Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Water partitioning between the bulk and an electrode surface
(USC Thesis Other)
Water partitioning between the bulk and an electrode surface
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
Water Partitioning between the Bulk and an Electrode Surface
by
Anuj K Pennathur
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(CHEMICAL PHYSICS)
December 2023
Copyright 2024 Anuj K Pennathur
To Appa and Amma
ii
Acknowledgements
I would like to thank my amazing lab mates, specially, Dr. Anwesha Maitra, Dr. Cindy Tseng, Dr.
Sevan Menacheckanian, Dr. Matthew Voegtle and Dr. Ryan Hunt who helped me develop both
personally and professionally.
Thanks to Dr. Joel Patrow and Dr. Angelo Montenegro for teaching me the basics of ultrafast
laser operation - from OPA alignment to setting up a SFG experiment.
All the students in the Bradforth, Benderskii and Resiler lab for their support, feedback during
group meetings that have made my research better and willingness to help if required - from loaning
optics to helping troubleshoot laser problems.
My advisors, professors, Jahan Dawlaty and Alexander Benderskii for their guidance and support. Jahan has always been a mentor to me since my second week at USC (long before I joined
the lab). He has helped me in ways too long to write down, I’ll mention a few here. He has always
pushed me to achieve higher standards than the expectations I had for myself - with regard to presentations, explanation and research, which helped better my scientific communication. Looking
back, I think my research was productive because of the way Jahan encouraged my outlandish
ideas (which more often than not failed) in a way which did not discourage me from coming up
with more. Another important lesson Jahan taught me was to spend time carving out the question
your project aims to address and to pursue the path for answers even if the experiment lies outside
our comfort zone.
Alex’s scientific insight is nothing short of aspirational. In most of my group meetings that
are the culmination of a couple months of thought and experiments Alex would usually have a
suggestion or question that clarifies the project or brings it to a more understandable regime.
iii
Professor Steve Bradforth has helped my progress a lot by somehow always asking the hardest
and pertinent questions in group meetings irrespective of the topic. Watching Steve give talks has
made me a better presenter.
I have learnt a lot from Professor Curt Wittig. Curt was the academic advisor for our year of
graduate students, we had to talk to him prior to deciding which classes we would take. We had
spirited debates on harnessing phonons for coherent control, me with the overconfidence of a new
phd student and him with the professor-who-knows-everything energy, he always listened to my
point of view despite me not knowing really what I was talking about. Recently, Curt graciously
offered us his old FTIR for our experiments.
Professor Steve Cronin for serving in my committee. His work on electrochmistry with SERS
as well as tuning plasmonic excitations has helped me design new ideas for experiments (some
presented in this thesis).
Family has been very important in this journey, they have constantly reminded me about taking
care of myself and travelling. In this regard, specific acknowledgements to : Revathi chitti who
would always remind me to take care of my health, Arjun chitappa who’s work ethic is inspirational, Baba chitappa for his very practical advise, Kumar chitappa for his very interesting takes on
politics, Revathy chitti, Padma pati, CV pati, Ganesh, Krithika, Krish, for their support. My parents for being part of this journey pretty much every step of the way. My father being an academic
was interested more in the details of my research while my mother was interested in my travels
and explorations - this helped keep me balanced.
Special thanks to my partner, Dr. Ketika Garg, who has been with me through the good and
bad times. She helps me grow constantly as a person. In my research endeavors, I have immensely
profited from her external perspective which keeps me grounded and serves as a powerful reminder
that there are other things to do in life.
iv
Table of Contents
Dedication ii
Acknowledgements iii
List of Figures vii
Abstract xiv
Chapter 1: Introduction 1
1.1 Thesis Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Chapter 2: Ionic Liquid - Water Binary Mixtures 3
2.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.3 Experimental Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.5 Conclusion and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Chapter 3: Regulating Interfacial Water Content using Surfactants 18
3.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.3 Experimental Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.4 Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Chapter 4: Photothermal Modulation of Silicon Optical Cavities Measured by Continuous
Mid-IR 36
4.1 abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
4.3 The Concept of Cavity Photothermal Effect . . . . . . . . . . . . . . . . . . . . . 38
4.4 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.4.1 Experimental Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.4.2 Computational Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4.5 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4.6 Conclusion and Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Chapter 5: Ongoing work and Future directions 56
v
References 58
Appendices 75
A Conversion between A and coverage . . . . . . . . . . . . . . . . . . . . . . . . . 94
B Calculations for converting SERIAS absorption signal to surface coverage . . . . . 98
C Scheme for interpolation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
C.1 Computational transient temperature profile . . . . . . . . . . . . . . . . . 102
C.2 MATLAB code for simulations . . . . . . . . . . . . . . . . . . . . . . . . 107
vi
List of Figures
2.1 Cartoon showing the concept of the experiments in this paper. The nitrile probe
frequency changes due to electrostatic environments and hydrogen bonding
variations at the interface. We track how the frequency varies between the limits
of pure ionic liquid and pure water. As will be shown, this variation is vastly
different from a linear interpolation between these limits, and informs us about the
tendency of ions to aggregate at the interface. . . . . . . . . . . . . . . . . . . . . 7
2.2 SEM images of the roughned silver surface. . . . . . . . . . . . . . . . . . . . . . 8
2.3 The chemical structures and acronyms for the ionic liquids used in this study. . . . 9
2.4 Representative FTIR spectra for the [EMIM][BF4] and [HMIM][BF4] dilution
series. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.5 (a) The nitrile frequency of benzonitrile in different ionic liquids as a function of
water content. The data presented is for [EMIM][BF4] and [HMIM][BF4]. The
dashed line in black represents a hypothetical ideal solution where ωCN changes
linearly as a function of water concentration. ωCN for pure water (χwater = 1) was
not collected as benzonitrile does not dissolve in pure water. All frequency shifts
are with respect to the nitrile frequency in a dry sample (in contact with air) with
no solution. The error bars are frm (b) A schematic of the solvation structure
around benzonitrile in low and high water concentrations. The data suggests that
in water-IL mixtures the probe molecule is favorably solvated by ions. . . . . . . . 10
2.6 Representative SERS spectra.(a): [EMIM][BF4] dilution series.(b) : [HMIM][BF4]
dilution series.(c): [DMIM][BF4] dilution series. . . . . . . . . . . . . . . . . . . 11
2.7 The nitrile frequency at the interface as a function of water mole fraction for three
different ionic liquids. The ionic liquids were [EMIM][BF4] and [HMIM][BF4]
and [DMIM][BF4]. The inserts display a magnified version of the same data.
All frequency shifts are with respect to the nitrile frequency in a dry sample (in
contact with air) with no solution.The error bars represent errors from fitting the
spectra to Gaussian line-shapes. The shaded rectangle represents the measurement
error arising from substrate variability. . . . . . . . . . . . . . . . . . . . . . . . . 12
vii
2.8 (a): SERS data of the central nitrile frequency of MBN as a function of EMIm
BF4 concentration, this is the same data from the main text. (b): SERS data of the
central nitrile frequency of MBN as a function of KCl concentration.The nitrile
stretching frequency is referenced to the nitrile frequency in the absence of any
solvent. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.9 The linewidths of the nitrile stretch (benzonitrile) as a function of molfraction for
EMIm BF4 and HMIm BF4. These are for the bulk dilution and were obtained
through FTIR. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.10 The linewidths of the nitrile stretch (4-mercaptobenzonitrile monolayer) in the
presence of different concentrations of EMIm BF4, HMIm BF4 and DMIm BF4.
For a point of reference, linewidths of the substrate in absence of any ILs are 8.5
cm−1
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.1 A cartoon depicting the concept of our experiment. The vibrational signatures of
the surfactants (green) and water (blue) were tracked as a function of potential and
bulk water mole fraction. The bromide anion of CTAB is not shown. . . . . . . . . 20
3.2 Interfacial spectral signatures of the three component system of 1 mM CTAB in
10 M water in dDMSO at 0 V vs. Ag/AgCl. (a) SO stretch of dDMSO, (b) CH
stretch of CTAB, and (c) OH stretch of water referenced to gold on ZnSe. . . . . . 23
3.3 Potential-dependent SEIRAS data for the CTAB-dDMSO-water (10 M) system
with the corresponding spectral signatures labelled. All spectra have been
displaced vertically. The grey dashed line represents the baseline for each
spectrum. The solid black line indicates 0 V verses Ag/AgCl. Enrichment of
CTAB corresponds to depletion of dDMSO and water, and vice versa. . . . . . . . 24
3.4 The differential integrated peak data for the 1mM CTAB-dDMSO-10 M water
system. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.5 Integrated CH stretches of CTAB for the 1 mM CTAB 10 M water dDMSO
mixture as a function of applied potential ranging from – 0.6 V to + 0.6 V vs
Ag/AgCl. In the left panel, the potential scanned from 0 to -0.6, then to +0.6 and
back to -0.6. In the right panel, the potential is scanned from 0 to +0.6, then to
-0.6 and back to +0.6. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.6 Left panel. Integrated CH peaks of CTAB in a 1 mM CTAB – 10 M water dDMSO
mixture with 100 mM KCl (blue) and without 100 mM KCl (red). Right panel.
The reversibility of 1 mM CTAB, 10 M water-dDMSO mixture with 100 mM
KCl. The potential was stepped from 0 V to + 0.6 V to – 0.6 V and back. . . . . . . 27
3.7 The CH stretch region at 0 V for different water concentrations (a) and the
corresponding ratio of integrated areas as function of applied potential (b). For the
pure water and 20 M water cases, application of reductive potentials past -0.6 V
and -0.8 V were not possible due to substrate degradation. . . . . . . . . . . . . . 28
viii
3.8 (a) CTAB coverage (inferred from the spectra) as a function of applied potential
and bulk mole fraction of water in the system. The dotted lines are experimental
data. (b) A contour view of Figure 3.8a Each contour line represents equal
coverage of CTAB at various potentials and bulk water mole fraction. The
colorbar indicates the CTAB coverages, red corresponding to higher coverages
than blue. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.9 A plot demonstrating the equivalence between the hydrophobic effect and the
electrochemical potential necessary for driving surfactants to the surface. The
hydrophobicity of the bulk is tuned by the water mole fraction. . . . . . . . . . . . 31
3.10 Integrated CH peaks of CTAB (red) and CTAC (blue) in a 1 mM CTAB/CTAC –
10M water dDMSO. The potential was cycled from + 0.6 V to – 0.6 V and back. . 32
3.11 CH and OH integrated peak data for the 1 mM surfactant dDMSO 10 M water
system. The surfactants used were CTAB (left panel) and CTAC (right panel).
These experiments were carried out using the same SEIRAS substrate. . . . . . . . 32
3.12 . (a). The OH stretch region at 0 V for different water concentrations in the
CTAB-dDMSO mixture. (b). The corresponding ratio of integrated areas as a
function of applied potential. The shaded gray region represents the integrated
area of water for the pure dDMSO and CTAB, due to residual water from gold
substrate preparation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
4.1 The general concept of our experiment. (a): Tunable CW IR from the QCL and
800 nm pulse incident on a silicon wafer (gray). The red portion inside the silicon
is the region of change in refractive index caused by 800 nm absorption. (b):
The manifestation of this refractive index change in the Fabry-Perot transmission
fringes. For comparison the transmission fringes in the absence of the pump are in
blue. (c): The change between the pumped and un-pumped fringes as a function of
time for a fixed wavelength. The timescales for the decay of the intensity change
is in the order of a few µs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
4.2 FTIR spectrum of the unpumped silicon wafer. The observed fringes arise from
Fabry-Perot interference. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
4.3 A schematic of the 800 nm pump incident on the silicon wafer. The pump is a
Gaussian beam with width w traveling in the z direction. Heat is generated by the
absorption of the pump inside silicon (along the z direction). For calculation of
dissipated heat as a function of space, the ring volume element defined at radius r,
width dr and thickness dz is needed as explained in the text. . . . . . . . . . . . . 42
4.4 Simulated Fabry-Perot fringes in a silicon wafer. The blue line represents the
fringes for un-pumped silicon (∆T = 0) and the red line corresponds to the fringes
when the silicon is pumped. The pumped fringes were simulated using equation 8
using the values of E0 and w as 10 µJ and 50 µm respectively. . . . . . . . . . . . 44
ix
4.5 A schematic of our experimental setup. The 800 nm pulse comes from a
Ti:Sapphire amplifier. It is attenuated using a half wave plate (HWP) and a
polarizer. The pulse is reflected off an angled silicon dichroic and focused by a
lens (L1) onto the silicon wafer sample. The CW IR comes from the QCL (black
box), goes through the angled silicon dichroic and is focused onto the silicon
wafer sample by lens (L1). The transmitted IR is collected by L2 and sent to the
MCT detector. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
4.6 The response of the MCT detector used in the experiment. This was obtained by
directing the ∼ 50 f s pulse (800 nm) light onto the detector. The decay was fit to
an exponential and the extracted time was ∼ 2 µs . . . . . . . . . . . . . . . . . . 47
4.7 The simulated temperature change inside silicon caused by the pump as a function
of time. This profile was obtained by using equation 5 and solving the diffusion
equation as explained in the text. The contours represent a slice of the temperature
change along the x-z plane. Here, z is oriented along the direction of the thickness
of the silicon wafer. The color axis indicates the change in temperature. . . . . . . 49
4.8 Representative data from our experiments. (a) : The change in fringes plotted
against IR wavelengths spanning from 2200 cm−1
to 2300 cm−1
. This slice was
taken at t = 10 µs. (b) : Time traces of ∆I at three representative wavelengths. . . . 50
4.9 (a) : Our raw experimental data. (b): The same fringes after filtered in Fourier
space. A Gaussian filter was used to filter out the effects of QCL laser fluctuations
and fringes arising from our silicon dichroic. . . . . . . . . . . . . . . . . . . . . . 51
4.10 A comparison between the unpumped fringes seen in blue (FTIR) and the filtered
pumped fringes from the previous figure in red. . . . . . . . . . . . . . . . . . . . 51
4.11 Simulated Fabry-Perot fringes. (a): Each spectrum was computed for a different
temperature profile corresponding to each time. (b): The time dependence of the
differential transmission (difference between pumped and un-pumped) for three
wavelengths. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
4.12 (a): Differential spectra measured at various pump powers at a fixed delay. (b):
The root mean square (RMS) of the spectra plotted against the corresponding
pump energies, demonstrating a linear relationship. . . . . . . . . . . . . . . . . . 54
5.1 A pyridinium surfactant molecule with two hydroxyl groups, the counter ion is
not shown in the figure. With the hydroxyl groups, the surfactant can hydrogen
bond and bring water along when it is pulled to the electrode. The second panel is
an analogy depicting the surfactant as a molecular level water carrier. . . . . . . . 57
5.2 Potential dependent SEIRAS spectra for dDMSO alone. All the potentials are
referenced with respect to Ag/AgCl. The solid black line represents 0 V with
respect to the reference and the dotted grey line represents the baseline of each
displaced spectrum. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
x
5.3 Potential dependent SEIRAS spectra for 1 mM CTAB in dDMSO. All the
potentials are referenced with respect to Ag/AgCl. The solid black line represents
0 V wrt to the reference and the dotted grey line represents the baseline of each
displaced spectrum. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
5.4 Potential dependent SEIRAS spectra of the CTAB-dDMSO-water system. The
concentrations of CTAB and water were 1 mM and 1 M respectively. All the
potentials are referenced with respect to Ag/AgCl. The solid black line represents
0 V wrt to the reference and the dotted grey line represents the baseline of each
displaced spectrum. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.5 Potential dependent SEIRAS spectra of the CTAB-dDMSO-water system. The
concentrations of CTAB and water were 1 mM and 20 M respectively. All the
potentials are referenced with respect to Ag/AgCl. The solid black line represents
0 V wrt to the reference and the dotted grey line represents the baseline of each
displaced spectrum. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
5.6 Potential dependent SEIRAS spectra for 1 mM CTAB in H2O. All the potentials
are referenced with respect to Ag/AgCl. The solid black line represents 0 V wrt
to the reference and the dotted grey line represents the baseline of each displaced
spectrum. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
5.7 Potential dependent SEIRAS spectra for 1 mM CTAB in D2O. All the potentials
are referenced with respect to Ag/AgCl. The solid black line represents 0 V wrt
to the reference and the dotted grey line represents the baseline of each displaced
spectrum. The peaks at 2850 cm−1
and 2915 cm−1
are from CHs in CTAB. The
identity of the peak at 2750 cm−1
is still uncertain; it is where the free OD should
manifest.[190, 191] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
5.8 : Potential dependent SEIRAS spectra of the 1 mM SDS in dDMSO. All the
potentials are referenced with respect to Ag/AgCl. The solid black line represents
0 V wrt to the reference and the dotted grey line represents the baseline of each
displaced spectrum. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
5.9 Potential dependent SEIRAS spectra of the 100 mM SDS in dDMSO. All the
potentials are referenced with respect to Ag/AgCl. The solid black line represents
0 V wrt to the reference and the dotted grey line represents the baseline of each
displaced spectrum. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
5.10 Potential dependent SEIRAS spectra for 10 mM SDS in water. All the potentials
are referenced with respect to Ag/AgCl. The solid black line represents 0 V wrt
to the reference and the dotted grey line represents the baseline of each displaced
spectrum. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
xi
5.11 SEIRAS spectra of 100 mM SDS in water as a function of applied potential. An
increase in the CH features of SDS corresponds to a decrease in water. All spectra
have been displaced vertically. The grey dashed line represents the baseline for
each spectrum. The solid black line indicates 0 V verses Ag/AgCl. Note that the
concentration SDS used here is 12 times the critical micelle concentration. Even
at such concentrations, the signal is much smaller than that of CTAB at its CMC
( 1 mM). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
5.12 The differential integrated peak data for the 100 mM SDS - water system. . . . . . 92
5.13 The current verses potential plot for 1 mM CTAB-dDMSO-water systems. a. 1
mM CTAB -dDMSO 0 M water. b. 1 mM CTAB -dDMSO 10 M water. c. 1
mM CTAB -dDMSO 20 M water. d. 1 mM CTAB -water. The potential was
scanned from +0.6 V to -1 V with respect to Ag/AgCl in increments of 0.2 V.
Each potential was held for roughly 120 s. . . . . . . . . . . . . . . . . . . . . . . 93
5.14 Data obtained from bulk FTIR measurements. A: The CH stretching region
of FCHT for different bulk concentrations of FCHT in dDMSO. The two
characteristic CH stretching peaks are at 2853 cm−1
and 2929 cm−1 This data
was obtained with a calcium fluoride FTIR cell, using a path length of 100 µm.
The spectra are baseline subtracted. B: integrated area of the peaks plotted against
bulk FCHT concentration. The region of integration was 2700 – 3000 cm−1
. The
obtained slope was dA/dc= εbl = 0.07644 mM−1 = 76.44 M−1 => εb = 7644 M−1
cm−1
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
5.15 The experimental results from the FCHT monolayer on the gold ZnSe SEIRAS
substrate. A: representative cyclic voltammogram of the FCHT monolayer
after subtraction of the capacitive response at scan rate of 200 mV/s. B: The
corresponding CH stretching region of the SEIRAS spectrum. Data adapted from
reference 6. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
5.16 Data obtained from bulk FTIR measurements. A: The CH stretching region of
CTAB for different bulk concentrations in dDMSO. This data was obtained with
a calcium fluoride FTIR cell, using a path length of 100 µm. The spectra are
baseline subtracted. B: integrated area of the peaks plotted against bulk CTAB
concentration. The region of integration was 2700 – 3000 cm−1
. The obtained
slope was dA/dc= εbl = 0.1102 mM−1 = 110.2 M−1 => εb = 11020 M−1
cm−1
. . . 97
5.17 Matlab script used for interpolation of experimental data. The initial lines describe
interpolation of experimental data to fill in points. The later stages describe
interpolation of potential data to find potentials that result in the same CTAB
coverages as those corresponding to a variation of the mole fraction. . . . . . . . . 101
5.18 Simulated temperature profile inside silicon for 0.5 µJ pump energy calculated at
different times. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
5.19 Simulated temperature profile inside silicon for 1 µJ pump energy calculated at
different times. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
xii
5.20 Simulated temperature profile inside silicon for 2.5 µJ pump energy calculated at
different times. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
5.21 Simulated temperature profile inside silicon for 5 µJ pump energy calculated at
different times. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
5.22 A comparison of the transmission fringes for different pumps. (a) : The
transmission fringes plotted against IR wavelength at 5 µs. (b): A comparison
between time traces taken at 2279.5 cm−1
. While there could be discernible
differences between the two time traces, we believe that these subtle variations
stem from differences in beam divergence, focusing conditions and power between
the 800 nm and 400 nm pumps. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
5.23 The refractive index as a function of space along the body of silicon (z direction).
This was calculated by considering the pump incident on the face of the silicon
wafer (at x=0) and subsequently calculating the temperature change induced by
pump absorption. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
5.24 A representative set of data. This data was obtained from a 800 nm pump of
energy 17 µJ. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
5.25 Our MATLAB code to simulate the heat temperature profile by solving the heat
diffusion equation at different times . . . . . . . . . . . . . . . . . . . . . . . . . 107
5.26 A few time traces from the simulations. This is the same as figure 8b from the
main text. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
xiii
Abstract
Most electrochemical reactions require delivery of protons, often from water, to surface adsorbed
species. However, water also acts as a competitor to many such processes by directly reacting with
the electrode, which necessitates using water in small. Controlling the water content and structure
near the surface is an important frontier in directing reactivity and selectivity of electrochemical
reactions. In this thesis, interfacial water is probed using surface spectroscopy in binary mixtures
of water and another solvent (Ionic liquid and DMSO).
For the water - Ionic liquid studies, we used a monolayer of mercaptobenzonitrle (MBN) as a
probe of the interfacial environment. We find that even up to a significant mole fraction of bulk
water (x ∼ 0.95) the nitrile frequency does not change from that indicative of pure ionic liquid for
[EMIM][BF4], indicating preferential aggregation of the ions near the surface. Since this behavior
is very similar to surfactants, we chose an imadazolium cation with a longer side chain which
resulted into behavior expected from a surfactant, with preferential layer of the ions on the surface
even in dilute water solutions (x ∼ 0.995). This observation indicates that even those ILs that are
not nominally categorized as surfactants, have a strong tendency to aggregate at the surface. Since
ILs serve as electrolytes in a range of electrochemical reactions, including those requiring water,
our results are likely useful for mechanistic understanding and tuning of such reactions.
In the water - DMSO investigation, we did not use MBN molecules, instead we used surfactants
which had a unique vibrational signature that could be monitored. Surfactants accumulate near
surfaces, and therefore can be used as agents to control interfacial water. Using mid-IR spectroelectrochemistry, we show that a modest concentration (1 mM) of the cationic surfactant CTAB
in mixtures of 10 M water in an organic solvent (dDMSO) has a large effect on the interfacial
xiv
water concentration, changing it by up to 35% in the presence of an applied potential. The major
cause of water content change is displacement due to accumulation or depletion of surfactants
driven by potential. Two forces drive the surfactants to the electrode – the applied potential and
the hydrophobic interactions with the water in the bulk. We have quantified their competition by
varying the water content in the bulk. To our knowledge, for the first time, we have identified the
electrochemical equivalent of the hydrophobic drive. For our system a change in applied potential
of 1 V has the same effect as adding 0.55 mole fraction of water to the bulk. This work illustrates
the significance of surfactants in partitioning of water between the bulk and the surface, and paves
the way towards engineering interfacial water structure for controlling electrochemical reactions.
xv
Chapter 1
Introduction
With anthropogenic energy consumption at an all time high, long term viable sources like renewable ones are gaining more appeal. The transition to these sustainable options will likely be a
gradual one - this would practically look something like weaning off our current non-renewable
energy source (fossil fuels). Utilizing fossil fuels for energy involves generating greenhouse gases
like methane and carbon dioxide as by-products which substantially contribute to global warming.[1, 2]
Many research endeavors aim to reduce these greenhouse gases/ re-purpose them for other
uses. Specifically, in the case of carbon dioxide, current approaches involve capture and storage or
reduction to greener compounds.[3, 4] Reduction of carbon dioxide can further be divided into two
general approaches ; electrochemical means or molecular catalysts.[5–8] Both approaches towards
carbon dioxide reduction require a source of protons. [9, 10]
The work presented in this thesis aims to understand the role of water at a metal-electrolyte
interface under externally applied potentials.
Water plays an important role in many electrochemical reactions. It has a dual purpose of a
solvent as well as a reactant. During CO2 reduction a source of H+ is required which is supplied
by water. Other relevant electrochemical reactions in which water serves as a reactant include :
nitrate reduction REF, alcohol oxidation REF and hydrogen evolution.
Controlling electrochemical reduction reactions which use water is challenging as water has a
competing reduction reaction (hydrogen evolution). There have been many endeavors to alter the
1
properties of water at the an electrode surface in order to favor specific electrochemical reaction
pathways. Such efforts very broadly consist of either changing the chemical composition of the
system or changing the properties of the metal which serves as the electrode.
Spectroscopy is a powerful tool for probing the electrochemical interface. It provides mechanistic insight into Faradaic processes.
1.1 Thesis Outline
This thesis is divided into three chapters. Chapter 1 deals with water partitioning at an electrode
surface. Water partioning is studied using binary mixtures of water and ionic liquids (ILs), this
allows us to regulate the water concentration in the bulk. The water at the surface was studied by
using a monolayer of probe molecules (mercaptobenzontriles). The nitrile frequency of the probe
molecule had different values in water and ILs. Thus, monitoring the nitrile frequency of MBN as
a function of water content in the IL-water served as a proxy for interfacial water. A caveat of this
study is the alteration of the metal surface in introducing our monolayer of probe molecules.
To surpass this limitation, a similar experiment using water - DMSO binary experiments was
performed. We added charged surfactants as a means to change the interfacial content of water and
DMSO when an external potential was applied. Instead of using a probe molecule, we observed
each component of this system directly using surface enhanced Infrared absorption spectroscopy
(SEIRAS). This is described in Chapter 2.
Chapter 3 describes the development of a photothermal spectroscopic technique that can be
extended to the electrochemical interface. We probed the heat generated in a silicon wafer using
a CW IR laser. The chapter describes the theory of the photothermal effect for the silicon wafer.
Photothermal spectroscopy like this could be used to study interfacial processes, we are currently
working towards extending it to ATR electrochemical processes.
2
Chapter 2
Ionic Liquid - Water Binary Mixtures
2.1 Summary
Understanding ionic structure and electrostatic environments near a surface has both fundamental
and practical value. In electrochemistry, especially when room temperature ionic liquids (ILs) are
involved, complex ionic structure near the interface is expected to crucially influence reactions.
Here we report evidence that even in dilute aqueous solutions of several ionic liquids, the ions
aggregate near the surface in ways that are qualitatively different from simple electrolytes. We
have used a vibrational probe molecule, 4-mercaptobenzonitrile, tethered to a metal surface to
monitor the behavior of the ionic layers. The characteristic nitrile vibrational frequency of this
molecule has distinct values in the presence of pure water (∼ 2232 cm−1
) and pure ionic liquid
(for example, (∼ 2226 cm−1
) for ethylmethyl imidazolium tetrafluoroborate [EMIM][BF4]). This
difference reflects the local electrostatic field and the hydrogen bonding variations between these
two limiting cases. We tracked this frequency shift as a function of IL concentration in water
all the way from pure water to pure IL. We report two important findings. First, only one nitrile
peak is observed for the entire concentration range, indicating that, at least on the length scale
of the probe molecule, water and ionic liquids do not phase separate within the interface and no
heterogeneously distinct electrostatic environments are formed. Second, and more importantly,
we find that even up to a significant mole fraction of bulk water (x ∼ 0.95) the nitrile frequency
does not change from that indicative of pure ionic liquid for [EMIM][BF4], indicating preferential
3
aggregation of the ions near the surface. Since this behavior is very similar to surfactants, we
chose an imadazolium cation with a longer side chain which resulted into behavior expected from
a surfactant, with preferential layer of the ions on the surface even in dilute water solutions (x
∼ 0.995). This observation indicates that even those ILs that are not nominally categorized as
surfactants, have a strong tendency to aggregate at the surface. Since ILs serve as electrolytes in
a range of electrochemical reactions, including those requiring water, our results are likely useful
for mechanistic understanding and tuning of such reactions.
2.2 Introduction
Room temperature ionic liquids (ILs) are organic salts that are liquid at room temperature. They
have a range of properties including high vapor pressure, high electrochemical stability, and the
ability to dissolve a wide range of substances, that make them interesting for many applications as
solvents and electrolytes [11–13]. Due to high ion concentration, non-negligible ionic sizes and
shapes, and strong Coulomb interactions, they exhibit some behavior that are at odds with conventional understanding of most dilute electrolytes and liquids [14]. Of particular interest is the
interaction of ILs with solid surfaces, which is directly important in their electrochemical applications. More specifically, aqueous solutions of ILs near an interface are of both fundamental and
practical interest for all electrochemical reactions that require water.
Due to packing and electrostatic constraints, the structure and energetics of ILs at the surface
can be different from the bulk. This is especially true near a metallic interface, where ions can interact with their image charges, in addition to being subjected to packing constraints and Coulomb
interaction between ions. These interactions result into complex interfacial structures that are often
difficult to measure and model. For mixtures of IL with water the nature of the interfacial phase
can become more complicated. For example, it is known that most ILs are hydrophilic, leading
them to absorb ample quantities of water vapor at ambient conditions [15, 16]. This gives rise to an
interesting question. Since the bulk of most IL can easily accommodate water, if a small quantity
4
of water is introduced in the IL, how much of it, if any, will appear at its interface with a metal?
Similarly if the situation is reversed and a small amount of ionic liquid is added to a large quantity
of water, will the ionic liquids preferentially aggregate near the surface? How much water should
be added to the bulk before the aggregation of ions near the surface is disfavored? Our work aims
to address such questions.
At a practical level, transport of water to the interface is important for electrochemical reactions
in which water is a reagent. For example, in most schemes of CO2 reduction, water serves as a
source of protons, making partitioning of water between bulk and interface critical to the reaction
mechanism [17]. Therefore, probing the physical and chemical environment of the interface can
help with rational design of electrolytes.
Here we focus on experiments that involve mixtures of ILs with other molecules, especially
water, at solid interfaces. A relatively recent review of water at ionic liquid interfaces [18] highlights the most recent developments. A range of experiments, mostly Atomic Force Microscopy
(AFM) and Surface Force Balance (SFB), have proven effective in identifying the layered structure
of the ions at the interface. In these experiments, the ionic structure manifests itself as oscillatory
force pattern with respect to distance as two surfaces are brought close to each other with IL layers
in between. Some experiments, point to the existence of two phases for IL near an interface as a
function of water concentration [19, 20]. At low water concentration, or the “water in IL phase”
the structure may be thought of as predominantely dictated by the properties of the IL, while at
the high water concentration, or the “IL in water phase”, one approaches the properties of a dilute
aqueous electrolyte. The existence of an abrupt phase change at the interface when IL are diluted
in a high dielectric solvent, propylene carbonate, has been reported by Perkin [21]. Using SFB
measurements, they show that addition of propylene carbonate does not affect the properties of the
interface significantly until a threshold of nearly 60% mole fraction of solvents is reached. After
that concentration, an abrupt change in the force profile indicates crossing from the layered ionic
phase to the dissolved state of the interface. At the junction of a metal one may expect stronger
forces between ions and their image charges which can enhance the stability of the interfacial
5
structure of pure IL. Therefore, similar to the case of propylene carbonate, it is plausible to expect
the interfacial ionic layers near a metal to persist even when their concentration in the bulk is quite
small.
The solid-like behavior of IL in nanoconfined spaces and its dependence on the water content
has been reported by others as well [20, 22]. In particular, it has been reported [20] that addition
of 15wt% water to [BMIM][BF4] disrupts the ionic structure at a silica interface and creates a dissolved phase. Near mica, however, such an effect was not observed. This difference points to the
fact, that even within dielectrics, ionic structuring can be quite varied. Therefore, while one may
hypothesize similar behavior near a metal, quantitative extrapolation from the case of dielectrics
should be avoided, since ionic interactions with their image charges adds an extra degree of complexity. We have previously reported aggregation of surfactants at very low bulk concentrations
near a SAM on metal and their electrostatic influence on the interface [23]. We will later show in
this work that the behavior of some ionic liquids is akin to such surfactants.
Infrared spectroscopy of IL-water mixtures have been carried out extensively, with attention
to both steady state and dynamic behavior [24–32]. Dynamics of probe molecules, such as CO2,
as reporters of their solvation environment in ionic liquids have been studied using 2D-IR [33–
37]. This literature highlights the complexity of solvation in ionic liquids in the bulk. One should
be cautious in extrapolating from the bulk studies to the interface, where a different balance of
packing and electrostatics complicates the picture. The interface of ionic liquids with air and with
electrodes have also been studied using Sum Frequency Generation (SFG) spectroscopy [38–44].
However, the behavior near a metal, in particular with explicit regards to measuring electrostatic
properties as a function of water content has not been reported previously.
In this work, both our problem and our approach are different and complementary to the other
methods described above. First, we are interested in the behavior at the metal-IL interface, in
contrast to dielectric-IL interface. Second, we use a well-known vibrational chromophore, 4-
mercaptobenzonitrile (MBN) covalently tethered at the metal, to study this elusive environment.
6
Since many properties of this chromophore, including its response to electrostatic fields and hydrogen bonding is well-documented by us and others [45–49] it provides a unique window for
understanding the interfacial electrostatics that is otherwise very difficult to access. Furthermore,
these chromophores are covalently bound at the interface, and there is no ambiguity about their
distance from the surface. Therefore, averaging the vibrational signal over several layers as is often the case in SFG spectroscopy of liquids at interfaces [50, 51] is not a concern. As will be shown
in this work, the nitrile vibration of the probe as a function of bulk water content, reveals that the
ionic liquids studied here favor the interface disproportionately even when their bulk concentration
is quite low.
--
-
A
ΔωCN
0 1
B
C
-
-
A: Pure IL B: IL-Water Mixture C: Pure Water
Figure 2.1: Cartoon showing the concept of the experiments in this paper. The nitrile probe frequency changes due to electrostatic environments and hydrogen bonding variations at the interface.
We track how the frequency varies between the limits of pure ionic liquid and pure water. As will
be shown, this variation is vastly different from a linear interpolation between these limits, and
informs us about the tendency of ions to aggregate at the interface.
7
2.3 Experimental Methods
Ag substrates for SERS were prepared by chemically etching the Ag foils [52] . Prior to etching,
the Ag foils were sonicated in HPLC water for 30 minutes. The washed Ag foils were treated with
28% NH4OH for 30 seconds that helped remove dirt and smoothen the surface. Next the foils were
immersed in ∼ 6 M HNO3 acid for 10 seconds that roughened the substrate and made it SERS
active. SEMs images of the roughened surface structures are shown in figure Figure 2.2. It is to be
noted that in a SERS experiment, majority of the signal stems from a small number of molecules
that ’enjoy’ significantly larger enhancement than the rest.
10 m 10 m 100 nm
Figure 2.2: SEM images of the roughned silver surface.
The SERS active Ag substrate was washed with water and soaked in 0.02 M solution of 4-
mercaptobenzonitrile in ethanol for overnight. After soaking in 4-MBN, the Ag substrates were
sonicated for 90s in ethanol, prior to using them for Raman experiments. The resolution for the
SERS and FTIR measurements were nominally 1 cm−1
and 0.25 cm−1
respectively.
The Raman excitation was a 532 nm laser (Ocean Optics Inc.). A dichroic mirror was used to
guide the Stokes shifted light to the detector which was a 320 mm focal length spectrometer with
1800 lines/mm gratings (Horiba iHR320) and a 1024 × 256 CCD array (Horiba Syncerity).
8
N N
+ B F
F
F
F
N N
+ B F
F
F
F
N
+ N B F
F
F
F
1-Ethyl-3-methylimidazolium tetrafluoroborate ( [EMIm][BF4
] )
1-Hexyl-3-methylimidazolium tetrafluoroborate ( [HMIm][BF4
] )
1-Dexyl-3-methylimidazolium tetrafluoroborate ( [DMIm][BF4
] )
Figure 2.3: The chemical structures and acronyms for the ionic liquids used in this study.
The ionic liquids used in this study all had [BF4]
− anions and 3-methyl-1-alkylimidazolium
anions, with alkyl chain 2-, 6-, and 10-carbons long ([EMIM]+, [HMIM]+, [DMIM]+). The
ionic liquids were purchased from Sigma Aldrich with purities higher than 98%.The ionic liquids
were stored under moisture-free air and dried before measurement using a microwave purification
method adapted from Ha et al [53]. This method was shown to remove water to levels below 0.5
wt% rapidly and without damage to the ions. In short, aliquots of ionic liquid were heated in a lab
microwave until ionic liquid temperatures reached 120 °C . A Vertex Bruker 80 FTIR instrument
was used for the bulk IR measurments.
2.4 Results
Prior to discussing the interaction of ionic liquids with SAM at the interface, it is important to
consider their interaction with benzonitrile in the bulk. The frequency shift of benzonitrile in
aqueous solutions of two ionic liquids, [EMIM][BF4] and [HMIM][BF4], is shown in figure Figure 2.3. Since in the range of concentrations studied, water phase separates when interacting with
[DMIM][BF4], we could not obtain any bulk data. Similarly, benzonitrile does not dissolve in pure
9
Figure 2.4: Representative FTIR spectra for the [EMIM][BF4] and [HMIM][BF4] dilution series.
-
-
-
-
-
-
-
-
Figure 2.5: (a) The nitrile frequency of benzonitrile in different ionic liquids as a function of
water content. The data presented is for [EMIM][BF4] and [HMIM][BF4]. The dashed line in
black represents a hypothetical ideal solution where ωCN changes linearly as a function of water
concentration. ωCN for pure water (χwater = 1) was not collected as benzonitrile does not dissolve in
pure water. All frequency shifts are with respect to the nitrile frequency in a dry sample (in contact
with air) with no solution. The error bars are frm (b) A schematic of the solvation structure around
benzonitrile in low and high water concentrations. The data suggests that in water-IL mixtures the
probe molecule is favorably solvated by ions.
10
Figure 2.6: Representative SERS spectra.(a): [EMIM][BF4] dilution series.(b) : [HMIM][BF4]
dilution series.(c): [DMIM][BF4] dilution series.
water, which limits the axis of the figure to 0.9 in mole fraction. All frequency shifts are with
respect to the nitrile frequency in a dry sample (in contact with air) with no solution.
As seen in the figure, the nitrile frequency in aqueous solutions of [EMIM][BF4] does not
obey a straight line interpolating between the characteristic frequency in water and in pure ionic
liquids. The shape of the line suggests that the IL tends to encapsulate the benzonitrile in a shell
which resists getting dissolved by the added water. Shortly, we will discuss that this effect is far
more pronounced at the interface, with the nitrile frequency not changing significantly until values
above 0.95 mole fraction of water. For [HMIM][BF4] in the bulk this caging effect is likely far
more pronounced. Even in the presence of 0.9 mole fraction of water, the probe frequency tends to
remain far below the expected frequency from solvation in water. With 6-carbon long side-chain,
the IL likely behaves as a surfactant and perhaps dissolves the nitrile in a micellar structure. As we
will discuss below, this behavior is extended to the surface as well.
The frequency of adsorbed MBN is very close to that of the pure benzonitrile, consistent with
previous observations (within 1 cm−1
) .Results for the surface measurements are shown in figure
Figure 2.4. From the figure insets it is observed that even at very high mole fractions of water in the
bulk the frequency shift for the nitrile lingers close to that of the pure IL. It takes an exceptionally
large amount of water (in excess of 0.95, 0.99 and 0.998 mole fraction of water for [EMIM][BF4],
[HMIM][BF4] and [DMIM][BF4] respectively), to enforce an environment at the interface that
even begins to have properties similar to water. This suggests strong aggregation of the IL at the
11
Figure 2.7: The nitrile frequency at the interface as a function of water mole fraction for three different ionic liquids. The ionic liquids were [EMIM][BF4] and [HMIM][BF4] and [DMIM][BF4].
The inserts display a magnified version of the same data. All frequency shifts are with respect to
the nitrile frequency in a dry sample (in contact with air) with no solution.The error bars represent
errors from fitting the spectra to Gaussian line-shapes. The shaded rectangle represents the measurement error arising from substrate variability.
12
1 M 0 M
KCl Concentration
Conventional
inorganic salt KCl
Ionic Liquid
[EMIM][BF4]
Pure 0 M
IL Concentration
(a) (b)
Figure 2.8: (a): SERS data of the central nitrile frequency of MBN as a function of EMIm BF4
concentration, this is the same data from the main text. (b): SERS data of the central nitrile
frequency of MBN as a function of KCl concentration.The nitrile stretching frequency is referenced
to the nitrile frequency in the absence of any solvent.
interface, even when the bulk is overwhelmingly composed of water with little IL. Expressed in
terms of concentration of IL, rather than water mole fraction, it seems that even in solutions as
dilute as 60 mM of [EMIM][BF4], the ions tend to favorably cover the interface.
A second observation is that pure ionic liquids in the bulk shift the nitrile frequency to a smaller
degree compared to their surface analogs. For [EMIM][BF4] and [HMIM][BF4] the frequency shift
in the bulk is ∆ωCN ∼ 1 cm−1 whereas it is ∆ωCN ∼ 3.5 cm−1
at the interface. This may also be
an indication of stronger solvation layer at the interface with higher ion density, compared to the
solvation shell in the bulk.
Another noteworthy comparison is between the effect an IL-water mixtures and a common
electrolyte like KCl dissolved in water. This is seen in figure Figure 2.8, which shows the effect
of both IL (EMIm BF4) and KCl on MBN as a function of water dilution. ILs provide a means of
accessing a wider regime of salt concentrations from the salt-in-water to water-in salt regimes.
13
2.5 Conclusion and Discussion
There are two major outcomes from the above results. The first important outcome is that even
ionic liquids that are not nominally considered surfactants, in this case [EMIM][BF4], can have
strong propensity to accumulate at the interface in dilute solutions. We note that when conventional
simple inorganic salts, such as KCl, are introduced in these experiments not a significant change
in nitrile frequency is measured for concentrations even as high as 1 M figure Figure 2.8. This
behavior is quite distinct from that of simple electrolytes. It is a reminder that ionic liquids not
only manifest non-traditional behavior when they are pure, i.e. large ionic concentration, but can
also exhibit unexpected behavior at low concentrations.
It is interesting to compare this behavior to surfactants. In a previous work, we have reported
accumulation of the well-known alkyl-ammonium family of surfactants, on a benzonitrile-covered
metal surface and have studied their influence on the interfacial electrostatic environment as probed
by nitrile frequency shifts. It was observed that R-dimethyl ammonium cations, with chain length
of R as 8, 12, and 16 carbons, all form fully assembled layers on the surface in concentrations
beyond 50 mM[23] Therefore, we expected the [DMIM] and [HMIM] to have some level of similarity to those surfactants. This was consistent with other studies [54, 55]. However, for the much
shorter chain cation [EMIM][BF4] to exhibit such strong surface presence was not anticipated by
us. Those employing ionic liquids in applications must note that even in small concentrations, ILs
can have surfactant-like behavior.
14
Figure 2.9: The linewidths of the nitrile stretch (benzonitrile) as a function of molfraction for
EMIm BF4 and HMIm BF4. These are for the bulk dilution and were obtained through FTIR.
Figure 2.10: The linewidths of the nitrile stretch (4-mercaptobenzonitrile monolayer) in the presence of different concentrations of EMIm BF4, HMIm BF4 and DMIm BF4. For a point of reference, linewidths of the substrate in absence of any ILs are 8.5 cm−1
.
The second important observation is the existence of a single peak, with no systematic and discernable change in line width as a function of ionic liquid concentration. seen in figures Figure 2.9
and Figure 2.10. This result seems counterintuitive, as one may expect some level of heterogeneity
in the interfacial electrostatic environment at low ionic concentrations. That is, some of the probe
molecules may interact directly with an ion, while some others may interact with water molecules.
