Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Optically triggered smart polymers for environmental monitoring
(USC Thesis Other)
Optically triggered smart polymers for environmental monitoring
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
Optically Triggered Smart Polymers for
Environmental Monitoring
By
Michele Elizabeth Lee
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Materials Science)
May 2017
Copyright 2017 Michele Lee
ii
In loving memory of Daw Kyi Kyi Win.
iii
Acknowledgements
Looking back on the incredible journey that was my PhD, I could not have succeeded
without the help and support of many friends and mentors. My greatest and most heartfelt thanks
goes out to my advisor, Professor Andrea Armani, for believing in me even when I didn’t believe
in myself, and encouraging me to take on challenges that I didn’t know I could tackle. As a result
of her mentorship and commitment to excellence, I have grown immensely as a researcher and
expanded the limits of my capabilities far beyond what I had previously imagined possible.
When I was applying to grad schools, I was told that your choice of advisor can make or break
your grad school experience. I had a good feeling about Andrea, so I anxiously signed on with
her for the next half-decade of my life. It would be an understatement to say that I chose wisely.
Andrea’s constant dedication to the success of her students and postdocs is truly inspirational.
I would also like to extend sincere thanks to all of the professors at USC who have taught
me about materials science and polymer chemistry, especially my thesis defense committee
members, Professor Barry Thompson, Professor Jayakanth Ravichandran, and Professor Mark
Thompson. Additionally, I would like to express my great appreciation for the USC staff and
faculty members who have taken the time to train me in using various instrumentation and
analysis techniques: Allan Kershaw, Travis Williams, Frank Devlin, Ralf Haiges, John Curulli,
Matthew Mecklenberg, Michael Nonezyan, and Shuxing Li.
Our research group is fortunate to have had incredibly knowledgeable postdocs who are
also committed to mentoring students. In particular, I would like to thank Dr. Eda Gungor for
taking the time to teach me so much about organic chemistry and also sharing her wisdom and
experience about life with me. I would also especially like to thank Dr. Xiaoqin Shen for
constantly making himself available to answer my chemistry questions and help me troubleshoot
iv
my experiments. Additionally, thank you to our previous group postdocs, Dr. Cecilia Zurita
Lopez, Dr. Rigo Castro Beltran, Dr. Tushar Rane, Dr. Rasheeda Hawk, Dr. Alper Nese, and Dr.
Jason Gamba. Your work in our lab has undoubtedly helped me become a better researcher.
I would like to give an extra special thank you to all of the past and current members of
the Armani Lab for enduring with me in this crazy ride we call grad school. All of you have truly
made my PhD experience so much more enjoyable, whether by sharing laughs with me in the
office, helping me design better experiments, exploring the vastness of LA with me, inspiring
and expanding my research, trying new foods with me, or countless other memorable
experiences I’ve had with you over the past six years. A huge thank you to Soheil Soltani, Erick
Moen, Samantha McBirney, Vinh Diep, Hyungwoo Choi, Alexa Hudnut, Dongyu Chen, Arsenii
Epishin, Rene Zeto, Lili Lash-Rosenberg, Danny Amchin, Omar Garcia, Lea Fang, Brock
Hudnut, Chase Choate, John Lazzeroni, Linda Xu, Ashley Maker, Kelvin Kuo, Simin
Mehrabani, Mark Harrison, Xiaomin Zhang, Ce Shi, Maria Chistiakova, Garrison Crouch, Tara
Assi, Gumi Sethi, Brian Rose, and Leah Tsui. A special thank you to the undergraduate students
who I have had the privilege of mentoring, Martin Siron and Spencer Gilbert. Your intellectual
curiosity and enthusiasm for research keeps me on my toes!
I’ve heard that friends are the siblings God never gave us, and as someone who doesn’t
have any siblings, that rings especially true for me. The friendships I have made at USC have
truly made me feel like I am part of a family, and I am ever so grateful to my friends for their
support. Thank you to Jonathan Lo, Joe Thackwell, Sarah Lee, Marissa Honda, Magdalene Ante,
John Mac, and Mary Boyd for the laughs, prayers, and hugs when I need them. And thank you
especially to my original Office Buddy, my spontaneous dance party partner, my Los Angeles
co-explorer, and my overall life homeslice, Victoria Sun.
v
Finally, I would like to thank the people who became intimately familiar with my
struggles in completing this PhD and who never wavered in providing the love and support that
was just what I needed. Thank you to my parents, and to Andre. You made this possible.
vi
Table of Contents
Acknowledgements ........................................................................................................................ iii
List of Figures ................................................................................................................................ ix
List of Tables ............................................................................................................................... xiv
Abstract ......................................................................................................................................... xv
Chapter 1. Introduction ................................................................................................................. 16
1.1 Motivation ........................................................................................................................... 16
1.2 Chapter Overview ............................................................................................................... 18
Chapter 2. Background ................................................................................................................. 21
2.1 o-Nitrobenzyl Alcohol Derivatives ..................................................................................... 22
2.1.1 oNB Photocleavage Mechanism .................................................................................. 22
2.1.2 Photoinduced Dimerization to Azo Compounds ......................................................... 24
2.1.3 oNB in Polymer Systems ............................................................................................. 25
2.2 Atom Transfer Radical Polymerization (ATRP) ................................................................ 26
2.2.1 Dynamic Equilibrium in CRP ...................................................................................... 27
2.2.2 ATRP Mechanism & Components .............................................................................. 29
2.2.3. Controlling Reaction Kinetics ..................................................................................... 30
2.3 Dose-Response Modeling ................................................................................................... 32
Chapter 3. Synthesis & Characterization of UV Cleavable Polymers .......................................... 36
3.1 Motivation & Background .................................................................................................. 36
3.2 Bifunctional Initiator Synthesis .......................................................................................... 37
3.3 Photocleavable Poly(methyl acrylate) Synthesis ................................................................ 41
3.3.1 General ATRP reaction conditions & experimental procedure ................................... 43
3.3.3 Catalyst ........................................................................................................................ 47
3.3.4 Solvent ......................................................................................................................... 50
3.3.5 Final Polymerization Conditions ................................................................................. 51
3.4 Polymer Characterization .................................................................................................... 54
3.4.1 oNB Verification .......................................................................................................... 55
3.4.2 Fluorescent Emission ................................................................................................... 57
Chapter 4. Photocleavage Kinetics Studies .................................................................................. 62
4.1 Experimental Set-Up and Sample Preparation ................................................................... 64
4.2 Data Analysis ...................................................................................................................... 65
4.2.1 Kinetic Constants ......................................................................................................... 65
vii
4.2.2 Monitoring Photocleavage Reaction Progress ............................................................. 67
4.2.3 Quantum Efficiency ..................................................................................................... 69
4.3 Photocleavage Kinetics Experimental Results .................................................................... 69
4.3.1 Photocleavage in Solution ............................................................................................ 70
4.3.2 Photocleavage in Film .................................................................................................. 72
4.3.3 Photocleavage in Solution vs. Film ............................................................................. 75
4.3.4 Photocleavage Rate Constants ..................................................................................... 76
4.3.5 Trends in Quantum Efficiency ..................................................................................... 78
4.3.6 Comparison with UV-Vis Absorbance Methods ......................................................... 79
Chapter 5. Wearable Ultraviolet Light Dosimeter ........................................................................ 84
5.1 Motivation ........................................................................................................................... 84
5.2 Experimental Details ........................................................................................................... 86
5.2.1 Simulated Sun Exposure .............................................................................................. 87
5.2.2 Quantifying Color Change ........................................................................................... 88
5.3 Wearable Sensor Fabrication .............................................................................................. 89
5.3.1 Bilayer Sensor Fabrication ........................................................................................... 91
5.3.2 Trilayer Sensor Fabrication .......................................................................................... 91
5.3.3 Unsuccessful Sensor Fabrication Methods .................................................................. 91
5.3.3.1 Transfer Printing ................................................................................................... 92
5.3.3.2 PDMS Sputter Coating & Alternative Solvents .................................................... 92
5.3.3.3 PET & Parylene C Coating ................................................................................... 93
5.4 Sensor Characterization ...................................................................................................... 93
5.4.1 Thickness Measurements ............................................................................................. 94
5.4.2 UV-Vis Absorption ...................................................................................................... 95
5.5 Wearable UV Dosimeter Response ................................................................................ 98
5.5.1 Sigmoidal Response ..................................................................................................... 99
5.5.2 PMA-ONB Concentration ........................................................................................... 99
5.5.3 Sensor Lifetime .......................................................................................................... 101
5.5.4 Bending ...................................................................................................................... 102
5.5.5 Water Exposure .......................................................................................................... 104
5.5.6 Sunscreen Compatibility ............................................................................................ 105
Chapter 6. Optically Tunable Frequency Combs ........................................................................ 109
6.1 Alkyne-functionalized oNB Synthesis .............................................................................. 111
6.2 Microsphere Surface Functionalization ............................................................................ 113
6.2.1 Microsphere Fabrication ............................................................................................ 114
viii
6.2.2 Three-Step Surface Functionalization ........................................................................ 114
6.3 Verification of Surface Groups ..................................................................................... 116
6.4 Frequency Comb Testing .................................................................................................. 121
6.4.1 Post-azidation and Post-click Testing ........................................................................ 122
6.4.2 Contaminated Spheres ............................................................................................... 124
Chapter 7. Future Work .............................................................................................................. 127
7.1 Solvent-polymer interaction .............................................................................................. 127
7.2 Structural control of azobenzene isomerization ................................................................ 128
7.3 Fluorescent UV Sensor ..................................................................................................... 129
7.4 Expanding the capabilities of the wearable UV dosimeter ............................................... 130
7.5 Sunscreen efficacy indicator ............................................................................................. 131
Appendix A. oNB Photocleavage with Different Polymers ....................................................... 133
A.1 Alternative monomers for ATRP ..................................................................................... 133
A.1.1 Polystyrene (PS) ATRP ............................................................................................ 133
A.1.2 PHEMA ATRP ......................................................................................................... 135
A.1.3 MPEGMA ATRP ...................................................................................................... 135
A.2 Polystyrene Photocleavage Kinetics ................................................................................ 137
Appendix B. Electro-optically Tunable Waveguides ................................................................. 139
B.1 Lithium Niobate ............................................................................................................... 139
B.2 Lithium Niobate Nanoparticle Synthesis ......................................................................... 140
B.3 Hybrid Silica-Lithium Niobate Films .............................................................................. 143
ix
List of Figures
Figure 2-1. Photoinduced intramolecular hydrogen abstraction of o-nitrobenzyl (oNB) ester to
produce nitrosobenzaldehyde and carboxylic acid. ...................................................................... 23
Figure 2-2. Photodimerization scheme of nitrosobenzaldehyde to azobenzene dimer. ............... 24
Figure 2-3. Examples of hybrid oNB-polymer microstructures and their photodegradable
functionalities. Photodegradable a) brush layer, b) hydrogel, c) micelle, and d) multilayer film. 26
Figure 2-4. General steps in controlled radical polymerization. .................................................. 28
Figure 2-5. General scheme of transition-metal-catalyzed ATRP. .............................................. 29
Figure 3-1. a) Photocleavage of centrally located oNB results in reduction of molecular weight
by half. b) Linear homopolymer with photocleavable ortho-nitrobenzyl (ONB) group in the
center. Upon UV exposure, this group cleaves to nitrosobenzaldehyde and carboxylic acid. ..... 36
Figure 3-2. Esterification reaction pathway to produce brominated ONB bifunctional initiator.
The molar ratio of HNBA:TEA:BiBB is 1:4:4. ............................................................................ 38
Figure 3-3.
1
H NMR Spectra of synthesized oNB initiator before (top) and after (bottom)
purification by column chromatography. ...................................................................................... 39
Figure 3-4.
1
H NMR spectrum of final bifunctional oNB initiator, post purification in CDCl
3
. . 40
Figure 3-5. Variations from standard behavior between monomer conversion and reaction time
in different scenarios. Image adapted from Matyjaszewski, Chem Rev. 2001
18
. ......................... 42
Figure 3-6. Schematic representation of the relationship between PDI and conversion. Image
adapted from Matyjaszewski, Chem Rev. 2001
18
. ........................................................................ 43
Figure 3-7. ATRP of methyl acrylate using bifunctional ONB initiator. Reaction times were
varied to produce n = 50 – 250. .................................................................................................... 44
Figure 3-8. GPC trace of polymer ML-03-24 with multiple peaks deconvoluted using a Gaussian
fit. Peak heights were normalized to account for differences in concentration. ........................... 46
Figure 3-9. Comparison of polymerization results with regular ATRP catalyst (CuBr) and
halogen exchange catalyst (CuCl). ............................................................................................... 48
Figure 3-10. Halogen-exchange-synthesized polymer exposed to various UV irradiation times.
Each peak is labeled with the maximum molecular weight in Da. Peaks heights were normalized
to account for differences in concentration. .................................................................................. 49
Figure 3-11. GPC chromatogram of polymer synthesized in 50% DMF solvent before (black
solid) and after (red dash) UV exposure at 385nm for 1h 42min. ................................................ 51
Figure 3-12. Kinetics of a single ATRP polymerization reaction with methyl acrylate and oNB
initiator using M:I:CuBr:L:solvent = 800:1:2:2:10%. .................................................................. 52
Figure 3-13. Table (left) lists reaction times and molecular weights for polymers used in kinetic
experiments. Graph (right) shows molecular weight distributions for same suite of polymers,
labeled with molecular weight (Da), PDI. .................................................................................... 53
x
Figure 3-14.
1
H NMR spectrum of oNB-modified PMA in THF-d
8
. Inset shows magnification of
the spectral region corresponding to aromatic and end group peaks. Solvent peaks were removed
from the spectrum for clarity. ....................................................................................................... 55
Figure 3-15. a) GPC traces of polymer before and after UV exposure for 79 hours. The
molecular weight reduces to half the original value. b) UV-Vis spectra of 38 kDa polymer before
(solid) and after (dotted) UV exposure (350nm, 1hour). .............................................................. 56
Figure 3-16. Comparison of
1
H NMR spectrum of photocleavable PMA in d-THF before
(bottom) and after (top) 1 hour of exposure to UV light. ............................................................. 57
Figure 3-17. Photocleaved PMA a) in dichloromethane solution and b) as a solid fluorescing
blue after 30 min UV exposure. Non UV exposed samples are next to each sample for
comparison. ................................................................................................................................... 58
Figure 3-18. Fluorescent emission of PMA solution exposed to UVA for 70 min (left) and PMA
thin film exposed to UVA for 80 min (right). ............................................................................... 58
Figure 3-19. Fluorescent emission of photocleavable PMA film on Si with increasing amounts
of UV exposure (given in exposure time). .................................................................................... 59
Figure 4-1. Illustration of polymer chain conformations in solvents with different polymer-
solvent interaction parameters (c). ............................................................................................... 63
Figure 4-2. Schematic of photoreactor set-up for liquid (left) and solid (right) samples. ........... 65
Figure 4.3. GPC Chromatograms of an oNB-PMA polymer sample in THF as it is exposed to
UV light for varying times. The exposure time is shown in the upper right. Blue and red dotted
fits represent cleaved and uncleaved polymer respectively. ......................................................... 68
Figure 4-4. Progression of molecular weight distribution for 38 kDa polymer in THF exposed to
increasing doses of UV irradiation. Left: GPC traces for polymer exposed to 1min (top), 35min
(middle), and 30h 52min (bottom) UV light. Right: Reaction progress plotted as % cleaved and
corresponding fit. .......................................................................................................................... 70
Figure 4-5. Photocleavage reaction progress for the full range of polymer molecular weights
cleaved in THF and their corresponding fits. Key on the top left describes polymer molecular
weight in Da. ................................................................................................................................. 71
Figure 4-6. Photocleavage reaction progress and corresponding fits of polymer in chloroform,
tetrahydrofuran, and toluene. ........................................................................................................ 72
Figure 4-7. Progression of molecular weight distribution for 38 kDa polymer exposed to
increasing doses of UV irradiation as a thin film. Left: GPC traces for polymer exposed to 46
seconds (top), 8 minutes (middle), and 54 minutes (bottom) of UV irradiation. Right: Reaction
progress plotted as % cleaved and corresponding fit. Hollow points were not included in the fit.
....................................................................................................................................................... 73
Figure 4-8. Photocleavage reaction progress for the full range of polymer molecular weights
cleaved as a solid thin film and their corresponding fits. Key on the top left describes polymer
molecular weight in Da. ................................................................................................................ 75
Figure 4-9. Comparison of photocleavage process of 38 kDa polymer in THF solution and as a
thin film spin coated onto a silicon wafer. .................................................................................... 76
xi
Figure 4-10. Dependence of normalized rate constants k
1
and k
2
on the polymer-solvent
interaction parameter. ................................................................................................................... 77
Figure 4-11. Dependence of normalized rate constants k
1
(solid) and k
2
(hollow) on polymer
molecular weight. .......................................................................................................................... 78
Figure 4-12. Dependence of quantum efficiency of photocleavage in solution on solvent type
and molecular weight. ................................................................................................................... 79
Figure 4-13. Comparison of photocleavage kinetics measured by change in absorption vs.
GPC % cleaved. a) Evolution of the UV-Vis absorbance spectrum for 25 kDa polymer cleaved in
THF. Key shows UV exposure time. b) % cleaved obtained by GPC plotted with change in
absorbance at 350nm. .................................................................................................................... 81
Figure 5-1. Wearable UV dosimeter strip on hand before (a) and after (b) UV exposure in a
photoreactor. ................................................................................................................................. 86
Figure 5-2. Left: Image of simulated sun exposure experimental set-up. White arrows indicate
the path of the simulated sunlight. Right: Schematic of light being applied for different amounts
of time through a circular aperture. Each time data point was taken on a different part of the
sample. .......................................................................................................................................... 88
Figure 5-3. Schematics and images of bilayer (left) and trilayer (right) wearable UV sensors. .. 90
Figure 5-4. Images of the trilayer sensor demonstrating the transparency and flexibility. .......... 94
Figure 5-5. Cross-section microscope images of the trilayer sensor fabricated with 10wt% PMA-
EtAc solution. Images indicate a) successful deposition of all three layers, b) deposition of only
PMA active layer and c) deposition of only PDMS coating. The scale bar is 100µm. ................ 95
Figure 5-6. UV-Vis Absorption spectrum of bilayer PMA-PDMS sensor with increasing
simulated sun exposure time. Shows the concurrent decrease in the peak at 269nm and increase
in the peak at 318nm over time as oNB groups are cleaved. ........................................................ 96
Figure 5-7. Change in absorbance of bilayer PMA-PDMS sensor at two specific wavelengths
(269nm and 318nm) monitored as a function of simulated sun exposure time. Inset shows images
of the sensor before and after 2 hours of simulated sun exposure. ............................................... 97
Figure 5-8 Schematic of color changing sensor concept. The yellow color change appears only
after exposure to UV light, and the intensity is correlated to UV exposure dose. ........................ 98
Figure 5-9. Change in yellowness index (DYI) with increasing simulated sun exposure of trilayer
sensors using 0 wt% (control), 5 wt% and 10 wt% PMA-ONB concentrations. Both non-control
concentrations exhibit a sigmoidal relationship with exposure time. ......................................... 100
Figure 5-10 Left: Color change response of sensor stored for 5 weeks (lifetime) vs. newly made
sensors (10wt%) with respect to simulated sun exposure time. Right: Images of trilayer sensor
before (top) and after (bottom) simulated sun exposure. Exposure times from left to right are 10,
20, 30, and 60 minutes. ............................................................................................................... 102
Figure 5-11 Left: Color change response of bent and non-deformed sensors as a function of
simulated sun exposure time. Right: Images of trilayer sensor before (top) and after (bottom)
simulated sun exposure. Exposure times from left to right are 10, 20, 30, and 60 minutes. ...... 103
Figure 5-12 Left: Color change response of water-exposed sensors and non-water exposed
sensors as a function of simulated sun exposure time. Right: Images of trilayer sensor before
xii
(top) and after (bottom) simulated sun exposure. Exposure times from left to right are 10, 20, 30,
and 60 minutes. ........................................................................................................................... 104
Figure 5-13 Left: Color change response of sensors with and without sunscreen as a function of
simulated sun exposure time. Right: Images of trilayer sensor before (top) and after (bottom)
simulated sun exposure. Exposure times from left to right are 10, 20, 30, and 60 minutes. ...... 106
Figure 6-1. Schematic of testing set-up used to measure optical frequency comb generation in
silica microresonators. Light is evanescently coupled into the resonator with a tapered optical
fiber. The fiber output is split between an optical spectrum analyzer (OSA) and a photodetector
(PD) to measure the optical power and transmission respectively. ............................................ 111
Figure 6-2. Reaction Scheme for alkyne functionalization of o-nitrobenzyl alcohol. ............... 112
Figure 6-3.
1
H NMR spectrum of alkyne functionalized oNB alcohol in CDCl
3
. ..................... 113
Figure 6-4. Surface Functionalization reaction scheme for a) azidation of silica surface and b)
Cu mediated click chemistry to attach oNB moiety to the surface. ............................................ 116
Figure 6-5. SEM images of microspheres that had undergone silanization (left) and surface
azidation + click chemistry (right). ............................................................................................. 117
Figure 6-6. Comparison of X-ray Photoelectron Spectra of post-azidation (Step 2) treated silica/
silicon wafers performed at different temperatures. Binding energy was scanned between 195 –
207 eV for Chlorine 2p. .............................................................................................................. 118
Figure 6-7. Comparison of X-ray Photoelectron Spectra of post-azidation (Step 2) and post-click
(Step 3) treated silica/silicon wafers with binding energy range optimized for a) Nitrogen 1s, b)
Carbon 1s, and c) Oxygen 1s. ..................................................................................................... 119
Figure 6-8. Schematic of UV light-induced changes to surface-attached oNB moiety. ............ 120
Figure 6-9. Comparison of X-ray Photoelectron Spectra of oNB-functionalized silica/silicon
wafers before and after 10 minutes of UV light exposure. Binding energy ranges optimized for a)
Nitrogen 1s, b) Carbon 1s, and c) Oxygen 1s. ............................................................................ 121
Figure 6-10. Transmission spectrum (left) and optical spectrum analyzer spectrum (right) of a
functionalized microsphere after azidation (Step 2) on resonance. ............................................ 123
Figure 6-11. Transmission spectrum (left) and optical spectrum analyzer spectrum (right) of a
functionalized microsphere after click chemistry (Step 3) on resonance. .................................. 124
Figure 6-12. Transmission spectrum (left) and optical spectrum analyzer spectrum (right) for a
representative contaminated microsphere. Q = 5.88 x 10
5
......................................................... 124
Figure 7-1. Structures of trans and cis isomers of azobenzene. Isomerization is typically
activated by light; however, the cis isomer will convert to trans over time. .............................. 128
Figure A-1. Reaction Scheme for ATRP of Styrene with oNB initiator. .................................. 134
Figure A-2. Comparison of deconvoluted peaks from GPC chromatograms of PS before (black)
and after (red) UV exposure in THF. There is a clear shift in molecular weight from 5 kDa to 2.5
kDa. ............................................................................................................................................. 134
Figure A-3. Reaction scheme for ATRP of 2-hydroxyethyl methacrylate (HEMA) with oNB
initiator. ....................................................................................................................................... 135
xiii
Figure A-4. Reaction scheme for ATRP of poly(ethylene glycol) methyl ether methacrylate
(MPEGMA) with oNB initiator. ................................................................................................. 136
Figure A-5. Comparison of GPC molecular weight spectra of MPEGMA before (black) and after
(red) 30 minutes of UV exposure ................................................................................................ 136
Figure A-6. Comparison of photocleavage kinetics of oNB-containing PS and PMA. Both
polymers exhibit similar sigmoidal behavior. ............................................................................. 137
Figure B-1. Overview of steps in sol-gel synthesis of silica-coated lithium niobate nanoparticles.
..................................................................................................................................................... 140
Figure B-2. XRD spectra for synthesized lithium niobate (red) and silica-coated lithium niobate
(black) nanoparticles. Each peak is labeled with its corresponding crystallographic planes. .... 141
Figure B-3. TEM Image of silica-coated lithium niobate nanoparticle clusters. Scale bar = 20nm.
..................................................................................................................................................... 143
Figure B-4. Fabrication steps of hybrid lithium niobate-silica films. ........................................ 144
Figure B-5. Difference in appearance of particles with (bottom row) and without (top row) the
polarization filter. a), b) pure lithium niobate powder; c), d) dust particle in pure TEOS film; e),
f) lithium niobate particle in lithium niobate doped TEOS film. ................................................ 145
Figure B-6. Left: EDX analysis of characteristic X-rays emitted at sample point designated by
red cross (inset). The characteristic peak for Nb at 2.2 keV verifies that the particle seen on the
surface is lithium niobate. Right: SEM micrograph of lithium niobate nanoparticles embedded in
silica film. Scale bar is 20µm. ..................................................................................................... 146
xiv
List of Tables
Table 3-1. Rates of propagation and termination for commonly used monomers
23,24
. ................ 45
Table 3-2. Compilation of reaction conditions and molar ratios (with respect to oNB initiator)
for successful methyl acrylate polymerizations using CuBr/ PMDETA. Halogen exchange
reactions and reactions that resulted in multimodal molecular weight distributions are not
included.
a
Reference name in lab notebook. ................................................................................. 53
Table 5-1 Coefficients in Gompertz Fitting Function for DYI of UV-Exposed Sensors ........... 101
Table 5.2 Coefficients in Gompertz fitting function for DYI of UV-exposed trilayer sensors
under various environmental conditions. The original trilayer samples are included in the table
for reference. ............................................................................................................................... 105
xv
Abstract
Stimuli-responsive materials are a rapidly growing area of research, with applications that
range from microfabrication to personalized medicine. Also known as smart materials, they are
engineered to exhibit a controlled property change in response to an external stimulus. Smart
materials are particularly well suited for the field of sensing because they can respond to very
small changes in their environment. The focus of this dissertation is to synthesize, characterize,
and develop applications for optically responsive smart polymer systems.
In the first part of this dissertation, a novel UV-cleavable polymer is synthesized. The UV
response is derived from the photocleavage of a centrally integrated ortho-nitrobenzyl (oNB)
moiety. When UV light is applied to the polymer, it breaks in half where the oNB moiety is
located. In the second part of this dissertation, the photocleavage rate behavior of these polymers
is characterized with respect to environment and molecular weight. The photocleavage is studied
in three different solvents and as a solid thin film. In the last part of this dissertation, two
different applications of the ortho-nitrobenzyl-based photocleavable system are presented: a
wearable UV dosimeter for preventing skin cancer and an optically tunable frequency comb.
16
Chapter 1. Introduction
1.1 Motivation
Some of the most sophisticated and complex materials known to man are those
beautifully engineered by nature, enabling biological function and sustaining life itself. These
materials are tailored with a precise structure that responds uniquely to its environment. In recent
years, the scientific community has taken a cue from nature, resulting in a surge of novel,
synthetically-created responsive materials
1,2
. Dubbed “smart” materials, their utility lies in the
dynamic nature of their properties. When the surrounding environment changes, the material
properties also change in response. Smart materials are utilized across a broad range of
applications, including sensing
3
, tissue engineering
4,5
, drug delivery
6,7
, and responsive surface
coatings
8
. The methods of stimulus across these applications are equally as varied, with the most
common being temperature, pH, redox-activity, chemical activity, and light
9,10
.
Among these stimuli, light is arguably the most versatile because of the many possible
methods of alteration. From the UV to the deep infrared, a wide variety of wavelength ranges are
available for use as triggers
11–13
. Additionally, the exposure time and intensity of light can be
altered to afford additional control. And perhaps most importantly, light can easily be focused to
a specific area, allowing for localized stimulation. In this dissertation, UV light-cleavable
polymers are designed, synthesized, and studied as a model system to probe various aspects of
smart material behavior.
In order to utilize these unique materials, we first need to understand the specific
relationships between material structure and response. Deciding where to place functionality in
the material structure has a significant impact on the overall stimuli response behavior. For
example, in a photoresponsive polymer hydrogel, choosing to incorporate the photolabile unit in
17
the primary vs secondary structure (in the polymer backbone vs in the pendant groups) can either
accelerate or slow down the photodegradation rate
14
. Alternatively, using a functionalized
initiator instead of a functionalized cross-linker can result in star polymers or linear polymers
respectively once photocleaved
15
. There are myriad possibilities in creating novel
macromolecular structures, each one resulting in a different triggered response. The first section
of this dissertation takes an in-depth look at the synthetic methods of creating UV-cleavable
polymers and offers specific methods to accurately ensure functional unit placement along the
polymer chain.
