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
/
Monitoring local electric fields, charge densities at electrode surfaces using in-situ graphene/surface enhanced Raman spectroscopy (GERS/SERS) based Stark-shifts
(USC Thesis Other)
Monitoring local electric fields, charge densities at electrode surfaces using in-situ graphene/surface enhanced Raman spectroscopy (GERS/SERS) based Stark-shifts
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
MONITORING LOCAL ELECTRIC FIELDS, CHARGE DENSITIES
AT ELECTRODE SURFACES USING IN-SITU
GRAPHENE/SURFACE ENHANCED RAMAN SPECTROSCOPY
(GERS/SERS) BASED STARK-SHIFTS
By
Haotian Shi
A Dissertation Presented to
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of
the Requirement for the Degree of
DOCTOR OF PHILOSOPHY
(Chemistry)
May. 2020
Copyright 2020 Haotian Shi
ii
Acknowledgements
I would like to start this part by thanking my advisor, Dr. Stephen B. Cronin, for his
guidance, support and inspiration throughout my five years’ Ph.D. life. He set a perfect example
for us as a teacher, a researcher and a mentor. I appreciate his passion and dedication to science
and work from the bottom of my heart.
My sincere thanks also go to my dissertation committee Dr. Alexander V. Benderskii, Dr.
Wei Wu and qualifying exam committee members Dr. Sri R. Narayan, Dr. Jahan M. Dawlaty
for their invaluable suggestions.
Before I joined our group, I had worked in Dr. Jahan M. Dawlaty’s group as a master student
for one and a half years. I had such a good time there and learned a lot from Dr. Dawlaty, who
is the best teacher I have ever met. This experience made my transition from a student to a
researcher much easier.
I’m quite lucky to work with the senior graduate students from Cronin research group.
Special thanks to Dr. Guangtong Zeng. He recruited me into this group and had been my mentor
and friend in my early Ph.D. life. Also, thanks to Dr. Jing Qiu, Dr. Zhen Li, Dr. Rohan Dhall,
Dr. Ioannis Chatzakis and Dr. Shun-Wen Chang. I won’t make it without their inspiring advises
and instructions.
Dr. Bingya Hou and Dr. Jihan Chen from Cronin research group started their doctoral
program almost at the same time as me and became close friends of mine. Thanks to their
supports and sharing of experiences both in research and life. I would like to thank all the
graduated and current colleagues and friends of mine from Cronin research group: Dr. Nirakar
Poudel, Dr. Lang Shen, Ms. Sisi Yang, Mr. Bo Wang, Mr. Yu Wang, Mr. Bofan Zhao, Mr. Zhi
iii
Cai, Ms. Indu A A and Mr. Boxin Zhang. I would also like to thank Dr. Yuqiang Ma, Dr. Anyi
Zhang, Dr. Zeyu Chen, Dr. Yi Wang, Dr. Boxiang Song, Dr. Angelo Montenegro, Mr. Zhen Li,
Mr. Mingrui Chen, Ms. Jie Ma, Ms. Nan Chen and all my colleagues and friends from other
groups in both EE department and Chemistry department for their help and support.
Ph.D. life is not easy. I feel so blessed for having lots of friends here in LA that we could
share happiness and sorrow, and they are Ms. Yiwei Liu, Ms. Yiyun Liu, Ms. Yue Sheng, Mr.
Bohao Li, Ms. Shu Huang, Mr. Aibo Li, Mr. Xinyu Yan, Mr. Yachen Yu, Ms. Xinyi Wang, Ms.
Lingyun Li, Dr. Yan Xiong, Ms. Yangtong Yang, Mr. Dawei Gu, Ms. Linxi Yan, Mr. Zhongqi
Cheng, Ms. Xiangyu Zhang, Ms. Xiangyi Cai, Ms. Danting Zhu, Ms. Shuang Wu, Ms. Xinran
Li, Mr. Guangzhou Zeng, Mr. Gang Zhang, Ms. Xuanrui Li, Ms. Yuchuan Zhang, Mr. Haoxi Li
and Ms. Linzhen Liu.
I take pride in dedicating this dissertation to my beloved parents.
iv
Table of Contents
Acknowledgments ................................................................................................................. ii
List of Figures ....................................................................................................................... vi
Abstract .................................................................................................................................. x
Chapter 1: Introduction .......................................................................................................... 1
1.1 Surface Enhanced Raman Spectroscopy (SERS) ............................................................. 1
1.1.1 Raman Spectroscopy ................................................................................................. 1
1.1.2 SERS Basics ............................................................................................................. 2
1.1.3 Electromagnetic Mechanism (EM) of SERS ............................................................ 4
1.1.4 Chemical Mechanism (CM) of SERS ...................................................................... 5
1.1.5 Gold-island Material System .................................................................................... 7
1.1.6 Electrostatic/Electrochemical SERS (E-SERS) ....................................................... 9
1.2 Graphene Enhanced Raman Spectroscopy (GERS) ....................................................... 10
1.2.1 Graphene & Its Raman spectroscopy ..................................................................... 10
1.2.2 Raman Studies on Gated Graphenes: Electron-Phonon Interaction ....................... 13
1.2.3 Application of Graphene in SERS .......................................................................... 17
1.2.4 Graphene-based Electrodes .................................................................................... 19
1.2.5 Chemical Vapor Deposition (CVD) of Graphene .................................................. 21
1.3 Vibrational Stark-shift Spectroscopy ............................................................................. 23
1.3.1 Stark-shift Effect .................................................................................................... 23
1.3.2 Related Molecular Systems .................................................................................... 25
1.3.3 Vibrational Sum Frequency Generation (VSFG) ................................................... 26
1.4 Photo-electrochemical (PEC) Measurements ................................................................ 28
1.4.1 Important Reactions in Artificial Photosynthesis ................................................... 28
1.4.2 Three-terminal Photo-electrochemical Setup ......................................................... 31
1.4.3 In-situ Raman Spectroscopy Combined with PEC ................................................. 32
Chapter 2: Sensing Local pH and Ion Concentration at Graphene
Electrode Surfaces using in-situ Raman Spectroscopy ........................................................ 34
2.1 Abstract .......................................................................................................................... 34
2.2 Introduction .................................................................................................................... 35
2.3 Experimental Details ...................................................................................................... 36
2.4 Results and Discussion ................................................................................................... 39
2.5 Conclusions .................................................................................................................... 51
2.6 Acknowledgements ........................................................................................................ 52
Chapter 3: Monitoring Local Electric Fields at Electrode Surfaces
v
Using SERS-based Stark-shift Spectroscopy During Hydrogen
Evolution Reactions ............................................................................................................. 53
3.1 Abstract .......................................................................................................................... 53
3.2 Introduction .................................................................................................................... 54
3.3 Experimental Details ...................................................................................................... 58
3.4 Results and Discussion ................................................................................................... 65
3.5 Conclusions .................................................................................................................... 70
3.6 Acknowledgements ........................................................................................................ 71
Chapter 4: Measuring Local Electric Fields and Local Charge
Densities at Electrode Surfaces Using Graphene-Enhanced
Raman Spectroscopy (GERS)-based Stark-shifts ................................................................ 72
4.1 Abstract .......................................................................................................................... 72
4.2 Introduction .................................................................................................................... 73
4.3 Experimental Details ...................................................................................................... 76
4.4 Computational Methods ................................................................................................. 79
4.5 Results and Discussion ................................................................................................... 79
4.6 Conclusions .................................................................................................................... 96
4.7 Acknowledgements ........................................................................................................ 97
Chapter 5: Future Work and Conclusions ............................................................................ 98
5.1 Monitoring Local Electric Field using Stark shifts on
Naphthyl-Nitrile Functionalized Silicon .............................................................................. 98
5.2 Monitoring Local pH using Benzoic Acid group on SERS Sample ............................ 103
5.3 Discussion of the Approaches ...................................................................................... 106
5.4 Conclusions .................................................................................................................. 108
Bibliography ....................................................................................................................... 109
vi
List of Figures
Figure 1.1. Energy level diagram for Raman scattering; (a) Stokes Raman
scattering (b) anti-Stokes Raman scattering ........................................................................... 2
Figure 1.2. Schematic of Surface Enhanced Raman Spectroscopy (SERS) ......................... 3
Figure 1.3. Schematic of (a) Electromagnetic Mechanism (EM) and
(b) Chemical Mechanism (CM) of Surface Enhanced Raman Spectroscopy (SERS) ........... 6
Figure 1.4. (a) SEM and (b) TEM images of 5nm thick Au island films, together
with the corresponding (c) electric field distribution calculated using the
FDTD method ........................................................................................................................ 8
Figure 1.5. The honeycomb lattice of graphene .................................................................. 11
Figure 1.6. (a) 2D electronic band structure of graphene (b) Measured monolayer
graphene Raman spectrum with 2.33 eV laser-excitation energy ........................................ 12
Figure 1.7. Linewidth and Frequency of the Raman G band as a function of
electron concentration which is proportional to the square of the Fermi level energy ........ 14
Figure 1.8. Fermi energy EF as a function of the relative frequency of the G mode ........... 16
Figure 1.9. Comparisons of Raman signals of R6G and PPP deposited on graphene
(red line) and on the SiO2/Si substrate (blue line) using vacuum evaporation at
514.5 nm excitation and 632.8 nm excitation ...................................................................... 18
Figure 1.10. Monolayer graphene with gold electrode on glass ......................................... 20
Figure 1.11. CVD-growth and wet transfer of monolayer graphene................................... 22
Figure 1.12. Computed regular (non-chaotic) Rydberg atom energy level spectra
of hydrogen in an electric field near n=15 for magnetic quantum number m=0 ................. 24
Figure 1.13. (a) Schematic diagram of the in-situ SERS measurements using a
water immersion lens. (b) SEM image of the 5nm Au film. (c) Cross-section
diagram of the sample structure ........................................................................................... 25
Figure 1.14. (a) Schematic diagram illustrating the SERS measurement of
thiolated-benzonitrile bound to an electrode for Stark-shift spectroscopy.
(b) Stark-shift of the C-N stretch vibrational mode obtained by SFG
spectroscopy plotted as a function of applied potential ....................................................... 27
Figure 1.15. (a) Schematic diagram of a photoelectrochemical cell for
photocatalytic water splitting (b) Photocatalytic current-potential curves
measured for GaP photocatalysts with various thicknesses of TiO2 under
1W/cm
2
532nm illumination in a 0.5M Na2SO4 pH=7 solution .......................................... 30
Figure 1.16. Schematic diagrams of the three-terminal photo-electrochemical
cell using a water immersion lens ........................................................................................ 32
Figure 1.17. Schematic diagrams of the in-situ Raman measurement
combined with the PEC set up and 3D plot of the graphene sample using
as the working electrode in a PEC cell ............................................................................... 33
Figure 2.1. (a) Schematic diagrams of the three-terminal photoelectrochemical
cell using a water immersion lens and (b) monolayer graphene electrode .......................... 38
Figure 2.2. (a) G-band Raman shift and corresponding charge concentration
and (b) G band Raman linewidth plotted as a function of the applied voltage
and Fermi energy measured in DI water. (b) Raw spectra showing the voltage-
vii
induced redshift of the G-band Raman mode ...................................................................... 40
Figure 2.3. (a) G-band Raman shift and corresponding carrier concentration
with error bar (b) Results for 5 different samples ................................................................ 41
Figure 2.4. (a) G band Raman shift and corresponding charge concentration
and (b) G band Raman linewidth plotted as a function of the applied voltage
and Fermi energy measured in 1M KCl solution. (c) Raw Raman spectra
showing the voltage-induced redshift of the G band Raman mode ..................................... 43
Figure 2.5. (a) Capacitance-voltage plot of the graphene-based electrode
measured in DI water and 1M KCl. Charge densities obtained from the
capacitance-voltage data (i.e., Q=CV) and Raman spectroscopy in
(b) DI water and (c) 1M KCl ................................................................................................ 45
Figure 2.6. (a) 1/C
2
versus applied voltage, (b) Double layer capacitance
versus applied voltage, (c) Charge accumulated in the double layer versus
applied voltage and (d) Carrier concentration in the double layer versus
applied voltage from electrochemical impedance measurement ......................................... 46
Figure 2.7. Local ion concentration plotted as a function of voltage for
(a) pure DI water and (c) 1M KCl. (b) Local pH plotted as a function of
voltage for pure DI water ..................................................................................................... 48
Figure 2.8. (a) Local pH plotted as a function of voltage for 5 solutions
with different pH. (b) Results of measured pH, Debye length of the solution,
and the known pH of the bulk solution ................................................................................ 49
Figure 2.9. Cyclic voltammetry of 50mM ferro/ferricynide redox couple in
(a) DI water and (b) 1M KCl .............................................................................................. 49
Figure 2.10. Cyclic voltammetry of graphene sample as working electrode
in DI water. ........................................................................................................................... 50
Figure 3.1. (a) Schematic diagrams of the three-terminal photoelectrochemical
cell using a water immersion lens. (b) Cross-section diagram of the sample
structure. (c) 3D plotting of the sample and the laser (d) SEM image of the
5nm Au film ....................................................................................................................... 59
Figure 3.2. (a) Schematic diagram of the in situ SERS measurements using
a water immersion lens. (b) SERS spectra taken of thiolated benzonitrile
on Au nano-islands deposited on a glass substrate both with and without an
underlying monolayer of graphene. (c) Cross-sectional diagram of the Au
nano-islands/graphene on glass sample structure ................................................................ 62
Figure 3.3. SERS spectra of thiolated benzonitrile on different regions
of the sample: Au nano-islands (AuNI) deposited on a glass substrate
both with and without an underlying monolayer of graphene (Gr) with
gold electrode (Au) .............................................................................................................. 63
Figure 3.4. (a) Raman shift of the nitrile stretch and electric field strength
plotted as a function of the applied electrochemical potential. (b) Schematic
diagram illustrating the SERS measurement of thiolated-benzonitrile bound
to an Au electrode for Stark-shift spectroscopy. (c) Current-voltage plotted
as a function of the electrochemical potential, indicating the onset potential.
(d) Electrochemical impedance spectroscopy obtained capacitance-voltage
viii
plot indicating the point of zero charge (PZC) ..................................................................... 64
Figure 3.5. Waterfall plot of the raw Raman spectra under various applied
potentials as indicated in the legend ..................................................................................... 66
Figure 3.6. (a) Frequency change as a function of the applied potential
and KCl ionic concentration. (b) Slice through this plot at a potential of
0.6V vs. Ag/AgCl reference electrode ................................................................................. 67
Figure 3.7. (a) Raman shift of the nitrile stretch and electric field strength
plotted as a function of the applied electrochemical potential in 0.01mol/L
KCl solution. (b) Current-voltage plotted as a function of the electrochemical
potential, indicating the onset potential. (c) V oltage drop across the solution
calculated from (b). (d) Raman shift of the nitrile stretch plotted as a function
of the corrected potential ...................................................................................................... 68
Figure 4.1. Schematic diagrams of the (a) three-terminal photoelectrochemical
cell using a water immersion lens to obtain Raman Spectra in situ and
(b) CuPc-coupled monolayer graphene electrode ................................................................ 78
Figure 4.2. (a) CuPc Stark-shift (black) and the electric field (red) plotted
as a function of the applied potential measured in DI water. (b) Graphene
G band Raman shift (black) and the free carrier concentration (red) plotted
as a function of the applied potential. (c) CuPc 1531cm
-1
peak plotted as a
function of electric field at the monolayer graphene electrode surface ............................... 82
Figure 4.3. Raman shift of different CuPc peaks plotted as a function of
the applied electrochemical voltage, measured in water (a)(b)(c) and in
1mol/L KCl solution (d)(e)(f)............................................................................................... 83
Figure 4.4. Cartesian coordinate axes orientation and bond lengths (in Å) ........................ 85
Figure 4.5. (a) CuPc Stark-shift (black) and the electric field (red) plotted
as a function of the applied potential measured in 1M KCl solution. (b)
Graphene G band Raman shift (black) and the free carrier concentration
(red) plotted as a function of the applied potential measured in 1M KCl
solution. (c) CuPc 1531cm
-1
peak plotted as a function of electric field at
the graphene electrode surface. (d) CuPc Stark-shift change plotted as a
function of the applied voltage, measured in DI water (red) and in KCl
(black) plotted together with their linear fits, which give the Stark tuning
rates (STRs) in units of cm
-1
/(10
6
V/cm) ............................................................................. 87
Figure 4.6. ADF-calculated (a) CuPc Raman spectra, (b) CuPc Raman
Shift 1546cm
-1
vibrational mode versus the applied electric field, (c) CuPc
1546cm
-1
peak plotted as a function of electric field with the Stark tuning
rate at the graphene electrode surface .................................................................................. 90
Figure 4.7. (a) Capacitance-voltage plot of the CuPc graphene-based
electrode measured in DI water, (b) carrier density obtained from the
capacitance-voltage data (i.e., Q=CV , in black) plotted together with
the carrier density obtained by Raman spectroscopy (red) .................................................. 91
Figure 4.8. Atomic diagrams of the five Raman active vibrational
modes calculated for CuPc at (a) 1546cm
-1
, (b) 1520cm
-1
, (c) 1428cm
-1
,
(d) 1418cm
-1
, (e) 1335cm
-1
vibrational modes ..................................................................... 91
ix
Figure 4.9. CuPc peaks different vibrational modes: (a) 1546cm
-1
,
(b) 1520cm
-1
, (c) 1428cm
-1
, (d) 1418cm
-1
, (e) 1335cm
-1
plotted as a
function of electric field with their calculated Stark tuning rates (STRs) ............................ 92
Figure 4.10. Experimental (black) and Simulated (red) CuPc Raman
spectra in DI water ............................................................................................................... 93
Figure 4.11. (a) All three main modes, (b) zoomed-in version of
1531cm
-1
mode (c) water fall plot of Stark-shift Raman spectra (633nm)
of CuPc are shown in DI water ............................................................................................ 94
Figure 4.12. (a) All three main modes, (b) zoomed-in version of G band
mode of graphene Raman spectra (532nm) in DI water ...................................................... 95
Figure 5.1. (a) Molecular structure of Naphthyl-Nitrile grafted on silicon
surface, (b) Raman spectra and (c) 3D structure of surface functionalized
silicon electrode covered by 5nm gold nano-island ........................................................... 100
Figure 5.2. Waterfall plot of the raw Raman spectra of Naphthyl-Nitrile
functionalized Si in (a) DI water and (b) 0.1M KCl under various applied
potentials as indicated in the legend, (c) Raman shift of the nitrile stretch
plotted as a function of the applied electrochemical potential in DI water
and 0.1M KCl solution ....................................................................................................... 102
Figure 5.3. Molecular structure of p-Mercaptobenzoic acid (p-MBA)
bounded to gold nano-island on glass ................................................................................ 103
Figure 5.4. (a) Raman spectra p-MBA on gold nano-island (AuNI) in
air, DI water, 0.01M KOH solution and 0.01M HCl solution, (b) In-situ
Raman spectra of p-MBA on AuNI under different electrochemical
potentials as indicated in the legend. (c) Acidic and (d) basic Raman peak
fitted intensities plotted as a function of electrochemical potentials ................................. 105
x
Abstract
This dissertation presents our efforts in monitoring local electric fields and
charge densities at electrode surfaces using in-situ graphene/surface enhanced Raman
spectroscopy (GERS/SERS) based stark-shifts. These approaches shed light on some
fundamental physics and chemistry in such experimental systems and could
potentially help to study the local pH, local ion concentration, reaction mechanisms,
and catalysis, which are crucial to electrochemical and photoelectrochemical reactions,
like CO2 reduction, hydrogen evolution reaction in water splitting and so on.
Chapter 1 provides introduction and background materials to aid in
understanding the experiments presented in the dissertation. The chapter starts with a
brief overview of the Raman Spectroscopy. Next, we briefly discuss about the
mechanisms, the material systems and setups we will use in surface/graphene-
enhanced Raman spectroscopy (SERS/GERS). Then, we also introduce the concept of
vibrational Stark-shift spectroscopy and our experimental set-up. Then we describe the
basics electrochemistry behind our setup. More detailed background materials for each
topic are covered in the relevant chapter.
Chapter 2 presents a new approach to probe the local ion concentration at
graphene/water interfaces using in situ Raman spectroscopy. This result provide an
accurate measurement of the charge on the graphene electrode under electrochemical
working conditions.