15
However, consistent with our previous work on surfactants [23], it seems that even in such circumstances a single benzonitrile molecule feels an electrostatic environment that effectively averages
the influence of several nearby ions and water molecules. The extent, i.e. the radius, of this
averaging can not be directly inferred from our measurements. We anticipate that detailed molecular dynamics simulations coupled with electronic structure and vibrational frequency calculations
could shed more light on this phenomenon.
It is important to note whether the drive for the ions to accumulate at the interface is partially
due to the benzonitrile SAM. It may be recommended to do away with the SAM layer entirely in
such experiments. We rely on the SAM, since we wish to measure the electrostatic environment
with a well-calibrated probe, and reasonable orientation information with respect to the surface.
One may directly measure the IL vibrational signatures at the surface, which is possible, albeit with
some degree of complexity and have been reported previously [56–61]. However, in that case, two
important factors are highly uncertain. The first is the orientation of the ionic liquid molecules
near the surface. Second the penetration depth of the signal into the bulk can cause ambiguities
about the number of molecular layers at the interface that contribute to the signal, even for the most
surface-sensitive measurements such as SFG. Therefore, all things considered, while using a SAM
of probe molecules may be thought of as intrusive, it is quite an incisive and complementary tool
to the existing methods to study the interface.
The interaction of ionic liquids with surface may comprise of at least three components. The
charge-metal (image charge) interactions, the charge-dipole interactions, and the hydrophobic interactions. The hydrophobic interaction between the ionic liquids and the SAM can be strong, and
it is possible for the ions to intercalate within the monolayer.
Our SAM monolayers has a dipole of nearly 4.5 D per molecule pointing away from the surface.
Part of the interactions of the ionic liquids with the interface may be due to a charge-dipole attraction, and could explain the accumulation of positively charged ions in all cases at the interface.
However, one must not neglect the attraction of the ions to their image charges in the metal, which
even though they are effectively farther away, contribute with a possibly stronger charge-charge
16
attraction. When an ion is near a metal interface, its solvation necessarily becomes asymmetric.
This is not to say that the liquid side of the solvation sphere becomes negligible. The ion is still
heavily stabilized by the liquid side of the interface. However, on the metal side, it necessarily has
a stabilizing interaction with the image charge. The solvation structure at the end is the result of
a balance between these forces (and other specific interactions). Comparing the charge-dipole and
charge-image charge interactions is quite important in this case, and without detailed simulations
along with carefully crafted experiments with varying SAM thickness and precise control over size
of ions is not possible.
Another noteworthy point regarding the degree to which the SAM changes the metal-IL interface can be seen in reference 50. In that work it was demonstrated that [EMIM][Cl] inhibits HER
on bare metal electrodes (Ag,Cu,Fe) at concentrations of 100 mM. This suggests that even in the
absence of SAM, some ILs tend to aggregate or perhaps adsorb on the surface.
While our work shows preferential accumulation of ILs on the surface, it is difficult to determine from our measurements the ionic arrangements near the monolayer. It has been shown that
small anions such as Cl− can adsorb on metals or penetrate SAMs [62], observation of the nitrile
probe alone can not deliver such detailed information.
In summary, this work highlights yet another intricate aspect of ionic structure at the interface. We hope that the knowledge from this work can help understand electrochemical reaction
mechanisms in the presence of ionic liquids or their aqueous solutions.
17
Chapter 3
Regulating Interfacial Water Content using Surfactants
3.1 Summary
Most electrochemical reactions require delivery of protons, often from water, to surface adsorbed
species. However, water also acts as a competitor to many such processes by directly reacting with
the electrode, which necessitates using water in small. Controlling the water content and structure
near the surface is an important frontier in directing reactivity and selectivity of electrochemical
reactions. Surfactants accumulate near surfaces, and therefore can be used as agents to control
interfacial water. Using mid-IR spectro-electrochemistry, we show that a modest concentration (1
mM) of the cationic surfactant CTAB in mixtures of 10 M water in an organic solvent (dDMSO) has
a large effect on the interfacial water concentration, changing it by up to 35% in the presence of an
applied potential. The major cause of water content change is displacement due to accumulation or
depletion of surfactants driven by potential. Two forces drive the surfactants to the electrode – the
applied potential and the hydrophobic interactions with the water in the bulk. We have quantified
their competition by varying the water content in the bulk. To our knowledge, for the first time,
we have identified the electrochemical equivalent of the hydrophobic drive. For our system a
change in applied potential of 1 V has the same effect as adding 0.55 mole fraction of water to the
bulk. This work illustrates the significance of surfactants in partitioning of water between the bulk
and the surface, and paves the way towards engineering interfacial water structure for controlling
electrochemical reactions.
18
3.2 Introduction
Water is ubiquitously important in chemistry, including electrochemical reactions.[63].It not only
serves as a solvent but also as a reactant, such as a Bronsted acid or base. Many electrochemical reactions of importance for energy conversion such as CO2 reduction,[64–66] alcohol oxidation,[67–
69] and nitrate reduction [70, 71] require water as either a source or sink of protons. In some cases,
water competes with an electrochemical reaction by directly interacting with the electrode through
hydrogen or oxygen evolution reactions.[72–74] Since water both acts as a needed reagent and as a
competitor, it is important to control its content near an interface. In principle, it is even necessary
to control the details of the water structure to affect its reactivity near the surface. The details of
water delivery to enzyme reactive centers is a vibrant field of biochemistry.[75–78] Concepts such
as proton wires,[79–82] structured water,[83–85] and pKa gradients,[86, 87] have been invoked
in the function of several enzymes. Therefore, it is reasonable to take steps towards controlling
water at electrochemical surfaces for the purpose of tuning the reactivity and selectivity of electrocatalysts.[87–101] Similarly, the details of water structure near metal oxide surfaces is known to
be important for environmentally relevant reactions and have been a subject of continued study of
nonlinear surface spectroscopy.[102–107] Using water as a minor component in binary mixtures
with organic solvents is a way of influencing electrochemical reactions.[103, 108–113] However,
controlling the partition of water between the bulk and the interface, especially in the presence of
electrochemical potential is a complicated challenge. An approach to address this problem is by
using surfactants, which often have high surface affinity at low concentrations and therefore are
good candidates to influence the chemistry of surfaces. Surfactants have been extensively studied
by electrochemists and surface scientists. It has been reported that they can control the electrostatics of the interface,[114, 115] the selectivity of chemical reactions [116–118] and suppression
of reactions such as corrosion.[119, 120] Therefore, as a natural extension we use surfactants to
control the water partition between the bulk and the interface.
We report the influence of surfactants with both positive and negative heads groups on the
partitioning of water between the bulk and the surface in binary mixtures of water with an organic
19
solvent dDMSO, as shown in the conceptual cartoon in figure Figure 3.1. Using surface enhanced
IR absorption spectroscopy (SEIRAS) [121–124] at an electrode surface, we could observe the
spectral signatures of all three components in the system – water, dDMSO, and surfactants. This
allowed us to track their relative enrichment as a function potential. We quantify the surfactant
accumulation/depletion and the correlated depletion/accumulation of water and dDMSO near the
surface. Furthermore, we provide experimentally obtained estimates of the surface coverages of
the surfactants. Our work is useful for the broad and important range of reactions where water is
used as a stoichiometric reactant in a background solvent in electrochemistry. It is the first step
towards controlling the water content, with the future goal of enforcing details of water structure
at the interface.
Figure 3.1: A cartoon depicting the concept of our experiment. The vibrational signatures of the
surfactants (green) and water (blue) were tracked as a function of potential and bulk water mole
fraction. The bromide anion of CTAB is not shown.
20
3.3 Experimental Methods
All FTIR and SEIRAS spectra were collected using a ThermoFisher Nicolet iS50 FTIR spectrometer. The spectra were averaged for 128 scans, collected at a resolution of 4 cm−1
and detected
using a liquid nitrogen cooled mercury cadmium telluride (MCT) detector. A Gamry 1010B potentiostat was used for the electrochemical measurements. For the potential dependent experiments,
each potential was held for 2 minutes and FTIR data was collected after the current had reached
steady state, ∼ 10 s after the potential step. The chronoamperometry graphs for all our SEIRAS
experiments are shown in the appendix, figure Figure 5.13.
The SEIRAS substrates were prepared using an adapted method from Bao et al. [125] Briefly,
the ZnSe crystal (Pike technologies, 60◦
cut) was polished with successively finer alumina slurries
- 1 µm diamond slurry for 1 minute and 0.05 µm diamond slurry for 1 minute. The crystal was
then thoroughly rinsed with HPLC water. For the gold deposition, the ZnSe crystal was heated to
∼ 80◦C on a hot plate and a 10 mM gold chloride solution was added on the top crystal face for ∼
20-30 s. The reaction was terminated by washing the crystal face with HPLC water. The substrate
was checked for conductivity with a multimeter. The typical conductivities for the substrates used
in our measurements were between 10-50 Ω across the length of the crystal face (∼ 1 cm). We
note that there is a trade-off between high spectroscopic signal and conductivity. A thinner, less
conductive gold layer has more pronounced SEIRAS signal compared to a thicker gold substrate
that has higher conductivity.
The SEIRAS experiments were carried out in an attenuated total reflectance (ATR) geometry
with a VeeMax III FTIR accessory (Pike Technologies). All the SEIRAS spectra were collected at
p polarization. Prior to the SEIRAS spectrochemical measurements, the substrates were checked
for opacity in the infrared by taking a spectrum of ethanol on the substrate. For substrates which
had a thick gold layer, no ethanol features were observable due to significantly diminished infrared
penetration through the gold, and therefore were not used. A spectrum of the SEIRAS substrate in
the absence of solvent was also collected and used as a background.
21
Cetyltrimethylammonium bromide (CTAB) and sodium dodecyl sulfate (SDS) were purchased
from Sigma Aldrich, 99.9% purity.Deuterated dimethyl sulfoxide (dDMSO) was obtained from
VWR ( 99.9% purity). MilliQ water was used for the experiments. Gold (III) chloride trihydrate
was purchased from Sigma Aldrich (99.9 % purity).
3.4 Results and discussion
To orient the reader, we will first discuss the characteristic spectral features of the three components
of our system - water, surfactant, and dDMSO. The SEIRAS spectra in the signature regions for
dDMSO, CTAB, and water at 0 V vs. Ag/AgCl are shown in figure Figure 3.2. The spectra are
referenced to the SEIRAS substrate (gold on ZnSe) in contact with air. For these spectra, the
concentration of water was 10 M and that of CTAB was 1 mM in dDMSO as the dominant solvent.
The water spectrum corresponds to the OH stretch and is blue shifted (3550 cm-1) with respect
to the expected peak for pure bulk water (3480 cm-1).[126, 127] This behavior is expected for
water diluted in organic solvents.[128] The signature of CTAB is the symmetric (2850 cm-1) and
antisymmetric (2929 cm-1) CH stretches of the aliphatic chain.[129] The dDMSO signature peaks
in the 1050 cm-1 region are previously assigned to the SO stretch and its combinations with other
modes in the molecule.[130]The dispersive feature in the lower frequency side does not change
with potential as will be seen shortly and will not be considered in our analysis. Observations of
such dispersive features in SEIRAS is common and has been reported and explained before.[131–
133]
Next, we present the potential dependence of these spectral features in figure figure Figure 3.3
for 10 M water and 1 mM CTAB in dDMSO. These spectra show the differential absorbance ∆A at
a given potential relative to absorbance at 0 V vs. Ag/AgCl. A negative ∆A indicates depletion of
the species while a positive ∆A implies its enrichment relative to 0 V. The CTAB signal distinctly
shows enrichment in negative potentials and depletion in positive potentials. This is consistent with
the positive charge of the CTAB head, which is expected to be pulled towards (or pushed away
22
(a) (b) (c)
Figure 3.2: Interfacial spectral signatures of the three component system of 1 mM CTAB in 10 M
water in dDMSO at 0 V vs. Ag/AgCl. (a) SO stretch of dDMSO, (b) CH stretch of CTAB, and (c)
OH stretch of water referenced to gold on ZnSe.
from) the interface at negative (or positive) potentials. Our interest, however, is mainly in how the
CTAB motion is correlated with water and dDMSO content. As is evident from the OH stretch
region, the enrichment/depletion of CTAB is anti-correlated with the water content. That is, when
CTAB is pulled into the surface, water is likely displaced from the interface. Therefore, CTAB
serves as a handle for controlling the interfacial water content in an organic-water mixture. Note
that this control of interfacial water is achieved by a modest concentration of 1 mM of surfactant.
Later in this paper, we will discuss how this control is affected by the total water content in the
mixture.
We cannot completely decouple the water potential-dependent re-orientation from changes in
water concentration at the surface. Previous work by Osawa and coworkers have shown that the
water spectrum in a simpler electrolyte (500 mM perchloric acid) shows different line shapes as a
function of potential, which was taken as evidence for water structuring and re-orientation [98]. In
our case, a much smaller concentration of surfactants (1 mM) can induce water spectral changes.
We also observe possible signatures of free OH, which can be taken as water restructuring. However, in general, our line shapes do not show drastic variations as a function of potential. Note
that this does not rule out water structuring and re-orientation. Water in the vicinity of organic
long chains is known to acquire a structure different from the bulk. Our data does not afford the
23
Wavenumber (cm-1
)
2800 3000 3200 3400 3600
+ 0.6 V
+ 0.4 V
+ 0.2 V
- 0.2 V
- 0.4 V
- 0.6 V
- 0.8 V
- 1.0 V
900 1000 1100 1200
Wavenumber (cm-1
)
dDMSO Surfactant
Water ∆
(
)
∆
(
)
10
−
10
−
Figure 3.3: Potential-dependent SEIRAS data for the CTAB-dDMSO-water (10 M) system with
the corresponding spectral signatures labelled. All spectra have been displaced vertically. The grey
dashed line represents the baseline for each spectrum. The solid black line indicates 0 V verses
Ag/AgCl. Enrichment of CTAB corresponds to depletion of dDMSO and water, and vice versa.
24
resolution or specificity to assign such structures. When we refer to water content or concentration, both orientational and concentration effects are included. Perhaps future work, with tailored
surfactants, can induce water structuring in a controlled and measurable way.
We also noted two small peaks at ∼ 3750 cm-1 that appear at positive potentials. In mixtures of
water with organics, there have been reports of peaks in this region which have been attributed to
‘free’ OH stretches as the diluted water loses hydrogen-bonding partners [134, 135]. To understand
the origin of these peaks, we repeated the experiment in deuterated water and saw similar features
in the OD stretching region (Appendix figure Figure 5.7). Therefore, it is plausible to conjecture
that these peaks could correspond to free OHs in different interfacial local environments. Further
experiments, perhaps with more surface sensitive techniques such as Sum Frequency Generation
(SFG), are required to verify this assignment.
The spectral signature of dDMSO is also anticorrelated with that of CTAB, indicating its displacement by the surfactant. There are no spectral shifts with potential, and as noted before, the
dispersive part of the spectrum (figure Figure 3.2) is not potential-dependent.
To assess the potential dependence of the three interfacial species, we integrated their spectral
signatures across their corresponding spectral regions. Regions of integration were 3150 cm-1 to
3590 cm-1 for water, 2800 cm-1 to 3000 cm-1 for CTAB, and 940 cm-1 to 1150 cm-1 for dDMSO.
The integrated spectral change is denoted by ∆A and when divided by the integrated spectrum A
at 0 V indicates the relative change of the population of a species as a function of potential. This
quantity is shown in figure Figure 3.4 for the three species. As seen in the figure, the relative
change in the CTAB population is the largest compared to the other components. At -1 V the ratio
of CTAB is ∆A
A ≈ 1, implying that the concentration of interfacial CTAB has nearly doubled. The
concentration of water is correspondingly reduced by 35%. Note that the bulk concentration of
CTAB and water are 1 mM and 10 M respectively. The important message of this figure is that
even such a modest concentration of surfactant can induce about 35% change in the interfacial
water content with application of an appropriate potential. The induced change in the dDMSO
25
content is about 11% at this potential. The opposite effect is observed at positive potentials, where
there is depletion of surfactant and increase of the water and dDMSO content.
Figure 3.4: The differential integrated peak data for the 1mM CTAB-dDMSO-10 M water system.
The reversibility of the potential scans is shown in figure Figure 3.5 . Overall, a slight hysteresis of the surfactant peak is observed. After negative potentials are applied, the overall surfactant
coverage increases. This may originate from the surfactants forming a more ordered layer at negative potentials which is harder to break with reversing the potential. Therefore, the double layer
structure is not uniquely determined by the applied potential, and depends on the history of the
scan. This observation highlights that assuming equilibrium theories for the double layer structure
in the presence of surfactants may be unsuitable.
The CTAB - dDMSO - 10 M water was repeated in the presence of a supporting electrolyte,
KCl. The coverage of CTAB was found to be slightly larger in this system compared to the case
where there was no KCl (figure Figure 3.6). This may seem counterintuitive, since the large quantity of KCl is expected to form the double layer and exclude CTAB from the interface. However,
as emphasized in this work, CTAB is also partially driven to the surface by the hydrophobic effect.
A recent study suggests that the hydrophilicity of the metal – liquid interface is affected by salts
and can be dramatically altered by changing the identity of the salts.[136] We hypothesize that KCl
26
Figure 3.5: Integrated CH stretches of CTAB for the 1 mM CTAB 10 M water dDMSO mixture as
a function of applied potential ranging from – 0.6 V to + 0.6 V vs Ag/AgCl. In the left panel, the
potential scanned from 0 to -0.6, then to +0.6 and back to -0.6. In the right panel, the potential is
scanned from 0 to +0.6, then to -0.6 and back to +0.6.
Figure 3.6: Left panel. Integrated CH peaks of CTAB in a 1 mM CTAB – 10 M water dDMSO
mixture with 100 mM KCl (blue) and without 100 mM KCl (red). Right panel. The reversibility
of 1 mM CTAB, 10 M water-dDMSO mixture with 100 mM KCl. The potential was stepped from
0 V to + 0.6 V to – 0.6 V and back.
27
may alter the hydrophilicity of the gold surface which in turn affects CTAB accumulation. Further
experiments are required to understand this effect more clearly.
To better understand the cross-dependence of the interfacial surfactant and water content, we
performed the experiments with varying bulk water concentrations of 0 M, 10 M, 20 M and 55.6
M (corresponding to mole fractions χw =0,0.4,0.6 and 1). The differential spectra are shown in the
Appendix (figures Figure 5.2 Figure 5.3,Figure 5.5). The CTAB spectra at 0 V and the integrated
∆A
A
are shown in figure Figure 3.7. We also tried 1 M water concentration, but it was difficult to
discern the O-H stretch in the differential spectra (see Appendix Figure 5.4).
(a) (b)
Figure 3.7: The CH stretch region at 0 V for different water concentrations (a) and the corresponding ratio of integrated areas as function of applied potential (b). For the pure water and 20 M water
cases, application of reductive potentials past -0.6 V and -0.8 V were not possible due to substrate
degradation.
The most immediate observation is that the amount of surfactants at the interface in pure
dDMSO is the smallest (blue spectrum in figure Figure 3.7a). However, it exhibits the highest
sensitivity to potential (figure Figure 3.7b) reaching more than 300 % at -1 V (relative to 0 V).
As the bulk water content is increased, more CTAB accumulates near the surface at 0 V (figure
Figure 3.7a). However, its sensitivity to potential is diminished significantly (figure Figure 3.7b).
To get a global view of the CTAB content at the interface as a function of both potential and
bulk water mole fraction, we present the data as a three-dimensional plot (figure Figure 3.8a). The
28
vertical axis is scaled to represent estimated CTAB coverage. Such surface coverages are challenging to obtain given the wide range of enhancement factors reported in the literature.[122, 137–
139] For this work, we used SEIRAS spectra of a redox active surface species, and determined the
surface coverage by Coulometry to estimate the enhancement factor as described in the Appendix.
This method was adapted from a prior publication of ours. [140] Our method of estimating the
enhancement factor has an advantage over previous approaches, since surface coverage is independently determined. Despite this advantage, the method relies on some assumptions, and the result
should be taken as a reasonable estimate and not an exact value. The two horizontal axes represent
variations in the water mole fraction and potential. Note that the surface is interpolated across our
experimental data (dotted lines).
(a) (b)
Figure 3.8: (a) CTAB coverage (inferred from the spectra) as a function of applied potential and
bulk mole fraction of water in the system. The dotted lines are experimental data. (b) A contour
view of Figure 3.8a Each contour line represents equal coverage of CTAB at various potentials and
bulk water mole fraction. The colorbar indicates the CTAB coverages, red corresponding to higher
coverages than blue.
This plot demonstrates that there are two forces that drive CTAB to the surface – the negative
potential and the hydrophobic interactions with the water in the bulk. The data shows that in the
absence of water (i.e. pure dDMSO), the surfactants are pulled to the surface with potential. The
same is true for pure water as well. However, in pure water, the surface coverage is always larger
than that of pure dDMSO across all potentials. The data also shows that the sensitivity of the
29
coverage to potential is the least in pure water and the largest in pure dDMSO. Note that the bulk
dielectric constant of water (∼ 78) is larger than that of DMSO (∼ 46), implying that water will
have a larger screening capability. However, prior research shows that dielectric constants at the
surface can be drastically different from the bulk. Specifically, the dielectric constant of interfacial
water is estimated to be ∼ 4-7.[141–143] The interfacial dielectric constant of DMSO has not been
reported, but it can be assumed to be smaller than the bulk. Due to this ambiguity, we can not
comment on the details of the influence of the dielectric constant on surfactant accumulation at
the interface. Therefore, the surface electrostatic effects should be more diminished in water. The
fact that more surfactants are present at the surface in water compared to dDMSO implies that the
drive for adsorption is predominantly the hydrophobic effect in water. We suspect that at more
negative potentials more surfactants can be driven to the electrode in the water-dDMSO mixtures.
However, the instability of the SEIRAS substrate, and the bubbles from hydrogen evolution in
water-containing mixtures, prevents us from exploring much larger potentials.