Another method of controlling the material response is by manipulating exposure to the
response trigger. Limiting stimuli quantities, such as the duration or intensity of the trigger,
profoundly affects the resulting change in material properties. However, in the specific field of
UV light-induced polymer degradation, little work has been done in this area and the relationship
between UV dose and photocleavage response remains largely unexplored. In the middle section
of this dissertation, an experimentally derived UV dose dependent kinetic photocleavage model
is presented that aims to begin filling that gap in knowledge. The experiments described are
designed to probe the relationship between UV exposure dose and polymer photocleavage rate.
An understanding of the factors that control the photocleavage rate would not only allow for
optimization of polymers for the current applications, but also introduce a new field of purely
kinetics-based applications.
Finally, the last section of this dissertation details the development and characterization
of two novel applications for UV-cleavable smart materials: a wearable, UV radiation dosimeter
(sun sensor) and a UV-light switchable optical frequency comb. The fact that these two
18
applications are very different from each other and yet use the same photoresponsive unit
illustrates the versatility and ubiquitous nature of smart materials.
1.2 Chapter Overview
Chapter 2 provides background information specific to the synthesis and characterization
of o-nitrobenzyl (oNB) containing UV responsive polymers. An overview of Atom Transfer
Radical Polymerization (ATRP) is presented with particular focus on the structural consequences
of changing each variable in the reaction (monomer, catalyst, ligand, temperature, time). The
oNB photocleavage mechanism is discussed along with secondary reactions that occur as a result
of continued UV exposure. Additionally, a brief overview of generalized dose-response models
is presented.
Chapter 3 details the experimental procedure for the synthesis of the photocleavable
initiator and desired polymer structure. In the initiator synthesis, oNB undergoes a double
esterification reaction to become a bifunctional initiator for ATRP. Polymerization was
conducted concurrently from each initiator end under specific conditions for well-controlled
growth and a suite of polymers with different chain lengths were synthesized. This chapter also
presents an analysis of the structural characteristics of the polymer via Gel Permeation
Chromatography (GPC), UV-Vis Spectroscopy, Fluorescence Spectroscopy, and Nuclear
Magnetic Resonance (NMR) Spectroscopy.
Chapter 4 explores the effect of the polymer environment and molecular weight on the
photocleavage kinetics
16
. Rates of polymer photocleavage were experimentally determined under
four different conditions: different molecular weights, in nonpolar solvent, in polar aprotic
solvent, and as a solid film. In addition to calculating the photocleavage rates, changes in
quantum efficiency of the reaction in each type of environment were also calculated. A strong
19
rate dependence on the solvent-monomer interaction parameter was evident from experimental
results.
Chapter 5 details the incorporation of oNB-containing polymer into a wearable UV light
dosimeter
17
. The dosimeter was fabricated in the form of a flexible, transparent patch composed
of three unique polymer layers: water-resistant PDMS coating, active oNB-polymer layer, and
flexible PEN substrate. Other transparent flexible polymers were used as substrates as well and
are discussed in this section. The dosimeter response was quantified by the change in ASTM
Yellowness Index, and its relationship with increasing UV light exposure was experimentally
determined and discussed. Finally, the sensor robustness and compatibility with sunscreen are
tested and compared with the original color change response results.
Chapter 6 describes how oNB molecules can be attached to the surface of a silica
microresonator, allowing the device to act as a UV-switchable frequency comb. Optical
frequency combs exhibit very narrow linewidths and precise frequency spacing, rendering them
very sensitive to small changes on their surface. By combining this sensitivity with a light-
responsive molecule, we could potentially modulate the frequency comb behavior with an
external light source. The click chemistry techniques used to covalently bond an oNB derivative
molecule to the surface of a silica frequency comb are detailed in this chapter. Preliminary
experimental testing results obtained in collaboration with Andre Kovach are also presented.
Chapter 7 discusses future projects that could be done to further explore many of the
topics discussed in this thesis. In particular, ultrafast spectroscopy is suggested as a method of
elucidating intermediate states in the oNB photocleavage and how they are affected by solvent
interaction. Further studies of the effect of polymer network structure on the isomerization of
azobenzene dimer are suggested. In addition, multiple ways to modify the UV dosimeter (such as
20
changing the UV-exposed color or tuning sensor responsiveness) and use the dosimeter in novel
applications (sunscreen efficacy detector) are presented.
Appendices A and B each detail projects that were started in collaboration with another
group member and that are still actively being researched. Appendix A describes the synthesis
and kinetic studies of additional ONB-functionalized polymers performed in collaboration with
Martin Siron. Appendix B describes the fabrication of an inorganic electro-optic waveguide
composed of lithium niobate and silica, performed in collaboration with Mark Harrison and
Brian Rose.
Chapter 1 References
(1) Stuart, M. a C.; Huck, W. T. S.; Genzer, J.; Müller, M.; Ober, C.; Stamm, M.;
Sukhorukov, G. B.; Szleifer, I.; Tsukruk, V. V; Urban, M.; Winnik, F.; Zauscher, S.;
Luzinov, I.; Minko, S. Nat. Mater. 2010, 9, 101–113.
(2) Esser-Kahn, A. P.; Odom, S. a.; Sottos, N. R.; White, S. R.; Moore, J. S. Macromolecules
2011, 44, 5539–5553.
(3) Anker, J. N.; Hall, W. P.; Lyandres, O.; Shah, N. C.; Zhao, J.; Van Duyne, R. P. Nat.
Mater. 2008, 7, 442–453.
(4) Liu, Z.; Calvert, P. Adv. Mater. 2000, 12, 288–291.
(5) Eglin, D.; Alini, M. Eur. Cells Mater. 2008, 16, 80–91.
(6) Gao, Z. G.; Lee, D. H.; Kim, D. I.; Bae, Y. H. J. Drug Target. 2005, 13, 391–397.
(7) Wilson, D. S.; Dalmasso, G.; Wang, L.; Sitaraman, S. V; Merlin, D.; Murthy, N. Nat.
Mater. 2010, 9, 923–928.
(8) Motornov, M.; Minko, S.; Eichhorn, K. J.; Nitschke, M.; Simon, F.; Stamm, M. Langmuir
2003, 19, 8077–8085.
(9) Schattling, P.; Jochum, F. D.; Theato, P. Polym. Chem. 2014, 5, 25.
(10) Rikkou-Kalourkoti, M.; Loizou, E.; Porcar, L.; Matyjaszewski, K.; Patrickios, C. S.
Polym. Chem. 2012, 3, 105.
(11) Bédard, M. F.; De Geest, B. G.; Skirtach, A. G.; Möhwald, H.; Sukhorukov, G. B. Adv.
Colloid Interface Sci. 2010, 158, 2–14.
(12) Griffin, D. R.; Kasko, A. M. ACS Macro Lett. 2012, 1, 1330–1334.
(13) DeForest, C. a.; Anseth, K. S. Nat. Chem. 2011, 3, 925–931.
(14) Zhu, C.; Bettinger, C. J. Macromolecules 2015, DOI: 10.1021/ma502372f.
(15) Johnson, J.; Finn, M.; Koberstein, J.; Turro, N. Macromolecules 2007, 40, 3589–3598.
(16) Lee, M. E.; Gungor, E.; Armani, A. M. Macromolecules 2015, 48, 8746–8751.
(17) Lee, M. E.; Armani, A. M. ACS Sensors 2016, 1, 1251–1255.
21
Chapter 2. Background
The ability to leverage light as an energy source in chemical reactions has become very
popular in recent years, particularly in the fields of protecting group and triggered release
chemistry
1–4
. By using light to induce a reaction, one often eliminates the need for additional
chemical reagents, allowing compatibility with a wide variety of different solid substrates.
Additionally, photochemical reactions typically occur at mild temperature and pH conditions,
enabling the study of particularly sensitive compounds
5–7
.
When light of a specific wavelength is applied to a photoresponsive molecule, the
absorbed photons convert the molecule into an electronically excited state. These excited states
are highly reactive and, depending on the particular chemistry, can undergo a wide variety of
chemical reactions such as cycloaddition, cyclization, electron transfer, hydrogen abstraction,
etc
1
. In triggered release applications, the photochemical reaction will result in permanent
cleavage of one or more bonds, releasing the target group to the environment. Some examples of
more commonly used photoremovable protecting groups include arylcarbonylmethyl groups,
nitroaryl groups, coumarin-4-ylmethyl groups, and arylmethyl groups
8–13
.
All of the work presented in this dissertation focuses on photoresponsive materials that
leverage the photoreactivity of the ortho-nitrobenzyl (o-NB or oNB) functional group. While the
specific applications and function of the different macromolecules vary between the projects
presented here, many of the synthetic methods and the basic photoreaction mechanism remain
the same. This chapter provides a brief overview of important characteristics of the oNB
photoreaction, polymerization methods, and kinetic characterization techniques relevant to the
overall theme. More detailed background information that is specific to each chapter is included
in the relevant chapter introduction.
22
2.1 o-Nitrobenzyl Alcohol Derivatives
Of the photoresponsive groups discovered to date, oNB derivatives remain the most
thoroughly researched class of photocleavable molecules in both academia and industry
14–18
.
oNB derivatives were first synthesized as a photolabile protecting group in organic small
molecule synthesis, and today are utilized in a wide variety of applications such as triggered
payload release, photoreconfigurable hydrogels, studying cell signaling, and
microfabrication
4,7,19–23
. oNB groups are so widely used because their photoresponse exhibits a
high quantum yield compared to other photocleavage reactions, and the resulting released
photoproduct can be tailored based on the starting molecule chemistry.
2.1.1 oNB Photocleavage Mechanism
The generalized photoreaction proceeds via an intramolecular hydrogen abstraction from
the o-alkyl substituent to the nitro group (Figure 2-1) to form nitrosobenzaldehyde and a leaving
group (carboxylic acid for the oNB ester shown in Fig. 2-1)
17,18
. The specific leaving group that
is released can be modified via substitutions on both the benzylic group and the ester. However,
these substitutions have effects on the corresponding intermediate states, and on the
photoreaction rates as well
4,17,24,
.
23
Figure 2-1. Photoinduced intramolecular hydrogen abstraction of o-nitrobenzyl (oNB) ester to produce
nitrosobenzaldehyde and carboxylic acid.
Absorption wavelengths for the oNB parent molecule and nitroso photoproduct vary by
50-100nm, making it easy to use absorption as a marker for reaction progress
18,24
. The oNB
parent molecule exhibits absorption around 300nm, and thus the photoreaction will proceed most
efficiently when exposed to UVB light. However because many applications require biological
compatibility, most studies do not use this wavelength range, but instead opt for less damaging
UVA light. In this dissertation, oNB photocleavage is observed at wavelengths up to 385nm. In
addition, the photocleavage wavelength can be shifted by substituting the aromatic ring or
changing the benzyl position of the linker
25
. Once photocleavage has occurred, the absorption of
the reaction solution shifts to a longer wavelength due to the formation of photoproducts, whose
absorption is closer to 400nm
24,26
.
24
2.1.2 Photoinduced Dimerization to Azo Compounds
In many cases, the photoreaction does not stop after formation of the nitroso product.
Upon exposure to UV light, nitrosobenzaldehyde can react with itself to form an azobenzene
dimer (Figure 2-2)
27,28
.
Figure 2-2. Photodimerization scheme of nitrosobenzaldehyde to azobenzene dimer.
Unlike the nitroso compound, the azobenzene dimer typically has a yellow-orange color
(depending on trans or cis photoisomerization of the azo bond), which serves as a visual
indicator of dimerization. Azo dimer formation is often an undesirable consequence of oNB
photocleavage because it can significantly reduce product yields. The fact that dimerization is
triggered by UV light coupled with the UV light absorption of the azo dimer itself causes it to act
as an internal light filter that prevents UV light from reaching oNB. Several studies have shown
that dimerization can be reduced or eliminated by substituting the α-position on the o-nitrobenzyl
group
29
or using 2-(2-nitrophenyl)-propoxycarbonyl to eliminate the formation of o-
nitrosobenzaldehyde entirely
30
. In addition, a radical scavenger such as butylated
hydroxytoluene can be added to the system to inhibit reactivity of the aci-nitro tautomers
responsible for dimerization.
25
2.1.3 oNB in Polymer Systems
When used on their own, oNB derivatives can function as a photoswitch to release
specific molecules upon applied UV light exposure. However much research today focuses on
oNB groups embedded in a hybrid system that allows functionality that is dependent on their
combined properties with a secondary matrix (Figure 2-3). Within the class of polymers, oNB
groups can be embedded into many different locations such as in cross-linkages, side chains, as
monolayers, and as separating junctions between blocks of block copolymers. As mentioned in
Chapter 1, the location of oNB placement dictates the final photocleaved structure and resulting
application of each unique material. In order to embed the oNB group in each type of location
along the polymer, it must be transformed into a different starting polymer component. The next
section addresses the function of each type of starting polymer component and provides a review
of the controlled radical polymerization method used in the experiments described in this
dissertation.
26
Figure 2-3. Examples of hybrid oNB-polymer microstructures and their photodegradable functionalities.
Photodegradable a) brush layer, b) hydrogel, c) micelle, and d) multilayer film.
2.2 Atom Transfer Radical Polymerization (ATRP)
Atom Transfer Radical Polymerization (ATRP) is a method of polymerization that
belongs to the overall class of controlled radical polymerization (CRP). CRP is characterized by
an active chain end which is a free radical that grows with the addition of monomers to this
radical end. Control over this type of polymerization was previously considered impossible
because of the extremely fast rate at which two radicals near each other will terminate. CRP
changed that notion by introducing the idea of dynamic equilibrium, in which radical coupling is
mitigated by continuous activation and deactivation of the propagating chain ends. In ATRP, the
propagating radicals are reversibly activated and deactivated to form active/ dormant states that
can be reactivated many times during the reaction via the change in oxidation state of an added
catalyst
31,32
. This concept of dynamic equilibrium is key to ATRP’s utility and popularity.
27
Because of this reversible and controllable activation/ deactivation, one can achieve very low
polydispersity values and a high degree of control over molecular architecture. As a result, many
new chain topologies such as stars, brushes, combs and networks can now be synthesized
33–36
.
Since its discovery in 1995, ATRP has been among the most widely studied methods of
polymerization and is used to produce ca. 100 million tons of polymers each year
37–39
.
In addition to new molecular architectures, ATRP also contributed greatly to the field of
functional polymer design. CRP is also known as “living” polymerization because the chain end
retains its functionality and polymerization can be restarted or brought back to life even after a
chain end has been terminated
37,40–42
. The ability to stop and restart polymer growth allows the
chemist to treat homopolymers as blocks that can be combined and functionalized at will, simply
by replicating the original reaction conditions to restart the propagating radical. This section will
discuss the primary methods of controlling polymer growth and the major defining metrics used
to characterize polymer chains.
2.2.1 Dynamic Equilibrium in CRP
As mentioned earlier, the defining characteristic of a CRP reaction is the dynamic
equilibrium between propagating radicals and dormant species. The rate constants describing this
equilibrium will ultimately determine many important characteristics of the end polymer, such as
molecular weight and polydispersity. Therefore, it is important to understand each step within
the polymerization and its corresponding rate constant.
CRP with reversible termination (living polymerization) can generally be divided into
three major steps: Initiation, Propagation, and Termination (Figure 2-4). During the initiation
process, a halogenated initiator undergoes homolytic bond cleavage to produce one reactive and
one stable free radical. The reactive radicals rapidly initiate polymerization with nearby
28
monomers, while the stable radicals will remain in solution. These stable radicals are also called
persistent radicals because they persist in solution with a much longer lifetime than the reactive
radicals, and in fact build up in to a concentration at least 4 orders of magnitude greater than the
reactive radicals over the course of the propagation reaction. The stable radicals function as
controlling or mediating agents by coupling with initiating and propagating radicals to reversibly
convert them to dormant, non-propagating species. The persistent radical effect also serves to
suppress undesired bimolecular coupling between two propagating radical chains.
Figure 2-4. General steps in controlled radical polymerization.
The critical equilibrium in chain propagation exists between the propagating radical
(active state) and the temporarily halogen-terminated dormant species (dormant state). As noted
in Fig. 2-4, the equilibrium highly favors the dormant species, often by several orders of
magnitude. The overall result is that chain propagation via the propagating radical occurs at a
29
controlled rate over the course of the reaction, allowing for the prediction of molecular weight
and growth of polymer chains that are roughly the same size.
2.2.2 ATRP Mechanism & Components
ATRP follows the same steps as CRP, with the specification that a transition metal
catalyst drives radical generation. A generalized ATRP mechanism, labeled with the major
equilibrium rate constants, is shown in Figure 2-5.
Figure 2-5. General scheme of transition-metal-catalyzed ATRP.
The major components of an ATRP reaction are: a transition metal catalyst with multiple
oxidation states, a ligand to aid in catalyst solubility, monomer, and a halogenated initiator.
Additionally, the reaction may be conducted with or without solvent (solution or bulk
polymerization, respectively). First, initiation occurs via a reversible redox process whereby the
catalyst, T
n
, abstracts a halogen atom, X, from the initiator. Consequently, the catalyst increases
in oxidation state. This process occurs with a rate constant of activation k
act
. Following
activation, the free radical on the initiator is now able to react with monomers in a propagation
process characterized by the rate constant of propagation, k
p
. After propagation, this radical is
reversibly terminated by reacting with the catalyst in its higher oxidation state, resulting in a
30
transfer of the halide back to the propagating chain end with the rate constant of deactivation
k
deact
. The atom transfer step between the catalyst and growing chain end is what gives ATRP its
name. The reaction rate K = k
act
/k
deact
is controlled by the continuous equilibrium between the
polymer-radical chain end (active state) and the polymer-halide chain end (dormant state). As
noted in Figure 2-4, the dormant state is highly favored in equilibrium so that radical
concentration, and consequently bimolecular coupling (governed by the rate constant of
termination, k
t
), is minimized.
Each component of the reaction plays an important role in the kinetics of the system, both
on an individual level and in conjunction with the other system components. For example, an
initiator must be chosen with a rate of activation much higher than the rate of propagation
(typically governed by the choice of catalyst/ligand complex), otherwise the chains will not all
begin at the same time and a very polydisperse sample will result. Also, monomer, solvents, and
ligands must be optimized for solubility in order to ensure adequate distribution throughout the
solution. In some cases, the final polymer is not soluble in its own monomer solution and the
addition of a solvent is necessary to facilitate solubility.
2.2.3. Controlling Reaction Kinetics
While the previous subsections discussed the effects of internal component choice on
reaction kinetics, this subsection focuses on changing reaction kinetics via externally controlled
variables i.e. reaction conditions. Temperature, reaction duration, and relative component
concentrations each play an important role in kinetics. But first, we will define some key terms in
discussing the reaction kinetics. The concentration of propagating radicals, [R], is defined as
𝑅∙ =
$ %['
(
]
['
( *+
]
(2-1)
31
where K = k
act
/ k
deact
, [I] is the concentration of initiator, and [T
n
] is the concentration of
transition metal catalyst. From Figure 2-4, we can assume that the rate of polymerization
(equivalent to the rate of propagation), R
p
, is equal to
𝑅
,
=𝑘
,
𝑅∙ [𝑀] (2-2)
where k
p
is the rate constant of propagation and [M] is the monomer concentration. Combining
(2-1) and (2-2) yields the polymerization rate as
𝑅
,
=
/
0
$ 1 %['
(
]
['
( *+
]
=𝐶 (2-3)
or a constant, C. In more general terms, the polymerization rate is also expressed as the change in
monomer concentration, ln([M
0
]/[M]), with time. Rearranging, we can see that
ln
[1 ]
5
[1 ]
=𝐶𝑡 (2-4)
or that the change in monomer concentration is linear with respect to reaction time. Physically,
this means that in an ideal ATRP reaction, propagation proceeds via first order kinetic behavior.
While many reactions in reality are far from ideal, the ATRP reaction often does exhibit this
linearity between change in monomer concentration and time. In fact, the degree of linearity
between ln([M
0
]/[M]) and reaction time is used as a universal metric to check for control of the
polymerization. This relationship is only observed if there is a constant concentration of the
active propagating radical, achieved by balancing the rates of activation and deactivation.
As a result of this defined relationship, one can pre-determine the required reaction time
to produce a desired molecular weight. This feature, along with preserved end group
functionality and typically narrow molecular weight distribution, make ATRP an ideal choice for
polymerization of many different types of functional polymers.
32
2.3 Dose-Response Modeling
Typically, the kinetics of rapid release agents like o-nitrobenzyl groups are characterized
by a quantity known as the appearance rate constant, k
app
. The term appearance refers to the
appearance of the desired free substrate, which in the case of the oNB ester shown in Figure 2-1
is R’-carboxylic acid. In a generic trigger-response reaction, k
app
is defined as the inverse of the
lifetime of the rate-determining intermediate, τ
rd
; k
app
= 1/ τ
rd
. However, this definition can be
approximated in the case of oNB photoisomerization without measuring intermediate state
lifetimes. As shown in Fig. 2-1, the rate-determining step for oNB is generally not the formation
of a reaction intermediate, but instead the release of leaving group (which is often slower by
several orders of magnitude)
17
. In this case k
app
can be approximated by directly monitoring the
appearance of the photoreaction products.
There are several ways to monitor the photoproduct formation. The most direct method is
to use ultrafast laser spectroscopy to measure rapid changes in absorption corresponding to aci-
nitro tautomer formation
43
. However, this method is both costly and time consuming. A similar
and simpler method is to use UV-Vis spectroscopy to monitor absorption changes on a much
longer time scale
26
. Yet, individual peaks are often difficult to separate in absorbance spectra,
making it difficult to directly monitor the formation of a single compound. In addition, both
methods are best suited for measuring the kinetics of an isolated oNB moiety, not an oNB group
in a complex polymer system. Recently, several groups have developed unique methods to study
the reaction kinetics of hybrid oNB-polymer systems. These methods track the progress of
secondary markers related to photocleavage such as changes in mechanical properties or mass
loss to estimate photoproduct formation
7,21
.
33
Chapter 2 References
(1) Hoffmann, N. Photochemical Reactions as Key Steps in Organic Synthesis. Chem. Rev.
(Washington, DC, U. S.) 2008, 108 (Copyright (C) 2013 American Chemical Society
(ACS). All Rights Reserved.), 1052–1103 DOI: 10.1021/cr0680336.
(2) Xuan, J.; Xiao, W. J. Visible-Light Photoredox Catalysis. Angewandte Chemie -
International Edition. 2012, pp 6828–6838.
(3) Klan, P.; Solomek, T.; Bochet, C. G.; Blanc, A.; Givens, R.; Rubina, M.; Popik, V.;
Kostikov, A.; Wirz, J. Photoremovable Protecting Groups in Chemistry and Biology:
Reaction Mechanisms and Efficacy. Chemical Reviews. 2013, pp 119–191.
(4) Pelliccioli, A. P.; Wirz, J. Photoremovable Protecting Groups: Reaction Mechanisms and
Applications. Photochem. Photobiol. Sci. 2002, 1 (7), 441–458 DOI: 10.1039/b200777k.
(5) Kloxin, A. M.; Tibbitt, M. W.; Anseth, K. S. Synthesis of Photodegradable Hydrogels as
Dynamically Tunable Cell Culture Platforms. Nat. Protoc. 2010, 5 (12), 1867–1887 DOI:
10.1038/nprot.2010.139.
(6) Kloxin, A. M.; Tibbitt, M. W.; Kasko, A. M.; Fairbairn, J. a.; Anseth, K. S. Tunable
Hydrogels for External Manipulation of Cellular Microenvironments through Controlled
Photodegradation. Adv. Mater. 2010, 22 (1), 61–66 DOI: 10.1002/adma.200900917.
(7) Tibbitt, M. W.; Kloxin, A. M.; Anseth, K. S. Modeling Controlled Photodegradation in
Optically Thick Hydrogels. J. Polym. Sci. A. Polym. Chem. 2013, 51 (9), 1899–1911 DOI:
10.1002/pola.26574.
(8) Kammari, L.; Solomek, T.; Ngoy, B. P.; Heger, D.; Klan, P. Orthogonal Photocleavage of
a Monochromophoric Linker. J. Am. Chem. Soc. 2010, 132 (33), 11431–11433 DOI:
10.1021/ja1047736.
(9) Klan, P.; Zabadal, M.; Heger, D. 2,5-Dimethylphenacyl as a New Photoreleasable
Protecting Group for Carboxylic Acids. Org. Lett. 2000, 2 (8), 1569–1571 DOI:
10.1021/ol005789x.
(10) Suzuki, A. Z.; Watanabe, T.; Kawamoto, M.; Nishiyama, K.; Yamashita, H.; Ishii, M.;
Iwamura, M.; Furuta, T. Coumarin-4-Ylmethoxycarbonyls as Phototriggers for Alcohols
and Phenols. Org. Lett. 2003, 5 (25), 4867–4870 DOI: 10.1021/ol0359362.
(11) Dore, T. M. Multiphoton Phototriggers for Exploring Cell Physiology. In Dynamic Studies
in Biology: Phototriggers, Photoswitches and Caged Biomolecules; 2005; pp 435–459.
(12) Pengfei, W.; Huayou, H.; Yun, W. Novel Photolabile Protecting Group for Carbonyl
Compounds. Org. Lett. 2007, 9 (8), 1533–1535 DOI: 10.1021/ol070346f.
(13) Wang, P.; Mondal, M.; Wang, Y. Photolabile Carbonyl Protecting Group: A New Tool for
Light-Controlled Release of Anticancer Agents. European J. Org. Chem. 2009, No. 13,
2055–2058 DOI: 10.1002/ejoc.200900093.
(14) Schattling, P.; Jochum, F. D.; Theato, P. Multi-Stimuli Responsive Polymers – the All-in-
One Talents. Polym. Chem. 2014, 5 (1), 25 DOI: 10.1039/c3py00880k.
(15) Rikkou, M.; Patrickios, C. Polymers Prepared Using Cleavable Initiators: Synthesis,
Characterization and Degradation. Prog. Polym. Sci. 2011, 36 (8), 1079–1097 DOI:
10.1016/j.progpolymsci.2011.01.004.
(16) Bochet, C. Photolabile Protecting Groups and Linkers. Journal of the Chemical Society,
Perkin Transactions 1. 2001, pp 125–142.
(17) Klán, P.; Šolomek, T.; Bochet, C. G.; Blanc, A.; Givens, R.; Rubina, M.; Popik, V.;
Kostikov, A.; Wirz, J. Photoremovable Protecting Groups in Chemistry and Biology:
34
Reaction Mechanisms and Efficacy. Chem. Rev. 2013, 113 (1), 119–191 DOI:
10.1021/cr300177k.
(18) Zhao, H.; Sterner, E. S.; Coughlin, E. B.; Theato, P. O-Nitrobenzyl Alcohol Derivatives:
Opportunities in Polymer and Materials Science. Macromolecules 2012, 45 (4), 1723–
1736.
(19) de Gracia Lux, C.; McFearin, C. L.; Joshi-Barr, S.; Sankaranarayanan, J.; Fomina, N.;
Almutairi, A. A Single UV or Near IR Triggering Event Leads to Polymer Degradation
into Small Molecules. ACS Macro Lett. 2012, 1 (7), 922–926.
(20) Bédard, M. F.; De Geest, B. G.; Skirtach, A. G.; Möhwald, H.; Sukhorukov, G. B.
Polymeric Microcapsules with Light Responsive Properties for Encapsulation and
Release. Adv. Colloid Interface Sci. 2010, 158 (1–2), 2–14 DOI:
10.1016/j.cis.2009.07.007.
(21) Zhu, C.; Bettinger, C. J. Light-Induced Disintegration of Robust Physically Cross-Linked
Polymer Networks. Macromol. Rapid Commun. 2013, 34, 1446–1451 DOI:
10.1002/marc.201300420.
(22) Zhu, C.; Bettinger, C. J. Photoreconfigurable Physically Cross-Linked Triblock
Copolymer Hydrogels: Photodisintegration Kinetics and Structure–Property
Relationships. Macromolecules 2015, DOI: 10.1021/ma502372f DOI:
10.1021/ma502372f.
