University of Southern California Dissertations and Theses

Elucidating local electric fields and ion adsorption using benzonitrile: vibrational spectroscopic studies at interfaces and in bulk

(USC Thesis Other)

Elucidating local electric fields and ion adsorption using benzonitrile: vibrational spectroscopic studies at interfaces and in bulk

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Elucidating Local Electric Fields and Ion Adsorption Using Benzonitrile :
Vibrational Spectroscopic Studies at Interfaces and in Bulk
by
Anwesha Maitra
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(CHEMISTRY)
December 2023
Copyright 2023 Anwesha Maitra

To my parents,
Ma & Babai
(Mrs. Tanusree Maitra, Mr. Basu Dev Maitra)
ii

Acknowledgements
I would like to start by thanking Prof. Jahan M. Dawlaty, Professor, Chemistry Department at
University of Southern California, my mentor and advisor for the last five years. I don’t think I
could have asked for anything more than what I have received from him as a guide in every step
of my PhD journey. He has answered all my questions, even the most humble ones, with the same
patience and enthusiasm. His continuous support and positive approach towards every problem
have kept me motivated at various difficult phases of my research.
I would like to thank my defense committee members - Prof. Oleg Prezhdo, Prof. Susumu
Takahashi and Prof. Stephen Cronin. I was fortunate to have Prof. Moh-el-Naggar and Late
Prof. Sri Narayan in my screening and qualification examinations. Along with them, during my
graduate school, I have come across many eminent professors. Long discussions and inputs from
Prof. Alexander Benderskii, Prof. Stephen Bradforth, Prof. Curt Wittig and Prof. Hanna Reisler
during our Wednesday group-meetings amongst the spectroscopy groups have immensely helped
me in shaping my thought process and broaden my views in science. Besides the professors, I
am also thankful to my wonderful labmates - Sohini, Yi, Ryan, Anuj, Cindy, Matt, Sevan, Elliott,
Tirthick, Sean, Berk, Ken, Danny and Thien - who have made Dawlaty group a great place to work.
I would like to mention Michele Dea, Claudia Cortez and Diana Cervantes for all the administrative helps and suggestions. I will be always grateful to Airforce Office of Scientific Research
(AFOSR) and Research Corporation of Science Advancement for financial support throughout my
research.
The course of graduate school can’t be completed without friends, who have become a family
away from home. I am fortunate enough to get some of my friends as my colleagues too. I started
iii

my research under the mentorship of Sohini. She made the initial struggles of graduate school
much easier to handle. I would always cherish the scientific discussions I had with Anuj. Working
together in the lab has been a pleasure with him. I am blessed to have friends like Pratyusha (also
a collaborator in one of my projects), Swetha, Sraddha, Samprita, Shivalee, Sourav, Sai, Mythreyi
and Sharon.
Next integral part of my graduate school is my family. My parents, Mrs. Tanusree Maitra and
Mr. Basu Dev Maitra, have immensely supported my each and every decision in life and without
their blessings I couldn’t have come this far. Long chats with my sister, Dr. Trisha Maitra, have
been my stress-buster all through my life and especially, during these last five years. Shalmoli,
Udita and Rochita are with me since my high school days and they are just a phone call away to
share all my ups and downs. Another person who deserves a special mention is my significant
other, Mr. Debmalya Dhar, who has always believed in me, sometimes more than I could.
In the end, I would like to express my sincere gratitude towards University of Southern California, for offering me the training ground to pursue my dream as a science-practitioner.
Anwesha Maitra
September 2023
iv

Table of Contents
Dedication ii
Acknowledgements iii
List of Figures viii
Abstract xvii
Chapter 1: Introduction 1
1.1 Brief Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 List of Publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Chapter 2: Taming the Electrostatic Forces in Chemistry 5
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2 Three Views of Electrostatics: Forces, Fields, and Potentials . . . . . . . . . . . . 7
2.3 Measuring Fields on the molecular Scales . . . . . . . . . . . . . . . . . . . . . . 11
2.4 Electric fields Inside a Liquid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.5 Electric Fields in Enzymes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.6 Electric Fields in Electrochemistry . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.7 Future Prospects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Chapter 3: Electric Fields Influence Intramolecular Vibrational Energy Relaxation and
Linewidths 23
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.2 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.3 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Chapter 4: Electric Fields at Metal-Surfactant Interfaces: A Combined Vibrational
Spectroscopy and Capacitance Study 37
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
4.2 Experimental Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.3 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
4.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Chapter 5: Mechanistic Insights about Electrochemical Proton-Coupled Electron Transfer
Derived from a Vibrational Probe 62
v

5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
5.2 Experimental Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
5.3 Computational Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
5.4 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
5.4.1 Reactants outside the double layer (0 to −0.4 V) . . . . . . . . . . . . . . 70
5.4.2 Entry into the double layer and the initial reaction (−0.4 to −0.7 V) . . . . 72
5.4.3 Current onset and continued Stark shift of the reactants and products (−0.7
to −1.2 V) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
5.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Chapter 6: Distinguishing between the Electrostatic Effects and Explicit Ion Interactions
in a Stark Probe 84
6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
6.2 Experimental Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
6.3 Computational Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
6.4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
6.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Chapter 7: Measuring the Electric Fields of Cations Captured in Crown Ethers 101
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
7.2 Experimental Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
7.2.1 Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
7.2.2 Condensed Phase Experiment : FT-IR Spectroscopy . . . . . . . . . . . . . 103
7.2.3 Gas phase Experiment . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
7.3 Computational methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
7.4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
7.4.1 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Chapter 8: Direct Spectrosocpic Observation of Ultraslow Ion Desolvation at an Interface 115
8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
8.2 Experimental Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
8.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
8.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Chapter 9: Conclusion and ongoing work 129
References 131
Appendices 152
A Supporting Information: Electric Fields Influence Intramolecular Vibrational
Energy Relaxation and Linewidths . . . . . . . . . . . . . . . . . . . . . . . . . . 153
A.1 Input Geometry of benzonitrile . . . . . . . . . . . . . . . . . . . . . . . . 153
A.2 Computed Frequencies of Benzonitrile at different electric fields . . . . . . 155
A.3 Third Order Anharmonic Couplings with CN (mode 28) and all other
modes (mode 1 to mode 27) . . . . . . . . . . . . . . . . . . . . . . . . . 155
A.3.1 At Electric Field -50 MV/cm . . . . . . . . . . . . . . . . . . . . 155
vi

A.3.2 At Electric Field -33 MV/cm . . . . . . . . . . . . . . . . . . . . 158
A.3.3 At Electric Field -22 MV/cm . . . . . . . . . . . . . . . . . . . . 161
A.3.4 At Electric Field -11 MV/cm . . . . . . . . . . . . . . . . . . . . 164
A.3.5 At Electric Field 0 MV/cm . . . . . . . . . . . . . . . . . . . . . 167
A.3.6 At Electric Field 11 MV/cm . . . . . . . . . . . . . . . . . . . . 170
A.3.7 At Electric Field 22 MV/cm . . . . . . . . . . . . . . . . . . . . 173
A.3.8 At Electric Field 33 MV/cm . . . . . . . . . . . . . . . . . . . . 176
A.3.9 At Electric Field 50 MV/cm . . . . . . . . . . . . . . . . . . . . 179
B Supporting Information: Electric Fields at Metal-Surfactant Interfaces: A
Combined Vibrational Spectroscopy and Capacitance Study . . . . . . . . . . . . . 182
C Supporting Information: Mechanistic Insights about Electrochemical ProtonCoupled Electron Transfer Derived from a Vibrational Probes . . . . . . . . . . . . 186
C.1 Cyclic Voltammetry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
C.2 Open Circuit Potential (OCP) Measurement to Determine the Equilibrium
Potential of Triethyl Amine and Triethyl Ammonium Redox Couple . . . . 187
D Supporting Information: Distinguishing between the Electrostatic Effects and
Explicit Ion Interactions in a Stark Probe . . . . . . . . . . . . . . . . . . . . . . . 189
D.1 Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
D.1.1 NMR Characterization . . . . . . . . . . . . . . . . . . . . . . . 190
D.2 IR spectra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
D.2.1 Effect of Solvent . . . . . . . . . . . . . . . . . . . . . . . . . . 193
D.2.2 IR spectra data analysis . . . . . . . . . . . . . . . . . . . . . . 194
D.3 Computational Details . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
D.3.1 Input file for geometry optimization and frequency calculation
of Benzonitrile . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
D.3.2 Example input file for single point energy and frequency
calculation of Benzonitrile in presence of point charges . . . . . . 196
D.3.3 Example input file for single point energy and frequency
calculation of Benzonitrile in presence of Zn2+ and negative
point charge . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
D.3.4 Table S1 : Nitrile frequencies of benzonitrile at different
distances in the two computaional models . . . . . . . . . . . . 201
E Supporting Information: Measuring the Electric Fields of Cations Captured in
Crown Ethers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
E.1 Frequency shift vs radius plot . . . . . . . . . . . . . . . . . . . . . . . . . 202
vii

List of Figures
2.1 (Left) Charge imbalance in water molecule (H2O), showing that hydrogens are
slightly more positive than the oxygen. (Right) A thought experiment carrying a
positive charge in a solution of table salt (NaCl) in water. . . . . . . . . . . . . . . 7
2.2 Electric field vectors emanating from a positive charge +q. A negative charge will
feel an attractive force towards the +q and a positive charge will feel a repulsive
force from +q as depicted by the arrows. . . . . . . . . . . . . . . . . . . . . . . . 9
2.3 Electric field between Two charged metallic plates at constant potential difference
depend on the separation of the plates. The potential can be thought of as the
height difference between the peak of a mountain and the valley. The electric field
is analogous to the slope of the mountain. . . . . . . . . . . . . . . . . . . . . . . 11
2.4 (a) Absorption of -CN stretching frequency in benzonitrile. (b) Shift of -CN
stretching frequency, ∆ν, under the influence of external electric field, ⃗E . . . . . . 12
2.5 (a) Caffeine molecule, dark grey – Carbon, red – Oxygen, blue – nitrogen, grey –
hydrogen. The red arrow shows the overall charge separation over the body of the
molecule. (b) Water molecules oriented around the Caffeine. Only the electrostatic
energy of the water molecules is considered in this picture and hydrogen-bonding
is excluded. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.6 The presence of K+ ion helps in Zn2+ binding in active site of deacetylase.[8] . . . 15
2.7 The Presence of polar functional groups easily detaches Penicillin (PenG) , but
less polar amino acid facilitates Avibactam (Avb) binding.[9] . . . . . . . . . . . . 15
2.8 The vibrational frequencies of C=O and C–D of the molecule, CXF-D studied in
solvents and in enzyme reveal differences in the electrostatic environment between
the solvent cavity and the enzyme active site.[12] . . . . . . . . . . . . . . . . . . 16
2.9 Electric double layer consisting of the positively charged electrode and the
adjacent layer of negative charge. The potential φ(x) has an exponential decay
profile from the electrode. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
viii

2.10 (a)Two examples of surfactants – CTAB with a positively charged head group and
SDS with a negatively charged head group. (b) Measuring the frequency shift
and electric fields from a layer of surfactants at metal-electrolyte interface using
vibrational Stark-shift spectroscopy.[13] . . . . . . . . . . . . . . . . . . . . . . . 18
2.11 (a) An example of Crown ether. The ring can be made larger by adding repeating
units as indicated by n. (b) Change in the electrostatic environment near a reaction
center changes the reaction rate. [16] (c) Trapping ions near a catalytic center.[15]
(d) Benzonitrile tethered to crown ether for detailed understanding electrostatics
in the vicinity of the trapped ion.[17] . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.12 product formation of Diels Alder reaction is catalyzed by potential at STM tip.[19] 21
2.13 Dissociation of O-O bond by external electric field.[20] . . . . . . . . . . . . . . . 21
3.1 (a) Experimental nitrile spectra and (b) linewidths of surface tethered benzonitrile
from Sum Frequency Generation experiments. . . . . . . . . . . . . . . . . . . . 26
3.2 A cartoon of the computational setup used in this study (adapted from reference
[62]) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.3 (a) Cubic coupling in the absence of electric field between the CN stretch and
combinations of modes β and γ. (b) The third order Fermi resonance (TFR) in the
absence of field is a scaled version of panel (a) to account for resonance (equation
3). Since the quantities are symmetric, only half of the coupling and TFR matrices
are shown for ease of visualization. . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.4 (a) The vibrational manifold of benzonitrile. (b) The variation of several
representative modes with electric field. The Stark shift of the CN stretch and the
lower frequency modes will affect the resonance condition for energy transfer.(c)
Third order Fermi resonance (TFR) for the three important channels of vibrational
relaxation as a function of electric field. . . . . . . . . . . . . . . . . . . . . . . . 32
3.5 (a)Variation of coupling and resonance for energy relaxation from the CN stretch
to the combination of modes 10 and 25. The motion of the atoms in the respective
modes are shown below. The detuning increases significantly with more positive
fields, giving rise to narrower linewidths.(b)Variation of coupling and resonance
for energy relaxation from the CN stretch to two quanta of mode 20. The
motion of the atoms in the respective modes are shown below. There is an
significant decrease in coupling with more positive field, giving rise to narrower
linewidths.(c)Variation of coupling and resonance for energy relaxation from the
CN stretch to two quanta of mode 21. The motion of the atoms in the respective
modes are shown below. The coupling increases significantly with more positive
field. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.6 Computed nitrile linewidths as a function of electric field due to all channels (blue
line) and the dominant channel (orange). Consistent with experimental results, the
linewidth narrows with more positive field. . . . . . . . . . . . . . . . . . . . . . . 35
ix

3.7 Relative amplitude of the displacement of shared carbon atom C1 as a function of
electric field. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.1 Circuit diagram used for fitting impedance data . . . . . . . . . . . . . . . . . . . 41
4.2 Chemical structures and names of the surfactants with their acronyms used in the
study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
4.3 a) Nitrile stretch spectra of 4-MBN tethered on roughened Ag in 100 mM solutions
of surfactants b) Nitrile peak frequencies as a function of surfactant concentrations. 44
4.4 Cartoon representation of hydration shell around a) hydrophobic quaternary
ammonium head group, b) hydrophilic sulfate head group. Weakly solvated
ammonium head group can readily desolvate and approach closer to the nitriles.
Strongly solvated sulfate head group retains its solvation shell which limits how
close the headgroup can approach the nitriles. . . . . . . . . . . . . . . . . . . . . 45
4.5 Experimentally determined central frequencies of nitrile stretch vs surfactant
concentrations in logarithmic scale. . . . . . . . . . . . . . . . . . . . . . . . . . 46
4.6 Double layer capacitance as a function of surfactant concentration for OTAB
(8 carbon), DTAB (12 Carbon) and CTAB (16 Carbon) carried out with Ag
foil-MBN. Due to the negative potential, the cationic surfactants are drawn to the
surface at a much smaller bulk concentration (less than 0.5 mM) compared to the
SERS experiments (larger than 10 mM). This gives rise to smaller capacitance for
the longest chain. Further decrease of the capacitance is likely due to displacing
the high dielectric solvent water and mobile supporting electrolyte ions with long
chains hydrocarbons. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
4.7 Representative figure showing that the probe molecules will sense zero field if the
surfactants head group and counter ions are treated as continuum sheets of charge.
The blue and the red plates indicate sheets formed by the surfactant headgroups
and the counterions respectively. . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.8 Evolution of a) ζ function with z/d, b) electric field with z. ζ decays rapidly and
asymptotically approaches 2π. The electric field value approaches ∼ 9V/nm for a
lattice of charge density ρ = 1e/nm2
. . . . . . . . . . . . . . . . . . . . . . . . . 53
4.9 Field profile from two oppositely charged lattices held at a given distance from
each other. The resultant field obtained by summing fields from the two lattices
leaks outside the two sheets and has a non-zero value at shorter distance but zero
value at larger distance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
x

4.10 The positively charged surfactant head is represented as a blue circle with
hydrophobic tail attached to it and the negatively charged counterion is
represented as a red circle. Cartoon representation for three motifs : a) surfactant
head facing the nitriles with counterions lingering near the head group or outside
the hydrophobic tail, b) surfactant tail facing the nitriles with the head group
towards the bulk and counterions adjacent to the headgroups, and c) counterions
adjacent to the monolayer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
4.11 Comparison of NaBr and CTAB induced change of Cdl on (a) Ag foil and (b)
Ag foil-MBN (c) Proposed model for surfactant accumulation in absence and in
presence of 4-MBN on the Ag surface.he positively charged surfactant head is
represented as a blue circle with hydrophobic tail attached to it and the negatively
charged counterion is represented as a red circle. . . . . . . . . . . . . . . . . . . 57
4.12 Calculated field vs z for CTAB, DTAB and OTAB where head groups are facing
the nitriles. The field values for the three surfactants of different chain lengths
overlap one another. The zoomed in part illustrates that the field value along the
body of the probe molecule is inhomogeneous. . . . . . . . . . . . . . . . . . . . 60
5.1 . Schematic of the PCET reaction studied using surface enhanced Raman
spectroscopy. The tertiary ammonium proton donor with a benzonitrile vibrational
tag enters the electric double layer and discharges its proton to the electrode,
yielding the amine form. The nitrile probe stretch frequency reports on the details
of this process, including the entry of reactants into the EDL, the electric field
within the EDL, and adsorption of the products to the surface. . . . . . . . . . . . 66
5.2 . Nitrile stretch spectra of the reactant MAMBN-H+ (blue) and the product
MAMBN (red) (a) in pure form, (b) dissolved in DMSO, and (c) in presence of
Ag electrode at open circuit potential. The nitrile stretch peak of the product is
always red shifted compared to that of the reactant. . . . . . . . . . . . . . . . . . 69
xi

5.3 Potential-dependent evolution of the nitrile stretch spectra during the (a) forward
and (c) reverse scans. (b) Steady state current versus potential, where the
horizontal dotted line at −0.7 V indicates the onset potential for steady state
current. The vertical offset of the spectra in panels (a) and (c) is chosen such
that each spectrum is aligned with their corresponding potential value in (b). The
blue and red dashed lines in (a) and (c) correspond to the Stark shifts attributed
to MAMBN-H+ and MAMBN, respectively. In the forward scan, initially only
the nitrile peak corresponding to the reactant MAMBN-H+ is present. It does
not undergo Stark shift, indicating that the reactants are outside the EDL and
are not polarized by the electrode. At −0.4 V, a second peak around 2200 cm−1
corresponding to the product MAMBN begins to appear, indicating entry of the
reactants into the EDL and formation of products at a potential below the onset
of steady state current. Both peaks undergo Stark shift to lower frequencies with
more negative potential, indicating that they arise from species within the EDL
because they respond to the polarization exerted by the electrode. The reverse scan
shows that the reaction is reversible. Frequency shifts as a function of potential
are shown in Figure 5.4. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
5.4 Nitrile frequency of the reactant MAMBN-H+ (shown in blue) and product
MAMBN (shown in red) as a function of applied potential during forward
and reverse scans. The grey shaded area represents the region where a steady
electrochemical current is measured. . . . . . . . . . . . . . . . . . . . . . . . . 72
5.5 Interactions between Ag(100) and either product MAMBN or reactant MAMBNH+ molecules. (a,b) Product MAMBN (a) chemisorbed to and (b) physisorbed
near Ag(100). (c) MAMBN-H+ physisorbed near Ag(100). Charge density
difference isosurfaces are shown with respect to the electrode PZFC, where the
cyan and yellow indicate increases and decreases in electron density, respectively;
the isosurface level is 0.00025 e–
/Bohr3
. Ag atoms are grey, N atoms are light
blue, C atoms are brown, and H atoms are light pink. Further details are provided
in the SI of [175]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
5.6 Potential dependent nitrile peak in a control experiment, where only the product
MAMBN was present. The peak around 2227 cm–1 does not Stark shift over a
very large range of potential, indicating that the neutral MAMBN is not drawn
into the EDL easily. It is not until a relatively high negative potential of ∼ –1.0
V that a second Stark-shifting peak corresponding to the adsorbed MAMBN is
observed. This peak appears at a much less negative potential when MAMBN is
not drawn from the solution, but rather is produced by the Volmer reaction from
MAMBN-H+, as seen in Figure 5.3. . . . . . . . . . . . . . . . . . . . . . . . . 76
xii

5.7 Frequency shifts of the reactant and the product from theory (left panel)
(corresponding to the geometries in Figure 5a and 5c) and from experiment (right
panel). The reactant frequency shifts are plotted in blue and the product frequency
shifts are plotted in red. The range of potentials in the experimental work is
narrower due to the stability of the electrode. The slopes of the middle regions
of the experimental curves are qualitatively similar to the slopes of the calculated
curves. Note that the calculated frequencies at the most negative potentials are
less reliable because of charge transfer from the electrode to the molecule (see
Figure S8 of [175]) and likely proton transfer from MAMBN-H+. . . . . . . . . . 80
5.8 Schematic figure showing the sequence of events for the reaction based on our
experimental and computational data. (a) Below −0.4 V, the reactant MAMBNH+ molecules, which are solvated by DMSO, reside outside the EDL and do not
feel the polarizing influence of the electrode and therefore do not exhibit a Stark
shift. (b) Within the potential range of –0.4 V to −0.7 V, the reactant molecules
partially shed their bulk solvation shell and enter the EDL, as evidenced by their
Stark shift. Some of the reactant MAMBN-H+ molecules undergo the Volmer
reaction and form MAMBN, as evidenced by the largely red-shifted peak in our
data. Lack of electrochemical current in this range suggests that this is a stationary
population and does not turn over. Computational and experimental observations
suggest that MAMBN is chemisorbed on the surface. (c) At potentials more
negative of −0.7 V, the reaction can turn over yielding a steady state current.
Correspondingly, a steady state population of MAMBN-H+ and MAMBN that
continue to Stark shift with potential is observed. . . . . . . . . . . . . . . . . . . 81
6.1 Representative cartoon for metal cation complex with crown-ether functionalized
benzonitrile in two different locations giving rise to two different effects on
−CN vibrational probe. n= 1 for Benzo-15-Crown-5-CN (B15C5-CN); n= 2 for
Benzo-18-Crown-6-CN (B18C6-CN). . . . . . . . . . . . . . . . . . . . . . . . . 87
6.2 Metal ion complexes with Benzo-15-Crown-5-CN (small crown) and Benzo-18-
Crown-6-CN (big crown) only showing electrostatic interactions (top panel) and
showing both electrostatic and Lewis interaction (bottom panel). . . . . . . . . . . 90
6.3 Lewis interaction of metal ions with -CN of Benzonitrile. (a) Benzonitrile with
M+, (b) Benzonitrile with M2+. (c) Benzonitrile with M3+. Propylene carbonate
is used as solvent in all cases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
6.4 Shift in -CN frequency after coordinating with metal ions in benzonitrile (blue)
and in Benzo-15-crown-5-CN (grey), in propylene carbonate . . . . . . . . . . . . 93
6.5 (a) Effect of counterions on the electrostatic and Lewis peaks in the crown etherZn2+ complexes. Both the electrostatic and Lewis peaks for the smaller anion Cl−
show smaller blue shift compared to the organic anions. (b) For benzonitrile, no
electrostatic effect and no variation from the counter ion is observed. The Lewis
peak shows similar behavior towards anions as the crown system. . . . . . . . . . 95
xiii

6.6 (a) IR absorption spectra of Cu(OTf)2 and Benzo-15-crown-5-CN at different
concentration ratios. (b) IR absorption spectra of Zn(TFSI)2 and Benzo-15-crown5-CN at different concentration ratios. (c) Integrated peak area corresponding to
different species present for Cu2+ extracted from IR spectra shown in (a). (d)
Nitrile frequency at different concentrations of Cu2+ for Stark peak (lower panel)
and Lewis peak (upper panel) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
6.7 (a) Benzonitrile in the presence of Zn2+ and negative point charge to study both
the Lewis and electrostatic effects. (b) Benzonitrile in the presence of positive and
negative point charges to study the electrostatic effect. (c) The nitrile frequency
shift as a function of proximity of the charged species - Zn2+ (blue) and positive
point charge (orange) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
6.8 Results from Natural Bond Orbital (NBO) analysis. (a) Change in occupancy of
nitrogen lone pair orbital. (b) Change in occupancy of 4s orbital of Zn2+ as a
function of distance of positive charge from nitrogen of benzonitrile. . . . . . . . . 98
7.1 Synthsized Crown-ethers with benzonitrile vibrational probe (a)Benzo-15-Crown5-CN (B-15C5-CN) (b) Benzo-18-Crown-6-CN (B-18C6-CN) (c) cartoon of
expected blue-shift of CN frequency after capturing a metal ion (Mn+) . . . . . . . 103
7.2 (a) IR absorption spectra of cation complexed Benzo-15-Crown-5-CN and (b)
Benzo-18-Crown-6-CN, (c) Maximum frequency shifts obtained in Benzo-15-
Crown-5-CN (blue circles) and Benzo-18-Crown-6-CN (red squares) . . . . . . . . 106
7.3 CN frequency shift with charge of cation complexed with (a)Benzo-15-Crown-5-
CN (small crown) (b) Benzo-18-Crown-6-CN (big crown) . . . . . . . . . . . . . 108
7.4 (a) CH stretching frequencies in gas phase experiment for M2+ complexes
with Benzo-15-crown-5-CN (smaller crown) indicating different co-ordination
geometry, (b) -CN stretching frequencies for M2+ complexes with Benzo-15-
crown-5-CN (smaller crown) in condensed phase (top) and in gas phase (bottom)
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
7.5 CN frequency shift for different metal ions in vacuum and in solvent . . . . . . . . 109
7.6 relative displacement of the crown complexes M+ metals and M2+ metals . . . . . 110
7.7 CN frequncy shift with the sum of distances between the 5 oxygens and the metal
ion (in small crown (B15C5-CN) . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
7.8 CN frequncy shift with (a),(b) the distance of the metal ion from the benzonitrile
plane for +2 and +1 metals respectively (c),(d) the angle of the metal ion relative
to the plane of -CN group in the small crown (B15C5-CN) . . . . . . . . . . . . . 113
7.9 comparing metal ion vs point charge for (a) CN frequency shifts (b) partial charge
differences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
xiv

8.1 (a) Representative cartoon of Lewis coordination with Zn2+ in bulk propylene
carbonate solvent (b) IR spectra of Lewis coordination of -CN group in
benzonitrile with metal cations of different charge, reproduced from our previous
work. [17] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
8.2 Representative cartoon for partial breaking of solvation shell of metal cation to
form Lewis coordination with mercapto benzonitrile (MBN) tethered on gold surface118
8.3 Solvents used in this work along with water, nit shown.(a) Dimethyl Carbonate
(DMC), (b) Propylene Carbonate (PC), (c) Dimethyl Sulfoxide (DMSO). . . . . . 119
8.4 Nitrile stretch in saturated solution of Cu(OTf)2 (blue), ZnCl2 (red), In(OTf)3
(yellow). All spectra are nomarlized with respect to CN stratch at 2227 cm−1
.
Only Zn2+ shows the most intense Lewis peak centered ∼2265cm−1
. . . . . . . . 120
8.5 Adsorption of Zn2+ at the interface in presence of (a) Dimethyl carbonate (DMC),
(c) Propylene Carbonate (PC) and (d) Water. No Lewis coordination observed in
(b) dimethyl sulfoxide (DMSO). Blue spectra indicates nitrile peak from the MBN
monolayer in air and red spectra indicates nitrile peak at the interface after adding
saturated solution of ZnCl2 of each solvent . . . . . . . . . . . . . . . . . . . . . . 122
8.6 Time evolution of Lewis peak in (a) Propylene Carbonate, and in (b) water.
Integrated Peak area for Zn2+ bound CN peak and free CN peak with time in (c)
propylene carbonate and in (d) water. . . . . . . . . . . . . . . . . . . . . . . . . . 123
8.7 A representative fitting with the kinetic model, the data corresponds to the plot
shown in Figure 8.6(d) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
8.8 Spectral Evolution of free CN peak and Zn2+ coordinated CN peak upon dilution
with Dimethyl Carbonate (DMC) . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
8.9 Integrated peak areas of Zn2+ coordinated CN peak upon dilution with different
solvents, Dimethyl Carbonate (DMC), propylene carbonate (PC) , water and
mixture of PC and water. The peak areas are normalized with respect to their
respective area at the saturated concentration. The black dotted line indicates
when the Lewis peak intensity is half of the initial peak area. . . . . . . . . . . . . 126
8.10 A proposed framework to design electrolytes to get optimum ion adsorption at the
an electrode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
9.1 SEM images of etched Ag substrate . . . . . . . . . . . . . . . . . . . . . . . . . 182
9.2 Sheet of discrete charges giving rise to electric field at z . . . . . . . . . . . . . . . 183
9.3 ζ (z/d) for lattices in which the positions of the charges are randomly shifted from
that of the square lattice. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
9.4 Cartoon representation of a) OTAB, b) DTAB and, c) CTAB when the hydrophobic
tails are pointing towards the nitriles and are penetrating the SAM layer . . . . . . 185
xv

9.5 Calculated field for CTAB, DTAB and OTAB vs z where surfactant tails are
pointing toward the surface (Figure Figure 9.4), possibly intercalating with the
benzonitrile monolayer, keeping the charged head along with its counter ion far
away from the surface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
9.6 Cyclic voltammogram (10 mV s−1
) of polycrystalline silver recorded in DMSO
containing 100 mM MAMBN-H+ (reactant), and 300 mM tetrabutylammonium
hexafluorophosphate (supporting electrolyte) . . . . . . . . . . . . . . . . . . . . 186
9.7 (a) The supporting electrolyte. (b,c) The redox couple. . . . . . . . . . . . . . . . 187
9.8 The open circuit potential as a function of the log of concentration ratio of the
deprotonated and protonated species. . . . . . . . . . . . . . . . . . . . . . . . . . 188
9.9 Synthesis of B15C5-CN and B18C6-CN . . . . . . . . . . . . . . . . . . . . . . 189
9.10 1H-NMR of (1) in CDCl3 at 25 °C and 400 MHz. . . . . . . . . . . . . . . . . . . 190
9.11 1H-NMR of (2) in CDCl3 at 25 °C and 500 MHz. . . . . . . . . . . . . . . . . . . 191
9.12 1H-NMR of B15C5-CN in CDCl3 at 25 °C and 400 MHz. . . . . . . . . . . . . . 191
9.13 1H-NMR of B18C6-CN in CDCl3 at 25 °C and 400 MHz. . . . . . . . . . . . . . 192
9.14 Metal ion complexes with Benzo-15-crown-5-CN in propylene carbonate . . . . . 192
9.15 Metal ion complexes with Benzo-18-crown-6-CN in propylene carbonate . . . . . 193
9.16 Effect of different solvents on the Lewis coordination of the Zn2+ with the
benzonitrile probe. 400 mM of benzonitrile concentration, Zn2+ salt are added.
(a) In DCM the cation is not well-solvated, and therefore are easily coordinated
with the probe, giving rise to a large Lewis peak. (b) DMSO has a higher dielectric
constant and is a Lewis base, and therefore, can solvate the cation much better
than DCM. Therefore, the probe cannot coordinate with the cation and no Lewis
signature is observed. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
9.17 CN frequency shift vs ionic radius of metal ions (a) Benzo-15-crown-5 (b)
Benzo-18-crown-6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
xvi

Abstract
Chemistry on the microscopic scale is strongly dominated by the electrostatic influence of charges
such as ions and molecular dipoles. This makes measuring and modeling of electric fields on
molecular scales an important priority in understanding chemical and electrochemical reactions.
Vibrational Stark shift spectroscopy is a useful tool to measure electric fields on nanometer scale.
This thesis elaborates on the use of benzonitrile molecules as a Stark probe which shows shifts in
their vibrational frequencies under the influence of external electric fields. This frequency shift
as a reporter of local electrostatics has given us insight about a variety of important chemical
phenomena - effect of surfactants at metal-electrolyte interfaces, tracking a proton donor from bulk
to the electric double layer in the presence of electrochemical bias and measuring fields produced
from cations dissolved in solvent can produce local electric fields. These studies will inspire ideas
about designing electric fields to control reaction rates and product selectivity. There are some
nuances in using this Stark probe. The systematic change in linewidth of nitrile with external
electric field is from intramolecular vibrational energy transfer in benzonitrile, hence cannot be
attributed completely to report the local environment. Sometimes the specific bonding interactions
create ambiguity in quantifying local electrostatics from frequency shifts. We have shown that this
specific interaction effect can also be utilized as an indicator for interfacial ion adsorption. This
effect may be extended to study local pH, a primary factor in protein functions and in controlling
over-potential in many electrochemical reactions.
xvii

Chapter 1
Introduction
1.1 Brief Overview
Electrostatic forces, i.e., attraction and repulsion between charges, are fundamentally important
in dictating chemical and biological structure, reactivity, and function. The simplest example is
organization of solvent dipoles around a solute in response to the local electric field. In nature,
we see that enzymes need a particular electrostatic environment to activate its function. In many
organic reactions, electric field catalyzes product formation and the effect is most prominent when
the intermediate is polar. Another prime example is a photovoltaic material. Controlling the electric field at the depletion zone of photovolatics causes manipulation of the depletion zone width ,
which in turn influences in the performance of the materials.
In Chapter 2, we review basic electrostatics followed by contemporary examples of controlling electrostatic effects in chemistry. Our introduction of electrostatics is intended for a general
audience without focusing on the mathematical complexities and with more emphasis on the microscopic details of electric fields at the molecular length scales. It is followed by describing electrostatics in solvents, enzymes, and electrochemistry, each with examples from recent advances these
areas. Finally, we provide a perspective of the current and upcoming research in molecular-scale
control of electrostatic interactions. Our goal is to inspire new researchers in the fields of physics,
chemistry, and biology to tackle the major challenges of molecular-scale electrostatic design.
1

Measuring, tailoring, and controlling electrostatic forces at the molecular level is at the frontiers
of modern research. This thesis focuses on how we can measure and control local electric fields
using vibrational Stark shift spectroscopy. Benzonitrile has been used as a Stark probe in all
of the work presented in this thesis. Except Chapter 3, all other subsequent chapters demonstrate
how benzonitrile can provide useful insights in various complex chemical phenomena.
Before going into the applications of benzonitrile as a Stark probe, Chapter 3 talks about a
fundamental property of benzonitrile. In a previous work from our group [1] a systematic change
in linewidth of nitrile stretch was observed along with the nitrile frequency shift under the influence
of external electric field. From computational studies we have shown that the externally applied
field can change anharmonic couplings and frequency detunings between nitrile stretching frequency and other modes. Our results are relevant for controlling intramolecular vibrational energy
relaxation (IVR) with external or internal fields and for gaining a more complete interpretation of
linewidths of vibrational Stark probes.
Chapter 4 talks about how vibrational Stark shift spectroscopy can be used to understand the
interfacial electric field at metal-surfactant interfaces. Surfactants modulate interfacial processes.
In electrochemical CO2 reduction, cationic surfactants promote carbon product formation and suppress hydrogen evolution. The interfacial field produced by the surfactants affect the energetics of
electrochemical intermediates, mandating their detailed understanding. Using a nitrile as a probe,
we found that at open circuit potentials, cationic surfactants produce larger effective interfacial
fields (∼ -1.25 V/nm) when compared to anionic surfactants (∼ 0.4 V/nm). At high bulk surfactant concentration, the surface field reaches a terminal value suggesting formation of a full layer,
which is also supported by electrochemical impedance spectroscopy (EIS). Our results help in designing tailored surfactants for influencing electrochemical reactions via the interfacial field effect.
In Chapter 5 we have studied electrochemical hydrogen evolution reaction (HER) using a
proton donor having a benzonitrile Stark probe tethered onto it. HER is a classic example of
Proton-coupled electron transfer (PCET) reactions which is a fundamental step in a wide range
of electrochemical processes. We have identified three main stages for the progress of the PCET
2

reaction as a function of applied potential. First, we have determined the potential necessary for
desolvation of the reactants and their entry into the polarizing environment of the electric double layer (EDL). Second, we have observed the appearance of product peaks prior to the onset of
steady state electrochemical current, indicating formation of a stationary population of products
that does not turn over. Finally, more negative of the onset potential, the electrode attracts additional reactants, displacing the stationary products and enabling steady state current. This work
shows that the integration of a vibrational Stark shift probe with a proton donor provides critical
insight into the interplay between interfacial electrostatics and heterogeneous chemical reactions.
Such insights cannot be obtained from electrochemical measurement alone.
Next, we delved into engineering electric field in a microscopic scale. Chapter 6 and 7 talk
about electric fields from ions captured in a crown ether functionalized with benzonitrile Stark
probe. Chapter 6 mostly focuses on the nuances of using benzonitrile Stark probe as the interpretation of frequency shift often gets complicated due to the specific interactions of the probe, such
as hydrogen bonding and Lewis bonding. Therefore, it is important to distinguish between the pure
electrostatic response and the response due to such specific interactions. The molecular system we
have used in this work is sensitive to both the Stark effect from a single ion, discussed in details
in Chapter 7 and the explicit Lewis bonding of ions with the probe, discussed in Chapter 6. The
crown captures cations of various charges, and the electric field from the ions is sensed by the
benzonitrile probe. Additionally, the lone pair of the benzonitrile can engage in Lewis interactions
with some of the ions by donating partial charge density to the ions. Our system exhibits both of
these effects and therefore is a suitable test bed for distinguishing between the pure electrostatic
and the Lewis interactions.
Chapter 8 is based on the results from Chapter 7. We found a few metal ions show strong
Lewis interaction with benzonitrile. We translated this property of benzonitrile from bulk to the
interface and come up with the idea of using benzonitrile as an indicator of adsorbed metal ions at
electrode-electrolyte interface. Chapter 8 is still an ongoing work and the analysis of the results
are not completed yet.
3

1.2 List of Publications
1. Direct Spectrosocpic Observation of Ultraslow Ion Desolvation at an Interface. Anwesha
Maitra, Anuj Pennathur, Elliott Chiang, Jahan M. Dawlaty (in preparation)
2. Taming the Electrostatic Forces in Chemistry. Anwesha Maitra, Jahan M. Dawlaty (submitted)
3. Measuring the Electric Fields of Cations Captured in Crown Ethers. Anwesha Maitra,
Pratyusha Das, A. Mohammed, William R. Lake, Barry. C. Thompson, Mark Johnson,
Sharon Hammes-Schiffer, Jahan M. Dawlaty (in preparation)
4. Distinguishing between the Electrostatic Effects and Explicit Ion Interactions in a Benzonitrile Probe. J. Phys. Chem. B, 127, 2511-2520 (2023), Anwesha Maitra, Pratyusha Das,
Barry C. Thompson, Jahan M. Dawlaty Link
5. Electric Fields Influence Intramolecular Vibrational Energy Relaxation and Linewidths. J.
Phys. Chem. Lett., 12, 7818-7825 (2021), Anwesha Maitra, Sohini Sarkar, David Leitner,
Jahan M. Dawlaty Link
6. Advances in Vibrational Stark Shift Spectroscopy for Measuring Interfacial Electric Fields.
Chapter 10, pp 199-224, ACS Symposium Series, Vol. 1398 (2021), Sohini Sarkar, Cindy
Tseng, Anwesha Maitra, Matthew J. Voegtle, Jahan M. Dawlaty Link
7. Mechanistic Insights about Electrochemical Proton-Coupled Electron Transfer Derived from
a Vibrational Probe. J. Am. Chem. Soc., 143, 8381–8390 (2021), Sohini Sarkar, Anwesha Maitra, William R. Lake, Robert E. Warburton, Sharon Hammes-Schiffer, Jahan M.
Dawlaty Link
8. Electric Fields at Metal-Surfactant Interfaces: A Combined Vibrational Spectroscopy and
Capacitance Study. J. Phys. Chem. B, 124, 1311-1321 (2020), Sohini Sarkar, Anwesha
Maitra, Soumyodip Banerjee, V. Sara Thoi, Jahan M. Dawlaty Link
4

Chapter 2
Taming the Electrostatic Forces in Chemistry
2.1 Introduction
There are four fundamental forces in nature: gravitational, electromagnetic, nuclear weak, and nuclear strong forces. The last two are in the realm of the size of the nuclei of atoms, and do not have
direct consequences for chemistry or biology. Of the first two, we are most well-accustomed to
gravity that we experience on a daily basis. However, gravity is far weaker than the electromagnetic
forces. The only reason it becomes apparent is the huge mass of the earth that exerts gravitational
attractions on earth-bound objects. For comparison, the gravitational attraction between two nuclei
of hydrogen atoms (protons) due to their masses is 1037 (that is ten trillion trillion trillion) times
smaller than the electrostatic repulsion between them due to their opposite charges. Therefore, for
most of chemistry and biology, which operates on the length scales of molecules, the dominant
force is electrostatics. The reason we, for the most part, do not feel significant electrostatic forces
in our daily life is because most objects are nearly neutral, containing an equal quantity of positive
and negative charges overall. Not surprisingly, any macroscopic imbalance of charges results into
strong forces, that will try to rebalance the charge disparity back to zero. Such imbalances on the
macroscopic scale are known as electric potential and is the subject of physics and electrical engineering. Its harnessing is the basis of a large number of modern technologies. However, if we zoom
in on the nanometer or molecular scale (10−9 m or one billionth of a meter), the electrostatic forces
are not in complete balance. Molecules obey the rules of quantum mechanics, which necessitate
5

that all matter behave like waves. Electrons, which are the lighter particles and carry a negative
charge have more wave-like behavior. They make the outer layers of atoms and molecules, with
the positively charged nuclei tucked in tightly inside them in the form of much smaller waves.
Rules of quantum mechanics prevent the electrons from “falling” into the nuclei, even though they
attract each other strongly. The quantum rules dictate the charge distribution within a molecule,
which often is inhomogeneous. Due to this granularity of matter on the molecular scale, charges
do not cancel each other down at the microscopic scale. These microscopic variations are at the
core of a vast number of chemical and biological processes.
Imagine yourself as a being smaller than the size of a molecule that meanders amongst the
molecules in a solution of table salt dissolved in water ( Figure 2.1). For the sake of argument, also
assume that you carry a small positive charge with you. Each water molecule is neutral overall,
but one end of it is slightly negative compared to the other (see Figure 2.1). The oxygen atom
pulls electrons towards itself more than the hydrogens, thereby giving the molecule this electrical
asymmetry, which is also known as a dipole. Since in this thought experiment, you are carrying a
positive charge, if you approach the water molecule from the oxygen side, you will feel a strong
attraction, but if you approach it from the hydrogen side, you will feel a strong repulsion. Similarly,
the table salt in water breaks into positive (sodium) and negative (chlorin) ions. If we approach an
ion, we will feel strongly repelled from or attracted to them. Note that such forces do not manifest
themselves on a macroscopic scale, i.e., a cup of water will not exert net repulsion or attraction on
anything. However, charged molecules or atom (ions) and asymmetry of charges (dipoles) on the
molecular scale will exert forces on similar-sized charged entities on the microscopic scales.
Furthermore, thermal motion constantly moves molecules around. In gases and liquids molecules
may move about, as well as rotate and vibrate, while in solids atoms mostly vibrate around fixed
positions. Consequently, the electrostatic forces between molecules fluctuate and change. Therefore, on the macroscopic scale the world may seem calm and quiet, but on the molecular scale it is
teeming with fantastically strong forces between charges, pushing and pulling on the constituents
of matter. To make it more interesting, the constituents of matter, especially the electrons, behave
6

Figure 2.1: (Left) Charge imbalance in water molecule (H2O), showing that hydrogens are slightly
more positive than the oxygen. (Right) A thought experiment carrying a positive charge in a
solution of table salt (NaCl) in water.
like standing waves and bend, bulge, slosh, or snap off and move in response to these forces. Such
forces, not surprisingly, are at the heart of chemical and biological reactivity. They contribute to
creation or breaking of chemical bonds, contributing to the stability of biomolecules such as DNA,
and function of enzymes. Controlling and directing electrostatic forces for driving chemical reactions is a modern frontier of chemistry, and the purpose of this treatise is to give the reader a
cursory overview of the fundamentals in this area, the importance of electric forces for catalysis
and enzymatic reactions, along with some recent developments in tuning and controlling them for
tailored purposes.
2.2 Three Views of Electrostatics: Forces, Fields, and Potentials
The first record of static electricity is found in the work of Greek philosopher Thales of Miletus
in 600 BC. However, it was not until the 18th century that their origin was ascribed as attraction
between two opposite charges and repulsion between two like charges. Electrons are negatively
charged, and protons have an equal but opposite positive charge. Most objects are neutral because
they contain the same number of electrons and protons. An imbalance of electrons and protons
endows an object with net charge. The fundamental law for the attraction or repulsion between
charges is known as the Coulomb’s law:
7

