University of Southern California Dissertations and Theses

Vapor phase deposition of dense and porous polymer coatings and membranes for increased sustainability and practical applications

(USC Thesis Other)

Vapor phase deposition of dense and porous polymer coatings and membranes for increased sustainability and practical applications

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Vapor Phase Deposition of Dense and

Porous Polymer Coatings and Membranes for Increased

Sustainability and Practical Applications

by

Nareh Movsesian

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the Requirements for the Degree
DOCTOR OF PHILOSOPHY
(CHEMICAL ENGINEERING)

August 2020

ii
Acknowledgements

I would like to thank my advisors Dr. Malancha Gupta and Dr. Noah Malmstadt for their
continuous guidance and support throughout my PhD studies. I am extremely grateful to be able
to work with such motivating and knowledgeable individuals. I would also like to express my
gratitude to my other committee member, Dr. Maral Mousavi for serving on my defense.
Moreover, I would like to thank my colleagues and labmates Dr. Golnaz Dianat and Dr. Lu Wang
for training and getting me started in both labs. My PhD journey was very pleasant because of you.
Finally, I would like to thank my family, specially my husband and my parents for their
unwavering support and encouragement throughout my academic journey. To my husband, thank
you for being the best support system. This would not be possible without you.

iii
Table of Contents
Acknowledgements ....................................................................................................................... ii
List of Tables ................................................................................................................................ vi
List of Figures ............................................................................................................................. vii
Abstract ....................................................................................................................................... xii
Chapter 1: Introduction ............................................................................................................... 1
1.1 Initiated Chemical Vapor Deposition of Dense Polymer Films .................................. 1
1.2 Macroporous Polymer Membranes by Vapor Deposition ........................................... 3
1.3 References ........................................................................................................................ 9
Chapter 2: Robust Vapor-Deposited Antifouling Fluoropolymer Coatings for Stainless Steel
Polymerization Reactor Components ....................................................................................... 15
2.1 Abstract ............................................................................................................................ 15
2.2 Introduction ..................................................................................................................... 16
2.3 Experimental Section ...................................................................................................... 19
2.3.1 Materials .................................................................................................................. 19
2.3.2 Fabrication of Coatings via iCVD .......................................................................... 19
2.3.3 Novec 7300 Solvent Test .......................................................................................... 21
2.3.4 Tape Test .................................................................................................................. 21
2.3.5 PVP Fouling Test ..................................................................................................... 21
2.3.6 Characterization ...................................................................................................... 22
2.4 Results and Discussion .................................................................................................... 24
2.5 Conclusion ........................................................................................................................ 37
2.6 Acknowledgements .......................................................................................................... 37
iv
2.7 References ........................................................................................................................ 38
Chapter 3: Downstream Monomer Capture and Polymerization During the Vapor Phase
Fabrication of Porous Polymers ................................................................................................ 43
3.1 Abstract ............................................................................................................................ 43
3.2 Introduction ..................................................................................................................... 44
3.3 Experimental Section ...................................................................................................... 46
3.3.1 Materials ...................................................................................................................46
3.3.2 Fabrication of Porous Polymers via iCVD ............................................................ 46
3.3.3 Fabrication of Hydrophobic Dense Coatings for Porous Polymers via iCVD ....49
3.3.4 Characterization ...................................................................................................... 49
3.4 Results and Discussion .................................................................................................... 50
3.5 Conclusion ........................................................................................................................ 59
3.6 Acknowledgements .......................................................................................................... 60
3.7 References ........................................................................................................................ 61
Chapter 4: Giant Lipid Vesicle Formation Using Vapor-Deposited Charged Porous
Polymers ...................................................................................................................................... 65
4.1 Abstract ............................................................................................................................ 65
4.2 Introduction ..................................................................................................................... 66
4.3 Experimental Section ...................................................................................................... 69
4.3.1 Materials .................................................................................................................. 69
4.3.2 Fabrication of Porous Membranes ......................................................................... 69
4.3.3 Giant Vesicle Formation ......................................................................................... 71
4.3.4 Lamellarity Measurement ...................................................................................... 72
v
4.3.5 Microscopy ............................................................................................................... 72
4.3.6 Image Processing and Data Analysis ..................................................................... 72
4.4 Results and Discussion .................................................................................................... 74
4.5 Conclusion ........................................................................................................................ 97
4.6 Acknowledgements .......................................................................................................... 98
4.7 References ........................................................................................................................ 99
Chapter 5: Concluding Remarks and Future Work .............................................................. 105
5.1 Conclusion ...................................................................................................................... 105
5.2 Future Work .................................................................................................................. 107
5.3 References ...................................................................................................................... 110

vi
List of Tables
Table 2-1. XPS atomic composition of the coatings as deposited and after the solvent test ....... 31
Table 2-2. Atomic composition of the fluorinated coatings before and after the PVP fouling test
for the as-deposited coatings and coatings that were solvent tested ...............................................35
Table 3-1. Polymer mass obtained on TEC1 and TEC2 as a result of monomer capture and
polymerization from the preceding surface at varying TEC2 temperatures ................................... 57
Table 4-1. Mass swelling ratios of the porous membranes measured after hydration with 185 mM
PBS buffer containing 200 mM Sucrose at pH=7.4 ...................................................................... 89
Table 4-2. Non-parametric size comparisons for sample pairs of the same thickness and deposition
temperature using Wilcoxon rank-sum test for (a) pure POPC (b) 20% POPG (c) 20% DOTAP
vesicles formed using different membranes upon hydration with 185 mM PBS buffer containing
200 mM sucrose except the case for 20% DOTAP vesicles ......................................................... 90
Table 4-3. Non-parametric size comparisons for sample pairs using Wilcoxon rank-sum test for
POPC vesicles formed using 36 μm thick membrane deposited at -20 ℃ at varying buffer ionic
strength and sucrose concentration ............................................................................................... 95

vii
List of Figures
Figure 1-1. Schematic of a conventional iCVD reactor ................................................................. 1
Figure 1-2. Representative pressure-temperature phase diagram and SEM images of dense and
porous polymers fabricated using different iCVD conditions ........................................................ 4
Figure 1-3. (a) Schematic of solid monomer deposition showing the growth on a molecular and
macroscopic scale. (b) Monomer is deposited as pillar like microstructures and is partially
polymerized. The excess unreacted monomer is then removed by sublimation resulting in dual
scale porosity polymer membranes. (c) morphology of the membranes is shown by the cross-
sectional SEM images ..................................................................................................................... 5
Figure 1-4. (a) Angled and (b) top down SEM images showing that the morphology of crosslinked
PMAA membranes can be tuned by varying the substrate temperature (0 °C, -10 °C, -20 °C)
during the monomer deposition step ............................................................................................... 6
Figure 1-5. Cross-sectional and zoomed in SEM images of the membranes formed at different
polymerization temperatures .......................................................................................................... 7
Figure 2-1. (a) Chemical structures of the perfluorinated monomers and the EGDA crosslinker.
(b) Schematic representation of the two types of coating configurations: crosslinked p(C8PFA-co-
EGDA) and p(C6PFA-co-EGDA) coatings and graded pC8PFA-pEGDA and pC6PFA-pEGDA
coatings ......................................................................................................................................... 24
Figure 2-2. FTIR spectrum of the pEGDA crosslinked base layer ............................................... 25
Figure 2-3. Representative 5 micron and 500 nm AFM force images of (a) plasma cleaned bare
stainless steel (S.S.), and stainless steel that is coated with (b) p(C8PFA-co-EGDA), (c) pC8PFA-
pEGDA, (d) p(C6PFA-co-EGDA), and (e) pC6PFA-pEGDA polymers as deposited. 5 micron
viii
AFM force images of (f) bare silicon wafer, and silicon wafers that are coated with (g) p(C8PFA-
co-EGDA) and (h) p(C6PFA-co-EGDA) polymers as deposited .................................................. 26
Figure 2-4. Representative 5 micron and 500 nm AFM force images of stainless steel (S.S.) coated
with (a) p(C8PFA-co-EGDA), (b) pC8PFA-pEGDA, (c) p(C6PFA-co-EGDA), and (d) pC6PFA-
pEGDA polymers as deposited and after tape test ........................................................................ 28
Figure 2-5. (a) Chemical structure of the fluorinated solvent Novec 7300. (b) Schematic
representation of the solvent test performed on the coated stainless steel substrates
....................................................................................................................................................... 29
Figure 2-6. Representative 5 micron and 500 nm AFM force images of stainless steel (S.S.) coated
with (a) p(C8PFA-co-EGDA), (b) pC8PFA-pEGDA, (c) p(C6PFA-co-EGDA), and (d) pC6PFA-
pEGDA polymers as deposited and after solvent test ................................................................... 30
Figure 2-7. Average static water contact angles (CA) and hysteresis on the coated stainless steel
substrates ...................................................................................................................................... 32
Figure 2-8. Representative XPS survey spectra of (a) uncoated stainless steel (S.S.), (b) p(C8PFA-
co-EGDA) coated stainless steel, (c) p(C8PFA-co-EGDA) coated stainless steel after solvent test,
and d) p(C8PFA-co-EGDA) coated stainless steel after PVP test ................................................ 34
Figure 3-1. Schematic representation of multiple TECs in the iCVD reactor .............................. 47
Figure 3-2. Stepwise process of capturing and polymerizing monomer from the preceding TEC
....................................................................................................................................................... 48
Figure 3-3. (a) In-situ optical images of the monomer deposited on TEC1, TEC2, and TEC3 at
-10 °C at once and the corresponding (b) optical and (c) SEM images of the porous membranes
after polymerization and sublimation of the unreacted monomer ................................................. 51
ix
Figure 3-4. (a) In-situ optical images of the monomer deposited on TEC1, TEC2, and TEC3 at
-20 °C at once and the corresponding (b) optical and (c) SEM images of the porous membranes
after polymerization and sublimation of the unreacted monomer ................................................. 52
Figure 3-5. (a) Top down in-situ optical images of monomer capture and polymerization on TEC2
as a function of time at temperatures of -20, -10, and 0 °C while polymerization on TEC1 occurs
at 10 °C in all cases. (b) Images of polymer membranes formed on TEC2 after sublimation of the
unreacted monomer ...................................................................................................................... 54
Figure 3-6. SEM images of the porous membranes formed on (a) TEC1 and (b) TEC2 as a result
of monomer capture and polymerization from the preceding surface ........................................... 55
Figure 3-7. (a) SEM images of the membranes formed on TEC2 after PPFDA coating and (b) their
corresponding static water contact angle ....................................................................................... 56
Figure 3-8. (a) Optical images and (b) the corresponding SEM images of the porous membranes
formed on TEC1, TEC2, and TEC3 as a result of monomer capture and polymerization from the
preceding surface .......................................................................................................................... 58
Figure 4-1. Schematic (not to scale) of the vapor phase fabrication of porous membranes in an
iCVD chamber proceeded by hydrogel-supported vesicle formation experiments. Steps a-c depict
the process of (a) monomer deposition at variable substrate temperatures, followed by (b)
polymerization in the presence of a crosslinking agent under the heated filament array and
ultimately, (c) sublimation of the unreacted monomer in the reactor. Steps d-f represent the
hydrogel-supported formation of vesicles: (d) applying the lipid film on the polymer membrane
and subsequently e) hydrating it with sucrose buffer and (f) harvesting the vesicles in glucose
buffer for observation with fluorescence microscopy ................................................................... 74
x
Figure 4-2. (a)
1
H-NMR spectra of crosslinked poly(methacrylic acid-co-ethylene glycol
diacrylate) after solvent extraction in deuterium oxide. (b)
1
H-NMR spectra of standard
poly(methacrylic acid) (Polysciences, Inc.) in deuterium oxide (13 mg in 1ml D2O) as the
control........................................................................................................................................... 76
Figure 4-3. (a) SEM micrographs of xPMAA porous membranes deposited at temperatures -20
℃, -10 °C, and 0 ℃ with three different thicknesses. The scale bars for the top-down and angled
micrographs are 150 μm and 20 μm, respectively. (b) Deposition time versus thickness of the
porous membranes measured at the three temperatures: circle, diamond and square correspond to
deposition temperatures -20 ℃, -10 °C, and 0 ℃ and red, blue and black refer to average
thicknesses of 36 ± 11μm, 82 ± 8 μm, and 124 ± 19 μm, respectively ......................................... 77
Figure 4-4. (a) Yield of vesicles formed from POPC lipid-coated porous membranes (membrane
labels correspond to the designations in Figure 4-3a) upon hydration with 185 mM PBS (137 mM
NaCl, 10 mM PO4
-3
, and 2.7 mM KCl) buffer containing 200mM sucrose at pH 7.4. (b) Yield of
vesicles formed from charged lipid mixture (20%POPG:80% POPC or 20%DOTAP:80% POPC).
The first four pairs of bars describe hydration with 185 mM PBS buffer containing 200mM sucrose
at pH 7.4. The final bar shows the results for 20%DOTAP:80% POPC hydrated with 740 mM PBS
containing 200 mM Sucrose at pH=7.4. (c) Schematic for the proposed charge screening
mechanism in oppositely charged polymer and lipid mixtures ..................................................... 79
Figure 4-5. (a) Spinning disk confocal micrograph of vesicles at 491 nm excitation before adding
α-hemolysin (b) Confocal micrograph of vesicles at 491 nm excitation after adding α-hemolysin.
(c) Confocal micrograph of vesicles at 561 nm excitation after adding α-hemolysin. (d)
Fluorescence intensity across the lines indicated in micrographs a and b. This experiment was
xi
repeated on another set of vesicles formed from the same membrane. Here, (e), (f), (g) and (h)
have analogous descriptions to panels (a), (b), (c), and (d), respectively ...................................... 83
Figure 4-6. Fluorescent micrograph of vesicles formed from 20% DOTAP lipid-coated 36μm
thick membrane fabricated at deposition temperature -20℃ and hydrated with 740 mM PBS
containing 200 mM sucrose .......................................................................................................... 85
Figure 4-7. (a) Box plots for natural log-scale size distribution of the vesicles formed from POPC-
coated porous membranes upon hydration with 185 mM PBS buffer containing 200mM sucrose at
pH 7.4. (b) and (c) Box plots for log-scale size distribution of the vesicles formed from 20%POPG
and 20% DOTAP coated porous membranes upon hydration with 185 mM PBS buffer containing
200mM sucrose at pH 7.4 (except if otherwise mentioned) .......................................................... 86
Figure 4-8. (a) Yield and (b) log- scale size distribution of the vesicles formed from POPC lipid-
coated 36 μm thick membrane deposited at -20 ℃ at varying hydration buffer ionic strength and
sucrose concentration to match osmolarity ................................................................................... 93
Figure 4-9. Fluorescent micrographs of vesicles formed from POPC lipid-coated 36μm thick
membrane fabricated at deposition temperature -20℃ (a) upon hydration with 18.5 mM PBS
containing 400 mM sucrose and (b) upon hydration with 0 PBS containing 440 mM sucrose at
pH=7.4 .......................................................................................................................................... 96

xii
Abstract
Polymers are widely used as surface coatings and membranes that are capable of imparting
functionality and surface chemistry desired for applications such as hydrophobic coatings and
water purification membranes. Solvent-free vapor phase polymerization methods are great for the
fabrication of polymer coatings and membranes since they eliminate surface tension and solvent
compatibility limitations associated with solution phase polymerization methods. Initiated
chemical vapor deposition (iCVD) is a vapor phase polymer fabrication technique that can be used
to deposit dense and porous polymers. In this dissertation, we demonstrate iCVD processes to
fabricate functional dense and porous polymers with practical applications.
Chapter 1 provides a background on the conventional iCVD process for the deposition of
dense polymers. The modification of the conventional iCVD process for the fabrication of porous
polymers with dual scale porosity is covered next. Chapter 2 is focused on robust vapor deposited
fluoropolymers with an industrial application as antifouling coatings for stainless steel
polymerization reactor components. In this section, we study and compare the chemical and
mechanical stability and the performance of long chain vs short chain perfluorinated iCVD
polymers as antifouling coatings. We show that the more environmentally friendly short chain
perfluorinated coating is a potential alternative to its long chain analog as an antifouling surface.
Chapter 3 is focused on process improvements of the modified iCVD process for the fabrication
of porous polymers which have applications in sensing, tissue engineering, separations and so on.
We demonstrate that downstream monomer capture and polymerization during vapor phase
fabrication of porous polymers is possible by incorporation of multiple cold traps inside the
reactor. The uniformity and morphology of the membranes across surfaces and the effects of
surface temperature on membrane morphology and polymerization kinetics are studied. Our ability
xiii
to capture and polymerize monomer across multiple surfaces provides a more sustainable route for
reducing monomer waste and improving conversion and throughput of the process. Chapter 4 is
focused on a biophysical application of vapor deposited hydrophilic porous polymers as hydrogel
substrates for the formation of giant lipid vesicles using the hydrogel assisted rehydration method.
Swelling of the lipid covered porous polymer membrane leads to the formation of giant lipid
vesicles. Polymer membranes with controlled morphologies, thicknesses, and surface chemistry
are fabricated to study the effects of the physicochemical properties of the polymer on vesicle
characteristics for method optimization. Lipid-polymer interactions and osmotic effects in addition
to the substrate morphology and surface charge are shown to be key factors affecting vesicle
formation. Chapter 5 concludes this dissertation by summarizing the key contributions and the
future work.

1
Chapter 1: Introduction
1.1 Initiated Chemical Vapor Deposition of Dense Polymer Films

Initiated chemical vapor deposition (iCVD) is a vapor phase surface polymerization
technique that is used to deposit a variety of functional polymer films via the free radical
polymerization mechanism in mild reactor conditions.
1
Vapor deposition techniques are desirable
due to the solventless nature of the process eliminating surface tension, dewetting, and solubility
issues associated with liquid phase methods.
2,3
This allows for conformal polymer deposition on
complex substrates such as porous materials and 3D printed microfluidic devices with thicknesses
in the nano and micron scale.
4,5
iCVD is a low energy process in which the thermal initiation and
polymer deposition sites are decoupled such that the chemical functionalities of the reactive
precursors are retained.
6

Figure 1-1. Schematic of a conventional iCVD reactor.
A typical iCVD reactor is shown in Figure 1-1. The reactor is covered with a quartz lead
which allows for visual observation and monitoring the polymer thickness via in-situ laser
interferometry. Backside recirculating chiller is connected to the bottom of the reactor to maintain
the stage temperature between 10 °C – 60 °C. Monomer(s) and initiator vapor are introduced into
the reactor through feed jars that are mounted on the lines connected to the reactor. Vapor pressure
stage
glass
filament
adsorption
monomer
free radical species
polymer
desorption
polymerization
vacuum pump
monomer
initiator
2
and the flow rates of the precursors are controlled by temperature of the heated jars, needle valves,
and mass flow controllers. An exhaust line on the other side of the reactor connects to the vacuum
pump and creates a uniform flow in the reactor. Reactor pressure is maintained by a pressure valve
which controlled by a baratron manometer. The reactor pressure is usually kept constant
throughout the deposition and is usually set between 70-600 mTorr depending on the monomer
processing conditions. An array of Nichrome filament array is suspended above the substrate and
is resistively heated during the depositions to cleave the initiator molecules into reactive free
radicals.
7

Psat (T2) = Psat (T1) exp [ (-
ΔHvap
R
) (
1
T 2
−
1
T 1
)] equation 1

PM = Preactor (
F m
F total
) equation 2

In the iCVD process, initiator and monomer molecules maintain the reactor pressure under
continuous flow. Monomer adsorbs on the surface of the cooled substrates and the heated filament
array creates the initiator free radicals which then start the polymerization with the vinyl bond of
the monomer. In this process, partial pressure of the monomer is kept below the its saturation
pressure (Pm/Psat mostly in the range 0.2-0.7) at a given substrate temperature in order to produce
uniform polymer films and avoid condensation.
8
Monomer saturation pressure (Psat) is calculated
using the Clausius- Clapeyron equation (equation 1) at a given substrate temperature and the partial
pressure of the monomer (Pm) is calculated using the measured flow rates and the reactor pressure
(equation 2). Process parameters such as reactor pressure, flow rates, substrate temperature, and
filament temperature have been used to tune the deposition rate, thickness, and molecular weight
of the deposited polymers.
9
Substrate temperature affects Pm/Psat and hence the monomer
concentration on the surface of the substrate. Filament temperature controls the free radical
concentration. It has been shown that deposition rate and molecular weight both increase by
3
decreasing the substrate temperatures due to an increase in Pm/Psat and monomer surface
concentration. However, monomer flow rate can be adjusted to keep Pm/Psat constant in order to
have constant monomer concentration. In this condition, decreasing the substrate temperature
decreases the deposition rate due to the slower reaction kinetics at lower temperatures.
10,11

The iCVD process is a continuous flow (CF) vapor deposition technique since the
monomer and initiator are constantly fed into the chamber and are exhausted by a vacuum pump.
Petruczok and coworkers proposed a closed-batch (CB) iCVD method in which the reactor
chamber is isolated both from the feed lines and the vacuum exhaust line after it is filled with the
reactants such that there is no advective flow during depositions. This process configuration can
be used to reproducibly fabricate ultrathin dense polymer films by providing excellent control over
deposition rates. In addition, the CB process improves the reaction yield while it significantly
reduces cost, material waste, and reactor maintenance downtime.
12