(23) Zhao, B.; Moore, J. S.; Beebe, D. J. Surface-Directed Liquid Flow inside Microchannels.
Science 2001, 291 (5506), 1023–1026 DOI: 10.1126/science.291.5506.1023.
(24) Il’ichev, Y. V; Schwörer, M. A.; Wirz, J. Photochemical Reaction Mechanisms of 2-
Nitrobenzyl Compounds: Methyl Ethers and Caged ATP. J. Am. Chem. Soc. 2004, 126
(14), 4581–4595 DOI: 10.1021/ja039071z.
(25) Holmes, C. P. Model Studies for New O-Nitrobenzyl Photolabile Linkers: Substituent
Effects on the Rates of Photochemical Cleavage. J. Org. Chem. 1997, 62 (8), 2370–2380
DOI: 10.1021/jo961602x.
(26) Schumers, J.; Gohy, J.; Fustin, C. A Versatile Strategy for the Synthesis of Block
Copolymers Bearing a Photocleavable Junction. Polym. Chem. 2010, 1 (2), 161 DOI:
10.1039/b9py00218a.
(27) Guillier, F.; Orain, D.; Bradley, M. Linkers and Cleavage Strategies in Solid-Phase
Organic Synthesis and Combinatorial Chemistry. Chem. Rev. 2000, 100 (6), 2091–2157
DOI: 10.1021/cr980040+.
(28) Zhou, H.; Lu, Y.; Qiu, H.; Guerin, G.; Manners, I.; Winnik, M. A. Photocleavage of the
Corona Chains of Rigid-Rod Block Copolymer Micelles. Macromolecules 2015, 48 (7),
2254–2262 DOI: 10.1021/acs.macromol.5b00238.
(29) Barltrop, J.A., Plant, P.J., Schofield, P. Photosensitive Protective Groups. Chem.
Commun. 1966, 822.
(30) Wöll, D.; Smirnova, J.; Galetskaya, M.; Prykota, T.; Bühler, J.; Stengele, K. P.; Pfleiderer,
W.; Steiner, U. E. Intramolecular Sensitization of Photocleavage of the Photolabile 2-(2-
Nitrophenyl)propoxycarbonyl (NPPOC) Protecting Group: Photoproducts and
Photokinetics of the Release of Nucleosides. Chem. - A Eur. J. 2008, 14 (21), 6490–6497
DOI: 10.1002/chem.200800613.
(31) Braun, D.; Ritter, H.; Cherdron, H.; Rehahn, M.; Voit, B. Polymer Synthesis: Theory and
Practice; 2005; Vol. 55.
(32) Odian, G. Principles of Polymerization; 2004; Vol. 58.
35
(33) Burdyńska, J.; Cho, H. Y.; Mueller, L.; Matyjaszewski, K. Synthesis of Star Polymers
Using ARGET ATRP. Macromolecules 2010, 43 (22), 9227–9229 DOI:
10.1021/ma101971z.
(34) Burdyńska, J.; Li, Y.; Aggarwal, A. V.; Höger, S.; Sheiko, S. S.; Matyjaszewski, K.
Synthesis and Arm Dissociation in Molecular Stars with a Spoked Wheel Core and
Bottlebrush Arms. J. Am. Chem. Soc. 2014, 136 (36), 12762–12770 DOI:
10.1021/ja506780y.
(35) Jeyaprakash, J. D.; Samuel, S.; Dhamodharan, R.; Rühe, J. Polymer Brushes via ATRP:
Role of Activator and Deactivator in the Surface-Initiated ATRP of Styrene on Planar
Substrates. Macromol. Rapid Commun. 2002, 23 (4), 277–281 DOI: 10.1002/1521-
3927(20020301)23:4<277::AID-MARC277>3.0.CO;2-U.
(36) Bolton, J.; Rzayev, J. Tandem RAFT-ATRP Synthesis of Polystyrene - Poly(methyl
Methacrylate) Bottlebrush Block Copolymers and Their Self-Assembly into Cylindrical
Nanostructures. ACS Macro Lett. 2012, 1 (1), 15–18 DOI: 10.1021/mz200003j.
(37) Xia, J.; Matyjaszewski, K. Controlled /“ Living ” Radical Polymerization . Atom Transfer
Radical Polymerization Using Multidentate Amine Ligands. 1997, 9297 (97), 7697–7700.
(38) Matyjaszewski, K.; Xia, J. Atom Transfer Radical Polymerization. Chem. Rev. 2001, 101
(9), 2921–2990.
(39) Matyjaszewski, K. Atom Transfer Radical Polymerization (ATRP): Current Status and
Future Perspectives. Macromolecules 2012, 45 (10), 4015–4039 DOI:
10.1021/ma3001719.
(40) Matyjaszewski, K.; Spanswick, J. Controlled/living Radical Polymerization. Materials
Today. 2005, pp 26–33.
(41) Braunecker, W. A.; Matyjaszewski, K. Controlled/living Radical Polymerization:
Features, Developments, and Perspectives. Progress in Polymer Science (Oxford). 2007,
pp 93–146.
(42) Zhang, H. Controlled/“living” radical Precipitation Polymerization: A Versatile
Polymerization Technique for Advanced Functional Polymers. European Polymer
Journal. 2013, pp 579–600.
(43) Schmierer, T.; Laimgruber, S.; Haiser, K.; Kiewisch, K.; Neugebauer, J.; Gilch, P.
Femtosecond Spectroscopy on the Photochemistry of Ortho-Nitrotoluene. Phys. Chem.
Chem. Phys. 2010, 12 (48), 15653–15664 DOI: 10.1039/c004025h.
36
Chapter 3. Synthesis & Characterization of UV Cleavable Polymers
3.1 Motivation & Background
Unique functionality in a photocleavable polymer system is directly correlated with
where the oNB group is integrated
1–7
. Therefore, the molecular architecture is tailored to fit the
desired application. For the experiments described in this dissertation, the primary goal was to
create a photodegradable material that is easy to study and that can be applied to many different
types of photoreactive systems due to its simplicity. The chosen architecture consists of a single
orthronitrobenzyl (oNB) unit embedded directly in the center of a homopolymer chain (Figure 3-
1). The oNB placement in the center of the chain is strategic in that the molecular weight is
reduced by half upon photocleavage. Because the molecular weight reduces to a predetermined
size, molecular weight can be used as a metric for reaction progress
8,9
. This molecular
architecture fits our purpose well because oNB photocleavage events can be easily distinguished
and monitored. In addition, by limiting the number of oNB units to one per chain, it is easy to
control the overall oNB concentration in the system by changing the polymer concentration.
Figure 3-1. a) Photocleavage of centrally located oNB results in reduction of molecular weight by half. b)
Linear homopolymer with photocleavable ortho-nitrobenzyl (ONB) group in the center. Upon UV
exposure, this group cleaves to nitrosobenzaldehyde and carboxylic acid.
h ν
a)
b)
NO
2
O
O
NO
O
H
O
HO
O
+
h
O
O
O
37
This chapter discusses the synthetic methods developed to create a polymer with an oNB
unit in the center. The oNB unit was first integrated into a halogenated bifunctional initiator for
ATRP. Reaction conditions were controlled such that polymerization propagated at an equal rate
from both sides. The reaction was terminated with air and Tetrahydrofuran (THF) once the
polymer reached its desired molecular weight. The methods described focus primarily on
synthesis of poly(methyl acrylate) (PMA) as opposed to other monomers because the methyl
acrylate monomer does not exhibit significant absorbance in the UV and therefore will not
interfere with the UV response of the oNB group
10,11
. Details on the synthesis of ATRP with
other monomers (polystyrene, PEGMA, PHEMA) can be found in Appendix A.
In addition to synthetic methods, polymer characterization also plays an important role in
creating the desired polymeric system. This chapter details methods to confirm oNB placement
in the center of the polymer chain as well as characterization of polymer molecular weight and
polydispersity.
3.2 Bifunctional Initiator Synthesis
Our system comprises a linear homopolymer with a photocleavable unit in the center of
the polymer chain (Figure 3-1). Often, functional unit placement in the center of a polymer chain
is achieved through multiple polymerization steps: for example, first polymerizing from an end
of an ATRP initiator to produce a macroinitiator and then addition of a second block to the other
end using a click chemistry reaction
1,5
. This method works well with block copolymers where
each polymer block must be synthesized separately. However, it adds an unnecessary step in the
synthesis of homopolymers. Instead, we can use a single-step, one-pot synthesis with a
bifunctional ATRP initiator to grow the pair of polymers simultaneously from both sides of the
oNB moiety. In this case, ensuring equal rates of propagation from each initiator end is
38
paramount in achieving central oNB placement along the chain. Methods for ensuring controlled
propagation rate will be discussed in further detail in Section 3.3.
As discussed in Chapter 2, ATRP proceeds via repeated halogen transfer between the
initiator and the catalyst. Thus, a bifunctional initiator must have two halogenated points from
which to initiate the reaction. In our reaction, 5-hydroxy-2-nitrobenzyl alcohol (HNBA, 97%,
Sigma Aldrich) was halogenated on both ends in an esterification reaction with excess
bromoisobutyryl bromide (BiBB, 98%, Sigma Aldrich) in dichloromethane (DCM) (Figure 3-
2)
12,13
. Triethylamine (TEA, 99.5%, Sigma Aldrich) was added in excess to trap free bromide in
solution and drive the reaction forward. BiBB was initially added dropwise over ice to minimize
HBr gas formation, and the reaction was slowly warmed to room temperature and stirred for 24-
72 hours. The reaction was also attempted in THF, however this reaction did not yield as much
product. All solvents mentioned in this chapter were ACS Reagent grade (>95%) and from Alfa
Aesar unless otherwise noted.
Figure 3-2. Esterification reaction pathway to produce brominated ONB bifunctional initiator. The molar
ratio of HNBA:TEA:BiBB is 1:4:4.
Longer reaction times worked best for this reaction to ensure that bromination occurred at
both hydroxyl groups. Bromination occurs first at the hydroxyl group closest to the nitro moiety
due to increased reactivity
14,15
. Some of these monofunctionalized products will still remain in
the reaction solution, even when the reaction is driven to completion. As a result, it was
39
necessary to separate out the desired bifunctional compound from the reaction solution using
base extraction and column chromatography. The product was first washed with a 1M HCl
solution to remove excess TEA and then washed with a saturated sodium bicarbonate solution to
remove weakly acidic byproducts as well as excess BiBB. Washing with a more basic solution of
1M NaOH was also tried, however this solution removed much of the desired oNB product.
Washing with multiple cycles of saturated aqueous sodium bicarbonate proved a better method
of removing the weakly acidic byproducts.
The resulting solution was filtered and dried over sodium sulfate. The crude product was
purified by column chromatography over silica gel eluting with a 90:10 hexane/EtOAc mixture.
Multiple cycles of base extraction and column chromatography were required to purify the
compound. The
1
H NMR spectrum in Figure 3-3 illustrates the removal of undesired
monofunctionalized byproduct by the reduction of the peak at 2.01ppm after column
chromatography.
Figure 3-3.
1
H NMR Spectra of synthesized oNB initiator before (top) and after (bottom) purification by
column chromatography.
40
The final ONB bifunctional initiator was a viscous dark amber liquid at 82% yield. The
molecular structure was verified using NMR spectroscopy (Figure 3-4). Peak shifts and integral
ratios agreed with the proposed initiator structure.
1
H NMR (600 MHz, CDCl
3
), δ (TMS, ppm):
8.24 (d, 1H, J = 9Hz, H
Ar
), 7.55 (s, 1H, H
Ar
), 7.29 (d, 1H, J = 9Hz, H
Ar
), 5.64 (s, 2H, −CH
2
−OCO−), 2.06 (s, 6H, CH
3
), 1.98 (s, 6H, CH
3
).
13
C NMR (600 MHz, CDCl ): δ170.8 (C=O),
169.2 (C=O), 154.4 (C
Ar
–O– ), 144.3 (C
Ar
-NO
2
), 134.7 (C
Ar
-CH
2
), 127.1 (H-C
Ar
), 121.3 (H-
C
Ar
), 121.0 (H-C
Ar
), 63.6 (-CH
2
-OCO-), 55.3 (C-Br), 54.7 (C-Br), 30.3 (CH
3
), 30.7 (CH
3
).
Figure 3-4.
1
H NMR spectrum of final bifunctional oNB initiator, post purification in CDCl
3
.
The purified initiator was stored under inert gas and protected from light. Over time,
some initiator batches changed in color and consistency from a dark amber brown viscous liquid
41
to a whitish-yellow opaque solid. The white color resulted from crystallization of reaction
byproducts that occurs as trace amounts of solvent slowly evaporated from the solution.
3.3 Photocleavable Poly(methyl acrylate) Synthesis
Our goal was placement of the ONB unit directly in the center of the final polymer chain;
therefore, the polymer growth must initiate simultaneously from each halogenated initiator end
and propagate at similar rates. Otherwise, the two polymer chains on either side of the initiator
will not be equal
16,17
. Designing this ATRP reaction was especially challenging because the
initiator is asymmetric; the nitro functional group increases reactivity of the adjacent initiator
end
12,14,15
. In order to account for the difference in initiation activities, the polymerization
reagents and conditions were carefully chosen to minimize this difference. Each of the
subsections in this chapter will discuss the optimization of a different aspect of the
polymerization conditions.
The goal of every ATRP reaction is control over the reaction kinetics, because this is
what allows the chemist to stop the reaction at a predetermined molecular weight with a
relatively monodisperse distribution. In general, a well-controlled ATRP reaction progresses via
first order reaction kinetics, i.e. the natural log of monomer conversion is linear with respect to
reaction time
18,19
(Eq 2-4). However, chain growth can deviate from ideal behavior in several
ways
20
(Figure 3-5). For example, if the rate of initiation is too low compared to the rate of
propagation, the molecular weight of growing polymer chains does not increase homogeneously
throughout the solution
21
. Or, if the concentration of propagating radicals grows too large in
relation to the concentration of persistent radicals, bimolecular coupling of active polymer chains
can occur as an undesired termination mechanism
22
. In both cases, the reaction is not considered
42
well-controlled, precluding the ability to predetermine molecular weight. In addition, there will
be a range of molecular weights and the PDI will be high.
Figure 3-5. Variations from standard behavior between monomer conversion and reaction time in
different scenarios. Image adapted from Matyjaszewski, Chem Rev. 2001
18
.
In the former case of slow initiation, it is typically easiest to use a different monomer or
initiator system. Each initiator and monomer possesses its own intrinsic activation rate, thus the
choice of initiator or monomer can be changed so that the initiator radical formation happens
instantaneously in relation to the monomer propagation
16,17
. Alternatively, the propagation rate
can be changed on its own by performing a halogen exchange reaction, which will be discussed
in further detail in Section 3.3.3.
In the latter case of early termination, the reaction conditions can be kept the same and
the reaction can be stopped early (before bimolecular termination occurs). However, the caveat
in stopping the reaction early is that the PDI can often be too high. While molecular weight
progresses linearly with respect to monomer conversion, the PDI does not decrease linearly but
instead depends on the relative rate of deactivation (Figure 3-6)
18,19
. Deactivation increases at
later reaction times (higher conversion) due to the higher concentration of persistent radicals/
monomer
conversion
time
fast
termination
slow initiation
first order
43
polymer product driving equilibrium in the other direction. From the experiments conducted in
this thesis, it was found that ATRP reactions are best stopped between 30-70% conversion, with
60% conversion being ideal in most cases.
Figure 3-6. Schematic representation of the relationship between PDI and conversion. Image adapted
from Matyjaszewski, Chem Rev. 2001
18
.
The following subsections describe the many variations of ATRP reaction conditions that
were attempted in order to obtain a suite of PMA homopolymers with a predetermined molecular
weight range. Each subsection heading indicates which reaction condition is discussed. A table
of all successful reaction conditions (Table 3-2) is found at the end of this section.
3.3.1 General ATRP reaction conditions & experimental procedure
Figure 3-7 shows the general synthetic procedure for the reactions described in this
section. Methyl acrylate monomer (MA, 99%, Sigma Aldrich) was purified by passing through a
basic alumina column to remove inhibitor before polymerization. All other reagents and solvents
were used as received.
Mw/Mn
Conversion
44
Figure 3-7. ATRP of methyl acrylate using bifunctional ONB initiator. Reaction times were varied to
produce n = 50 – 250.
MA, oNB initiator, N, N, N’, N”, N”- pentamethyldiethylenetriamine (PMDETA, 99%,
Sigma Aldrich) and Dimethylformamide (DMF) were added to a dry Schlenk tube. The reaction
mixture was degassed by three freeze−pump−thaw cycles, and CuBr catalyst (99.999%, Sigma
Aldrich) was quickly added to the reaction solution under Argon. After thawing and mixing, the
flask was submerged in a 70 °C oil bath and stirred for various lengths of time under Argon. To
stop the polymerization process, the solution was diluted with Tetrahydrofuan (THF), passed
through a neutral alumina column to remove the catalyst, and precipitated in hexane. To remove
any remaining solvent, the product was dried in a vacuum oven at 35 °C for 12 - 24 h. Resulting
polymers were sticky solids that appeared opaque to translucent white depending on the
molecular weight.
3.3.2 Monomer
Methyl acrylate was selected as the monomer for its low absorbance in the UV.
Additionally, the ratio of the rate of deactivation of methyl acrylate is a benefit; specifically, the
rate of propagation is high enough to allow stable dormant states, but not so high as to hamper
the reaction time
23,24
(Table 3-1).
45
Table 3-1. Rates of propagation and termination for commonly used monomers
23,24
.
Monomer k
p
(M
-1
s
-1
) k
t
(M
-1
s
-1
)
Methyl acrylate 47400 1.1 x 10
8
Methyl methacrylate 1300 9.0 x 10
7
Styrene 665 1.1 x 10
8
The molar ratio of monomer to initiator is a very important parameter in ATRP. This
ratio will dictate the degree of polymerization at each step of conversion. The degree of
polymerization, DP, is linearly related to the monomer: initiator ratio, [M]
0
/ [I], by the equation
𝐷𝑃 =
[𝑀]
9
[𝐼]
× 𝑐𝑜𝑛𝑣𝑒𝑟𝑠𝑖𝑜𝑛 (3-1)
Using this equation, specific molecular weights were obtained by stopping the reaction at a
predetermined conversion value. As mentioned in Section 3.3, the final conversion values should
not be too low (high PDI) or too high (bimolecular coupling).
Initially, the goal was to synthesize a range of polymers M
n
= 5,000 – 500,000 Da or
approximately DP = 60 – 6,000 for methyl acrylate. Each molecular weight was to be obtained
by varying both [M]
0
/ [I] and reaction time. However, it was found that the higher molecular
weight polymers (M
n
> 50,000) were more difficult to synthesize with low PDI because using
high [M]
0
/ [I] often resulted in long reaction times and more susceptibility to undesired side/
termination reactions. Many different [M]
0
/ [I] ratios were tested, and the results can be found in
Table 3-2.
As an example, we can study the case of [M]
0
/ [I] 2000:1, with the goal of stopping the
reaction around 50% conversion for M
n
= 86,000 Da. The reaction was conducted in 10% DMF
solvent with 1:3 monomer: catalyst/ligand ratio. Figure 3-8 shows the GPC trace of the final
46
polymer with a 17.5h reaction time. While the reaction proceeded quicker than predicted, the
final molecular weight (116,000 Da) is not that far from the desired M
n
. However the
polydispersity is fairly large at 1.18, and several shoulder peaks at different molecular weights
can be observed by deconvoluting the main peak using Origin software.
Figure 3-8. GPC trace of polymer ML-03-24 with multiple peaks deconvoluted using a Gaussian fit. Peak
heights were normalized to account for differences in concentration.
The low molecular weight peak is approximately half the value of the main molecular
weight peak, indicating either blocked initiation from one initiator end or the presence of
monofunctional initiator impurities in the starting initiator. The initiator was remade and purified
to a higher degree to resolve this problem. The high molecular weight peak is approximately
double the value of the main peak, indicating bimolecular coupling as a termination mechanism.
This issue was solved two different ways: shrinking the desired molecular weight range to M
n
<
50,000 Da (consequently decreasing conversion and [M]
0
/ [I]) and decreasing the relative
amount of catalyst to shift equilibrium toward the dormant state (discussed in Section 3.3.3). The
47
final suite of polymers varied in molecular weight from 9,000 – 32,000 Da and was synthesized
using [M]
0
/ [I] = 1:800.
3.3.3 Catalyst
The catalytic component in an ATRP reaction is often regarded as the most important
component of the system. This is because the catalyst determines the primary dynamics of
exchange between the active and dormant species, and consequently the reaction equilibrium
25
.
In acrylate polymerizations, the most commonly used catalysts are cuprous halides
26–28
. The
equilibrium constant for each halide follows the opposite trend of Cu-X bond strength. For
example, CuBr exhibits a higher equilibrium constant than CuCl due to the weaker CuBr bond.
The difference in equilibrium constants can be an advantage in reactions where the rates of
initiation and/or propagation need to be manipulated. It is possible to start from a bromo-
terminated chain by using a bromine containing initiator and to switch to a chloro-terminated
chain after initiation by using a CuCl catalyst. This type of reaction, called a halogen exchange
reaction, increases the rate of initiation relative to the rate of propagation
29,30
.
In Figure 3-8, the low molecular weight peak at ½ molecular weight suggests that one
end of the initiator could be failing to undergo initiation, possibly due to the difference in
reactivities resulting from the asymmetric placement of the nitro group. To account for this
inequality, a halogen exchange ATRP reaction was attempted to dramatically increase the rate of
initiation relative to the rate of propagation. This increase could effectively normalize the
difference in reactivities by ensuring rapid initiation through exchange of the more reactive
halogen on the initiator with the less reactive halogen in the catalyst.
In the halogen exchange reaction, CuCl was used as the catalyst instead of CuBr. A
comparison of the molecular weight distributions (Figure 3-9) shows that using CuCl results in a
48
more symmetric distribution than CuBr. However even though the molecular weight distribution
is more symmetric, the PDI using each catalyst was still large at 1.3.
Figure 3-9. Comparison of polymerization results with regular ATRP catalyst (CuBr) and halogen
exchange catalyst (CuCl).
To further investigate the structure of this polymer synthesized via halogen exchange, we
conducted several UV photocleavage experiments. The polymer was dissolved in THF and
placed in a cuvette that was exposed to UV light for different amounts of time. Each time point
sample was analyzed for molecular weight distribution using GPC. Figure 3-10 displays the
molecular weight distribution after the longest exposure times.
49
Figure 3-10. Halogen-exchange-synthesized polymer exposed to various UV irradiation times. Each peak
is labeled with the maximum molecular weight in Da. Peaks heights were normalized to account for
differences in concentration.
Instead of reduction to half the original molecular weight, it curiously reduced by 1/3
instead (Figure 3-10). In addition, there was not a single transition from one molecular weight to
another, but instead a multi-step progressive decrease down to this molecular weight. This slow
degradation indicates that the polymer was slowly cleaving into smaller pieces over time. This
theory is validated by the appearance of small MW (< 1kDa) peaks in the spectra of polymers
exposed to UV light for longer times. It is suggested that the CuCl slowed the propagation rate
too much in relation to the initiation rate, resulting in an overabundance of active radical species
early on in the reaction. Because there were not enough persistent radicals to deactivate these
species, bimolecular coupling of these oligomeric actively propagating chains occurred as a
termination mechanism.
After the halogen exchange proved unsuccessful, we returned to using CuBr catalyst.
This time, we used a 2:1 CuBr: Initiator ratio, instead of 3:1, in an effort to shift the reaction
equilibrium toward the dormant state. We found that by changing the molar ratio of CuBr in the
50
solution, we could slow propagation and achieve more control over the growth rate. The
resulting synthesized polymers did not exhibit as large of a low molecular weight peak than
before.
3.3.4 Solvent
Initially, the reaction was conducted in a 50% (v/v) DMF solution. The resulting
polymers had a bimodal molecular weight distribution, with the lower molecular weight shoulder
(70 kDa) being approximately half the molecular weight of the main peak (125 kDa) (Figure 3-
11). To understand the resulting polymer structure further, we exposed the polymer to UV light
(385nm) in a THF solution for 1 hour and 42 minutes. Following exposure, the GPC
chromatogram exhibited a single peak at 70 kDa (Fig 3-11). The change in peak shape from a
double peak to single peak shows that photocleavage only occurred for one of the original
molecular weights. The low molecular weight shoulder (70 kDa) did not undergo significant
structural changes upon UV exposure, thus ONB photocleavable unit was not anywhere near the
center of that polymer chain. We concluded that there must be a significantly different rate of
propagation from each initiator end, with one end either growing very slowly or not at all.
51
Figure 3-11. GPC chromatogram of polymer synthesized in 50% DMF solvent before (black solid) and
after (red dash) UV exposure at 385nm for 1h 42min.
We then decided to decrease the amount of solvent in solution to either a minimal amount
(10% v/v) or no solvent (bulk polymerization). Resulting polymers presented similar molecular
weight distributions, however the bulk polymerization results were more varied between
different initiator batches. We hypothesize that a small amount of solvent can reduce the effect of
residual water from the initiator on terminating radicals, and thus proceeded with 10% solvent in
subsequent ATRP reactions. Overall, we found that a solvent concentration around 10% by
volume was optimal.
3.3.5 Final Polymerization Conditions
From the previously described experiments, we were able to optimize the ATRP reaction
conditions for solvent ratio, monomer: initiator ratio, reaction time, and choice of catalyst. The
final choice of reaction conditions was M:I:CuBr:L:solvent = 800:1:2:2:10%. We verified that
the monomer conversion followed first order kinetics by sampling the molecular weight at
increasing time points during the reaction (Figure 3-12).
52
Figure 3-12. Kinetics of a single ATRP polymerization reaction with methyl acrylate and oNB initiator
using M:I:CuBr:L:solvent = 800:1:2:2:10%.
The reaction times and molecular weight distributions for the full suite of polymers
synthesized is shown in Figure 3-13. Note that these reaction times differ slightly from those
shown in Figure 3-12. These differences to produce the same molecular weight are likely caused
by inconsistencies in oxygen content (poor degassing) or in initiator ratios. Some of these
reactions were conducted very early on in the project before full control was obtained over these
two parameters. In addition, there were slight differences in activity between initiator batches
resulting from the purification process. The higher molecular weight polymers (32 kDa, 35 kDa,
38 kDa) did exhibit both low and high molecular weight shoulders, but these accounted for a
small percentage of the overall distribution (as evidenced by the low PDI). In addition, these
peaks were accounted for via normalization in the photocleaving kinetics experiments.
53
Figure 3-13. Table (left) lists reaction times and molecular weights for polymers used in kinetic
experiments. Graph (right) shows molecular weight distributions for same suite of polymers, labeled with
molecular weight (Da), PDI.
Table 3-2. Compilation of reaction conditions and molar ratios (with respect to oNB initiator)
for successful methyl acrylate polymerizations using CuBr/ PMDETA. Halogen exchange
reactions and reactions that resulted in multimodal molecular weight distributions are not
included.
a
Reference name in lab notebook.
Name
a
Monomer
Ratio
Catalyst
Ratio
Ligand
Ratio
Solvent
(%)
Time
(h)
M
p
,
GPC
PDI
ML0156 15000 2 2 20 7.3 600000
ML0152 8000 4 4 10 14 400000 1.17
ML0144 2000 3 3 10 16 80000 1.15
ML0337 2000 3 3 10 4 90000 1.04
ML0347 2000 2 2 0 3.3 22000 1.04
ML0175 1000 3 3 10 9 60000
ML0176 1000 3 3 10 8.3 40000 1.16
ML03117 800 2 2 10 5 40100 1.08
ML03119 800 2 2 10 5 38800 1.07
ML0377 800 2 2 10 5 20000 1.06
ML0382 800 2 2 10 5 37000 1.05
ML03108 800 2 2 10 4 41000
ML03124 800 2 2 10 3 43000
ML03106 800 2 2 10 2 35000
ML03129 800 2 2 10 1.5 26000
ML03134 800 2 2 10 1.5 21000 1.04
Time (h) M
n, GPC
M
w
/M
n
0.33 8600 1.08
0.58 14700 1.07
0.75 19400 1.05
1.5 24600 1.04
2.0 32400 1.06
5.0 34900 1.05
4.0 37900 1.09
!