An insightful way of understanding the combined effects of water content and potential on
CTAB is to view the data as a contour plot (figure Figure 3.8.b). Each contour line is a line of constant CTAB coverage and shows that a constant coverage can be maintained by a combination of
potential and water mole fraction. This result presents an important opportunity to connect the hydrophobic effect that drives the surfactant to the surface in the presence of water to the electrostatic
effect of the electrode in a non-aqueous environment. To our knowledge, these quantities have not
been related in a clear way before. For example, the contour line crossing the potential axis near
-0.5 V (shown by the arrow in figure 6.b) crosses the vertical axis near χwater = 0.65. Therefore,
one concludes that the hydrophobic effect of 0.65 mole fraction of water in dDMSO is equivalent
to application of -0.5 V of potential in dDMSO. Moving along this contour, one can maintain the
same surfactant concentration by varying a combination of potentials and mole fractions. To quantify the equivalence between the hydrophobic drive and potential, we took two slices of the data in
figure Figure 3.8a. One slice is along the potential axis at χwater = 0 , and a second slice is along
the mole fraction axis V=+0.6 V (vs. Ag/AgCl) corresponding to the least amount of surfactants
30
in pure dDMSO. Then, we paired points of equal coverage on these two slices by interpolation
described in the Appendix. The result (figure Figure 3.9) demonstrates the correlation between the
hydrophobic and potential drives, and is one of the main takeaways of our study. A linear fit across
this data has a slope of 0.55 ( χwater
V
), and can be interpreted as a change of 1 V in potential having
an equivalent effect as 0.55 mole fraction of water for driving the surfactants to the surface.
Figure 3.9: A plot demonstrating the equivalence between the hydrophobic effect and the electrochemical potential necessary for driving surfactants to the surface. The hydrophobicity of the bulk
is tuned by the water mole fraction.
We caution that figure Figure 3.9 should not be interpreted as a free energy assignment to the
hydrophobic effect. The adsorption process is non-Faradaic, and governed by the electrostatics of
the interface. The potential felt by the surfactant is not the same as the applied potential, because
of screening by the solvents and the ions. This, and several other factors, complicate determination
of the energy of adsorption, at least based on the data from this work. Therefore, figure Figure 3.9
should be interpreted as an empirical correlation between the applied potential and the hydrophobic
forces.
To understand the influence of anions on surfactant accumulation near the surface, we used
the surfactant CTAC, where the bromide ion is replaced with chloride. We found that the CTAC
behaved similar to CTAB, both in terms of accumulation near the surface in response to potential
(figure Figure 3.10), and its influence on water spectrum (figure Figure 3.11).
31
Figure 3.10: Integrated CH peaks of CTAB (red) and CTAC (blue) in a 1 mM CTAB/CTAC – 10M
water dDMSO. The potential was cycled from + 0.6 V to – 0.6 V and back.
Figure 3.11: CH and OH integrated peak data for the 1 mM surfactant dDMSO 10 M water system.
The surfactants used were CTAB (left panel) and CTAC (right panel). These experiments were
carried out using the same SEIRAS substrate.
32
Next, we discuss the differential water spectra as a function of potential and bulk water mole
fraction. The water absorption spectra at 0 V with respect to gold on ZnSe is shown in figure
Figure 3.12a. We note that the preparation of the electrode results into capturing a small amount
of water in the porous gold. Some of the water is dissolved away when pure dDMSO is added,
explaining the small negative feature (blue line in figure Figure 3.12a). Expectedly, as more water
is added to dDMSO, the water signal increases. The potential dependence of these spectra is
shown in the Appendix, and the summary of the differential change in integrated peaks is shown
in figure Figure 3.12b. The integrated OH since not all of the interfacial water due to the substrate
preparation is removed. This variation is denoted by the shaded bar in the figure. At higher water
concentrations, there is a clear potential dependence in the opposite direction of surfactants (figure
Figure 3.8b), and therefore implies exchange of water with surfactants in the presence of potential.
The largest variation (∼ 60 %) in population of water across the entire potential range is seen
at 10 M bulk water concentration (corresponding to χwater =0.4). This suggests that, for inducing
the largest change in water population via potential, water should be present at bulk mole fractions
of 40% in solution. For higher bulk mole fractions of water, more current was drawn at higher
negative potentials (hydrogen evolution), followed by the substrate losing conductivity. Therefore,
we only interpret the data in the lower potential range. The sensitivity of water to potential in this
range is relatively small.
We extended this study to the anionic surfactant SDS, anticipating an opposite effect as that
of the positively charged CTAB. The spectral signatures monitored for SDS were the same CH
stretches as that of CTAB. Although we chose the same concentration of SDS as that of CTAB
(1 mM), we did not observe any signal from the CH region (Appendix, Figure 5.8) in dDMSO.
Upon increasing the concentration of SDS, we noticed that even at 100 mM concentrations, the
CH stretches of SDS could not be observed (Appendix figure Figure 5.9). Even in water, it was
necessary to use a high concentration (100 mM) to observe the potential dependence of SDS, and
can be seen in the Appendix, figures Figure 5.10 and Figure 5.11.
33
(a) (b)
Figure 3.12: . (a). The OH stretch region at 0 V for different water concentrations in the CTABdDMSO mixture. (b). The corresponding ratio of integrated areas as a function of applied potential.
The shaded gray region represents the integrated area of water for the pure dDMSO and CTAB,
due to residual water from gold substrate preparation.
3.5 Conclusion
Surfactants are powerful agents for controlling the amount of interfacial water. Even when present
in small concentrations (1 mM) the can influence interfacial water content in a 10 M solution of
water in an organic solvent. Based on the choice of surfactants, the amount of interfacial water
can be enriched/depleted at positive/negative potentials. Upon comparing different ways of obtaining the same surfactant coverage, a correlation between addition of water to the system and
applying potential was drawn. This enabled translating the hydrophobic drive in the electrochemical language of applied potential. It was found that a change of 0.5 bulk mole fraction of water
has the same effect as changing the applied potential by 1 V. An important message of work to
the electrochemistry community is highlighting the appropriate ratio of water-organic mixtures for
controlling interfacial water content. We found that at low mole fractions of water, surfactants had
a larger influence on water content.
This opens avenues for delievering water to the interface for electrochemical reactions. While
our work focused on using relatively simple surfactants, a vast area of surfactant design remains to
be explored for enforcing structure on the interfacial water. The broader goal will be to not only
34
control the quantity of water, but also control and measure the structural and chemical details of
the interfacial water such as orientation, clustering, and acidity in water-organic mixtures.
35
Chapter 4
Photothermal Modulation of Silicon Optical Cavities Measured
by Continuous Mid-IR
4.1 abstract
Photothermal effects, or changes in physical properties of materials in response to heat delivered by
light, have a number of applications including the recent success in sub-diffraction IR microscopy.
Optical cavities, which are at the core of numerous technologies, are also sensitive to thermal
effects. Here we bring these ideas together by demonstrating and modeling short pulse photothermal modulation of a Fabry-Perot cavity formed between the two polished faces of a single silicon
wafer. We used near-IR ultrafast pump pulses to transiently heat a silicon cavity while monitoring
its transmission with a continuous wave mid-IR probe. The temperature change induced by the
pump pulse changes the refractive index of silicon, which in turn shifts the cavity resonances and
the transmission spectrum. We model both the magnitude and the temporal behavior of the photothermal response of the cavity. We envisage applications of photothermal modulation of cavities
in the area of optical switches, and in photothermal microscopy of thin films.
36
4.2 Introduction
Photothermal effects are observed when thermal energy delivered by light changes the physical
and chemical properties of materials. These effects give rise to applications such as photothermal
detection of light [144, 145], photoacoustic spectroscopy [146, 147], photothermal calorimetry
[148–150], and photothermal therapy [151–153]. An important recent example is photothermal
microscopy [154–156], where sub-diffraction IR microscopy has been made possible by detecting
photothermal expansion of the IR absorbing regions in a sample using visible light. Optical cavities
are susceptible to thermal effects due to either expansion or refractive index change [157, 158].
They are essential components in a wide range of applications spanning thin film interference
filters [159], lasers, and cavity enhanced spectroscopy [160, 161]. Therefore, photothermal control
of optical cavities is technologically relevant and promising.
In this work, we present experimental evidence and a simple model to describe pump induced
thermal effects on cavity resonances. This work is both of fundamental and practical relevance
to several areas of optical spectroscopy. These results can be extrapolated to waveguides used
in silicon photonics. Controlling the optical transmission of a cavity waveguide with light is the
fundamental element of a photonics logic circuit. Photothermal control of a waveguide is a means
of achieving this goal. For miniaturized photonic elements, the required energy for such switching
can be quite small. Another area where these results are applicable is in the spectroscopy of
thermoelectric thin films[162, 163]. Characterization of such materials requires precise delivery of
heat and measurement of the ensuing temperature change. Extension of our results can be of value
for researchers in that field.
An important future step in this direction is to perform extremely sensitive cavity enhanced
pump-probe measurements. For example, as of now transient spectroscopy of a monolayer is quite
difficult[164–166]. We envisage that placing a single monolayer in a very high finesse cavity may
produce a large enough photothermal effect to shift the cavity resonances, and thereby open a new
way of probing the dynamics of monolayers.
37
Here we demonstrate and model the photothermal effects in a simple silicon Fabry-Perot optical
cavity using a continuous wave (CW) infra-red (IR) laser source from a quantum cascade laser
(QCL). The heat is delivered by an ultrafast near IR pulse, which is absorbed in the material within
the cavity. The ensuing temperature rise and refractive index change shifts the cavity resonances
and modulates the transmission of the CW mid-IR probe. The effect persists until the heat diffuses
away from the photoexcited volume. The general concept of our experiment is shown in Figure
Figure 4.1.
This work is also relevant to the field of all-optical switching [167, 168] as there are many
tunable parameters such as choice of radiation profile, cavity finesse and size which can modulate
the transmission. Similarly, the effect can be exploited in photothermal microscopy of thin films,
where in addition to thermal expansion, the thermally induced refractive index change may shift
the thin film optical resonances. While this work is demonstrated in silicon, the model is general
and opens opportunities for exploring other types of materials, cavity geometries, and wavelength
ranges. These include but are not limited to, whispering gallery modes, integrated waveguides and
nanowires [169, 170].
This paper is arranged as follows. First we describe the concept of cavity photothermal effect
and derive the essential equations and Figures of merit that govern the response of a cavity to
optical heating. Second, we describe the experimental methods along with computational methods
of estimating heat diffusion from a photoexcited volume. Third, we present the experimental
results and compare them to computations. Finally, we discuss possible extensions of the work
and its future potentials.
4.3 The Concept of Cavity Photothermal Effect
A simple Fabry-Perot optical cavity [171] is formed by two reflective surfaces placed at distance d
from each other. Interference of multiple reflections within this cavity gives rise to resonances that
38
v
v
Tunable
CW IR
Near IR
fs pulse
D
IR
pumped un-pumped
ΔI
ΔI
~ 5 μs
decay
(a) (b)
(c)
Figure 4.1: The general concept of our experiment. (a): Tunable CW IR from the QCL and 800
nm pulse incident on a silicon wafer (gray). The red portion inside the silicon is the region of
change in refractive index caused by 800 nm absorption. (b): The manifestation of this refractive
index change in the Fabry-Perot transmission fringes. For comparison the transmission fringes in
the absence of the pump are in blue. (c): The change between the pumped and un-pumped fringes
as a function of time for a fixed wavelength. The timescales for the decay of the intensity change
is in the order of a few µs.
39
manifest themselves as peaks and dips in the transmission spectrum. The transmission function of
the cavity, as found in most standard optics texts, is expressed as [172].
T =
I
I0
=
1
1+F sin2(Φ)
(4.1)
Here Φ = nk d cosθt
is the phase accumulated by light upon traversing the cavity, with k =
2π/λ the free space wavevector of light, n the refractive index of the medium, and θt
the angle
with respect to normal within the cavity. For our work, we will consider θt = 0. The parameter
F, known as the finesse of the cavity, is related to the reflectivity R, defined in terms of intensity,
of the facets F =
4R
(1−R)
2
. The transmission spectrum has an oscillatory pattern - known as fringes
with peaks corresponding to resonances in the cavity. The width of the peaks are dictated by the
finesse F, and the spacing between the peaks are dictated by the net phase ∆Φ = 2π accumulated
across the cavity.
For silicon the refractive index is reported to be between 3.42−3.54 [173, 174]. For this work,
we used n = 3.4 which gives rise to R = 0.31 at the air interface. Correspondingly, the finesse for
a silicon slab with two polished facets is F = 2.5. The spectral fringes in the transmission function
of a silicon wafer with d = 450µm in the mid IR range have peaks with spacing of ∼ 3 cm−1
. A
typical transmission spectrum is shown in Figure Figure 4.2.
The central idea of photothermal modulation in such a cavity, is excitation by a visible or
near IR pulse, which is completely absorbed in the first few tens of microns within the wafer.
The photoexcited electron hole pairs will first thermalize and then recombine, releasing all of the
absorbed energy as heat within the material within a few nanoseconds [175, 176]. This gives rise
to a transient temperature rise that persists until the heat is conducted away from the photoexcited
volume. The temperature dependence of the refractive index of silicon is well-documented and
reported as [177]
n = n0 +η ∆T (4.2)
40
Figure 4.2: FTIR spectrum of the unpumped silicon wafer. The observed fringes arise from FabryPerot interference.
where η = 1.9 × 10−4/K[148]. Therefore, the temperature rise in the photoexcited volume
causes a refractive index change. If that area is monitored by transmission of a CW IR probe
beam, the net phase Φ accumulated by the probe will change due to this temperature jump. In
other words, the transmission fringes of the heated cavity are shifted with respect to the unheated
cavity. Such fringe shifts will cause an increase in transmission for some wavelength and a decrease
for others. As heat is conducted away from the photoexcited volume, the cavity returns to its initial
state. This process can be monitored by the change in transmission as a function of time.
The above is the core idea for photothermal modulation of a simple silicon cavity. Below, we
will analytically derive the photothermal response of such a cavity immediately after the pump
pulse (at t = 0). The time dependence of the response that arises from heat diffusion away from
the excitation volume will be treated computationally as described under methods.
First, we consider the spatial profile of a pump/heating pulse with total energy E0 and a Gaussian profile impinged on the cavity, which is aligned along the z axis as shown in Figure Figure 4.3.
E(r,z) = E0
2π w2
e
−
r
2
2w2 e
−α z
(4.3)
41
Figure 4.3: A schematic of the 800 nm pump incident on the silicon wafer. The pump is a Gaussian
beam with width w traveling in the z direction. Heat is generated by the absorption of the pump
inside silicon (along the z direction). For calculation of dissipated heat as a function of space, the
ring volume element defined at radius r, width dr and thickness dz is needed as explained in the
text.
42
Here w is the width of the Gaussian profile and α is the absorption coefficient of the material
and dictates an exponential decay of the light energy as the pulse enters the absorbing material. It
can be verified that for α = 0, integrating the above with respect to r yields the total energy E0.
The dimension of E(r,z) is energy per unit area.
We are interested in the energy lost by light within the volume of a ring of thickness ∆z and
radius r
∆E =
dE
dz
∆z2π rdr (4.4)
where dr is the width of the ring in the radial direction as seen in Figure Figure 4.3. The heat
dissipated in the material is the negative of the above quantity ∆Q = −∆E. The volume of the ring
is ∆z2π rdr and its mass is m = ρ ∆z2πr dr, where ρ is the density of the material. If the heat
capacity of the material is C, the temperature rise within the ring is:
∆T =
∆Q
mC
= −
1
ρC
dE
dz
=
α E0
2π ρC w2
e
−
r
2
2w2 e
−α z
(4.5)
If we assume that the probe width is much smaller than width of the pump pulse, we will
only focus on the influence of this temperature profile on the refractive index in the middle of the
illuminated volume (at r = 0). Then, based on equation 2 the refractive index profile along z is:
n(z) = n0 +
α η
2π ρC
E0
w2
e
−α z
(4.6)
To find the total phase accumulated across the heated cavity, we integrate the above profile:
Φ =
Z d
0
k n(z)dz = Φ0 +kβ
E0
w2
1−e
−α d
(4.7)
Here Φ0 = n0 k d is the accumulated phase in the unpumped cavity. The value β =
η
2π ρC
is
the cavity photothermal figure of merit. For silicon η = 1.9 × 10−4/K [148], ρ = 2.33 g
cm3 and
4
Figure 4.4: Simulated Fabry-Perot fringes in a silicon wafer. The blue line represents the fringes
for un-pumped silicon (∆T = 0) and the red line corresponds to the fringes when the silicon is
pumped. The pumped fringes were simulated using equation 8 using the values of E0 and w as
10 µJ and 50 µm respectively.
C = 0.7
J
gK [173], resulting into β = 1.8 × 10−5 cm3
J
. When the thickness of the cavity is much
larger than the characteristic absorption length (d ≫ 1/α), the above is simplified to:
Φ = Φ0 +kβ
E0
w2
(4.8)
This accumulated phase shift can be used to calculate the transmission spectrum of the pumped
cavity using equation 4.1, as shown in Figure Figure 4.4 for typical experimental parameters. Note
that the above is based on the refractive index profile right after pumping (at t = 0) prior to heat
transfer away from the pumped volume.
Next we consider the dynamics of the photothermal modulation. After the establishment of the
initial temperature profile, heat diffuses away from the photoexcited volume and the cavity returns
to its initial state. To understand the temporal evolution of the cavity response, one must solve the
heat diffusion equation given the temperature profile at t = 0. While analytical solutions for simplified cases of heat diffusion into a semi-infinite slab exist, a closed form solution for a temperature
profile that is Gaussian radially that exponentially decays along z does not exist. However, this
44
problem lends itself to solution by Fourier transforms, as is well-established [178] and a numerical
solution can be easily obtained. This is described in more detail in the computational methods.
Note that the effective path length difference can also be caused by thermal expansion of the
material as well. For the case described in this work the length of the cavity (450µm) is much
larger than the heated part of the cavity (∼ 10µm). Given the relatively small thermal expansion
coefficient of silicon (2.6 × 10−6/K)[179], the net phase accumulated due to expansion from a
typical temperature rise (50 K) is relatively small (∆d = 0.013µm). However, when working with
smaller cavities and larger temperature changes, one may also need to include this effect in the
model.
4.4 Methods
4.4.1 Experimental Methods
A 1 kHz regeneratively amplified Ti:Sapphire laser (Coherent) was used to generate ultrafast near
IR pulses (800 nm). A portion of this 800 nm pulse (2 W) was attenuated by a half-wave plate and
polarizer and then used to pump the silicon wafer. The pump wavelengths used for this experiment
were 800 nm and 400 nm. The 400 nm was generated by frequency doubling 800 nm (with BBO).
The pump powers used ranged from 1−17 mW corresponding to an energy of 1−17 µJ per pulse.
The silicon wafers used for the experiment were purchased from University Wafer, which were
undoped, doubly polished and 450 µm thick.
The IR source was a mid-infrared quantum cascade laser (MIRcat-QCL) from DRS Daylight
Solutions. In the past few years, the quantum cascade laser (QCL) has been utilized in many
applications due its many advantages [180–183]. With careful engineering of layered quantum
wells and using the electron tunneling effect cascading down supperlattices [184, 185], QCLs
provide high intensity IR laser powers that exceed the limits of other IR sources such as the FTIR
glowbar. This IR source can be tuned with a resolution of 0.1 cm−1
from 1500 cm−1
to 2300 cm−1
.
In our experiments, we used a wavelength region from 2200 cm−1
to 2300 cm−1
. We collected 200
45
Figure 4.5: A schematic of our experimental setup. The 800 nm pulse comes from a Ti:Sapphire
amplifier. It is attenuated using a half wave plate (HWP) and a polarizer. The pulse is reflected
off an angled silicon dichroic and focused by a lens (L1) onto the silicon wafer sample. The CW
IR comes from the QCL (black box), goes through the angled silicon dichroic and is focused onto
the silicon wafer sample by lens (L1). The transmitted IR is collected by L2 and sent to the MCT
detector.
46
data points in this region, corresponding to a 0.5 cm−1
increment. LabView was used to interface
with the QCL and the oscilloscope.
We noted that the visible and IR beams were distinctly focusing on two different spots along
the optical axis. The sample was intentionally placed at the focus of the IR beam instead of the
focus of the visible beam to avoid overheating. The spot sizes of the IR and visible beam at the
sample position was 76 µm and 635 µm, respectively.
Figure 4.6: The response of the MCT detector used in the experiment. This was obtained by
directing the ∼ 50 f s pulse (800 nm) light onto the detector. The decay was fit to an exponential
and the extracted time was ∼ 2 µs
The signal was detected with a nitrogen cooled mercury cadmium telluride detector (InfraRed
Associates Inc. MCT-13-1.00) with an operational amplifier (InfraRed Associates Inc MCT-1000).
The response of the MCT detector consists of a rapid rise followed by a decay (figure Figure 4.6).
The time constant for the decay component was determined to be 1.6 µs and the error bars are
reported in the Appendix (Table Table 5.8). This was achieved by directing the 50 fs pulse (800
nm) onto the detector and extracting the decay time of the obtained time trace. It is noteworthy to
mention that the MCT detector used in this experiment was a photo-conductive (PC) detector that
can only detect AC signals, or changes in the IR light, rather than the average power of the IR.
Therefore, the signals detected by the MCT are the changes of the optical fringes between pumped
47
and un-pumped silicon. The signal was measured with an oscilloscope (Keysight Technology,
54855A Infiniium Oscilloscope) and MATLAB was used for signal processing and simulations.
