Additionally, exploration in the delivery of the hyd rogels in three dimensional (3D) printing would be ideal as well. The precise construction of hydrogels in a spatial matrix is one of the components in the dream of creating artificial tissues and organs.

PAGE 93

93 APPENDIX SUPPLEMENTARY FIGURES AND SCHEMES A.1 Gel Permeation Chromatography Traces Figure A 1. GPC trace s of A) 10% P(2APBAE co DMA) and B) 15% P(2APBAE co DMA)

PAGE 94

94 Figure A 2 GPC trace of A) 10% P(DOPA Am co DMA) and B) 15% P(DOPA Am co DMA)

PAGE 95

95 Figure A 3 GPC trace s of A) 5 mol % P( APAPBAE co DMA) B) 10 mol% P(APAPBAE co DMA) C) 20 mol% P(APAPBAE co DMA) Figure A 4 GPC trace s of A) 5 mol % P( APAPBAE co NIPAM B) 10 mol% P(APAPBAE co NIPAM )

PAGE 96

96 A.2 N uclear M agnetic R esonance Spectra Figure A 5 19 F NMR spectra of A ) P(PFPA co DMA) a nd B ) P(PFPA co DMA) after substitution with dopamine, showing cleavage of the pentafluorophenolate group

PAGE 97

97 Figure A 6 Typical 1 H NMR plot of DMA/2APBAE reactivity ratio study. The first spectrum is for t = 0 min and the second spectrum is for t = 120 min. The letters a, b, and c represent the peaks for 2APBAE, DMA, and 2,6 dimethylaniline respectively. The 90/10 monomer ratio is shown here.

PAGE 98

98 Figure A 7 Typical 1 H NMR plot of DMA/APAPBAE reactivity ratio study. The first spectrum is for t = 0 min and the second spectrum is for t = 120 min. The letters a, b, and c represent the peaks for 2,6 dimethylaniline, DMA, and APAPBAE respectively. The 90/10 monomer ratio is shown here.

PAGE 99

99 A.3 Rheom etry Figure A 8 Frequency sweeps of hydrogels formed from A) P(2APBA co DMA) (10 mol% and 15 mol% 2APBA) with PVA at pH 7.0 and B) P(2APBA co DMA) ( 10 mol% ) with P(DOPA Am co DMA) hydrogels at pH 7.0 and 4.0 The concentration of the original polymer s olutions was 10 wt%.

PAGE 100

100 Figure A 9 Strain sweeps of A) 10 mol% P(2APBA co DMA) with P(DOPA Am co DMA) hydrogel and B) 10 mol% P(2APBA co DMA) with PVA hydrogel in DI water The concentration of the original polymer solutions was 10 wt%.

PAGE 101

101 Figure A 10 Strain sweep A) and frequency sweep B) of 10 mol% P(2APBA co DMA) with P(DOPA Am co DMA) hydrogel in DI water oxidized for 24 h The concentration of the original polymer solutions was 10 wt%.

PAGE 102

102 Figure A 11 Strain sweeps of 10 wt %, 10 mol % P(APAPBA co DMA) hydrogel at pH 3, 4, 5, 6, and 7 Figure A 12 Strain sweeps of 10 wt %, 20 mol % P(APAPBA co DMA) hydrogel at pH 3, 4, 5, 6, and 7

PAGE 103

103 Figure A 13 Moduli recovery of hydrogels. A) Strain sweep and modulus recovery of 10 wt% 10 mol% P(APAPBA co DMA) hydrogel in pH 4 0.1 M ac etate buffer (10 rad/sec). The original modulus is restored within approximately 60 seconds. B) The same test for the same hydrogel in pH 6 0.1 M acetate buffer. Notice that the modulus does not recover to the original value within the given timeframe, alt hough the modulus continues to increase over time.