⃗F =
1
4πε0
Qq
|r|
2
rˆ (2.1)
Where is the force between charges Q and q that are separated by distance r in vacuum. The
indicates that the force is directed along a straight line between the two charges, and 0, also known
as the permittivity of free space, is a constant of nature (8.854 ×10−12 F m−1
) which is essentially
a scaling factor that converts between charges and the mechanical force. The unit of charge is
Coulomb (C). Two objects each carrying 1 C of charge and held at a distance of 1 m will repel
each other with a force of 8.988×109 N. For comparison, a mass of 1 Kg is attracted towards the
earth with the force of about 9.8 N. However, note that the molecular scale for charge is different.
The charge of an electron is much smaller ( e =-1.602×10−19 C). An electron and a proton held at
a distance of 1 nm will exert a force of 1.44×10−10 N upon each other. This may seem small but
note that it also acts on extremely small entities. In practice, the above law is used in a modified
and more convenient way.
⃗F = ⃗Eq (2.2)
This is nothing but restating equation (1), with the electric field defined as ⃗E =
1
4πε0
Q
|r|
2
. However, this representation carries a significant conceptual leap. The charge Q is the source of field ⃗E
that fills the space around it, regardless of whether another charge q is present or not. Any charge
that approaches Q will feel the field surrounding it and will interact with it according to equation
(2) . Therefore, equation (2) gives the electric fields their own reality. They fill the space around
all charges and dictate the interactions between them. The electric fields are vector quantities and
have a direction and a magnitude in each location in space. Via equation (2), they tell us the forces
that act upon a charge that is placed at a location that contains a field. The electric field vectors
emanate from positive charges and terminate on negative ones. Positive charges will feel a force
aligned with these vectors, while the negative charges will feel an opposite force ( Figure 2.2). The
8

strength of the electric field vectors decays with the square of the distance from the source charge,
consistent with Coulomb’s law.
Figure 2.2: Electric field vectors emanating from a positive charge +q. A negative charge will
feel an attractive force towards the +q and a positive charge will feel a repulsive force from +q as
depicted by the arrows.
Sometimes it is more convenient to consider the energy of interaction, rather than the force
between particles. If we push a positive charge towards a positive charge, we will feel the repulsion
and it will feel like rolling a rock uphill in a mountain slope. Larger electric fields (which are
proportional to forces) will be analogous to steeper slopes. The same way that we need energy
to push a rock uphill, we will need energy to push the two charges together. The same way that
an elevated rock has stored potential energy and can release it if it rolls back, the two positive
charges pushed together will have some potential energy stored in them. This inspires us to come
up with a unit of electric interaction that is analogous to the mountain height and is known as
potential φ. Therefore, if we push one unit of charge 1 C between two points and we spend 1
J of energy, the “height difference” or potential difference between the two points is said to be
1 Volt (abbreviated as V). In this view, the electrostatic environment is like a series of mountain
peaks and valleys. Positive charges constitute the mountaintops, while negative charges are the
deepest points in the valleys. A test positive charge naturally rolls towards the valleys and slides
away from the mountain slopes. The electric forces are analogous to the slopes of the mountains
and valleys. Pushing a positive charge up the valley costs energy and rolling it towards the valley
9

releases energy. The height in this landscape is measured in units of V, and naturally the electric
field is the slope in this terrain,
⃗E = −
dφ
dx
(2.3)
The negative sign in this expression ensures that the direction of the forces is consistent with
Coulomb’s law. Consequently, the unit of electric field is V/m. This way of thinking about the
electric field is most common since setting objects at a fixed potential and thinking about energy
loss and gain is more convenient than thinking about forces.
To give a more concrete picture, imagine two metallic plates held 1 cm apart, and each one
connected to the opposite terminals of a common household 12 V battery. The positive terminal
resides at a “mountain top” that is 12 V higher than the ground or zero terminal. If 1 C of charge
falls from one plate to the other 12 J of energy will be released. The electric field between the
two plates is proportional to the slope of the potential change according to equation 3 and will
be 12 V/cm ( Figure 2.3). Bringing the plates closer makes the slope in φ steeper. Because the
potential drop occurs over a smaller distance, the corresponding electric field is larger. The above
also inspires a new unit for energy on the molecular scale, known as electron-volt (eV). 1 eV is the
energy gained by one electron when it falls across a potential difference of 1 V. Given the charge
of electron, 1 eV = 1.602×10−19 J.
Armed with the above understanding, now let us consider an example on the atomic scales.
Consider the simplest atom, which is the hydrogen atom with one proton and one electron in its
most stable state. The wave nature of the electron necessitates that the peak of the density of the
electron remain on average about 0.0529 nm away from the proton in the nucleus, a distance also
known as the Bohr radius. It is also known that it takes 13.6 eV of energy to ionize the hydrogen
atom, that is to strip the electron away from the nucleus and move it to a very large distance. This
means that the electron in the hydrogen atom was experiencing a potential of 13.6 V. Since the
potential peak originates at the nucleus and is felt as 13.6 V at Bohr radius, the “slope” of potential
or the electric field is 13.6V
0.0529nm = 257 V/nm, which is equivalent to 2.57×109 V/cm. Compare this
10

Figure 2.3: Electric field between Two charged metallic plates at constant potential difference
depend on the separation of the plates. The potential can be thought of as the height difference
between the peak of a mountain and the valley. The electric field is analogous to the slope of the
mountain.
with the battery thought experiment in the previous paragraph (12 V/cm). The effective electric
field between the nucleus of a hydrogen atom and the electron is about 100 million times larger
than the field between the two plates held at 1 cm distance and connected to a 12 V battery!
2.3 Measuring Fields on the molecular Scales
From the previous explanations, it seems that a voltmeter would be an important device for measuring the electric potential. This is possible in our macroscopic realm. However, at the length
scales of molecules (1 nm = 10−9 m), measuring the variations of potential is not easy, because the
measurement apparatus needs to be about as big as the molecules. Fortunately, there is a method to
achieve this as will be described below. Molecules can vibrate at a variety of distinct frequencies.
In this way, they are similar to strings of a guitar that are tuned to different notes. Sometimes, when
light passes through a sample, if the frequency of the light matches that of an appropriate vibration,
light can drive that mode of vibration. Therefore, one can learn about vibrations of molecules by
measuring the absorption of various frequencies of light after it has passed through a medium. A
spectrum is a plot of the absorption of light as a function of the light frequencies. The example
11

in Figure 2.4a shows the absorption spectrum of benzonitrile due to the nitrile bond stretching
vibration as shown in the figure.
Figure 2.4: (a) Absorption of -CN stretching frequency in benzonitrile. (b) Shift of -CN stretching
frequency, ∆ν, under the influence of external electric field, ⃗E
The same way that one can tune the frequency (i.e., the note) of a guitar string by adjusting
its tension, some vibrations in molecules respond to external factors, such as electric fields, and
change their frequencies (ν). The change of vibrational frequency due to electric field is known
as the Stark effect, and is the basis of measuring electric fields on the molecular scale. [2, 3] The
change in frequency ∆ν is approximately proportional to the electric field ⃗E.
∆ν = −∆µ · ⃗E (2.4)
with the proportionality constant ∆µ, which is a tuning parameter specific to a molecule.
Armed with this knowledge, when facing the problem of measuring an unknown electric field
in a chemical environment, one may insert a molecule as a probe with a well-known Stark tuning rate in that environment and measure its absorption spectrum. The changes in the absorption
frequency, after accounting for some other factors, can be related to the electric field experienced
by the molecule. A simple, but intuitive explanation of this effect is that electrons are like blobs
of glue that hold atoms together to make bonds. Electric fields can push and pull on this glue,
12

and slightly move it around. Therefore, there is no surprise that sometimes the electric fields can
change the stiffness of the bonds, which just like a guitar string, affect their vibrational frequency.
An example of such a Stark probe molecule is benzonitrile, which changes its frequency by
approximately 0.36 cm−1
for every 1 MV/cm of electric field. Other vibrations that have been used
for this purpose are carbonyl (-C=O), azide (-N=N=N), alkyne (-C≡C), and thiocyanate (SCN−).
The above approach is known as Stark shift spectroscopy and is used in a number of chemical and
biological systems as will be elaborated below.
2.4 Electric fields Inside a Liquid
Dissolving a molecule within a solvent, for example caffeine from your coffee or tea in water, may
seem trivial. However, on the microscopic scale the molecules of the solute (e.g., caffeine) will
be surrounded by the water molecules in a specifically organized way Figure 2.5. At the simplest
level of description, both the caffeine and water molecules have dipole moments. The water dipoles
will orient themselves around caffeine in such a way to lower the total electrostatic energy. As a
result of this orientation, caffeine will feel a net sum of electric fields from surrounding water
molecules. This field experienced by the solute molecule due to the orientation of the solvent is
called the solvation electric field. If the solute molecule has a well-defined vibration with Stark
response, it is possible to measure the solvation field by observing the absorption frequency shift
in the spectrum.
Benzonitrile can be used as a probe molecule to measure the solvation field in a number of
solvents. The field is estimated to range from 2 MV/cm to 10 MV/cm. [4] The idea of solvation
field was recently extended to the junction between a metal and a liquid, which is important for
electrochemistry and surface science. Solvation electric fields on the order of 10-20 MV/cm were
measured for some common solvents.[5] In summary, the act of dissolving a chemical in a solvent
is associated with electrostatic fields. In some cases, the effective energy of such interactions is
quite large. Dissolving molecules into or removing them from a solvent requires working with or
13

Figure 2.5: (a) Caffeine molecule, dark grey – Carbon, red – Oxygen, blue – nitrogen, grey –
hydrogen. The red arrow shows the overall charge separation over the body of the molecule. (b)
Water molecules oriented around the Caffeine. Only the electrostatic energy of the water molecules
is considered in this picture and hydrogen-bonding is excluded.
against such electrostatic interactions. For example, table salt easily dissolves in water because the
dipoles of the water molecule can stabilize the sodium (Na) and chlorine (Cl) ions. In solid table
salt the attraction between these ions is so strong that it will require an excess of 8000C to melt
it. But simply adding water stabilizes the ions individually and easily dissolves an otherwise very
strong crystal.
2.5 Electric Fields in Enzymes
Electrostatic environment in the reactive site of enzymes is crucial for their proper function.[6,
7] For example, the presence of a potassium ion is important in the function of the deacetylase
enzyme. It does not directly participate in the reaction but induces structural changes due to its
charge that promote binding of a zinc ion to the active site. This, in turn affects reactivity of the
active site ( Figure 2.6). [8]
Another important role of Electric fields is found in antibiotic resistance of proteins. Recent
studies have shown that in the absence of highly stabilizing electric field in the active site, antibiotic
binding is facilitated. However, in presence of high electric fields covalent linkages needed for
antibiotic attachment easily break off and inhibit the binding of the drug to the protein. This
inspires new drug designs so that the electric field in the active site remains small to stabilize the
14

Figure 2.6: The presence of K+ ion helps in Zn2+ binding in active site of deacetylase.[8]
protein-drug linkages ( Figure 2.7).[9] Another study shows that the intrinsic field in catalytic site
of ketosteroid isomerase (KSI) helps in preorganization of the geometry of the ground state of the
reactant such that the transition state gets easily stabilized i.e., the geometric change during the
reaction is very small.[10]
Figure 2.7: The Presence of polar functional groups easily detaches Penicillin (PenG) , but less
polar amino acid facilitates Avibactam (Avb) binding.[9]
Recently, Boxer and coworkers have performed experiments using a molecule (CXF-D) with
two vibrational modes that are sensitive to local electric field ( Figure 2.8).[11, 12] This molecule
is studied in a wide range of solvents, and it was noted that the change in the frequencies of the
two vibrations correlated with each other in a linear way. However, when the same molecule was
placed inside an enzyme, they discovered that the frequencies of the two modes did not follow
the correlation that was found in the solvents. With support from further computational work,
they inferred that the inner environment of an enzyme is fundamentally and completely different
than that of a cavity formed by solvent molecules. Such fundamental difference between a solvent
15

environment and an enzyme cavity is one of the reasons for the exceptional catalytic powers of
enzymes that drive the chemistry of life.
Figure 2.8: The vibrational frequencies of C=O and C–D of the molecule, CXF-D studied in
solvents and in enzyme reveal differences in the electrostatic environment between the solvent
cavity and the enzyme active site.[12]
2.6 Electric Fields in Electrochemistry
Unarguably, the field of electrochemistry is extremely important in modern technology. All batteries are electrochemical devices, many industrial processes rely on electrochemistry, and electrochemical methods help with detection and analysis of molecules. In brief, electrochemistry is
the science of coupling electrical and chemical energy. Often, electrodes are placed in a liquid
containing salts (electrolytes) and connected to a potential source. Depending on the sign and the
magnitude of the potential, electrons may traverse from the solid electrode into molecules on or
near the surface. Removing electrons from a molecule is known as oxidation and while delivering
an electron is known as reduction.
Most of the electrochemical action occurs at the junction of a solid and a liquid. Therefore,
understanding the physics and chemistry of that junction is very important. When an electrode
is charged positively, negative ions from the solution tend to accumulate near the surface. This
negative layer, effectively, reduces the influence of the electrode on the remainder of the liquid
because the positive charge is partially cancelled by the negative ions ( Figure 2.9). This effect is
known as screening and causes the potential from the electrode to diminish as we move away from
the surface. Molecules that are far away in the bulk, do not feel any electrostatic field from the
16

electrode. As we have discussed in a previous section, the slope of the potential drop is proportional
to the electric field. Therefore, an electric field is expected near the surface where the potential
from the electrode drops due to screening. This field can influence the structure and reactivity of
the molecules in that region. The positive charge in the electrode, and the net negative charge in
the nearby liquid in a combined way are known as the electric double layer (EDL).
Figure 2.9: Electric double layer consisting of the positively charged electrode and the adjacent
layer of negative charge. The potential φ(x) has an exponential decay profile from the electrode.
Understanding the structure and function of the EDL, and control over its molecular behavior is
a major frontier of electrochemistry. Since the EDL is formed from ions, and their corresponding
screening, the choice of ions affects the structure and behavior of the EDL. A frontier of research
is using surfactants as ions to create EDLs for specific purposes. Surfactants are often long-chain
organic molecules with a charged head. Two examples are shown in Figure 2.10a. Their long
chain organic parts make them soluble in organic solvents, while their charged heads make them
soluble in water. Because of this dual solubility, they are the functional components of soaps,
allowing them to dissolve grease into water. Beyond their widespread use in other field, they find
special use in electrochemistry. Because their long organic tails do not dissolve well in water, in
aqueous electrolytes they are pushed to the surfaces. Because of their charged heads a significant
net charge is accumulated near the electrode. Even modest concentrations of surfactants can lead
17

to significant charging of the surface. The electric field generated by this process between the
surfactants and the electrode has been recently measured using a vibrational Stark probe as shown
in Figure 2.10b.[13] It is shown that concentrations as low as 50 mM can result into electric fields
in excess of 10 MV/cm.
Figure 2.10: (a)Two examples of surfactants – CTAB with a positively charged head group
and SDS with a negatively charged head group. (b) Measuring the frequency shift and electric
fields from a layer of surfactants at metal-electrolyte interface using vibrational Stark-shift spectroscopy.[13]
It is likely that such electrostatic modification allows surfactants to control the selectivity of
electrochemical reactions. One of the most important electrochemical reactions is reduction of
CO2 to valuable products. Naturally, leaves of green plants absorb CO2 from the air and use sunlight and their sophisticated photosynthetic apparatus to deliver electrons and protons into CO2
to convert it to molecules useful for life. Performing this task using electrochemistry and at an
economical and sustainable scale is a modern frontier of science and holds promise for the future
of the planet. However, there are other reactions that compete with reduction of CO2 that need to
be suppressed. Hydrogen generation, although is important in its own right, is one such competing reaction. Recent work by electrochemists have shown that surfactants can suppress hydrogen
production in favor of CO2 reduction.[14] The molecular details of such suppression are still being
investigated. It is likely that the surfactants positive head groups prevent the positive protons (the
prime ingredient for making hydrogen) from approaching the electrode from the solution. Additionally, the excess electric field generated by surfactants may also help with the CO2 reduction
18

steps. Using both the charge and functionalization of surfactants as handles to control chemical reactivity at electrode surfaces remain an active area of research. Since organic chemistry can afford
creation of a wide range of surfactant molecules, it is hoped that tailored surfactants for specific
reactions will be developed in the future.
2.7 Future Prospects
From the above description, it is clear that managing the magnitude and direction of fields on the
molecular scale by appropriate positioning of charged entities is a frontier of modern chemistry.
While synthetic chemistry is quite advanced, the problem of placing charges at fixed locations can
still present major challenges. Strong electrostatic forces affect structure in complicated ways and
fluctuations of the charged entities lead to variation of fields. Nonetheless, major advances have
been made in creating motifs for exerting control on electrostatics of reactive centers.
An example of the recent advances is using crown ethers as a trap for charge at a well-defined
location relative to a catalyst center. Crown ethers are cyclic molecules with repeating units of
carbon and oxygens ( Figure 2.11) and are known for several decades. The ring-like arrangement
of oxygens enable crown ethers to capture positive metal ions. The crowns can be chemically
attached to a reactive center. When such a crown is loaded with the ion, the reactive center will feel
the electric field exerted by the ion. Therefore, one can affect the reactivity of the catalyst by the
choice of the ions captured in the crown.[15] An example of such motif, by Yang and coworkers,
is shown in Figure 2.11b, and shows that the rate of nitrogen formation in a bimolecular coupling
reaction gets affected by the choice of the ion in the crown.[16]
Benzonitrile stark probes attached to crown ethers have also been reported.[17] It has been
shown that the electric field from the cations can induce a frequency shift on the nitrile. However,
the quantity of the shift depends not only on the charge of the ion, but also the ion size, the nature
of the solvent, and likely the attachment geometry of the ion in the crown. All this point to the
fact that engineering fields on the microscopic scale requires attention to numerous details. One
19

Figure 2.11: (a) An example of Crown ether. The ring can be made larger by adding repeating
units as indicated by n. (b) Change in the electrostatic environment near a reaction center changes
the reaction rate. [16] (c) Trapping ions near a catalytic center.[15] (d) Benzonitrile tethered to
crown ether for detailed understanding electrostatics in the vicinity of the trapped ion.[17]
may envision future motifs where such details are further understood and that both positive and
negative ions can be used for exerting electrostatic influence on reactions.
Another exciting direction is using tailored surfactants for exerting control on the electrostatics
of interfaces.[18] A large variety of surfactants can be synthesized. However, the design rules for
their surface electrostatics are still being developed. Often water molecules surround the charged
head of surfactants, thereby screening their charge. Similarly, variation in their organic tail affect
their organization on the surfaces. Understanding the electrostatics and their structure at surfaces
is also complicated because surface spectroscopy and measurements of such monolayers is often
difficult. Nonetheless, it is hopeful that families of designer surfactants will be developed for controlling the yield and selectivity of surface and electrochemical reactions. Mimicking enzymes has
been a continuous challenge for decades in chemistry. Often, when synthetic chemists mimic the
inner core of the enzyme, without consideration of the environment in which the core is supposed to
work, the resulting structures cannot compete with natural enzymes. By now, it is well-understood
that the inner environment of the enzymes is fundamentally important to the reaction, both in terms
of its electrostatics and its hydration structure. Therefore, a significant research effort is expected
in the direction of mimicking the enzymes in this holistic way.
20

Finally, a series of experiments using Scanning Tunneling Microscopy (STM) have been conducted that suggest that chemical reactions can be controlled via direct application of an electric
potential between a metallic tip and a planar metal. The STM technique works based on tunneling
of electrons between a sharp metallic tip and the planar metal. It is extensively used for atomicscale microscopy of surfaces. However, its application to chemistry is relatively new. Recent work
has shown that the sign of the potential can increase the rate of the common organic Diels-Alder
reaction ( Figure 2.12).[19] Venkataraman and coworkers have demonstrated that external electric
fields applied at STM junction drives bond cleavage at ambient temperatures in solution without
any other activators( Figure 2.13).[20] The external electric fields helps decrease O-O bond dissociation energy. Another interesting study at STM junction along with other spectroscopic techniques
reveals important mechanistic details of how interfacial electric fields at roughened gold electrodes
catalyze chemical reaction.[21]
Figure 2.12: product formation of Diels Alder reaction is catalyzed by potential at STM tip.[19]
Figure 2.13: Dissociation of O-O bond by external electric field.[20]
21

Given that electric fields and electrostatic interactions are essential for chemistry, we anticipate
that new frontiers for engineering fields will open in the coming decade. Some of the ideas developed in academic research laboratories will find their way to the industrial scale reactions and will
perhaps facilitate energy-demanding processes in an environmentally friendly way.
22

Chapter 3
Electric Fields Influence Intramolecular Vibrational Energy
Relaxation and Linewidths
This chapter is based on the publication by A. Maitra, S. Sarkar, D. M. Leitner and J. Dawlaty,
”Electric Fields Influence Intramolecular Vibrational Energy Relaxation and Linewidths”, J. Phys.
Chem. Lett., 12, 7818-7825 (2021). Prof. David Leitner helped us in parts of the calculation and
in plotting Figure 3.6.
3.1 Introduction
The importance of electric fields and electrostatic interactions at the molecular length scales in
influencing the thermodynamics and kinetics of reactions can not be overstated. The relevance of
electric fields in enzyme catalysis [6, 22], electrocatalysis [23, 24], and thermocatalysis [25, 26],
is becoming more evident with new experimental and computational evidence. [27] Control over
fields at interfaces and micro-environments for enhanced reactivity is a major research frontier.
[28, 29]
A critical tool for measuring electric fields on the molecular scale is vibrational Stark shift
spectroscopy.[30–32] A well-defined localized vibration in a probe molecule is first calibrated in
electric fields and reference environments. The frequency shift of the reporter molecule is often
linearly related to the local electric field. Even though specific interactions of such molecules,
23

such as hydrogen bonding, are more complicated and must be accounted for, they remain valuable
reporters of chemical microenvironments such as enzyme cavities [33] and near electrochemical
interfaces.[13, 34]
Benzonitrile is a useful Stark probe because it is relatively unreactive and its nitrile stretch
frequency near 2230 cm−1
is isolated from other common vibrations of organic molecules. Frequency shift of nitrile has been used by us and others for understanding a variety of problems
ranging from electrostatic fields within proteins[35, 36], dielectric solvation [5] and ionic structure
near interfaces.[13]
A natural issue in interpretation of nitrile stretch spectra is understanding their linewidths. One
may envisage at least two sources contributing to the linewidth. First, the molecule reports on the
fluctuations and inhomogeneities of the environment as an antenna, and thereby encodes the environmental variations in its linewidth. Second, the Stark-active vibration is often coupled to other
low frequency modes within the same molecule and can dissipate energy into those modes thereby
contributing to the linewidth.[37, 38] When the second mechanism is operative, interpreting the
linewidths as merely a representation of the environmental fluctuations via an antenna effect becomes problematic. In this work, we show that not only is there a significant intramolecular source
of line broadening in the benzonitrile Stark probe molecule, but also that such intramolecular pathways of energy dissipation change as a function of applied electric field. The second observation
is the main point of our work and helps with isolating intra- and inter-molecular effects when interpreting Stark shift spectroscopy linewidths. In the broader context of intramolecular vibrational
energy relaxation, our work emphasizes that such relaxation pathways can be altered by external
electric fields, either applied by an electrode or due to a nearby ionic structure. Therefore, these results are applicable when considering reactive pathways that depend on intramolecular vibrational
relaxation.
The linewidth of nitriles has been discussed in the literature previously. It is well-known that
not only the nitrile frequency but also the linewidth is altered by the solvent [39–42] and temperature [39, 41]. Gai and coworkers in studying nitrile derivatized amino acids have reported red-shift
24

and narrowing in tetrahydrofuran, and blue-shift and broadening in water. Temperature-dependent
FTIR measurements show narrower bandwidth with increase in temperature and has been attributed
to more a homogeneous environment at higher temperatures.[39] Hydrogen-bonding solvents give
rise to a blue shift and line broadening. Several decades ago, work on ortho, meta, para substituted acetophenone by Mueller and coworkers revealed that steric and inductive effects of the
substituents play an important role in determining the linewidth of the carbonyl. [43]
Recent experimental and theoretical studies have shown nitrile frequency-shift changes the
nitrile vibrational lifetime due to intramolecular coupling with other low frequency modes present
in the molecule. This is due to alteration of the vibrational energy transfer rate from nitrile to other
overtones of the low frequency modes.[38] This intramolecular energy dissipation phenomena for
the CN mode has been observed in isotopically substituted nitriles. In this case the perturbing factor
that changes the intramolecular relaxation is mass. IR pump-probe measurements of isotopically
substituted p-cyano-phenylalanine (Phe-CN) by Rodgers et al. revealed a nonmonotonic lifetime
trend with respect to reduced mass.[37] As the electronic potential of the molecule does not change
due to isotopic substitution, the contributing factor to the nitrile lifetime was thus found to be due
to the intramolecular vibrational relaxation. [38] 2DIR experiments by Tucker et al. showed that
by tuning the intramolecular coupling it is possible to tune the vibrational energy flow in paraazidobenzonitrile (PAB) and para-(azidomethyl)benzonitrile (PAMB).[44]
Outside of the context of Stark shift spectroscopy, long-lived vibrations are important in mediating chemical reactions in gas and condensed phase reactions and in biomolecules.[45–47] There
is interest in vibrational energy and thermal transport through molecules and molecular junctions[48–53], including hydrocarbon chains and polyethylene glycol oligomers, in some studies
with the aim of designing molecular systems that rectify vibrational energy [54–58] and thermal
transport [59–61]. The nature and rate of energy and thermal transport in molecular systems are
influenced by vibrational energy relaxation.
Recently, we have identified a systematic line-width change in the nitrile stretch of benzonitrile
with application of an electrochemical potential. [1] Sum frequency generation experiments on
25

tethered 4-mercaptobenzonitrile show line-narrowing with oxidative potentials and line-broadening
with reductive potentials ( Figure 7.1a, Figure 7.1b). One may suspect that such changes are due
to changes in the fluctuations or structure of ions near the electrode. However, no a priori reason
suggests that in the mentioned experiment either sign of the electric field should result into reduced
or enhanced fluctuations or inhomogeneities.
(a) (b)
Figure 3.1: (a) Experimental nitrile spectra and (b) linewidths of surface tethered benzonitrile
from Sum Frequency Generation experiments.
We have also reported a similar trend as a function of the electron-donation strength of a number of neutral substituents. The more electron donating groups (emulating a negatively biased
electrode), in general, cause line-broadening[1]. The above experimental results led us to hypothesize that an intramolecular source of line-broadening, especially one that is sensitive to the
externally applied field, is likely the reason for these observations.
In this work we seek to test the hypothesis of electric field-dependent vibrational energy transfer. To our knowledge, intramolecular effects in linebroadening from the perspective of a continuously variable external electric field has not been explored before. Two factors govern dissipation
of energy from nitrile to lower frequency modes: anharmonic couplings, and resonances between
the nitrile and combinations and overtones of lower frequency modes. Correspondingly, we compute how these factors change as the applied electric field on a molecule is varied. Using a standard
26

energy transfer formalism, we show that the combination of these two factors predict line narrowing for negative fields and linebroadening for positive fields, consistent with the experimental
observations.
3.2 Methods
To compute the molecular vibrations and anharmonicities in the presence of an electric field, we
use a previously reported method [62]. In brief, two meshes of point charges are created to mimic
positive and negative sides of a capacitor. The size of the capacitor and the mesh is selected such
that the field is uniform along the body of the molecule. Geometry optimization and frequency calculations were carried out by placing a benzonitrile molecule centered between the two capacitor
plates ( Figure 3.2). Density functional theory methods were used for geometry optimization and
frequency calculation with B3LYP functional and 6-311G(d) basis set using the Q-chem package
[63]. IQMol was used as visualization software. All the computations were performed in vacuum.
For anharmonic frequencies of benzonitrile VCI(2) (vibrational configuration interaction theory)
was used. We have reported the effect of electric fields ranging from -50 MV/cm to +50 MV/cm.
Third order anharmonic couplings are calculated for geometry optimized structures. These
cubic anharmonicities and resonance conditions (or their detuning from CN mode) are crucial in
determining the intramolecular effects in the linewidth of nitrile stretch. To third order in the
anharmonic coupling, the rate of energy relaxation from the nitrile, W, is given by the sum of two
terms, W = Wd +Wc, where[64]
Wd(ωα) =
16ωα
∑
β,γ
|Φαβ γ |
2
(1+nβ +nγ )
ωβωγ
×
(Γα +Γβ +Γγ )
(ωα −ωβ −ωγ )
2 + (Γα +Γβ +Γγ )
2/4
(1)
27

-100 0 +100
z (Å)
+

-
Mesh with 1 Å
separation between
point charges
45
Å
Figure 3.2: A cartoon of the computational setup used in this study (adapted from reference [62])
Wc(ωα) =
8ωα
∑
β,γ
|Φαβ γ |
2
(nβ −nγ )
ωβωγ
×
(Γα +Γβ +Γγ )
(ωα +ωβ −ωγ )
2 + (Γα +Γβ +Γγ )
2/4
(2)
In the above, the subscript α refers to the nitrile vibration, while β and γ refer to the modes that
accept energy from nitrile. The coefficients of the cubic terms in the expansion of the interatomic
potential in normal coordinates is Φαβ γ . The detuning between donor and acceptor modes is
∆ω = ωα −ωβ −ωγ . The broadening of each mode involved in the transfer is Γ, while the thermal
population in the relevant modes is n = (exp(ω/kBT)−1)
−1
.
Relaxation of the nitrile stretch into two vibrational quanta, either into two different modes or
into one mode, is given by Eq. (1), and is referred to as decay with rate given by Wd. Eq. (2)
corresponds to the combination of a quantum of the nitrile stretch with a vibrational quantum of
a second mode to yield a quantum of vibration in a third mode, with a rate given by Wc. This
second process makes negligible contribution to the total rate, W, and thus to the linewidth of the
28

nitrile. The frequency of the nitrile stretching mode, roughly 2300 cm−1
, lies well below the next
group of vibrational frequencies, which start above 3000 cm−1
. To reach beyond 3000 cm−1
a
vibrational quantum of greater than 700 cm−1
is required, and such modes have a low probability
to be populated at 300 K. Furthermore, the CH stretch vibrations above 3000 cm−1
tend to be fairly
localized and couple only very weakly to the nitrile stretch. We thereby only consider Eq. (1) in
the calculation of the linewidth, so that W ≈ Wd. The lifetime contribution to the linewidth of the
nitrile is given by W/2.
The rate of energy relaxation, W, is given by the product of the square of the coupling and the
local density of states coupled via cubic anharmonic interactions to the nitrile stretch. The density
of states depends on the broadening of all the locally coupled states, represented by Γj for mode
j. The value of the broadening depends on relaxation rates for all modes due to coupling to other
modes and to the solvent. We do not know the solvent contribution, and we have taken the sum of
Γj
in Eq. (1) to be 24 cm−1
for all the calculations that we report. The rate increases somewhat
when we select larger values of this sum, but we are interested here in how the linewidth varies
with the strength of the field, and what gives rise to that variation. Changes in the sum to a different
value, say 20 cm−1 or 50 cm−1
, does not affect the trend.
Of course, higher order anharmonic coupling terms may contribute to the rate of relaxation
of the nitrile and thus to the linewidth. In past work, calculation of the nitrile lifetimes of four
isotopomers of cyanophenylalanine in solution that incorporated only cubic anharmonic interactions yielded results[38] that compared well with experimental measurements carried out by Gai
and coworkers[37]. Not only did the rates of energy relaxation compare well, but trends in the
rates with isotopic substitution also reproduced the trends found in the experiments. In general,
we expect that the anharmonic contribution to rates of energy relaxation in molecules of moderate
size, as considered in this study and in the earlier study of cyanophenylalanine, or larger to be on
average be dominated by cubic terms [46]. We carried out calculations of the rate of energy transfer from the nitrile, and thereby the linewidth, using the mode frequencies and cubic anharmonic
coupling at 9 different field strengths.
29

3.3 Results and Discussions
As shown in Equation 1, two factors determine energy transfer from nitrile to lower frequency
modes: anharmonic couplings and resonance between the coupled modes. We will begin by discussing the third order anharmonic couplings between the nitrile vibration (mode 28) and lower
frequency modes (modes 3 to 27) wihtout an external field. Figure 3.3a shows the magnitude of
coupling |ΦCN,β,γ
| as a function of mode number β and γ. The two lowest frequency modes are
not included because coupling to them could not be calculated by the electronic structure method
used by us. This is not significantly prohibitive since these modes are largely off resonant.
|Anharmonic Coupling| / (cm-1
)
(a) (b) TFR
Figure 3.3: (a) Cubic coupling in the absence of electric field between the CN stretch and combinations of modes β and γ. (b) The third order Fermi resonance (TFR) in the absence of field is a
scaled version of panel (a) to account for resonance (equation 3). Since the quantities are symmetric, only half of the coupling and TFR matrices are shown for ease of visualization.
The couplings in Figure 3.3a alone are not enough to determine energy transfer rates. The
resonance condition ∆ω = ωCN −ωβ −ωγ must also be considered. For that reason, we evaluated
an effective dimensionless third order Fermi resonance (TFR) parameter [38] that accounts for
both coupling and resonance as defined below:
30

T FR =
|ΦCN,β,γ
|
∆ω
(3)
Figure 3.3b shows a plot of this parameter, and exhibits a few combinations that dominate
energy transfer to low frequency modes. The first of these combinations involves energy transfer
from nitrile to one quantum of mode 10 and one quantum of mode 25 (ω10 = 776 cm−1
and
ω25 = 1517 cm−1
). The second dominant contribution is energy transfer to two quanta of mode 21
(ω21 = 1218 cm−1
). Beyond that, there are a few other modes with similar TFR, however, some
of them did not show significant variation with electric field. One channel with moderate TFR
involves energy transfer to two quanta of mode 20 (ω20 = 1206 cm−1
), which as we will show
later does have dependence on electric field.
When the electric field is turned on, the couplings and resonances discussed above are expected
to change. Several of the low frequency modes exhibit Stark shifts as a function of field, which can
affect their resonance with the nitrile. To exhibit this point, some of the Stark shifting modes are
presented in Figure 3.4a, b. The couplings are also expected to vary. To understand the combined
effect of coupling and resonance in the presence of electric field, we have analyzed the TFRs for
the above mentioned three dominant channels ( Figure 3.4c), and have noticed that these channels
remain dominant for all fields. However, their contributions to energy transfer varies as the electric
field is changed either due to change in resonance or the coupling.
For the first channel (mode 10, 25) the coupling does not change significantly as a function
of electric field ( Figure 3.5a). However, the resonance condition ∆ω varies significantly from
the negative electric field (8.6 cm−1
at -50 MV/cm) to positive field (40.4 cm−1
at +50 MV/cm).
As this channel becomes more off-resonant with increasing field values, energy transfer becomes
less favorable, and consequently its contribution to linewidth decreases. This is consistent with
experimental observation as will be discussed in detail shortly.
For the second channel (mode 20) there is no considerable change in resonance condition as a
function of field ( Figure 3.5b). However, the coupling decreases significantly from the negative
31

CN Stretch
Ring Modes
(a) (b) (c)
Electric Field / (MV/cm)
Frequency / (cm-1
)
Frequency / (cm-1
)
Electric Field / (MV/cm)
Figure 3.4: (a) The vibrational manifold of benzonitrile. (b) The variation of several representative
modes with electric field. The Stark shift of the CN stretch and the lower frequency modes will
affect the resonance condition for energy transfer.(c) Third order Fermi resonance (TFR) for the
three important channels of vibrational relaxation as a function of electric field.
32

|Coupling|/ cm
-
1
|Coupling|/ cm
-
1

Electric Field / (MV/cm)
(a) (b) (c)

Frequency/ cm
-
1

+

Electric Field / (MV/cm)
∆

∆
Frequency / cm
-
1
|Coupling|/ cm-1

∆
|Coupling|/ cm
-
1
Electric Field / (MV/cm)

Frequency / cm
-
1
Figure 3.5: (a)Variation of coupling and resonance for energy relaxation from the CN stretch
to the combination of modes 10 and 25. The motion of the atoms in the respective modes are
shown below. The detuning increases significantly with more positive fields, giving rise to narrower linewidths.(b)Variation of coupling and resonance for energy relaxation from the CN stretch
to two quanta of mode 20. The motion of the atoms in the respective modes are shown below.
There is an significant decrease in coupling with more positive field, giving rise to narrower
linewidths.(c)Variation of coupling and resonance for energy relaxation from the CN stretch to
two quanta of mode 21. The motion of the atoms in the respective modes are shown below. The
coupling increases significantly with more positive field.
33

field (29.3 cm−1
at -50 MV/cm) to positive field (3.0 cm−1
at +50 MV/cm). This observation is
also consistent with decreased linewidth with increasing field.
In contrast to the previous channel, for the third channel (mode 21) the resonance condition
does not change much as a function of field ( Figure 3.5c). However, there is an appreciable increase in coupling from the negative field (37.2 cm−1
at -50 MV/cm) to positive field (55.8 cm−1
at
+50 MV/cm). This third channel if evaluated separately would result in larger linewidths with increasing field, which is inconsistent with the experimental observations. However, its contribution
is largely overwhelmed by the other two channels described above.
To identify the contributions of all combinations, including the above three channels, to the
linewidth we evaluated equation 1. Our results are presented in Figure 3.6. The figure shows that
the linewidth decreases monotonically from negative to positive fields, consistent with experimental observations. The figure also shows that the contribution of the first channel (coupling to modes
10 and 25) by far is the largest and dominates the trend. The contribution to the linewidth from the
second and third dominant channels (mode 20, mode 21) are far smaller and not shown.
Note that the span of electric fields studied computationally is larger than the experimentally
achievable range. The controllable parameter in the experimental results shown in figure 1 is the
applied potential, which ranges from -1.2 V to + 0.6 V with respect to Ag/AgCl. As explained
in detail in reference [1] this range corresponds to effective field values spanning ∼ 25 MV/cm.
Therefore, within the range of fields that are achieved in the experiments, the computational results
( Figure 3.6) show a nearly linear decrease in linewidth consistent with the experimental observations ( Figure 7.1b). We note that the computed linewidth at zero electric field is comparable
to the linewidth at zero applied potential ( Figure 7.1) and the previously reported homogeneous
linewidth for benzonitrile derivatives as determined by 2D IR spectroscopy.[65]
Given that the choice of the coupling linewidths in equation 1 (Γα,Γβ and Γγ ) can influence
the total predicted linewidth, we do not anticipate quantitatively matching the linewidths between
computation and experiments. Our goal is to demonstrate that the experimental trend in linewidths
is reproduced computationally.
34

For channel 2 and channel 3, the anharmonic coupling is largely varying with electric fields and
they show opposite trends. To explain how the the electric field influences the coupling between
the CN stretch and the modes in these channels, we proposed that the motion of the shared carbon
atom (C1, shown in Figure 3.7) must be influenced by the electric field. This atom serves as a
bridge connecting nitrile to benzene ring. As seen in Figure 3.7, the amplitude of the C1 atom in
the nitrile stretch is small, but remains nearly constant across all fields. However, the amplitude of
C1 atom in mode 20 decreases, while in mode 21 increases with positive field. This justifies the
change in coupling shown in Figure 3.5 (b) and (c).
Electric Field / (MV/cm)
Nitrile Linewidth /(cm-1
)
Contribution from combination mode 10,25
Contribution from all modes
Figure 3.6: Computed nitrile linewidths as a function of electric field due to all channels (blue line)
and the dominant channel (orange). Consistent with experimental results, the linewidth narrows
with more positive field.
3.4 Conclusions
In conclusion, our major finding is demonstrating that IVR and linewidth can depend on the local
electric field felt by the molecule. The electric field affects IVR via changing the anharmonic couplings and/or the resonances between high and low frequency modes. This finding is important for
interpreting the vibrational linewidth of molecules that are used as Stark shift probes. It may be
convenient to assign changes in linewidth of such molecules to fluctuations and variations of electric field in their environments, as has been reported [39–42]. Here we point out that intramolecular
35

CN stretch
Electric Field /(MV/cm)
|Displacement Vector|
Figure 3.7: Relative amplitude of the displacement of shared carbon atom C1 as a function of
electric field.
sources of line broadening, and more importantly, their electric-field dependence should not be ignored. Of course, the trend of decreasing linewidth with increasing field is a consequence of the
resonances and couplings for this particular molecule. It should not be generalized to systems that
are very different from benzonitrile. For other molecules such effects may be present, however,
need to be studied separately.
Our work is also relevant to the area of controlling vibrational energy relaxation, and other
phenomena that are influenced by vibrational resonances. Note that the source of electric field
that influences molecular phenomena need not arise from an electrode. Molecular scale electric
fields can also be controllably exerted by the choice of nearby ions [13, 66] and solvents. The
influence of field on tuning vibrations must be evaluated to understand a broad range of chemical
phenomena. A few examples include the role of bridging groups in transferring vibrational energy
[67, 68], electrons [69], and protons [68]. In such cases, one may use electric fields to tune the
frequency and couplings of vibrations that control these processes.
36

Chapter 4
Electric Fields at Metal-Surfactant Interfaces: A Combined
Vibrational Spectroscopy and Capacitance Study
This chapter is based on the publication by S. Sarkar, A. Maitra, S. Banerjee, S. Thoi and J.
Dawlaty, ”Electric Fields at Metal-Surfactant Interfaces: A Combined Vibrational Spectroscopy
and Capacitance Study”, J. Phys. Chem. B, 124, 1311-1321 (2020). We have used two complementary tools - vibrational Stark shift spectroscopy which probes interfacial fields at molecular
length scales and electrochemical impedance spectroscopy (EIS) which probes the entire double
layer - to study the electric fields at metal-surfactant interfaces. EIS studies are performed by our
collaborators Soumyadeep Banerjee in Prof. Sara Thoi’s group at Johns Hopkins University. Figure 4.1, Figure 4.6, Figure 4.11 are also provided by Soumyadeep Banerjee.
37

4.1 Introduction
Surfactants are amphiphilic molecules, often with polar head groups and long hydrophobic chains
that have predominant presence and influence near a surface. Their importance is well recognized
in a wide range of diverse disciplines such as biology, medicine, environment, and petroleum
industry. [70–73] Another important example of surfactants is in the field of corrosion science
and engineering, where they are used as anticorrosion agents [74, 75]. Despite their importance,
some of the mechanistic details of their action, in particular their modification of the electrostatic
environment of the interface, remain elusive.
The influence of surfactants on electrochemical reactions has been studied for several decades
[76–78] and has gained recent attention [14, 79]. Of particular interest is the ability of surfactants
to affect the selectivity of electrochemical reactions. Uncovering the possible relation between
selectivity and interfacial fields is the broader motivation for this work. A prime example is the
electrochemicalCO2 reduction reaction (CO2RR), which is both of fundamental and practical value
due to its relevance to energy and climate. This reaction, however, is known to be outcompeted by
H2 evolution (proton reduction), which occurs at a lower potential [80]. It has been reported that
cationic surfactants, for example cetyltrimethylammonium bromide (CTAB), suppress H2 evolution in favor of CO2 reduction [14, 79, 81, 82]. It is important to identify the origin of such
selectivity. The two prevailing ideas are that when positive head groups adsorb on the electrode
surface, approach of H
+ or positively charged proton carriers to the surface is disfavoured, leading
to suppression of proton reduction [14, 83]. They also influence the concentration of other ions,
which in turn can influence the overall transport barrier across the interface. Secondly, the hydrophobic nature of the surfactant tails interferes with the local water concentration, which in turn
can affect reaction selectivity for water-requiring reactions [79]. These two scenarios are closely
related and in either one, interfacial electric fields influence the delivery of reactants to the surface. Some evidence from literature suggests that the presence of cations at the interface may not
just impede hydrogen evolution, but rather actively promote the first steps of CO2 reduction by
38

lowering the energies of the intermediates formed during CO2 reduction [84–87]. Therefore, the
interfacial fields that cationic surfactants exert may play a key role in promoting CO2RR.
Measuring the local electrostatic environment within a few monolayers of a surface is a difficult task. A common approach is impedance spectroscopy, in which one measures the response
of the interface to small oscillatory potentials at various frequencies, often in the range of a few
Hz to a few KHz. Then the data is fitted to an assumed equivalent circuit model for the interface
and the results are related to the chemical and physical properties of the interface. The oscillatory
potential of the electrode can penetrate into the solution through several layers, until it is screened
by the interfacial ionic structure. Depending on the electrolyte type and concentration, this penetration depth can be several tens of nanometers. Therefore, impedance spectroscopy has a probing
length scale in the order of the screening length which is often larger than the molecular size. The
advantages of this method are its convenience and versatility, which have made it an ideal tool for
studying various interfacial processes such as transport, adsorption, and ionic structure. However,
a disadvantage is that it measures the overall electrical response of the interface, and it can be
difficult to uniquely tease out the details of structure and dynamics of specific interfacial layers
without heavily relying on assumed equivalent circuit models.
An important contribution of this work is to complement impedance spectroscopy with a more
local measurement of the electrostatic environment based on vibrational frequency shift of probe
molecules in the presence of an electric field. A molecule, bearing a well-defined vibrational
signature, is tethered at the interface and the frequency of the vibration is measured as a function
of changing the interfacial environment. As will be shown further in this work, the vibrational
probe is sensitive to the field profile in the vicinity of the surfactant layer over a length scale in
the order of the molecular size of the probe, ∼ 1 nm. We will show that the ionic layers that
are far away from the probe by several nm will not directly influence its frequency. Therefore,
this approach is far more myopic and local compared to impedance spectroscopy. Nevertheless,
the advantage is that such local fields on the molecular scale likely matter more for reactivity. A
disadvantage, however, is that it requires insertion of the probe molecule at the interface. This can
39

perturb the properties of the interface as we will show later in this work. Our work combines these
two approaches for understanding the interfacial environment between a surfactant and a metal.
4.2 Experimental Section
Silver foils of 0.1 mm thickness and 99.9% purity, cetyltrimethylammonium bromide (CTAB) (purity 99%), dodecyltrimethylammonium bromide (DTAB) (purity 98%), trimethyloctylammonium
bromide (OTAB) (purity 98%), sodium dodecylsulfate (SDS)(purity 99%), and Triton X-100 were
purchased from Sigma Aldrich. The surfactants were used without further purification. Surfactant
solutions were made in HPLC water. Ag substrates for SERS were prepared by chemically etching
the Ag foils [88]. Prior to etching, the Ag foils were sonicated in HPLC water for 30 minutes.
The washed Ag foils were treated with 28% NH4OH for 30 seconds to clean and smoothen the
surface. Next the foils were immersed in ∼ 6 M HNO3 acid for 10 seconds that roughened the
substrates and made them SERS active. The SEM images of the Ag substrates are shown in figure
S1 of B. The SERS active Ag substrates were washed with water and soaked in 0.02 M solution
of 4-mercaptobenzonitrile in ethanol overnight. After soaking in 4-MBN, the Ag substrates were
sonicated for 90s in ethanol, prior to using them for Raman experiments. A laser source from
Ocean Optics Inc. emitting 532 nm was used to excite the 4-MBN tethered on the SERS active Ag
substrates. Raman probe from IncPhotonics was used to excite and collect backscattered Raman
signal from the samples. Raman signals were detected using a 320 mm focal length spectrometer
with 1800 g/mm gratings (Horiba iHR320) and a 1340 × 100 CCD array (Princeton Instruments
Pixis).
EIS experiments were carried out in a vial type electrochemical cell. 0.1 M NaHCO3 was
used as electrolyte and was purged with N2 for 15 mins prior to experiments to remove any dissolved oxygen. An Ag/AgCl (3 M NaCl) reference electrode and graphite counter electrode and an
Ivium multichannel potentiostat were used for all electrochemical measurements. All impedance
measurements were taken at polarizing potential of -1.0 V vs Ag/AgCl, 1 mV amplitude and 107
40