Chapter 3 shows our investigations of using SERS to measure the vibrational
Stark shifts of surface-bound thiolated-benzonitrile molecules bounded to nano-
xi
structured gold electrode surface during hydrogen evolution reactions.
Chapter 4 reports spectroscopic measurements of the local electric fields and
local carrier densities at electrode surfaces using GERS based on the Stark-shifts of
surface-bound molecules and the G band frequency shift in graphene. This general
approach can be applied to a wide range of molecules that can be used to report various
aspects of charge transfer, local pH, and ion concentration.
Finally, in Chapter 5, future work related to in-situ graphene/surface enhanced
Raman spectroscopy (GERS/SERS) approach is discussed. Also, there is a conclusion
of this thesis.
1
Chapter 1: Introduction
1.1 Surface Enhanced Raman Spectroscopy (SERS)
1.1.1 Raman Spectroscopy
Raman spectroscopy, as an important tool for molecule identification and
characterization, has been thoroughly studied and vastly used for decades. It is
extremely useful in studying chemical bonding and intramolecular bonds, since
vibrational frequencies are specific to a molecule’s symmetry and chemical bonds. The
Raman spectrum are plotted from the energies of the transitions, or so called “bands”,
which can be used as the “molecular fingerprint” to identify the observed molecules.
There are vibrational transitions are allowed by both Raman and IR with different
signal intensities. But some of the vibrational transitions are forbidden in IR and can
be seen in Raman. Thus, Raman has been used as important supplementary in the
practical spectroscopic characterization methods.
Photons scattered from a molecule can be classified into elastically scattered
photons and inelastically scattered photons. If the scattered photon has the same energy
or wavelength as the incident photon, we call it an elastically scattered photon.
Otherwise, the photon with higher or lower energy than the incident photon is called
inelastically scattered photon. To note that, the ratio of inelastically scattered photon is
quite small (approximately 1 in 10
7
photons). The process of inelastically scatter is
called Raman effect, named after C.V. Raman in 1928.
1
As is shown in Figure 1.1, if
the photon energy is lost during the inelastically scattering, the process is called Stokes
Raman scattering. Otherwise, it is called anti-Stokes Raman scattering. Although the
2
Stokes and anti-Stokes spectra contain the same frequency information, most
experiments only focus on Stokes-shifted shift due to its dominance at room
temperature.
Figure 1.1. Energy level diagram for Raman scattering; (a) Stokes Raman
scattering (b) anti-Stokes Raman scattering.
1.1.2 SERS Basics
However, the Raman scattering cross-section of most molecules is extremely
small, which generally limits its potential uses. The intensities of certain Raman-active
vibrational signals were found to be increased by a factor of 10
2
to 10
4
when the
exciting laser wavelength is within the spectrum of a molecule. This effect is called
resonance enhancement.
2
The Raman scattering signal from small amount materials
Energy
Incident
Photon
Stokes
Scatter
Anti-Stokes
Scatter
Incident
Photon
Final
Final Initial
Initial
Vibrational
Levels
Virtual
State
(a) (b)
3
adsorbed on structured metal surfaces can be drastically increased. This phenomenon
is called surface-enhanced Raman scattering (SERS), discovered in the late 1970s
3-7
.
SERS has the best performance on silver, gold and copper for common excitation
sources. Up to now, the highest enhancement factor reported is as much as 10
14
, which
makes single-molecule detection possible, as is shown in Figure 1.2.
8-13
Measuring
nanoscale materials like graphene, carbon nanotubes and transition metal
dichalcogenides has been its main applications in nanotechnology.
14-19
Figure 1.2. Schematic of Surface Enhanced Raman Spectroscopy (SERS)
The mechanism of the SERS enhancement effect is still under debate in the
community. Two primary theories are generally accepted are an electromagnetic
mechanism (EM) and a chemical enhancement mechanism (CE). EM arises from the
enhanced electromagnetic field locally on metal nanoparticles or rough metal surfaces,
while CM originates from the formation of charge-transfer complexes between the
Substrate
Gold/Silver Nanoparticles
Analyte
Nanoparticles = nano amplifier
Raman Scattering
× 10
6
- 10
14
4
metal and the adsorbed target molecule.
20-23
Although the two mechanisms are
completely different, it is not that straightforward to distinguish them experimentally.
1.1.3 Electromagnetic Mechanism (EM) of SERS
The substantial enhancement of Raman signal of the adsorbed molecules is a
result of the enhancement in the electric field provided by the rough surfaces or
particular-structured surfaces. The photons in the incident light hit the surface and
excite the localized surfaced plasmons. When the wavelength of the incident light
matches the plasmon frequency, the field enhancement is the largest.
24-26
To note that,
only perpendicular oscillations of the plasmon can make SERS happen. If the
oscillations are in plane with the surface, there will be no SERS effect.
1
A two-stage
enhancement occurs in the SERS effect, as is shown in Figure 1.3a. First, the enhanced
field magnifies the incident light intensity, which excite the Raman modes of the
adsorbed molecule, hence the Raman signal is increased. Then it is further enhanced
by the structured surface due to the same mechanism, which also enlarges the Raman
signal, leading to a greater increase in the total output. At each stage the electric field
is increased by the factor of 2, for a total enhancement of E
4
.
27
The surface metals are
chosen according to the plasmon resonance frequency. Incident light in SERS are
mostly visible and near-infrared radiation (NIR). Gold, silver, copper, platinum and
palladium have the proper plasmon resonance frequencies, which fall within NIR
wavelength ranges. Gold and silver are most widely used in SERS and can produce
fields as much as 3 orders of magnitude.
24-26
5
Unlike EM, CM only applies for the system in which the metal and the
molecules have charge transfer or form chemical bonds. Thus, CM cannot explain the
cases when the molecules only physically adsorbed on the structured metal surfaces.
Interestingly, SERS enhancement has been reported even when molecules and metal
surfaces are far apart,
28
which is seen as a strong evidence of the EM mechanism.
1.1.4 Chemical Mechanism (CM) of SERS
In principle, the EM can affect all Raman modes, while the CM is highly related
to the symmetry of each specific vibrational mode. This is the reason that SERS spectra
differs from the Raman spectra of bulk solution. In many cases, the magnitude of the
Raman enhancement cannot be explained by the model of EM only. For instance, the
highest order of magnitude mentioned in section 1.1.2 is 10
14
cannot be reached by EM
mechanism solely.
In fact, the charge transfer can easily happen, since a lot of molecules involved
in SERS contain the atoms which have lone pair of electrons. This type of lone-pair
electrons tends to fill in the empty orbitals of the transition metals. In another word,
charge transfer occurs between the structured surface metals and the chemisorbed
molecules, see Figure 1.3b. To note that, the chemical process works with EM
mechanism to carry out the boost of the Raman signal, that makes single molecule
detection possible.
29
6
Figure 1.3. Schematic of (a) Electromagnetic Mechanism (EM) and (b)
Chemical Mechanism (CM) of Surface Enhanced Raman Spectroscopy
(SERS)
Substrate
Gold/Silver Nanoparticles
Substrate
Gold/Silver Nanoparticles
7
1.1.5 Gold-island Material System
Due to its good conductivity, gold has been used as an important electrode
material for decades. In our previous work, we have studied a structured surface
platform which has plasmonic enhancement effect.
24, 25, 26, 30
5nm thin film of gold were
deposited on different substrates by electron-beam evaporation in vacuum. To note that,
5nm is not thick enough to form continuous films. Instead, it has island-like structures
that have been reported to exhibit strong plasmon resonances.
26, 30, 31
SEM and TEM
images of Au island films with a nominal thickness of 5nm are shown in Figures 1.4a
and 1.4b. To figure out its physical properties, we conducted calculation on the electric
field distribution across these films using Finite-difference time-domain (FDTD)
method, as shown in Figure 1.4c. These simulations were carried out on the
supercomputers in our campus, with a grid spacing of 2Å. The electric field response
of these films is dominated by the small gaps, for instance less than 2nm, between the
islands. These gaps can produce intense “hot spots” in electric field, as illustrated in
Figure 1.4c. The electric field intensity has been increased 10
3
times compared to the
incident electric field in these localized “hot” regions, causing the strong
electromagnetic SERS enhancement.
24, 25
In spite of its strong plasmonic enhancement
effect, gold atom can bond to sulfur covalently. The choice of surface-bound thiolated
molecules is the basis of this spectroscopic approach.
8
Figure 1.4. (a) SEM and (b) TEM images of 5nm thick Au island films,
together with the corresponding (c) electric field distribution calculated
using the FDTD method.
(a)
SE
9
1.1.6 Electrostatic/Electrochemical SERS (E-SERS)
Measurement of SERS spectra under applied electrostatic/electrochemical
potentials is referred to as Electrostatic/electrochemical SERS (E-SERS). This
approach has been applied by a handful of groups, most of which reported irreversible
changes in the Raman spectra.
32-37
Natelson team have reported substantial bias driven
Stark shifts reversibly using Phenyl-C60-butyric acid methyl ester (PCBM)/C60-gold
system to study SERS in the presence of electrostatic fields.
38, 39
While voltage
dependent Raman peaks have been reported, a rigorous theoretical understanding of its
mechanism is still lacking. Up to now, it is generally believed that it can be explained
by the CM and mostly depends on the surface potential. Not only the overall intensities
are enhanced, but also the relative peaks intensities are changed as well.
33, 34
From the
Raman spectra of the biased surface molecules, we can tell that their chemical
structures are obviously changed.
35
It can only prove that relative changes by the
symmetric or anti-symmetric nature of the vibrational modes. But its quantitatively
explanation is still lacking.
40-42
A solid theoretical picture supported by quantum-
mechanical modeling can largely help up to understand bias dependence.
10
1.2 Graphene Enhanced Raman Spectroscopy (GERS)
1.2.1 Graphene & Its Raman spectroscopy
Graphene (here we refer to monolayer graphene simply as ‘‘graphene’’ in this
thesis) is a single two-dimensional (2D) sheet of covalent bonded carbon atoms in a
honeycomb lattice,
43
which is the basic structural element of a couple of allotropes,
including carbon nanotubes, graphite and so on. Rolled up graphene sheet is carbon
nanotube, and multilayer stacked graphene layers with weak interlayer Van der Waals
force is called graphite. The graphene system is exactly one atomic monolayer thick,
and carrier dynamics is also confined within one strict 2D layer. So it is considered as
the most perfect two dimensional electronic material in nature.
44
Figure 1.5 shows the honeycomb lattice of graphene, which consists
parallelogram unit cells. Each parallelogram unit cell has two carbon atoms, designated
A (blue) and B (red). Carbon A belongs to one cell itself. However, each carbon B is
shared by four cells. Thus, there is 1/4*4=1 carbon B in one cell. There are one s and
three p orbitals in one carbon atom, in which sp
2
hybridization happens in one plane
and one p orbital is perpendicular to the plane. Within the plane, s orbitals and in-plane
p orbitals are shared with each other to form strong covalent bonds. The electrons can
also be shared in the remaining p orbitals that are oriented perpendicular to the plane,
which hybridizes to form π (valence) and π* (conduction) bands, as shown in Figure
1.6. On one hand, it improves graphene’s in-plane mechanical strength. On the other
hand, sharing of the electrons in the π-system is responsible for graphene’s conductivity.
In an undoped graphene sheet, the Fermi energy is defined as the energy where
11
valence and conduction bands meet (or neutrality point). We can see that the π and π*
bands (blue bands in Figure 1.6) are decoupled from the red σ and σ* bands. Occupied
and empty states are separated by the Fermi energy. The bands form conical valleys
that meet at two of the high-symmetry points, labeled as K and K', in the Brillouin
zone (inset plot in Figure 1.6). Near these points the energy is linearly related to the
magnitude of momentum measured from the Brillouin-zone corners.
44
Figure 1.5. The honeycomb lattice of graphene
Figure 1.6b is the measured monolayer graphene Raman spectrum with 2.33 eV
(532nm) laser excitation energy in our lab. The most prominent peaks in the Raman
spectra of monolayer graphene are the G band at 1580 cm
-1
and the 2D band at about
2700 cm
-1
. If the graphene sample is disordered or the Raman spectra is taken at the
edge of it, we can also see the disorder-induced D band, at about half of the frequency
of the 2D band (around 1350 cm
-1
using laser excitation at 2.33 eV).
A
B B
A
B B B
12
Figure 1.6. (a) 2D electronic band structure of graphene. (The blue conical
illustration shows that the conduction band and valence band intersect at
point K and point K’ in Brillouin zone. Adapted from Geim, A. K.;
MacDonald, A. H. Physics Today 60, 8, 35 (2007).
44
) (b) Measured
monolayer graphene Raman spectrum with 2.33 eV laser excitation energy
G
2D
D
G
*
(b)
13
1.2.2 Raman Studies on Gated Graphene: Electron-Phonon Interaction
The interaction between electrons and phonons is crucial for understanding most
of the physical properties of graphene’s, such as electronic and thermal conductivity.
Since graphene has a zero band gap, the lattice vibrations are partially screened by the
electrons.
14
In 1959, Kohn showed that this screening effect changes dramatically at
special wave vectors, leading to a divergence in the phonon dispersion at these special
points,
45
which is known as the Kohn anomaly. In graphene, the special wave vectors
are related to Fermi wavevectors k1 and k2, which also correspond to the corners of the
first Brillouin zone (namely the K and K’ vectors in inset of Fig. 1.6a). Piscanec et al.
46
reported the Kohn anomaly occurs for special phonons at the Γ and K points of
graphene (inset of Fig 1.6a), where the phonon branches exhibiting the Kohn anomaly
are those associated with the Raman G band and 2D band, respectively.
46
The physical
origin for the Kohn anomaly in graphene is as follows: firstly, an electron is excited
from valence band to the conduction band by absorbing a phonon, creating an electron
hole-pair; the electron and hole then recombine, emitting a phonon.
47
The
renormalization of the phonon energy is strongly dependent on the Fermi level
position, which is tunable by doping graphene with electrons or holes.
14
Lazzeri et al.
48
performed a theoretical calculation of the phonon energy and
linewidth as a function of the electron concentration, as is shown in Figure 1.7. This
calculation was conducted for a non-adiabatic process, considering the electron-
phonon interaction is strong in graphene. There is a logarithm divergence of the phonon
frequency, especially at low temperature, which is smoothed out at room temperature.
14
Figure 1.7. Linewidth and Frequency of the Raman G band as a function
of the electron concentration which is proportional to the square of the
Fermi level energy. (Adapted from Ref.
48
)
Electron-phonon interactions are investigated via electrostatic doping, chemical
doping and electrochemical doping. For electrostatic doping, usually solid-state
graphene field-effect transistors (FETs) are used, namely a Si substrate as a back-gate
and a SiO2 epilayer as a gate dielectric. However, Raman response of graphene can be
only studied in the vicinity of the Dirac point (|EF| = 0-300 meV).
49-52
To reach higher
doping levels, other methods based on chemical doping
53-56
and electrochemical
gating
57-59
have been studied.
15
In this thesis, the linear dispersion relation and constant density of electronic
states of graphene, are used to provide a convenient system with linear relationships
between the G band frequency (∆ωG) and its Fermi energy (EF) , with slightly different
slopes observed for electron and hole doping, as is shown in Figure 1.8.
60
Also, a
quadratic relationship between Fermi energy and doping concentration are reported.
Thus, using the relation between Fermi energy (EF) and doping concentration (n)
obtained by Das Sarma et al, EF=11.65 √ 𝑛𝑛 � (meV, in which 𝑛𝑛 �= n/ 10
10
cm
-2
)
43
. With
these two relations combined, we can get a direct conversion from the graphene G band
frequency shift to the doping concentrations on the graphene sheet, which can provide
some information of the system, such as local ion concentration or electric field on the
surface.
16
Figure 1.8. Fermi energy E
F
as a function of the relative frequency of the G
mode ∆ω
G
. Measurements on five different devices are presented with
different symbols. The dashed and solid lines correspond to the two fitted lines
for electrons and holes with the equations in the graph. (Adapted from Ref
60
)
17
1.2.3 Application of Graphene in SERS
Raman signal of several adsorbed molecules on monolayer graphene were
reported to be drastically enhanced, such as rhodamine 6G (R6G), phthalocyanine (Pc),
protoporphyin IX (PPP) and crystal violet (CV).
61, 62
Figure 1.9 shows a comparison
of the Raman intensity of R6G on graphene and on a bare Si/SiO2 substrate. This
enhancement, called graphene-enhanced Raman scattering (GERS), originated from
charge transfer between graphene and the adsorbed molecules on it. This is exactly
from chemical enhancement mechanism, which is mode specific and dependent on the
symmetry of vibrations of the molecules. Sub-monolayer (1 or 2 Å) of the organic dye
molecules on graphene surface was deposited using thermal evaporation under the
pressure of 10
-4
Pa. The organic dye molecules can also be deposited by simply soaking
the Si/SiO2 substrate with graphene in diluted solution (e.g. 10
-5
~10
-4
M), although with
less control over the uniformity. To note that, graphene is largely transparent in the
visible range (with only 2.3% absorption). And its plasmon resonances lie in the
infrared wavelength range.
63, 64
Thus EM enhancement is minimal in graphene, which
makes GERS a perfect system to study CM, as introduced in section 1.1.4.
25, 26
It should
be noted that the underlying chemical enhancement mechanism associated with this
GERS phenomenon is still not fully understood.
18
Figure 1.9. Comparisons of Raman signals of R6G and PPP deposited on
graphene (red line) and on the SiO
2
/Si substrate (blue line) using vacuum
evaporation at 514.5 nm excitation and 632.8 nm excitation. The peak
marked by the star (*) is the G band of graphene. (Adapted from Ref
65
)
19
1.2.4 Graphene-based Electrodes
Graphene has quite unique properties, such as single atomic layer thickness,
large Raman scattering cross-section, strong electron-phonon coupling, and small
density of electronic states, that makes its vibrational modes sensitive to its charge
density. It can be used as an electrode under electrochemical working conditions,
providing measurable charge induced in the graphene.
66
The graphene-electrolyte interface also has other advantages over traditional
electrodes: 1.) It is electronic conductive and optical transparent over the visible and
IR wavelength ranges. 2.) It is chemically “clean” compared with other transparent
electrodes such as indium tin oxide (ITO). 3.) Its Raman spectra provides in-situ
measure of the local charge density via the frequency shift of the G band (∆ωG) (vide
infra).
66
4.) Graphene can be used as a conductive substrate for photo-catalytical
materials, including metal oxides, metal nanostructures, and molecular catalysts
attached using a huge database of diazonium chemistry to functionalize the surface of
graphene.
67, 68
Figure 1.10 is a monolayer graphene sample made by me. Beside its
potential use in electrochemistry, the monolayer graphene surface can also be used for
other electronic devices, such as sensors,
69
photodetectors
70
and solar cells.
71-75
20
Figure 1.10. Monolayer graphene with gold electrode on glass
(b)
21
1.2.5 Chemical Vapor Deposition (CVD) of Graphene
Graphene can be fabricated using multiple methods. In this thesis, the monolayer
graphene was mainly synthesized via chemical vapor deposition (CVD).
76
Surface
etched and cleaned copper foil was loaded in a quartz tube in the oven. Then the quartz
tube was connected to the inlet and outlet of the stainless-steel tubes of the sealed gas
system. The pressure of system was pumped down to 1-1.5 Torr. Then H2 gas was
flowed into the tube to make a chemically reductive environment. At the same time,
the temperature was gradually increased to 1000
o
C within 40 minutes. At 1000
o
C,
methane was then flowed in at a 1/7 flow rate compared to hydrogen gas for 1 hour.
After that, the methane flow was stopped, and the temperature was decreased to room
temperature for 30 minutes. H2 gas should be kept on until room temperature.
76
The
copper foil with graphene was then spin-coated with a protection layer of PMMA-A6
at 3000 rpm for 30s and then baking at 170
o
C for 5 minutes. We used copper etchant
to etch away the copper from the downside of the sandwich-like structure. A
graphene/PMMA film was floating on the surface of the copper etchant solution. Here,
the PMMA film acted as a supporting layer. It is then cleaned with 10% hydrochloric
acid in DI water.
77
The PMMA-graphene layer was then scooped up on the target
substrate. The sample was then baked at 150
o
C for 15 minutes, then 80
o
C for 30
minutes to improve adhesion. After this, the PMMA layer was removed by soaking in
acetone overnight. The quality of the graphene can be monitored by measuring the
resistance or taking Raman spectra on the surface. The schematic is shown in Figure
1.11.
