Photoinduced Wetting Kinetics of Water on
Immersed Nanoporous Titania Surfaces with
Application to Oil-Water Separation
ARCHIVES
MASSACHU
by
OCT 012015
Divya Panchanathan
LIBRARIES
Submitted to the Department of Mechanical Engineering
in partial fulfillment of the requirements for the degree of
Master of Science
at the
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
September 2015
@ Massachusetts Institute of Technology 2015. All rights reserved.
Signature redacted
Author..............
Department of Mechanical Engineering
, . 2015
Signatu re redacted
Certified by.............
School of Engineerin
.
Certified by...
Gareth "..IcKinley
Professor of Teaching Innovation
Tiesis Supervisor
Signature redacted
....
I
KAV)a1.Varanasi
Associate Professor of Mechanical Engineering
7')
Accepted by ............
S INSTITUTE
OF TE-CHNOLGY
,
TyaieSugisor
Signature redacted
David E. Hardt
Chairman, Department Committee on Graduate Theses
.
.................... .......
Photoinduced Wetting Kinetics of Water on Immersed
Nanoporous Titania Surfaces with Application to Oil-Water
Separation
by
Divya Panchanathan
Submitted to the Department of Mechanical Engineering
on Aug 6, 2015, in partial fulfillment of the
requirements for the degree of
Master of Science
Abstract
We study the self-cleaning ability of titania surfaces in oil-water environments for
fouling recovery in oil-water separation applications. A number of previous studies
have shown that meshes and porous structures can be used to separate oil/water mixtures through careful control of surface energy and preferential wettability, however
these structured surfaces are prone to fouling by oil and dirt. The photocatalytic and
hydrophilic nature of titania coatings can be exploited to ensure preferential wetting
of water over oil under ultraviolet (UV) irradiation and this provides a mechanism
for recovery of fouling. Titania nanoporous surfaces were prepared by depositing
TiO 2 nanoparticles onto flat substrates using Layer-by-Layer (LBL) assembly, and
were then impregnated with oil to simulate typical fouling conditions experienced
in oil-water separation applications. The resulting hydrophobic surfaces were irradiated with UV light in an oil-water environment to photocatalytically decompose the
organic pollutants and restore hydrophilicity. The kinetics of this conversion from hydrophobicity to hydrophilicity were studied in situ under various UV intensities using
goniometric measurements and a simple adsorption-photocatalysis model is proposed
to describe the observed data.
Thesis Supervisor: Gareth H. McKinley
Title: School of Engineering Professor of Teaching Innovation
Thesis Supervisor: Kripa K. Varanasi
Title: Professor of Mechanical Engineering
3
4
Acknowledgments
I would like to thank my two advisors Prof.
Gareth McKinley and Prof.
Kripa
Varanasi who helped me shape this work and taught me how I should go about
thinking and solving a research problem. I must admit that I couldn't have reached
this goal without the constant help and support of Dr. Gibum Kwon who taught me
how to overcome obstacles in experiments and how I can best present my research.
I would like to thank my sponsor KFUPM for funding this work and my collaborators Prof.Mohammed Gondal, Subkhi Sadullah and Talal Qahtan for their valuable
insights and feedback throughout the process. I would like to mention Kishor Nayar
from whom I learnt how I could be proactive, committed and organized in my research
while working with him on another project.
I express my gratitude to the members of both the NNF and Varanasi group for
providing me with a wonderful environment to conduct my research. Their constant
help and feedback was important during every step of this work. A special mention
to Bavand Keshavarz and Aditya Jaishankar for helping me cope up with stressful
situations in research and shaping my thinking about life in general.
I must thank my friends from both within MIT and outside who offered their
moral support and helped me feel at home in this new country. These include friends
from my high school, undergraduate college, the a cappella group MIT Ohms, roommates and the general Indian community here.
They helped me find my balance
between work and social life which was important for my overall self-development.
In particular, I would like to mention Vaibhav Unhelkar who encouraged me to stay
focused on whatever I did, taught me to love my work and who was a great friend in
general. My parents and my brother have been supportive to me during every step
of my life and this one is no exception. I am so glad that I decided to pursue my
graduate studies at MIT and I know that this place and people will never stop to
inspire me in every way possible.
5
Contents
Need for Oil-Water Separation . . . . . . . . . . . . . . . . . . . . .
18
1.2
Need for Membrane Separation
. . . . . . . . . . . . . . . . . . . .
20
1.3
Need for Fouling Remediation . . . . . . . . .. . . . . . . . . . . . .
23
.
.
.
1.1
Use of Titanium dioxide
27
Basics of Wetting . . . . . . . . . . . . . . .
29
2.2
Hydrophilicity and Underwater Oleophobicity
31
2.3
Photocatalytic Property of Titania . . . . .
34
2.4
Titania and Hydrophilicity . . . . . . . . . .
38
2.5
Study of In-situ wetting on titania . . . . . .
42
.
.
2.1
.
2
17
Introduction
.
1
47
3 Experiment Section
4
Fabrication of Titania Surfaces
. . . .
47
.
3.1
3.1.1
Preparation of nanoporous titania films
48
3.1.2
Preparation of flat titania films
51
3.2. Characterization of Fabricated Surfaces
51
Nanoporous TiO 2 Surface
. . .
52
3.2.2
Flat TiO 2 Surface . . . . . . . .
52
.
.
3.2.1
Contamination of TiO 2 coating
3.4
Contact Angle and In-situ UV Exposure
.
. . . .
. . . . . . . . . . . . . . .
.
3.3
54
55
61
Results and Discussion
7
5
4.1
Wettability change under oil contamination . . . . . . . . . . . . . . .
63
4.2
Wetting under UV Irradiation . . . . . . . . . . . . . . . . . . . . . .
63
4.3
Modeling Kinetics of Wetting of Water under Oil
. . . . . . . . . . .
66
4.4
A pplications . . . . . . .... . . . . . . . . . . . . . . . . . . . . . . .
72
Conclusions and Future Work
75
A Tables
77
B Figures
79
8
List of Figures
1-1
Average annual releases(1990-1999) of petroleum hydrocarbons (in kilotonnes) from natural seeps and activities associated with crude oil or
refined products into the marine environment [7] . . . . . . . . . . . .
18
1-2
Average anuual releases (1990-1999) of petroleum by source (in kilotonnes)[7
1-3
Comparisons of current treatment technologies for produced water
22
treatment in oil and gas industry [221 . . . . . . . . . . . . . . . . . .
1-4
Regimes of operation for various membrane filtration processes (Amer. . . . . . . . . . . . .
24
2-1
A schematic of various applications of TiO 2 by Nakata et al. [67] . . .
28
2-2
Balance of interfacial forces at a three phase contact line and definition
ican Water Works Association (AWWA)) [191
of contact angle 0..
2-3
.......
......
...................
.30
The different water and oil wettabilities achievable by modifying surface energy of solid . . . . . . . . . . . . . . . . . . . . . . . . . . . .
32
2-4
Different regimes of wetting 142] . . . . . . . . . . . . . . . . . . . . .
33
2-5
Prediction of underwater oil contact angles assuming interfacial tensions of water-air, oil-air and water-oil to be 72 mN/m, 27 mN/m and
51 mN/m respectively
2-6
. . . . . . . . . . . . . . . . . . . . . . . . . .
33
Oil-water separation using titania-coated meshes exploiting underwater
oleophobicity; (a) TiO 2 nanotube arrays in Ti mesh [70], (b) SWCNT/TiO
2
composite film [291, (c) LBL (silicate/TiO 2 ) 2o coated SS mesh [124], (d)
uncoated mesh (left) and TiO 2 nanoparticles spray coated on SS mesh
(right) [311 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9
34
19
2-7
Possible pathways for the e--h+ pairs on the TiO 2 surface and bulk
after light excitation 128] . . . . . . . . . . . . . . . . . . . . . . . . .
36
2-8
TiO 2 Crystal Structures 1531 . . . . . . . . . . . . . . . . . . . . . . .
37
2-9
Photoinduced hydrophilicity on titania under UV irradiation and recovery under dark conditions; (a) Water spreads as a thin film on the
TiO 2 surface under UV, maintaining transparency 11051, (b) Water
contact angle evolution with time on TiO 2 surface upon exposure to
UV (top) and storage in the dark (bottom) in air (dashed line) and in
oxygen atmosphere (solid line) [1061 . . . . . . . . . . ..
. . . . . . .
39
2-10 Mechanism of (a) dissociative water adsorption on TiO 2 [110] surface without defects (top) and with oxygen vacancies(bottom) [46],
(b) molecular water adsorption on TiO 2 surface (plausible structure
based on NMR study) [72] . . . . . . . . . . . . . . . . . . . . . . . .
41
2-11 Evolution of water contact angle in air with time upon UV irradiation
and storage in the dark; (a) UV light 0.1-50 mW/cm 2 (left) and dark
storage (right) [95], (b) UV light 40 mW/cm 2 on [110] face (solid line)
and 10011 face (dashed line) 11091, (c) UV light 2 mW/cm 2 (left) and
dark storage (right) [83], (d) UV light 2.0 mW/cm 2 [65], (e) cos 0 under
sunlight 10 mW/cm 2 (left) and dark storage (right) [82] . . . . . . . .
43
2-12 (a) Reversing UV induced superhydrophilicity by ultrasonication [691
(b) UV induced superhydrophilicity on Titania surface under flowing
water conditions. 1711 (c) Repeatability of contamination-UV recovery
cycles on TiO 2 demonstrated using 0,,, and 6w,, [70]
. . . . . . . . .
44
2-13 Self cleaning of TiO 2 ; (a) Oil drop displaced by water under UV irradiation 127], (b) Integration of oil-water separation with UV flow-through
photocatalysis on TiO 2 meshes [52] . . . . . . . . . . . . . . . . . . .
45
2-14 Wetting kinetics on TiO 2 ; (a) Modeling using inverse of contact angle
as a function of time 1851, (b) Modeling using cosine of contact angle
as a function of time (left) and rate constants as a function of UV
intensity [88]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
10
45
3-1
Schematic representing the deposition of a single bilayer of (PAH/TiO 2)
during the Layer-by-layer(LBL) deposition process.
. . . . . . . . . .
48
3-2
(a)Thickness and (b)roughness of TiO 2 LBL film vs. number of bilayers 52
3-3
Micrograpghs of (a)LBL (25 BL) and (b)ALD (100 nm) TiO 2 films
3-4
Height profile obtained through AFM; (a) LBL TiO 2 (30 BL), (b) ALD
T iO 2 (100nm ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3-5
53
Water contact angle measured under oil (hexadecane) for ALD TiO 2
substrates immersed for different amounts of time . . . . . . . . . . .
3-6
53
55
Experimental setup schematic: Water in oil contact angle imaging is
performed on a goniometer using a quartz cell. Sample is placed inside
the quartz cell and irradiated with UV light using a 5 mm diameter
probe. ........
3-7
...................................
56
Experimental setup: The UV probe is inserted into the quartz cell to
irradiate the sample with UV light. This configuration is imaged using
a goniometer to obtain contact angle changes.
3-8
. . . . . . . . . . . . .
UV source calibration: absolute intensities at various shutter openings.
A linear fit was used to find intermediate intensities . . . . . . . . . .
3-9
57
58
UV attenuation coefficients of air, water and dodecane calculated by
fitting intensity data measured using a radiometer (inside a quartz cell)
at various distances from the source . . . . . . . . . . . . . . . . . . .
4-1
Water contact angle measured under oil (dodecane) on LBL TiO 2 substrates contaminated for different amounts of time . . . . . . . . . . .
4-2
64
Contact angle evolution with time for water drop immersed in dodecane
environment on calcined titania LBL sample (I = 160 mW/cm 2 ) . . .
4-4
64
Contact angle evolution with time for water drop immersed in dodecane
environment on titania LBL sample (I = 160 mW/cm 2 ) . . . . . . . .
4-3
58
65
Contact angle of water in oil environment against time of UV irradiation at various intensities . . . . . . . . . . . . . . . . . . . . . . . . .
11
65
4-5
Contact angle evolution with time for water drop immersed in dodecane
environment on flat ALD titania sample (I = 200 mW/cm 2 ) . . . . .
4-6
Evolution of water-in-oil contact angle with time of immersion in dodecane along with the fit . . . . . . . . . . . . . . . . . . . . . . . . .
4-7
71
Cosine of contact angle of water in oil environment against time of UV
irradiation at various intensities . . . . . . . . . . . . . . . . . . . . .
4-8
66
First order wetting rate constants (kp) vs. UV intensity.
1/2
72
power law
fit was applied to the data in accordance with photocatalysis studies
at high intensities.
4-9
. . . . . . . . . . . . . . . . . . . . . . . . . . . .
73
Contact angle evolution with time for oil drop immersed in water environment on calcined titania LBL sample(I = 240 mW/cm 2 ) . . . . .
73
4-10 Evolution of oil-in-water contact angle with time of UV irradiation(I = 240 mW/cm 2
in water environment . . . . . . . . . . . . . . . . . . . . . . . . . . .
74
B-i X-ray diffraction of the calcined LBL TiO 2 film on a glass slide. Anatase
phase was confirmed. . . . . . . . . . . . . . . . . . . . . . . . . . . .
B-2 Spectra of the Omnicure S2000 Lamp.
Brochure........
B-3 UV Pass Filter
79
Courtesy: Omnicure S2000
..................................
80
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
80
B-4 UV source calibration: Variation of absolute intensity at the immediate
output of the probe with respect to the percentage shutter opening
.
81
B-5 UV source calibration: Variation of the absolute intensity at a point
on the central axis with respect to percentage shutter opening (S.0) .
81
B-6 UV source calibration: Variation of the absolute intensity on the center
axis at various shutter openings . . . . . . . . . . . . . . . . . . . . .
82
B-7 UV source calibration: Radial variation of the absolute intensity at a
constant center distance of 4 mm from the optical filter . . . . . . . .
12
82
2
)
4-11 Oil beading up from a contaminated mesh under UV irradiation(I = 300 mW/cm
74
List of Tables
1.1
Particle size removal capabilities of different technologies [251 . . . . .
1.2
Compilation of different techniques used to achieve oil-water separation
using m embranes . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1
23
25
In-air contact angle measurements on the flat ALD and LBL titania
film s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
13
62
Abbreviations
Symbol
Meaning
OY
Young's contact angle
Neo
Interfacial tension of 'a' and '/' phases (a and
Oa,3
Contact angle of a drop of 'a' in an environment of '#'
foo
Interfacial area between phases 'a' and '0'
Sa,,
Spreading coefficient of phase 'a' on '/'
e-
Electron
h+
Hole
UV
Ultraviolet
LBL
Layer-by-layer
ALD
Atomic layer deposition
Ao,
Attenuation coefficient in phase a
r
Roughness
S
Uncontaminated titania surface
C
Contaminant
P
Photocatalysis products
cS
Concentration of uncontaminated sites
cc
Concentration of contaminated sites
fS
fc
Fraction of uncontaminated sites
kc
Rate constant for adsorption of contaminant
k-c
Rate constant for desorption of contaminant
kP
Rate constant for photocatalysis for contaminant
/
Fraction of contaminated sites
14
are defined in the subscripts section)
Subscript
Meaning
a
Air
S
Solid
1
Liquid
v
Vapor
w
Water
0
Oil
15
16
Chapter 1
Introduction
Production, transportation and utilization of petroleum and its products have led to
unwanted release of oil into the environment causing pollution of fresh and marine
waters. Although accidental oil spills are the ones which generate a lot of attention due
to immediate and concentrated impact on the environment, they account for only 10%
of all the oil that enters the ocean. The other sources are from motor oil run-offs from
non-tank vessels, storm water run-offs from paved urban areas, untreated industrial
waste streams and natural sources
[7]. Figure 1-1 shows the relative contribution of
each of these sources towards oil pollution and Figure 1-2 shows the various types
of sources in each of the categories. The major reasons for lack of control over oil
release are the lack of affordable and accessible technologies for treating effluents
to meet environmental standards.
This work is focused on designing coatings for
membranes with photocatalytic cleaning ability for oil-water separation.
This has
potential to reduce costs incurred from membrane cleaning and replacement due to
oil fouling.