HZ to 0.1 HZ frequency range. Resulting Nyquist plots were analyzed with Ivium software using
the following equivalent circuit where, R1 is the solution resistance and R2 is the charge transfer resistance. CPE1 is due to the contact capacitance and CPE2 is due to diffused double layer
capacitance.
R1 R2
CPE1 CPE2
Figure 4.1: Circuit diagram used for fitting impedance data
All Cdl values were extracted from CPE2 by using the formula:
Cdl =

(R2)
(1−N)
(CPE)
(1/N)
(4.1)
where N is the fitting parameter which is also extracted by simulating the Nyquist plots.
4.3 Results and Discussions
We begin with explaining the details of using vibrational spectroscopy in measuring interfacial
electrostatics. Vibrational Stark shift spectroscopy has been used extensively to reveal the local
electric field in a range of problems [89–92]. This approach has been used by us and others for
studying the interfacial electric fields [1, 34, 93–98]. A common molecule for interfacial vibrational Stark shift spectroscopy is 4-mercaptobenzonitrile (4-MBN), which is covalently attached to
a metal surface. The nitrile vibrational frequency of this molecule is known to be sensitive to the local electric field [4, 34, 93]. Within some approximations, the frequency shift is related to the local
electric field according to the linear Stark equation hν = −∆⃗µ · ⃗F, where ⃗F is the electric field and
41

∆⃗µ, also known as the Stark tuning rate, is the change in dipole moment between the ν0 and ν1 vibrational levels. The probe molecule 4-MBN has a Stark tuning rate of ∆⃗µ ∼ 0.36cm−1/(MV/cm)
and has been studied in a variety of contexts, both for measuring electrostatic fields [93, 99], and
for understanding its specific interactions such as hydrogen bonding and Lewis acid-base bonding
[4, 35, 94, 100, 101]. While such specific interactions are important, as will become apparent later,
the major influence on the nitrile frequency of 4-MBN in this work is the electrostatic fields created
by the surfactant molecules. We have also previously studied the influence of applied fields and
substituents of various electron-withdrawing power on benzonitrile [1], and through that work have
identified the limitations of using linear Stark shift approximation. The frequency shifts reported
in this work, lie reasonably within the linear regime. Finally, as will be shown later, the field profile
along the body of the probe molecule near a surfactant layer will be inhomogeneous. Therefore,
one must be aware that the field reported by the nitrile is not the same as the homogeneous field in
an ideal capacitor, but rather an effective field averaged over the body of the molecule. Conversion
from frequency shift to field values using the Stark shift equation noted above should be done with
this caution in mind. We address this nuance in interpretation of the frequency shift in detail later.
We have used surface-enhanced Raman scattering (SERS) spectroscopy, a sensitive technique
for observing vibrational signatures from monolayers on metal substrates, to study shifts in the
nitrile frequency as a function of surfactant concentration [102, 103]. We have chosen to work
with archetypal surfactants representing cationic, anionic, and neutral head groups, which have
very different influences on the interfacial field. Different surfactants with their structures used in
this study are shown in Figure 4.2.
This paper is arranged as follows. First, we present the vibrational spectroscopy and impedance
spectroscopy results, followed by explanation of the different behavior of cationic and anionic surfactants. We then present an electrostatic model in which the surfactants are represented as discretely charged sheets which influence the probe molecule through their near field effect. Based on
this model, and informed by the experiments, we present a picture of how the surfactants modify the interfacial electric fields, and how this modification manifests itself in vibrational and
42

impedance spectroscopy measurements. Finally, we comment on how these insights can be used
to understand design concepts and limitations of surfactants for achieving desired interfacial field
profiles.
(a) Cetyltrimethylammonium bromide (CTAB)
(c) Octyltrimethylammonium bromide (OTAB)
(b) Dodecyltrimethylammonium bromide (DTAB)
(d) Sodium dodecyl sulfate (SDS)
(e) Polyethylene glycol tert-octylphenyl ether (Triton)
Figure 4.2: Chemical structures and names of the surfactants with their acronyms used in the study
Nitrile stretch spectra of 4-MBN tethered on Ag substrate in water and 100 mM aqueous solutions of CTAB (cation), SDS (anionic) and Triton (neutral) are shown in Figure 4.3a. The nitrile
stretch frequencies νCN were retrieved by fitting the nitrile peaks to gaussian lineshapes, and are
shown in Figure 4.3b as a function of surfactant concentrations in the bulk. The frequency undergoes red shift, no shift, and a small blue shift with increasing concentration of CTAB, triton and
SDS respectively. Beyond a certain cationic surfactant concentration, the frequency does not shift
with increasing concentration, which corresponds to formation of a complete layer on the surface.
Expectedly, the nitrile probe is hydrogen bonded prior to adding any surfactant molecules.
However, hydrogen bonding does not prevent the probe from responding to an external field induced by the charged head groups of the surfactants. The sensitivity of the hydrogen bonded
molecule to the applied external field is expected to be different from that in vacuum or in low dielectric non-hydrogen bonding environments and must be assessed separately. In a previous work
43

(a) (b)
Figure 4.3: a) Nitrile stretch spectra of 4-MBN tethered on roughened Ag in 100 mM solutions of
surfactants b) Nitrile peak frequencies as a function of surfactant concentrations.
by our group the nitrile frequency shift was modeled as a function of both potential and ionic concentration in an aqueous electrolyte medium and a Stark tuning rate was retrieved from it [34].
This Stark tuning rate is used in the present work.
A shift of 4.5 cm−1
corresponds to an effective average field of -1.25 V/nm at the interface
of metal and cationic surfactant. We emphasize that this value should be interpreted as an effective average field, the field profile caused by the surfactant layer is expected to be inhomogeneous
within the length scale of the molecular probe. The effective average electric field at metal-anionic
surfactant interface is only 0.4 V/nm, even though the positive head group of CTAB and negative
headgroup of SDS both bear unit charge. This observation can be explained if we consider the
hydration shell around the quaternary ammonium and sulfate headgroups. The quaternary ammonium head group is hydrophobic and is weakly solvated.[104] The sulfate head group is more
hydrophilic and has a strong solvation shell around it. [104, 105] When the surfactant head group
comes close to the Ag surface, which is rendered hydrophobic by the nitrile layer, the weakly solvated ammonium head group can desolvate its water more easily and come closer to the surface.
On the other hand, the strongly solvated sulfate head group will carry its solvation shell and will
remain farther from the surface. Furthermore, the intervening water molecules between the sulfate
44

and the surface screen the field from the head group, resulting into a much smaller effective field.
Figure 4.4 is a cartoon representation of this effect.
d
weakly solvated
d
strongly solvated
(a)
(b)
Figure 4.4: Cartoon representation of hydration shell around a) hydrophobic quaternary ammonium head group, b) hydrophilic sulfate head group. Weakly solvated ammonium head group can
readily desolvate and approach closer to the nitriles. Strongly solvated sulfate head group retains
its solvation shell which limits how close the headgroup can approach the nitriles.
The cationic CTAB and anionic SDS surfactants also have different hydrophobic chain lengths.
The hydrophobic chain length determines the drive for ordering near the surface. Therefore, it is
reasonable to study its effect on the interfacial field. For this purpose, in addition to CTAB that
has a tail of 16 C, we studied the field in the presence of DTAB and OTAB which have 12 and 8
C respectively in their hydrophobic tails. As seen in Figure 4.5, for very low concentrations (10-
100 µM), there is no noticeable nitrile vibrational frequency shift, implying very small surfactant
population on the surface. Beyond that range, the nitrile frequency initially decreases for all three
cationic surfactants at different rates with respect to increasing concentration. However, beyond ∼
50 mM, they all induce the same frequency shift of nearly 4.5 cm−1
. This interesting observation
is quite telling about both the ordering and the orientation of the surfactants near the surface. First,
the fact that the longest chain surfactant, CTAB, creates a significant shift at much smaller bulk
concentrations, indicates that its accumulation on the surface is far more favored compared to
45

its short-chain counterparts. This observation is not very surprising and consistent with the idea
that longer chains make better surfactants. Beyond 50 mM, a full layer of surfactants has formed
for all three cases and any further increase in the bulk does not seem to affect the interfacial field.
However, the fact that the final value of shift produced by all three surfactants is the same indirectly
informs us about the orientation of the surfactant layer on the surface. In brief, this observation
is consistent with the charged head, instead of the hydrophobic tail, of the surfactant pointing to
the surface. This effect will be discussed in more detail shortly after introducing an electrostatic
model for interfacial field caused by surfactants.
It is important to note that all three surfactants form micelles in the bulk. The critical micelle
concentration (CMC) for CTAB, DTAB and OTAB are ∼ 1 mM, ∼ 12 mM, and ∼ 130 mM respectively [106]. As our data shows, the nitrile frequency shifts sharply with increasing CTAB
concentration below the CMC of CTAB. Similarly, both DTAB and OTAB produce frequency
shifts at concentrations far lower than their respective bulk CMC values. It is not surprising to
have surfactant accumulation on a surface prior to CMC values, which is consistent with our observations.
Figure 4.5: Experimentally determined central frequencies of nitrile stretch vs surfactant concentrations in logarithmic scale.
46

Electrochemical impedance spectroscopy (EIS) was used as a complementary technique to the
SERS studies. It is commonly understood that upon applying a potential on a surface, solvated
molecules of opposing charges accumulate at the electrode-electrolyte interface, generating an
electrochemical double layer capacitance between the electrode and ionic layers [107–109]. This
double layer capacitance, Cdl, thus represents a quantitative parameter to elucidate changes to the
molecular structure at the surface. Capacitance, which is defined as C = dQ/dV, reports how
much charge accumulates at the interface for a differential change in potential. Naturally, if more
mobile ions are available in an electrolyte, the capacitance will be larger. Furthermore, for an ideal
double plate capacitor, capacitance is directly proportional to dielectric constant of the medium in
the intervening space between the plates and inversely proportional to the distance between them.
This also applies to the effective double layer capacitance of the interface. If the dielectric constant
of the medium between the ions and the electrode is reduced, the capacitance will drop. Similarly,
if the mobile ions are kept at a larger distance from the electrode, the measured capacitance will
be smaller. In EIS, the current response is measured as a function of small oscillatory potentials
of varying frequencies. The results are modeled as arising from an equivalent circuit consisting
of resistive and capacitive components representing the interface. This approach is very common
and powerful in electrochemistry [110–112], and reveals various processes such as formation of
the double layer, ionic transport, and adsorption.
We carried out EIS studies on 4-MBN modified Ag foil as a function of surfactant concentration in the bulk and retrieved the capacitance as outlined in the experimental section. A plot of
capacitance as a function of surfactant concentration is shown in Figure 4.6. The most immediate
observation is that at low surfactant concentrations, the double layer capacitance decreases rapidly
for all of the three surfactants. This is similar to trends of capacitance on Cu electrodes reported
previously [79] and is consistent with formation of a surfactant layer on the surface. The surfactant layer has lower dielectric constant due to its long organic chains compared to the background
aqueous electrolyte that it displaces. Furthermore, the surfactant layer keeps the free ions of the
background electrolyte and the surfactant counter ions that contribute to capacitance farther away
47

from the surface [113]. Therefore, both by the dielectric constant and distance arguments, the
capacitance of the interface is expected to decline after formation of a surfactant layer, which is
consistent with the measured data.
At higher concentrations, however, the double layer capacitance starts to increase with concentration. We propose that this increase arises due to the increase of the net ionic strength of the
electrolyte. At these concentrations, addition of surfactants involves introduction of both anions
and cations to the solution at concentration that are becoming comparable to that of the background
electrolyte, which is kept constant. Therefore, as explained above, it is expected that increasing
the number of mobile ions in the solution will increase the capacitance.
These results are consistent with our SERS observation: accumulation of the surfactant molecules
occurs at low concentrations and reaches a terminal value of interfacial field after formation of a
surfactant layer. We remind the reader that for performing reliable EIS experiments, it is ideal if
a negative potential is applied, which for our experiments, was -1 V vs. Ag/AgCl. This is in contrast to our SERS measurements, which were performed at open circuit potential (OCP) to avoid
mixing the frequency shift induced by the applied potential from that created by the surfactants.
Note that it is possible for the metal to carry net charge at OCP, and it may be plausible to think
that it could influence the probe frequency. However, the probe frequency as observed at OCP in
an electrolyte is very close to that observed in air (within 2 cm−1
). This change is also known
to arise from hydrogen bonding. Therefore, to the best of our knowledge, if any change to the
vibrational frequency is exerted by the charge in the electrode at OCP, it is likely far smaller than
this value. Therefore, the threshold bulk concentration at which the surfactant layers will form on
the surface differs in these two experiments. In the EIS experiments, due to the negative potential,
the cationic surfactants are drawn to the surface at a much smaller bulk concentration (less than 0.5
mM) compared to the SERS experiments (larger than 10 mM). Therefore, the very first two points
in the capacitance in Figure 4.6 already correspond to nearly complete surface coverage concentrations. The capacitances for the three surfactants also follow the expected inverse relation with
chain length. Any drop of capacitance beyond that point is due to further changes in the interface
48

such as displacing high dielectric solvent water with long chains hydrocarbons and displacing the
more mobile supporting electrolyte ions. Bilayer formation could also be the cause of this drop
in capacitance.[114] Nonetheless, both techniques confirm formation of surfactant layers on the
surface.
The frequency shift data reveals that once surfactant ordering is complete, the local field induced by the surfactants of different chain lengths reach a terminal value and do not respond to
further increase in concentration. The capacitance, on the other hand, shows an initial drop, and
then further increase with increasing concentration. The reason for this difference is that the nitrile
shift exclusively reports local field variations at molecular level near the surface while EIS probes
the entire diffused layer. Therefore, in thinking about the interfacial field near electrodes, it is
important to distinguish between the longer length scale probed by EIS and the shorter molecular
length scale probed by vibrational spectroscopy. Both of these quantities are valuable in their own
rights. For example, the behavior of the entire diffused layer may be important for understanding
ionic transport from the bulk to the interface, while the local field felt by a reactant very close to
the interface may be important for its reactivity. Our experiments highlight the importance of both
of these pictures.
Given the importance of near-field profile close to surfactant layers, we create a simple electrostatic model to understand its behavior. Most importantly, we will reveal that the discreteness
of the surfactant charges and lateral separation between surfactant heads affect this near-field profile. To set the stage, we begin with a well-known classical electrostatics result for electric field
produced by a sheet of charge. [115, 116] If an infinitely large sheet carries a continuous surface
charge density ρ, the electric field at a distance z from the sheet will always be perpendicular to
the sheet and will have magnitude:
E =
ρ
2ε0
(4.2)
Importantly, this electric field is constant and independent of z (i.e. the same at all distances
from the sheet). For instance, a sheet carrying a charge density of ρ = 1e/nm2 will support a field
49

0.1 1 1 0 1 00
50
1 00
1 50
200
250
300
C
a
p
a
cit
a
n
c
e
(
m
F
c
m
-
2
)
CTAB con cen trati on ( mM)
OTAB
DTAB
CTAB
- 1 . 0 V vs Ag/AgCl
Ag - MBN
Figure 4.6: Double layer capacitance as a function of surfactant concentration for OTAB (8 carbon), DTAB (12 Carbon) and CTAB (16 Carbon) carried out with Ag foil-MBN. Due to the negative potential, the cationic surfactants are drawn to the surface at a much smaller bulk concentration (less than 0.5 mM) compared to the SERS experiments (larger than 10 mM). This gives rise
to smaller capacitance for the longest chain. Further decrease of the capacitance is likely due to
displacing the high dielectric solvent water and mobile supporting electrolyte ions with long chains
hydrocarbons.
of ∼ 9 V/nm at any distance. This gives rise to the classical capacitor result. If two oppositely
charged sheets are held at a distance from each other, the fields will add in the interim space
between the sheets, while outside the fields will always perfectly cancel. Therefore, the field in
between two sheets carrying ρ = 1e/nm2
and ρ = −1e/nm2 will be ∼ 18 V/nm, while no field
will exist outside the plates.
We emphasize that such a continuum approach is inadequate for the problem of surfactants,
especially at close distances. If a layer of surfactants along with its counter ions were to be treated
as two continuum sheets of charge, no field would be expected outside (see Figure 4.7). The
consequence for our experiment would be that no field would be sensed by the probe molecule,
which is contrary to the observations. This would be true, regardless of the image charges for the
ions.
Thus, one may not apply the continuum electrostatic charge distribution model to this problem.
Given that the average spacing between surfactant molecules is in the order of ∼ 1 nm and we are
also interested in finding out the field profiles over similar distances from the surfactant layer, it
50

Sheet of image charges

=
+
= 0
= 0
Figure 4.7: Representative figure showing that the probe molecules will sense zero field if the
surfactants head group and counter ions are treated as continuum sheets of charge. The blue and
the red plates indicate sheets formed by the surfactant headgroups and the counterions respectively.
is necessary to treat the electric field as a sum over spatially localized discrete set of charges as
opposed to an integrated sum over a continuum of charges that gives rise to equation 4.2.
The electric field at a distance z from a discrete square lattice of charges with average charge
density ρ is:
E =
ρ
4πε0
ζ (z/d) (4.3)
Here, d is the lattice spacing. In the above, one can identify a constant part ρ/4πε0, which
is analogous to the continuum result. The function ζ (z/d) represents the contribution that arises
from summing over the discrete charges.
ζ (z/d) = z
d
∞
∑
j=−∞
∞
∑
k=−∞
1

(z/d)
2 + j
2 +k
2
3/2
(4.4)
This function only depends on summing over lattice indices j, k and the distance from the sheet
in units of the lattice constant z/d. It determines the deviation of the field arising due to discrete
assembly of charges from that of a continuum. Therefore, we study its properties separately. Figure 4.8 shows a plot of this function with respect to z/d computed over a large grid. As can be seen
in the figure, this function decays very rapidly as z/d increases, and asymptotically approaches

the value of 2π for large z/d. The asymptotic value corresponds to the far field and retrieves the
continuum limit when inserted in equation 4.3. This is physically justified and conveys the idea
that if we move far away from a discrete lattice, for all practical purposes, the field looks as if it
arises from a continuum and details of the discreteness of the lattice is blurred. However, the more
important result is the consequences of discrete charges for the near field, or the length scale over
which the function ζ (z/d) decays. The figure shows that the asymptotic value is practically approached with a good approximation for a distance of z/d ∼ 2. Therefore, as we move away from
a lattice of charges with spacing d, by the time we are about ∼ 2d away, the field from all points
on the lattice have merged and blurred as if the field was arising from a continuum distribution.
However, if we reside closer than about two lattice constants to a lattice of charges, we are in the
near field regime and we must account for the discreteness of charges.
Since ζ (z/d) is in normalized units, it is scale-free and once it is computed with adequate
resolution, it will hold for any lattice spacing. Then, for convenient calculations, ζ (z/d) can
be interpolated for an arbitrary distance, and multiplied by the constant part of equation 4.3 to
retrieve the electric field. Figure 4.8 shows the field of a discrete lattice of unit charges with
d = 1nm giving rise to charge density of ρ = 1e/nm2
. The field from the lattice will approach the
constant ∼ 9V/nm for large z/d values. Note that in the above, field is calculated along a line that
is directly above a point charge. The results are slightly different if this line is not coincident with
a point charge on the lattice, but rather shifted from it. This is discussed in details in the SI ( B).
Surfactant layers near the surface can be modeled as sheets of discrete charges. However,
unlike the above, they do not necessarily reside on a well defined square lattice. Even when the
positions of the charges are randomly shifted from that of the square lattice, the main features of
the field profile as discussed above are conserved. In particular, the approach to a continuum value
within z/d ∼ 1−2 remains true. For details, see Figure 9.3 in SI ( B).
With the above results about the near field behavior for a single discrete lattice, we will consider
the field from two such lattices carrying opposite charges and held at a distance of s from each other
as shown in Figure 4.9. Consider s ≫ d, that is the distance between plates is several times the
52

(a) (b)
2π
Expected constant value for
a sheet of charges with r = 1e-
/1nm2
Figure 4.8: Evolution of a) ζ function with z/d, b) electric field with z. ζ decays rapidly and
asymptotically approaches 2π. The electric field value approaches ∼ 9V/nm for a lattice of charge
density ρ = 1e/nm2
.
spacing between charges in the lattice. The total field is the sum of fields from each lattice. At a
large distance from the plates, the far field values from both lattices will be equal and opposite to
each other, making the total field zero, as expected for two continuously charged capacitor plates.
However, very importantly, the total field very close to one of the sheets is not zero. For example,
near the positive sheet, the near field of the positive sheet adds to the far field of the negative sheet,
and since they do not have the same magnitude, they give rise to a non-zero value. Therefore,
unlike what is expected from two continuously charged infinite plates where the field is entirely
contained within the two plates, for the two discretely charged sheets, the electric field ‘leaks’ out
over some distance outside the two sheets (see Figure 4.9).
It is this near field that is sensed by the vibrational probe near the surface. Furthermore, this
leaking field has a decaying profile along the z direction and therefore is not homogeneous along
the body of our vibrational probe. The relevance of this field inhomogeneity in interpretation of
the frequency shift is discussed shortly.
We mention that it is also customary to calculate the field from a distribution of charges, especially periodic ones, using summation in the frequency domain, for example the Ewald summation
[117], and Wolf summation methods [118]. Often the advantage of resorting to frequency domain
53

Enearfield ≠
Efarfield =

−
Figure 4.9: Field profile from two oppositely charged lattices held at a given distance from each
other. The resultant field obtained by summing fields from the two lattices leaks outside the two
sheets and has a non-zero value at shorter distance but zero value at larger distance.
summation is computational efficiency, as is demanded for example in molecular dynamics simulations. Although they can be applied to our problem relatively easily, they do not introduce any
extra benefit or insight, since direct summation in the space domain is not difficult in this case.
In light of the electrostatic model, we shall interpret the observed frequency shifts of the field
probe. Three major factors affect the interfacial field produced by the surfactant layers. These
factors are surface density, their distance from the vibrational probe, and finally their orientation
and arrangement near the surface. The density of surfactants, especially the quaternary ammonium
family, has been extensively reported in the literature [119–123]. We will examine the field for an
average surfactant density of 1e/nm2
, which is justified based on literature values [119, 120].
For the surfactant orientation when the layers are fully formed and the terminal value of frequency shift is reached, three main motifs may be considered ( Figure 4.10). First is the surfactant
head pointing towards the surface with the tail towards the solution and the counter ion either held
in the solution or intercalating along with its solvation shell within the surfactant layer. The second
motif is the surfactant tail pointing toward the surface, possibly intercalating with the benzonitrile
54

monolayer, keeping the charged head along with its counter ion far away from the surface. The
third motif is the anion (Br−) mediated head group adsorption.
-
+
-
-
+
-
-
-
-
-
-
+
(a) (b) (c)
Figure 4.10: The positively charged surfactant head is represented as a blue circle with hydrophobic tail attached to it and the negatively charged counterion is represented as a red circle. Cartoon
representation for three motifs : a) surfactant head facing the nitriles with counterions lingering
near the head group or outside the hydrophobic tail, b) surfactant tail facing the nitriles with the
head group towards the bulk and counterions adjacent to the headgroups, and c) counterions adjacent to the monolayer
If the third scenario, or the Br− near the surface motif, was true, we would expect a significant blue shift in the nitrile frequency, which is inconsistent with the observation. Therefore, we
can comfortably rule out the third scenario in Figure 4.10. To explore this idea further, we performed EIS experiments under two conditions: a) when the electrode surface was decorated with
SAM layers, and b) bare Ag foil. When we compared their capacitive behavior, vastly different
trends were observed. The double layer capacitance ( Figure 4.11a) was observed to rise at all
CTAB concentrations on the bare Ag foil. The double layer capacitance (Cdl) was observed to
decrease and then rise with CTAB concentrations for 4-MBN bound Ag foil ( Figure 4.11b). The
different capacitance-concentration profiles can be explained by considering two different CTAB
interaction mechanisms in the presence and absence of dipolar 4-MBN on Ag surface. We propose Br− mediated CTA+ adsorption mechanism on bare Ag due to strong covalent interaction
of Br− with Ag. This kind of counterion mediated surfactant adsorption has been reported earlier
[119, 124–127]. In the case of bare Ag foil, probably at extremely low surfactant concentration,
the surface gets covered with Br− and subsequently surfactants and the increase in capacitance
thereafter is due to increase in ionic strength of the electrolyte. On the other hand, similar Br−
55

adsorption is unlikely on SAM modified Ag because of scarcity of active sites due to the presence
of the SAM. Under these circumstances, the CTA+ interaction would be more favorable due to
possible hydrophobic and favorable ion-dipole interaction between cationic headgroup and nitrile
group. Some additional EIS control experiments were performed with varying concentrations of
NaBr that further validated our proposed model. The Cdl of 4-MBN covered Ag was weakly influenced by NaBr. Conversely, both for NaBr and CTAB on bare silver foil Cdl kept increasing with
concentration ( Figure 4.11a). This observation supports our proposed model ( Figure 4.11c) about
bromide mediated cationic surfactant adsorption on bare Ag foil. On the other hand, very different
capacitance-concentration profile ( Figure 4.11b) for NaBr and CTAB on 4-MBN modified Ag foil
most likely originated from direct CTA+ interaction ( Figure 4.11c) in presence of nitrile probes.
With the third motif in Figure 4.10c ruled out, we discuss the other two. Our measurements
of the CN frequency shift at high surfactant concentrations is consistent with the first motif ( Figure 4.10a). If the second structure ( Figure 4.10b), with the tails pointing towards the surface were
true, the surfactant head groups would be held at a large distance from the vibrational probe. The
counterions would lie adjacent to the head group. The longer the chain length, the farther away
the surfactant head would be held from the surface. Based on the electrostatic model above and
assuming 1e/nm2 density, the second motif would only produce frequency shift of 0.4 cm−1
for
OTAB, and even less for DTAB and CTAB (see SI). This is in contradiction with experimental observations. Furthermore, if this motif was correct, the shorter surfactants would hold the positive
charges closer to the surface compared to longer ones. Therefore, a chain length dependence in the
final value of frequency shift would be expected. This is also in contradiction with observations,
which show a chain length-independent 4.5cm−1
shift for all cationic surfactants ( Figure 4.5).
These observations are consistent with the head-to-surface model, or the first motif in Figure 4.10.
Within this model, the heads of the surfactants point toward the probe molecule, while the counterions are held either in the bulk or partially penetrating the surfactant layer. Within this model,
the nitriles perceive the near field of the positively charged head groups. However, even for the
shortest of the surfactants, the counter ion location is adequately far such that it is perceived mostly
56

0 20 40 60 80 100
200
400
600
800
1000 Capacitance (F cm-2
) Concentration (mM)
NaBr
CTAB
- 1.0V vs Ag/AgCl
Ag Foil
0 20 40 60 80 100
80
120
160
200 Capacitance (F cm-2
) Concentration (mM)NaBr CTAB
Ag - MBN
- 1.0 V vs Ag/AgCl
-
-
-
-
-
-
-
-
-
+
-
-
-
-
-
-
-
-
-
-
-
-
In absence of SAM layer In presence of SAM layer
(a) (b)
(c)
Figure 4.11: Comparison of NaBr and CTAB induced change of Cdl on (a) Ag foil and (b) Ag foilMBN (c) Proposed model for surfactant accumulation in absence and in presence of 4-MBN on the
Ag surface.he positively charged surfactant head is represented as a blue circle with hydrophobic
tail attached to it and the negatively charged counterion is represented as a red circle.
57

as a far field effect. This is similar to the two discrete sheets of charges discussed above. The far
fields of the surfactant heads and the counter ions cancel each other. But, the near field effect of
the surfactant heads remain. For a surface concentration of 1e/nm2
and a counter ion location at
the tail end of the surfactant (after accounting for the estimated solvation radius of the ion based
on the literature)[128], this near field profile can be estimated using the approach outlined in the
previous section.
Since our probe senses the near field of the interfacial ions over length scales smaller than 1 nm
as inferred from the electrostatic model, the probe cannot sense if any ionic structure beyond this
distance away from the interface is formed. For example, it is possible that at higher concentrations,
surfactant bilayers could form at the surface. This can impact electrochemical behavior such as
ionic transport. However, the myopic nature of our probe only senses the field near the surface.
One may expect the EIS study to allude to such larger scale ionic structure. It is possible that
part of the decline of the capacitance with increasing concentration is due to bilayer formation,
since bilayers will effectively correspond to a capacitor with larger plate separation Figure 4.6.
However, we cannot be certain about this effect from the EIS data. It is desirable to design local
probes that protrude away from the surface to explore such ionic structure.
In our electrostatic model we did not include water because we are unsure about the geometry,
and electrostatic properties of the hypothesized water or water layers. Understanding interfacial
water structure on a hydrophobic surface (SAM layer) in the presence of hydrophobic cations
is a phenomenally complicated experimental question to answer with certainty, as is very wellknown and heavily debated within the water structure community and those interested in water near
electrodes [129–131]. However, as can be seen from our model, the field experienced by the nitriles
is a near field and therefore the surfactant head groups cannot reside very far from the nitriles. This
suggests that if there was any hypothesized water layer between the nitriles and the surfactant head
groups, its maximum thickness could not exceed 1 nm, because no field would be sensed by the
nitrile at that distance ( Figure 4.9). However, a more reasonable limit of thickness is likely 0.7 nm.
Based on previous literature on the hydration shell of the surfactant head group we have kept the
58

minimum distance between the N of the nitrile probe and ammonium head group to be 0.4 nm.[104,
105] Figure 4.12 shows the field profile corresponding to OTAB, DTAB and CTAB as a function
of changing the distance between the N of nitrile probe and the positive charge emulating the
surfactant head. As can be seen, for similar surface coverages, corresponding to high concentration
limit of the experimental results, all of the cationic surfactants in the series are expected to produce
the same field. In addition, the field values at the location where the nitrile group is estimated to
reside, within the Stark shift approximation, roughly corroborate experimental results.
Therefore, both vibrational spectroscopy and EIS, support the model in Figure 4.10a, in which
the surfactant head group is pointing towards the surface. An implication of this finding is that
when important reactants or reaction intermediates with dipoles similar to that of nitrile (e.g. intermediates in CO2RR) adsorb on a surface, surfactants may orient themselves with their positive
head groups towards that adsorbed dipole.
Here we address the important question of how to interpret the frequency shifts in terms of field
values. To claim, for example, that the field value under the surfactant layer is a given number of
V/nm, would be misleading, since it would imply that a homogeneous field of that value resides
under the layer. Even the simplest estimates of the field profile, as done in our model, show that
the field is not homogeneous along the body of the molecule ( Figure 4.12).
Therefore, assigning a single value to the field is not appropriate. Nevertheless, it does not
take away from the utility of the Stark shift reporter in revealing the electrostatic environment
of the interface. The spatial resolution of the field is limited by the size of the reporter, that
means the reporter cannot pinpoint the field to within spatial resolution much smaller than its body.
We have emphasized the above in our previous work, and it is imperative for understanding and
contextualizing the current work as well.[1] The probe molecule simply reports a frequency shift.
From that frequency shift, it is impossible to completely infer the field distribution to resolution
smaller than its size. For example, a benzonitrile molecule subjected to a homogeneous field
may produce a frequency shift of -4.5 cm−1
. From the linear Stark shift approximation, this would
correspond to a field of -1.25 V/nm. It is also possible that the molecule may produce the same -4.5
59

- +
~ 0.4-0.6 nm
Enearfield ≠
Efarfield =
Figure 4.12: Calculated field vs z for CTAB, DTAB and OTAB where head groups are facing the
nitriles. The field values for the three surfactants of different chain lengths overlap one another.
The zoomed in part illustrates that the field value along the body of the probe molecule is inhomogeneous.
cm−1
frequency shift if the field profile varies along its body, and most importantly along its long
axis. The latter is likely the scenario in our case, as shown in Figure 4.12. In the second situation,
the applied field is effectively averaged by the inherent electronic structure of the molecule, and
the net result manifests itself as the frequency shift. Hence, one may at best conclude that the
field in the second scenario, when averaged by the molecule, produces the same shift as if the
molecule resided in a homogeneous field of -1.25 V/nm. Further electronic structure studies of
this molecule, especially in inhomogeneous fields, would help in making the interpretation of
the frequency shifts more quantitative. Note that this should not be considered a deficiency or a
drawback of the method. Rather, expecting a single observation of a frequency shift to report on
both field value and the spatial profile of the field separately is unreasonable. All this concern
aside, observation of the frequency shift using such probe molecules is still very useful, provided
that the interpretation is carried out with caution.
60

4.4 Conclusion
In conclusion, we demonstrate how two different types of measurements of interfacial electrostatics can be brought together to understand the field created by aggregation of surfactants near
an electrode. We observe that cationic surfactants, in particular, CTAB produce large fields with
effective average value of -1.25 V/nm near the surface, which could be the reason for their effectiveness in CO2RR and in HER suppression. Both the experimental work and the electrostatic
model shows that to design the most effective cationic surfactants, one must focus on enhanced
surface concentration, and longer chain lengths to keep the counter ions farther away from the surface. Surfactants with partly unsaturated chains for better stacking, or even laterally polymerizing
the adsorbed surfactants, are possible solutions to enhance surface concentration. Using zwitterionic surfactants, where the anion-cation distance can be controlled is also a promising approach.
We hope that our work can inspire further studies of this kind, along with application of surfactants
for field design and enhancement near electrodes.
61

Chapter 5
Mechanistic Insights about Electrochemical Proton-Coupled
Electron Transfer Derived from a Vibrational Probe
This chapter is based on the publication by S. Sarkar, A. Maitra, S. Banerjee, W. R. Lake, R. E.
Warburton, S. H. Schiffer and J. Dawlaty, ”Mechanistic Insights about Electrochemical ProtonCoupled Electron Transfer Derived from a Vibrational Probe”, J. Am. Chem. Soc., 143, 8381-
8390 (2021). All of the computational investigations are performed by our collaborators Robert
Warburton and William Lake from Prof. Sharon Hammes Schiffer’s group at Yale university.
Figure 5.5 and Figure 5.7a are also provided by the theoretical collaborators.
5.1 Introduction
Reaction of a proton and an electron is one of the most elementary chemical processes. At the
electrode surface, this reaction produces an adsorbed hydrogen atom and is called the Volmer reaction,[132] which is the precursor to electrochemical hydrogen evolution. Beyond the Volmer
reaction, proton-coupled electron transfer (PCET) is also an elementary step for a variety of electrocatalytic reactions,[133–137] including water oxidation,[138] CO2 reduction,[139–141] and alcohol oxidation in fuel cells.[142] Therefore, understanding the transfer of a proton across the
electrical double layer (EDL) and its discharge at an electrode surface is of both fundamental and
practical importance.
62

Although proton discharge from water or H3O
+ is significant and prevalent for many electrocatalytic reactions, the complex hydrogen bonding network in an aqueous environment makes
direct spectroscopic characterization of the Volmer reaction challenging. Using a non-aqueous
proton donor that mitigates these hydrogen bonding effects can help elucidate properties of the
interfacial environment that are relevant to catalysis. Organic ammonium salts in non-aqueous solvents have been used for this purpose[143] and have revealed the importance of nuclear quantum
effects in interfacial PCET reactions.[143, 144] However, revealing the electrostatic environments
that are experienced by the proton donor in the EDL remains challenging even in non-aqueous
solutions. A greater understanding of the electrostatic profile and proton transport across the EDL
is critical for the rational control of interfacial electrochemical reactions.
Mechanistic understanding of interfacial PCET is often derived from current–voltage measurements.[134, 143, 145] While such techniques can be quite powerful, spectroscopic measurements can provide additional mechanistic details of electrochemical reactions.[146] Operando
spectroscopy, in particular, can aid in the detection of chemical or physical changes near the
electrode prior to the onset potential; such insights cannot be obtained from current-voltage relationships alone.[147, 148] For example, electrochemical impedance spectroscopy is sensitive
to capacitance but averages the behavior over the entire double layer and is not sensitive to the
molecular details near the electrode.
Vibrational spectroscopy can be a useful tool to elucidate the electrostatic environment of the
electrochemical interface by using electrolyte molecules as vibrational probes of the local electric field.[92, 96, 148–158] In particular, benzonitriles are vibrational Stark-shift probes wherein
the nitrile vibrational frequency is modified by changes in the electric field experienced by the
molecule. Benzonitriles have been used as probes of electrostatics and hydrogen bonding within
enzyme cavities.[22, 32, 92, 149, 159, 160] Moreover, they have been previously used by us to
reveal the details of the structure and polarization of a range of interfacial environments. We have
used benzonitrile probes to measure the interfacial dielectric solvation reaction fields,[5] interfacial
fields in the presence of aqueous electrolytes and applied potential,[34] fields in the presence of
63

surfactants,[13] and fields and hydrogen bonding in the presence of ionic liquid solutions.[1, 161,
162]
Our recent computational work[163] illustrated the variation of the electrostatic potential on a
sub-molecular length scale across a benzonitrile vibrational probe attached to an electrode. This
computational work showed that the nitrile probe vibrational frequency reports changes in the
electronic structure of the molecule arising from the electrostatic environment. More specifically,
the polarizable benzene ring of the benzonitrile contributes to the sensitivity of the nitrile stretch
to the applied potential.
To further understand the Volmer reaction in the context of the structure and dynamics of
the interface, we designed an experiment in which a tertiary amine proton donor is tagged with
a benzonitrile vibrational Stark-shift probe. The proton donor, 4-[(dimethylammonium)methyl]
benzonitrile (abbreviated here as MAMBN-H+), is shown in Figure 5.1. The benzonitrile tag
is crucial because the protonated reactant (MAMBN-H+) and deprotonated product (MAMBN)
forms have distinct nitrile vibrational frequencies, providing a spectroscopic signal that reports on
the protonation state of the donor. Moreover, the benzonitrile groups of both MAMBN-H+ and
MAMBN molecules can Stark shift if they reside within the polarizing environment of the EDL.
Therefore, changes in the nitrile stretching frequency can provide spectroscopic evidence about the
presence of reactants and products as a function of applied potential, in addition to whether these
molecules reside within or outside of the EDL. Installation of a well-defined vibrational probe on
the reactant rather than covalently bound to the surface reveals crucial electrochemical phenomena
that would not be detectable by electrochemical measurements alone or by direct spectroscopy of
the mode involved in proton transfer.
Our aim is to describe changes in the interfacial distribution of reactants and products in the
EDL with applied potential during the Volmer reaction, including the effects of solvation and interfacial field on MAMBN-H+ proton donors and orientation of reactants and products relative to the
electrode surface. These insights are relevant to broader fundamental electrochemical challenges
64

such as understanding the desolvation threshold for ionic species, the electrostatics of the interface, and electrode processes prior to the current onset. Moreover, this fundamental knowledge
has implications to other fields. For example, ion intercalation in battery electrodes is impeded
by desolvation energy,[164] modification of the EDL by surfactants can suppress hydrogen evolution,[79, 165] and electrosorption of molecules prior to the thermodynamic onset is relevant to
underpotential deposition phenomena.[166, 167]
In this work, we use operando surface enhanced Raman spectroscopy (SERS) to measure the
nitrile stretch frequency as a function of potential. These measurements are augmented by computational modeling to describe the protonated ammonium (reactant, MAMBN-H+) and the deprotonated amine (product, MAMBN) near a charged Ag electrode in dielectric continuum solvent.
A schematic representation of the process studied herein is shown in Figure 5.1. Using insights
from experimental measurements and theoretical computations, we report three major findings.
First, we report the desolvation threshold potential for the proton donor and its entry into the EDL.
Second, we identify the appearance of products prior to the onset of steady state electrochemical
current, indicating that the reaction can start below the onset for steady state current and yield a
static population of products that cannot turn over continuously. Finally, we find that the product
remains within the EDL after it is produced and continues to exhibit a Stark shift. The relevance
of each finding to the broader range of electrochemical problems will also be discussed.
5.2 Experimental Methods
To prepare the ammonium salt, concentrated HCl was added to 4-(dimethylaminomethyl)benzonitrile
(MAMBN) obtained from Combi-Blocks. This resulted in formation of a white precipitate of protonated form of 4-(dimethylammonium methyl)benzonitrile (MAMBN-H+) which was air dried.
The electrolyte comprised of 100 mM of the reactant and 300 mM of tetrabutylammonium hexaflurophosphate (Sigma Aldrich) dissolved in DMSO, which is the ideal choice for solvent as it
can dissolve both the reactant and the product. Tetrabutylammonium hexaflurophosphate was used
65

Figure 5.1: . Schematic of the PCET reaction studied using surface enhanced Raman spectroscopy.
The tertiary ammonium proton donor with a benzonitrile vibrational tag enters the electric double
layer and discharges its proton to the electrode, yielding the amine form. The nitrile probe stretch
frequency reports on the details of this process, including the entry of reactants into the EDL, the
electric field within the EDL, and adsorption of the products to the surface.
as supporting electrolyte. Silver foils of 0.1 mm thickness and 99.9% purity (Sigma Aldrich) were
etched electrochemically and used as the working electrode. Electrochemically etched silver was
used as the working electrode because it is an ideal substrate for surface enhanced Raman spectroscopy (SERS).46 The reference electrode was Ag/AgCl (Gamry), and a Pt wire was used as the
counter electrode. A Gamry Reference 3000 potentiostat was used for step scan voltametry. Etched
Ag electrode, reference electrode and Pt wire were placed in-side a 3 mL quartz cuvette. A laser
source from Ocean Optics Inc. emitting 532 nm was used to excite the SERS active Ag substrates.
A Raman probe from InPhotonics held in longitudinal alignment was used to excite and collect
backscattered Raman signals from the samples. The Raman signals were sent to a spectrometer
(HORIBA iHR320) with 1800 g/mm gratings with a CCD camera (Syncerity) for spectral analysis. SERS spectra were collected as a function of applied potential. Applied potential was scanned
from 0 V to -1.2 V relative to Ag/AgCl reference electrode. Since all experiments were performed
in the reductive range of potentials (0 to -1.2 V vs Ag/AgCl), formation of oxide on silver is not
expected. After application of each potential step, a transient capacitive current was observed. This
transient current was allowed to decay prior to acquiring SERS spectra under steady state electrode
condition. At each potential, three SERS spectra were collected with 30 seconds integration times.
66

The nitrile stretch frequencies were retrieved by fitting the nitrile peaks to Gaussian line-shapes
using MATLAB fitting toolbox.
5.3 Computational Methods
This section is contributed by Prof. Sharon Hammes Schiffer’s group at Yale university.
We performed periodic density functional theory (DFT) calculations with the dispersion-corrected
PBE-D3 functional in Quantum ESPRESSO.[168–170] Additional calculations were performed
using the PBE50 functional, which demonstrated similar trends. Solvent effects were described
using a dielectric continuum model via the Environ patch in Quantum ESPRESSO.[171] In this
approach, a modified Poisson equation is solved self-consistently with the Kohn-Sham equations to
determine solvation contributions to the total energy. A dielectric constant of 47 was used to model
DMSO.[172] The electrode potential was modified by varying the number of electrons in the system, which is compensated by a homogeneous background charge. The potential was determined
according to previously described methods.[145, 163, 173, 174] Four-layer Ag(111), (100), or
(322) surface slabs were employed as models for the Ag electrode. The nitrile stretching frequencies were calculated using a grid-based approach that includes anharmonicity.[163] In this implementation, the grid for the surface-molecule system was generated along the approximate normal
mode vector corresponding to the nitrile stretch. A one dimensional potential energy curve was
obtained by performing single-point energy calculations at regular intervals along this grid. Subsequently, the Fourier grid Hamiltonian method was used to solve the one dimensional Schr ¨odinger
equation for the nitrile stretch mode represented by this potential, and the nitrile stretch frequency
was obtained from the difference between the lowest two vibrational energy levels. For each geometry studied, this method was repeated at different electrode potentials to investigate the vibrational
Stark effect. Additional computational details are provided in the Supporting Information of [175].
67