22
Figure 1.11. CVD-growth and wet transfer of monolayer graphene
23
1.3 Vibrational Stark-shift Spectroscopy
1.3.1 Stark-shift Effect
When applied an external electric field, the spectral lines of atoms and molecules
will be shifted and split, this is called Stark effect.
78, 79
The Stark effect is linear
(proportional to the applied electric field) or quadratic with a high accuracy for most
spectral lines, and can be observed for both emission and absorption lines. To
understand the mechanism of this effect, we can think of applying an electric field from
left to right to a system with a nuclei and electrons. The nuclei tend to be pushed to the
right and electrons tend to be dragged to the left. For an electronic state in this system,
its energy is lowered when its electron cloud tends to be on the left, and of course raised
when its electron cloud tends to be on the right.
More importantly, Stark effect can lead to splitting of degenerate energy levels.
Under an electric field, the energies of originally hybrid orbitals (e.g. one s orbital and
three p orbitals in sp
3
hybridization) will be different because in which electron tends
to be in different directions. Thus, the formerly degenerate energy levels will split into
slightly lower and slightly higher energy levels, as is shown in Figure 1.12.
24
Figure 1.12. Computed regular (non-chaotic) Rydberg atom energy level
spectra of hydrogen in an electric field near n=15 for magnetic quantum
number m=0. Each n level consists of n-1 degenerate sublevels;
application of an electric field breaks the degeneracy.
79
(Michael Courtney
[CC BY-SA 3.0 (https://creativecommons.org/licenses/by-sa/3.0)])
25
1.3.2 Related Molecular Systems
In this thesis, we also measured the electric fields at the surface of electrodes other
than graphene-based electrodes under electrochemical working conditions using in-situ
SERS spectroscopy. On these electrodes, Stark-shifted molecules are bounded to their
surface by soaking the nano-structured sample in 0.03M 4-mercaptobenzonitrile
solution in ethanol. As demonstrated in Figure 1.13a,c, the surface of Au nano-islands
was connected to 4-mercaptobenzonitrile by Au-S bond. SERS-enhanced Raman shift
of the nitrile (i.e. C-N) stretch was measured.
80
The Figure 1.13b showed the gold
island structure mentioned earlier in Section 1.1.5.
Figure 1.13. (a) Schematic diagram of the in-situ SERS measurements
using a water immersion lens. (b) SEM image of the 5nm Au film. (c)
Cross-section diagram of the sample structure.
200nm
(b)
(c)
26
1.3.3 Vibrational Sum Frequency Generation (VSFG)
Sum frequency generation (SFG) is a 2
rd
-order nonlinear optical process based
on the annihilation of two incident photons at angular frequencies ω1 and ω2 and
simultaneously, one photon at frequency ω3 is generated.
81
This phenomenon can only
occur under conditions where high intensity light (mostly pulsed laser) is interacting
with the material (namely the interfaces or surfaces) asymmetrically.
82
Thus, SFG is
extremely useful in getting spectroscopic information at the surfaces.
Using sum frequency generation (SFG) spectroscopy, Dawlaty et. al conducted
a systematic study of the interfacial solvation effects on the C-N frequency shift of
MBN bound to Au electrode surfaces.
83, 84
Here, they used bulk gold electrode pad
deposited by electron beam evaporator, as is shown in Figure 1.14. Since it was the
first time this Stark-shift spectroscopy have been used to study the interfacial fields, it
was quite difficult to perform the experiments. Absorption of the IR beam is quite
strong in water (electrolyte), which limits the optical pathlength to thin micro-fluidic
channels/cells (about 25μm). Also, the signal produced by the surface layer of
molecules is quite small and can be easily obscured by the large background signal
from other components of the system. Despite of these difficulties, they were able to
perform Stark-shift spectroscopy using SERS) which is much easier to perform and
has a couple of advantages over SFG spectroscopy. SERS eliminates the need for
micro-cell reactors, thus, no large voltage drops associated with the series resistance of
the electrolyte. What is more important, SERS has much higher signals that SFG,
enabling spectra to be acquired in seconds rather than minutes, and with higher signal-
27
to-noise ratios. SERS-based vibrational Stark shifts have also been applied to probe
interfacial electric fields by Harris et al.
85, 86
and Hildebrandt et al.
87
using roughened
metal electrodes.
Figure 1.14. (a) Schematic diagram illustrating the SERS measurement of
thiolated-benzonitrile bound to an electrode for Stark-shift spectroscopy.
(b) Stark-shift of the C-N stretch vibrational mode obtained by SFG
spectroscopy plotted as a function of applied potential.
Solvent
E
(a)
(b)
28
1.4 Photo-electrochemical (PEC) Measurements
1.4.1 Important Reactions in Artificial Photosynthesis
The two most important chemical reactions in semiconductor photocatalysis
using solar energy conversion are photocatalytic water splitting and CO2 reduction. In
this thesis, we are mostly focused on the water splitting reaction.
Water splitting, namely producing hydrogen (H2) and oxygen (O2) from water by
electrochemical potentials with the help of the solar energy. As an important alternative
to hydrocarbon fuels, hydrogen fuel has drawn intensive attention since 1970s, when
the photocatalytic water splitting under ultraviolet radiation was first come up with.
88
Figure 1.15a is a schematic diagram of photoelectrochemical cell, in which the water
splitting reaction is performed. A thin charge layer is formed near the surface of the
photoelectrode in the electrolyte to establish the thermodynamic equilibration with the
ions in solution. If the photon of incident light has the energy larger than the band gap,
electron-hole pairs are generated and can be separated in the space charge layer. The
photo-generated electrons are then transferred to the cations on surface of the working
electrode, that forms the hydrogen, which is called hydrogen evolution reaction (HER).
At the meantime, the photo-generated holes are effectively transferred to the interface
of the counter electrode and the electrolyte, where oxygen is produced form the O
element from the water, namely oxygen evolution reaction (OER).
89, 90
By applying either positive or negative electrochemical potentials, we can visit
the HER or OER. By examining the current density-voltage (j-v) curve, we can have
an idea of the performance of the catalytic materials. In Figure 1.15b, there is an
29
example of the application of the photocatalytic HER measurement using the Gamry
potentiostat by our research group. We used GaP, a III-V semiconductor material, as
the photo-catalytic material for the photocathode, on which the HER is happened.
Photocatalytic current-potential curves measured for GaP photocatalysts with various
thicknesses of TiO2 under 1W/cm
2
532nm illumination in a 0.5M Na2SO4 pH=7
solution were plotted.
91
As we can see from the plot, the so-called turn on voltage of
the curve using different thickness of the TiO2 on the GaP were quite different. We can
see a 0.5V shift (from -0.7V to -0.2V) of the turn-on voltage. Compared to bare GaP,
the GaP with 10nm TiO2 on the surface seems to be turned on with minimal
electrochemical potential added. That means, the energy from the photons are used
more efficiently if there is a 10nm TiO2 on the GaP surface. The addition of the TiO2
layer makes the GaP a better HER catalyst in this case.
30
Figure 1.15. (a) Schematic diagram of a photoelectrochemical cell for
photocatalytic water splitting (b) Photocatalytic current-potential curves
measured for GaP photocatalysts with various thicknesses of TiO
2
under
1W/cm
2
532nm illumination in a 0.5M Na
2
SO
4
pH=7 solution
91
Potentiostat
Argon in
Gas out
to GC
Light
Photocathode Pt Ref.
(a)
(b)
31
1.4.2 Three-terminal Photo-electrochemical Setup
Photoelectrochemical measurements are performed using a three terminal
potentiostat (Gamry, Inc.), as illustrated in Figure 1.16. There are three electrodes
needed: counter electrode (CE) reference electrode (RE) and working electrode (WE).
The CE is used to close the current circuit loop in the electrochemical cell. It is
generally fabricated using an inert material (e.g. Pt, glassy carbon) and usually it does
not participate in the electrochemical reaction, namely no net reactions happen on the
CE. The current is flowing between the WE and the CE. To avoid CE being the limiting
factor in the kinetics of the process, the total surface of CE should be higher than that
of WE. RE has a stable and well-stablished potential and is used as a reference point
for controlling and measuring the potential of the RE. Inside the RE, there will always
be a redox system with buffered or saturated ions with stable concentrations. Here, the
current flow through the RE is kept around zero by using CE to close the current loop
and a high input impedance on the electrometer (> 100 GOhm). The WE is where the
reaction of interest is happening. All three electrodes are immersed in the same
electrolytes in liquid phase as the media.
92
As is shown in Figure 1.16, in our case, a
platinum wire and a Ag/AgCl reference electrode were used as the counter electrode
and reference electrode, respectively. The electrodes are immersed in a pH=0, 0.5 M
sulfuric acid solution, which the can supply enough protons for the HER and has a
stable anion, SO4
2-
as well.
32
Figure 1.16. Schematic diagrams of the three-terminal photo-
electrochemical cell using a water immersion lens
1.4.3 In-situ Raman Spectroscopy combined with PEC
Using the Renishaw Raman system in our lab, we are also able to use the water
immersion lens to measure the in-situ Raman spectra, while doing the PEC
measurements. Raman spectra of the graphene electrode were taken under 633nm
wavelength excitation under applied electrochemical potentials in pure DI water using
a water immersion lens, as illustrated in Figure 1.17a. To note that, a 13µm thick Teflon
sheet (American Durafilm, Inc.) is used to wrap the water immersion lens, to avoid the
direct contact of the lens and the solution, which might be corrosive. And also, there
should always be one drop of DI water in between the Teflon sheet and the tip of the
lens, since the water immersion lens is calibrated within the water environment.
Gamry
Potentiostat
Pt
Counter
Electrode
5nm TiO
2
p-Si
200nm
Al
Ag/AgCl
Reference
Electrode
532nm
laser
0.5 M
H
2
SO
4
solution
33
Figure 1.17. Schematic diagrams of the in-situ Raman measurement
combined with the PEC set up and 3D plot of the graphene sample using
as the working electrode in a PEC cell.
Gamry
Potentiostat
Ag/AgCl
Reference
Electrode
Glassy
carbon
Counter
Electrode
Graphene
Working
Electrode
Water
Immersion
Lens
633nm
Laser
V
A
DI
water
(a)
(b)
34
CHAPTER 2: Sensing Local pH and Ion Concentration at
Graphene Electrode Surfaces Using In-situ Raman
Spectroscopy
This chapter is similar to Shi et al.
66
, published in Nanoscale.
2.1 Abstract
We report a novel approach to probe the local ion concentration at
graphene/water interfaces using in situ Raman spectroscopy. Here, the upshifts
observed in the G band Raman mode under applied electrochemical potentials are used
to determine the charge density in the graphene sheet. For voltages up to ±0.8V vs.
NHE, we observe substantial upshifts in the G band Raman mode by as much as 19cm
-
1
, which corresponds to electron and hole carrier densities of 1.4×10
13
cm
-2
and Fermi
energy shifts of ±430meV. The charge density in the graphene electrode is also
measured independently using the capacitance-voltage characteristics (i.e., Q=CV),
and is found to be consistent with those measured by Raman spectroscopy. From charge
neutrality requirements, the ion concentration in solution per unit area must be equal
and opposite to the charge density in the graphene electrode. Based on these charge
densities, we estimate the local ion concentration as a function of electrochemical
potential in both pure DI water and 1M KCl solutions, which span a pH range from 3.8
to 10.4 for pure DI water and net ion concentrations of ±0.7mol/L for KCl under these
applied voltages.
35
2.2 Introduction
The local pH (or ion concentration) at the surface of an electrode can differ
from that of the bulk solution by several orders of magnitude, affecting the energetics,
reaction field strength, and kinetics of electrochemical reactions. Sensing the ion
concentration at electrode surfaces is potentially important in controlling the selectivity
of electrochemical and photoelectrochemical reactions, like CO2 reduction, which
normally compete with the hydrogen evolution reaction. There have been several
previous attempts to quantify the local ion concentration at electrode surfaces. Using a
rotating platinum disc electrode, Auinger et al. measured the near-surface ion
distribution and buffer effects in electrochemical reactions including hydrogen
oxidation (HOR) and hydrogen evolution reactions (HER).
93-95
While the ideal
conditions utilized in this fundamental study cannot be directly applied to real scenarios,
they do provide a basic understanding of the concept of surface pH for more complex
heterogeneous reactions. Gupta et al. calculated the cathode surface concentrations in
the electrochemical reduction of CO2 in KHCO3 solutions.
96
In this work, the authors
predict a pH variation of 4.25 within 30µm of the electrode surface. Deligianni et al.
performed the first in situ measurement of surface pH during electrolysis using a
rotating pH electrode.
97, 98
Here, their spatial resolution was on the order of tens of
microns. Using an optical approach, Leenheer and Atwater performed imaging of
water-splitting electrocatalyst surfaces using pH-sensing confocal fluorescence
microscopy.
99
Here, small concentrations of a pH indicator dye added to the aqueous
electrolyte enabled ratiometric fluorescence sensing for quantitative pH detection over
36
a relatively small pH range from 5.3 to 7.5, but with minimal perturbation to the local
environment. While interesting, these previous attempts to establish the local pH near
electrode surfaces have been inaccurate, invasive, and/or do not provide small enough
spatial resolution to be considered “close” enough to the electrode to be relevant to the
reaction of interest.
Our approach exploits the following unique properties of graphene: 1.) single
atomic layer thickness, 2.) high Raman scattering cross-section, 3.) small density of
electronic states, and 4.) large electron-phonon coupling strength, which make
graphene’s vibrational modes quite sensitive to charge density. All these combined
properties provide an accurate measurement of the charge induced in the graphene
electrode under electrochemical working conditions.
2.3 Experimental Details
Monolayer graphene is grown by chemical vapor deposition (CVD) on copper foil
at 1000
o
C in methane and H2 gas at a reduced pressure of 1-1.5 Torr.
76
After growth,
the copper foil is spin-coated with PMMA-A6 at 2000 rpm for 45s and then baked at
150℃ for 5 minutes. The copper foil is then etched away in copper etchant, and the
graphene with PMMA is “scooped” out and rinsed in 10% HCl and DI water. Next, the
graphene monolayer is transferred to the target substrate with the same scooping
method and baked at 120℃ for 5 minutes to improve adhesion. After this, the PMMA
layer is removed with a 5-minute acetone dip.
100
Prior to transferring the graphene, two
gold electrodes are deposited on a glass slide by electron-beam evaporation using a
37
shadow mask. After transferring the graphene to these electrodes, copper wires are
connected to the gold electrodes and epoxy is used to cover the surface of the gold
electrode so that only the graphene electrode is in contact with the electrolytic solution.
After the protection of the epoxy the active graphene area was about 1cm
2
. Raman
spectra (532nm excitation) of the graphene electrode under applied electrochemical
potentials are taken in liquid solutions consisting of pure DI water and 1M KCl using
a water immersion lens, as illustrated in Figure 2.1a. In order to protect the lens from
the solution, a 13µm thick Teflon sheet (American Durafilm Inc.) was used to cover
the lens. Three terminal potentiostat (Gamry, Inc.), as illustrated in Figure 2.1a, is used
to add bias between our graphene working electrode and the reference electrode. A
glassy carbon (SPI, Inc.) and an Ag/AgCl reference electrode were used as the counter
electrode and reference electrode, respectively. Three-terminal potentiostat setup
eliminates errors associated with voltage drops across the low conductivity DI water.
Electrochemical impedance measurements were also carried out using the
electrochemical impedance spectroscopy the three-terminal potentiostat (Gamry, Inc).
38
Figure 2.1. (a) Schematic diagrams of the three-terminal
photoelectrochemical cell using a water immersion lens and (b) monolayer
graphene electrode.
copper wire
glass
substrate
Au contact
monolayer
graphene
copper wire
epoxy
(b)
Gamry Potentiostat
Ag/AgCl
Reference
Electrode
Glassy
carbon
Counter
Electrode
Graphene
Working
Electrode
Water
Immersion
Lens
532nm
Laser
V
A
DI
water
39
2.4 Results and Discussion
Figure 2.2 shows the G band Raman shift measured as a function of the applied
voltage measured in a three-terminal electrochemical cell. Here, we observe substantial
upshifts in the G band Raman frequency for both positive and negative applied
potentials. Using the relation between Fermi energy (EF) and doping concentration (n)
obtained by Das Sarma et al, EF=11.65 √ 𝑛𝑛 � (meV, in which 𝑛𝑛 �= n/ 10
10
cm
-2
)
43
, and the
linear relation between EF and ∆ωG (with slightly different slopes observed for electron
and hole doping)
60
, we also plot the charge concentration in the graphene as a function
of the applied voltage on the right axis of Figure 2.2a.
49
Here, we see that the charge
neutrality point occurs at 0.0V vs. NHE. For voltages above this charge neutrality
point, the ions in solution are OH
-
ions, and for voltages below this, the ions are H3O
+
.
At the charge neutrality point, we expect the ion concentration at the surface of the
electrode to reach a minimum (i.e., the point of minimum charge in solution). In this
plot, we also indicate the Fermi energy of the graphene on the top horizontal axis,
which spans a range from -430meV to +390meV. It should be noted that graphene,
because of its linear dispersion relation and constant density of electronic states,
provides a convenient system with linear relationships between the G band frequency,
applied voltage, and Fermi energy, and a quadratic relationship between Fermi energy
and doping concentration.
40
Figure 2.2. (a) G-band Raman shift and corresponding charge
concentration and (b) G band Raman linewidth plotted as a function of the
applied voltage and Fermi energy measured in DI water. (b) Raw spectra
showing the voltage-induced redshift of the G-band Raman mode.
(c)
(b)
41
Furthermore, because of graphene’s small electron density of states, the Fermi energy
of this material can be tuned over a large range of ±430meV. Results of more than one
samples shows the range of charge neutrality point is less than 0.1V. See Figure 2.3(b).
Figure 2.3. (a) G-band Raman shift and corresponding carrier
concentration with error bar (b) Results for 5 different samples
Figure 2.2b shows the G band linewidth plotted as a function of applied voltage, which
varies from 15.2cm
-1
at the charge neutrality point to 6.3cm
-1
for heavily doped
graphene. These results are consistent with previous reports of Yan et al.
49, 101
and arise
)
42
from a Kohn anomaly,
48, 102
which causes the G band to be broadened and downshift
near the charge neutrality point of graphene.
103-107
Figure 2.4 shows the corresponding Raman data taken with the monolayer
graphene electrode in a 1M KCl solution. Again, we see substantial upshifts (up to
18cm
-1
) in the G band Raman frequency under both positive and negative applied
potentials, as observed in the DI water solution (Figure 2.2). However, here, the
charge neutrality point occurs at +0.2V vs. NHE, indicating doping of the graphene due
to the ions in solution at zero applied potential. Here, the charge density on the
graphene reaches ±1.2×10
13
cm
-2
under the applied potentials in this range. Again,
we see a substantial drop in the G band linewidth from 15.2cm
-1
at the charge neutrality
point to 6.3cm
-1
when heavily doped, corresponding to doping-induced suppression of
the Kohn anomaly.
We also quantified the graphene charge concentration (and hence concentration
of ions) using electrochemical impedance (EIS) measurements, which provide the
capacitance-voltage relation at the electrode/electrolyte interface, as shown in Figure
2.5a. The charge (and carrier density) can then be obtained from these capacitance-
voltage characteristics (i.e. Q=CV). See figure 2.6 for further details. The Figure 2.5a
shows the capacitance-voltage plot of the graphene electrode measured in both DI
water and 1M KCl. Here, the capacitance-voltage
43
Figure 2.4. (a) G band Raman shift and corresponding charge
concentration and (b) G band Raman linewidth plotted as a function of the
applied voltage and Fermi energy measured in 1M KCl solution. (c) Raw
Raman spectra showing the voltage-induced redshift of the G band Raman
mode.
(c)
(b)
44
profiles both exhibit a dip near the minimum charge point, as observed with Hg
electrode measured in both DI water and 1M KCl. Here, the capacitance-voltage
profiles both exhibit a dip near the minimum charge point, as observed with Hg
electrodes.
108-110
The charge density per unit area obtained from the product of the
capacitance and voltage (i.e., Q=CV) is plotted as a function of the applied potential
for the DI water and 1M KCl solutions in Figures 2.5b and 2.5c, respectively. Here, the
charge density is plotted together with the charge density obtained by Raman
spectroscopy, and we observe excellent agreement between these two independent
measurements of charge in this graphene electrode system. The agreement is
particularly good near the charge neutrality point. However, discrepancies occur at
relatively high potentials, likely due to non-linearities in the ∆n/∆ωG relation for
heavily doped graphene. Here, the capacitance is roughly a factor of two larger in the
KCl solution than in pure DI water, as expected due to the short ion distribution (i.e.,
Debye length) in the ionic solution.
45
Figure 2.5. (a) Capacitance-voltage plot of the graphene-based electrode
measured in DI water and 1M KCl. Charge densities obtained from the
capacitance-voltage data (i.e., Q=CV) and Raman spectroscopy in (b) DI
water and (c) 1M KCl.
(b)
(c)
46
Figure 2.6. (a) 1/C
2
versus applied voltage, (b) Double layer capacitance
versus applied voltage, (c) Charge accumulated in the double layer versus
applied voltage and (d) Carrier concentration in the double layer versus
applied voltage from electrochemical impedance measurement.
1/C
2
vs V plots
Q vs V plots
C vs V plots
Carrier vs V plots
(d)
47
In order to estimate the local ion concentration at the electrode surface, we
assume that the ions in solution follow an exponential distribution as a function of
distance from the electrode surface with a decay constant given by the Debye length of
the solution. For DI water, λ
𝐷𝐷 = 961nm and for 1M KCl, λ
𝐷𝐷 = 0.3nm. From charge
neutrality requirements, we set the ion concentration in solution per unit area equal
(and opposite) to the charge density in the graphene electrode, i.e., nG
= ∫ 𝐴𝐴 𝑜𝑜 exp �−
𝑧𝑧 λ
𝐷𝐷 � 𝑑𝑑𝑑𝑑 =
∞
0
𝐴𝐴 𝑜𝑜 λ
𝐷𝐷 , where nG is the charge density in the graphene
electrode and 𝐴𝐴 𝑜𝑜 is the charge density at the interface, which can be obtained by the
ratio nG/ λ
𝐷𝐷 . Figure 2.7 shows the local ion concentration plotted as a function of the
applied potential for both pure DI water and 1M KCl solution. From these ion
distributions, we determine the local pH, which spans a range from 3.8 to 10.4 for pure,
pH neutral water under these applied voltages as plotted in Figure 2.7b. For the 1M
KCl solution, the net local ion concentration spans a range of ±0.78mol/L under these
applied voltages, as plotted in Figure 2.7c. It is important to note that, in DI water, this
approach provides a direct measure of the local pH. In electrolytic solutions (e.g., KCl),
however, the charge on the electrode (i.e. graphene) will be a sum of the ions in solution
(i.e. H
+
, K
+)
. However, in 1M KCl, the H
+
ion concentration is many (approximately
seven) orders of magnitude smaller than K
+
and can be neglected. The same
measurement was also carried out in 5 different solutions with pH values ranges from
0.58 to12.70, as shown in Figure 2.8a. The comparison of the results of the local pH
values calculated from the Raman measurements with defined pH are shown in Figure
2.8 as well.
48
Figure 2.7. Local ion concentration plotted as a function of voltage for (a)
pure DI water and (c) 1M KCl. (b) Local pH plotted as a function of
voltage for pure DI water.
DI Water
1M KCl
(a)
(b)
(c)
49
Figure 2.8. (a) Local pH plotted as a function of voltage for 5 solutions with
different pH. (b) Results of measured pH, Debye length of the solution, and
the known pH of the bulk solution.
Figure 2.9. Cyclic voltammetry of 50mM ferro/ferricynide redox couple in (a)
DI water and (b) 1M KCl.
Known pH Debye Length(nm)
bias
0.58 0.60 1.77
2.66 6.50 2.70
7.00 961 6.07
10.6 15.6 10.8
12.7 1.36 12.0
(a) (b)
(a) (b)
50
Figure 2.10. Cyclic voltammetry of graphene sample as working electrode
in DI water
As we can see, the measured pH values are very close to the defined pH values,
which verifies the applicability of the proposed pH sensing method. To calibrate the
Ag/AgCl reference electrode, we performed cyclic-voltammetry in a 50mM solution
of a known redox couple of K3Fe(III)(CN)6 and K4Fe(II)(CN)6 in both DI water and
1M KCl solution, as shown in Figure 2.9. For each calibration, we take the average of
the cathodic and anodic peak positions and compared these with the standard potential
of the ferro/ferricyanide redox couple (i.e., 0.436V versus NHE). We then shift our
applied voltages vs. NHE based on this calibration. We also performed cyclic
voltammograms obtained during the change of the potential (Figure 2.10). No
hydrogen evolution was observed at negative potentials, with the measured current
remaining below 5µA even at negative potentials. The reason for this low current (i.e.,
HER) is the low concentration of H
+
ions and the relatively high overpotential required
-1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0
-5
-4
-3
-2
-1
0
1
2
Current (µA)
Voltage vs NHE (V)
51
for HER on the graphene electrode. In fact, we chose the voltage range carefully to
avoid the formation of H2 bubbles, which will affect our Raman measurements.
2.5 Conclusions
In summary, we have developed a technique for measuring the local ion
concentration at graphene/water interfaces using Raman spectroscopy. Here, the charge
density in the graphene sheet is determined from the upshifts in the G band Raman
frequency. We observe upshifts as large as 19 cm
-1
under applied potentials of ±0.8V
vs. NHE, which corresponds to charge densities of ±1.4×10
13
cm
-2
and Fermi energy
shifts of ±430meV . An independent measurement of the charge density in the graphene,
based on the capacitance-voltage characteristics (i.e., Q=CV) is found to be consistent
with the Raman measurements, particularly near the charge neutrality point of
graphene. The local ion concentration at the water/graphene interface is estimated
assuming the charge per unit area is equal and opposite to the charge density in the
graphene electrode. The G band upshifts and, hence, ion concentrations are measured
as a function of voltage in both pure DI water and 1M KCl solutions. This type of local
probing of ion concentration is potentially important in controlling the selectivity of
electrochemical and photoelectrochemical reactions, like CO2 reduction, which
normally compete with the hydrogen evolution reaction.
52
2.6 Acknowledgements
The authors would like to thank Drs. Joel Ager, Sri Narayan and Rehan Kapadia
for valuable discussions. This research was supported by the Army Research Office
ARO Award No. W911NF-14-1-0228 (H.S.), Air Force Office of Scientific Research
Grant No. FA9550-15-1-0184 (B.H.), NSF Award No. CBET-1512505 (L.S.), and
ACS-PRF grant #55993-ND5 (N.P.).
53
Chapter 3: Monitoring Local Electric Fields at Electrode
Surfaces Using SERS-based Stark-shift Spectroscopy During
Hydrogen Evolution Reactions
This chapter is similar to Shi et al.
80
published in ACS Appl. Mater. Interfaces.
3.1 Abstract
We report the use of surface enhanced Raman scattering (SERS) to measure the
vibrational Stark shifts of surface-bound thiolated-benzonitrile molecules bounded to
an electrode surface during hydrogen evolution reactions. Here, the electrode surface
consists of Au nano-islands deposited both with and without an underlying layer of
monolayer graphene on a glass substrate. The Stark shifts observed in the nitrile (C-N)
stretch frequency (around 2225 cm
-1
) are used to report the local electric field strength
at the electrode surface under electrochemical working conditions. Under positive (i.e.,
oxidative) applied potentials (vs. NHE), we observe blueshifts of up to 7.6 cm
-1
, which
correspond to local electric fields of 22 MV/cm. Under negative applied potentials (vs.
NHE), the C-N stretch frequency is redshifted by only about 1 cm
-1
. This corresponds
to a regime in which the electrochemical current increases exponentially in the
hydrogen evolution process. Under these finite electrochemical currents, we estimate
the voltage drop across the solution (V=IR). Correcting for this voltage drop results in
a highly linear electric field versus applied electrochemical voltage relation. Here, the
onset potential for the HER lies around 0.2V vs. NHE and the point of zero charge
(PZC) occurs at 0.04V vs. NHE, based on the capacitance-voltage (C-V) profile. The
solution field is obtained by comparing the C-N stretch frequency in solution with that
54
obtained in air. By evaluating the local electric field strength at the PZC and onset
potential, we can separate the solution field from the reaction field (i.e., electrode field),
respectively. At the onset of HER, the solution field is -0.8MV/cm and the electrode
field is -1.2MV/cm. At higher ion concentrations, we observe similar electric field
strengths and more linear E-field versus applied potential behavior due to the relatively
low resistance of the solution, which results in negligible voltage drops (V=IR).
3.2 Introduction
Electrochemical reactions at solid-liquid interfaces represent complex processes,
which include the application of an applied potential between the working and counter
electrodes, binding of reactant and intermediate species to the electrode surface,
electrostatic fields within the double layer, diffusion of reactants in solution, and
ultimately charge transfer to the ions in solution. There are loss mechanisms associated
with each of these key components in the overall electrochemical efficiency. This
complex process is often oversimplified in order to provide a basic interpretation of
experimental data, largely because simple methods for separating these key
components do not yet exist. In the work presented here, we apply spectroscopic tools,
i.e., Stark-shift surface enhanced Raman scattering (SERS) spectroscopy, to study and
separate the key components of the electrochemical process in order to improve our
understanding of energy loss in this important energy conversion system. Large
electrostatic fields exist at the surface of electrodes largely due to the electrochemical
double layer at this solid/liquid interface.
84
Many chemical reactions involve charged
55
or polarized reactants, transition states, and products. Electric fields can have a
significant effect on any of these and, therefore, fields, reaction mechanisms, and
catalysis are intimately related to the local electric field strength.
Using sum frequency generation (SFG) spectroscopy, Dawlaty and coworkers
carried out a systematic study of the interfacial solvation effectss on the nitrile
frequency shift of MBN bound to gold electrode surfaces.
83, 84
While these prior Stark-
shift spectroscopy measurements have elucidated the interfacial fields for the first time,
there are several difficulties associated with SFG spectroscopy, chief of which is the
strong absorption of the IR beam in water (electrolyte), which limits the optical
pathlength to approximately 25µm-thick micro-fluidic channels/cells. Also, substantial
non-linear optical mixing by other components of the system create a large background
signal that can obscure the relatively small signal-produced by the surface layer of
molecules. As a result of these difficulties, there are only a handful of groups in the
world that are capable of performing SFG measurements reliably. In the work presented
here, we demonstrate the ability to perform Stark-shift spectroscopy using surface
enhanced Raman scattering (SERS), which is a far easier measurement to perform and
has several additional advantages over SFG spectroscopy. In typical SFG
measurements, the optical pathlength through the solution must be limited to 25 μm or
less in order to minimize absorption of the IR beam, requiring elaborate micro-cell
reactors to be employed. SERS eliminates the need for these micro-cell reactors, thus,
eliminating the large voltage drops associated with the series resistance of the
electrolyte. SERS has much higher signals that SFG, enabling spectra to be acquired
56
in seconds rather than minutes, and with higher signal-to-noise ratios.
SERS-based vibrational Stark shifts have been used previously to probe interfacial
electric fields by Harris et al.
85, 86
and Hildebrandt et al.
87
using roughened metal
electrodes. In the work presented here, we deposit Au films nominally 5nm in thickness
by electron-beam evaporation in vacuum which are not thick enough to form
continuous films and instead create island-like structures that are known to exhibit
strong plasmon resonances, as shown in Figure 3.1d.
26, 30, 31
The Cronin group has
studied these films extensively using the finite difference time domain (FDTD)
method.
24, 25
The electric field response of these films is dominated by the small gaps
between the islands, which produce intense electric field “hot spots”. In these localized
hot spot regions, the electric field intensity is 1000 times larger than the incident
electric field, resulting in strong electromagnetic SERS enhancement.
24, 25
Natelson and
his coworkers have large, reversible, bias driven Stark shifts using Phenyl-C61-butyric
acid methyl ester (PCBM)/C60-gold system to study SERS in the presence of
electrostatic fields.
38, 39
The use of surface-bound thiolated molecules is a key
component of this spectroscopic approach, as well as the use of a water immersion lens
to measure these SERS spectra in situ under electrochemical working conditions.
Measurement of SERS spectra under applied electrochemical or electrostatic potentials
is referred to as “E-SERS,” and there are only a handful of papers on this topic, mostly
reporting irreversible changes in the spectra and results that have not been reproduced
by other groups.
32-37
In these previous SERS studies, the Stark-shifted fields were not
correlated to the electrochemical current and point of zero charge, nor were solution
57
fields taken into account. We use Stark-shift SERS spectroscopy to establish the
electrostatic contribution to the reaction fields that take place at electrode surfaces in
an electrochemical process. These interfacial fields have been difficult to establish
experimentally but play an important role in the electrochemical process through the
initiation of charge transfer and in determining the overpotentials required to drive
reactions. Every molecule in a medium polarizes its surrounding environment, and the
induced polarization creates a field that is, in turn, felt by the molecule. This is known
as the solvation field, and this has important consequences for charge transfer at
interfaces. By elucidating the electrostatics at the electrode/electrolyte interface, we
develop a better understanding of the solvation fields near the interface. The point of
zero charge corresponds to the electrochemical potential above which the electrode
becomes positively charged and drives oxidation half-reactions and below which it
becomes negative and drives reduction half-reactions.
111
The surrounding electrolyte
acquires an equal and opposite charge (i.e., double layer and Debye layer) forming a
capacitor. At the point of zero charge, the electrode and the surrounding electrolyte
become charge neutral, and the capacitance is a minimum. What is fundamentally
interesting about this study is the fact that the electric field at the electrode surface does
not go zero at the charge neutrality point, as would be the case in a solid-state junction
(e.g., pn-junciton). This is due to the complex nature of the water/electrode interface.
We attempted to graft benzonitrile onto graphene directly from diazonium salts
using electrochemical reducing deposition. But we cannot get uniform layer of
molecules. Thus, we came up with the following method.
58
3.3 Experimental Details
The electrodes used in our work presented here utilize an underlying monolayer
of graphene grown by chemical vapor deposition (CVD) to electronically connect the
5nm Au nanoislands. Monolayer graphene was grown by chemical vapor deposition
(CVD) at 1000
o
C in methane and H2 gas on copper foil at a reduced pressure of 1-1.5
Torr.
76
The copper foil with graphene was then coated with a thin layer of PMMA-A6
by spin-coating at 2000 rpm for 45s and then baking at 150
o
C for 5 minutes. Copper
etchant was used to etch away the copper from the bottom of the “sandwich” structure,
resulting in a graphene/PMMA film that is left floating on the surface of the liquid
copper etchant. This film is then cleaned with 10% hydrochloric acid in DI water.
100
Prior to transferring the graphene, two gold electrodes are deposited on a glass substrate
by electron-beam evaporation using a shadow mask to serve as the target substrate. The
PMMA-graphene layer was then scooped up on the target substrate connecting both
gold electrodes. The sample was then baked at 120
o
C for 5 minutes to improve
adhesion. After this, the PMMA layer was removed with a 5-minute acetone dip. A 5nm
(nominal thickness) Au film was then deposited on the sample using electron-beam
evaporation. A thin layer of 4-mercaptobenzonitrile (4-MBN) molecules was deposited
by soaking the sample in a 0.03 mol/L solution of 4-MBN in ethanol for 24 hours,
83, 112
resulting in the structure illustrated in Figure 3.1b. We attached copper wires to both
gold electrodes as our working electrode in a three-terminal potentiostat setup. The
contact area between the copper wires and gold electrodes (coated with silver paint)
were then encapsulated in epoxy so that they were not in contact with electrolyte.
59
Figure 3.1. (a) Schematic diagrams of the three-terminal
photoelectrochemical cell using a water immersion lens. (b) Cross-section
diagram of the sample structure. (c) 3D plotting of the sample and the laser
(d) SEM image of the 5nm Au film.
Gamry
Potentiostat
Ag/AgCl
Reference
Electrode
Glassy
carbon
Counter
Electrode
Graphene
Working
Electrode
Water
Immersion
Lens
633nm
Laser
V
A
DI
water
b)
60
Raman spectra of 4-MBN were taken with 633nm wavelength excitation (see
Figure 3.2b), which is resonant with this molecule’s absorption, under applied
electrochemical potentials in pure DI water using a water immersion lens, as illustrated
in Figure 3.1a. In order to protect the lens from the solution, it was covered with a
13µm-thick Teflon sheet (American Durafilm, Inc.). A three terminal potentiostat
(Gamry, Inc.) was used to apply various electrochemical potentials to the graphene
working electrode with respect to the reference electrode. A silver/silver chloride
reference electrode and glassy carbon electrode (SPI, Inc.) were used as the reference
and counter electrodes, respectively. The monolayer graphene electrode (1 cm × 1 cm)
typically has an in-plane resistance of 1-2kΩ. Figure 3.2 shows SERS spectra taken of
thiolated benzonitrile on Au nano-islands deposited on a glass substrate both with and
without the underlying graphene monolayer. We observe a strong Raman signal from
the C-N nitrile stretch mode around 2225 cm
-1
with a 20-second integration time and
0.1mW laser power. It is also important to note that rectification of plasmon resonance
may also occur when using plasmonic structures to study Stark shifts.
113
Kwasnieski
and coworkers reported that the CN stretch of MBN varies as function of optical
power.
114
This implies that optical rectification of the plasmon driven field can be used
to change the surface potential. Nelson et al. has shown that this effect can even alter
surface reactivity.
115
However, according to Nelson’s laser power-CN stretch relation,
the laser power used in our work reported here (0.1mW) is on the low end of powers
inducing any appreciable surface optical rectification effects. At such low powers, the
modulation optical rectification is expected to play a very limited role in the reaction
61
on the surface.
115
Here, the conducting graphene layer reduces the SERS enhancement
by a factor of approximately 5X. For Au nano-islands deposited on a continuous gold
film (50 nm thick), however, the SERS effect is completely quenched and there are no
detectible Raman features of the MBN molecules on these structures. As such, the Au
nano-island system and the graphene/Au nano-island system present a unique
opportunity to study in-situ SERS spectra at electrochemical interfaces. SERS spectra
of thiolated benzonitrile on different regions of the sample: Au nano-islands (AuNI)
deposited on a glass substrate both with and without an underlying monolayer of
graphene (Gr) with gold electrode (Au). Raman signals from different regions are
compared in Figure 3.3. As we can see, the Au nano-island with and without graphene
gives the most enhanced Raman Signals, which confirms the validity of this approach.
62
Figure 3.2. (a) Schematic diagram of the in situ SERS measurements using
a water immersion lens. (b) SERS spectra taken of thiolated benzonitrile on
Au nano-islands deposited on a glass substrate both with and without an
underlying monolayer of graphene. (c) Cross-sectional diagram of the Au
nano-islands/graphene on glass sample structure.
Gold nanoisland
on Graphene
Gold nanoisland
C-N Stretch
C-N Stretch
In water
With Graphene
Without Graphene
(b)
(c)
Au Au Au
63
500 1000 1500 2000 2500 3000
0
2000
4000
6000
8000
10000
AuNI/Gr
AuNI/Gr/Au
AuNI/Au
AuNI
Au
Counts
Raman Shift cm
-1
Figure 3.3. SERS spectra of thiolated benzonitrile on different regions of
the sample: Au nano-islands (AuNI) deposited on a glass substrate both
with and without an underlying monolayer of graphene (Gr) with gold
electrode (Au).
64
Figure 3.4. (a) Raman shift of the nitrile stretch and electric field strength
plotted as a function of the applied electrochemical potential. (b) Schematic
diagram illustrating the SERS measurement of thiolated-benzonitrile bound
to an Au electrode for Stark-shift spectroscopy. (c) Current-voltage plotted
as a function of the electrochemical potential, indicating the onset potential.
(d) Electrochemical impedance spectroscopy obtained capacitance-voltage
plot indicating the point of zero charge (PZC).
Solvent
E
65
3.4 Result and Discussion
Figure 3.4a shows the Stark shifts of the C-N stretch mode of MBN bound to
5nm Au nano-island films (shown in Figure 3.1) using surface enhanced Raman
spectroscopy. Figure 3.4a shows the voltage-induced shifts of the nitrile stretch mode
(around 2225 cm
-1
) plotted as a function of the applied electrochemical potential in
pure water. These Stark shifts can be converted to electric field using the E=∆ωC-N/0.36
((MV/cm)/cm
-1
) relation put forth by Boxer
116, 117
and Dawlaty
83, 84
, as indicated on the
right axis of the plot. These Stark-shifted fields can be correlated to the onset potential
and point of zero charge (PZC), obtained from the current-voltage (I-V) and
capacitance-voltage (C-V) characteristics, as shown in Figures 3.4c and 3.4d,
respectively. This is the first time these three datasets have been obtained
simultaneously on the same electrode surface. Taken together, they enable us to
separate the solvation (i.e., reaction) fields from the electrostatic fields produced by the
electrode. Here, the onset potential for the hydrogen evolution reaction (HER) lies
around 0V vs. NHE (-0.2V vs. Ag/AgCl) and the point of zero charge (PZC) occurs at
0.04V vs. NHE (-0.16V vs. Ag/AgCl), based on the capacitance-voltage profile. Based
on the local electric field strength at the PZC (Esolution=-0.8 MV/cm) and the onset
potential (Eonset=-2.0 MV/cm), we can separate the solution field from the reaction field,
respectively. Interestingly (or coincidentally), the solvation field -1.4MV/cm (blue-
shifted with respect to the C-N molecule in air) is quite close to the electrode field (-
1.2MV/cm) at the onset potential. Figure 3.5 is a waterfall plot showing the evolution
the entire Raman spectrum as a function of applied potential. In addition to changes in
66
the C-N mode position, the integrated intensity decreases and the line width increases
with positively applied potentials, which is in good agreement with the results
Hildebrandt’s work.
118
Other modes (other than C-N mode) also shift under the applied
potentials. However, these shifts are substantially smaller than the C-N stretch mode,
which is known to be particularly sensitive to externally applied electric fields.
Figure 3.5. Waterfall plot of the raw Raman spectra of CN-functionalized
gold nano-island electrode under various applied potentials (voltages vs.
NHE), as indicated in the legend.
67
Figure 3.6. (a) Frequency change as a function of the applied potential and
KCl ionic concentration. (b) Slice through this plot at a potential of 0.6V
vs. Ag/AgCl reference electrode.
We also studied the ionic strength dependence of these Stark-shifts. Here, we
repeated the voltage-dependent C-N stretch Stark-shift measurements in potassium
chloride (KCl) solutions with different concentrations (0, 0.01, 0.05, 0.1 mol/L). Figure
3.6 shows the electrochemical potential dependence of the Raman Shifts and
corresponding electric fields of the MBN molecule in various ionic strengths. In the
3D plot (Figure 3.6a), we observe larger Stark-shifts (∆ν) in solutions with higher ionic
strength. In pure water, the Stark-shift is around 2.5 cm
-1
when a potential of +0.6V vs.
Ag/AgCl is applied. However, in 0.1M KCl solution, we observe Stark-shifts (∆ν) up
to 7.6 cm
-1
at +0.6V vs. Ag/AgCl, which can be seen more clearly in Figure 3.6b,
depicting a slice of the data shown in Figure 3.6a at an applied potential of +0.6V vs.
Ag/AgCl. Empirically, we found that the MBN molecules were not stable at potentials
above +0.6V vs. Ag/AgCl.
(b) (a)
68
Figure 3.7. (a) Raman shift of the nitrile stretch and electric field strength
plotted as a function of the applied electrochemical potential in 0.01mol/L KCl
solution. (b) Current-voltage plotted as a function of the electrochemical
potential, indicating the onset potential. (c) Voltage drop across the solution
calculated from (b). (d) Raman shift of the nitrile stretch plotted as a function
of the corrected potential.
(c) (d)
69
Figures 3.7a and 3.7b show the original Stark-shift vs. voltage dependence and
current-voltage (I-V) dependence obtained in 0.01mol/L KCl solution. Here, a
deviation from the linear relation between Raman shift and applied potential can be
seen when the electrochemical current increases below approximately -0.3V vs.
Ag/AgCl. This is due to the finite voltage drop across the solution (i.e., V=IR). Figure
3.7c shows an estimate of the voltage drop across the solution (Vsolution) obtained by
multiplying the electrochemical current in Figure 3.7b by the finite resistance of the
electrolyte. Figure 3.7d plots the Raman shift as a function of the corrected voltage
(Vapplied - Vsolution), which shows a more consistent linear dependence of the Raman
Stark shift on electrode potential.
70
3.5 Conclusion
We observe substantial shifts in the Raman spectra of surface-bound MBN
molecules under various applied electrochemical potentials. The C-N stretch frequency
is blueshifted by as much as 7.6 cm
-1
under an applied potential of 0.6V (vs. NHE),
which correspond to local electric fields of 22 MV/cm. Under -0.8V applied potentials
(vs. NHE), the C-N stretch frequency is red shifted for only about 1 cm
-1
, which is
much less compared to the positive case. This corresponds to a regime in which the
electrochemical current increases exponentially in the hydrogen evolution process. We
believe that this saturation field corresponds to a substantial voltage drop across the
solution (V=IR) under finite currents in the hydrogen evolution reaction. Based on the
C-V measurement, the point of zero charge (PZC) occurs at 0.04V vs. NHE. And the I-
V measurements shows the onset potential for the HER lies around 0.2V vs. NHE.
Comparing the C-N stretch frequency in solution with that obtained in air, we can
calculate the solution field. Thus, the solution field is separated from the reaction field
(i.e., electrode field) by assessing the local electric field strength at the onset potential
and the PZC. At higher ion concentrations, similar electric field strengths and more E-
field saturation at the onset potential are observed, which is due to the relatively low
resistance of the solution (V=IR).
71
3.6 Acknowledgements
This research was supported by NSF Award No. 1708581 (H.S.), Army Research
Office ARO Award No. W911NF-14-1-0228 (Yi.W.), NSF CAREER Award No.
1454467 (Joel.P., Jahan,D.) and Air Force Office of Scientific Research Grant No.
FA9550-15-1-0184 (B.Z.).
72
Chapter 4: Measuring Local Electric Fields and Local Charge
Densities at Electrode Surfaces Using Graphene-Enhanced
Raman Spectroscopy (GERS)-based Stark-shifts
This chapter is similar to Shi et al.
119
published in ACS Appl. Mater. Interfaces.
4.1 Abstract
We report spectroscopic measurements of the local electric fields and local
charge densities at electrode surfaces using graphene-enhanced Raman spectroscopy
(GERS) based on the Stark-shifts of surface-bound molecules and the G band
frequency shift in graphene. Here, monolayer graphene is used as the working electrode
in a three-terminal potentiostat while Raman spectra are collected in situ under applied
electrochemical potentials using a water immersion lens. First, a thin layer (1Å) of
copper (II) phthalocyanine (CuPc) molecules are deposited on monolayer graphene by
thermal evaporation. GERS spectra are then taken in an aqueous solution as a function
of the applied electrochemical potential. The shifts in vibrational frequencies of the
graphene G band and CuPc are obtained simultaneously and correlated. The upshifts
in the G band Raman mode are used to determine the free carrier density in the
graphene sheet under these applied potentials. Of the three dominant peaks in the
Raman spectra of CuPc (i.e., 1531, 1450, and 1340 cm
-1
), only the 1531 cm
-1
peak
exhibits Stark-shifts and can, thus, be used to report the local electric field strength at
the electrode surface under electrochemical working conditions. Between applied
electrochemical potentials from -0.8V to 0.8V vs. NHE, the free carrier density in the
graphene electrode spans a range from -4×10
12
cm
-2
to 2×10
12
cm
-2
. Corresponding
73
Stark-shifts in the CuPc peak around 1531 cm
-1
are observed up to 1.0 cm
-1
over a range
of electric field strengths between -3.78×10
6
and 1.85×10
6
V/cm. Slightly larger Stark-
shifts are observed in a 1M KCl solution, compared to those observed in DI water, as
expected based on the higher ion concentration of the electrolyte. Based on our data,
we determine the Stark shift tuning rate to be 0.178 cm
-1
/ (10
6
V/cm), which is
relatively small due to the planar nature of the CuPc molecule, which largely lies
perpendicular to the electric field at this electrode surface. Computational simulations
using density functional theory (DFT) predict similar Stark shifts and provide a detailed
atomistic picture of the electric field-induced perturbations to the surface-bound CuPc
molecules.
4.2 Introduction
Raman spectroscopy is an important tool for measuring the vibrational
signatures of molecules for chemical detection and for characterizing nanoscale
materials like graphene, transition metal dichalcogenides, and carbon nanotubes.
14-19
The discovery of surface-enhanced Raman scattering (SERS) in the 1970s
3-7
was, in
fact, an early manifestation of nanoscale enhancement with reports of increases in the
Raman signal by as much as 10
14
and robust demonstrations of single-molecule
detection.
8-13
In SERS, there are two mechanisms of enhancement that are generally
accepted: 1) an electromagnetic mechanism (EM), which originates from the local
electromagnetic field enhancement on rough metal surfaces and metal nanoparticles,
and 2) a chemical enhancement mechanism (CE), which arises from the dynamic
74
transfer of charge through interaction between the target molecule and underlying
metal.
20-23
The EM enhancement typically arises when the incident light matches a
plasmon resonance within the metal nanostructure, which can produce fields as much
as 1000 times higher than the incident electromagnetic fields for Au and Ag
nanoparticles.
24-26
While the EM enhancement produces an overall uniform
enhancement of all the Raman modes, the chemical enhancement depends on the
symmetry of each vibrational mode, and therefore produces a vibrational-mode-
specific enhancement, which makes SERS spectra appear quite different from bulk
solution Raman spectra.
In 2010, monolayer graphene was shown to enhance the Raman signal of
adsorbed molecules by as much as 2 orders of magnitude, including phthalocyanine
(Pc), rhodamine 6G (R6G), protoporphyin IX (PPP), and crystal violet (CV).
61, 62
Since graphene is largely optically inert with only 2.3% absorption in the visible range
and plasmon resonances that lie in the infrared wavelength range,
63, 64
it is believed that
EM enhancement is minimal in this material system.
25, 26
For GERS, this distinction
is further supported by the fact that molecules with large GERS enhancement factors
can be correlated with their symmetry groups.
65
The Raman signals of molecules on
multilayer graphene show no enhancement and are even weaker than those molecules
on a Si/SiO2 substrate.
65,120,36
It should be noted that the underlying chemical
enhancement mechanism associated with this GERS phenomenon is still not fully
understood.
In our previous work, we measured Raman spectra at a graphene/water interface
75
as a function of electrochemical potential.
111
Large electrostatic fields exist at the
surface of these electrodes due to the double layer at this solid/liquid interface.
84
Various chemical reactions entail the transfer of charged or polarized reactants across
this interface. Electric fields can, therefore, have a significant effect on these reaction
mechanisms and catalysis, which are intimately related to the local electric field
strength. In a previous study by Dawlaty et al., sum frequency generation (SFG)
spectroscopy was used to explore the interfacial solvation effects on the C-N frequency
shift of 4-mercaptobenzonitrile (4-MBN) bound to gold electrode surfaces.
83, 84
Lian
and coworkers also used in situ SFG probes to characterize the electric fields at
solid/liquid interfaces.
121
In a more recent study, Shi et al. demonstrated the local
electric fields generated at graphene electrode surfaces can be obtained using surface
enhanced Raman scattering, which is a far more facile technique than SFG
spectroscopy.
80
In the work reported here, we explore the effect of applied electrochemical
potential and the associated electric fields on a GERS system using CuPc molecules.
This approach takes advantage of the unique ability of monolayer graphene to host
adsorbed molecules of copper phthalocyanine (CuPc) and provide
amplification/enhancement of its Raman signal via the GERS phenomena. The
monolayer graphene also serves as a highly-conductive electrode material, enabling us
to externally apply an electrochemical potential, while monitoring CuPc’s vibrational
modes in situ. Despite the sparse coverage of the surface adsorbed molecules, the
Raman spectra of the CuPc molecules can easily be observed using a water immersion
76
lens. In addition, the G band Raman shift of the graphene is used to report the charge
density on the electrode, which is proportional to the local electric field strength. This
general approach can be applied to a wide range of molecules that can be used to report
various aspects of charge transfer, local pH, and ion concentration.
4.3 Experimental Details
Monolayer graphene was grown by chemical vapor deposition (CVD) at 1000
o
C
in methane and H2 gas on copper foil at a reduced pressure of 1-1.5 Torr.
76
The copper
foil with graphene was then coated with a thin protection layer of PMMA-A6 by spin-
coating at 2000 rpm for 45s and then baking at 150
o
C for 5 minutes. Copper etchant
was used to etch away the copper from the bottom of the “sandwich” structure,
resulting in a graphene/PMMA film that is left floating on the surface of the liquid
copper etchant. This is then cleaned with 10% hydrochloric acid in DI water.
77
Prior to
transferring the graphene, two gold electrodes are deposited on an oxidized silicon
substrate (300nm SiO2) by electron-beam evaporation using a shadow mask to serve
as the target substrate. The PMMA-graphene layer was then scooped up on the target
substrate connecting both gold electrodes. The sample was then baked at 120
o
C for 5
minutes to improve adhesion. After this, the PMMA layer was removed by soaking in
acetone for 5 minutes. Copper (II) phthalocyanine (CuPc) molecules are then deposited
on the graphene by thermal evaporation of a 1Å nominal thickness, which provides
sub-monolayer coverage. Copper wires are attached to both gold electrodes enabling
this graphene layer to serve as a working electrode in a three-terminal potentiostat setup.
77
In order to monitor any potential electrochemical degradation of this electrode, we also
measure the in-plane resistance of the graphene, which usually lies in the range
between 1000 and 3000Ω. The gold contacts are then covered with epoxy so that they
are not in direct contact with the electrolyte. The active area of the graphene was about
0.25cm
2
after protection with the epoxy, as illustrated in Figure 4.1b.
Raman spectra of the graphene electrode were taken with 633nm and 532nm
wavelength excitation under applied electrochemical potentials in pure DI water using
a water immersion lens, as illustrated in Figure 4.1a. In order to protect the lens from
the solution, it was covered with a 13µm thick Teflon sheet (American Durafilm, Inc.).
A three terminal potentiostat (Gamry, Inc.) was used to apply a potential to the
graphene working electrode with respect to the reference electrode, as illustrated in
Figure 4.1a. A silver/silver chloride reference electrode and glassy carbon electrode
(SPI, Inc.) were used as the reference and counter electrodes, respectively. As a
comparison, we also performed the same measurements in a 1M KCl as solution.
GERS spectra were collected as a function of the applied electrochemical potential.
Since the CuPc molecules are not water soluble, they remain stably bound to the
graphene surface in solution during these in situ measurements. The Raman spectra of
the CuPc were collected with a 633nm wavelength laser, which is resonant with this
molecule’s absorption. The Raman spectra of the underlying graphene electrode were
collected with a 532nm wavelength laser, without being obscured by the CuPc peaks.
The shifts in the G band Raman frequency are then used to provide a measure of the
electrochemical doping in the graphene monolayer, as described in the section 4.5.
78
Figure 4.1. Schematic diagrams of the (a) three-terminal
photoelectrochemical cell using a water immersion lens to obtain Raman
Spectra in situ and (b) CuPc-coupled monolayer graphene electrode.
Gamry
Potentiostat
Ag/AgCl
Reference
Electrode
Glassy
carbon
Counter
Electrode
Graphene
Working
Electrode
Water
Immersion
Lens
633nm
Laser
V
A
DI
water
Si
Copper wire
300nm SiO
2
Graphene
Gold electrodes
covered by epoxy
CuPc
molecule
Water
immersion
lens
633nm
Excitation
light
(b)
79
4.4 Computational Methods
We performed computational simulations of the Stark shift tuning rate using the
Amsterdam Density Functional (ADF) program package.
122-124
The geometry
optimizations and frequency calculations were carried out using the Becke–Perdew
125,
126
(BP86) exchange-correlation functional with dispersion correction (DFT-D3-BJ) by
Grimme
127
and triple-ζ polarized (TZP) Slater-type basis. A hydrogen-terminated
graphene sheet was optimized with a large frozen core, and the sheet was forced to be
planer by specifying Cs symmetry. In vacuum, the CuPc was optimized with a large
frozen core. The optimized CuPc was then placed on the graphene sheet and was
allowed to relax while holding the graphene sheet fixed in electric fields in both plus
and minus z-directions of various field strengths (-0.001au, -0.0005au, 0au, 0.0005au,
0.001au, in which 1 au=5.14×10
9
V/cm). For each of the optimized geometries, the
frequencies of the CuPc vibrational modes were calculated using the Mobile Block
Hessian
128, 129
where the graphene sheet had its internal degrees of freedom removed.
4.5 Result and Discussion
The G band shift of the underlying graphene is plotted in Figure 4.2a as a function
of reference potential (the left y-axis in Figure 4.2a). Here, the free carrier
concentration is obtained from the G band Raman shift using following relations put
forth by Das Sarma and Berciaud: 𝑛𝑛 e = [(21∆ωG+75)/11.65]
2
•10
10
cm
-2
and 𝑛𝑛 h= [(-
18∆ωG-83)/11.65]
2
•10
10
cm
-2
, where 𝑛𝑛 e and 𝑛𝑛 h are the two-dimensional charge
densities of electrons and holes, respectively, and ∆ωG is the change in the G band
80
Raman frequency with respect to the charge neutrality point with units of cm
-1
.
43, 60
This 2D charge density on the graphene sheet (in electrons or holes per cm
2
) is plotted
as a function of the applied voltage on the right axis of Figure 4.2a.
49, 111
Here, we have
indicate the regions over which there are electrons and holes in the graphene. Under
the applied electrochemical potentials (-0.8V to 0.8V vs. NHE), the free carrier density
in the graphene electrode, obtained from the shifts in the G band frequency, spans a
range from -4×10
12
cm
-2
to 2×10
12
cm
-2
.
Figure 4.2b shows the vibrational Stark shift of the 1531 cm
-1
vibrational mode
of CuPc plotted as a function of the reference potential, which increases monotonically
with increasing potential. The righthand axis plots the local electric field calculated
from the free carrier concentration in the graphene, obtained from the graphene G band
Raman shift (the right y-axis in Figure 4.2a). We calculate the local electric field at the
electrode surface using the formula ε=e σ/2ϵ0 ϵr for an infinitely charged plane, where ε
is electric field strength, e is the charge of one electron, ϵr is the relative dielectric
constant, and ϵ0 is the permittivity of free space. We also assume that the electric field
is zero for the sample in air in order to establish the righthand y-axis in Figure 4.2b,
which spans a range from -3.78×10
6
to 1.85×10
6
V/cm. Here, the CuPc Raman mode
around 1531 cm
-1
exhibits Stark shifts up to 1.0 cm
-1
. From the data in Figures 4.2a and
4.2b, we can relate the Stark shift of the CuPc 1531 cm
-1
peak to the electric field, as
plotted in Figure 4.2c, which exhibits a Stark tuning rate of 0.178 cm
-1
/ (10
6
V/cm). It
is interesting to note that there is an inflection point near zero electric field in the CuPc
Stark shift vs. voltage plot taken in DI water (Figure 4.2c). This is likely related to the
81
differing abilities of the cations (H
+
) and anions (OH
-
) to produce a local electric field
in the water environment. However, to the 1st order approximation, we treat this as a
simple linear relation, which is consistent with our simulation results.
Interestingly, the 1531 cm
-1
peak is the only Raman mode that exhibits a Stark
shift. Figure 4.3 shows the Raman shift of three different CuPc peaks (1531 cm
-1
, 1450
cm
-1
and 1340cm
-1
) plotted as a function of the applied electrochemical potential. From
this data, it is clear that only the 1531 cm
-1
vibrational mode shows a Stark shift.
According to Basova, et al., these three modes are assigned to different bonds with B1g
symmetry: the 1531 cm
-1
peak corresponds to the Cα-Nβ band, 1450 cm
-1
is from Cβ-
Cβ, Cβ-Cγ-H, and 1340cm
-1
is from Cβ-Cβ, Cα-Cβ-Cβ, Cγ-Cδ and Cβ-Cγ.
130, 131
These
bonds are labeled in detail in Figure 4.4.
82
Figure 4.2. (a) CuPc Stark-shift (black) and the electric field (red) plotted
as a function of the applied potential measured in DI water. (b) Graphene
G band Raman shift (black) and the free carrier concentration (red) plotted
as a function of the applied potential. (c) CuPc 1531cm
-1
peak plotted as a
function of electric field at the monolayer graphene electrode surface.
)
)
)
83
Figure 4.3. Raman shift of different CuPc peaks plotted as a function of
the applied electrochemical voltage, measured in water (a)(b)(c) and in
1mol/L KCl solution (d)(e)(f).