This chapter introduces the need for self-cleaning membrane technologies and explains why titania is a suitable material for the same. Chapter 2 gives an overview of
titania properties and illustrates the need to study the photocatalytic wetting properties of titania in an oil-water environment. Experimental procedures are described
in detail in Chapter 3 and the results of this study are presented in Chapter 4. The
final chapter summarizes the conclusions of this study and presents opportunities for
17
WorldWide Marine Waters
North American Marine Waters
B
A
84
480
600
160
9.1
160
1
38
* Natural Seeps
* Extraction of Petroleum
OlVansportation of Petroleum
OConsumption of Petroleum
Figure 1-1: Average annual releases(1990-1999) of petroleum hydrocarbons (in kilotonnes) from natural seeps and activities associated with crude oil or refined products
into the marine environment
171
future work.
1.1
Need for Oil-Water Separation
The current world crude oil production is about 4,200 megatonne per year [11 and
it is predicted to be on the rise for the next two decades [2].
As oil supply and
demand continues to increase with time, the necessity for treatment of waste produced
during production, transportation and utilization of petroleum and its products is also
increasing. One of the major steps in the treatment of this wastewater is oil-water
separation. The following passages elaborate on the different sources of waste streams
and the need for oil-water separation in each of these.
'Produced water' is the largest wastewater stream in oil and gas production. This
refers to the mixture of water, residual oil and other dissolved components which come
out along with bulk oil stream during oil extraction. The produced water production
is estimated to be about 71 billion bbl per year from all over the world with the
U.S.A alone contributing 30% of the total 1201. The ratio of water to oil in produced
water increases with the age of the oil well and the current water-oil ratio hovers
18
Worldwide
North Americe
Natural Seeps
Best Est.
Regionsb
160
160
Min.
Max.
s0
240
150
120
6.1
93
260
37
130
72
2.4
0.2
15
traced
19
2.6
2.2
1.1
0.03
0.03
9.1
1.0
2000
1900
9
1.4
0.30
0.30
81
4.4
480
140
130
6.8
6000
5000
110
2300
1300
43
0.18
Transportatiom of Petroleum
Pipeline spills
Tank vessel spills
Operational discharges (cargo washings)
Coastal facility spills
Atmospheric deposition
9.1
1.9
5.3
usW
1.9
0.01
74
1.7
4.0
na
1.7
0.01
7.4
Constmption of Petroleum
Land-based (river and runoff)
Recreational marine vessel
Spills (non-tank vessela)
Operational discharges (vesselsa100 GT)
Operational discharges (vessels<100 GT)
Atmospheric deposition
Jettisoned aircraft fuel
54
54
5.6
1.2
0.10
0.12
21
1.5
1.7
4.0
na
1.7
250
2000
11
2.1
6.4
na
2.2
0.02
2.3
0.15
0.07
2.1
260
200
62
1.4
2.6
58
3.0
0.15
0.12
2.7
Total
Max.
20
0.29
0.38
19
3.0
0.16
0.12
2.7
54
5.6
0.91
0.10
0.12
21
1.5
600
Min.
38
0.86
1.3
36
Extractic of Petroleum
Platforns
Atmospheric deposition
Produced waters
83
Best Est.
0.45
3.7
12
100
36
4.9
0.4
18
nd
nde
7.1
270
ndf
52
7.5
6.5
90
nd
23
5.0
470
I
nd
8.8
810
nd
200
22
8300
ONumbers are reported to no more than two significant figures.
6"Regions" refers to 17 zones or regions of North American waters for which estimates were prepared. These are discussed later in this chapter.
washing is not allowed in U.S. waters, but is not restrided in international waters. Thus, it was assumed that this practice does not occur frequently
<Cargo
in U.S. waters (see Chapter 3 and Appendix E).
'Estimated loads of less than 10 tonnes per year reported as "trace."
eWorldwide populations of recreational vessels were not available (see Chapter 3 and Appendix F).
Ansufficient data were available to develop estimates for this class of vessels (see Chapter3 and Appendix E).
Figure 1-2: Average anuual releases (1990-1999)
kilotonnes) [71
19
of petroleum by source (in
around 2:1 to 3:1 worldwide. Most of the produced water is currently re-injected
back into underground formations, re-used for secondary recovery or discharged into
the environment. The general legislation for oil levels in discharged produced water is
around 40 ppm but the stringency is increasing over time with increase in technological
advances [39]. Thus, there is a lot of effort going into produced water purification to
re-use this water for domestic and agricultural purposes in areas with water scarcity
and to prevent environmental pollution by direct disposal of these waters.
Apart from produced water, there are oil spill accidents which cause huge release
of oil into the oceans during transportation of oil. Some of the major (greater than
5000 Tonnes) oil spills in the past 5 years include Gulf of Mexico(2010), Xingang Port
oil spill (Yellow Sea, China), ExxonMobil(Nigeria, 2010) and Bonga Field(Nigeria)
and most of the times, these spills take months to clean up. Currently used biological
and chemical mitigation measures have known to cause harm to the environment.
Hence there a need for more environment friendly measures to take care of accidents.
It is important to note that human activities are not the only causes of oil release
in the oceans [7]. Natural seeps of oil into the marine waters from highly pressurized
seafloor rock account for about 40 % of the total oil released into the ocean and
typically, there are natural evolved bacteria which eat away the hydrocarbons in the
environment. However in case of oil spills, the flora and fauna of the surroundings
are adversely affected because they are usually not adapted to digest the oil. Hence,
it is essential to find ways to clean man-made water pollution.
1.2
Need for Membrane Separation
Current oil-water separation technologies can be classified into three categories based
on the size and nature of oil content being treated. The separation technologies listed
below are focused on treating produced water but are not limited to it 122, 39, 201.
1. Primary treatment
Primary treatment aims at separation of bulk oil content from water by using
gravitational and centrifugal forces.
20
These include plate separators, hydro-
cyclones and centrifuges.
2. Secondary treatment
Secondary treatment aims at conditioning the produced water for re-injection,
overboard discharge or further complex treatment. These include gas flotation,
coalescence, disk stack centrifuges.
3. Tertiary treatment
Tertiary treatment is the final stage of treatment which is aimed at achieving
high quality separation and elimination of soluble organics as well. The type of
tertiary treatment selected depends on the type of wastewater and size of particles to be removed. This includes adsorption (modified zeolites, activated carbon, activated alumina, organoclays), media filtration (nutshell), macro-porous
polymer extraction technology, membrane filtration, thermal oxidation, chemical oxidation, biological aerated filters etc.
A combination of the above methods can be used to create more efficient systems
and save process costs. Figure 1-3 summarizes the advantages and limitations of the
various technologies available for produced water treatment. While there are many
techniques available to remove bulk oil and micron-sized oil droplets from water, there
is still a need for cheap and efficient methods to remove dispersed oil (especially submicron sized drops) from water (See Table 1.1). Many of the chemical and biological
methods use toxic chemicals, need high installation space, cause secondary pollution
and incur high costs of treatment. Membrane filtration can eliminate many of these
disadvantages as discussed below.
Membrane filtration includes microfiltration (MF) which uses pore sizes ranging
from 0.1-3 pim and ultrafiltration (UF) which uses 0.01-0.1 Am. UF is a more effective
method for removing residual oil content when compared to MF as it can remove nanosized droplets as well (See Figure 1-4). Typical recovery using UF/MF ranges from
85-95 % and flux rates range from 2-100 gallons per day per square foot [34]. They
have lighter weight and smaller space requirements compared to other technologies.
Being a barrier technology, the treated water is less susceptible to contamination from
21
A1
11
ND~ t~
NDiIiii
CD
CD
'
IL'
1141 E11
IL
22
fill
It
I
rI
tit[II 111it1
I
ILI
i~
a t~it'~
16
1 1I
Table 1.1: Particle size removal capabilities of different technologies [25]
Technology
API gravity separator
Corrugated plate separator
Induced gas flotation without chemical
addition
Induced gas flotation with chemical addition
Hydrocyclone
Mesh coalescer
Media filter
Centrifuge
Membrane filter
Removal Capacity
by Particle Size
(Um)
150
40
25
3-5
10-15
5
5
2
0.01
upstream. Thus, membrane filtration is ideal for separating oil-water nanoemulsions
but one of the major problems faced during membrane separation is flux decline
due to fouling. The following section describes the need to design and study fouling
resistant membranes and fouling recovery processes.
1.3
Need for Fouling Remediation
Fouling is the adsorption or accumulation of unwanted components from the feed
onto the membrane surface or in the membrane pores, eventually decreasing flux at
constant pressure drop or increasing pressure drop at constant flux. Chemical or mechanical means can be used to decrease fouling. Dynamic shear enhanced membrane
filtration creates mechanical shear on the membranes by using a rotating disk or by
rotating or vibrating the membrane module [401. Although this is effective for reducing fouling, it can be highly energy consuming. Thus, using surface active membranes
which have preferential wettability towards either the oil or water phase is a more
preferred method.
Membranes can be of two types- water passing and oil-passing. By argument of
surface energies of surfaces, it is intuitive and easy to obtain organophilic materials
for oil-water separation in which case the oil will pass through the membrane and the
23
Ionic ROW
I
htbi t a
I
Ok~um
M, M'
IhUMvIhr
I
0ol,
Rww
OS
I
t'p
Nbco PutdoRKng
i
Twily
kw'-w
Ge~-,
*tw
p-
54
~wki
Pak
sa&i 1rdwt
*
su.
~i
~
,aPo"
ucs
Figure 1-4: Regimes of operation for various membrane filtration processes (American
Water Works Association (AWWA)) [19]
water will be rejected. Examples of these materials include kapok, carbon based materials, hydrophobic aerogels, PTFE coating/embedding, PDMS coating/embedding,
polydivinylbezene materials and cross-linked polymer gels 1117]. However, oleophilic
membranes can get easily fouled due to the accumulation of viscous oil in the pores and
so it is desirable to use hydrophilic materials instead. Some of the main chemistries
used for water-passing and oil-passing membranes have been summarized in Table 1.2.
Some of these are currently employed commercially while others are still in development. However, both of them are susceptible to organic fouling in a real application
[74, 5, 411.
Currently, backwashing is used in all membrane filtration applications to remove
foulants while non-backwashable foulants are removed by chemical cleaning
131.
In
most of the cases, the membrane has to also be removed periodically for extensive
cleaning. Although these methods are effective for recovering most of the flux, they
typically cost high energy, clean water and time. Therefore, it is preferable to have
some method of self-cleaning to remove foulants in-situ i.e. under oil-water environment without having to stop or reverse the flow feed.
In order to remove the organic foulants in-situ, it is necessary to use methods
24
Table 1.2: Compilation of different techniques used to achieve oil-water separation
using membranes
Principle
Materials
Oil/water
passing
1
Electric field
2
Hygro-responsive
chemistry
Oleophilic
chemistry
waterpassing
waterpassing
oilpassing
water-
4
pH responsive
FluorodecylPOSSa+xPDMSb[47
FluorodecylPOSS+xPEGDAc[44]
PTFE[24,
PDMS[51J,
PSd/PETe[108],nperfluoroeicosane[421,
SWCNT'network[90],
Fe 2 03/C
[121, Fluoropolymer[118, Lauric
acid [1151, F-PBZ9[89],OTSh[911
P2VPz-PDMS [123], PAAi[14
5
Hydrophilic chem-
Hydrogel
istry (Organic)
PMAPS[125], Hydrogel PEDOT-
3
PAMk[ 11 1 7 ],
water/oil
passing
passing
PSS [101] , PDA-NDM"[8
6
7
Hydrophilic chemistry (Inorganic)
Temperature
re-
TiO 2 [291121][711, ZnO [1031, Zeolite 11111, Cu(OH) 2 56][1211
PNIPAAmo [116] 1921
sponsive
8
9
Photo-responsive
Amphiphilic
waterpassing
oil/water
passing
Azobenzene,Spiropyran[107
Cellulose/Fluorinated
Silica
particles 1301,PVDFP [1001
oligomeric silsesquioxane
bcross-lined polydimethylsiloxane
ccross-linked poly(ethylene glycol) diacrylate
apolyhedral
dPolystyrene
epolyethylene terephthalate
fSingle walled Carbon Nanotube
9fluorinated polybenzoxazine
hoctadecyltrichlorosilane
tpoly(2-vinylpyridine)
Jpoly(acrylic acid)
kpolyacrylamide
'poly(3-(N-2- methacryloxyethyl-N,N-dimethyl) ammonatopropanesultone
poly(3,4-ethylenedioxythiophene)-poly(styrenesulfinate)
npolydopamine n-dodecyl mercaptan
Opoly(N -isopropylacrylamide)
Ppoly(vinylidene fluoride)
m
25
??
oil/water
passing
other than mechanical means to break or weaken the adhesion of organic foulants
on the membrane surface. Photocatalytic materials are good choice in this regard
as they can break down organic pollutants in the presence of light of appropriate
wavelengths and thus rid themselves from foulants. Titanium dioxide is one such
material which is cheaply available that can be employed for this purpose. This work
studies the recovery of Titanium dioxide nanoporous surfaces from fouling under oilwater environment in the presence of UV light. This can be applied to fouling recovery
in ceramic membranes that are made of or coated with titania and used for oil-water
separation.
26
Chapter 2
Use of Titanium dioxide
The element Titanium is the ninth most abundant element in the earth's crust (0.63%
by mass) [6] and the largest use of this element is in the form of its oxide, Titanium
dioxide (TiO 2 ). The commercially important ores of TiO 2 are rutile (TiO 2 ) and ilmenite (FeTiO 3 ).
Although the first laboratory synthesis of pure Titanium dioxide
from its ore was done during the early 1800s, it took almost a century to scale up the
process for mass production. During the early years of industrial scale production, Titanium dioxide was majorly used as a pigment in paints, plastics, tiles and paper[110]
for its brightness and opacity. Over the recent years, it has begun to be used extensively in the nanoparticle form in the food, cosmetics and drugs industry as well to
impart color and texture. The US Food and Drug Administration (FDA) allows upto
1% concentration of TiO 2 in food products if used as an inactive ingredient[43] but it
is still unknown if exposure to TiO 2 nanoparticles can have adverse long term effects
on human health and environment [1191.
The use of TiO 2 was restricted to pigments until the 1970s when the discovery of its
water-splitting property in the presence of UV light by Fujishima and Honda [26] gave
rise to new fields of research exploring the various photo-responsive applications[13
of TiO 2 including photo-watersplitting for hydrogen production, photocatalysis for
water purification, photoinduced-superhydrophilicity for self cleaning surfaces, photoelectrochromic devices, photo-voltaics for dye sensitized solar cells etc. Due to its high
chemical stability, cheap availability and non-toxicity, TiO 2 has a lot of potential for
27
Figure 2-1: A schematic of various applications of TiO 2 by Nakata et al. [67]
commercial applications [681 in both energy production and environment protection
as shown in Figure 2-1.
The major application of titania has been in photocatalytic air/water treatment
to remove organic fouling agents and microbial content [59, 73, 15]. Recently, TiO 2
has been gaining attention in the fields of anti-fogging and oil-water separation using
surface modified membranes due a combination of its intrinsic and photo-induced hydrophilic properties. As mentioned in the previous chapter, one of the major problems
with current methods of membrane separation is membrane fouling. Owing to the
photocatalytic and hydrophilic nature of titania, it is possible to create self-cleaning
membranes for oil-water separation. This work is aimed at proving and characterizing
in-situ recovery of fouled titania surfaces under UV light in an oil-water environment.
Instead of using chemical methods to study recovery of surfaces from fouling,
water wettability was used as a probe to study contamination and self-cleaning. In
order to study wetting behavior of titania, it is important to understand some basic
terms, concepts and equations related general wetting phenomena and photocatalysis.
28
Sections 2.1 and
2.2 introduce the basic terms and equations related to wetting
physics. Sections
2.3 and 2.4 describe the mechanisms proposed in literature for
photocatalysis and hydrophilicity on TiO 2 respectively.
2.1
Basics of Wetting
Designing membranes for oil-water separation requires the understanding of basic
principles involved in wetting and interfacial phenomena. An interface is the boundary between two or more entities which could be solid, liquid or gas. When there is
just one uniform phase which extends to infinity there are no interfaces. The introduction of another phase/medium introduces interfaces immediately which are 2-D
in nature. When a third component is introduced and all three phases are in contact
with each other, it happens along a thin line called the contact line. Introducing
any more phases can cause the components to meet at the contact line or contact
points. Even though this boundary may look clear and sharp at the macroscale, the
two phases gradually pass into one another at the molecular level. Gibbs defined a
theoretical boundary called "dividing surface" for convenience and simplicity while
mathematically modeling these systems [37].