5.4 Results and Discussions
We first describe the nitrile vibrational frequencies of the protonated (MAMBN-H+, reactant) and
deprotonated (MAMBN, product) forms of the proton donor. The spectra of neat MAMBN-H+
(solid) and neat MAMBN (liquid) are shown in Figure 5.2a. The nitrile stretch of MAMBN-H+ is
blue shifted significantly compared to that of MAMBN. This shift is expected and consistent with
previous observations of Stark shift of benzonitrile self-assembled monolayers.[1, 34, 96, 176] A
positive charge at the terminal amine produces an electric field at the nitrile similar to that produced by an oxidizing potential, resulting in a blue shift of the nitrile stretch frequency as reported
previously.[1] Upon dissolving the molecules in DMSO, the nitrile frequencies of both MAMBNH+ and MAMBN experience red shifts Figure 5.2b, which is the expected solvatochromic shift in
a high dielectric constant medium.[5] It is also seen that MAMBN-H+ experiences a much larger
red shift compared to MAMBN, indicating that the positive charge of the ammonium is heavily
screened by the solvent. Figure 5.2c shows the nitrile stretch frequencies of MAMBN-H+ and
MAMBN in the presence of a silver electrode at open circuit potential. Both reactants and products experience a red shift with respect to their dissolved forms in the absence of the Ag electrode.
This likely arises due to the participation of the metal in the solvation environment of the molecule,
consistent with our previous work on the dielectric solvation near a metal interface.[5]
Calculations of the nitrile frequencies are qualitatively consistent with the above observations
in that they show a blue shift from the product to the reactant and a red shift of both peaks upon
solvation (Supporting information, [175]. These trends were observed with both plane wave and
localized Gaussian basis sets using four different functionals. This consistency between the theoretical calculations and the experimental data suggests that the level of theory used in the DFT
calculations is appropriate to capture key trends in the nitrile vibrational frequencies. To summarize, the data in Figure 5.2 show that the protonated MAMBN-H+ is always blue shifted relative to
the deprotonated MAMBN, and that both are subject to solvatochromic red shifts in the bulk and
near the electrode. These spectra serve as a baseline for interpreting the spectroelectrochemical
data, which is discussed next.
68

Figure 5.2: . Nitrile stretch spectra of the reactant MAMBN-H+ (blue) and the product MAMBN
(red) (a) in pure form, (b) dissolved in DMSO, and (c) in presence of Ag electrode at open circuit
potential. The nitrile stretch peak of the product is always red shifted compared to that of the
reactant.
69

Figure 5.3 shows the spectral evolution of the nitrile stretch region as a function of applied
potential during the forward ( Figure 5.3a.) and reverse ( Figure 5.3c.) scans. Figure 5.3b. shows
the applied potential and the steady state electrochemical current. The onset of electrochemical
current is near –0.7 V vs Ag/AgCl and is in close agreement with the triethylammonium systems
in previous studies.[143] The frequencies associated with the centers of the dominant two peaks
are shown in Figure 5.4.
Below we will use Figure 5.3 and Figure 5.4, along with input from computational work,
to describe the desolvation of MAMBN-H+, its entry into the EDL, and its eventual conversion
into MAMBN following interfacial PCET. To guide the analysis, we discuss the characteristic
processes in three distinct potential ranges.
5.4.1 Reactants outside the double layer (0 to −0.4 V)
For the most oxidizing potentials in Figure 5.3 (0 to −0.4 V vs. Ag/AgCl), only a single peak
is observed, corresponding to the reactant MAMBN-H+. Previous work by us and others has
shown that self-assembled monolayers of covalently bound benzonitrile are within the electrostatic
influence of the electrode and exhibit a significant Stark shift in this potential range.[1, 176] In the
present work, however, this peak shows no Stark shift, indicating that the molecules do not feel the
polarizing influence of the electrode in this potential range. Therefore, we propose that the nonStark shifting peaks in this potential range arise from MAMBN-H+ molecules that are outside the
EDL, fully solvated, and shielded from the polarizing influence of the electrode. Note that SERS
can detect molecules outside of the EDL because the penetration depth of the SERS technique into
the solution is several nanometers,[177] while the characteris-tic thickness of the EDL based on a
continuum approxima-tion for DMSO is 4–8 A˚.[163, 178] This observation further highlights the
utility of the Stark reporter, where even its lack of frequency shift is informative and can report on
the location of the molecules relative to the EDL.
70

Figure 5.3: Potential-dependent evolution of the nitrile stretch spectra during the (a) forward and
(c) reverse scans. (b) Steady state current versus potential, where the horizontal dotted line at
−0.7 V indicates the onset potential for steady state current. The vertical offset of the spectra in
panels (a) and (c) is chosen such that each spectrum is aligned with their corresponding potential
value in (b). The blue and red dashed lines in (a) and (c) correspond to the Stark shifts attributed
to MAMBN-H+ and MAMBN, respectively. In the forward scan, initially only the nitrile peak
corresponding to the reactant MAMBN-H+ is present. It does not undergo Stark shift, indicating
that the reactants are outside the EDL and are not polarized by the electrode. At −0.4 V, a second
peak around 2200 cm−1
corresponding to the product MAMBN begins to appear, indicating entry
of the reactants into the EDL and formation of products at a potential below the onset of steady
state current. Both peaks undergo Stark shift to lower frequencies with more negative potential,
indicating that they arise from species within the EDL because they respond to the polarization
exerted by the electrode. The reverse scan shows that the reaction is reversible. Frequency shifts
as a function of potential are shown in Figure 5.4.
71

5.4.2 Entry into the double layer and the initial reaction (−0.4 to −0.7 V)
At potentials more negative than −0.4 V, the nitrile frequency of the reactant MAMBN-H+ Stark
shifts to lower values ( Figure 5.3a and Figure 5.4). We propose that at this potential the reactant
MAMBN-H+ molecules partially shed their bulk solvation shell and enter the EDL, where they
experience the polarizing influence of the electrode. Thus, −0.4 V marks the desolvation potential
for the protonated form in DMSO. The desolvation threshold of ions near a surface is critically
important to many electrochemical processes. A prime example is desolvation of ions in batteries
prior to intercalation, where the cost of desolvation is especially high for multivalent ions.[164,
179] Our observation of the desolvation threshold using a vibrational reporter could be applicable
to such problems.
Figure 5.4: Nitrile frequency of the reactant MAMBN-H+ (shown in blue) and product MAMBN
(shown in red) as a function of applied potential during forward and reverse scans. The grey shaded
area represents the region where a steady electrochemical current is measured.
Since the entry and exit of the reactants and products into and out of the EDL is central to
our study, we will briefly highlight the definitions and approximations that go into this understanding. The EDL is the ionic structure near an electrode that forms in response to the applied
potential, which, in turn, screens the applied potential. Within the EDL, the electrostatic potential
decays with distance from the electrode surface,[178] and the resulting electric field can polarize
molecules such as Stark-shift probes used in this work. As mentioned above, the Stark response of
72

the benzonitrile probe is influenced by the polarizability of the entire benzene ring conjugated to
the nitrile.[13]
For our experiments, the characteristic length of the EDL ( 4–8 A˚) is comparable to the size
of the molecule ( 10 A˚). Furthermore, when the probe is oriented with the nitrile pointing away
from the electrode and the ammonium end toward the electrode (i.e., prior to the onset of the
Volmer reaction, Figure 5.1), the nitrile bond will reside far from the most polarizing region.
Nonetheless, we will demonstrate that the effect of the field will be communicated to that bond via
the polarization of the benzene ring.
At −0.4 V, a new peak around 2200 cm−1
appears and is red shifted compared to the reactant peak. We assign this peak to the deprotonated product MAMBN based on experimental and
computational evidence, as explained below. The first piece of evidence for assigning this peak
to MAMBN is that this peak continues to exist at more negative potentials, where equilibrium favors the products of the reaction. Second, the frequency is red shifted relative to the reactants, as
expected based on the discussion above ( Figure 5.2 and Supporting Information, Table S-T1 in
[175]). However, the red shift is much larger (20 cm−1
) than in the bulk solution ( 4 cm−1
). This
difference indicates that upon formation, the product is closely interacting with the silver electrode
surface, potentially chemisorbing through the lone pairs of the amine, as discussed further below.
We investigated the interactions between the electrode and MAMBN/MAMBN-H+ with computational methods. To account for the polycrystalline Ag electrode used in the experimental portion of this work, we considered several different surface models. In particular, DFT calculations
of MAMBN and MAMBN-H+ were carried out near three different Ag surfaces. Figure 5.5 shows
optimized geometries of these molecules near the Ag(100) surface. These calculations employed a
4x4 unit cell, corresponding to a coverage of 0.72 molecules nm−2
. We performed similar analyses
on close-packed Ag(111) and stepped Ag (322) surfaces. Further details are provided in the SI of
[175], Figures S2–S4.
On Ag(100), we identified a local minimum for chemisorption of MAMBN at a Ag–N distance
of rAg−N ∼2.5 A˚ ( Figure 5.5a). This distance is consistent with previously calculated bond lengths
73

of amines on Ag surfaces.[180, 181] The presence of a covalent Ag–N bond in this configuration is
further supported by the charge transfer between Ag(100) and the amine, as indicated by the charge
density difference relative to the potential of zero charge (PZFC) ( Figure 5.5a). A covalent bond
is also supported by the observed population of Ag–N bonding orbitals below the Fermi level,
as determined from crystal orbital Hamilton population analysis (SI, Figure S5 of [175]). The
adsorption energies for chemisorbed MAMBN are calculated at PZFC using an uncharged unit
cell. The adsorption energy of −0.79 eV suggests that the bond between MAMBN and Ag(100) is
energetically favorable. The adsorption energy was insensitive to higher MAMBN coverage with
a computed adsorption energy of −0.77 eV using a 3×3 unit cell.
Figure 5.5: Interactions between Ag(100) and either product MAMBN or reactant MAMBN-H+
molecules. (a,b) Product MAMBN (a) chemisorbed to and (b) physisorbed near Ag(100). (c)
MAMBN-H+ physisorbed near Ag(100). Charge density difference isosurfaces are shown with
respect to the electrode PZFC, where the cyan and yellow indicate increases and decreases in
electron density, respectively; the isosurface level is 0.00025 e–
/Bohr3
. Ag atoms are grey, N
atoms are light blue, C atoms are brown, and H atoms are light pink. Further details are provided
in the SI of [175].
We find that MAMBN adsorption is modestly stronger on Ag(322) steps (∆Eads = −0.98 eV,
Figure S6 of [175]). This trend aligns with the decreased bond order of Ag atoms on steps.[182,
183] We also find that MAMBN can interact with Ag(100) through π interactions between the
benzonitrile and the surface at a distance of ∼ 3.3 A˚ (SI of [175], Figure S4). However, this
structure does not correspond to the product directly after proton transfer from MAMBN-H+.
74

The MAMBN-H+ reactant must be oriented in a similar manner as depicted in Figure 5.5 to
transfer its proton to the Ag electrode. In contrast to MAMBN, the MAMBN-H+ reactant does not
chemisorb to the surface, as indicated by a chemical bonding analysis (Supporting information of
[175], Figure S7). This lack of chemisorption is attributed to the lack of a lone pair on the fully
coordinated sp3 nitrogen. These calculations point to the likelihood that, following desolvation of
the reactant MAMBN-H+ and product formation at −0.4 V vs. Ag/AgCl (see Figure 5.3), the
product MAMBN may be directly interacting by chemisorption with the electrode surface.
The next piece of evidence for the assignment of the 2200 cm−1 peak to adsorbed MAMBN on
the surface comes from a control experiment, where only the deprotonated MAMBN was subjected
to the spectroelectrochemical study. The results are shown in Figure 5.6. Since this experiment
is carried out in the absence of protons, the Volmer reaction is not possible, and thus the peaks
for the solvated MAMBN are observed until the potential is biased more negative than −1.0 V vs.
Ag/AgCl. This observation indicates that unlike the positively charged MAMBN-H+ molecules,
which are more electrostatically attracted to a negatively charged electrode, these molecules remain
outside the EDL up to a much more negative potential. Below −1.0 V, the Stark shifting 2200 cm–1
peak, which was also seen in the Volmer reaction in refFig3, begins to appear. This observation
suggests that MAMBN can be drawn from the solution and adsorbed on the silver electrode. In
the case of the electrochemical Volmer reaction, however, the charged MAMBN-H+ is first drawn
to the surface, and the deprotonated product is readily generated near the surface and available
for adsorption. Based on the experimental and computational evidence discussed above, we can
confidently assign the 2200 cm–1 peak to the chemisorbed MAMBN product on the surface.
Figure 5.3 and Figure 5.4 show that the product peak appears at around −0.4 V, less negative
than the onset potential for sustained steady state electrochemical current at −0.7 V. The explanation for this observation is revealed by spectroscopy and otherwise would remain hidden with
electrochemical measurements alone. The appearance of the product peaks at less negative potentials than the onset for steady state current indicates that soon after entry into the EDL, PCET from
some reactant MAMBN-H+ molecules has started to occur, turning over to the product MAMBN.
75

Figure 5.6: Potential dependent nitrile peak in a control experiment, where only the product
MAMBN was present. The peak around 2227 cm–1 does not Stark shift over a very large range
of potential, indicating that the neutral MAMBN is not drawn into the EDL easily. It is not until
a relatively high negative potential of ∼ –1.0 V that a second Stark-shifting peak corresponding
to the adsorbed MAMBN is observed. This peak appears at a much less negative potential when
MAMBN is not drawn from the solution, but rather is produced by the Volmer reaction from
MAMBN-H+, as seen in Figure 5.3.
76

This results in an adsorbed population of products on the surface. However, this population is
stationary and does not turn over to yield a sustained electrochemical current. The net current
due to a single layer of such population on the electrode surface is too small to be observable
electrochemically at slow scan rates and can go undetected. However, spectroscopic observation
of the surface reveals a buildup of this stationary population prior to the onset and sheds light on
this otherwise hidden reaction. The spectroscopic signature of the new species observed between
−0.4 and −0.7 V is consistent with a deprotonated molecule. Furthermore, the DMSO solvent is
unlikely to be protonated, and the underpotential deposition (UPD) of hydrogen atoms on metals
is well-known.[166] Thus, the most likely scenario is that the proton is adsorbed on the surface as
a hydrogen atom after PCET from the donor molecule. Interestingly, this peak slightly diminishes
in intensity after the onset potential, indicating that fewer MAMBN molecules are built up near the
surface once the reaction turns over and is generating steady state current. In that scenario, fresh
positively charged reactants are continuously brought to the negatively charged surface, displacing
the adsorbed products built up near the electrode surface. It is important to find out whether the
−0.4 V potential at which products begin to appear spectroscopically is above or below the thermodynamic equilibrium potential E0
for the reaction. Appearance of a monolayer or a few layers
of product prior to the equilibrium potential is known as underpotential deposition in electrochemistry and is very well studied.[166] To identify the E0 of the reaction, we conducted a series of
open circuit potential measurements on a related amine (see C),[184] and determined that E0
is
near –0.06 V vs Ag/AgCl, which is far from the −0.4 V value. Therefore, our observation of the
products at −0.4 V is not due to UPD. It is deposition at more negative potentials than E0
, but less
negative than the steady current threshold. We conjecture that such subthreshold deposition may
be more common than currently known because its electrochemical signature or current is rather
small and easy to miss, especially at slow scan speeds. The above analysis shows that the potential
for desolvation and formation of the stationary layer can be a significant fraction of the overpotential of the reaction. We hope that our study can inspire spectroscopic searching for adsorbed
77

reaction products prior to current onset in a larger range of reactions to test the generality of this
proposal.
5.4.3 Current onset and continued Stark shift of the reactants and products
(−0.7 to −1.2 V)
Finally, we will discuss the Stark shift of both the products and reactants starting from −0.4 V
and continuing all the way to the end of our scan at −1.2 V. A potential of −0.7 V corresponds
to the onset of steady state current corresponding to hydrogen evolution, as shown in Figure 5.3
and Figure 5.4. A significant Stark shift is observed for both reactant MAMBN-H+ and product
MAMBN, indicating that both reside within the EDL and therefore are influenced by the electrode. Interestingly, the amount of shift per unit potential is not dramatically different between
the reactant and the product, with MAMBN exhibiting a slightly larger Stark shift compared to
MAMBN-H+ (14 cm−1 versus 10 cm−1 over the potential range of −0.4 to −1.0V).
The comparable sensitivity of the nitrile frequency to applied potential for MAMBN-H+ and
MAMBN can be explained with input from our computational work. The isosurfaces in Figure 5.5
show that upon a potential bias negative of PZFC, the charge density differences across the benzene
ring and the nitrile are qualitatively similar between MAMBN-H+ and both the chemisorbed and
physisorbed MAMBN. Bader analysis[185–187] indicates minimal charge transfer from the electrode to the MAMBN or MAMBN-H+ (see SI of [175], Figure S9). These analyses suggest that the
nitrile frequency of MAMBN-H+ will respond similarly to that of chemisorbed and physisorbed
MAMBN with applied potential. The small differences in these responses are most likely due to
differences in the polarizability of the benzene ring.
Although MAMBN-H+ and MAMBN exhibit similar Stark behavior, the small difference between them that is observed experimentally is also supported by computations, as presented in
Figure 5.7. A possible reason for the slightly larger shift for MAMBN is that it can approach the
electrode more closely and chemisorb, as shown in Figure 5.5. In contrast, MAMBN-H+ cannot
78

chemisorb to the surface. Comparison of the vibrational shifts for MAMBN at two different distances from the electrode (see SI of [175], Figure S8) shows that the Stark shift is greater when
the molecule is closer to the electrode, as expected based on the significance of the polarizable
benzene ring. We also calculate a Stark shift for a configuration where MAMBN interacts with Ag
through π interactions because the charged electrode can polarize charge within the benzonitrile
(see SI of [175], Figure S4).
Our spectra show that the product MAMBN molecules adsorb on the surface after the reaction
with an adequately long residence time to create a spectroscopically observable steady state population. Of course, this population is continuously exchanged for fresh reactant MAMBN-H+ to
maintain steady state current. The adsorbed population consumes some surface area of the electrode and affects the apparent kinetics of the reaction. Our spectroscopic evidence for such product
accumulation near the surface will help inform and augment existing electrochemical kinetic models for this reaction.
Since adsorbed MAMBN will occupy some electrode surface area, it will reduce the availability of surface sites for PCET and impede the reaction. Adsorption and desorption of the reactant
and product, respectively, can be tuned by the choice of the donor molecule, the solvent, or the electrode. In aqueous hydrogen evolution, for example, adsorption of deprotonated donor molecules
can either promote or limit reaction rates.[188]
The reverse scan provides information about whether the product MAMBN is permanently
adsorbed on the electrode surface. While running the reverse scan from –1.2 V to 0 V, both the
reactant MAMBN-H+ and the product MAMBN peaks backtrack their frequency shifts, until the
MAMBN peak disappears, and only the MAMBN-H+ peak remains without any Stark shift. This
observation indicates that the product reversibly adsorbs and desorbs to the electrode, and that the
conversion between reactants and products is consistent between forward and reverse scans, with
no detectable formation of side products. It is also noteworthy that in the reverse scan a significant
amount of the product that was produced at more negative potentials remains visible even after
the current no longer flows, indicating that sub-threshold interaction between the products and the
79

electrode surface discussed earlier is perhaps a function of the potential and not the direction of
the scan.
Finally, we comment on the variation of spectral linewidths as a function of potential. As seen
in Figure 5.3, we observe differences in linewidth in the spectra of reactants residing outside the
EDL and the chemisorbed products. Such differences may arise from variation of the electrostatic
environments for the two cases. Linewidth variation as a function of applied potential was previously observed by us for covalently bound benzonitriles as a function of potential.[1] Given that
previous literature has identified intramolecular sources of damping for the nitrile vibration,[38]
however, we caution against an interpretation of linewidths as solely arising from the electrostatic
variations.
Figure 5.7: Frequency shifts of the reactant and the product from theory (left panel) (corresponding
to the geometries in Figure 5a and 5c) and from experiment (right panel). The reactant frequency
shifts are plotted in blue and the product frequency shifts are plotted in red. The range of potentials
in the experimental work is narrower due to the stability of the electrode. The slopes of the middle
regions of the experimental curves are qualitatively similar to the slopes of the calculated curves.
Note that the calculated frequencies at the most negative potentials are less reliable because of
charge transfer from the electrode to the molecule (see Figure S8 of [175]) and likely proton transfer from MAMBN-H+.
5.5 Conclusion
A summary of the Volmer reaction based on insights from the spectroelectrochemical experiments
and theoretical calculations is presented in Figure 5.8. From our joint experimental and theoretical
80

Figure 5.8: Schematic figure showing the sequence of events for the reaction based on our experimental and computational data. (a) Below −0.4 V, the reactant MAMBN-H+ molecules, which
are solvated by DMSO, reside outside the EDL and do not feel the polarizing influence of the electrode and therefore do not exhibit a Stark shift. (b) Within the potential range of –0.4 V to −0.7 V,
the reactant molecules partially shed their bulk solvation shell and enter the EDL, as evidenced by
their Stark shift. Some of the reactant MAMBN-H+ molecules undergo the Volmer reaction and
form MAMBN, as evidenced by the largely red-shifted peak in our data. Lack of electrochemical
current in this range suggests that this is a stationary population and does not turn over. Computational and experimental observations suggest that MAMBN is chemisorbed on the surface. (c)
At potentials more negative of −0.7 V, the reaction can turn over yielding a steady state current.
Correspondingly, a steady state population of MAMBN-H+ and MAMBN that continue to Stark
shift with potential is observed.
81

analysis, we have categorized three main stages for the progression of the reaction as a function of
applied potential. These interpretations were facilitated by an understanding that molecular species
positioned within the EDL will Stark shift with applied potential, while those outside of the EDL
will not. First, the reactant MAMBN-H+ molecules remain solvated outside of the EDL prior to
the desolvation potential of −0.4 V vs. Ag/AgCl ( Figure 5.8a). Next, after the molecule enters the
EDL, the interfacial Volmer reaction leads to the formation of a stationary population of products
( Figure 5.8b). Finally, at potentials negative of −0.7 V vs. Ag/AgCl, the negatively charged
surface attracts additional reactant MAMBN-H+ molecules, displacing the stationary MAMBN
molecules near the electrode surface and enabling the onset of steady state current ( Figure 5.8c).
By combining a Stark shift vibrational probe with a proton donor, we have revealed several
features of the Volmer reaction through operando vibrational spectroscopy that were not accessible from electrochemical measurements alone. We anticipate that many of the concepts elucidated
by this work are general and applicable to a wide range of electrochemical reactions. Specifically,
formation of a stationary layer of products prior to the onset of steady state current is quite likely to
occur for other reactions. However, its detection by conventional electrochemical methods, especially in slow scans, may not be possible. Surface sensitive spectroscopic probes, such as the one
used here, can reveal such hidden processes. A significant focus in electrochemistry is understanding the onset of reactions and the processes controlling overpotentials. Our finding of separate
onsets for the formation of a stationary layer of products and the attainment of steady current may
inspire electrochemists and surface scientists to further explore and exploit such concepts.
Our work inspires the following design principles for further understanding and optimizing interfacial PCET reactions. The goal is to favor adsorption of the reactant and disfavor adsorption of
the product. For example, the double layer may be optimized by the choice of tailored surfactants
to facilitate entry and alignment of the proton donors during the reaction. In addition, the interface
could be designed to ensure that the discharged carrier does not adsorb on the surface but instead
is solvated by the surfactant.
82

While these experiments were performed on silver electrodes, we believe that several of the
results presented here are generalizable to other metallic electrodes. The adsorption of products
on the electrode surface will vary for different electrodes and proton donors. Understanding the
relationships between different electrode-donor combinations and PCET reaction mechanisms is
an important topic for further study. Moreover, rational design strategies to tailor the desolvation
threshold of proton donors may be developed by tuning the properties of the solvent and electrolyte,
as well as the PZFC of the electrode, to impact the electrostatics of the EDL. We conjecture that
an analogous desolvation threshold may exist for the hydronium ion, although further work is
needed to measure and model such systems. This study highlights the insights that can be gained
by attaching well-defined vibrational probes to electrochemical reactants or catalysts.
83

Chapter 6
Distinguishing between the Electrostatic Effects and Explicit
Ion Interactions in a Stark Probe
This chapter is based on the publication by A. Maitra, P. Das, B. Thompson and J. Dawlaty, ”Distinguishing between the Electrostatic Effects and Explicit Ion Interactions in a Stark Probe”, J.
Phys. Chem. B, 127, 2511-2520 (2023). Pratysha Das from Prof. Barry C. Thompson’s group at
University of Southern California did the synthesis and characterization of the benzonitrile functionalized 15-Crown-5 and 18-crown-6 used in this study.
6.1 Introduction
Knowledge of electric fields in microenvironments, such as protein cavities and electric double
layers, is fundamental to controlling reaction rates and product selectivity. [24, 28, 29, 189] Often,
these fields vary over nanometer length scales, which necessitates molecular scale probes for their
measurement. Vibrational Stark shift probes, such as carbonyls and nitriles, have been successfully
used for this purpose. [30–32, 92, 190] Ideally, vibrational probes are expected to interact with the
surroundings only via electrostatic interactions. They must be chemically inactive and should not
engage in hydrogen bonding or other types of explicit interactions such as Lewis bonding. In
practice, it is found that most probes are not completely innocent to their environments and that
they participate in specific interactions.[191–196]
84

Benzonitrile is a vibrational probe with a sharp absorption at 2230 cm−1 which is well-isolated
from other common organic vibrations. It has been used as a probe for measuring fields in solvents,
protein cavities, and the electrode - electrolyte interface.[1, 5, 13, 35, 36, 197] Just like many other
probes, it is also subject to specific interactions such as hydrogen bonding [100, 197–199] and
other donor -acceptor type interactions[62] which affect its frequency and line width. When just
the electrostatics of the environment is operative, the frequency shift of the probe is purely due to
the polarization of electrons, without involvement of direct orbital overlap with the surrounding
species. In hydrogen bonding, and other type of explicit interactions such as Lewis bonding with
electron-deficient species like BF3 and metal ions, the lone pair orbital of the probe donates partial
charge density to an acceptor. The lone pair electron donation results into a blue shift, as has been
explained previously [2, 195, 200], and mimics a field that is oriented from the nitrogen to the
carbon.
In many cases, such the electrode - electrolyte interface, it is possible to encounter both the
electrostatic and Lewis interactions simultaneously, which makes it difficult to report the local
electric field unequivocally. To understand such issues, it is necessary to study the complexation of
nitrile with ions. While the hydrogen bonding [100, 197–199] and interaction with classic Lewis
acids such as BF3 have been studied before [62], in this work we focus on association of the
benzonitrile lone pair with a large number of metal cations.
An important feature of our work is our choice of the molecules, which allows for observation
of both the long-range electrostatic effect, and the Lewis interactions. Yang and co-workers have
shown that fields from non-redox active metal centers complexed with crown ethers affect nearby
reaction centers.[16, 201, 202] Inspired by these results, we have synthesized two crown ethers
functionalized by benzonitriles (Benzo-15-Crown-5-CN and Benzo-18-Crown-6-CN). We chose
the 15-C-5 crown system because it is very-well documented in the literature and can coordinate
a number of small cations [203–208]. We noted that some of the larger cations (e.g. K+, Eu3+,
Ba2+) did not fit in the crown, as evidenced by a negligible or small shift in the nitrile frequency.
Therefore, we decided to synthesize the larger 18-C-6 crown to accommodate some of the larger
85

cations.[205, 209, 210] The binding constants for both 15-C-5 and 18-C-6 crown systems have
been extensively studied in the literature, [203, 211–214] and they depend on the choice of solvents
and cannot be exactly extrapolated to our work. The concentration-dependent shifts of the nitrile,
discussed later in this paper, show the concentrations of ions that are necessary to saturate the
crown sites. The electrostatic field from the captured ions in the crown cavity is sensed by the
adjacent benzonitrile. Additionally the nitrile lone pair can explicitly interact with some of the
ions. The main purpose of choosing such a designed crown-ether probe system is to identify and
isolate the Lewis interaction from the pure electrostatic effect of metal cations ( Figure 8.2). To
our knowledge, these two interactions of different origins have not been systematically studied
before. We find that the electrostatic and Lewis interactions are quite distinct from each other,
with the latter causing a large shift of 40 - 60 cm−1
. The electrostatic variations show a rough
correlation with the charge of the ion, but are strongly affected by the solvation environment and
the coordination geometry of ions. Explaining the details of the electrostatic variations is the
subject of an upcoming study.
Propylene carbonate was chosen as a solvent for this study for three reasons. First it can
dissolve all of the metal salts and the crown ethers within the necessary concentration ranges used
in this study. Second, it is an aprotic solvent, and does not hydrogen bond to the probe. Finally, it
is a solvent of interest for battery electrolytes. Understanding solvation of ions within propylene
carbonate is highly desirable for battery electrolyte research, such as developing multivalent ion
batteries.[215–217]
Our work will serve two purposes. First, it will help those involved in measuring local fields
in distinguishing between the Stark and Lewis effects in complex environments. Second, it may
inspire further work to use benzonitrile as a probe of the strength of Lewis interactions, and perhaps
as an analytical tool for measuring certain cations.
86

Mn+
Mn+
Lone pair orbital
donated to metal cation
(Lewis Interaction)
Figure 6.1: Representative cartoon for metal cation complex with crown-ether functionalized
benzonitrile in two different locations giving rise to two different effects on −CN vibrational probe.
n= 1 for Benzo-15-Crown-5-CN (B15C5-CN); n= 2 for Benzo-18-Crown-6-CN (B18C6-CN).
6.2 Experimental Methods
All reactions were carried out under dry N2 in oven- dried glassware, unless otherwise noted.
Solvents and inorganic reagents were purchased from commercial sources through VWR and
used as received unless otherwise noted. Anhydrous dichloromethane (DCM) (Sigma-Aldrich)
was purchased and used as received. Potassium carbonate (K2CO3) was dried at 120 °C in a
vacuum oven overnight prior to use. Acetonitrile (EMD), Tetraethylene glycol (Sigma-Aldrich),
Pentaethylene glycol (Sigma-Aldrich), p-Toluenesulfonyl chloride (TsCl) (Alfa Aesar), and 3,4-
Dihydroxybenzonitrile (Sigma-Aldrich) were purchased and used as received.
Crown ethers Benzo-15-Crown-5-CN (B15C5-CN) and Benzo-18-Crown-6-CN (B18C6-CN)
were prepared according to the literature procedure.[218] To a 500 mL three-necked round bottom
flask equipped with a condenser, a mixture of 3,4-dihydroxybenzonitrile (3 ) (1.08 g, 7.96 mmol)
and tosylated tetraethylene glycol (1 ) (4 g, 7.96 mmol) or tosylated pentaethylene glycol (2 ) (4.35
g, 7.96 mmol), K2CO3 (4.4 g, 31.83 mmol) in acetonitrile (CH3CN) (310 mL) were added and
stirred under nitrogen. The reaction mixture was refluxed for 48 hours under N2. Post completion
of the reaction, the above mixture was filtered, and the acetonitrile solvent was removed under
reduced pressure. The organic residue was dissolved in dichloromethane and was washed with
water and brine. The combined organic phase was concentrated to afford B15C5-CN (2.2 g, 94.2%)
or B18C6-CN (2.1 g, 78.2%).
87

Details of the reaction schemes, synthesis of the tosylate precursors (1,2), and NMR characterizations can be found in the Supporting Information ( D).
The FT-IR spectra presented here were recorded for 400 mM of crown ethers with and without various metal salts in a Thermo Scientific Nicolet iS50 spectrometer. A demountable FT-IR
cell from International Crystal Laboratories was used to hold two calcium fluoride windows separated by a Teflon spacer, and FT-IR spectra were acquired in transmission mode with liquid nitrogen cooled MCT detector. Two different thicknesses, 15 and 100 µm are used as spacers in this
work. Highly polar propylene carbonate (dielectric constant: 64.9) has been used as solvent, as
this can dissolve all the metal salts and the crown ethers. Spectra for the metal ions complexed
with crown ethers are obtained using a 1:1 mixture of metal salts and crown ethers. The 4−5
cm−1
red shift of the empty crown systems relative to free benzonitrile in the same solvent is explainable by the polarizing influence of the ether functionalities of the crown. They are electron
donating, and as previous studies have suggested, they result into a red shift.[1] The metal salts
are used as purchased from Sigma-Aldrich. Trifluoromethanesulfonate ( OTf−, [CF3SO3]
−) and
bis(trifluoromethane)sulfonimide (TFSI−, [(CF3SO2)2N]−) are used as counteranions. The OTf−
and TFSI− anions are larger in size and they can remain dissociated from the metal ions. The metal
salts used in this work are LiTFSI, NaTFSI, KOTf, Mg(TFSI)2, Ba(OTf)2, Zn(TFSI)2, Zn(OTf)2,
Zn(Cl)2, Mn(OTf)2, Ni(OTf)2, Cu(OTf)2, Eu(OTf)3, Y(OTf)3, Yb(OTf)3 and In(OTf)3. The spectra were baseline corrected. To study the counteranion dependence on frequency shifts of nitrile,
we have used a 1:1 concentration ratio for Zn2+ salts to the probe except for Zn(TFSI)2, where a
1:2 concentration ratio of crown ether to salt was used to get better signal.
The estimated error in frequency measurements was ≤1 cm−1
. The error in the measured volume using VWR micropipettes was 0.2% and the error in the analytical balance while measuring
the masses of the salts was 0.1%, resulting an error of 0.3% in concentration. The estimated error
in FTIR cell thickness is at most 1%, resulting in 1.3% error in the absorbance.
88

6.3 Computational Methods
Density functional theory was used for geometry optimization of benzonitrile using B3LYP functional and LANL2DZ basis set (effective core potential fit-LANL2DZ) using the Q-chem package
[63]. Then distances between N of benzonitrile and point charge pair or Zn2+ and negative point
charge pair separated by 2 A˚ have been varied from 1.65 to 4.05 A˚. At each distance, single point
energy and frequencies are calculated with the same level of theory and basis set. Natural bond
orbital (NBO) analysis has been carried out at each location. [219–221] IQMol was used as visualization software. All the computations were performed in vacuum.
6.4 Results and Discussion
Figure 6.2 shows the interactions of various metal ions with the two crown ethers. We have
organized the data such that the ions that do not give rise to Lewis interactions are in the top panel
and those that give rise to Lewis peaks are in the bottom panel. To avoid clutter in the figure, we
have reported the data for the trivalent ions in the SI ( D, Figure 9.14 and Figure 9.15), where some
show the Lewis peak and some do not. In all of these spectra the cations and crowns are in 1:1 mol
ratios in propylene carbonate, and spectra for coordination with both small and large crowns are
shown. The spectra in the bottom panel prominently show a blue-shifted peak (∼ 40−60 cm−1
)
that is absent in the top panel. Both sets of spectra show smaller blue shifts of about ∼ 4−6 cm−1
with respect to the empty crowns. We assign the smaller shift to the electrostatic influence of the
captured ion causing a Stark shift in the nitrile. Cho et al. have proposed that the π hydrogen bond
interaction with the triple bond can cause a red shift on the nitrile frequency.[222] Note that we do
not observe any red shift upon addition of salts to the crown system in our experiments. Therefore,
we do not believe that such structures contribute to frequency shifts in this system. The larger
frequency shift arises from the Lewis interaction, i.e., direct coordination of the nitrile lone pair,
with the ions. Our assignment is further confirmed by spectra of benzonitrile (without a crown)
interacting with the ions in the same solvent as will be described shortly.
89

Normalized Absorbance (a.u.)
wavenumber / cm-1
Normalized Absorbance (a.u.)
complex with
small crown
complex with
big crown
complex with
small crown
complex with
big crown
2210
Electrostatic peak
Electrostatic peak
Lewis peak
Non-coordinating ions
Coordinating ions
Figure 6.2: Metal ion complexes with Benzo-15-Crown-5-CN (small crown) and Benzo-18-
Crown-6-CN (big crown) only showing electrostatic interactions (top panel) and showing both
electrostatic and Lewis interaction (bottom panel).
90

Note that the ion that is captured in the crown resides on the opposite of side of nitrile ( Figure 8.2). The field from this ion is expected to blue-shift the nitrile, analogous to an oxidative
potential in an electrode for surface bound nitriles as reported by us previously [1]. Our experimental results are consistent with this picture since an overall blue shift is measured for all cations.
However, the field is subject to the details of solvent screening, coordination geometries, and
counterion positions. Figure 6.2 shows that there is a variation in the electrostatic peak frequency
(∼ 4−6 cm−1
) and furthermore the same ion can produce different electrostatic shifts in the two
crowns. The reason is likely due to the coordination geometry of the ion in the crown. Since the
crowns are rather soft and floppy, a complete description of the effect requires further theoretical
work that is already underway. Here, we mainly focus on distinguishing these effects from the
Lewis coordination of the nitrile with the ions. The origin of the large blue shift is due to the partial donation of electron density from the nitrile lone pair to the Lewis acidic cations. Of course,
this direct Lewis interaction will depend on the Lewis acidity of the metal cations relative to the
lone pair of nitrile. Some of the metal ions do not have orbitals at the right energy to accept the
electron density, and therefore do not exhibit the largely blue-shifted peaks. Note that the origin of
the Lewis peak is not due to long-range electrostatics, but rather orbital overlap. The direction of
the shift is consistent with our previous work where benzonitrile interacts with a classic Lewis acid
such as BF3 [62]. The electronic origin of the blue shift due to such bonding has been explained
previously. [2, 5, 195]
Next, we wanted to confirm that the Lewis interaction was indeed due to the nitrile and not
a consequence of the crown. Therefore, we mixed the same metal salts with benzonitrile (without crown) in propylene carbonate in 1:1 molar ratios. Interestingly, we did not observe any shift
similar to the Stark effect in the benzonitrile peak. All of the salt mixtures produced a peak of
nearly the same frequency at 2230 cm−1
. In contrast, the ions that were identified as Lewis acidic
in the crown experiments, produced an additional peak that was much farther blue-shifted ( Figure 6.3). This supports the fact that the origin of this peak is Lewis acid-base type interaction
between metal ions and the CN group. It furthermore emphasizes that the crown is necessary to
91

Absorbance
Benzonitrile Peak
No Lewis Peak
(a)Absorbance
Lewis Peak
(b)
wavenumber / cm-1
Absorbance
Lewis Peak
(c)
Benzonitrile
Benzonitrile + Li+
Benzonitrile + K+
Benzonitrile + Mn2+
Benzonitrile + Zn2+
Benzonitrile + Cu2+
Benzonitrile
Benzonitrile + In3+
Benzonitrile + Eu3+
Benzonitrile + Y3+
Benzonitrile + Yb3+
Benzonitrile
Figure 6.3: Lewis interaction of metal ions with -CN of Benzonitrile. (a) Benzonitrile with M+,
(b) Benzonitrile with M2+. (c) Benzonitrile with M3+. Propylene carbonate is used as solvent in
all cases.
92

hold the ion in the correct orientation relative to the nitrile for producing the Stark effect. The lack
of Stark shift in benzonirile salt soltuions shows that, on their own volition, the ions are likely to
be solvated away by propylene carbonate and are too far from benzonitrile to exert an electrostatic
influence analgous to the crown case. As an aside, this observation highlights the importance of
ion capture by crowns for creating local electrostatic environments that are not permitted by the
innate solvent structure. Figure 6.4 compares the Lewis effect in benzonitrile and crown-ether
functionalized benzonitrile. Note that the Lewis peaks in the two systems correlate very well with
each other. Metal cations showing stronger Lewis effect in benzonitrile are likely to show stronger
Lewis effect in the crown-ether functionalized benzonitrile. The deviations between the two systems is relatively small and may arise from the small differences in the electronic structure of the
crown-appended benzonitrile relative to benzonitrile, or the details of the solvent structure.
Figure 6.4: Shift in -CN frequency after coordinating with metal ions in benzonitrile (blue) and in
Benzo-15-crown-5-CN (grey), in propylene carbonate
We performed experiments using three different anions: two organic anions TFSI− and OTf−,
and one inorganic anion Cl−. It is expected that the larger organic anions will have a weaker
tendency for ion pairing compared to the smaller and more compact Cl−.[223] We performed
experiments using these ions both with the crown system and the benzonitrile. First, note that
the electrostatic peak varies as a function of the anion choice ( Figure 6.5a). The Cl− produces a
smaller electrostatic shift compared to the organic anions. This is in conformity with the idea that
Cl− resides closer to the cation that is captured in the crown, and therefore reduces its electrostatic
influence on the probe. The electrostatic shift does not exist in benzonitrile, indicating that both
93

the cation and the anion are fully solvated and far away from the uncoordinated benzonitrile probe.
This further highlights the importance of the crown for inducing the oriented field on nitrile.
We also note that the Lewis peak varies for different anions for both the crown and benzonitrile
systems ( Figure 6.5a, b). This result indicates that the Lewis complex is subject to the electrostatic
influence of the anion. Similar to the electrostatic peak, it is expected that the Cl− anion will reside
closer to the cation that is Lewis-coordinated with the probe. However, explaining the behavior of
the Lewis peak in response to the anion as merely a pure Stark effect is not adequate to explain
the data. The presence of a nearby anion can produce a field that affects the probe directly via the
electrostatic effects, but can also influence the strength of the Lewis interaction by affecting the
charge density and overlap of both the donor and the acceptor orbitals. Our results of observing
a red shift for the Cl− anion relative to the organic anions, may indicate that the closely spaced
anion may reduce the extent of Lewis bonding. We hope this inspires further theoretical work on
the intricate balance between the two effects. Finally, we see about 8 cm−1 blue-shifted Lewis peak
for the organic anions relative to Cl− for the crown-ether complex ( Figure 6.5 a) and about 5 cm−1
blue-shift for benzonitrile. ( Figure 6.5 b). This difference may arise from the Stark effect of the
ion that is captured in the crown, causing an extra blue shift that is not observable for benzonitrile.
Note that for the case of crown-functionalized nitriles, the Lewis acidic ions have an opportunity to coordinate either with the crown or with the nitrile. The crown, with several oxygen groups,
in general, can accommodate the ions better. However, a competition between the two sites is expected, which can be studied by concentration-dependent spectra. For that reason, we have chosen
crown ether complexes with Zn2+ and Cu2+ for further study. Note that propylene carbonate can
also act as a Lewis base towards Zn2+ and Cu2+ [224]. Therefore, the Lewis basic nitrile is in
competition with the solvent for the cations. At low concentrations of cations, the population of
Lewis-coordinated probe is small since the solvent which is at much higher concentration solvates
the cations. As the concentration of cations is increased, the signature of the Lewis-coordinated
probe gradually emerges. Since the solvent is the same throughout the entire study, the interaction
of ions with the probe can be compared relative to each other. They show the most intense Lewis
94

Electrostatic peak Electrostatic peak
5 cm-1
8 cm-1
Lewis Peak
OTF- Cl- OTF- ClTFSITFSI-
(a) (b)
Lewis Peak
Figure 6.5: (a) Effect of counterions on the electrostatic and Lewis peaks in the crown ether-Zn2+
complexes. Both the electrostatic and Lewis peaks for the smaller anion Cl− show smaller blue
shift compared to the organic anions. (b) For benzonitrile, no electrostatic effect and no variation
from the counter ion is observed. The Lewis peak shows similar behavior towards anions as the
crown system.
peaks, along with distinct Stark shifts, implying that both coordinating sites are accessible to the
ions. At a fixed concentration of the crown ether (400 mM) in propylene carbonate, Cu(OTf)2
( Figure 6.6.a) and Zn(TFSI)2 ( Figure 6.6.b) were added at increasing concentration ratios. Initially at low concentration ratios, the electrostatic peak shifts and decreases in intensity for both
cations. Then the Lewis peak emerges and increases as the metal ion concentration is increased.
An interesting observation is that for Cu2+, the appearance of the Lewis peak starts at 0.5 metal ion
to crown ether concentration ratio. This means the nitrile group is quite competitive for coordination with Cu2+, and even when some crown cavities are still empty, Cu2+ engages in coordinating
with the -CN. Therefore, the curve showing the dependence of the electrostatic peak on concentration, is affected by the Lewis affinity of the ion. But, for Zn2+, the Lewis peak starts to appear
when the metal ion concentration is double of that of the crown ether ( Figure 6.6.b). Figure 6.6.c
shows integrated peak area for Electrostatic and Lewia peak regions at each concentration ratio
for Cu(OTf)2 complexation with crowns. At least four types of species can be expected in this
experiment. The first species is the empty crown (species 1, Figure 6.6.c) and as Cu2+ concentration increases, species 2 (Cu2+ coordinated in the crown) starts to form. At concentration ratio
95

[Cu2+]/[Benzo-15-crown-5-CN]
Integrated Peak Area
(c)
(d)
Absorbance
[+]
[ ]
(b)
wavenumber / cm-1
[Cu2+]/[Benzo-15-crown-5-CN]
5 cm-1
57 cm-1
Mn+
Mn+
-CN stretch / cm-1
-CN stretch / cm-1
Absorbance
[+]
[ ]
wavenumber / cm-1
(a)
Mn+
Species 1
Species 2 Species 3
Species 4
Figure 6.6: (a) IR absorption spectra of Cu(OTf)2 and Benzo-15-crown-5-CN at different concentration ratios. (b) IR absorption spectra of Zn(TFSI)2 and Benzo-15-crown-5-CN at different
concentration ratios. (c) Integrated peak area corresponding to different species present for Cu2+
extracted from IR spectra shown in (a). (d) Nitrile frequency at different concentrations of Cu2+
for Stark peak (lower panel) and Lewis peak (upper panel)
96