4.4.2 Computational Methods
To calculate the time evolution of the refractive index change in the cavity, the heat equation for
∆T(t) was solved by Fourier transformation method. First the problem was symmetrized by removing the constraint of the slab and extending the slab material to both positive and negative z
directions. The initial temperature profile was defined over a grid of points in the x, y and z coordinates ∆T0(x, y,z). This profile was also symmetrized by reflecting it around the xy plane. Since the
temperature profile monotonically decreases in either direction of z, heat will not diffuse from one
semi-infinite side to the other. The MATLAB multidimensional Fourier transform function was
used to Fourier transform the temperature profile into kx, ky, kz variables ∆T˜
0(kx, ky, kz). The action
of diffusion in this domain is multiplication by a Gaussian filter ∆T˜(k,t) = ∆T˜
0 e
−Dt k2
where D is
the heat diffusion coefficient and k
2 = k
2
x + k
2
y + k
2
z
. The resulting function ∆T˜(k,t) was inverse
Fourier transformed back into the space domain ∆T(x, y,z,t). The MATLAB code for this operation is in the Appendix. Figure Figure 4.7 shows a few snapshots of the temperature change inside
the silicon slab at different times. To find the phase shift along the z direction at r = 0, the resulting
profile of temperature was converted to refractive index change and numerically integrated along
the z direction.
For silicon, the diffusion coefficient is D = 0.8
cm2
s
[177]. For pump pulses similar to those
used in the experiments, diffusion of heat continues for several microseconds.
4.5 Results and Discussions
Experimental results showing the transient transmissions are shown in Figure Figure 4.8. The maximum transient change as a function of frequency is plotted in Figure Figure 4.8 a. Note that the
48
Figure 4.7: The simulated temperature change inside silicon caused by the pump as a function of
time. This profile was obtained by using equation 5 and solving the diffusion equation as explained
in the text. The contours represent a slice of the temperature change along the x-z plane. Here, z
is oriented along the direction of the thickness of the silicon wafer. The color axis indicates the
change in temperature.
49
Figure 4.8: Representative data from our experiments. (a) : The change in fringes plotted against
IR wavelengths spanning from 2200 cm−1
to 2300 cm−1
. This slice was taken at t = 10 µs. (b) :
Time traces of ∆I at three representative wavelengths.
observed fringes correspond to the differences in pumped and un-pumped spectra. As explained
before, this difference can either be positive, negative, or zero depending on the wavelength. The
fringes shown in Figure Figure 4.8 a are filtered, where the frequency components due to fluctuations in the QCL laser and additional etalon effects from the silicon dichroic are removed. A
comparison between the raw data and the filtered version can be seen in the Appendix (Figure Figure 4.9). Additionally, we have shown the pumped and un-pumped spectra in Figure Figure 4.10).
50
Figure 4.9: (a) : Our raw experimental data. (b): The same fringes after filtered in Fourier space.
A Gaussian filter was used to filter out the effects of QCL laser fluctuations and fringes arising
from our silicon dichroic.
Figure 4.10: A comparison between the unpumped fringes seen in blue (FTIR) and the filtered
pumped fringes from the previous figure in red.
51
The fringe patterns are not uniform across the wavenumber region. We believe this is an effect
from a combination of QCL laser fluctuations and fringes arising from the silicon dichroic. These
additional frequency components can be filtered out as seen in the Appendix (Figure Figure 5.24).
Time traces for three such wavelengths are shown in Figure Figure 4.8b. The decay parts of the
transmissions were fitted to exponentials and time constants of about 6 µs were retrieved.
To explain the observed dynamics, we started by solving the heat diffusion equation as discussed earlier. This yielded spatial temperature profiles for different times as seen in Figure Figure 4.7. From the temperature profiles, we calculated the refractive index along the z axis n(z) at
x, y = 0 as shown in the Appendix (Figure Figure 5.23), which was then used to calculate Φ(z) and
the transmission spectra according to equation 8. This process was repeated for different temperature profiles at various times and simulated fringes were obtained as shown in Figure Figure 4.11.
Not only does a change in refractive index affect the phase, but also the reflectivity of the facet,
and therefore the finesse. However, the magnitude of the change in finesse is small. Taking the
largest thermally induced refractive index from the Appendex (Figure Figure 5.26), we calculated
∆R and a corresponding change in finesse to be in the order of 0.002%. Therefore, we conclude
that the finesse is not affected significantly by heat.
52
Figure 4.11: Simulated Fabry-Perot fringes. (a): Each spectrum was computed for a different
temperature profile corresponding to each time. (b): The time dependence of the differential transmission (difference between pumped and un-pumped) for three wavelengths.
The time dependence of transmission at three selected wavelengths is shown in Figure Figure 4.11b. These traces were obtained by subtracting the simulated pumped and un-pumped spectra at different wavelengths. Exponential fits to such traces reveal time constants in the range of
4−8 µs , which is in excellent agreement with the experimental values of 6 µs. There is a slower
decay component in the simulated dynamics, resulting into significant differential signal at times
as long as 80 µs. The experimental differential signals, however, completely vanish beyond 25 µs.
This discrepancy is likely due to the limited size of the simulation box, which prevents heat escape
and rapid cooling.
The signal dependence on power was also studied as demonstrated in Figure Figure 4.12. The
transient IR spectra at four different pump powers were measured and the root mean square value
of the fringes over the 100 cm−1
spectral range was calculated and plotted as a function of power.
There is a clear linear dependence on pump power, as expected for a photothermal effect.
53
Figure 4.12: (a): Differential spectra measured at various pump powers at a fixed delay. (b):
The root mean square (RMS) of the spectra plotted against the corresponding pump energies,
demonstrating a linear relationship.
Additionally, we also measured the effects of pump pulses centered at 400 nm. The resulting
signal was very similar to the 800 nm pump, as shown in the Appendix. This can be explained
based on equation 7, where the total phase accumulated across the silicon wafer does not have
dependence on the absorption coefficient as long as all of the light is absorbed within the sample.
The absorption depth of silicon at 800 nm is 10 µm and for 400 nm is 0.1 µm [186]. Even though
the absorption depth is different for the two pump wavelengths, the net heat deposited is a function
of the total energy of the pulse alone.
4.6 Conclusion and Outlook
Our work demonstrates the concepts, models, and experimental results for photothermal cavity
modulation. Optical control over transmission of light, or optical switching, has been an active
area of research in photonics.[167, 168] In our work, the cavity was naturally formed between
54
the two polished faces of a silicon wafer. If the reflectivity of the cavity facets is enhanced, for
example, by depositing optical coatings, the lines in the transmission spectra will be narrower
and correspondingly the photothermal modulation factor will be larger. This effect can be implemented in optical waveguides and micro-fabricated optical resonators, where the mode volume,
finesse, and free spectral range can be precisely engineered to optimize the photothermal effect.
In these cases, it is possible to engineer the transmission such that an ultrafast visible pulse can
shutter the continuous wave IR to create mid IR pulses limited in width by the thermal response of
the cavity. We can estimate the heat that would be required to induce similar changes for silicon
photonics applications. The cross section of a typical silicon waveguide is in the order of ∼ 25 µm
2
[169],which is approximately 400 times smaller than the pumped area in our experiment. Therefore, one can anticipate the energy needed to induce similar refractive index changes for a silicon
waveguide to be correspondingly 400 times smaller. The pump powers in our experiment were
in the order of ∼ 10 µJ, which means for silicon waveguide applications, a substantially smaller
magnitude of power would be required (∼ 25 nJ), as shown by Zhang et al [187] in photonic crystal microresonators. Our work suggests that such switching may be possible without the need for
nanofabricated photonic crystals.
Photothermal microscopy is an emergent field with a wide range of applications[188]. Photothermal microscopy of transparent thin film materials will necessarily include optical interference and cavity effects. Our model and experimental work will help increase the understanding of
imaging contrast in such cases. Finally, there have been recent studies on understanding whether
coupling of mid-IR cavity modes with vibrations of a molecule affect chemical reactions. Since
thermal effects are closely related to such phenomena, this work may be of interest to the polariton
chemistry community as well[168, 189].
55
Chapter 5
Ongoing work and Future directions
Currently, we are working towards time resolved surface spectroscopy applied to electrodes in the
presence of external potentials. Specifically, our aim is to combine electrochemical impedance
with vibrational spectroscopy. In an electrochemical impedance experiment, a sinusodial potential
is applied and the resulting current is measured. This process is repeated for different frequencies
of the sinusodial potential wave. From the differences in phase and magnitude of the current with
respect to the applied potential, inferences about the impedance of the entire system is made. Circuit diagrams can be conjured to match the impedance obtained in the electrochemical experiment,
multiple different circuit diagrams can have the same impedance. To justify the particular choice
of circuit, electrochemists resort to various control experiments (and have devised very creative
experiments). Having spectroscopic information would help in ascertaining actual components of
the circuit (differentiating between capacitative and resistive components).
The system we envision comprises of surfactants and another salt with a unique vibrational
probe. Using lock-in detection and either step scan FTIR or monochromatic IR excitation with a
quantum cascade laser we can isolate vibrational signatures of the system that arise ONLY due to
the oscillatory potential. The expectation is that at some frequencies surfactants respond faster than
the other salt. Perhaps with clever supporting electrolyte concentrations and smaller amplitudes of
potentials, the double layer can be optically sectioned.
56
With regard to future directions, the potential SEIRAS study with surfactants can be readily
extended to surfactants that have hydrogen bonding capabilities and can recycle water to the interface upon application of the correct external potential. We can also study electrochemical reactions
involving the water.
Figure 5.1: A pyridinium surfactant molecule with two hydroxyl groups, the counter ion is not
shown in the figure. With the hydroxyl groups, the surfactant can hydrogen bond and bring water
along when it is pulled to the electrode. The second panel is an analogy depicting the surfactant as
a molecular level water carrier.
Monitoring the current would give additional information about the reduction of carbon dioxide.
57
Bibliography
1. Schmalensee, R., Stoker, T. M. & Judson, R. A. World carbon dioxide emissions:
1950–2050. Review of Economics and Statistics 80, 15–27 (1998).
2. Ritchie, H., Roser, M. & Rosado, P. CO and greenhouse gas emissions. Our world in data
(2020).
3. D’Alessandro, D. M., Smit, B. & Long, J. R. Carbon dioxide capture: prospects for new
materials. Angewandte Chemie International Edition 49, 6058–6082 (2010).
4. Gao, W., Liang, S., Wang, R., Jiang, Q., Zhang, Y., Zheng, Q., et al. Industrial carbon
dioxide capture and utilization: state of the art and future challenges. Chemical Society
Reviews 49, 8584–8686 (2020).
5. Jitaru, M. Electrochemical carbon dioxide reduction-fundamental and applied topics.
Journal of the University of chemical Technology and Metallurgy 42, 333–344 (2007).
6. Chen, C., Kotyk, J. F. K. & Sheehan, S. W. Progress toward commercial application of
electrochemical carbon dioxide reduction. Chem 4, 2571–2586 (2018).
7. Collin, J. & Sauvage, J. Electrochemical reduction of carbon dioxide mediated by
molecular catalysts. Coordination chemistry reviews 93, 245–268 (1989).
8. Sun, L., Reddu, V., Fisher, A. C. & Wang, X. Electrocatalytic reduction of carbon dioxide:
opportunities with heterogeneous molecular catalysts. Energy & Environmental Science
13, 374–403 (2020).
9. Amatore, C. & Saveant, J. M. Mechanism and kinetic characteristics of the
electrochemical reduction of carbon dioxide in media of low proton availability. Journal of
the American Chemical Society 103, 5021–5023 (1981).
10. Bachmeier, A. & Armstrong, F. Solar-driven proton and carbon dioxide reduction to
fuels—lessons from metalloenzymes. Current opinion in chemical biology 25, 141–151
(2015).
58
11. Plechkova, N. V. & Seddon, K. R. Applications of ionic liquids in the chemical industry.
Chemical Society Reviews 37, 123–150 (2008).
12. Ohno, H. Electrochemical aspects of ionic liquids (John Wiley & Sons, 2011).
13. Van Aken, K. L., Beidaghi, M. & Gogotsi, Y. Formulation of ionic-liquid electrolyte to
expand the voltage window of supercapacitors. Angewandte Chemie - International
Edition 54, 4806–4809. doi:10.1002/anie.201412257 (2015).
14. Esperanc¸a, J. M., Tariq, M., Pereiro, A. B., Araujo, J. M., Seddon, K. R. & Rebelo, L. P. N. ´
Anomalous and Not-So-Common Behaviour in Common Ionic Liquids and Ionic
Liquid-containing Systems. Frontiers in chemistry 7, 450 (2019).
15. Tran, C. D., De Paoli Lacerda, S. H. & Oliveira, D. Absorption of water by
room-temperature ionic liquids: Effect of anions on concentration and state of water.
Applied Spectroscopy 57, 152–157. doi:10.1366/000370203321535051 (2003).
16. Anthony, J. L., Maginn, E. J. & Brennecke, J. F. Solution thermodynamics of
imidazolium-based ionic liquids and water. Journal of Physical Chemistry B 105,
10942–10949. doi:10.1021/jp0112368 (2001).
17. Li, J., Li, X., Gunathunge, C. M. & Waegele, M. M. Hydrogen bonding steers the product
selectivity of electrocatalytic CO reduction. Proceedings of the National Academy of
Sciences 116, 9220–9229 (2019).
18. Broderick, A. & Newberg, J. T. in Ionic Liquids: Current State and Future Directions
227–249 (ACS Publications, 2017).
19. Cui, T., Lahiri, A., Carstens, T., Borisenko, N., Pulletikurthi, G., Kuhl, C., et al. Influence
of Water on the Electrified Ionic Liquid/Solid Interface: A Direct Observation of the
Transition from a Multilayered Structure to a Double-Layer Structure. The Journal of
Physical Chemistry C 120, 9341–9349. doi:10.1021/acs.jpcc.6b02549 (2016).
20. Sakai, K., Okada, K., Uka, A., Misono, T., Endo, T., Sasaki, S., et al. Effects of water on
solvation layers of imidazolium-type room temperature ionic liquids on silica and mica.
Langmuir 31, 6085–6091 (2015).
21. Smith, A. M., Perkin, S., et al. Switching the structural force in ionic liquid-solvent
mixtures by varying composition. Physical review letters 118, 096002 (2017).
22. Jurado, L. A., Kim, H., Rossi, A., Arcifa, A., Schuh, J. K., Spencer, N. D., et al. Effect of
the environmental humidity on the bulk, interfacial and nanoconfined properties of an ionic
liquid. Physical Chemistry Chemical Physics 18, 22719–22730 (2016).
59
23. Sarkar, S., Maitra, A., Banerjee, S., Thoi, V. S. & Dawlaty, J. M. Electric Fields at
Metal-Surfactant Interfaces: A Combined Vibrational Spectroscopy and Capacitance
Study. Journal of Physical Chemistry B 124, 1311–1321.
doi:10.1021/acs.jpcb.0c00560 (2020).
24. Yamada, T., Tominari, Y., Tanaka, S. & Mizuno, M. Infrared Spectroscopy of Ionic Liquids
Consisting of Imidazolium Cations with Different Alkyl Chain Lengths and Various
Halogen or Molecular Anions with and without a Small Amount of Water. Journal of
Physical Chemistry B 121, 3121–3129. doi:10.1021/acs.jpcb.7b01429 (2017).
25. Yaghini, N., Pitawala, J., Matic, A. & Martinelli, A. Effect of water on the local structure
and phase behavior of imidazolium-based protic ionic liquids. Journal of Physical
Chemistry B 119, 1611–1622. doi:10.1021/jp510691e (2015).
26. Jurado, L. A., Kim, H., Rossi, A., Arcifa, A., Schuh, J. K., Spencer, N. D., et al. Effect of
the environmental humidity on the bulk, interfacial and nanoconfined properties of an ionic
liquid. Physical Chemistry Chemical Physics 18, 22719–22730.
doi:10.1039/c6cp03777a (2016).
27. Motobayashi, K. & Osawa, M. Potential-dependent condensation of Water at the Interface
between ionic liquid [BMIM][TFSA] and an Au electrode. Electrochemistry
Communications 65, 14–17. doi:10.1016/j.elecom.2016.01.018 (2016).
28. Kramer, P. L., Giammanco, C. H. & Fayer, M. D. Dynamics of water, methanol, and
ethanol in a room temperature ionic liquid. Journal of Chemical Physics 142, 212408.
doi:10.1063/1.4914156 (2015).
29. Gomez, I. J., Arnaiz, B., Cacioppo, M., Arcudi, F. & Prato, M. Nitrogen-doped Carbon
Nanodots for bioimaging and delivery of paclitaxel. Journal of Materials Chemistry B 6,
5540–5548. doi:10.1039/x0xx00000x (2018).
30. Ren, Z., Brinzer, T., Dutta, S. & Garrett-Roe, S. Thiocyanate as a Local Probe of Ultrafast
Structure and Dynamics in Imidazolium-Based Ionic Liquids: Water-Induced
Heterogeneity and Cation-Induced Ion Pairing. Journal of Physical Chemistry B 119,
4699–4712. doi:10.1021/jp512851v (2015).
31. Thomaz, J. E., Lawler, C. M. & Fayer, M. D. The Influence of Water on the Alkyl Region
Structure in Variable Chain Length Imidazolium-Based Ionic Liquid/Water Mixtures.
Journal of Physical Chemistry B 120, 10350–10357. doi:10.1021/acs.jpcb.6b07853
(2016).
32. Sturlaugson, A. L., Fruchey, K. S. & Fayer, M. D. Orientational dynamics of room
temperature ionic liquid/water mixtures: Water-induced structure. Journal of Physical
Chemistry B 116, 1777–1787. doi:10.1021/jp209942r (2012).
60
33. Brinzer, T., Daly Jr, C. A., Allison, C., Garrett-Roe, S. & Corcelli, S. A. Modeling Carbon
Dioxide Vibrational Frequencies in Ionic Liquids: III. Dynamics and Spectroscopy. The
Journal of Physical Chemistry B 122, 8931–8942 (2018).
34. Ren, Z. & Garrett-Roe, S. Reorientation-induced spectral diffusion of non-isotropic
orientation distributions. The Journal of chemical physics 147, 144504 (2017).
35. Tamimi, A., Bailey, H. E. & Fayer, M. D. Alkyl Chain Length Dependence of the
Dynamics and Structure in the Ionic Regions of Room-Temperature Ionic Liquids. The
Journal of Physical Chemistry B 120, 7488–7501 (2016).
36. Giammanco, C. H., Kramer, P. L., Yamada, S. A., Nishida, J., Tamimi, A. & Fayer, M. D.
Carbon dioxide in an ionic liquid: Structural and rotational dynamics. The Journal of
chemical physics 144, 104506 (2016).
37. Fayer, M. D. Dynamics and structure of room temperature ionic liquids. Chemical Physics
Letters 616, 259–274 (2014).
38. Penalber, C. Y., Baker, G. A. & Baldelli, S. Sum frequency generation spectroscopy of ˜
imidazolium-based ionic liquids with cyano-functionalized anions at the solid salt–liquid
interface. The Journal of Physical Chemistry B 117, 5939–5949 (2013).
39. Rivera-Rubero, S. & Baldelli, S. Surface characterization of 1-butyl-3-methylimidazolium
Br-, I-, PF6-, BF4-,(CF3SO2) 2N-, SCN-, CH3SO3-, CH3SO4-, and (CN) 2N-ionic liquids
by sum frequency generation. The Journal of Physical Chemistry B 110, 4756–4765
(2006).
40. Baldelli, S. Surface structure at the ionic liquid- electrified metal interface. Accounts of
chemical research 41, 421–431 (2008).
41. Rivera-Rubero, S. & Baldelli, S. Influence of water on the surface of hydrophilic and
hydrophobic room-temperature ionic liquids. Journal of the American Chemical Society
126, 11788–11789 (2004).
42. Garc´ıa, N. & Dlott, D. Vibrational Sum Frequency Study of the Influence of Water-Ionic
Liquid Mixtures in the CO2 Electroreduction on Silver Electrodes in 70th International
Symposium on Molecular Spectroscopy: June 22-26, 2015 at The University of Illinois at
Urbana-Champaign. Talk FB05 1 (2015).
43. Garc´ıa Rey, N. & Dlott, D. D. Structural Transition in an Ionic Liquid Controls CO2
Electrochemical Reduction. Journal of Physical Chemistry C 119, 20892–20899.
doi:10.1021/acs.jpcc.5b03397 (2015).
61
44. Rosen, B. A., Haan, J. L., Mukherjee, P., Braunschweig, B., Zhu, W., Salehi-Khojin, A.,
et al. In situ spectroscopic examination of a low overpotential pathway for carbon dioxide
conversion to carbon monoxide. The Journal of Physical Chemistry C 116, 15307–15312
(2012).
45. Sorenson, S. A., Patrow, J. G. & Dawlaty, J. M. Solvation reaction field at the interface
measured by vibrational sum frequency generation spectroscopy. Journal of the American
Chemical Society 139, 2369–2378 (2017).
46. Patrow, J. G., Sorenson, S. A. & Dawlaty, J. M. Direct spectroscopic measurement of
interfacial electric fields near an electrode under polarizing or current-carrying conditions.
The Journal of Physical Chemistry C 121, 11585–11592 (2017).
47. Staffa, J. K., Lorenz, L., Stolarski, M., Murgida, D. H., Zebger, I., Utesch, T., et al.
Determination of the local electric field at Au/SAM interfaces using the vibrational Stark
effect. The Journal of Physical Chemistry C 121, 22274–22285 (2017).
48. Bagchi, S., Fried, S. D. & Boxer, S. G. A solvatochromic model calibrates nitriles’
vibrational frequencies to electrostatic fields. Journal of the American Chemical Society
134, 10373–10376. doi:10.1021/ja303895k (2012).
49. Aschaffenburg, D. J. & Moog, R. S. Probing hydrogen bonding environments:
Solvatochromic effects on the CN vibration of benzonitrile. Journal of Physical Chemistry
B 113, 12736–12743. doi:10.1021/jp905802a (2009).
50. Eisenthal, K. Liquid interfaces probed by second-harmonic and sum-frequency
spectroscopy. Chemical Reviews 96, 1343–1360 (1996).
51. Hayes, P. L., Malin, J. N., Jordan, D. S. & Geiger, F. M. Get charged up: Nonlinear optical
voltammetry for quantifying the thermodynamics and electrostatics of metal cations at
aqueous/oxide interfaces. Chemical Physics Letters 499, 183–192 (2010).
52. Wijesuriya, S., Burugapalli, K., Mackay, R., Ajaezi, G. & Balachandran, W. Chemically
roughened solid silver: A simple, robust and broadband SERS substrate. Sensors 16, 1742
(2016).
53. Ha, S. H., Mai, N. L. & Koo, Y.-M. Microwave-assisted separation of ionic liquids from
aqueous solution of ionic liquids. Journal of Chromatography A 1217, 7638–7641 (2010).
54. Jiang, W., Wang, Y. & Voth, G. A. Molecular Dynamics Simulation of Nanostructural
Organization in Ionic Liquid/Water Mixtures. The Journal of Physical Chemistry B 111.
PMID: 17388424, 4812–4818. doi:10.1021/jp067142l (2007).
55. Evans, K. O. Room-temperature ionic liquid cations act as short-chain surfactants and
disintegrate a phospholipid bilayer. Colloids and Surfaces A: Physicochemical and
Engineering Aspects 274, 11–17 (2006).
62
56. Deng, G. H., Li, X., Guo, Y. Q., Liu, S. L., Lu, Z. & Guo, Y. Orientation and structure of
ionic liquid cation at air/[bmim][BF 4] aqueous solution interface. Chinese Journal of
Chemical Physics 26, 569–575. doi:10.1063/1674-0068/26/05/569-575 (2013).
57. Deng, G., Guo, Y., Li, X., Zhang, Z., Liu, S., Lu, Z., et al. Surface of room temperature
ionic liquid [bmim][PF6] studied by polarization- and experimental
configuration-dependent sum frequency generation vibrational spectroscopy. Science
China Chemistry 58, 439–447. doi:10.1007/s11426-014-5241-5 (2015).
58. Rivera-Rubero, S. & Baldelli, S. Influence of water on the surface of the water-miscible
ionic liquid 1-butyl-3-methylimidazolium tetrafluoroborate: A sum frequency generation
analysis. Journal of Physical Chemistry B 110, 15499–15505. doi:10.1021/jp062694p
(2006).
59. Xu, S., Xing, S., Pei, S. S., Ivanistˇ sev, V., Lynden-Bell, R. & Baldelli, S. Molecular ˇ
Response of 1-Butyl-3-Methylimidazolium Dicyanamide Ionic Liquid at the Graphene
Electrode Interface Investigated by Sum Frequency Generation Spectroscopy and
Molecular Dynamics Simulations. Journal of Physical Chemistry C 119, 26009–26019.
doi:10.1021/acs.jpcc.5b08736 (2015).
60. Feaster, J. T., Jongerius, A. L., Liu, X., Urushihara, M., Nitopi, S. A., Hahn, C., et al.
Understanding the Influence of [EMIM]Cl on the Suppression of the Hydrogen Evolution
Reaction on Transition Metal Electrodes. Langmuir 33, 9464–9471 (2017).
61. Urushihara, M., Chan, K., Shi, C. & Nørskov, J. K. Theoretical Study of EMIM+
Adsorption on Silver Electrode Surfaces. Journal of Physical Chemistry C 119,
20023–20029 (2015).
62. Varela, A. S., Ju, W., Reier, T. & Strasser, P. Tuning the Catalytic Activity and Selectivity
of Cu for CO2 Electroreduction in the Presence of Halides. ACS Catalysis 6, 2136–2144
(2016).
63. McHugh, P. J., Stergiou, A. D. & Symes, M. D. Decoupled Electrochemical Water
Splitting: From Fundamentals to Applications. Advanced Energy Materials 10, 2002453.
doi:10.1002/aenm.202002453 (2020).
64. Chan, K. A few basic concepts in electrochemical carbon dioxide reduction. Nature
Communications 11, 2–5. doi:10.1038/s41467-020-19369-6 (2020).
65. Dunwell, M., Lu, Q., Heyes, J. M., Rosen, J., Chen, J. G., Yan, Y., et al. The Central Role
of Bicarbonate in the Electrochemical Reduction of Carbon Dioxide on Gold. Journal of
the American Chemical Society 139, 3774–3783. doi:10.1021/jacs.6b13287 (2017).
63
66. Kas, R., Ayemoba, O., Firet, N. J., Middelkoop, J., Smith, W. A. & Cuesta, A. In-Situ
Infrared Spectroscopy Applied to the Study of the Electrocatalytic Reduction of CO2:
Theory, Practice and Challenges. ChemPhysChem 20, 2904–2925.
doi:10.1002/cphc.201900533 (2019).
67. Garcia, A. C., Kolb, M. J., Van Nierop Y Sanchez, C., Vos, J., Birdja, Y. Y., Kwon, Y.,
et al. Strong Impact of Platinum Surface Structure on Primary and Secondary Alcohol
Oxidation during Electro-Oxidation of Glycerol. ACS Catalysis 6, 4491–4500.
doi:10.1021/acscatal.6b00709 (2016).
68. Rodriguez, P., Kwon, Y. & Koper, M. T. The promoting effect of adsorbed carbon
monoxide on the oxidation of alcohols on a gold catalyst. Nature Chemistry 4, 177–182.
doi:10.1038/nchem.1221 (2012).
69. Santasalo-Aarnio, A., Kwon, Y., Ahlberg, E., Kontturi, K., Kallio, T. & Koper, M. T.
Comparison of methanol, ethanol and iso-propanol oxidation on Pt and Pd electrodes in
alkaline media studied by HPLC. Electrochemistry Communications 13, 466–469.
doi:10.1016/j.elecom.2011.02.022 (2011).
70. Lu, X., Song, H., Cai, J. & Lu, S. Recent development of electrochemical nitrate reduction
to ammonia: A mini review. Electrochemistry Communications 129, 107094.
doi:10.1016/j.elecom.2021.107094 (2021).
71. McEnaney, J. M., Blair, S. J., Nielander, A. C., Schwalbe, J. A., Koshy, D. M.,
Cargnello, M., et al. Electrolyte engineering for efficient electrochemical nitrate reduction
to ammonia on a titanium electrode. ACS Sustainable Chemistry and Engineering 8,
2672–2681. doi:10.1021/acssuschemeng.9b05983 (2020).
72. Dubouis, N., Lemaire, P., Mirvaux, B., Salager, E., Deschamps, M. & Grimaud, A. The
role of the hydrogen evolution reaction in the solid–electrolyte interphase formation
mechanism for “Water-in-Salt” electrolytes. Energy & Environmental Science 11,
3491–3499 (2018).
73. Jamesh, M. I. & Sun, X. Recent progress on earth abundant electrocatalysts for oxygen
evolution reaction (OER) in alkaline medium to achieve efficient water splitting – A
review. Journal of Power Sources 400, 31–68. doi:10.1016/j.jpowsour.2018.07.125
(2018).
74. Mohammed-Ibrahim, J. & Xiaoming, S. Recent progress on earth abundant
electrocatalysts for hydrogen evolution reaction (HER) in alkaline medium to achieve
efficient water splitting – A review. Journal of Energy Chemistry 34, 111–160.
doi:10.1016/j.jechem.2018.09.016 (2019).
75. Antonini, E., Carrea, G. & Cremonesi, P. Enzyme catalysed reactions in water - Organic
solvent two-phase systems. Enzyme and Microbial Technology 3, 291–296.
doi:10.1016/0141-0229(81)90002-8 (1981).
64
76. Savelli, G., Spreti, N. & Di Profio, P. Enzyme activity and stability control by amphiphilic
self-organizing systems in aqueous solutions. Current Opinion in Colloid and Interface
Science 5, 111–117. doi:10.1016/S1359-0294(00)00043-1 (2000).
77. Hahn-Hagerdal, B. Water activity: a possible external regulator in biotechnical processes. ¨
Enzyme and Microbial Technology 8, 322–327. doi:10.1016/0141-0229(86)90129-8
(1986).
78. Pocker, Y. Water in enzyme reactions: Biophysical aspects of hydration-dehydration
processes. Cellular and Molecular Life Sciences 57, 1008–1017.
doi:10.1007/PL00000741 (2000).
79. Miyake, T. & Rolandi, M. Grotthuss mechanisms: From proton transport in proton wires to
bioprotonic devices. Journal of Physics Condensed Matter 28.
doi:10.1088/0953-8984/28/2/023001 (2016).
80. Nagle, J. F. & Tristram-Nagle, S. Hydrogen bonded chain mechanisms for proton
conduction and proton pumping. The Journal of Membrane Biology 74, 1–14.
doi:10.1007/BF01870590 (1983).
81. Buch-Pedersen, M. J., Pedersen, B. P., Veierskov, B., Nissen, P. & Palmgren, M. G. Protons
and how they are transported by proton pumps. Pflugers Archiv European Journal of
Physiology 457, 573–579. doi:10.1007/s00424-008-0503-8 (2009).
82. Kocsis, I., Sorci, M., Vanselous, H., Murail, S., Sanders, S. E., Licsandru, E., et al.
Oriented chiral water wires in artificial transmembrane channels. Science Advances 4.
doi:10.1126/sciadv.aao5603 (2018).
83. Pal, S. K. & Zewail, A. H. Dynamics of water in biological recognition. Chemical Reviews
104, 2099–2123. doi:10.1021/cr020689l (2004).
84. Despa, F. Biological water: Its vital role in macromolecular structure and function. Annals
of the New York Academy of Sciences 1066, 1–11. doi:10.1196/annals.1363.023
(2006).
85. Israelachvili, J. & Wennerstrom, H. Role of hydration and water structure in biological and ¨
colloidal interactions. Nature 379, 219–225 (1996).
86. Asokan, A. & Cho, M. J. Exploitation of intracellular pH gradients in the cellular delivery
of macromolecules. Journal of Pharmaceutical Sciences 91, 903–913.
doi:10.1002/jps.10095 (2002).
87. Hohner, R., Aboukila, A., Kunz, H. H. & Venema, K. Proton gradients and ¨
proton-dependent transport processes in the chloroplast. Frontiers in Plant Science 7, 1–7.
doi:10.3389/fpls.2016.00218 (2016).
65
88. Sanders, S. E. & Petersen, P. B. Heterodyne-detected sum frequency generation of water at
surfaces with varying hydrophobicity. Journal of Chemical Physics 150.
doi:10.1063/1.5078587 (2019).
89. Garrett, B. C., Dixon, D. A., Camaioni, D. M., Chipman, D. M., Johnson, M. A.,
Jonah, C. D., et al. Role of water in electron-initiated processes and radical chemistry:
Issues and scientific advances. Chemical Reviews 105, 355–389. doi:10.1021/cr030453x
(2005).
90. Reddy, S. K., Thiraux, R., Wellen Rudd, B. A., Lin, L., Adel, T., Joutsuka, T., et al. Bulk
Contributions Modulate the Sum-Frequency Generation Spectra of Water on Model
Sea-Spray Aerosols. Chem 4, 1629–1644. doi:10.1016/j.chempr.2018.04.007 (2018).
91. Eftekhari-Bafrooei, A. & Borguet, E. Effect of electric fields on the ultrafast vibrational
relaxation of water at a charged solid-liquid interface as probed by vibrational sum
frequency generation. Journal of Physical Chemistry Letters 2, 1353–1358.
doi:10.1021/jz200194e (2011).
92. Motobayashi, K. & Osawa, M. Recent advances in spectroscopic investigations on ionic
liquid/electrode interfaces. Current Opinion in Electrochemistry 8, 147–155.
doi:10.1016/j.coelec.2018.05.003 (2018).
93. Gunathunge, C. M., Li, J., Li, X. & Waegele, M. M. Surface-Adsorbed CO as an Infrared
Probe of Electrocatalytic Interfaces. ACS Catalysis 10, 11700–11711.
doi:10.1021/acscatal.0c03316 (2020).
94. Poburko, D. & Demaurex, N. Regulation of the mitochondrial proton gradient by cytosolic
Ca 2+ signals. Pflugers Archiv European Journal of Physiology 464, 19–26.
doi:10.1007/s00424-012-1106-y (2012).
95. Liu, D., Ma, G. & Allen, H. C. Adsorption of 4-picoline and piperidine to the hydrated
SiO2 surface: Probing the surface acidity with vibrational sum frequency generation
spectroscopy. Environmental Science and Technology 39, 2025–2032.
doi:10.1021/es0482280 (2005).
96. Chen, X., Hua, W., Huang, Z. & Allen, H. C. Interfacial water structure associated with
phospholipid membranes studied by phase-sensitive vibrational sum frequency generation
spectroscopy. Journal of the American Chemical Society 132, 11336–11342.
doi:10.1021/ja1048237 (2010).
97. Osawa, M., Tsushima, M., Mogami, H., Samjeske, G. & Yamakata, A. Structure of water ´
at the electrified platinum-water interface: A study by surface-enhanced infrared
absorption spectroscopy. Journal of Physical Chemistry C 112, 4248–4256.
doi:10.1021/jp710386g (2008).
66
98. Ataka, K. I., Yotsuyanagi, T. & Osawa, M. Potential-dependent reorientation of water
molecules at an electrode/electrolyte interface studied by surface-enhanced infrared
absorption spectroscopy. Journal of Physical Chemistry 100, 10664–10672.
doi:10.1021/jp953636z (1996).
99. De Beer, A. G., Samson, J. S., Hua, W., Huang, Z., Chen, X., Allen, H. C., et al. Direct
comparison of phase-sensitive vibrational sum frequency generation with maximum
entropy method: Case study of water. Journal of Chemical Physics 135.
doi:10.1063/1.3662469 (2011).
100. Baldelli, S., Mailhot, G., Ross, P. N. & Somorjai, G. A. Potential-dependent vibrational
spectroscopy of solvent molecules at the Pt(111) electrode in a water/acetonitrile mixture
studied by sum frequency generation. Journal of the American Chemical Society 123,
7697–7702. doi:10.1021/ja016063e (2001).
101. Laaser, J. E. & Zanni, M. T. Extracting structural information from the polarization
dependence of one- and two-dimensional sum frequency generation spectra. Journal of
Physical Chemistry A 117, 5875–5890. doi:10.1021/jp307721y (2013).
102. Ricks, A. M., Anfuso, C. L., Rodrıguez-Cordoba, W. & Lian, T. Vibrational relaxation ´
dynamics of catalysts on TiO2 Rutile (1 1 0) single crystal surfaces and anatase
nanoporous thin films. Chemical Physics 422, 264–271 (2013).
103. Backus, E. H., Bonn, D., Cantin, S., Roke, S. & Bonn, M. Laser-heating-induced
displacement of surfactants on the water surface. Journal of Physical Chemistry B 116,
2703–2712. doi:10.1021/jp2074545 (2012).
104. Calegari Andrade, M. F., Ko, H. Y., Car, R. & Selloni, A. Structure, Polarization, and Sum
Frequency Generation Spectrum of Interfacial Water on Anatase TiO 2. Journal of
Physical Chemistry Letters 9, 6716–6721. doi:10.1021/acs.jpclett.8b03103 (2018).
105. Wang, C. Y., Groenzin, H. & Shultz, M. J. Comparative study of acetic acid, methanol, and
water adsorbed on anatase TiO2 probed by sum frequency generation spectroscopy.
Journal of the American Chemical Society 127, 9736–9744. doi:10.1021/ja051996m
(2005).
106. Achtyl, J. L., Vlassiouk, I. V., Dai, S. & Geiger, F. Interactions of organic solvents at
graphene/α-Al2O 3 and graphene oxide/α-Al2O3 interfaces studied by sum frequency
generation. Journal of Physical Chemistry C 118, 17745–17755. doi:10.1021/jp5047547
(2014).
107. Vanselous, H., Videla, P. E., Batista, V. S. & Petersen, P. B. Distinct Binding of Rhenium
Catalysts on Nanostructured and Single-Crystalline TiO2 Surfaces Revealed by
Two-Dimensional Sum Frequency Generation Spectroscopy. Journal of Physical
Chemistry C 122, 26018–26031. doi:10.1021/acs.jpcc.8b08423 (2018).
67
108. McDaniel, J. G. & Son, C. Y. Ion correlation and collective dynamics in BMIM/BF4-based
organic electrolytes: From dilute solutions to the ionic liquid limit. The Journal of Physical
Chemistry B 122, 7154–7169 (2018).
109. Callaghan, A., Doyle, R., Alexander, E. & Palepu, R. Thermodynamic Properties of
Micellization and Adsorption and Electrochemical Studies of Hexadecylpyridinium
Bromide in Binary Mixtures of 1,2-Ethanediol with Water. Langmuir 9, 3422–3426.
doi:10.1021/la00036a016 (1993).
110. Ferreira Batista, F. H., Amaral Siqueira, L. J., Costa, L. T. & Ferreira, T. L. Understanding
the Influence of the Electrochemical Behavior of Potassium Hexacyanoferrate(II/III) on
Acetonitrile/Water Binary Mixtures: An Experimental and Molecular Dynamics Study.
Journal of Physical Chemistry C 122, 2717–2725. doi:10.1021/acs.jpcc.7b10989
(2018).
111. Lee, A. A., Perez-Martinez, C. S., Smith, A. M. & Perkin, S. Underscreening in
concentrated electrolytes. Faraday Discussions 199, 239–259. doi:10.1039/c6fd00250a
(2017).
112. Davis, J. G., Gierszal, K. P., Wang, P. & Ben-Amotz, D. Water structural transformation at
molecular hydrophobic interfaces. Nature 491, 582–585. doi:10.1038/nature11570
(2012).
113. Baldelli, S., Schnitzer, C., Jane Shultz, M. & Campbell, D. J. Sum frequency generation
investigation of water at the surface of H2O/H2SO4 and H2O/Cs2SO4 binary systems.
Chemical Physics Letters 287, 143–147. doi:10.1016/S0009-2614(98)00161-4 (1998).
114. Sarkar, S., Maitra, A., Banerjee, S., Thoi, V. S. & Dawlaty, J. M. Electric fields at
metal–surfactant interfaces: a combined vibrational spectroscopy and capacitance study.
The Journal of Physical Chemistry B 124, 1311–1321 (2020).
115. Banerjee, S., Zhang, Z.-Q., Hall, A. S. & Thoi, V. S. Surfactant perturbation of cation
interactions at the electrode–electrolyte interface in carbon dioxide reduction. ACS
Catalysis 10, 9907–9914 (2020).
116. Banerjee, S., Gerke, C. S. & Thoi, V. S. Guiding CO2RR Selectivity by Compositional
Tuning in the Electrochemical Double Layer. Accounts of Chemical Research 55,
504–515. doi:10.1021/acs.accounts.1c00680 (2022).
117. Zhang, Z. Q., Banerjee, S., Thoi, V. S. & Shoji Hall, A. Reorganization of Interfacial Water
by an Amphiphilic Cationic Surfactant Promotes CO2Reduction. Journal of Physical
Chemistry Letters 11, 5457–5463. doi:10.1021/acs.jpclett.0c01334 (2020).
68
118. Gutierrez-S ´ anchez, O., Daems, N., Bulut, M., Pant, D. & Breugelmans, T. Effects of ´
Benzyl-Functionalized Cationic Surfactants on the Inhibition of the Hydrogen Evolution
Reaction in CO2Reduction Systems. ACS Applied Materials and Interfaces 13,
56205–56216. doi:10.1021/acsami.1c17303 (2021).
119. Malik, M. A., Hashim, M. A., Nabi, F., AL-Thabaiti, S. A. & Khan, Z. Anti-corrosion
ability of surfactants: A review. International Journal of Electrochemical Science 6,
1927–1948 (2011).
120. Zhu, S., Jiang, B., Cai, W. B. & Shao, M. Direct Observation on Reaction Intermediates
and the Role of Bicarbonate Anions in CO2 Electrochemical Reduction Reaction on Cu
Surfaces. Journal of the American Chemical Society 139, 15664–15667.
doi:10.1021/jacs.7b10462 (2017).
121. Osawa, M., Ataka, K. I., Ikeda, M., Uchihara, H. & Nanba, R. Surface enhanced infrared
absorption spectroscopy: Mechanism and application to trace analysis. Analytical Sciences
7, 503–506. doi:10.2116/analsci.7.Supple_503 (1991).
122. Osawa, M. Dynamic Processes in Electrochemical Reactions Studied by Surface-Enhanced
Infrared Absorption Spectroscopy (SEIRAS) 1997. doi:10.1246/bcsj.70.2861.
123. Osawa, M., ichi Ataka, K., Yoshii, K. & Yotsuyanagi, T. Surface-enhanced infrared ATR
spectroscopy for in situ studies of electrode/electrolyte interfaces. Journal of Electron
Spectroscopy and Related Phenomena 64-65, 371–379.
doi:10.1016/0368-2048(93)80099-8 (1993).
124. Jensen, T. R., Van Duyne, R. P., Johnson, S. A. & Maroni, V. A. Surface-enhanced infrared
spectroscopy: A comparison of metal island films with discrete and nondiscrete surface
plasmons. Applied Spectroscopy 54, 371–377. doi:10.1366/0003702001949654 (2000).
125. Bao, W. J., Li, J., Li, J., Zhang, Q. W., Liu, Y., Shi, C. F., et al. Au/ZnSe-Based Surface
Enhanced Infrared Absorption Spectroscopy as a Universal Platform for Bioanalysis.
Analytical Chemistry 90, 3842–3848. doi:10.1021/acs.analchem.7b04505 (2018).
126. Coker, D., Reimers, J. & Watts, R. The Infrared Absorption Spectrum of Water. Australian
Journal of Physics 35, 623. doi:10.1071/ph820623 (1982).
127. Senior, W. A. & Thompson, W. K. Assignment of the infra-red and raman bands of liquid
water [9]. Nature 205, 170. doi:10.1038/205170a0 (1965).
128. Elderderi, S., Sacre, P. Y., Wils, L., Chourpa, I., Elbashir, A. A., Hubert, P., ´ et al.
Comparison of Vibrational Spectroscopic Techniques for Quantification of Water in
Natural Deep Eutectic Solvents. Molecules 27, 4819. doi:10.3390/molecules27154819
(2022).
69
129. Sung, W., Kim, D. & Shen, Y. R. Sum-frequency vibrational spectroscopic studies of
Langmuir monolayers. Current Applied Physics 13, 619–632.
doi:10.1016/j.cap.2012.12.002 (2013).
130. Skripkin, M. Y., Lindqvist-Reis, P., Abbasi, A., Mink, J., Persson, I. & Sandstrom, M. ¨
Vibrational spectroscopic force field studies of dimethyl sulfoxide and hexakis(dimethyl
sulfoxide)scandium(III) iodide, and crystal and solution structure of the hexakis(dimethyl
sulfoxide)scandium(III) ion. Dalton Transactions, 4038–4049. doi:10.1039/b413486a
(2004).
131. Yi, J., You, E. M., Ding, S. Y. & Tian, Z. Q. Unveiling the molecule-plasmon interactions
in surface-enhanced infrared absorption spectroscopy. National Science Review 7,
1228–1238. doi:10.1093/NSR/NWAA054 (2020).
132. Andvaag, I. R., Lins, E. & Burgess, I. J. An Effective Medium Theory Description of
Surface-Enhanced Infrared Absorption from Metal Island Layers Grown on Conductive
Metal Oxide Films. Journal of Physical Chemistry C 125, 22301–22311.
doi:10.1021/acs.jpcc.1c06627 (2021).
133. Krauth, O., Fahsold, G. & Pucci, A. Asymmetric line shapes and surface enhanced infrared
absorption of CO adsorbed on thin iron films on MgO(001). Journal of Chemical Physics
110, 3113–3117. doi:10.1063/1.477907 (1999).
134. Tainter, C. J., Ni, Y., Shi, L. & Skinner, J. L. Hydrogen bonding and oh-stretch
spectroscopy in water: Hexamer (Cage), liquid surface, liquid, and ice. Journal of Physical
Chemistry Letters 4, 12–17. doi:10.1021/jz301780k (2013).
135. Lock, A. J., Gilijamse, J. J., Woutersen, S. & Bakker, H. J. Vibrational relaxation and
coupling of two OH-stretch oscillators with an intramolecular hydrogen bond. Journal of
Chemical Physics 120, 2351–2358. doi:10.1063/1.1637576 (2004).
136. Dorchies, F., Serva, A., Crevel, D., De Freitas, J., Kostopoulos, N., Robert, M., et al.
Controlling the Hydrophilicity of the Electrochemical Interface to Modulate the
Oxygen-Atom Transfer in Electrocatalytic Epoxidation Reactions. Journal of the American
Chemical Society 144, 22734–22746 (2022).
137. Killian, M. M., Villa-Aleman, E., Sun, Z., Crittenden, S. & Leverette, C. L. Dependence of
surface-enhanced infrared absorption (SEIRA) enhancement and spectral quality on the
choice of underlying substrate: A closer look at silver (Ag) films prepared by physical
vapor deposition (PVD). Applied Spectroscopy 65, 272–283 (2011).
138. Levering, L. M., Hayes, C. J., Callahan, K. M., Hadad, C. M. & Allen, H. C. Non-aqueous
solvation of n-octanol and ethanol: Spectroscopic and computational studies. Journal of
Physical Chemistry B 110, 6325–6331. doi:10.1021/jp0570465 (2006).
70
139. Brown, L. V., Zhao, K., King, N., Sobhani, H., Nordlander, P. & Halas, N. J.
Surface-enhanced infrared absorption using individual cross antennas tailored to chemical
moieties. Journal of the American Chemical Society 135, 3688–3695 (2013).
140. Tseng, C., Pennathur, A. K., Blauth, D., Salazar, N. & Dawlaty, J. M. Direct Determination
of Plasmon Enhancement Factor and Penetration Depths in Surface Enhanced IR
Absorption Spectroscopy. Langmuir 39, 3179–3184 (2023).
141. Taylor, C. D., Wasileski, S. A., Filhol, J.-S. & Neurock, M. First principles reaction
modeling of the electrochemical interface: Consideration and calculation of a tunable
surface potential from atomic and electronic structure. Physical Review B 73, 165402
(2006).
142. De Souza, E., Ceotto, G. & Teschke, O. Dielectric constant measurements of interfacial
aqueous solutions using atomic force microscopy. Journal of Molecular Catalysis A:
Chemical 167, 235–243 (2001).
143. Varghese, S., Kannam, S. K., Hansen, J. S. & Sathian, S. P. Effect of hydrogen bonds on
the dielectric properties of interfacial water. Langmuir 35, 8159–8166 (2019).
144. Vargas-Rodr´ıguez, E. & Rutt, H. N. Design of CO, CO2 and CH4 gas sensors based on
correlation spectroscopy using a Fabry-Perot interferometer. Sensors and Actuators, B:
Chemical 137, 410–419. doi:10.1016/j.snb.2009.01.013 (2009).
145. Ding, T. Controllable tuning of Fabry-Perot cavities: Via laser printing. Journal of
Materials Chemistry C 8, 10825–10830. doi:10.1039/d0tc01662d (2020).
146. Rosencwaig, A. Photoacoustic Spectroscopy. Annual Review of Biophysics and
Bioengineering 9. PMID: 6994590, 31–54.
doi:10.1146/annurev.bb.09.060180.000335 (1980).
147. Snook, R. D. & Mitchem, L. Photoacoustic Spectroscopy. Encyclopedia of Analytical
Science: Second Edition 797, 174–180. doi:10.1016/B0-12-369397-7/00464-7 (2004).
148. Xu, D.-X., Delage, A., Verly, P., Janz, S., Wang, S., Vachon, M., ˆ et al. Empirical model for
the temperature dependence of silicon refractive index from O to C band based on
waveguide measurements. Optics Express 27, 27229. doi:10.1364/oe.27.027229 (2019).
149. McCaulley, J. A., Donnelly, V. M., Vernon, M. & Taha, I. Temperature dependence of the
near-infrared refractive index of silicon, gallium arsenide, and indium phosphide. Physical
Review B 49, 7408–7417. doi:10.1103/PhysRevB.49.7408 (1994).
150. Zhang, S., Zhao, Z., Chen, N., Pang, F., Chen, Z., Liu, Y., et al. Temperature characteristics
of silicon core optical fiber Fabry–Perot interferometer. Optics Letters 40, 1362.
doi:10.1364/ol.40.001362 (2015).
71
151. Vines, J. B., Yoon, J. H., Ryu, N. E., Lim, D. J. & Park, H. Gold nanoparticles for
photothermal cancer therapy 2019. doi:10.3389/fchem.2019.00167.
152. Li, X., Lovell, J. F., Yoon, J. & Chen, X. Clinical development and potential of
photothermal and photodynamic therapies for cancer 2020.
doi:10.1038/s41571-020-0410-2.
153. Chen, Q., Xu, L., Liang, C., Wang, C., Peng, R. & Liu, Z. Photothermal therapy with
immune-adjuvant nanoparticles together with checkpoint blockade for effective cancer
immunotherapy. Nature Communications 7. doi:10.1038/ncomms13193 (2016).
154. Selmke, M., Braun, M. & Cichos, F. Photothermal single-particle microscopy: Detection of
a nanolens. ACS Nano 6, 2741–2749. doi:10.1021/nn300181h (2012).
155. Nedosekin, D. A., Galanzha, E. I., Dervishi, E., Biris, A. S. & Zharov, V. P.
Super-resolution nonlinear photothermal microscopy. Small 10, 135–142.
doi:10.1002/smll.201300024 (2014).
156. Gaiduk, A., Yorulmaz, M., Ruijgrok, P. V. & Orrit, M. Room-Temperature Detection of a
Single Molecule’s Absorption by Photothermal Contrast. Science 330, 353–356.
doi:10.1126/science.1195475 (2010).
157. Singer, C., Goetz, A., Prasad, A. S., Becker, M., Rothhardt, M. & Skoff, S. M. Thermal
tuning of a fiber-integrated Fabry-Perot cavity. ´ Optics Express 29, 28778.
doi:10.1364/oe.433094 (2021).
158. Liu, G. & Han, M. Fiber-optic gas pressure sensing with a laser-heated silicon-based
Fabry–Perot interferometer. Optics Letters 40, 2461. doi:10.1364/ol.40.002461 (2015).
159. Macleod, H. A. & Macleod, H. A. Thin-film optical filters (CRC press, 2010).
160. Tian, Y., Guo, Z., Liu, Z., Lin, H., Li, X., Chen, J., et al. Efficiently enhanced the
visible-light absorption of monolayer WS2 by constructing an asymmetric Fabry-Perot
cavity. Materials Today Nano 14. doi:10.1016/j.mtnano.2021.100112 (2021).
161. Chen, M., Pang, S., Zhou, J., Wu, H., Hirokazu, M. & Kiyoshi, T. High-accuracy spectral
interferometer with multi-Fabry–Perot Etalon for thickness measurement of the silicon
wafer. Optics Communications 501, 1–8. doi:10.1016/j.optcom.2021.127346 (2021).
162. Hori, T. Structural optimization of silicon thin film for thermoelectric materials. Scientific
Reports, 1–9. doi:10.1038/s41598-021-01855-6 (2021).
163. Venkatasubramanian, R., Siivola, E., Colpitts, T. & Quinn, B. O. Thin- ® lm
thermoelectric devices with high room-temperature ® gures of merit, 597–602 (2001).
72
164. Hess, C., Wolf, M., Roke, S. & Bonn, M. Femtosecond time-resolved vibrational SFG
spectroscopy. 503, 304–312 (2002).
165. Ricks, A. M., Anfuso, C. L., Rodr´ıguez-cordoba, W. & Lian, T. Vibrational relaxation ´
dynamics of catalysts on TiO 2 Rutile ( 1 1 0 ) single crystal surfaces and anatase
nanoporous thin films. Chemical Physics 422, 264–271.
doi:10.1016/j.chemphys.2013.04.001 (2013).
166. Inoue, K.-i., Ishiyama, T., Nihonyanagi, S., Yamaguchi, S., Morita, A. & Tahara, T. E ffi
cient Spectral Di ff usion at the Air/Water Interface Revealed by Femtosecond
Time-Resolved Heterodyne-Detected Vibrational Sum Frequency Generation
Spectroscopy. doi:10.1021/acs.jpclett.6b00701 (2016).
167. Sasikala, V. & Chitra, K. All optical switching and associated technologies: a review.
Journal of Optics 47, 307–317 (2018).
168. Chai, Z., Hu, X., Wang, F., Niu, X., Xie, J. & Gong, Q. Ultrafast all-optical switching.
Advanced Optical Materials 5, 1600665 (2017).
169. Maker, A. J. & Armani, A. M. Low-loss silica-on-silicon waveguides (2011).
170. Soltani, S. & Armani, A. M. Optothermal transport behavior in whispering gallery mode
optical cavities. Applied Physics Letters 105. doi:10.1063/1.4892169 (5 2014).
171. Perot, A. & Fabry, C. On the Application of Interference Phenomena to the Solution of
Various Problems of Spectroscopy and Metrology. The Astrophysical Journal 9, 87.
doi:10.1086/140557 (1899).
172. Hecht, E. Optics (Pearson, 2012).
173. Tilli, M., Motooka, T., Airaksinen, V. M., Franssila, S., Paulasto-Krockel, M. & ¨
Lindroos, V. Handbook of Silicon Based MEMS Materials and Technologies: Second
Edition (2015).
174. Frey, B. J., Leviton, D. B. & Madison, T. J. Temperature-dependent refractive index of
silicon and germanium. Optomechanical Technologies for Astronomy 6273, 62732J.
doi:10.1117/12.672850 (2006).
175. Kudrna, J., Maly, P., Trojanek, F., ´ Stˇ epˇ anek, J., Lechner, T., Pelant, I., ´ et al. Ultrafast
carrier dynamics in undoped microcrystalline silicon. Materials Science and Engineering
B: Solid-State Materials for Advanced Technology 69, 238–242.
doi:10.1016/S0921-5107(99)00302-5 (2000).
176. Recombination of an electron-hole plasma in silicon under the action of femtosecond laser
pulses. JETP Letters 79. doi:10.1134/1.1787099 (11 2004).
73
177. Ghosh, G. Handbook of optical constants of solids: Handbook of thermo-optic coefficients
of optical materials with applications (Academic Press, 1998).
178. Baron Fourier, J. B. J. The analytical theory of heat (The University Press, 1878).
179. Semiconductor, V. Basic Mechanical and Thermal Properties of Silicon. Growth
(Lakeland) 22401, 1–4 (2000).
180. Hugi, A., Maulini, R. & Faist, J. External cavity quantum cascade laser. Semiconductor
Science and Technology 25. doi:10.1088/0268-1242/25/8/083001 (2010).
181. Spott, A., Peters, J., Davenport, M. L., Stanton, E. J., Merritt, C. D., Bewley, W. W., et al.
Quantum cascade laser on silicon at 4.8 µm. 2016 Conference on Lasers and
Electro-Optics, CLEO 2016 3. doi:10.1364/cleo_si.2016.sth3l.4 (2016).
182. Phillips, M. C., Craig, I. M. & Blake, T. A. Reflection-absorption infrared spectroscopy of
thin films using an external cavity quantum cascade laser in. 8631 (2013).
doi:10.1117/12.2002849.
183. Michel, A. P. M., Morrison, A. E., Colson, B. C., Pardis, W. A., Moya, X. A., Harb, C. C.,
et al. Quantum cascade laser-based reflectance spectroscopy: a robust approach for the
classification of plastic type. Optics Express 28. doi:10.1364/oe.393231 (12 2020).
184. Faist, J., Capasso, F., Sivco, D. L., Sirtori, C., Hutchinson, A. L. & Cho, A. Y. Quantum
cascade laser. Science 264, 553–556 (1994).
185. Faist, J., Capasso, F., Sivco, D. L., Sirtori, C., Albert, L. & Cho, A. Y. Quantum Cascade
Laser Published by : American Association for the Advancement of Science Stable URL :
http://www.jstor.org/stable/2883703. 264, 553–556 (2008).
186. Green, M. A. & Keevers, M. J. Optical properties of intrinsic silicon at 300 K. Progress in
Photovoltaics: Research and Applications 3. doi:10.1002/pip.4670030303 (3 1995).
187. Zhang, R., He, Y., Zhang, Y., An, S., Zhu, Q., Li, X., et al. Ultracompact and
low-power-consumption silicon thermo-optic switch for high-speed data. Nanophotonics
10, 937–945. doi:10.1515/nanoph-2020-0496 (2020).
188. Li, M., Razumtcev, A., Yang, R., Liu, Y., Rong, J., Geiger, A. C., et al.
Fluorescence-Detected Mid-Infrared Photothermal Microscopy. arXiv preprint
arXiv:2104.02900 (2021).
189. Imperatore, M., Asbury, J. & Giebink, N. English (US). Journal of Chemical Physics 154.
doi:10.1063/5.0046307 (2021).
74
190. Stiopkin, I. V., Weeraman, C., Pieniazek, P. A., Shalhout, F. Y., Skinner, J. L. &
Benderskii, A. V. Hydrogen bonding at the water surface revealed by isotopic dilution
spectroscopy. Nature 474, 192–195. doi:10.1038/nature10173 (2011).
191. Miranda, P., Xu, L., Shen, Y. & Salmeron, M. Icelike water monolayer adsorbed on mica at
room temperature. Physical review letters 81, 5876 (1998).
192. Hsu, W.-C., Zhen, C. & Wang, A. X. Electrically tunable high-quality factor silicon
microring resonator gated by high mobility conductive oxide. ACS Photonics 8,
1933–1936 (2021).
75
Appendices
Table 5.1: The volumes of water needed for different mole fractions in the case of EMIm BF4.
This was calculated using vw =
ρILMMw
ρwMMIL
χ
1−χ
vIL Where vw is the volume of water for a particular
mole fraction χ and volume of ionic liquid (vIL), ρIL is the density of the IL (EMIm BF4 ; 1.29
g/ml), MMIL is the molecular mass of the IL (EMIm BF4 ;197.97 g/mol), ρw is the density of
water and MMw is the molecular mass of water. The following result is for vIL = 0.1 ml.
χwater Volume of water added (in ml)
0.1 0.0013
0.2 0.0029
0.3 0.0050
0.4 0.0078
0.5 0.0118
0.6 0.0176
0.7 0.0275
0.8 0.471
0.9 0.106
0.9174 (5M Ionic Liquid) 0.131
0.9488 (3M Ionic Liquid) 0.218
0.9823 (1M Ionic Liquid) 0.653
0.9911 (0.5M Ionic Liquid) 1.31
76
Table 5.2: Summary of the EMImBF4 SERS results from the main text. We have included the
mole fraction, average values ∆ωCN,average values ΓCN, along with error bars for measurements.
The values of ∆ωCN and ΓCN for pure water (χwater=1) are ∼ + 2 cm−1
and ∼ 11 cm−1
.
χwater (∆ωCN)(cm−1
) (ΓCN)(cm−1
)
0 (pure Ionic Liquid) -3.36 ± 0.22 4.43 ± 0.40
0.1 -3.64±0.19 5.52±0.43
0.2 -2.87±0.19 4.92±0.28
0.3 -2.93±0.22 6.47±0.14
0.4 -3.19±0.15 6.60±0.17
0.5 -3.32±0.21 4.93±0.31
0.6 -4.54±0.69 5.02±0.41
0.7 -4.12±0.25 7.42±0.24
0.8 -2.19±0.5 5.15±0.10
0.9 -2.75±0.4 5.06±0.19
0.9174 (5M Ionic Liquid) -1.43±0.54 7.23±0.15
0.9488 (3M Ionic Liquid) -1.59±0.40 7.18±0.14
0.9823 (1M Ionic Liquid) -1.2±0.54 8.99±0.85
0.9911 (0.5M Ionic Liquid) 0.05±0.29 8.03±0.12
77
Table 5.3: Summary of the HMImBF4 SERS results from the main text. We have included the
mole fraction, average values ∆ωCN,average values ΓCN, along with error bars for measurements.
χwater (∆ωCN)(cm−1
) (ΓCN)(cm−1
)
0 (pure Ionic Liquid) -3.06 ± 1.26 8.32 ± 0.38
0.1 -4.35±0.55 7.84±0.97
0.2 -4.18±0.32 7.61±0.94
0.3 -5.33±0.26 6.97±0.48
0.4 -3.05±0.16 7.40±0.34
0.5 -3.16±0.53 8.53±0.14
0.6 -2.20±0.34 7.85±0.55
0.7 -3.60±1.6 7.69±0.86
0.9882 (0.6M Ionic Liquid) -1.74±0.30 7.24±0.42
0.9919 (0.4 M Ionic Liquid) -0.19±2.69 7.70±0.36
78
Table 5.4: Summary of the DMImBF4 SERS results from the main text. We have included the
mole fraction, average values ∆ωCN,average values ΓCN, along with error bars for measurements.
χwater (∆ωCN)(cm−1
) (ΓCN)(cm−1
)
0 (pure Ionic Liquid) -8.65 ± 0.35 8.16 ± 0.37
0.9982 (100 mM of Ionic Liquid) -6.52±0.20 7.94±0.38
0.9991 (50 mM of Ionic Liquid) -6.16±0.43 5.84±0.9991
0.9996 (25 mM of Ionic Liquid) -5.41±0.29 6.02±0.74
0.9998 (25 mM of Ionic Liquid) -5.41±0.29 10.83±0.39
0.9999 (10 mM of Ionic Liquid) -5.17±0.81 9.55±0.15
0.999982 (5 mM of Ionic Liquid) -3.88±0.20 8.48±0.16
0.9999982 (1 mM of Ionic Liquid) -2.99±0.20 14.87±0.64
79
Table 5.5: Summary of the EMImBF4 FTIR results from the main text. We have included the mole
fraction, average values ∆ωCN,average values ΓCN, along with error bars for measurements.The
following values are from fitting the peak to a gaussian.
χwater (∆ωCN)(cm−1
) (ΓCN)(cm−1
)
0 (pure Ionic Liquid) -1.13 ± 0.01 5.13 ± 0.01
0.1 -0.92±0.01 5.19±0.01
0.2 -0.90±0.01 5.18±0.01
0.3 -0.75±0.01 5.20±0.01
0.4 -0.54±0.01 5.23±0.02
0.5 -0.29±0.01 5.31±0.02
0.6 0.08±0.01 5.46±0.02
0.7 0.52±0.02 5.63±0.03
0.8 1.27±0.02 5.92±0.04
0.9 2.46±0.02 6.25±0.05
80
Table 5.6: Summary of the HMImBF4 FTIR results from the main text. We have included the mole
fraction, average values ∆ωCN,average values ΓCN, along with error bars for measurements.The
following values are from fitting the peak to a gaussian.
χwater (∆ωCN)(cm−1
) (ΓCN)(cm−1
)
0 (pure Ionic Liquid) -1.40 ± 0.01 4.72 ± 0.04
0.1 -1.30±0.01 5.12±0.02
0.2 -1.63±0.01 5.14±0.02
0.3 -1.17±0.01 5.15±0.02
0.4 -1.05±0.01 4.99±0.02
0.5 -0.80±0.01 5.27±0.02
0.6 -0.52±0.01 5.28±0.02
0.7 -0.25±0.02 5.59±0.03
0.8 -0.21±0.02 5.69±0.03
0.9 -0.07±0.02 5.61±0.03
81
Figure 5.2: Potential dependent SEIRAS spectra for dDMSO alone. All the potentials are referenced with respect to Ag/AgCl. The solid black line represents 0 V with respect to the reference
and the dotted grey line represents the baseline of each displaced spectrum.
82
Figure 5.3: Potential dependent SEIRAS spectra for 1 mM CTAB in dDMSO. All the potentials
are referenced with respect to Ag/AgCl. The solid black line represents 0 V wrt to the reference
and the dotted grey line represents the baseline of each displaced spectrum.
83
Figure 5.4: Potential dependent SEIRAS spectra of the CTAB-dDMSO-water system. The concentrations of CTAB and water were 1 mM and 1 M respectively. All the potentials are referenced
with respect to Ag/AgCl. The solid black line represents 0 V wrt to the reference and the dotted
grey line represents the baseline of each displaced spectrum.
84
Figure 5.5: Potential dependent SEIRAS spectra of the CTAB-dDMSO-water system. The concentrations of CTAB and water were 1 mM and 20 M respectively. All the potentials are referenced
with respect to Ag/AgCl. The solid black line represents 0 V wrt to the reference and the dotted
grey line represents the baseline of each displaced spectrum.