54
ML03126 800 2 2 10 0.75 21000
ML03131 800 2 2 10 0.58 16000 1.07
ML03150 800 2 2 10 0.5 7700 1.22
ML0367 800 2 2 0 4.8 18500 1.06
ML0349 800 2 2 0 2.6 15500 1.06
ML03111 600 2 2 10 0.33 9000
ML0379 400 2 2 10 1.5 15000 1.07
ML0384 400 2 2 10 1.5 18000 1.04
ML0342 400 2 2 0 1.5 13000
ML0344 200 2 2 0 0.75 12000 1.07
3.4 Polymer Characterization
The overall polymer structure was verified using a combination of spectroscopy methods.
First,
1
H NMR was used to verify that the spectrum of the polymer agreed with expected spectra
for poly(methyl acrylate). Initially the spectrum was taken in deuterated chloroform, however
because the polymer did not fully dissolve in this solvent, it was difficult to see the aromatic
nitrobenzyl initiator peaks. The solvent was then changed to deuterated tetrahydrofuran. In the
resulting
1
H NMR spectrum, the largest peaks will correspond to the polymer backbone (PMA).
As shown in Figure 3-14, the four characteristic PMA peaks between 1.5 – 2.5ppm are present
31
.
Additionally, the aromatic peaks from the nitrobenzyl initiator can be clearly seen between 7.0 –
8.5ppm. There is a small peak in the spectrum at 10.81 that likely corresponds to –OH. This
alcohol could result from either cleaved polymer or unpurified monofunctional initiator
impurities.
1
H NMR (600MHz, THF-d8) δ 10.81 (1H, s), 8.18 (1H, d), 7.44 (1H, s), 7.37-7.35
(1H, m), 5.43 (2H, d), 4.37-4.31 (2H, m), 3.61 (3H, s), 2.41-2.28 (1H, m), 1.93-1.85 (1H, m),
1.74-1.63 (1H, m), 1.60-1.44 (1H, m), 1.33-1.27 (6H, m), 1.21-1.15 (6H, m).
55
Figure 3-14.
1
H NMR spectrum of oNB-modified PMA in THF-d
8
. Inset shows magnification of the
spectral region corresponding to aromatic and end group peaks. Solvent peaks were removed from the
spectrum for clarity.
3.4.1 oNB Verification
In addition to overall structure verification, the placement of the oNB was also verified to
be at the center of the polymer chain. This location was validated using
1
H NMR, GPC, and UV-
Vis spectroscopy. First, the polymer was dissolved in a THF solution and exposed to UV light
for an extended period of time (79 hours). After UV exposure, we observed that the molecular
weight reduces to almost exactly half of the original value (Figure 3-15a), indicating that the
oNB unit cleaved in the middle of the polymer chain.
56
Figure 3-15. a) GPC traces of polymer before and after UV exposure for 79 hours. The molecular weight
reduces to half the original value. b) UV-Vis spectra of 38 kDa polymer before (solid) and after (dotted)
UV exposure (350nm, 1hour).
To verify that photocleavage was occurring via the proposed mechanism in Figure 3-1b,
spectroscopy was conducted on photocleaved samples. First, UV-Vis spectra were taken of a
polymer in THF solution before and after 1 hour of UV exposure (Figure 3-15b). The
characteristic ONB absorption peak at 300nm broadens with the formation of new peaks
corresponding to carboxylic acid and aldehyde photoreaction products, whose characteristic
peaks are closer to 350 nm
12,32
. Additionally, a
1
H NMR spectrum was also taken of a polymer in
deuterated THF that was exposed to UV light for 1 hour (Figure 3-16). After UV exposure, a
strong decrease in the ethyl group signal at 5.4 ppm indicates photocleavage of that bond. The
broad spreading and decrease of aromatic peaks between 7.0 – 8.5 ppm indicates conversion of
the original oNB group to other aromatic derivatives, i.e. the nitrosobenzaldehyde and azo dimer
photoproducts.
(b
)
57
Figure 3-16. Comparison of
1
H NMR spectrum of photocleavable PMA in d-THF before (bottom) and
after (top) 1 hour of exposure to UV light.
3.4.2 Fluorescent Emission
Upon photocleavage the oNB group produces multiple photoproducts, including an
azobenzene conjugated structure (Section 2.2.1). Azobenzene compounds, though conjugated,
rarely exhibit fluorescence because they undergo photoisomerization when in the photoexcited
state. Recently, it has been suggested that fluorescence can be induced by modifying the
aromatic structure to prevent photoisomerization
33
. In our samples, we observed bright blue
fluorescent emission when the oNB-PMA solution was excited with 350 nm light (Figure 3-17a).
While we did not modify the polymer to restrict photoisomerization movement, the polymer
structure itself may restrict photoisomerization. Further experimentation needs to be done to
determine the cause of fluorescence. In addition to the solution, fluorescence also occurs in the
58
solid state polymer, although only in larger polymer chunks and not in film (Figure 3-17b). The
lack of visible fluorescence in film is likely due to low concentration of azo units.
Figure 3-17. Photocleaved PMA a) in dichloromethane solution and b) as a solid fluorescing blue after 30
min UV exposure. Non UV exposed samples are next to each sample for comparison.
When observed using a spectrofluorometer, emission is very clearly defined in both
solution and in film after photocleavage by a peak around 420 nm (Figure 3-18). This peak only
appears when excited at 350 nm. When excited at 385 nm, no emission peaks (for film) were
observed. When the ONB group is cleaved alone (without polymer), the emission peak is
observed at 430 nm.
Figure 3-18. Fluorescent emission of PMA solution exposed to UVA for 70 min (left) and PMA thin film
exposed to UVA for 80 min (right).
a)
b)
59
Emission appears to be dependent on the homogeneity of the film at the surface. With
very rough samples, no emission was observed via fluorimetry. A time-dependent study of the
emission was conducted in THF. As the UV exposure time increased, the emission peak became
more defined from the background (Figure 3-19).
Figure 3-19. Fluorescent emission of photocleavable PMA film on Si with increasing amounts of UV
exposure (given in exposure time).
Chapter 3 References
(1) Zhao, H.; Gu, W.; Sterner, E.; Russell, T. P.; Coughlin, E. B.; Theato, P. Highly Ordered
Nanoporous Thin Films from Photocleavable Block Copolymers. Macromolecules 2011,
44 (16), 6433–6440.
(2) Chandra, B.; Subramaniam, R.; Mallik, S.; Srivastava, D. K. Formulation of
Photocleavable Liposomes and the Mechanism of Their Content Release. Org. Biomol.
Chem. 2006, 4 (9), 1730–1740 DOI: 10.1039/b518359f.
(3) Kloxin, A. M.; Tibbitt, M. W.; Kasko, A. M.; Fairbairn, J. a.; Anseth, K. S. Tunable
Hydrogels for External Manipulation of Cellular Microenvironments through Controlled
Photodegradation. Adv. Mater. 2010, 22 (1), 61–66 DOI: 10.1002/adma.200900917.
(4) De Gracia Lux, C.; McFearin, C. L.; Joshi-Barr, S.; Sankaranarayanan, J.; Fomina, N.;
Almutairi, A. Single UV or near IR Triggering Event Leads to Polymer Degradation into
Small Molecules. ACS Macro Lett. 2012, 1 (7), 922–926 DOI: 10.1021/mz3002403.
(5) Zhu, C.; Bettinger, C. J. Light-Induced Disintegration of Robust Physically Cross-Linked
Polymer Networks. Macromol. Rapid Commun. 2013, 34, 1446–1451 DOI:
60
10.1002/marc.201300420.
(6) Schumers, J.; Gohy, J.; Fustin, C. A Versatile Strategy for the Synthesis of Block
Copolymers Bearing a Photocleavable Junction. 2009, 1–10.
(7) Gumbley, P.; Koylu, D.; Pawle, R. H.; Umezuruike, B.; Spedden, E.; Staii, C.; Thomas, S.
W. Wavelength-Selective Disruption and Triggered Release with Photolabile
Polyelectrolyte Multilayers. Chem. Mater. 2014, 26 (3), 1450–1456 DOI:
10.1021/cm403979p.
(8) Rikkou-Kalourkoti, M.; Loizou, E.; Porcar, L.; Matyjaszewski, K.; Patrickios, C. S. End-
Linked, Amphiphilic, Degradable Polymer Conetworks: Synthesis by Sequential Atom
Transfer Radical Polymerization Using a Bifunctional, Cleavable Initiator. Polym. Chem.
2012, 3 (1), 105 DOI: 10.1039/c1py00349f.
(9) Tsarevsky, N. V.; Matyjaszewski, K. Combining Atom Transfer Radical Polymerization
and Disulfide/thiol Redox Chemistry: A Route to Well-Defined (Bio)degradable
Polymeric Materials. Macromolecules 2005, 38 (8), 3087–3092 DOI:
10.1021/ma050020r.
(10) Chiantore, O.; Trossarelli, L.; Lazzari, M. Photooxidative Degradation of Acrylic and
Methacrylic Polymers. Polymer (Guildf). 2000, 41 (5), 1657–1668 DOI: 10.1016/S0032-
3861(99)00349-3.
(11) Fox, R. B.; Isaacs, L. G.; Stokes, S.; Kagarise, R. E. Photodegradation of Poly(methyl
Acrylate). J. Polym. Sci. Part A Polym. Chem. 1964, 2 (September 1962), 2085–2092.
(12) Schumers, J.; Gohy, J.; Fustin, C. A Versatile Strategy for the Synthesis of Block
Copolymers Bearing a Photocleavable Junction. Polym. Chem. 2010, 1 (2), 161 DOI:
10.1039/b9py00218a.
(13) Lee, M. E.; Gungor, E.; Armani, A. M. Photocleavage of Poly(methyl Acrylate) with
Centrally Located O-Nitrobenzyl Moiety: Influence of Environment on Kinetics.
Macromolecules 2015, 48 (24), 8746–8751 DOI: 10.1021/acs.macromol.5b01496.
(14) Cabane, E.; Malinova, V.; Meier, W. Synthesis of Photocleavable Amphiphilic Block
Copolymers: Toward the Design of Photosensitive Nanocarriers. … Chem. Phys. 2010,
211 (17), 1847–1856 DOI: 10.1002/macp.201000151.
(15) Gungor, E.; Armani, A. M. Photocleavage of Covalently Immobilized Amphiphilic Block
Copolymer: From Bilayer to Monolayer. Macromolecules 2016, 49 (16), 5773–5781 DOI:
10.1021/acs.macromol.6b01609.
(16) Tang, W.; Matyjaszewski, K. Effects of Initiator Structure on Activation Rate Constants in
ATRP. Macromolecules 2007, 40 (6), 1858–1863 DOI: 10.1021/ma062897b.
(17) Kwak, Y.; Matyjaszewski, K. Effect of Initiator and Ligand Structures on ATRP of
Styrene and Methyl Methacrylate Initiated by Alkyl Dithiocarbamate. Macromolecules
2008, 41 (18), 6627–6635 DOI: 10.1021/ma801231r.
(18) Matyjaszewski, K.; Xia, J. Atom Transfer Radical Polymerization. Chem. Rev. 2001, 101
(9), 2921–2990.
(19) Matyjaszewski, K. Atom Transfer Radical Polymerization (ATRP): Current Status and
Future Perspectives. Macromolecules 2012, 45 (10), 4015–4039 DOI:
10.1021/ma3001719.
(20) Bergenudd, H.; Jonsson, M.; Malmström, E. Predicting the Limit of Control in the ATRP
Process: Results from Kinetic Simulations. Macromol. Theory Simulations 2011, 20 (9),
814–825 DOI: 10.1002/mats.201100031.
(21) Krys, P.; Ribelli, T. G.; Matyjaszewski, K.; Gennaro, A. Relation between Overall Rate of
61
ATRP and Rates of Activation of Dormant Species. Macromolecules 2016, 49 (7), 2467–
2476 DOI: 10.1021/acs.macromol.6b00058.
(22) Matyjaszewski, K.; Paik, H.; Zhou, P.; Diamanti, S. J. Determination of Activation and
Deactivation Rate Constants of Model Compounds in Atom Transfer Radical
Polymerization 1. Macromolecules 2001, 34 (15), 5125–5131 DOI: 10.1021/ma010185+.
(23) Beuermann, S.; Buback, M. Rate Coefficients of Free-Radical Polymerization Deduced
from Pulsed Laser Experiments. Prog. Polym. Sci. 2002, 27 (2), 191–254 DOI:
10.1016/S0079-6700(01)00049-1.
(24) Barner-Kowollik, C.; Buback, M.; Egorov, M.; Fukuda, T.; Goto, A.; Olaj, O. F.; Russell,
G. T.; Vana, P.; Yamada, B.; Zetterlund, P. B. Critically Evaluated Termination Rate
Coefficients for Free-Radical Polymerization: Experimental Methods. Prog. Polym. Sci.
2005, 30 (6), 605–643 DOI: 10.1016/j.progpolymsci.2005.02.001.
(25) Hong, S. C.; Matyjaszewski, K. Fundamentals of Supported Catalysts for Atom Transfer
Radical Polymerization (ATRP) and Application of an Immobilized/soluble Hybrid
Catalyst System to ATRP. Macromolecules 2002, 35 (20), 7592–7605 DOI:
10.1021/ma020054o.
(26) Matyjaszewski, K.; Nanda, A. K.; Tang, W. Effect of [Cu] on the Rate of Activation in
ATRP. Macromolecules 2005, 38 (Scheme 2), 2015–2018 DOI: 10.1021/ma047531i.
(27) Matyjaszewski, K.; Nanda, A. K.; Tang, W. Effect of [Cu II] on the Rate of Activation in
ATRP. Macromolecules 2005, 38 (5), 2015–2018 DOI: 10.1021/ma047531i.
(28) Tang, H.; Arulsamy, N.; Radosz, M.; Shen, Y.; Tsarevsky, N. V.; Braunecker, W. A.;
Tang, W.; Matyjaszewski, K. Highly Active Copper-Based Catalyst for Atom Transfer
Radical Polymerization. J. Am. Chem. Soc. 2006, 128 (50), 16277–16285 DOI:
10.1021/ja0653369.
(29) Matyjaszewski, K.; Shipp, D. A.; Wang, J.-L.; Grimaud, T.; Patten, T. E. Utilizing Halide
Exchange To Improve Control of Atom Transfer Radical Polymerization. Macromolecules
1998, 31 (20), 6836–6840 DOI: 10.1021/ma980476r.
(30) Peng, C. H.; Kong, J.; Seeliger, F.; Matyjaszewski, K. Mechanism of Halogen Exchange
in ATRP. Macromolecules 2011, 44 (19), 7546–7557 DOI: 10.1021/ma201035u.
(31) Brar, a. S.; Goyal, A. K.; Hooda, S. Two-Dimensional NMR Studies of Acrylate
Copolymers. Pure Appl. Chem. 2009, 81 (3), 389–415 DOI: 10.1351/PAC-CON-08-06-
01.
(32) Zhao, H.; Sterner, E. S.; Coughlin, E. B.; Theato, P. O-Nitrobenzyl Alcohol Derivatives:
Opportunities in Polymer and Materials Science. Macromolecules 2012, 45 (4), 1723–
1736.
(33) Yoshino, J.; Kano, N.; Kawashima, T. Fluorescent Azobenzenes and Aromatic Aldimines
Featuring an N-B Interaction. Dalton Trans. 2013, 42 (45), 15826–15834 DOI:
10.1039/c3dt51689j.
62
Chapter 4. Photocleavage Kinetics Studies
Photodegradable materials have gained widespread popularity in recent years due to the
increasing demand for trigger-responsive or “smart” materials. They have been utilized across a
wide variety of applications, for example in triggered payload release
1–5
, microscale surface
patterning
6–9
, and tunable hydrogel matrices
10–15
. In most of these applications, light of a
predetermined wavelength is applied to the sample until photodegradation is observed. Yet little
research has been conducted in investigating the photodegradation rate. Determining what
factors affect the photodegradation rate could help optimize the light intensities and exposure
times needed to observe photodegradation. In addition, it would open up a new set of purely
kinetics-based applications – one example of which is a wearable UV sensor described in
Chapter 5. In this chapter, we take a closer look at the photodegradation kinetics using a model
system based on oNB photocleavage. We present a set of experiments that probe the effects of
solvent environment and polymer molecular weight on the photocleavage reaction.
Previously, it has been observed that the rate of oNB photocleavage is dependent on its
benzylic substituents and solvent environment
16–18
. By modulating these variables, hydrogen
abstraction associated with photocleavage can be accelerated. However, these studies were
conducted on the oNB molecule in isolation and may not accurately represent the trends
observed for oNB photocleavage when it is integrated into a polymer. In conducting our studies,
we expected that the interaction between the polymer and its environment would have an effect
on the oNB unit photocleavage rate.
The interaction between a polymer and the surrounding solvent is governed by the
change in Gibbs free energy when mixing the two components. This change is directly correlated
63
to the difference in chemical potential of the solvent after mixing with the polymer
19
. This
difference can be expressed as the polymer-solvent interaction parameter, c
sp
, and defined as
𝜒
F,
=
(HIH
5
)
K
L
0
M
N'
(4-1)
where µ is the chemical potential of the solvent, f
p
is the volume fraction of polymer in the
solution, R is the gas constant and T is temperature. c
sp
values can be used to define three general
types of solvents: good solvents, theta solvents, and poor solvents (Figure 4-1). In general,
solvents with smaller c
sp
values (c
sp
< 0.5) are better at fully dissolving polymers. The low
energetic cost of mixing allows for more interactions between the polymer and solvent, leading
to a stretched or expanded chain conformation in solution. c
sp
= 0.5 is the defining value
separating good solvents from poor solvents. When c
sp
~ 0.5, there are minimal interactions
between the polymer and solvent and the polymer coil can act as an ideal chain in solution. This
conformation is modeled as a random walk coil, constrained only by chain self-avoidance. When
c
sp
> 0.5, interactions between the polymer and itself are preferred and the polymer coil contracts
to form a shrunken conformation.
Figure 4-1. Illustration of polymer chain conformations in solvents with different polymer-solvent
interaction parameters (c).
χ
< 0.5
χ
= 0.5
χ
> 0.5
good solvent solvent poor solvent θ
64
In some polymer-solvent pairs, c
sp
can have a negative value. Physically, the negative
value indicates that there are net attractive interactions instead of repulsive interactions between
the polymer and solvent. Other factors that can affect c include the polymer molecular weight,
concentration in solution, and solution temperature. In general, c increases with increasing
concentration and temperature
20
. The relationship between c and molecular weight is less clearly
defined, with the exception of very high molecular weights that change the solution viscosity.
In this study, we characterized the role of the environment and the polymer molecular
weight (MW) on the photocleavage behavior of an oNB subunit. The oNB unit was incorporated
into the center of a linear homopolymer chain as described in Chapter 3. The photocleavage rates
of the polymer both in solution and in film were measured upon exposure to an ultraviolet light
(350nm) source, and a significant difference in photocleavage rates and in the MW-dependence
was observed. The behavior modeled in this ideal system provides a framework for
understanding the kinetic behavior of more complicated photo-responsive polymer networks.
4.1 Experimental Set-Up and Sample Preparation
We exposed samples to light using a Luzchem ICH-2 photoreactor with temperature
control and a rotating sample stage (Figure 4-2). The temperature was set to 22°C although
sometimes rose up to 27°C because the internal fan could not cool the chamber fast enough.
Liquid samples were placed in a quartz cuvette in the center of the stage and illuminated with 4
bulbs from each side. Solid samples were exposed with 8 total bulbs directly from the top.
Hitachi FL8BL-B and Sylvania cool white fluorescent bulbs were used for UV and visible
exposure experiments respectively.
65
Figure 4-2. Schematic of photoreactor set-up for liquid (left) and solid (right) samples.
While the bulb emission spectrum was fairly broad across the UVA range, the highest
peak centered around 350nm. A Smart Sensor digital lux meter measured typical intensities
between 4 – 6 mW/cm
2
at the exposure positions. These intensities are close to typical exposure
intensities on a very sunny day in Southern California, making these experiments relevant for
future real-world applications.
Liquid samples were prepared by dissolving polymer in the chosen solvent at 0.3mM.
Solid samples were prepared via spin coating a 10mg/mL polymer in THF solution onto a
polished silicon wafer with a 1µm thermal oxide layer. Spin coating parameters were optimized
at 4000 rpm for 1 minute to produce smooth polymer films. The films were dried on a hot plate
at 50°C for 5 minutes and then a vacuum oven at 60°C for 1 hour. Film thicknesses were
measured as approximately 900nm using a Gaertner Variable Angle Stokes Waferskan
ellipsometer.
4.2 Data Analysis
4.2.1 Kinetic Constants
As mentioned in the dose-response background theory in Section 2.3, it is expected that
the photocleavage kinetics will follow first order kinetic behavior. First order behavior dictates
that the reactant concentration, [R], will decrease exponentially with time, or
66
𝑅 =[𝑅]
9
𝑒
I/O
(4-2)
From this equation, one could assume that cleaved product formation follows the same trend: as
reactant (uncleaved polymer) is used up, product (cleaved polymer) concentration increases
exponentially with time. However, that is not necessarily the case. Because oNB photocleavage
is not the only process that occurs in the reaction solution, we cannot assume that cleaved
product formation is only governed by the decrease in reactant concentration. Side processes
such as photoinduced dimerization of nitrosobenzaldehyde groups (Section 2.1.2) can also occur
simultaneously with oNB photocleavage, affecting the overall cleaved product concentration.
Thus, we modified the kinetic equation by adding another exponential term to account for
additional processes that affect photocleaved product (P) formation:
𝑃 =[𝑃]
9
+𝐴
R
𝑒
I/
+
O
+𝐴
S
𝑒
I/
M
O
(4-3)
where [P] is the concentration of photocleaved polymer chains, k
1
& k
2
are the normalized rate
constants of photocleavage and side processes respectively, and A
1
& A
2
are environment
specific constants.
In quantifying and comparing reaction kinetics between data sets, we were primarily
interested in changes in the reaction rate constant k. Because our primary kinetic equation (Eq 4-
3) contains multiple rate constants for each system, we can determine information about potential
multi-step kinetic mechanisms. If the reaction is instead single-step, the rate constants k
1
and k
2
will be equal to each other. In Equation 4-3, the rate constant value has been normalized to apply
regardless of experimental set-up (i.e. UV exposure in solution vs in film). While typical rate
constants are in units of time
-1
, the normalized rate constants determined in our experiments have
units of m
2
/ (kW s) to account for variation in UV lamp exposure intensity.
67
4.2.2 Monitoring Photocleavage Reaction Progress
The concentration of cleaved polymer chains at any given time can be measured a
number of ways. As mentioned in the background section, the most common method of
monitoring photoreaction progress in the literature is comparing the UV-Vis absorbance spectra
of the polymer after different UV exposure times
21,22
. We have explored this method and found it
to be ineffective, particularly for monitoring the photocleavage kinetics as a solid film (Section
4.3.6). Instead, we determined the concentration of cleaved polymer chains in the sample using
Gel Permeation Chromatography (GPC). Typically, GPC is used as a method of separating
polymers and oligomers based on their relative hydrodynamic radius in the eluent solvent.
Following separation, the presence of each group of differently sized polymer chains is detected
by changes in light scattering, refractive index, or absorption. Our GPC system is equipped with
four Waters Styragel separation columns (HR5E, HR4E, HR4, HR1) in THF and a refractive
index detector. The detector signal is roughly proportional to the relative amount of the polymer
in the sample solution. Using GPC, we can determine the concentration of cleaved polymer by
finding the relative concentrations of full molecular weight (uncleaved) and half molecular
weight (cleaved) polymer chains. These concentrations are represented by the corresponding
peak heights. Photocleavage process was defined by ratio of cleaved to uncleaved polymer
chains in the sample. This value, % cleaved was represented in the GPC data as
% 𝑐𝑙𝑒𝑎𝑣𝑒𝑑 =
X
YZ[\][^
X
YZ[\][^
_X
`(YZ[\][^
(4-4)
where h
cleaved
and h
uncleaved
represent the relative peak heights corresponding to cleaved and
uncleaved polymer molecular weight respectively.
To illustrate, we can look at the progression of a sample GPC chromatogram over
different UV light exposure times. Figure 4-3 shows the GPC chromatogram of an oNB PMA
68
sample after 0 hours (left), 1 hour (center), and 23 hours of UV exposure (right). The multipeak
data was fit to a bi-Gaussian distribution to determine the height of each individual peak. The fits
shown in blue and red represent the cleaved and uncleaved polymer respectively. Each GPC
chromatogram was normalized by dividing by the maximum to account for slight variation in
polymer concentration.
Figure 4.3. GPC Chromatograms of an oNB-PMA polymer sample in THF as it is exposed to UV light
for varying times. The exposure time is shown in the upper right. Blue and red dotted fits represent
cleaved and uncleaved polymer respectively.
Before any UV exposure, there was a large peak at 19 kDa corresponding to the
uncleaved polymer molecular weight. We took the height of this peak (approximately 0.986) to
represent the relative concentration of uncleaved polymer in the sample. Additionally, there was
a small peak at 10 kDa corresponding to the cleaved polymer molecular weight. The height of
this peak was 0.035. Using Eq 4-4, the % cleaved at this time point is 3.43%. However, because
this is the zero time point, we used this % cleaved value to normalize the following time points.
We continued our kinetic study for multiple time points until the ratio of peak heights no longer
changed, which for this polymer was 23 hours. Most samples reached at least 95% cleaved by
0 h
1 h 23 h 0 h
69
this time point. The remaining uncleaved percentage could be attributed to dimerization of the
cleaved polymer chains falsely appearing to be uncleaved polymer.
Photocleavage reaction progress (% cleaved) was then plotted against an intensity-
normalized time value that we defined as Radiant Exposure. We found that all data sets fit well
to the kinetic model presented in Equation 4-4 with R
2
>99%. From each experimentally
determined fit, we obtained the rate constants k
1
& k
2
for comparison across data sets.
4.2.3 Quantum Efficiency
In addition to reaction kinetics, we also sought to investigate the quantum efficiency, f,
of the reaction. The quantum efficiency represents how effectively incoming light is used by the
oNB moiety in completing the photocleavage reaction. f was quantitatively determined by
computing the concentration of photocleavage products formed per input photon of UV light. f
can be represented by
𝜙 =
b
YZ[\]
b
0cde
=
Nf∙g/f
ij5
%klmno∙b
p(pe
(4-5)
where m
cleav,
m
phot
, m
init
are the # of moles of cleaved product, photons, and initial polymer, RE is
radiant exposure, A is exposure area, and E
350
is the energy of a 350nm photon.
4.3 Photocleavage Kinetics Experimental Results
In our kinetics experiments, we investigated the effects of changing three major
variables: 1) molecular weight, 2) solvent-polymer interaction, and 3) physical environment. For
the first variable, we synthesized a suite of different polymer molecular weights ranging from 9 –
38 kDa. For the second variable, we utilized three different solvents of varying polarity and
polymer-solvent interaction with PMA. The solvents were toluene (nonpolar, c
12
=0.624),
tetrahydrofuran (polar aprotic, c
12
=0.425), and chloroform (nonpolar, c
12
=-0.222). For the third
70
variable, we conducted experiments on polymers both in solution and spin coated onto silicon
wafers as thin polymer films.
4.3.1 Photocleavage in Solution
As expected, significantly more of the polymer was cleaved in solution with increasing
exposure time. The molecular weight distributions of the uncleaved and cleaved peaks were
similar in PDI, indicating that polymer did not degrade under UV light and only cleaved at the
oNB central junction. We then created a dose-response curve by compiling the % cleaved
measurements at a range of UV exposure time points (Figure 4-4). The experimental data was
sigmoidal in general, and fit well to a multi-exponential model as presented in Eq 4-4. The
kinetics curve can be split into three general sections, each with a different linear slope. The first
slope change was associated with the onset of photocleavage (%cleaved > 1%) and typically
occurred between 2-5 minutes. The second slope change signified that photocleavage was no
longer the dominant process in solution and typically occurred after 3 hours.
Figure 4-4. Progression of molecular weight distribution for 38 kDa polymer in THF exposed to
increasing doses of UV irradiation. Left: GPC traces for polymer exposed to 1min (top), 35min (middle),
and 30h 52min (bottom) UV light. Right: Reaction progress plotted as % cleaved and corresponding fit.
71
Photocleavage progress was plotted for all the polymer molecular weights (9 – 38 kDa)
on a single graph for comparison (Figure 4-5). In general, we observed that the photocleavage
kinetics did not vary significantly with respect to molecular weight. The rate constants and the
photocleavage onset/ decline times were approximately constant across the molecular weight
range tested. In Figure 4-5 it may appear as though the lower molecular weight polymers (9k,
15k) cleave slightly slower than polymers of higher molecular weight, however this is not the
case. This effect is a product of the processing method; multipeak GPC chromatograms are more
difficult to deconvolute when the peaks are closer together, which is the case for peaks
corresponding to molecular weight <10 kDa. Because both the 9 kDa and 15 kDa polymer GPC
data sets include polymer molecular weights <10 kDa, the % cleaved values for these two data
sets are moderately affected.
Figure 4-5. Photocleavage reaction progress for the full range of polymer molecular weights cleaved in
THF and their corresponding fits. Key on the top left describes polymer molecular weight in Da.
We also compared the data sets from photocleavage in different solvents, and observed a
trend relating to the solvent-polymer interaction parameter, c
12
. Figure 4-6 shows a comparison
72
of the photocleavage progress for the same polymer cleaved in three different solvents. From this
graph, it is apparent that both the onset of photocleavage and the linear rate of the main
photocleavage region differ between solvents. As the c
12
increases, photocleaavage both begins
and continues to progress fastest (with the least UV radiant exposure) in chloroform as compared
to THF and toluene. This difference is further quantified in Section 4.3.3 when comparing the
kinetic constants of the reaction in each solvent. Polymers of different molecular weights were
also cleaved in each of the three solvents. Similar results were obtained to the THF experiments
discussed previously and no correlation between molecular weight and photocleavage kinetics
was observed.
Figure 4-6. Photocleavage reaction progress and corresponding fits of polymer in chloroform,
tetrahydrofuran, and toluene.
4.3.2 Photocleavage in Film
A similar series of experiments were performed for samples in film. Polymers of each
molecular weight were spin coated onto polished silicon wafers and dried in the oven, resulting
73
in solid thin films approximately 900nm in thickness. Figure 4-7 contains the results from the 38
kDa polymer thin film to enable direct comparison with Figure 4-4. As the thin film polymer is
cleaved under UV irradiation, the molecular weight reduces by half without a change in the PDI.
Figure 4-7. Progression of molecular weight distribution for 38 kDa polymer exposed to increasing doses
of UV irradiation as a thin film. Left: GPC traces for polymer exposed to 46 seconds (top), 8 minutes
(middle), and 54 minutes (bottom) of UV irradiation. Right: Reaction progress plotted as % cleaved and
corresponding fit. Hollow points were not included in the fit.
The photocleavage reaction progress in film adheres to the multiexponential fit for the
majority of the reaction, but appears to regress at long radiant exposure times (hollow squares in
Figure 4-7). Because the photocleavage reaction itself is not reversible, the apparent regression
must be due to a different reaction involving photocleaved products. As mentioned in the
introduction, o-nitrosobenzaldehyde can react with itself upon UV excitation to form azobenzene
dimers
23–25
. The molecular weight of the dimerized polymer chains is almost identical to that of
the uncleaved polymer, and cannot be distinguished from the latter in GPC chromatograms.
Therefore, the decrease observed in the hollow data points in Figure 4-7 is an artifact and the %
74
cleaved can no longer be calculated by simply evaluating the ratio of the two peaks.
Accordingly, the fit was made without these three points.
The dimerization occurs more prominently in polymer films than in a solution because of
the restricted mobility of a radical scavenging additive: butylated hydroxytoluene (BHT). BHT is
typically added to THF solvents as a preservative that prevents peroxide formation. In our
experiments, BHT also serves as a radical scavenger, preventing dimerization reactions in many
cases
26
. BHT was present in all the solvent samples and in the film samples, however has less
mobility in film because there is no solvent to facilitate diffusion. As a result, dimerization
occurs at an accelerated rate in film.
Similar results and fits for thin film photocleavage kinetics were obtained for all of the
other polymers with varying molecular weights, and are shown in Figure 4-8. All of the
polymers tested exhibited either a decrease or plateau in the % cleaved after long UV exposure
times. This trend reflects the dimerization that is occurring in all of the polymer thin film
samples. In light of this secondary photo-initiated process, it is not possible to determine the
final % cleavage in film. However, by observing the trends, it is clear that 100% photocleavage
is not likely. This decrease in efficiency with respect to the results in solution is likely due to the
formation of an optical protecting layer comprising cleaved reaction products and azobenzene
dimers. These photoproducts absorb light near the exposure wavelength range, attenuating the
incoming UV light and decreasing its penetration efficacy throughout the entire film
thickness
22,27,28
.
75
Figure 4-8. Photocleavage reaction progress for the full range of polymer molecular weights cleaved as a
solid thin film and their corresponding fits. Key on the top left describes polymer molecular weight in Da.
4.3.3 Photocleavage in Solution vs. Film
In Figure 4-9, the photocleavage reaction progress is plotted for the same polymer in
THF solution and as a thin film for direct comparison. The radiant exposure values were
normalized by UV light intensity to account for the differences in UV lamp experimental set up
between the two experiments. It is immediately apparent that both the onset of photocleavage
and the rate of photocleavage are accelerated in the film sample. Photocleavage begins and
progresses with less incident UV exposure when the polymer is in the form of a solid film. This
result was expected because there is no solvent in a film sample to attenuate incoming UV light.
Therefore, the photocleavage process was more efficient overall with respect to UV exposure.
76
Figure 4-9. Comparison of photocleavage process of 38 kDa polymer in THF solution and as a thin film
spin coated onto a silicon wafer.
4.3.4 Photocleavage Rate Constants
A more quantitative comparison between the photocleavage kinetics of polymer in
different environments can be made by calculating the rate constants for the reaction (Eq 4-3).
These constants were calculated using the multiexponential fits of each data set. Figures 4-9 and
4-10 show the rate constants for photocleavage reactions compared across polymer-solvent
interaction parameters and molecular weights. Interestingly, the two rate constants k
1
and k
2
exhibit different trends with respect to c
12
. k
1
does not appear to have a clear dependence on c
12
,
whereas k
2
exhibits an inverse relationship with c
12
(Figure 4-10). Because k
1
and k
2
represent
different photoinitiated processes (or sets of processes) in the sample solution, we can conclude
that one process is solvent interaction-dependent while the other is not. However, discerning
what physical processes these rate constants represent is more difficult and would require an in-
depth spectroscopic analysis of the photoinduced reaction processes occurring in real time – an
experiment that we are not equipped to conduct. What we can conclude from our data is that
77
there appears to be a single rate constant for photocleavage of oNB-PMA in THF and toluene,
while there are multiple rate constants for photocleavage in chloroform. This trend does not
relate to solvent polarity, as both chloroform and toluene are nonpolar solvents, but instead
appears to relate to the nature of polymer-solvent interaction. Chloroform exhibits a negative
interaction parameter with PMA, hinting that strong polar interactions between the polymer and
solvent may play a role in increasing the rate of photocleavage. This trend is further explored as
it relates to quantum efficiency in Section 4.3.5.
Figure 4-10. Dependence of normalized rate constants k
1
and k
2
on the polymer-solvent interaction
parameter.
Figure 4-11 shows the dependence of kinetic rate constants on molecular weight for both
the solution and film samples. Across all data sets, there was no clear trend between polymer
molecular weight and photocleavage rate constant value. The rate constants of cleaving as a
polymer film were consistently larger than those for cleaving in solution, confirming the visual
toluene chloroform THF
Interaction Parameter
78
trend that we qualitatively observed in the multi-exponential fits discussed earlier. Cleaving in
film exhibited very similar k
1
and k
2
values, indicating that there was a single dominant
photocleaving process for the majority of the reaction. This was also the case for cleaving in
THF and in toluene. However, the k
1
& k
2
values for cleaving in chloroform exhibited a different
trend. The k
2
values for cleaving in chloroform were very similar to the rate constants for
cleaving as a polymer film, suggesting that some photoinitiated processes of oNB-PMA in
chloroform are similar to those in film.
Figure 4-11. Dependence of normalized rate constants k
1
(solid) and k
2
(hollow) on polymer molecular
weight.
4.3.5 Trends in Quantum Efficiency
Quantum efficiency (f) values for photocleavage in each solvent were calculated
according to Equation 4-5 and are plotted according to polymer molecular weight in Figure 4-12.
f was not calculated for photocleavage in film because we did not know the polymer density as a
film and thus were not able to estimate the total exposure area. The results shown in Figure 4-12
Molecular Weight (kDa)
toluene chloroform THF fi lm
79
are similar to the trends exhibited in the photocleavage kinetic experiments. Across all solvents,
there is not a strong relationship between polymer molecular weight and quantum efficiency.
However, there is a twofold increase in the quantum efficiency of photocleavage in chloroform
over photocleavage in toluene or THF. We hypothesize that this increase may be due to strong
polar interactions between PMA and chloroform that allow photons to more effectively interact
with the oNB unit.
Figure 4-12. Dependence of quantum efficiency of photocleavage in solution on solvent type and
molecular weight.
4.3.6 Comparison with UV-Vis Absorbance Methods
The typical method of tracking photocleavage progress is by monitoring the change in the
UV absorbance spectrum. We conducted a study to compare the kinetic results obtained using
the UV-Vis absorbance method against the GPC method described previously. A 25 kDa oNB
PMA polymer was dissolved in THF and exposed to increasing amounts of UV exposure. Figure
80
4-13a shows the progression of the UV-Vis spectrum of the same polymer solution sample with
increasing UV exposure time. To obtain a kinetics curve from the absorption spectrum, we first
calculated the absorbance values at a fixed wavelength for different UV exposure times. We
chose 350 nm as the fixed wavelength because o-nitrosobenzaldehyde absorbs light near this
value. We plotted the changes in absorbance with increasing UV exposure and overlaid the
changes in % cleaved calculated using GPC for comparison (Figure 4-13b).
As expected, the absorbance peak shifts to longer wavelengths as more photocleaved o-
nitrosobenzaldehyde product is formed. However after long exposure times, there is an
unexpected result. The absorption begins to decrease after exposure times exceeding
approximately 5 hours. The decrease of this peak can be attributed to secondary photoinitiated
reactions of photocleaved products. As these secondary reactions occur, the concentration of o-
nitrosobenzaldehyde photoproduct is decreased due to their participation in the reactions. This
decrease in concentration results in a decrease of absorbance near 350 nm. The absorption
decrease is only observed at long exposure times, because photocleavage rates slow down at
these times and there are fewer o-nitrosobenzaldehyde photoproducts concurrently being formed.
When comparing the kinetic results obtained using GPC and UV-Vis, the progression of
photocleavage appears very similar initially but deviates significantly at long UV exposure
times. The sharp decrease in absorbance values at large radiant exposures illuminates a weakness
of using the changes in absorbance to monitor photocleavage kinetics. The absorbance spectrum
is much more sensitive to secondary photoinitiated reactions than the GPC, making it difficult to
accurately record the photoproduct formation at long UV exposure times. Thus, the GPC is better
suited to examine the entire photocleavage reaction while the UV-Vis has limitations at long
exposure times.
81
Figure 4-13. Comparison of photocleavage kinetics measured by change in absorption vs. GPC %
cleaved. a) Evolution of the UV-Vis absorbance spectrum for 25 kDa polymer cleaved in THF. Key
shows UV exposure time. b) % cleaved obtained by GPC plotted with change in absorbance at 350nm.
Chapter 4 References
(1) Chandra, B.; Subramaniam, R.; Mallik, S.; Srivastava, D. K. Formulation of
Photocleavable Liposomes and the Mechanism of Their Content Release. Org. Biomol.
Chem. 2006, 4 (9), 1730–1740 DOI: 10.1039/b518359f.
(2) Esser-Kahn, A. P.; Odom, S. a.; Sottos, N. R.; White, S. R.; Moore, J. S. Triggered
Release from Polymer Capsules. Macromolecules 2011, 44 (14), 5539–5553 DOI:
10.1021/ma201014n.
(3) Wirkner, M.; Alonso, J. M.; Maus, V.; Salierno, M.; Lee, T. T.; García, A. J.; del Campo,
A. Triggered Cell Release from Materials Using Bioadhesive Photocleavable Linkers.
Adv. Mater. 2011, 23 (34), 3907–3910 DOI: 10.1002/adma.201100925.
(4) Gumbley, P.; Koylu, D.; Pawle, R. H.; Umezuruike, B.; Spedden, E.; Staii, C.; Thomas, S.
W. Wavelength-Selective Disruption and Triggered Release with Photolabile
Polyelectrolyte Multilayers. Chem. Mater. 2014, 26 (3), 1450–1456 DOI:
10.1021/cm403979p.
(5) Bédard, M. F.; De Geest, B. G.; Skirtach, A. G.; Möhwald, H.; Sukhorukov, G. B.
Polymeric Microcapsules with Light Responsive Properties for Encapsulation and
Release. Adv. Colloid Interface Sci. 2010, 158 (1–2), 2–14 DOI:
10.1016/j.cis.2009.07.007.
(6) García-Fernández, L.; Herbivo, C.; Arranz, V. S. M.; Warther, D.; Donato, L.; Specht, A.;
Del Campo, A. Dual Photosensitive Polymers with Wavelength-Selective Photoresponse.
Adv. Mater. 2014, 26 (29), 5012–5017 DOI: 10.1002/adma.201401290.
(7) Zhao, B.; Moore, J. S.; Beebe, D. J. Surface-Directed Liquid Flow inside Microchannels.
Science 2001, 291 (5506), 1023–1026 DOI: 10.1126/science.291.5506.1023.
a) b)
82
(8) Cui, J.; Miguel, V. S.; Del Campo, A. Light-Triggered Multifunctionality at Surfaces
Mediated by Photolabile Protecting Groups. Macromol. Rapid Commun. 2013, 34 (4),
310–329 DOI: 10.1002/marc.201200634.
(9) Zhao, H.; Gu, W.; Sterner, E.; Russell, T. P.; Coughlin, E. B.; Theato, P. Highly Ordered
Nanoporous Thin Films from Photocleavable Block Copolymers. Macromolecules 2011,
44 (16), 6433–6440.
(10) Pasparakis, G.; Manouras, T.; Argitis, P.; Vamvakaki, M. Photodegradable Polymers for
Biotechnological Applications. Macromol. Rapid Commun. 2012, 33 (3), 183–198 DOI:
10.1002/marc.201100637.
(11) Kloxin, A. M.; Tibbitt, M. W.; Anseth, K. S. Synthesis of Photodegradable Hydrogels as
Dynamically Tunable Cell Culture Platforms. Nat. Protoc. 2010, 5 (12), 1867–1887 DOI:
10.1038/nprot.2010.139.
(12) DeForest, C. a.; Anseth, K. S. Cytocompatible Click-Based Hydrogels with Dynamically
Tunable Properties through Orthogonal Photoconjugation and Photocleavage Reactions.
Nat. Chem. 2011, 3 (12), 925–931 DOI: 10.1038/nchem.1174.
(13) Kloxin, A. M.; Tibbitt, M. W.; Kasko, A. M.; Fairbairn, J. a.; Anseth, K. S. Tunable
Hydrogels for External Manipulation of Cellular Microenvironments through Controlled
Photodegradation. Adv. Mater. 2010, 22 (1), 61–66 DOI: 10.1002/adma.200900917.
(14) Zhu, C.; Bettinger, C. J. Light-Induced Disintegration of Robust Physically Cross-Linked
Polymer Networks. Macromol. Rapid Commun. 2013, 34, 1446–1451 DOI:
10.1002/marc.201300420.
(15) Griffin, D. R.; Kasko, A. M. Photoselective Delivery of Model Therapeutics from
Hydrogels. ACS Macro Lett. 2012, 1 (11), 1330–1334 DOI: 10.1021/mz300366s.
(16) Holmes, C. P. Model Studies for New O-Nitrobenzyl Photolabile Linkers: Substituent
Effects on the Rates of Photochemical Cleavage. J. Org. Chem. 1997, 62 (8), 2370–2380
DOI: 10.1021/jo961602x.
(17) Hasan, A.; Stengele, K.; Giegrich, H. Photolabile Protecting Groups for Nucleosides:
Synthesis and Photodeprotection Rates. Tetrahedron 1997, 53 (12), 4247–4264.
(18) Bochet, C. Photolabile Protecting Groups and Linkers. Journal of the Chemical Society,
Perkin Transactions 1. 2001, pp 125–142.
(19) Orwoll, R. a; Arnold, P. a. Polymer – Solvent Interaction Parameter X. Phys. Prop.
Polym. Handb. 2007, 50 (3), 451 DOI: 10.1007/978-0-387-69002-5_14.
(20) Rubinstein, M.; Colby, R. H. Polymer Physics. Polymer International. 2003, p 440.
(21) Kuckling, D.; Harmon, M.; Frank, C. Photo-Cross-Linkable Pnipaam Copolymers. 1.
Synthesis and Characterization of Constrained Temperature-Responsive Hydrogel.
Macromolecules 2002, 6377–6383.
(22) Schumers, J.; Gohy, J.; Fustin, C. A Versatile Strategy for the Synthesis of Block
Copolymers Bearing a Photocleavable Junction. Polym. Chem. 2010, 1 (2), 161 DOI:
10.1039/b9py00218a.
(23) Guillier, F.; Orain, D.; Bradley, M. Linkers and Cleavage Strategies in Solid-Phase
Organic Synthesis and Combinatorial Chemistry. Chem. Rev. 2000, 100 (6), 2091–2157
DOI: 10.1021/cr980040+.
(24) Zhao, H.; Sterner, E. S.; Coughlin, E. B.; Theato, P. O-Nitrobenzyl Alcohol Derivatives:
Opportunities in Polymer and Materials Science. Macromolecules 2012, 45 (4), 1723–
1736.
(25) Zhou, H.; Lu, Y.; Qiu, H.; Guerin, G.; Manners, I.; Winnik, M. A. Photocleavage of the
83
Corona Chains of Rigid-Rod Block Copolymer Micelles. Macromolecules 2015, 48 (7),
2254–2262 DOI: 10.1021/acs.macromol.5b00238.
(26) Fujisawa, S.; Kadoma, Y.; Yokoe, I. Radical-Scavenging Activity of Butylated
Hydroxytoluene (BHT) and Its Metabolites. Chem. Phys. Lipids 2004, 130 (2), 189–195
DOI: 10.1016/j.chemphyslip.2004.03.005.
(27) Beharry, A. A.; Sadovski, O.; Woolley, G. A. Azobenzene Photoswitching without
Ultraviolet Light. J. Am. Chem. Soc. 2011, 133 (49), 19684–19687 DOI:
10.1021/ja209239m.
(28) Il’ichev, Y. V; Schwörer, M. A.; Wirz, J. Photochemical Reaction Mechanisms of 2-
Nitrobenzyl Compounds: Methyl Ethers and Caged ATP. J. Am. Chem. Soc. 2004, 126
(14), 4581–4595 DOI: 10.1021/ja039071z.
84
Chapter 5. Wearable Ultraviolet Light Dosimeter
5.1 Motivation
Over 20% of the American population is predicted to develop skin cancer during their
lifetime
1
. This number is staggering, considering that overexposure to UV radiation from
sunlight is the disease’s primary risk factor
2
. The prevalence of skin cancer brings to light an
important fact – that it is difficult to measure and control the amount of UV radiation exposure
that you receive. One cannot simply avoid sunlight all together because it is necessary for
vitamin D synthesis and immune system benefits
3,4
. Therefore, individuals must balance the
maximum exposure limit with a minimum recommended dose. Clearly, the need exists for a
simple yet effective method of maintaining healthy UV light exposure levels, while preventing
overexposure.
Current prevention measures recommend avoiding the risk entirely by restricting UV
light exposure (via the sun). However, broadly defined preventative measures like these are
unnecessarily restrictive because they do not take into account individual and situational
differences in risk
5
. Skin melanin content, UV index, and reflective environments are examples
of factors that impact the relative risk-associated UV exposure dose that a person receives. Given
how important specific situation parameters are, the ideal solution would be a UV monitoring
device or alert that is exposed to the same environmental conditions as the individual, for
example, a personalized, wearable UV dosimeter.
The design and implementation of a wearable UV dosimeter poses several fundamental
challenges
6
. (1) The sensor must be portable and able to conform easily to different body areas.
(2) The external material should be able to perform under various environmental conditions, like
heat and water exposure. (3) The sensor should be able to differentiate UV doses between 30 –
85
300 J/m
2
to accurately prevent sun-induced damage
7,8
. (4) The sensor should also have an alert
system that is straightforward and noticeable to the user.
Existing sensors must balance accuracy with ease of use, with most options falling into
one of two categories: electronic sensors that are precise yet costly and bulky
9–12
, or disposable
sensors that are inexpensive yet inaccurate
13,14
. Recent progress in wearable monitors has
focused primarily on making electronic sensors smaller and cheaper; however, few advances
have been made in developing a highly accurate power-free sensor. In addition, many sensors
only measure instantaneous UV light intensity, rather than providing the cumulative dose
15
.
Research in skin cancer has shown that the cumulative UV exposure received plays a more
important role in development of the disease than instantaneous UV light intensity, making a
cumulative UV measurement more clinically relevant for prevention of skin cancer
16
.
One potential route is to use a UV responsive material to act as an indicator of cumulative
UV exposure. There have been several papers that form the foundation of this field of polymer
research. In 1976, Davis et al. demonstrated the first such UV indicator comprising a
polysulphone film that exhibited a change in absorbance as a function of incident ultraviolet
dose
17
. While still used widely used today, this method required the use of a spectrophotometer
to analyze absorbance change, and was impractical for a portable device
18
. Mills et al.
demonstrated a lightweight UV-indicator strip with an independent color change alert based on a
two-step mechanism of UV-driven acid release and pH indicating dye
19
. The multicomponent
dosimeter incorporated different compositions of an added base to the indicator solution and
created both a fast-acting and delayed-response system on the same indicator. The multi-kinetic
system was unique; however, it involved the incorporation of malachite green pH dye, which is
toxic. Khibani et al. demonstrated a paper-based UV sensor utilizing the photocatalytic
86
properties of titanium dioxide to degrade food dyes
20
. The degradation results in a gradual loss of
film color, alerting the user to UV exposure dose. This sensor was non-toxic, but there was
relatively low contrast between pre- and post- UV exposed state makes it difficult to read.
The work presented in this chapter outlines the development of a power-free wearable
UV dosimeter that withstands typical environmental use conditions, and whose measurements
are accurate and reproducible
21
. The dosimeter is based on the UV-induced color change
mechanism that occurs in oNB-substituted polymer films as a result of azo dimerization. The
appearance and degree of color change from colorless to yellow-orange alerts the wearer to
potentially harmful exposure doses of UV light (Figure 5-1). We present an environmentally safe
fabrication method using FDA-approved polymer backing and coating and non-toxic solvents.
Additionally, we show the dosimeter’s consistent performance when subjected to various
environmental tests. The technology presented here can serve as a basis for future smart
polymer-based UV sensors and ultimately contribute to the widespread prevention of skin
cancer.
Figure 5-1. Wearable UV dosimeter strip on hand before (a) and after (b) UV exposure in a photoreactor.
5.2 Experimental Details
While we did not perform any experimental tests involving human subjects wearing the
sensors, the experiments conducted in this project were designed to mimic anticipated use and
a) b)
87
natural sun exposure as closely as possible. Because the sensor is intended to be a wearable
device, it would likely be subjected to mechanical deformation during use. We subjected the
sensors to various bending and water exposure tests to simulate conditions beyond the lab. For
the bending tests, the samples were wrapped around a pen with radius 3.75mm and subjected to
20 bending and unbending cycles before response kinetics testing. Water exposure tests involved
spraying a continuous stream of DI water on the samples for several seconds and letting the
samples air dry before response kinetics testing. We did not conduct temperature experiments
where the temperature was deliberately controlled and recorded against UV response; however,
the temperature was unrestricted during simulated sun exposure and was observed to have
increased during experiments. Using a rough touch estimate, we predicted that the temperature of
the sample increased between 10 – 20 degrees Celsius above room temperature when exposed to
simulated sunlight for extended periods of time.
5.2.1 Simulated Sun Exposure
In order to simulate sun exposure as accurately as possible, we matched both the spectral
output and intensity of the sun at the Earth’s surface. Spectral matching was achieved by using a
HAL-320 Class A solar spectrum lamp from Asahi Spectra. It is important to note that the HAL-
320 lamp has an AM1.5 filter (Air Mass 1.5), which effectively mimics atmospheric scattering
due to 1.5x the thickness of the Earth’s atmosphere. Atmospheric scattering tends to be largest at
higher frequencies and strongly depends on the angle of the sun to an observer on the earth (solar
zenith angle). This is why the sky appears more pink at sunrise and sunset; the blue and green
frequencies are being scattered over longer atmospheric paths as compared to when the sun is
directly overhead. AM 1.5 corresponds to a solar zenith angle of approximately 48°, and is most
accurate for representing sunlight attenuation at many of the world’s most populous cities
88
located at mid-latitudes. To match intensity, we calibrated the lamp output with a Si photodiode
before each exposure experiment to be one Sun, or the irradiance of the sun on the outer
atmosphere at a distance of one AU (astronomical unit).
The physical set-up of the sample and the solar simulator path is shown in Figure 5-2.
Light from the solar simulator was reflected off a mirror up through a circular aperture to the
sample. The diameter of this aperture was variable across experiments and is specified in detail
for each experiment.
Figure 5-2. Left: Image of simulated sun exposure experimental set-up. White arrows indicate the path of
the simulated sunlight. Right: Schematic of light being applied for different amounts of time through a
circular aperture. Each time data point was taken on a different part of the sample.
5.2.2 Quantifying Color Change
To compare the color change across samples, we decided to use the yellowness index
(YI) as a defining metric. YI is commonly used when determining the color change in plastics as
10 min 1 hour
89
they yellow over time
22
. YI is given by the ASTM D1925 Standard Test Method for Yellowness
Index of Plastics through the following equation
𝑌𝐼 =100×
R.SuNIR.9vw
x
(5.1)
where R, G, B are the standard primary color values. Because the initial sensor has some
yellowness value prior to UV exposure, we will primarily be looking at the change in YI, or DYI,
in the following experiments.
The RGB values were determined using Adobe Lightroom software to analyze
photographs of each sample. Each sample photograph was taken under the same lighting
conditions with a 50mm prime lens on a Canon 60D camera. The RGB values were sampled at
three different points in each sample image and averaged to obtain each YI value. The error in
each measurement was calculated using the standard deviation.
5.3 Wearable Sensor Fabrication
In fabricating a robust wearable sensor, we considered several conditions. First, the
sensor must be flexible enough to conform to and move with the skin, allowing it to be placed
wherever on the body is most exposed to sunlight. A polymer-based sticker or patch meets this
condition because most polymers are pliable enough at room temperature to be flexible. Second,
the sensor must be waterproof to allow use at the pool or beach, where most intense UV
exposure occurs. We met this requirement by encapsulating the sensor in a layer of
polydimethylsiloxane (PDMS), which keeps water from interacting with the main active layer
polymer. Third, the sensor should be fabricated from materials that are safe for human exposure.
PDMS is FDA approved as safe for human contact, and it is even in some foods. Polyethylene
napthalate (PEN), which is used as the sensor backing, is also generally regarded as safe by the
FDA. The solvent used for deposition of the active layer polymer was Ethyl Acetate, which is
90
relatively non-toxic should any remain on the sensor after vacuum drying. Fourth, the coating
polymer must be transparent, allowing UV light to pass through and also the resulting color
change to be visible through the top layer. This condition was met by using PDMS, which is
transparent to visible light and has low absorption in the UV wavelength range.
Two types of sensors were fabricated: one with a PDMS backing (bilayer) and one with a
PEN backing/ PDMS top layer (trilayer). Schematics and images of both are shown in Figure 5-
3. The PDMS-backed bilayer sensor was optimal for UV-Vis transmission measurements
because PDMS is transparent to UV light, whereas PEN is not. However, active polymer thin
film layers were more difficult to produce consistently on PDMS due to its highly hydrophobic
nature. For this reason, we used PEN as an alternate backing in the trilayer sensors to ensure
uniformity of active layer film thicknesses. In the colorimetric testing experiments, the lack of
UV light transmission through PEN was not an issue because transmission through the sample
was not measured.
Figure 5-3. Schematics and images of bilayer (left) and trilayer (right) wearable UV sensors.
PMA-ONB
PDMS
PEN
PMA-ONB
PDMS
Bilayer Trilayer
91
5.3.1 Bilayer Sensor Fabrication
First, solutions of photocleavable PMA (synthesis described in Chapter 4) were made as
5wt% and 10wt% solutions in Ethyl Acetate. PDMS was prepared using Sylgard 184 elastomer
with a base:curing agent ratio of 10:1 and cured at 75°C for 1 hour. Once fully cured, the PDMS
was cut into 1cm x 1cm squares. Equal volumes of the PMA solutions (175µL) were deposited
directly onto the small squares of PDMS via drop coating and dried in a vacuum oven overnight.
5.3.2 Trilayer Sensor Fabrication
Solutions of 5wt% and 10wt% photocleavable PMA were prepared in Ethyl Acetate.
Teonex PEN films were first stripped of adhesive, cut into small squares (1cm
2
), and then
underwent an oxygen plasma treatment to enhance the hydrophilicity of the surface. The PMA-
Ethyl Acetate solutions were drop cast onto PEN in an evacuated chamber, such that the solvent
evaporation and surface diffusion rates were balanced to avoid a coffee ring effect. Once
evacuated, the chamber was rotated at small angles to ensure adequate dispersion of the liquid
across the surface during drying. The vacuum was released after approximately 3 minutes, or
when the liquid on the surface of PEN appeared mostly evaporated. The films were dried first on
a hot plate at 50°C for 5 minutes, then in a vacuum oven at 40 – 50°C overnight to fully remove
the Ethyl Acetate. The top protective layer of PDMS was deposited using a spin coating method
at 2000 rpm for 1 minute. PDMS was prepared in a 10:1 ratio as described previously, and cured
on the surface of the sample for 30 minutes at 70°C. The resulting layer thicknesses were
analyzed using a Keyence VHE 5000 Digital Microscope.
5.3.3 Unsuccessful Sensor Fabrication Methods
A number of alternate methods of UV sensor fabrication were attempted. These less
reproducible methods are detailed in the following sections.
92
5.3.3.1 Transfer Printing
In Chapter 4, uniformly thick polymer thin films were deposited onto silica/silicon wafers
via spin coating. One method attempted was to transfer these spin coated thin films to PDMS
using an existing transfer printing method
23
. In this method, PDMS was pressed on top of the
sample such that complete conformal contact was achieved and water was used to separate the
polymer film from the wafer surface, allowing the PMA to adhere to the PDMS surface and be
peeled off the wafer. Ideally, water would have been able to penetrate the polymer film and silica
interface due to the hydrophobic and hydrophilic mismatch of the surfaces. However, the
hydrophobicity of our system differed from that of the paper: PMA is less hydrophobic than
polystyrene (PS). Thus, our polymer did not detach so readily when exposed to water. Multiple
film thicknesses and PDMS curing agent ratios to curing times were attempted, but the polymer
film ultimately proved inseparable from the wafer surface.
5.3.3.2 PDMS Sputter Coating & Alternative Solvents
Other solvents aside from Ethyl Acetate were used as a deposition solvent for
photocleavable PMA. Tetrahydrofuran, toluene, chloroform, and acetone were all tested to
optimize the ratio of drying time to surface diffusion. The major problem encountered with most
early drop coating attempts was that polymer diffusion within the solution to the pinned edges of
the solution drop would occur much faster than the solvent evaporation rate, causing
inconsistencies in the final dried film (coffee ring effect)
24
. By trying solvents with varying
evaporation rates, we sought to increase the evaporation rate so that it was faster than diffusion
and the resulting film would be smooth. However, many of these solvents had a strong
interaction with the PDMS itself, causing it to contract and bend undesirably.
93
We then sought to create a barrier between the solvent and the PDMS that was still
flexible but did not allow the solvent to deform the PDMS. We sputter coated platinum and
palladium directly onto the PDMS to create this layer. However, the layer was not resistant to
bending and formed cracks upon slight mechanical deformation of the PDMS. Thus, when the
PMA solution was deposited, the solvent still reached the PDMS through the cracks in the
coating and caused the PDMS to deform.
5.3.3.3 PET & Parylene C Coating
Two alternative polymers to PDMS were investigated as potential coating materials to
form a waterproof barrier over the active polymer layer in the trilayer system: polyethylene
terephthalate (PET) and Parylene C. Both of these materials were obtained as thin polymer films.
We attempted to stretch the polymer films over the surface of the deposited PMA layer and seal
the films around the side of each sample. However, we found that these films were not
structurally sound against mechanical bending and would permanently wrinkle upon flexing the
sensor, detaching entirely in some cases.
5.4 Sensor Characterization
The finished wearable sensor is transparent, flexible, and can be temporarily attached to
the skin’s surface using uncured PDMS (Figure 5-4). The sensor was cut into rectangular strips
to facilitate different time points of increasing sun exposure.
94
Figure 5-4. Images of the trilayer sensor demonstrating the transparency and flexibility.
5.4.1 Thickness Measurements
Cross-sections of trilayer samples were cut using a razor and imaged to determine layer
uniformity and adhesion. Figure 5-5 shows that each layer smoothly adhered to the adjacent
layer without any air gaps. Additionally, we determined that the active PMA layers varied
between 30 – 60 µm across samples. The sample cross section pictured in Figure 5-5a has
thicknesses of 35µm (PDMS), 57µm (PMA-ONB), and 111µm (PEN). Microscope images were
used to confirm these findings and helped elucidate some problems with the deposition. At the
edges of some of the samples, one or more of the polymer layers was missing. In some samples,
no PDMS layer was observed, suggesting that the PDMS layer delaminated from the surface
(Figure 5-5b). In other samples, no PMA layer was observed (Figure 5-5c), suggesting that the
drop coating procedure was not spreading effectively over the whole surface. All of the structural
inconsistencies were observed near the edges of samples, and possibly resulted from the edge
effects of drying or susceptibility to peeling. Using these results, we modified all future samples
by cutting the edges off before testing and using only the middle of sample pieces for UV
exposure experiments.
95
Figure 5-5. Cross-section microscope images of the trilayer sensor fabricated with 10wt% PMA- EtAc
solution. Images indicate a) successful deposition of all three layers, b) deposition of only PMA active
layer and c) deposition of only PDMS coating. The scale bar is 100µm.
5.4.2 UV-Vis Absorption
One way to quantify both the color change response and formation of photoproducts as a
result of UV light exposure is to monitor changes in the absorption spectrum of the sensor
25
. We
conducted these absorption measurements on the bilayer sensor rather than the trilayer sensor
because there is less background absorption due to the non-active sensor layers. We monitored
the absorbance spectrum over 2 hours of simulated sun exposure and the results are shown in
Figure 5-6.
Between 0 - 1 hour of simulated sun exposure, we observed a gradual decrease in the
absorption peak at 269nm. The decrease in height of this peak corresponds to the decrease in
concentration of nitrobenzyl groups as they are being converted to nitroso groups through
photocleavage
26
. Within the same time frame, we also observed the growth of an absorption peak
at 318nm. After 1 hour, the height of this 318nm peak appeared to stabilize. This peak
corresponds to the formation of the primary nitroso photoproduct
26,25
.
a) b) c)
PEN
PMA-ONB
PDMS
96
Figure 5-6. UV-Vis Absorption spectrum of bilayer PMA-PDMS sensor with increasing simulated sun
exposure time. Shows the concurrent decrease in the peak at 269nm and increase in the peak at 318nm
over time as oNB groups are cleaved.
We took a more detailed look at the simultaneous decrease and growth of these two
absorbance peaks by overlaying plots of the change in absorbance over time (Figure 5-7). The
exponential decrease in the absorbance at 269nm corresponds exactly with the exponential
increase in the absorbance at 318nm, validating the relationship between the two. Both values
appeared to stabilize after approximately 1 hour of UV exposure, suggesting completion of the
photocleavage reaction after this time.
97
Figure 5-7. Change in absorbance of bilayer PMA-PDMS sensor at two specific wavelengths (269nm and
318nm) monitored as a function of simulated sun exposure time. Inset shows images of the sensor before
and after 2 hours of simulated sun exposure.
Additionally, many small absorption peaks between 350 – 500nm appeared along with
the growth of the 318nm peak (Figure 5-6). These peaks appear in the UV-Vis absorbance
spectrum as an overall broad increase between 350 – 500nm rather than a collection of small
peaks due to lack of resolution. Interestingly, these peaks continue to grow in value beyond 2
hours of simulated sun exposure. It is likely that these peaks correspond to the formation of
secondary photoproducts, which continue to form after the primary photocleavage reaction is
complete.
One of these photoproducts is azobenzene, which is formed through UV-induced
dimerization of nitroso photocleavage products (Section 2.1.2). Due to the isomerization of
azobenzene, the absorption spectra contains peaks around 330nm and 430nm
27
. Because this
absorbance extends into the visible range of wavelengths, the resulting material exhibits a
0h 2h
98
yellow-orange color. This color change from clear to yellow due to azobenzene formation serves
as the basic mechanism behind the UV dosimeter response.
5.5 Wearable UV Dosimeter Response
For the wearable UV dosimeter, we utilized the trilayer polymer sensor instead of the
bilayer sensor because the active PMA-ONB material is fully encapsulated by the two other
layers, protecting it from the environment. Since the top layer of the trilayer (PEN) is transparent
and colorless, the color change of the active layer can be seen. The degree of saturation of this
color change indicates the cumulative UV exposure dose, or time-normalized UV intensity. A
higher degree of saturation correlates to a larger UV exposure dose (Figure 5-8).
Figure 5-8 Schematic of color changing sensor concept. The yellow color change appears only after
exposure to UV light, and the intensity is correlated to UV exposure dose.
In section 5.4, we characterized this color change by monitoring how the absorbance
changed at the corresponding wavelengths. However, this method requires the use of a
spectrophotometer and is not feasible for a wearable sensor. Therefore, in testing the wearable
99
trilayer sensor, we relied on a visual observation of the color change, quantified by the change in
yellowness index.
The following sections describe experiments probing the color change response of the
sensor under various environmental and extended lifetime conditions. The color change response
kinetics were characterized in response to four major variables: drop coating concentration,
lifetime, bending, and water exposure. To test feasibility in intended use the sensor was tested in
conjunction with a commercially available sunscreen to examine both compatibility and the
change in color response.
5.5.1 Sigmoidal Response
The DYI behavior of all sensors tested exhibited a sigmoidal increase with time. This
type of behavior is typical for dose-response systems and can be modeled using a Gompertz
sigmoidal function
𝑌𝐼 =𝑎𝑒
Im
yz({y{
Y
)
(5.2)
where a is the maximum DYI, k is the response rate of the PMA-ONB, and x
c
is the exposure
time at which the sensor reaches half of its maximum DYI value. Each of these constants was
calculated by fitting the experimental data to the generalized form of Equation 5.2. We used
these experimentally calculated constants as metrics to compare kinetic behavior across data sets.
5.5.2 PMA-ONB Concentration
In the fabrication of the trilayer sensor, two different concentrations of PMA-ONB in
Ethyl Acetate solvent were tested: 5 wt% and 10 wt%. The same volume of liquid was dispensed
for each sample. Therefore, the main difference in the resulting samples was film thickness.
Films made with 10wt% solution were slightly thicker than those made with 5wt% solution due
to the increased mass of polymer deposited. DYI of the trilayer sensors was plotted as a function
100
of simulated sun exposure time in Figure 5-9. Three separate points at each exposure time were
sampled in calculation of DYI and the standard deviation of these values is plotted as the error
for each data point. DYI of a trilayer sensor with no active PMA-ONB layer (only PDMS-PEN)
was also measured and plotted in Figure 5-9 as a control. With respect to the control sample, all
of the data sets exhibit sigmoidal behavior. However, the 10% samples exhibit a steeper initial
increase in DYI than the 5% samples. Additionally, the final DYI values reached by the 10%
samples is larger than that of the 5% samples.
Figure 5-9. Change in yellowness index (DYI) with increasing simulated sun exposure of trilayer sensors
using 0 wt% (control), 5 wt% and 10 wt% PMA-ONB concentrations. Both non-control concentrations
exhibit a sigmoidal relationship with exposure time.
The trends can be compared quantitatively using the coefficients of the fitted sigmoidal
curves (Table 5-1). The average response rate k has a lower value for the 5wt% trilayer than the
10wt% trilayer, supporting the visual observation in Figure 5-10 that the increase in DYI of the
5wt% is slower. The slower rate is also evident when comparing x
c
, which occurs later in the
5wt% trilayer than the 10wt% trilayer. These observations are correlated to the fact that the final
101
DYI value, represented by a in Equation 5.2, is larger on average for 10wt% trilayer than the
5wt% trilayer. Physically, this means that the 10wt% trilayer sensor exhibits a larger color
change to a more saturated yellow/orange than the 5wt% trilayer sensor. The deeper color
change makes sense because in a thicker film there are more color producing azo units per unit
area. The higher azo unit density in the 10wt% sensor also explains the increased color change
response rate when compared to the 5wt% sensor.
Table 5-1 Coefficients in Gompertz Fitting Function for DYI of UV-Exposed Sensors
a k (h
-1
) x
c
(h)
Trilayer, 5% 23.278 ± 1.028 9.697 ± 1.571 0.212 ± 0.011
Trilayer 5% 28.238 ± 0.457 6.172 ± 0.338 0.222 ± 0.006
Trilayer, 10% 28.792 ± 0.400 8.622 ± 1.009 0.185 ± 0.006
Trilayer, 10% 33.000 ± 1.221 8.180 ± 0.850 0.202 ± 0.014
In light of the more rapid and deeper color change, we decided to use the 10wt% trilayer
sensors for the majority of the response rate experiments. The more saturated color change was
easier to observe and characterize visually.
5.5.3 Sensor Lifetime
In order for the wearable sensor to be a commercially viable product, it must be shelf-
stable for an extended period of time after fabrication. We explored the stability of the trilayer
sensor by characterizing the color change response after storage for 5 weeks in an ambient
environment. Figure 5-10 shows the response of the 5 weeks stored sensor in comparison to the
102
freshly fabricated sensor. The older sensor clearly exhibited the characteristic sigmoidal increase
in DYI, with a comparable response rate to the newer sensors.
Figure 5-10 Left: Color change response of sensor stored for 5 weeks (lifetime) vs. newly made sensors
(10wt%) with respect to simulated sun exposure time. Right: Images of trilayer sensor before (top) and
after (bottom) simulated sun exposure. Exposure times from left to right are 10, 20, 30, and 60 minutes.
The observation that the sensor remains active after five weeks was further validated by
the calculating the coefficients of the sigmoidal fit, found in Table 5.2. Values of a, k, and x
c
were found to be similar within error for the 5 weeks stored sensor and the new sensors. From
these results, we concluded that long term storage of the trilayer sensor of up to 5 weeks did not
significantly affect the sensor performance.
5.5.4 Bending
The trilayer sensor was designed to be a wearable product, which must be able to
maintain performance under repeated mechanical deformation. To simulate this environmental
condition, we performed 20 bending and unbending cycles of the sensor around a pen with radius
3.75mm. We characterized the bent sensor response and compared it to non-deformed sensors in
103
Figure 5-11. The bent sensor exhibited a slightly different sigmoidal curve than the non-
deformed sensors, due to a higher DYI at 10 minutes of exposure and a lower DYI at 30 minutes
of exposure. These values resulted in a larger error when calculating the Gompertz fitting
coefficients (Table 5.2). The variability in yellowness change may have resulted from non-
uniform deformation during bending. Some parts of the PMA-ONB film may have been bent
more than others, specifically near the edges of the sample. Regardless of the potential bending
inconsistencies, the a, k, and x
c
values for the bent sample were all calculated to be within the
error of the non-deformed trilayer samples and are presented in Table 5.2. From this experiment,
we can conclude that bending does not significantly affect the sensor performance.
Figure 5-11 Left: Color change response of bent and non-deformed sensors as a function of simulated sun
exposure time. Right: Images of trilayer sensor before (top) and after (bottom) simulated sun exposure.
Exposure times from left to right are 10, 20, 30, and 60 minutes.
After
104
5.5.5 Water Exposure
The highest doses of UV exposure often occur near bodies of water, for example at the
pool or the beach, so the sensor must maintain its performance with water exposure. We
simulated this environmental condition by spraying DI water on the sensor for a few minutes,
followed by air-drying. The color change response of the water-exposed sensors is compared
with non-water exposed sensors in Figure 5-12. The initial color change of the water-exposed
sensor (10 minutes sun exposure) is less than the other sensors, yet overall it maintains the
approximately the same response rate and final DYI value. Comparison of the Gompertz fitting
coefficients further illustrates this trend. The a, k, and x
c
values for both the water-exposed and
non-water exposed sensors were very similar (Table 5.2).
Figure 5-12 Left: Color change response of water-exposed sensors and non-water exposed sensors as a
function of simulated sun exposure time. Right: Images of trilayer sensor before (top) and after (bottom)
simulated sun exposure. Exposure times from left to right are 10, 20, 30, and 60 minutes.
105
Table 5.2 Coefficients in Gompertz fitting function for DYI of UV-exposed trilayer sensors
under various environmental conditions. The original trilayer samples are included in the table
for reference.
a k (h
-1
) x
c
(h)
Trilayer, 10% 28.792 ± 0.400 8.622 ± 1.009 0.185 ± 0.006
Trilayer, 10% 33.000 ± 1.221 8.180 ± 0.850 0.202 ± 0.014
Lifetime 30.322 ± 0.474 7.073 ± 0.370 0.218 ± 0.005
Bending 31.896 ± 2.603 5.962 ± 1.538 0.171 ± 0.016
Water Exposure 30.870 ± 0.279 8.267 ± 0.259 0.229 ± 0.003
Sunscreen 9.047 ± 2.121 2.049 ± 0.731 0.393 ± 0.118
5.5.6 Sunscreen Compatibility
One existing method that many people have adopted to prevent UV overexposure is the
use of a topically applied sunscreen. Combining the sensor alert system with current preventative
measures involving sunscreen would prove a powerful tool in the fight against skin cancer. Thus,
we characterized the sensor performance when a commercially available sunscreen was applied
over it before exposure to UV. In this experiment, Neutrogena Ultra Sheer SPF 100+ spray on
sunscreen was applied as a thin layer over the sensor and allowed to air dry for several minutes.
The color change response of the sensor was then recorded and plotted as a function of simulated
sun exposure time in Figure 5-13.
There was a clear difference in the color change behavior between the sunscreen-coated
and bare sensors. The sunscreen-coated sensor did not follow a sigmoidal trend, but instead more
closely followed a linear increase. However, this increase in DYI was slow, resulting in a final
106
DYI value after 1 hour that was less than the initial DYI value of the bare sensor after 10 minutes.
The faintness of the color change can be clearly seen in Figure 5-13b.
Figure 5-13 Left: Color change response of sensors with and without sunscreen as a function of simulated
sun exposure time. Right: Images of trilayer sensor before (top) and after (bottom) simulated sun
exposure. Exposure times from left to right are 10, 20, 30, and 60 minutes.
The sunscreen-coated sensor data was fitted to the Gompertz sigmoidal equation and the
coefficients are listed in Table 5.2. The error in these calculated coefficients is relatively large,
because the data does not follow a sigmoidal trend. However, it can be clearly seen that the a, k,
and x
c
values vary significantly from the coefficient values of all of the uncoated sensors. The
low k and high x
c
show that the sunscreen effectively acts as a UV filter, preventing color-
causing azo units from being formed in the active PMA-ONB layer. From these results, we
concluded that our sensor is compatible with this type of UV filter and can detect the UV-
blocking capabilities of sunscreen.
107
Chapter 5 References
(1) Robinson, J. K. Sun Exposure, Sun Protection, and Vitamin D. J. Am. Med. Assoc. 2005,
294 (12), 1541–1543.
(2) Narayanan, D. L.; Saladi, R. N.; Fox, J. L. Ultraviolet Radiation and Skin Cancer.
International Journal of Dermatology. 2010, pp 978–986.
(3) Reichrath, J. The Challenge Resulting from Positive and Negative Effects of Sunlight:
How Much Solar UV Exposure Is Appropriate to Balance between Risks of Vitamin D
Deficiency and Skin Cancer? Prog. Biophys. Mol. Biol. 2006, 92 (1), 9–16 DOI:
10.1016/j.pbiomolbio.2006.02.010.
(4) Rasmussen, L. B. Vitamin D Deficiency. N Engl J Med 2007, 357 (43), 266–281 DOI:
357/3/266 [pii] 10.1056/NEJMra070553.
(5) Roy, C. R.; Gies, H. P.; Lugg, D. J.; Toomey, S.; Tomlinson, D. W. The Measurement of
Solar Ultraviolet Radiation. Mutat. Res. 1998, 422 (1), 7–14.
(6) Diffey, B. L. Sources and Measurement of Ultraviolet Radiation. Methods 2002, 28 (1),
4–13 DOI: 10.1016/S1046-2023(02)00204-9.
(7) Fanselow, D. L.; Ph, D.; Pathak, M. A.; Ph, D. Reusable Ultraviolet Monitors: Design,
Characteristics, and Efficacy. J. Am. Acad. Dermatol. 1983, 9 (5), 714–723 DOI:
10.1016/S0190-9622(83)70185-4.
(8) Australian Radiation Protection and Nuclear Safety Agency. Radiation Protection
Standard; Occupational Exposure to Ultraviolet Radiation; 2006; Vol. 12.
(9) Wu, W.; Bai, S.; Yuan, M.; Qin, Y.; Wang, Z. L.; Jing, T. Lead Zirconate Titanate
Nanowire Textile Nanogenerator for Wearable Energy-Harvesting and Self-Powered
Devices. ACS Nano 2012, 6 (7), 6231–6235 DOI: 10.1021/nn3016585.
(10) Chang, H.; Sun, Z.; Ho, K. Y.-F.; Tao, X.; Yan, F.; Kwok, W.-M.; Zheng, Z. A Highly
Sensitive Ultraviolet Sensor Based on a Facile in Situ Solution-Grown ZnO
Nanorod/graphene Heterostructure. Nanoscale 2011, 3 (1), 258–264 DOI:
10.1039/c0nr00588f.
(11) Xi, M.; Wang, X.; Zhao, Y.; Zhu, Z.; Fong, H. Electrospun ZnO/SiO2 Hybrid
Nanofibrous Mat for Flexible Ultraviolet Sensor. Appl. Phys. Lett. 2014, 104 (13), 133102
DOI: 10.1063/1.4870296.
(12) Hooke, R.; Pearson, A.; O’Hagan, J. Autonomous Portable Solar Ultraviolet
Spectroradiometer (APSUS) - A New CCD Spectrometer System for Localized, Real-
Time Solar Ultraviolet (280-400 Nm) Radiation Measurement. Photochem. Photobiol.
2014, 90 (4), 903–910 DOI: 10.1111/php.12251.
(13) Ales, T. M. et al. UV Detection Devices and Methods. 8044363 B2, 2011.
(14) Goudjil, K. Ultraviolet Active Wristband. 5914197, 1999.
(15) Corrêa, M. de P.; Godin-Beekmann, S.; Haeffelin, M.; Brogniez, C.; Verschaeve, F.;
Saiag, P.; Pazmiño, A.; Mahé, E. Comparison between UV Index Measurements
Performed by Research-Grade and Consumer-Products Instruments. Photochem.
Photobiol. Sci. 2010, 9 (4), 459–463 DOI: 10.1039/b9pp00179d.
(16) Karagas, M. R.; Zens, M. S.; Nelson, H. H.; Mabuchi, K.; Perry, A. E.; Stukel, T. A.;
Mott, L. A.; Andrew, A. S.; Applebaum, K. M.; Linet, M. Measures of Cumulative
Exposure from a Standardized Sun Exposure History Questionnaire: A Comparison with
Histologic Assessment of Solar Skin Damage. Am. J. Epidemiol. 2007, 165 (6), 719–726
DOI: 10.1093/aje/kwk055.
108
(17) Davis, A.; Deane, G. H. W.; Diffey, B. L. Possible Dosimeter for Ultraviolet Radiation.
Nature 1976, 261 (5556), 169–170 DOI: 10.1038/261169a0.
(18) Lester, R. a; Parisi, a V; Kimlin, M. G.; Sabburg, J. Optical Properties of poly(2,6-
Dimethyl-1,4-Phenylene Oxide) Film and Its Potential for a Long-Term Solar Ultraviolet
Dosimeter. Phys. Med. Biol. 2003, 48 (22), 3685–3698 DOI: 10.1088/0031-
9155/48/22/005.
(19) Mills, A.; McDiarmid, K.; McFarlane, M.; Grosshans, P. Flagging up Sunburn: A
Printable, Multicomponent, UV-Indicator That Warns of the Approach of Erythema.
Chem. Commun. 2009, No. 11, 1345–1346 DOI: 10.1039/b900569b.
(20) Khiabani, P. S.; Soeriyadi, A. H.; Reece, P. J.; Gooding, J. J. Paper-Based Sensor for
Monitoring Sun Exposure. ACS Sensors 2016, 1 (6), 775–780 DOI:
10.1021/acssensors.6b00244.
(21) Lee, M. E.; Armani, A. M. Flexible UV Exposure Sensor Based on UV Responsive
Polymer. ACS Sensors 2016, 1 (10), 1251–1255 DOI: 10.1021/acssensors.6b00491.
(22) ASTM, D. 70- Standard Test Method for Yellowness Index of Plastics. In Annual Book of
ASTM Standards; 1970; pp 167–169.
(23) Li, H.; Wu, J.; Huang, X.; Yin, Z.; Liu, J.; Zhang, H. A Universal, Rapid Method for
Clean Transfer of Nanostructures onto Various Substrates. ACS Nano 2014, 8 (7), 6563–
6570 DOI: 10.1021/nn501779y.
(24) Deegan, R. D.; Bakajin, O.; Dupont, T. F. Capillary Flow as the Cause of Ring Stains
from Dried Liquid Drops. 1997, 827–829.
(25) Schumers, J.; Gohy, J.; Fustin, C. A Versatile Strategy for the Synthesis of Block
Copolymers Bearing a Photocleavable Junction. Polym. Chem. 2010, 1 (2), 161 DOI:
10.1039/b9py00218a.
(26) Il’ichev, Y. V; Schwörer, M. A.; Wirz, J. Photochemical Reaction Mechanisms of 2-
Nitrobenzyl Compounds: Methyl Ethers and Caged ATP. J. Am. Chem. Soc. 2004, 126
(14), 4581–4595 DOI: 10.1021/ja039071z.
(27) Bédard, M. F.; De Geest, B. G.; Skirtach, A. G.; Möhwald, H.; Sukhorukov, G. B.
Polymeric Microcapsules with Light Responsive Properties for Encapsulation and
Release. Adv. Colloid Interface Sci. 2010, 158 (1–2), 2–14 DOI:
10.1016/j.cis.2009.07.007.
109
Chapter 6. Optically Tunable Frequency Combs
Optical frequency combs generated by ultra high-Q microresonators are a promising new
area of research that has spurred significant progress in precision spectroscopy measurements
and the development of all-optical atomic clocks
1–5
. An optical frequency comb is defined by a
series of sharp spectral emission lines with a predetermined spacing between comb modes
1,6
. The
optical resonance cycles responsible for these combs allow scientists to access frequencies up to
1000 GHz, enabling the realization of extremely high precision measurements
7
. Ultra high-Q
optical microresonators made from third order nonlinearity materials like silica can generate
frequency combs through the process of parametric frequency conversion
3,7–9
.
The goal of this project was to fabricate an optical microresonator that changed its
frequency comb behavior in response to UV light exposure. The key functionality of such a
device is the ability to externally tune optical emission through a user-controlled light source.
This device could be used as an optical switch in integrated optical circuits, or as an indicator of
UV light exposure since the change in optical behavior is non-reversible.
The optical responsiveness of the microresonator stems from surface functionalization
with a UV-light responsive oligomeric layer. The specific light-responsive moiety is ortho-
nitrobenzyl alcohol (oNB), which undergoes an intramolecular hydrogen abstraction upon
exposure to UV light
10,11
. This rearrangement of electrons causes a change in the chemical
functional groups present on the surface of the microresonator, which in turn affects the optical
polarizability at the surface of the device. We propose that changing the surface in this way will
affect the nonlinear optical properties of the microresonator, and can be observed by changes in
the parametric optical frequency comb generation.
