University of Southern California Dissertations and Theses

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

(USC Thesis Other)

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

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Copyright 2022 Nicholas Alexander Welchert
CONTROLLING POLYMER FILM PATTERNING, MORPHOLOGIES, AND CHEMISTRY
USING VAPOR DEPOSITION

by

Nicholas Alexander Welchert

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

May 2023
ii

Table of Contents
List of Tables ................................................................................................................................. iii
List of Figures ................................................................................................................................ iv
Abstract .......................................................................................................................................... vi
Chapter 1: Introduction ....................................................................................................................1
1.1 initiated Chemical Vapor Deposition ................................................................................1
1.2 Oblique Angle Deposition .................................................................................................2
1.3 High Throughput Processing and Computational Modeling for Chemical Vapor
Deposition ....................................................................................................................................2
1.4 Silicon-based Inorganic Membranes Production Utilizing iCVD.....................................3
1.5 Deposition onto Liquid Substrates ....................................................................................4
1.6 Low Energy Plasma Enhanced Chemical Vapor Deposition (PECVD) and In Situ
Pyrolysis .......................................................................................................................................5
Chapter 2: Oblique Angle Initiated Chemical Vapor Deposition for Patterning Film Growth .......7
2.1 Introduction .......................................................................................................................7
2.2 Materials and Experimental Procedure .............................................................................8
2.3 Results and Discussion ....................................................................................................12
2.4 Conclusions .....................................................................................................................24
2.5 Acknowledgements .........................................................................................................25
Chapter 3: Branched Nozzle Oblique Angle Flow for Initiated Chemical Vapor Deposition ......26
3.1 Introduction .....................................................................................................................26
3.2 Materials and Experimental Procedure ...........................................................................28
3.3 Results and Discussion ....................................................................................................33
3.4 Conclusions .....................................................................................................................43
3.5 Acknowledgements .........................................................................................................44
Chapter 4: Vapor Deposition of Silicon-Containing Microstructured Polymer Films onto
Silicone Oil Substrates ...................................................................................................................45
4.1 Introduction .....................................................................................................................45
4.2 Materials and Experimental Procedure ...........................................................................46
4.3 Results and Discussion ....................................................................................................48
4.4 Conclusions .....................................................................................................................58
4.5 Acknowledgements .........................................................................................................59
4.6 Future Work: In Situ Pyrolysis of Vapor Deposited Polymer Films for the Formation
of inorganic Silicon-Based Inorganic Materials.........................................................................60
References ......................................................................................................................................62
iii

List of Tables
Table 1. Deposition conditions for samples. ....................................................................................9
Table 2. Deposition conditions for all samples. .............................................................................30
Table 3. Deposition conditions for the various samples. ...............................................................50
Table 4. XPS atomic composition of the top and bottom surfaces of the p(V4D4-co-
EGDA) films deposited on silicone oil. .................................................................................53
Table 5. XPS atomic composition of the top and bottom surfaces of the p(V4D4-co-
EGDA) films deposited on silicone oil for 45, 90, and 120 minutes at a V4D4 to
EGDA flow rate ratio of 1:1. .................................................................................................57

iv

List of Figures
Figure 1. Image of the PnBA thickness profile (nm) on a silicon wafer for a deposition (a)
without a monomer inlet extension and (b) using a monomer inlet extension at θ =
50° at the same deposition conditions. Thickness measurements along (c) the x-axis
about the geometric center and (d) the y-axis in line with the thickest measured point
on the x-axis for the deposition at θ = 50°. ........................................................................... 12
Figure 2. (a) Thickness measurements along the x-axis at varying deposition times and
(b) corresponding images of the PnBA films. (c) Thickness measurements along the
x-axis as a function of substrate temperature and (d) corresponding images of the
PnBA films. ........................................................................................................................... 14
Figure 3. (a) Thickness measurements along the x-axis for PnBA depositions at varying
monomer flow rates and (b) corresponding images. (c) Thickness measurements
along the x-axis for PMAA depositions at varying monomer flow rates and (d)
corresponding images (e) Thickness measurements along the x-axis for a PHEMA
deposition at a monomer flow rate of 0.9 sccm and (f) the corresponding image of
the PHEMA film. .................................................................................................................. 16
Figure 4. (a) Thickness measurements along the x-axis for PMAA depositions at varying
reactor pressures and (b) corresponding images. (c) Thickness measurements along
the x-axis for PHEMA depositions at varying reactor pressures and (d)
corresponding images. .......................................................................................................... 18
Figure 5. (a) Reactor model with a representative velocity profile. The velocity profile
shown is a horizontal slice 0.01 mm above the stage. Velocity and pressure profiles
along the x-axis of the reactor stage for θ = 50° and corresponding images of the
polymer films indicating location of maximum thicknesses for (b) HEMA at a flow
rate of 0.9 sccm, (c) nBA at a flow rate of 4.5 sccm, and (d) MAA at a flow rate of
12.6 sccm. ............................................................................................................................. 20
Figure 6. Images placed in order of the monomer flow rates showing the evolution of the
indentation point and the movement of the maximum thickness point for PHEMA,
PnBA, and PMAA depositions using θ of 50°. ..................................................................... 22
Figure 7. (a) PMAA thickness measurements along the x-axis at θ of 30°, 50°, and 70°
and (b) corresponding images of PMAA. (c) PHEMA thickness measurements along
the x-axis at θ of 30°, 50°, and 70° and (b) corresponding images of PHEMA. .................. 23
Figure 8. (a) Angled top-down image of the iCVD reactor with a branched monomer inlet
extension with two nozzles separated by an angle of 92º. (b) Images of the nozzles
for the monomer inlet extension N1, N2, and N3. (c) Schematic of the nozzle of the
monomer inlet extension. (d) Example of the left nozzle x-axis, right nozzle x-axis,
and y-axis associated with each sample. ............................................................................... 33
Figure 9. Top down images of the experimental results for a) N1-1, b) N2-1, c) N3-1, d)
N1-5, e) N2-5, f) N3-5, g) N1-10, h) N2-10, and i) N3-10. .................................................. 33
Figure 10. Thickness measurements along the a) left nozzle x-axis, b) right nozzle x-axis,
and c) y-axis for samples N1-1, N2-1, and N3-1. ................................................................. 35
Figure 11. Thickness measurements along the a) left nozzle x-axis, b) right nozzle x-axis,
and c) y-axis for samples N1-5, N2-5, and N3-5. ................................................................. 36
Figure 12. Thickness measurements along the a) left nozzle x-axis, b) right nozzle x-axis,
and c) y-axis for samples N1-10, N2-10, and N3-10. ........................................................... 37
v

Figure 13. Thickness measurements along the a) left nozzle x-axis, b) right nozzle x-axis,
and c) y-axis for samples N3-5-0.5I, N3-5, and N3-5-2I. ..................................................... 38
Figure 14. (a) Right nozzle x-axis and (b) y-axis velocity and pressure data for N1-5. (c)
Right nozzle x-axis and (d) y-axis velocity and pressure data for N2-5. (e) Right
nozzle x-axis and (f) y-axis velocity and pressure data for N3-5. The x-axis data is
shifted to the geometric center (i.e., x = 0, GC) and the y-axis data is shifted to be in
line with the monomer inlet stem (i.e., y = 0). The velocity is indicated by the blue
line and the pressure is indicated by the dashed orange line. ............................................... 39
Figure 15. (a) Right nozzle x-axis and (b) y-axis velocity and pressure data for N2-1. (c)
Right nozzle x-axis and (d) y-axis velocity and pressure data for N2-5. (e) Right
nozzle x-axis and (f) y-axis velocity and pressure data for N2-10. The x-axis data is
shifted to the geometric center (i.e., x = 0, GC) and the y-axis data is shifted to be in
line with the monomer inlet stem (i.e., y = 0). The velocity is indicated by the blue
line and the pressure is indicated by the dashed orange line. ............................................... 41
Figure 16. FTIR spectra of pV4D4 and pEGDA films and of copolymer films deposited
on reference wafers placed in the reactor during deposition of samples A, B, and C;
dashed lines indicate the position of the C=O and Si-O-Si peak .......................................... 51
Figure 17. Schematic depicting the top and bottom surfaces of the p(V4D4-co-EGDA)
film deposited onto silicone oil. ............................................................................................ 51
Figure 18. Top-down SEM images of the bottom surface of the pEGDA film and
copolymer films (samples A, B, and C) deposited onto 1000 cst silicone oil. ..................... 54
Figure 19. (a) SEM image of the top surface of the p(V4D4-co-EGDA) film (Sample A)
showing an example of the densely-packed and loosely-packed regions. (b) Higher
magnification image of the interface between the densely-packed and loosely-
packed regions. ..................................................................................................................... 55
Figure 20. SEM images of the densely-packed regions of the top surface of the pEGDA
film and copolymer films (samples A, B, and C) deposited onto 1000 cst silicone oil. ....... 56
Figure 21. SEM images of the loosely-packed regions of the top surface of the pEGDA
film and copolymer films (samples A, B, and C) deposited onto 1000 cst silicone oil. ....... 56
Figure 22. (a) SEM images of the cross-sections of the pEGDA film and copolymer films
(samples A, B, and C) deposited onto 1000 cst silicone oil. (b) SEM images of the
cross-section of p(V4D4-co-EGDA) film deposited on 1000 cst silicone oil for 45,
90, and 120 minutes at a V4D4 to EGDA flow rate ratio of 1:1. ......................................... 58

vi

Abstract
This work is split into two sections in which unique processes are used in order to control
the patterning of thin polymer films and the ability to deposit silicon containing polymers onto
high viscosity liquids for use in hydrogen production via methane steam reforming. Chapter 1 will
go into the fundamentals of initiated chemical vapor deposition (iCVD) and the motivation and
background for the works in Chapters 2, 3, and 4.
Chapter 2 discusses how iCVD can be used to deposit thin polymer films on a variety of
substrates. In this work, the monomer precursor was introduced at an oblique angle to the substrate
using an inlet extension and the pattern of the resulting polymer film was studied as a function of
the deposition time, substrate temperature, monomer flow rate, reactor pressure, and vapor flow
angle. The polymerization of n-butyl acrylate (nBA), methacrylic acid (MAA), and 2-
hydroxyethyl methacrylate (HEMA) was examined to determine the generality of the trends across
several monomers. It was found that the monomer flow rate significantly affected the pattern of
the deposited polymer by shifting the location of the thickest point in the films. Increasing the
deposition time, decreasing the substrate temperature, and increasing the reactor pressure increased
the polymer deposition rate consistent with conventional iCVD, however the pattern of the
deposited polymer did not vary with these parameters. Computational analysis was used to
determine how the inlet extension affects the pressure and velocity profiles within the reactor. The
data demonstrated that the introduction of the monomer precursor at an oblique angle can be used
to pattern polymer films during iCVD.
Chapter 3 is a follow-up work to chapter 2 and discusses work in which monomer precursor
flow was introduced at an oblique angle to the substrate at two locations during the iCVD process

vii

using a branched nozzle inlet extension. The polymerization of MAA was systematically studied
as a function of the nozzle length and the monomer flow rate. Our experimental data showed the
evolution of two distinct symmetrical thickness profiles as the flow rate and nozzle length
increased. The maximum thickness moved downstream along the axes of both nozzles as the flow
rate and nozzle length increased. Computational models were used to study the effects of the nozzle
length and the monomer flow rate on the velocity and pressure profiles within the reactor.
Increasing the monomer flow rate and the nozzle length resulted in increases in both the velocity
and pressure profile ranges and the movement of the location of the maximum velocity and
pressure. This velocity and pressure data provided insight for explaining the trends found in the
experimental results. The data demonstrated the ability to use a branched nozzle inlet extension
to control the location of polymer deposition during the iCVD process.
Finally, in chapter 4 we discuss an experiment where a silicon-containing crosslinked
polymer, p(2,4,6,8-tetravinyl-2,4,6,8-tetramethyl cyclotetrasiloxane-co-ethylene glycol
diacrylate) (p(V4D4-co-EGDA)), was deposited onto high viscosity silicone oil using initiated
chemical vapor deposition (iCVD). The ratio of the feed flow rate of V4D4 to EGDA was
systematically studied and the chemical composition and morphology of the top and bottom
surfaces of the films were analyzed. The films were microstructured and the porosity and thickness
of the films increased with increasing V4D4 content. The top of the film was comprised of densely-
packed and loosely-packed microstructured regions. X-ray photoelectron spectroscopy on the top
and bottom surfaces of the films showed a heterogeneous chemical composition along the
thickness of the film, with higher silicon content on the top surface compared to the bottom surface.
To our knowledge, this is the first study of iCVD deposition of a silicon-containing polymer films
onto silicone oil. The results of this study can be used for the synthesis of polymer precursors films

viii

for the fabrication, via pyrolysis, of silicon-based inorganic membranes for use in hydrogen
production.
In chapter 4 we discuss the future directions for this work, including the low energy plasma
enhanced chemical vapor deposition of siloxane polymers, namely pV4D4, followed by the in situ
pyrolysis to form silicon-based inorganic materials for use in hydrogen production, transportation,
and storage.

1

Chapter 1: Introduction
1.1 initiated Chemical Vapor Deposition
Initiated chemical vapor deposition (iCVD) is a solvent-free polymerization technique that
can be used to deposit thin, uniform films on a variety of substrates.
1,2,3
Since the precursors are in
the vapor phase, this is not a line of sight process making it an ideal process for conformal coatings
on complex substrates. The iCVD process has been used to deposit polymer coatings onto solid
substrates such as silicon wafers,
4
membranes,
5
wires,
6
carbon nanotubes,
7
and fibers.
8
In the
iCVD process, monomer and initiator vapors are introduced into a vacuum chamber where the
initiator is thermally cleaved by a heated filament array to produce free radicals. The monomer
molecules and initiator radicals adsorb to the surface of the cooled substrate, and polymerization
occurs via a free-radical mechanism, producing a thin film. The monomer can be varied to produce
polymer films with a variety of properties including hydrophobicity,
9
hydrophilicity,
10
and stimuli-
responsiveness.
11
Higher deposition rates and molecular weights can be achieved by increasing the
monomer surface concentration which is governed by the ratio of the monomer partial pressure
(Pm) to the saturation pressure (Psat).
12,13
This ratio can be controlled by varying processing
parameters such as the monomer flow rate, the substrate temperature, and the reactor pressure.
14,15

The following equation gives the ratio of Pm to Psat as a function of the flow rate of the monomer
(Fm), the reactor pressure (Ptotal), and the Clausius-Clapeyron equation as a function of substrate
temperature (Tsubstrate).

𝑃 𝑚 𝑃 𝑠𝑎𝑡

=
F
m
*P
total

F
total
A𝑒 (∆H
vap
/RT
substrate
)
(1)

2

1.2 Oblique Angle Deposition
Oblique angle deposition (OAD) has been used to introduce vapor flow at an oblique angle
to the substrate during inorganic depositions, causing shadowing effects which lead to films with
columnar microstructures.
16
This method has been used to create structured porous films for
applications such as organic photovoltaics,
17,18
fuel cells,
19,20
and microfluidics.
21,22
Studies have
shown that porosity within the films increases as the vapor flow becomes perpendicular to the flat
substrate since the columns orient in the direction of the vapor flow.
23
During glancing angle
deposition (GLAD), changing the angular rotation of the substrate while introducing the precursor
vapor flow at an oblique angle can produce helical, zig zag, and s-shaped columns.
24
While OAD
and GLAD have been commonly applied to the deposition of inorganic materials such as silica,
25

titanium nitride,
26
and cerium oxide,
27
advancements have also been made to incorporate these
processes to the deposition of organic materials. For example, OAD and GLAD of Parylene C
have been used to fabricate thin, porous films composed of sculptured polymer nanowires.
28,29

1.3 High Throughput Processing and Computational Modeling for Chemical
Vapor Deposition
High throughput processing in the field of chemical vapor deposition (CVD) has been
widely studied due to the need for low cost manufacturing processes than enable mass
production.
30
Increasing throughput has been achieved by increasing sample loading
number,
31,32,33
manipulating processing parameters,
34,35,36
substrate placement,
37
and roll-to-roll
processing.
38,39,40,9
Our results also enable the understanding of multi-dimensional flow and
transport phenomena in the iCVD process via computational modeling. Computational modeling
of the CVD process has been previously studied for design, analysis, and optimization.
41
Fluid

3

dynamics modeling in particular is a primary part in modeling multiple CVD processes as it allows
for a more realistic description of the process.
42,43
Although computational models of the iCVD
process have been previously presented,
44
there is need for more research to increase the accuracy
and generality in iCVD specific modeling.

1.4 Silicon-based Inorganic Membranes Production Utilizing iCVD
Silicon-based nanoporous asymmetric inorganic membranes have been shown to have
excellent hydrogen separation properties
45,46,47
and are, thus, ideal for use in hydrogen production
through catalytic methane steam reforming or biomass and coal gasification. In comparison to
their polymer counterparts, these membranes have been shown to be stable at high temperatures
and pressures and can withstand the corrosive environments that are involved in hydrogen
production. Such inorganic membranes are often fabricated by the application of a pre-ceramic
silicon-containing polymer film on a mechanically strong porous ceramic substrate and its
subsequent pyrolysis.
48,49,50
A uniform polymer film on top of the porous substrate is ideal for the
formation of high-quality ceramic membranes.
51,52,53
The dense polymer film deposition methods
to date, typically, employ solvent-based techniques, such a slip-casting
54
, spin-coating
55,56
, and
dip-coating
57
, which use potentially environmentally harmful solvents such as toluene, benzene,
and tetrahydrofuran.
The development of vapor phase polymer deposition methods for the fabrication of pre-
ceramic silicon-containing polymer films shows promise for reducing solvent use. In particular,
the iCVD method is a solvent-free polymerization technique that is traditionally used to deposit
thin films on a variety of solid substrates.
1,2,3
In this process, monomer and initiator vapors are
introduced into a vacuum chamber where the initiator is thermally cleaved by a heated filament

4

array to produce free radicals. The monomer molecules and initiator radicals adsorb onto the
surface of a substrate where polymerization occurs via a free-radical mechanism, producing a
dense film. Tert-butyl peroxide (TBPO) is, typically, used as the initiator.
58
We recently used
iCVD to deposit a crosslinked silicon-containing polymer poly(2,4,6,8-tetravinyl-2,4,6,8-
tetramethyl cyclotetrasiloxane) (pV4D4) onto a macroporous support and subsequently pyrolyzed
the film to form a silica membrane.
59
These asymmetric membranes utilized a macroporous silicon
carbide (SiC) substrate with a mesoporous SiC top layer, prepared via the dipcoating and
subsequent pyrolysis of a pre-ceramic polymer film, in order to prevent infiltration, during
polymerization, of the pV4D4 polymer into the macroporous substrate, which would then result in
clogging of the pores and waste of the monomer and initiator precursors.
50

1.5 Deposition onto Liquid Substrates
Low vapor pressure liquids such as silicone oils and ionic liquids have recently been used
as substrates in iCVD processes.
60
It has been shown that using this method can produce a thin
film on the surface of the liquid or particles at the liquid surface or within the liquid itself.
61,62

Previous work has shown that the formation of a thin film or particles is dependent on the
interactions between the polymer deposited and the liquid substrate, namely, the spreading
coefficient (S) seen here:
𝑆 = 𝛾 𝐿𝑉
∗ (1 + cos 𝜃 ) − 2𝛾 𝑃𝑉
(2)
which is a measure of the free energy required for the polymer to spread over the surface of the
liquid, where γLV is the liquid−vapor surface tension, γPV is the polymer−vapor surface tension,
and θ is the advancing contact angle of the liquid on the polymer.
63
Work has also been done to

5

deposit p(ethylene glycol diacrylate) (p(EGDA)) onto silicone oils and ionic liquids. It has been
shown to produce microstructured thin films on silicone oils,
64
and used as a crosslinker for
copolymer systems on ionic liquids.
65
Bradley and coworkers hypothesized that these
microstructures formed on silicone oil were the result of chemical cross-linking, which helped to
form polymer networks. These microstructures grew then by simultaneous polymer diffusion and
aggregation and wetting of the growing aggregates by the liquid.

1.6 Low Energy Plasma Enhanced Chemical Vapor Deposition (PECVD)
and In Situ Pyrolysis
The iCVD method is a solvent-free polymerization technique that is used to deposit thin
films on a variety of solid substrates.
1,2,3
In the iCVD process, control of the substrate temperature
is required in order to influence precursor adsorption.
13,14
In contrast to iCVD, plasma enhanced
chemical vapor deposition (PECVD) is not a substrate temperature dependent process,
66
and has
been used for the preparation of polymers, such as the previously described pV4D4,
67,68
which can
be used for the synthesis of silica and silicon carbide.
69,70
Low-energy PECVD can be chosen as
an alternative method to iCVD when substrate temperature cannot be controlled. In situ pyrolysis
of a polymer such as pV4D4 prevents exposing the sample to atmospheric conditions, which
reduces the likelihood of external contamination. An in situ pyrolysis system will increase the
sample throughput of the vapor deposited silicon-containing polymers since both the deposition
and pyrolysis steps are done in a single batch process, therefore reducing the time required for
processing. The results of creating an in situ pyrolysis system of a vapor deposited film can be
used to further the research of silicon-based inorganic membranes for use in hydrogen production,
transportation, and storage by combining the vapor deposition and pyrolysis steps into a

6

streamlined process. Furthermore, the same system can be used for carrying out in situ other heat-
treatment processes of vapor-deposited polymer films needed for crosslinking,
71,72
increasing
crystallinity,
73,74,75
and driving self-assembly for patterning.
76,77,78

7

Chapter 2: Oblique Angle Initiated Chemical Vapor Deposition for
Patterning Film Growth
1

Initiated chemical vapor deposition (iCVD) can be used to deposit thin polymer films on a
variety of substrates. In this work, the monomer precursor was introduced at an oblique angle to
the substrate using an inlet extension and the pattern of the resulting polymer film was studied as
a function of the deposition time, substrate temperature, monomer flow rate, reactor pressure, and
vapor flow angle. The polymerization of n-butyl acrylate (nBA), methacrylic acid (MAA), and 2-
hydroxyethyl methacrylate (HEMA) was examined to determine the generality of the trends across
several monomers. It was found that the monomer flow rate significantly affected the pattern of
the deposited polymer by shifting the location of the thickest point in the films. Increasing the
deposition time, decreasing the substrate temperature, and increasing the reactor pressure increased
the polymer deposition rate consistent with conventional iCVD, however the pattern of the
deposited polymer did not vary with these parameters. Computational analysis was used to
determine how the inlet extension affects the pressure and velocity profiles within the reactor. The
data demonstrate that the introduction of the monomer precursor at an oblique angle can be used
to pattern polymer films during iCVD.
2.1 Introduction
In this study, we applied the concept of OAD to the iCVD process and studied the pattern
and thickness of the resulting polymer films. We introduced the monomer vapor flow at different
angles (30°, 50°, 70°) using a monomer inlet extension. We studied the polymerization of n-butyl
acrylate (nBA), methacrylic acid (MAA), and 2-hydroxyethyl methacrylate (HEMA) to examine
the generality of the trends across several monomers. We studied how iCVD parameters such as

1
Welchert et. al, J. Vac. Sci. Technol. A 38, 063405 (2020).

8

the monomer flow rate, the deposition time, the substrate temperature, and the reactor pressure
affect the pattern and thickness of the films. COMSOL Multiphysics was used to model the effect
of the monomer inlet extension on the steady-state velocity and pressure profiles within the reactor.
Our data showed that shape of the deposited polymer was significantly influenced by the monomer
vapor flow angle (θ) and the monomer flow rate. These results allow for polymer film thickness
profiles to be tailored which is useful for cell and tissue engineering studies
79,80
and high-
throughput polymer thin film characterization
81,82
and experiments.
83,84,85,86

2.2 Materials and Experimental Procedure

Methacrylic acid (MAA) (Aldrich, 99%), tert-butyl peroxide (TBPO) (Aldrich, 98%), 2-
hydroxyethyl methacrylate (HEMA) (Aldrich, 97%), and n-butyl acrylate (nBA) (Aldrich, 99%)
were used as received without further purification. Polymer films were deposited on silicon wafers
(Wafer World, 100 mm) using a custom-designed pancake-shaped iCVD vacuum chamber that is
250 mm in diameter and 48 mm in height (GVD Corporation). A nichrome filament array (Omega
Engineering, 80%/20% Ni/Cr) inside the reactor was heated during the deposition process to
thermally decompose the TBPO initiator. The reactor pressure was maintained using a throttle
valve (MKS 153D) and measured using a capacitance manometer (MKS Baratron 622A01TDE).
The temperature of the stage was maintained using a recirculating chiller (Thermo Scientific
NESLAB RTE 7). The nBA, MAA, and HEMA flow rates were metered using a needle valve.
TBPO flow rate was metered using a mass flow controller (MKS 1479A) and was maintained at
0.9 sccm for all experiments. Table 1 lists the reactor conditions for each experiment.

9

Table 1. Deposition conditions for samples.
Sample Set Monomer θ (°) Flow Rate
(sccm)
Substrate
Temperature
(°C)
Reactor
Pressure
(mTorr)
Deposition
Time (min)
A nBA 50 4.5 10 450 15, 30, 45
B nBA 50 4.5 10, 15, 20 450 30
C nBA 50 4.6, 7.8, 10.1 20 450 30
D MAA 50 9.1, 10.7, 12.6 20 600 25
E HEMA 50 0.9 25 70, 90, 120 35
F MAA 50 10 20 300, 400, 500 25
G MAA 30, 50,
70
10 20 600 25
H HEMA 30, 50,
70
0.9 25 120 35

To direct the monomer vapor flow, extensions for the monomer inlet (3/8 inch inner
diameter) were fabricated using 1/4 inch inner diameter stainless steel tubing (Figure 1a).
Polytetrafluoroethylene tape (1/4 inch, TaegSeal) was used to seal 1 inch of the monomer inlet
extension into the existing monomer line. The monomer inlet extensions had bend angles of 150°,
130°, and 110° which correspond to monomer vapor flow angles (θ) of 30°, 50°, and 70°,
respectively. A 4 cm × 5 cm piece of silicon wafer was placed with the left edge 8 cm from the
left side of the reactor for experiments A–F and 7.25-7.5 cm from the left for experiments G–H as
shown in Table 1. As shown in Figure 1b, the height (H) from the monomer inlet extension to the
substrate is 1.2 cm which corresponds to a line-of-sight distance (D) of 2.3, 1.3, and 0.8 cm from
the monomer inlet extension to the substrate surface for θ of 30°, 50°, and 70°, respectively. The
location on the substrate corresponding to the line-of-sight impingement point from the monomer
inlet extension is referred to as the geometric center. The silicon wafer was given a 2D coordinate
system with the x-axis in line with the monomer inlet extension and the y-axis perpendicular to the
x-axis. The thickness of the polymer was measured after deposition using a Stokes Waferskan
Ellipsometer (Gaertner L3W15S405.830). The measurements were taken at 0.5 cm intervals from

10

the left side of the wafer. These measurements were then normalized to the geometric center for
graphical representation. Measurements at y = -1.0, -0.5, 0.5, and 1.0 cm were taken in line with
thickest measured point on the x-axis to confirm symmetry. Photos of the polymer films were taken
under consistent lighting and placement with an iPhone 6s Plus.
A 3D model of the iCVD chamber with the monomer inlet extension was built using
COMSOL Multiphysics® (Version 5.5). The laminar flow physics model was applied since the
Reynolds number for all experiments was calculated to be under 2100. The velocity profile and
pressure profile were derived using a steady-state incompressible Navier-Stokes equation. Gas
viscosities were estimated using the temperature dependent Sutherland formula.
87
Gas densities
were estimated by using the ideal gas law. A 1D line graph of the velocity and pressure was taken
along the x-axis of the reactor. Since there is a no-slip boundary condition at the surfaces, we
reported the velocity profile 0.01 mm above the surface while the pressure profile exerted on the
walls of the reactor was reported.

11

Figure 2. (a) Angled top-down photo of the iCVD reactor with a monomer inlet extension at a
monomer vapor flow angle of 50°. (b) Schematic of the iCVD reactor using the monomer inlet
extension showing the location of the geometric center on the substrate.

12

2.3 Results and Discussion

Figure 1. Image of the PnBA thickness profile (nm) on a silicon wafer for a deposition (a) without
a monomer inlet extension and (b) using a monomer inlet extension at θ = 50° at the same
deposition conditions. Thickness measurements along (c) the x-axis about the geometric center
and (d) the y-axis in line with the thickest measured point on the x-axis for the deposition at θ =
50°.

13

We deposited PnBA without using the monomer inlet extension for 30 minutes at a reactor
pressure of 450 mTorr, nBA flow rate of 4.5 sccm, and substrate temperature of 10 °C. The
thickness measurements across the polymer film demonstrate that the film is uniform with no
distinct maximum or minimum (Figure 2a). To study the effect of using the monomer inlet
extension, PnBA was deposited using θ = 50° at the same deposition conditions. The thickness
profile of the deposited PnBA film (Figure 2b–d) shows that the thickest point is to the right of the
geometric center with decreasing thickness in all directions and there is symmetry about the x-
axis. This data indicates that the localized introduction of the monomer vapor flow at an oblique
angle modifies the iCVD process to produce a patterned thin film. In the conventional iCVD
process, the deposition rate is governed by the adsorption of the monomer to the substrate. The
adsorption of the monomer follows the Brunauer–Emmett–Teller (BET) isotherm,
13,88
which
assumes a uniform convective flow. The ratio of the rate of convective mass transfer to diffusive
mass transfer can be estimated using the dimensionless Peclet (Pe) number where Pe = UL/D in
which U is the average velocity, L is the characteristic length, and D is the mass diffusivity.
89
For
both the case with and without the monomer inlet extension, L is the height from the filament array
to the substrate and D is determined by the kinetic theory of gases. By using the monomer inlet
extension, the inlet diameter decreases by 40% and therefore the average velocity is 2.8 times
higher resulting in an increase of the Pe number by a factor of 2.8. This increase in Pe causes the
monomer surface concentration to be governed by convective forces (i.e. the non-uniform flow
profile) resulting in a non-uniform thickness profile. The thickness at all points on the wafer in
Figure 2a is greater with an extension than without, indicating that increasing localized velocity
leads to an increased monomer concentration at the surface. The shape of the thickness profile
observed in the deposition of PnBA with θ = 50°

is consistent with experimental and theoretical

14

studies of oblique jet impingement flow profiles onto a flat substrate,
90,91,92
validating that the use
of the monomer inlet extension at an oblique angle leads to a non-uniform flow profile which leads
to a non-uniform thickness profile. For example, Floss et al. showed that an oblique jet flow profile
of air at θ = 45° on a flat substrate results in a point of maximum velocity to the right of the
geometric center, with decreasing velocity moving away from this point in all directions,
90

consistent with the thickness profile seen in Figure 2b.

Figure 2. (a) Thickness measurements along the x-axis at varying deposition times and (b)
corresponding images of the PnBA films. (c) Thickness measurements along the x-axis as a
function of substrate temperature and (d) corresponding images of the PnBA films.

15

To study the effect of the deposition time on the thickness profile, we deposited PnBA for
15, 30, and 45 minutes at θ = 50° while keeping all other conditions constant (Sample Set A).
Figure 3a shows that the overall shape of the thickness profile did not vary, and the thickest point
remained at the same position to the right of the geometric center for all deposition times. The film
thickness at all locations increased with longer deposition times, which is consistent with
conventional iCVD. Since the shape of the thickness profile remained constant with varying
deposition times, the deposition times were kept constant within individual sample sets as shown
in Table 1. To study the effect of substrate temperature on the thickness profile, we deposited
PnBA at substrate temperatures of 10, 15, and 20 °C for 30 minutes at θ = 50°, keeping all other
conditions constant (Sample Set B). To determine whether the use of the extension affects the
substrate temperature, a thermocouple was placed at the geometric center of the silicon wafer to
measure the temperature during deposition with and without the monomer inlet extension. It was
found that there was no significant difference in the substrate temperature with and without the
use of the extension. As shown in Figure 3c, the substrate temperature does not significantly affect
the shape of the thickness profile. The film thickness increases at all locations with decreasing
substrate temperature which is consistent with conventional iCVD.

Decreasing the substrate
temperature leads to a lower saturation pressure and therefore a higher Pm/Psat which leads to a
higher deposition rate.
13,40
This dependence of deposition rate on substrate temperature confirms
that the deposition rate of iCVD with the monomer inlet extension is governed by the monomer
surface concentration.

16

Figure 3. (a) Thickness measurements along the x-axis for PnBA depositions at varying monomer
flow rates and (b) corresponding images. (c) Thickness measurements along the x-axis for PMAA
depositions at varying monomer flow rates and (d) corresponding images (e) Thickness
measurements along the x-axis for a PHEMA deposition at a monomer flow rate of 0.9 sccm and
(f) the corresponding image of the PHEMA film.

To study the effect of the monomer flow rate on the thickness profile, PnBA was deposited
using monomer flow rates of 4.6, 7.8, and 10.1 sccm at a substrate temperature of 20 °C for 30

17

minutes using θ = 50° (Sample Set C). Increasing the monomer flow rate increases the thickness
at all points due to an increased Pm/Psat and increased localized velocity (Figure 4a). At a flow rate
of 10.1 sccm, there is visible condensation of the monomer on the wafer during the deposition due
to the increase in Pm/Psat (Figure 4b). Figure 4a shows that for the higher flow rate of 10.1 sccm,
the thickest point of the film moves farther to the right of the geometric center relative to the lower
flow rates of 4.6 and 7.8 sccm and a visible indentation is produced to the left of the geometric
center (Figure 4b). The location of the visible indentation corresponds to the stagnation point in
the flow. At this stagnation point, the flow is brought to rest, converting kinematic energy to static
pressure.
91
O’Donovan et al. showed computationally and experimentally that the stagnation point
is located upstream of the geometric center for all angles of oblique impinging flow,
92
which is
consistent with our thickness profile at 10.1 sccm. To further study the presence of the visible
indentations at higher flow rates, we studied the deposition of poly (methacrylic acid) (PMAA) at
flow rates of 9.1, 10.7, and 12.6 sccm at θ = 50° while keeping all other conditions constant
(Sample Set D). We chose MAA since it does not condense at higher flow rates. Figure 4c shows
that with increasing monomer flow rate, the thickest point of the film moves further to the right of
the geometric center and the film thickness at all points increases, trends that are consistent with
those for PnBA. These results demonstrate that for the iCVD process with a monomer inlet
extension, the monomer flow rate predominately determines the shape of the thickness profile (i.e.
the location of the thickest point and the appearance of an indentation) regardless of the monomer
used. To study the deposition at a low flow rate, PHEMA was deposited for 35 min using θ = 50°,
a substrate temperature of 25 °C, a monomer flow rate of 0.9 sccm, and a reactor pressure of 120
mTorr. In contrast to higher flow rate depositions, the thickest point of the film is located to the
left of the geometric center as shown in Figure 4f. Since the flowrate of HEMA is an order of

18

magnitude lower than that of the other monomers used, Pe is lower and therefore the monomer
surface concentration is governed by diffusion. The highest concentration of monomer is at the
outlet of the extension and the concentration decreases as it travels farther away from this point
due to dispersion, resulting in the thickest point being closer to the extension outlet.

Figure 4. (a) Thickness measurements along the x-axis for PMAA depositions at varying reactor
pressures and (b) corresponding images. (c) Thickness measurements along the x-axis for PHEMA
depositions at varying reactor pressures and (d) corresponding images.

To study the effect of reactor pressure, PMAA was deposited at reactor pressures of 300,
400, and 500 mTorr at θ = 50° and a constant flow rate of 10 sccm (Sample Set E). Figure 5a–b
shows that the thickest point of the film falls to the right of the geometric center due to a high flow

19

rate and subsequently high Pe, and the shape of the thickness profile is similar for all PMAA
depositions. As the reactor pressure increases, the film thickness increases at all points, consistent
with conventional iCVD. Higher reactor pressures lead to higher monomer partial pressures and
therefore higher deposition rates.
93
PHEMA was deposited at reactor pressures of 70, 90, and 120
mTorr at θ = 50° and a constant flow rate of 0.9 sccm (Sample Set F). Figures 5c–d show that the
thickest point of the film falls to the left of the geometric center due to a low flow rate and
subsequently low Pe, and the shape of the thickness profile does not change with increasing reactor
pressures. In addition, the film thickness increases as the reactor pressure increases. These results
are consistent with the data for varying substrate temperatures which showed that higher Pm/Psat
leads to higher deposition rates, consistent with conventional iCVD.

20

Figure 5. (a) Reactor model with a representative velocity profile. The velocity profile shown is a
horizontal slice 0.01 mm above the stage. Velocity and pressure profiles along the x-axis of the
reactor stage for θ = 50° and corresponding images of the polymer films indicating location of
maximum thicknesses for (b) HEMA at a flow rate of 0.9 sccm, (c) nBA at a flow rate of 4.5 sccm,
and (d) MAA at a flow rate of 12.6 sccm.

21

To further study the effects of using the monomer inlet extension in the iCVD process, we
used COMSOL Multiphysics to build a laminar flow model to analyze the velocity at a slice 0.01
mm above the substrate and the pressure data at the surface of the substrate for θ = 50° (Figure 6a)
Figures 6b–d show that the maximum pressure and maximum velocity shift slightly to the right as
the velocity of the monomer increases. At a HEMA flow rate of 0.9 sccm (Figure 6b), the location
of the thickest point on the deposited film corresponds to the location of the highest pressure since
the flow rate and Pe are low and therefore the monomer surface concentration is determined by
the concentration profile. For the deposition of PnBA at 4.5 sccm (Figure 6c), the location of the
thickest point of the deposited film corresponds to the location of the highest velocity. In chemical
vapor deposition processes, the gas molecules slow down as they approach the surface of the
substrate due to viscous friction and the transport of the molecules to the surface can only occur
by diffusion through a stationary fluid boundary layer. The magnitude of the film deposition
depends on the local molecular flux to the surface, which is inversely proportional to the boundary
layer thickness distribution,
94
which is inversely proportional to the square root of the localized
Reynolds number near the surface of the substrate.
95,96
Since the Reynolds number is proportional
to velocity, we can conclude that the areas of higher velocity have higher deposition rates. This
data supports that at higher flow rates and Pe, the monomer surface concentration is governed by
the flow profile. All the thickness profiles produced at flow rates of 4.5–7.8 sccm similarly follow
the same trend observed here. For the deposition of PMAA at 12.6 sccm, the thickest point of the
deposited film is located downstream from both the pressure maximum and the velocity maximum
which might be due to deviations from the laminar flow model. For example, researchers have
studied impinging jet flow at oblique angles and found that increasing velocity near the surface
can lead to increases in velocity fluctuations and turbulence downstream.
97,98
The location of the

22

visible indentation point aligns to the local velocity minimum which corresponds to the stagnation
point. At this location, the boundary layer is largest leading to the lowest deposition rate.

Figure 6. Images placed in order of the monomer flow rates showing the evolution of the
indentation point and the movement of the maximum thickness point for PHEMA, PnBA, and
PMAA depositions using θ of 50°.

Figure 7 illustrates the progression of the shape of the thickness profile with increasing
monomer flow rate for PHEMA, PnBA, and PMAA depositions at a constant monomer vapor flow
angle (θ = 50°). The movement of the maximum thickness point to the right irrespective of the
monomer used suggests that the monomer flow rate is the primary parameter that affects the
resulting thickness profile. The thickness profiles fall into three regimes: slow flow rates (regime
I), medium flow rates (regime II), and high flow rates (regime III). In regime I, the thickest point
is to the left of the geometric center and is at the location of maximum pressure. In regime II, the
thickest point is to the right of the geometric center and is at the location of the highest velocity.
In regime III, the thickest point moves farther downstream of the geometric center and an

23

indentation appears to the left of the geometric center at the location of the minimum local velocity.

Figure 7. (a) PMAA thickness measurements along the x-axis at θ of 30°, 50°, and 70° and (b)
corresponding images of PMAA. (c) PHEMA thickness measurements along the x-axis at θ of 30°,
50°, and 70° and (b) corresponding images of PHEMA.

To study the effect of the monomer vapor flow angle (θ) on the polymer thickness profile,
PMAA was deposited using vapor flow angles of 30°, 50°, and 70° at a flow rate of 10 sccm while
keeping all other conditions constant (Sample Set G). The resulting thickness profiles show that
the thickest point of the film falls to the right of the geometric center for all θ which is expected
due to the high flow rate and subsequently high Pe (Figure 8a–b). As θ decreases, the profile shape
broadens in the x-direction. This is likely due to the broadening of the velocity profile in the x-
direction near the surface of the substrate which was confirmed by COMSOL modelling. The

24

thickness of the thickest point is observed to decrease as θ decreases because the distance to travel
to the substrate is longer, thereby decreasing the localized monomer velocity near the substrate
which was confirmed by COMSOL modelling. To study the effect of varying θ on the PHEMA
thickness profile, PHEMA was deposited using vapor flow angles of 30°, 50°, and 70° at a flow
rate of 0.9 sccm (Sample Set H). The resulting thickness profiles show that the thickest point on
the film falls to the left of the geometric center for all θ (Figure 8c–d), consistent with the profile
observed for the previous PHEMA depositions at a low flow rate and low Pe. As θ decreases, the
thickness of the thickest point decreases, consistent with the data for PMAA, however the
magnitude of change is much larger for PHEMA. Our COMSOL model has shown that the
localized pressure decreases as θ decreases. This variation can lead to significant changes in the
deposition rate for PHEMA due to the low saturation pressure of HEMA at the operating
temperature (160 mTorr) because diffusive forces dominate during the PHEMA depositions at
these low flow rates.

2.4 Conclusions

The monomer precursor was introduced at an oblique angle relative to the substrate using
an inlet extension and the resulting polymer patterns were studied. The polymerizations of n-butyl
acrylate (nBA), methacrylic acid (MAA), and 2-hydroxyethyl methacrylate (HEMA) were
examined to determine the generality of the trends across several monomers. It was found that the
flow rate of the monomer is the primary parameter that affects the resulting thickness profile. At
slower flow rates and subsequently lower Pe, the thickest point in the polymer film is to the left of
the geometric center. As the flow rate and Pe increase, the thickest point moves farther to the right

25

of the geometric center. At very high flow rates, there is a visible indentation that appears to the
left of the geometric center. At high monomer flow rates, decreasing the monomer vapor flow
angles causes the profile shape to broaden in the x-direction. The deposition time, the substrate
temperature, and the reactor pressure do not alter the pattern of the deposited polymer, however
the changes in the thicknesses within the pattern are consistent with the trends in conventional
iCVD. These results provide fundamental groundwork toward future exploration of using the
modified iCVD process to create specialized polymer thin film patterns and gradients.

2.5 Acknowledgements

N. A. W. is supported by the USC Viterbi School of Engineering/Chevron Corporation University
Partnership Program Ph.D. Fellowship in Energy Resources.

26

Chapter 3: Branched Nozzle Oblique Angle Flow for Initiated
Chemical Vapor Deposition

In this study, Monomer precursor flow was introduced at an oblique angle to the substrate at
two locations during the initiated chemical vapor deposition (iCVD) process using a branched
nozzle inlet extension. The polymerization of methacrylic acid was systematically studied as a
function of the nozzle length and the monomer flow rate. Our experimental data showed the
evolution of two distinct symmetrical thickness profiles as the flow rate and nozzle length
increased. The maximum thickness moved downstream along the axes of both nozzles as the flow
rate and nozzle length increased. Computational models were used to study the effects of the nozzle
length and the monomer flow rate on the velocity and pressure profiles within the reactor.
Increasing the monomer flow rate and the nozzle length resulted in increases in both the velocity
and pressure profile ranges and the movement of the location of the maximum velocity and
pressure. This velocity and pressure data provided insight for explaining the trends found in the
experimental results. The data demonstrates the ability to use a branched nozzle inlet extension to
control the location of polymer deposition during the iCVD process.
3.1 Introduction

Initiated chemical vapor deposition (iCVD) is a technique used to produce thin, uniform
polymer films which can be applied to a wide range of planar and non-planar substrates.
1,2,3,99
In
this solvent-free polymerization process, initiator and monomer vapors are introduced into a
vacuum chamber reactor and the initiator molecules are thermally cleaved to produce initiator
radicals. The initiator radicals and monomer molecules adsorb to the surface of a cooled substrate
where polymerization by a free-radical mechanism occurs leading to the growth of a thin polymer

27

film. By varying the monomer feed, the characteristics of the polymer film produced can alter the
surface properties of a given substrate, such as changing hydrophobic surfaces into hydrophilic
surfaces,
10
or hydrophilic surfaces into hydrophobic surfaces.
9
Other industrially useful surface
properties such as stimuli-responsiveness
11
and adhesion
100
have also been shown to be achieved
through the iCVD process. In the iCVD process, the polymer deposition rate has been shown to
be controlled by processing parameters such as substrate temperature, reactor pressure, and
monomer flow rate.
14,15
Careful control of these parameters can result in higher monomer surface
concentrations which result in higher deposition rates.
12,13
In our previous study, we demonstrated
that the pattern of polymer deposition could be tuned by introducing the monomer precursor flow
at an oblique angle relative to the substrate.
101
This use of an oblique angle during the iCVD
process resulted in a gradient in polymer film thickness, which could have applications in cell and
tissue engineering
79,80
and high-throughput polymer thin film experiments
83,84,85,86
and
characterization.
81,82
In this study, we examine how the growth of the polymer films during the
iCVD process can be affected by simultaneously introducing two separate monomer flows both at
oblique angles relative to the substrate.
In this study, we created a branched nozzle monomer inlet extension which allowed for the
introduction of the monomer precursor flow to be delivered at an oblique angle relative to the
substrate at two separate locations within the iCVD reactor simultaneously. Three different nozzle
lengths and three different flow rates were studied. The results showed the evolution of a thickness
profile with two distinct symmetrical thickness maximums as the flow rate and nozzle length
increased. As the flow rate and the nozzle length increased, the two symmetrical maximum
thicknesses moved downstream. A COMSOL Multiphysics fluid dynamics model was created
which analyzed the effects of the various flow rates and nozzle lengths on the velocity and pressure

28

profiles within the reactor. There was an overall increase in velocity and pressure as flow rate and
nozzle length increased, and movement of the velocity and pressure maximum downstream as the
nozzle length increased. As flow rate increased at a constant nozzle length, maximum pressure
remained at the same location. This computational velocity and pressure data provided insight for
explaining the trends found in the experimental results.
Our results show how oblique angle deposition can be used to the control the location of
polymer deposition which can be beneficial for the development of high throughput processing.
High throughput processing in the field of chemical vapor deposition (CVD) has been widely
studied due to the need for low cost manufacturing processes than enable mass production.
30

Increasing throughput has been achieved by increasing sample loading number,
31,32,33

manipulating processing parameters,
34,35,36
substrate placement,
37
and roll-to-roll
processing.
38,39,40,9
Our results also enable the understanding of multi-dimensional flow in the
iCVD process.
41,42,43
Although computational models of the iCVD process have been previously
presented,
44
there is need for more research to increase the accuracy and generality in iCVD
specific modeling.
3.2 Materials and Experimental Procedure

Methacrylic acid (MAA) (Aldrich, 99%) and tert-butyl peroxide (TBPO) (Aldrich, 98%) were
used as received without further purification. The polymer films were deposited on silicon wafers
(Wafer World, 100 mm) using a custom-designed iCVD vacuum chamber reactor that is 4.8 cm in
height and 25 cm in diameter (GVD Corporation). A nichrome filament (Omega Engineering,
80%/20% Ni/Cr) inside the reactor was heated to 250ºC by applying an electrical current during
the process to thermally cleave the TBPO initiator. The reactor pressure was measured using a

29

capacitance manometer (MKS Baratron 622A01TDE) and maintained using a throttle valve (MKS
153D). The stage temperature was maintained using a recirculating chiller (Thermo Scientific
NESLAB RTE 7). The TBPO flow rate was metered using a mass flow controller (MKS 1479A)
for all experiments and flown through the existing reactor inlet line found on the side of the reactor
[Fig. 1(a)]. The MAA flow rate was metered manually using a needle valve and the MAA jar was
heated to 25 ºC in order to vaporize the monomer.
Three extensions for the monomer inlet (3/8 inch inner diameter) were fabricated using 1/4
inch inner diameter stainless steel tubing. As shown in Figure 8a, the stem of the monomer inlet
extension extended 8 cm from the side of the reactor. Two nozzles branched from the stem at top
down angles of 46º either above or below the extension stem, creating a top down angle of 92º
between the two nozzles [Fig. 1 (b)]. The monomer inlet extensions had symmetrical nozzle
lengths of either 1, 1.5, and 2 cm, which corresponds to monomer inlet extension identifications
as N1, N2, and N3, respectively [Fig. 1 (b)]. Each nozzle had a bend angle of 130º normal to the
angle of nozzle branching, which corresponds to a monomer vapor flow angle relative to the
substrate of 50º [Fig. 1 (c)]. The height (H) is the shortest geometrical distance from each nozzle
to the substrate, and is 1.2, 0.9, and 0.5 cm for corresponding monomer inlet extensions N1, N2,
and N3, respectively. The line-of-sight distance (D) is 1.6, 1.1, and 0.6 cm for corresponding
monomer inlet extensions N1, N2, and N3, respectively. The location on the substrate
corresponding to the line-of-sight impingement point from the nozzle is referred to as the
geometric center (GC). Since there are two nozzles, there are two corresponding GCs. Due to the
symmetry of the system, the schematic shown in Figure 8(c) can be applied to both nozzles. To fit
the stem of the monomer inlet extension into the existing monomer line,
polytetrafluoroethylene tape (1/4 inch, TaegSeal) was used as a seal. A silicon wafer was placed

30

under the monomer inlet extension with the left edge 7 cm from the left side of the reactor for all
experiments. The edges of the silicon wafer were taped to the stage using Kapton tape to secure
the substrate location and improve contact of the silicon wafer to the stage. Table 2 lists the reactor
conditions for each sample.

Table 2. Deposition conditions for all samples.
Sample ID Nozzle(s)
Length
(cm)
MAA Flow
Rate (sccm)
Substrate
Temperature
(°C)
Reactor
Pressure
(mTorr)
Deposition
Time (min)
Initiator
Flow
Rate
(sccm)
N1-1 1 1 20 550 50 1
N1-5 1 5 20 600 25 1
N1-10 1 10 25 575 30 1
N2-1 1.5 1 20 450 50 1
N2-5 1.5 5 20 500 35 1
N2-10 1.5 10 25 600 40 1
N3-1 2 1 25 650 45 1
N3-5 2 5 20 500 30 1
N3-10 2 10 25 600 30 1
N2-5-2I 1.5 5 20 500 35 2
N2-5-0.5I 1.5 5 20 500 35 0.5

The thickness of the polymer deposited onto the silicon wafer was measured using a Stokes
Waferskan Ellipsometer (Gaertner L3W15S405.830). As seen in Figure 8d, the samples were
given two different x-axes which corresponded to be in line with either the left or right nozzle. The
measurements were taken at 0.5 cm intervals from the left edge of the wafer along both the left
and right nozzles x-axes and then shifted to x = 0 at the geometric center for each nozzle for
graphical representation. Each sample was also given a y-axis [Fig. 1 (d)] which aligned with the
thickest point measured along both the left and right nozzle x-axes, which due to symmetry was at
the same location along both x-axes. Measurements were taken at 0.5 cm intervals from the top

31

edge of the silicon wafer to the bottom edge of the silicon wafer. Photos of the polymer films were
taken under consistent lighting and placement with an iPhone 10 XR.
COMSOL Multiphysics® (Version 5.5) was used to build 3D models of the three monomer
inlet extension systems. The geometry of the reactor matched the physical reactor chamber, while
the monomer inlet extension matched the geometry of N1, N2, and N3. The Reynolds number for
all flow rates and reactor pressures was calculated to be under 2100, allowing for the use of a
laminar flow model. This model utilized the Navier-Stokes equation assuming steady-state and
incompressible flow. The gas viscosity of MAA was estimated using the temperature dependent
Sutherland formula,
87
and the gas density was estimated using the ideal gas law at a temperature
of 25 ºC, which was in the range of values of stage temperatures for the experiments. We used a
thermocouple to measure the temperature at the GC using an unbranched nozzle and found that
the temperature was 21 ºC (±2 ºC) when the stage temperature was set to 20 ºC, the jar temperature
was 25 ºC, and the filament temperature was set to 250 ºC, and that slight of variations in
temperature have little effect on the gas density and viscosity, and therefore little effect on the
velocity and pressure profile. Since there are two nozzles for each extension, the total monomer
flow rate in the system was halved, converted to a velocity using the area of each nozzle outlet,
and used as an input for each nozzle outlet in the model. The velocity and pressure profiles were
then computed for each model. Two 1D line graphs of the velocity and pressure were calculated
along two different x-axes of the reactor stage which corresponded to the x-axes in which thickness
measurements were taken as seen in Figure 8d. These line graphs were shifted to x = 0 at the GC
of each nozzle for graphical representation. A 1D line graph of the velocity, which intersects with
the maximum velocity points along both the left and right nozzle x-axes and is perpendicular to
the monomer inlet extension stem, was calculated and a 1D line graph of the pressure, which

32

intersects with the maximum pressure points along both the left and right nozzle x-axes and is
perpendicular to the monomer inlet extension stem, was also calculated. These line graphs were
then shifted to an axis in line with the monomer inlet extension stem, indicated by y = 0, for
graphical representation, and both were displayed on a single graph. Since there is a no-slip
boundary condition at the surfaces, the velocity profile 0.01 mm above the surface was reported
while the pressure profile exerted on the reactor stage was reported.

33

Figure 8. (a) Angled top-down image of the iCVD reactor with a branched monomer inlet
extension with two nozzles separated by an angle of 92º. (b) Images of the nozzles for the monomer
inlet extension N1, N2, and N3. (c) Schematic of the nozzle of the monomer inlet extension. (d)
Example of the left nozzle x-axis, right nozzle x-axis, and y-axis associated with each sample.
3.3 Results and Discussion

Figure 9. Top down images of the experimental results for a) N1-1, b) N2-1, c) N3-1, d) N1-5, e)
N2-5, f) N3-5, g) N1-10, h) N2-10, and i) N3-10.

The three extensions N1, N2, and N3 were used to deposit poly(methacrylic acid) (pMAA)
onto a silicon wafer at flow rates of 1, 5, and 10 sccm, resulting in nine samples. Thickness
measurements were made along the x-axis of the left and right nozzle as seen in Figure 8d.
Measurements were also taken along the y-axis which intersects both the left and right nozzle x-

34

axes [Fig. 8d]. The top down images for all nine samples are shown in Figure 9. Parameters such
as the reactor pressure, the substrate temperature, and the deposition time for each sample is
reported in Table 2. These parameters were chosen to prevent monomer condensation while
providing a sufficient deposition rate in order to deposit a thickness profile which was easily
visible. Our previous study using an unbranched nozzle showed that the reactor pressure, the
substrate temperature, and the deposition time only have an effect on the deposition rate and do
not have an effect on the shape of the thickness profile.
101

We first studied the effect of varying nozzle lengths at a constant monomer flow rate of 1
sccm [Fig 9(a-c)]. As the nozzle length increases, there is an evolution in the thickness profile
resulting in two distinct thickness profiles, each associated with either the left or right nozzle. At
all the nozzle lengths, the maximum thickest points of both the left and right nozzle x-axis are
located upstream of the GC. As the nozzle length increases from N1 to N3, the thickest point along
the left and right nozzle x-axes shifts closer to the GC. This movement of the maximum thickness
along the x-axes as the nozzle length increases from N1 to N3 also leads to an increase in the
distance between the two thickness maximums along the y-axis [Fig. 10(c)]. The movement of the
maximum thickness is due to the movement of the nozzle outlet relative to the substrate as the
nozzle length increases, indicated by the point on the substrate associated with distance H in Figure
8c, which is directly under each nozzle outlet, upstream of the GC. Since the stem size is constant,
the shortest distance from each nozzle to the substrate moves closer to the GC as the nozzle length
increases from N1 to N3.

35

Figure 10. Thickness measurements along the a) left nozzle x-axis, b) right nozzle x-axis, and c)
y-axis for samples N1-1, N2-1, and N3-1.

We next studied the three nozzle lengths at a constant flow rate of 5 sccm (samples N1-5,
N2-5, and N3-5) which resulted in two distinct symmetric thickness profiles [Fig. 9(d-f)]. The
thickest points along the left and right nozzle x-axes for samples N1-5 and N2-5 are located
upstream of the GC and move closer towards the GC as the nozzle length increases from N1 to N2
[Fig. 11(a-b)]. For sample N3-5, the maximum thickness along the left and right nozzle x-axes is
located downstream of the GC as seen in Figure 4. Our previous study
101
showed that the
movement of the thickest point from upstream of the GC to downstream on the GC indicated a
transition from a diffusively dominant mass transport regime to a convectively dominant mass
transport regime as the flow rate increases. The movement of the maximum thicknesses for all
three samples along the x-axes corresponds to an increase in the distance between the two thickness
maximums along the y-axis as the nozzle length increases [Fig. 11c].

36

Figure 11. Thickness measurements along the a) left nozzle x-axis, b) right nozzle x-axis, and c)
y-axis for samples N1-5, N2-5, and N3-5.

We next studied the three nozzle lengths at a constant flow rate of 10 sccm (samples N1-
10, N2-10, and N3-10) which also resulted in two distinct symmetric thickness profiles [Fig. 9(g-
i)]. For sample N1-10, the thickness maximum along the left and right nozzle x-axes is located
downstream of the GC, similar to sample N3-5. The thickness profiles for samples N2-10 and N3-
10 have a maximum thickness farther downstream of the GC compared to the other samples, and
an indentation, or minimum thickness, to the left of the GC which aligns with the shortest
geometrical distance from each nozzle to the substrate. Since the thickest point falls downstream
of the GC along the x-axes for N2-10 and N3-10, there is a corresponding increase in the distance
between the two thickness maximums along the y-axis compared to N1-10 [Fig. 12c].

37

Figure 12. Thickness measurements along the a) left nozzle x-axis, b) right nozzle x-axis, and c)
y-axis for samples N1-10, N2-10, and N3-10.

In order to study the effect of the initiator flow rate relative to the monomer flow rate,
experiments were conducted using N3-5 with initiator flow rates of 0.5 to 2 sccm indicated by
sample N3-5-0.5I and N3-5-2I found in Table 2. The thickness measurements in Figure 13 indicate
that the overall shape of the thickness profile is maintained at all initiator flow rates, with two
distinct symmetric thickness profiles with the maximum thickness falling to the right of the GC
along the x-axes of both nozzles. Since the deposition time was held constant for all three samples,
the overall thickness for N3-5-05I is much less than that of the other two samples in the series. It
is also observed that at the highest initiator flow rate, the thickness far away from the nozzle outlets
is increased.

38

Figure 13. Thickness measurements along the a) left nozzle x-axis, b) right nozzle x-axis, and c)
y-axis for samples N3-5-0.5I, N3-5, and N3-5-2I.

To study the velocity and pressure profiles produced using the branched nozzle monomer
inlet extension, we used COMSOL Multiphysics to build a laminar flow model to analyze the
velocity at a slice 0.01 mm above the substrate and the pressure at the surface of the substrate along
the left nozzle x-axis, right nozzle x-axis, and y-axes within the reactor. These axes aligned with
the same axes where thickness measurements were taken. The three extensions were constructed
in COMSOL to match the same geometry of the physical extensions constructed for the
experimentation. Monomer flow rates of 1, 5, and 10 sccm were analyzed for each monomer inlet
extension, resulting in a total of nine computational models which matched the nine experiments
labeled in table 2. The models assumed a reactor pressure of 600 mTorr which was in the range of
pressures used experimentally and allows for direct comparison between the nine models. Using
pressures that aligned with the exact experimental pressures only changed the overall pressure but

39

did not result in changes to the shapes of the velocity and pressure profiles produced. The data
taken from the models shows that the x-axis velocity and pressure profiles for the left and right
nozzle mirror each other due to the symmetry of the system in all the computational models studied
and therefore only the right nozzle x-axis data is reported in the following analysis. These profiles
were also shifted to x = 0 at the GC, similar to the thickness measurements reported above. The
data also shows the velocity and pressure profiles along the y-axis, which intercepts the left and
right nozzle x-axes.

Figure 14. (a) Right nozzle x-axis and (b) y-axis velocity and pressure data for N1-5. (c) Right
nozzle x-axis and (d) y-axis velocity and pressure data for N2-5. (e) Right nozzle x-axis and (f) y-
axis velocity and pressure data for N3-5. The x-axis data is shifted to the geometric center (i.e., x
= 0, GC) and the y-axis data is shifted to be in line with the monomer inlet stem (i.e., y = 0). The
velocity is indicated by the blue line and the pressure is indicated by the dashed orange line.

To study the effect of increasing the nozzle length, we first compared the monomer inlet
extensions N1, N2, and N3 at a constant flow rate of 5 sccm. For each nozzle, as exemplified in
Figure 14(a, c, e), the maximum velocity is located to the right, or downstream, of the geometric
center (GC), and the maximum pressure is located to the left, or upstream, of the GC. The location

40

of the maximum pressure aligns to a local velocity minimum which corresponds to the stagnation
point, which matches our previous work using COMSOL Multiphysics on a single nozzle
system.
101
The stagnation point is a location where flow is brought to rest, converting kinetic
energy to static pressure.
91
It has been shown computationally and experimentally that the
stagnation point is located upstream of the geometric center for all angles of oblique impinging
flow.
92,102
The y-axis data, exemplified in Figure 9(b, d, f), shows that the maximum velocity of
each nozzle is separated by a region of decreased velocity and the maximum pressure of each
nozzle is separated by a region of decreased pressure, both associated with the separation distance
between the nozzles. There is a greater distance between the velocity maximums compared to the
pressure maximums because for both the left and right nozzles, the maximum velocity is located
downstream of the GC, while the maximum pressure is located upstream of the GC. As we increase
the nozzle length [Fig. 14(a, c, e)], there is an increase in the velocity and the pressure, and the
maximum velocity moves farther downstream of the GC while the maximum pressure moves
closer to the GC. There is also an increase in the difference between the local minimum and the
maximums along the y-axis for both velocity and pressure [Fig. 14(b, d, f)]. There is an increase
in the distance along the y-axis between the maximum velocities and pressures as the nozzle length
increases, which is consistent with the x-axes data since there is a movement downstream for the
velocity maximum, and movement of the pressure maximum closer to the GC, along the left and
right nozzle x-axes as the nozzle length increase. We also analyzed the velocity and pressure
profiles at a constant nozzle length while increasing the monomer flow rate. As the monomer flow
rate increases, for instance using N2 [Fig. 15(a, c, e)], there is an increase in the velocity and
pressure profiles and the maximum velocity moves farther downstream of the GC, while the
pressure maximum remains in the same location indicating that the location of the monomer outlet

41

relative to the substrate (i.e., which nozzle is used) is an important factor for determining the
location of the pressure maximum. As the flow rate increases [Fig. 15(b, d, f)] there is an increase
in the difference between the localized minimum and two maximums along the y-axis for both
velocity and pressure. There is also an increase in the distance between the maximum velocities as
the monomer flow increases [Fig. 15(b, d, f)] however, the distance between the pressure
maximums along the y-axis remains the same since the location of the pressure maximum does not
change with increasing monomer flow. The monomer inlet extensions N1 and N3 were studied at
flow rates of 1 sccm, 5 sccm, and 10 sccm and it was found that the trends associated with the
velocity and pressure profiles were consistent for all nozzle lengths and flow rates.

Figure 15. (a) Right nozzle x-axis and (b) y-axis velocity and pressure data for N2-1. (c) Right
nozzle x-axis and (d) y-axis velocity and pressure data for N2-5. (e) Right nozzle x-axis and (f) y-
axis velocity and pressure data for N2-10. The x-axis data is shifted to the geometric center (i.e.,
x = 0, GC) and the y-axis data is shifted to be in line with the monomer inlet stem (i.e., y = 0).
The velocity is indicated by the blue line and the pressure is indicated by the dashed orange line.

Although these models do not take into account the adsorption and reaction of the
monomer, which act like a monomer sink, it is interesting to note that the location of the maximum
thicknesses align with the maximum pressures for samples N2-1, N3-1, N1-5 and N2-5. The

42

alignment of the thickness maximums to the pressure maximums for these samples is likely
because in conventional iCVD the adsorption of the monomer, and therefore deposition, is
governed by the diffusively dominant mass transport regime and follows the Brunauer–Emmett–
Teller (BET) isotherm, where monomer adsorption increases as the pressure increases.
88
The
highest pressure, and therefore concentration of monomer, in the reactor is located at the outlet of
each nozzle of the extension and that concentration decreases as it travels farther away from this
point due to dispersion. This results in the location of the highest monomer surface concentration
being the location of the shortest distance from the outlet to the substrate (distance H) which is
directly under each nozzle outlet, upstream of the GC. As the nozzle length increases, the point
directly under each nozzle outlet moves closer to the GC, resulting in the maximum thickness
points moving closer to the GC as the nozzle length increases. The maximum thicknesses for
samples N1-10 and N3-5 match the location of the maximum velocities, which are downstream of
the GC. This movement of the maximum thickness downstream of the GC is likely due to the
transition from the diffusively dominant mass transport regime to a convectively dominant mass
transport regime as the flow rate, and therefore local velocity near the surface, increases. This leads
to the monomer surface concentration, and therefore deposition, to be governed by the velocity
profile which was shown in our previous study.
101
The velocity profile for an obliquely impinging
jet on a flat substrate has the maximum velocity located downstream of the GC with decreasing
velocity in all directions,
90,91,92
which match our current velocity profiles along the left and right
nozzle x-axes. In vapor deposition processes, film deposition depends on the local molecular flux
to the surface of the substrate. As precursor molecules approach the surface, they slow down due
to viscous friction forces, and the remaining transport of the molecules to the surface occurs by
diffusion through a stationary fluid boundary layer. It has previously been shown that the local

43

molecular flux, is inversely proportional to the thickness distribution of the boundary layer,
94

which is inversely proportional to the square root of the velocity near the surface of the
substrate.
95,96
This association between areas of high velocity and small boundary layer thickness
for oblique impinging jets at low Reynolds numbers has been confirmed experimentally and
computationally.
102
For samples N2-10 and N3-10, the maximum thickness does not align to the
pressure or velocity profile which may be due to divergences from our assumption of laminar flow
since other researchers have studied laminar impinging jet flow at oblique angles and found that
high local velocities near the surface can lead to velocity fluctuations and turbulence downstream
of the impingement point.
103,104

3.4 Conclusions

Branched nozzle inlet extensions were used to introduce monomer precursor flow at an
oblique angle relative to the substrate at two different locations during the initiated chemical vapor
deposition process. Experiments were performed by depositing poly(methacrylic acid) using three
branched nozzle inlet extensions. Three different nozzle lengths and three different monomer flow
rates were studied. The thickness profiles of the resulting films were analyzed. As the nozzle length
and monomer flow rate increased, the evolution of two distinct symmetrical thickness profiles was
observed. The maximums for each thickness profile were located upstream of the GC for shorter
nozzle lengths and slower monomer flow rates whereas the maximums were located downstream
of the GC at higher nozzle lengths and monomer flow rates. At the highest flow rates and nozzle
lengths, the resulting film had a maximum thickness farther downstream of the GC relative to the
other samples and an indentation in the film upstream of the GC. The effect of the initiator
concentration on the resulting thickness gradient profile was systematically studied and was

44

determined to have no effect on the profile shape. Computational models were created to analyze
the velocity and pressure profiles of the three different nozzle lengths and three different monomer
flow rates. The computational models showed that the velocity and pressure increased as the nozzle
length increased at a constant monomer flow rate, and the velocity maximum moved downstream
of the GC while the pressure maximum moved closer to the GC. Samples N2-1, N3-1, N1-5, and
N2-5 had thickness profiles which aligned with the pressure profile, indicating a diffusively
dominant mass transport regime while samples N3-5 and N1-10 had thickness profiles which
aligned with the velocity profile, indicating a convectively dominant mass transport regime. These
results show the ability to control the location of deposition using a branched nozzle inlet
extension.

3.5 Acknowledgements

USC Viterbi School of Engineering and the Gabilan Distinguished Professorship in
Science and Engineering was the source of funding for this work.

45

Chapter 4: Vapor Deposition of Silicon-Containing Microstructured
Polymer Films onto Silicone Oil Substrates
2

In this study, a silicon-containing crosslinked polymer, p(2,4,6,8-tetravinyl-2,4,6,8-
tetramethyl cyclotetrasiloxane-co-ethylene glycol diacrylate) (p(V4D4-co-EGDA)), was
deposited onto high viscosity silicone oil using initiated chemical vapor deposition (iCVD). The
ratio of the feed flow rate of V4D4 to EGDA was systematically studied and the chemical
composition and morphology of the top and bottom surfaces of the films were analyzed. The films
were microstructured and the porosity and thickness of the films increased with increasing V4D4
content. The top of the film was comprised of densely-packed and loosely-packed microstructured
regions. X-ray photoelectron spectroscopy on the top and bottom surfaces of the films showed a
heterogeneous chemical composition along the thickness of the film, with higher silicon content
on the top surface compared to the bottom surface. To our knowledge, this is the first study of
iCVD deposition of a silicon-containing polymer films onto silicone oil. The results of this study
can be used for the synthesis of polymer precursors films for the fabrication, via pyrolysis, of
silicon-based inorganic membranes for use in hydrogen production.

4.1 Introduction
In this work, we study for the first time the iCVD deposition of silicon-containing polymer
films onto low surface tension, high viscosity silicone oils, serving the same role as the SiC
mesoporous scaffolds above of confining the polymer film growth on the top of the macroporous
substrates, and thus avoiding infiltration during the preparation of asymmetric inorganic

2
Welchert et al. Langmuir, 37, 13859 (2021).

46

membranes. Our group has previously shown that the iCVD process can, indeed, be used to deposit
other types of polymers onto low vapor pressure liquid substrates, such as ionic liquids and silicone
oils.
60,105,106
The surface tension interactions between the deposited polymer and the liquid
substrate determine whether a dense polymer film
63
or polymer particles
61,62
form, the latter being
an undesirable outcome when preparing inorganic membrane and sensor films.
In this study, we systematically investigate the iCVD deposition of copolymer p(2,4,6,8-
tetravinyl-2,4,6,8-tetramethyl cyclotetrasiloxane-co-ethylene glycol diacrylate) (p(V4D4-co-
EGDA)) films, and study their properties as a function of their V4D4 content. For this study, the
ratio of the V4D4 to the EGDA feed flow rates into the iCVD reactor was systematically varied,
and the resulting chemical composition and morphology of the top and bottom surfaces of the
polymer film were analyzed. The crosslinker EGDA was used because the deposition of a
homopolymer pV4D4 film on the silicone oil surface at typical reactor conditions did not result in
film formation. The results of this study can be used to guide future research into the fabrication
of silicon-based inorganic membranes for use in hydrogen production, employing silicone oil as
an easy-to-apply barrier to reduce pre-ceramic polymer infiltration into the underlying
macroporous supports.

4.2 Materials and Experimental Procedure

Silicone oil (1000 cSt, Sigma-Aldrich), 2,4,6,8-tetravinyl-2,4,6,8-tetramethyl
cyclotetrasiloxane (V4D4) (Gelest, Inc.), ethylene glycol diacrylate (EGDA) (97%
MonomerPolymer), tert-butyl peroxide (98% Sigma-Aldrich), and hexane (98% VWR) were all
used as received. All polymer film depositions were performed using a custom-designed pancake-

47

shaped iCVD vacuum chamber that is 250 mm in diameter and 48 mm in height (GVD
Corporation). For polymer film depositions onto liquid, 0.5 mL of 1000 cst silicone oil was
pipetted onto a silicon wafer (Wafer World, 100 mm). The silicon wafers were then placed on the
reactor stage that was maintained at a constant temperature using a recirculating chiller (Thermo
Scientific NESLAB RTE 7). A nichrome filament array (Omega Engineering, 80%/20% Ni/Cr)
inside the reactor was heated during the deposition process to thermally decompose the TBPO
initiator. The reactor pressure was maintained using a throttle valve (MKS 153D) and measured
using a capacitance manometer (MKS Baratron 622A01TDE). The V4D4 and EGDA flow rates
were metered using a needle valve. The TBPO flow rate was metered using a mass flow controller
(MKS 1479A) and was maintained at 1 standard cubic centimeters per minute (sccm) for all
experiments. The polymer deposition rates were monitored on a reference silicon wafer using an
in situ 633 nm helium−neon laser interferometer (Industrial Fiber Optics). Table 3 lists the reactor
conditions for each sample.
A quartz crystal microbalance (QCM) (Sycon Instruments) with a 6 MHz gold-plated
crystal was used to compare the adsorption of the precursor onto a bare gold crystal versus the
adsorption and absorption of the precursor when a thin layer of 1000 cst silicone oil is added to
the crystal. A glass pipette was used to transfer silicone oil onto the gold crystal and the oil spread
to cover the entire surface of the crystal. The experiments were performed at pressures,
temperatures, and flow rates consistent with the range of deposition conditions. The uptake of each
precursor (V4D4, EGDA, and TBPO) was studied independently of each other. The mass uptake
of each precursor was allowed to equilibrate, which occurred within 2 minutes. The presence of
absorption was determined by measuring the difference in mass uptake of the precursor on the bare
gold crystal versus the gold crystal with a thin layer of silicone oil on the top of it.

48

The chemical composition of the deposited polymer films on the reference silicon wafers
was analyzed using Fourier transform infrared (FTIR) spectroscopy (Nicolet iS10, Thermo
Scientific). 32 scans were collected between 4000 and 500 cm
−1
with a resolution of 4 cm
−1
. In
order to confirm the chemical composition of the films deposited on the silicone oil, X-ray
photoelectron spectroscopy (XPS) survey scans of the removed copolymer films were taken at
both the top and bottom surfaces of the films. A Kratos Axis Ultra DLD XPS spectrometer
equipped with a magnetic immersion lens, a charge neutralization system, and a monochromator
Al X-ray source was used to collect the survey spectra. Survey spectra were taken from 800 eV to
0 eV in 1 eV steps and averaged over seven scans. The copolymer films were removed from the
liquid after deposition by placing the sample in a hexane bath for 15 minutes to separate the film
from the silicone oil. The films were then cut into two samples and were placed on two separate
clean silicon wafer mounts, with the one sample having the top surface and the other sample having
the bottom surface exposed. Once the samples were mounted, they were soaked in hexane for
another 15 minutes to remove any residual silicone oil. The samples were then removed from the
hexane bath and placed into a vacuum chamber at room temperature for 12 hours. The morphology
of the cross-section, top and bottom surfaces of the films was imaged by scanning electron
microscopy (SEM) (JEOL-4500) using a 25 kV acceleration voltage. Prior to imaging, the samples
were sputter-coated for 60 s with gold.

4.3 Results and Discussion

For our studies, we used 1000 cst silicone oil as our substrate. The absorption of monomer
molecules in the liquid substrate has been shown to affect the morphology of the deposited

49

polymer.
107,108
A quartz crystal microbalance (QCM) was used in this research to study 2,4,6,8-
tetravinyl-2,4,6,8-tetramethyl cyclotetrasiloxane (V4D4), ethylene glycol diacrylate (EGDA), and
tert-butyl peroxide (TBPO) absorption in the silicone oil at the relevant reactor conditions. We
have previously shown
62,109
that QCM measurements can be used as a sensitive indicator of
whether a certain precursor only adsorbs on the surface of a liquid substrate versus absorbing in
its bulk. By first measuring the mass uptake on a bare gold surface and then on the same surface
coated by a thin layer of liquid, one can readily distinguish whether only adsorption versus
absorption takes place,
109
since in the former case the mass uptakes are quite similar and the
adsorbed amounts on a bare gold surface and on a liquid-coated gold surface are, typically, close
to each other. Significant differences in the mass uptakes measured, on the other hand, offer strong
indication that absorption does, indeed, take place.
The measured mass uptakes of V4D4, EGDA and TBPO on a bare gold surface were 0.060,
0.044, and 0.016 µg/cm
2
, respectively. After applying a thin layer of 1000 cst silicone oil, the
corresponding mass uptakes of V4D4, EGDA and TBPO were 6.282, 0.049, and 0.012 µg/cm
2
,

respectively. The mass uptakes of EGDA and TBPO with and without silicone oil are similar,
likely indicating that there is no absorption in the silicone oil, whereas the significant (2 orders of
magnitude) increase in V4D4 mass uptake in the presence of silicone oil indicates that there is
V4D4 absorption in the silicone oil. Since the V4D4 monomer absorbs in the silicone oil, it would
be unlikely that one could identify reactor conditions under which a homopolymer pV4D4 film
would form at the surface of the silicone oil.
Since the QCM tests indicated that the EGDA monomer does not absorb in the silicone oil and
remains instead on the surface of the liquid, we investigated whether copolymerizing EGDA and
V4D4 would help lead to the formation of p(V4D4-co-EGDA) films on the surface of the silicone

50

oil. In these experiments, we systematically varied the ratio (r) of the reactor feed flow rate of
V4D4 to that of EGDA from 1:1.5 to 1:1 and finally to 1.5:1 in order to prepare three different
samples A, B, and C, as described in Table 3. All three different feed flow rate ratios resulted in
the deposition of a thin polymer film on the surface of the silicone oil.

Table 3. Deposition conditions for the various samples.

Sample V4D4 Flow
Rate
(sccm)
EGDA
Flow Rate
(sccm)
V4D4:EGDA
Flow Ratio (r)
Stage
Temperature
(°C)
Reactor
Pressure
(mTorr)
Deposition
Rate
(
nm
min.
)
A 0.4 0.6 1:1.5 40 75 4.2
B 0.4 0.4 1:1 40 110 4.7
C 0.6 0.4 1.5:1 40 110 4.2
D - 0.4 - 40 55 4.6

To keep the mass of polymer deposited on the liquid surface consistent among the three
different samples prepared, the polymerization reaction was allowed to proceed until a 400 nm
thick film, measured in situ by laser interferometry, was deposited on a reference silicon wafer
placed in the same reactor. A control deposition of homopolymer pEGDA onto the silicone oil was
also conducted (sample D), which resulted in the formation of a thin polymer film on the surface
of the silicone oil. To determine the chemical composition of the copolymer films, Fourier
transform infrared (FTIR) analysis was conducted on the films deposited on reference silicon
wafers and the spectra were compared to homopolymer pV4D4 and homopolymer pEGDA films
also deposited on reference silicon wafers. As shown in Figure 16, the spectra of the three different
copolymer films (A, B, and C, see Table 3) contain the characteristic C=O bond peak at 1735 cm
-
1
, which indicates the presence of EGDA; in addition, all three spectra contain the characteristic

51

Si-O-Si peak at 1065 cm
-1
, which indicates the presence of V4D4, thus confirming
copolymerization. The ratio of the intensity of the Si-O-Si peak to that of the C=O peak in samples
A, B, and C is equal to 0.6, 0.9, and 1.2, respectively, thus indicating an increase in the fraction of
V4D4 incorporated into the copolymer films as the feed flow rate ratio of V4D4 to that of EGDA
increases.

Figure 16. FTIR spectra of pV4D4 and pEGDA films and of copolymer films deposited on
reference wafers placed in the reactor during deposition of samples A, B, and C; dashed lines
indicate the position of the C=O and Si-O-Si peak

Figure 17. Schematic depicting the top and bottom surfaces of the p(V4D4-co-EGDA) film
deposited onto silicone oil.

52

The chemical composition of the top and bottom surfaces of the copolymer films deposited
on the silicone oil, as shown in Figure 17, were analyzed via X-ray photoelectron spectroscopy
(XPS). The deposited samples were placed in a hexane bath to remove the silicone oil and the
films were then carefully removed, placed on a separate clean silicon wafer, and dried under
vacuum at room temperature. As shown in Table 4, there is an increase in the silicon content both
at the top and bottom surfaces of the film as the V4D4 flow rate increases relative to the EGDA
flow rate. This indicates greater incorporation of V4D4 into the copolymer, consistent with the
FTIR data of the films deposited on the reference silicon wafers. It is also observed that for all
samples, the silicon content at the top surface is greater than the content at the bottom surface.
Since V4D4 does not polymerize into a thin film at typical reaction conditions due to the fact that
it absorbs into the silicone oil, the copolymer films are likely formed by the EGDA first
polymerizing on the surface of the liquid to form a base layer before its reaction with V4D4
commences. This copolymer film formation mechanism is consistent with the observed
nonuniform chemical composition along the thickness of the film, see Table 4, with less
incorporation of V4D4 found at the bottom surface.

53

Table 4. XPS atomic composition of the top and bottom surfaces of the p(V4D4-co-EGDA)
films deposited on silicone oil.
Atomic Composition (%)
Sample Carbon Oxygen Silicon
Top Surface
A 68.92 18.02 13.69
B 58.95 23.96 17.09
C 54.99 25.57 19.44
Bottom Surface
A 67.67 24.78 7.55
B 71.99 19.75 8.25
C 68.62 20.87 10.51
Reference Homopolymer
pEGDA 65.63 30.5 3.87
pV4D4 50.51 26.19 23.3

Scanning electron microscopy (SEM) was used to visualize the bottom surface of the pEGDA
and p(V4D4-co-EGDA) films, which are all composed of microstructures (see top-down images
in Figure 18). Our previous work
64
has shown that homopolymer pEGDA films formed on silicone
oils of various viscosities via iCVD deposition are characterized by similar microstructures; we
speculated in the earlier study that this is due to chemical cross-linking which helps to form
polymer networks. The microstructures grow by simultaneous polymer diffusion and aggregation
and wetting of the growing aggregates by the liquid. Higher viscosity liquids result in slower
diffusion and aggregation of the wetted polymer,
63
leading to more dense microstructured pEGDA
films like the one shown in Figure 18. The bottom surfaces of all three p(V4D4-co-EGDA) films
(Sample A, B, and C) appear highly porous, characterized by three-dimensional structures; the
porosity and three-dimensional character of the film both increase as the content of V4D4 in the
copolymer increases relative to EGDA. The morphology of the copolymer films is, in fact, similar

54

to that observed in our previous study with pEGDA films deposited on lower viscosity silicone
oils (5-500 cst).
64
In that case, lower viscosity liquids cause greater diffusion and aggregation of
the wetted polymer, leading to more porous films. The impact that V4D4 has on the characteristics
of the polymer films formed may suggest, therefore, that the incorporation of V4D4 increases
polymer diffusion. We hypothesize that the increase in polymer diffusion is likely due to the
chemical similarity of V4D4 which also explains its higher solubility in the silicone oil.
110

Figure 18. Top-down SEM images of the bottom surface of the pEGDA film and copolymer films
(samples A, B, and C) deposited onto 1000 cst silicone oil.

SEM images of the top surface of the pEGDA and the copolymer films were also taken. Both
the pEGDA and copolymer films are macroscopically non-uniform with both densely-packed and
loosely-packed microstructured regions (an example for the copolymer case is shown in Figure
19). The densely-packed regions are characterized by larger size microstructures with lower

55

porosity and three-dimensional character relative to the loosely-packed regions. Figure 20 shows
images of the densely-packed regions for both the pEGDA and the copolymer films. The densely-
packed region for pEGDA is a completely dense, non-porous film. As more V4D4 is incorporated
in the copolymer, the densely-packed regions of the films become more porous and have distinct,
larger size microstructures. The increase in the microstructure size and porosity of the film, as the
V4D4 content in the polymer increases relative to EGDA, is likely due to an increase in polymer
diffusion and aggregation at higher concentrations of V4D4, consistent with the trend found on
the bottom of the film. Figure 21 shows images of the loosely-packed regions for both the pEGDA
and the copolymer films. These regions are characterized by a higher porosity, more prominent
three-dimensional nature, and smaller size microstructures relative to the densely-packed regions.
As more V4D4 is incorporated into the copolymer, the porosity, three-dimensional nature, and
microstructure size of the loosely-packed regions of the films all increase, similarly to the behavior
shown by the densely-packed regions.

Figure 19. (a) SEM image of the top surface of the p(V4D4-co-EGDA) film (Sample A) showing
an example of the densely-packed and loosely-packed regions. (b) Higher magnification image of
the interface between the densely-packed and loosely-packed regions.

56

Figure 20. SEM images of the densely-packed regions of the top surface of the pEGDA film and
copolymer films (samples A, B, and C) deposited onto 1000 cst silicone oil.

Figure 21. SEM images of the loosely-packed regions of the top surface of the pEGDA film and
copolymer films (samples A, B, and C) deposited onto 1000 cst silicone oil.

SEM images of the cross-sections of the films are shown in Figure 22. The pEGDA has a
relatively dense, microstructured cross-section similar to what has been observed in our previous
work using high viscosity liquids.
64
The thickness of all the p(V4D4-co-EGDA) films is larger
than the thickness of the pEGDA film. The cross-sections of the copolymer films have columnar
microstructures and increase in thickness as more V4D4 is incorporated into the film, which
indicates faster polymer diffusion and aggregation of the wetted polymer, which is consistent with
the SEM images of the top and bottom surfaces of the films. In order to observe how reaction time

57

affects the thickness of the films, p(V4D4-co-EGDA) was deposited with a flow rate ratio of 1:1
for three different deposition times of 45, 90, and 120 minutes. As Figure 22b indicates, there is
an increase in film thickness as deposition time increases. This indicates that as deposition
continues, polymer diffusion, aggregation, and wetting continuously occur throughout the time of
deposition, thus creating thicker films. XPS was conducted on the top and bottom surfaces of the
three samples generated with different deposition times. Similar to samples A, B, and C, see Table
5, the bottom surface showed a lower silicon content than the top side, which again is explained
by the fact that the EGDA polymerizes first to form the base layer of the film. The chemical
composition of the top surface did not show a similar trend, indicating that deposition time has no
effect on the silicon content in the top of the film.

Table 5. XPS atomic composition of the top and bottom surfaces of the p(V4D4-co-EGDA)
films deposited on silicone oil for 45, 90, and 120 minutes at a V4D4 to EGDA flow rate ratio of
1:1.

Atomic Composition (%)
Deposition Time
(min.)
Carbon Oxygen Silicon
Top Surface
45 59.83 21.83 18.34
90 58.95 23.96 17.09
120 56.16 24.42 19.42
Bottom Surface
45 71.65 19.67 8.69
90 71.99 19.75 8.25
120 69.57 20.63 9.79

58

Figure 22. (a) SEM images of the cross-sections of the pEGDA film and copolymer films
(samples A, B, and C) deposited onto 1000 cst silicone oil. (b) SEM images of the cross-section
of p(V4D4-co-EGDA) film deposited on 1000 cst silicone oil for 45, 90, and 120 minutes at a
V4D4 to EGDA flow rate ratio of 1:1.

4.4 Conclusions
The deposition of p(V4D4-co-EGDA) copolymer films onto 1000 cst silicone oil via iCVD
was systematically studied by varying the flow rates of the two precursor monomers. The
morphology and chemical composition of the copolymer films were compared to homopolymer
pEGDA films. All the films studied were microstructured. The copolymer films had larger
microstructures, higher porosity, and higher thickness compared to the homopolymer pEGDA
films, indicating an increase in polymer diffusion and aggregation as V4D4 is incorporated into
the film. The films were removed from the liquid surface and SEM was used to visualize the
morphology at both their top and bottom surfaces. At the bottom surface of the film, there is greater
porosity and three-dimensional character as the concentration of V4D4 in the copolymer film

59

increases. At the top surface of the copolymer films, there are both densely-packed and loosely-
packed microstructured regions. As the V4D4 content in the polymer increases, it is observed that
there is an increase in the microstructure size and in porosity of both the densely-packed and
loosely-packed microstructured regions. The cross-sections of all the copolymer films show an
increase in thickness relative to the pEGDA film. XPS analysis showed that the copolymer films
had a higher silicon content at the top surface relative to the bottom surface. This is likely due to
EGDA polymerizing first on the surface of the liquid to form a base layer before reacting with
V4D4, thus resulting in an inhomogeneous chemical composition along the thickness of the film.
As the V4D4 flow rate into the reactor increased relative to the EGDA, an increase in silicon
content at both the top and bottom surfaces of the film was observed. These results indicate that
the flow rate ratio controls the morphology and chemical composition of the copolymer films
formed.

4.5 Acknowledgements

This work was supported by the National Science Foundation Award Number CMMI- 2012196.

60

4.6 Future Work: In Situ Pyrolysis of Vapor Deposited Polymer Films for
the Formation of inorganic Silicon-Based Inorganic Materials

The future work will focus on the iCVD polymerization of siloxane monomers with
subsequent in situ pyrolysis for the fabrication of silicon-based membranes. We developed a
system to study the deposition of a polymer film, p(1,3,5,7-tetravinyl-1,3,5,7-
tetramethylcyclotetrasiloxane) (pV4D4), using low energy plasma enhanced chemical vapor
deposition (PECVD), followed by pyrolysis of the film using a heating element placed within the
vacuum chamber reactor. Low-energy PECVD was chosen as an alternative method to iCVD,
since the presence of the microheater system inside the reactor did not allow for cooling of the
substrate temperature during deposition, as required by the iCVD method. In situ pyrolysis of the
polymer prevents exposing the sample to atmospheric conditions, which reduces the likelihood of
external contamination. The in situ pyrolysis system can also increase the sample throughput of
the vapor deposited silicon-containing polymers since both the deposition and pyrolysis steps are
done in a single batch process, therefore reducing the time required for processing. We plan to
analyze the in situ pyrolyzed film to a film which was exposed to atmosphere and subsequently
pyrolyzed. This work will also look at depositing other siloxane polymers which can be analyzed
with and without exposure to atmosphere. The results of this study can be used to further the
research of silicon-based inorganic membranes for use in hydrogen production, transportation, and
storage by combining the vapor deposition and pyrolysis steps into a streamlined process.
Furthermore, the same system can be used for carrying out in situ other heat-treatment processes
of vapor-deposited polymer films needed for crosslinking,
71,72
increasing crystallinity,
73,74,75
and
driving self-assembly for patterning.
76,77,78
To our knowledge, this will be the first time the vapor
deposition of a polymer film followed by in situ pyrolysis in a single batch process was preformed,

61

thus avoiding the exposure of the polymer to ambient air, which may lead to oxidation during the
pyrolysis step.

62

References

1
S. G. Im and K. K. Gleason, AIChE J. 57, 276 (2011).

2
S. H. Baxamusa and K. K. Gleason, Chem. Vapor Depos. 14, 313 (2008).

3
S. H. Baxamusa, S. G. Im and K. K. Gleason, Phys. Chem. Chem. Phys. 11, 5227 (2009).

4
K. K. S. Lau and K. K. Gleason, Adv. Mater. 18, 1972 (2006).

5
M. Gupta, V. Kapur, N. M. Pinkterton, and K. K. Gleason, Chem. Mater. 20, 1646 (2008).

6
W. S. O’Shaughnessy, S. K. Murthy, D. J. Edell, and K. K. Gleason, Biomacromolecules 8, 2564
(2007).

7
M. E. Alf, P. D. Godfrin, T. A. Hatton, and K. K. Gleason, Macromol. Rapid Commun. 31, 2166 (2010).

8
M. Ma, Y. Mao, M. Gupta, K. K. Gleason, and G. C. Rutledge, Macromolecules 38, 9742 (2005).

9
C. Cheng, and M. Gupta, Ind. Eng. Chem. Res. 57, 11675 (2018).

10
C. Cheng, and M. Gupta, Beilstein J. Nanotechnol. 8, 1629 (2017).

11
P. D. Haller, C. A. Flowers, and M. Gupta, Soft Matter 7, 2428 (2011).

12
M. Gupta and K. K. Gleason, Langmuir 22, 10047 (2006).

13
K. K. S. Lau and K. K. Gleason, Macromolecules 39, 3695 (2006).

14
K. K. S. Lau and K. K. Gleason, Macromolecules 39, 3688 (2006).

15
G. Ozaydin-Ince and K. K. Gleason, J. Vac. Sci. Technol. A 27, 1135 (2009).

16
M. M. Hawkeye and M. J. Brett, J. Vac. Sci. Technol. A 25, 1317 (2007).

63

17
D. A. Rider, Nanotechnology 22, 085706 (2011).

18
M. K. Fung, Y. C. Sun, A. Ng, A. M. C. Ng, A. B. Djurisic, H. T. Chan, and W. K. Chan, ACS Appl.
Mater. Interfaces 3, 522 (2011).

19
M. D. Gasda, G. A. Eisman, and D. J. Gall, Electrochem. Soc. 157, B437 (2010).

20
W. J. Khudhayer, A. U. Shaikh, and T. Karabacak, Nanosci. Nanotechnol. Lett. 4, 999 (2012).

21
G. K. Kannarpady, K. R. Khedir, H. Ishihara, J. Woo, O. D. Oshin, and S. Trigwell, ACS Appl. Mater.
Interfaces 3, 2332 (2011).

22
M. K. Dawood, H. Zheng, N. A. Kurniawan, K. C. Leong, Y. L. Foo, and R. Rajagopalan, Soft Matter
8, 3549 (2012).

23
A. Barranco, A. Borras, A. R. Gonzalez-Elipe, and A. Palmero, Prog. Mater. Sci. 76, 59 (2016).

24
K. Robbie, J. C. Sit, and M. J. Brett, J. Vac. Sci. Technol. A 16, 1115 (1998).

25
S. Sarkar and S. K. Pradhan, Surf. Eng. 35, 982 (2019).

26
R. K. Khan, A. A. Farghaly, T. A. Silva, D. Ye, and M. M. Collinson, ACS Appl. Nano. Mater. 2, 1562
(2019).

27
T. An, X. Deng, Y. Gao, S. Liu, C. Dou, and J. Ju, Mater. Lett. 216, 147 (2018).

28
S. Pursel, M. W. Horn, M. C. Demirel, and A. Lakhtakia, Polymer 46, 9544 (2005).

29
N. A. Malvadkar, M. J. Hancock, K. Sekeroglu, W. J. Dressick, and M. C. Demirel, Nat. Mater. 9, 1023
(2010).

30
H. Xin and W. Li, Appl. Phys. Rev. 5, 031105 (2018).

31
B. Huet, X. Zhang, J. M. Redwing, D. W. Snyder, and J. P. Raskin, 2D Mater. 6, 045032 (2019).

64

32
W. Kern, and G. L. Schinable, IEEE Trans. Electron Devices 26, 647 (1979).

33
D. Dobkin, and M. K. Zuraw, Principles of chemical vapor deposition. (Springer Science & Business
Media, 2003).

34
D. M. Gill, B. A. Block, C. W. Conrad, B. W. Wessels, and S. T. Ho, Appl. Phys. Lett. 69, 2968 (1996).

35
T. H. Bointon, M. D. Barnes, S. Russo, and M. F. Craciun, Adv. Mater. 27, 4200 (2015).

36
Y. Wang, X. Xu, J. Lu, M. Lin, Q. Bao, B. Ozyilmaz, and K. P. Loh, ACS Nano. 4, 6146 (2010).

37
R. Hawaldar, P. Merino, M. R. Correia, I. Bdikin, J. Gracio, J. Mendez, J. A. Martin-Gago, and M. K.
Singh, Sci. Rep. 2, 682 (2012).

38
M. R. Wertheimer, J. E. Klemberg-Sapieha, J. Cerny, and S. Liang, Plasmas Polym. 3, 151 (1998).

39
P. Kovacik, G. del Hierro, W. Livernois, and K. K. Gleason, Mater. Horiz. 2, 221 (2015).

40
M. Gupta and K. K. Gleason, Thin Solid Films 515, 1579 (2006).

41
N. Cheimarios, G. Kokkoris, and A. G. Boudouvis, Arch. Computat. Methods Eng. 28, 637 (2021).

42
C. R. Kleijn, Thin Solid Films 365, 294 (2000).

43
N. Cheimarios, E. D. Koronaki, and A. G. Boudouvis, Chem. Eng. J. 181-182, 516 (2012).

44
J. Brcka, “Model of a Filament Assisted CVD reactor”, Proc. COMSOL Conference, Boston, MA,
(2009).

45
N. W. Ockwig and T. M. Nenoff, Chem. Rev. 107, 4078 (2007).

46
B. Elyassi, M. Sahimi, and T. Tsotsis, Encycl. Chem. Process. Des. 1:1, 1 (2009).

47
S. J. Khatib, S. T. Oyama, K. de Souza and F. B. Noronha, Membr. Sci. Technol. 14, 25 (2011).

65

48
N. A. Z. Khalib, IOP Conf. Ser. Mater. Sci. Eng. 290 (2018).

49
V. Boffa, D. H. A. Blank and J. E. Elshof, J. Memb. Sci. 319 (1–2), 256 (2008).

50
S. Dabir, W. Deng, M. Sahimi and T. T. Tsotsis, J. Membrane Sci. 537, 239 (2017).

51
Y. Iwamoto, K. Sato, T. Kato, T. Inada and T. Kubo, J. Eur. Ceram. Soc. 25, 257 (2005).

52
S. Adhikari and S. Fernando, Ind. Eng. Chem. Res. 45, 875 (2006).

53
T. Konegger, C. C. Tsai and R. K. Bordia, Mater. Sci. Forum 825, 645 (2015).

54
S. Sano, T. Banno, M. Maeda, K. Oda and Y. Shibasaki, J. Ceram. Soc. Jpn. 104, 10, 984 (1996).

55
Z. Ren, S. B. Mujib and G. Singh, Materials 14 (3), 614 (2021).

56
A. Klausmann, K. Morita, K. Johanns, C. Fasel, K. Durst, G. Mera, R. Riedel and E. Ionescu, J. Eur.
Ceram. Soc. 35, 3771 (2015).

57
R. Hauser, S. Nahar-Borchard, R. Riedel, Y. Ikuhara and Y. Iwamoto, J. Ceram. Soc. Jpn. 114, 6, 524
(2006).

58
J. Xu and K. K. Gleason, ACS Appl. Mater. Interfaces 3, 2410 (2011).

59
B. Nguyen, S. Dabir, T. Tsotsis and M. Gupta., Ind. Eng. Chem. Res. 58, 15190 (2019).

60
M. M. De Luna, P. Karandikar and M. Gupta, Mol. Syst. Des. Eng. 5, 15 (2019).

61
P. D. Haller and M. Gupta, Macromol. Rapid Commun. 35, 2000 (2014).

62
R. J. Frank-Finney and M. Gupta, Langmuir 32, 11014 (2016).

63
P. D. Haller, L. C. Bradley and M. Gupta, Langmuir 29, 11640 (2013).

66

64
L. C. Bradley and M Gupta, Langmuir 31, 7999 (2015).

65
L. C. Bradley and M. Gupta, Langmuir 29, 10448 (2013).

66
S. R. Droes, T. T. Kodas and M. J. Hampden-Smith, Carbide, Nitride and Boride Materials Synthesis
and Processing (Chapman & Hall, 1997).

67
D. A. Ondo, F. Loyer, and N. D. Boscher, Plasma Process Polym. 18, 2000222 (2021).

68
D. A. Ondo, F. Loyer, F. Werner, R. Leturcq, P. J. Dale, and N. D. Boscher, ACS Appl. Polym. Mater.
1, 3304 (2019).

69
E. Chen, G. Du, Y. Zhang, X. Qin, H. Lai, and W. Shi, Ceram. Int. 40, 9791 (2014).

70
S. Cogan, D. Edell, A. Guzelian, Y. Liu, and Edell, J. Biomed. Mater. Res. 67A, 3, 856 (2003).

71
Y. Mao and K. K. Gleason, Macromolecules 39, 3895 (2006).

72
J. L. Yagü e and K. K. Gleason, Macromolecules 46, 6548 (2013).

73
B. A. Abdul-Majeed, H. K. Hussain, and N. A. Kafel Al-Sultanee, Iraqi Journal of Chemical and
Petroleum Engineering 13, 29 (2012).

74
T. J. Shin, and M. Ree, J. Phys. Chem. B 111, 13894 (2007).

75
J. Hsu, L. Rieth, S. Kammer, M. Orthner, and F. Solzbacher, Sens. Mater. 20, 87 (2008).

76
D. W. Janes, J. M. Katzenstein, K. Shanmuganathan, and C. J. Ellison, J. Polym. Sci., Part B: Polym.
Phys. 51, 535 (2013).

77
S. Y. Chou and L. Zhuang, J. Vac. Sci. Technol. B 17, 3197 (1999).

78
C. K. Shelton and T. H. Epps, Macromolecules 48, 4572 (2015).

79
E. M. Benetti, M. K. Gunnewiek, C. A. van Blitterswijk, G. J. Vancso, and L. Moroni, J. Mater. Chem.
B 4, 4244 (2016).

67

80
L. Li, Y. Zhu, B. Li, and C. Gao, Langmuir 24, 13632 (2008).

81
K. E. Roskov, T. H. Epps III, B. C. Berry, S. D. Hudson, M. S. Tureau, and M. J. Fasolka, J. Comb.
Chem. 10, 966 (2008).

82
J. C. Meredith, A. P. Smith, A. Karim, and E. J. Amis, Macromolecules 33, 9747 (2000).

83
X. Li, and B. D. Vogt, Polymer 50, 4182 (2009).

84
M. Luo, J. E. Seppala, J. N. L. Albert, R. L. Lewis III, N. Mahadevapuram, G. E. Stein, and T. H. Epps
III, Macromolecules 46, 1803 (2013).

85
J. M. Rathfon, R. W. Cohn, A. J. Crosby, J. R. Rothstein, and G. N. Tew, Macromolecules 44, 5436
(2011).

86
K. G. Yager, B. C. Berry, K. Page, D. Patton, A. Karim, and E. J. Amis, Soft Matter, 5, 622 (2009).

87
Crane Co., Flow of Fluids Through Valves, Fittings and Pipe. Technical Paper No. 410 (TP 410)
(Crane Co., U.S.A. 1988).

88
W. E. Tenhaeff and K. K. Gleason, Adv. Funct. Mater. 18, 979 (2008).

89
T. DebRoy, K. Tankala, W. A. Yardbrough, and R. Messier, J. Appl. Phys. 68, 2424 (1990).

90
J. F. Foss, AIAA J. 17, 801 (1979).

91
L. J. Clancy, Aerodynamics (Pitman Publishing Limited, London. 1979).

92
T. S. O’Donovan, and D. B. Murray, Int. J. Heat Mass Transfer 51, 6169 (2008).

93
T. P. Martin, and K. K. Gleason, Chem. Vapor Depos. 12, 685 (2006).

94
A. C. Jones, and M. L. Hitchman, Chemical Vapour Deposition Precursors, Processes and
Applications (Royal Society of Chemistry, Cambridge. 2009).

68

95
D. Dandy, and J. Yun, J. Mater. Res. 12, 1112 (1997).

96
N. Curle, The Laminar Boundary Layer Equations (Courier Dover Publications, New York. 2017).

97
A. Mishra, H. Yadav, L. Djenidi, and A. Agrawal, Exp. Fluids. 61, 90 (2020).

98
R. Gardon, and J. C. Akfirat, Int. J. Heat Mass Transfer., 8, 1262 (1965).

99
P. Kwong and M. Gupta, Anal. Chem. 84, 10129 (2012).

100
G. Dianat, K. Gao, S. Prevoir, and M. Gupta, Appl. Polym. Mater. 2, 3339 (2020).

101
N. A. Welchert, C. Cheng, P. Karandikar, and M. Gupta, J. Vac. Sci. Technol. A 38, 063405 (2020).

102
X. Zhang, S. Yarusevych, and S. D. Peterson, Exp. Fluids. 60, 11 (2019).

103
A. Mishra, H. Yadav, L. Djenidi, and A. Agrawal, Exp. Fluids. 61, 90 (2020).

104
R. Gardon, and J. C. Akfirat, Int. J. Heat Mass Transfer. 8, 1262 (1965).

105
M. M. De Luna, P. Karandikar, and M. Gupta, ACS Appl. Nano Mater. 1, 6575 (2018).

106
P. Karandikar and M. Gupta, Langmuir 33, 7701 (2017).

107
P. D. Haller, R. J. Frank-Finney, and M. Gupta, Macromolecules 44, 2653 (2011).

108
P. Karandikar and M. Gupta, Thin Solid Films 635, 17 (2017).

109
R. J. Frank-Finney, L. C. Bradley and M. Gupta, Macromolecules 46, 6852 (2013).

110
C. H. Kim, C. Joo, H. J. Chun, B. R. Yoo, D. I. Noh and Y. B. Shim, Appl. Surf. Sci. 262, 146 (2012).

Abstract (if available)

Abstract This work is split into two sections in which unique processes are used in order to control the patterning of thin polymer films and the ability to deposit silicon containing polymers onto high viscosity liquids for use in hydrogen production via methane steam reforming. Chapter 1 will go into the fundamentals of initiated chemical vapor deposition (iCVD) and the motivation and background for the works in Chapters 2, 3, and 4.
Chapter 2 discusses how iCVD can be used to deposit thin polymer films on a variety of substrates. In this work, the monomer precursor was introduced at an oblique angle to the substrate using an inlet extension and the pattern of the resulting polymer film was studied as a function of the deposition time, substrate temperature, monomer flow rate, reactor pressure, and vapor flow angle. The polymerization of n-butyl acrylate (nBA), methacrylic acid (MAA), and 2-hydroxyethyl methacrylate (HEMA) was examined to determine the generality of the trends across several monomers. It was found that the monomer flow rate significantly affected the pattern of the deposited polymer by shifting the location of the thickest point in the films. Increasing the deposition time, decreasing the substrate temperature, and increasing the reactor pressure increased the polymer deposition rate consistent with conventional iCVD, however the pattern of the deposited polymer did not vary with these parameters. Computational analysis was used to determine how the inlet extension affects the pressure and velocity profiles within the reactor. The data demonstrated that the introduction of the monomer precursor at an oblique angle can be used to pattern polymer films during iCVD.
Chapter 3 is a follow-up work to chapter 2 and discusses work in which monomer precursor flow was introduced at an oblique angle to the substrate at two locations during the iCVD process using a branched nozzle inlet extension. The polymerization of MAA was systematically studied as a function of the nozzle length and the monomer flow rate. Our experimental data showed the evolution of two distinct symmetrical thickness profiles as the flow rate and nozzle length increased. The maximum thickness moved downstream along the axes of both nozzles as the flow rate and nozzle length increased. Computational models were used to study the effects of the nozzle length and the monomer flow rate on the velocity and pressure profiles within the reactor. Increasing the monomer flow rate and the nozzle length resulted in increases in both the velocity and pressure profile ranges and the movement of the location of the maximum velocity and pressure. This velocity and pressure data provided insight for explaining the trends found in the experimental results. The data demonstrated the ability to use a branched nozzle inlet extension to control the location of polymer deposition during the iCVD process.
Finally, in chapter 4 we discuss an experiment where a silicon-containing crosslinked polymer, p(2,4,6,8-tetravinyl-2,4,6,8-tetramethyl cyclotetrasiloxane-co-ethylene glycol diacrylate) (p(V4D4-co-EGDA)), was deposited onto high viscosity silicone oil using initiated chemical vapor deposition (iCVD). The ratio of the feed flow rate of V4D4 to EGDA was systematically studied and the chemical composition and morphology of the top and bottom surfaces of the films were analyzed. The films were microstructured and the porosity and thickness of the films increased with increasing V4D4 content. The top of the film was comprised of densely-packed and loosely-packed microstructured regions. X-ray photoelectron spectroscopy on the top and bottom surfaces of the films showed a heterogeneous chemical composition along the thickness of the film, with higher silicon content on the top surface compared to the bottom surface. To our knowledge, this is the first study of iCVD deposition of a silicon-containing polymer films onto silicone oil. The results of this study can be used for the synthesis of polymer precursors films for the fabrication, via pyrolysis, of silicon-based inorganic membranes for use in hydrogen production.
In chapter 4 we discuss the future directions for this work, including the low energy plasma enhanced chemical vapor deposition of siloxane polymers, namely pV4D4, followed by the in situ pyrolysis to form silicon-based inorganic materials for use in hydrogen production, transportation, and storage.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Novel processing of liquid substrates via initiated chemical vapor depostion

PDF

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

PDF

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

PDF

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

PDF

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

PDF

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

PDF

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

PDF

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

PDF

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

PDF

Understanding structural changes of polymer films during vapor phase deposition

PDF

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

PDF

Selective deposition of polymer coatings onto structured surfaces

PDF

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

PDF

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

PDF

Controlled synthesis of large grain single crystal graphene by chemical vapor deposition

PDF

Biogas reforming: conventional and reactive separation processes and the preparation and characterization of related materials

PDF

Chemical vapor deposition of graphene and two-dimensional materials: synthesis, characterization, and applications

PDF

A flow-through membrane reactor for destruction of a chemical warfare simulant

PDF

Fabrication, deposition, and characterization of size-selected metal nanoclusters with a magnetron sputtering gas aggregation source

PDF

Synthesis of high-quality nanoparticles using microfluidic platforms

Asset Metadata

Creator Welchert, Nicholas Alexander (author)

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

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Chemical Engineering

Degree Conferral Date 2023-05

Publication Date 01/30/2023

Defense Date 11/28/2022

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

Tag chemical vapor deposition,OAI-PMH Harvest,polymers,thin film polymers,thin films

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Gupta, Malancha (committee chair), Nakano, Aiichiro (committee member), Tsotsis, Theodore (committee member)

Creator Email nwelchert@arizona.edu,welchert@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC112723824

Unique identifier UC112723824

Identifier etd-WelchertNi-11450.pdf (filename)

Legacy Identifier etd-WelchertNi-11450

Document Type Dissertation

Format theses (aat)

Rights Welchert, Nicholas Alexander

Internet Media Type application/pdf

Type texts

Source 20230201-usctheses-batch-1005 (batch), 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 author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

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

Repository Email cisadmin@lib.usc.edu