85
Figure 5.6: Potential dependent SEIRAS spectra for 1 mM CTAB in H2O. All the potentials are
referenced with respect to Ag/AgCl. The solid black line represents 0 V wrt to the reference and
the dotted grey line represents the baseline of each displaced spectrum.
86
Figure 5.7: Potential dependent SEIRAS spectra for 1 mM CTAB in D2O. All the potentials are
referenced with respect to Ag/AgCl. The solid black line represents 0 V wrt to the reference and
the dotted grey line represents the baseline of each displaced spectrum. The peaks at 2850 cm−1
and 2915 cm−1
are from CHs in CTAB. The identity of the peak at 2750 cm−1
is still uncertain; it
is where the free OD should manifest.[190, 191]
87
Figure 5.8: : Potential dependent SEIRAS spectra of the 1 mM SDS in dDMSO. All the potentials
are referenced with respect to Ag/AgCl. The solid black line represents 0 V wrt to the reference
and the dotted grey line represents the baseline of each displaced spectrum.
88
Figure 5.9: Potential dependent SEIRAS spectra of the 100 mM SDS in dDMSO. All the potentials
are referenced with respect to Ag/AgCl. The solid black line represents 0 V wrt to the reference
and the dotted grey line represents the baseline of each displaced spectrum.
89
Figure 5.10: Potential dependent SEIRAS spectra for 10 mM SDS in water. All the potentials are
referenced with respect to Ag/AgCl. The solid black line represents 0 V wrt to the reference and
the dotted grey line represents the baseline of each displaced spectrum.
90
Figure 5.11: SEIRAS spectra of 100 mM SDS in water as a function of applied potential. An
increase in the CH features of SDS corresponds to a decrease in water. All spectra have been
displaced vertically. The grey dashed line represents the baseline for each spectrum. The solid
black line indicates 0 V verses Ag/AgCl. Note that the concentration SDS used here is 12 times
the critical micelle concentration. Even at such concentrations, the signal is much smaller than that
of CTAB at its CMC ( 1 mM).
91
Figure 5.12: The differential integrated peak data for the 100 mM SDS - water system.
92
a b
c d
e
Figure 5.13: The current verses potential plot for 1 mM CTAB-dDMSO-water systems. a. 1 mM
CTAB -dDMSO 0 M water. b. 1 mM CTAB -dDMSO 10 M water. c. 1 mM CTAB -dDMSO 20
M water. d. 1 mM CTAB -water. The potential was scanned from +0.6 V to -1 V with respect to
Ag/AgCl in increments of 0.2 V. Each potential was held for roughly 120 s.
93
A Conversion between A and coverage
Here we describe the process of converting our SEIRAS data from absorbance units to coverage
units (molecules/ nm2). This was achieved using a Beer’s law type of analysis. Beer’s law states:
A = εcl
Where A is the absorbance of the species, ε is the molar absorptivity of the species, c is the
concentration of the species and l is the path length of the sample. ε is typically calculated by
measuring the absorbance while varying concentration for a constant l, as seen below in figure
S11. Once ε is known, when we measure A, the ratio A/ε is cl, which can be expressed in terms
of molecules/nm2
.
cl = (mol
L
)(cm) => (
mol
103cm3
)(cm) = mol
103cm2
=
10−3mol ∗NA(
molecules
mol )
1014nm2
=> cl = 6.02×106molecules
nm2
(A.1)
For surface species, the bulk molar absorptivity (εb) can be enhanced by factors ranging from
10-1000.[137, 139, 192] So, merely converting from Asur f ace to cl using εb is insufficient to provide realistic estimates of the surface coverage.
To surpass this limitation, we devised an experiment to measure the molar absorptivity at the
surface (εs) . We used a redox active species that formed monolayers via thiol linkages to our
gold electrode (FCHT - ferrocenylhexanethiol). From the monolayer of FCHT, we had access to
the surface coverage from integrating peak areas in cyclic voltammagrams as well as measuring
the absorbance through SEIRAS. Having both A and cl in (molecules
nm2
) allowed us to determine εs
through their ratio. Furthermore, the ratio of εs
to εb would provide information on the degree of
94
enhancement of molar absorptivity at the surface.
Figure 5.14: Data obtained from bulk FTIR measurements. A: The CH stretching region of FCHT
for different bulk concentrations of FCHT in dDMSO. The two characteristic CH stretching peaks
are at 2853 cm−1
and 2929 cm−1 This data was obtained with a calcium fluoride FTIR cell, using a
path length of 100 µm. The spectra are baseline subtracted. B: integrated area of the peaks plotted
against bulk FCHT concentration. The region of integration was 2700 – 3000 cm−1
. The obtained
slope was dA/dc= εbl = 0.07644 mM−1 = 76.44 M−1 => εb = 7644 M−1
cm−1
.
From bulk FTIR absorption experiments of FCUT in dDMSO, we were able to determine the
bulk molar absorptivity (εb) to be 7644 M−1
cm−1
(for 2700 – 3000 cm−1
). We then measured
the absorption of a FCHT monolayer on gold using SEIRAS. The SEIRAS spectrum of the FCHT
monolayer and the corresponding CV is shown in figure S12.
Through the cyclic voltammogram and SEIRAS spectrum, we obtained the surface coverage
and absorption respectively. With these, we were able to calculate the molar absorptivity at the
surface (εs). Below is a table that summarizes the results.
From our experiments, we obtain an enhancement factor which is the ratio of εs
to εb. This
enhancement factor pertains to the CH stretches of the aliphatic tail in FCHT. CTAB and SDS also
have these CH stretches present. It follows then, if we were to determine εb for the surfactants,
we could assume the same degree of enhancement to obtain εs
. This provides us with means of
95
Figure 5.15: The experimental results from the FCHT monolayer on the gold ZnSe SEIRAS substrate. A: representative cyclic voltammogram of the FCHT monolayer after subtraction of the
capacitive response at scan rate of 200 mV/s. B: The corresponding CH stretching region of the
SEIRAS spectrum. Data adapted from reference 6.
96
Table 5.7: A summary of the experimentally calculated enhancement factors based on comparing
coverage from CVs and absorption signal from SEIRAS. The integrated absorbance was obtained
by integration from 2800 – 3000 cm−1
.
SNo. Coverage (molecules/nm2
) Integrated Absorbance (O.D.) Enhancement factor (f)
1 0.13 0.16 914
2 0.60 0.49 645
3 1.53 1.74 895
Average 818
transforming our SEIRAS data from absorbance units to coverage. The calculations for this transformation are provided later.
Figure 5.16: Data obtained from bulk FTIR measurements. A: The CH stretching region of CTAB
for different bulk concentrations in dDMSO. This data was obtained with a calcium fluoride FTIR
cell, using a path length of 100 µm. The spectra are baseline subtracted. B: integrated area of the
peaks plotted against bulk CTAB concentration. The region of integration was 2700 – 3000 cm−1
.
The obtained slope was dA/dc= εbl = 0.1102 mM−1 = 110.2 M−1 => εb = 11020 M−1
cm−1
.
97
B Calculations for converting SERIAS absorption signal to surface
coverage
ASEIRAS = εscl =>
ASEIRAS
εs
= cl (B.1)
Where ASEIRAS is the integrated absorbance from 2800 – 3000 cm−1
, εs
is the molar absorptivity
of the CH integrated from 2800 -3000 cm−1
at the surface, c is the concentration of surfactant and
l is the path length, in our case we will express c l in units of molecules/nm2
. We can express εs =
f εb, we know f from table Table 5.7 and εb from figure Figure 5.16. Using this information,from
equation B.1 we obtain:
=> cl =
ASEIRAS
f εb
=
ASEIRAS
818×11020M−1cm−1
= 1.109×10−7 ×ASEIRASMcm
(B.2)
From equation A.1 we know how cl can be expressed in terms of molecules/ nm2
. Applying
this to equation B.2 :
cl(in
molecules
nm2
) = 1.109×10−7 ×ASEIRAS6.02×106molecules
nm2
cl(
molecules
nm2
) => 0.67ASEIRAS
molecules
nm2
98
So, from an integration of the SEIRAS absorption for the CH region for CTAB, we can estimate
the coverage of CTAB.
99
C Scheme for interpolation
In our experiment for the CTAB-dDMSO-water system, potential was varied for four different bulk
water mole fractions. The potential verses Ag/AgCl was typically varied from -1 V to + 0.6 V in
increments of 0.2 V and the bulk water mole fraction values were 0, 0.2 ,0.6 and 1. Using our
experimentally obtained absorbance data at these combinations of potentials and mole fractions,
we linearly interpolated values of absorbances for combinations of mole fractions that were not
explored.
The interpolation increments selected for the potential and mole fraction were chosen to be
0.05 V and 0.05 χ respectively. Multiple combinations of V and χ correspond to identical CTAB
coverages. The same CTAB coverage can be obtained by a variation of potential while the mole
fraction is held constant or a variation of the mole fraction at a constant potential. Thus, by comparing points of CTAB coverages obtained by a variation of potential to the same CTAB coverage
obtained by a variation of mole fraction, a parallel between voltage and mole fraction is drawn.
This is practically achieved by interpolating the potential data to find the exact value of potential
that corresponds to the same CTAB coverage as the coverage obtained from a particular water mole
fraction value. The MATLAB code demonstrating this process is shown below.
100
Z = [intA_dDMSOc ; intA1_10Mc; intA1_20Mc ; intA1_water2c ];%putting all the absorbance
data together
chiq = chi(1):0.05:chi(4);%defining desired mole fraction resolution (increments of
0.05)
potq = t(1):-0.05:t(9); %defining desited potential resolution (increments of 0.05)
%interpolating a function of two variable requires them to be meshed
%together
[T, CHI] = meshgrid(t,chi);%meshing the experimentally obtained potential (t) and mole
fraction (chi)
[POTQ, CHIQ] = meshgrid(potq,chiq);%meshing the desired resolution of potential (potq)
and mole fraction (chiq)
Zq = interp2(T, CHI, Z, POTQ, CHIQ, 'linear');%interpolating the experimental data
mesh to the number of points of
%the desired mesh
slicechi = Zq(:,1); %taking a slice in chi along potential = 0.6 V
sliceV = Zq(1,:); % taking a slice in V along chi = 0 V
potinterp = interp1(sliceV, potq, slicechi); %interpolating to find the values of
potential that correspond to the
%same CTAB coverages as those obtained from variation of water mole
%fraction while the potential was held at 0.6 V
Figure 5.17: Matlab script used for interpolation of experimental data. The initial lines describe
interpolation of experimental data to fill in points. The later stages describe interpolation of potential data to find potentials that result in the same CTAB coverages as those corresponding to a
variation of the mole fraction.
101
C.1 Computational transient temperature profile
Figure 5.18: Simulated temperature profile inside silicon for 0.5 µJ pump energy calculated at
different times.
102
Figure 5.19: Simulated temperature profile inside silicon for 1 µJ pump energy calculated at
different times.
Figure 5.20: Simulated temperature profile inside silicon for 2.5 µJ pump energy calculated at
different times.
103
Figure 5.21: Simulated temperature profile inside silicon for 5 µJ pump energy calculated at
different times.
Table 5.8: Fitting results for the MCT detector decay component. The equation used for fitting
was a ∗ exp(−b ∗t) +c.
Name a σa b (1/s) σb c σc τ(s) στ R
2
MCT decay -1.23 ×106 5.80 ×105 6.27×105 2.03 ×104 0.038 0.0018 1.62 ×10−6 5.17 ×10−8 0.995
104
Figure 5.22: A comparison of the transmission fringes for different pumps. (a) : The transmission
fringes plotted against IR wavelength at 5 µs. (b): A comparison between time traces taken at
2279.5 cm−1
. While there could be discernible differences between the two time traces, we believe
that these subtle variations stem from differences in beam divergence, focusing conditions and
power between the 800 nm and 400 nm pumps.
Table 5.9: Fits for the experimental time traces, corresponding to an 400 nm pump. The equation
used for fitting was a ∗ exp(−b ∗t) +c.
Wavenumber (cm−1
) a σa b (1/s) σb c σc τ(s) στ R
2
2223 1.07 -0.100 1.68×105
-5.09 ×103
-0.00617 -2.28 ×10−4 5.95 ×10−6 9.02 ×10−8 0.988
2224 5.62 -0.230 1.60×105
-2.23 ×103
-0.0205 -6.72 ×10−4 6.24 ×10−6 4.34 ×10−8 0.997
2231 -0.648 -0.127 1.61×105
-1.06 ×104
-0.00528 -3.52 ×10−4 6.22 ×10−6 2.05 ×10−7 0.946
Table 5.10: Fits for the experimental time traces, corresponding to an 800 nm pump. The equation
used for fitting was a ∗ exp(−b ∗t) +c.
Wavenumber (cm−1
) a σa b (1/s) σb c σc τ(s) στ R
2
2223 106 9.08 1.73×105
-0.0522 0.00110 5.80 ×10−6 2.85 ×103 4.78 ×10−8 0.996
2224 -82.1 12.8 1.87×105
-0.0285 8.42 ×10−4 5.36 ×10−6 5.16 ×103 7.42 ×10−8 0.989
105
Figure 5.23: The refractive index as a function of space along the body of silicon (z direction).
This was calculated by considering the pump incident on the face of the silicon wafer (at x=0) and
subsequently calculating the temperature change induced by pump absorption.
Figure 5.24: A representative set of data. This data was obtained from a 800 nm pump of energy
17 µJ.
106
C.2 MATLAB code for simulations
% diffusion modeling
clear all
close all
% units
J = 1;K = 1;W = 1;s = 1;
cm = 1e-2;micron = 1e-6; microJ = 1e-6;micros = 1e-6;
g = 1e-3;
% properties of silicon
D = 0.8*(cm^2)/s; % diffusion constant for silicon
C = 0.7*J/(g*K); % specific heat
rho = 2.33*g/(cm^3); % density
eta = 1.9e-4/K; % change in refractive index as a function of temperature
n0=3.4; %refractive index of silicon
alpha=1/(10*micron); % extension of silicon at 800nm (10 um) in units of length
d = 450*micron; % wafer thickness
% properties of pump light
E0 = 10*microJ; %energy of the pump
w = 50*micron; %radius of the pump spot at focus
% properties of probe light
lambda = 4.3478*micron:0.0001*micron:4.5455*micron; % IR wavelengths used
invcm = 10000*micron./lambda; % above range in cm-1
% =========================================
% The initial temperature difference at z=0, r=0.
DelT0 = (alpha/(2*pi*rho*C))*(E0/(w^2))
% create a 3D temperature profile
dz = 1*micron; dx = dz; dy = dz;
z = (-450*micron: dz :450*micron);
x = (-250*micron: dx :250*micron);
y = (-250*micron: dy :250*micron);
[Z, X , Y] = meshgrid(z, x, y);
DelT = DelT0*exp(-abs(Z)/(10*micron)).* exp(-(X.^2+Y.^2)/(50*micron).^2) ;
contourf(z,x,DelT(:,:,200), 50, 'edgecolor', 'none')
% FT the temperature profile
DelTk = fftshift(fftn(DelT));
kx = 2*pi*fftaxis(x);
ky = 2*pi*fftaxis(y);
kz = 2*pi*fftaxis(z);
[Kz , Kx, Ky] = meshgrid(kz, kx , ky);
Ksqr = Kx.^2 + Ky.^2 + Kz.^2;
% figure(2)
%contourf(z,x,abs(DelTk), 50, 'edgecolor', 'none')
figure(3)
% applying the diffusion frequency filter at a future time
t=(0:1:80)*micros;
Phi = zeros(length(lambda),length(t));
T = zeros(size(Phi));
Figure 5.25: Our MATLAB code to simulate the heat temperature profile by solving the heat
diffusion equation at different times
107
Figure 5.26: A few time traces from the simulations. This is the same as figure 8b from the main
text.
Table 5.11: Fits for the time traces corresponding to the pumped spectra, from simulation. The
equation used for fitting was a ∗ exp(−b ∗t) +c+d ∗ exp(−f ∗t).
Wavenumber (cm−1
) a b (1/s) c τ1(s) στ1 d f (1/s) τ2(s) στ2 R
2
2221.88 0.00224 2.28×105
-0.00384 4.39 ×10−6 4.10 ×10−7 0.0235 4.35×104 2.30 ×10−5 1.98 ×10−6 0.999
2223.11 -0.0156 1.29×105
-0.00913 7.78 ×10−6 1.68 ×10−6 0.00355 9.62×103 1.04 ×10−4 0.34 ×10−4 0.994
108
Abstract (if available)
Abstract
Most electrochemical reactions require delivery of protons, often from water, to surface adsorbed species. However, water also acts as a competitor to many such processes by directly reacting with the electrode, which necessitates using water in small. Controlling the water content and structure near the surface is an important frontier in directing reactivity and selectivity of electrochemical reactions. In this thesis, interfacial water is probed using surface spectroscopy in binary mixtures of water and another solvent (Ionic liquid and DMSO).
For the water - Ionic liquid studies, we used a monolayer of mercaptobenzonitrle (MBN) as a probe of the interfacial environment. We find that even up to a significant mole fraction of bulk water (x ∼ 0.95) the nitrile frequency does not change from that indicative of pure ionic liquid for [EMIM][BF4], indicating preferential aggregation of the ions near the surface. Since this behavior is very similar to surfactants, we chose an imadazolium cation with a longer side chain which resulted into behavior expected from a surfactant, with preferential layer of the ions on the surface even in dilute water solutions (x ∼ 0.995). This observation indicates that even those ILs that are not nominally categorized as surfactants, have a strong tendency to aggregate at the surface. Since ILs serve as electrolytes in a range of electrochemical reactions, including those requiring water, our results are likely useful for mechanistic understanding and tuning of such reactions.
In the water - DMSO investigation, we did not use MBN molecules, instead we used surfactants which had a unique vibrational signature that could be monitored. Surfactants accumulate near surfaces, and therefore can be used as agents to control interfacial water. Using mid-IR spectroelectrochemistry, we show that a modest concentration (1 mM) of the cationic surfactant CTAB in mixtures of 10 M water in an organic solvent (dDMSO) has a large effect on the interfacial xivwater concentration, changing it by up to 35% in the presence of an applied potential. The major cause of water content change is displacement due to accumulation or depletion of surfactants driven by potential. Two forces drive the surfactants to the electrode – the applied potential and the hydrophobic interactions with the water in the bulk. We have quantified their competition by varying the water content in the bulk. To our knowledge, for the first time, we have identified the electrochemical equivalent of the hydrophobic drive. For our system a change in applied potential of 1 V has the same effect as adding 0.55 mole fraction of water to the bulk. This work illustrates the significance of surfactants in partitioning of water between the bulk and the surface, and paves the way towards engineering interfacial water structure for controlling electrochemical reactions.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Elucidating local electric fields and ion adsorption using benzonitrile: vibrational spectroscopic studies at interfaces and in bulk
PDF
Monitoring local electric fields, charge densities at electrode surfaces using in-situ graphene/surface enhanced Raman spectroscopy (GERS/SERS) based Stark-shifts
PDF
Interfacial polarization and photobasicity: spectroscopic studies of the electrode environment and photochemical effects
PDF
Molecular orientation and structure of hydrogen bonds at water interfaces
PDF
Probing charge transfer and electric fields at the bulk and interface using vibrational spectroscopy
PDF
Vibrational spectroscopy and molecular orientation at interfaces: correlating the structure of a material's surface to its properties
PDF
Potential-induced formation and dissociation of adducts at interfaces and adsorption of phenols onto metal electrodes
PDF
Molecular orientation from sum frequency generation spectroscopy: case study of water and methyl vibrations
PDF
Asymmetric response and the reorientation of interfacial water with respect to applied electric fields
PDF
Understanding organic and inorganic semiconductor surfaces by coherent nonlinear spectroscopy
PDF
Ultrafast dynamics of excited state intra- and intermolecular proton transfer in nitrogen containing photobases
PDF
Hot electron driven photocatalysis and plasmonic sensing using metallic gratings
PDF
Vibrational sum frequency spectroscopy of molecules on metal, semiconductor, and aqueous surfaces
PDF
Non-linear surface spectroscopy of photoswitches annd photovoltaics
PDF
Dissociation energy and dynamics of HCl and aromatic-water clusters
PDF
Plasma emission across metal nanoparticle surfaces and semiconductor -liquid interface characterizations
PDF
Advancing lithium batteries and related electrochemical technologies for a sustainable future
PDF
Understanding the structure-property relationship in electrode materials for electrochemical energy storage
PDF
Targeted cell imaging by in-situ assembly and activation of hot spot SERS nanoprobes using split-fluorescent protein scaffolds
PDF
Concentration and size partitioning of trace metals in surface waters of the global ocean and storm runoff
Asset Metadata
Creator
Pennathur, Anuj Krishnasundar
(author)
Core Title
Water partitioning between the bulk and an electrode surface
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Chemistry (Chemical Physics)
Degree Conferral Date
2023-12
Publication Date
12/14/2023
Defense Date
10/18/2023
Publisher
Los Angeles, California
(original),
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
electrochemistry,OAI-PMH Harvest,photothermal spectroscopy,Raman spectroscopy,Surface Enhanced IR Absorption Spectroscopy (SEIRAS),Surface Enhanced Raman spectroscopy (SERS),surfactants,vibrational spectroscopy,Water
Format
theses
(aat)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Dawlaty, Jahan Mansoor (
committee chair
), Benderskii, Alexander (
committee member
), Cronin, Stephen (
committee member
)
Creator Email
anujpennathur@gmail.com,pennathu@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-oUC113797225
Unique identifier
UC113797225
Identifier
etd-PennathurA-12561.pdf (filename)
Legacy Identifier
etd-PennathurA-12561
Document Type
Thesis
Format
theses (aat)
Rights
Pennathur, Anuj Krishnasundar
Internet Media Type
application/pdf
Type
texts
Source
20231218-usctheses-batch-1116
(batch),
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright.
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Repository Email
cisadmin@lib.usc.edu
Tags
electrochemistry
photothermal spectroscopy
Raman spectroscopy
Surface Enhanced IR Absorption Spectroscopy (SEIRAS)
Surface Enhanced Raman spectroscopy (SERS)
surfactants
vibrational spectroscopy