Surface tension is a force arising from the effective summation of short-range attractive forces exerted on a molecule by the neighboring molecules. When a molecule
is brought from the bulk to the surface these forces are not balanced on all sides
giving rise to an excess energy which is called surface energy.
A simple mathematical model for a three phase contact line was proposed by
Thomas Young in 1890 and since then this has been the basis for all studies in this
field. By balancing the interfacial surface tension forces at the contact line as shown
in Figure 2-2, we can obtain the Young- Dupre equation [1201.
cos OY = Ys'v - NY'
(2.1)
'Ylv
where 7ys,v, ys,l and
YI,,v
represent the interfacial tensions of the solid-vapor, solid-
29
il,
y
s,v
Figure 2-2: Balance of interfacial forces at a three phase contact line and definition
of contact angle Q,
liquid and liquid-vapor interfaces respectively. Oy refers to the angle formed by the
liquid-vapor and liquid-solid surface tension lines and it is called Young's contact angle. Since the surface tension force (Force/Length) can also be interpreted as surface
energy per unit area (Force x Length/Area), we can obtain the above equation from
energy arguments too. However, this equation is defined for ideal surfaces which are
physically and chemically homogeneous. In reality, there are microscopic asperities
even on surfaces which look smooth on the macroscale and there are small domain of
chemical discontinuities depending the surface preparation. In order to account for
this, Wenzel [112] and, Cassie and Baxter [10] came up with new models which accounted for topographical heterogeneity of surfaces and calculated the corresponding
apparent contact angles depending on the wetting regime.
Wenzel predicted that the contact angle is dependent on the roughness of a given
surface if the water drop impregnates the surface features. Roughness (r) is defined
as the total surface area per unit project area of surface. Hence r is always greater
than or equal to one.
cos , = r cos O,
(2.2)
Cassie however discovered that another state is possible where the drop could sit
suspended on the asperities without impregnating them. The energy balance for this
state gives the following relation.
cos Ob =
fs,i cos Oy - fia
30
(2.3)
where f8, and fi,a are the areas of the solid-liquid and liquid-air interfaces on an unit
projected area of solid.
2.2
Hydrophilicity and Underwater Oleophobicity
With definitions of contact angles in place, the different wettability regimes on a solid
surface can be studied using contact angle as a parameter. In general, a substrate is
said to be hydrophilic if it the contact angle of water on the substrate is 0 < 0 < 90'.
It is said to be superhydrophilic if the water drop spreads on the surface forming a
negligible contact angle (less than 100). Similarly, a substrate is said to be oleophobic
when the contact angle of oil is greater than 900. Another parameter used to study
wetting is the spreading coefficient 'S'.
5
5
w,a
where S,
,a
o'a
=
Ys'a
-
('Ys,w
=
'Ys,a
-
(7s,o
+
'Yw,a)
'Yo,a)
is the spreading coefficient of water on solid in air medium and So,a is the
spreading coefficient of oil on solid in air medium. When S > 0, the liquid wets the
surface completely and when S < 0, there is partial wetting or complete non-wetting
of the liquid on the solid surface.
Figure 2-3 shows how the wettability of a solid surface can be tuned by modifying its surface energy relative to those of oil and water.
solid
('s,a >> (-Ys,,
+
-Yw,a))
A high surface energy
would tend to be wet by both water and oil i.e, the
surface would be hydrophilic-oleophilic. This is generally observed for everyday materials like metals and glass. When the surface energy of a solid is extremely low
(7Ys,a
<< (7Ys,w +
'Yw,a)),
the surface would be hydrophobic-oleophobic in air. This
is generally difficult to achieve unless the surface is functionalized with Fluorine or
Chlorine groups like those in fluorodecyl POSS [61]. Similarly, an intermediate solid
surface energy can give rise to hydrophobic-oleophilic surfaces like Teflon and other
plastics. From the above description, it can be seen that there is no case where the
31
Ywa= 7 2 mN.m'
Yo,a= 2 5 mN.m-1
Oleophobic
Hydrophobic
Oleophillic
Hydrophobic
Oleophillic
Ys,a
Hydrophilic
Figure 2-3: The different water and oil wettabilities achievable by modifying surface
energy of solid
surface is hydrophilic-oleophobic in air. This case is difficult to achieve in general
unless the surface is functionalized in some way to enhance polar interactions. Introducing surface groups which reorient themselves to attract water molecules like PEG
groups[44] or cellulose[75] can create these kind of surfaces.
However, it is quite possible to achieve hydrophilic-oleophobic surfaces in an oilwater environment and this is termed as "Underwater oleophobicity". This phenomena was first reported by Jiang's and Bhushan's groups 142, 55] while studying oil
repellency of fish scales underwater. The following sequence of equations describe
how the underwater contact angle of oil can be derived in terms of the respective
contact angles in air and interfacial tensions.
cos Ow,a
-
Cos Oo,a
w,w
s,a -
-
7s,a
-
Yw,a
-
7sO
'Ys ,a
cos O0 ,
7s__e
-
s,o
N0,W
Cos 00,W
-
-7o,acOS
0
o,a -
Yw,a COS Ow,a
(2.4)
No,w
Figure 2-4 shows the various possible scenarios for the solid-water-oil interface and
Figure 2-5 shows the graphical representation of Equation 2.4 assuming interfacial
tensions of water-air, oil-air and water-oil to be 72 mN/m, 27 mN/m and 51 mN/m
respectively. The yellow boxed region represents the currently accessible wettability
32
Solid-air-water interface
Solid-water-oil interface
Hydrophilic
Oleophobic if bAcos% < NAcosOw
Oleophilic if -bAcos8o> WAcos Ow
Solid-air-water interface
Hydrophobic
Solid-air-oil interface
Solid-water-oil interface
Oleophobic if 7 < g
Oleophobic if bAcosdo > WAcos Ow
Oleophilic if ?tACO < 1ACOV
0
Oleqphilic if A > )to
(F&A < nw)
Oleophilic
-
Figure 2-4: Different regimes of wetting [42]
regimes as the surface energy of solids practically possible today cannot produce
contact angles beyond the box region for flat surfaces. The bottom-right quarter of
the box represents the region of underwater superoleophobicity.
180
180
160
150
140
120
120
100
90
80
60
00W
60
30
0
0
30
60
90
120
150
180
0
0o,a
Figure 2-5: Prediction of underwater oil contact angles assuming interfacial tensions
of water-air, oil-air and water-oil to be 72 mN/m, 27 mN/m and 51 mN/m respectively
Thus, even if a material is amphiphilic in air it can be oleophobic underwater.
A membrane coated with this kind of material can be used for oil-water separation
as demonstrated by a number of studies in the recent years - TiO 2 [29, 124, 71, 31,
52], ZnO [103], Zeolite [1111, Cu(OH) 2 [56, 121]. Figure 2-6 shows some of the oilwater separation demonstration using titania coatings.
33
In order to understand the
mechanism behind the hydrophilic nature of titania, is it essential to understand the
origin of its photocatalytic behavior as well.
(b)
(a)
Ir%
(d)
(c)
Figure 2-6: Oil-water separation using titania-coated meshes exploiting underwater
oleophobicity; (a) TiO 2 nanotube arrays in Ti mesh [70], (b) SWCNT/TiO 2 composite
film 129], (c) LBL (silicate/TiO 2 ) 2o coated SS mesh [1241, (d) uncoated mesh (left)
and TiO 2 nanoparticles spray coated on SS mesh (right) [311
2.3
Photocatalytic Property of Titania
The photocatalytic properties of TiO 2 arise from its semiconductor nature.
The
band gap of TiO 2 is about 3.2 eV and its excitement requires a light spectrum with
frequencies in the ultraviolet (UV) region. When UV irradiation is incident upon
TiO 2 , the electrons in the valence band get excited to the conduction band and
electron (e-) - hole (h+) pairs are created.
34
Photoexcitation: TiO 2 + hv
-
e-+ h+
These electron - hole pairs help with oxidation and reduction reactions on the surface
of TiO 2 leading to photocatalysis [15][451.
There are a number pathways that can be taken after the electron-hole pairs are
formed. Figure 2-7 is a schematic showing the reactions that take place on the surface
and bulk of TiO 2 in the presence of the appropriate exciting wavelength. The e - and
h+ diffuse through the material and get trapped at surface trapping sites.
Charge-carrier trapping of e-: eCB-* e-R
Charge-carrier trapping of h+: h+
-+ h+R
where CB,TR and VB represent conduction band, trapped and valence band electrons/holes respectively. Here, the e- can either react with acceptor molecules (path
d) or recombine with holes to get to the ground state (path a).
Electron-hole recombination: eTR+ h+R -4 eCB+ heat
Photoexcited e- scavenging:
(O2)ads
+ e- -
02
Similarly, the h+ can react with donor molecules (path c) or can recombine with
free trapped electrons (path a).
Sometimes, electron-hole recombination can take
place in the bulk too releasing heat in the process (path b).
Oxidation of hydroxyls: OH-+ h+
-
OH'
The hydroxyl radicals attack long chain organic molecules and break them into smaller
molecules using a series of free radical initiation, propagation and termination reactions. A number of intermediate species like oxy- and peroxy- radicals are formed
during the reaction which result in chain scission [80] of long chain organic molecules.
Photodegradation by OH* R-H + OH' 35
R'*+H 2 0
hv
+0
A
C':'-+'02
+0+
egr
Wh r + RH -+ R* + H+
+ H20 -+*O, + H+
I
(d)
02 -+ 02
eI + H2 02 --. 0, + OH-
h
+
D
eg +_R'+ H+ -+ RH
AS=
electron
a= hole
Figure 2-7: Possible pathways for the e--h+ pairs on the TiO
2
surface and bulk after
light excitation 1281
TiO 2 is called a catalyst because it provides a low energy pathway for reactions
to proceed efficiently but does not get consumed in the reaction.
Further, it is a
photocatalyst since it acts an catalyst only when photons of energy greater than or
equal to the band gap strike the material. One of the major problems with using TiO 2
extensively in applications is that its large band gap requires light of wavelengths 400
nm or lower for excitation which means that natural sunlight will not be self-sufficient
to have optimal efficiency (since UV constitutes only 7.6 % of the Solar Constant[35]).
This issue can be overcome by extending the excitation wavelength to visible light
range by various techniques like metal oxide doping and non-metal oxide doping 11021.
Another factor that determines the efficiency of the catalyst is the electron-hole
recombination rate (path a and b in Figure 2-7). Minimizing recombination of e -h+ pairs can ensure better use of absorbed energy and the crystalline structure of
TiO 2 plays an important role in this process. The most studied crystalline phases of
titania are rutile and anatase of which the former is more thermodynamically stable.
Figure 2-8 shows the structure of two major crystalline phases of TiO 2 . The band
gaps of anatase and rutile phases are 3.2 eV and 3.0 eV respectively [87] and it has
been reported that anatase form is more photocatalytically active compared to the
rutile form. A number of researchers have proposed reasons for the higher activity of
36
A
Rutile TiO, (110)
B
Rutile T102 (101)
110111011
C
1011
10111
110~
Anatase TiO,(110)
D
Anatase TiO,(101)
101011
11101
TV*
02-
Figure 2-8: TiO 2 Crystal Structures [531
anatase. Ahmed et al. [41 compared the activity of the single crystal anatase (101)
surface to rutile (001),(100) and (110) surfaces and concluded that anatase (101)
surface is the most photocatalytically active of them all. Since (101) is the most
thermodynamically stable facet of the anatase phase, he concluded that the higher
activity of anatase is due to the availability of high number of (101) faces. Other
studies [57, 231 on the synthesis of anatase phase with exposed high energy (001)
facets show that this facet can influence the photocatalytic properties more than the
(101) facet. Recently, Luttrell et al. [60] attributed the higher photocatalytic activity
of anatase TiO 2 films to charge carriers being excited at greater depths in the bulk.
Zhang et al. [1221 found that anatase's indirect band gap (rutile has direct band gap)
enhances the lifetime of the photogenerated e -- h+ pairs and the lighter effective mass
of electrons and holes facilitates faster migration of charge carriers to the surface trap
sites.
Apart from the crystalline phase, different structural designs of TiO 2 such as
spherical nanoparticles, cylindrical nanotubes and planar nanosheets can contribute
37
uniquely towards different applications in terms of surface area to volume ratio, fabrication ease, adhesion to substrate and structural strength [67]. As discussed in the
beginning of the chapter, titania also possesses superhydrophilic property under UV
irradiation which is of interest for oil-water separation.
The following section de-
scribes important qualitative and quantitative studies on superhydrophilicity of TiO
2
and the proposed mechanisms.
2.4
Titania and Hydrophilicity
Titanium dioxide converts from a hydrophilic to a superhydrophilic surface under UV
irradiation. This was reported first by Wang et al. [1051 and since then, this phenomena has been studied extensively by various groups with interest in anti-fogging and
self-cleaning applications [541. Figure 2-9a shows the induction of superhydrophilicity upon UV irradiation and the recovery under dark storage. Other inorganic oxides
such as Zinc oxide, Vanadium oxide, Tungsten oxide etc have also been found to show
switchable wettability towards water under UV irradiation [65]. The reason behind
this phenomena has been under debate for many years and most of the proposed theories are divided between surface hydroxyl group formation upon UV irradiation and
photocatalytic degradation of surface contaminating species. The various mechanisms
are elaborated belowRemoval of Organic Species
A hydrophilic surface can be rendered hydrophobic by a thin layer of carbonaceous
species on the surface. If this organic species is removed from the surface by some
cleansing method, then the surface can be restored to its natural hydrophilic state.
Therefore many studies on photo-induced wetting attributed the change in wettability
to the inherent photocatalytic nature of the material. The claim was that any organic
species on the surface of the substrate would be degraded in the presence of UV light
and would render the surface clean and hydrophilic 182, 62, 126]. Upon storage in
the dark or heat treatment 1931, the surface regains its hydrophobicity due to the
adsorption of organic species.
38
(a)
(b)
0
(a)
sI0
-
-..........
ci
20
0
6
15
10
20
25
30
Time of UV t1hu. (min.)
(b)
tM
i
K
a
0
(
-
*
-' s
r
tm (hour)
2
*4
Figure 2-9: Photoinduced hydrophilicity on titania under UV irradiation and recovery
under dark conditions; (a) Water spreads as a thin film on the TiO 2 surface under
UV, maintaining transparency [1051, (b) Water contact angle evolution with time on
TiO 2 surface upon exposure to UV (top) and storage in the dark (bottom) in air
(dashed line) and in oxygen atmosphere (solid line) [1061.
39
The counter argument for this was that certain inorganic oxides like SrTiO 3 which
exhibited photocatalytic activity did not show photoinduced hydrophilicity and vice
versa [65].
Also, careful decontamination of TiO 2 and ZnO surfaces with NaOH
solution did not induce superhydrophilicity [95]. These arguments led to the idea that
the mechanisms must be different for photocatalysis and photo-induced hydrophilicity.
Dissociative adsorption of water molecules
Highfield and Gratzel [38] observed that presence of water vapor delayed the
process of electron-hole recombination. Lo et al. [581 showed that the photogeneration
of Ti3 + species on TiO 2 was important for the chemisorption of 02, H 2 and H 2 0
molecules on the (100) surface. Wang et al. [104] tested the effects of UV irradiation
and heat treatment under vacuum and Argon environment to confirm that water
vapor played an important role in the hydrophilicity conversion. All of these studies
lead to the hypothesis of dissociative adsorption of water molecules (vapor or in
solution) on photogenerated oxygen vacancies on the surface of the TiO 2 upon UV
irradiation [95, 106, 96, 84]. It was also observed that upon storage in the dark or heat
treatment, the surface is restored to its not so hydrophilic state by adsorbing oxygen
molecules from atmosphere [114, 107, 9]. Wang et al. also proved that recovery in
the dark in enhanced by the presence of oxygen as shown in Figure 2-9b. Figure 210a illustrates the mechanism of dissociative adsorption of water molecules on TiO 2
surface with and without oxygen defects.
The following equations describe steps
leading to the creation of oxygen vacancies.
Trapping of e-: e-+Ti 4+
Creation of
02
-
e-+ h+
-+
Ti3
vacancies: 2h++ 0-
+
Photoexcitation: TiO 2 + hv
--
O2
Molecular adsorption of water molecules
Some studies have reported the growth of multilayers of water molecules on TiO 2
surface with UV irradiation by molecular adsorption of water molecules without
dissociation[72, 66] as shown in Figure 2-10b.
40
Their argument is that if water is
(a)
O(b)
a
H
H
,Q
Mer I
H
Ver{HiXj VI
/
H,
H
HH H
H
V H. JH
otQ,
H
Ck
Q
H HH H H
H
[1101 surface
without defects (top) and with oxygen vacancies(bottom) 1461, (b) molecular water
adsorption on TiO 2 surface (plausible structure based on NMR study) 1721
Figure 2-10: Mechanism of (a) dissociative water adsorption on TiO 2
dissociatively adsorbed, the resulting hydroxyl groups must not be easily removed
by oxygen exposure as seen in experiments. Henderson
1361
and Kurtz
[461
observed
both dissociative and molecular adsorption at different conditions. Recently, Lee et
al.
1501
studied temporal growth dynamics of the water layer on TiO 2 under visible,
near-infrared and UV light conditions and proved that the superwetting of titania
is due to water-layer growth via delocalised surface electrons and "water wets water
mechanism".
Other mechanisms
Some studies say that it is combination of both photocatalysis and water adsorption which leads to photo-induced wetting [33, 49j. Takeuchi [98J claims that the
surface tension of H 2 0 clusters reduces due to desorption of water molecules from
the TiO 2 surface by heating from the light source and causes further wetting. Rico
et al.
1811
say that different mechanisms come into play for photocatalysis (which
is dependent on crystalline phase) and wetting (which is independent of crystalline
41
phase) and that they cannot be directly correlated.
Although TiO 2 is amphiphilic (wettable by both oil and water) in air, it is underwater superoleophobic which has been exploited by many groups for achieving
oil-water separation. Although, there are a number of quantitative studies on wetting of titania in air which can be used for anti-fogging applications, there have been
no quantitative studies on wetting under oil environment which is the case in oil-water
separation applications. Thus, there is a need for studying in-situ wetting of titania
in liquid-liquid environments to understand self-cleaning capabilities of titania.
2.5
Study of In-situ wetting on titania
Early studies on TiO 2 were focused on proving that the water wettability on titania
changed with UV illumination [95, 65, 1061. Some specifically studied the wetting
properties of different faces of single crystalline titania such as (110), (100) etc under
UV irradiation [1091. Sahoo et al. studied wetting kinetics on visible light active N2
doped TiO 2 [821. All of these studies were performed in air medium and Figure 2-11
shows the data obtained under UV irradiation in some of these studies.
There are a few studies which measure contact angles after exposure to water or
organic contaminants. Sakai et al. 1831 observed that the fluid friction on a surface
coated with TiO 2 reduced under UV illumination (See Figure 2-11c). Nishimoto et
al. studied the contact angle change on a oleic acid contaminated surface of TiO
2
in the presence of UV irradiation [701 and flowing water [71]. Sakai et al. [861 and
Nishimoto [691 reported that the hydrophilicity induced by UV irradiation could be
reversed by ultrasonication in pure water. Some of these are shown in Figure 2-12.
Although these studies measure the change in wetting of water on TiO 2 flat surfaces,
it is unclear in most of the studies if the surface was irradiated in-situ with the water
drop already present or if the drop was placed on the surface after irradiation for the
contact angle measurement. The kinetics could be greatly different between the two
cases. Also, none of these studies were performed using an oil environment. In order
to have self-cleaning membranes for oil-water separation, it is important to achieve
42
(a)
(b)
1.20
120
so
-80-0-zoo
100
0
1000
to-(b)
I
V
0
00
40
20
0
40
20
60
60
so
10
-+-(110)fm.e
60-
0
(001)6w.e
-OO
0
4
6
9
Storage time (day)
2
Blumination dme (min.)
1o
12
0
s
200
1
IO
10 6
(e)
(d)
To~l
30
t
2t
1:
-
60
0
W1
0
20
Timi I.)ia
00
10so
rm"
0
C
Figure 2-11: Evolution of water contact angle in air with time upon UV irradiation
and storage in the dark; (a) UV light 0.1-50 mW/cm 2 (left) and dark storage (right)
[951, (b) UV light 40 mW/cm 2 on [1101 face (solid line) and 1001] face (dashed line)
[109], (c) UV light 2 mW/cm 2 (left) and dark storage (right) 1831, (d) UV light 2.0
mW/cm 2 [651, (e) cos 0 under sunlight 10 mW/cm 2 (left) and dark storage (right) [821
photocatalytic cleaning in an oil-water environment. An in-situ study would mimic
an actual application where both oil and water are present in the environment and
UV light is irradiated under liquid medium for self-cleaning through photocatalysis.
Fujishima et al. reported detachment of oil from titania surface under water in
the presence of UV as depicted in Figure2-13a 1271 and Gondal et al. 132] reported
the wetting of water on TiO 2 under oil. Li et al. used hierarchical TiO 2 nanotubes
for integration of oil-water separation, flow-through photocatalysis and self-cleaning
[52] (See Figure 2-13b. However, all of these studies did not rigorously quantify the
wetting behavior.
Besides quantification, very few studies have tried to model the wetting kinetics
of TiO 2 films under UV irradiation. Sakai et al. obtained data for wetting of water
on dipcoated anatase TiO 2 thin films in air and modeled the kinetics using inverse
of contact angle as their main parameter as shown in Figure 2-14a but this did not
have any physical significance[851. Later, Seki et al. remodeled the same data using
cosine of the contact angle as the main parameter as shown in Figure 2-14b 1881. This
43
. UV
Ul
(a)
1
+
40
Ultrasonic +
150
100
+ UV
00
t
il'
0
(b)
1
2
5
3
4
Cycles
(c)
6
-I
120
40
so
~40
0
1
2
3
4
40
5
1
2
1
0 1
4A3wa7
2 3 4
5
7
UV inadiation time/ h
Figure 2-12: (a) Reversing UV induced superhydrophilicity by ultrasonication 1691 (b)
UV induced superhydrophilicity on Titania surface under flowing water conditions.
1711 (c) Repeatability of contamination-UV recovery cycles on TiO 2 demonstrated
using Oo,w and Ow,a [701
44
(b)
(a)
4
&*
0
0
1
3
2
Cyac
4
Wi
Figure 2-13: Self cleaning of TiO 2 ; (a) Oil drop displaced by water under UV irradiation [271, (b) Integration of oil-water separation with UV flow-through photocatalysis
on TiO 2 meshes 152]
(b)
(a)
0.98
_73_
_ _A0.96
12
I
0.92
-
0.01
0.90
(b )
-
OM
-
_14
0.001
0.001
0.84,
jAM
0
o~o?
_L
3L30 4L 3 0IL103
1020
0~~
4 30
o 640 0.01 0.1 1201030
UV Irradiation TM /iin
~
i
i
~
.
-
0.10-
O08
0.1
0.94-
-
-
-
-
&1-
I0.0001
4Z0-
Tkne/mm
UV Intensity /mW cm
Figure 2-14: Wetting kinetics on TiO 2 ; (a) Modeling using inverse of contact angle
as a function of time [851, (b) Modeling using cosine of contact angle as a function of
time (left) and rate constants as a function of UV intensity [88]
had some physical significance as the interfacial energies can be directly related to
the cosine of contact angles using Young's relation. They also gave a model for the
dependence of the extracted rate constants as a function of UV intensity which was
based on electron-hole pair generation and recombination kinetics (See Figure 2-14b).
Later Sahoo et al. used the same model to in their work to explain water wetting on
visible light active N 2 doped titania films [821.
Thus there is a lack of literature which both characterize the in-situ photo-active
wetting on titania surfaces under oil-water environments and provide a model for the
same. Most oil-water membrane separation happens in submerged environments and
hence there is a need for studying wetting of titania in liquid-liquid environments
to understand in-situ self-cleaning capabilities of titania for fouling recovery. This
45
work studies the recovery kinetics of TiO 2 nanoporous surfaces from oil fouling under
oil-water environment in the presence of UV light but unlike other photocatalysis
studies, it uses contact angle as a probe to study this phenomena. A relation between
wetting and photocatalysis was developed by modeling the processes under dark and
various UV intensities.
While this work focuses on analyzing the kinetics of the
wetting, further studies need to be conducted using spectroscopic methods to clarify
the intrinsic mechanism of water wetting on titania surfaces.
46
Chapter 3
Experiment Section
In order to study the self-cleaning of titania, nanoporous films of titania were pre-
pared using Layer by layer (LBL) assembly of TiO 2 nanoparticles and photo-induced
wetting properties of these films were studied using continuous contact angle measurements on a goniometer in the presence of UV light. The following sections describe
the procedures for sample preparation, characterization, contamination, UV light illumination and contact angle measurement.
3.1
Fabrication of Titania Surfaces
A number methods can be used to create both flat and nanostructured films of TiO
2
[13] including Sol-gel method, Sol method, Hydrothermal method, Chemical Vapor
Deposition, Atomic Layer deposition, Electrodeposition, Direct oxidation method,
Magnetron sputtering [99], layer by layer assembly etc. Some of these methods can
be tuned to produce specific crystalline phase of TiO
2
by modifying the process con-
ditions while some of them produce amorphous phase of TiO
2
which then needs to
be annealed in appropriate conditions to get anatase or rutile phase.
In order to
analyze the effect of texture on the wetting of titania, flat and nanoporous films were
fabricated. The flat surfaces were fabricated using Atomic Layer Deposition (ALD)
technique and the nanoporous surfaces were fabricated using Layer-by-layer (LBL)
self assembly of TiO 2 nanoparticles.
47
2. Rinse
1. Cation
(PAH)
I
4. Rinse
3. Anion (TiO 2
)
N2dr$
dry
Figure 3-1: Schematic representing the deposition of a single bilayer of (PAH/TiO
during the Layer-by-layer(LBL) deposition process.
3.1.1
2
)
jt
Preparation of nanoporous titania films
Layer by layer (LBL) deposition method was used to assemble TiO 2 nanoparticles
onto flat substrates to generate nanoporous surfaces. LBL is a self-assembly technique which has been gaining attention over the last 1.5 decades [48, 11J. It is the
process of sequential adsorption of oppositely charged species onto a substrate by
electrostatic attraction in order to create uniformly deposited bilayers or multilayers
of these species. The adsorbed species could be polyelectrolytes or inorganic nanoparticles/nanosheets/nanotubes etc and the method of substrate solution-contact can be
dip coating or spray coating. Figure 3-1 shows the LBL process used for the deposition of titania nanoparticles. The detailed procedure followed for LBL assembly is
described belowSubstrate Preparation
The films were deposited on glass (VWR Vistavision Microscope Slides) or silicon
48
_-
liw
,
- , ,
-
"
141
substrates (2" Test Wafer, University WAFER 1). After achieving uniform deposition
on these substrates, the same deposition protocol was extended to metal or polymer substrates depending on the application. The glass/silicon substrates were first
washed with DI water (Millipore De-ionized water 18 MQ) and then sonicated in a
soap solution (Alconox Powder Detergent) for 1 hour to remove organics and other
contaminants on the surface. Then they were rinsed with DI water and sonicated
in a 10% by weight Sodium hydroxide solution for 1 hour to strip off the remaining
organic contaminants to expose the pristine surface. Finally, they were washed and
stored in DI water to prevent any contamination from air. The substrates were removed from the water just prior to LBL deposition and treated in oxygen plasma to
induce a mild negative charge on the surface. This negative charge is created because
the Si-O-H bonds on the surface are broken by the bombardment of ions during the
plasma treatment. Simultaneously, residual organic contaminants are also oxidized
and removed. Thus, clean charged substrates were prepared for attracting the first
layer of charged species in the LBL process.
Dispersion Preparation
The protocol followed for the fabrication of titania nanoparticle thin films was sim-
ilar to the one used by Cohen et al. in their work on LBL TiO 2 for anti-fingerprint
surfaces [171. In this work, TiO 2 nanoparticles served as the negative species while a
positive polyelectrolyte, Poly(allyamine Hydrochloride) (PAH) was used as the positive species.
1. Positive species: PAH was obtained in powder form (Sigma Aldrich, M.W.
58,000) and dissolved in DI water (Millipore De-ionized water 18 M Q) by using
a magnetic stirrer. The pH of the solution was adjusted to 7.5
M Sodium hydroxide solution (Sigma Aldrich 2 ).
0.1 using 0.1
It is essential to adjust the
pH of both the nanoparticle and polymer solutions so that there is sufficient
electrostatic attraction between the layers and optimal growth of the film.
'Product No.:155125, P Type, B doped, <100> Orientation, 1-100 Q-cm, Thickness 320-350 pm,
SSP Polish
2
Product No.:319481 FLUKA, Sodium hydroxide solution, volumetric, 0.1 M NaOH 0.1 N)
49
2. Negative species:
Anatase Titanium dioxide nanoparticles were preferred
for this work due to their higher photocatalytic activity. Commercially prepared concentrated Anatase TiO 2 nanoparticle dispersion was obtained (Svaya
Nanotechnologies 3 ) and diluted down to 0.03% by wt. in DI water (Millipore
De-ionized water 18 MQ) by using a magnetic stirrer. The average particle size
was quoted to be 20
5 nm. The pH of this dispersion was adjusted to 9.0
0.1 using 0.1 M Hydrochloric acid (Sigma Aldrich 4 ).
The pH adjustment
was important for the nanoparticle dispersion because the zeta potential of the
dispersion must be away from the iso-electric point to prevent agglomeration
of particles. The iso-electric point of TiO 2 nanoparticles is approximately between 5 and 7 [79][97] and this depends on the ionic strength of the solution and
also the particle size. Similarly, rinse water solutions were prepared for both
the cationic and anionic LBL solutions using DI water and their respective pH
values were adjusted to 7.5
0.2 and 9.0 t 0.2 respectively.
3. pH adjustment: pH of newly prepared solutions take time for reaching equilibrium and incorrect readings may be obtained if not measured in the right
way.
Polyelectrolyte and nanoparticles were first well-dissolved/dispersed in
aqueous medium before starting the pH adjustment. In order to adjust the pH,
acid/base was added while stirring the solution and it was allowed to mix well.
The pH reading was taken while stirring the solution at 200 rpm (substrate rotation speed used during the LBL dipping step) and it was taken multiple times
to ensure that the reading was stabilized and that there was no drift over time.
This process was repeated a couple of times (by adding acid or base) in order
to get the pH close to the required value. It was preferred that pH be adjusted
slowly to approach the required value rather than overshoot it because the addition of neutralizing acid/base causes formation of salts which could affect the
growth rate of the films.
3
Titanium dioxide nanoparticles (20
5 nm diameter), 10% by wt. aqueous solution
4 Product No.:94015 FLUKA, Hydrochloric acid solution, volumetric, 0.1 M HCl (0.1 N)
50
Dip coating
The substrates were dip-coated in the prepared LBL dispersions by using an automated dipcoating machine (Nanostrata Inc.). 8 beakers were used in total comprising
of 1 cationic solution, 1 anionic solution and 3 rinse solutions for each. The dipping
time in the cationic and anionic solutions were 10 min each with 30 s of drying between each layer. The dipping times in the rinse solutions were 2 min, 1 min and 1
min for each of the layers. The substrate was spun in all solutions and during drying
steps as well. The drying steps helped in removing residual water from the substrate
before beginning the deposition of every layer, thus avoiding cross-contamination.
Calcination
Since the LBL process used a polyelectrolyte for the intermediate layers between
TiO 2 layers, it was necessary to remove this organic content because the polyelectrolyte layers made the coating hydrophobic while the final application needed hydrophilic coatings. In order to remove the polymer layer, samples were calcined (heat
treated) in air at 400'C for 2 hours. Temperatures above 550'C cause the anatase
phase to convert into rutile phase and this was not preferred.
3.1.2
Preparation of flat titania films
In order to have a comparison of the wetting behavior of the nanoporous films to a flat
titania surface, Atomic Layer Deposition (ALD) was used to obtain smooth thin films
of controlled thickness. The films were fabricated by CambridgeNanoTech (Waltham,
MA 02453) according to the following specifications. The films were deposited on
clean glass (Microscope slides) at 250 'C with target thickness of 100 nm.
3.2
Characterization of Fabricated Surfaces
The fabricated films were characterized for their thickness, morphology and titania
phase using profilometry, Atomic Force Microscopy (AFM), Scanning Electron Microscopy (SEM) and X-ray Diffraction (XRD).