110
The oNB moiety was covalently attached to a silica sphere microresonator using a three-
step process. In preparation for the reaction, the oNB moiety (in the form of 5-hydroxy-2-
nitrobenzyl alcohol, HNBA) was modified to contain an alkyne functional group that could
participate in a CuAAC click chemistry reaction. To prepare the silica surface, it was first
halogenated to contain a reactive chlorine end group (step 1) and then underwent azidation to
form a CuAAC azide reactive group (step 2). The click reaction was conducted in solution to
attach the modified oNB to the surface (step 3). Each of these steps is described in more detail in
the synthesis section below.
The quality factors and frequency comb generation of the functionalized microspheres
were tested using the setup shown in Figure 6-1. A tapered optical fiber was used to couple light
into the resonator from a 1550nm tunable laser. The light coupled back into the fiber was
analyzed simultaneously by a photodetector and an optical spectrum analyzer (OSA) to measure
transmission and emission respectively. Measurements were taken after steps 2 and 3 of the
functionalization process in order to compare the effect of oNB moieties on the surface of the
spheres. From our results, we observed a noticeable difference in the emission of the sphere
before and after oNB surface functionalization. Future work on this project includes reproducing
these results with more spheres and characterizing the change in behavior after application of UV
light.
111
Figure 6-1. Schematic of testing set-up used to measure optical frequency comb generation in silica
microresonators. Light is evanescently coupled into the resonator with a tapered optical fiber. The fiber
output is split between an optical spectrum analyzer (OSA) and a photodetector (PD) to measure the
optical power and transmission respectively.
6.1 Alkyne-functionalized oNB Synthesis
In this project, we covalently attached oNB to the surface of a silica sphere. To perform
this attachment, we modified the one end of oNB with an alkyne functionality to allow it to
participate in a CuAAC click chemistry reaction. All solvents and reagents were purchased from
Sigma Aldrich and are ACS Reagent grade unless otherwise noted.
Alkyne functionalization of oNB alcohol was performed via reaction with propargyl
bromide in the presence of potassium carbonate at elevated temperature (Figure 6-2)
12,13
. HNBA
(0.856g, 1 equiv) was added to a flask with potassium carbonate (2.15g, 3 equiv) and 10.5mL
Dimethylformamide (DMF). The mixture was stirred at 60°C for 1 hour. Propargyl bromide
(0.7mL of 80%mix in toluene, 1.2 equiv) was added to the solution dropwise to avoid excess
formation of hydrogen bromide gas. The mixture was stirred at 60°C overnight.
Laser
Resonator
OSA
PD
90%
10%
112
Figure 6-2. Reaction Scheme for alkyne functionalization of o-nitrobenzyl alcohol.
The resulting dark brown solution was first washed with multiple Dichloromethane/
Water extractions to remove DMF solvent. However, it was observed that the aqueous wash
solution was still light brown after repeated washes indicating residual product loss. To recover
the product, further Ethyl Acetate/ Water extractions were performed. The final solution was
then dried over magnesium sulfate to remove trace water. Removal of Ethyl Acetate under
reduced pressure yielded a yellow-brown solid, indicating bifunctionalized impurities.
1
H NMR
analysis of the product confirmed the presence of these impurities.
Further purification was done through repeated recrystallization in Cyclohexane/ Ethyl
Acetate. A small amount of the compound was dissolved as a saturated solution in Cyclohexane
heated to approximately 80°C. Ethyl Acetate was slowly added to fully dissolve all solids. The
solution was then removed from heat and left to cool and slowly recrystallize overnight. The
resulting solids were clear iridescent crystals, indicating that the impurities had been removed.
The solids were filtered using a Buchner funnel and rinsed with cold cyclohexane. The molecular
structure was verified using
1
H NMR spectroscopy (Figure 6-3).
113
Figure 6-3.
1
H NMR spectrum of alkyne functionalized oNB alcohol in CDCl
3
.
The total reaction yield was less than 50% due to impurities in the starting HNBA
reactant. These impurities were not discovered until after the reaction had been completed, and
GC/MS elucidated the impurity structure. GC/MS signal peaks corresponding to azo-dimerized
compounds indicate that the reactant had previously underwent hydrogen abstraction and formed
dimers as a result. The total percentage of impurities was around 50%, accounting for the low
product yield.
6.2 Microsphere Surface Functionalization
To attach the oNB moiety to the surface of the sphere, we decided to utilize a click
chemistry reaction. Click chemistry is a fairly recent term for a class of synthetic reactions
developed by Sharpless and coworkers in the early 2000s
14,15
. Their goal was to develop a set of
modular building blocks for making carbon-heteroatom bonds that could be used across a wide
variety of different conditions. In order to be defined as click chemistry, the reactions involving
these building blocks must use readily available materials and reagents, simple reaction
114
conditions that are ideally not sensitive to air or water exposure, and yield products that require
few or no purification methods to separate from the reaction solution
14,15
. As a result of their
reliably high yields and simplicity, click chemistry reactions have been used extensively in many
different applications including drug delivery and materials science
15–17
.
The particular click reaction chosen for the surface attachment described in this chapter is
the Huisgen 1,3-dipolar cycloaddition of alkynes and azides to form 1,4-disubstituted- 1,2,3-
triazoles. Also known as copper(I)-catalyzed azide alkyne cycloaddition (CuAAC), this reaction
can be performed at ambient temperature and requires no purification to obtain the final
product
18,19
. In addition, its high yield makes this reaction an ideal candidate for our purposes. In
the following section, we describe the preparation of the silica surface for CuAAC and the
reaction performance on the surface.
6.2.1 Microsphere Fabrication
Silica microspheres were fabricated from optical fibers by first stripping the outer
polymer cladding and then using a CO
2
laser to melt the tip of the fiber. Microspheres with
diameters between 100 – 200 µm were used in these experiments. Prior to surface
functionalization, the microspheres were exposed to an oxygen plasma treatment to both increase
the number of hydroxyl groups present on the surface and also clean the surface of organic
contaminants.
6.2.2 Three-Step Surface Functionalization
The full surface functionalization procedure is shown in Figure 6-3. The plasma-treated
microspheres were first placed in a vacuum chamber with several drops of chloromethyl
trichlorosilane to introduce chlorine reactant groups to the surface (Step 1). The deposition time
was found to affect the final silane layer thickness. Initially, the vapor deposition was conducted
115
for 10 minutes. However, this process resulted in a cloudy residual film on the surface of the
glass. The vapor deposition was then conducted for 5 minutes to minimize the layer thickness.
Following silane layer deposition, the microresonators were placed in a 6mM solution of
sodium azide in DMF and the reaction was stirred overnight (Step 2). Initially, this reaction was
conducted at room temperature. However, later studies with X-Ray Photoelectron Spectroscopy
(XPS) showed that this method resulted in low azide conversion, as indicated by chlorine groups
that remained on the surface after this step. As a result, the temperature of this reaction was
increased to 80°C. Several microspheres were reserved after this step for testing and emission
characterization (Section 6.4).
For the final functionalization step (Step 3), the oNB moiety was attached to the surface
using two different methods. The first method was a copper-mediated process in solution at room
temperature. Alkyne-functionalized oNB (0.015g, 0.75equiv) was weighed into a flask with
copper bromide (0.02g, 1 equiv) and ascorbic acid (0.024g, 1.2 equiv). DMF (3mL) was then
added to the flask along with the azide-functionalized microspheres. The reaction was stirred
overnight at room temperature. Functionalized samples were washed with THF and Acetone,
then dried in a vacuum oven at 60°C until testing.
116
Figure 6-4. Surface Functionalization reaction scheme for a) azidation of silica surface and b) Cu
mediated click chemistry to attach oNB moiety to the surface.
6.3 Verification of Surface Groups
Initially, analysis of samples functionalized using the CuAAC method showed low signal
corresponding to the aromatic group on the oNB moiety, indicating low conversion. In addition,
scanning electron microscope images of these samples showed high levels of surface
contamination (Figure 6-5).
117
Figure 6-5. SEM images of microspheres that had undergone silanization (left) and surface azidation +
click chemistry (right).
In an effort to both increase click reaction efficiency and decrease surface contamination
from solution, we then tried the click reaction with minimal solution at an elevated temperature.
This experiment was performed by making a concentrated solution of alkyne-functionalized oNB
(10mg/mL) in THF and dropping two drops of this solution on the microresonator surface. The
solvent was allowed to evaporate, leaving large concentrations of the oNB moiety on the surface.
The reaction was accelerated by placing the microresonators in an oven at 80°C overnight.
Following this reaction, the microresonators were rinsed with THF and Acetone and stored in the
vacuum oven until testing. Imaging of these samples with an optical microscope revealed a slight
overall decrease in surface contamination, however many contaminant particles were still
observed.
In order to verify the attachment of oNB groups to the silica surface, we used silicon
wafers with a 2µm oxide layer as a control surface. Initially, glass microscope slides were used
as a substrate, however the glass was found to have a coating on the surface that interfered with
compositional measurements. Only results for the silica/ silicon wafers will be presented in this
118
section. The wafers were functionalized with oNB moieties using the same three-step procedure
as the microspheres.
X-ray Photoelectron Spectroscopy (XPS) was performed using a Kratos Axis Ultra DLD
X-ray photoelectron spectrometer at the USC Center for Electron Microscopy and Microanalysis
(CEMMA) with the help of Dr. Xiaoqin Shen. Prior to XPS sampling, the wafers were dried
under vacuum for several hours to remove any volatile contaminants.
The first set of XPS measurements was performed to confirm the silanization and
azidation steps. Previously, XPS measurements had shown a strong chlorine signal after step 2
(azidation). This peak should have been reduced due to the chlorine reaction with sodium azide,
yet its presence indicated unreacted chlorine. After modifying the procedure by increasing
temperature and stirring rate, we observed a reduction in the chlorine signal post-azidation,
indicating a more complete conversion to azide than previously observed (Figure 6-6).
Figure 6-6. Comparison of X-ray Photoelectron Spectra of post-azidation (Step 2) treated silica/ silicon
wafers performed at different temperatures. Binding energy was scanned between 195 – 207 eV for
Chlorine 2p.
200
400
600
800
195 200 205
Intensity (CPS)
Binding Energy (eV)
Chlorine 2p
80 C
RT
119
Then, we compared XPS spectra from samples after steps 2 & 3 to verify the addition of
oNB moiety. Comparing the signal of Nitrogen 1s did not show much difference between before
and after the click reaction (Figure 6-7a). This makes sense, as the nitrogen present in both steps
is part of a conjugated system and does not change very much. In contrast, the Carbon 1s and
Oxygen 1s spectra show a clear shift to lower binding energies after the click step 3 occurred
(Figure 6-7b, 6-7c).
Figure 6-7. Comparison of X-ray Photoelectron Spectra of post-azidation (Step 2) and post-click (Step 3)
treated silica/silicon wafers with binding energy range optimized for a) Nitrogen 1s, b) Carbon 1s, and c)
Oxygen 1s.
Another way that we can test oNB presence on the surface is to observe the UV light-
induced response of the oligomeric coating. After exposure to UV, we propose that the original
moiety will undergo hydrogen abstraction as presented in Figure 6-8. This reaction will produce
different chemical groups on the surface, which should be reflected in the obtained XPS spectra.
a)
b)
c)
120
However, as shown in Figure 6-9, little change was detected in the XPS spectra after UV
exposure. These results are likely due to the fact that the oNB moiety doesn’t change the surface
groups to significantly different structures, therefore causing them to appear similar in XPS.
Figure 6-8. Schematic of UV light-induced changes to surface-attached oNB moiety.
121
Figure 6-9. Comparison of X-ray Photoelectron Spectra of oNB-functionalized silica/silicon wafers
before and after 10 minutes of UV light exposure. Binding energy ranges optimized for a) Nitrogen 1s, b)
Carbon 1s, and c) Oxygen 1s.
6.4 Frequency Comb Testing
Following surface functionalization, the optical properties of the microspheres were
characterized using the testing setup pictured in Figure 6-1. More specifically, the quality factor
and the optical power output of the microsphere were measured after Step 2, Step 3, and UV
exposure.
The quality factor Q of a resonator describes its ability to confine and store light
20–22
.
Higher Q values correspond to longer circulating photon lifetimes within the resonant cavity. Q
is an important characteristic for frequency comb generation because long interaction lengths and
high circulating intensities are needed for nonlinear optical effects to occur with reasonably low
a)
c)
b)
122
input powers
7,23
. As will be shown in our results, microresonators with low Q did not exhibit any
parametric frequency conversion.
The quality factor can be calculated using the linewidth of the resonance dip in
transmission through a coupler (Dl) and the central wavelength of the resonance l
0
in the
following equation
𝑄 =
}
5
∆}
(6-1)
Light was coupled from a continuous wave tunable laser (Newport, Velocity TLB6700) into the
microresonator using a tapered optical fiber. Transmission through the fiber was monitored using
a photodetector (Thorlabs) and oscilloscope (National Instruments, NI PCI-5114). A dip in the
transmission indicated that light of a particular wavelength was resonating in the cavity. The dip
was fitted to a Lorenzian curve and the linewidth (Dl) was measured as the full width at half
maximum.
6.4.1 Post-azidation and Post-click Testing
The transmission spectrum for a microsphere after Step 2 (azidation) is shown in Figure
6-10. Mode splitting of the resonance peak was observed in the transmission spectrum due to the
propagation of both clockwise and counterclockwise resonance modes. The Q values of each
mode were Q
left
= 3.32E7 and Q
right
= 3.36E7. These quality factors were calculated by
measuring the quality factor of the device at decreasing coupling strengths and then fitting the
data to a linear slope. These Q values were high enough to observe parametric four wave mixing
in the OSA spectrum.
123
Figure 6-10. Transmission spectrum (left) and optical spectrum analyzer spectrum (right) of a
functionalized microsphere after azidation (Step 2) on resonance.
After testing, the microsphere was functionalized with oNB (Step 3) and re-tested using
the same setup. The transmission and OSA spectra are shown in Figure 6-11. We observed a
change in the Q of both resonance modes: Q
left
decreased (Q
left
= 2.91E7), however Q
right
increased (Q
right
= 4.81E7). The decrease in Q
left
was expected because the additional surface
modifications may have altered the device surface, increasing the amount of loss. Yet, the
increase in Q
right
was unexpected, and potentially due to error in fitting the data for intrinsic Q.
Additionally, even though Q remained in the 10
7
range, four wave mixing was no longer
observed for the microsphere (Figure 6-11, right). We hypothesize that the oNB moiety may be
quenching emission from the sphere due to its conjugated electronic structure. Further testing
with additional samples is needed to confirm this theory.
124
Figure 6-11. Transmission spectrum (left) and optical spectrum analyzer spectrum (right) of a
functionalized microsphere after click chemistry (Step 3) on resonance.
6.4.2 Contaminated Spheres
As detailed in Section 6.3 and shown in Figure 6-5, surface contamination was observed
on many of the spheres after the functionalization process. This contamination on the surface of
the resonant cavity decreased the quality factor of the device
21
. The contaminated microspheres
typically exhibited Q values around 10
4
– 10
5
, and did not show any emission in the OSA
spectrum (Figure 6-12).
Figure 6-12. Transmission spectrum (left) and optical spectrum analyzer spectrum (right) for a
representative contaminated microsphere. Q = 5.88 x 10
5
UV Sensitive Comb Testing
Quality Factors: Sphere 1
1552.032 1552.034 1552.036 1552.038 1552.040 1552.042
0.658
0.752
0.846
0.940
1.034
Transmission
Wavelength (nm)
-56
-28
0
28
1400 1500 1600 1700
-56
-28
0
28
Optical Power (dBm)
Optical Power (dBm)
Wavelength (nm)
• Loaded Q = 5.88x10^5
• Tunable Laser: 1550 nm
125
Chapter 6 References
(1) Udem, T.; Holzwarth, R.; Hänsch, T. W. Optical Frequency Metrology. Nature 2002, 416
(6877), 233–237 DOI: 10.1038/416233a.
(2) Udem, T. Spectroscopy: Frequency Comb Benefits. Nat. Photonics 2009, 3 (2), 82–84
DOI: 10.1038/nphoton.2008.284.
(3) Del’Haye, P.; Schliesser, A.; Arcizet, O.; Wilken, T.; Holzwarth, R.; Kippenberg, T. J.
Optical Frequency Comb Generation from a Monolithic Microresonator. Nature 2007, 450
(7173), 1214–1217 DOI: 10.1038/nature06401.
(4) Del’Haye, P.; Arcizet, O.; Schliesser, A.; Holzwarth, R.; Kippenberg, T. J. Full
Stabilization of a Microresonator-Based Optical Frequency Comb. Phys. Rev. Lett. 2008,
101 (5) DOI: 10.1103/PhysRevLett.101.053903.
(5) Papp, S. B.; Beha, K.; Del’Haye, P.; Quinlan, F.; Lee, H.; Vahala, K. J.; Diddams, S. a.;
Del’Haye, P.; Quinlan, F.; Lee, H.; Vahala, K. J.; Diddams, S. a. Microresonator
Frequency Comb Optical Clock. Optica 2014, 1 (1), 10–14 DOI:
10.1364/OPTICA.1.000010.
(6) Cundiff, S. T.; Ye, J. Colloquium: Femtosecond Optical Frequency Combs. Rev. Mod.
Phys. 2003, 75 (1), 325–342 DOI: 10.1103/RevModPhys.75.325.
(7) Kippenberg, T. J.; Holzwarth, R.; Diddams, S. A. Microresonator-Based Optical
Frequency Combs. Science (80-. ). 2011, 332 (6029), 555–559 DOI:
10.1126/science.1193968.
(8) Savchenkov, A. A.; Matsko, A. B.; Ilchenko, V. S.; Solomatine, I.; Seidel, D.; Maleki, L.
Tunable Optical Frequency Comb with a Crystalline Whispering Gallery Mode Resonator.
Phys. Rev. Lett. 2008, 101 (9) DOI: 10.1103/PhysRevLett.101.093902.
(9) Foster, M. A.; Levy, J. S.; Kuzucu, O.; Saha, K.; Lipson, M.; Gaeta, A. L. Silicon-Based
Monolithic Optical Frequency Comb Source. arXiv 2009, 0912.4890v1 DOI:
10.1364/OE.19.014233.
(10) Klán, P.; Šolomek, T.; Bochet, C. G.; Blanc, A.; Givens, R.; Rubina, M.; Popik, V.;
Kostikov, A.; Wirz, J. Photoremovable Protecting Groups in Chemistry and Biology:
Reaction Mechanisms and Efficacy. Chem. Rev. 2013, 113 (1), 119–191 DOI:
10.1021/cr300177k.
(11) Zhao, H.; Sterner, E. S.; Coughlin, E. B.; Theato, P. O-Nitrobenzyl Alcohol Derivatives:
Opportunities in Polymer and Materials Science. Macromolecules 2012, 45 (4), 1723–
1736.
(12) Gungor, E.; Armani, A. M. Photocleavage of Covalently Immobilized Amphiphilic Block
Copolymer: From Bilayer to Monolayer. Macromolecules 2016, 49 (16), 5773–5781 DOI:
10.1021/acs.macromol.6b01609.
(13) Schumers, J.; Gohy, J.; Fustin, C. A Versatile Strategy for the Synthesis of Block
Copolymers Bearing a Photocleavable Junction. Polym. Chem. 2010, 1 (2), 161 DOI:
10.1039/b9py00218a.
(14) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Click Chemistry: Diverse Chemical Function
from a Few Good Reactions. Angewandte Chemie - International Edition. 2001, pp 2004–
2021.
(15) Kolb, H. C.; Sharpless, K. B. The Growing Impact of Click Chemistry on Drug
Discovery. Drug Discov. Today 2003, 8 (24), 1128–1137 DOI: 10.1016/S1359-
6446(03)02933-7.
126
(16) Moses, J. E.; Moorhouse, A. D. The Growing Applications of Click Chemistry. Chem.
Soc. Rev. 2007, 36 (8), 1249–1262 DOI: 10.1039/b613014n.
(17) Binder, W. H.; Sachsenhofer, R. “Click” Chemistry in Polymer and Materials Science.
Macromolecular Rapid Communications. 2007, pp 15–54.
(18) Liang, L.; Astruc, D. The copper(I)-Catalyzed Alkyne-Azide Cycloaddition (CuAAC)
“click” reaction and Its Applications. An Overview. Coord. Chem. Rev. 2011, 255 (23–
24), 2933–2945 DOI: 10.1016/j.ccr.2011.06.028.
(19) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. A Stepwise Huisgen
Cycloaddition Process: Copper(I)-Catalyzed Regioselective “ligation” of Azides and
Terminal Alkynes. Angew. Chemie - Int. Ed. 2002, 41 (14), 2596–2599 DOI:
10.1002/1521-3773(20020715)41:14<2596::AID-ANIE2596>3.0.CO;2-4.
(20) Kippenberg, T. J.; Spillane, S. M.; Armani, D. K.; Min, B.; Yang, L.; Vahala, K. J.
Fabrication, Coupling and Nonlinear Optics of Ultra-High-Q Micro-Sphere and Chip-
Based Toroid Microcavities. Opt. Microcavities 2004, 177–238 DOI:
10.1142/9789812565730_0005.
(21) Gorodetsky, M. L.; Savchenkov, a a; Ilchenko, V. S. Ultimate Q of Optical Microsphere
Resonators. Opt. Lett. 1996, 21 (7), 453–455 DOI: 10.1364/OL.21.000453.
(22) Zhang, X.; Choi, H. S.; Armani, A. M. Ultimate Quality Factor of Silica Microtoroid
Resonant Cavities. Appl. Phys. Lett. 2010, 96 (15) DOI: 10.1063/1.3398008.
(23) Braginsky, V. B.; Gorodetsky, M. L.; Ilchenko, V. S. Quality-Factor and Nonlinear
Properties of Optical Whispering-Gallery Modes. Phys. Lett. A 1989, 137 (7–8), 393–397
DOI: 10.1016/0375-9601(89)90912-2.
127
Chapter 7. Future Work
The work presented in this thesis lays the foundation for novel applications of the ortho-
nitrobenzyl moiety. The previous chapters detail the synthetic routes and characteristic behavior
of oNB-modified macromolecules, leaving much room open for exploration of the application of
these techniques and knowledge. In this chapter, we propose potential future projects for further
probing the behavior and application potential of oNB-modified materials.
7.1 Solvent-polymer interaction
In Chapter 4, we investigated the effects of solvent environment on photocleavage
kinetics of an oNB-containing polymer (PMA-oNB). In this work, we used tetrahydrofuran,
toluene, and chloroform as solvents for the reaction. In the future, it would be interesting to
explore the effects of using additional solvents with a wider range of interaction parameters (c)
to more fully characterize the effect of solvent on oNB photocleavage response. With a wide
enough range of solvents, one could develop a predictive model for how solvent affects the
photocleavage rate based on specific solvent-polymer parameters.
Previous work has shown that the photocleavage rate of the o-nitrobenzyl moiety alone
(without polymer) can be accelerated or slowed by changing the solvent
1–3
. Most of these studies
were conducted in aqueous media and focused on the effect of pH. They found that different pH
environments can stabilize or destabilize the intermediate states between the uncleaved and fully
cleaved oNB derivative. Performing a similar study with oNB- polymers would require the use
of ultrafast spectroscopy techniques to characterize the reaction intermediate states in detail
1
.
Appendix A starts this study through the use of additional solvents and polymers. By
changing the polymer, one is also changing the c value even if the solvent remains the same.
Investigations into the solvent-polymer interaction could lead to better control in designing
128
photocleavage reactions in the future. Applications where the photodegradation rate is important
include cascaded triggered release, hydrogel cell matrices and photoinduced micropatterning
4–6
.
7.2 Structural control of azobenzene isomerization
In Chapter 2, we described the dimerization of nitrosobenzaldehyde photoreaction
products into azobenzene, a visible light absorbing compound that is commonly used in organic
dyes. These azobenzene dimers form more readily in polymer films than in solution. In Chapter
5, we used the color change caused by the formation of azobenzene dimers as an indicator of UV
light exposure. As a focus of future studies, we propose the characterization of the effect of
polymer environment on azobenzene isomerization
7
.
Azobenzene can exist in two conformations: the trans conformation and the cis
conformation (Figure 7-1)
8,9
. The trans isomer has a near-planar conformation and a dipole
moment near zero
10
. As a result, it is more stable than the bent cis conformation and found at
much higher concentrations (>99%) at ambient conditions
11
. The cis conformation can be
induced by irradiation with UV light (l = 340nm). Its half-life varies greatly from milliseconds
to days depending on the substituents of the benzene ring
11
. The trans isomer can be regenerated
upon irradiation with visible light (l = 450nm) or via thermal relaxation over time upon removal
of the UV light.
Figure 7-1. Structures of trans and cis isomers of azobenzene. Isomerization is typically activated by
light; however, the cis isomer will convert to trans over time.
129
The trans and cis isomers have differing absorbance spectra spanning the visible and UV
regions, and thus will appear different colors
11,12
. In general, the trans isomer appears orange and
the cis isomer appears yellow. We observed these color changes in the UV dosimeter after
samples had been stored in the dark for longer than one day. Initially, the UV-exposed
dosimeters were yellow on the same day of exposure. Over time, the yellow gradually changed
to a dark orange color, with a noticeable color change occurring in as little as 24 hours.
There are a number of variables that could potentially affect the half-life of the cis
isomer, for example the polymer molecular weight, composition, and film thickness. It would be
interesting to study which variables have an effect on the thermal relaxation and why. The results
of this research could inform better design of azobenzene-based photoswitches and enable
applications where a longer cis isomer stable state is needed.
7.3 Fluorescent UV Sensor
In Chapter 3, we showed that oNB-containing polymers fluoresce bright blue after
extended exposure to UV light. The fluorescence was observed in both solution and as a film. An
interesting follow-up project to this observation would be the development of a blue light-
emitting UV sensor.
Our goal in characterizing the fluorescent emission was to create such a sensor, however
we encountered multiple problems. Fluorescent emission only weakly occurred in the polymer
film, whereas the emission was very bright in solution. Ideally, the UV sensor would not contain
any liquid components, so the emission in the solid state would need to be enhanced. We found
that using a thicker sample caused the blue color to appear deeper, likely due to the increased
concentration of fluorophores. Additionally, the UV sensor would fluoresce in response to low
levels of UV light. Currently, emission was only observed upon significantly high intensities of
130
UV light exposure. Optimization of emission process could potentially focus on decreasing the
fluorescent quenching that is likely occurring in the polymer film due to the conjugated nature of
the molecules in the film. In solvent-based studies of similarly fluorescing molecules, the
environment was shown to have an effect on the quantum yield
13
.
7.4 Expanding the capabilities of the wearable UV dosimeter
In Chapter 5, we detailed the development of a wearble UV dosimeter based on the UV
dose-dependent formation of colored azobenzene moieties. We showed that the dosimeter could
be modified with an additional UV filter (sunscreen) to change the sensitivity of its response.
Additional research could be done in combining the dosimeter with other types of optical filters
to controllably modulate the sensitivity. Having a range of different response sensitivities would
be extremely useful in tailoring the UV dosimeter to different populations of people. Some
groups of people are at a higher risk than others for UV-induced skin damage, for example
patients undergoing chemotherapy or taking medications that increase their sun sensitivity. In
these cases, being able to tune the response of the UV dosimeter to be more sensitive would be
better able to prevent skin damage.
Additionally, the color change itself could be further optimized. In Chapter 5, we showed
a color change of the dosimeter from clear to yellow/ orange. However, yellow/ orange is not
immediately visible on all skin types. A more suitable alert color would be one that is instantly
noticeable, for example green or blue. Previous studies have shown that the substituents in the
ortho- and para- positions in oNB can change the photocleavage behavior, including the
absorption spectra
1,3,14,15
. Adding different substitutuents at these benzylic positions could
potentially change the color of the final cleaved moiety.
131
7.5 Sunscreen efficacy indicator
Commercially available sunscreens vary greatly in their sun protection capabilities, and
sometimes do not provide adequate UV protection
16–18
. These inconsistencies can make it
difficult for a consumer to make an informed choice in sunscreen purchasing or, in the latter
case, fail to ensure appropriate skin protection. We propose that the UV dosimeter presented in
Chapter 5 can serve as a sunscreen efficacy tester to help the consumer more appropriately
compare different sunscreen formulations. In Chapter 5, we showed how our wearable UV
dosimeter changed response with the application of a topical sunscreen. Specifically, the
sunscreen-coated sensor color changed slower in comparison to the bare sensor, showing the
UV-blocking effect of the sunscreen. From these results, we can extrapolate that the dosimeter
could be used to test sunscreen efficiency. A potential project would include testing various
brands and compositions of sunscreen using the UV dosimeter to measure how quickly the
sunscreen degraded or stopped providing protection. A more effective sunscreen would prevent
color change from occurring over a longer period of time.
Chapter 7 References
(1) Il’ichev, Y. V; Schwörer, M. A.; Wirz, J. Photochemical Reaction Mechanisms of 2-
Nitrobenzyl Compounds: Methyl Ethers and Caged ATP. J. Am. Chem. Soc. 2004, 126
(14), 4581–4595 DOI: 10.1021/ja039071z.
(2) Klán, P.; Šolomek, T.; Bochet, C. G.; Blanc, A.; Givens, R.; Rubina, M.; Popik, V.;
Kostikov, A.; Wirz, J. Photoremovable Protecting Groups in Chemistry and Biology:
Reaction Mechanisms and Efficacy. Chem. Rev. 2013, 113 (1), 119–191 DOI:
10.1021/cr300177k.
(3) Bochet, C. Photolabile Protecting Groups and Linkers. Journal of the Chemical Society,
Perkin Transactions 1. 2001, pp 125–142.
(4) Piloto, A. M.; Costa, S. P. G.; Gonçalves, M. S. T. Wavelength-Selective Cleavage of O-
Nitrobenzyl and Polyheteroaromatic Benzyl Protecting Groups. Tetrahedron 2014, 70 (3),
650–657 DOI: 10.1016/j.tet.2013.11.100.
(5) Kloxin, A. M.; Tibbitt, M. W.; Anseth, K. S. Synthesis of Photodegradable Hydrogels as
Dynamically Tunable Cell Culture Platforms. Nat. Protoc. 2010, 5 (12), 1867–1887 DOI:
132
10.1038/nprot.2010.139.
(6) Li, L.; Deng, X.; Li, Z.; Du, F.; Li, Z. Multifunctional Photodegradable Polymers for
Reactive Micropatterns. Macromolecules 2014.
(7) Mita, I.; Horie, K.; Hirao, K. Photochemistry in Polymer Solids. 9. Photoisomerization of
Azobenzene in a Polycarbonate Film. Macromolecules 1989, 22, 558–563 DOI:
10.1021/ma00192a008.
(8) Kumar, G. S.; Neckers, D. C. Photochemistry of Azobenzene-Containing Polymers.
Chem. Rev. 1989, 89 (8), 1915–1925 DOI: 10.1021/cr00098a012.
(9) Bandara, H. M. D.; Burdette, S. C. Photoisomerization in Different Classes of
Azobenzene. Chem. Soc. Rev. 2012, 41 (5), 1809–1825 DOI: 10.1039/c1cs15179g.
(10) Fliegl, H.; Kohn, A.; Hattig, C.; Ahlrichs, R. Ab Initio Calculation of the Vibrational and
Electronic Spectra of Trans- and Cis-Azobenzene. J. Am. Chem. Soc. 2003, 125 (32),
9821–9827.
(11) Beharry, A. A.; Woolley, G. A. Azobenzene Photoswitches for Biomolecules. Chem Soc
Rev 2011, 40 (8), 4422–4437 DOI: 10.1039/c1cs15023e.
(12) Beharry, A. A.; Sadovski, O.; Woolley, G. A. Azobenzene Photoswitching without
Ultraviolet Light. J. Am. Chem. Soc. 2011, 133 (49), 19684–19687 DOI:
10.1021/ja209239m.
(13) Cheng, S.; Song, P.; Yang, S.; Yin, H.; Han, K. Fluorescence and Solvent-Dependent
Phosphorescence Studies of O-Nitrobenzaldehyde: A Combined Experimental and
Theoretical Investigation. Phys. Chem. Chem. Phys. 2010, 12 (31), 9067–9074 DOI:
10.1039/c002270e.
(14) Zhao, H.; Sterner, E. S.; Coughlin, E. B.; Theato, P. O-Nitrobenzyl Alcohol Derivatives:
Opportunities in Polymer and Materials Science. Macromolecules 2012, 45 (4), 1723–
1736.
(15) Guillier, F.; Orain, D.; Bradley, M. Linkers and Cleavage Strategies in Solid-Phase
Organic Synthesis and Combinatorial Chemistry. Chem. Rev. 2000, 100 (6), 2091–2157
DOI: 10.1021/cr980040+.
(16) Haywood, R. Relevance of Sunscreen Application Method, Visible Light and Sunlight
Intensity to Free-Radical Protection: A Study of Ex Vivo Human Skin. Photochem.
Photobiol. 2006, 82 (4), 1123–1131 DOI: 10.1562/2006-02-08-RA-799.
(17) Wright, M. W.; Wright, S. T.; Wagner, R. F. Mechanisms of Sunscreen Failure. J. Am.
Acad. Dermatol. 2001, 44 (5), 781–784 DOI: 10.1067/mjd.2001.113685.
(18) Dennis, L. K.; Freeman, L. E. B.; VanBeek, M. J. Sunscreen Use and the Risk for
Melanoma: A Quantitative Review. Annals of Internal Medicine. 2003, p 966–978+I16.
133
Appendix A. oNB Photocleavage with Different Polymers
In the photocleavage kinetic studies described in Chapter 4, we presented results for a
single polymer: poly(methyl acrylate) (PMA). Over the course of our research, several other
polymers were also synthesized using the oNB initiator. This Appendix describes the synthetic
details for these polymers as well as preliminary kinetic testing for oNB-containing polystyrene
photocleavage. The synthetic experiments and kinetic testing were performed in collaboration
with Martin Siron. By using different types of polymers, our aim was to vary the polymer-
solvent interaction parameter over a wider range of values and develop a more comprehensive
model of the effect of solvent interaction on polymer photocleavage
1,2
.
Additionally, some of the polymers described in this section (poly(MPEGMA), PHEMA)
are water-soluble
3,4
. Integrating photocleavable functionality into a water-soluble system widens
the range of potential applications, particularly in the biomedical field
5–7
.
A.1 Alternative monomers for ATRP
A.1.1 Polystyrene (PS) ATRP
ATRP of styrene followed a similar reaction to PMA (Aldrich, 99.9%, stored at -5°C)
8
.
Prior to the reaction, styrene was allowed to warm to room temperature. Styrene (300 equiv.)
was then added to a schlenk flask with oNB initiator (1 equiv), PMDETA (0.5 equiv), and
anisole (10 % volume). The flask was degassed using three freeze-pump-thaw cycles and CuBr
catalyst (0.5 equiv) was added while the solution was frozen. The flask was thawed, back-filled
with argon and placed in a 100°C oil bath to initiate the reaction. The reaction was stopped after
4 hours and 40 minutes by opening the flask to air and adding THF. The resulting conversion
was 12.7% by NMR and the molecular weight was M
n
= 4500 Da, PDI = 1.01. The polymer was
precipitated in methanol and dried under vacuum.
134
Figure A-1. Reaction Scheme for ATRP of Styrene with oNB initiator.
Verification of the oNB placement in the center of the polymer chain was done by
comparing the GPC chromatograms before and after UV exposure in a photoreactor. First, we
deconvoluted the GPC spectra by fitting it to a bi-Gaussian model. The two peaks corresponding
to cleaved and uncleaved PS are shown in Figure A-2. There is a clear shift in the molecular
weight from approximately 5 kDa to 2.5 kDa after 1 hour of UV exposure. The reduction to half
the original molecular weight confirms the placement of oNB in the center of the chain.
Figure A-2. Comparison of deconvoluted peaks from GPC chromatograms of PS before (black) and after
(red) UV exposure in THF. There is a clear shift in molecular weight from 5 kDa to 2.5 kDa.
135
A.1.2 PHEMA ATRP
We also used 2-hydroxyethyl methacrylate (HEMA) as a monomer in an ATRP
procedure that is similar to the procedure for PMA and PS
3
(Figure A-3). HEMA (100 equiv.)
was added to a schlenk flask with oNB initiator (1 equiv), PMDETA (2 equiv), and propanol
(75 % volume). The schlenk flask was degassed using three freeze-pump-thaw cycles and CuBr
catalyst (2 equiv) was added while the solution was frozen. The flask was further degassed and
backfilled with Ar gas. The flask was thawed and mixed at 90°C for 5 hours. After this time, the
solution was very viscous, indicating that the polymer may have cross-linked or grown too large
in molecular weight. We were unable to remove the polymer from solution or dissolve enough in
THF to obtain a GPC sample.
Figure A-3. Reaction scheme for ATRP of 2-hydroxyethyl methacrylate (HEMA) with oNB initiator.
A.1.3 MPEGMA ATRP
As a further attempt to synthesize a water soluble polymer, we used poly(ethylene glycol)
methyl ether methacrylate (MPEGMA, avg M
n
= 300, Aldrich) as a monomer. Prior to the
reaction, MPEGMA was allowed to warm to room temperature. MPEGMA (50 equiv.) was then
added to a schlenk flask with oNB initiator (1 equiv), PMDETA (2 equiv), and anisole (50 %
volume). The schlenk flask was degassed using three freeze-pump-thaw cycles and CuBr catalyst
(2 equiv) was added while the solution was frozen. The flask was thawed, filled with argon, and
136
placed in a 60°C oil bath for 1.5 hourS. The polymer was precipitated in cold methanol and dried
under vacuum.
Figure A-4. Reaction scheme for ATRP of poly(ethylene glycol) methyl ether methacrylate (MPEGMA)
with oNB initiator.
We checked the oNB placement along the polymer chain by comparing the GPC
chromatograms before and after 30 minutes of UV exposure in a photoreactor. The polymer
molecular weight reduced from 20 kDa to 14.5 kDa, indicating that the oNB group was not
centrally located.
Figure A-5. Comparison of GPC molecular weight spectra of MPEGMA before (black) and after (red) 30
minutes of UV exposure
Molecular Weight (kDa)
0 20 40 60 80 100
137
A.2 Polystyrene Photocleavage Kinetics
In Chapter 4, we characterized the relationship between polymer-solvent interaction
parameter c and photocleavage rate by changing the solvent environment. In this section, we
varied c by changing the monomer used to synthesize the photocleavable polymer. Polymer
samples of 0.05mM were prepared in acetone and exposed to UV light for a range of times using
a photoreactor. For reference, c
PMA-acetone
= 0.482 and c
PS-acetone
= 0.6
1
. The photocleavage of
oNB-containing PS was compared with oNB-containing PMA by generating kinetic curves
similar to those presented in Chapter 4. In Figure A-6, we show that the kinetic curves are very
similar for both PS and PMA in acetone. While we cannot draw a conclusion yet from this single
set of data, it provides an interesting starting point for future work.
Figure A-6. Comparison of photocleavage kinetics of oNB-containing PS and PMA. Both polymers
exhibit similar sigmoidal behavior.
PS
PMA
138
Appendix A References
(1) Orwoll, R. a; Arnold, P. a. Polymer – Solvent Interaction Parameter X. Phys. Prop.
Polym. Handb. 2007, 50 (3), 451 DOI: 10.1007/978-0-387-69002-5_14.
(2) Masaro, L.; Zhu, X. X. Physical Models of Diffusion for Polymer Solutions, Gels and
Solids; 1999; Vol. 24.
(3) Beers, K. L.; Boo, S.; Gaynor, S. G.; Matyjaszewski, K. Atom Transfer Radical
Polymerization of 2-Hydroxyethyl Methacrylate. Macromolecules 1999, 32 (18), 5772–
5776 DOI: Doi 10.1021/Ma990176p.
(4) Yu, Q.; Zeng, F.; Zhu, S. Atom Transfer Radical Polymerization of Poly(ethylene Glycol)
Dimethacrylate. Macromolecules 2001, 34 (6), 1612–1618 DOI: 10.1021/ma001665o.
(5) Casadio, Y. S.; Brown, D. H.; Chirila, T. V.; Kraatz, H. B.; Baker, M. V. Biodegradation
of poly(2-Hydroxyethyl Methacrylate) (PHEMA) and poly{(2-Hydroxyethyl
Methacrylate)-Co-[Poly(ethylene Glycol) Methyl Ether Methacrylate]} Hydrogels
Containing Peptide-Based Cross-Linking Agents. Biomacromolecules 2010, 11 (11),
2949–2959 DOI: 10.1021/bm100756c.
(6) Schramm, O. G.; Pavlov, G. M.; Van Erp, H. P.; Meier, M. A. R.; Hoogenboom, R.;
Schubert, U. S. A Versatile Approach to Unimolecular Water-Soluble Carriers: ATRP of
PEGMA with Hydrophobic Star-Shaped Polymeric Core Molecules as an Alternative for
PEGylation. Macromolecules 2009, 42 (6), 1808–1816 DOI: 10.1021/ma8024738.
(7) Sun, L.; Baker, G. L.; Bruening, M. L. Polymer Brush Membranes for Pervaporation of
Organic Solvents from Water. Macromolecules 2005, 38 (6), 2307–2314 DOI:
10.1021/ma047510o.
(8) Matyjaszewski, K.; Xia, J. Atom Transfer Radical Polymerization. Chem. Rev. 2001, 101
(9), 2921–2990.
139
Appendix B. Electro-optically Tunable Waveguides
B.1 Lithium Niobate
Due to its unique crystal structure, lithium niobate (LiNbO
3
) has been shown to display
piezoelectricity, the electro-optic effect, and nonlinear optical polarizability, making it an
attractive compound for use in the optics industry
1–3
. At room temperature, the crystal structure
of lithium niobate is composed of planar sheets of oxygen atoms in a distorted hexagonal close-
packed configuration. This structure exhibits three-fold rotation symmetry about the c-axis,
classifying it as part of the trigonal crystal system
4
. Upon application of an electric field,
fluctuations in the electric displacement of the material are observed to induce changes in the
refractive index throughout the material
2
. Due to this observed electro-optic effect, lithium
niobate has been used in many different types of external modulators for fiber-optic transmission
systems.
While much research has focused on the characterization of lithium niobate single
crystals and single crystal thin films, little research has evolved in the study of nanocrystalline
lithium niobate. It has been shown that nanocrystalline lithium niobate exhibits very different
transport properties than the bulk material
5
. In addition, lithium niobate thin films have a higher
power density than bulk wave guides, making this field a very interesting area of study.
In this project, we aimed to combine the electro-optic properties of lithium niobate with a
low-loss material like silica to create electrically switchable optical devices. The ability to
electrically control optical properties in a low-loss medium has many potential applications in
integrated optical circuits and telecommunications systems. Previously, our group has shown that
doping inorganic elements into a silica matrix during a sol-gel process results in a material that
140
reflects the combined properties of both components
6,7
. We utilized a similar sol-gel
incorporation method to introduce lithium niobate into silica.
B.2 Lithium Niobate Nanoparticle Synthesis
Nanoparticles of lithium niobate were synthesized using a sol-gel double alkoxide
procedure
8,9
. Niobium ethoxide and lithium ethoxide were refluxed in ethanol at 80°C overnight,
then water was added to hydrolyze the solution and begin the gelling process. Tetraethyl
orthosilicate (TEOS) was added to some of the samples at the outset of this reaction to form
silica-coated nanoparticles, which would be more compatible with a silica film
9
. The resulting
gel was dried and ground using a mortar and pestle. Nanoparticles were formed through high
temperature calcination. A schematic of the steps for this procedure is shown in Figure 3-1.
Figure B-1. Overview of steps in sol-gel synthesis of silica-coated lithium niobate nanoparticles.
141
The calcination temperature and time determines the crystallite size. Calcination at higher
temperatures has been shown to increase crystalline content
9
. We characterized the crystalline
content of our samples using powder X-Ray Diffraction (XRD) and the results are shown in
Figure B-2. We noticed that the spectra of samples with TEOS added during the sol-gel
procedure (TEOS LN) exhibited overall peak broadening when compared to samples without
TEOS. This peak broadening can be correlated to crystallite size τ using the Scherrer Equation:
𝜏 =
9.}
(B.1)
where λ is the X-ray wavelength, L is the full width at half max of the XRD peaks, and θ is the
Bragg angle. (Figure B-2). Because the crystallite size (often equated to grain size) is inversely
related to XRD peak width, we concluded that the broadened peaks of TEOS LN indicate the
restriction of grain growth during the calcination process. From Eq B.1, we calculated that the
crystallite sizes of lithium niobate particles with TEOS and without TEOS were 12nm and 20nm
respectively.
Figure B-2. XRD spectra for synthesized lithium niobate (red) and silica-coated lithium niobate (black)
nanoparticles. Each peak is labeled with its corresponding crystallographic planes.
(012)
(104)
(110)
(006)
(202)
(024)
(116)
(122)
(214)
142
We further characterized the size and structure of the TEOS coated lithium niobate
nanoparticles by imaging them using Transmission Electron Microscopy, TEM (JEOL JEM 2100
LaB6). Samples were dispersed in ethanol prior to imaging. Because there were no surfactants in
solution, the nanoparticles aggregated together, making them difficult to image individually.
Figure B-3 shows that at high magnification, the lattice fringe from the lithium niobate
crystalline structure can be observed. These sections of lattice fringe further verify that
crystalline lithium niobate was formed during the calcination process. The clusters of lattice
fringe range in size from 10 - 25 nm, which agrees with the grain size calculated using XRD.
Additionally, the silica coating can be observed around the edges of the lithium niobate particle
clusters. This coating around the outside of groups of nanoparticles instead of single
nanoparticles indicates that the addition of TEOS needs to be added at a different step in the
process in order to stabilize the nanoparticles.
143
Figure B-3. TEM Image of silica-coated lithium niobate nanoparticle clusters. Scale bar = 20nm.
B.3 Hybrid Silica-Lithium Niobate Films
The silica sol-gel was prepared using an existing procedure
7
, whereby a metal alkoxide
precursor (tetraethylorthosilicate, or TEOS) is hydrolyzed by an acid catalyst (HCl) in a solution
of water and ethanol. Upon aging of this solution, cross-links between siloxane groups will
develop to form a porous silica glass network. We can introduce dopants to this network by
adding compounds to the liquid precursor before the sol-gel network becomes too dense. Lithium
niobate (LN) nanocrystals were introduced to the sol-gel mixture at a ratio of 1:1000 LN:TEOS
Lattice Fringe
Silica Coating
144
at 5 minutes after the acid catalyst is added, and allowed to age for 24-48 hours at room
temperature. After the aging process, the sol-gels were spin coated onto a silicon substrate and
annealed at 1000°C for 1 hour, resulting in films approximately 350nm thick. As a preliminary
study, films were made with commercially synthesized LN (Aldrich) that was ground into a
powder via high energy ball-milling (Zr milling balls, 1h) at HRL Laboratories (Figure B-4).
Figure B-4. Fabrication steps of hybrid lithium niobate-silica films.
The presence of lithium niobate in the silica films was verified through polarized light
microscopy and energy dispersive X-ray spectroscopy (EDX). Because of the crystalline
structure of lithium niobate, the interaction of light with the compound is directionally
dependent. We utilized this anisotropic property to verify the presence of LN nanoparticles by
viewing films under polarized and non-polarized light (Figure B-5).
145
Figure B-5. Difference in appearance of particles with (bottom row) and without (top row) the
polarization filter. a), b) pure lithium niobate powder; c), d) dust particle in pure TEOS film; e), f) lithium
niobate particle in lithium niobate doped TEOS film.
Energy-dispersive X-Ray Spectroscopy (EDX) was also used to compositionally identify
the lithium niobate in the sol-gel films. Figure B-6 shows the composition of the nanoparticle
sample as determined through analysis of the characteristic X-rays emitted at that point.
Samples were also tested for Zirconium contamination, which may have occurred through
abrasion of the ball milling media, but no zirconium was detected.
146
Figure B-6. Left: EDX analysis of characteristic X-rays emitted at sample point designated by red cross
(inset). The characteristic peak for Nb at 2.2 keV verifies that the particle seen on the surface is lithium
niobate. Right: SEM micrograph of lithium niobate nanoparticles embedded in silica film. Scale bar is
20µm.
Several problems were encountered in creating a uniform size distribution and a
homogeneous spatial distribution of LN nanoparticles within the silica film. As shown in the
SEM image in Figure B-6, nanoparticle sizes range from 0.2-5µm. Exact particle sizing was
attempted using Dynamic Light Scattering (DLS), however the nanoparticles did not stay in the
carrier solution (ethanol) long enough for measurements to be collected. We hypothesize that the
agglomeration of nanoparticles causes them to fall out of solution, also affecting their spatial
distribution during spin coating.
The silica-coated lithium niobate nanoparticles were not integrated into a sol-gel and
would thus be an interesting future project. We anticipate that the silica coating may help
stabilize the nanoparticles in solution, facilitating a more homogeneous final distribution in the
film.
SE1 2µm
Matrix: 256x200
Data Type: SE1(ADC)
Magnification: 4000x
Image Size: 0.0309x0.0241mm
kV: 5.0
Tilt: 0
C:\EDAX32\GENESIS\GENMAPS.SPC
kV:5.0 Tilt:0.00 Tkoff:34.32 Det:Det:SUTW Reso:132.6 Amp.T:51.20
FS : 3231 LSec : 14.1 Prst:None 27-Aug-2012 15:47:54
0.4k
0.8k
1.2k
1.6k
2.0k
2.4k
2.8k
Counts
C
O
Si
Zr
Nb
0.40 0.80 1.20 1.60 2.00 2.40 2.80 3.20 3.60 keV
SE1 2µm
Matrix: 256x200
Data Type: SE1(ADC)
Magnification: 4000x
Image Size: 0.0309x0.0241mm
kV: 5.0
Tilt: 0
C:\EDAX32\GENESIS\GENMAPS.SPC
kV:5.0 Tilt:0.00 Tkoff:34.32 Det:Det:SUTW Reso:132.6 Amp.T:51.20
FS : 3231 LSec : 14.1 Prst:None 27-Aug-2012 15:47:54
0.4k
0.8k
1.2k
1.6k
2.0k
2.4k
2.8k
Counts
C
O
Si
Zr
Nb
0.40 0.80 1.20 1.60 2.00 2.40 2.80 3.20 3.60 keV
147
Appendix B References
(1) Wooten, E.; Kissa, K. A Review of Lithium Niobate Modulators for Fiber-Optic
Communications Systems. Sel. Top. … 2000, 6 (1), 69–82.
(2) Chen, F. S.; Lamacchia, J. T.; Fraser, D. B. Holographic Storage in Lithium Niobate.
Appl. Phys. Lett. 1968, 13 (7), 223–225 DOI: 10.1063/1.1652580.
(3) Lawrow, a.; Pannell, C. N.; Negoita, M.; Russell, P. S.; Webjorn, J. Focused Acoustic
Wave Acousto-Optic Device Using a Planar Domain-Inverted Lithium Niobate
Transducer. 1997, 8 (97), 6–9 DOI: 10.1016/S0030-4018(97)00458-6.
(4) Weis, R.; Gaylord, T. Lithium Niobate: Summary of Physical Properties and Crystal
Structure R. Appl. Phys. A Mater. Sci. Process. 1985, 37 (4), 191–203 DOI:
10.1007/BF00614817.
(5) An, C.; Tang, K.; Wang, C.; Shen, G.; Jin, Y.; Qian, Y. Characterization of LiNbO3
Nanocrystals Prepared via a Convenient Hydrothermal Route. Mater. Res. Bull. 2002, 37
(11), 1791–1796 DOI: 10.1016/S0025-5408(02)00869-3.
(6) Maker, A. J.; Rose, B. a.; Armani, A. M. Tailoring the Behavior of Optical Microcavities
with High Refractive Index Sol-Gel Coatings. Opt. Lett. 2012, 37 (14), 2844 DOI:
10.1364/OL.37.002844.
(7) Rose, B. a.; Maker, A. J.; Armani, A. M. Characterization of Thermo-Optic Coefficient
and Material Loss of High Refractive Index Silica Sol-Gel Films in the Visible and near-
IR. Opt. Mater. Express 2012, 2 (5), 671 DOI: 10.1364/OME.2.000671.
(8) Pooley, M. J., Chadwick, A. V. The Synthesis and Characterisation of Nanocrystalline
Lithium Niobate. Radiat. Eff. defects solids 2003, 158, 197–201.
(9) Chadwick, a. V.; Pooley, M. J.; Savin, S. L. P. Lithium Ion Transport and Microstructure
in Nanocrystalline Lithium Niobate. Phys. Status Solidi 2005, 2 (1), 302–305 DOI:
10.1002/pssc.200460170.
Abstract (if available)
Abstract
Stimuli-responsive materials are a rapidly growing area of research, with applications that range from microfabrication to personalized medicine. Also known as smart materials, they are engineered to exhibit a controlled property change in response to an external stimulus. Smart materials are particularly well suited for the field of sensing because they can respond to very small changes in their environment. The focus of this dissertation is to synthesize, characterize, and develop applications for optically responsive smart polymer systems. ❧ In the first part of this dissertation, a novel UV-cleavable polymer is synthesized. The UV response is derived from the photocleavage of a centrally integrated ortho-nitrobenzyl (oNB) moiety. When UV light is applied to the polymer, it breaks in half where the oNB moiety is located. In the second part of this dissertation, the photocleavage rate behavior of these polymers is characterized with respect to environment and molecular weight. The photocleavage is studied in three different solvents and as a solid thin film. In the last part of this dissertation, two different applications of the ortho-nitrobenzyl-based photocleavable system are presented: a wearable UV dosimeter for preventing skin cancer and an optically tunable frequency comb.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Development of organic-inorganic optical microcavities for studying polymer thin films
PDF
Development of hybrid microsensors for environmental monitoring and biodetection
PDF
Engineering solutions for biomaterials: self-assembly and surface-modification of polymers for clinical applications
PDF
Materials development and characterization of optically-active core-shell nanomaterials
PDF
Surface-enhanced on-chip whispering gallery mode resonators for improved nonlinear optics
PDF
Nonlinear optical nanomaterials in integrated photonic devices
PDF
Structure and behavior of nano metallic multilayers under thermal and mechanical loading
PDF
Fabrication and characterization of yoroidal resonators for optical process improvement
PDF
Non-traditional architectures for semiconducting polymers
PDF
Development of novel optical materials for whispering gallery mode resonator for nonlinear optics
PDF
Perovskite chalcogenides: emerging semiconductors for visible to infrared optoelectronics
PDF
Smart buildings: synergy in structural control, structural health monitoring and environmental systems
PDF
Optical simulation and development of novel whispering gallery mode microresonators
PDF
Real time surface analysis of complex oxide thin films during pulsed laser deposition
PDF
Direct C−H arylation for the synthesis of conjugated polymers
PDF
Synthesis and characterization of 3-hexylesterthiophene based random and semi-random polymers and their use in ternary blend solar cells
PDF
In situ process monitoring for modeling and defect mitigation in composites processing
PDF
Developing improved silica materials and devices for integrated optics applications
PDF
Formation of polymer gels, films, and particles via initiated chemical vapor deposition onto liquid substrates
PDF
Mechanics and additive manufacturing of bio-inspired polymers
Asset Metadata
Creator
Lee, Michele Elizabeth
(author)
Core Title
Optically triggered smart polymers for environmental monitoring
School
Viterbi School of Engineering
Degree
Doctor of Philosophy
Degree Program
Materials Science
Publication Date
04/12/2017
Defense Date
03/02/2017
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
cancer prevention,light-responsive,OAI-PMH Harvest,photocleavable,skin cancer,smart polymer,stimuli-responsive,UV radiation,UV sensor,wearable sensor
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Armani, Andrea (
committee chair
), Ravichandran, Jayakanth (
committee member
), Thompson, Barry (
committee member
), Thompson, Mark (
committee member
)
Creator Email
lee853@usc.edu,michele.elizabeth.lee@gmail.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c40-352783
Unique identifier
UC11255609
Identifier
etd-LeeMichele-5172.pdf (filename),usctheses-c40-352783 (legacy record id)
Legacy Identifier
etd-LeeMichele-5172.pdf
Dmrecord
352783
Document Type
Dissertation
Rights
Lee, Michele Elizabeth
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
cancer prevention
light-responsive
photocleavable
skin cancer
smart polymer
stimuli-responsive
UV radiation
UV sensor
wearable sensor