0.5, Cu2+ will start to bind with -CN directly (species 3). With increasing Cu2+ concentration, the
frequency of the Lewis peak continues to blue shift ( Figure 6.6.d), implying that a fourth species
may be involved, where both the crown and the nitrile sites are occupied by the ions ( Figure 6.6.c,
species 4). Given the distance between the ions and screening of the propylene carbonate solvent,
such doubly bound species is plausible. The concentration dependence of both the electrostatic and
the Lewis peak are shown Figure 6.6.d. The electrostatic peak plateaus at 2231 cm−1
, whereas
the Lewis peak starts at 2276 cm−1
and reaches to 2283 cm−1
. The Lewis peak starts to appear at
concentration ratio of 0.5, implying that it is possible for Cu2+ to bind the nitrile of those crowns
that do not contain an ion within them. Note that the linewidths of the Lewis peaks are much larger
than those of the electrostatic peaks. This may arise from the larger sensitivity of the Lewis interaction to distance, because it requires orbital overlap between the nitrile and the cations. Small
fluctuations and variations in the environment are magnified further for the Lewis peak compared
to the electrostatic peak. The line shapes may also have concentration dependence. Perhaps a more
in-depth analysis of the linewidth can be performed by 2D-IR spectroscopy in the future. To understand the competition between the solvent and the probe for the cation, we extended this study to
two other solvents with one representative salt: dimethyl sulfoxide (DMSO), and dichloromethane
(DCM) (see SI, D, Figure 9.16). In brief, we noted that the Lewis peak forms very readily in
DCM, but does not form at all in DMSO. The explanation for this effect is described in the SI ( D).
It may be questioned that the ion that is coordinating with the nitrile will also have an overall
electrostatic effect as well. This electrostatic effect is expected to result into a red shift, since
the field from the ion will be, in general, in the opposite direction to the field from the crownbound ion. To decipher the balance between the electrostatic and the Lewis influence of the ion,
we performed some calculations. The best way to approach this problem is to vary the distance
between the ion and the nitrile as shown in Figure 6.7. To avoid an overwhelmingly large field, a
compensating point charge is always kept at a distance of 2 A relative to the ion. Then we consider ˚
two scenarios. First, is a scenario where the ion’s electronic structure is explicitly included in the
calculation, and therefore its bonding with nitrile is permitted (Fig. Figure 6.7.a). In the second
97

Allowing for Bonding
with Zn2+
Electrostatic
Effect only (point
charge)
Distance between positive charge and N of –CN
/ Angstrom
-CN frequency Shift / (cm-1
)
Zn2+
Distance
(1.65 Å – 4.05 Å)
-2 point charge (to
balance the positively
charged ion)
Model system for both Lewis effect and electrostatic effect
Model system for electrostatic effect
-2 point
charge
+2 point
charge
Distance
(1.65 Å – 4.05 Å)
(a)
(b)
(c)
Figure 6.7: (a) Benzonitrile in the presence of Zn2+ and negative point charge to study both the
Lewis and electrostatic effects. (b) Benzonitrile in the presence of positive and negative point
charges to study the electrostatic effect. (c) The nitrile frequency shift as a function of proximity
of the charged species - Zn2+ (blue) and positive point charge (orange)
Distance between positive charge and N of –CN / Angstrom
Occupancy of Zn2+ 4s orbital
Zn
2+ 4s
Distance between positive charge and N of –CN / Angstrom
Occupancy of Nitrogen
Lone Pair orbital
(a) (b)
Figure 6.8: Results from Natural Bond Orbital (NBO) analysis. (a) Change in occupancy of
nitrogen lone pair orbital. (b) Change in occupancy of 4s orbital of Zn2+ as a function of distance
of positive charge from nitrogen of benzonitrile.
98

scenario, just the electrostatics is considered and the ion is treated as a point charge ( Figure 6.7.b).
Then we scanned the distance of the nitrile relative to the ion in each one of these scenarios, and
calculated the frequency of the nitrile at each distance as shown in Figure 6.7.c. Note that at
large distances, both the explicit ion and the artificial point charge behave similarly and give rise
to the expected red shift. However, at shorter distances, where orbital overlap becomes possible,
they have drastically different effects. The point charge continues to red-shift the nitrile, while
the explicit ion begins to accept electron density from nitrile and blue-shifts the frequency. In
our model calculations, we have observed that at around 2.5 A the metal cation starts to behave ˚
as a Lewis acid because the nitrile frequency blue-shifts beyond this point ( Figure 6.7.c). This
distance is within the order of H-bonding distance for nitrile as has been reported previously.[92]
These model calculations show that the same metal cation can exhibit an electrostatic effect when
it is far away, and a Lewis effect when it is within orbital overlap distance with the nitrile. We
also performed a natural bond orbital (NBO) analysis of the two systems at each distance, and
found that for the explicit ions, at shorter distances, the occupancy of lone-pair orbital of nitrogen
in benzonitrile decreases and the occupancy of Zn2+ 4s orbital increases ( Figure 6.8). In contrast,
we see a much smaller monotonic increase in the occupancy of the lone-pair orbital where only
point charge is considered instead of the explicit ions ( Figure 6.8.a). This trend supports the
qualitative description of blue-shift in nitrile frequency due to charge transfer from nitrogen lone
pair orbital to the acceptor ion. Decrease in the charge density of the nitrogen lone pair orbital
stiffens the nitrile bond since the lone-pair orbital has some antibonding character, as has been
reported before. [5, 200]
6.5 Conclusion
We have used a crown ether with an appended benzonitrile to distinguish the long-range electrostatic influence from the explicit Lewis bonding of several cations to the lone pair of nitrile. We
report that the Lewis bonding causes a frequency shift that is much larger than the Stark shift due
99

to the crown-bound ion. Furthermore, we show that at large distances the influence of the ion on
frequency shift is via electrostatics. At shorter distances, where orbital overlap becomes possible,
frequency shift due to the Lewis bonding dominates. Our work is expected to help in interpretation
of spectra in complex ionic environments, where both of these effects may be operative. Frequency
shift induced by the Lewis bonding of metal ions may also inspire applications of the benzonitrile
probe as a reporter of local ion concentration at electrode-electrolyte interfaces.
100

Chapter 7
Measuring the Electric Fields of Cations Captured in Crown
Ethers
This chapter is based on the draft manuscript by A. Maitra, P. Das, A. Mohamed, W. R. Lake,
S. H. Schiffer, M. Johnson, B. Thompson and J. M. Dawlaty, ”Measuring the Electric Fields of
Cations Captured in Crown Ethers”. Crown ethers (B-15-C-5-CN and B-18-C-6-CN) synthesized
by Pratysha Das from Prof. Barry Thompson’s group at University of Southern California are
again used in this work. Gas phase experiments are performed by Ahmed Mohamed from Prof.
Mark Johnson’s group at Yale university. Figure 7.4a is provided by Ahmed Mohamed. William
R. Lake from Prof. Sharon Hammes Schiffer’s group at Yale university has done some of the
computational studies presented here. Figure 7.5 is plotted by William R. Lake.
7.1 Introduction
Inherent electric fields experienced by a molecule arise from orientation of solvent dipoles, organization of ions, and the presence of nearby polarizable entities. Such fields influence the thermodynamics and kinetics of a vast range of reactions, especially if the reactants, products, or transitions
states are charged or dipolar. The magnitude and direction of such fields are a natural consequence
of the molecular environment. To control and direct chemical reactions, it is often necessary to
101

tailor the local electrostatics. Such engineering of the chemical microenvironments is an important
frontier of contemporary chemistry.
The effects of electric fields on reactions is manifested in a variety of scenarios, especially
in electrochemical, molecular, and enzymatic catalysis. A few examples are through space substitution effect on iron(0) tetraphenylporphyrins to catalyze CO2 to CO electrochemical conversion [225], and electron-catalyzed dehydrogenation reaction of ethylene [226]. Surendranath and
coworkers have shown that spontaneous electric fields generated by both interfacial proton and
electron transfer in different solvents can affect hydrogenation reactions on a Pt/C catalyst.[227]
Similarly polarizing fields emanating from charged entities, such as titratable aminoacids, in an
enzyme[159, 228, 229] also influence chemical reactions. To control the reactivity of molecular catalysts, a number of studies have shown that installing a charge at a well-defined locations
relative to a reactive center can control the reaction mechanisms and selectivity [230–234].
The influence of these electric fields on reactions can be due to a number of factors, including
pre-organization, electrostatic attraction or repulsion in bimolecular reactions, and direct polarization of a chemical bond.[8, 16, 235] For example, the presence of a potassium ion in deacetylase active site helps pre-oraganize the residues to activate the protein function [8], whereas the
presence of higher cationic charges can reduce the rate of N2 fomation in bimolecular coupling
reaction [16]. The presence of cationic charge in a copper catalyst was shown to affect C-H activation by polarizing the C-H bond.[235] Electric fields from ions can also affect excited state
processes, for example by alteration of the electronic excited states in photodissociation reactions
[236]. Therefore, it is important to quantify the localized electric fields emanating from ions more
systematically at molecular length-scales.
Inspired by the work by Yang and coworkers [16, 201, 202], we have designed a model
molecule which helps quantifying the electric fields emanating from cations. Our molecule is a
crown ether that bears a covalently attached benzonitrile ( Figure 7.1a, Figure 7.1b) - which is
a known vibrational probe for sensing local electric fields [1, 5, 13, 35, 36, 92, 197]. The probe
102

which resides within a nanometer of the captured ion reports the effective electric field from the
cation.
We present vibrational spectroscopy results in the condensed phase for the crown ethers loaded
with a number of cations with +1, +2, and +3 charges. To better isolate the effect of solvents
and counter ions and to understand the geometry of the cations in the crown we report vibrational
spectra in the gas phase. Finally, with support from computational work, we describe the factors
that affect the electric field generated by the ions and sensed by the probe.
(a) (b)

CN
CN
Free Crown Ether
Metal ion complexed Crown Ether
Mn+
(c)
Figure 7.1: Synthsized Crown-ethers with benzonitrile vibrational probe (a)Benzo-15-Crown-5-
CN (B-15C5-CN) (b) Benzo-18-Crown-6-CN (B-18C6-CN) (c) cartoon of expected blue-shift of
CN frequency after capturing a metal ion (Mn+)
7.2 Experimental Methods
7.2.1 Synthesis
Detailed description of the synthesis of the crown ethers is given in Chapter 6.
7.2.2 Condensed Phase Experiment : FT-IR Spectroscopy
Detailed description of the Fourier Transform Infra red (FT-IR) spectroscopy is given in Chapter 6.
103

7.2.3 Gas phase Experiment
This section is provided by Ahmed Mohamed from Prof. Mark Johnson’s group.
The gas-phase spectra presented here were recorded using a custom-built, tandem photofragmentation mass spectrometer which has been described in detail previously. In short, the crown
ether complexes were extracted from solution and ionized using electrospray ionization (ESI). The
ions so produced were introduced into vacuum through multiple stages of differential pumping and
held in a radio frequency Paul trap, where they were equilibrated at the experimental temperature
of 20 K through collisions with helium buffer gas. The buffer gas was doped with 10% D2, which
was employed as a mass tag. At the cryogenic temperatures maintained in the ion trap, the D2
mass tag forms a weakly-bound complex with the ion under study. Infrared spectra are recorded
by scanning a pulsed laser ( 10 mJ, 10 ns, 10 Hz repetition rate) over the reported frequency range.
When the laser pulse is resonant with a vibrational mode of the ion under study, absorption of a
single photon is sufficient to effect photodissociation of the mass tag through intramolecular vibrational redistribution. Monitoring photofragmentation of the weakly bound complex as a function
of laser frequency thus yields infrared action spectra in a linear, single-photon regime.
7.3 Computational methods
Density functional theory was used for geometry optimization and frequency calculation of the
Benzo-15-Crown-5-CN using B3LYP functional and LANL2DZ basis set (effective core potential
fit-LANL2DZ) using the Q-chem package [63]. IQMol was used as visualization software. All the
computations were performed in vacuum.
7.4 Results and Discussion
We have used 16 cations of charges +1, +2, and +3 in this study. For simplicity, first it is useful to
only plot the cations that produced the largest shifts for their charge. The IR spectra in the nitrile
104

range of the two crown ethers bearing these cations is shown in Figure 7.2a-b. The figures show
that the frequencies of the free crown ethers for the small and large crowns are nearly identical
(2225 cm−1
). As cations are inserted in the crowns, the frequency blue shifts as expected for a
field emanating from the cation. The ratio of metal ion to crown ether was 1:1 in all of these
measurements, except for Sc3+, where a higher ratio of 2:1 was used. The reason is the smaller
affinity of Sc3+ for the crown ether. Even at this ratio, only partial occupancy was achieved as
evidenced by the two peaks in the spectrum. We will discuss the possible reasons for the frequency
shift of the nominally unoccupied crown, and the concentration-dependence of occupancy later in
this paper.
The frequency shifts relative to free crown ether are summarized in Figure 7.2c, which shows
that larger charges produce larger shifts in both crowns. However, the frequency shift is more
sensitive to charge for the smaller crown. The origin of this difference is likely that the cations
are held closer to the Stark probe in the smaller crown. Although the field from the cation is
expectedly inhomogeneous across the body of the benzonitrile probe, it is still useful to estimate
the approximate or effective value of this field from the known Stark tuning rate of the benzonitrile
molecule (∼ 3.6 cm−1/V/nm) [2]. The right vertical axis in the figure shows this effective field
felt by the nitrile, which varies from 1−3.5 V/nm.
As an aside, we also observed IR peaks that were blue-shifted by a very large value (40-50
cm−1
) for some of the cations (e.g. Zn2+, Cu2+, and Y3+). These spectra are shown in the [17],
and do not arise from the Stark effect of the cation, but rather from direct Lewis coordination of the
nitrile lone pair with the cations. Similar effects have been observed for acetonitrile before [195].
Such peaks also arise when the cations are placed in pure benzonitrile. This phenomena has been
studied in detail in [17].
The data for all of the studied cations is shown in Figure 7.3, which exhibits a significant range
of scatter in the frequency shift induced by cations of similar charge. For the smaller crown ether
( Figure 7.3a) the shift produced by Li+ is much larger than that of K+, even though they both have
105

Normalized Intensity of CN Normalized Intensity of CN
(a)
(b)
wavenumber / cm-1
∆νCN / cm-1
Charge of cation
Li+
K
+
Zn2+
Ca2+
Sc3+
Eu3+
Empty
Electric Field / (V/nm)
(c)
small crown
big crown
small crown
big crown
Empty Crown
Empty Crown
Figure 7.2: (a) IR absorption spectra of cation complexed Benzo-15-Crown-5-CN and (b) Benzo18-Crown-6-CN, (c) Maximum frequency shifts obtained in Benzo-15-Crown-5-CN (blue circles)
and Benzo-18-Crown-6-CN (red squares)
106

the same charge. Similarly, a large range of shifts is observed for the doubly charged cations. For
the triply charged cations, it seems that most of them induce either a small shift or no shift at all.
It is known that matching the cavity size to the ionic radius is important for binding to the
crown. The reported cavity radius of 15-crown-5 based on X-ray crystallography is 90 - 110 pm
[237]. The ions that induce the maximum shift in the smaller crown ether also have sizes that
nearly match these numbers (90 pm Li+, 88 pm Zn2+, 88.5 pm Sc3+). However, we did not
identify a clear monotonic correlation between the ionic radii and the measured frequency shifts
(see SI, E). Successful coordination of many of these and other ions with crown ethers have been
reported in the literature. However, most reports that are based on conductometric and calorimetric
measurements can not decipher the binding geometry. Some of the ions may bind loosely with the
crown, and may be subject to solvent screening and interactions with counter ions. Such effects
can result into attenuation of the field that is perceived by the probe. Our computational resulted,
presented later in this paper, addresses some of the effects of complexation on frequency shift.
The bigger 18-crown-6 has a reported cavity size of 130-140 pm [237]. We expected this
crown to house the larger cations that could not be accommodated by the smaller one. The results
for frequency shifts are presented in Figure 7.3b. As can be seen in the figure, several of the larger
cations that produced small or no shift in the smaller crown (e.g. K+, Ca2+, Eu3+), exert a larger
frequency shift in the big crown. However, it should be noted that the shifts measured in the big
crown are smaller than the shifts in the small crown. A possible reason for this may be that in the
big crown the ions are held at a larger distance relative to the probe. This suggests that accounting
for the size of the crown is important for engineering the electrostatic environment of the ion.
The results presented above strongly suggest that a number of factors including complexation geometries, solvent effects, and counter ions affect the field perceived by the probe. In the
condensed phase, it is impossible to neatly isolate these effects. Therefore we resorted to low temperature gas phase ion spectroscopy, which allows to isolate the spectrum of a crown-ion complex
without the involvement of solvents and counter ions. This approach is a powerful tool for discerning the inherent field produced by the ion, and in conjunction with the condensed phase data can
107

Li+
K
+
Free
Mn2+
Ba2+
Ca2+
Ni2+ Sc3+
Al3+
Eu3+
In3+
Y
3+
∆νCN / cm-1
Charge of Cation
∆νCN / cm-1
Li+
Na+
K
+
Free
Zn2+
Mg2+
Ba2+
Ca2+
Ni2+
Sc3+
Al3+
Eu3+
In3+
Y
3+
Charge of Cation
Cu2+
(b) (a)
Na+
Mn2+
Yb3+
Zn2+
Cu2+
Mg2+
Yb3+
Figure 7.3: CN frequency shift with charge of cation complexed with (a)Benzo-15-Crown-5-CN
(small crown) (b) Benzo-18-Crown-6-CN (big crown)
reveal the net influence of solvent and counter ion screening. Furthermore, we performed computations to reveal the details of complexation geometries and their effects on the field perceived by
the probe.
Representative gas phase spectra in the CN stretch region for the small crown complexes with
2+ ions (Mg2+, Zn2+, Ca2+, Ba2+) are shown in Figure 7.4b). The CN stretch could not be
observed for the +1 ion complexes. The corresponding C-H stretch spectra for these 2+ ions
are shown in Figure 7.4a. Even though all these ions carry the same charge, there are distinct
variations in the C-H stretch frequencies. There are five peaks in the aliphatic C-H stretch region,
as distinctly observed for the Ca2+ complex. The frequencies of these peaks vary significantly
for the other complexes ( Figure 7.4a), indicating that their coordination geometries to the crown
are different. However, interestingly, the nitrile frequencies for all of these complexes are nearly
identical (2253 cm−1
) and largely blue shifted (∼ +22 cm−1
) compared to the condensed phase
values ( Figure 7.4b). Investigation with Sc3+ in gas phase is still going on by Prof. Mark Johnson’s
group.
With support from computations, two important conclusions can be drawn from these measurements. First, the large blue shift of ∼ +22 cm−1 with respect to the condensed phase shows that
108

C-H stretches (a) (b)
wavenumber / cm-1
+22 cm-1
2253 cm-1
2227 - 2231 cm-1
Liquid
Phase
Gas
Phase
wavenumber / cm-1
Mg2+
Ca2+
Ba2+
2900 3000 3100
Zn2+
Figure 7.4: (a) CH stretching frequencies in gas phase experiment for M2+ complexes with Benzo15-crown-5-CN (smaller crown) indicating different co-ordination geometry, (b) -CN stretching
frequencies for M2+ complexes with Benzo-15-crown-5-CN (smaller crown) in condensed phase
(top) and in gas phase (bottom)
2240
2245
2250
2255
2260
2265
2270
2275
2280
2285
2290
0 1 2 3 4
Charge of Cation
Vacuum PCM

/
−

Zn
Ca
Y
Al
Sc
Y
Sc
Al
~24 cm-1
Empty
Empty
Li
Na
K
K
Na
Li
Zn, Ca
Ba
Ba
Figure 7.5: CN frequency shift for different metal ions in vacuum and in solvent
109

the inherent field experienced by the probe is much larger in the absence of the solvent and counter
ions. To our knowledge, it is the first time that the net influence of the solvents and counter ions is
distinguished from the inherent field of an ion. To estimate this effect, we performed computations
comparing frequency shifts between the 2+ ions in vacuum and in a polarizable continuum model
(PCM) with the dielectric constant of 64.9 ( Figure 7.5). The computed shift is ∼ + 24 cm−1 which
is reasonably within the range of the experimental observation. We can compare this with a crude
estimate from Coulomb’s law for field E at a distance of 0.7 nm, which is roughly the distance
between the nitrile and the ion. At such a distance the difference between field in vacuum and field
in a medium with a dielectric constant of 64.9 is about ∆E = 5.8 V/nm. If a linear Stark effect is
assumed (with ∆µ = 3.6 cm−1/(V/nm)), this will correspond to a frequency shift of ∼ 21cm−1
.
This value is, interestingly, in the same range of the experimental and computational results. However, one must be cautious about its interpretation, since the field is inhomogeneous around the
ion, and the Stark response may also deviate from linearity at such values.
Li+ Na+
K
+
Zn2+ Mg2+ Ca2+ Ba
2+
Figure 7.6: relative displacement of the crown complexes M+ metals and M2+ metals
110

Second, the possible variations in the complexation geometry, as suggested by the C-H stretch
region, does not seem to affect the frequency of the nitrile. Our computational results show that
the 2+ ion complexes in Figure 7.6 have different complexation geometries. The smaller Mg2+
ion (86 pm radius) fits well in the cavity, while Ba2+ ion (149 pm radius) hovers over the crown
and does not fit well. A measure of this fit is the sum S of the distances MO1 − MO5 from the
complexed ion to the five oxygen atoms in the crown. The computed frequency shifts as a function
of this parameter is shown in Figure 7.7. The value of S is nearly 30% (from 10.3 A˚ to 13.6 A˚)
between Mg2+ and Ba2+. The computed frequency shift associated with this change is ∼ 1.75
cm−1
, which is a small fraction of the maximum frequency shift of ∼ 24 cm−1
. A similar trend
is observed for the 1+ ions as well. This parameter, while a reasonable reporter of coordination,
is not necessarily indicative of the orientation of the ion with respect to the probe. As seen in
the structures ( Figure 7.6), the ring of the crown is somewhat flexible relative to the plane of the
benzonitrile, and assumes different geometries for various ions. Exploration of multiple geometries
with nearly equal energies are still being explored by William R. Lake from Prof. Sharon Hammes
Schiffer’s group at Yale university. Therefore, to understand the effect of the ion on the probe, a
better parameter is the perpendicular distance of the ion from the plane of the benzene ring d. This
ranges from 0.15 A˚ to 1.25 A˚ and correlates well with the computed frequency shifts as shown in
Figure 7.8a,b. A second parameter is the angle θ between the nitrile bond and the line connecting
the mid point of the bond to the ion, which ranges from -3 ◦
to +9 ◦
. This parameter also correlates
with the computed frequency shifts ( Figure 7.8c,d). From an elementary understanding of the
Stark effect, the frequency shift is the dot product of the angle between the field and the dipole
moment difference of the probe. If it is assumed that field lines emanate radially from the ion, then
the frequency shift should vary as the cosine of the angle θ. We have plotted this behavior as the
dotted line in the figure. It is clearly seen that the computed shifts are significantly larger than that
expected from the geometric estimate of the field projection on the bond.
Hence, merely considering the geometric projection of the electric field vectors from the metal
ion on the -CN probe cannot explain the variability of the frequency shift in the same charge group
111

of the ions obtained in experiment. More investigations are required to determine if the variability
we see in the theoretical calculations for vacuum are within computational error or not. From
Figure 7.5, we see a larger spread in frequencies for the same charge family qualitatively matching
with the experimental trend ( Figure 7.3 when a solvent model is considered (grey circles). This
might be indicative of the fact that in presence of solvent field, the specifics of the coordination
geometry will matter more.
Finally, it is important to consider whether the influence of the ions is entirely through their
electrostatic effects on benzonitrile, or does the explicit Lewis bonding of the ions with the oxygen atoms is partly responsible for the frequency shift. To answer this, we performed geometry
optimization and frequency computations on representative ions complexes Li+, Zn2+, and Sc3+.
Then we kept the optimized geometry and replaced the ions with point charges +1, +2, and +3
respectively, followed by frequency computations. The computed frequencies (figure 9.a) show
nearly no difference between the two scenarios. The computed charge density differences induced
by either the ions or the point charges (Figure 9.b) also are nearly identical to each other. These
results indicate that the ions have a direct electrostatic influence on benzonitrile and the nature of
bonding of the ions with oxygen atoms does not majorly affect the frequency shift.
112

/Å

/
−

/
−

= + + + +
Ba2+
Ca2+
Mg2+
Zn2+
Li+
Na+
K
+
Figure 7.7: CN frequncy shift with the sum of distances between the 5 oxygens and the metal ion
(in small crown (B15C5-CN)

/
−

d/Å
Li+
Na+
K
+

/
−

Ba2+
Ca2+
Mg2+
Zn2+

/
−

Ba2+
Ca2+
Mg2+
Zn2+

/
−

Li+
Na+
K
+
Angle ()/ °
(a) (c)
(b) (d)
Figure 7.8: CN frequncy shift with (a),(b) the distance of the metal ion from the benzonitrile plane
for +2 and +1 metals respectively (c),(d) the angle of the metal ion relative to the plane of -CN
group in the small crown (B15C5-CN)
113

0
(
−
)

0 1 2 3
charge of cation
(b)

/
−

0 1 2 3
charge of cation
Crown ether + Metal ion
Crown ether + Point Charge
(a)
Figure 7.9: comparing metal ion vs point charge for (a) CN frequency shifts (b) partial charge
differences
7.4.1 Conclusion
We have quantified the effective fields exerted by cations held in crown ethers on an adjacent
vibrational Stark shift probes. The gas phase experiments tell us about the contribution of the bare
cations alone as these experiments are free from the screening of the field of solvent and counteranions. Hence, combining the results from condensed phase and gas phase experiments, we can
conclude that the contribution of the solvent and counter-anion together is about ∼24 cm−1
. Also,
We have found that the details of the coordination geometry of the cation in the crown affects the
field significantly when solvent field is present. We believe, our results are relevant for the new area
in organometallic catalysis where cations captured in crown ethers are used for creating oriented
electric fields for influencing reactivity of crown-bound catalysts.
114

Chapter 8
Direct Spectrosocpic Observation of Ultraslow Ion Desolvation
at an Interface
The work presented here is based on the preliminary results of an ongoing project of understanding
ion adsorption process using surface enhanced IR absorption spectroscopy. This chapter is the first
draft of an ’in prepearation’ manuscript ”Direct Spectrosocpic Observation of Ion Desolvation at
an Interface” by Anwesha Maitra, Anuj Pennathur, Elliott Chiang and Jahan M. Dawlaty.
8.1 Introduction
Ion intercalation batteries are the modern day energy storage unit. Both desolvation of the ions and
then ion diffusion inside the porous electrodes determine how fast a battery can charge or discharge.
Starting from ion desolvation from the solvent, interactions between solvents and counter ions to
ion diffusion within the porous electrode - all of the factors will determine the overall performance
of an ion intercalation batttery. Besides Li-ion batteries, a lot of studies are going on to establish
multivalent ion battery devises with more earth-abundant metals in non-aqueous electrolytes. One
of the major difficulty in using multivalent ions (e.g. Mg2+, Ca2+, Zn2+) as charge carriers are their
extremely strong solvation shells. Although selecting the electrode material and the electrolyte
containing the charge carriers together will affect the rate of charging and discharging of a battery,
115

the main focus of our work is to investigate ion migration from bulk solvent to the electrodeelectrolyte interface. Both for ion adsorption on a non-porous solid surface and ion intercalation
into a porous electrode, ions need to shed some of their solvation shell. Hence, by looking into
only desolvation process in presence of different solvents, we will get more fundamental insights
on one of the main controlling factors of battery function. This can further help in the choice of
correct electrolytes for future rechargeable devises.
Now, to get the molecular level picture for ion adsorption effect using spectroscopic method we
need some surface specific signals. Of course, this becomes extremely difficult as most of the cases
ions do not have their own vibrational signatures. In this work we have used a a surface tethered
benzonitrile monolayer to study ion adsorption process with vibrational spectroscopy. We have
formed a self-assembled monolayer (SAM) of 4-Mercapto benzonitrile (4-MBN), a well-known
vibrational probe which attaches with gold via -SH functional group. This benzonitrile monolayer
is acting as a Lewis base for the suitable metal ion which can act as a Lewis acid to study this ion
adsorption process. Thus we ensure to get information of ions only specific to the surface.
From our previous work, we found that several metal ions show strong Lewis interaction peak
with benzonitrile ( Figure 8.1).[17] This inspired us to use this Lewis-coordination peak as an indicator of adsorbed metal ions at a metal-electrolyte interface. To specifically interact with the
lone pair of nitrogen of the benzonitrile, the metal ions need to break their solvation shell. Similar
scenario holds true at metal-electrolyte interfaces. We have studied the Lewis peak formed with
metal ions at the 4-MBN SAM tethered on a gold electrode. Previously, surface tethered benzonitrile has been used as a Lewis base for a classic Lewis base BF3, [62] but to our knowledge Lewis
interaction between benzonitrile with a charged species at an interface has not been studied before.
8.2 Experimental Methods
In this work, we have used a surface sensitive FT-IR technique - surface enhanced IR absorption
spectroscopy (SEIRAS). A layer of gold were deposited using electroless deposition method on a
116

(a) (b)
Figure 8.1: (a) Representative cartoon of Lewis coordination with Zn2+ in bulk propylene carbonate solvent (b) IR spectra of Lewis coordination of -CN group in benzonitrile with metal cations of
different charge, reproduced from our previous work. [17]
117

Detector ZnSe Crystal
Au
Figure 8.2: Representative cartoon for partial breaking of solvation shell of metal cation to form
Lewis coordination with mercapto benzonitrile (MBN) tethered on gold surface
60◦
cut ZnSe ATR crystal (Pike Technologies) following the procedure of Bao et al.[238] First the
crystals were polished with 0.05 µm alumina slurry for 10-15 minutes followed by 10 -15 minutes
ultrasonic cleaning in a HPLC water bath. Then the crystals were heated ∼90-95◦ on hot plate and
∼30 mM gold(III)chloride (HAuCl4) solution in water was added on top of the crystal by using a
transfer pipette. Once the solid gold deposition starts, we waited for 15-20 seconds before washing
off excess gold chloride solution by spraying HPLC water on the crystal. Then the crystals were
air-dried before soaking them overnight in a 20 mM 4-Mercapto benzonitrile (4-MBN) solution
in ethanol to make the self-assembled monolayers (SAMs). After sonicating in ethanol for 10-15
minutes and air drying, the gold deposited ZnSe crystals with 4-MBN monolayer were used for
further experiments. The crystal was held in attenuated total internal reflection (ATR) geometry
with spectroelectrochemical Teflon cell compatible with the VeeMAXIII (PikeTechnologies). All
FT-IR spectra have been obtained with liquid nitrogen cooled MCT detector in a Thermo Scientific
Nicolet iS50 spectrometer. We have used 128 scans with a resolution of 0.94 cm−1
for all FT-IR
spectra presented in this study. All of the spectrum are collected with p-polarized IR light as the
118

monolayer signal was more sensitive to p-polarization. A schematic of the experimental set-up is
given in Figure 8.2.
4-Mercaptobenzonitrile (4-MBN) was obtained from Combi-blocks Inc. Zinc(II) Chloride
(ZnCl2) (anhydrous, powder, 99.9 %), Indium(III) trifluoromethanesulfonate (In(OTf)3), Copper(II) trifluoromethanesulfonate (Cu(OTf)2), Dimethyl Carbonate (DMC) and Propylene Carbonate (PC) ( Figure 8.3) were bought from Sigma-Aldrich, Dimethyl Sulfoxide (DMSO) and HPLC
water were purchased from VWR.
(a) (b) (c)
Figure 8.3: Solvents used in this work along with water, nit shown.(a) Dimethyl Carbonate (DMC),
(b) Propylene Carbonate (PC), (c) Dimethyl Sulfoxide (DMSO).
8.3 Results and Discussion
Figure 8.4 shows the Lewis interaction peak for ion-coordinated 4-MBN SAMs at the goldelectrolyte interface. We found that metal ions which showed prominent Lewis peak in the bulk
can show similar effect at the interface even when saturated concentrations of the respective metal
ions in propylene carbonate are added (see Figure 8.4 for three representative spectra). This is
likely due to the loss of required coordination geometry with benzonitrile at the interface compared to bulk. In presence of Zn2+, most intense Lewis peak with benzonitrile arises at the surface,
which is centered at ∼2265 cm−1
. For this reason, Zn2+ was chosen for further explorations as a
representative divalent ion.
Next, we will discuss the effect of solvent on appearance of lewis peak at the interface. We have
used three different solvents saturated with ZnCl2 ( Figure 8.3) in addition to water (ε = 78, 1.84
119

wavenumber / cm-1
Figure 8.4: Nitrile stretch in saturated solution of Cu(OTf)2 (blue), ZnCl2 (red), In(OTf)3 (yellow).
All spectra are nomarlized with respect to CN stratch at 2227 cm−1
. Only Zn2+ shows the most
intense Lewis peak centered ∼2265cm−1
.
D). Aqueous electrolyte for Zn2+ ion batteries are reported to have the problem of nonuniform
zinc dendrite formation on electrodes and easy decomposition of water impedes the coulombic
efficiencies. Organic solvents are used as cosolvents with aqueous solutions in Zn2+ ion batteries
to suppress side-reactions in water.[239] These cosolvents change solvation structures around Zn2+
and hence, it will change the desolvation threshold at the electrode surface. Carbonate class of
organic solvents are a popular choice in ion-intercalation battery.[239, 240] We have used Dimethyl
Carbonate (DMC, ε = 3.087, 0.91 D) and Propylene Carbonate (PC, ε = 64.9, 4.94 D) as two
examples of this class of organic solvents and another high dielectric solvent Dimethyl supfoxide
(DMSO, ε = 46.6, 3.96 D) which is also reported to scavenge water molecules around the Zn2+
solvation shell.[241]
After adding 0.5 ml of the solution on 4-MBN monolayer, we have waited for 40-50 minutes
before reporting the final spectra in Figure 8.5. Our results show that Lewis peak at the interface
appears almost at the similar location ∼2265 cm−1
(for dimethyl carbonate (DMC) and propylene
120

carbonate (PC) and ∼2268 cm−1
for water in presence of saturated solutions of ZnCl2 the respective solvents and a corresponding decrease in free -CN peak. In DMSO, ZnCl2 does not show any
Lewis peak at the interface. Previous studies show that even in the bulk DMSO, benzonitrile does
not form Lewis peak with ZnCl2.[17] In DMSO, we only see the expected solvatochromic shift of
the -CN frequency by ∼4 cm−1
.[93] This observation follows the trend of Gutmann donor number
of different solvents, which is the measure of their Lewis basicity. For DMC, PC, water and DMSO
the reported donor numbers are 17.2, 15.1, 18, 29.8 respectively. [241–243] Lewis basicity of benzonitrile which is acting as a Lewis base for the metal ions is smaller compared to the solvents
(Gutmann donor number for benzonitrile 11.9).[244] Much stronger Lewis interaction in DMSO
can explain our observation of no Lewis peak at saturated concentration of ZnCl2 in DMSO. The
increasing solubility order of ZnCl2 in these solvents are DMC, DMSO , PC and water. From
our results, we see that the specific interactions (donor number) of the solvent molecules with the
Zn2+ ions plays a bigger role than solubility in determining the extent of Lewis co-ordination at
the interface.
The next interesting result from this work is that the observed maximum intensity of the Lewis
peak at the interface is not instantaneous. Figure 8.6 shows the time evolution of the Lewis peak
in propylene carbonate and in water. We believe that the timescale for the formation of the Lewis
bond is coming from the very high activation energy of overcoming the solvation and we will
show next that how we arrive at the conclusion that the adsorption process is rate limited by the
desolvation step from the solvent. The bottom panel of Figure 8.6 shows the time traces for the
respective solvents.
The time traces were fitted to the following equation
[Zn2+]ads = [Zn2+]solv(1−e
−k1t
) +[Zn2+
0
]ads (8.1)
where [Zn2+
0
]ads depicts adsorbed Zn2+ concentration in the beginning and [Zn2+]ads is maximum adsorbed Zn2+ concentration which can be correlated to the integrated areas extracted from
the baseline corrected spectra for the respective solvents. It is reported that in absence of any
121

No Lewis Peak
DMSO added
wavenumber / cm-1 wavenumber / cm-1 Absorbance / mOD
Absorbance / mOD
(a) (b)
wavenumber / cm-1
Absorbance / mOD
(c)
wavenumber / cm-1
Absorbance / mOD
(d)
.
Figure 8.5: Adsorption of Zn2+ at the interface in presence of (a) Dimethyl carbonate (DMC), (c)
Propylene Carbonate (PC) and (d) Water. No Lewis coordination observed in (b) dimethyl sulfoxide (DMSO). Blue spectra indicates nitrile peak from the MBN monolayer in air and red spectra
indicates nitrile peak at the interface after adding saturated solution of ZnCl2 of each solvent
122

resistive layers (cathode/electrolyte interphase, CEI), desolvation is the rate limiting step in ion
transfer process into porous electrodes. [245] In our system, we do not apply any potential bias to
actively passivate the electrode with any CEI. Hence, we conjecture that the ion adsorption process
might be happening in two steps. The first step is a slow equilibrium step between solvated and
desolvated Zn2+ and then a fast adsorption step as given below:
[Zn2+]solv
k1
⇌
k−1
[Zn2+]desolv +[solventf ree] (8.2)
[Zn2+]desolv +[CNf ree] →
k2
[Zn2+ −NC]ads (8.3)
-CN Zn2+
Free CN
wavenumber / cm-1
Absorbance (mOD)
-CN Zn2+
Free CN
wavenumber / cm-1
Absorbance (mOD)
Normalized Integrated Area
Time / minute
Normalized Integrated Area
Time / minute
(a) (b)
(c) (d)
Figure 8.6: Time evolution of Lewis peak in (a) Propylene Carbonate, and in (b) water. Integrated
Peak area for Zn2+ bound CN peak and free CN peak with time in (c) propylene carbonate and in
(d) water.
123

We have to keep in mind that in our experiment we have not applied any external bias to drive
the ions to the surface. This indicates that on their own volition if the ions were to overcome the
solvation energy barrier, the timescale will be in the order of minutes. From multiple trials on each
solvent, the order of desolvation time are approximately 2 minutes, 6 minutes and 12 minutes for
DMC, PC and water respectively. This matches with the solubility order of ZnCl2 in the solvents.
We have found out that the solubility limit of DMC, PC and water are approximately 153 mM,
1 M and 32 M. Lower solubility of the salt encourages faster saturation of the nitrile monolayer
with the adsorbed Zn2+. This result indicates that for multivalent ions, when the solubility is high,
the solvent will pay a much bigger role and by careful design of mixture of solvents, lowering the
solubility will help to lower the activation energy barrier for the breaking of the solvation shell
faster.
= −
− + ,

= .
Figure 8.7: A representative fitting with the kinetic model, the data corresponds to the plot shown
in Figure 8.6(d)
Once the adsorbed ion layer is equilibrated, we checked that if we can reversibly break the
Lewis bond at the interface or not. We start with an adsorbed layer of Zn2+ ions on the surface
and then keep adding solvent to break the Lewis bond at the interface. Figure 8.8 (representative
spectra dilution with DMC) shows that as we are increasing solvent, the intensity of the Lewis
124

peak decreasing and correspondingly the free -CN peak intensity is increasing. We repeated the
same procedure for water, propylene carbonate and water-propylene carbonate mixture at 1:3 mole
ratio.
Figure 8.8: Spectral Evolution of free CN peak and Zn2+ coordinated CN peak upon dilution with
Dimethyl Carbonate (DMC)
The integrated peak areas for the Lewis peak at different dilutions are extracted from the baseline corrected spectra and plotted all of them together in Figure 8.9. The x-axis indicates mole
fraction of the Zn2+ ions in the respective solvents. The starting concentrations are different for
different solvents as the solubility limit is different. For DMC, the solubility limit is 153 mM, but
the rate of decrease upon adding more solvent is smaller than PC and water. Solvents who can hold
more Zn2+, it will require a much larger bulk ion concentration to show the ion adsorption effect
as for those solvents the drive to make the Lewis bond at the interface is much smaller.
125

0.15 M 1 M 32 M
Saturation Concentration
~4-5 M
+
Figure 8.9: Integrated peak areas of Zn2+ coordinated CN peak upon dilution with different solvents, Dimethyl Carbonate (DMC), propylene carbonate (PC) , water and mixture of PC and water.
The peak areas are normalized with respect to their respective area at the saturated concentration.
The black dotted line indicates when the Lewis peak intensity is half of the initial peak area.
126

Note that we cannot directly calculate the population of the ion-coordinated species at the
interface as the absorption cross-section of the free -CN and Zn2+ coordinated species are different,
but the integrated peak areas are still important to understand the stability of the Lewis coordination
at different bulk concentration limit. From Figure 8.9, we can estimate the mole fraction of the
Zn2+ in the respective solvent below which the Lewis peak becomes almost half of its initial value
(denoted by the black dotted line in Figure 8.9).
From these experimental observations, we propose a framework in Figure 8.10 which will be
helpful to design electrolyte for optimum ion adsorption. The y-axis in Figure 8.10 denotes the
mole-fraction of solvated Zn2+ ions. The blue dots indicates the saturation mole fractions for Zn2+
ions at which we should get the maximum Lewis peak intensity for the respective solvents and the
line orange dots are those mole fraction points where half of the Lewis peak intensity is lost. Hence,
we can say the surface adduct is stable in the region above the red dotted line. So, to get decent
enough ion adsorption we have to be in the limit of concentration for the Zn2+ ions which stays
in between the blue and red dotted line for different solvent.Development of non-aqueous Zn-ion
battery is an ongoing research area and to prevent unwanted side reactions from water splitting,
co-solvents will play an important role. Figure 8.10 is specific for propylene carbonate and water
mixture, but the type of phase diagrams can be formed with other solvents which are miscible or
semi-miscible with water.
8.4 Conclusion
We have two major findings from this work. First, the ultra-slow kinetics (order of minutes) of the
ion adsorption at the electrode-electrolyte interface. From our experimental results, we conclude
that the desolvation step is the rate limiting step for the specific adsorption process to take place.
The rate of desolvation is slowest for water, where the solubility is the highest and fastest for
dimethyl carbonate (lowest solubility for ZnCl2). Second, we found that we can manipulate the
population of the adsorbed species at the interface by changing the solubility of the ions and it is
127

+
100 % water 100 % PC
0.36
0.23
Solubility limit
Surface adduct
dissolution

+

25 % water
75% PC
Figure 8.10: A proposed framework to design electrolytes to get optimum ion adsorption at the an
electrode
important to find a balance between the drive for giving up the ions to the surface as opposed to
holding the ions in the solvation shell.
128

Chapter 9
Conclusion and ongoing work
This thesis is mainly focused on how benzonitrile Stark probe is a powerful handle to decipher
complicated chemistry taking place both in bulk and at electrode-electrolyte interfaces. We have
also discussed about some limitations in interpretations of this probe - such as in conjunction with
the electrostatics of the environment, inherent property (intramolecular vibrational energy transfer)
of the molecule itself can show systematic broadening in the nitrile peak. Another example which
can lead to ambiguous interpretations is hydrogen bonding effect showing frequency shifts in benzonitrile in the same range of electrostatic effect. Hence, some cautions need to be considered
while using benzonitrile as a reporter of the local electrostatics.
So far benzonitrile has found its usefulness mostly as a reporter of local electrostatics. I would
like to emphasize on the fact that The distinctive spectral signatures due to specific Lewis acidbase type interactions shown by benzonitrile can be explored more. In this context, we have already used this Lewis coordination effect as an indicator of surface adsorbed metal ions (Chapter
8). Different combinations of aqueous and non-aqueous solvents and additives relevant for ion
intercalation batteries are already being investigated in battery research. We believe vibrational
spectroscopy can give more molecular level picture in understanding their functions. In the Zn2+
adsoroption study in Chapter 8 we have observed that ion desolvation process is in the order of
minute timescales. Capacitance measurements for the same systems will give information about
evolution of the electric double layer at the interface.
129

This specific interaction effect can be further explored to determine interfacial pKa. We have
preliminary results on surface tethered benzonitrile and in bulk benzonitrile which shows almost
15-20 cm−1
shift in presence of various acidic alcohols (e.g. fluorinated alcohols), hence can be
used as a probe for local acidity. These spectroscopic measurements in conjunction with electrochemical impedance experiments can give us insight about point of zero free charge in otherwise
neutral electrode surface. This will have applications in understanding mechanisms of pH controlled electrochemistry and other reactions of biological importance.
130

Bibliography
1. Sarkar, S., Patrow, J. G., Voegtle, M. J., Pennathur, A. K. & Dawlaty, J. M. Electrodes as
Polarizing Functional Groups: Correlation between Hammett Parameters and
Electrochemical Polarization. J. Phys. Chem. C 123, 4926–4937 (2019).
2. Sarkar, S., Tseng, C., Maitra, A., Voegtle, M. J. & Dawlaty, J. M. in Emerging Trends in
Chemical Applications of Lasers 199–224 (American Chemical Society, 2021).
doi:10.1021/bk-2021-1398.ch010.
3. Bublitz, G. U. & Boxer, S. G. Stark spectroscopy: applications in chemistry, biology, and
materials science. Annual review of physical chemistry 48, 213–242 (1997).
4. Bagchi, S., Fried, S. D. & Boxer, S. G. A Solvatochromic Model Calibrates Nitriles’
Vibrational Frequencies to Electrostatic Fields. J. Am. Chem. Soc. 134, 10373–10376
(2012).
5. Sorenson, S. A., Patrow, J. G. & Dawlaty, J. M. Solvation Reaction Field at the Interface
Measured by Vibrational Sum Frequency Generation Spectroscopy. J. Am. Chem. Soc 139,
2369–2378. doi:10.1021/jacs.6b11940 (2017).
6. Warshel, A., Sharma, P. K., Kato, M., Xiang, Y., Liu, H. & Olsson, M. H. M. Electrostatic
Basis for Enzyme Catalysis. Chem. Rev. 106, 3210–3235. doi:10.1021/cr0503106
(2006).
7. Warshel, A. Electrostatic Origin of the Catalytic Power of Enzymes and the Role of
Preorganized Active Sites. Journal of Biological Chemistry 273, 27035–27038.
doi:10.1074/jbc.273.42.27035 (1998).
8. Nechay, M. R., Gallup, N. M., Morgenstern, A., Smith, Q. A., Eberhart, M. E. &
Alexandrova, A. N. Histone Deacetylase 8: Characterization of Physiological Divalent
Metal Catalysis. The Journal of Physical Chemistry B 120. PMID: 26996235, 5884–5895.
doi:10.1021/acs.jpcb.6b00997 (2016).
131