51
_194
-
- -
-
-
-
- --
-
-
-
(b)
(a)
-- 5
n
0
_
_
IM149 -
- -
20
A
-
250:
225
200
E
C175
n 150
a. 125
-Y 100 -0
'
75I25
-
<15-
A
C
-
10
A
o
A
10
20
30
40
50
0
60
No. of Bilayers (PAH/TiO 2)
10
20
30
40
50
60
No. of Bilayers (PAH/TiO 2
)
--
Figure 3-2: (a)Thickness and (b)roughness of TiO 2 LBL film vs. number of bilayers
3.2.1
Nanoporous TiO 2 Surface
LBL process can be tuned to achieve a specific thickness of coating based on the
number of bilayers deposited. This film growth characteristics is shown in Figure 3-2
and it was observed that the growth of film is approximately linear with respect to
increase in the number of bilayers. It was independently verified that the removal of
polymer from the film by calcination affects the film thickness by less than 10 nm
for the range of thicknesses studied. The crystalline phase of the TiO 2 nanoparticles
was anatase and remained so even after calcination at 400 'C as verified by XRD
analysis shown in Figure B-1. Figure 3-3a shows both the top and cross-sectional
views of the film which illustrate the nano-porosity of these films.Figure 3-4a shows
the surface morphology of the nanoporous samples obtained through AFM and the
surface roughness was found to be 15.1 pm. Samples with 30 bilayers of TiO 2 /PAH
were prepared and calcined for experiments with UV light. The average thickness of
these calcined films was approximately 120 nm.
3.2.2
Flat TiO 2 Surface
The thickness of the flat ALD film was interpreted from the SEM. It was found to be
approximately 100 nm. The micrographs of the sample in Figure 3-3b show that the
deposition is uniform but has some non-uniform domains formed during the atomic
52
-
- - -
- -
---
--
depositon process. Figure 3-4b shows the surface morphology of the uncalcined flat
ALD samples obtained through AFM and the surface roughness is 7.6 Am. Although
the roughness of the flat films were lower than that of nanoporous films, there were
tiny domains on the flat surface which could act as pinning sites. Hence the wetting
behavior of these films showed multiple pinning-depinning transitions (compared to
smooth wetting on nanoporous films) which were unrepeatable. The films were also
found to be amorphous in nature and conversion to anatase phase by annealing was
not repeatable. Therefore, these films were finally not used for quantitative studies.
(b)
(a)
Top view 3:Top
viewoB)
films
2
Figure 3-3: Micrograpghs of (a)LBL (25 BL) and (b)ALD (100 nm) Ti02 films
(a)
0
(b)
200.0 nm
Height
0
1 pm!
50.0 nm
Height
1 pm
Figure 3-4: Height profile obtained through AFM; (a) LBL TiO 2 (30 BL), (b) ALD
TiO 2 (100nm)
53
3.3
Contamination of TiO 2 coating
In order to observe the underwater/underoil self-cleaning effect of titania, it was necessary to contaminate the surface before the UV experiment. Some of the essential
properties of the contaminant to conduct this study were immicibility with water and
oil, stability under UV irradiation in the absence of titania, reproducibility of coating
across samples, hydrophobic nature and ability of be degraded by titania in the presence of UV. Silanes were initially thought be ideal for this study. However, it was seen
that although they satisfied most of the conditions they were difficult/inconsistent to
break down under UV light in the presence of titania. This maybe due to the fact
that silanes repelled water strongly and prevented it from reaching the nanoparticle
surface [321. Since photocatalysis requires the creation of hydroxyl species in-situ, the
blocking of access to water prevented/slowed down the reaction process. Hence these
were found to be ineffective for the study of wetting dynamics. On the other hand,
pure alkane contamination allowed for partial wetting of the water on the surface and
hence allowed for immediate reaction on the surface. Dodecane (anhydrous >99%,
Sigma Aldrich) was chosen as the contaminant oil for these experiments.
The calcined samples were first washed in acetone, IPA/ethanol and dried in air.
The samples were immersed in the model contaminant oil for 18-24 hours before water
contact and UV exposure. Some of these substrates were sampled out at different
times and water droplets of constant volume
3 pl were inserted on these surfaces
under oil (dodecane) to observed the advancing contact angle. The advancing angles
were observed to increase with the time of contamination and it is explained in detail
in the next chapter.
The time of immersion was chosen to be 18-24 hours instead of greater than 24
hours because water drop starts rolling on a completely repelling surface and it was
difficult to have control for contact angle studies. Also, a completely repelled water
droplet did not wet the surface immediately after UV irradiation as the photocatalytic
process is inhibited by the covering oil layer just as in silane contamination. Thus,
in order to ensure that the drop can be pinned to the surface, the time of immersion
54
150
A adv
w'o
grec
120!
90
90-
AA
O
A
AA
AAA
A
A
60
30
0
30
60
90
120
150
timmersion (min)
Figure 3-5: Water contact angle measured under oil (hexadecane) for ALD TiO
substrates immersed for different amounts of time
2
was kept under 24 hours. A similar experiment was done for the flat amorphous
ALD surfaces using hexadecane as the contaminant oil and it was observed that the
advancing and receding contact angles saturated at 1180 and 40' respectively (See
Figure3-5). The existence of a finite hysteresis even after long exposure explains why
the receding contact angles were zero for the nanoporous surface.
3.4
Contact Angle and In-situ UV Exposure
In-situ measurements of liquid-liquid contact angle were performed using a quartz
cell (1"x2"x2") on a goniometer stage (ram6-hart model 590-Fl). High power density
UV irradiation was provided through a fiber optic cable from a Mercury Vapor Short
Arc UV lamp source (Omnicure S2000). Figure 3-6 shows a schematic of the setup
where the sample is placed in the quartz cell and UV light is irradiated from the top.
It was chosen to irradiate the surface from the top in order to eliminate intensity
losses in the light path from the source to the target surface. A small tubular cap was
fabricated to prevent the liquid from touching the fiber optic cable wires. Since the
light spectrum of the lamp source had some visible light component in the 400-450 nm
55
Quart ceflUV
source
Peak: 365
5 nm
Power: 95 mW @ 365 rn
Light source
Camera
Figure 3-6: Experimental setup schematic: Water in oil contact angle imaging is
performed on a goniometer using a quartz cell. Sample is placed inside the quartz
cell and irradiated with UV light using a 5 mm diameter probe.
range (See Figure B-2), it interfered with the silhouette imaging on the goniometer
which depended on strong back-lighting for clear resolution of droplet boundaries. In
order to circumvent this problem, a UV Pass filter (Edmund Optics 46-081
-
Filter
Band U-360 12 MM Dia) was used at the outlet of the fiber optic cable to eliminate all
wavelengths of light above 400 nm (See Figure B-3) which allowed for clear imaging.
Figure 3-7 shows a picture of the setup where the UV light is being irradiated on a
sample.
The distance between the UV filter and drop was 3 mm for all experiments. In
order to know the intensity of UV at the surface of TiO 2 , the source was calibrated in
air environment (with the filter). Figure 3-8 shows the variation of absolute intensity
of UV light in air. The intensities were found in the liquid environments by using
attenuation coefficients of oil and water. The attenuation coefficients (A) were found
using a simple characterization experiment for air, water and oil. Figure 3-9 shows the
decay of intensities under air, water and oil environments from which the attenuation
coefficients were extracted. The attenuation coefficients of UV in water and dodecane
56
Backlight
Sample
UV probe
Camera
Figure 3-7: Experimental setup: The UV probe is inserted into the quartz cell to
irradiate the sample with UV light. This configuration is imaged using a goniometer
to obtain contact angle changes.
were found to be 1.24 x 102 m- 1 and 1.06 x 102 m
1
respectively. This implies that 28
% of UV is transmitted through 3 mm of water or oil to reach the substrate. This is
much higher than the literature value (A,
= 2 x 10-2 m-'[18]) probably because the
light beam is not exactly parallel (See Figure B-7). Here, the divergence of the beam
is taken into account in the attenuation coefficient. Also, it seems like the attenuation
coefficient of air is much larger. This maybe due to Fresnel losses from the visible
light filter. Thus the intensity at the substrate for the water drop in oil experiments
were calculated by the following formula.
Isurface
where 1 = 3 mm, A,
=
Ipeakeo
(3.1)
1.06 x 10 2 m- 1 and Ieak is calculated from Figure 3-8.
In UV irradiation experiments, contact angle was recorded for a few seconds before
UV irradiation was started in order to ensure that the drop was stable prior to UV
57
800
*
700
Data
IP/k= 11-1 PShutterOpen
MW/CM 2
600
500
0/
a'
I>'400
-
300
1200
S
100
10 20 30 40 50 60
Percentage Shutter Opening (%)
70
Figure 3-8: UV source calibration: absolute intensities at various shutter openings.
A linear fit was used to find intermediate intensities
0 Air
1
o Water
0
Dodecane
-
0.8
0.8
0.7
-
.
00
0.6
00-
0.5
0.4
0
0.3
00o
0.2
0.1
0
1
2
4
5
6
3
Axial distance (mm)
7
8
Figure 3-9: UV attenuation coefficients of air, water and dodecane calculated by
fitting intensity data measured using a radiometer (inside a quartz cell) at various
distances from the source
58
exposure. The measurement was stopped after the contact angle drop was less than
0.05' s-. 6 different intensities were used for the experiments.
All the fits were
performed using MATLAB. The outcome of these experiments is described in the
next section and a simple model has been proposed to explain the observations.
59
60
Chapter 4
Results and Discussion
Titania is intrinsically hydrophilic and once the surface is wet by water, it is difficult to
displace the water with oil which is why titania is oleophobic underwater. However,
when the surface is exposed to some contamination by hydrocarbons, oil tends to
stick to the surface of titania even in the presence of water and this leads to the
hydrophilic surface becoming hydrophobic with time. This is where photocatalytic
cleaning becomes important to recover the surface from contamination.
Since the
material is intrinsically hydrophilic, the contact angle of water is an indication of how
clean the surface is. If the surface is pristine, the water would completely spread on
the surface in the presence of oil but if there are contaminants the water would tend to
bead up. The degree of beading up indicates the level of contamination. This study
examines the evolution of contact angles of water on contaminated TiO 2 surfaces in
oil environment under UV irradiation.
As a base experiment, contact angles were measured for both flat and nanoporous
surfaces in air environment. As seen in Table 4.1, the contact angle of water and oil
on the LBL surfaces is much lower than the same on flat TiO 2 surfaces due to the
porous nature of the film.
61
Table 4.1: In-air contact angle measurements on the flat ALD and LBL titania films
TiO 2 Surfaces
ia
6a[
LBL TiO 2 (Calcined)
0
0
0
0
Flat ALD Amorphous (Vacuum stored)
57
3
4
5
wa
Q a
There were three reasons for performing 'water drop in oil' experiments instead
of 'water drop in air' experiments.
" Contaminant: The oil itself acts as like a contaminant and its removal from
the surface can be studied.
" In-situ cleaning: For application to oil-water separation, it would be ideal to
have in-situ self-cleaning in an water-oil environment. Hence it is more useful
to study kinetics of wetting on these surfaces under water-oil environment.
" Uncertainties in air environment: Water in air experiments are subject to
the uncertainties which arise due to evaporation of water drop into the environment. Unless the humidity of the chamber is well -controlled these experiments
are difficult to perform especially with a heat generating UV source. Also if an
oil (like hexadecane/dodecane) is used as the contaminant, it tends to cloak the
water drop in air.
Therefore, in order to study the self-cleaning properties of nanoporous titania
films, LBL TiO 2 samples were contaminated with dodecane and irradiated with UV
light in the presence of a water drop under oil environment to observe the fouling recovery of the surface. In order to measure the kinetics of wetting under contamination
and photocatalysis, the contact angle of water drop under oil (0,,o) was monitored
both under dark condition (UV absent) and in the presence of UV light on various
prepared samples. Under dark condition, the surface was observed to become increasingly non-wetting towards water as the time of contamination increased while under
UV irradiation, the surface tended towards superhydrophilicity and oleophobicity.
62
_Ud"
4.1
IQ
__
-
4W
i
Wettability change under oil contamination
The advancing contact angle of water 0, on a clean TiO 2 surface was observed to
be almost zero even after immersion under oil due to the intrinsic hydrophilicity of
titania. But this angle was found to increase with increasing time of contamination
as seen in 4-1. The corresponding receding contact angles were found to be 00 for
all the times. It is interesting to note that the surface does not become repelling
to water immediately after immersion under oil but instead takes finite time for it
to change its wetting behavior. This means that there is some kinetics involved in
the contamination process. When the surface is completely covered by adsorbed oil
molecules, the contact angle reaches 180' i.e. water is completely repelled from the
surface. When an oil molecule is adsorbed onto the surface, it is difficult for water to
displace it even though titania is intrinsically hydrophilic.
Although it takes finite time for adsorption of oil onto the TiO 2 surface, this is
not the case for water adsorption on a clean TiO
2
surface. If a clean surface is first
immersed in water and an oil drop is inserted immediately over it, the surface is nonwetting towards oil because water adsorbs strongly onto the TiO 2 surface. This is
why titania meshes impregnated with water are suitable for oil rejection in oil-water
separation.
In these studies, the TiO 2 mesh is generally impregnated with water
before introducing the oil-water mixture. However, any kind of oil contact would
destroy the hydrophilic property of the mesh.
4.2
Wetting under UV Irradiation
Water was found to preferentially wet a TiO 2 surface under UV light even in the
presence of an oil background as seen in Figure 4-2. Figure 4-3 shows the evolution
of the water contact angle and drop width with time of UV irradiation for a single
intensity. The increasing width shows that the contact angle decrease is due to water
drop spreading and not evaporation. When the intensity of UV light was increased,
the rate of contact angle change also increased as seen in Figure 4-4. An interesting
63
180-'
O.-
150120
0
0%MO
90
-
30
30
0O
0
300
600
900
1200
1500
1800
ttcontamination~mn
(mi n)
Figure 4-1: Water contact angle measured under oil (dodecane) on LBL TiO 2 substrates contaminated for different amounts of time
observation here is that the timescale for contamination is much higher than the time
scale for fouling recovery. While it take many hours to completely contaminate a
surface, it takes only minutes to recover the hydrophilicity. This is useful in a real
application where we want to minimize the ratio of recovery time with respect to
fouling time.
The flat amorphous titania samples also showed similar behavior where the water
wet the surface preferentially over oil but it had a lot of pinning-depinning transitions
Water drop in
dodecane
t
=
0 min
tUV= 0.5 min
I min
2m
MM
3 min
5 min
7 min
F
.
mi
)
Figure 4-2: Contact angle evolution with time for water drop immersed in dodecane
environment on titania LBL sample (I = 160 mW/cm 2
64
,
.
.I
.
,
,
,
140- I
120-
000000 -4
0000
100-
0
00
80-
00
60-
0
0
0o
-3
20-
*
*
Conta,ct angle
Width
0
,
T
E
E
60
,
i
00"W%
-2o
o
40-
0-
ooooooooooooooooo
0~o*
0
0
-5
I. 5-I-I
100
50
200
150
250
UV irradiation (S)
)
Figure 4-3: Contact angle evolution with time for water drop immersed in dodecane
160 mW/cm 2
environment on calcined titania LBL sample (I
I
I
9
120
I =40 mW/cm 2
0 I = 120 mW/cm 2
0
.0 I = 200 mW/cm 2
0.
.%
90
0
0
'-I-
1*.
60
0
p
p
p
0
200
400
UV
*
301
600
irradiation (S)
Figure 4-4: Contact angle of water in oil environment against time of UV irradiation
at various intensities
65
-3.5
140120
3
100
10
CZ
80
2.5
-
-'
60
40K
-
--
2
202
01
''1.5
0
200
400
600
800 1000 1200 1400
1
tIrradiation(S)
)
Figure 4-5: Contact angle evolution with time for water drop immersed in dodecane
environment on flat ALD titania sample (I = 200 mW/cm 2
(as shown in Figure 4-5)which made reproducibility difficult. Hence, these sample
were not used for quantitative analysis.
4.3
Modeling Kinetics of Wetting of Water under Oil
In order to model the kinetics, we need to understand the various physicochemical
processes taking place under UV light and dark (contamination) conditions. When
the titania sample is immersed under oil, although the oil spreads on the surface instantly, it does not adsorb on the surface immediately 1113]. There is an adsorptiondesorption equilibrium between the bulk oil phase and the surface which causes the
delay in adsorption.
If a water drop is inserted before complete surface coverage,
the water can partially wet the surface occupying the uncovered sites. The degree
of partial wetting is dependent on the surface coverage of oil which is a function of
time as seen in Figure 4-1. The reported 0,
angles are a measure of advancing
contact angle on the LBL TiO 2 surface while the receding angles are always zero due
66
to penetration of water drop into the texture. When the surface is contaminated long
enough (tcontamination > 24 hrs), the water drop is completely repelled from the surface
with an advancing contact angle of 1800 and zero. Therefore the change in wettability under dark conditions can be explained by change in surface chemistry through
adsorption of oil on the surface. In the presence of UV irradiation, it is observed
that water preferentially wets over oil. While the mechanism of water wetting over
oil could be photodegradation of adsorbed oil molecules or simply displacement of oil
molecules by water molecules in the nanoporous layer, the ultimate effect of these two
processes can be seen as creation of hydrophilic sites on a contaminated hydrophobic
surface. Photodegradation of dodecane on TiO 2 in aqueous medium has been studied
previously [76, 771. The conditions under which the 'water drop in oil' experiments
were performed were favorable for photocatalysis as well. Hence, the rates of wetting
under UV were characterized by kinetics of titania photocatalysis.
The Langmuir-Hinshelwood (LH) mechanism [16] has been used to model heterogeneous photocatalytic degradation of organic contaminants on TiO 2 in many studies
[63, 21, 64]. In a typical case, the reactant adsorbs onto the solid catalyst, the reaction
takes place on the surface of the catalyst (in the presence of light of appropriate wavelength for photocatalysis) and the products detach from the catalyst surface allowing
for new reactant to adsorb at the same site. The rates of degradation is therefore
determined by the intrinsic reactivity of the photoactivated surface, the concentration
of reacting species and the Langmuir adsorption-desorption equilibrium of the reactant on the catalyst. Since photocatalysis is expected to be the mechanism behind
the water wetting, the LH mechanism was used to model the processes taking place
in our experiments. Here, the important species in the reaction were recognized to
be the contaminant (C), substrate (S) and the degradation products (P). Equation
4.1 represents a general equation in heterogeneous photocatalysis.
S + C ke SC kp'
CS
k--
Csc
>S + P
(4.1)
CS
where kc, kc and kp are the adsorption, desorption and photocatalytic rate constants
67
in the reaction. kp is a function of light intensity while k, and k-c were assumed to be
independent of the same. c. and c,, are the concentration of free sites and contaminant
occupied sites on the substrate respectively.
The concentrations were calculated by assuming a Langmuir model for surface
coverage of oil molecules on the substrate. In this model, the surface is assumed
to be a two dimensional and comprising of a fixed number of sites, each of which
can hold at most one molecule of adsorbent (monolayer coverage). It also assume
that all sites are equivalent and that there are no interactions between the adsorbed
molecules. In our case, the symbols 'S' and 'C' represent the substrate TiO 2 and
the contaminant dodecane respectively. Therefore there are two types of sites on the
surface - clean site (S) and contaminated site (SC). Let the surface fraction of clean
sites and contaminated sites be f8 and
fS
f,,
+
respectively. Then,
fc
(4.2)
= 1
It has been reported in literature that n-alkane adsorption on gold surface follows
first order kinetics based on Langmuir adsorption model [113].
It has also been
reported that dodecane degradation by TiO 2 in UV in aqueous medium follows first
order reaction kinetics [94, 78] and that it is irreversible. Therefore, first order kinetics
was used to obtain the rate of change of concentration of surface species, 'S' and 'SC'
during adsorption and photocatalysis in the titania-oil-water system. Zeroth order
dependency was assumed with respect to the concentration of reactant bulk phase
'C' because the oil, dodecane was used in its pure form. Also, 'C' was held constant
in all experiments. Thus, the surface fraction of hydrophobic sites was obtained as a
function time of UV irradiation.
--
dt
+tcsc(t )
=kecs
+fsc(t )
=
ke fs
68
-
k~cc8 e - kcc(4.3)
--
~
e
-
kpfc(4.4)
Substituting Eqn. 4.2 and solving the first order differential equation,
f8 c(t)
=
k
kc+ k-c + k
-
kc
kc +k-c + kp
-
fS
-(k+k-c+kp)t
ec(O)
0
(4.5)
Since the surface species were not directly monitored in our experiments, we
needed to relate the wetting data to the surface composition. Although contact angle
was the measured parameter during the water drop-in-oil UV wetting experiments, it
did not directly relate to any physical phenomena. Its cosine however could be related
to the interfacial energies of the surface, liquid and the medium assuming topographical and chemical homogeneity during equilibrium and it is given by Equation 4.6.
This was also done in Seki's analysis[88] while characterizing UV induced wetting of
water on titania in air.
cos 0", 0 = 7S'O
-
7,W
(4.6)
7'I',o
For a composite surface, the effective contact angle was derived by Cassie using
surface energies of each component
[10].
Thus knowing the surface density of the
individual species on TiO 2 , we can find the effective contact angle on such a chemically
.
heterogeneous surface by using the Cassie-Baxter equation
n
cos
fi cos 0'
=
-'f
(4.7)
i=1
where
6 eff,fi
and 6 are the effective contact angle on composite surface, surface
fractions of component 'i' and contact angle on component 'i' respectively.
important to note that 0'
It is
is the effective contact angle of water under oil at each
component 'i'. In the TiO 2 -water-oil system, the hydrophilic sites are bare TiO 2 sites
(S) and the hydrophobic sites are TiO 2 covered by a monolayer of oil contaminant
(SC). The contact angles of water under oil on each of sites are assumed to be 00 and
180' respectively since the bare titania is superhydrophilic and oil covered titania is
completely water repelling.
69
-
+ f+ cos 0"
cos
=
f, cos 0
cos
=
-"ff
f, cos(00 ) + fc cos(180')
-
1 - 2f,,
cos O'f
-"f
(4.8)
(4.9)
(4.10)
Thus we now have a relation between the measured parameter, 6Oef and surface
species concentrations. Assuming quasistatic processes under contamination and UV
conditions, the cosine of the contact angle was obtained as function of time of contamination and UV irradiation by substituting equation 4.5 into 4.10. We call this
the Langmuir Hinshelwood Cassie Baxter Young (LHCBY) relation. It can be seen
that the variation of cosine of the contact angle with time follows an exponential
behavior.
cos
W'O
k+= ( -2 k k k - fsc(0) e-(kc+k-c+kp)t (4.11)
1ke + k-c + kP
(k, + k-c + kp
Figure 4-6 and Figure 4-7 show the cosines of the contact angles of water in oil
plotted against the time of contamination and UV irradiation respectively. A range
of UV intensities were tested to observe the effects of UV intensity on the kinetics
of wetting.
The LHCBY model was fit to this data to obtain the rate constants
kc and k-, under dark conditions and kp for each UV intensity.
and desorption rate constants were found to be 3.54 x 10-
The adsorption
s-1 and 3.58 x 10- s-1
respectively. Thus the adsorption rate constant was about 100 times greater than the
desorption rate constant implying that the system tended towards complete surface
coverage of contaminant at equilibrium.
Although it has been observed that the
,
Langmuir adsorption constants are dependent on the UV illumination conditions
it has been assumed to be equal to the obtained values from dark conditions. The
initial surface fraction of hydrophobic sites was calculated from the initial contact
angle (before UV illumination) before each UV experiment. An assumption in these
experiments is that the water adsorption on bare titania is quick and irreversible.
70
1.0
Contamination in
dodecane
- - - Exp. fit
k= 3.5 x10s
0
.
0.5-
k= 3.5 x 10 -7 s-
00
C-
0.0 -
Cl)
0
-0.5-1.0 -,
0
0
300
600
600
900
1200 1500
1800
ttcontamination (mm)
(in
Figure 4-6: Evolution of water-in-oil contact angle with time of immersion in dodecane
along with the fit
This means that a water molecule which comes in contact with bare titania which
was UV irradiated, would immediately bond to the surface due to photoinduced
superhydrophilicity.
The photocatalysis rate constants kp were found to increase with UV light intensity. Effect of UV light intensity on rate of photocatalytic degradation has been
studied in literature and it has been observed that the rate of photocatalytic reaction
is a conditional function of UV intensity (I) [64, 21].
r oc
I,
at low intensities
I5,
at high intensities
110
(4.12)
at very high intensities
At low intensities, the rate is proportional to I as is limited by the photon incidence
rate but at high intensities the rate is proportional to
11/2
as e--h+ pair recombination
starts to decrease the efficiency of the reaction. The transition between these two
regimes is approximately around 1-10 mW/cm 2 121]. Since the intensities used were
much higher that this limit, the rates kp were found to follow the
71
j1/2
power law as
1.0
CDa
1= 40 mW/cm 2
-0.5
0
200
)
x
I
---
-
-1.0
10 3 s-1
120 mW/cm 2
(k, = 14.4 x 10-3s 1 )
200 mW/cm 2
(k = 49.5 x 10-3s1
(k = 7.1
-
- Fit
* 1=
- - Fit
I=
*
- - Fit
-
0
)
*
0.0
400
UV irradiation
600
(S)
Figure 4-7: Cosine of contact angle of water in oil environment against time of UV
irradiation at various intensities
shown in Figure 4-8. This is interesting as we can see that photoinduced wetting also
follows kinetics similar to that of photocatalysis. This is a useful as findings in one
field can be interchangeably applied to the other in future.
4.4
Applications
In the situations where oil-water separation needs to be done with water rich emul-
sions, it more useful to have kinetics of oil drop wetting on titania underwater. Figure 4-9 shows an oil drop detaching from a contaminated titania surface under UV
irradiation in water environment.
The oil drop beads up from the surface and sat-
urates at a high contact angle as shown in Figure 4-10. We observe kinetics similar
to water-in-oil experiments in the plot of contact angle
O6'
verses time. Although
the drop does not detach from the surface on its own, a slight mechanical vibration
causes the drop to detach from the mesh surface due to the weakened adhesion at
contact line. In order to simulate a practical situation, we fouled an LBL coated ti-
72
101
5x10-2
2x10-2
10-2
5x10-3
*
2x1 0733
21
10'
---
5x101
2x101
k (I)
1/2 power law fit
102
2x10 2
5x102
UV Intensity (mW/crr)
Figure 4-8: First order wetting rate constants (kp) vs. UV intensity. I1/2 power law fit
was applied to the data in accordance with photocatalysis studies at high intensities.
Dodecane drop
tuv = 0 min
tuy=
tUy = 0.5
10 min
min
tu= 12 min
tuy= 2 min
t v=
2 mdisturbance
2~
ty= 5 min
15 min Oil detaches upon
small mechanical
t=
18 min
mm
)
Figure 4-9: Contact angle evolution with time for oil drop immersed in water environment on calcined titania LBL sample(I = 240 mW/cm 2
tania stainless steel mesh (pore size 70 pm) with dodecane and observed its recovery
under UV irradiation under water.
In Figure 4-11, we can see that the oil dewets
from the pores and beads up on the surface. Again, the oil drop leaves the surface
upon obtaining sufficient buoyancy or upon small mechanical vibration. This can be
applied for in-situ photocatalytic cleaning of membranes coated with titania under
water. Other applications that can take advantage of this competitive oil-water wetting include precise control oil-water drop shape on a surface, water drop coalescence
on a surface, directional transport of oil-water across membrane etc.
73
16(
3
14( -tb0
0
0
E00%
00
12(
10(
2
0*
00
E
00
E
68(
00
*
6
00000060
4 0
-
0 Width
2(
0
200
400
600
800
-- A
1000 1200' 1400
UV irradiation (S)
contact
angle
with
time
of UV
)
Evolution of oil-in-water
Figure 4-10:
irradiation(I = 240 mW/cm 2
Oil beading up from a contaminated
Figure 4-11:
irradiation(I = 300 mW/cm 2 ) in water environment
74
mesh
under
UV
Chapter 5
Conclusions and Future Work
This work has quantified and modeled the fouling and recovery kinetics of titanium
dioxide nanoporous surfaces under UV irradiation. The main findings of this study
are as follows. Although titania is hydrophilic and underwater oleophobic by nature,
this property is masked after fouling by oil. The intrinsic hydrophilicity can be recovered by UV irradiation as water preferentially wets titania over oil under UV light.
The photoinduced wetting changes are much faster than the fouling-induced wetting
changes and hence UV light can be an effective method for fouling recovery. The
kinetics of wetting were modeled by using a combination of the first order LangmuirHinshelwood equation and Cassie-Baxter equation.
The analysis showed that the
rates of photoinduced wetting increased with UV intensity in a non-linear fashion
similar to those in photocatalysis studies. It was also demonstrated that this fouling
recovery from oil can be extended to mesh like structures coated with nanoporous
titanium dioxide which have application in membrane oil-water separation.
This work has helped us understand the photocatalytic wetting behavior of titanium dioxide in an water-oil environment but at the same time has also raised a
number of questions that need to be answered by future studies. For example, these
experiments were performed on Layer-by-layer assembled TiO
2
nanoparticle surfaces
and hence the assumptions of Cassie-Baxter equation were justified. However, further
studies need to be done on flat titanium dioxide surfaces to obtain information on
the evolution of Young's contact angle of water drop in oil environment under UV
75
irradiation. Although photocatalysis has been proposed as the main mechanism for
wetting, further experiments need to be conducted to find the exact mechanism of
water wetting in the presence of oil. UV attenuation losses through the liquid medium
need to be minimized by designing innovative light delivery systems. UV light can
be eliminated by using visible light active nanoparticles/surfaces instead.
By connecting the two surface phenomena - photocatalysis and wetting, this work
will enable us to understand the less understood interfacial wetting behavior of surfaces with the help of well-known photocatalysis mechanisms. While this work focuses specifically on the material titanium dioxide, concepts demonstrated here can
be extended to many other catalyst and liquid interface systems and hence has great
potential for further research and applications.
76
Appendix A
Tables
77
78
Appendix B
Figures
Ll
3000:
L
L
L
L
30
35
40
45
2500Q
.
2000
1500
1000
3
500
O10
15
20
25
50
55
60
20['](CuK
Figure B-1: X-ray diffraction of the calcined LBL TiO 2 film on a glass slide. Anatase
phase was confirmed.
79
Power Spectra of OmniCuree S2000 Standard Lamps
140
120
100
op.%
80
2
60
A
40
20
0
50
350
300
450
400
500
550
600
650
Wavelength (nm)
Figure B-2: Spectra of the Omnicure S2000 Lamp.
Brochure
Courtesy:
Omnicure S2000
FGUV5M
d I~E~
I 0U
80
0
8'
E)
*0
a*-
60
--
40
20
-6000
0
2 '00
400
600
800
10
12010
1000 1200 1400 1600 1800
Wavelength (nm)
Figure B-3: UV Pass Filter
80
Intensity Vs. Shutter Opening
Absolute Intensity I
.
22 r
20 .1 S
E 18
U
16
14
12
4aC 10
8
0
0
-
U)
00
4
2
0
C 10 20 30 40 50 60 70 80 90 100
Percentage of shutter open (%)
Figure B-4: UV source calibration: Variation of absolute intensity at the immediate
output of the probe with respect to the percentage shutter opening
1100
1000
900
800
700
E 600
500
'C 400
.4300
200
100
---- x= 3 mm
S-x=4 mm
-"--x=5mm
-x=6 mm
* -x=7mm
* x=8mm
A1W
0
20
40
60
80
Percentage Shutter Opening (%)
Figure B-5: UV source calibration: Variation of the absolute intensity at a point on
the central axis with respect to percentage shutter opening (S.0)
81
Cf2
1 = 907e-175x mW/cm
*
*
*
800
2
1 = 454e-184 mW/cm
-
0
1000
2
1 = 1376e-175x mW/cm2
1 = 1695e-168x mW/cm
2
600
-4
400
-- 4
200
-00
L
4
3
r
6
5
8
7
(
Distance from source x10-
3
m)
Figure B-6: UV source calibration: Variation of the absolute intensity on the center
axis at various shutter openings
1000
%
-S.O = 20
--- S.O = 40
+---S.0= 60
700
+S.0 =80
%
800
%
%
900
600-
4-0
:t:
*
500
S
0~
~0
a
400K
9
4
\
0
300
U)
N
*
E
S
/*
200
0
e
100
-5
-
0
4
4
-3
-2
0
-1
1
2
3
4
5
Distance from Central Axis (mm)
Figure B-7: UV source calibration: Radial variation of the absolute intensity at a
constant center distance of 4 mm from the optical filter
82
Bibliography
Technical report, International Energy
[1] Key World Energy STATISTICS.
Agency, 2013.
12] World Oil Outlook 2013. Technical report, Organization of the Petroleum Exporting Countries, 2013.
[3] A. Abdelrasoul, H. Doan, and A. Lohi. Fouling in membrane filtration and
remediation methods. Mass Transfer-Advances in Sustainable Energy and Environment Oriented Numerical Modeling, pages 195-218, 2013.
[4] Amira Y. Ahmed, Tarek A. Kandiel, Torsten Oekermann, and Detlef Bahnemann. Photocatalytic activities of different well-defined single crystal TiO 2
surfaces: anatase versus rutile. The Journal of Physical Chemistry Letters,
2(19):2461-2465, 2011.
[5] Salem Alzahrani and Abdul Wahab Mohammad.
Challenges and trends in
membrane technology implementation for produced water treatment: a review.
Journal of Water Process Engineering, 4:107-133, 2014.
[6] J Barksdale. Titanium. In The Encyclopedia of the Chemical Elements, pages
732-738. Reinhold Book Corporation:Skokie, IL, 1968.
[7] Marine Board and Ocean Studies Board. Oil in the Sea III: Inputs, Fates, and
Effects. national academies Press, 2003.
18] Yingze Cao, Xiaoyong Zhang, Lei Tao, Kan Li, Zhongxin Xue, Lin Feng, and
Yen Wei. Mussel-inspired chemistry and Michael addition reaction for efficient
oil/water separation. ACS applied materials & interfaces, 5(10):4438-4442,
2013.
[9] Gianvito Caputo, Roberto Cingolani, Pantaleo Davide Cozzoli, and Athanassia Athanassiou. Wettability conversion of colloidal TiO 2 nanocrystal thin
films with UV-switchable hydrophilicity. Physical Chemistry Chemical Physics,
11(19):3692-3700, 2009.
[10] A. B. D. Cassie and S. Baxter. Wettability of porous surfaces. Transactions of
the Faraday Society, 40:546-551, 1944.
83
[11] Fevzi Q Cebeci, Zhizhong Wu, Lei Zhai, Robert E. Cohen, and Michael F.
Rubner. Nanoporosity-driven superhydrophilicity: a means to create multifunctional antifogging coatings. Langmuir, 22(6):2856-2862, 2006.
[12] Ning Chen and Qinmin Pan. Versatile fabrication of ultralight magnetic foams
and application for oil-Water separation. A CS nano, 7(8):6875-6883, July 2013.
[13] Xiaobo Chen and Samuel S. Mao. Titanium dioxide nanomaterials: synthesis,
properties, modifications, and applications. Chemical reviews, 107(7):28912959, 2007.
[14] Qunfeng Cheng, Mingzhu Li, Fu Yang, Mingjie Liu, Lin Li, Shutao Wang, and
Lei Jiang. An underwater pH-responsive superoleophobic surface with reversibly
switchable oil-adhesion. Soft Matter, 8(25):6740-6743, 2012.
115] Meng Nan Chong, Bo Jin, Christopher WK Chow, and Chris Saint. Recent
developments in photocatalytic water treatment technology: a review.
research, 44(10):2997-3027, May 2010.
[16] Alfred Clark.
York, 1970.
Water
The theory of adsorption and catalysis. Academic Press, New
f17] Robert E. Cohen, Michael F. Rubner, Gareth H. Mckinley, George Barbastathis,
Hyungryul Johnny Choi, Kyoo Chul Park, and Hyomin Lee. Anti-fingerprint
photocatalytic nanostructure for transparent surfaces., U.S. Patent Application
14/274,642, 2014.
[18] Elizabeth A. Costner, Brian K. Long, Carlos Navar, Steffen Jockusch, Xuegong
Lei, Paul Zimmerman, Alan Campion, Nicholas J. Turro, and C. Grant Willson.
Fundamental optical properties of linear and cyclic alkanes: Vuv absorbance and
index of refraction. The Journal of Physical Chemistry A, 113(33):9337-9347,
2009.
[191 Raul Dores, Altaf Hussain, Mary Katebah, and Samer S. Adham. Using Advanced Water Treatment Technologies To Treat Produced Water From The
Petroleum Industry. In SPE InternationalProduction and Operations Conference and Exhibition. Society of Petroleum Engineers, 2012.
[20] Howard Duhon. Produced Water Treatment : Yesterday , Today , and Tomorrow. Oil and Gas Facilities,pages 29-31, February 2012.
[211 Alexei V. Emeline, Vladimir Ryabchuk, and Nick Serpone. Factors affecting
the efficiency of a photocatalyzed process in aqueous metal-oxide dispersions:
prospect of distinguishing between two kinetic models. Journal of Photochemistry and Photobiology A: Chemistry, 133(1):89-97, 2000.
122] Ahmadun Fakhru'l-Razi, Alireza Pendashteh, Luqman Chuah Abdullah,
Dayang Radiah Awang Biak, Sayed Siavash Madaeni, and Zurina Zainal Abidin.
84
1-
1 .. I
,
-
,
t
kweALd "ia
Review of technologies for oil and gas produced water treatment. Journal of
Hazardous Materials, 170(2):530-551, 2009.
[231 Wen Qi Fang, Xue-Qing Gong, and Hua Gui Yang. On the unusual properties
of anatase TiO 2 exposed by highly reactive facets. The Journal of Physical
Chemistry Letters, 2(7):725-734, 2011.
[24] Lin Feng, Zhongyi Zhang, Zhenhong Mai, Yongmei Ma, Biqian Liu, Lei Jiang,
and Daoben Zhu. A super-hydrophobic and super-oleophilic coating mesh film
for the separation of oil and water. Angewandte Chemie InternationalEdition,
43(15):2012-2014, 2004.
[25] T. Frankiewicz. Understanding the Fundamentals of Water Treatment, the
Dirty Dozen - 12 Common Causes of Poor Water Quality. In 11th Produced
Water Seminar, Houston, TX, January 2001.
[26] Akira Fujishima. Electrochemical photolysis of water at a semiconductor electrode. Nature, (238):37-38, 1972.
[27] Akira Fujishima, Tata N. Rao, and Donald A. Tryk. Titanium dioxide photocatalysis. Journal of Photochemistry and Photobiology C: Photochemistry
Reviews, 1(1):1-21, 2000.
[28] Akira Fujishima, Xintong Zhang, and Donald A. Tryk. TiO 2 photocatalysis
and related surface phenomena. Surface Science Reports, 63(12):515-582, 2008.
[291 Shou Jian Gao, Zhun Shi, Wen Bin Zhang, Feng Zhang, and Jian Jin. Photoinduced Superwetting Singel-Walled Nanotube/ TiO 2 Ultrathin Network Films
for Ultrafast Separation of Oil-in-Water Emulsions. A CS nano, 8(6):6344-6352,
2014.
[30] Dengteng Ge, Lili Yang, Chenbo Wang, Elaine Lee, Yongquan Zhang, and
Shu Yang. A multi-functional oil-water separator from a selectively pre-wetted
superamphiphobic paper. Chemical Communications, 51(28):6149-6152, 2015.
&
[31] Mohammed A. Gondal, Muhammad S. Sadullah, Mohamed A. Dastageer,
Gareth H. Mckinley, Divya Panchanathan, and Kripa K. Varanasi. Study of
Factors Governing Oil-Water Separation Process Using TiO 2 Films Prepared
by Spray Deposition of Nanoparticle Dispersions. ACS applied materials
interfaces, 2014.
[321 Mohammed. A. Gondal, Muhammad. S. Sadullah, Gareth. H. McKinley,
Kripa. K. Varanasi, and Divya. Panchanathan. Photo-induced in situ switching
of surface wettability of Titania films under air and oil environment. In High
Capacity Optical Networks and Enabling Technologies (HONET-CNS), pages
151-154. IEEE, December 2013.
85
[33] Kaishu Guan. Relationship between photocatalytic activity, hydrophilicity and
self-cleaning effect of TiO 2 / Si0 2 films.
Surface and Coatings Technology,
191(2):155-160, 2005.
[34] Katie Guerra, Katharine Dahm, and Steve Dundorf. Oil and Gas Produced
Water Management and Beneficial Use in the Western United States. Technical Report No. 157, U.S. Department of the Interior, Bureau of Reclamation,
September 2011.
[35] Christian A. Gueymard. The sun's total and spectral irradiance for solar energy
applications and solar radiation models. Solar Energy, 76(4):423-453, 2004.
[36] Michael A. Henderson. An HREELS and TPD study of water on TiO 2 (110): the
extent of molecular versus dissociative adsorption. Surface Science, 355(1):151166, 1996.
[37] Paul C. Hiemenz and R. Rajagopalan. Principles of Colloid and Surface Chemistry, revised and expanded., volume 14. CRC Press, 1997.
[38] James G. Highfield and Michael Gratzel. Discovery of reversible photochromism
in titanium dioxide using photoacoustic spectroscopy: implications for the investigation of light-induced charge-separation and surface redox processes in
titanium dioxide. The Journal of Physical Chemistry, 92(2):464-467, 1988.
[391 Ebenezer T. Igunnu and George Z. Chen. Produced water treatment technologies. InternationalJournal of Low-Carbon Technologies, 0:1-21, 2012.
[401 Michel Y. Jaffrin. Dynamic shear-enhanced membrane filtration: A review of
rotating disks, rotating membranes and vibrating systems.
brane Science, 324(1-2):7-25, 2008.
Journal of Mem-
[41] Hao Ju, Bryan D. McCloskey, Alyson C. Sagle, Yuan-Hsuan Wu, Victor A.
Kusuma, and Benny D. Freeman. Crosslinked poly(ethylene oxide) fouling resistant coating materials for oil/water separation. Journal of Membrane Science,
307(2):260-267, 2008.
[42] Yong Chae Jung and Bharat Bhushan. Wetting behavior of water and oil
droplets in three-phase interfaces for hydrophobicity/philicity and oleophobicity/philicityaAg. Langmuir, 25(24):14165-14173, December 2009.
1431
Prashant V Kamat. TiO 2 nanostructures : recent physical chemistry advances.
The Journal of Physical Chemistry C, 116(22):11849-11851, 2012.
[44] Arun K. Kota, Gibum Kwon, Wonjae Choi, Joseph M. Mabry, and Anish Tuteja.
Hygro-responsive membranes for effective oil-water separation. Nature communications, 3:1025, 2012.
86
[45] S. Girish Kumar and L. Gomathi Devi. Review on modified TiO 2 photocatalysis under UV/visible light: selected results and related mechanisms on interfacial charge carrier transfer dynamics. The Journal of Physical Chemistry A,
115(46):13211-13241, 2011.
[46] Richard L. Kurtz, Roger Stock-bauer, Theodore E. Msdey, Elisa Romiln, and
Jos6L. de Segovia. Synchrotron radiation studies of H 2 0 adsorption on TiO 2
(110). Surface Science, 218(1):178-200, 1989.
[47J Gibum Kwon, Arun Kota, Yongxin Li, Ameya Sohani, Joseph M. Mabry, and
Anish Tuteja. On-demand separation of oil-water mixtures. Advanced materials,
24(27):3666-3671, July 2012.
[481 Daeyeon Lee, Michael F. Rubner, and Robert E. Cohen. All-nanoparticle thinfilm coatings. Nano letters, 6(10):2305-2312, 2006.
149]
Fuk Kay Lee, Gaeelle Andreatta, and Jean-Jacques Benattar. Role of water
adsorption in photoinduced superhydrophilicity on TiO 2 thin films. Applied
Physics Letters, 90(18), 2007.
[50] Kunyoung Lee, QHwan Kim, Sangmin An, JeongHoon An, Jongwoo Kim,
Bongsu Kim, and Wonho Jhe. Superwetting of TiO 2 by light-induced waterlayer growth via delocalized surface electrons. Proceedings of the National
Academy of Sciences, 111(16):5784-5789, May 2014.
[51] Kan Li, Jie Ju, Zhongxin Xue, Jie Ma, Lin Feng, Song Gao, and Lei Jiang.
Structured cone arrays for continuous and effective collection of micron-sized
oil droplets from water. Nature Communications, 4, August 2013.
[52] Lin Li, Zhaoyue Liu, Qianqian Zhang, Chenhui Meng, Tierui Zhang, and Jin
Zhai. Underwater superoleophobic porous membrane based on hierarchical TiO 2
nanotubes: multifunctional integration of oil-water separation, flow-through
photocatalysis and self-cleaning. Journal of Materials Chemistry A, 3(3):12791286, 2015.
[53] Yong Li and Wenjie Shen. Morphology-dependent nanocatalysts: Rod-shaped
oxides. Chemical Society Reviews, 43(5):1543-1574, 2014.
[54] Kesong Liu, Moyuan Cao, Akira Fujishima, and Lei Jiang. Bio-inspired titanium dioxide materials with special wettability and their applications. Chemical
reviews, 114(19):10044-10094, October 2014.
[55] Mingjie Liu, Shutao Wang, Zhixiang Wei, Yanlin Song, and Lei Jiang. Bioinspired design of a superoleophobic and low Ashesive water/solid interface. Advanced Materials, 21(6):665-669, February 2009.
[56] Na Liu, Yuning Chen, Fei Lu, Yingze Cao, Zhongxin Xue, Kan Li, Lin Feng,
and Yen Wei. Straightforward oxidation of a copper substrate produces an
87
underwater superoleophobic mesh for oil/water separation.
14(15):3489-3494, October 2013.
ChemPhysChem,
[57] Shengwei Liu, Jiaguo Yu, and Mietek Jaroniec. Anatase TiO 2 with dominant
high-energy {001} facets : synthesis, properties, and applications. Chemistry
of Materials, 23(18):4085-4093, 2011.
[58] Wei Jen Lo, Yip Wah Chung, and G. A. Somorjai. Electron spectroscopy studies
of the chemisorption of 02, H 2 and H 2 0 on the TiO 2 (100) surfaces with varied
stoichiometry: Evidence for the photogeneration of Ti+ 3 and for its importance
in chemisorption. Surface Science, 71(2):199-219, 1978.
[591 Max Lu. Photocatalysis and water purification: from fundamentals to recent
applications. John Wiley & Sons, 2013.
[60] Tim Luttrell, Sandamali Halpegamage, Junguang Tao, Alan Kramer, Eli Sutter,
and Matthias Batzill. Why is anatase a better photocatalyst than rutile?-Model
studies on epitaxial TiO 2 films. Scientific reports, 4, 2014.
[61] Adam J. Meuler, Shreerang S. Chhatre, Amarilys Rivera Nieves, Joseph M.
Mabry, Robert E. Cohen, and Gareth H. McKinley. Examination of wettability and surface energy in fluorodecyl POSS/polymer blends. Soft Matter,
7(21):10122-10134, 2011.
[62] Andrew Mills and Matthew Crow. A study of factors that change the wettability
of titania films. InternationalJournal of Photoenergy, 2008.
[631 Andrew Mills and Stephen Le Hunte. An overview of semiconductor photocatalysis. Journal of Photochemistry and Photobiology A: Chemistry, 108(1):1-35,
1997.
[641 Andrew Mills, Jishun Wang, and David F. Ollis. Dependence of the kinetics of liquid-phase photocatalyzed reactions on oxygen concentration and light
intensity. Journal of Catalysis, 243(1):1-6, 2006.
[65] Masahiro Miyauchi, Akira Nakajima, Toshiya Watanabe, and Kazuhito
Hashimoto. Photocatalysis and photoinduced hydrophilicity of various metal
oxide thin films. Chemistry of Materials, 14(6):2812-2816, 2002.
[66] Ryuhei Nakamura, Kazuhiro Ueda, and Shinri Sato. In situ observation of
the photoenhanced adsorption of water on TiO 2 films by surface-enhanced IR
absorption spectroscopy. Langmuir, 17(8):2298-2300, 2001.
[67] Kazuya Nakata and Akira Fujishima. TiO 2 photocatalysis: design and applications. Journal of Photochemistry and Photobiology C: Photochemistry Reviews,
13(3):169-189, 2012.
88
[68] Meng Ni, Michael KH Leung, Dennis YC Leung, and K. Sumathy. A review and
recent developments in photocatalytic water-splitting using TiO 2 for hydrogen
production. Renewable and Sustainable Energy Reviews, 11(3):401-425, 2007.