(a)
(f)
84
Unlike 1450 cm
-1
and 1340cm
-1
, which correspond to bonds inside the macrocycle,
the contribution to the 1531 cm
-1
vibration is mainly given by the displacements of the
Cα-Nβ-C’α bridge bonds between the macrocycles.
41
It is possible that the bottom
macrocycle vibrational modes lie in the plane of the graphene electrode, which is
perpendicular to the electric field direction. However, one of the macrocycles might
rotate along one half of the bridge bond Cα-Nβ (C’α-Nβ) pulling the other half of the
bridge bond Nβ-C’α (Nβ-Cα) slightly out of the macrocycle plane, which can thereby
feel the effects of the applied electric fields. In our understanding, when
electrochemical potential is applied to the graphene-CuPc electrode, either cations (H
+
)
or anions (OH
-
) are accumulated on the surface. They will either attract or repulse the
electron-rich isoindole rings in CuPc. If the CuPc planar structure is not perfectly in
parallel with the graphene surface, adjacent isoindole rings will have different
interaction with the accumulated ions, which leads to the displacement and rotation of
the bridge bond. Steric effects may also be the reasons of the rotation of the bridge
bonds.
85
Figure 4.4. Cartesian coordinate axes orientation and bond lengths (in Å)
(replotted from Figure 1a of the work by Basova et al.
130, 131
) with the
assigned bonds that correspond to the three modes.
1450 cm
-1
86
Figure 4.5 shows the corresponding results of the CuPc-on-graphene electrode
obtained in a 1M KCl solution. From Figure 4.5a, we can also see upshifts (up to 12
cm
-1
) in the G band frequency under negative applied potentials, which is
approximately 50% larger the upshifts observed in DI water (8 cm
-1
). Here, the
carrier concentration on the sample surface reaches -8×10
12
cm
-2
under -0.8V vs. NHE,
which is almost two times as large as that obtained in DI water. However, under positive
applied potentials, G band upshift is comparable to that of the DI water and the carrier
concentration is even smaller. By correlating the CuPc Raman data to the electric field
derived from the graphene G band Raman data, we find that both the Stark shift and
electric field increase monotonically as a function of the reference potential in a 1M
KCl solution, as plotted in Figure 4.5b. The CuPc peak exhibits Stark shifts up to 1.56
cm
-1
, over a range of electric field strengths from -7.5×10
6
to 1.2×10
6
V/cm. Figure
4.5c shows the direct relation between the Stark shift of the CuPc 1531 cm
-1
peak and
the electric field in a 1M KCl solution, which exhibits a Stark tuning rate of 0.184 cm
-
1
/ (10
6
V/cm) . Figure 4.5d shows a comparison of the Stark shift difference of the
1531cm
-1
peak of CuPC plotted as a function reference potential in DI water and 1M
KCl solution with their fitted line, the slopes of which give the Stark tuning rate (STR)
in units of cm
-1
/ (10
6
V/cm). The difference between the Stark tuning rates in these
different electrolytes is less than 3%, which verifies the validity of this approach. Thus,
the local electric field can be directly related to Stark-shift of the 1531cm
-1
CuPc
Raman peaks.
From the computational simulation results, we observe that, for an electric field
87
Figure 4.5. (a) CuPc Stark-shift (black) and the electric field (red) plotted
as a function of the applied potential measured in 1M KCl solution. (b)
Graphene G band Raman shift (black) and the free carrier concentration
(red) plotted as a function of the applied potential measured in 1M KCl
solution. (c) CuPc 1531cm
-1
peak plotted as a function of electric field at
the graphene electrode surface. (d) CuPc Stark-shift change plotted as a
function of the applied voltage, measured in DI water (red) and in KCl
(black) plotted together with their linear fits, which give the Stark tuning
rates (STRs) in units of cm
-1
/(10
6
V/cm).
applied perpendicularly out of the surface, the outer phenyl rings moved away from the
surface while the inner carbon-nitrogen ring moved closer to the surface. For an electric
field applied perpendicularly into the surface, the outer phenyl rings moved closer to
c) (d)
STR=0.178
STR=0.184
88
the surface while the inner carbon-nitrogen ring moved away from the surface. In the
simulated Raman spectra of CuPc in vacuum, there were five Raman active modes near
the 3 peaks seen in the experimental spectra. The calculated Raman spectrum is plotted
in Figure 4.6a, exhibiting peaks at 1546cm
-1
, 1520cm
-1
, 1428cm
-1
, 1418cm
-1
and
1335cm
-1
. Atomic scale diagrams of the vibrational modes under applied electric field
are shown in Figures 4.8a-e, respectively. The 1418cm
-1
and 1428cm
-1
modes are the
symmetric and asymmetric version of phenyl ring vibration, while the 1520cm
-1
and
1546cm
-1
are the symmetric and asymmetric version of carbon-nitrogen stretching. Of
these modes, only the 1546cm
-1
mode exhibits an appreciable Stark shift. The Stark
shift of this mode is plotted as a function of the applied electric field in Figure 4.6c and
exhibits a Stark tuning rate of 0.141cm
-1
/ (10
6
V/cm). The same can be done for the
other modes, as shown in Figures 4.9b-e. The other 4 modes result in Stark tuning rates
of 0.027 cm
-1
/(10
6
V/cm), 0.016cm
-1
/(10
6
V/cm), 0.006cm
-1
/(10
6
V/cm), and 0.020cm
-
1
/(10
6
V/cm) for the 1520cm
-1
, 1428cm
-1
, 1418cm
-1
, and 1335cm
-1
modes, respectively,
which are at least one order of magnitude smaller than that of 1546cm
-1
mode. Full
experimental and calculated Raman spectra are plotted against each other in Figure
4.10. The simulation results of the Stark shifts calculated for the 1546cm
-1
mode are
consistent with our experimental measurements of the 1531cm
-1
mode. Here, it is
important to note that the simulated spectra correspond to the gas phase Raman spectra
of CuPc, Therefore, a direct comparison with the experimental GERS spectra is not
valid, since the Raman scattering selection rules are changed when the molecules are
bound to a surface. In addition, we only used the gas phase spectra to identify the
89
normal modes of the graphene-CuPc complex that were likely to be Raman active and
give support that the 1546cm
-1
mode corresponded to the mode that we found in
experiment. In our experiment, only the 1531cm
-1
mode shows an apparent Stark shift-
electric field dependence, and more importantly, the STR 0.141 is reasonably close to
the experimental data 0.178, which verifies the applicability of this approach. The raw
Raman spectra taken under 633nm and 532nm excitations are shown in Figure 4.11
and Figure 4.12, clearly showing the Stark shift of CuPc and the Raman shift of
graphene G band.
We would like to point out that our previous work using graphene alone to
monitor local electric fields via the relation E=σ/2ε, relied on the assumption that ε is
known, which is true to bulk water but not necessarily at the electrode surface.
111
It is,
therefore, important to develop other probes to report the local electric field at electrode
surfaces. Here, we chose CuPc for this study because it is a strongly GERS-enhanced
molecule.
25
However, its planar structure results in relatively small Stark tuning rates.
One of the key insights provided by the DFT calculations is a slight bond rotation out
of plane that makes this particular Raman mode Stark sensitive.
90
Figure 4.6. ADF-calculated (a) CuPc Raman spectra, (b) CuPc Raman
Shift 1546cm
-1
vibrational mode versus the applied electric field, (c) CuPc
1546cm
-1
peak plotted as a function of electric field with the Stark tuning
rate at the graphene electrode surface. Calculated in Prof. Lasse Jensen’s
group at Penn State University.
(a)
(b)
(c)
STR=0.141
= C
= N
= Cu
= Vibrate direction
= Rotate along
91
Figure 4.7. (a) Capacitance-voltage plot of the CuPc graphene-based
electrode measured in DI water, (b) carrier density obtained from the
capacitance-voltage data (i.e., Q=CV , in black) plotted together with the
carrier density obtained by Raman spectroscopy (red).
Figure 4.8. Atomic diagrams of the five Raman active vibrational modes
calculated for CuPc at (a) 1546cm
-1
, (b) 1520cm
-1
, (c) 1428cm
-1
, (d) 1418cm
-
1
, (e) 1335cm
-1
vibrational modes.
b)
(a) (c) (b)
(d) (e)
= C
= N
= Cu
= Vibrate direction
= Rotate along
92
Figure 4.9. CuPc peaks different vibrational modes: (a) 1546cm
-1
, (b)
1520cm
-1
, (c) 1428cm
-1
, (d) 1418cm
-1
, (e) 1335cm
-1
plotted as a function
of electric field with their calculated Stark tuning rates (STRs).
)
STR=0.141
STR=0.027
STR=0.016
STR=0.020
93
1200 1300 1400 1500 1600 1700
1531 cm
-1
1546 cm
-1
Raman Shift (cm
-1
)
Experimental Raman
Intensity
Simulated Raman
Intensity
Figure 4.10. Experimental (black) and Simulated (red) CuPc Raman
spectra in DI water
94
Figure 4.11. (a) All three main modes, (b) zoomed-in version of 1531cm
-1
mode (c) water fall plot of Stark-shift Raman spectra (633nm) of CuPc are
shown in DI water
(b)
(a)
(c)
95
Figure 4.12. (a) All three main modes, (b) zoomed-in version of G band mode
of graphene Raman spectra (532nm) in DI water
(b)
96
4.6 Conclusion
In summary, we report spectroscopic measurements of the local electric fields
and local charge densities at electrode surfaces using graphene-enhanced Raman
spectroscopy (GERS) based on the Stark-shifts of surface-bond molecules and the G
band frequency in monolayer graphene. GERS spectra are then taken in an aqueous
solution systematically as a function of the applied electrochemical potential. The
information from the shifts in both the graphene G band and CuPc vibrational
frequencies are obtained simultaneously and correlated. The carrier density in the
graphene sheet is used to determine the upshifts in the G band Raman mode under these
applied potentials. Of the three dominant peaks in the Raman spectra of CuPc (at 1531,
1450, and 1340cm
-1
), only the 1531 cm
-1
peak exhibits Stark-shifts and is used to report
the local electric field strength at the electrode surface under electrochemical working
conditions. Under the applied electrochemical potentials between -0.8V to 0.8V vs.
NHE, the free carrier density in the graphene electrode is obtained from the shifts in
the G band frequency and spans a range from -4×10
12
cm
-2
to 2×10
12
cm
-2
.
Corresponding Stark-shifts in the CuPc peak (around 1531 cm
-1
) are observed up to 1.0
cm
-1
over a range of electric field strengths between -3.78×10
6
to 1.85×10
6
V/cm. About
56% larger Stark-shifts and 54% larger local electric fields are observed in a 1M KCl
solution, compared to those observed in DI water, as expected based on the higher ion
concentration of the electrolyte. Despite these larger shifts and larger electric fields, the
Stark tuning rates observed in DI water and 1M KCl solutions agree within 3% of each
other, which verifies the validity of this approach. Density functional theory
97
calculations predict 5 Raman active modes for CuPc-on-graphene, only one of which
exhibits a Stark shift. The vibrational frequency and Stark tuning rate of this mode
agrees well with the experimental values.
4.7 Acknowledgements
This research was supported by the National Science Foundation (NSF) award
no. 1708581 (H.S.) and CBET-1512505 (J.C.), Air Force Office of Scientific Research
(AFOSR) grant no. FA9550-15-1-0184 (B.Z.), Army Research Office (ARO) award no.
W911NF-17-1-0325 (Z.C.), U.S. Department of Energy, Office of Science, Office of
Basic Energy Sciences, and award no. DE-SC0019322 (Y.W.). We would like to thank
Prof. Mark Thompson for a valuable discussion and use of deposition facilities. L.J.
and M.B. acknowledge the support from the National Science Foundation Grant CHE-
1707657 and NRT-1449785. Portions of this work were conducted with Advanced
Cyberinfrastructure computational resources provided by The Institute for Cyber-
Science at The Pennsylvania State University (https://ics.psu.edu/).
98
Chapter 5: Future Work and Conclusion
5.1 Monitoring Local Electric Field using Stark-shifts on Naphthyl-
Nitrile Functionalized Silicon
Using the local electric field-Stark-shifts approach, we can also conduct
spectroscopic measurements on electrodes made of other materials, such as
functionalized silicon surfaces. Dr. Neale’s group from National Renewable Energy
Laboratory (NREL) has been able to graft different materials on silicon surface.
Recently, they have successfully performed the synthesis and surface immobilization
of 2-cyano-3-[4-(dimethylamino) naphthalen-1-yl] prop-2-enoic acid (1) on p-type
silicon, as shown in Figure 5.1a. The synthesis procedures are described as followed.
In a procedure adapted from a Knoevenagel condensation published
previously,
132
4-dimethlyamino-1-napthaldehyde (Aldrich, 97%; 1 equiv., 5 mmol,
0.996 g) and cyanoacetic acid (Aldrich, 99%; 1 equiv., 5 mmol, 0.423 g) were dissolved
in a minimum amount of ethanol (HPLC grade, Pharmoco) and added to a round
bottom flask. Next, piperidine (Aldrich, ReagentPlus; 0.12 mol%, 0.6 mmol, 51 mg)
was added dropwise to the solution. The stirred solution was refluxed overnight.
Afterward, the ethanol was removed by distillation and the remaining red solid was
washed with cold methanol (J.T. Baker, Reagent Grade; 0 ºC) and placed under vacuum
for ~3 hrs. After purification by column chromatography using a silica stationary phase
and ethyl acetate (J.T. Baker, reagent grade) as an eluent, the product (1) was reacted
with the silicon surface. To prepare the silicon for functionalization, the wafer (TopSil,
Si (100), 1–5 Ω cm, 255–305 μm thick) was sonicated for 10 min. in acetone (electronic
99
grade, J.T. Baker), isopropanol (Fischer), and deionized water. The oxide-terminated
silicon was then etched in 2 M hydrofluoric acid (Sigma, 48% Reagent Grade) and
immediately brought into an Ar glovebox (0.1 ppm O2, < 0.5 ppm H2O). To chlorinate
the Si–H sites on the wafer, a 10 mL chlorobenzene (Aldrich, HPLC grade, distilled
over CaH2) solution saturated with phosphorus pentachloride (Fluka, >98%) and 5-10
mg of radical initiator (1,1′-Azobis(cyclohexanecarbonitrile), 98%) was prepared. The
wafer was soaked in the chlorination solution for 1 hour at 90 ºC in a sealed vial. After
thoroughly rinsing the wafer with chlorobenzene and toluene (J.T. Baker, reagent grade,
distilled over CaH2), the wafer was reacted with a saturated solution of (1) in dry
toluene at 100ºC overnight in a sealed vial. Afterward, the wafer was rinsed thoroughly
with and sonicated (10 min.) in dry toluene without exposure to air.
We have measured Raman spectra under three different wavelength (532, 633,
785nm) excitation lasers and no C-N Raman mode has been detected. Given the proof
that 5nm gold nano-island can largely enhance the Raman signal of the surface bounded
materials in Chapter 3, we have tried to deposit 5nm gold on the surface and C-N
Raman feature under 633nm excitation is shown in Figure 5.1b. Then we have
connected a copper wire to the back of the Ohmic contact layer on silicon and
encapsulate it with epoxy to make the electrode, which is shown in Figure 5.1c.
Then we performed in-situ Raman measurement of the electrode in a three-
terminal set-up under different electrochemical potentials, similar to Chapter 3. We
then plotted the waterfall plot of the raw Raman spectra of Naphthyl-Nitrile
functionalized Si in DI water (Figure 5.2a) and 0.1M KCl (Figure 5.2b) under various
100
Figure 5.1. (a) Molecular structure of Naphthyl-Nitrile grafted on silicon
surface, (b) Raman spectra and (c) 3D structure of surface functionalized silicon
electrode covered by 5nm gold nano-island.
Al Back
Ohmic
Contact
Copper
wire
Si
Naphthyl-Nitrile
Functionalization
Layer
5nm Gold
Island
structured layer
101
applied potentials. As we can see from the plots, the C-N stretch has red shifted under
negative electrochemical potentials. Also, we fit the C-N peak and we plotted the
Raman shift of the nitrile stretch as a function of the applied electrochemical potential
in DI water and 0.1M KCl solution (Figure 5.2c). The Raman shift change is larger in
0.1M KCl (8.55cm
-1
) than in DI water (4.71cm
-1
).
Next, we will try to figure out the local electric field on the C-N molecule on
the surface. However, unlike the case in Chapter 4 with graphene underlying the
material, we don’t have a direct measure of the charge. We can use EIS Capacitance-
voltage measurement to get the capacitance. Then we can integrate the capacitance
over the voltage range to get the charge. Thus, we can calculate the local electric field
at the electrode surface using the formula ε=e σ/2ϵ0 ϵr for an infinitely charged plane,
where ε is electric field strength, e is the charge of one electron, ϵr is the relative
dielectric constant, and ϵ0 is the permittivity of free space. Then we can compare the
Stark-tuning rate (STR) calculated from this method to the typical C-N STR value.
102
Figure 5.2. Waterfall plot of the raw Raman spectra of Naphthyl-Nitrile
functionalized Si in (a) DI water and (b) 0.1M KCl under various applied
potentials as indicated in the legend, (c) Raman shift of the nitrile stretch
plotted as a function of the applied electrochemical potential in DI water
and 0.1M KCl solution.
(a)
(b)
(c)
103
5.2 Monitoring Local pH using Benzoic Acid group on SERS Sample
Inspired by the method from Chapter 3, we also performed in-situ Raman spectra
on another SERS sample, which is p-Mercaptobenzoic acid bonded on 5nm gold nano-
island (Figure 5.3).
A 5nm (nominal thickness) Au film was then deposited on a glass slide using
electron-beam evaporation. A thin layer of p-Mercaptobenzoic acid (p-MBA)
molecules was deposited by soaking the sample in a 0.03 mol/L solution of p-MBA in
ethanol for 24 hours. Then we have connected a copper wire to the surface gold and
encapsulate it with epoxy to make the electrode.
Figure 5.3. Molecular structure of p-Mercaptobenzoic acid (p-MBA)
bounded to gold nano-island on glass.
Glass
Au Nano-island
S
O OH
S
O OH
S
O OH
104
Then we got the Raman spectra of the sample in air, DI water, 0.01M KOH
solution and 0.01M HCl solution, as are shown in Figure 5.4a. As we can see, we can
see the two major peaks of p-MBA under all conditions. But we can only see acidic
peak feature (COOH, 1697cm
-1
) in 0.01M HCl solution (pH=2) and the basic peak
feature (COO
-
, 1413cm
-1
) in 0.01M KOH solution (pH=12).
133
As we have discussed in Chapter 2, local pH can be measured using the in-situ
Raman probe. We want to see if we can change the local pH electrochemically which
can be also tested by the p-MBA method. So we performed in-situ Raman measurement
of the electrode in a three-terminal set-up under different electrochemical potentials.
And the result is in Figure 5.4b. Using DI water, apparent acidic peak and basic peak
feature can be shown under a relative negative (-0.6V) and positive (0V)
electrochemical potentials, respectively. And also, we can plot their Raman intensity as
a function of electrochemical potential by fitting the acidic peak feature (Figure 5.4c,
COOH, 1697cm
-1
) and the basic peak feature (Figure 5.4d, COO
-
, 1413cm
-1
). Under
the potentials from the negative to positive, the acidic peak intensity is decreasing
monotonically, while the basic peak intensity is increasing monotonically.
Next, we will compare these results with the local pH values got from the method
from Chapter 2. So, samples of 5nm AuNI with p-MBA samples should be made on
the monolayer of graphene and do the similar measurements. Thus, we can compare
the local pH values using different methods to see the validity of this probe.
105
Figure 5.4. (a) Raman spectra p-MBA on gold nano-island (AuNI) in air,
DI water, 0.01M KOH solution and 0.01M HCl solution, (b) In-situ Raman
spectra of p-MBA on AuNI under different electrochemical potentials as
indicated in the legend. (c) Acidic and (d) basic Raman peak fitted
intensities plotted as a function of electrochemical potentials.
COOH
COO
-
(b)
(d)
106
5.3 Discussion of the Approaches
Although the validity of the approaches that are discussed in this thesis have
been proved by different methods. There are definitely some limitations of them.
In Chapter 2 and 4, graphene is chosen as the substrate material, because of its
constant density of electronic states and linear dispersion relation, provides a
convenient conversion between the G band frequency and Fermi energy, and a
quadratic relationship between Fermi energy and doping concentration. We used this
approach to determine the local ionic concentration which provides more details about
the HER half reaction of the water splitting process. Although we are studying the HER
reaction, no hydrogen evolution was observed at negative potentials in our condition,
with the measured current absolute value remaining below 5µA. This is due to the low
concentration of H
+
ions and the relatively high overpotential required for HER on the
graphene electrode. What’s more, low negative potentials can avoid the formation of
H2 bubbles, which will destroy the graphene layer and affect our Raman measurements.
Also, positive potential can degrade the graphene rapidly due to oxidation reaction,
broadening Raman feature peaks. So, we have to choose the voltage range carefully
when using this probe, typically from -1V to 0.6V versus Ag/AgCl reference electrode.
In Chapter 3 and 4, the probes involve the use of SERS molecules (4-MBN and
CuPc). Similar to graphene, these molecules are easy to be destroyed when applied
positive potentials. Also, high negative potentials can lead to bubble formation. In my
experience, -0.8V to 0.4V versus Ag/AgCl reference electrode should be a reasonable
range for this probe.
107
Another concept I want to discuss is the charge neutrality point or the point of
zero charge. The point of zero charge is defined as the electrochemical potential above
which the electrode becomes positively charged and below which it becomes negative.
It is considered as a boundary condition between oxidation and reduction half
reactions.
111
However, the electric field at the electrode surface is not always zero at
the charge neutrality point, as would be the case in a solid-state junction. This can be
due to the complex nature of the water/electrode interface.
Last but not least, I’d like to point out that, in the thesis, graphene alone to
monitor local electric fields via the relation E=σ/2ε, relied on the assumption that ε is
known, which is true to bulk water but not necessarily at the electrode surface.
111
Also,
charge can be calculated form the capacitance-voltage plots from EIS measurement,
which can only be true under the infinity planar capacitor assumption. It is, therefore,
important to develop other probes to report the local electric field at electrode surfaces.
108
5.4 Conclusion
We have developed a novel approach to measure the local ion concentration at
graphene/water interfaces using Raman spectroscopy. Also, we are also able to use
surface enhanced Raman scattering (SERS) to measure the vibrational Stark shifts of
surface-bound thiolated-benzonitrile molecules bounded to an electrode surface during
hydrogen evolution reactions. Last but not least, we report spectroscopic measurements
of the local electric fields and local charge densities at electrode surfaces using
graphene-enhanced Raman spectroscopy (GERS) based on the Stark-shifts of surface-
bond molecules and the G band frequency in monolayer graphene. These approaches
shed light on some fundamental physics and chemistry in such experimental systems
and could potentially help to study the local pH, local ion concentration, reaction
mechanisms, and catalysis, which are crucial to electrochemical and
photoelectrochemical reactions, like CO2 reduction, hydrogen evolution reaction in
water splitting and so on.
109
Bibliography
1. Raman, C., A New Radiation. Indian Association for the Cultivation of Science, 2, 387 (1927).
2. Kerker, M., D.S. Wang and H. Chew, Surface enhanced Raman scattering (SERS) by molecules
adsorbed at spherical particles. Applied Optics, 19, 3373-3388 (1980).
3. Campion, A. and P. Kambhampati, Surface-enhanced Raman scattering. Chemical Society
Reviews, 27, 241-250 (1998).
4. Moskovits, M., Surface-enhanced Raman spectroscopy: a brief retrospective. Journal of Raman
Spectroscopy, 36, 485-496 (2005).
5. Otto, A., I. Mrozek, H. Grabhorn and W. Akemann, Surface-enhanced Raman scattering.
Journal of Physics: Condensed Matter, 4, 1143 (1992).
6. Efremov, E.V., F. Ariese and C. Gooijer, Achievements in resonance Raman spectroscopy:
Review of a technique with a distinct analytical chemistry potential. Analytica Chimica Acta,
606, 119-134 (2008).
7. Petry, R., M. Schmitt and J. Popp, Raman Spectroscopy—A Prospective Tool in the Life
Sciences. ChemPhysChem, 4, 14-30 (2003).
8. Nie, S. and S.R. Emory, Probing Single Molecules and Single Nanoparticles by Surface-
Enhanced Raman Scattering. Science, 275, 1102 (1997).
9. Kneipp, K., Y. Wang, H. Kneipp, L.T. Perelman, I. Itzkan, R. Dasari and M.S. Feld, Single
molecule detection using surface-enhanced Raman scattering (SERS). Physical Review Letters,
78, 1667-1670 (1997).
10. Dieringer, J.A., R.B. Lettan, K.A. Scheidt and R.P. Van Duyne, A frequency domain existence
proof of single-molecule surface-enhanced Raman Spectroscopy. Journal of the American
Chemical Society, 129, 16249-16256 (2007).
11. Jiang, K. Bosnick, M. Maillard and L. Brus, Single Molecule Raman Spectroscopy at the
Junctions of Large Ag Nanocrystals. The Journal of Physical Chemistry B, 107, 9964-9972
(2003).
12. Xu, H., E. J. Bjerneld, M. Käll and L. Börjesson, Spectroscopy of Single Hemoglobin Molecules
by Surface Enhanced Raman Scattering. V ol. 83. 1999. 4357-4360.
13. Michaels, A.M., Jiang and L. Brus, Ag Nanocrystal Junctions as the Site for Surface-Enhanced
Raman Scattering of Single Rhodamine 6G Molecules. The Journal of Physical Chemistry B,
104, 11965-11971 (2000).
14. Malard, L.M., M.A. Pimenta, G. Dresselhaus and M.S. Dresselhaus, Raman spectroscopy in
graphene. Physics Reports, 473, 51-87 (2009).
15. Dresselhaus, M.S., A. Jorio, M. Hofmann, G. Dresselhaus and R. Saito, Perspectives on Carbon
Nanotubes and Graphene Raman Spectroscopy. Nano Letters, 10, 751-758 (2010).
16. Dresselhaus, M.S., A. Jorio and R. Saito, Characterizing Graphene, Graphite, and Carbon
Nanotubes by Raman Spectroscopy. Annual Review of Condensed Matter Physics, 1, 89-108
(2010).
17. Dhall, R., M.R. Neupane, D. Wickramaratne, M. Mecklenburg, Z. Li, C. Moore, R.K. Lake and
S. Cronin, Direct Bandgap Transition in Many-Layer MoS2 by Plasma-Induced Layer
Decoupling. Advanced Materials, 27, 1573-1578 (2015).
18. Dhall, R., Z. Li, E. Kosmowska and S.B. Cronin, Charge neutral MoS2 field effect transistors
through oxygen plasma treatment. Journal of Applied Physics, 120, 195702 (2016).
110
19. Dhall, R., K. Seyler, Z. Li, D. Wickramaratne, M.R. Neupane, I. Chatzakis, E. Kosmowska,
R.K. Lake, X. Xu and S.B. Cronin, Strong Circularly Polarized Photoluminescence from
Multilayer MoS2 Through Plasma Driven Direct-Gap Transition. ACS Photonics, 3, 310-314
(2016).
20. Persson, B.N.J., K. Zhao and Z. Zhang, Persson, Zhao, and Zhang Reply. Physical Review
Letters, 97, 199702 (2006).
21. Le Ru, E.C. and P.G. Etchegoin, Comment on ``Chemical Contribution to Surface-Enhanced
Raman Scattering''. Physical Review Letters, 97, 199701 (2006).
22. Wu, D.-Y., X.-M. Liu, S. Duan, X. Xu, B. Ren, S.-H. Lin and Z.-Q. Tian, Chemical
Enhancement Effects in SERS Spectra: A Quantum Chemical Study of Pyridine Interacting
with Copper, Silver, Gold and Platinum Metals. The Journal of Physical Chemistry C, 112,
4195-4204 (2008).
23. Sun, M., S. Liu, M. Chen and H. Xu, Direct visual evidence for the chemical mechanism of
surface-enhanced resonance Raman scattering via charge transfer. Journal of Raman
Spectroscopy, 40, 137-143 (2009).
24. Pavaskar, P., J. Theiss and S.B. Cronin, Plasmonic hot spots: nanogap enhancement vs.
focusing effects from surrounding nanoparticles. Optics Express, 20, 14656-14662 (2012).
25. Pavaskar, P., I.K. Hsu, J. Theiss, W.H. Hung and S.B. Cronin, A microscopic study of strongly
plasmonic Au and Ag island thin films. Journal of Applied Physics, 113 (2013).
26. Liu, Z., W. Hou, P. Pavaskar, M. Aykol and S. Cronin, Plasmon Resonant Enhancement of
Photocatalytic Water Splitting Under Visible Illumination. Nano Letters, 11, 1111 (2011).
27. Kneipp, K., M. Moskovits and H. Kneipp, Surface-enhanced Raman scattering: physics and
applications. V ol. 103. 2006: Springer Science & Business Media.
28. Kukushkin, V.I., A.B. Van’kov and I.V. Kukushkin, Long-range manifestation of surface-
enhanced Raman scattering. JETP Letters, 98, 64-69 (2013).
29. Ding, S.-Y., E.-M. You, Z.-Q. Tian and M. Moskovits, Electromagnetic theories of surface-
enhanced Raman spectroscopy. Chemical Society Reviews, 46, 4042-4076 (2017).
30. Kumar, R., H. Zhou and S.B. Cronin, Surface-enhanced Raman spectroscopy and correlated
scanning electron microscopy of individual carbon nanotubes. Applied Physics Letters, 91,
223105 (2007).
31. Kneipp, K., H. Kneipp, P . Corio, S.D.M. Brown, K. Shafer, J. Motz, L.T. Perelman, E.B. Hanlon,
A. Marucci, G. Dresselhaus and M.S. Dresselhaus, Surface-enhanced and normal Stokes and
anti-Stokes Raman spectroscopy of single-walled carbon nanotubes. Physical Review Letters,
84, 3470-3473 (2000).
32. Harroun, S.G., T.J. Abraham, C. Prudhoe, Y. Zhang, P.J. Scammells, C.L. Brosseau, C.C. Pye
and R.D. Singer, Electrochemical surface-enhanced Raman spectroscopy (E-SERS) of novel
biodegradable ionic liquids. Physical Chemistry Chemical Physics, 15, 19205-19212 (2013).
33. Kudelski, A. and B. Pettinger, Fluctuations of surface-enhanced Raman spectra of CO
adsorbed on gold substrates. Chemical physics letters, 383, 76-79 (2004).
34. Sanchez, L.A., R.L. Birke and J.R. Lombardi, Surface-enhanced Raman scattering of
piperidine. The effect of electrode potential on intensity. The Journal of Physical Chemistry, 88,
1762-1766 (1984).
35. Abdelsalam, M., P.N. Bartlett, A.E. Russell, J.J. Baumberg, E.J. Calvo, N.G. Tognalli and A.
Fainstein, Quantitative electrochemical SERS of flavin at a structured silver surface. Langmuir,
111
24, 7018-7023 (2008).
36. Xu, H., L. Xie, H. Zhang and J. Zhang, Effect of Graphene Fermi Level on the Raman
Scattering Intensity of Molecules on Graphene. ACS Nano, 5, 5338-5344 (2011).
37. Sriram, S., M. Bhaskaran, S. Chen, S. Jayawardhana, P.R. Stoddart, J.Z. Liu, N.V. Medhekar,
K. Kalantar-Zadeh and A. Mitchell, Influence of Electric Field on SERS: Frequency Effects,
Intensity Changes, and Susceptible Bonds. Journal of the American Chemical Society, 134,
4646-4653 (2012).
38. Li, Y., P. Doak, L. Kronik, J.B. Neaton and D. Natelson, Voltage tuning of vibrational mode
energies in single-molecule junctions. Proceedings of the National Academy of Sciences, 111,
1282 (2014).
39. Li, Y ., P. Zolotavin, P. Doak, L. Kronik, J.B. Neaton and D. Natelson, Interplay of Bias-Driven
Charging and the Vibrational Stark Effect in Molecular Junctions. Nano Letters, 16, 1104-1109
(2016).
40. Lombardi, J.R. and R.L. Birke, A unified approach to surface-enhanced Raman spectroscopy.
The Journal of Physical Chemistry C, 112, 5605-5617 (2008).
41. Lombardi, J.R. and R.L. Birke, A unified view of surface-enhanced Raman scattering. Accounts
of chemical research, 42, 734-742 (2009).
42. Lombardi, J.R. and R.L. Birke, The theory of surface-enhanced Raman scattering. The Journal
of Chemical Physics, 136, 144704 (2012).
43. Das Sarma, S., S. Adam, E.H. Hwang and E. Rossi, Electronic transport in two-dimensional
graphene. Reviews of Modern Physics, 83, 407-470 (2011).
44. Geim, A.K. and A.H. MacDonald, Graphene: Exploring carbon flatland. Physics Today, 60,
35-41 (2007).
45. Kohn, W., Image of the Fermi Surface in the Vibration Spectrum of a Metal. Physical Review
Letters, 2, 393-394 (1959).
46. Piscanec, S., M. Lazzeri, F. Mauri, A.C. Ferrari and J. Robertson, Kohn Anomalies and
Electron-Phonon Interactions in Graphite. Physical Review Letters, 93, 185503 (2004).
47. Lazzeri, M., S. Piscanec, F. Mauri, A.C. Ferrari and J. Robertson, Phonon linewidths and
electron-phonon coupling in graphite and nanotubes. Physical Review B, 73, 155426 (2006).
48. Lazzeri, M. and F. Mauri, Nonadiabatic Kohn Anomaly in a Doped Graphene Monolayer.
Physical Review Letters, 97, 266407 (2006).
49. Yan, J., Y. Zhang, P. Kim and A. Pinczuk, Electric Field Effect Tuning of Electron-Phonon
Coupling in Graphene. Physical Review Letters, 98, 166802 (2007).
50. Pisana, S., M. Lazzeri, C. Casiraghi, K.S. Novoselov, A.K. Geim, A.C. Ferrari and F. Mauri,
Breakdown of the adiabatic Born-Oppenheimer approximation in graphene. Nat Mater, 6, 198-
201 (2007).
51. Yan, J., E.A. Henriksen, P. Kim and A. Pinczuk, Observation of Anomalous Phonon Softening
in Bilayer Graphene. Physical Review Letters, 101, 136804 (2008).
52. Araujo, P .T., D.L. Mafra, K. Sato, R. Saito, J. Kong and M.S. Dresselhaus, Phonon Self-Energy
Corrections to Nonzero Wave-Vector Phonon Modes in Single-Layer Graphene. Physical
Review Letters, 109, 046801 (2012).
53. Jung, N., N. Kim, S. Jockusch, N.J. Turro, P. Kim and L. Brus, Charge Transfer Chemical
Doping of Few Layer Graphenes: Charge Distribution and Band Gap Formation. Nano Letters,
9, 4133-4137 (2009).
112
54. Zhao, W., P. Tan, J. Zhang and J. Liu, Charge transfer and optical phonon mixing in few-layer
graphene chemically doped with sulfuric acid. Physical Review B, 82, 245423 (2010).
55. Howard, C.A., M.P.M. Dean and F. Withers, Phonons in potassium-doped graphene: The
effects of electron-phonon interactions, dimensionality, and adatom ordering. Physical Review
B, 84, 241404 (2011).
56. Chen, Z., P. Darancet, L. Wang, A.C. Crowther, Y . Gao, C.R. Dean, T. Taniguchi, K. Watanabe,
J. Hone, C.A. Marianetti and L.E. Brus, Physical Adsorption and Charge Transfer of Molecular
Br2 on Graphene. ACS Nano, 8, 2943-2950 (2014).
57. Lu, C., Q. Fu, S. Huang and J. Liu, Polymer Electrolyte-Gated Carbon Nanotube Field-Effect
Transistor. Nano Letters, 4, 623-627 (2004).
58. Kim, S.H., K. Hong, W. Xie, K.H. Lee, S. Zhang, T.P. Lodge and C.D. Frisbie, Electrolyte-
Gated Transistors for Organic and Printed Electronics. Advanced Materials, 25, 1822-1846
(2013).
59. DasA, PisanaS, ChakrabortyB, PiscanecS, S.K. Saha, U.V. Waghmare, K.S. Novoselov, H.R.
Krishnamurthy, A.K. Geim, A.C. Ferrari and A.K. Sood, Monitoring dopants by Raman
scattering in an electrochemically top-gated graphene transistor . Nat Nano, 3, 210-215 (2008).
60. Froehlicher, G. and S. Berciaud, Raman spectroscopy of electrochemically gated graphene
transistors: Geometrical capacitance, electron-phonon, electron-electron, and electron-defect
scattering. Physical Review B, 91, 205413 (2015).
61. Ling, X., L.M. Xie, Y . Fang, H. Xu, H.L. Zhang, J. Kong, M.S. Dresselhaus, J. Zhang and Z.F.
Liu, Can Graphene be used as a Substrate for Raman Enhancement? Nano Letters, 10, 553-
561 (2010).
62. Huang, S.X., X. Ling, L.B. Liang, Y . Song, W.J. Fang, J. Zhang, J. Kong, V . Meunier and M.S.
Dresselhaus, Molecular Selectivity of Graphene-Enhanced Raman Scattering. Nano Letters, 15,
2892-2901 (2015).
63. Brar, V .W., M.S. Jang, M. Sherrott, J.J. Lopez and H.A. Atwater, Highly Confined Tunable Mid-
Infrared Plasmonics in Graphene Nanoresonators. Nano Letters, 13, 2541-2547 (2013).
64. Yan, H., X. Li, B. Chandra, G. Tulevski, Y. Wu, M. Freitag, W. Zhu, P. Avouris and F. Xia,
Tunable infrared plasmonic devices using graphene/insulator stacks. Nature Nanotechnology,
7, 330 (2012).
65. Ling, X., L. Xie, Y. Fang, H. Xu, H. Zhang, J. Kong, M.S. Dresselhaus, J. Zhang and Z. Liu,
Can Graphene be used as a Substrate for Raman Enhancement? Nano Letters, 10, 553-561
(2010).
66. Shi, H., N. Poudel, B. Hou, L. Shen, J. Chen, A. V . Benderskii and S. B Cronin, Sensing Local
pH and Ion Concentration at Graphene Electrode Surfaces using In Situ Raman Spectroscopy.
2018.
67. Pinson, J. and F. Podvorica, Attachment of organic layers to conductive or semiconductive
surfaces by reduction of diazonium salts. Chemical Society Reviews, 34, 429-439 (2005).
68. Delamar, M., R. Hitmi, J. Pinson and J.M. Saveant, Covalent modification of carbon surfaces
by grafting of functionalized aryl radicals produced from electrochemical reduction of
diazonium salts. Journal of the American Chemical Society, 114, 5883-5884 (1992).
69. Suvarnaphaet, P. and S. Pechprasarn, Graphene-Based Materials for Biosensors: A Review.
Sensors, 17 (2017).
70. Sun, Y., M. Sun and D. Xie, 5 - Graphene Electronic Devices, in Graphene, H. Zhu, et al.,
113
Editors. 2018, Academic Press. p. 103-155.
71. Yang, F., M. Shtein and S.R. Forrest, Controlled growth of a molecular bulk heterojunction
photovoltaic cell. Nature Materials, 4, 37-41 (2005).
72. Liu, Z., Q. Liu, Y . Huang, Y . Ma, S. Yin, X. Zhang, W. Sun and Y . Chen, Organic Photovoltaic
Devices Based on a Novel Acceptor Material: Graphene. Advanced Materials, 20, 3924-3930
(2008).
73. Cheng, W.-Y., C.-C. Wang and S.-Y. Lu, Graphene aerogels as a highly efficient counter
electrode material for dye-sensitized solar cells. Carbon, 54, 291-299 (2013).
74. Bae, S., H. Kim, Y. Lee, X. Xu, J.-S. Park, Y. Zheng, J. Balakrishnan, T. Lei, H. Ri Kim, Y .I.
Song, Y.-J. Kim, K.S. Kim, B. Özyilmaz, J.-H. Ahn, B.H. Hong, and S. Iijima, Roll-to-roll
production of 30-inch graphene films for transparent electrodes. Nature Nanotechnology, 5,
574 (2010).
75. Yang, N., J. Zhai, D. Wang, Y. Chen and L. Jiang, Two-Dimensional Graphene Bridges
Enhanced Photoinduced Charge Transport in Dye-Sensitized Solar Cells. ACS Nano, 4, 887-
894 (2010).