9. Ji, Z. & Boxer, S. G. -Lactamases Evolve against Antibiotics by Acquiring Large
Active-Site Electric Fields. Journal of the American Chemical Society 144. PMID:
36399691, 22289–22294. doi:10.1021/jacs.2c10791 (2022).
10. Wu, Y., Fried, S. D. & Boxer, S. G. A Preorganized Electric Field Leads to Minimal
Geometrical Reorientation in the Catalytic Reaction of Ketosteroid Isomerase. Journal of
the American Chemical Society 142. PMID: 32378409, 9993–9998.
doi:10.1021/jacs.0c00383 (2020).
11. Zheng, C., Mao, Y., Kozuch, J., Atsango, A. O., Ji, Z., Markland, T. E., et al. A
two-directional vibrational probe reveals different electric field orientations in solution and
an enzyme active site. Nature Chemistry 14, 891–897 (2022).
12. Pennathur, A. & Dawlaty, J. Electric field maps in enzymes. Nature Chemistry 14,
845–846 (2022).
13. Sarkar, S., Maitra, A., Banerjee, S., Thoi, V. S. & Dawlaty, J. M. Electric fields at
metal–surfactant interfaces: a combined vibrational spectroscopy and capacitance study.
The Journal of Physical Chemistry B 124, 1311–1321 (2020).
14. Chen, L., Li, F., Zhang, Y., Bentley, C. L., Horne, M., Bond, A. M., et al. Electrochemical
reduction of carbon dioxide in a monoethanolamine capture medium. ChemSusChem 10,
4109–4118 (2017).
15. Idris, N. S., Barlow, J. M., Chabolla, S. A., Ziller, J. W. & Yang, J. Y. Synthesis and redox
properties of heterobimetallic Re (bpyCrown-M)(CO) 3Cl complexes, where M= Na+, K+,
Ca2+, and Ba2+. Polyhedron 208, 115385 (2021).
16. Chantarojsiri, T., Reath, A. H. & Yang, J. Y. Cationic Charges Leading to an Inverse
Free-Energy Relationship for NN Bond Formation by MnVI Nitrides. Angewandte Chemie
International Edition 57, 14037–14042.
doi:https://doi.org/10.1002/anie.201805832 (2018).
17. Maitra, A., Das, P., Thompson, B. C. & Dawlaty, J. M. Distinguishing between the
Electrostatic Effects and Explicit Ion Interactions in a Stark Probe. The Journal of Physical
Chemistry B 127, 2511–2520 (2023).
18. Banerjee, S., Zhang, Z.-Q., Hall, A. S. & Thoi, V. S. Surfactant perturbation of cation
interactions at the electrode–electrolyte interface in carbon dioxide reduction. ACS
Catalysis 10, 9907–9914 (2020).
19. Aragones, A. C., Haworth, N. L., Darwish, N., Ciampi, S., Bloomfield, N. J.,
Wallace, G. G., et al. Electrostatic catalysis of a Diels–Alder reaction. Nature 531, 88–91
(2016).
132

20. Zhang, B., Schaack, C., Prindle, C. R., Vo, E. A., Aziz, M., Steigerwald, M. L., et al.
Electric fields drive bond homolysis. Chemical Science 14, 1769–1774 (2023).
21. Stone, I. B., Starr, R. L., Hoffmann, N., Wang, X., Evans, A. M., Nuckolls, C., et al.
Interfacial electric fields catalyze Ullmann coupling reactions on gold surfaces. Chemical
Science 13, 10798–10805 (2022).
22. Fried, S. D. & Boxer, S. G. Electric Fields and Enzyme Catalysis. Annu. Rev. Biochem. 86,
387–415. doi:10.1146/annurev-biochem-061516-044432 (2017).
23. Rhodes, C. J. Electric Fields in Zeolites: Fundamental Features and Environmental
Implications‡. Chem 70, 4–21. doi:doi:10.1515/chempap-2015-0160 (2016).
24. Che, F., Gray, J. T., Ha, S., Kruse, N., Scott, S. L. & McEwen, J.-S. Elucidating the Roles
of Electric Fields in Catalysis: A Perspective. ACS Catal. 8, 5153–5174.
doi:10.1021/acscatal.7b02899 (2018).
25. Das, S., Perez-Ram ´ ´ırez, J., Gong, J., Dewangan, N., Hidajat, K., Gates, B. C., et al.
Core–shell Structured Catalysts for Thermocatalytic, Photocatalytic, and Electrocatalytic
Conversion of CO2. Chem. Soc. Rev. 49, 2937–3004. doi:10.1039/C9CS00713J (10
2020).
26. Wesley, T. S., Roman-Leshkov, Y. & Surendranath, Y. Spontaneous Electric Fields Play a ´
Key Role in Thermochemical Catalysis at MetalLiquid Interfaces. ACS Cent. Sci. 7,
1045–1055. doi:10.1021/acscentsci.1c00293 (2021).
27. Suydam, I. T., Snow, C. D., Pande, V. S. & Boxer, S. G. Electric Fields at the Active Site of
an Enzyme: Direct Comparison of Experiment with Theory. Science 313, 200–204.
doi:10.1126/science.1127159 (2006).
28. B´ım, D. & Alexandrova, A. N. Local Electric Fields As a Natural Switch of Heme-Iron
Protein Reactivity. ACS Catal. 11, 6534–6546. doi:10.1021/acscatal.1c00687 (2021).
29. Shaik, S., Mandal, D. & Ramanan, R. Oriented electric fields as future smart reagents in
chemistry. Nature chemistry 8, 1091–1098 (2016).
30. Walker, D. M., Hayes, E. C. & Webb, L. J. Vibrational Stark Effect Spectroscopy reveals
Complementary Electrostatic Fields created by Protein–protein Binding at the Interface of
Ras and Ral. Phys. Chem. Chem. Phys. 15, 12241–12252. doi:10.1039/C3CP51284C (29
2013).
31. Park, E. S., Andrews, S. S., Hu, R. B. & Boxer, S. G. Vibrational Stark Spectroscopy in
Proteins: A Probe and Calibration for Electrostatic Fields. J. Phys. Chem. B 103,
9813–9817. doi:10.1021/jp992329g (1999).
133

32. Slocum, J. D. & Webb, L. J. Measuring Electric Fields in Biological Matter Using the
Vibrational Stark Effect of Nitrile Probes. Annu. Rev. Phys. Chem. 69, 253–271.
doi:10.1146/annurev-physchem-052516-045011 (2018).
33. Fafarman, A. T., Webb, L. J., Chuang, J. I. & Boxer, S. G. Site-Specific Conversion of
Cysteine Thiols into Thiocyanate Creates an IR Probe for Electric Fields in Proteins. J.
Am. Chem. Soc 128, 13356–13357. doi:10.1021/ja0650403 (2006).
34. Patrow, J. G., Sorenson, S. A. & Dawlaty, J. M. Direct spectroscopic measurement of
interfacial electric fields near an electrode under polarizing or current-carrying conditions.
J. Phys. Chem. C 121, 11585–11592 (2017).
35. Deb, P., Haldar, T., Kashid, S. M., Banerjee, S., Chakrabarty, S. & Bagchi, S. Correlating
nitrile IR frequencies to local electrostatics quantifies noncovalent interactions of peptides
and proteins. J. Phys. Chem. B 120, 4034–4046 (2016).
36. Jo, H., Culik, R. M., Korendovych, I. V., DeGrado, W. F. & Gai, F. Selective Incorporation
of Nitrile-Based Infrared Probes into Proteins via Cysteine Alkylation. Biochemistry 49,
10354–10356. doi:10.1021/bi101711a (2010).
37. Rodgers, J. M., Zhang, W., Bazewicz, C. G., Chen, J., Brewer, S. H. & Gai, F. Kinetic
Isotope Effect Provides Insight into the Vibrational Relaxation Mechanism of Aromatic
Molecules: Application to Cyano-phenylalanine. J. Phys. Chem. Lett 7, 1281–1287.
doi:10.1021/acs.jpclett.6b00325 (2016).
38. Pandey, H. D. & Leitner, D. M. Vibrational States and Nitrile Lifetimes of
Cyanophenylalanine Isotopomers in Solution. J. Phys. Chem. A 122, 6856–6863.
doi:10.1021/acs.jpca.8b06300 (2018).
39. Getahun, Z., Huang, C.-Y., Wang, T., De Leon, B., DeGrado, W. F. & Gai, F. Using ´
Nitrile-Derivatized Amino Acids as Infrared Probes of Local Environment. J. Am. Chem.
Soc 125, 405–411. doi:10.1021/ja0285262 (2003).
40. Lindquist, B. A. & Corcelli, S. A. Nitrile Groups as Vibrational Probes: Calculations of the
CN Infrared Absorption Line Shape of Acetonitrile in Water and Tetrahydrofuran. J. Phys.
Chem. B 112, 6301–6303. doi:10.1021/jp802039e (2008).
41. Maienschein-Cline, M. G. & Londergan, C. H. The CN Stretching Band of Aliphatic
Thiocyanate is Sensitive to Solvent Dynamics and Specific Solvation. J. Phys. Chem. A
111, 10020–10025. doi:10.1021/jp0761158 (2007).
42. Mojica, E.-R. E., Abbas, N., Wyan, L. O., Vedad, J. & Desamero, R. Z. B. in Raman
Spectroscopy in the Undergraduate Curriculum 181–197 (ACS Symposium Series, 2018).
doi:10.1021/bk-2018-1305.ch011.
134

43. Jones, R. N., Forbes, W. F. & Mueller, W. A. The Infrared Carbonyl Stretching band of
Ring substituted Acetophenones. Can. J. Chem. 35, 504–514. doi:10.1139/v57-071
(1957).
44. Schmitz, A. J., Pandey, H. D., Chalyavi, F., Shi, T., Fenlon, E. E., Brewer, S. H., et al.
Tuning Molecular Vibrational Energy Flow within an Aromatic Scaffold via Anharmonic
Coupling. J. Phys. Chem. A 123, 10571–10581. doi:10.1021/acs.jpca.9b08010 (2019).
45. Leitner, D. M. Influence of quantum energy flow and localization on molecular
isomerization in gas and condensed phases. International journal of quantum chemistry 75,
523–531 (1999).
46. Leitner, D. M. Quantum ergodicity and energy flow in molecules. Advances in Physics 64,
445–517 (2015).
47. Yadav, P. K. & Keshavamurthy, S. Breaking of a bond: when is it statistical? Faraday
Discussions 177, 21–32 (2015).
48. Rubtsova, N. I. & Rubtsov, I. V. Vibrational Energy Transport in Molecules studied by
Relaxation-assisted Two-dimensional Infrared Spectroscopy. Ann. Rev. Phys. Chem. 66,
717–738. doi:10.1146/annurev-physchem-040214-121337 (2015).
49. Yue, Y., Qasim, L. N., Kurnosov, A. A., Rubtsova, N. I., Mackin, R. T., Zhang, H., et al.
Band-selective Ballistic energy Transport in Alkane Oligomers: Toward Controlling the
Transport Speed. J. Phys. Chem. B 119, 6448–6456. doi:10.1021/acs.jpcb.5b03658
(2015).
50. Rubtsov, I. V. & Burin, A. L. Ballistic and diffusive vibrational energy transport in
molecules. The Journal of Chemical Physics 150 (2019).
51. Segal, D. & Agarwalla, B. K. Vibrational heat transport in molecular junctions. Annual
review of physical chemistry 67, 185–209 (2016).
52. AGARWALLA, B. K., Segal, D. & Kalantar, N. On the definitions and simulations of
vibrational heat transport in nanojunctions (2020).
53. Dinpajooh, M. & Nitzan, A. Heat conduction in polymer chains with controlled end-to-end
distance. The Journal of Chemical Physics 153 (2020).
54. Leong, T. X., Qasim, L. N., Mackin, R. T., Du, Y., Pascal, R. A. & Rubtsov, I. V.
Unidirectional Coherent Energy Transport via Conjugated Oligo(p -phenylene) Chains. J.
Chem. Phys 154, 134304. doi:10.1063/5.0046932 (2021).
55. Pein, B. C., Sun, Y. & Dlott, D. D. Controlling vibrational energy flow in liquid
alkylbenzenes. The Journal of Physical Chemistry B 117, 10898–10904 (2013).
135

56. Pein, B. C., Sun, Y. & Dlott, D. D. Unidirectional vibrational energy flow in nitrobenzene.
The Journal of Physical Chemistry A 117, 6066–6072 (2013).
57. Pein, B. C. & Dlott, D. D. Modifying vibrational energy flow in aromatic molecules:
Effects of ortho substitution. The journal of physical chemistry A 118, 965–973 (2014).
58. Leitner, D. M. & Pandey, H. D. Asymmetric energy flow in liquid alkylbenzenes: A
computational study. The Journal of Chemical Physics 143 (2015).
59. Segal, D. & Nitzan, A. Heat rectification in molecular junctions. The Journal of Chemical
Physics 122 (2005).
60. Leitner, D. M. Thermal boundary conductance and thermal rectification in molecules. The
Journal of Physical Chemistry B 117, 12820–12828 (2013).
61. Reid, K. M., Pandey, H. D. & Leitner, D. M. Elastic and inelastic contributions to thermal
transport between chemical groups and thermal rectification in molecules. The Journal of
Physical Chemistry C 123, 6256–6264 (2019).
62. Patrow, J. G., Wang, Y. & Dawlaty, J. M. Interfacial Lewis Acid–Base Adduct Formation
Probed by Vibrational Spectroscopy. J. Phys. Chem. Lett 9, 3631–3638.
doi:10.1021/acs.jpclett.8b00470 (2018).
63. Shao, Y., Gan, Z., Epifanovsky, E., Gilbert, A. T., Wormit, M., Kussmann, J., et al.
Advances in Molecular Quantum Chemistry Contained in the Q-Chem 4 Program Package.
Mol. Phys. 113, 184–215. doi:10.1080/00268976.2014.952696 (2015).
64. Fabian, J. & Allen, P. B. Anharmonic Decay of Vibrational States in Amorphous Silicon.
Phys. Rev. Lett. 77, 3839–3842. doi:10.1103/PhysRevLett.77.3839 (1996).
65. Ma, J., Pazos, I. M. & Gai, F. Microscopic Insights into the Protein-stabilizing Effect of
Trimethylamine N-oxide (TMAO). Proc. Natl. Acad. Sci. U.S.A. 111, 8476–8481.
doi:10.1073/pnas.1403224111 (2014).
66. Thammavongsy, Z., Cunningham, D. W., Sutthirat, N., Eisenhart, R. J., Ziller, J. W. &
Yang, J. Y. Adaptable Ligand Donor Strength: Tracking Transannular Bond Interactions in
Tris(2-pyridylmethyl)-azaphosphatrane (TPAP). Dalton Trans. 47, 14101–14110.
doi:10.1039/C8DT03180K (39 2018).
67. Rafiq, S., Bezdek, M. J., Chirik, P. J. & Scholes, G. D. Dinitrogen coupling to a
terpyridine-molybdenum chromophore is switched on by Fermi resonance. Chem 5,
402–416 (2019).
68. Lin, Z., Lawrence, C. M., Xiao, D., Kireev, V. V., Skourtis, S. S., Sessler, J. L., et al.
Modulating Unimolecular Charge Transfer by Exciting Bridge Vibrations. J. Am. Chem.
Soc 131, 18060–18062. doi:10.1021/ja907041t (2009).
136

69. Delor, M., Scattergood, P. A., Sazanovich, I. V., Parker, A. W., Greetham, G. M.,
Meijer, A. J. H. M., et al. Toward Control of Electron Transfer in Donor-acceptor
Molecules by Bond-specific Infrared Excitation. Science 346, 1492–1495.
doi:10.1126/science.1259995 (2014).
70. Schramm, L. L. Surfactants: fundamentals and applications in the petroleum industry
(Cambridge University Press, 2000).
71. Negin, C., Ali, S. & Xie, Q. Most common surfactants employed in chemical enhanced oil
recovery. Petroleum 3, 197–211 (2017).
72. Florence, A. T. & Gillan, J. M. Biological implications of the use of surfactants in
medicines: and the biphasic effects of surfactants in biological systems. Pestic. Sci. 6,
429–439 (1975).
73. Knepper, T. P. & Berna, J. L. Surfactants: properties, production, and environmental
aspects. Compr. Anal. Chem. 40, 1–49 (2003).
74. Malik, M. A., Hashim, M. A., Nabi, F., Al-Thabaiti, S. A. & Khan, Z. Anti-corrosion
ability of surfactants: a review. Int. J. Electrochem. Sci 6, 1927–1948 (2011).
75. Ma, H., Chen, S., Yin, B., Zhao, S. & Liu, X. Impedance spectroscopic study of corrosion
inhibition of copper by surfactants in the acidic solutions. Corros. Sci. 45, 867–882 (2003).
76. Rusling, J. F. Controlling electrochemical catalysis with surfactant microstructures. Acc.
Chem. Res. 24, 75–81 (1991).
77. Wang, J., Zeng, B., Fang, C. & Zhou, X. The influence of surfactants on the
electron-transfer reaction at self-assembled thiol monolayers modifying a gold electrode. J.
Electroanal. Chem. 484, 88–92 (2000).
78. Vittal, R., Gomathi, H. & Kim, K.-J. Beneficial role of surfactants in electrochemistry and
in the modification of electrodes. Adv. Colloid Interface Sci. 119, 55–68 (2006).
79. Banerjee, S., Han, X. & Thoi, V. S. Modulating the Electrode-Electrolyte Interface with
Cationic Surfactants in Carbon Dioxide Reduction. ACS Catal. 9, 5631–5637 (2019).
80. Sun, Z., Ma, T., Tao, H., Fan, Q. & Han, B. Fundamentals and challenges of
electrochemical CO2 reduction using two-dimensional materials. Chem 3, 560–587 (2017).
81. Quan, F., Xiong, M., Jia, F. & Zhang, L. Efficient electroreduction of CO2 on bulk silver
electrode in aqueous solution via the inhibition of hydrogen evolution. Appl. Surf. Sci. 399,
48–54 (2017).
82. Oh, Y. & Hu, X. Organic molecules as mediators and catalysts for photocatalytic and
electrocatalytic CO 2 reduction. Chem. Soc. Rev. 42, 2253–2261 (2013).
137

83. Ovalle, V. J. & Waegele, M. M. Understanding the Impact of N-Arylpyridinium Ions on
the Selectivity of CO2 Reduction at the Cu/Electrolyte Interface. J. Phys. Chem. C 123,
24453–24460 (2019).
84. Liu, M., Pang, Y., Zhang, B., De Luna, P., Voznyy, O., Xu, J., et al. Enhanced
electrocatalytic CO 2 reduction via field-induced reagent concentration. Nature 537, 382
(2016).
85. Resasco, J., Chen, L. D., Clark, E., Tsai, C., Hahn, C., Jaramillo, T. F., et al. Promoter
effects of alkali metal cations on the electrochemical reduction of carbon dioxide. J. Am.
Chem. Soc. 139, 11277–11287 (2017).
86. Chen, L. D., Urushihara, M., Chan, K. & Nrskov, J. K. Electric field effects in
electrochemical CO2 reduction. ACS Catal. 6, 7133–7139 (2016).
87. Gunathunge, C. M., Ovalle, V. J. & Waegele, M. M. Probing promoting effects of alkali
cations on the reduction of CO at the aqueous electrolyte/copper interface. Phys. Chem.
Chem. Phys. 19, 30166–30172 (2017).
88. Wijesuriya, S., Burugapalli, K., Mackay, R., Ajaezi, G. & Balachandran, W. Chemically
roughened solid silver: A simple, robust and broadband SERS substrate. Sensors 16,
1742–1762 (2016).
89. Chattopadhyay, A. & Boxer, S. G. Vibrational Stark effect spectroscopy. J. Am. Chem. Soc.
117, 1449–1450 (1995).
90. Andrews, S. S. & Boxer, S. G. Vibrational stark effects of nitriles I. Methods and
experimental results. J. Phys. Chem. A 104, 11853–11863 (2000).
91. Andrews, S. S. & Boxer, S. G. Vibrational Stark effects of nitriles II. Physical origins of
stark effects from experiment and perturbation models. J. Phys. Chem. A 106, 469–477
(2002).
92. Fried, S. D. & Boxer, S. G. Measuring electric fields and noncovalent interactions using
the vibrational Stark effect. Acc. Chem. Res. 48, 998–1006 (2015).
93. Sorenson, S. A., Patrow, J. G. & Dawlaty, J. M. Solvation reaction field at the interface
measured by vibrational sum frequency generation spectroscopy. J. Am. Chem. Soc. 139,
2369–2378 (2017).
94. Patrow, J. G., Wang, Y. & Dawlaty, J. M. Interfacial Lewis Acid–Base Adduct Formation
Probed by Vibrational Spectroscopy. J. Phys. Chem. Lett 9, 3631–3638 (2018).
95. Clark, M. L., Ge, A., Videla, P. E., Rudshteyn, B., Miller, C. J., Song, J., et al. CO2
reduction catalysts on gold electrode surfaces influenced by large electric fields. J. Am.
Chem. Soc. 140, 17643–17655 (2018).
138

96. Ge, A., Videla, P. E., Lee, G. L., Rudshteyn, B., Song, J., Kubiak, C. P., et al. Interfacial
structure and electric field probed by in situ electrochemical vibrational Stark effect
spectroscopy and computational modeling. J. Phys. Chem. C 121, 18674–18682 (2017).
97. Ge, A., Rudshteyn, B., Videla, P. E., Miller, C. J., Kubiak, C. P., Batista, V. S., et al.
Heterogenized Molecular Catalysts: Vibrational Sum-Frequency Spectroscopic,
Electrochemical, and Theoretical Investigations. Acc. Chem. Res. 52, 1289–1300 (2019).
98. Staffa, J. K., Lorenz, L., Stolarski, M., Murgida, D. H., Zebger, I., Utesch, T., et al.
Determination of the local electric field at Au/SAM interfaces using the vibrational Stark
effect. J. Phys. Chem. C 121, 22274–22285 (2017).
99. Bagchi, S., Boxer, S. G. & Fayer, M. D. Ribonuclease S Dynamics Measured Using a
Nitrile Label with 2D IR Vibrational Echo Spectroscopy. J. Phys. Chem. B 116,
4034–4042 (2012).
100. Aschaffenburg, D. J. & Moog, R. S. Probing hydrogen bonding environments:
solvatochromic effects on the CN vibration of benzonitrile. J. Phys. Chem. B 113,
12736–12743 (2009).
101. Biava, H., Schreiber, T., Katz, S., Voller, J.-S., Stolarski, M., Schulz, C., ¨ et al. Long-Range
Modulations of Electric Fields in Proteins. J. Phys. Chem. B 122, 8330–8342 (2018).
102. Oklejas, V., Sjostrom, C. & Harris, J. M. SERS detection of the vibrational stark effect
from nitrile-terminated SAMs to probe electric fields in the diffuse double-layer. J. Am.
Chem. Soc. 124, 2408–2409 (2002).
103. Shi, H., Cai, Z., Patrow, J., Zhao, B., Wang, Y., Wang, Y., et al. Monitoring Local Electric
Fields at Electrode Surfaces Using Surface Enhanced Raman Scattering-Based Stark-Shift
Spectroscopy during Hydrogen Evolution Reactions. ACS Appl. Mater. Interfaces 10,
33678–33683 (2018).
104. Marcus, Y. Ionic radii in aqueous solutions. Chem. Rev. 88, 1475–1498 (1988).
105. Jimenez- ´ Angeles, F. & Firoozabadi, A. Hydrophobic Hydration and the Effect of NaCl ´
Salt in the Adsorption of Hydrocarbons and Surfactants on Clathrate Hydrates. ACS Cent.
Sci. 4, 820–831 (2018).
106. Mukerjee, P. & Mysels, K. J. Critical micelle concentrations of aqueous surfactant systems
tech. rep. (National Standard reference data system, 1971).
107. Lasia, A. in Modern aspects of electrochemistry 143–248 (Springer, 2002).
108. Orazem, M. E. & Tribollet, B. Electrochemical impedance spectroscopy (John Wiley &
Sons, 2017).
139

109. Yuan, X.-Z. R., Song, C., Wang, H. & Zhang, J. Electrochemical impedance spectroscopy
in PEM fuel cells: fundamentals and applications (Springer Science & Business Media,
2009).
110. Park, S.-M. & Yoo, J.-S. Peer reviewed: electrochemical impedance spectroscopy for
better electrochemical measurements 2003.
111. Chang, B.-Y. & Park, S.-M. Electrochemical impedance spectroscopy. Annu. Rev. Anal.
Chem. 3, 207–229 (2010).
112. Gomes, W. & Vanmaekelbergh, D. Impedance spectroscopy at semiconductor electrodes:
review and recent developments. Electrochim. Acta 41, 967–973 (1996).
113. Garg, S., Li, M., Rufford, T. E., Ge, L., Rudolph, V., Knibbe, R., et al. Catalyst-electrolyte
Interactions in Aqueous Reline Solutions for Highly Selective Electrochemical CO2
Reduction. ChemSusChem (2019).
114. Lee, S., Anderson, L. J., Payne, C. M. & Hafner, J. H. Structural transition in the surfactant
layer that surrounds gold nanorods as observed by analytical surface-enhanced Raman
spectroscopy. Langmuir 27, 14748–14756 (2011).
115. Griffiths, D. J. Introduction to electrodynamics 2005.
116. Jackson, J. D. Classical electrodynamics 1999.
117. Toukmaji, A. Y. & Board Jr, J. A. Ewald summation techniques in perspective: a survey.
Comput. Phys. Commun 95, 73–92 (1996).
118. Wolf, D., Keblinski, P., Phillpot, S. & Eggebrecht, J. Exact method for the simulation of
Coulombic systems by spherically truncated, pairwise r- 1 summation. J. Chem. Phys. 110,
8254–8282 (1999).
119. Boschkova, K. & Stalgren, J. Cationic and nonionic surfactant adsorption on thiol surfaces ˚
with controlled wettability. Langmuir 18, 6802–6806 (2002).
120. Silva, J. A. d., Dias, R. P., da Hora, G. C., Soares, T. A. & Meneghetti, M. R. Molecular
dynamics simulations of cetyltrimethylammonium bromide (CTAB) micelles and their
interactions with a gold surface in aqueous solution. J. Braz. Chem. Soc. 29, 191–199
(2018).
121. Liu, S., Wu, D. & Yang, X. Coarse-grained molecular simulation of self-assembly
nanostructures of CTAB on nanoscale graphene. Mol Simul. 42, 31–38 (2016).
122. Viana, R. B., da Silva, A. B. & Pimentel, A. S. Infrared spectroscopy of anionic, cationic,
and zwitterionic surfactants. Adv. Chem. Phys. 2012, 1–14 (2012).
140

123. Campbell, R. A., Parker, S. R., Day, J. P. & Bain, C. D. External reflection FTIR
spectroscopy of the cationic surfactant hexadecyltrimethylammonium bromide (CTAB) on
an overflowing cylinder. Langmuir 20, 8740–8753 (2004).
124. Jaschke, M., Butt, H.-J., Gaub, H. & Manne, S. Surfactant aggregates at a metal surface.
Langmuir 13, 1381–1384 (1997).
125. Vivek, J. & Burgess, I. J. Quaternary ammonium bromide surfactant adsorption on
low-index surfaces of gold. 1. Au (111). Langmuir 28, 5031–5039 (2012).
126. Endo, O., Matsumura, D., Kohdate, K., Kiguchi, M., Yokoyama, T. & Ohta, T. In-situ
XAFS studies of Br adsorption on the silver (111) electrode. J. Electroanal. Chem. 494,
121–126 (2000).
127. Varela, A. S., Ju, W., Reier, T. & Strasser, P. Tuning the catalytic activity and selectivity of
Cu for CO2 electroreduction in the presence of halides. ACS Catal. 6, 2136–2144 (2016).
128. D’Angelo, P., Migliorati, V. & Guidoni, L. Hydration properties of the bromide aqua ion:
the interplay of first principle and classical molecular dynamics, and X-ray absorption
spectroscopy. Inorg. Chem. 49, 4224–4231 (2010).
129. Yamakata, A. & Osawa, M. Destruction of the hydration shell around tetraalkylammonium
ions at the electrochemical interface. J. Am. Chem. Soc. 131, 6892–6893 (2009).
130. Yamakata, A. & Osawa, M. Destruction of the water layer on a hydrophobic surface
induced by the forced approach of hydrophilic and hydrophobic cations. J. Phys. Chem.
Lett 1, 1487–1491 (2010).
131. Yamakata, A., Soeta, E., Ishiyama, T., Osawa, M. & Morita, A. Real-time observation of
the destruction of hydration shells under electrochemical force. J. Am. Chem. Soc. 135,
15033–15039 (2013).
132. Santos, E., Hindelang, P., Quaino, P., Schulz, E. N., Soldano, G. & Schmickler, W.
Hydrogen Electrocatalysis on Single Crystals and on Nanostructured Electrodes.
ChemPhysChem 12, 2274 (2011).
133. Hammes-Schiffer, S. Theory of Proton-Coupled Electron Transfer in Energy Conversion
Processes. Acc. Chem. Res. 42, 1881 (2009).
134. Hammes-Schiffer, S. & Soudackov, A. V. Proton-Coupled Electron Transfer in Solution,
Proteins, and Electrochemistry. J. Phys. Chem. B 112, 14108 (2008).
135. Weinberg, D. R., Gagliardi, C. J., Hull, J. F., Murphy, C. F., Kent, C. A., Westlake, B. C.,
et al. Proton-Coupled Electron Transfer. Chem. Rev. 112, 4016 (2012).
141

136. Warren, J. J., Tronic, T. A. & Mayer, J. M. Thermochemistry of Proton-Coupled Electron
Transfer Reagents and Its Implications. Chem. Rev. 110, 6961 (2010).
137. Darcy, J. W., Koronkiewicz, B., Parada, G. A. & Mayer, J. M. A Continuum of
Proton-Coupled Electron Transfer Reactivity. Acc. Chem. Res. 51, 2391 (2018).
138. Meyer, T. J., Huynh, M. H. V. & Thorp, H. H. The Possible Role of Proton-Coupled
Electron Transfer (PCET) in Water Oxidation by Photosystem II. Angew. Chem., Int. Ed.
46, 5284 (2007).
139. Bhugun, I., Lexa, D. & Saveant, J.-M. Catalysis of the Electrochemical Reduction of ´
Carbon Dioxide by Iron(0) Porphyrins. Synergistic Effect of Lewis Acid Cations. The
Journal of Physical Chemistry 100, 19981–19985. doi:10.1021/jp9618486 (1996).
140. Keith, J. A., Grice, K. A., Kubiak, C. P. & Carter, E. A. Elucidation of the Selectivity of
Proton-Dependent Electrocatalytic CO2 Reduction by Fac-Re(Bpy)(CO)3Cl. J. Am. Chem.
Soc. 135, 15823 (2013).
141. Mondal, B., Rana, A., Sen, P. & Dey, A. Intermediates Involved in the 2e–/2H+ Reduction
of CO2 to CO by Iron(0) Porphyrin. J. Am. Chem. Soc. 137, 11214 (2015).
142. Nocera, D. G. Chemistry of Personalized Solar Energy. Inorg. Chem. 48, 10001 (2009).
143. Jackson, M. N. & Surendranath, Y. Donor-Dependent Kinetics of Interfacial
Proton-Coupled Electron Transfer. J. Am. Chem. Soc. 138, 3228 (2016).
144. Goldsmith, Z. K., Lam, Y. C., Soudackov, A. V. & Hammes-Schiffer, S. Proton Discharge
on a Gold Electrode from Triethylammonium in Acetonitrile: Theoretical Modeling of
Potential-Dependent Kinetic Isotope Effects. J. Am. Chem. Soc. 141, 1084 (2019).
145. Warburton, R. E., Hutchison, P., Jackson, M. N., Pegis, M. L., Surendranath, Y. &
Hammes-Schiffer, S. Interfacial Field-Driven Proton-Coupled Electron Transfer at
Graphite-Conjugated Organic Acids. J. Am. Chem. Soc. 142, 20855 (2020).
146. Mathew, R., Kayal, S. & Yapamanu, A. L. Excited State Structural Dynamics of
4-Cyano-4’-Hydroxystilbene: Deciphering the Signatures of Proton-Coupled Electron
Transfer Using Ultrafast Raman Loss Spectroscopy. Phys. Chem. Chem. Phys. 21, 22409
(2019).
147. Frydendal, R., Seitz, L. C., Sokaras, D., Weng, T.-C., Nordlund, D., Chorkendorff, I., et al.
Operando Investigation of Au-MnOx Thin Films with Improved Activity for the Oxygen
Evolution Reaction. Electrochim. Acta 230, 22 (2017).
148. Gunathunge, C. M., Li, J., Li, X. & Waegele, M. M. Surface-Adsorbed Co as an Infrared
Probe of Electrocatalytic Interfaces. ACS Catal. 10, 11700 (2020).
142

149. Waegele, M. M., Culik, R. M. & Gai, F. Site-Specific Spectroscopic Reporters of the Local
Electric Field, Hydration, Structure, and Dynamics of Biomolecules. J. Phys. Chem. Lett.
2, 2598 (2011).
150. Waegele, M. M., Gunathunge, C. M., Li, J. & Li, X. How Cations Affect the Electric
Double Layer and the Rates and Selectivity of Electrocatalytic Processes. J. Chem. Phys.
151, 160902 (2019).
151. Gunathunge, C. M., Li, J., Li, X., Hong, J. J. & Waegele, M. M. Revealing the
Predominant Surface Facets of Rough Cu Electrodes under Electrochemical Conditions.
ACS Catal. 10, 6908 (2020).
152. Shaw, S. K., Berna, A., Feliu, J. M., Nichols, R. J., Jacob, T. & Schiffrin, D. J. Role of ´
Axially Coordinated Surface Sites for Electrochemically Controlled Carbon Monoxide
Adsorption on Single Crystal Copper Electrodes. Phys. Chem. Chem. Phys. 13, 5242
(2011).
153. Rey, N. G. & Dlott, D. D. Studies of Electrochemical Interfaces by Broadband Sum
Frequency Generation. J. Electroanal. Chem. 800, 114 (2017).
154. Baldelli, S. Probing Electric Fields at the Ionic Liquid-Electrode Interface Using Sum
Frequency Generation Spectroscopy and Electrochemistry. J. Phys. Chem. B 109, 13049
(2005).
155. Schultz, Z. D., Shaw, S. K. & Gewirth, A. A. Potential Dependent Organization of Water at
the Electrified Metal-Liquid Interface. J. Am. Chem. Soc. 127, 15916 (2005).
156. Shaw, S. K., Lagutchev, A., Dlott, D. D. & Gewirth, A. A. Electrochemically Driven
Reorientation of Three Ionic States of P-Aminobenzoic Acid on Ag(111). J. Phys. Chem.
C 113, 2417 (2009).
157. Heidary, N., Ly, K. H. & Kornienko, N. Probing CO2 conversion chemistry on
nanostructured surfaces with operando vibrational spectroscopy. Nano Lett. 19, 4817–4826
(2019).
158. Kornienko, N., Ly, K. H., Robinson, W. E., Heidary, N., Zhang, J. Z. & Reisner, E.
Advancing Techniques for Investigating the Enzyme-Electrode Interface. Acc. Chem. Res.
52, 1439 (2019).
159. Fried, S. D., Bagchi, S. & Boxer, S. G. Extreme Electric Fields Power Catalysis in The
Active Site of Ketosteroid Isomerase. Science 346, 1510–1514 (2014).
160. Hammes-Schiffer, S. Catalytic Efficiency of Enzymes: A Theoretical Analysis.
Biochemistry 52, 2012 (2013).
143

161. Pennathur, A. K., Voegtle, M. J., Menachekanian, S. & Dawlaty, J. M. Strong Propensity of
Ionic Liquids in Their Aqueous Solutions for an Organic-Modified Metal Surface. J. Phys.
Chem. B 124, 7500 (2020).
162. Voegtle, M. J., Pal, T., Pennathur, A. K., Menachekanian, S., Patrow, J. G., Sarkar, S., et al.
Interfacial Polarization and Ionic Structure at the Ionic Liquid-Metal Interface Studied by
Vibrational Spectroscopy and Molecular Dynamics Simulations. J. Phys. Chem. B 125,
2741 (2021).
163. Goldsmith, Z. K., Secor, M. & Hammes-Schiffer, S. Inhomogeneity of Interfacial Electric
Fields at Vibrational Probes on Electrode Surfaces. ACS Cent. Sci. 6, 304 (2020).
164. Lapidus, S. H., Rajput, N. N., Qu, X., Chapman, K. W., Persson, K. A. & Chupas, P. J.
Solvation Structure and Energetics of Electrolytes for Multivalent Energy Storage. Phys.
Chem. Chem. Phys. 16, 21941 (2014).
165. Guo, C., Ran, J., Vasileff, A. & Qiao, S.-Z. Rational design of electrocatalysts and photo
(electro) catalysts for nitrogen reduction to ammonia (NH3)underambientconditions.
Energy & Environmental Science 11, 45–56 (2018).
166. Herrero, E., Buller, L. J. & Abruna, H. D. Underpotential Deposition at Single Crystal ˜
Surfaces of Au, Pt, Ag and Other Materials. Chem. Rev. 101, 1897 (2001).
167. Schouten, K. J. P., van der Niet, M. J. T. C. & Koper, M. T. M. Impedance Spectroscopy of
H and OH Adsorption on Stepped Single-Crystal Platinum Electrodes in Alkaline and
Acidic Media. Phys. Chem. Chem. Phys. 12, 15217 (2010).
168. Grimme, S., Antony, J., Ehrlich, S. & Krieg, H. A Consistent and Accurate Ab Initio
Parametrization of Density Functional Dispersion Correction (DFT-D) for the 94 Elements
H-Pu. J. Chem. Phys. 132, 154104 (2010).
169. Giannozzi, P. Quantum Espresso: A Modular and Open-Source Software Project for
Quantum Simulations of Materials. J. Phys.: Condens. Matter 21, 395502 (2009).
170. Giannozzi, P. Advanced Capabilities for Materials Modelling with Quantum Espresso. J.
Phys.: Condens. Matter 29, 465901 (2017).
171. Andreussi, O., Dabo, I. & Marzari, N. Revised Self-Consistent Continuum Solvation in
Electronic-Structure Calculations. J. Chem. Phys. 136, 064102 (2012).
172. Lu, Z., Manias, E., Macdonald, D. D. & Lanagan, M. Dielectric Relaxation in Dimethyl
Sulfoxide/Water Mixtures Studied by Microwave Dielectric Relaxation Spectroscopy. J.
Phys. Chem. A 113, 12207 (2009).
144

173. Kastlunger, G., Lindgren, P. & Peterson, A. A. Controlled-Potential Simulation of
Elementary Electrochemical Reactions: Proton Discharge on Metal Surfaces. J. Phys.
Chem. C 122, 12771 (2018).
174. Ge, A., Kastlunger, G., Meng, J., Lindgren, P., Song, J., Liu, Q., et al. On the Coupling of
Electron Transfer to Proton Transfer at Electrified Interfaces. J. Am. Chem. Soc. 142,
11829 (2020).
175. Sarkar, S., Maitra, A., Lake, W. R., Warburton, R. E., Hammes-Schiffer, S. &
Dawlaty, J. M. Mechanistic Insights about Electrochemical Proton-Coupled Electron
Transfer Derived from a Vibrational Probe. Journal of the American Chemical Society 143,
8381–8390. doi:10.1021/jacs.1c01977 (2021).
176. Schkolnik, G., Salewski, J., Millo, D., Zebger, I., Franzen, S. & Hildebrandt, P. Vibrational
Stark Effect of the Electric-Field Reporter 4-Mercaptobenzonitrile as a Tool for
Investigating Electrostatics at Electrode/SAM/Solution Interfaces. Int. J. Mol. Sci. 13,
7466–7482 (2012).
177. Kovacs, G. J., Loutfy, R. O., Vincett, P. S., Jennings, C. & Aroca, R. Distance Dependence
of Sers Enhancement Factor from Langmuir-Blodgett Monolayers on Metal Island Films:
Evidence for the Electromagnetic Mechanism. Langmuir 2, 689 (1986).
178. Bard, A. J., Faulkner, L. R. & White, H. S. Electrochemical methods: fundamentals and
applications (John Wiley & Sons, 2022).
179. Chmiola, J., Largeot, C., Taberna, P.-L., Simon, P. & Gogotsi, Y. Desolvation of Ions in
Subnanometer Pores and Its Effect on Capacitance and Double-Layer Theory. Angew.
Chem., Int. Ed. 47, 3392 (2008).
180. Stachiotti, M. G. First-Principles Study of the Adsorption of NH3 on Ag Surfaces. Phys.
Rev. B: Condens. Matter Mater. Phys. 79, 115405 (2009).
181. Mart´ınez, A. Bonding Interactions of Metal Clusters [Mn (M= Cu, Ag, Au; N = 1–4)] with
Ammonia. Are the Metal Clusters Adequate as a Model of Surfaces? J. Braz. Chem. Soc.
16, 337 (2005).
182. Abild-Pedersen, F., Greeley, J., Studt, F., Rossmeisl, J., Munter, T. R., Moses, P. G., et al.
Scaling Properties of Adsorption Energies for Hydrogen-Containing Molecules on
Transition-Metal Surfaces. Phys. Rev. Lett. 99, 016105 (2007).
183. Calle-Vallejo, F., Loffreda, D., Koper, M. T. M. & Sautet, P. Introducing Structural
Sensitivity into Adsorption-Energy Scaling Relations by Means of Coordination Numbers.
Nat. Chem. 7, 403 (2015).
145

184. Wise, C. F., Agarwal, R. G. & Mayer, J. M. Determining Proton-Coupled Standard
Potentials and X-H Bond Dissociation Free Energies in Nonaqueous Solvents Using
Open-Circuit Potential Measurements. J. Am. Chem. Soc. 142, 10681 (2020).
185. Tang, W., Sanville, E. & Henkelman, G. A Grid-Based Bader Analysis Algorithm without
Lattice Bias. J. Phys.: Condens. Matter 21, 084204 (2009).
186. Sanville, E., Kenny, S. D., Smith, R. & Henkelman, G. Improved Grid-Based Algorithm
for Bader Charge Allocation. J. Comput. Chem. 28, 899 (2007).
187. Henkelman, G., Arnaldsson, A. & Jonsson, H. A Fast and Robust Algorithm for Bader ´
Decomposition of Charge Density. Comput. Mater. Sci. 36, 354 (2006).
188. McCrum, I. T. & Koper, M. T. M. The Role of Adsorbed Hydroxide in Hydrogen
Evolution Reaction Kinetics on Modified Platinum. Nat. Energy 5, 891 (2020).
189. Gorin, C. F., Beh, E. S., Bui, Q. M., Dick, G. R. & Kanan, M. W. Interfacial Electric Field
Effects on a Carbene Reaction Catalyzed by Rh Porphyrins. Journal of the American
Chemical Society 135, 11257–11265. doi:10.1021/ja404394z (2013).
190. Kim, H. & Cho, M. Infrared Probes for Studying the Structure and Dynamics of
Biomolecules. Chemical Reviews 113, 5817–5847. doi:10.1021/cr3005185 (2013).
191. Okur, H. I., Kherb, J. & Cremer, P. S. Cations Bind Only Weakly to Amides in Aqueous
Solutions. Journal of the American Chemical Society 135, 5062–5067.
doi:10.1021/ja3119256 (2013).
192. Buchberger, A. R., Danforth, S. J., Bloomgren, K. M., Rohde, J. A., Smith, E. L.,
Gardener, C. C. A., et al. Condensed-Phase Effects on the Structural Properties of
FCH2CN–BF3 and ClCH2CN–BF3: A Matrix-Isolation and Computational Study. The
Journal of Physical Chemistry B 117, 11687–11696. doi:10.1021/jp405368a (2013).
193. Swanson, B. & Shriver, D. F. Vibrational spectra, vibrational analysis and bonding in
acetonitrile-boron trifluoride. Inorganic Chemistry 9, 1406–1416 (1970).
194. Eigner, A. A., Rohde, J. A., Knutson, C. C. & Phillips, J. A. IR Spectrum of CH3CNBF3
in Solid Neon: Matrix Effects on the Structure of a Lewis AcidBase Complex. The Journal
of Physical Chemistry B 111, 1402–1407. doi:10.1021/jp0656375 (2007).
195. Dereka, B., Lewis, N. H. C., Keim, J. H., Snyder, S. A. & Tokmakoff, A. Characterization
of Acetonitrile Isotopologues as Vibrational Probes of Electrolytes. The Journal of
Physical Chemistry B 126, 278–291. doi:10.1021/acs.jpcb.1c09572 (2022).
146