1.2 Macroporous Polymer Membranes by Vapor Deposition

Porosity, controllable structure, and chemical functionality of porous polymers make them
desirable for a vast range of applications in separations
13-15
, sensing
16,17
, biocatalysis
18,19
, and tissue
engineering
20,21
. Traditional solution-phase methods such as high internal phase emulsion
22,23
,
cryopolymerization
24-26
, phase separation
27,28
, and solvent casting and particulate leaching
29,30
offer fine control over porosity but the solubility requirements limit the range of substrates and the
chemical functionalities that can be achieved by these techniques. Solventless processes offer a
green alternative which do not rely on the choice of solvent to satisfy solubility requirements.
Novel solventless processes that can solve the shortcomings of the current fabrication techniques
have the potential to promote new areas of research by expanding the applicability of porous
polymers.
4

Figure 1-2. Representative pressure-temperature phase diagram and SEM images of dense and
porous polymers fabricated using different iCVD conditions.
31

Solventless techniques for the fabrication of porous polymer membranes have previously
been introduced in initiated chemical vapor deposition (iCVD) vacuum chambers by Tao, et al
32

and our group
33
. Tao and coworkers demonstrated that porous polymer films can be formed by
concurrent phase separation and polymerization during vapor deposition. This process involves
the use of a condensable inert species that phase separates from the reactive components and is
removed after polymerization. Our modified iCVD technique involves the introduction of
monomer and initiator vapor in a vacuum chamber where the monomer partial pressure and
substrate temperature are kept below the triple point pressure and temperature of the monomer,
such that the monomer vapor deposits as solid microstructures on the surface of the substrate
(Figure 1-2). The polymerization of the solid monomer occurs via the free radical polymerization
mechanism, initiated by the free radicals that are formed by the heated filament array suspended
above the substrate. The excess unreacted monomer is then sublimated resulting in dual scale
porosity polymers. (Figure 1-3) Similar to the typical iCVD process, here the site of monomer
deposition and thermal initiation are decoupled, enabling direct growth of morphologically
controlled polymer membranes with retained chemical functionality on delicate substrates.
34
The
bottom-up, all dry nature of this technique renders the process substrate independent by avoiding
surface tension and solvent compatibility issues.
5

Figure 1-3. (a) Schematic of solid monomer deposition showing the growth on a molecular and
macroscopic scale. (b) Monomer is deposited as pillar like microstructures and is partially
polymerized. The excess unreacted monomer is then removed by sublimation resulting in dual
scale porosity polymer membranes. (c) morphology of the membranes is shown by the cross-
sectional SEM images.
33

In this technique, monomer deposition and polymerization can occur simultaneously or
sequentially. In the simultaneous process, monomer deposition and polymerization happen
concurrently by heating the filament array during monomer deposition
33-38
, while in the sequential
process polymerization starts after monomer deposition
39-41
. Decoupling the monomer deposition
and polymerization steps in the sequential process grants proper adjustment of the reactor
conditions for the introduction of other functional monomers and allows for systematic studies of
partial polymerization
monomer free radical species polymer
(b) monomer deposition and polymerization
small-scale pores
Deposition
monomer
monomer
removal
(a) monomer deposition
large-scale pores
monomer
removal
200 µm
large-scale pores
20 µm
small-scale pores
(c) resultant structures
6
each process parameter in addition to reducing material waste associated with continuous
monomer flow during polymerization.
39,40

Figure 1-4. (a) Angled and (b) top down SEM images showing that the morphology of crosslinked
PMAA membranes can be tuned by varying the substrate temperature (0 °C, -10 °C, -20 °C)
during the monomer deposition step.
31,41

0
∘
C -10
∘
C -20
∘
C
100µm
(a)
(b)
7

Figure 1-5. Cross-sectional and zoomed in SEM images of the membranes formed at different
polymerization temperatures.
40
In both the simultaneous and the sequential processes, polymer membranes have shown to
form dual scale pores. The morphology and the large scale porosity (tens to hundreds of microns)
are controlled by the substrate temperature during monomer deposition (Figure 1-4) and the small
scale pores that are formed within the microstructures (hundreds of nanometers to a few microns)
are controlled by the polymerization temperature which affects the sublimation rate of the
monomer in addition to the reaction kinetics (Figure 1-5).
40
Lower deposition temperatures result
in higher nucleation and formation of 3D pillar like microstructures whereas higher deposition
temperatures result in 2D web like microstructures (Figure 1-4). Higher polymerization
temperatures lead to higher monomer sublimation rate during polymerization resulting in larger
small scale pores within the microstructures (Figure 1-5).
We have studied the kinetics of solid monomer polymerization with a vapor phase initiator.
The porous membranes fabricated at all processing conditions have a bimodal molecular weight
distribution studied by gel permeation chromatography (GPC).
40
Shorter polymer chains with
-20 °C -10 °C 0 °C 10 °C
150 µm
20 µm
8
number-averaged molecular weight (Mn) of approximately ten thousand Da form at the vapor-
solid interface due to the high concentration of the initiator radicals resulting in faster termination.
Longer polymer chains in the range of hundreds of thousands to a few millions of Da form inside
the solid monomer microstructures because of slower diffusion of the initiator free radicals inside
the solid monomer microstructures leading to a slower termination rate. The Mn of the high
molecular weight fraction increases with polymerization time due to the addition of solid monomer
to the growing polymer chains, while the Mn of the low molecular weight fraction remains
relatively constant with time as a result of the continuous introduction of initiator radicals at the
vapor-solid interface. Higher polymerization temperatures also result in a drastic increase in the
Mn of the larger chains due to the increased mobility of the solid monomer within the
microstructures.
Polymer conversion increases with increasing the substrate temperature during
polymerization which is associated with higher monomer mobility and faster kinetics as well as
faster monomer sublimation rates.
31
Although increasing polymerization time increases monomer
conversion, polymer mass growth rate slows down after ~120 minutes of polymerization in this
process.
40
This is likely due to the diffusional barrier for the initiator as the polymer structures
densify over time.

At our processing conditions, the highest monomer to polymer conversion is
approximately 12% which is obtained at the polymerization temperature of 10 °C for a
polymerization time of 240 minutes. As with the previously reported solid phase polymerization
techniques
42-44
, this vapor phase initiator-solid monomer polymerization method results in low
conversions (approximately 10 wt % or less) even with extended polymerization times.
40

9
1.3 References

[1] Tenhaeff, W.E.; Gleason, K.K. Initiated and Oxidative Chemical Vapor Deposition of
Polymeric Thin Films: iCVD and oCVD. Adv. Funct. Mater. 2008, 18(7), 979−992.
[2] Lau, K.K.S.; Bico, J.; Teo, K.B.K.; Chhowalla, M.; Amaratunga, G.A.J.; Milne, W. I.;
McKinley, G.H.; Gleason, K.K. Superhydrophobic Carbon Nanotube Forests. Nano Lett. 2003,
3(12),1701−1705.
[3] Kwong, P.; Flowers, C.A.; Gupta, M. Directed Deposition of Functional Polymers onto Porous
Substrates Using Metal Salt Inhibitors. Langmuir 2011, 27(17), 10634−10641.
[4] Riche, C.T.; Marin, B.C.; Malmstadt, N.; Gupta, M. Vapor Deposition of Cross-linked
Fluoropolymer Barrier Coatings onto Pre-assembled Microfluidic Devices. Lab on a Chip, 2011,
11(18), 3049-3052.
[5] Ma, M.; Gupta, M.; Li, Z.; Zhai, L.; Gleason, K.K.; Cohen, R.E.; Rubner, M.F.; Rutledge, G.C.
Decorated Electrospun Fibers Exhibiting Superhydrophobicity. Advanced Materials 2007, 19(2),
255-259.
[6] Ma, M.; Mao, Y.; Gupta, M.; Gleason, K.K.; Rutledge, G.C. Superhydrophobic Fabrics
Produced by Electrospinning and Chemical Vapor Deposition. Macromolecules 2005, 38(23),
9742-9748.
[7] Seidel, S.; Riche, C.; Gupta, M. Chemical Vapor Deposition of Polymer Films. Encyclopedia
of Polymer Science and Technology. 2011, 4.
[8] Alf, M.E.; Asatekin, A.; Barr, M.C.; Baxamusa, S.H.; Chelawat, H.; Ozaydin‐Ince, G.;
Petruczok, C.D.; Sreenivasan, R.; Tenhaeff, W.E.; Trujillo, N.J.; Vaddiraju, S. Chemical Vapor
Deposition of Conformal, Functional, and Responsive Polymer Films. Advanced Materials
2010, 22(18), 1993-2027.
10
[9] Chan, K.; Gleason, K.K. A Mechanistic Study of Initiated Chemical Vapor Deposition of
Polymers: Analyses of Deposition Rate and Molecular Weight. Macromolecules 2006, 39(11),
3890-3894.
[10] Ozaydin-Ince, G.; Gleason, K.K. Transition Between Kinetic and Mass Transfer Regimes in
the Initiated Chemical Vapor Deposition from Ethylene Glycol Diacrylate. Journal of Vacuum
Science & Technology A: Vacuum, Surfaces, and Films 2009, 27(5), 1135-1143.
[11] Moni, P.; Mohr, A.C.; Gleason, K.K. Growth Rate and Cross-linking Kinetics of Poly (divinyl
benzene) Thin Films Formed via Initiated Chemical Vapor Deposition. Langmuir 2018, 34(23),
6687-6696.
[12] Petruczok, C.D.; Chen, N.; Gleason, K.K. Closed Batch Initiated Chemical Vapor Deposition
of Ultrathin, Functional, and Conformal Polymer Films. Langmuir 2014, 30(16), 4830-4837.
[13] Svec, F.; Frechet, J.M. New Designs of Macroporous Polymers and Supports: from Separation
to Biocatalysis. Science 1996, 273(5272), 205-11.
[14] Li, Z.; Wang, B.; Qin, X.; Wang, Y.; Liu, C.; Shao, Q.; Wang, N.; Zhang, J.; Wang, Z.; Shen,
C.; Guo, Z. Superhydrophobic/Superoleophilic Polycarbonate/Carbon Nanotubes Porous
Monolith for Selective Oil Adsorption from Water. ACS Sustainable Chemistry & Engineering.
2018, 6(11), 13747–13755.
[15] Zhang, X.; Wang, X.; Liu, X.; Lv, C.; Wang, Y.; Zheng, G.; Liu, H.; Liu, C.; Guo, Z.; Shen,
C. Porous Polyethylene Bundles with Enhanced Hydrophobicity and Pumping Oil-Recovery
Ability via Skin-Peeling. ACS Sustainable Chemistry & Engineering 2018, 6(10), 12580-12585.
[16] Xie, Z.; Ma, L.; deKrafft, K.E.; Jin, A.; Lin, W. Porous Phosphorescent Coordination
Polymers for Oxygen Sensing. J. Am. Chem. Soc. 2009, 132(3), 922-923.
11
[17] Novotney, J.L.; Dichtel, W.R. Conjugated Porous Polymers for TNT Vapor Detection. ACS
Macro Letters 2013, 2(5), 423-426.
[18] Cantone, S.; Ferrario, V.; Corici, L.; Ebert, C.; Fattor, D.; Spizzo, P.; Gardossi, L. Efficient
Immobilisation of Industrial Biocatalysts: Criteria and Constraints for the Selection of Organic
Polymeric Carriers and Immobilisation Methods. Chem Soc Rev. 2013, 42(15), 6262-6276.
[19] Kimmins, S.D.; Wyman, P.; Cameron, N.R. Amine-Functionalization of Glycidyl
Methacrylate-Containing Emulsion-Templated Porous Polymers and Immobilization of Proteinase
K for Biocatalysis. Polymer 2014, 55(1), 416-425.
[20] Hentze, H.P.; Antonietti, M.; Porous Polymers and Resins for Biotechnological and
Biomedical Applications. J Biotechnol. 2002, 90(1), 27-53.
[21] Rezwan, K.; Chen, Q.Z.; Blaker, J.J.; Boccaccini, A.R. Biodegradable and Bioactive Porous
Polymer/Inorganic Composite Scaffolds for Bone Tissue Engineering. Biomaterials 2006, 27(18),
3413-3431.
[22] Hayward, A.S.; Sano, N.; Przyborski, S.A.; Cameron, N.R. Acrylic‐acid‐Functionalized
PolyHIPE Scaffolds for Use in 3D Cell Culture. Macromol. Rapid Commun. 2013, 34(23-24),
1844-1849.
[23] Pulko, I.; Krajnc, P. High Internal Phase Emulsion Templating–a Path to Hierarchically
Porous Functional Polymers. Macromol. Rapid Commun. 2012, 33(20), 1731-1746.
[24] Xue, W.; Hamley, I.W.; Huglin, M.B. Rapid Swelling and Deswelling of Thermoreversible
Hydrophobically Modified Poly (N-isopropylacrylamide) Hydrogels Prepared by Freezing
Polymerisation. Polymer. 2002, 43(19), 5181-5186.
12
[25] Strachota, B.; Šlouf, M.; Hodan, J.; Matějka, L. Advanced Two-Step Cryopolymerization to
Form Superporous Thermosensitive PNIPA/Clay Gels with Unique Mechanical Properties and
Ultrafast Swelling-Deswelling Kinetics. Colloid and Polymer Science 2018, 296(4), 753-769.
[26] Perez, P.; Plieva, F.; Gallardo, A.; San Roman, J.; Aguilar, M.R.; Morfin, I.; Ehrburger- Dolle,
F.; Bley, F.; Mikhalovsky, S.; Galaev, I.Y.; Mattiasson, B. Bioresorbable and Nonresorbable
Macroporous Thermosensitive Hydrogels Prepared by Cryopolymerization. Role of the Cross-
linking Agent. Biomacromolecules 2007, 9(1), 66-74.
[27] Ikem, V.O.; Menner, A.; Bismarck, A. High-Porosity Macroporous Polymers Sythesized from
Titania-Particle-Stabilized Medium and High Internal Phase Emulsions. Langmuir 2010, 26(11),
8836-8841.
[28] Sukitpaneenit, P.; Chung, T.S. Molecular Elucidation of Morphology and Mechanical
Properties of PVDF Hollow Fiber Membranes from Aspects of Phase Inversion, Crystallization
and Rheology. J. Membr. Sci. 2009, 340(1-2), 192-205.
[29] Sa-nguanruksa, J.; Rujiravanit, R.; Supaphol, P.; Tokura, S. Porous Polyethylene Membranes
by Template-Leaching Technique: Preparation and Characterization. Polym. Test. 2004, 23(1),
91−99.
[30] Mikos, A. G.; Thorsen, A. J.; Czerwonka, L. A.; Bao, Y.; Langer, R.; Winslow, D. N.;
Vacanti, J. P. Preparation and Characterization of Poly(L-Lactic Acid) Foams. Polymer 1994,
35(5), 1068−1077.
[31] Dianat, G.; Movsesian, N.; Gupta, M. Vapor Deposition of Functional Porous Polymer
Membranes. ACS Applied Polymer Materials 2020, 2(2), 98-104.
[32] Tao, R.; Anthamatten, M. Porous Polymers by Controlling Phase Separation During Vapor
Deposition Polymerization. Macromol Rapid Commun. 2013, 34(22), 1755-1760.
13
[33] Seidel, S.K.; Kwong, P.; Gupta, M. Simultaneous Polymerization and Solid Monomer
Deposition for the Fabrication of Polymer Membranes with Dual-Scale Porosity. Macromolecules
2013, 46(8), 2976-2983.
[34] Kwong, P.; Seidel, S.; Gupta, M. Solventless Fabrication of Porous-on-Porous Materials. ACS
Appl. Mater. Interfaces. 2013, 5(19), 9714-9718.
[35] Seidel, S.; Gupta, M. Systematic Study of the Growth and Morphology of Vapor Deposited
Porous Polymer Membranes. J. Vac. Sci. Technol. A. 2014, 32(4), 041514.
[36] Seidel, S.; Chu Cheong, C.; Kwong, P.; Gupta, M. All‐Dry Fabrication of Poly (methacrylic
acid)‐Based Membranes with Controlled Dissolution Behavior. Macromol. Mater. Eng. 2015,
300(11), 1079-1084.
[37] Seidel, S.; Dianat, G.; Gupta, M. Formation of Porous Polymer Coatings on Complex
Substrates Using Vapor Phase Precursors. Macromol. Mater. Eng. 2016, 301(4), 371-376.
[38] Dianat, G.; Seidel, S.; De Luna, M.M.; Gupta, M. Vapor Phase Fabrication of Hydrophilic
and Hydrophobic Asymmetric Polymer Membranes. Macromol. Mater. Eng. 2016, 301(9), 1037-
1043.
[39] Dianat, G.; Gupta, M. Sequential Deposition of Patterned Porous Polymers Using Poly
(dimethylsiloxane) Masks. Polymer 2017, 126, 463-469.
[40] Dianat, G.; Movsesian, N.; Gupta, M. Process-Structure-Property Relationships for Porous
Membranes Formed by Polymerization of Solid Monomer by a Vapor-Phase Initiator.
Macromolecules 2018, 51(24), 10297-10303.

14
[41] Movsesian, N.; Tittensor, M.; Dianat, G.; Gupta, M.; Malmstadt, N. Giant Lipid Vesicle
Formation Using Vapor-Deposited Charged Porous Polymers. Langmuir 2018, 34(30), 9025-
9035.
[42] Eastmond, G.C.; Solid-State Polymerization. Progress in Polymer Science 1970, 2, 1-46.
[43] Bamford, C. H.; Jenkins, A. D.; Ward, J.C.; Polymerization in the Solid and Near‐Solid States.
Journal of Polymer Science Part A: Polymer Chemistry 1960, 48(150), 37-51.
[44] Bamford, C.H.; Eastmond, G.C.; Ward JC. Studies in polymerization XIV. The Solid-State
Polymerization of Acrylic and Methacrylic Acids. Proc. R. Soc. Lond. A 1963, 271(1346), 357-
378.

15
Chapter 2: Robust Vapor-Deposited Antifouling Fluoropolymer
Coatings for Stainless Steel Polymerization Reactor Components
2.1 Abstract

Initiated chemical vapor deposition (iCVD) offers a solventless and scalable method to
apply thin polymer coatings on a variety of substrates. In this report, we systematically compare
the robustness, hydrophobicity, and antifouling properties of coatings synthesized using long chain
1H,1H,2H,2H-perfluorodecyl acrylate (PFDA or C8PFA) and short chain 1H,1H,2H,2H-
perfluorooctyl acrylate (PFOA or C6PFA) precursors. Our results show that the incorporation of
ethylene glycol diacrylate (EGDA) into the coatings is critical for the retention of the chemical
functionality and hydrophobicity of the coatings after sonication in a fluorinated solvent. The
adhesion of polyvinylpyrrolidone polymer onto these surfaces was tested at simulated mixing
conditions to determine the applicability of these coatings for preventing fouling during industrial
polymerization processes. Graded coatings that were synthesized by polymerizing EGDA prior to
polymerizing the fluorinated monomer performed better than coatings that were synthesized by
copolymerization of EGDA with the fluorinated monomer. Based on our findings, the short chain
coatings are a potential alternative to long chain coatings as an antifouling surface. Our data on
the robustness and antifouling behavior of the fluoropolymer coatings on stainless steel substrates
provides guidelines for designing functional coatings for industrial polymerization reactors.

16
2.2 Introduction

Fluorinated coatings can be used to fabricate self-cleaning
1,2
, anti-bacterial
3,4
, anti-icing
5,6
,

and anti-corrosion
7,8
surfaces for a range of applications in textiles
9
, microelectronics
10
, and
microfluidics
11
due to their hydrophobicity, low surface energy, low polarizability, and high
chemical stability.
12,13
Initiated chemical vapor deposition (iCVD) is a one-pot vapor phase free
radical polymerization technique that can be used to deposit conformal fluorinated polymer
coatings on a variety of textured substrates including cellulose paper
14
, PDMS microchannels
15
,
porous membranes
16
, and 3D-printed objects
17
due to the solventless nature of the process. In the
iCVD process, monomer and initiator molecules are introduced into the reactor in the vapor phase
and a filament array is resistively heated to a temperature of 200-300 °C to cleave the peroxide
initiator molecules into free radicals. The monomer molecules adsorb onto a cooled substrate and
the free radicals initiate polymerization via a free radical mechanism. In contrast to plasma
enhanced chemical vapor deposition (PECVD), iCVD operates at mild reactor conditions to
selectively cleave the initiator while offering full retention of the chemical functionality of the
monomer precursors as demonstrated by Fourier transform infrared spectroscopy (FTIR) and X-
ray photoelectron spectroscopy (XPS).
18,19

The iCVD process is particularly desirable for depositing fluorinated coatings since strict
solubility requirements lead to challenges in finding a compatible solvent for solution phase
polymerization methods.
20
Although long chain poly(perfluoroalkyl acrylate)s such as
poly(1H,1H,2H,2H-perfluorodecyl acrylate) (pPFDA or pC8PFA) with eight perfluorinated
pendant groups show superior hydrophobicity and crystallinity,
21,22
their degradation products
such as perfluorooctanoic acid and perfluorooctanesulfonic acid can accumulate in the
environment, wildlife, and human tissue.
23
The byproducts of perfluorinated acrylate polymers
17
with pendant groups of six carbons or less such as poly(1H,1H,2H,2H-perfluorooctyl acrylate)
(pPFOA or pC6PFA) do not accumulate as significantly.
24
However, the short chain perfluorinated
polymers have high chain mobility due to non-crystallinity which results in lower hydrophobicity
as a consequence of extensive molecular reorganization and exposure of the hydrophilic carbonyl
functional groups upon contact with water.
25
The contact angle hysteresis and mechanical
properties of pC6PFA and pC8PFA polymers can be enhanced via incorporation of a
crosslinker
26,27
, thermal annealing after crosslinking
28
, and enhanced crystallinity via grafting to
the substrate
29
. For example, Liu et al. showed that the contact angle hysteresis for polymers
composed of C6PFA can be significantly lowered by restricting the movement of the fluorinated
polymer chains by either copolymerization with divinyl benzene (DVB) crosslinker or by grafting
pC6PFA on top of a crosslinked base layer of poly(divinyl benzene) (pDVB).
26
Sojoudi et al.
similarly showed that pC8PFA deposited on a crosslinked base layer has enhanced mechanical
properties due to the graded polymer network.
27
Furthermore, Cocilte et al. showed that grafting
pC8PFA chains on a silicon substrate increases the formation of crystalline structures and reduces
polymer chain mobility.
29
The ability to coat stainless steel with fluoropolymers has many potential applications due
to high demands in chemical plants and reactor equipment (e.g., continuously stirred-tank reactors,
mixers, millireactors, and heat exchangers). The performance of these steel components may be
disrupted by surface adsorption of reaction materials leading to fouling.
30,31
Stable antifouling
perfluorinated coatings have the potential to maintain the properties of industrial reactors (e.g.,
flow, mixing pattern, and reactor volume) and prevent direct economic loss due to a halt in
production, costly cleaning processes, and material deterioration.
30
Several iCVD grafting
methods have been introduced to coat silicon and metallic substrates such as copper, titanium, and
18
stainless steel with perfluorinated acrylic polymers to reduce ice adhesion and enhance heat
transfer properties.
27,32,33
The environmental and health hazards of the long chain perfluorinated
polymers requires the exploration and use of shorter chain perfluorinated analogs.

In this report,
we systematically study the robustness of C8 and C6 iCVD polymer coatings on plasma cleaned
stainless steel substrates. The polymer coatings contain ethylene glycol diacrylate (EGDA)
crosslinker in two different configurations: 1) the EGDA is copolymerized with the fluorinated
monomer to form crosslinked p(C8PFA-co-EGDA) and p(C6PFA-co-EGDA) coatings and 2) the
EGDA is polymerized prior to polymerization of the fluorinated monomer to form graded
pC8PFA-pEGDA and pC6PFA-pEGDA coatings. The robustness of these coatings was tested via
soaking and sonication in Novec 7300, a hydrofluoroether solvent. The integrity of the films was
characterized using tape tests, X-ray photoelectron spectroscopy (XPS), atomic force microscopy
(AFM), and contact angle goniometry. The adsorption of polyvinylpyrrolidone (PVP) onto these
coatings at simulated mixing conditions were studied for potential applications as antifouling
surfaces for polymerization reactor mixers.

19
2.3 Experimental Section

2.3.1 Materials

1H,1H,2H,2H-perfluorodecyl acrylate (PFDA or C8PFA) (SynQuest Labs, Inc. 97%),
1H,1H,2H,2H-perfluorooctyl acrylate (PFOA or C6PFA) (SynQuest Labs, Inc. 97%), ethylene
glycol diacrylate (EGDA) (Polysciences, Inc.), tert-butyl peroxide (TBPO) (Sigma Aldrich, 98%),
polyvinylpyrrolidone (PVP, average MW: 40kDa) (Sigma Aldrich), acetone (VWR), and the
fluorinated solvent 3M
TM
Novec
TM
7300 (Gallade Chemical) were used as received. Scotch tape
(3M) was used for the tape test. Grade 304 stainless steel sheets (McMaster) and silicon wafer
(Wafer World 119) were used as the substrates.
2.3.2 Fabrication of Coatings via iCVD

The coatings were fabricated in a custom designed initiated chemical vapor deposition
(iCVD) chamber of 25 cm diameter and 5 cm height (GVD Corporation). A rotary vane vacuum
pump (Edwards E2M18) was used to achieve vacuum and the reactor pressure was controlled by
a throttle valve (MKS 153D) regulated by a capacitance manometer (MKS Baratron 622C01TDE).
A nichrome filament array (Omega Engineering, 80/20% Ni/Cr) suspended 4.6 cm above the stage
was used as the heating source for initiation during the depositions. The stage temperature was
maintained using a recirculating chiller (Thermo Scientific NESLAB RTE 7). Prior to the iCVD
depositions, the stainless steel substrates were rinsed with acetone and treated with a XEI
CombiClean air plasma cleaner for 10 minutes at 20 W. The thickness of the coatings was
monitored on a reference silicon wafer by an in-situ interferometer using a He-Ne laser (Industrial
Fiber Optics, 633 nm). C8PFA, C6PFA, and EGDA feed jars were mounted onto the reactor and
heated to achieve the desired vapor pressures and the flow rates were metered using needle valves.
The TBPO jar was kept at 23°C throughout all the depositions and a mass flow controller (MKS
20
1152 C) was used to meter the TBPO flow rate. During the depositions, the stage was kept at 30
°C and the nichrome wire was heated to 250 °C to thermally cleave the TBPO into free radicals.
For the deposition of the control pC8PFA homopolymer coating, the C8PFA jar was heated to
50°C. C8PFA and TBPO were introduced into the reactor at 0.1 and 0.7 sccm, respectively, and
the reactor pressure was maintained at 70 mtorr. The filament array was then resistively heated for
30 minutes to achieve a polymer thickness of approximately 300 nm. For the deposition of the
control pC6PFA homopolymer coating, the C6PFA jar was heated to 50°C. C6PFA and TBPO
were introduced into the reactor at 1.6 and 0.7 sccm, respectively, and the reactor pressure was
maintained at 70 mtorr. The filament array was resistively heated for 16 minutes to achieve a
thickness of approximately 300 nm. For the deposition of the p(C8PFA-co-EGDA) coating,
C8PFA and EGDA were heated to 50°C and 35°C, respectively. C8PFA, EGDA, and TBPO were
introduced into the reactor at flow rates of 0.05, 0.4, and 0.7 sccm, respectively, and the reactor
pressure was maintained at 70 mtorr. The filament array was resistively heated for 12 minutes to
achieve a thickness of approximately 300 nm. For the deposition of the p(C6PFA-co-EGDA)
coating, the C6PFA and EGDA were heated to 25°Cand 35°C, respectively. C6PFA, EGDA, and
TBPO were introduced into the reactor at flow rates of 0.2, 0.4, and 0.7 sccm, respectively, and
the reactor pressure was maintained at 70 mtorr. The filament array was resistively heated for 47
minutes to achieve a thickness of approximately 300 nm. For the deposition of the pC8PFA-
pEGDA coating, C8PFA and EGDA were heated to 50°C and 40°C, respectively. The pEGDA
layer was deposited first by introducing EGDA and TBPO at 0.1 and 0.6 sccm, respectively, and
the reactor pressure was maintained at 100 mtorr. The filament array was heated for 3 minutes in
order to achieve a thickness of 100 nm. Next, the filament array was turned off, the EGDA and
TBPO flows were halted, and the reactor was allowed to pump down to base pressure for
21
approximately 5 mins. The pC8PFA homopolymer was then deposited on top of the pEGDA layer
by introducing C8PFA and TBPO at 0.1 and 0.7 sccm, respectively, at a reactor pressure of 70
mtorr and heating the filament array for 35 minutes in order to achieve a thickness of 300 nm. For
the deposition of the pC6PFA-pEGDA coating, C6PFA and EGDA were heated to 50°C and 40°C,
respectively. The pEGDA layer was deposited as described above. The pC6PFA homopolymer
was then deposited on top of the pEGDA layer by introducing C6PFA and TBPO at 1.6 and 0.7
sccm, respectively, at a reactor pressure of 70 mtorr and heating the filament array for 11 minutes
in order to achieve a thickness of 300 nm.

2.3.3 Novec 7300 Solvent Test

The adhesion of the coatings on stainless steel was tested by immersion in a
hydrofluoroether solvent, 3M
TM
Novec
TM
7300. The protocol for the solvent test in this study
involved soaking the coated stainless steel substrates in Novec 7300 for two hours followed by
visual inspection to determine if the coating was present. The substrates were then immersed back
into Novec 7300 and sonicated at 75 °C for 30 minutes. The samples were then left in the solvent
overnight and air dried the next day.
2.3.4 Tape Test

Scotch tape (Magic
TM
) was used to test the adhesion of the coatings on stainless steel. The
tape was pressed onto the coated steel samples and a 2 lb weight was placed on top. The weight
was removed after 5 minutes while the tape remained on the samples overnight before removal.
2.3.5 PVP Fouling Test

The antifouling properties of the coatings were tested in a 1% PVP solution in DI water.
The coated samples were glued on a magnetic stir bar and placed in the 1% PVP solution in a
22
closed container at 85 °C. The samples were rotated at 100 rpm for 1 hour. The substrates were
then rinsed with deionized water and air dried overnight.
2.3.6 Characterization

The atomic compositions of the coated and uncoated stainless steel substrates were
characterized using the X-ray photoelectron spectrometer Kratos AXIS Ultra DLD (Kratos,
Manchester, UK) with a monochromatic Al Kα source. Survey spectra were taken on three spots
on each sample, from 1200 to 0 eV in 1 eV steps, and the spectrum for each spot was an average
of five scans with 100 ms dwell time each. The XPS spectra were analyzed using the CasaXPS
software Version 2.3.23rev1.2C (Casa Software Ltd., UK). Kratos sensitivity factors and Shirley
backgrounds were applied to the spectral regions for all elements but for the spectral regions of
the transition metals (Cr 2p, Mn 2p, Fe 2p, Ni 2p) where the U2 Tougaard background was applied.
The error within each spectrum was calculated based on the Monte Carlo method as implemented
in CasaXPS and the reported error bars were found by calculating the standard deviation over the
three measurements.
The water repellency of the coatings was characterized using contact angle goniometry
(ramé-hart model 290, ramé-hart instrument co New Jersey, USA). The reported static contact
angles were measured using 5 μL droplets of deionized water and the reported contact angles for
each coating are an average of ten measurements. The advancing and receding contact angle
measurements had a starting droplet volume of 5 μL with a step size of 1 μL and a maximum
droplet volume of 11 μL. Hysteresis was measured as the difference between the maximum and
minimum contact angles and averaged over three measurements on each sample. Error bars were
found by calculating the standard deviation.
23
Fourier transform infrared spectroscopy (FTIR) was conducted with a Nicolet iS 10
(Thermo Fisher Scientific Inc, Waltham, Massachusetts, USA) on a 1 μm thick pEGDA film
deposited on a silicon wafer via iCVD. The spectrum was baseline corrected on the SpectraGryph
software (version 1.2.14). The surface topography of the coatings was measured using atomic force
microscopy (AFM) (Dimension Icon, Bruker, Santa Barbara, California, USA) via peakforce
tapping (ScanAsyst mode) using ScanAsyst-air probes (Bruker, Santa Barbara, California, USA).
AFM force images were collected and roughness average (Ra) was estimated based on the height
profile using NanoScope Analysis (version 2).

24
2.4 Results and Discussion

Figure 2-1. (a) Chemical structures of the perfluorinated monomers and the EGDA crosslinker. (b)
Schematic representation of the two types of coating configurations: crosslinked p(C8PFA-co-
EGDA) and p(C6PFA-co-EGDA) coatings and graded pC8PFA-pEGDA and pC6PFA-pEGDA
coatings.
Figure 2-1a shows the chemical structures of the long chain (C8) and short chain (C6)
perfluorinated monomers and the crosslinker used in this study. Figure 2-1b shows the schematic
representation of the two coatings studied: crosslinked p(C8PFA-co-EGDA) and p(C6PFA-co-
EGDA) coatings and graded pC8PFA-pEGDA and pC6PFA-pEGDA coatings. To fabricate the
crosslinked coatings, the fluorinated monomer and the crosslinker were simultaneously introduced
into the reactor to allow for the crosslinker to be incorporated into the polymer backbone. To
fluoropolymer

substrate
substrate
p(C8PFA-co-EGDA)
p(C6PFA-co-EGDA)
pC8PFA-pEGDA
pC6PFA-pEGDA

(a)
(b)
crosslinked fluoropolymer
graded fluoropolymer
C8PFA
C6PFA
EGDA
pEGDA
O
O
(CF
2
)
7
CF
3
O
O
(CF
2
)
5
CF
3
O
O
O
O
EGDA

25
fabricate the graded coatings, first EGDA was polymerized to form a base layer of poly(ethylene
glycol diacrylate) (pEGDA) and then the fluorinated monomer was polymerized on top of the base
layer. This grading strategy promotes grafting of the fluorinated polymer chains onto the pEGDA
crosslinked base layer by reaction with unreacted vinyl bonds (Figure 2-2).

Figure 2-2. FTIR spectrum of the pEGDA crosslinked base layer. The assigned peaks correspond
to the vibrations of the unreacted vinyl bonds.
The stainless steel substrates were plasma cleaned prior to the deposition of the coatings.
The thickness of the p(C8PFA-co-EGDA) and p(C6PFA-co-EGDA) coatings was 300 nm as
measured by in-situ interferometry. For the graded coatings, the thickness of the pEGDA base
layer was 100 nm and the thickness of the pC8PFA and pC6PFA top layers was 300 nm. The AFM
images show the overall higher roughness (approximately 70 to 130 nm) is attributed to the large
features on the steel surface compared to the roughness on a smooth silicon wafer (approximately
1 to 3 nm) (Figure 2-3). The 500 nm AFM scans of the crosslinked and graded coatings on the
stainless steel substrates also show that the surface roughness ranges from 1 to 4 nm and therefore
the polymer is relatively smooth.

26

(a)
p(C8PFA-co-EGDA) coated S.S.
plasma cleaned bare S.S.
(g)
1 µm
1 µm 100 nm
100 nm
p(C8PFA-co-EGDA) coated silicon wafer
1 µm
(b)
1 µm
bare silicon wafer
(f)
pC8PFA-pEGDA coated S.S.
(c)
p(C6PFA-co-EGDA) coated S.S.
1 µm 100 nm
(d) (h)
p(C6PFA-co-EGDA) coated silicon wafer
1 µm
1 µm
pC6PFA-pEGDA coated S.S.
(e)
1 µm
27
Figure 2-3. Representative 5 micron and 500 nm AFM force images of (a) plasma cleaned bare
stainless steel (S.S.), and stainless steel that is coated with (b) p(C8PFA-co-EGDA), (c) pC8PFA-
pEGDA, (d) p(C6PFA-co-EGDA), and (e) pC6PFA-pEGDA polymers as deposited. 5 micron
AFM force images of (f) bare silicon wafer, and silicon wafers that are coated with (g) p(C8PFA-
co-EGDA) and (h) p(C6PFA-co-EGDA) polymers as deposited.

We tested the adhesion of the fluorinated polymer coatings on the stainless steel substrates
using tape test. As a control study, pC8PFA and pC6PFA homopolymer coatings that do not
contain the crosslinker were also deposited on the substrates. The tape partially removed the
homopolymers coatings. In contrast, the crosslinked p(C6PFA-co-EGDA) and p(C8PFA-co-
EGDA) coatings were fully intact after tape removal and there were no marks on the film. Similar
to the crosslinked fluoropolymers, there was no visible change in the graded pC8PFA-pEGDA
coating after the tape removal. However, part of the top layer of the graded pC6PFA-pEGDA
fluoropolymer was visibly peeled off after the tape test which is likely due to delamination of the
excess ungrafted pC6PFA chains. The AFM results confirm the unchanged morphology of the
crosslinked and graded coatings on the substrates after the tape test (Figure 2-4). These results are
consistent with studies that have shown that the adhesion of acrylic polymers is improved by
crosslinking.
34
In addition to increasing the robustness of the polymer coatings, the incorporation
of EGDA into the coatings allows for hydrogen bonding between the ethylene glycol polar
functionalities and the hydroxyl groups of the metal oxide layer of the stainless steel.
28
Figure 2-4. Representative 5 micron and 500 nm AFM force images of stainless steel (S.S.) coated
with (a) p(C8PFA-co-EGDA), (b) pC8PFA-pEGDA, (c) p(C6PFA-co-EGDA), and (d) pC6PFA-
pEGDA polymers as deposited and after tape test.

p(C8PFA-co-EGDA) coated S.S.
1 µm
after tape test
100 nm
1 µm 100 nm
as deposited
pC8PFA-pEGDA coated S.S.
after tape test
as deposited
(b)
p(C6PFA-co-EGDA) coated S.S.
as deposited
1 µm 100 nm
1 µm 100 nm
(c)
after tape test
pC6PFA-pEGDA coated S.S.
as deposited
after tape test
(d)
1 µm
1 µm
(a)
1 µm
1 µm
29

Figure 2-5. (a) Chemical structure of the fluorinated solvent Novec 7300. (b) Schematic
representation of the solvent test performed on the coated stainless steel substrates.

We also tested the adhesion between the coatings and the stainless steel substrate by
soaking and sonicating the samples in a hydrofluoroether solvent (Novec 7300) which is a good
solvent to dissolve fluorinated polymers
35
(Figure 2-5a). The coated stainless steel samples were
soaked in Novec 7300 at room temperature for 2 hours followed by sonication at 75 °C for 30
minutes. The samples were then left in the solvent overnight and were characterized by visual
inspection (Figure 2-5b). Both the C8 and C6 crosslinked and graded fluoropolymers were visually
present on the substrates after the solvent test. The AFM images indicate that mild topological
changes occur due to swelling in the solvent as shown in the 500 nm scan of the p(C8PFA-co-
EGDA) fluoropolymer coating (Figure 2-6). In contrast, the control pC8PFA and pC6PFA
homopolymer coatings without the EGDA crosslinker dissolved. The pC6PFA homopolymer
coating was not visually present after soaking in the solvent whereas the pC8PFA homopolymer
coating dissolved after sonication. These results combined with the tape tests demonstrate the
importance of incorporating EGDA into the polymer coating to prevent delamination.

soaking
ultrasonication
2 hours 30 mins
soaking
overnight
drying
visual analysis
drying
visual and chemical
analysis
Novec 7300
(a) (b)
F
3
C CF
3
O
F
F
F
CF
3
F
30

Figure 2-6. Representative 5 micron and 500 nm AFM force images of stainless steel (S.S.) coated
with (a) p(C8PFA-co-EGDA), (b) pC8PFA-pEGDA, (c) p(C6PFA-co-EGDA), and (d) pC6PFA-
pEGDA polymers as deposited and after solvent test.

p(C8PFA-co-EGDA) coated S.S.
after solvent test
1 µm 100 nm
1 µm 100 nm
(a)
as deposited
pC8PFA-pEGDA coated S.S.
after solvent test
as deposited
(b)
1 µm
1 µm
p(C6PFA-co-EGDA) coated S.S.
as deposited
1 µm 100 nm
after solvent test
1 µm 100 nm
(c)
pC6PFA-pEGDA coated S.S.
as deposited
after solvent test
(d)
1 µm
1 µm
31
Table 2-1. XPS atomic composition of the coatings as deposited and after the solvent test. The
error values represent the standard deviation within three positions in a sample.

pC8PFA p(C8PFA-co-EGDA) pC8PFA-pEGDA
atomic
composition
as
deposited
solvent
tested
as
deposited
solvent
tested
as
deposited
solvent
tested
O 4.6 ± 0.1 6.5 ± 2.3 9.7 ± 0.5 10.1 ± 0.2 5.7 ± 1.5 6.1 ± 0.2
C 34.1 ± 0.8 34.2 ± 0.4 38.2 ± 5.5 37.3 ± 5.2 32.8 ± 4.2 39.5 ± 2.8
F 61.1 ± 0.8 58.7 ± 2.7 52 ± 5.9 52.1 ± 4.8 61.3 ± 2.6 53.9 ± 2.7
pC6PFA p(C6PFA-co-EGDA) pC6PFA-pEGDA
atomic
composition
as
deposited
solvent
tested
as
deposited
solvent
tested
as
deposited
solvent
tested
O 6.2 ± 0.1 14.8 ± 1.0 19.6 ± 0.6 17.8 ± 0.1 6.2 ± 0.1 9.8 ± 0.2
C 36.9 ± 0.0 31.5 ± 1.5 47.1 ± 3.8 43.1 ± 4.1 35.4 ± 1.1 44.2 ± 4.1
F 56.8 ± 0.1 46.1 ± 0.9 32.8 ± 4.7 37.1 ± 3.8 58.2 ± 1.2 45.4 ± 4.3

We used XPS to characterize the chemical composition of the top 5-10 nm of the coatings
before and after the solvent test (Table 2-1). The atomic composition of the plasma cleaned
stainless steel was measured to be 41.7 ± 0.7 % O, 22.3 ± 6.4% C, 1.5 ± 0.3% N, 23.8 ± 4.9% Fe,
and 4.3 ± 1 % Cr with trace amounts (<2 %) of Na, S, Si, Ca, Ni, and F (Figure 2-8a). Although
the pC8PFA homopolymer coating was not visually present after the solvent test, the XPS data
shows full retention of the atomic composition likely because of the presence of a thin (>5nm)
polymer layer. The steel substrate was exposed in the pC6PFA homopolymer coating after solvent
testing as indicated by the decreases in the fluorine and carbon concentrations, and the increase in
the oxygen concentration in addition to the appearance of the elements associated with steel: 2.9
± 0.4% Fe, 1.2 ± 0.2% Cr, and 2.8 ± 0.7% Na. The crosslinked fluoropolymers both have similar
oxygen, carbon, and fluorine atomic percentages before and after the solvent test, confirming the
robustness of these coatings (Table 2-1, Figure 2-8). It is important to note that the degree of
crosslinking of the p(C8PFA-co-EGDA) and p(C6PFA-co-EGDA) coatings is crucial to obtain
robust films that are stable after solvent exposure and sonication. Solvent tests conducted on
p(C8PFA-co-EGDA) and p(C6PFA-co-EGDA) with flow rate ratios of EGDA/C8PFA=4 and
32
EGDA/C6PFA=1 partially delaminated and dissolved due to low crosslinking densities. Therefore,
the flow rate ratios were increased to EGDA/C8PFA=8 and EGDA/C6PFA=2 respectively to
prevent dissolution. Although the flow rate ratio was lower during the fabrication of p(C6PFA-co-
EGDA), the ratio of the EGDA to fluorinated monomer incorporated in the coating was estimated
to be 1.2 for p(C6PFA-co-EGDA) versus 0.3 for p(C8PFA-co-EGDA) based on the XPS data. For
the graded fluoropolymers, the fluorine content goes down from 61.3 ± 2.6% to 53.9 ± 2.7% for
pC8PFA-pEGDA and from 58.2 ± 1.2% to 45.4 ± 4.3% for pC6PFA-pEGDA after the solvent test.
This decrease is likely due to some of the ungrafted polymer chains washing away in the solvent.
Although the FTIR spectra of the pEGDA base layer shows the presence of vinyl peaks at 810,
1408, and 1632 cm
-1
which can lead to grafting of the fluoropolymer chains to the pEGDA base
layer (Figure 2-2), there are ungrafted chains that can be washed away by Novec 7300.

Figure 2-7. Average static water contact angles (CA) and hysteresis on the coated stainless steel
substrates. Hysteresis measurements after the solvent test were only conducted for the crosslinked
and graded fluoropolymers. The static contact angles are the average of ten measurements and the
33
hysteresis is the average of three measurements within each sample. The error bars are
representative of the standard deviation.

The contact angle of the uncoated stainless steel before and after plasma cleaning was
measured to be 92.5 ± 2.6° and 35.7 ± 2.5°, respectively. Consistent with literature, the pC6PFA
homopolymer coating has a smaller water contact angle and larger hysteresis (difference between
the advancing and receding contact angles) compared to the pC8PFA homopolymer coating due
to higher chain mobility
25
(Figure 2-7). The graded polymer coatings show the same trend with
respect to chain length since the pC6PFA-pEGDA coating has a smaller water contact angle and
larger hysteresis compared to the pC8PFA-pEGDA coating. The water contact angle significantly
decreases for the pC8PFA and pC6PFA homopolymers after the solvent test confirming that most
of the homopolymer coating is washed away which is consistent with the XPS results. In contrast,
the crosslinked C8 and C6 coatings show similar static contact angles (>100°) and hysteresis
before and after the solvent test demonstrating fluoropolymer retention consistent with the XPS
data. The graded fluoropolymers also show high contact angles (>115°) even though the XPS
results showed a decrease in the fluorine content after the solvent test. For reference, the pEGDA
crosslinked base layer has a static water contact angle of 45.1 ± 3.4° on stainless steel. The slight
decrease in the hysteresis of the graded fluoropolymers after the solvent test, specifically in the
case of pC6PFA-pEGDA coating, is likely a result of the lower chain mobility due to the decrease
in the thickness of the fluorinated top layer.
26

34

Figure 2-8. Representative XPS survey spectra of a) uncoated stainless steel (S.S.), b) p(C8PFA-
co-EGDA) coated stainless steel, c) p(C8PFA-co-EGDA) coated stainless steel after solvent test,
and d) p(C8PFA-co-EGDA) coated stainless steel after PVP test.

35
Table 2-2. Atomic composition of the fluorinated coatings before and after the PVP fouling test
for the as-deposited coatings and coatings that were solvent tested. The error values represent the
standard deviation from three spots in a sample.

p(C8PFA-co-EGDA) pC8PFA-pEGDA
as deposited solvent tested as deposited solvent tested
atomic
composition
after
PVP test
after
PVP test
after
PVP test
after
PVP test
O 9.7 ± 0.5 12.4 ± 0.5 10.1 ± 0.2 11.3 ± 0.3 5.7 ± 1.5 6 ± 0.8 6.1 ± 0.2 7.4 ± 0.1
C 38.2 ± 5.5 46.1 ± 0.5 37.3 ± 5.2 47.8 ± 1.2 32.8 ± 4.2 39.4 ± 2.8 39.5 ± 2.8 40.7 ± 1.1
F 52 ± 5.9 39.3 ± 0.2 52.1 ± 4.8 38.9 ± 1.6 61.3 ± 2.6 53.9 ± 4 53.9 ± 2.7 51.3 ± 1.2
N ND 1.8 ± 0.2 0.1 ± 0.1 1.8 ± 0.1 0 ± 0.1 0.7 ± 0.4 0.2 ± 0.1 0.4 ± 0.1
p(C6PFA-co-EGDA) pC6PFA-pEGDA
as deposited solvent tested as deposited solvent tested
atomic
composition
after
PVP test
after
PVP test
after
PVP test
after
PVP test
O 19.6 ± 0.6 17.6 ± 0.4 17.8 ± 0.1 15.4 ± 0.2 6.2 ± 0.1 6.3 ± 0.3 9.8 ± 0.2 9.3 ± 0.5
C 47.1 ± 3.8 59.8 ± 4 43.1 ± 4.1 53.6 ± 0.9 35.4 ± 1.1 36.6 ± 0.3 44.2 ± 4.1 46.5 ± 6
F 32.8 ± 4.7 20 ± 4.7 37.1 ± 3.8 28.6 ± 1 58.2 ± 1.2 57.1 ± 0.1 45.4 ± 4.3 43.1 ± 7
N ND 2.2 ± 0.4 ND 2.0 ± 0.2 0.1 ± 0.1 ND 0.1 ± 0.1 0.5 ± 0.2

Fouling and aggregation on metallic reactor mixers occur during the production of
polymeric materials which not only induces clogging and flow variations but also leads to cleaning
difficulties and decreased operation time. Here we study the fouling behavior of PVP polymer on
the crosslinked and graded fluoropolymer coatings. PVP is a water soluble polymer that is
produced in pharmaceutical and biomedical industries and therefore provides a good model
foulant.
36,37
The coated stainless steel substrates were rotated at 100 rpm in a 1% wt PVP solution
at 85 °C m in order to mimic the reaction temperature and the motion of the mixers in an industrial
setting. The theoretical atomic composition of PVP is 12.5 % N, 12.5 % O, and 75% C. The
uncoated stainless steel after the PVP test has a composition of 31.4 ± 3 % O, 50 ± 5.8 % C, 3.5 ±
0.3 % N, 8.7 ± 0.7 % Fe, and 4.6 ± 1.4 % Cr with trace amounts (<1 %) of Mn, S, Si, and Ni. The
increase in nitrogen from 1.5 ± 0.3% to 3.5 ± 0.3 % and the corresponding changes in C, O, and
Fe indicate adsorption and fouling of the PVP polymer on the steel surface. The crosslinked and
36
graded coatings that were tested in the Novec7300 solvent prior to the PVP test show similar
fouling compared to the as deposited coatings, confirming that the solvent test does not
mechanically or chemically change the functionality of the coatings. The XPS data shows that
compared to the uncoated stainless steel, both p(C8PFA-co-EGDA) and p(C6PFA-co-EGDA)
show a similar change in the nitrogen concentration from 0.1 ± 0.1 % to 1.8 ± 0.1 % and from 0
% to 2.0 ± 0.2 % after PVP testing, respectively (Figure 2-8d) . Noticeably, both crosslinked
fluoropolymers show similar fouling regardless of the different degrees of crosslinking and
fluorine concentration. (Table 2-2) In contrast, both of the graded fluoropolymers show minimal
fouling with a change in nitrogen concentration from 0.2 ± 0.1 % to 0.4 ± 0.1 % and from 0.1 ±
0.1 % to 0.5 ± 02 %, for pC8PFA-pEGDA and pC6PFA-pEGDA, respectively. This antifouling
capability is likely because the surface that is exposed to the hydrophilic PVP foulant is solely
composed of the perfluorinated moieties which have very weak Van der Waals intermolecular
interactions.
10
In contrast, the backbone of the crosslinked fluoropolymers is composed of the
fluorinated monomers and the EGDA crosslinker consisting of polarizable functionalities, leading
to more PVP adhesion. The XPS results indicate that the graded pC6PFA-pEGDA has similar
antifouling capabilities (N~0.5 ± 02 %) compared to pC8PFA-pEGDA (N~0.4 ± 0.1 %) and
therefore the C6 polymer coatings can be used as a more environmentally friendly alternative to
C8 coatings.

37
2.5 Conclusion

We have shown that incorporation of ethylene glycol diacrylate crosslinker into C8 and C6
fluoropolymer coatings improved the robustness of the coatings. The crosslinker can either be
incorporated into the polymer backbone via copolymerization or it can be used to form a base layer
that serves to adhere the fluoropolymer to the substrate. The coatings were applied to plasma
cleaned stainless steel substrates and contact angle measurements and XPS analysis demonstrated
the robustness of the coatings after sonication in a fluorinated solvent. The graded pC8PFA-
pEGDA and pC6PFA-pEGDA coatings were more effective at preventing fouling by PVP
compared to the p(C8PFA-co-EGDA) and p(C6PFA-co-EGDA) coatings at simulated mixing
conditions. The C6 graded fluoropolymer coating showed analogous antifouling properties
compared to the C8 graded fluoropolymer coating which is promising for sustainability since short
chain perfluorinated polymers have been shown to have less environmental accumulation.
Additional investigation of the chain mobility of the graded coatings should allow for further
development of antifouling coatings for stainless steel components for industrial applications.

2.6 Acknowledgements

This work was supported by the California Research Alliance (CARA) by BASF. I would
like to thank Dr. Sabine Hirth and Dr. Joshua Speros for the collaboration and useful discussions.
I would also like to thank Dr. Shuxing Li and the USC NanoBiophysics Core Facility for assisting
with the AFM measurements.

38
2.7 References

[1] Wu, W.; Zhu, Q.; Qing, F.; Han, C.C. Water Repellency on a Fluorine-Containing Polyurethane
Surface: Toward Understanding the Surface Self-Cleaning Effect. Langmuir 2009, 25,17-20.
[2] Yang, X.; Zhu, L.; Zhang, Y.; Chen, Y.; Bao, B.; Xu, J.; Zhou, W. Surface Properties and
Self-Cleaning Ability of the Fluorinated Acrylate Coatings Modified with Dodecafluoroheptyl
Methacrylate Through two Adding Ways. Applied surface science 2014, 295, 44-49.
[3] Lin, J.; Chen, X.; Chen, C.; Hu, J.; Zhou, C.; Cai, X.; Wang, W.; Zheng, C.; Zhang, P.; Cheng,
J.; Guo, Z. Durably Antibacterial and Bacterially Antiadhesive Cotton Fabrics Coated by Cationic
Fluorinated Polymers. ACS applied materials & interfaces 2018, 10, 6124-6136.
[4] Krishnan, S.; Ward, R.J.; Hexemer, A.; Sohn, K.E.; Lee, K.L.; Angert, E.R.; Fischer, D.A.;
Kramer, E.J.; Ober, C.K. Surfaces of Fluorinated Pyridinium Block Copolymers with Enhanced
Antibacterial Activity. Langmuir 2006, 22, 11255-11266.
[5] Zhan, X.; Yan, Y.; Zhang, Q.; Chen, F. A Novel Superhydrophobic Hybrid Nanocomposite
Material Prepared by Surface-Initiated AGET ATRP and its Anti-Icing Properties. Journal of
Materials Chemistry A 2014, 2, 9390-9399.
[6] Puretskiy, N.; Chanda, J.; Stoychev, G.; Synytska, A.; Ionov, L. Anti‐Icing Superhydrophobic
Surfaces Based on Core‐Shell Fossil Particles. Advanced Materials Interfaces 2015, 2, 1500124.
[7] Huang, T.C.; Yeh, L.C.; Lai, G.H.; Huang, B.S.; Yang, T.I.; Hsu, S.C.; Lo, A.Y.; Yeh, J.M.
Advanced Superhydrophobic Electroactive Fluorinated Polyimide and its Application in
Anticorrosion Coating. International journal of green energy 2017, 14, 113-120.
[8] Weng, C.J.; Chang, C.H.; Lin, I.L.; Yeh, J.M.; Wei, Y.; Hsu, C.L.; Chen, P.H. Advanced
Anticorrosion Coating Materials Prepared from Fluoro-Polyaniline-Silica Composites with
Synergistic Effect of Superhydrophobicity and Redox Catalytic Capability. Surface and Coatings
39
Technology 2012, 207, 42-49.
[9] Ferrero, F.; Periolatto, M. Application of Fluorinated Compounds to Cotton Fabrics via Sol–
Gel. Applied Surface Science 2013, 275, 201-207.
[10] Maier, G. Low Dielectric Constant Polymers for Microelectronics. Progress in polymer
science 2001, 26, 3-65.
[11]

Begolo, S.; Colas, G.; Viovy, J.L.; Malaquin, L. New Family of Fluorinated Polymer Chips
for Droplet and Organic Solvent Microfluidics. Lab on a Chip 2011, 11, 508-512.
[12] Imae, T. Fluorinated Polymers. Current opinion in colloid & interface science 2003, 8,
307-314.
[13]

Giannetti, E. Semi‐Crystalline Fluorinated Polymers. Polymer international 2001, 50, 10-26.
[14] Chen, B.; Kwong, P.; Gupta, M. Patterned Fluoropolymer Barriers for Containment of
Organic Solvents within Paper-Based Microfluidic Devices. ACS applied materials & interfaces
2013, 5, 12701-12707.
[15] Riche, C.T.; Marin, B.C.; Malmstadt, N.; Gupta, M. Vapor Deposition of Cross-linked
Fluoropolymer Barrier Coatings onto Pre-assembled Microfluidic Devices. Lab on a Chip
2011, 11, 3049-3052.
[16] Movsesian, N.; Dianat, G.; Gupta, M. Downstream Monomer Capture and Polymerization
during Vapor Phase Fabrication of Porous Membranes. Industrial & Engineering Chemistry
Research 2019, 58, 9908-9914.
[17] Cheng, C.; Gupta, M. Surface Functionalization of 3D-Printed Plastics via Initiated chemical
Vapor Deposition. Beilstein journal of nanotechnology 2017, 8, 1629-1636.
[18] Chan, K.; Kostun, L.E.; Tenhaeff, W.E.; Gleason, K.K. Initiated Chemical Vapor Deposition
of Polyvinylpyrrolidone-Based Thin Films. Polymer 2006, 47, 6941-6947.
40
[19] Tenhaeff, W.E.; Gleason, K.K. Initiated and Oxidative Chemical Vapor Deposition of
Polymeric Thin Films: iCVD and oCVD. Advanced Functional Materials 2008, 18, 979-992.
[20] Anton, D. Surface‐Fluorinated Coatings. Advanced Materials 1998, 10, 1197-1205.
[21] Coclite, A.M.; Shi, Y.; Gleason, K.K. Controlling the Degree of Crystallinity and Preferred
Crystallographic Orientation in Poly‐Perfluorodecylacrylate Thin Films by Initiated Chemical
Vapor Deposition. Advanced Functional Materials 2012, 22, 2167-2176.
[22] Perrotta, A.; Christian, P.; Jones, A.O.; Muralter, F.; Coclite, A.M. Growth Regimes of Poly
(perfluorodecyl acrylate) Thin Films by Initiated Chemical Vapor Deposition. Macromolecules
2018, 51, 5694-5703.
[23] Buck, R.C.; Franklin, J.; Berger, U.; Conder, J.M.; Cousins, I.T.; De Voogt, P.; Jensen, A.A.;
Kannan, K.; Mabury, S.A.; van Leeuwen, S.P. Perfluoroalkyl and Polyfluoroalkyl Substances in
the Environment: Terminology, Classification, and Origins. Integrated environmental assessment
and management 2011, 7, 513-541.
[24]

Liu, J.; Wang, N.; Buck, R. C.; Wolstenholme, B. W.; Folsom, P. W.; Sulecki, L. M.; Bellin,
C. A. Aerobic Biodegradation of [
14
C] 6:2 Fluorotelomer Alcohol in a Flow-Through Soil
Incubation System. Chemosphere 2010, 80, 716−723.
[25] Soto, D.; Ugur, A.; Farnham, T.A.; Gleason, K.K.; Varanasi, K.K. Short‐Fluorinated iCVD
Coatings for Nonwetting Fabrics. Advanced Functional Materials 2018, 28, 1707355.
[26] Liu, A.; Goktekin, E.; Gleason, K.K. Cross-linking and Ultrathin Grafted Gradation of
Fluorinated Polymers Synthesized via Initiated Chemical Vapor Deposition to Prevent Surface
Reconstruction. Langmuir 2014, 30, 14189-14194.
[27] Sojoudi, H.; McKinley, G.H.; Gleason, K.K. Linker-free Grafting of Fluorinated Polymeric
41
Cross-linked Network Bilayers for Durable Reduction of Ice Adhesion. Materials Horizons
2015, 2, 91-99.
[28] Yagüe, J.L.; Gleason, K.K. Enhanced Cross-linked Density by Annealing on Fluorinated
Polymers Synthesized via Initiated Chemical Vapor Deposition to Prevent Surface Reconstruction.
Macromolecules 2013, 46, 6548-6554.
[29] Coclite, A.M.; Shi, Y.; Gleason, K.K. Grafted Crystalline Poly‐perfluoroacrylate Structures
for Superhydrophobic and Oleophobic Functional Coatings. Advanced Materials 2012, 24,
4534-4539.
[30] Geddes, K.R. Process for Reducing Reactor Fouling During Polymerization in an Aqueous
Medium. U.S. Patent 4,957,982. 1990.
[31] Stephenson, T.; Kubis, A.; Derakhshesh, M.; Hazelton, M.; Holt, C.; Eaton, P.; Newman, B.;
Hoff, A.; Gray, M.; Mitlin, D. Corrosion-Fouling of 316 Stainless Steel and Pure Iron by Hot
Oil. Energy & Fuels 2011, 25, 4540-4551.
[32] Khalil, K.; Soto, D.; Farnham, T.; Paxson, A.; Katmis, A.U.; Gleason, K.; Varanasi, K.K.
Grafted Nanofilms Promote Dropwise Condensation of Low-Surface-Tension Fluids for High-
Performance Heat Exchangers. Joule 2019, 3, 1377-1388.
[33] Paxson, A.T.; Yagüe, J.L.; Gleason, K.K.; Varanasi, K.K. Stable Dropwise Condensation for
Enhancing Heat Transfer via the Initiated Chemical Vapor Deposition (iCVD) of Grafted Polymer
Films. Advanced Materials 2014, 26, 418-423.
[34] Peykova, Y.; Lebedeva, O. V.; Diethert, A.; Müller-Buschbaum, P.; Willenbacher, N.
Adhesive Properties of Acrylate Copolymers: Effect of the Nature of the Substrate and Copolymer
Functionality. International journal of adhesion and adhesives 2012, 34, 107-116.
[35] Cheng, D.F.; Masheder, B.; Urata, C.; Hozumi, A. Smooth Perfluorinated Surfaces with
42
Different Chemical and Physical Natures: Their Unusual Dynamic Dewetting Behavior Toward
Polar and Nonpolar Liquids. Langmuir 2013, 29, 11322-11329.
[36] Marks, J.A.; Wegiel, L.A.; Taylor, L.S.; Edgar, K.J. Pairwise Polymer Blends for Oral Drug
Delivery. Journal of pharmaceutical sciences 2014, 103, 2871-2883.
[37]

Teodorescu, M.; Bercea, M. Poly (vinylpyrrolidone)–a Versatile Polymer for Biomedical and
Beyond Medical Applications. Polymer-Plastics Technology and Engineering 2015, 54, 923-943.

43
Chapter 3: Downstream Monomer Capture and Polymerization
During the Vapor Phase Fabrication of Porous Polymers

Publication citation: Movsesian, N.; Dianat, G.; Gupta, M. Downstream Monomer Capture and
Polymerization during Vapor Phase Fabrication of Porous Membranes. Industrial & Engineering
Chemistry Research 2019, 58, 9908-9914.
3.1 Abstract

Porous membranes can be formed by polymerization of solid monomer by a vapor-phase
initiator followed by sublimation of the unreacted monomer. This versatile bottom-up process can
be used to deposit porous polymer membranes on a variety of substrates including planar, curved,
and structured surfaces. In this paper, we incorporate additional thermoelectric coolers (TECs) into
the reactor in order to study downstream monomer capture and polymerization. Better uniformity
across the TECs is achieved by capturing the unreacted monomer from the preceding surface
instead of capturing the monomer onto multiple surfaces at once. We show that the surface
temperature of the downstream TEC affects the membrane morphology and the kinetics of
polymerization. Lower capture temperatures lead to better surface coverage while higher
temperatures promote the polymerization rate. The morphological differences are further
confirmed by coating the membranes with a fluoropolymer to study variations in hydrophobicity.
Our ability to capture and polymerize monomer across multiple TECs provides a sustainable route
for enhancing the throughput of the process which is desirable for practical applications.
44
3.2 Introduction

Initiated chemical vapor deposition (iCVD) is a solventless free radical polymerization
technique that can be used to fabricate functional polymer films on a variety of substrates.
1,2

Monomer and initiator vapors are introduced into a vacuum chamber and the initiator is cleaved
into free radicals with a heated filament array. The monomer adsorbs to the surface of the substrate,
which is usually kept at room temperature, and polymerization occurs via a free radical
mechanism, leading to the formation of a dense film.

Our group has recently modified the iCVD
process to fabricate polymer membranes with dual scale porosity by maintaining the monomer
partial pressure and the substrate temperature below the triple point pressure and temperature of
the monomer using a thermoelectric cooler (TEC).
3-6
At these unconventional conditions, the
monomer deposits as solid microstructures. Polymerization occurs at both the vapor-solid interface
and within the solid structure and the excess unreacted monomer is subsequently sublimated.
7-9
There is no unreacted monomer in the final polymer membranes as confirmed by Fourier-
transform infrared spectroscopy analysis.
3
Compared to solution phase processes for fabricating
porous polymer membranes, such as thermally induced phase separation
10,11
and solvent casting
and particulate leaching,
12,13
our bottom up solventless approach allows deposition of membranes
onto a variety of substrates including porous membranes for the fabrication of porous-on-porous
structures.
14

In this paper, we introduce multiple TECs in the reactor to study deposition onto larger
areas. The iCVD process is commonly run at continuous flow where the monomer and initiator
vapors are constantly fed into the chamber and are exhausted by a vacuum pump.
15
CVD systems
can also be operated as batch systems to decrease material waste.
16-19
For example, Jankowska-
Kuchta et al. deposited tungsten in a non-flowing batch reactor to reduce the use of high purity
45
and toxic gases.
16
Kukovitsky et al. used a closed batch CVD configuration for the fabrication of
carbon nanotubes.
17
Petruczok et al. recently developed a closed-batch iCVD method in which the
reactor chamber was first filled with reactants and then isolated from the feed and exhaust lines
which led to an improved reaction yield and less accumulation of waste.
19
In our current study, we
run our process under a semibatch processing mode where the initiator flows during both the
monomer deposition and polymerization steps but there is no additional monomer flow during
polymerization in order to reduce material waste. The continuous initiator flow provides steady
introduction of free radicals during polymerization. Here we study two different systems for
monomer deposition where the additional TECs serve as downstream cold traps: the deposition of
monomer onto multiple TECs at once and the capture of sublimating monomer from the preceding
surface.
We systematically study how the temperature of the downstream cold traps affects the
morphology of the membranes and the kinetics of polymerization. An advantage of our fabrication
process is the ability to modify the surface properties of the porous membranes with a top coating
by transitioning from the conventional iCVD process to the unconventional iCVD process.
20
We
used this technique to apply a dense fluorinated coating on the membranes which allowed us to
study morphological variations as a function of processing temperature by assessing water
repellency. Combinatorial methods for the synthesis and screening of a vast range of materials
have long attracted interest because they can accelerate discovery and lead to high throughput
production and optimization of new materials.
21-23
The series of cold traps in our system allows
for combinatorial synthesis of porous polymer coatings in a single run. Our platform adds to
previously reported combinatorial vapor phase deposition methods for the fabrication of organic
46
and inorganic material libraries with compositional gradients such as antimicrobial polymers
24
,
amorphous and microcrystalline silicon
25,26
, and thin films of magnesium and aluminum
27
.

3.3 Experimental Section

3.3.1 Materials
Methacrylic acid (MAA) (Aldrich, 99%), tert-butyl peroxide (TBPO) (Aldrich, 98%), and
1H,1H,2H,2H-perfluorodecyl acrylate (PFDA) (SynQuest Laboratories, 97%) were used as
received.
3.3.2 Fabrication of Porous Polymers via iCVD
Poly(methacrylic acid) (PMAA) membranes were fabricated in a custom designed initiated
chemical vapor deposition (iCVD) chamber of 250 mm diameter and 48 mm height (GVD
Corporation) containing a nichrome filament array (Omega Engineering, 80/20% Ni/Cr). The stage
temperature was maintained at 20 °C using a recirculating chiller (Thermo Scientific Haake A25).
A rotary vane vacuum pump (Edwards E2M40) was used to achieve vacuum and the reactor
pressure was controlled by a throttle valve (MKS 153D) which was regulated by a capacitance
manometer (MKS Baratron 622A01TDE). MAA and TBPO feed jars were mounted onto the
reactor and maintained at 35 and 28 °C, respectively. A needle valve was used to meter the MAA
flow and a mass flow controller (MKS 1479 A) was used to meter the TBPO flow. The 4´4 cm
2
thermoelectric coolers (TECs) (TE Technology) were placed onto the reactor stage approximately
1 cm apart from each other with TEC1 closest to the feed lines. Silicon wafers (Wafer World 119)
were cut into 3´3 cm
2
pieces and placed on top of the TECs. The temperature of each TEC was
regulated using an adjustable DC power supply (Volteq HY3010D). Since methacrylic acid freezes
at approximately 16 ± 1 °C, all TEC temperatures were kept at 10 °C or below for all experiments.
47

Figure 3-1. Schematic representation of multiple TECs in the iCVD reactor.

Figure 3-1 shows a schematic of the iCVD reactor with multiple TECs. Silicon wafers were
placed on the TECs for sample collection. For deposition and polymerization on only TEC1, the
temperature of TEC1 was first maintained at -10 °C. TEC2 and TEC3 were kept off the whole time.
TBPO was introduced into the chamber at 0.7 sccm to achieve a total pressure of 650 mTorr. Then
MAA was introduced into the chamber at a flow rate of 0.13 sccm for 5 minutes to cause monomer
deposition as solid microstructures. The temperature of TEC1 was then increased to 10°C and the
filament array was heated to 240 °C to start polymerization for 30 mins. For the experiments with
deposition onto multiples TECs at once, monomer deposition on TEC1 and TEC2 and on TEC1,
TEC2, and TEC3 were performed with all TECs set to either -10 °C or -20 °C during the 5 minutes
of monomer deposition and subsequently the temperature of all TECs was increased to 10 °C
during 30 minutes of polymerization. After polymerization, the filament was turned off, the TBPO
flow was halted, and the TECs used during each of the experiments were set to 5 °C to allow for
sublimation of the remaining unreacted monomer which was confirmed by the reactor returning to
a base pressure of 16-18 mTorr. Cracks form when the monomer sublimation is too fast, therefore
we decreased the sublimation rate by decreasing the temperature of the TEC from 10 °C to 5 °C.
48

Figure 3-2. Stepwise process of capturing and polymerizing monomer from the preceding TEC.

We also studied a system where the monomer is captured from the preceding TEC surface
instead of being captured onto all TECs at once. Figure 3-2 shows the schematic of a system of
two TECs. In step 1, monomer deposition on TEC1 followed the same reactor conditions described
above while TEC2 and TEC3 were kept off. After this step, no additional monomer was introduced
into the process. In step 2, the temperature of TEC1 was increased to 10 °C, the temperature of
TEC2 was set to -20°C, -10°C, or 0°C, and the filament array was heated to 240 °C to cleave the
initiator into free radicals. Polymerization occurred on both TECs for 30 minutes and TEC2 served
as a cold trap to partially capture and polymerize monomer that sublimated from TEC1. In step 3,
polymerization was stopped, and sublimation of the unreacted monomer occurred at 5°C. For the
experiments with an additional TEC (TEC3), steps 1 and 2 were identical to those shown in Figure
2. Monomer capture and polymerization on TEC2 occurred at -20°C for 30 mins. Next, the
temperature of TEC2 was increased to 10°C and the temperature of TEC3 was turned to -20°C for
an additional 30 minutes of monomer capture and polymerization. In step 4, polymerization was
stopped and all the TECs were set to 5 °C to sublimate the remaining unreacted monomer. The
49
reactor pressure, the initiator flow rate, the monomer flow rate during deposition, and the
temperature of TEC1 during deposition and polymerization are fixed for all the experiments. All
polymer masses were measured on a Metter AE 160 scale and averaged over three samples.
3.3.3 Fabrication of Hydrophobic Dense Coatings for Porous Polymers via iCVD
The poly(1H,1H,2H,2H-perfluorodecyl acrylate) (PPFDA) depositions onto the PMAA
membranes were performed using conventional iCVD parameters. The reactor stage was kept at
30 °C and the PFDA jar was maintained at 50 °C. PFDA and TBPO were introduced into the
reactor at 0.2 sccm and 1.5 sccm, respectively. The reactor pressure was maintained at 70 mTorr
and the nichrome filament array was resistively heated to 240 °C for 45 minutes. The deposition
rate was 2 nm/min which was monitored on a reference silicon wafer by in-situ interferometry
using He−Ne laser (Industrial Fiber Optics, 633 nm).
3.3.4 Characterization
The surface properties of the PPFDA coated porous membranes were characterized using
contact angle goniometry (rame-́hart 290). The reported static contact angles were averaged over
10 measurements with a droplet volume of 10 μL of deionized water and the advancing and
receding contact angle measurements had a starting droplet volume of 5 μL with a step size of 1
μL and a maximum droplet volume of 10 μL. Hysteresis was measured as the difference between
the maximum and minimum contact angles and was averaged over 3 measurements. Error bars
were found by calculating the standard deviation. All samples were imaged by scanning electron
microscopy (SEM; Topcon Aquila) at 20 kV accelerating voltage. Prior to imaging, gold was
sputtered onto the polymer samples for 30 s at 20 mA to avoid charging.

50
3.4 Results and Discussion

In order to understand the effect of adding multiple TECs into the reactor, we compared
the morphology and mass of the membranes for a system of one, two, and three TECs. The TEC
temperature was kept at -10 °C during 5 min of monomer deposition and was increased to 10 °C
during 30 min of polymerization. Initiator was introduced at a constant flow rate during both the
deposition and polymerization steps, however there was no additional monomer introduced during
the polymerization step. Unreacted monomer was sublimated after polymerization and the masses
of the membranes were measured. For the case of polymerization on only one TEC, the mass was
29.4 ± 3.3 mg. For the case of polymerization onto two TECs at once, the mass was 20 ± 0.3 mg
and 22 ± 4.1 mg on TEC1 and TEC2, respectively. For the case of polymerization onto three TECs
at once, the mass was 17.2 ± 2.5 mg, 14.8 ± 2.2 mg, and 8.4 ± 2.3 mg for TEC1, TEC2, and TEC3,
respectively. In addition, the membrane thickness was shown to decrease as more TECs were
added. For the case of polymerization on only one TEC, the thickness was 900-1000 μm. For the
case of polymerization onto two TECs at once, the thickness was 700-900 μm on both TEC1 and
TEC2. For the case of polymerization onto three TECs at once, the thickness was 200-850 μm,
200-600 μm, and 80-200 μm for TEC1, TEC2, and TEC3, respectively. Based on these results, we
can conclude that the mass and thickness per area decreases as more TECs are added into the
system. In our process, the deposited monomer serves as the template for polymerization and
therefore affects the morphology of the resulting membrane.
3,4
51

Figure 3-3. (a) In-situ optical images of the monomer deposited on TEC1, TEC2, and TEC3 at
-10 °C at once and the corresponding (b) optical and (c) SEM images of the porous membranes
after polymerization and sublimation of the unreacted monomer.

Figure 3-3a shows the in-situ optical images of the monomer prior to polymerization on
the three TECs. It can be seen that the deposition of monomer onto TEC3 is not uniform. The
corresponding optical and SEM images of the final porous membranes in Figures 3-3b and 3-3c
confirm the loss of uniformity and pillar-like morphology on TEC2 and TEC3. In order to determine
if the deposition temperature affects the uniformity of the membranes, we also deposited monomer
onto all three TECs at -20 °C for 5 minutes and polymerized at 10 °C for 30 minutes.
52

Figure 3-4. (a) In-situ optical images of the monomer deposited on TEC1, TEC2, and TEC3 at
-20 °C at once and the corresponding (b) optical and (c) SEM images of the porous membranes
after polymerization and sublimation of the unreacted monomer.
Figure 3-4a shows that the monomer deposits more uniformly on all the TECs compared
to Figure 3-3a and therefore the resulting membranes in Figure 3-4b are also more uniform than
in Figure 3-3b. As shown in the SEM images in Figure 3-4c, these membranes have pillar-like
structures due to greater surface nucleation at lower temperatures and dual-scale porosity,
consistent with our previous studies.
3,4,7
The larger pores form due to the void spaces between the
solid microstructures and smaller pores form within the microstructures due to the sublimation of
the monomer. At both deposition temperatures (-10 °C and -20 °C), the membrane formed on
TEC3 has significant structural loss. Since all three TECs are kept at the same temperature during
monomer deposition and polymerization, the nonuniformity is likely caused by the monomer flow
pattern across the reactor.
In order to test whether we could achieve better uniformity across multiple TECs by
changing the flow pattern of the monomer, we studied a system where the unreacted monomer is
53
captured from the preceding surface as shown in Figure 3-2 instead of capturing the monomer onto
multiple surfaces at once. For our preliminary studies, we focused on a system of two TECs.
Monomer was deposited onto TEC1 in the first step. After this step, no additional monomer was
introduced into the process. Our motivation for this setup was that our data from Figure 3-3a
indicated that the use of only one TEC resulted in the highest polymer mass per area. In the second
step, the temperature of TEC1 was increased to 10 °C and the temperature of TEC2 was set to
-20°C, -10°C, or 0°C. The monomer that sublimated from TEC1 was partially captured as solid
microstructures on TEC2. During this step, the filament array was heated to cleave the initiator into
free radicals and polymerization occurred on both TECs for 30 minutes. The temperature of TEC1
was increased from -10 °C to 10 °C to enable a faster polymerization rate and faster monomer
sublimation, both of which increase the conversion of the process.
8

54

Figure 3-5. (a) Top down in-situ optical images of monomer capture and polymerization on TEC2
as a function of time at temperatures of -20, -10, and 0 °C while polymerization on TEC1 occurs
at 10 °C in all cases. (b) Images of polymer membranes formed on TEC2 after sublimation of the
unreacted monomer.

The in-situ optical images in Figure 3-5a show monomer transfer and polymerization on
TEC2 within the reactor as a function of time. A lower TEC2 temperature (-20 °C) leads to more
nucleation spots on the silicon surface resulting in uniform coverage whereas a higher TEC2
temperature (0 °C) results in larger patterns and less coverage. After sublimation, the membranes
were removed from the reactor and imaged (Figure 3-5b). The macroscopic structures of the final
polymer membranes are consistent with the in-situ images inside the reactor. The membranes
55
formed at -20 °C have better surface coverage than the membranes formed at 0 °C which have
cracks and voids.

Figure 3-6. SEM images of the porous membranes formed on (a) TEC1 and (b) TEC2 as a result
of monomer capture and polymerization from the preceding surface.
The SEM images in Figure 3-6a show that the pillar microstructures formed on TEC1 are
not affected by the temperature variations on TEC2. The SEM images in Figure 3-6b show that the
membranes formed at higher TEC2 temperatures have denser structures than the membranes
formed at lower temperatures due to lack of nucleation, which is consistent with the macroscopic
images shown in Figures 3-5a and 3-5b. An advantage of our fabrication process is that we can
vary processing parameters in-situ to apply a thin polymer coating onto the membranes without
affecting their morphology by transitioning from unconventional iCVD parameters to
conventional iCVD parameters.
20
The fluorinated polymer PPFDA was deposited onto the
hydrophilic PMAA membranes and the resulting hydrophobicity was studied using contact angle
goniometry. PPFDA is a hydrophobic polymer that has a water contact angle of approximately
120 ° on a flat surface.
28
High surface roughness enhances hydrophobicity and therefore contact
angle measurements provide information about the roughness of the membranes.
29

56

Figure 3-7. (a) SEM images of the membranes formed on TEC2 after PPFDA coating and (b) their
corresponding static water contact angle.

The SEM images after PPFDA coating in Figure 3-7a show a similar morphology to those
in Figure 3-6b, indicating no significant change in structure after coating. The membranes formed
on TEC1 are very rough due to the pillar microstructures and therefore water droplets immediately
roll off after coating with PPFDA, indicating superhydrophobicity similar to the Lotus Leaf
effect
30
. The membranes formed on TEC2 were less hydrophobic, indicating less roughness as
shown in Figure 3-6a. The water contact angle decreases with increasing TEC2 temperature due to
the denser structures and the hysteresis (differences in the maximum advancing and minimum
receding contact angles) increases with increasing substrate temperature and was measured to be
10.2 ± 2.5 °, 20.2 ± 2.1 °, and 27.2 ± 2.9 ° for membranes formed at -20 °C, -10 °C, and 0 °C,
respectively (Figure 3-7b).

(b)
120.8 ± 11.6 ° 134.8 ± 6.0 ° 147.4 ± 4.6 °
TEC
2
membranes after PPFDA coating
0
∘
C
-10
∘
C -20
∘
C
(a)
100 μm 100 μm 100 μm
static contact angle
57
Table 3-1. Polymer mass obtained on TEC1 and TEC2 as a result of monomer capture and
polymerization from the preceding surface at varying TEC2 temperatures.
TEC1 Temperature (
∘
C) TEC2 Temperature (
∘
C) Polymer Mass (mg)
deposition polymerization simultaneous deposition and polymerization TEC1 TEC2
-10 10 -20 20.4 ± 4.9 5.2 ± 1.6
-10 10 -10 22.4 ± 1.3 10.2 ± 0.5
-10 10 0 23.7 ± 3.0 9.4 ± 0.9

Table 3-1 shows that the polymer mass on TEC1 is relatively constant at varying TEC2
temperatures likely due to the high initial monomer concentration on TEC1 (~600 mg) in addition
to the slow solid phase polymerization kinetics (approximately 10 wt % or less of monomer is
converted)
8

which is consistent with previously reported solid phase polymerization techniques.
31-
33
The polymer mass on TEC2 increases from -20 °C to -10 °C due to faster polymerization rates
at higher temperatures. This is consistent with our recent studies that show that increasing the
polymerization temperature increases the sublimation rate of the monomer, leading to increased
polymer mass at the vapor-solid interface.

However, there is no significant change from -10 °C to
0 °C. Although the polymerization rate is higher at 0 °C, the membrane formed at -10 °C has a
larger surface area which allows for more polymerization at the vapor-solid interface.
8

We further studied the capability to capture and polymerize monomer from the preceding
surfaces by adding a third TEC. After monomer capture and polymerization on TEC2 at -20 °C for
30 min as described in Figure 3-2, the temperature of TEC2 was increased to 10 °C and TEC3 was
set to -20 °C for an additional 30 min of capture and polymerization. The remaining unreacted
monomer in the system was then sublimated. We performed the monomer capture and
polymerization on TEC2 and TEC3 at -20 °C in order to obtain the best surface coverage.
58

Figure 3-8. (a) Optical images and (b) the corresponding SEM images of the porous membranes
formed on TEC1, TEC2, and TEC3 as a result of monomer capture and polymerization from the
preceding surface.

Figure 3-8 shows that the optical and SEM images of the membranes have uniform patterns
across each silicon wafer. The membranes formed on TEC1 and TEC2 have similar morphologies
as those formed in the two TEC system for the case of TEC2 at -20 °C in Figure 3-5b, however
changing the temperature of TEC2 from -20 °C to 10 °C promotes a faster sublimation rate which
leads to enlargement of the smaller scale pores within the microstructures. Monomer capture and
polymerization on TEC3 extends the polymerization time to 60 mins for the membranes formed
on TEC1 and TEC2 which results in polymer masses of 35.6 ± 2.7 mg and 37.7 ± 1.4 mg,
respectively. The polymer mass on TEC3 as a result of 30 min of monomer capture and
polymerization at -20 °C is 4.9 ± 0.6 mg which is consistent with the polymer mass obtained on
TEC2 at a capture and polymerization temperature of -20 °C in the system of two TECs (Table 3-
1). The morphology across each TEC was more uniform compared to deposition onto all three
TECs at once, indicating that downstream capture and polymerization from the preceding surface
is a possible route toward achieving uniformity across multiple TECs.
59
In addition to the controllable morphologies obtained in this process, the efficiency of the
technique is increased by making the sublimating monomer in the system a resource by continual
capture and polymerization on new surfaces where the monomer surface area for polymerization
is increased. The addition of two extra TECs as downstream cold traps during the 60 minutes of
polymerization (period for the introduction of initiator free radicals) results in a total polymer mass
of 78.2 ± 3.1 mg which is approximately 49% higher than the polymer mass of 52.4 ± 4.1 mg,
obtained by polymerization of the solid monomer deposited on TEC1 for 60 minutes. This leads to
a 4% increase in polymer conversion from 9% to 13%. Based on these results, downstream
monomer capture and polymerization can increase the efficiency of our vapor phase porous
polymer fabrication by reducing monomer waste and improving conversion and throughput.
Additional cold traps can be used for further capture and polymerization until the complete
depletion of all the unreacted monomer in the system.

3.5 Conclusion

We have shown that multiple TECs can be incorporated in an iCVD reactor. We showed
that deposition of monomer onto multiple TECs at once results in less uniformity than capturing
and polymerizing the monomer from the preceding surface. We studied the effect of temperature
on the reaction kinetics and the morphology of the membranes formed as a result of downstream
capture. Lower temperatures lead to better surface coverage and increased surface roughness.
Higher temperatures lead to more mass, however there is less nucleation leading to poor surface
coverage. Based on our results, it is beneficial to capture monomer from the preceding surfaces at
lower temperatures to achieve better uniformity across multiple surfaces. Since there was structural
60
loss in the pillar morphology across the length of the reactor, future work will focus on developing
new reactor configurations that can enable the production of more consistent morphologies.

3.6 Acknowledgements

This work was supported by the National Science Foundation CAREER Award CMMI-1252651.

61
3.7 References
[1] Im, S.G.; Gleason, K.K. Solvent‐Free Modification of Surfaces with Polymers: The Case for
Initiated and Oxidative Chemical Vapor Deposition (CVD). AIChE journal 2011, 57(2), 276-285.
[2] Baxamusa, S.H; Gleason, K.K. Thin Polymer Films with High Step Coverage in Microtrenches
by Initiated CVD. Chemical Vapor Deposition 2008, 14(9‐10), 313-318.
[3] Seidel, S.K.; Kwong, P.; Gupta, M. Simultaneous Polymerization and Solid Monomer
Deposition for the Fabrication of Polymer Membranes with Dual-Scale Porosity. Macromolecules
2013, 46, 2976-2983.
[4] Seidel, S.; Gupta, M. Systematic Study of the Growth and Morphology of Vapor Deposited
Porous Polymer Membranes. J. Vac. Sci. Technol. A. 2014, 32(4), 041514.
[5] Seidel, S.; Chu Cheong, C.; Kwong, P.; Gupta, M. All‐Dry Fabrication of Poly (methacrylic
acid)‐Based Membranes with Controlled Dissolution Behavior. Macromol. Mater. Eng. 2015,
300(11), 1079-1084.
[6] Seidel, S.; Dianat, G.; Gupta, M. Formation of Porous Polymer Coatings on Complex
Substrates Using Vapor Phase Precursors. Macromol. Mater. Eng. 2016, 301(4), 371-376.
[7] Dianat, G.; Gupta, M. Sequential Deposition of Patterned Porous Polymers Using Poly
(dimethylsiloxane) masks. Polymer 2017, 126, 463-469.
[8] Dianat, G.; Movsesian, N.; Gupta, M. Process-Structure-Property Relationships for Porous
Membranes Formed by Polymerization of Solid Monomer by a Vapor-Phase Initiator.
Macromolecules 2018, 51(24), 10297-10303.
[9] Movsesian, N.; Tittensor, M.; Dianat, G.; Gupta, M.; Malmstadt, N. Giant Lipid Vesicle
Formation Using Vapor-Deposited Charged Porous Polymers. Langmuir 2018, 34(30), 9025-
9035.
62
[10] Caneba, G.T.; Soong, D.S.; Polymer Membrane Formation Through the Thermal-Inversion
Process. 1. Experimental Study of Membrane Structure Formation. Macromolecules 1985, 18,
2538-2545.
[11] Matsuyama, H.; Yuasa, M.; Kitamura, Y.; Teramoto, M.; Lloyd, D.R.; Structure Control of
Anisotropic and Asymmetric Polypropylene Membrane Prepared by Thermally Induced Phase
Separation. Journal of Membrane Science 2000, 179, 91-100.
[12] Mikos, A.G.; Thorsen, A.J.; Czerwonka, L.A.; Bao, Y.; Langer, R.; Winslow, D.N.; Vacanti,
J.P.; Preparation and Characterization of Poly (L-lactic acid) Foams. Polymer 1994, 35, 1068-
1077.
[13] Yang, Q.; Chen, L.; Shen, X.; Tan, Z.; Preparation of Polycaprolactone Tissue Engineering
Scaffolds by Improved Solvent Casting/Particulate Leaching Method. Journal of Macromolecular
Science, Part B: Physics 2006, 45, 1171-1181.
[14] Kwong, P.; Seidel, S.; Gupta, M.; Solventless Fabrication of Porous-on-Porous Materials.
ACS Applied Materials and Interfaces 2013, 5, 9714-9718.
[15] Alf, M.E.; Asatekin, A.; Barr, M.C.; Baxamusa, S.H.; Chelawat, H.; Ozaydin‐Ince, G.;
Petruczok, C.D.; Sreenivasan, R.; Tenhaeff, W.E.; Trujillo, N.J.; Vaddiraju, S.; Chemical Vapor
Deposition of Conformal, Functional, and Responsive Polymer Films. Advanced Materials
2010, 22(18), 1993-2027.
[16] Jankowska-Kuchta, E. B.; McConica, C.; Moss, D.; Kozlowski, J. X-Ray Measurement of
Tungsten Films Grown in a Nonflowing Laser-Induced CVD Process, Proc. SPIE 3274, Laser
Applications in Microelectronic and Optoelectronic Manufacturing III, San Jose, CA, January 24,
1998, 331−340.
63
[17] Kukovitsky, E. F.; Lvov, S. G. Increased Carbon Chemical Vapor Deposition and Carbon
Nanotube Growth on Metal Substrates in Confined Spaces. ECS J. Solid State Sci. Technol. 2013,
2, M1−M8.
[18] Tang, J.; Fan, G.; Li, Z.; Li, X.; Xu, R.; Li, Y.; Zhang, D.; Moon, W.-J.; Kaloshkin, S. D.;
Churyukanova, M. Synthesis of Carbon Nanotube/Aluminium Composite Powders by Polymer
Pyrolysis Chemical Vapor Deposition. Carbon 2013, 55, 202−208.
[19] Petruczok, C.D.; Chen, N.; Gleason, K.K. Closed Batch Initiated Chemical Vapor Deposition
of Ultrathin, Functional, and Conformal Polymer Films. Langmuir 2014, 30(16), 4830-4837.
[20] Dianat, G.; Seidel, S.; De Luna, M.M.; Gupta, M. Vapor Phase Fabrication of Hydrophilic
and Hydrophobic Asymmetric Polymer Membranes. Macromol. Mater. Eng. 2016, 301(9), 1037-
1043.
[21] Jandeleit, B.; Schaefer, D.J.; Powers, T.S.; Turner, H.W.; Weinberg, W.H. Combinatorial
Materials Science and Catalysis. Angewandte Chemie International Edition 1999, 38(17), 2494-
2532.
[22] Thorstenson, J.B.; Petersen, L.K.; Narasimhan, B. Combinatorial/High Throughput Methods
for the Determination of Polyanhydride Phase Behavior. Journal of combinatorial chemistry
2009, 11(5), 820-828.
[23] Atwater, J. Mattes, D.S. Streit, B. von Bojničić‐Kninski, C. Loeffler, F.F. Breitling, F. Fuchs,
H. Hirtz, M. Combinatorial Synthesis of Macromolecular Arrays by Microchannel Cantilever
Spotting (µCS). Advanced Materials 2018, 30(31), 1801632.
[24] Martin, T.P.; Gleason, K.K. Combinatorial Initiated CVD for Polymeric Thin
Films. Chemical Vapor Deposition 2006, 12(11), 685-691.
64
[25] Wang, Q.; Yue, G.; Li, J.; Han, D. A Combinatorial Study of Materials in Transition from
Amorphous to Microcrystalline Silicon. Solid State Communications 1999, 113(3), 175-178.
[26] Wang, Q. Combinatorial Hot-Wire CVD Approach to Exploring Thin-Film Si Materials and
Devices. Thin Solid Films 2003, 430(1-2), 78-82.
[27] Garcia, G.; Doménech-Ferrer, R.; Pi, F.; Santiso, J.; Rodríguez-Viejo, J.; Combinatorial
Synthesis and Hydrogenation of Mg/Al Libraries Prepared by Electron Beam Physical Vapor
Deposition. Journal of combinatorial chemistry 2007, 9(2), 230-236.
[28] Gupta, M.; Gleason, K.K. Initiated Chemical Vapor Deposition of Poly (1H, 1H, 2H, 2H-
perfluorodecyl acrylate) Thin Films. Langmuir 2006, 22(24), 10047-10052.
[29] Gupta, M.; Kapur, V.; Pinkerton, N.M.; Gleason, K.K. Initiated Chemical Vapor Deposition
(iCVD) of Conformal Polymeric Nanocoatings for the Surface Modification of High-Aspect-Ratio
Pores. Chemistry of Materials 2008, 20(4), 1646-1651.
[30] Latthe, S.; Terashima, C.; Nakata, K.; Fujishima, A. Super Hydrophobic Surfaces Developed
by Mimicking Hierarchial Surface Morphology of Lotus Leaf. Molecules 2014 19(4), 4256-4283.
[31] Eastmond, G.C.; Solid-State Polymerization. Progress in Polymer Science 1970, 2, 1-46.
[32] Bamford, C. H.; Jenkins, A. D.; Ward, J.C.; Polymerization in the Solid and Near‐Solid States.
Journal of Polymer Science Part A: Polymer Chemistry 1960, 48, 37-51.
[33] Bamford, C.H.; Eastmond, G.C.; Ward JC. Studies in Polymerization XIV. The Solid-State
Polymerization of Acrylic and Methacrylic Acids. Proc. R. Soc. Lond. A 1963, 271, 357-378.

65
Chapter 4: Giant Lipid Vesicle Formation Using Vapor-Deposited
Charged Porous Polymers

Publication citation: Movsesian, N.; Tittensor, M.; Dianat, G.; Gupta, M.; Malmstadt, N. Giant
Lipid Vesicle Formation Using Vapor-Deposited Charged Porous Polymers. Langmuir 2018, 34,
9025-9035.
4.1 Abstract

In this study, we prepare giant lipid vesicles using vapor-deposited charged microporous
poly(methacrylic acid-co-ethylene glycol diacrylate) polymer membranes with different
morphologies and thicknesses. Our results suggest that vesicle formation is favored by thinner,
more structured porous hydrogel substrates. Electrostatic interactions between the polymer and the
lipid head groups affect vesicle yield and size distribution. Repulsive electrostatic interactions
between the hydrogel and the lipid head groups promote vesicle formation; attractive electrostatic
interactions suppress vesicle formation. Ionic strength and sugar concentration are also major
parameters affecting the yield and size of giant vesicles. The presence of both ions and sugars in
the hydration buffer results in increased vesicle yields. These results indicate that lipid-polymer
interactions and osmotic effects in addition to the substrate morphology and surface charge are
key factors affecting vesicle formation. Our data suggest that surface chemistry should be designed
to tune electrostatic interactions with the lipid mixture of interest to promote vesicle formation.
This vapor-deposited hydrogel fabrication technique offers tunability over physicochemical
properties of the hydrogel substrate for the production of giant vesicles with different sizes and
compositions.

66
4.2 Introduction
Giant unilamellar vesicles (GUVs) are commonly employed as practical cell models to
examine various aspects of biomembranes
1,2
such as interactions with nanoparticles
3,4
, membrane
dynamics
5
, incorporation of membrane proteins and their activities
6-8
, lipid bilayer permeability
9
,
and transport of ions through the membrane
10
. They have also been explored as potential drug
delivery vehicles.
11
Emerging techniques for GUV fabrication are opening new research vistas by
facilitating the construction of systems that better mimic biology.
Unlike the traditional electroformation
12
and gentle hydration
13
techniques for giant vesicle
formation, rehydration of lipids using hydrophilic hydrogels has proven efficient to form vesicles
over a short timescale under physiological conditions at high yields.
14
Even though
electroformation using longer hydration times and high field frequencies allows vesicle formation
at high buffer ionic strength, the yield is low and undesired lipid peroxidation and hydrolysis
reactions occur.
15,16
Microfluidic techniques, on the other hand, produce monodisperse vesicles of
various compositions at high yields; yet they often require specialized devices and leave oil
impurities in the lipid membranes.
17-19

Hydrogel-assisted rehydration easily forms vesicles made of charged lipid mixtures under
physiological conditions, overcoming one of the main disadvantages of electroformation. Recent
studies on the dynamics of GUV formation on agarose hydrogels have given insight into the
formation mechanism.
20
Small vesicles form upon rehydration and coalesce on the surface of
agarose to form larger vesicles. Higher buffer ionic strength results in Debye screening of vesicle
surface charges, leading to increased rate of coalescence and larger average vesicle size regardless
of lipid type. In addition, higher agarose density increases the rate of hydrogel rehydration and
67
GUV coalescence due to water retention in the hydrogel which allows for faster preorientation and
self-assembly of the lipid lamellae.

Following the introduction of agarose as the first effective hydrogel substrate for the
rehydration method, submillimeter glass beads
21,22
, cellulose paper
23,24
, and synthetic hydrogel
materials such as crosslinked polyacrylamide
14
, poly(vinyl alcohol)
25
, and crosslinked
dextran(ethylene glycol)
26
have been introduced. Altered vesicle mechanical properties due to
polymer incorporation in the bilayer system arising from contamination of vesicles by agarose and
poly(vinyl alcohol) chains have been reported.
27,28
Cellulose and synthetic crosslinked hydrogels,
however, are insoluble and therefore do not result in vesicle contamination from the polymer
substrate. Lopez Mora and coworkers recently introduced chemically crosslinked dextran (poly
ethylene glycol) (DexPEG) to control vesicle size and production by controlling crosslinking
density.
26
They further showed that chemical functionality and crosslinking conditions affecting
surface roughness of the hydrogel significantly impacts the yield of GUV production. Moreover,
weaker interactions between the polymer and lipid were shown to favor the formation of vesicles.
29

Clearly, physical and chemical properties of the hydrogels are major control parameters for vesicle
formation. Here we introduce a method of GUV formation based on swelling from a covalently
crossed-linked, vapor-deposited, porous, hydrophilic polymer with well controlled surface
chemistry and morphology.

In this study, porous hydrophilic polymers with distinct morphologies and thicknesses are
fabricated by a modified initiated chemical vapor deposition (iCVD) technique and are used as
hydrogel substrates for GUV production. The iCVD process is a solventless polymerization
method that is typically used to deposit dense polymer films via a free radical polymerization
mechanism.
30
In this report, our modified iCVD technique was used to form porous polymer
68
membranes by introducing the gaseous monomer into a vacuum chamber where the monomer
partial pressure and substrate temperature are kept below the triple point pressure and freezing
temperature of the monomer. Hence, the vapor undergoes a gas to solid phase change on the
surface of the substrate and forms solid microstructures.
31-34
Polymerization subsequently occurs
under a heated filament array in the presence of an initiator and a crosslinking agent is used to
render the polymer insoluble in aqueous media.
35
Following polymerization, unreacted solid
monomer is removed via sublimation resulting in membranes with dual-scale porosity. The
morphology and large-scale porosity (tens to hundreds of microns) of the membranes are
dependent on the shape of the deposited solid monomer which is controlled by substrate
temperature and monomer deposition time while smaller scale pores (hundreds of nanometers to a
few microns) are formed during sublimation of the unreacted monomer. We previously
demonstrated that monomer deposition at low temperatures (-20°C) results in 3D pillar-like
microstructures while deposition at higher temperatures (0 °C) results in 2D web-like structures.
36

Here we report on the fabrication of crosslinked poly(methacrylic acid-co-ethylene glycol
diacrylate) (xPMAA) porous membranes with distinct morphologies and thicknesses to form GUVs
from zwitterionic and charged lipid mixtures. PMAA is a pH-responsive polymer (pKa = 5.7)
37

and is negatively charged at physiological pH. We show that the yield and the size distribution of
the vesicles are controlled by the morphology and charge of the porous hydrogels. Furthermore,
we study the effect of buffer ionic strength and charge of the lipid on vesicle formation. Our results
show that the morphology and charge of the polymer both play significant roles in the yield of
vesicles, specifically for charged lipid mixtures.

69
4.3 Experimental Section
4.3.1 Materials
1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC), 1-palmitoyl-2-oleoyl-sn-
glycero-3-phospho-(1'-rac-glycerol) (POPG), 1,2-dioleoyl-3-trimethylammonium-propane
(DOTAP), and 1,2-dipalmitoyl-sn-glycero-3-phosphoethanolamine-N-(Liss Rhod PE) were
purchased from Avanti polar lipids (USA) and used without purification. Liss Rhod PE was used
as a fluorescent lipid probe. Bovine serum albumin (BSA), chloroform (CHCl3), phosphate-
buffered saline (PBS), sucrose, glucose, calcein, α-hemolysin from staphylococcus aureus,
methacrylic acid (MAA; 99%), and tert-butyl peroxide (TBPO; 98%) were purchased from Sigma
Aldrich and used as received. Ethylene glycol diacrylate (EGDA) was purchased from Polyscience
was used as a crosslinking agent. Deuterium oxide (D2O, 99.9%) was purchased from Cambridge
Isotope Laboratories, Inc. 18.2 MΩ-cm milli-Q water (EMD Millipore, USA), Sykes-Moore
chambers (Bellco, USA), and standard 25 mm no. 1 glass coverslips (ChemGlass, USA) were used
in the GUV formation experiments.
4.3.2 Fabrication of Porous Membranes
The membranes were made in a custom-designed pancake-shaped iCVD vacuum chamber
of 48 mm in height and 250 mm in diameter (GVD Corporation) containing a nichrome filament
array (Omega Engineering, 80%/20% Ni/Cr). A thermoelectric cooler (TEC) was incorporated
onto the chamber stage for temperature control. Vacuum was achieved using a rotary vane pump
(Edwards E2M40) and maintained with a throttling valve (MKS 153D) and capacitance
manometer (MKS Baratron 622A01TDE). MAA and EGDA feed jars were maintained at 30 °C
and 35 °C, respectively. Needle valves were used to meter MAA and EGDA flows, and a mass
flow controller (MKS 1479 A) was used to meter TBPO flow. Porous polymer membranes were
70
deposited onto 1.5 cm x 1 cm silicon wafer substrates (Wafer World) that were located on top of
the TEC.
To synthesize xPMAA porous membranes, TBPO was first introduced into the reactor at
0.7 sccm to achieve a total pressure of 650 mTorr. Then MAA monomer was introduced at 3.5
sccm while keeping the reactor pressure at 650 mTorr. The deposition time was varied between 10
s to 2 min and the TEC temperature was varied between and -20 °C and 0 °C during MAA
deposition. MAA flow was then halted and the reactor pressure was decreased to 200 mTorr with
TBPO flowing at 1.0 sccm. The TEC temperature was increased to 0 °C and EGDA was introduced
into the reactor at 0.2 sccm. Subsequently, polymerization started by heating the filament to 230
°C for 30 minutes for all depositions. Finally, the TBPO and EGDA flows were stopped and the
samples were pumped down to allow the unreacted excess MAA to sublimate which was
confirmed by the system returning to base pressure (Figure 1). Sublimation time across the samples
was variable and lasted from 10 to 40 minutes. In order to confirm that polymer membranes did
not leach any components into vesicles during GUV formation, we performed solvent extraction
on the crosslinked polymer membranes (xPMAA) in deuterium oxide (Cambridge Isotope
Laboratories, Inc. D, 99.9%). We repeated this for 3 different samples of various thicknesses and
the
1
H-NMR spectra were collected using a Varian Mercury 400 instrument.
Samples were imaged using scanning electron microscopy (SEM; Topcon Aquila) at 20
kV accelerating voltage. ImageJ was used to process the SEM images for estimation of the
thickness and pore sizes of the membranes. Gold was sputter-coated on polymer samples for 30 s
prior to imaging in order to avoid charging.
71
4.3.3 Giant Vesicle Formation
Glass coverslips were sonicated in MeOH at 38 °C for 30 minutes, dried at 45°C, and
plasma treated in a PDC-32G benchtop plasma cleaner (Harrick Plasma, USA) for 10 minutes.
The wafers with the porous polymer membranes were glued on coverslips using epoxy and held
in Sykes-Moore chambers. For GUV formation, lipid solution was applied by syringe on the
porous polymer substrate, dried with a gentle stream of nitrogen, and subsequently hydrated with
buffer for 1 hour. 2 mg/ml of the lipid solutions containing 0.4 mole % Liss Rhod PE (POPC, 20
POPG: 80 POPC (mol%) and 20 DOTAP: 80 POPC (mol%)) were prepared in CHCl3. For each
vesicle formation experiment, 10 μl of the lipid solution was added dropwise in 2 μl drops on the
polymer membranes and quickly dried with a gentle flow of nitrogen gas to evaporate the solvent
and form a uniform lipid film on the substrate. Following the lipid film formation process on the
substrate, 600 μl hydration buffer was added to each sample and the system was allowed to hydrate
for 1 hour. To determine the yield and size distribution of the vesicles, GUVs formed in sucrose-
containing PBS buffers (in 200 mM, 400 mM, and 440 mM sucrose concentrations) were
transferred into an isosmotic PBS buffer containing glucose. Osmolarities of the buffers were
measured using an Osmomat 3000 basic freezing point osmometer. GUVs were harvested and
transferred to 700 μl glucose solution and were allowed to settle for 40 minutes. From the bottom
of the harvested GUV solution, 400 μl was gently withdrawn and transferred to an observation
chamber for viewing. Observation chambers were pre-treated by incubation with 1 mg/ml BSA
solution in milliQ water to passivate their surfaces and avoid vesicle rupture. After 40 minutes of
further settling down in the observation chambers, samples were imaged using fluorescence
microscopy and vesicle count and size were determined using ImageJ.
72
4.3.4 Lamellarity Measurement
Unilamellarity of the vesicles were tested using α-hemolysin as the pore forming protein
and calcein as the fluorescent dye. Vesicles were prepared using -20 ℃_36 μm membrane in 185
mM PBS buffer containing 200 mM sucrose as described previously and were harvested in185
mM PBS buffer containing 200 mM glucose and 0.1 mM Calcein. Final alpha-hemolysin
concentration was 14 μg/ml and images were taken 30 minutes after adding the protein.
4.3.5 Microscopy
Fluorescent micrographs were acquired on an Axio Observer Z1 (Zeiss, Germany) inverted
microscope using an EC Plan-Neofluar 40× objective (numerical aperture of 0.75) with 1.6×
optovar magnification and equipped with a Hamamatsu CMOS camera (Hamamatsu, Japan). Each
image taken contains 2048×2048 pixels (pixel size of 0.102 µm). Illumination was provided by a
Colibri 2 LED Illumination System with a 120 V LED illuminator (Zeiss, Germany). Liss Rhod
PE was illuminated using a green filter (555/25 nm bandpass, double band emission filter 575/98
nm). For the lamellarity experiments, imaging was performed using a Nikon TI-E inverted
microscope with a spinning-disk CSUX confocal head (Yokogawa, Japan), a S Plan-Fluor 40×
objective (numerical aperture of 0.6) and a 16-bit Cascade II 512 electron-multiplied charge-
coupled device camera (Photometrics, US). 50 mW solid-state lasers (Coherent, US) at 491 nm
(emission filter centered at 525 nm) and 561 nm (emission filter centered at 595 nm) were used as
the illumination sources for calcein and rhodamine, respectively.
4.3.6 Image Processing and Data Analysis

Image processing was conducted in ImageJ by utilizing a batch macro to process and
collect vesicle count and size data from samples. Particle size measurements were performed by
making the images binary using the triangle thresholding enabling the built-in particle analyzer
73
function to operate. Processed vesicle size data was then transferred to JMP for statistical analysis.
Particulate structures with diameters less than 3 micron were not easily distinguishable as vesicles
and hence were eliminated from data analysis. These structures were not frequently observed.

74
4.4 Results and Discussion

Figure 4-1. Schematic (not to scale) of the vapor phase fabrication of porous membranes in an
iCVD chamber proceeded by hydrogel-supported vesicle formation experiments. Steps a-c depict
the process of (a) monomer deposition at variable substrate temperatures, followed by (b)
polymerization in the presence of a crosslinking agent under the heated filament array and
ultimately, (c) sublimation of the unreacted monomer in the reactor. Steps d-f represent the
hydrogel-supported formation of vesicles: (d) applying the lipid film on the polymer membrane
and subsequently (e) hydrating it with sucrose buffer and (f) harvesting the vesicles in glucose
buffer for observation with fluorescence microscopy.
75

Understanding the role that polymer structure and chemistry plays in GUV swelling from
hydrogel substrates is an important step in optimizing this increasingly important technique. Here,
we use a vapor-phase polymerization technique to create polymer substrates with controlled
structure and chemistry. By varying these parameters, we are able to study their impacts on GUV
formation. To form controlled polymer hydrogel substrates, we used a modified iCVD technique.
Porous polymer membranes of xPMAA were formed in a two-step process. First, MAA monomer
vapor was deposited as solid microstructures on a cooled silicon substrate by keeping the substrate
temperature and monomer partial pressure below the triple point temperature and pressure of the
monomer, respectively. This solid MAA was then copolymerized with a crosslinking agent
(EGDA) from the vapor phase in the presence of initiator free radicals that were formed under the
heated filament array (Figure 4-1). Excess unreacted monomer was subsequently sublimated after
polymerization by decreasing the reactor pressure.
Decoupling the monomer deposition and the polymerization steps allows for changes in
process parameters (e.g., temperature and pressure) that allowed for crosslinking and parametric
studies of the relationships between hydrogel properties and GUVs formed from these hydrogels.
Additionally, solvent extraction and
1
H-NMR confirm that the membranes were not soluble in
aqueous media. No signal pertaining to the polymer was detected after solvent extraction on
several membranes (Figure 4-2).
76

Figure 4-2. (a)
1
H-NMR spectra of crosslinked poly(methacrylic acid-co-ethylene glycol
diacrylate) after solvent extraction in deuterium oxide (b)
1
H-NMR spectra of standard
poly(methacrylic acid) (Polysciences, Inc.) in deuterium oxide (13 mg in 1ml D2O) as the control.
These were collected using a Varian Mercury 400 instrument. No polymer signal was detected for
the crosslinked samples.
77

Figure 4-3. (a) SEM micrographs of xPMAA porous membranes deposited at temperatures -20
℃, -10 °C, and 0 ℃ with three different thicknesses. Note that deposition time is varied with
78
temperature to maintain a fixed thickness. Insets show the angled-view of the corresponding
membranes. The scale bars for the top-down and angled micrographs are 150 μm and 20 μm,
respectively. (b) Deposition time versus thickness of the porous membranes measured at the three
temperatures: circle, diamond and square correspond to deposition temperatures -20 ℃, -10 °C,
and 0 ℃ and red, blue and black refer to average thicknesses of 36 ± 11μm, 82 ± 8 μm, and 124 ±
19 μm, respectively. Error bars for each of the depositions are calculated from thickness measured
at three different locations in each sample. The average thicknesses are then calculated by taking
the average of the measured mean thicknesses for the three temperatures.

The morphology and thickness of the xPMAA hydrogels are the physical properties that
were varied by changing the substrate temperature and the monomer deposition time. The
micrographs in Figure 2a correspond to membranes with average thicknesses of 36 ± 11μm, 82 ±
8 μm, and 124 ± 19 μm fabricated by deposition at substrate temperatures of -20, -10, and 0 °C.
Membranes of all thicknesses were prepared by varying the deposition time at each of the
deposition temperatures (Figure 4-3b). SEM images indicate that lower temperatures result in both
closer spacing of nucleation sites and shorter deposition periods for a given thickness. Higher
temperatures require longer deposition times to achieve the same thickness. Moreover, the
microstructures increase in width from several microns to tens of microns when the total
membrane thickness is increased at -20 °C. Increasing the deposition temperature to -10 °C and
further to 0 °C forms structures with considerably fewer void spaces. These data are consistent
with previously demonstrated monomer depositions at temperatures of -20 °C and 0 °C resulting
in 3D pillar-like and 2D web-like microstructures, respectively.

79

80

81

Figure 4-4. (a) Yield of vesicles formed from POPC lipid-coated porous membranes (membrane
labels correspond to the designations in Figure 4-3a) upon hydration with 185 mM PBS (137 mM
NaCl, 10 mM PO4
-3
, and 2.7 mM KCl) buffer containing 200mM sucrose at pH 7.4. (b) Yield of
vesicles formed from charged lipid mixture (20%POPG:80% POPC or 20%DOTAP:80% POPC).
The first four pairs of bars describe hydration with 185 mM PBS buffer containing 200mM sucrose
at pH 7.4. The final bar shows the results for 20%DOTAP:80% POPC hydrated with 740 mM PBS
containing 200 mM Sucrose at pH=7.4. (c) Schematic for the proposed charge screening
mechanism in oppositely charged polymer and lipid mixtures. Increasing hydration buffer ionic
strength decreases electrostatic interactions between the polymer and the lipid and promotes
vesicle growth. The Debye length in 185 mM PBS buffer is 0.7 nm. Increasing the ionic strength
by 4 fold should decrease the Debye length by half.

82
Effects of the physicochemical properties of xPMAA hydrogels on giant vesicle formation
were studied in terms of vesicle yield and size distribution. Vesicle formation experiments were
performed using the hydrogel assisted method using xPMAA porous membranes in Figure 4-3a.
Figure 4-4 shows the count for zwitterionic and net-charged lipid GUVs formed upon hydration
of the porous membranes. Vesicle count declines at higher deposition temperatures and thicker
membranes regardless of the lipid type; recall that both of these parameters are correlated with
longer deposition times. The increase in POPC vesicle count at lower deposition temperatures is
likely due to increased structures in low temperature depositions which provide more nucleation
sites for the lipid film to grow into vesicles (Figure 4-4a). Increasing thickness leads to a decrease
in vesicle yield due to structural densification and a corresponding decrease in nucleation sites.
Vesicle yield for membranes with higher average thicknesses of 82 ± 8 μm and 124 ± 19 μm at
deposition temperatures of -20 and -10 °C are on the same order of magnitude while membranes
formed at 0 °C result in a very poor vesicle yield. The surface roughness of crosslinked DexPEG
hydrogels was previously shown to alter the yield of vesicles in a similar manner.
29

Net-negatively charged and net-positively charged lipid compositions (20% POPG or 20%
DOTAP with the remainder POPC) show similar trends in vesicle yield. For both lipids, decreased
yield is associated with higher deposition temperatures and thicker membranes as shown in Figure
4-4b. However, DOTAP yields were overall much lower. This indicates that electrostatic
interactions between the negatively charged polymer and the charged lipid head groups are crucial
to vesicle formation under physiological conditions (185 mM PBS with 200mM Sucrose at
pH=7.4). Electrostatic repulsion between the anionic POPG lipid mixture and the negatively
charged polymer favors formation of vesicles while electrostatic attraction between the cationic
DOTAP lipid mixture and the polymer has an adverse impact on vesicle yield. For the case of
83
positively charged DOTAP lipid mixture, increasing ionic strength of the hydration buffer 4-fold
from 185 mM to 740 mM results in a drastic increase in vesicle yield, as shown in Figure 4-4b.
Shielding the negative charges of the polymer by higher ionic strength weakens lipid-polymer
interactions and allows the charged lipid film to swell and grow into vesicles (Figure 4-4c).

84

Figure 4-5. (a) Spinning disk confocal micrograph of vesicles at 491 nm excitation before adding
α-hemolysin (b) Confocal micrograph of vesicles at 491 nm excitation after adding α-hemolysin.
(c) Confocal micrograph of vesicles at 561 nm excitation after adding α-hemolysin. (d)
Fluorescence intensity across the lines indicated in micrographs a and b. These images show that
these vesicles allowed calcein transport into the interior of the vesicles after incubation with α-
hemolysin and therefore are unilamellar. This experiment was repeated on another set of vesicles
formed from the same membrane. Here, (e), (f), (g) and (h) have analogous descriptions to panels
(a), (b), (c), and (d), respectively. The images and the line scans show that lumen intensity of the
vesicles did not increase even after 1 hour of incubation (decrease in overall fluorescence intensity
is due to photobleaching). This shows that both unilamellar and multilamellar vesicles are formed
in this method. Scales bar are 10 μm.
85
In addition, the lamellarity of several POPC vesicles was tested using α-hemolysin as a
membrane pore forming protein and calcein as a fluorescent dye that does not passively cross the
lipid bilayer.
38,39
Lumen intensity of some vesicles increased after incubation with α-hemolysin
(due to pore formation and calcein transport into the vesicles). However, lumen intensity of other
vesicles did not increase in the presence of α-hemolysin (Figure 4-5). This indicates that both
unilamellar and multilamellar vesicles are formed in this method. We also observed a mixture of
unilamellar and multilamellar vesicles, as determined by relative fluorescent intensity, under
various conditions (Figure 4-6).

Figure 4-6. Fluorescent micrograph of vesicles formed from 20% DOTAP lipid-coated 36μm
thick membrane fabricated at deposition temperature -20℃ and hydrated with 740 mM PBS
containing 200 mM sucrose. Multilamellar vesicles indicated by the higher fluorescence intensity.

86

87

88

Figure 4-7. (a) Box plots for natural log-scale size distribution of the vesicles formed from POPC-
coated porous membranes upon hydration with 185 mM PBS buffer containing 200mM sucrose at
pH 7.4. (b) and (c) Box plots for log-scale size distribution of the vesicles formed from 20%POPG
and 20% DOTAP coated porous membranes upon hydration with 185 mM PBS buffer containing
200mM sucrose at pH 7.4 (except if otherwise mentioned). Note that the x axes refer to the polymer
89
iCVD deposition temperature and nominal thickness of the membranes. The central box spans the
interquartile range with the confidence diamond containing the mean (horizontal line passing in
the middle) and the upper and lower 95% of the mean. The line segment inside each box shows
the median for each set. Whiskers (lines connected to the box) extend from each end of the box to
the outermost data point that falls within the range computed as follows: 1st quartile -
1.5*(interquartile range) and 3rd quartile + 1.5*(interquartile range). All data points outside the
range are considered outliers. (d) and (e) Fluorescent micrographs of the vesicles formed from
pure POPC and 20% POPG lipid-coated 36 μm porous membranes (deposition temperatures of -
20, -10, and 0 ℃) upon hydration with 185 mM PBS containing 200 mM Sucrose at pH=7.4. (f)
Fluorescent micrographs of the vesicles formed from 20% DOTAP lipid-coated 36 μm thick
porous membrane at deposition temperature -20 ℃ upon hydration with 185 mM and 740 mM
PBS containing 200 mM Sucrose at pH=7.4. False color was added and the contrast in all the
images was enhanced for better visualization. All scale bars are 10 μm.

Table 4-1. Mass swelling ratios of the porous membranes measured after hydration with 185 mM
PBS buffer containing 200 mM Sucrose at pH=7.4. Error is calculated from swelling ratio obtained
from two separate polymer depositions. The equilibrium mass swelling ratios of the polymers were
calculated by taking the ratio of the mass of the polymer in the swollen state over that in the dry
state. The mass of each silicon wafer was measured prior to depositions. Polymer masses were
measured after the depositions. Polymer samples were subsequently hydrated with 185 mM PBS
buffer containing 200 mM sucrose at pH 7.4. The masses of the swollen polymer samples were
measured after equilibrium swelling was achieved.

90
Porous Membrane Type Mass Swelling Ratio Standard Error
-20℃_36μm 23.19 1.97
-10℃_36μm 24.69 5.11
0℃_36μm 23.49 4.13
-20℃_82μm 21.12 3.93
-10℃_82μm 26.97 8.74
0℃_82μm 43.74 2.98
-20℃_124μm 67.01 8.19
-10℃_124μm 58.45 14.76
0℃_124μm 72.30 12.71

Figure 4-7a shows the log-scale size distribution of vesicles formed upon hydration of
POPC-coated porous membranes. For a given thickness, increasing deposition temperature leads
to smaller vesicles. However, there is not a strong dependence of vesicle size on membrane
thickness except in the case of the 124 µm membranes, which produce significantly smaller
vesicles (Figure 4-7a, Table 4-2a). The size and yield results do not correlate with the measured
equilibrium mass swelling data (Table 4-1). Higher swelling in the case of thicker and denser
membranes does not result in larger vesicles. We surmise that polymer substrates resulting in high
vesicle yields (such as the 36 μm thick membrane deposited at -20 °C) also result in larger vesicles;
this is likely due to coalescence when many vesicles are present in a given area. Similar to vesicle
bulge merging in electroformation
40
, vesicle coalescence via merging through the surface of
polymer has been shown for both agarose and dextran PEG hydrogels.
20,29

Table 4-2. Non-parametric size comparisons for sample pairs of the same thickness and deposition
temperature using Wilcoxon rank-sum test for (a) pure POPC (b) 20% POPG (c) 20% DOTAP
vesicles formed using different membranes upon hydration with 185 mM PBS buffer containing
200 mM sucrose except the case for 20% DOTAP vesicles. This method was used to determine
whether two independent data sets come from the same distribution. Unlike the t-test, this test does
91
not require the assumption of normal distribution. The null hypothesis is that the data for each pair
comes from the same distribution. p values smaller than the calculated Z reject the hypothesis.
Stars represent significance in an ascending order: p<1*, p<0.1**, p<0.01***, and p<0.0001****.
A significance comparison for multiple trials at a fixed set of experimental conditions result in p-
values of 0.001 or greater; therefore, we only consider p-values less than 0.0001 in Table 4-2 to
be significant.
Table 4-2(a)

92
Table 4-2(b)

Table 4-2(c)

In the case of negatively charged POPG, there is no trend relating yield and size. This is
consistent with charge-charge repulsion inhibiting coalescence of these vesicles (Figure 4-7b,
Table 4-2b). Given the small number of DOTAP vesicles formed at low ionic strength, no
statistically significant correlation between growth conditions and size can be identified. However,
higher buffer ionic strength increases the average DOTAP vesicle size significantly and produces
multilamellar vesicles indicated by higher fluorescence intensities (Figure 4-7c, Table 4-2c, Figure
4-6). This is likely due to screening of charges on the positively charged lipid lamella and the
negatively charged polymer surfaces facilitating vesicle growth and coalescence (Figure 4-4c).
93

94

Figure 4-8. (a) Yield and (b) log- scale size distribution of the vesicles formed from POPC lipid-
coated 36 μm thick membrane deposited at -20 ℃ at varying hydration buffer ionic strength and
sucrose concentration to match osmolarity. Stars represent the hydration buffers with the same
osmolarity. All buffers are at pH 7.4. See figure 4-7 for explanation on box plots.

95
Table 4-3. Non-parametric size comparisons for sample pairs using Wilcoxon rank-sum test for
POPC vesicles formed using 36 μm thick membrane deposited at -20 ℃ at varying buffer ionic
strength and sucrose concentration. See Table 1 for explanation on the method. Stars represent
significance in an ascending order: p<1*, p<0.1**, p<0.01***, and p<0.0001****.

We further studied the effect of hydration buffer ionic strength on POPC GUV formation.
The goal of these studies was to decouple ionic effects from osmotic effects. 36 μm thick
membranes deposited at -20 ℃ were hydrated with PBS buffer containing 200 mM sucrose at
varying ionic strengths of PBS (1X PBS with a total ionic strength of 185 mM contains 137 mM
NaCl, 10 mM PO4
-3
and 2.7 mM KCl). We then repeated these experiments with the concentration
of sucrose varied such that buffer osmolarity was constant across varying ionic strengths. At
constant concentrations of sucrose, GUV count is significantly greater in the presence of salt.
However, increasing ionic strength above 18.5 mM does not significantly increase yield (Figure
4-8a). Average vesicle size increases in the presence of PBS (Figure 4-8b, Table 4-3) in a trend
that mirrors that observed for vesicle count: there is a significant increase upon the addition of salt
but adding salt above 18.5 mM results in no additional increases. In the isosmotic case, more
96
vesicles are formed in the higher sugar concentrations, and a combination of both sugar and salt in
the buffer results in the highest observed yield.

Figure 4-9. Fluorescent micrographs of vesicles formed from POPC lipid-coated 36μm thick
membrane fabricated at deposition temperature -20℃ (a) upon hydration with 18.5 mM PBS
containing 400 mM sucrose and (b) upon hydration with 0 PBS containing 440 mM sucrose at
pH=7.4. All scale bars are 10 μm.

97
Together with the high vesicle yields observed in the presence of sucrose, high sucrose
concentrations also result in the formation of tubules and GUVs that are accumulated inside each
other (Figure 4-9). A large difference in osmotic pressure and enhanced repulsion between lipid
lamellae through gentle hydration of sugar-containing dried lipid films have previously shown to
promote GUV formation.
41
Moreover, Tubule formation has been reported in systems with high
sucrose concentrations and charged lipid mixtures due to excess membrane area.
42-44
High
concentrations of sugar have shown to cause instabilities by affecting interactions and lateral
expansion of the membranes.
45
These results suggest that high osmolarity or, possibly, direct
interactions between the lipids and sugars, is more important in controlling vesicle characteristics
than ionic strength.
Giant vesicle yields in this method are comparable to the vesicle yields produced using the
agarose rehydration method. Other phenomena such as ionic, osmotic, surface interactions and
polymer morphological effects are also consistent with previous reports.

The high surface area
provided by the porous hydrogel network similar to the large areas in the submillimeter sized glass
beads and the fibrous cellulose paper lead to larger number of vesicles produced in this method.
20-
24,26

4.5 Conclusion
In comparison to previously introduced hydrogel supports for giant vesicle preparation,
this vapor phase fabrication method offers control over different process parameters such as
temperature, pressure, and time that can be independently adjusted to modify the characteristics of
the resulting polymer for vesicle production. These characteristics include crosslinking density,
thickness, morphology, and surface chemistry of the polymer. In this report we show that largest
98
vesicle yields are achieved by highly porous thin hydrogel films, lipid films that have weak
interactions with the polymer substrate, and the presence of ions and sugars in the hydration buffer.
Electrostatic interactions between the lipid and the polymer also contribute to vesicle yield. Our
results indicate that minimized electrostatic attractions between the polymer and the lipid head
groups favor vesicle formation. Electrostatic interactions between the lipid and the underlying
hydrogel and its impact on vesicle size and yield are further borne out by the effects of hydration
buffer ionic strength. Both higher ionic strength and sugar concentration lead to increased vesicle
yields. In order to produce vesicles of certain size and composition, surface chemistry can be
modified to minimize attractive interactions with the lipid mixture of interest. Manipulation of
surface chemistry can be achieved by the application of a polymer top coat on the porous scaffolds
or copolymerization with various functionalities. As with the previous hydrogel assisted methods,
porous xPMAA forms polydisperse vesicles. Further studies regarding the effect of surface
chemistry on vesicle growth could give insights in designing hydrogel conditions for producing
low polydispersity vesicles at higher yields. These findings set important guidelines toward the
design of novel hydrogels in practical applications that involve engineering surface interactions.

4.6 Acknowledgements

This work was supported by a Viterbi School of Engineering Ph.D. Fellowship and NSF CAREER
Award CMMI-1252651.

99
4.7 References

[1] Walde, P.; Cosentino, K.; Engel, H.; Stano, P. Giant Vesicles: Preparation and
Applications. ChemBioChem 2010, 11 (7), 848-865.
[2] Fenz, S. F.; Sengupta, K. Giant Vesicles as Cell Models. Integr. Biol. 2012, 4 (9), 982–995.
[3] Li, S.; Malmstadt, N. Deformation and Poration of Lipid Bilayer Membranes by Cationic
Nanoparticles. Soft Matter 2013, 9 (20), 4969-4976.
[4] Wang, L.; Malmstadt, N. Interactions Between Charged Nanoparticles and Giant Vesicles
Fabricated from Inverted-Headgroup Lipids. J. Phys. D: Appl. Phys. 2017, 50 (41), 415402.
[5] Sankhagowit, S.; Wu, S.H.; Biswas, R.; Riche, CT.; Povinelli, M.L.; Malmstadt, N. The
Dynamics of Giant Unilamellar Vesicle Oxidation Probed by Morphological Transitions. Biochim.
Biophys. Acta, Biomembr. 2014, 1838 (10), 2615-2624.
[6] Hansen, J. S. Thompson, J. R. Helix-Nielsen, C. Malmstadt, N. Lipid Directed Intrinsic Membrane
Protein Segregation. J. Am. Chem. Soc. 2013, 135 (46), 17294-17297.
[7] Gutierrez, M.G.; Malmstadt, N. Human Serotonin Receptor 5-HT1A Preferentially Segregates
to the Liquid Disordered Phase in Synthetic Lipid Bilayers. J. Am. Chem. Soc. 2014, 136 (39),
13530-13533.
[8] Gutierrez, M.G.; Mansfield, K.S.; Malmstadt, N. The Functional Activity of the Human
Serotonin 5-HT1AReceptor Is Controlled by Lipid Bilayer Composition. Biophys. J. 2016, 110
(11), 2486-2495.
[9] Runas, K.A.; Malmstadt. N. Low Levels of Lipid Oxidation Radically Increase the Passive
Permeability of Lipid Bilayers. Soft Matter 2015, 11 (3), 499-505.
100
[10] Valkenier, H.; López Mora, N.; Kros, A.; Davis, A.P. Visualization and Quantification of
Transmembrane Ion Transport into Giant Unilamellar Vesicles. Angew. Chem., Int. Ed. 2015, 54
(7), 2137-2141.
[11] Tsai, F. C.; Stuhrmann, B.; Koenderink, G. H. Encapsulation of Active Cytoskeletal Protein
Networks in Cell-Sized Liposomes. Langmuir 2011, 27 (16), 10061–10071.
[12] Angelova, M.I.; Dimitrow, D.S. Liposome Electro formation. Faraday Discuss. Chem. Soc.
1986, 81, 303_311.
[13] Reeves, J.P.; Dowben, R. M. Formation and Properties of Thin-Walled Phospholipid Vesicles.
J. Cell. Physiol. 1969, 73 (1), 49-60.
[14] Horger, K.S.; Estes, D. J.; Capone, R.; Mayer, M. Films of Agarose Enable Rapid Formation
of Giant Liposomes in Solutions of Physiologic Ionic Strength. J. Am. Chem. Soc. 2009, 131 (5),
1810-1819.
[15] Pott, T.; Bouvrais, H.; Meleard, P. Giant Unilamellar Vesicle Formation Under Physiologically
Relevant Conditions. Chem. Phys. Lipids 2008, 154 (2), 115–119.
[16] Montes, L. R.; Alonso, A.; Goni, F. M.; Bagatolli, L. A. Giant Unilamellar Vesicles
Electroformed from Native Membranes and Organic Lipid Mixtures under Physiological
Conditions. Biophys. J. 2007, 93 (10), 3548–3554.
[17] van Swaay, D.; and deMello, A. Microfluidic Methods for Forming Liposomes. Lab Chip
2013, 13 (5), 752–767.
[18] Funakoshi, K.; Suzuki, H.; and Takeuchi, S. Formation of Giant Lipid Vesiclelike
Compartments from a Planar Lipid Membrane by a Pulsed Jet Flow. J. Am. Chem. Soc. 2007, 129
(42), 12608–12609.
101
[19] Kirchner, S. R.; Ohlinger, A.; Pfeiffer, T.; Urban, A. S.; Stefani, F. D.; Deak, A.; Lutich, A.A.
Feldman, J. Membrane Composition of Jetted Lipid Vesicles: a Raman Spectroscopy Study. J.
Biophotonics 2012, 5 (1), 40–46.
[20] Peruzzi, J.; Gutierrez, M.G.; Mansfield, K.; Malmstadt, N. Dynamics of Hydrogel-Assisted
Giant Unilamellar Vesicle Formation from Unsaturated Lipid Systems. Langmuir 2016, 32 (48),
12702-12709.
[21] Nourian, Z.; Roelofsen, W; Danelon, C. Triggered Gene Expression in Fed‐Vesicle
Microreactors with a Multifunctional Membrane. Angew. Chem., Int. Ed. 2012, 51 (13), 3114-
3118.
[22] Tanasescu, R.; Mettal, U.; Colom, A.; Roux, A.; Zumbuehl, A. Facile and Rapid Formation
of Giant Vesicles from Glass Beads. Polymers 2018, 10 (1), 54.
[23] Kresse, KM.; Xu, M.; Pazzi, J.; García-Ojeda, M.; Subramaniam, AB. Novel Application of
Cellulose Paper as a Platform for the Macromolecular Self-Assembly of Biomimetic Giant
Liposomes. ACS Appl. Mater. Interfaces 2016, 8 (47), 32102-32107.
[24] Li, A.; Pazzi, J.; Xu, M.; Subramaniam, A.B. Cellulose Abetted Assembly and Temporally
Decoupled Loading of Cargo into Vesicles Synthesized from Functionally Diverse Lamellar Phase
Forming Amphiphiles. Biomacromolecules 2018, 19 (3), 849-859.
[25] Weinberger, A.; Tsai, F. C.; Koenderink, G. H.; Schmidt, T. F.; Itri, R.; Meier, W.; Schmatko,
T.; Schroder, A.; Marques, C. Gel-Assisted Formation of Giant Unilamellar Vesicles. Biophys. J.
2013, 105 (1), 154–164.
[26] Mora, N. L.; Hansen, J. S.; Gao, Y.; Ronald, A. A.; Kieltyka, R.; Malmstadt, N.; Kros, A.
Preparation of Size Tunable Giant Vesicles from Cross-linked Dextran(ethylene glycol) Hydrogels.
Chem. Commun. 2014, 50 (16), 1953–1955.
102
[27] Lira, R. B.; Dimova, R.; Riske, K. A. Giant Unilamellar Vesicles Formed by Hybrid Films of
Agarose and Lipids Display Altered Mechanical Properties. Biophys. J. 2014, 107 (7), 1609–1619.
[28] Dao, T.P.T.; Fauquignon, M.; Fernandes, F.; Ibarboure, E.; Vax, A.; Prieto, M.; Le Meins,
J.F. Membrane Properties of Giant Polymer and Lipid Vesicles Obtained by Electroformation and
PVA Gel-Assisted Hydration Methods. Colloids Surf., A 2017. 533. 347-353.
[29] Mora, N.L.; Lopez, N.; Gao, Y.; Gutierrez, M. G.; Peruzzi, J.; Bakker, I.; Peters, R.J.; Siewert.
B.; Sylvestre, B.; Kieltyka, R.E.; van Hest, J.C. M.; Malmstadt, N.; Kros, A. Evaluation of
Dextran(ethylene glycol) Hydrogel Films for Giant Unilamellar Lipid Vesicle Production and Their
Application for the Encapsulation of Polymersomes. Soft Matter 2017, 13 (33), 5580-5588.
[30] Tenhaeff, W.E.; Gleason, K.K. Initiated and Oxidative Chemical Vapor Deposition of
Polymeric Thin Films: iCVD and oCVD. Adv. Funct. Mater. 2008, 18 (7), 979-992.
[31] Seidel, S.; Kwong, P.; Gupta, M. Simultaneous Polymerization and Solid Monomer
Deposition for the Fabrication of Polymer Membranes with Dual-Scale Porosity. Macromolecules
2013, 46 (8), 2976-2983.
[32] Seidel, S.; Cheong, C.C.; Kwong, P.; Gupta, M. All-Dry Fabrication of Poly(methacrylic
acid)-Based Membranes with Controlled Dissolution Behavior. Macromol. Mater. Eng. 2015, 300
(11), 1079-1084.
[33] Dianat, G.; Seidel, S.; De Luna, M.M.; Gupta, M. Vapor Phase Fabrication of Hydrophilic
and Hydrophobic Asymmetric Polymer Membranes. Macromol. Mater. Eng. 2016, 310 (9), 1037-
1043.
[34] Seidel, S.; Dianat, G.; Gupta, M. Formation of Porous Polymer Coatings on Complex
Substrates Using Vapor Phase Precursors. Macromol. Mater. Eng. 2016, 301 (4), 371-376.
103
[35] Dianat, G.; Gupta., M. Sequential Deposition of Patterned Porous Polymers Using
Poly(dimethylsiloxane) Masks. Polymer 2017, 126, 463-469.
[36] Seidel, S.; Gupta. M. Systematic Study of the Growth and Morphology of Vapor Deposited
Porous Polymer Membranes. J. Vac. Sci. Technol., A 2014, 32 (4), 041514.
[37] Porasso, R.D.; Benegas, J.C.; van den Hoop, M. A. G. T. Chemical and Electrostatic
Association of Various Metal Ions by Poly(acrylic acid) and Poly(methacrylic acid) As Studied by
Potentiometry. J. Phys. Chem. B 1999, 103 (13), 2361-2365.
[38] Song, L.; Hobaugh, M.R.; Shustak, C.; Cheley, S.; Bayley, H.; Gouaux, J.E. Structure of
Staphylococcal α-hemolysin, a Heptameric Transmembrane Pore. Science 1996, 274 (5294),
1859-1865.
[39] Robinson, T.; Kuhn, P.; Eyer, K.; Dittrich, P.S. Microfluidic Trapping of Giant Unilamellar
Vesicles to Study Transport Through a Membrane Pore. Biomicrofluidics 2013, 7 (4), 44105.
[40] Micheletto, Y.M.S.; Marques, C.M.; Silveira, N.P.D.; and Schroder, A.P. Electroformation
of Giant Unilamellar Vesicles: Investigating Vesicle Fusion Versus Bulge
Merging. Langmuir 2016, 32 (32), 8123-8130.
[41] Tsumoto, K.; Matsuo, H.; Tomita, M.; and Yoshimura, T. Efficient Formation of Giant
Liposomes Through the Gentle Hydration of Phosphatidylcholine Films Doped with
Sugar. Colloids Surf., B 2009, 68 (1), 1, 98-105.
[42] Li, Y.; Lipowsky, R.; Dimova, R. Membrane Nanotubes Induced by Aqueous Phase
Separation and Stabilized by Spontaneous Curvature. Proc. Natl. Acad. Sci. U. S. A. 2011, 108
(12), 4731-4736.
104
[43] Roux, A.; Cuvelier, D.; Nassoy, P.; Prost, J.; Bassereau, P.; Goud, B. Role of Curvature and
Phase Transition in Lipid Sorting and Fission of Membrane Tubules. EMBO J. 2005, 24 (8), 1537-
1545.
[44] Spector, M. S.; Selinger, J.V.; Singh, A.; Rodriguez, J.M.; Price, R.R.; Schnur J.M.
Controlling the Morphology of Chiral Lipid Tubules. Langmuir 1998, 14 (13), 3493-3500.
[45] Andersen, H. D.; Wang, C.; Arleth, L.; Peters, G. H.; Westh, P. Reconciliation of Opposing
Views on Membrane-Sugar Interactions. Proc. Natl. Acad. Sci. U. S. A. 2011, 108 (5), 1874-1878.

105
Chapter 5: Concluding Remarks and Future Work
5.1 Conclusion
In this dissertation, we presented a sustainable solvent free polymer fabrication technique
that is used to fabricate dense and porous polymer coatings and membranes. We have demonstrated
the robustness and functionality of perfluorinated dense coatings for industrial stainless steel
substrates that are fabricated using two iCVD crosslinking strategies. The application of the
perfluorinated coatings as antifouling surfaces for polymerization reactor components (e.g.,
mixers) is successfully demonstrated. The graded pC8PFA-pEGDA and pC6PFA-pEGDA
coatings were more effective at preventing fouling by PVP compared to the p(C8PFA-co-EGDA)
and p(C6PFA-co-EGDA) coatings at the simulated mixing conditions. In addition, we have
discovered that the more ecofriendly short chain perfluorinated coating is a potential alternative to
their long chain analogues as an antifouling surface, despite its lower hydrophobicity and higher
chain mobility.
Furthermore, we covered our modified iCVD process for the vapor phase fabrication of
porous polymers. We introduced modifications in the vapor phase porous polymer fabrication
technique in order to increase the conversion and throughput of the process while decreasing
material waste. We showed that during the polymerization of solid monomer the sublimating
monomer that is otherwise exhausted by the vacuum pump can be captured using downstream cold
traps. We demonstrated the impacts of surface temperature on reaction kinetics and the
morphology of the fabricated membranes. Lower temperatures lead to better surface coverage and
increased surface roughness. The higher mass obtained at higher capture and polymerization
temperatures is due to faster reaction kinetics and higher monomer sublimation rates regardless of
lack of nucleation and poor surface coverage. For practical applications, depending on the size of
106
the vacuum chamber several additional cold traps can be added to the system for the complete
consumption of the unreacted monomer through capture and polymerization.
Finally, we demonstrated an application for the porous polymers as crosslinked hydrogels
to serve as polymer substrates for the formation of giant lipid vesicles using the hydrogel assisted
rehydration method. Understanding the role that the physicochemical characteristics of the
polymer plays in giant vesicle swelling from hydrogel substrates is an important step in optimizing
this increasingly important technique. Here we showed that largest vesicle yields are achieved by
highly porous thin hydrogel films, lipid films that have weak interactions with the polymer
substrate, and the presence of ions and sugars in the hydration buffer. Electrostatic interactions
between the lipid and the polymer also contribute to vesicle yield indicating that minimized
electrostatic attractions between the polymer and the lipid head groups favor vesicle formation.
Electrostatic interactions between the lipid and the underlying hydrogel and its impact on vesicle
size and yield are further borne out by the effects of hydration buffer ionic strength. Both higher
ionic strength and sugar concentration lead to increased vesicle yields. These findings set
important guidelines toward the design of novel hydrogels in practical applications that involve
engineering surface interactions.

107
5.2 Future work
In chapter 2 of this dissertation, the short chain C6 graded fluoropolymer coating showed
analogous antifouling properties compared to the long chain C8 graded fluoropolymer and
therefore additional investigation of the chain mobility of the graded coatings should allow for
further development of short chain fluoropolymers for industrial applications. Liu et al. showed
that the contact angle hysteresis for polymers composed of C6PFA can be lowered by decreasing
the thickness of the pC6PFA layer deposited on top of a crosslinked base layer of poly(divinyl
benzene) (pDVB). Tuning the thickness of the fluorinated top layer should aid in understanding of
the effect of chain mobility and surface reorganization on the antifouling characteristics.
Moreover, shorter chain fluoropolymers (with perfluorinated pendant groups of 6 carbons or less)
should also be investigated. Multiple foulant testing cycles will also asses the performance of the
coatings after several exposures.
In chapter 3 of this dissertation, fabrication of porous polymers using our modified iCVD
process will focus on additional reactor modifications to further reduce the monomer waste and
increase the polymer conversion. In the typical iCVD process, only a small amount of the monomer
flown is deposited as polymer film and the rest is exhausted by the vacuum pump. Similarly, in
our modified iCVD process for the fabrication of porous polymers, the solid monomer that is
captured on the TEC surface is approximately 15-25 % of the monomer feed that is flown into the
reactor. As mentioned in the introduction, only 2-12% of the deposited monomer is then converted
to porous polymer due to the slow kinetics of solid phase polymerization and loss of monomer
through sublimation. Therefore, we would like to modify the reactor by adding an extension
through the monomer feed line in order to directionalize the flow to increase the monomer capture
rate which will then result in higher efficiency and decreased monomer waste. Multiple
108
downstream cold traps can be incorporated to further use the sublimating monomer for
polymerization. The directionalized flow can also affect the growth direction and morphology of
the porous membranes.
In chapter 4 of this dissertation, direct observation of the vesicles forming on the hydrogel
by microscopy techniques was not achievable in all cases since the vapor deposited xPMAA
porous polymers are opaque. In order to better understand and tune the vesicle formation process,
different ways should be explored to render the polymer transparent. The macroporous structure
of the membranes causes Mie-scattering while structures with pores sizes smaller than the visible
light wavelength (<400nm) are transparent. Fabrication process parameters should be tuned to
decrease thickness, microstructural width, and pore size of the membranes in order to reduce
scattering and opacity. Furthermore, polymer color and opaqueness have been associated with
crystallinity of polymers which can be tuned for membranes fabricated at varying processing
parameters (e.g., deposition and polymerization temperatures). In order to expand the generality
of this hydrogel swelling technique for the synthesis of artificial cells, integration of membrane
proteins into giant vesicle should be studied using this bottom-up xPMAA porous polymer
swelling method. Protein membranes have been shown to integrate into the bilayer of the vesicles
if incorporated in the agarose mixture during hydrogel preparation.
2,3
Proteins are mixed with the
hydrogel mixture after dissolution of agarose in water at a temperature above the melting point of
agarose. The mixture is then applied on a glass substrate and cooled to form a protein-hydrogel
composite. The lipid film is then applied on the hydrogel and the system is hydrated to swell and
form protein incorporated vesicles. Mixing of the protein and the polymer plays an important role
in the incorporation of the protein in the lipid membrane. Since no solvent is used in the vapor
deposition process, diffusion and mixing of the protein with the polymer network should be
109
maximized before the lipid is applied. Hence, hydrogel porosity and crosslinking density of the
xPMAA porous polymer should be tuned to maximize protein diffusion.

110
5.3 References
[1] Liu, A.; Goktekin, E.; Gleason, K.K. Cross-linking and Ultrathin Grafted Gradation of
Fluorinated Polymers Synthesized via Initiated Chemical Vapor Deposition to Prevent Surface
Reconstruction. Langmuir 2014, 30, 14189-14194.
[2] Hansen, J. S. Thompson, J. R. Helix-Nielsen, C. Malmstadt, N. Lipid Directed Intrinsic Membrane
Protein Segregation. J. Am. Chem. Soc. 2013, 135(46), 17294-17297.
[3] Gutierrez, M.G.; Malmstadt, N. Human Serotonin Receptor 5-HT1A Preferentially Segregates
to the Liquid Disordered Phase in Synthetic Lipid Bilayers. J. Am. Chem. Soc. 2014, 136(39),
13530-13533.

Abstract (if available)

Abstract Polymers are widely used as surface coatings and membranes that are capable of imparting functionality and surface chemistry desired for applications such as hydrophobic coatings and water purification membranes. Solvent-free vapor phase polymerization methods are great for the fabrication of polymer coatings and membranes since they eliminate surface tension and solvent compatibility limitations associated with solution phase polymerization methods. Initiated chemical vapor deposition (iCVD) is a vapor phase polymer fabrication technique that can be used to deposit dense and porous polymers. In this dissertation, we demonstrate iCVD processes to fabricate functional dense and porous polymers with practical applications. ❧ Chapter 1 provides a background on the conventional iCVD process for the deposition of dense polymers. The modification of the conventional iCVD process for the fabrication of porous polymers with dual scale porosity is covered next. Chapter 2 is focused on robust vapor deposited fluoropolymers with an industrial application as antifouling coatings for stainless steel polymerization reactor components. In this section, we study and compare the chemical and mechanical stability and the performance of long chain vs short chain perfluorinated iCVD polymers as antifouling coatings. We show that the more environmentally friendly short chain perfluorinated coating is a potential alternative to its long chain analog as an antifouling surface. ❧ Chapter 3 is focused on process improvements of the modified iCVD process for the fabrication of porous polymers which have applications in sensing, tissue engineering, separations and so on. We demonstrate that downstream monomer capture and polymerization during vapor phase fabrication of porous polymers is possible by incorporation of multiple cold traps inside the reactor. The uniformity and morphology of the membranes across surfaces and the effects of surface temperature on membrane morphology and polymerization kinetics are studied. Our ability to capture and polymerize monomer across multiple surfaces provides a more sustainable route for reducing monomer waste and improving conversion and throughput of the process. Chapter 4 is focused on a biophysical application of vapor deposited hydrophilic porous polymers as hydrogel substrates for the formation of giant lipid vesicles using the hydrogel assisted rehydration method. Swelling of the lipid covered porous polymer membrane leads to the formation of giant lipid vesicles. Polymer membranes with controlled morphologies, thicknesses, and surface chemistry are fabricated to study the effects of the physicochemical properties of the polymer on vesicle characteristics for method optimization. Lipid-polymer interactions and osmotic effects in addition to the substrate morphology and surface charge are shown to be key factors affecting vesicle formation. Chapter 5 concludes this dissertation by summarizing the key contributions and the future work.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Understanding structural changes of polymer films during vapor phase deposition

PDF

Scale-up of vapor-phase deposition of polymers: towards large-scale processing

PDF

Fabrication of functional porous membranes via polymerization of solid monomer by a vapor-phase initiator

PDF

Simultaneous monomer deposition and polymerization at low substrate temperatures for the formation of porous polymer membranes

PDF

Inorganic/organic hybrid materials and grafted coatings via vapor phase deposition processes

PDF

Novel processing of liquid substrates via initiated chemical vapor depostion

PDF

Controlling polymer film patterning, morphologies, and chemistry using vapor deposition

PDF

Fabrication of polymer films on liquid substrates via initiated chemical vapor deposition: controlling morphology and composition

PDF

The patterning of polymer thin films on porous substrates via initiated chemical vapor deposition

PDF

Vapor phase deposition of polymers in the presence of low vapor pressure liquids

PDF

Selective deposition of polymer coatings onto structured surfaces

PDF

Formation of polymer gels, films, and particles via initiated chemical vapor deposition onto liquid substrates

PDF

Fabrication of silicon-based membranes via vapor-phase deposition and pyrolysis of organosilicon polymers

PDF

Chemical vapor deposition of graphene: synthesis, characterization, and applications

PDF

Fabrication and study of organic and inorganic optoelectronics using a vapor phase deposition (VPD)

PDF

Improving the sustainability of conjugated polymer synthesis via direct arylation polymerization

PDF

Imaging molecular transport across and nanomaterial interaction with lipid membranes

PDF

High-throughput nanoparticle fabrication and nano-biomembrane interactions

PDF

Synthesis of high-quality nanoparticles using microfluidic platforms

PDF

Engineering solutions for biomaterials: self-assembly and surface-modification of polymers for clinical applications

Asset Metadata

Creator Movsesian, Nareh (author)

Core Title Vapor phase deposition of dense and porous polymer coatings and membranes for increased sustainability and practical applications

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Chemical Engineering

Publication Date 01/30/2021

Defense Date 06/11/2020

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag chemical vapor deposition,coatings,membranes,OAI-PMH Harvest,porous polymers,thin films

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Gupta, Malancha (committee chair), Malmstadt, Noah (committee chair), Mousavi, Maral (committee member)

Creator Email Nareh.movsesian@usc.edu,Narehmovsesian@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-354753

Unique identifier UC11666102

Identifier etd-MovsesianN-8855.pdf (filename),usctheses-c89-354753 (legacy record id)

Legacy Identifier etd-MovsesianN-8855.pdf

Dmrecord 354753

Document Type Dissertation

Rights Movsesian, Nareh

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