PAGE 104

104 Figure A 14 Frequency sweeps of 10 wt%, 10 mol % P( APAPBA co DMA) hydrogels at pH 3, 4, 5, 6, and 7 Figure A 15 Frequency sweeps of 10 wt %, 20 mol % P( APAPBA co DMA) hydrogels at pH 3, 4, 5, 6, and 7

PAGE 105

105 Figure A 16 Frequency sweeps of P( APAPBA co DMA)/PVOH hydrogel from 25 C to 40 C Figure A 17 Stress relaxation graphs at different temperatures A) P( APAPBA co NIPAM) (10 mol% APAPBA)/PVOH hydrogel stress relaxation from 25 C to 40 C. B) P( APAPBA co NIPAM) (5 mol% APAPBA)/PVOH hyd rogel stress relaxation from 20 C to 40 C. C) P( APAPBA co NIPAM) (10 mol% APAPBA)/PVOH hydrogel at pH 4 stress relaxation from 25 C to 40 C. D) P( APAPBA co NIPAM) (10 mol% APAPBA)/LMW PVOH hydrogel stress relaxation from 25 C to 40 C.

PAGE 106

106 A.4 Supplemen tal Figures and Schemes Figure A 18 X ray structure of 2 acrylamidophenylboronic acid pinacol ester (2APBAE) showing the chelation of the carbonyl oxygen, O1, to the boron center B1. Figure A 19 Two halves of a cut 10 mol% P(APAPBA co DMA)/PVOH hyd rogel after 4 hours of contact. No cut interface is seen indicating successful healing. Photo courtesy of author.

PAGE 107

107 Figure A 20 R elaxation times ( ) versus temperature for APAPBA/NIPAM and APAPBA/DMA hydrogels

PAGE 108

108 LIST OF REFERENCES (1) Bapat, A. P.; Roy, D.; Ray, J. G.; Savin, D. A ; Sumerlin, B. S. J. Am. Chem. Soc. 2011 133 19832. (2) Pablos, J. L.; Vallejos, S.; Ibeas, S.; Muoz, A.; Serna, F.; Garca, F. C.; Garca, J. M. ACS Mac ro Lett. 2015 979. (3) Zhao, Z.; Yao, X.; Zhang, Z.; Chen, L.; He, C.; Chen, X. Macromol. Biosci. 2014 14 1609. (4) Hall, D. G. Structure Properties and Preparation Of Boronic Acid Derivatives Overview of Their Reactions and Applications ; 2005 (5) Lorand, J. P.; Edwards, J. O. J. Org. Chem. 1959 24 769. (6) Springsteen, G.; Wang, B. Tetrahedron 2002 58 5291. (7) Miyamoto, C.; Suzuki, K.; Iwatsuki, S.; Inamo, M.; Takagi, H. D.; Ishihara, K. Inorg. Chem. 2008 47 1417. (8) Tomsho, J W.; Benkovic, S. J. J. Org. Chem. 2012 77 2098. (9) Rowan, S. J.; Cantrill, S. J.; Cousins, G. R. L.; Sanders, J. K. M.; Stoddart, J. F. Angew. Chemie Int. Ed. 2002 41 898. (10) Deuel, von H.; Neukom, H. Die Makromol. Chemie 1949 3 13. (11) Kitano, S.; Kataoka, K.; Koyama, Y.; Okano, T.; Sakurai, Y. Makromol. Chem. Rapid Commun. 1991 12 227. (12) Kikuchi, A.; Suzuki, K.; Okabayashi, O.; Hoshino, H.; Kataoka, K.; Sakurai, Y.; Okano, T. Anal. Chem. 1996 68 823. (13) Lee, M. C.; Kabilan, S.; Hussain, A.; Yang, X.; Blyth, J.; Lowe, C. R. Anal. Chem. 2004 76 5748. (14) Sartain, F. K.; Yang, X.; Lowe, C. R. Anal. Chem. 2006 78 5664. (15) Kataoka, K.; Miyazaki, H.; Bunya, M.; Okano, T.; Sakurai, Y. J. Am. Chem. Soc. 1998 120 12694. (16) Zhang, X.; Guan, Y.; Zhang, Y. J. Mater. Chem. 2012 22 16299. (17) Barnes, H. A. An Introduction to Rheology ; 1989 (18) Gould, P. L. Introduction to Linear Elasticity ; 2013

PAGE 109

109 (19) Mezger, T. G. The Rheology Handbook: For Users of Rotational a nd Oscillatory Rheometers ; 2006 (20) Barnes, H. A. A Handbook of Elementary Rheology ; 2000 (21) Zhao, X.; Huebsch, N.; Mooney, D. J.; Suo, Z. J. Appl. Phys. 2010 107 63509. (22) Chaudhuri, O.; Gu, L.; Klumpers, D.; Darnell, M.; Bencherif, S. A.; Weaver, J. C.; Huebsch, N.; Lee, H.; Lippens, E.; Duda, G. N.; Mooney, D. J. Nat. Mater. 2015 15 326. (23) Ivanov, A. E.; Larsson, H.; Galaev, I. Y.; Mattiasson, B. Polymer. 2004 45 2495. (24) Roberts, M. C.; Hanson, M. C.; Massey, A. P.; Karren, E A.; Kiser, P. F. Adv. Mater. 2007 19 2503. (25) He, L.; Fullenkamp, D. E.; Rivera, J. G.; Messersmith, P. B. Chem. Commun. 2011 47 7497. (26) He, L.; Szopinski, D.; Wu, Y.; Luinstra, G. A.; Theato, P. ACS Macro Lett. 2015 4 673. (27) Hussain, S.; Bharadwaj, S. K.; Chaudhuri, M. K.; Kalita, H. European J. Org. Chem. 2007 374. (28) Yesilyurt, V.; Webber, M. J.; Appel, E. A.; Godwin, C.; Langer, R.; Anderson, D. G. Adv. Mater. 2016 28 86. (29) Yan, J.; Springsteen, G.; Deeter, S.; Wang, B. Tetrahedron 2004 60 11205. (30) Babcock, L.; Pizer, R. Inorg. Chem. 1980 19 56. (31) Wulff, G. Pure Appl. Chem. 1982 54 2093. (32) Kim, K. T.; Cornelissen, J. J. L. M.; Nolte, R. J. M.; van Hest, J. C. M. J. Am. Chem. Soc. 2009 131 13908. ( 33) Li, H.; Liu, Y.; Liu, J.; Liu, Z. Chem. Commun. 2011 47 8169. (34) Negri, G. E.; Deming, T. J. Macromol. Biosci. 2016 1. (35) Piest, M.; Zhang, X.; Trinidad, J.; Engbersen, J. F. J. Soft Matter 2011 7 11111. (36) Liu, X. C.; Hubbard, J. L.; Scouten, W. H. J. Organomet. Chem. 1995 493 91. (37) Hughes, M. P.; Smith, B. D. J. Org. Chem. 1997 62 4492. (38) Jen, A. C.; Wake, M. C.; Mikos, A. G. Biotechnol. Bioeng. 1996 50 357.

PAGE 110

110 (39) Hoffman, A. S. Adv. Drug Deliv. Rev. 2012 64 18. (4 0) Drury, J. L.; Mooney, D. J. Biomaterials 2003 24 4337. (41) Boateng, J. S.; Kerr, M. H.; Stevens, H. N. E.; Eccleston, G. M. J. Pharm. Sci. 2008 97 2892. (42) Vermonden, T.; Censi, R.; Hennink, W. E. Chem. Rev. 2012 112 2853. (43) Seliktar, D. Science 2012 336 1124. (44) Kharkar, P. M.; Kiick, K. L.; Kloxin, A. M. Chem. Soc. Rev. 2013 42 7335. (45) Biegun, M. K.; Werten, M. W. T.; Stuart, M. A. C.; De Wolf, F. A.; De Vries, R. Biomacromolecules 2014 15 6 99. (46) Basak, S.; Nanda, J.; Banerjee, A. Chem. Commun. 2014 50 2356. (47) Kakuta, T.; Takashima, Y.; Nakahata, M.; Otsubo, M.; Yamaguchi, H.; Harada, A. Adv. Mater. 2013 25 2849. (48) Liu, J.; Song, G.; He, C.; Wang, H. Macromol. Rapid Commun. 2013 34 1002. (49) Huang, Y.; Lawrence, P. G.; Lapitsky, Y. Langmuir 2014 30 7771. (50) Buerkle, L. E.; Rowan, S. J. Chem. Soc. Rev. 2012 41 6089. (51) Hager, M. D.; Greil, P.; Leyens, C.; van der Zwaag, S.; Schubert, U. S. Adv. Mater. 2010 2 2 5424. (52) Lee, H.; Dellatore, S. M.; Miller, W. M.; Messersmith, P. B. Science 2007 318 426. (53) Wojtecki, R. J.; Meador, M. A.; Rowan, S. J. Nat. Mater. 2011 10 14. (54) Cambre, J. N.; Sumerlin, B. S. Polymer. 2011 52 4631. (55) Roy, D. ; Cambre, J. N.; Sumerlin, B. S. Prog. Polym. Sci. 2010 35 278. (56) De, P.; Gondi, S. R.; Roy, D.; Sumerlin, B. S. Macromolecules 2009 42 5614. (57) Cheng, F.; Jkle, F. Polym. Chem. 2011 2 2122. (58) Guan, Y.; Zhang, Y. Chem. Soc. Rev. 2013 42 8106. (59) Roy, D.; Cambre, J. N.; Sumerlin, B. S. Chem. Commun. 2008 6 2477.

PAGE 111

111 (60) Roy, D.; Cambre, J. N.; Sumerlin, B. S. Chem. Commun. 2009 7345 2106. (61) Cambre, J. N.; Roy, D.; Sumerlin, B. S. J. Polym. Sci. Part A Polym. Chem. 2012 50 3373. (62) Tarus, D.; Hachet, E.; Messager, L.; Catargi, B.; Ravaine, V.; Auzly Velty, R. Macromol. Rapid Commun. 2014 35 2089. (63) Roy, D.; Sumerlin, B. S. ACS Macro Lett. 2012 1 529. (64) Yang, X.; Lee, M. C.; Sartain, F.; Pan, X.; Lowe, C. R. Chem. A Eur. J. 2006 12 8491. (65) Bhat, K. L.; Howard, N. J.; Rostami, H.; Lai, J. H.; Bock, C. W. J. Mol. Struct. THEOCHEM 2005 723 147. (66) Neugebauer, D.; Matyjaszewski, K. Macromolecules 2003 36 2598. (67) Vatankhah Varnoosfaderani, M.; GhavamiNejad, A.; Hashmi, S.; Stadler, F. J. Chem. Commun. 2013 49 4685. (68) Herlinger, E.; Jamesonb, R. F.; Linerta, W. J. Chem. Soc. Perkin Trans. 2 1995 259. (69) Faure, E.; Falentin Daudr, C.; Jrme, C.; Lyskawa, J.; Fournier, D.; Woisel, P.; Detrembleur, C. Prog. Polym. Sci. 2013 38 236. (70) Boyer, C.; Davis, T. P. Chem. Commun. 2009 6029. (71) Eberhardt, M.; Mruk, R.; Zentel, R.; Thato, P. Eur. Polym. J. 2005 41 1569. (72) Yu, M.; Hwang, J.; Deming, T. J. J. Am. Chem. Soc. 1 999 121 5825. (73) Wei, Z.; Yang, J. H.; Zhou, J.; Xu, F.; Zrnyi, M.; Dussault, P. H.; Osada, Y.; Chen, Y. M. Chem. Soc. Rev. 2014 43 8114. (74) Huebsch, N.; Kearney, C. J.; Zhao, X.; Kim, J.; Cezar, C. A.; Suo, Z.; Mooney, D. J. Proc. Natl. Acad. Sci. 2014 111 9762. (75) McKinnon, D. D.; Domaille, D. W.; Cha, J. N.; Anseth, K. S. Adv. Mater. 2014 26 865. (76) Tseng, T. C.; Tao, L.; Hsieh, F. Y.; Wei, Y.; Chiu, I. M.; Hsu, S. H. Adv. Mater. 2015 27 3518. (77) Mukherjee, S.; Hill, M. R.; Sumerlin, B. S. Soft Matter 2015 11 6152. (78) Brooks, W. L. A.; Sumerlin, B. S. Chem. Rev. 2016 116 1375.

PAGE 112

112 (79) Deng, C. C.; Brooks, W. L. A.; Abboud, K. A.; Sumerlin, B. S. ACS Macro Lett. 2015 4 220. (80) Mori, Y.; Tokura, H.; Yoshikawa, M. J. Mater. Sci. 1997 32 491. (81) Gong, J. P. Soft Matter 2010 6 2583. (82) Deng, G.; Ma, Q.; Yu, H.; Zhang, Y.; Yan, Z.; Liu, F.; Liu, C.; Jiang, H.; Chen, Y. ACS Macro Lett. 2015 467. (83) Chen, B.; Bai, Y.; Xiang, F.; Sun, J. Y.; Mei Chen, Y.; Wang, H.; Zhou, J.; Suo, Z. J. Polym. Sci. Part B Polym. Phys. 2014 52 1055. (84) Nicodemus, G. D.; Bryant, S. J. Tissue Eng. Part B Rev. 2008 14 149. (85) Geckil, H.; Xu, F.; Zhang, X.; Moon, S.; Demirci, U. Nanomedicine 2010 5 469. (86) Came ron, A. R.; Frith, J. E.; Cooper White, J. J. Biomaterials 2011 32 5979. (87) Khetan, S.; Guvendiren, M.; Legant, W. R.; Cohen, D. M.; Chen, C. S.; Burdick, J. A. Nat. Mater. 2013 12 458. (88) Wiskur, S. L.; Lavigne, J. J.; Metzger, A.; Tobey, S. L .; Lynch, V.; Anslyn, E. V. Chem. A Eur. J. 2004 10 3792. (89) Marinaro, W. A.; Prankerd, R.; Kinnari, K.; Stella, V. J. J. Pharm. Sci. 2015 104 1399. (90) Schultz, R. K.; Myers, R. R. Macromolecules 1969 2 281. (91) Koga, K.; Takada, A.; Nem oto, N. Macromolecules 1999 32 8872.

PAGE 113

113 BIOGRAPHICAL SKETCH Chris Deng was born in Taipei, Taiwan in a distant eon Three years after spawning from the primordial ooze, he immigrated to New Orleans in the United States of America. Even at such a young age he could discern the corrupt nature of the city and its vulnerability to hurricanes, so he moved to Pompano Beach, Florida. A few short years later, the young Deng would move to what he considered his childhood home, Vero Beach Florida When he was stil l in middle school, a visiting cousin brought her chemistry textbook along to do homework. His casual perusal of this tome sparked a fascination with the central science. In high school, chemistry was his favorite course and the one he most excelled at Wi th these factors in mind he made the rational decision to initially choose computer science as his major at the Naval Academy. In time, he came to understand that playing video games did not make one adept at or enjoy programming. As he could not switch t o chemistry at that point, he graduated with a Bachelor of Science degree in oceanography. After six years of wondering why he did not quit when he had a chance, Chris left the Navy in 2010. Not desiring any occupation with his skillset, he decided to tak e advantage of his government welfare and pursue an advanced degree in chemistry. He had to take the requisite undergraduate chemistry courses to apply to graduate school. At the same time, he worked part time in a biochemistry laboratory which led to his decision not to study biochemistry Somehow he was accepted to the University of Florida as a graduate student. He was fortunate enough to join Dr. the compet ing applicant was thoroughly unlikable After five years h e graduated in 2017 with his doctorate in chemistry. His research focused on the synthesis and improvement of self healing boronate ester hydrogels.

PAGE 114

114 Chris has never been married He does not intend to have children because the idea frightens him, and the world is overpopulated anyways His hobbies include reading, video games, pretending t o work out, and cars. His goals include owning a Porsche, traveling the world, becoming a streamer on Twitch, and remembering his name when he i s old(er).