196. Fulfer, K. D., Galle Kankanamge, S. R., Chen, X., Woodard, K. T. & Kuroda, D. G.
Elucidating the mechanism behind the infrared spectral features and dynamics observed in
the carbonyl stretch region of organic carbonates interacting with lithium ions. The Journal
of Chemical Physics 154, 234504. doi:10.1063/5.0049742 (2021).
197. Zhang, S., Zhang, Y., Ma, X., Lu, L., He, Y. & Deng, Y. Benzonitrile as a Probe of Local
Environment in Ionic Liquids. The Journal of Physical Chemistry B 117, 2764–2772.
doi:10.1021/jp312251q (2013).
198. Błasiak, B., Ritchie, A. W., Webb, L. J. & Cho, M. Vibrational solvatochromism of nitrile
infrared probes: beyond the vibrational Stark dipole approach. Phys. Chem. Chem. Phys.
18, 18094–18111. doi:10.1039/C6CP01578F (27 2016).
199. Xu, J., Deng, G., Wang, Y.-T., Guo, H.-Y., Kalhor, P. & Yu, Z.-W. Local Acid Strength of
Solutions and Its Quantitative Evaluation Using Excess Infrared Nitrile Probes. The
Journal of Physical Chemistry Letters 11, 1007–1012.
doi:10.1021/acs.jpclett.9b03804 (2020).
200. Purcell, K. F. & Drago, R. S. Studies of the Bonding in Acetonitrile Adducts1. Journal of
the American Chemical Society 88, 919–924. doi:10.1021/ja00957a010 (1966).
201. Kang, K., Fuller, J., Reath, A. H., Ziller, J. W., Alexandrova, A. N. & Yang, J. Y.
Installation of internal electric fields by non-redox active cations in transition metal
complexes. Chem. Sci. 10, 10135–10142. doi:10.1039/C9SC02870F (43 2019).
202. Reath, A. H., Ziller, J. W., Tsay, C., Ryan, A. J. & Yang, J. Y. Redox Potential and
Electronic Structure Effects of Proximal Nonredox Active Cations in Cobalt Schiff Base
Complexes. Inorganic Chemistry 56, 3713–3718. doi:10.1021/acs.inorgchem.6b03098
(2017).
203. Chen, L., Bos, M., Grootenhuis, P., Christenhusz, A., Hoogendam, E., Reinhoudt, D., et al.
Stability constants for some divalent metal ion/crown ether complexes in methanol
determined by polarography and conductometry. English. Analytica chimica acta 201,
117–125. doi:10.1016/S0003-2670(00)85330-1 (1987).
204. Alizadeh, N. A comparison of complexation of Li+ ion with macrocyclic ligands
15-crown-5 and 12-crown-4 in binary nitromethane–acetonitrile mixtures by using
lithium-7 NMR technique and ab initio calculation. Spectrochimica Acta Part A:
Molecular and Biomolecular Spectroscopy 78, 488–493.
doi:https://doi.org/10.1016/j.saa.2010.11.014 (2011).
205. Armentrout, P. B., Austin, C. A. & Rodgers, M. T. Alkali Metal Cation Interactions with
15-Crown-5 in the Gas Phase: Revisited. The Journal of Physical Chemistry A 118,
8088–8097. doi:10.1021/jp4116172 (2014).
147

206. Yao, D., Ge, T., Xu, L., Chen, G., Yao, C., Yang, C., et al. Complexation mechanism of
crown ether with indium in the presence of KI: Toward efficient recovery of indium from
secondary resources. Separation and Purification Technology 308, 122936.
doi:https://doi.org/10.1016/j.seppur.2022.122936 (2023).
207. Atwood, D. A., Bott, S. G. & Atwood, J. L. Preparation and structure of the
[YbCl2·15-CROWN-5]+ cation, an example of seven-coordinate ytterbium. Journal of
Coordination Chemistry 16, 93–96 (1987).
208. Willey, G. R., Lakin, M. T. & Alcock, N. W. Crown ether complexation of scandium(III),
yttrium(III) and lanthanum(III) halides. Synthesis and spectroscopic characterisation of
anhydrous cationic metal-oxacrown species and crystal structure of a scandium
benzocrown. J. Chem. Soc., Dalton Trans., 3407–3411. doi:10.1039/DT9930003407 (22
1993).
209. Inokuchi, Y., Ebata, T. & Rizzo, T. R. Solvent Effects on the Encapsulation of Divalent
Ions by Benzo-18-Crown-6 and Benzo-15-Crown-5. The Journal of Physical Chemistry A
119, 8097–8105. doi:10.1021/acs.jpca.5b04450 (2015).
210. Starynowicz, P. Europium(II) complexes with unsubstituted crown ethers. Polyhedron 22,
337–345. doi:https://doi.org/10.1016/S0277-5387(02)01358-X (2003).
211. More, M. B., Ray, D. & Armentrout, P. B. Intrinsic Affinities of Alkali Cations for
15-Crown-5 and 18-Crown-6: Bond Dissociation Energies of Gas-Phase M+Crown Ether
Complexes. Journal of the American Chemical Society 121, 417–423.
doi:10.1021/ja9823159 (1999).
212. Buschmann, H.-J. The formation of 1:1 and 2:1 complexes of crown ethers and cryptands
with K+ in acetonitrile and propylene carbonate. Polyhedron 7, 721–724.
doi:https://doi.org/10.1016/S0277-5387(88)80044-5 (1988).
213. Buschmann, H.-J. & Schollmeyer, E. Complexation of alkaline earth (Group 2A) cations
by non-cyclic, macrocyclic, and macrobicyclic ligands in propylene carbonate solutions.
Thermochimica Acta 211, 13–20.
doi:https://doi.org/10.1016/0040-6031(92)87002-R (1992).
214. Sinta, R., Rose, P. S. & Smid, J. Formation constants of complexes between crown ethers
and alkali picrates in apolar solvents. Application of crown ether network polymers.
Journal of the American Chemical Society 105, 4337–4343. doi:10.1021/ja00351a036
(1983).
215. Kakoty, B., Vengarathody, R., Mukherji, S., Ahuja, V., Joseph, A., Narayana, C., et al. Two
for one: propylene carbonate co-solvent for high performance aqueous zinc-ion batteries –
remedies for persistent issues at both electrodes. J. Mater. Chem. A 10, 12597–12607.
doi:10.1039/D2TA01501C (23 2022).
148

216. Baranski, A. & Fawcett, W. The Formation of Lithium-Aluminum Alloys at an Aluminum
Electrode in Propylene Carbonate. Journal of The Electrochemical Society 129, 901
(1982).
217. Ito, S., Sugimasa, M., Toshimitsu, Y., Orita, A., Kitagawa, M. & Sakai, M. Formation of a
hydrophobic polyiodide complex during cathodic oxidation of iodide in the presence of
propylene carbonate in aqueous solutions, and its application to a zinc/iodine redox flow
battery. Electrochimica Acta 319, 164–174 (2019).
218. Zhang, Y., Ouyang, Y., Luo, Z. & Dong, S. Convenient, Column Chromatography-Free,
and Effective Synthesis of Benzo-21-crown-7 and Its Derivatives. European Journal of
Organic Chemistry 2019, 4741–4744.
doi:https://doi.org/10.1002/ejoc.201900712 (2019).
219. Reed, A. E., Curtiss, L. A. & Weinhold, F. Intermolecular interactions from a natural bond
orbital, donor-acceptor viewpoint. Chemical Reviews 88, 899–926 (1988).
220. Foster, J. P. & Weinhold, F. Natural hybrid orbitals. Journal of the American Chemical
Society 102, 7211–7218. doi:10.1021/ja00544a007 (1980).
221. Reed, A. E. & Weinhold, F. Natural bond orbital analysis of near-Hartree–Fock water
dimer. The Journal of Chemical Physics 78, 4066–4073. doi:10.1063/1.445134 (1983).
222. Choi, J.-H., Oh, K.-I., Lee, H., Lee, C. & Cho, M. Nitrile and thiocyanate IR probes:
Quantum chemistry calculation studies and multivariate least-square fitting analysis. The
Journal of Chemical Physics 128, 134506. doi:10.1063/1.2844787 (2008).
223. Borchers, N., Clark, S., Horstmann, B., Jayasayee, K., Juel, M. & Stevens, P. Innovative
zinc-based batteries. Journal of Power Sources 484, 229309.
doi:https://doi.org/10.1016/j.jpowsour.2020.229309 (2021).
224. Ming, F., Zhu, Y., Huang, G., Emwas, A.-H., Liang, H., Cui, Y., et al. Co-Solvent
Electrolyte Engineering for Stable Anode-Free Zinc Metal Batteries. Journal of the
American Chemical Society 144, 7160–7170. doi:10.1021/jacs.1c12764 (2022).
225. Azcarate, I., Costentin, C., Robert, M. & Saveant, J.-M. Through-Space Charge Interaction ´
Substituent Effects in Molecular Catalysis Leading to the Design of the Most Efficient
Catalyst of CO2-to-CO Electrochemical Conversion. Journal of the American Chemical
Society 138, 16639–16644. doi:10.1021/jacs.6b07014 (2016).
226. Chen, H., Jiang, F., Hu, C., Jiao, Y., Chen, S., Qiu, Y., et al. Electron-Catalyzed
Dehydrogenation in a Single-Molecule Junction. Journal of the American Chemical
Society 143, 8476–8487. doi:10.1021/jacs.1c03141 (2021).
149

227. Wesley, T. S., Roman-Leshkov, Y. & Surendranath, Y. Spontaneous Electric Fields Play a ´
Key Role in Thermochemical Catalysis at MetalLiquid Interfaces. ACS Central Science 7.
PMID: 34235265, 1045–1055. doi:10.1021/acscentsci.1c00293 (2021).
228. Ondrechen, M. J., Briggs, J. M. & McCammon, J. A. A Model for EnzymeSubstrate
Interaction in Alanine Racemase. Journal of the American Chemical Society 123,
2830–2834. doi:10.1021/ja0029679 (2001).
229. Sissi, C. & Palumbo, M. Effects of magnesium and related divalent metal ions in
topoisomerase structure and function. Nucleic Acids Research 37, 702–711.
doi:10.1093/nar/gkp024 (2009).
230. Siddiqui, S. A. & Dubey, K. D. Can the local electric field be a descriptor of catalytic
activity? A case study on chorismate mutase. Phys. Chem. Chem. Phys. 24, 1974–1981.
doi:10.1039/D1CP03978D (4 2022).
231. Shaik, S., Ramanan, R., Danovich, D. & Mandal, D. Structure and reactivity/selectivity
control by oriented-external electric fields. Chem. Soc. Rev. 47, 5125–5145.
doi:10.1039/C8CS00354H (14 2018).
232. Shaik, S., Danovich, D., Joy, J., Wang, Z. & Stuyver, T. Electric-Field Mediated
Chemistry: Uncovering and Exploiting the Potential of (Oriented) Electric Fields to Exert
Chemical Catalysis and Reaction Control. Journal of the American Chemical Society 142,
12551–12562. doi:10.1021/jacs.0c05128 (2020).
233. Shaik, S., de Visser, S. P. & Kumar, D. External Electric Field Will Control the Selectivity
of Enzymatic-Like Bond Activations. Journal of the American Chemical Society 126,
11746–11749. doi:10.1021/ja047432k (2004).
234. Yue, L., Wang, N., Zhou, S., Sun, X., Schlangen, M. & Schwarz, H. The Electric Field as a
“Smart” Ligand in Controlling the Thermal Activation of Methane and Molecular
Hydrogen. Angewandte Chemie International Edition 57, 14635–14639.
doi:https://doi.org/10.1002/anie.201805718 (2018).
235. Geng, C., Li, J., Weiske, T., Schlangen, M., Shaik, S. & Schwarz, H. Electrostatic and
Charge-Induced Methane Activation by a Concerted Double C–H Bond Insertion. Journal
of the American Chemical Society 139. PMID: 28051294, 1684–1689.
doi:10.1021/jacs.6b12514 (2017).
236. Marlton, S. J. P., McKinnon, B. I., Hill, N. S., Coote, M. L. & Trevitt, A. J.
Electrostatically Tuning the Photodissociation of the Irgacure 2959 Photoinitiator in the
Gas Phase by Cation Binding. Journal of the American Chemical Society 143, 2331–2339.
doi:10.1021/jacs.0c11978 (2021).
150

237. Chen, L., Bos, M., Grootenhuis, P., Christenhusz, A., Hoogendam, E., Reinhoudt, D., et al.
Stability constants for some divalent metal ion/crown ether complexes in methanol
determined by polarography and conductometry. English. Analytica chimica acta 201,
117–125. doi:10.1016/S0003-2670(00)85330-1 (1987).
238. Bao, W.-J., Li, J., Li, J., Zhang, Q.-W., Liu, Y., Shi, C.-F., et al. Au/ZnSe-based surface
enhanced infrared absorption spectroscopy as a universal platform for bioanalysis.
Analytical chemistry 90, 3842–3848 (2018).
239. Kakoty, B., Vengarathody, R., Mukherji, S., Ahuja, V., Joseph, A., Narayana, C., et al. Two
for one: propylene carbonate co-solvent for high performance aqueous zinc-ion
batteries–remedies for persistent issues at both electrodes. Journal of Materials Chemistry
A 10, 12597–12607 (2022).
240. Poder, K.-S., Eskusson, J., Lust, E. & J ˜ anes, A. Non-Aqueous Zn-Ion Hybrid ¨
Supercapacitors: Acetonitrile vs Propylene Carbonate Based Electrolyte. Journal of The
Electrochemical Society 170, 060501 (2023).
241. Cao, L., Li, D., Hu, E., Xu, J., Deng, T., Ma, L., et al. Solvation structure design for
aqueous Zn metal batteries. Journal of the American Chemical Society 142, 21404–21409
(2020).
242. Su, C.-C., He, M., Amine, R., Rojas, T., Cheng, L., Ngo, A. T., et al. Solvating power
series of electrolyte solvents for lithium batteries. Energy & Environmental Science 12,
1249–1254 (2019).
243. Sekhon, S., Arora, N. & Singh, H. P. Effect of donor number of solvent on the conductivity
behaviour of nonaqueous proton-conducting polymer gel electrolytes. Solid State Ionics
160, 301–307 (2003).
244. Fawcett, W. R., Liu, G. & Kessler, T. E. Solvent-induced frequency shifts in the infrared
spectrum of acetonitrile in organic solvents. The Journal of Physical Chemistry 97,
9293–9298 (1993).
245. Chekushkin, P. M., Merenkov, I. S., Smirnov, V. S., Kislenko, S. A. & Nikitina, V. A. The
physical origin of the activation barrier in Li-ion intercalation processes: the overestimated
role of desolvation. Electrochimica Acta 372, 137843 (2021).
246. Wise, C. F., Agarwal, R. G. & Mayer, J. M. Determining proton-coupled standard
potentials and X–H bond dissociation free energies in nonaqueous solvents using
open-circuit potential measurements. Journal of the American Chemical Society 142,
10681–10691 (2020).
247. Tshepelevitsh, S., Kutt, A., L ¨ okov, M., Kaljurand, I., Saame, J., Heering, A., ˜ et al. On the
basicity of organic bases in different media. European Journal of Organic Chemistry 2019,
6735–6748 (2019).
151

248. Stauffert, F., Serra-Vinardell, J., Gomez-Grau, M., Michelakakis, H., Mavridou, I., ´
Grinberg, D., et al. Stereodivergent synthesis of right- and left-handed iminoxylitol
heterodimers and monomers. Study of their impact on -glucocerebrosidase activity. Org.
Biomol. Chem. 15, 3681–3705. doi:10.1039/C7OB00443E (17 2017).
152

Appendices
A Supporting Information: Electric Fields Influence Intramolecular
Vibrational Energy Relaxation and Linewidths
A.1 Input Geometry of benzonitrile
Atom x y z
C 0.0000000 0.0000000 0.6027410
C 1.2152965 0.0000000 -0.0912771
H 2.1491027 0.0000000 0.4610900
C 1.2093941 0.0000000 -1.4823904
H 2.1495317 0.0000000 -2.0245159
C 0.0000000 0.0000000 -2.1771439
H 0.0000000 0.0000000 -3.2629541
C -1.2093941 0.0000000 -1.4823904
H -2.1495317 0.0000000 -2.0245159
C -1.2152965 0.0000000 -0.0912771
H -2.1491027 0.0000000 0.4610900
C 0.0000000 0.0000000 2.0463618
N 0.0000000 0.0000000 3.2060088
153

Electric Field
(MV/cm) 1 2 3 4 5 6 7 8 9 10 11 12
-50 158.82 185 406.81 416.86 472.3 584.62 587.63 645.22 714.7 771.12 806.8 884.42
-33 156.54 182.5 404.65 417.64 472.7 581.26 585.35 645.27 716.06 772.84 805.88 886.73
-22 155.23 181.05 403.12 418.17 472.92 579.13 584.04 645.26 716.86 773.89 805.19 887.78
-11 153.93 179.59 401.36 418.73 473.14 577.02 582.83 645.23 717.56 774.89 804.2 888.46
0 152.45 177.79 399.23 419.37 473.34 574.84 581.64 645.18 718.18 775.86 802.88 888.64
11 151.61 176.9 397.49 419.93 473.51 572.98 580.79 645.05 718.77 776.7 801.48 888.88
22 150.62 175.72 395.43 420.61 473.69 571.09 579.97 644.94 719.37 777.53 799.94 888.67
33 149.88 174.89 393.44 421.37 473.93 569.37 579.37 644.87 720.06 778.37 798.42 888.36
50 148.85 173.47 390.19 422.8 474.29 566.91 578.7 644.74 721.23 779.43 795.95 887.46
Mode Number
Electric Field
(MV/cm) 13 14 15 16 17 18 19 20 21 22 23 24
-50 969.32 1014.05 1016.15 1044.61 1048.56 1104.22 1199.94 1192.95 1207.28 1314 1342.96 1466.93
-33 968.07 1014.42 1014.64 1039.03 1048.76 1104.32 1196.99 1192.88 1218.77 1312.95 1343.85 1466.61
-22 967.04 1013.78 1014.37 1035.5 1048.76 1104.26 1194.96 1207.62 1218.02 1312.2 1344.21 1466.19
-11 965.75 1013.04 1014.05 1031.99 1048.8 1104.05 1192.86 1206.67 1217.73 1311.44 1357.37 1465.4
0 964.13 1012.23 1013.26 1028.45 1048.81 1103.64 1190.78 1206.21 1217.69 1310.71 1356.21 1464.06
11 962.57 1012.03 1012.25 1025.48 1048.71 1103.25 1188.64 1206.28 1218.12 1309.85 1355.5 1469.18
22 960.72 1011.78 1022.14 1022.84 1048.58 1102.7 1186.8 1206.24 1218.65 1309.06 1354.71 1467.83
33 958.89 1011.93 1020.77 1021.15 1048.44 1102.17 1185.62 1206.29 1219.47 1308.42 1354.25 1466.76
50 955.68 1012.5 1018.58 1019.95 1048.15 1101.06 1178.64 1206.06 1220.57 1307.27 1353.77 1465.23
Mode Number
Electric Field
(MV/cm) 25 26 27 28 29 30 31 32 33
-50 1516.69 1597.3 1622.04 2296.45 3094.16 3126.27 3128.57 3134.11 3144.54
-33 1516 1599.1 1624.45 2309.93 3094.31 3126 3130.14 3135.66 3144.77
-22 1516.11 1599.77 1626.26 2315.4 3093.88 3124.95 3129.94 3137.28 3144.72
-11 1516.42 1600.33 1627.84 2320.52 3093.37 3123.64 3128.8 3139.41 3144.65
0 1516.74 1600.89 1628.96 2325.9 3092.94 3122.26 3126.49 3142.18 3144.78
11 1517.04 1600.74 1630.14 2329.08 3078.94 3120.35 3123.01 3143.64 3144.44
22 1517.27 1600.65 1630.63 2332.3 3076.16 3118.17 3119.33 3144.47 3145.1
33 1517.53 1600.44 1631.14 2334.62 3072.23 3115.43 3115.82 3144.36 3146.13
50 1517.63 1599.69 1630.92 2337.48 3064.3 3110.77 3110.98 3144.27 3147.52
Mode Number
154

A.2 Computed Frequencies of Benzonitrile at different electric fields
A.3 Third Order Anharmonic Couplings with CN (mode 28) and all other
modes (mode 1 to mode 27)
A.3.1 At Electric Field -50 MV/cm
Mode
Number 3 4 5 6 7 8 9 10 11
3 153.4086 2.69E-07 0 3.4E-08 143.6842 6.07E-08 0.329077 7.56E-08 32.17904
4 0 0.395717 4.46E-08 5.17E-08 2.05E-07 4.22E-08 2.67E-07 3.07E-08 3.79E-08
5 0 0 11.99505 5.6E-09 2.23E-08 1.27E-07 3.12E-08 19.35013 1.03E-07
6 0 0 0 155.8069 8.28E-08 69.18588 5.78E-08 1.08E-07 6.68E-08
7 0 0 0 0 115.0362 5.64E-08 1.853231 3.41E-08 18.80955
8 0 0 0 0 0 30.01894 2.95E-08 5.27E-08 6.83E-08
9 0 0 0 0 0 0 4.539553 5.71E-08 1.516878
10 0 0 0 0 0 0 0 22.93379 3.78E-08
11 0 0 0 0 0 0 0 0 2.963184
Mode
Number 12 13 14 15 16 17 18 19 20
3 1.77E-08 8.715982 9.51E-08 1.45E-08 1.965802 2.36E-08 0 1.56E-08 8.14E-08
4 1.653948 0 0.505737 7.94E-08 3.8E-08 5.76E-08 1.67E-08 1.9E-07 1.65E-08
5 1.06E-07 5.4E-08 6.3E-08 0.756391 4.13E-08 4.646648 1.92E-07 1.86E-06 18.28553
6 3.26E-08 3E-08 2.92E-08 2.4E-07 3.83E-08 5.08E-08 7.953377 1.112554 5.83E-08
7 1.15E-07 6.158066 2.58E-08 2.36E-08 0.942423 6.72E-08 8.95E-08 1.52E-07 6.62E-08
8 1.11E-07 3.07E-08 1.99E-08 1.55E-07 4.89E-08 2.22E-08 4.263663 0.137619 1.87E-07
9 8.03E-08 0.526667 7.19E-08 8.22E-08 1.270936 6.7E-08 9.36E-08 2.65E-07 1.54E-08
10 2.15E-08 1.98E-08 1.25E-07 0.541255 5.68E-08 3.859406 4.43E-07 2.76E-06 27.43908
11 1.2E-07 0.802068 7.13E-08 8.2E-09 0.486079 1.33E-08 1.55E-08 1.75E-08 2.29E-07155

Mode
Number 12 13 14 15 16 17 18 19 20
12 0.804256 3.97E-07 0.185513 1.3E-07 1.2E-07 6.06E-08 1.23E-07 4.39E-07 3.47E-08
13 - 3.592924 2.99E-07 3.42E-08 0.054022 1.53E-07 4.86E-08 7.35E-08 3.2E-08
14 - - 1.533693 1.58E-07 1.61E-07 1.35E-08 1.58E-07 1.25E-07 1.4E-07
15 - - - 2.501986 8.2E-09 0.324029 1.44E-07 1.63E-08 1.142787
16 - - - - 1.082519 9.33E-08 0 1.23E-07 4.59E-08
17 - - - - - 2.360062 1.18E-07 4.8E-07 4.667257
18 - - - - - - 2.271394 0.90983 3.1E-07
19 - - - - - - - 0.784995 2.9E-06
20 - - - - - - - - 29.26301
Mode
Number 21 22 23 24 25 26 27
3 0 2.47E-08 4.92E-08 6.14E-08 1.01E-08 1.94E-08 3.74E-08
4 1.5E-07 0 7.49E-08 9.97E-08 9.84E-08 0 7.58E-08
5 23.32657 9.82E-08 4.89E-08 6.8E-09 8.948181 5.13E-08 7.372814
6 4.81E-08 2.19734 0.556424 0.997072 6.2E-08 3.946411 7.6E-09
7 5.45E-08 4.69E-08 5E-08 4.16E-08 4.93E-08 2.62E-08 2.02E-08
8 1.4E-07 0.088312 1.987327 1.638224 0 0.561887 8.98E-08
9 1.78E-07 9.3E-09 2.79E-08 0 5.73E-08 2.2E-08 1.41E-08
10 32.0592 2.5E-08 5.23E-08 3.11E-08 14.19182 3.9E-09 11.59586
11 1.13E-07 5.56E-08 2.77E-08 4.61E-08 1.14E-08 7.98E-08 1.4E-08156

Mode
Number 21 22 23 24 25 26 27
12 7.16E-08 5.27E-08 1.58E-07 2.62E-08 2.59E-08 2.48E-08 2.39E-08
13 7.91E-08 1.07E-07 8.7E-08 3.62E-08 4.76E-08 7.6E-08 1.47E-08
14 1.15E-07 3.78E-08 2.82E-08 3.52E-08 9.27E-08 5.18E-08 5E-08
15 2.403339 7.77E-08 1.29E-07 1.5E-07 2.067419 9.13E-08 0.599304
16 1.01E-07 3.71E-08 9.7E-08 5.77E-08 6.83E-08 2.91E-08 3.51E-08
17 5.321022 0 8.41E-08 7.88E-08 2.359963 1.65E-08 4.23531
18 3.89E-07 2.646337 2.948654 5.513716 1.71E-07 3.799088 9.91E-08
19 3.25E-06 0.628958 0.650486 0.869101 1.14E-06 0.740852 9.9E-07
20 33.33407 7.27E-08 4.83E-08 1E-08 12.05131 1.9E-08 9.783105
Mode
Number 21 22 23 24 25 26 27
21 37.18857 5.99E-08 8.95E-08 2.48E-08 15.49631 9.92E-08 17.15495
22 - 7.536741 0.770903 4.894666 9.03E-08 8.127531 5.93E-08
23 - - 0.376282 2.973916 1.71E-07 0.608539 9.41E-08
24 - - - 7.910206 2.24E-08 6.55873 1.38E-07
25 - - - - 5.422853 9.4E-09 5.236004
26 - - - - - 4.748317 5.8E-09
27 - - - - - - 2.046151157

A.3.2 At Electric Field -33 MV/cm
Mode
Number 3 4 5 6 7 8 9 10 11
3 -164.981 1.926599 0.000349 0.024915 -144.027 0.001449 0.864873 -0.00167 32.57521
4 - 0.278823 -3.6E-05 -0.00019 1.677191 3.26E-05 -0.01969 0.000258 -0.38019
5 - - -11.9872 0.178147 0.000411 0.093031 1.9E-05 19.19466 0.000804
6 - - - -159.175 -0.00898 -66.3063 -0.00108 0.29386 -0.00239
7 - - - - -107.784 -0.01058 2.943287 -0.001 17.73849
8 - - - - - -26.8241 -0.00033 0.120376 0.000193
9 - - - - - - 4.152272 3.81E-05 -1.85392
10 - - - - - - - -22.8214 -0.00093
11 - - - - - - - - -2.47125
Mode
Number 12 13 14 15 16 17 18 19 20
3 0.045068 9.597269 0.017928 -0.00016 2.201354 -1.3E-05 -5E-05 3.51E-05 -8.1E-06
4 1.445625 -0.11045 0.438007 -9.4E-06 -0.02395 3.73E-05 -3.9E-05 1.32E-05 -0.0004
5 -0.0003 2.11E-06 -0.00014 0.758554 7.57E-05 -4.54661 0.008782 0.102894 -16.0434
6 4.41E-05 -0.00083 3.97E-05 0.007104 -0.00028 -0.05045 7.928858 1.004935 -0.03697
7 0.023708 6.225762 0.015778 -8.9E-05 1.152064 -3.2E-06 0.001328 0.000141 8.55E-05
8 8.5E-06 0.000181 7.32E-06 0.002729 1.4E-05 -0.01246 4.089722 -0.16427 -0.02869
9 0.006614 -0.68392 -0.00416 -0.00011 1.175881 -4.7E-06 0.000148 -5.6E-05 7.63E-05
10 0.000425 -2.3E-05 0.000104 0.675198 7.87E-05 3.748227 -0.03815 -0.16181 24.61751
11 -0.00624 -0.66685 -0.0033 8.17E-05 -0.5294 0.000168 9.9E-05 -7.6E-05 0.001159158

Mode
Number 12 13 14 15 16 17 18 19 20
12 0.761514 0.006082 0.337891 -9.2E-05 0.002198 5.97E-05 -3E-05 1.8E-05 -0.00058
13 - -3.13505 -0.00082 6.41E-06 -0.12507 -1.2E-05 0.000136 -7.6E-05 7.8E-05
14 - - -1.28005 -3.4E-05 0.000913 0.000125 -6.1E-06 1.36E-06 -0.00021
15 - - - 2.533196 0.000322 -0.40232 -0.00189 -0.00639 1.148028
16 - - - - -0.95312 -4E-05 4.3E-05 3.52E-05 0.000123
17 - - - - - -2.13417 0.020679 0.023817 -4.12175
18 - - - - - - 2.100027 -0.84436 0.001551
19 - - - - - - - -0.59287 0.157439
20 - - - - - - - - -23.532
Mode
Number 21 22 23 24 25 26 27
3 7.4E-06 4.52E-05 0.000119 -6.6E-05 6.64E-06 -4.1E-05 -1E-05
4 0.000356 6.61E-06 2.01E-05 8.37E-06 2.11E-06 1.61E-05 -0.00011
5 24.19242 -0.01736 0.010258 -0.03528 8.745428 -0.00798 -7.19057
6 0.126738 -2.48172 0.638617 0.908379 0.044501 3.962788 -0.03753
7 -9.5E-05 -0.00028 0.000152 0.000123 -2.9E-05 0.000567 1.77E-05
8 0.054623 0.006523 1.977949 -1.72553 0.006735 0.392278 -0.00858
9 -5.5E-05 -0.0001 -8.2E-05 5.87E-05 -5.7E-06 -4.7E-05 -1E-06
10 -33.6596 0.058143 -0.03428 0.054962 -14.0847 -0.01373 11.65487
11 -0.00156 -3.1E-05 -9.2E-05 0.000105 -0.00064 6.4E-06 0.000527159

Mode
Number 21 22 23 24 25 26 27
12 0.000457 1.14E-05 4.24E-05 -1.8E-05 0.000105 8.72E-06 2.1E-05
13 -6.5E-05 -4.9E-05 -0.00024 3.75E-05 2.03E-06 -5.5E-05 4.81E-06
14 0.000127 2.71E-05 3.33E-05 -1.6E-05 -9.7E-05 -8.8E-06 4.17E-06
15 -2.42241 0.011474 -0.00041 0.006718 -2.22936 -0.00411 -0.17557
16 -0.00024 -6.3E-05 -7.5E-05 -9.3E-06 -0.0002 -1.3E-05 -7.1E-06
17 5.592976 -0.00569 -0.00142 -0.01179 2.440047 -0.00407 -4.15601
18 -0.03934 2.71283 2.754163 -5.25315 -0.04325 -3.79107 0.025385
19 -0.20768 -0.47496 -0.752 0.874547 -0.05702 -0.61357 0.047633
20 31.42724 -0.05902 0.040529 -0.03428 10.4101 0.019012 -8.30053
Mode
Number 21 22 23 24 25 26 27
21 -41.2367 0.075356 -0.03839 0.066147 -15.9347 -0.01228 17.33144
22 - 8.224436 0.649156 -4.96644 0.010842 -8.23321 -0.01557
23 - - 0.312795 -2.72697 -0.00657 -0.8359 -0.00243
24 - - - 7.666844 0.044844 6.652429 -0.02589
25 - - - - -5.15923 0.02816 5.158704
26 - - - - - 5.243673 -0.00941
27 - - - - - - 1.489643
160

A.3.3 At Electric Field -22 MV/cm
Mode
Number 3 4 5 6 7 8 9 10 11
3 -172.411 1.463723 0.000406 0.009552 -143.821 0.001376 1.589601 -0.00217 32.75078
4 - 0.212566 -3.6E-05 -1.3E-05 1.216314 2.29E-05 -0.02252 0.000251 -0.2768
5 - - -11.9679 0.172021 0.000454 0.085147 1.75E-05 19.08247 0.001017
6 - - - -160.843 -0.00391 -64.5846 -0.00076 0.245518 -0.0005
7 - - - - -102.909 -0.00342 3.54167 -0.00129 16.99586
8 - - - - - -25.0769 -0.00031 0.098904 0.000176
9 - - - - - - 3.903953 7.75E-05 -2.02652
10 - - - - - - - -22.7226 -0.00117
11 - - - - - - - - -2.15245
Mode
Number 12 13 14 15 16 17 18 19 20
3 0.033043 10.14631 0.010604 2.448011 6.88E-05 -8.7E-06 -5.5E-05 2.74E-05 1.08E-05
4 1.315654 -0.08449 0.387745 -0.01905 -5.6E-06 2.72E-05 -3.5E-05 1.13E-05 -0.00037
5 -0.00028 -6.6E-06 -0.00015 -9.1E-06 0.738655 -4.48466 0.009156 0.032736 -14.4507
6 0.000037 -0.00017 4.31E-05 -0.00015 0.007216 -0.04561 7.904631 0.91695 -0.02241
7 0.020846 6.232585 0.011453 1.339828 3.14E-05 1.3E-05 0.000475 2.88E-05 0.000118
8 4.46E-06 0.000175 6.95E-06 2.45E-05 0.002362 -0.00982 4.000436 -0.18672 -0.0219
9 0.006028 -0.78556 -0.00208 1.12188 7.73E-06 -5.7E-06 0.000141 -5.6E-05 6.7E-05
10 0.000407 -8.1E-06 0.000134 1.83E-05 0.784952 3.670615 -0.0358 -0.05394 22.55324
11 -0.00536 -0.58616 -0.00325 -0.53579 4.11E-05 0.000209 8.87E-05 -6.3E-05 0.001331161

Mode
Number 12 13 14 15 16 17 18 19 20
12 0.746098 0.004008 0.440009 0.002295 -8.2E-05 5.35E-05 -2.8E-05 1.43E-05 -0.00052
13 - -2.79114 0.000773 -0.25955 -7.7E-06 -1.2E-05 0.000129 -7.6E-05 6.56E-05
14 - - -1.09913 0.000701 -3.3E-05 0.000114 -7.1E-06 1.57E-06 -0.00022
15 - - - -0.85683 -3.8E-05 5.33E-07 4.82E-05 2.58E-05 1.32E-06
16 - - - - 2.552424 -0.46604 -0.00218 -0.00183 1.108736
17 - - - - - -1.99973 0.020283 0.005353 -3.73762
18 - - - - - - 1.998256 -0.81398 0.00258
19 - - - - - - - -0.4761 0.052839
20 - - - - - - - - -19.7674
Mode
Number 21 22 23 24 25 26 27
3 -2E-05 3.91E-05 0.000115 -6.1E-05 2.77E-06 -4E-05 -8E-06
4 0.000367 6.85E-06 1.93E-05 6.86E-06 1.05E-05 1.47E-05 -0.00011
5 24.74381 -0.01582 0.008885 -0.03213 8.616271 -0.0086 -7.08447
6 0.109467 -2.70201 0.699589 0.852527 0.040335 3.976961 -0.03555
7 -0.00017 -4.1E-05 8.79E-05 2.57E-05 -5E-05 0.00015 3.36E-05
8 0.049249 0.071997 1.974592 -1.78559 0.005811 0.290574 -0.00801
9 -5.3E-05 -9.3E-05 -7.6E-05 5.36E-05 -4.3E-06 -4.3E-05 -2E-06
10 -34.6984 0.053946 -0.03223 0.052703 -14.0122 -0.00914 11.69743
11 -0.00202 -2E-05 -8.4E-05 9.26E-05 -0.00081 6.88E-06 0.000671162

Mode
Number 21 22 23 24 25 26 27
12 0.000465 5.64E-06 3.98E-05 -1.5E-05 0.000104 1.13E-05 1.35E-05
13 -5.7E-05 -4.2E-05 -0.00022 3.64E-05 5.66E-06 -5.2E-05 9.59E-07
14 0.000165 2.52E-05 3.32E-05 -1.6E-05 -8.1E-05 -7.3E-06 -2.8E-06
15 1.72E-05 -6.1E-05 -7.8E-05 -9.9E-06 3.6E-05 -1.7E-05 2.41E-06
16 -2.41046 0.010096 -0.00019 0.006972 -2.31183 -0.00353 0.072954
17 5.763878 -0.00447 -0.00151 -0.0119 2.483898 -0.00616 -4.11382
18 -0.03873 2.76114 2.638683 -5.10171 -0.04182 -3.7917 0.025719
19 -0.07191 -0.39959 -0.80685 0.885606 -0.01146 -0.53219 0.011802
20 29.73836 -0.0503 0.040293 -0.03182 9.27457 0.019388 -7.2621
Mode
Number 21 22 23 24 25 26 27
21 -43.9857 0.073568 -0.03838 0.063971 -16.2078 -0.00841 17.4798
22 - 8.648568 0.58893 -5.0217 0.009433 -8.30832 -0.01116
23 - - 0.259138 -2.56963 -0.00418 -0.94991 -0.00389
24 - - - 7.523757 0.041407 6.695313 -0.02514
25 - - - - -5.00036 0.028264 5.113148
26 - - - - - 5.540749 -0.0107
27 - - - - - - 1.174075163

A.3.4 At Electric Field -11 MV/cm
Mode
Number 3 4 5 6 7 8 9 10 11
3 -179.664 1.208035 0.000541 0.004508 -143.333 0.001331 2.314526 -0.0052 32.89515
4 - 0.131296 -2.8E-05 5.75E-05 0.95901 3.27E-05 -0.02458 0.000256 -0.21934
5 - - -11.9327 0.18045 0.000537 0.086822 1.48E-05 18.96303 0.002717
6 - - - -162.214 -0.00201 -62.9466 -0.00064 0.247332 0.000115
7 - - - - -98.0465 -0.00114 4.084884 -0.00273 16.25266
8 - - - - - -23.5103 -0.0003 0.095417 0.000179
9 - - - - - - 3.6622 0.000287 -2.17135
10 - - - - - - - -22.6094 -0.00309
11 - - - - - - - - -1.83531
Mode
Number 12 13 14 15 16 17 18 19 20
3 0.02603 10.66226 0.00271 2.794847 5.5E-05 -9.4E-06 -6.1E-05 2.85E-05 1.1E-05
4 1.193617 -0.07018 0.331559 -0.01714 9.56E-07 2.49E-05 -3.5E-05 1.01E-05 -0.00034
5 -0.00027 -1.7E-05 -0.00016 -2.8E-06 0.713944 -4.42081 0.008496 0.011594 -12.78
6 4.63E-05 6.34E-05 4.43E-05 -9.5E-05 0.008132 -0.04654 7.880665 0.821825 -0.01896
7 0.019122 6.203599 0.006793 1.578958 2.29E-05 1.31E-05 0.000192 -6.3E-06 0.000104
8 8.2E-06 0.000177 6.59E-06 3.24E-05 0.00238 -0.00915 3.927389 -0.21207 -0.0189
9 0.00516 -0.89129 0.000363 1.070612 5.85E-08 -5E-06 0.000144 -5.7E-05 6.36E-05
10 0.000406 3.69E-05 0.000156 6.4E-05 0.899998 3.590215 -0.03586 -0.02038 20.34658
11 -0.00486 -0.51699 -0.00344 -0.52631 0.000133 0.000534 8.25E-05 -5.8E-05 0.003055164

Mode
Number 12 13 14 15 16 17 18 19 20
12 0.741415 0.002282 0.542815 0.002008 -7.7E-05 5.09E-05 -2.7E-05 1.33E-05 -0.00048
13 - -2.405 0.001804 -0.40297 -6.6E-06 -1.2E-05 0.000126 -8E-05 6.04E-05
14 - - -0.9034 0.000625 -4E-05 0.000115 -7.8E-06 1.55E-06 -0.00023
15 - - - -0.7567 -1.6E-05 -1.9E-06 5.37E-05 1.87E-05 1.38E-05
16 - - - - 2.574266 -0.52472 -0.00238 -0.00047 1.053734
17 - - - - - -1.87584 0.019813 -0.00041 -3.33296
18 - - - - - - 1.910975 -0.79335 0.002036
19 - - - - - - - -0.36865 0.022506
20 - - - - - - - - -16.113
Mode
Number 21 22 23 24 25 26 27
3 -2.2E-05 4.62E-05 0.000115 -6.3E-05 3.71E-06 -4.3E-05 -1.1E-05
4 0.000377 4.16E-06 1.96E-05 8.15E-06 9.99E-06 1.6E-05 -0.00011
5 25.251 -0.01317 0.008667 -0.0311 8.495171 -0.01072 -6.99044
6 0.115028 -2.9431 0.769478 0.797532 0.041475 3.992822 -0.03828
7 -0.00017 4.54E-05 6.56E-05 -5.4E-06 -4.6E-05 8.16E-06 3.01E-05
8 0.049728 0.143536 1.97279 -1.84775 0.005867 0.192174 -0.00821
9 -5.6E-05 -8.7E-05 -7.6E-05 5.11E-05 -4E-06 -4.3E-05 -1.8E-06
10 -35.689 0.051876 -0.03206 0.051852 -13.9475 -0.006 11.74447
11 -0.00534 -6.1E-06 -8.4E-05 9.12E-05 -0.00208 7.65E-06 0.001746165

Mode
Number 21 22 23 24 25 26 27
12 0.000488 3.87E-06 3.84E-05 -1.6E-05 0.000107 9.37E-06 1.08E-05
13 -5.8E-05 -3.8E-05 -0.00022 4.31E-05 5.85E-06 -5.1E-05 9.01E-07
14 0.0002 2.59E-05 3.48E-05 -1.7E-05 -7.6E-05 -7.1E-06 7.52E-07
15 -8E-06 -6.4E-05 -9E-05 -5.7E-06 2.01E-05 -2E-05 3.56E-06
16 -2.39132 0.009299 -0.00014 0.007082 -2.38146 -0.00327 0.303666
17 5.909737 -0.00384 -0.00123 -0.01196 2.513668 -0.00752 -4.0808
18 -0.03917 2.81622 2.52745 -4.96619 -0.04046 -3.79675 0.026154
19 -0.02847 -0.35519 -0.85282 0.903203 0.001947 -0.446 0.001894
20 27.67559 -0.04448 0.03898 -0.02833 8.11043 0.018414 -6.1843
Mode
Number 21 22 23 24 25 26 27
21 -46.6986 0.073602 -0.04033 0.063817 -16.4647 -0.00536 17.63926
22 - 9.051723 0.527843 -5.08712 0.008694 -8.3873 -0.00715
23 - - 0.199882 -2.40772 -0.00325 -1.02466 -0.00385
24 - - - 7.390677 0.03895 6.72591 -0.0256
25 - - - - -4.85857 0.028123 5.073559
26 - - - - - 5.80464 -0.01232
27 - - - - - - 0.895358166

A.3.5 At Electric Field 0 MV/cm
Mode
Number 3 4 5 6 7 8 9 10 11
3 -186.887 -6.3E-07 -1.2E-07 2.9E-08 -142.556 -2.8E-07 3.078367 33.02818 -1.8E-07
4 - 0.047568 -1.7E-08 -1.6E-08 1.03E-06 2.75E-07 -6.4E-07 -2.4E-07 -3.5E-07
5 - - -11.881 2.2E-07 -2.8E-08 1.1E-07 -9E-08 8.47E-08 18.8397
6 - - - -163.368 -2.3E-07 61.32364 2.68E-07 4.7E-09 6.72E-08
7 - - - - -93.1137 -2.5E-07 4.596213 15.5123 -9.1E-08
8 - - - - - -22.0435 -7.6E-08 4.3E-08 -7.9E-09
9 - - - - - - 3.423897 -2.29373 -1.9E-08
10 - - - - - - - -1.52411 5.02E-08
11 - - - - - - - - -22.4969
Mode
Number 12 13 14 15 16 17 18 19 20
3 5.23E-07 -11.1517 2.46E-07 3.255791 -8.4E-08 -4.6E-08 2.8E-07 3.8E-07 1.87E-07
4 -1.07968 1.44E-07 -0.26694 -3.9E-08 -6.1E-08 -2.1E-08 -1.2E-07 1.61E-07 -1.1E-07
5 2.27E-07 -2.4E-07 8.87E-08 8.43E-08 0.683002 4.358772 -2.8E-07 1.19E-07 -11.0064
6 -2.1E-07 8.35E-08 1.22E-07 -8.9E-08 -6.7E-08 -4.9E-07 7.862694 0.727236 2.06E-07
7 2.22E-07 -6.13828 -1.3E-07 1.87074 1.38E-07 -3.2E-07 1.79E-07 2.99E-07 1.67E-07
8 5.42E-07 -3.5E-07 -3.4E-07 1.01E-08 -3E-08 7.5E-09 -3.86424 0.236424 -1.5E-07
9 -3.4E-07 1.005813 9.66E-07 1.019162 -1.8E-07 2.75E-08 2.32E-07 7.29E-08 -4.7E-07
10 1.35E-07 0.468552 2.33E-07 -0.50052 5.91E-08 2.78E-07 1.95E-07 8.56E-08 -3.3E-07
11 2.13E-08 9.61E-08 3.2E-07 -1.2E-07 1.017118 -3.50951 4.56E-08 7.58E-08 17.96509167