76. Soo Min, K., H. Allen, L. Yi-Hsien, D. Mildred, P. Tomás, K. Ki Kang and K. Jing, The effect
of copper pre-cleaning on graphene synthesis. Nanotechnology, 24, 365602 (2013).
77. Chen, C.-C., C.-C. Chang, Z. Li, A.F.J. Levi and S.B. Cronin, Gate tunable graphene-silicon
Ohmic/Schottky contacts. Applied Physics Letters, 101, 223113 (2012).
78. Epstein, P.S., The Stark Effect from the Point of View of Schroedinger's Quantum Theory.
Physical Review, 28, 695-710 (1926).
79. Courtney, M., N. Spellmeyer, H. Jiao and D. Kleppner, Classical, semiclassical, and quantum
dynamics in the lithium Stark system. Physical Review A, 51, 3604-3620 (1995).
80. Shi, H., Z. Cai, J. Patrow, B. Zhao, Y. Wang, Y. Wang, A. Benderskii, J. Dawlaty and S.B.
Cronin, Monitoring Local Electric Fields at Electrode Surfaces Using Surface Enhanced
Raman Scattering-Based Stark-Shift Spectroscopy during Hydrogen Evolution Reactions. ACS
Applied Materials & Interfaces, 10, 33678-33683 (2018).
81. Morita, A., Theory of Sum Frequency Generation Spectroscopy. 2018: Springer Singapore.
82. Boyd, R.W., Nonlinear Optics: Edition 3. 2008, Elsevier. p. 14.
83. Patrow, J.G., S.A. Sorenson and J.M. Dawlaty, Direct Spectroscopic Measurement of
Interfacial Electric Fields near an Electrode under Polarizing or Current-Carrying Conditions.
Journal of Physical Chemistry C, 121, 11585-11592 (2017).
84. Sorenson, S.A., J.G. Patrow and J.M. Dawlaty, Solvation Reaction Field at the Interface
Measured by Vibrational Sum Frequency Generation Spectroscopy. Journal of the American
Chemical Society, 139, 2369-2378 (2017).
85. Oklejas, V., C. Sjostrom and J.M. Harris, SERS Detection of the Vibrational Stark Effect from
Nitrile-Terminated SAMs to Probe Electric Fields in the Diffuse Double-Layer. Journal of the
American Chemical Society, 124, 2408-2409 (2002).
86. Oklejas, V., C. Sjostrom and J.M. Harris, Surface-Enhanced Raman Scattering Based
Vibrational Stark Effect as a Spatial Probe of Interfacial Electric Fields in the Diffuse Double
Layer. The Journal of Physical Chemistry B, 107, 7788-7794 (2003).
87. Schkolnik, G., J. Salewski, D. Millo, I. Zebger, S. Franzen and P . Hildebrandt, Vibrational Stark
Effect of the Electric-Field Reporter 4-Mercaptobenzonitrile as a Tool for Investigating
Electrostatics at Electrode/SAM/Solution Interfaces. International Journal of Molecular
114
Sciences, 13 (2012).
88. Fujishima, A. and K. Honda, Electrochemical Photolysis of W ater at a Semiconductor Electrode.
Nature, 238, 37-38 (1972).
89. Cheng, X., Z. Shi, N. Glass, L. Zhang, J. Zhang, D. Song, Z.-S. Liu, H. Wang and J. Shen, A
review of PEM hydrogen fuel cell contamination: Impacts, mechanisms, and mitigation. Journal
of Power Sources, 165, 739-756 (2007).
90. Ogden, J.M., M.M. Steinbugler and T.G. Kreutz, A comparison of hydrogen, methanol and
gasoline as fuels for fuel cell vehicles: implications for vehicle design and infrastructure
development. Journal of Power Sources, 79, 143-168 (1999).
91. Qiu, J., G. Zeng, P. Pavaskar, Z. Li and S.B. Cronin, Plasmon-enhanced water splitting on
TiO2-passivated GaP photocatalysts. Physical Chemistry Chemical Physics, 16, 3115-3121
(2014).
92. Bard, A.J. and L.R. Faulkner, Electrochemical Methods: Fundamentals and Applications, 2nd
Edition. 2000: Wiley Textbooks.
93. Auinger, M., I. Katsounaros, J.C. Meier, S.O. Klemm, P.U. Biedermann, A.A. Topalov, M.
Rohwerder and K.J.J. Mayrhofer, Near-surface ion distribution and buffer effects during
electrochemical reactions. Physical Chemistry Chemical Physics, 13, 16384-16394 (2011).
94. Katsounaros, I., J.C. Meier, S.O. Klemm, A.A. Topalov, P .U. Biedermann, M. Auinger and K.J.J.
Mayrhofer, The effective surface pH during reactions at the solid-liquid interface.
Electrochemistry Communications, 13, 634-637 (2011).
95. Rossrucker, L., A. Samaniego, J.P. Grote, A.M. Mingers, C.A. Laska, N. Birbilis, G.S. Frankel
and K.J.J. Mayrhofer, The pH Dependence of Magnesium Dissolution and Hydrogen Evolution
during Anodic Polarization. Journal of the Electrochemical Society, 162, C333-C339 (2015).
96. Gupta, N., M. Gattrell and B. MacDougall, Calculation for the cathode surface concentrations
in the electrochemical reduction of CO2 in KHCO3 solutions. Journal of Applied
Electrochemistry, 36, 161-172 (2006).
97. Deligianni, H. and L.T. Romankiw, In situ surface pH measurement during electrolysis using a
rotating pH electrode. IBM Journal of Research and Development, 37, 85-95 (1993).
98. Romankiw, L.T., TECHNIQUE FOR MEASURING PH AT ELECTRODES DURING
ELECTROLYSIS. Journal of the Electrochemical Society, 117, C118 (1970).
99. Leenheer, A.J. and H.A. Atwater, Imaging Water-Splitting Electrocatalysts with pH-Sensing
Confocal Fluorescence Microscopy. Journal of The Electrochemical Society, 159, H752-H757
(2012).
100. Chen, C.C., C.C. Chang, Z. Li, A.F.J. Levi and S.B. Cronin, Gate tunable graphene-silicon
Ohmic/Schottky contacts. Applied Physics Letters, 101 (2012).
101. Bruna, M., A.K. Ott, M. Ijas, D. Yoon, U. Sassi and A.C. Ferrari, Doping Dependence of the
Raman Spectrum of Defected Graphene. Acs Nano, 8, 7432-7441 (2014).
102. Pisana, S., M. Lazzeri, C. Casiraghi, K.S. Novoselov, A.K. Geim, A.C. Ferrari and F. Mauri,
Breakdown of the adiabatic Born-Oppenheimer approximation in graphene. Nature Materials,
6, 198-201 (2007).
103. Sasaki, K., M. Y amamoto, S. Murakami, R. Saito, M.S. Dresselhaus, K. Takai, T. Mori, T. Enoki
and K. Wakabayashi, Kohn anomalies in graphene nanoribbons. Physical Review B, 80,
155450 (2009).
104. Sasaki, K., H. Farhat, R. Saito and M.S. Dresselhaus, Kohn anomaly in Raman spectroscopy of
115
single wall carbon nanotubes. Physica E-Low-Dimensional Systems & Nanostructures, 42,
2005-2015 (2010).
105. Yoon, D., D. Jeong, H.J. Lee, R. Saito, Y .W. Son, H.C. Lee and H. Cheong, Fano resonance in
Raman scattering of graphene. Carbon, 61, 373-378 (2013).
106. Binder, J., J.M. Urban, R. Stepniewski, W. Strupinski and A. Wysmolek, In situ Raman
spectroscopy of the graphene/water interface of a solution-gated field-effect transistor:
electron-phonon coupling and spectroelectrochemistry. Nanotechnology, 27, 045704 (2016).
107. Das, A., S. Pisana, B. Chakraborty, S. Piscanec, S.K. Saha, U.V. Waghmare, K.S. Novoselov,
H.R. Krishnamurthy, A.K. Geim, A.C. Ferrari and A.K. Sood, Monitoring dopants by Raman
scattering in an electrochemically top-gated graphene transistor. Nature Nanotechnology, 3,
210-215 (2008).
108. Grahame, D.C., Differential Capacity of Mercury in Aqueous Sodium Fluoride Solutions. I.
Effect of Concentration at 25°. Journal of the American Chemical Society, 76, 4819-4823
(1954).
109. Balents, L. and M.P .A. Fisher, Correlation effects in carbon nanotubes. Physical Review B, 55,
11973-11976 (1997).
110. Sato, N., Electrochemistry at Metal and Semiconductor Electrodes. 1998: Elsevier.
111. Shi, H., N. Poudel, B. Hou, L. Shen, J. Chen, A.V. Benderskii and S.B. Cronin, Sensing local
pH and ion concentration at graphene electrode surfaces using in situ Raman spectroscopy.
Nanoscale, 10, 2398-2403 (2018).
112. Humbert, C., B. Busson, C. Six, A. Gayral, M. Gruselle, F. Villain and A. Tadjeddine, Sum-
frequency generation as a vibrational and electronic probe of the electrochemical interface and
thin films. Journal of Electroanalytical Chemistry, 621, 314-321 (2008).
113. Banik, M., P .Z. El-Khoury, A. Nag, A. Rodriguez-Perez, N. Guarrottxena, G.C. Bazan and V .A.
Apkarian, Surface-Enhanced Raman Trajectories on a Nano-Dumbbell: Transition from Field
to Charge Transfer Plasmons as the Spheres Fuse. ACS Nano, 6, 10343-10354 (2012).
114. Kwasnieski, D.T., H. Wang and Z.D. Schultz, Alkyl-nitrile adlayers as probes of plasmonically
induced electric fields. Chemical Science, 6, 4484-4494 (2015).
115. Nelson, D.A. and Z.D. Schultz, Influence of Optically Rectified Electric Fields on the
Plasmonic Photocatalysis of 4-Nitrothiophenol and 4-Aminothiophenol to 4,4-
Dimercaptoazobenzene. The Journal of Physical Chemistry C, 122, 8581-8588 (2018).
116. Fried, S.D. and S.G. Boxer, Measuring Electric Fields and Noncovalent Interactions Using the
Vibrational Stark Effect. Accounts of Chemical Research, 48, 998-1006 (2015).
117. Bagchi, S., S.D. Fried and S.G. Boxer, A Solvatochromic Model Calibrates Nitriles’ Vibrational
Frequencies to Electrostatic Fields. Journal of the American Chemical Society, 134, 10373-
10376 (2012).
118. Schkolnik, G., J. Salewski, D. Millo, I. Zebger, S. Franzen and P . Hildebrandt, Vibrational Stark
Effect of the Electric-Field Reporter 4-Mercaptobenzonitrile as a Tool for Investigating
Electrostatics at Electrode/SAM/Solution Interfaces. International Journal of Molecular
Sciences, 13, 7466 (2012).
119. Shi, H., B. Zhao, J. Ma, M.J. Bronson Jr, Z. Cai, J. Chen, Y. Wang, M. Cronin, L. Jensen and
S.B. Cronin, Measuring Local Electric Fields and Local Charge Densities at Electrode
Surfaces Using Graphene-Enhanced Raman Spectroscopy (GERS)-Based Stark-Shifts. ACS
Applied Materials & Interfaces, (2019).
116
120. Kröger, I., B. Stadtmüller, C. Stadler, J. Ziroff, M. Kochler, A. Stahl, F. Pollinger, T.-L. Lee, J.
Zegenhagen, F. Reinert and C. Kumpf, Submonolayer growth of copper-phthalocyanine on
Ag(111). New Journal of Physics, 12, 083038 (2010).
121. Ge, A., P.E. Videla, G.L. Lee, B. Rudshteyn, J. Song, C.P. Kubiak, V.S. Batista and T. Lian,
Interfacial Structure and Electric Field Probed by in Situ Electrochemical Vibrational Stark
Effect Spectroscopy and Computational Modeling. The Journal of Physical Chemistry C, 121,
18674-18682 (2017).
122. te Velde, G., F.M. Bickelhaupt, E.J. Baerends, C. Fonseca Guerra, S.J.A. van Gisbergen, J.G.
Snijders and T. Ziegler, Chemistry with ADF . Journal of Computational Chemistry, 22, 931-967
(2001).
123. Fonseca Guerra, C., J.G. Snijders, G. te Velde and E.J. Baerends, Towards an order-N DFT
method. Theoretical Chemistry Accounts, 99, 391-403 (1998).
124. E.J. Baerends, T.Z., A.J. Atkins, J. Autschbach, O. Baseggio, D. Bashford, A. Bérces, F.M.
Bickelhaupt, C. Bo, P.M. Boerrigter, L. Cavallo, C. Daul, D.P. Chong, D.V. Chulhai, L. Deng,
R.M. Dickson, J.M. Dieterich, D.E. Ellis, M. van Faassen, L. Fan, T.H. Fischer, C. Fonseca
Guerra, M. Franchini, A. Ghysels, A. Giammona, S.J.A. van Gisbergen, A. Goez, A.W. Götz,
J.A. Groeneveld, O.V . Gritsenko, M. Grüning, S. Gusarov, F.E. Harris, P . van den Hoek, Z. Hu,
C.R. Jacob, H. Jacobsen, L. Jensen, L. Joubert, J.W. Kaminski, G. van Kessel, C. König, F.
Kootstra, A. Kovalenko, M.V. Krykunov, E. van Lenthe, D.A. McCormack, A. Michalak, M.
Mitoraj, S.M. Morton, J. Neugebauer, V.P. Nicu, L. Noodleman, V .P. Osinga, S. Patchkovskii,
M. Pavanello, C.A. Peeples, P .H.T. Philipsen, D. Post, C.C. Pye, H. Ramanantoanina, P . Ramos,
W. Ravenek, J.I. Rodríguez, P. Ros, R. Rüger, P.R.T. Schipper, D. Schlüns, H. van Schoot, G.
Schreckenbach, J.S. Seldenthuis, M. Seth, J.G. Snijders, M. Solà, M. Stener, M. Swart, D.
Swerhone, V. Tognetti, G. te Velde, P. Vernooijs, L. Versluis, L. Visscher, O. Visser, F. Wang,
T.A. Wesolowski, E.M. van Wezenbeek, G. Wiesenekker, S.K. Wolff, T.K. Woo, A.L. Y akovlev,
ADF2019. 2019: Theoretical Chemistry, Vrije Universiteit, Amsterdam, The Netherlands.
125. Becke, A.D., Density-functional exchange-energy approximation with correct asymptotic
behavior. Physical Review A, 38, 3098-3100 (1988).
126. Perdew, J.P., Density-functional approximation for the correlation energy of the
inhomogeneous electron gas. Physical Review B, 33, 8822-8824 (1986).
127. Grimme, S., S. Ehrlich and L. Goerigk, Effect of the damping function in dispersion corrected
density functional theory. Journal of Computational Chemistry, 32, 1456-1465 (2011).
128. Ghysels, A., D. Van Neck, V. Van Speybroeck, T. Verstraelen and M. Waroquier, Vibrational
modes in partially optimized molecular systems. The Journal of Chemical Physics, 126, 224102
(2007).
129. Ghysels, A., D. V an Neck and M. Waroquier, Cartesian formulation of the mobile block Hessian
approach to vibrational analysis in partially optimized systems. The Journal of Chemical
Physics, 127, 164108 (2007).
130. Basova, T.V., V.G. Kiselev, B.-E. Schuster, H. Peisert and T. Chassé, Experimental and
theoretical investigation of vibrational spectra of copper phthalocyanine: polarized single-
crystal Raman spectra, isotope effect and DFT calculations. Journal of Raman Spectroscopy,
40, 2080-2087 (2009).
131. Li, D., Z. Peng, L. Deng, Y . Shen and Y . Zhou, Theoretical studies on molecular structure and
vibrational spectra of copper phthalocyanine. Vibrational Spectroscopy, 39, 191-199 (2005).
117
132. Kroupa, D.M., M. Vörös, N.P. Brawand, N. Bronstein, B.W. McNichols, C.V. Castaneda, A.J.
Nozik, A. Sellinger, G. Galli and M.C. Beard, Optical Absorbance Enhancement in PbS
QD/Cinnamate Ligand Complexes. The Journal of Physical Chemistry Letters, 9, 3425-3433
(2018).
133. Michota, A. and J. Bukowska, Surface-enhanced Raman scattering (SERS) of 4-
mercaptobenzoic acid on silver and gold substrates. Journal of Raman Spectroscopy, 34, 21-
25 (2003).
Abstract (if available)
Abstract
This dissertation presents our efforts in monitoring local electric fields and charge densities at electrode surfaces using in-situ graphene/surface enhanced Raman spectroscopy (GERS/SERS) based stark-shifts. These approaches shed light on some fundamental physics and chemistry in such experimental systems and could potentially help to study the local pH, local ion concentration, reaction mechanisms, and catalysis, which are crucial to electrochemical and photoelectrochemical reactions, like CO₂ reduction, hydrogen evolution reaction in water splitting and so on. ❧ Chapter 1 provides introduction and background materials to aid in understanding the experiments presented in the dissertation. The chapter starts with a brief overview of the Raman Spectroscopy. Next, we briefly discuss about the mechanisms, the material systems and setups we will use in surface/graphene-enhanced Raman spectroscopy (SERS/GERS). Then, we also introduce the concept of vibrational Stark-shift spectroscopy and our experimental set-up. Then we describe the basics electrochemistry behind our setup. More detailed background materials for each topic are covered in the relevant chapter. ❧ Chapter 2 presents a new approach to probe the local ion concentration at graphene/water interfaces using in situ Raman spectroscopy. This result provide an accurate measurement of the charge on the graphene electrode under electrochemical working conditions. ❧ Chapter 3 shows our investigations of using SERS to measure the vibrational Stark shifts of surface-bound thiolated-benzonitrile molecules bounded to nano-structured gold electrode surface during hydrogen evolution reactions. ❧ Chapter 4 reports spectroscopic measurements of the local electric fields and local carrier densities at electrode surfaces using GERS based on the Stark-shifts of surface-bound molecules and the G band frequency shift in graphene. This general approach can be applied to a wide range of molecules that can be used to report various aspects of charge transfer, local pH, and ion concentration. ❧ Finally, in Chapter 5, future work related to in-situ graphene/surface enhanced Raman spectroscopy (GERS/SERS) approach is discussed. Also, there is a conclusion of this thesis.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Water partitioning between the bulk and an electrode surface
PDF
Elucidating local electric fields and ion adsorption using benzonitrile: vibrational spectroscopic studies at interfaces and in bulk
PDF
Nonlinear spectroscopy of chemical phenomena at interfaces: carrier trapping and local electric fields
PDF
Probing charge transfer and electric fields at the bulk and interface using vibrational spectroscopy
PDF
Asymmetric response and the reorientation of interfacial water with respect to applied electric fields
PDF
Targeted cell imaging by in-situ assembly and activation of hot spot SERS nanoprobes using split-fluorescent protein scaffolds
PDF
Artificial photosynthesis on titanium oxide passivated III-V semiconductors
PDF
Vibrational spectroscopy and molecular orientation at interfaces: correlating the structure of a material's surface to its properties
PDF
Non-linear surface spectroscopy of photoswitches annd photovoltaics
PDF
Electromagnetic modeling of plasmonic nanostructures
PDF
Electrochemical pathways for sustainable energy storage and energy conversion
Asset Metadata
Creator
Shi, Haotian
(author)
Core Title
Monitoring local electric fields, charge densities at electrode surfaces using in-situ graphene/surface enhanced Raman spectroscopy (GERS/SERS) based Stark-shifts
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Chemistry
Publication Date
04/21/2020
Defense Date
10/22/2019
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
graphene,OAI-PMH Harvest,photocatalytic reactions,Stark-shifts,surface enhanced Raman spectroscopy
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Cronin, Stephen (
committee chair
), Benderskii, Alexander (
committee member
), Wu, Wei (
committee member
)
Creator Email
haotians@usc.edu,shihaotian9018@gmail.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c89-283286
Unique identifier
UC11674760
Identifier
etd-ShiHaotian-8279.pdf (filename),usctheses-c89-283286 (legacy record id)
Legacy Identifier
etd-ShiHaotian-8279.pdf
Dmrecord
283286
Document Type
Dissertation
Rights
Shi, Haotian
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
graphene
photocatalytic reactions
Stark-shifts
surface enhanced Raman spectroscopy