[691 Shunsuke Nishimoto, Yuichi Mori, Yoshikazu Kameshima, and Michihiro
Miyake. Reversible Control of Underwater Oil Wettability of a Titanium Dioxide Surface through Ultraviolet and Ultrasonic Irradiation. Chemistry Letters,
44(3):262-264, 2015.
[70] Shunsuke Nishimoto, Yusuke Sawai, Yoshikazu Kameshima, and Michihiro
Miyake. Underwater Superoleophobicity of TiO 2 Nanotube Arrays. Chemistry
Letters, 43(4):518-520, 2014.
171] Shunsuke Nishimoto, Sana Tomoishi, Yoshikazu Kameshima, Eiji Fujii, and
Michihiro Miyake. Self-cleaning efficiency of titanium dioxide surface under
simultaneous UV irradiation of various intensities and water flow. Journal of
the Ceramic Society of Japan, 122(1426):513-516, 2014.
[72] Atsuko Y. Nosaka, Eiichi Kojima, Toshimichi Fujiwara, Hiromasa Yagi, Hideo
Akutsu, and Yoshio Nosaka. Photoinduced changes of adsorbed water on a
TiO 2 photocatalytic film as studied by 1H NMR spectroscopy. The Journal of
Physical Chemistry B, 107(44):12042-12044, 2003.
[73] Thomas Oppenhinder. Photochemical Purification of Water and Air: Advanced
Oxidation Processes (AOPs) - Principles, Reaction Mechanisms, Reactor Concepts. John Wiley & Sons, 2003.
[74] M. Padaki, R. Surya Murali, M.S. Abdullah, N. Misdan, A. Moslehyani, M.A.
Kassim, N. Hilal, and A.F. Ismail. Membrane technology enhancement in oilwater separation. A review. Desalination,357:197-207, 2015.
[75] Shuaijun Pan, Rui Guo, and Weijian Xu. Durable superoleophobic fabric surfaces with counterintuitive superwettability for polar solvents. AIChE Journal,
60(8):2752-2756, 2014.
[76] E. Pelizzetti and C. Minero. Mechanism of the photo-oxidative degradation of
organic pollutants over TiO 2 particles. Electrochimica acta, 38(1):47-55, 1993.
[77] E. Pelizzetti, C. Minero, and M. Vincenti. Photocatalytic degradation of organic contaminants. In Technologies for Environmental Cleanup: Toxic and
Hazardous Waste Management, pages 101-138. Springer Netherlands, 1994.
[78] Ezio Pelizzetti, Claudio Minero, Valter Maurino, Hisao Hidako, Nick Serpone,
and Rita Terzian. Photocatalytic degradation of dodecane and of some dodecyl
derivatives. Annali di Chimica, 80:81-87, 1990.
[79] Tajana Preoeanin and Nikola Kallay. Point of Zero Charge and Surface Charge
Density of TiO 2 in Aqueous Electrolyte Solution as Obtained by Potentiometric
Mass Titration. Croatica Chemica Acta, 79(1):95-106, 2006.
89
1801 Jan F. Rabek. Polymer photodegradation: mechanisms and experimental methods. Springer Science & Business Media, 1995.
[81] V. Rico, P. Romero, J. L. Hueso, J. P. Espinos, and A. R. GonzAlez-Elipe.
Wetting angles and photocatalytic activities of illuminated TiO 2 thin films.
Catalysis Today, 143(3):347-354, 2009.
[821 Madhusmita Sahoo, Tom Mathews, Rajini P. Antony, D. Nandagopala Krishna, Sitaram Dash, and Ashok Kumar Tyagi. Physico-chemical Processes
and Kinetics of Sunlight-Induced Hydrophobic - Superhydrophilic Switching of
Transparent N-Doped TiO 2 Thin Films. ACS applied materials & interfaces,
5(9):3967-3974, 2013.
1831 Munetoshi Sakai, Masaki Nishimura, Yasushi Morii, Tsutomu Furuta, Toshihiro
Isobe, Akira Fujishima, and Akira Nakajima. Reduction of fluid friction on the
surface coated with TiO 2 photocatalyst under UV illumination. Journal of
Materials Science, 47(23):8167-8173, 2012.
[84] Nobuyuki Sakai, Akira Fujishima, Toshiya Watanabe, and Kazuhito Hashimoto.
Enhancement of the photoinduced hydrophilic conversion rate of TiO 2 film electrode surfaces by anodic polarization. The Journal of Physical Chemistry B,
105(15):3023-3026, 2001.
1851 Nobuyuki Sakai, Akira Fujishima, Toshiya Watanabe, and Kazuhito Hashimoto.
Quantitative evaluation of the photoinduced hydrophilic conversion properties
of TiO 2 thin film surfaces by the reciprocal of contact angle. Journal of Physical
Chemistry B, 107(4):1028-1035, 2003.
[86] Nobuyuki Sakai, Rong Wang, Akira Fujishima, Toshiya Watanabe, and
Kazuhito Hashimoto. Effect of ultrasonic treatment on highly hydrophilic TiO 2
surfaces. Langmuir, 14(20):5918-5920, 1998.
[87] David 0. Scanlon, Charles W. Dunnill, John Buckeridge, Stephen A. Shevlin,
Andrew J. Logsdail, Scott M. Woodley, C. Richard A. Catlow, Michael J. Powell, Robert G. Palgrave, Ivan P. Parkin, Graeme W. Watson, Thomas W. Keal,
Paul Sherwood, Aron Walsh, and Alexey A. Sokol. Band alignment of rutile
and anatase TiO 2 . Nature materials, 12(9):798-801, 2013.
[88] Kazuhiko Seki and M. Tachiya. Kinetics of photoinduced hydrophilic conversion
processes of Ti0 2 surfaces. The Journal of Physical Chemistry B, 108(15):48064810, 2004.
1891 Yanwei Shang, Yang Si, Aikifa Raza, Liping Yang, Xue Mao, Bin Ding, and
Jianyong Yu. An in situ polymerization approach for the synthesis of superhydrophobic and superoleophilic nanofibrous membranes for oil-water separation.
Nanoscale, 4(24):7847-7854, 2012.
90
[901 Zhun Shi, Wenbin Zhang, Feng Zhang, Xia Liu, Dong Wang, Jian Jin, and Lei
Jiang. Ultrafast Separation of Emulsified Oil/Water Mixtures by Ultrathin FreeStanding Single-Walled Carbon Nanotube Network Films. Advanced materials,
25(17):2422-2427, May 2013.
[911 Brian R. Solomon, Md Nasim Hyder, and Kripa K. Varanasi. Separating OilWater Nanoemulsions using Flux-Enhanced Hierarchical Membranes. Scientific
reports, 4, January 2014.
192]
Wenlong Song, Fan Xia, Yubai Bai, Fengqi Liu, Taolei Sun, and Lei Jiang. Controllable water permeation on a poly(N-isopropylacrylamide)-modified nanostructured copper mesh film. Langmuir, 23(1):327-331, January 2007.
[93] Milena Stepien, Jarkko J. Saarinen, Hannu Teisala, Mikko Tuominen, Mikko
Aromaa, Janne Haapanen, Jurkka Kuusipalo, Jyrki M. Makela, and Martti
Toivakka. ToF-SIMS analysis of UV-switchable TiO 2 -nanoparticle-coated paper surface. Langmuir, 29(11):3780-3790, March 2013.
[94] Michela Sturini, Federica Soana, and Angelo Albini. Reaction paths in the titanium dioxide photocatalysed degradation of dodecane and some of its derivatives. Tetrahedron, 58(15):2943-2950, 2002.
[95] Ren-de Sun, Akira Nakajima, Akira Fujishima, Toshiya Watanabe, and
Kazuhito Hashimoto. Photoinduced surface wettability conversion of ZnO and
TiO 2 thin films. Journal of Physical Chemistry B, 105(10):1984-1990, 2001.
[961 Wei Sun, Shuxue Zhou, Bo You, and Limin Wu. A facile method for the fabrication of superhydrophobic films with multiresponsive and reversibly tunable
wettability. Journal of Materials Chemistry A, 1(9):3146-3154, 2013.
[97] Komkrit Suttiponparnit, Jingkun Jiang, Manoranjan Sahu, Sirikalaya Suvachittanont, Tawatchai Charinpanitkul, and Pratim Biswas. Role of surface area,
primary particle size, and crystal phase on titanium dioxide nanoparticle dispersion properties. Nanoscale Research. Letters, 6(1):27, 2011.
[98] Masato Takeuchi, Kenji Sakamoto, Gianmario Martra, Salvatore Coluccia, and
Masakazu Anpo. Mechanism of photoinduced superhydrophilicity on the TiO 2
photocatalyst surface. Journal of Physical Chemistry B, 109(32):15422-15428,
2005.
[991 Sakae Tanemura, Lei Miao, Wilfried Wunderlich, Masaki Tanemura, Yukimasa
Mori, Shoichi Toh, and Kenji Kaneko. Fabrication and characterization of
anatase/rutile - TiO 2 thin films by magnetron sputtering: A review. Science
and Technology of Advanced Materials, 6(1):11-17, 2005.
1100] Mimi Tao, Lixin Xue, Fu Liu, and Lei Jiang. An intelligent superwetting PVDF
membrane showing switchable transport performance for oil/water separation.
Advanced Materials, 26(18):2943-2948, 2014.
91
[1011 Chao Teng, Xianyong Lu, Guangyuan Ren, Ying Zhu, Meixiang Wan, and
Lei Jiang. Underwater Self-Cleaning PEDOT-PSS Hydrogel Mesh for Effective Separation of Corrosive and Hot Oil/Water Mixtures. Advanced Materials
Interfaces, 1(6), 2014.
[102] Tracy L. Thompson and John T. Yates Jr. TiO 2 -based photocatalysis: surface
defects, oxygen and charge transfer. Topics in Catalysis, 35(3-4):197-210, 2005.
11031 Dongliang Tian, Xiaofang Zhang, Xiao Wang, Jin Zhai, and Lei Jiang. Micro/nanoscale hierarchical structured ZnO mesh film for separation of water
and oil. Physical Chemistry Chemical Physics, 13(32):14606-14610, 2011.
1104] Junwei Wang, Baodong Mao, James L. Gole, and Clemens Burda. Visible-lightdriven reversible and switchable hydrophobic to hydrophilic nitrogen-doped titania surfaces: correlation with photocatalysis. Nanoscale, 2(10):2257-2261,
2010.
[105] Rong Wang, Kazuhito Hashimoto, Akira Fujishima, Makota Chikuni, Eiichi
Kojima, Atsushi Kitamura, Mitsuhide Shimohigoshi, and Toshiya Watanabe.
Light-induced amphiphilic surfaces. Nature, 388:431-432, 1997.
[1061 Rong Wang, Nobuyuki Sakai, Akira Fujishima, Toshiya Watanabe, and
Kazuhito Hashimoto. Studies of surface wettability conversion on TiO 2 singlecrystal surfaces. Journal of Physical Chemistry B, 103(12):2188-2194, 1999.
[107] Shutao Wang, Yanlin Song, and Lei Jiang. Photoresponsive surfaces with controllable wettability. Journal of Photochemistry and Photobiology C: Photochemistry Reviews, 8(1):18-29, 2007.
[1081 Suhao Wang, Mei Li, and Qinghua Lu. Filter paper with selective absorption
and separation of liquids that differ in surface tension. A CS applied materials
& interfaces, 2(3):677-683, 2010.
[1091 T. Watanabe, A. Nakajima, R. Wang, M. Minabe, S. Koizumi, A. Fujishima,
and K. Hashimoto. Photocatalytic activity and photoinduced hydrophilicity of
titanium dioxide coated glass. Thin solid films, 351(1):260-263, 1999.
[110] Alex Weir, Paul Westerhoff, Lars Fabricius, Kiril Hristovski, and Natalie von
Goetz. Titanium dioxide nanoparticles in food and personal care products.
Environmental science & technology, 46(4):2242-2250, 2012.
1111] Qiang Wen, Jiancheng Di, Lei Jiang, Jihong Yu, and Ruren Xu. Zeolite-coated
mesh film for efficient oil-water separation. Chemical Science, 4(2):591-595,
2013.
1112] Robert N. Wenzel. Resistance of solid surfaces to wetting by water. Industrial
& Engineering Chemistry, 28(8):988-994, 1936.
92
[113] S. M. Wetterer, D. J. Lavrich, T. Cummings, S. L. Bernasek, and G. Scoles.
Energetics and kinetics of the physisorption of hydrocarbons on Au (111). The
Journal of Physical Chemistry B, 102(46):9266-9275, 1998.
[114] Fan Xia, Ying Zhu, Lin Feng, and Lei Jiang. Smart responsive surfaces switching
reversibly between super-hydrophobicity and super-hydrophilicity. Soft Matter,
5(2):275-281, 2009.
[1151 Li-Ping Xu, Xiuwen Wu, Jingxin Meng, Jitao Peng, Yongqiang Wen, Xueji
Zhang, and Shutao Wang. Papilla-like magnetic particles with hierarchical
structure for oil removal from water. Chemical Communications, 49(78):87528754, 2013.
[116] Baolong Xue, Longcheng Gao, Yongping Hou, Zhiwen Liu, and Lei Jiang. Temperature controlled water/oil wettability of a surface fabricated by a block
copolymer: application as a dual water/oil on-off switch. Advanced Materials, 25(2):273-277, January 2013.
1117] Zhongxin Xue, Shutao Wang, Ling Lin, Li Chen, Mingjie Liu, Lin Feng, and
Lei Jiang. A novel sujperhydrophilic and underwater superoleophobic hydrogelcoated mesh for oil/water separation. Advanced Materials, 23(37):4270-4273,
October 2011.
[118] Hao Yang, Xiaojing Hu, Rong Chen, Shantang Liu, Pihui Pi, and Zhuo-ru
Yang. Fluoropolymer/ SiO 2 composite films with switchable superoleophilicity
and high oleophobicity for 'on-off' oil permeation. Applied Surface Science,
280:113-116, 2013.
[119] Yu Yang, Kyle Doudrick, Xiangyu Bi, Kiril Hristovski, Pierre Herckes, Paul
Westerhoff, and Ralf Kaegi. Characterization of food-grade titanium dioxide:
The presence of nanosized particles. Environmental science & technology Technology, 48(11):6391-6400, 2014.
11201 Thomas Young. An essay on the cohesion of fluids. Philosophical Transactions
of the Royal Society of London, 95:65-87, 1805.
[121] Feng Zhang, Wen Bin Zhang, Zhun Shi, Dong Wang, Jian Jin, and Lei Jiang.
Nanowire-Haired Inorganic Membranes with Superhydrophilicity and Underwater Ultralow Adhesive Superoleophobicity for High-Efficiency Oil/Water Separation. Advanced materials, 25(30):4192-4198, 2013.
[1221 Jinfeng Zhang, Peng Zhou, Jianjun Liu, and Jiaguo Yu. New understanding
of the difference of photocatalytic activity among anatase, rutile and brookite
TiO 2. Phys. Chem. Chem. Phys., 16(38):20382-20386, 2014.
11231 Lianbin Zhang, Zhonghai Zhang, and Peng Wang. Smart surfaces with switchable superoleophilicity and superoleophobicity in aqueous media: toward controllable oil/water separation. NPG Asia Materials, 4(2):e8, 2012.
93
[124] Lianbin Zhang, Yujiang Zhong, Dongkyu Cha, and Peng Wang. A self-cleaning
underwater superoleophobic mesh for oil-water separation. Scientific reports, 3,
2013.
[125] Yuzhang Zhu, Feng Zhang, Dong Wang, Xian Feng Pei, Wenbin Zhang, and
Jian Jin. A novel zwitterionic polyelectrolyte grafted PVDF membrane for
thoroughly separating oil from water with ultrahigh efficiency. Journal of Materials Chemistry A, 1(18):5758-5765, 2013.
[126] Tykhon Zubkov, Dirk Stahl, Tracy L. Thompson, Dimitar Panayotov, Oliver
Diwald, and John T. Yates. Ultraviolet light-induced hydrophilicity effect on
TiO 2 (110)(1 x 1). Dominant role of the photooxidation of adsorbed hydrocarbons causing wetting by water droplets. The Journal of Physical Chemistry B,
109(32):15454-15462, 2005.
94