Mode
Number 12 13 14 15 16 17 18 19 20
12 0.743871 -2.7E-07 0.643819 -4.5E-07 6.45E-08 -7.6E-08 -4.7E-07 -5E-08 3.69E-07
13 - -1.97116 2.36E-07 0.548121 3.98E-07 -2.5E-07 5.68E-07 1.53E-07 -7.7E-08
14 - - -0.69453 3.56E-07 -9.9E-08 2.3E-07 1.57E-07 -1.8E-07 2.97E-07
15 - - - -0.66048 -1.2E-07 -2.7E-08 -2.5E-07 -9E-09 3.42E-08
16 - - - - 2.594929 0.585256 1.29E-07 -1.3E-07 0.981917
17 - - - - - -1.76213 -9.4E-08 -6.7E-09 2.906026
18 - - - - - - 1.832258 -0.77633 5.08E-07
19 - - - - - - - -0.27269 -1.3E-07
20 - - - - - - - - -12.5887
Mode
Number 21 22 23 24 25 26 27
3 2.66E-07 1.97E-07 -3.1E-07 -9.8E-08 1.21E-07 1.35E-07 2.14E-07
4 2.85E-08 1.02E-07 -2.8E-07 -1.3E-07 -1.4E-07 3.23E-07 7.02E-08
5 25.70559 2.27E-08 4.61E-08 3.67E-08 8.383212 -6E-08 -6.91247
6 1.43E-07 -3.20223 0.855092 0.742608 3.29E-07 4.014078 -5.7E-09
7 5.75E-08 1.1E-07 2.21E-07 -2.6E-07 -3E-07 -2.2E-08 1.41E-07
8 -1.5E-07 -0.21725 -1.97469 1.90858 6.93E-08 -0.09707 2.32E-08
9 3.85E-08 2.81E-07 -1.7E-07 1.71E-07 -3.8E-07 1.1E-07 3.92E-08
10 -2.5E-07 -4.6E-08 -2.8E-07 3.32E-08 -1.3E-07 7.23E-08 -2.7E-08
11 -36.6294 3.24E-08 2.61E-07 -1.5E-08 -13.8948 -7.6E-08 11.80375168

Mode
Number 21 22 23 24 25 26 27
12 -3.9E-07 0 2.59E-07 9.69E-08 -1E-07 1.13E-07 -2E-07
13 2.72E-08 -1.3E-07 7.3E-09 -3E-07 3.43E-08 -1.8E-08 -1.8E-08
14 9.73E-08 7.89E-08 1.82E-07 1.91E-07 -2.5E-07 -3.3E-07 -1.6E-07
15 -2.2E-08 1.42E-07 -9.5E-08 8.77E-08 5.8E-09 8.73E-08 1.34E-07
16 -2.36301 1E-08 6.95E-08 2.37E-08 -2.43656 1.17E-07 0.512343
17 -6.03876 -2E-07 1.96E-07 -1.6E-07 -2.5375 -7.3E-08 4.058197
18 -8.9E-08 2.863665 2.425702 -4.83993 -9.9E-08 -3.80143 -1.1E-07
19 1.02E-07 -0.33608 -0.88845 0.91933 5.64E-08 -0.36137 -8.3E-08
20 25.18121 -3.8E-07 -4.1E-07 5.4E-09 6.905445 2.03E-08 -5.05357
Mode
Number 21 22 23 24 25 26 27
21 -49.3539 -3.4E-08 8.7E-09 2.35E-07 -16.7104 1.79E-07 17.81766
22 - 9.394252 0.469095 -5.149 3.59E-08 -8.45653 -5.1E-09
23 - - 0.138754 -2.24869 4.6E-09 -1.0755 5.6E-09
24 - - - 7.261562 2.92E-08 6.74991 -7.6E-08
25 - - - - -4.74277 1.25E-07 5.05211
26 - - - - - 6.044711 -9.8E-08
27 - - - - - - 0.635408169

A.3.6 At Electric Field 11 MV/cm
Mode
Number 3 4 5 6 7 8 9 10 11
3 -193.355 0.890749 0.000692 -0.00026 -141.546 0.001271 3.6606 33.078 0.00269
4 - -0.06544 -1.4E-05 0.000119 0.647036 4.53E-05 -0.02593 -0.14996 0.000193
5 - - -11.8248 0.199629 0.000588 0.091636 7.88E-06 -0.00181 18.7109
6 - - - -164.074 0.000259 -60.0048 -0.00049 0.000553 0.254912
7 - - - - -88.5117 0.000895 4.969724 14.78974 0.001032
8 - - - - - -20.912 -0.0003 0.00015 0.090627
9 - - - - - - 3.217937 -2.37579 -0.00025
10 - - - - - - - -1.20321 0.001936
11 - - - - - - - - -22.3418
Mode
Number 12 13 14 15 16 17 18 19 20
3 0.018975 -11.5282 0.002624 3.885815 4.37E-05 -1.1E-05 -7.5E-05 3.05E-05 9.37E-06
4 0.977068 0.052392 -0.19852 -0.01659 1.24E-05 1.93E-05 -3.5E-05 8.33E-06 -0.00029
5 -0.00026 2.52E-05 0.00018 8.85E-07 0.636331 -4.29066 0.006721 -0.00097 -9.43095
6 5.81E-05 -0.00029 -4.7E-05 -6.7E-06 0.009673 -0.04862 7.827764 0.606678 -0.00847
7 0.017344 -5.99779 -0.00519 2.242286 1.63E-05 9.04E-06 -7.6E-05 -3.4E-05 7.26E-05
8 1.27E-05 -0.00018 -7E-06 5.18E-05 0.002361 -0.00808 3.826407 -0.27763 -0.01297
9 0.002927 1.125429 -0.00666 0.963722 -4E-06 -3.8E-06 0.000151 -6.3E-05 5.54E-05
10 -0.00402 0.433104 0.004985 -0.45819 -0.0001 -0.00031 8.58E-05 -4.7E-05 -0.00141
11 0.000415 7.18E-05 -0.0002 -6.3E-05 1.157325 3.418444 -0.03584 0.001437 15.81816170

Mode
Number 12 13 14 15 16 17 18 19 20
12 0.756106 -0.00062 -0.74945 -0.00074 -6.9E-05 4.15E-05 -2.6E-05 1.2E-05 -0.00042
13 - -1.46529 0.000162 0.686101 4.07E-06 1.07E-05 -0.00012 8.71E-05 -5.4E-05
14 - - -0.46327 -0.0007 5.04E-05 -0.00012 7.6E-06 -1.9E-06 0.000237
15 - - - -0.58038 -3.8E-06 -4.4E-06 6.58E-05 4.11E-06 2.53E-05
16 - - - - 2.620622 -0.64268 -0.00264 0.000501 0.915695
17 - - - - - -1.65999 0.018839 -0.00422 -2.52495
18 - - - - - - 1.777399 -0.7825 -0.00082
19 - - - - - - - -0.17982 0.005937
20 - - - - - - - - -9.76491
Mode
Number 21 22 23 24 25 26 27
3 -2.4E-05 5.84E-05 0.000116 -6.8E-05 4.82E-06 -4.9E-05 -1.4E-05
4 0.000401 6.08E-07 1.93E-05 9.95E-06 1.01E-05 1.8E-05 -0.00011
5 26.00935 -0.00811 0.008728 -0.02929 8.274449 -0.01339 -6.84095
6 0.126975 -3.48047 0.930323 0.68693 0.043955 4.028564 -0.04291
7 -0.00015 0.000145 3.87E-05 -3E-05 -3.5E-05 -0.00013 2.03E-05
8 0.050585 0.295741 1.969948 -1.97784 0.005943 0.009452 -0.00832
9 -6.1E-05 -7.4E-05 -7.5E-05 4.44E-05 -4.7E-06 -4.3E-05 -1.2E-06
10 0.003345 -8.4E-07 -6.9E-05 6.88E-05 0.001241 1.13E-05 -0.00107
11 -37.3083 0.048499 -0.03268 0.050306 -13.8319 -0.00266 11.85094171

Mode
Number 21 22 23 24 25 26 27
12 0.000549 1.53E-06 3.51E-05 -1.7E-05 0.000124 5.71E-06 -1.6E-06
13 6.54E-05 2.88E-05 0.000215 -5.4E-05 -2.4E-06 4.77E-05 -2.5E-06
14 -0.00027 -2.8E-05 -3.8E-05 1.95E-05 6.61E-05 8.46E-06 -5.6E-06
15 -3.5E-05 -7.4E-05 -0.00012 4.56E-06 1.06E-05 -2.6E-05 6.92E-06
16 -2.29465 0.007944 -0.00022 0.007171 -2.4742 -0.00267 0.698867
17 6.106399 -0.00275 -0.00065 -0.01178 2.535939 -0.00899 -4.03648
18 -0.03952 2.934874 2.320305 -4.7442 -0.03793 -3.822 0.02625
19 0.000843 -0.36005 -0.92362 0.963662 0.009663 -0.26094 -0.00318
20 22.69833 -0.03449 0.035367 -0.02062 5.829954 0.017414 -4.03423
Mode
Number 21 22 23 24 25 26 27
21 -51.3774 0.073711 -0.0452 0.063274 -16.8301 -0.00337 17.92607
22 - 9.74277 0.413434 -5.21693 0.007847 -8.54652 -0.002
23 - - 0.077787 -2.08943 -0.00224 -1.08717 -0.0034
24 - - - 7.159666 0.034799 6.7492 -0.02541
25 - - - - -4.62061 0.026742 5.02374
26 - - - - - 6.240426 -0.0144
27 - - - - - - 0.450566172

A.3.7 At Electric Field 22 MV/cm
Mode
Number 3 4 5 6 7 8 9 10 11
3 -199.741 0.788994 0.000758 -0.00154 -140.302 0.001232 4.272285 33.13613 0.001257
4 - -0.18007 -6.9E-06 0.000142 0.548957 5.24E-05 -0.02599 -0.1284 0.000192
5 - - -11.7537 0.210508 0.000604 0.094771 4.14E-06 -0.00101 18.58111
6 - - - -164.628 0.001067 -58.6878 -0.00045 0.000671 0.260626
7 - - - - -83.8968 0.001393 5.3203 14.09016 0.000372
8 - - - - - -19.8313 -0.00031 0.000152 0.089136
9 - - - - - - 3.019657 -2.43979 -0.00015
10 - - - - - - - -0.88082 0.001028
11 - - - - - - - - -22.1905
Mode
Number 12 13 14 15 16 17 18 19 20
3 -0.0183 -11.8431 0.011121 -4.67006 4.22E-05 -1.1E-05 -8.3E-05 3.22E-05 9E-06
4 -0.8844 0.046807 0.120322 0.017249 1.68E-05 1.84E-05 -3.4E-05 7.69E-06 -0.00027
5 0.000252 3.01E-05 -0.00019 -1.2E-06 0.583844 -4.22366 0.005624 -0.0026 -7.85856
6 -6.4E-05 -0.00035 4.78E-05 -4.3E-05 0.010294 -0.04978 7.798772 0.48845 -0.00243
7 -0.01705 -5.79531 0.014205 -2.66976 1.59E-05 6.91E-06 -0.00015 -3.9E-05 6.18E-05
8 -1.5E-05 -0.00018 6.94E-06 -6.5E-05 0.002334 -0.00767 3.795046 -0.31963 -0.01012
9 -0.00148 1.260342 0.010798 -0.89809 -3.9E-06 -3.5E-06 0.000155 -6.5E-05 5.19E-05
10 0.003697 0.434116 -0.00638 0.404119 -5.9E-05 -0.00015 8.07E-05 -4.5E-05 -0.00063
11 -0.00042 5.11E-05 0.00022 4.53E-05 1.298834 3.326903 -0.03576 0.005195 13.65214173

Mode
Number 12 13 14 15 16 17 18 19 20
12 0.772951 0.000569 -0.85188 -0.00438 6.56E-05 -3.5E-05 2.57E-05 -1.2E-05 0.000386
13 - -0.89163 0.003866 -0.79419 3.03E-06 1.07E-05 -0.00011 8.99E-05 -4.8E-05
14 - - -0.21536 0.000583 -5.5E-05 0.000119 -8.7E-06 1.63E-06 -0.00023
15 - - - -0.52003 2.77E-06 5.84E-06 -7.4E-05 3.7E-06 -2.9E-05
16 - - - - 2.644731 -0.70113 -0.00271 0.000762 0.848658
17 - - - - - -1.56773 0.018344 -0.00489 -2.14895
18 - - - - - - 1.731238 -0.79263 -0.00237
19 - - - - - - - -0.09661 0.004107
20 - - - - - - - - -7.26055
Mode
Number 21 22 23 24 25 26 27
3 -2.8E-05 6.4E-05 0.000116 -6.9E-05 4.33E-06 -5.2E-05 -1.4E-05
4 0.000408 -6.6E-07 1.87E-05 1.06E-05 1.02E-05 1.85E-05 -0.00011
5 26.23379 -0.00569 0.008959 -0.02852 8.175111 -0.01406 -6.78669
6 0.133263 -3.77156 1.021763 0.628824 0.045291 4.046794 -0.04486
7 -0.00014 0.000181 2.89E-05 -3.4E-05 -3.1E-05 -0.00017 1.72E-05
8 0.050885 0.374863 1.967556 -2.04579 0.005962 -0.07586 -0.00826
9 -6.2E-05 -6.7E-05 -7.2E-05 4.04E-05 -4.8E-06 -4.3E-05 -1.1E-06
10 0.001762 9.27E-06 -6.6E-05 6.77E-05 0.00064 1.29E-05 -0.00056
11 -37.8832 0.047227 -0.03342 0.049646 -13.7807 -0.00233 11.91124174

Mode
Number 21 22 23 24 25 26 27
12 -0.00057 4.17E-07 -3.4E-05 1.69E-05 -0.00013 -4.4E-06 6.75E-06
13 6.5E-05 2.27E-05 0.000207 -5.9E-05 -1E-07 4.59E-05 -2.8E-06
14 0.000311 2.91E-05 3.89E-05 -2E-05 -6.2E-05 -8.8E-06 5.64E-06
15 4.62E-05 7.87E-05 0.000142 -1E-05 -9.7E-06 2.99E-05 -8.6E-06
16 -2.21094 0.007378 -0.00033 0.007158 -2.49735 -0.00241 0.862501
17 6.15039 -0.00226 -0.00036 -0.01156 2.527654 -0.00917 -4.02417
18 -0.0394 2.995212 2.223397 -4.65728 -0.03675 -3.84205 0.025974
19 0.005997 -0.41243 -0.94988 1.008057 0.010664 -0.15984 -0.00358
20 20.0197 -0.03011 0.033489 -0.01716 4.778517 0.01726 -3.02057
Mode
Number 21 22 23 24 25 26 27
21 -53.146 0.073804 -0.04796 0.062872 -16.9214 -0.00418 18.0421
22 - 10.01683 0.357815 -5.27304 0.007767 -8.62235 -0.00079
23 - - 0.018534 -1.93449 -0.00212 -1.08083 -0.00304
24 - - - 7.060896 0.033073 6.742315 -0.02486
25 - - - - -4.52365 0.025603 5.018075
26 - - - - - 6.412001 -0.01468
27 - - - - - - 0.280623175

A.3.8 At Electric Field 33 MV/cm
Mode
Number 3 4 5 6 7 8 9 10 11
3 -205.734 0.507043 -0.09793 0.012838 -138.89 -0.00028 4.817127 33.18593 -0.02833
4 - -0.30631 0.000347 -0.00872 0.340855 -0.0031 -0.01376 -0.08187 7.99E-05
5 - - -11.6684 0.060863 -0.07343 0.025419 0.001427 0.052237 18.44767
6 - - - -164.935 -0.0089 -57.5028 -0.00079 -0.00063 0.058097
7 - - - - -79.4585 -0.00592 5.610165 13.42747 0.004379
8 - - - - - -18.9014 -0.00016 0.00029 0.019921
9 - - - - - - 2.84352 -2.47977 0.002912
10 - - - - - - - -0.54324 -0.04537
11 - - - - - - - - -22.0253
Mode
Number 12 13 14 15 16 17 18 19 20
3 -0.02504 -12.0501 0.004133 -5.60976 0.00258 0.00212 -3.2E-05 0.000113 -0.00178
4 -0.80367 0.030624 0.033942 0.013026 2.44E-05 -7.5E-05 0.000532 0.000395 -0.00017
5 0.000208 -0.00711 -6.8E-05 -0.00522 0.527826 4.159004 0.010541 -0.02516 -6.38975
6 8.52E-05 0.000354 -0.00525 0.000308 0.00527 0.012211 7.760406 0.359782 0.002861
7 -0.01388 -5.50728 0.007735 -3.14451 0.001645 0.002764 0.000805 7.51E-06 -0.00427
8 -0.0001 -0.00015 -0.0018 -5.5E-05 -0.00122 -0.00176 3.775583 -0.37205 -0.00269
9 0.000812 1.408917 0.006525 -0.8181 0.000713 -0.00064 0.000156 -6.1E-05 -0.00023
10 0.00058 0.477485 -0.00365 0.344373 0.002185 -0.00627 2.42E-05 2.02E-05 0.022073
11 -0.00041 0.002529 7.61E-05 0.003351 1.444789 -3.23763 -0.01637 0.032931 11.61095176

Mode
Number 12 13 14 15 16 17 18 19 20
12 0.795872 0.000635 -0.95263 -0.00524 4.36E-05 3.12E-05 0.000203 -0.00017 0.000416
13 - -0.24465 0.00361 -0.84691 0.000787 -0.0001 -9.7E-05 7.8E-05 -0.00112
14 - - 0.054182 0.000464 -0.00013 -0.00016 -0.00041 0.000359 -0.00023
15 - - - -0.47901 0.001319 0.000992 -4.9E-05 -4.1E-06 -0.00085
16 - - - - 2.67143 0.755796 0.001331 0.000766 0.793634
17 - - - - - -1.48944 -0.0098 0.006935 1.803813
18 - - - - - - 1.698363 -0.8163 0.002957
19 - - - - - - - -0.01731 -0.01364
20 - - - - - - - - -5.20822
Mode
Number 21 22 23 24 25 26 27
3 0.007325 0.00035 0.000126 -0.00017 -0.00017 -0.00015 0.001322
4 0.000166 0.001109 -0.00022 -0.00051 -6.3E-05 -0.00069 -4.7E-05
5 26.36248 0.052801 -0.00261 -0.00436 8.082729 -0.01544 -6.74893
6 0.046533 -4.0803 1.120555 0.570577 0.026945 4.068282 -0.00977
7 0.01531 -0.00014 9.52E-05 8.23E-07 0.002018 0.000246 -0.00157
8 0.014501 0.454869 1.962998 -2.11624 0.003937 -0.15652 -0.007
9 -0.00115 -3.9E-05 -0.00014 6.88E-05 -0.00072 -5E-05 6.9E-05
10 -0.07259 -0.00013 -4.2E-05 8.82E-05 -0.0248 7.02E-05 0.021728
11 -38.3085 -0.07226 -0.0045 0.011104 -13.7338 0.021826 11.98333177

Mode
Number 21 22 23 24 25 26 27
12 -0.00066 -0.00081 0.000791 -0.00053 -0.00018 0.000323 8.32E-05
13 0.001999 4.28E-05 0.000244 -0.00013 0.000698 9.01E-06 -0.00121
14 0.000279 0.000232 -0.00086 0.001433 -3.2E-05 0.000592 2.75E-05
15 0.00265 0.000132 0.000213 -9.1E-05 -3.5E-05 -3E-06 -0.00188
16 -2.11234 -0.00238 0.002191 0.000742 -2.51042 0.004916 1.009942
17 -6.16939 -0.01912 0.003706 0.001379 -2.50694 0.005557 4.023144
18 -0.03014 3.059863 2.128867 -4.58784 -0.02701 -3.87086 0.002311
19 0.050402 -0.50541 -0.97286 1.068328 0.023133 -0.04751 -0.01996
20 17.35458 0.036013 0.0119 -0.00982 3.805432 -0.00778 -2.06338
Mode
Number 21 22 23 24 25 26 27
21 -54.5207 -0.11306 -0.01554 0.023121 -16.9525 0.032158 18.15122
22 - 10.24838 0.294348 -5.3222 -0.03858 -8.70607 0.000771
23 - - -0.03994 -1.77841 0.000947 -1.04529 -0.01425
24 - - - 6.974219 0.018374 6.726698 0.021578
25 - - - - -4.4356 0.021546 5.027519
26 - - - - - 6.561811 0.010137
27 - - - - - - 0.145081178

A.3.9 At Electric Field 50 MV/cm
Mode
Number 3 4 5 6 7 8 9 10 11
3 -214.214 -7.9E-08 9.52E-08 3.28E-08 -136.397 -6.7E-08 5.588145 33.28146 1.11E-07
4 - -0.51402 -1.1E-07 -5.2E-08 3.03E-07 -1.3E-07 -2.7E-07 1.6E-07 0
5 - - -11.5301 -1.7E-07 5.84E-08 2.9E-08 8.62E-08 2.92E-08 -18.2548
6 - - - -165.131 7.9E-09 -55.7828 -2E-08 -6.5E-08 7.31E-08
7 - - - - -72.9663 3.23E-08 5.962822 12.52456 6.89E-08
8 - - - - - -17.6158 1.04E-08 8.48E-08 1.6E-07
9 - - - - - - 2.604813 -2.49808 -9.9E-09
10 - - - - - - - 0.000393 0
11 - - - - - - - - -21.7706
Mode
Number 12 13 14 15 16 17 18 19 20
3 1.51E-07 -12.2525 1.21E-07 -7.15391 -4.1E-08 -5.6E-08 5.22E-08 -1.8E-07 -8.8E-08
4 0.704412 -2.5E-07 -0.11127 -2.1E-07 -5.2E-08 -5.7E-08 -6.6E-08 0 1.11E-07
5 -5.8E-08 1.04E-08 -5.2E-08 -5.4E-08 0.412579 4.049258 2.26E-07 5.21E-08 -4.53276
6 1.1E-08 5.94E-08 -9.9E-08 9.16E-08 2.97E-07 8.15E-08 7.715011 0.17023 3.85E-08
7 3.24E-08 -4.94882 1.6E-07 -3.84071 -1.3E-08 -2.2E-08 -6.3E-08 1.3E-07 -7.1E-08
8 -5.6E-08 8.08E-08 -4E-08 8.31E-08 1.79E-07 6.81E-08 3.759697 -0.45821 1.47E-07
9 -1.5E-07 1.651975 -1.9E-08 -0.67463 -3.4E-08 -1.4E-07 1.14E-07 7.48E-08 -7.3E-08
10 5.67E-08 0.653855 -1.7E-07 0.265518 3.08E-08 8.88E-08 -4.4E-08 -8.5E-08 -8.2E-08
11 3.21E-08 0 -3.8E-08 -9.9E-09 -1.68912 3.08158 3.4E-07 8.61E-08 -9.00371179

Mode
Number 12 13 14 15 16 17 18 19 20
12 0.838133 2.03E-08 1.09739 -3.1E-07 0 3.03E-08 8.77E-08 -1.4E-07 -1.8E-07
13 - 0.807047 1.82E-08 -0.7731 -3.3E-08 -1.4E-08 6.29E-08 -5.4E-08 1.06E-07
14 - - 0.494429 1.12E-07 8.3E-09 -1.4E-08 3.15E-08 0 0
15 - - - -0.40943 6.79E-08 8.39E-08 8.09E-08 1.87E-08 -1.3E-07
16 - - - - 2.704593 0.843182 0 -1.7E-07 0.729203
17 - - - - - -1.38387 -2.6E-07 -2E-07 1.368894
18 - - - - - - 1.672101 -0.86645 9.17E-08
19 - - - - - - - 0.090041 -5.3E-08
20 - - - - - - - - -3.00744
Mode
Number 21 22 23 24 25 26 27
3 1.6E-08 -4.9E-08 -6.1E-08 -1.9E-08 1.88E-08 -1.8E-08 -7E-08
4 -7.1E-08 -3.5E-08 4.62E-08 -2.4E-08 -1.2E-07 7.5E-09 1.46E-08
5 26.40165 1.95E-08 3.39E-08 -6.6E-08 7.959226 1.23E-08 -6.71818
6 -1.6E-08 -4.54296 1.284472 0.464827 1.85E-08 4.084929 3.8E-09
7 5.4E-08 -6.8E-08 1.41E-07 -2E-07 -1.1E-07 9.69E-08 -4.5E-08
8 -2.2E-07 0.574929 1.954292 -2.22432 -9.5E-08 -0.2746 2.33E-08
9 3.99E-08 -2.6E-08 -1.5E-08 7.1E-08 1.17E-08 7.37E-08 4.33E-08
10 -3.6E-08 4.74E-08 -4.1E-08 0 3.17E-08 6.7E-09 0
11 38.66754 -4.5E-09 -6.2E-08 -9.7E-08 13.668 3.77E-08 -12.0884180

Mode
Number 21 22 23 24 25 26 27
12 -1.1E-08 -5.7E-08 3.29E-08 1.15E-07 -1.3E-07 3.19E-08 7.8E-09
13 -7.7E-08 -8.5E-09 -5.9E-08 -2.3E-08 -1.1E-08 6.43E-08 5.6E-08
14 -8.7E-08 1.61E-07 -5.9E-08 9.19E-08 0 1.07E-07 -7E-09
15 6.96E-08 -4.4E-08 -1.5E-07 2.36E-08 -5.8E-08 0 -2.2E-08
16 -1.8837 1.19E-07 8.7E-08 -2E-07 -2.48543 -1.4E-07 1.166228
17 -6.11821 6.3E-09 1.1E-08 -2.6E-08 -2.44804 -6.4E-08 4.020882
18 -3.4E-07 3.132116 2.001793 -4.50894 -1.3E-07 -3.91071 1.82E-08
19 -1.4E-07 -0.69518 -0.99201 1.164234 3.39E-08 0.123034 1.05E-07
20 13.79612 5.76E-08 2.58E-07 -5.6E-08 2.59384 -6.9E-09 -0.82664
Mode
Number 21 22 23 24 25 26 27
21 -55.8284 -4.5E-08 2.35E-08 1.47E-07 -16.9188 -7.2E-08 18.30495
22 - 10.4466 0.222317 -5.3603 -1.1E-07 -8.78581 -3.6E-08
23 - - -0.11984 -1.56265 -1.1E-07 -0.97432 -5.7E-09
24 - - - 6.860033 4.3E-08 6.670439 6.63E-08
25 - - - - -4.34056 8.91E-08 5.079179
26 - - - - - 6.704887 8.2E-09
27 - - - - - - -0.01194181

B Supporting Information: Electric Fields at Metal-Surfactant
Interfaces: A Combined Vibrational Spectroscopy and Capacitance
Study
10 mm 10 mm 100 nm
Figure 9.1: SEM images of etched Ag substrate
182

x
y
Point charge
r
f
z
Figure 9.2: Sheet of discrete charges giving rise to electric field at z
ζ (z/d,n,m) = z
d
∞
∑
j=−∞
∞
∑
k=−∞
1

(z/d)
2 + (j +m)
2 + (k +n)
2
3/2
(B.1)
We have evaluated the ζ (z/d) for lattices in which the positions of the charges are randomly
shifted from that of the square lattice. This will necessitate that we generalize the ζ (z/d) function
to ζ (z/d,n,m), where n and m assume values between 0 and 1 and indicate the displacement from
the line along which the field is calculated with respect to the lattice. A few instances of this
function are plotted in Figure Figure 9.3, and shows that for large values of z/d regardless of how
we are aligned with the lattice, the function approaches the continuous limit 2π. However, for
small values of z/d depending on whether one is directly aligned above a charge (m = n = 0) or
misaligned (m ̸= 0;n ̸= 0) the asymptotic value is either approached from above or below.
1

Figure 9.3: ζ (z/d) for lattices in which the positions of the charges are randomly shifted from that
of the square lattice.
184

-
-
-
a)
b)
c)
1.014 nm
1.522 nm
2.028 nm
0.7 nm
Figure 9.4: Cartoon representation of a) OTAB, b) DTAB and, c) CTAB when the hydrophobic
tails are pointing towards the nitriles and are penetrating the SAM layer
Figure 9.5: Calculated field for CTAB, DTAB and OTAB vs z where surfactant tails are pointing
toward the surface (Figure Figure 9.4), possibly intercalating with the benzonitrile monolayer,
keeping the charged head along with its counter ion far away from the surface.
185

C Supporting Information: Mechanistic Insights about Electrochemical
Proton-Coupled Electron Transfer Derived from a Vibrational
Probes
C.1 Cyclic Voltammetry
Figure 9.6: Cyclic voltammogram (10 mV s−1
) of polycrystalline silver recorded in DMSO containing 100 mM MAMBN-H+ (reactant), and 300 mM tetrabutylammonium hexafluorophosphate
(supporting electrolyte)
186

C.2 Open Circuit Potential (OCP) Measurement to Determine the Equilibrium
Potential of Triethyl Amine and Triethyl Ammonium Redox Couple
We followed the procedure described previously[246] to measure the equilibrium potential for the
triethylamine/triethylammonium couple, which is closely related to the redox couple studied in
this work. 50 mM Triethylamine and 125 mM triethyl ammonium chloride were used as stock
solution. 300 mM tetrabutylammonium hexafluorophosphate (as supporting electrolyte), 2.5 M
1:1 Pyridine: Pyridinium chloride (as buffer, to overcome the hydrogen bonding effect within
the redox couple) were added to both the stock solutions. DMSO was used as solvent. To a 10
ml solution of 50 mM triethyl amine, 500 µl of 125 mM triethyl ammonium chloride solution
was added successively, and open circuit potential was measured after each addition using Gamry
Reference 3000 potentiostat. In this measurement, Ag/AgCl was used as reference electrode and
250 µm silver foil as working electrode. Data was recorded every 1 second over a duration of 10
minutes for each addition of the protonated species. Stability of the measurements was set such
that the drift of the potential is not more than 0.005 mV/s. There was continuous N2 flow and
stirring at a constant rate during the experiment. Before using as the working electrode in the OCP
experiment, the silver foil was held at -0.5 V for ∼3 minutes in 3 M KCl solution to remove trace
amount of silver oxide, if any.
Figure 9.7: (a) The supporting electrolyte. (b,c) The redox couple.
From each of the OCP vs time plots, potential values are averaged over last 5 minutes and used
to make OCP vs mole-ratio plot ( Figure 9.8). The reaction potential can be obtained using the
following:[246]
187

E = E
0 −
0.0592
n
log [Et3N]
[Et3N+]
−
0.0592
n
log [Pyr]
[PyrH+]
−0.0592pka (C.1)
The potential at 1:1 mole ratio of the protonated and deprotonated species represents the Yintercept of Figure 9.8:
E = E
0 −0.0592pKa (C.2)
The Nernstian portion of the data in the figure and the pKa of pyridine/pyridinium buffer in
DMSO[247] was used to estimate E0
.
E
0 = 0.05997pKa +Yintercept = 0.05997 ∗ 3.4−0.26501 = −0.0611V (C.3)
Figure 9.8: The open circuit potential as a function of the log of concentration ratio of the deprotonated and protonated species.
188

D Supporting Information: Distinguishing between the Electrostatic
Effects and Explicit Ion Interactions in a Stark Probe
D.1 Synthesis
Figure 9.9: Synthesis of B15C5-CN and B18C6-CN
Synthesis of (1) and (2): Compounds (1) and (2) were synthesized according to literature
procedure.[248] To a 100 mL three-necked round bottom flask, a solution of tetraethylene glycol
(5 g, 25.74 mmol) in CH2Cl2 (25 mL), 4-Toluenesulfonyl chloride (TsCl) (9.82 g, 51.49 mmol) was
added and the mixture was cooled to 0°C. Then KOH (11.56 g, 205.94 mmol) was added portion
wise maintaining the temperature below 5 °C. The reaction mixture was stirred at 0 °C for 4 h.
After the completion of the reaction, a mixture of ice and water was added, followed by extraction
with DCM (x 3). The combined organic layers were washed with brine, dried over MgSO4, filtered
and concentrated under reduced pressure to afford (1) (12.9 g, 99.9%) as a colorless oil. Compound
(2) was prepared in a similar fashion using pentaethylene glycol (6.9 g, 100%).
189

(1) 1H-NMR (CDCl3, 400 MHz, room temperature): δ 2.44 (s, 6H), 3.52-3.58 (m, 8H),
3.65-3.69 (m, 4H), 4.12-4.18 (m, 4H), 7.30-7.36 (d, 4H), 7.76-7.82 (d, 4H).
(2) 1H-NMR (CDCl3, 500 MHz, room temperature): δ 2.44 (s, 6H), 3.56-3.58 (m, 8H), 3.59
(s, 4H), 3.66-3.69 (m, 4H), 4.12-4.18 (m, 4H), 7.30-7.36 (d, 4H), 7.76-7.82 (d, 4H).
B15C5-CN: 1H-NMR (400 MHz, CDCl3, CDCl3, room temperature): δ 7.25 (d, 1H), 7.07
(s, 1H), 6.86 (d, 1H), 4.1-4.19 (m, 4H), 3.88-3.94 (m, 4H), 3.72-3.77 (m, 8H).
B18C6-CN: 1H-NMR (400 MHz, CDCl3, room temperature): δ 7.25 (d, 1H), 7.08 (s, 1H),
6.87 (d, 1H), 4.19 (m, 2H), 4.15 (m, 2H), 3.93 (m, 4H), 3.77 (m, 4H), 3.71 (m, 4H), 3.67 (m, 4H).
D.1.1 NMR Characterization
NMR spectra for all compounds were recorded at 25 °C using CDCl3 on either a Varian Mercury
400 MHz or Varian VNMRS-500 MHz. All spectra were referenced to CHCl3 (7.26 ppm).
Figure 9.10: 1H-NMR of (1) in CDCl3 at 25 °C and 400 MHz.
190

Figure 9.11: 1H-NMR of (2) in CDCl3 at 25 °C and 500 MHz.
Figure 9.12: 1H-NMR of B15C5-CN in CDCl3 at 25 °C and 400 MHz.
191

Figure 9.13: 1H-NMR of B18C6-CN in CDCl3 at 25 °C and 400 MHz.
D.2 IR spectra
Normalized Absorbance (a.u.)
wavenumber / cm-1
Figure 9.14: Metal ion complexes with Benzo-15-crown-5-CN in propylene carbonate
192

Normalized Absorbance (a.u.)
wavenumber / cm-1
Figure 9.15: Metal ion complexes with Benzo-18-crown-6-CN in propylene carbonate
D.2.1 Effect of Solvent
We have studied the effect of solvents with one representative cation - Zn2+ with benzonitrile in
DMSO and DCM. DMSO and DCM have vastly different properties, and therefore, are useful to
highlight the competition between the solvent and the probe. DMSO is a high dielectric solvent
(46.68), which can also Lewis-coordinate with the cations, while DCM is a low dielectric solvent
(8.93) with no significant Lewis basicity. Therefore, DMSO is expected to strongly solvate the
cations compared to DCM. When the nitrile probe is present in these two solvents, it should have
a difficult time to desolvate a cation from DMSO and coordinate with it, while it should easily
coordinate with cations in DCM. Our results (Fig.S Figure 9.16) confirm the above expectation.
The figure shows that there is no Lewis peak for the probe in DMSO, while a very strong Lewis
peak is observed in DCM. A further extension of this study could be using the probe to understand
relative solvation of cations in various solvents.
193

(a)
Lewis Peak
(b)
No Lewis Peak
Zn x 2+ Zn2+
Figure 9.16: Effect of different solvents on the Lewis coordination of the Zn2+ with the benzonitrile probe. 400 mM of benzonitrile concentration, Zn2+ salt are added. (a) In DCM the cation is
not well-solvated, and therefore are easily coordinated with the probe, giving rise to a large Lewis
peak. (b) DMSO has a higher dielectric constant and is a Lewis base, and therefore, can solvate
the cation much better than DCM. Therefore, the probe cannot coordinate with the cation and no
Lewis signature is observed.
D.2.2 IR spectra data analysis
Third order polynomial has been used for baseline corrections for all the IR spectra presented in
the paper: y = p1x
3 + p2x
2 + p3x+ p4
Fig.6.c shows integrated peak area for electrostatic (red) and Lewis peak (blue). For the electrostatic peak the integration range is from 2205.7 cm−1
to 2245.2 cm−1
and for the Lewis peak it
is from 2260.2 cm−1
to 2305 cm−1
. They are nomalized to the peak area at the first data point (at
400 mM B15C5-CN and 0 mM Cu2+ concentration).
In Fig. 6.d, the frequencies for the electrostatic peak (lower panel) and Lewis peak (top panel)
are extracted from the central frequency of the respective peaks at full width half maxima of spectra
at each data point.
194

D.3 Computational Details
D.3.1 Input file for geometry optimization and frequency calculation of Benzonitrile
$molecule
0 1
C 2.04659 -0.00014 -0.00001
C 0.61448 -0.00008 -0.00001
C -0.09223 1.21938 -0.00000
H 0.45591 2.15486 -0.00000
C -1.48819 1.21347 0.00000
H -2.02927 2.15367 0.00000
C -2.18772 0.00009 0.00000
H -3.27282 0.00016 0.00001
C -1.48834 -1.21338 0.00000
H -2.02953 -2.15351 0.00000
C -0.09238 -1.21946 -0.00000
H 0.45564 -2.15501 -0.00000
N 3.22098 0.00007 0.00001
$ end
$ rem
BASIS = LANL2DZ
ECP = fit-LANL2DZ
GUI = 2
METHOD = B3LYP
NBO = 2
SCF MAX CYCLES = 200
$ end
@@@
195

$molecule
read
$end
$rem
BASIS = LANL2DZ
ECP = fit-LANL2DZ
GUI = 2
JOB TYPE = Frequency
METHOD = B3LYP
RAMAN = 1
SCF CONVERGENCE = 8
SCF MAX CYCLES = 200
$ end
D.3.2 Example input file for single point energy and frequency calculation of Benzonitrile
in presence of point charges
$molecule
0 1
C 2.04659 -0.00014 -0.00001
C 0.61448 -0.00008 -0.00001
C -0.09223 1.21938 -0.00000
H 0.45591 2.15486 -0.00000
C -1.48819 1.21347 0.00000
H -2.02927 2.15367 0.00000
C -2.18772 0.00009 0.00000
H -3.27282 0.00016 0.00001
196

C -1.48834 -1.21338 0.00000
H -2.02953 -2.15351 0.00000
C -0.09238 -1.21946 -0.00000
H 0.45564 -2.15501 -0.00000
N 3.22098 0.00007 0.00001
$end
$external charges
+5.47098 +0.00007 +0.00001 2
+7.47098 +0.00007 +0.00001 -2
$end
$ rem
BASIS = LANL2DZ
ECP = fit-LANL2DZ
GUI = 2
METHOD = B3LYP
NBO = 2
SCF MAX CYCLES = 200
$ end
@@@
$molecule
read
$end
$rem
BASIS = LANL2DZ
ECP = fit-LANL2DZ
GUI = 2
JOB TYPE = Frequency
197

METHOD = B3LYP
RAMAN = 1
SCF CONVERGENCE = 8
SCF MAX CYCLES = 200
$end
$external charges
+5.47098 +0.00007 +0.00001 2
+7.47098 +0.00007 +0.00001 -2
$end
D.3.3 Example input file for single point energy and frequency calculation of Benzonitrile
in presence of Zn2+ and negative point charge
$molecule
2 1
C 2.04659 -0.00014 -0.00001
C 0.61448 -0.00008 -0.00001
C -0.09223 1.21938 -0.00000
H 0.45591 2.15486 -0.00000
C -1.48819 1.21347 0.00000
H -2.02927 2.15367 0.00000
C -2.18772 0.00009 0.00000
H -3.27282 0.00016 0.00001
C -1.48834 -1.21338 0.00000
H -2.02953 -2.15351 0.00000
C -0.09238 -1.21946 -0.00000
H 0.45564 -2.15501 -0.00000
198

N 3.22098 0.00007 0.00001
Zn +5.47098 +0.00007 +0.00001
$end
$external charges
+7.47098 +0.00007 +0.00001 -2
$end
$rem
BASIS = LANL2DZ
ECP = fit-LANL2DZ
GUI = 2
METHOD = B3LYP
NBO = 2
SCF MAX CYCLES = 200
$ end
@@@
$molecule
read
$end
$rem
BASIS = LANL2DZ
ECP = fit-LANL2DZ
GUI = 2
JOB TYPE = Frequency
METHOD = B3LYP
RAMAN = 1
SCF CONVERGENCE = 8
SCF MAX CYCLES = 200
199

$end
$external charges
+7.47098 +0.00007 +0.00001 -2
$end
200

D.3.4 Table S1 : Nitrile frequencies of benzonitrile at different distances in the two computaional
models
Distance (Å) CN frequency (cm-1
)
(in presence of point charge)
CN frequency (cm-1
)
(In presence of Zn2+)
1.65 2269.73 2359
1.8 2272.25 2316.12
1.95 2278.64 2295.24
2.1 2284.68 2285.63
2.25 2289.13 2282.21
2.4 2292.10 2280.42
2.55 2294.02 2282.32
2.7 2295.28 2284.3
2.85 2296.12 2285.27
3 2296.72 2287.04
3.15 2297.16 2289.25
3.3 2297.5 2290.49
3.45 2297.78 2290.98
3.6 2298.02 2292.84
3.75 2298.23 2294.43
3.9 2298.42 2294.70
4.05 2298.59 2295.14
201

E Supporting Information: Measuring the Electric Fields of
Cations Captured in Crown Ethers
E.1 Frequency shift vs radius plot
(b) (a) ∆νCN / cm-1 ∆νCN / cm-1
Li+
Na+
K
+
Zn Mn2+ 2+
Ba2+
Ca2+
Cu2+
Ni2+
Sc3+
Al3+ Eu3+
Y
3+
In3+
Eu3+
In3+
Li+
Na+
K
+
Ba Ca 2+
2+
Al3+
Sc3+
Mg2+
Yb3+
Zn2+ Cu2+
Mg2+
Ni2+
Crown
Ether
Crown
Ether
Y
3+
Yb3+
Mn2+
Figure 9.17: CN frequency shift vs ionic radius of metal ions (a) Benzo-15-crown-5 (b) Benzo18-crown-6
202

Abstract (if available)

Abstract Chemistry on the microscopic scale is strongly dominated by the electrostatic influence of charges such as ions and molecular dipoles. This makes measuring and modeling of electric fields on molecular scales an important priority in understanding chemical and electrochemical reactions. Vibrational Stark shift spectroscopy is a useful tool to measure electric fields on nanometer scale. This thesis elaborates on the use of benzonitrile molecules as a Stark probe which shows shifts in their vibrational frequencies under the influence of external electric fields. This frequency shift as a reporter of local electrostatics has given us insight about a variety of important chemical phenomena - effect of surfactants at metal-electrolyte interfaces, tracking a proton donor from bulk to the electric double layer in the presence of electrochemical bias and measuring fields produced from cations dissolved in solvent can produce local electric fields. These studies will inspire ideas about designing electric fields to control reaction rates and product selectivity. There are some nuances in using this Stark probe. The systematic change in linewidth of nitrile with external electric field is from intramolecular vibrational energy transfer in benzonitrile, hence cannot be attributed completely to report the local environment. Sometimes the specific bonding interactions create ambiguity in quantifying local electrostatics from frequency shifts. We have shown that this specific interaction effect can also be utilized as an indicator for interfacial ion adsorption. This effect may be extended to study local pH, a primary factor in protein functions and in controlling over-potential in many electrochemical reactions.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Probing charge transfer and electric fields at the bulk and interface using vibrational spectroscopy

PDF

Monitoring local electric fields, charge densities at electrode surfaces using in-situ graphene/surface enhanced Raman spectroscopy (GERS/SERS) based Stark-shifts

PDF

Water partitioning between the bulk and an electrode surface

PDF

Nonlinear spectroscopy of chemical phenomena at interfaces: carrier trapping and local electric fields

PDF

Asymmetric response and the reorientation of interfacial water with respect to applied electric fields

PDF

Vibrational sum frequency generation studies of electrode interfaces

PDF

Interfacial polarization and photobasicity: spectroscopic studies of the electrode environment and photochemical effects

PDF

Potential-induced formation and dissociation of adducts at interfaces and adsorption of phenols onto metal electrodes

PDF

Understanding electrostatic effects in the function of biological systems

PDF

Molecular orientation from sum frequency generation spectroscopy: case study of water and methyl vibrations

PDF

Raman spectroscopy and electrical transport in suspended carbon nanotube field effect transistors under applied bias and gate voltages

PDF

Computational spectroscopy in gas and condensed phases

PDF

Infrared spectroscopy of carbocations in helium droplets

PDF

The adsorption and selective deposition of molecular and nanocluster ions on carbon based devices

PDF

Two-coordinate coinage metal complexes for solar fuels and organic LED chemistry

PDF

Plasma emission across metal nanoparticle surfaces and semiconductor -liquid interface characterizations

PDF

Experiments in electrostatic deflection of doped helium nanodroplets

PDF

High-frequency and high-field magnetic resonance spectroscopy of diamond

PDF

Vibrational sum frequency spectroscopy of molecules on metal, semiconductor, and aqueous surfaces

PDF

Dissociation energy and dynamics of HCl and aromatic-water clusters

Asset Metadata

Creator Maitra, Anwesha (author)

Core Title Elucidating local electric fields and ion adsorption using benzonitrile: vibrational spectroscopic studies at interfaces and in bulk

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Chemistry

Degree Conferral Date 2023-12

Publication Date 11/16/2023

Defense Date 11/16/2023

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

Tag IR,local electric fields,nitrile,OAI-PMH Harvest,Raman,Stark shift,vibrational spectroscopy

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Cronin, Stephen (committee member), Dawlaty, Jahan M. (committee member), Prezhdo, Oleg (committee member), Takahashi, Susumu (committee member)

Creator Email amaitra@usc.edu,anweshamaitra79@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC113774574

Unique identifier UC113774574

Identifier etd-MaitraAnwe-12476.pdf (filename)

Legacy Identifier etd-MaitraAnwe-12476

Document Type Thesis

Format theses (aat)

Rights Maitra, Anwesha

Internet Media Type application/pdf

Type texts

Source 20231120-usctheses-batch-1107 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu