This is to certify that the
thesis entitled

DIAMOND OPTICALLY TRANSPARENT ELECTRODES FOR
USE IN IR TRANSMISSION SPECTROELECTROCHEMICAL
MEASUREMENTS

presented by

Yingrui Dai

has been accepted towards fulﬁllment
of the requirements for the

MS. degree in Chemistry

 

 

deameém

UMajor Professor’s Signature

Mag /0, 2M7

Date

MSU is an afﬂnnative-action, equal-opportunity employer

 

 

LIBRARY
Michigan State
University

 

 

 

..._-.-.-.-.—o-—n---‘--.-.-.-.-.—.-.-.~.-—‘-.—.-.-.--

u-—»-.-.—.-.-.-u-.-v—.-.-.-.-.-.-.a—.-t—— ..

I

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 p:/ClRC/DateDue.indd-p.1

 

DIAMOND OPTICALLY TRANSPARENT ELECTRODES
FOR USE IN IR TRANSMISSION
SPECTROELECTROCHEMICAL MEASUREMENTS

By

Yingrui Dai

A THESIS
Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of
MASTER OF SCIENCE
Department of Chemistry

2007

ABSTRACT

DIAMOND OPTICALLY TRANSPARENT ELECTRODES FOR USE IN IR
TRANSMISSION SPECTROELECTROCHEMICAL MEASUREMENTS

By
Yingrui Dai

The goal of the research was to develop boron-doped diamond optically
transparent electrodes for use in the IR transmission spectroelectrochemical
measurements. The diamond ﬁlm was deposited by microwave-assisted chemical vapor
deposition for 4 h using a 0.5% CH4/H2 with 2 ppm B2H6 source gas mixture at 45 torr
and 1000 W. The diamond/Si OTE transmits approximately 40-60% of IR light (2000-
700 cm'l). SEM, Raman spectroscopy and XRD results conﬁrmed the presence of a thin,
high quality boron-doped diamond ﬁlm. The low ﬁlm resistivity (0.1-0.4 Q-cm) and good
electrode activity produced relatively rapid electron-transfer reaction kinetics for both
aqueous (F e(CN)6'3/ '4, methyl viologen, Ru(NH3)5+3/+2 and IrC16'2/ '3) and non-aqueous
(ferrocene) redox couples. Difference FTIR spectra (oxidized minus reduced) for 10 mM
Fe(CN)6'3M in l M KCl recorded at various applied potentials showed that the CN
vibrational mode at 2039 cm'1 for Fe(CN)6'4 reversibly shifted upon oxidation to 2116
cm'1 as expected. From the difference FTIR spectra of 20 mM ferrocene in 0.1 M
TBABF4/CH3CN, the Shift of the OH bending mode perpendicular to the plane of the
cyclopentadienyl ring from 823 cm'1 to 857 cm'1 showed the conversion of the ferrocene
to the ferricenium ion. The results showed that the diamond/Si OTE possesses good
electrical conductivity and optical transparency enabling rapid, reproducible and

quantitative spectroelectrochemical measurements in the mid- and far- IR.

ACKNOWLEDGEMENTS

The past three years at Michigan State University is a valuable experience in my
life. I have Ieamed a lot: how to be a researcher, how to drive, and how to love.

The ﬁrst person I want to thank is my advisor, Dr. Greg Swain. Thanks for his
guidance and encouragement over the last three years. I have been fortunate to work with
Dr. Jerzey Zak and Dr. Denis A. Proshlyakov. I thank you sincerely for helping me
develop the skills critical to being a scientist through the discussions. Thanks to my
committee members, Dr. Gary Blanchard, and Dr. David Weliky for your helpful
comments about my work. I have never had the opportunity to work with a welcoming,
encouraging group of people in the Swain Lab. Words alone can not express my
gratefulness to Elizabeth, Yang, Shihua, Anne, Jason, Veronika, Martin, Dasa, Luther;
Matt, Isao, Ayten and Aihua. You guys are the best.

I also want to thank my best ﬁ'iends: Xin, Xiaoyu, Chang and Meng. It is nice to
have you in my life that saw me through the process. You always show up when I need
you. That is a friend indeed.

Finally, and most importantly, I would like to say thank you to my parents and
liangxing. Without your love and support, I would not be here today. Thank you for

always being there with me. Love you forever.

iii

TABLE OF CONTENTS

List of Tables .......................................................................... vi
List of Figures ........................................................................ vii
CHAPTER 1 — Introduction ........................................................ 1
1.1 Spectroelectrochemistry .............................................................. 1
1.2 Optically Transparent Electrodes (OTB) ........................................... 4
1.3 Development of Optically Transparent Boron-Doped Diamond Electrode. . . ...6
1.3.1 Boron-Doped Diamond Electrode ............................................ 6
1.3.2 Optically Transparent Boron-Doped Diamond Electrode ..................... 7
1.4 Outline of Thesis ................................................................... 10
1.5 References .......................................................................... 11
CHAPTER 2 — Experimental Methods ......................................... 14
2.1 Diamond Thin Film Deposition ................................................... 14
2.2 Material Characterization .......................................................... 16
2.2.1 Scanning Electron Microscopy (SEM) ....................................... 16
2.2.2 Raman Spectroscopy ........................................................ 16
2.2.3 X-Ray Diffraction Spectroscopy (XRD) .................................... 17
2.2.4 Four-Point Probe Measurements ............................................. 17
2.2.5 Optical Transmission Measurements ........................................ 18
2.2.6 Electrochemical Characterization ........................................... 18
2.3 Spectroelectrochemical Measurements ............................................. 20
2.3.1 Thin-Layer Spectroelectrochemical Cell ..................................... 20
2.3.2 Potential Step Method ....................................................... 22
2.3.3 Temperature Control ........................................................ 23
2.4 Chemicals ......................................................................... 24
2.5 References ........................................................................ 25

CHAPTER 3 — Optical, Electrical, Physical and Electrochemical

Properties of the Boron-Doped Diamond Electrodes ........................ 26
3.1 Introduction ....................................................................... 26
3.2 Results and Discussion ............................................................ 29
3.2.1 Optical Properties of Diamond/Si OTB-IR .................................. 29
3.2.2 Electrical Properties of Diamond/Si OTE .................................... 32
3.2.3 Physical Properties of Diamond/Si OTE .................................... 33

iv

3.2.4 Electrochemical Properties of Diamond OTE ............................... 36
3.3 Conclusions ........................................................................ 45
3.4 References ......................................................................... 47

CHAPTER 4 — IR Transmission Spectroelectrochemical Measurements
With an Optically Transparent Boron-Doped Diamond Electrode. . ....49

4. 1 Introduction ........................................................................ 49
4.2 Results and Discussion ............................................................. 52
4.3 Conclusions ........................................................................ 61
4.4 References ......................................................................... 63
CHAPTER 5 — Key Findings and Perspectives ................................ 64
5.1 Conclusions ........................................................................ 64
5.2 Future Research Directions ........................................................ 66

5.2.1 Metalloproteins .............................................................. 66

5.2.2 F ar-IR Spectroelectrochemical Measurements of Redox Metalloproteins....67

Table 3-1

Table 3-2

Table 3—3

Table 3-4

Table 3-5

LIST OF TABLES

Effect of growth parameters on the properties of diamond/Si OTE ......... 28

Comparison between measured XRD data for a diamond/Si OTE and values
from ASTM (6-0675) cubic diamond standard ............................. 36

Summary of cyclic voltammetric data for diamond/Si OTEs deposited for 4 h
using 2 ppm BZHG. The scan rate was 0.1 V/s .............................. 40

Concentration dependence of the cyclic voltammetric data at a diamond/Si
OTE ......................................................................... 44

Sean rate dependence of the cyclic voltammetric data at a diamond/Si
OTE ........................................................................ 45

vi

Figure 1-1

Figure 1-2

Figure 1-3

Figure 2-1

Figure 2-2

Figure 2-3

Figure 2-4

Figure 2-5

Figure 3-1

Figure 3-2

Figure 3-3

LIST OF FIGURES

Difference types of Spectroelectrochemical techniques ..................... 2

Two conventional OTEs for transmission measurements: (A) a metal
minigrid and (B) a thin ﬁlm supported on an optically transparent
substrate ..................................................................... 4

Schematic of bonding arrangement of sp3- bonded diamond ................ 6

A schematic of a microwave plasma—assisted CVD reactor used to grow
boron-doped diamond thin-ﬁlm electrodes ................................. 14

Design of the single-compartment, three—electrode electrochemical cell with
(a) the diamond ﬁlm working electrode, (b) O-ring seal, (c) nickel foil and
rubber spacer, ((1) reference electrode inside a glass capillary tube with a
cracked tip, (e) carbon rod counter electrode, and (f) input for nitrogen

purge gas .................................................................. 19

Diagram of the IR transmission thin-layer Spectroelectrochemical cell. . . ..21

Potential-time proﬁle for the recording of Spectroelectrochemical
measurements. An absorbance spectrum was recorded at each potential as

indicated by the arrows .................................................... 22

Principle of absorbance difference spectroscopy ........................... 23

IR transmission spectra for different types of boron—doped diamond
electrodes: (A) a CVD “white” (type IIa) diamond disc, (B) undoped and
polished Si substrate, (C) boron-doped diamond thin ﬁlm deposited on the
undoped Si using 2 ppm BzH6 and 4-h growth time, (D) boron-doped
diamond thin ﬁlm deposited on the undoped Si using 5 ppm B2H6 and 4-h
growth time, and (E) boron-doped diamond thin ﬁlm deposited on the
undoped Si using 2 ppm B2H6 and a 10-h growth
time ........................................................................ 30

IR absorption spectrum of boron-doped diamond/Si OTE deposited for 4 h
using 2 ppm B2H6 .......................................................... 31

Four-point probe electrical resistivity measurements for ﬁve different
diamond/Si OTEs deposited for 4 h using 2 ppm B2H6 ..................... 32

vii

Figure 3-4 Scanning electron micrographs of a diamond/Si OTE: (A) top view (B) side
section view. The ﬁlm was deposited for 4 h using 2 ppm B2H6 ............ 33

Figure 3-5 Raman spectrum for a diamond/Si OTE deposited for 4 h using 2 ppm
BZHG ....................................................................... 34

Figure 3-6 Powder X-ray diffraction spectrum of a diamond/Si OTE deposited for 4 h
using 2 ppm BzH6 .......................................................... 36

Figure 3-7 Cyclic voltammetric i-E curve in 1 M KCl for a diamond/Si OTE deposited
for 4 h using 2 ppm B2H6. Scan rate = 0.1 V/s ............................. 37

-3/4
Figure 3-8 Cyclic voltammetric i-E curves for 0.1 mM Fe(CN")6 , methyl viologen,

+3/ 2 we
Ru(NH3)6 + and IrCl6 , all in 1 M KCl, at a diamond/Si OTE deposited

for 4 h using 2 ppm B2H6. Scan rate = 0.1 V/s. Electrode geometric area =
0.2 cm2 ..................................................................... 39

Figure 3-9 Cyclic voltammetric i-E curve for 0.1 mM ferrocene in 0.1 M
TBABF4/CH3CN at a diamond/Si OTE deposited for 4 h using 2 ppm BZHG,

Scan rate = 0.1 V/s. Electrode geometric area = 0.2 cm2 ................... 39

Figure 3-10 Cyclic vo/ltammetric i-E curves for 0. 05, 0.1, 0. 5, and 1mM (A)
Fe(CN)63 ,(B) Ru(NH3 )6+ ,(C) 11'C162 3,and (D) methyl viologen, all in

1 M KCl, with a diamond/Si OTE deposited for 4 h with 2 ppm Bsz. Scan
rate = 0.1V/s. Electrode geometric area = 0.2 cm2 ......................... 43

Figure 4-1 Cyclic voltammetric data for the diamond/Si OTE with 7.5 pm spacer. (A)
Background-corrected thin-layer cyclic voltammetric i-E curve for 1 mM
Fe(CN)6'4 in 1 M KC1. The scan rate was 1 mV/s. (B) Plots of the oxidation
(square) and reduction (triangular) peak current vs. scan rate for 1 mM

F e(CN)6'4 in 1 M KCl. Data are shown for scan rates from 1 to 5 mV/s. . ..53

Figure 4-2 IR Spectroelectrochemical difference absorbance spectra of Fe(CN)6 3” at
the diamond/Si OTE with a Can window. The optical pathlength was 7. 5
pm. The analyte solution was 10 mM Fe(CN)(,4 in 1 M KC1, The potential
was stepped from O to 0.1 V, 0.2 V, 0.25 V, 0.3 V, 0.35 V, 0.4 V, and 0.5 V
vs. Ag/AgCI. Each spectrum was generated by subtracting the reference
spectrum obtained at 0 V ................................................... 55

Figure 4-3 Nemst plots for 10 mM Fe(CN)(, in 31 M KC1. (A) the Fe(CN)6 absorption
at 2039 cm and (B) the Fe(CN)6'3 absorption at 2116 cm .Plots were

viii

Figure 4-4

Figure 4-5

Figure 4-6

generated from spectroelectrochemical data obtained with a diamond/Si
OTE as presented in Figure 4-257

IR spectroelectrochemical absorbance spectra for Fe(CN)6‘4 (solid line) and
electrogenerated Fe(CN)6'3 (dashed line) at the diamond/Si OTE with a Can
window. The optical pathlength was 7.5 pm. The analyte solution was 10
mM Fe(CN)(,4 in 1 M KCl, Oxidized minus reduced (dashed line) and
reduced minus oxidized (solid line) spectra were generated from the
oxidized and reduced spectra collected at 0.5 and O V during a forward and

reverse potential step for 60 s .............................................. 58
Difference absorbance-potential curves for Fe(CN)6'3M at the diamond/Si
OTE with a Can window. The optical pathlength was 7.5 pm. The
measurement was made for a solution of 10 mM Fe(CN)(,’4 in 1 M KCl. The
data were collected at a scan rate of 2 mV/s. The IR spectra were collected
within a range of wavenumbers during the scanning, the ﬁnal difference
absorbance at each potential was actually an average of a small range of

potential. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

IR spectroelectrochemical difference absorbance spectra of ferrocene at the
diamond/Si OTE with a Si window. The optical pathlength was 75 pm. The
analyte solution was 20 mM ferrocene in 0.1 M TBABF4/CH3CN, The
potential was stepped from 0.1 to 0.9 V vs. Pt-QRE in increments of 0.1 V.
Each spectrum was generated by subtracting the reference spectrum for the

fully reduced form at 0.1 V... . . . . . .. . . . 61

ix

CHAPTER 1

Introduction

1.1 Spectroelectrochemistry

Spectroelectrochemistry is a hybrid of two techniques: spectroscopy and
electrochemistry. The technique (transmission mode) can be used to record the change in
the spectroscopic signature of an analyte associated with a change in redox state brought
about by electron transfer with an optically transparent electrode (OTE). It provides more
information about a redox reaction mechanism than that can be acquired using traditional
electrochemical methods alone. Spectroelectrochemistry has been used for nearly 40
years in the study of inorganic, organic, and biological redox systems. 1’ 2

Considerable development in Spectroelectrochemistry has taken place since its
inception in 1964 when Kuwana introduced this hybrid analytical method.3 The coupling
of a variety of spectroscopic and electrochemical techniques has enabled new and
powerful combinations to be applied to complex redox systems.4'6 In fact,
Spectroelectrochemistry has been used with electromagnetic radiation from almost every

region of the electromagnetic spectrum? 7’ 8 Some commonly used spectroscopic

methods that have been coupled with electrochemistry are listed in Figure 1-1.

 

  
   
 

  

 

 

 

 

ABSORPTION
LUMINESCENCE
SPECTROSCOPY AND
SCATTERING
UV: VISIBLE. IR SPECTROSCOPIES
H 0.2 mm Raman scattering

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A. Thin-layer cell C. Specular reflectance
I—I 1cm
l H \
U ' ’ [I 1‘ / --
Fluorescence
B. Conventional cell D. Internal reflectance F. Fluorescence

 

 

 

 

 

 

 

 

Figure 1-1 Different types of spectroelectrochemical techniques. (adapted with
permission from references 8)

Absorption spectroscopy in the ultraviolet, visible, or infrared regions is most
often employed. In the transmission mode (Figure 1-1 A, B), the optical beam is directed
orthogonally through an OTE and the adjacent solution layer. Specular reﬂectance
spectroscopy (Figure 1-1, C) involves passing light through a solution layer and
measurement of the light reﬂected from the electrode surface. In the internal reﬂection
mode (Figure 1-1, D), the light beam is introduced through the back side of an OTE at an
angle greater than the critical angle so that the beam is totally internally reﬂected. The
light interacts with species present adjacent to the electrode surface. The sensitivity can
be increased by causing multiple reﬂections to occur. In normal and resonance Raman

spectroscopy (Figure 1-1, B), a laser beam is directed through the solution at an electrode

and the Raman back-scattering is observed. A beam of excitation light can be passed
through an electrochemical cell and the resulting luminescence by electrogenerated
species can be recorded (Figure 1-1, F). Many other techniques, such as, surface plasmon
resonance (SPR), mass spectrometry (MS), electron spin resonance (EPR), nuclear
magnetic resonance (NMR), X—ray diffraction (XRD), have also been coupled with
electrochemistry. In summary, the transmission mode provides information about the
electrochemical reaction in the solution, the reﬂectance mode is more focused on the
electrode/electrolyte interface, and the scattering mode is useful for the investigation of
electrogenerated species adsorbed on the electrode surface.

The most frequently used spectroelectrochemical measurement mode is the
transmission one through a thin-layer cell. One of the main advantages of the thin-layer
cell design is that the electroactive analyte present can be exhaustively electrolyzed in a
very short time (typically 20-120 5 depending on the cell thickness). In the design, there
is ﬁnite diffusion, which means the mathematical description of the electrochemical
process can be simpliﬁed. This thesis will focus on transmission thin-layer
spectroelectrochemical measurements in the IR region using an optically transparent
diamond electrode. IR offers an advantage over other spectroscopies that have been
coupled to electrochemistry (e.g., UV/vis), because it provides information about the

identity and molecular structure of a reactive species.

1.2 Optically Transparent Electrodes (OTE)

Transmission spectroelectrochemical measurements require an electrode that is
transparent to the electromagnetic radiation. An ideal OTE would possess good electrical

conductivity and high optical transparency.

A. Minigrid OTE B. Film-Based OTE

 

 

 

. PIE. .. U
Substrate

 

 

 

 

Figure 1-2 Two conventional OTEs for transmission measurements: (A) a metal
mini grid and (B) a thin ﬁlm supported on an optically transparent substrate.

Electrode architecture that would work for such a measurement is a porous
electrode. One example is a minigrid electrode (e.g., Au, Ni, Ag or Cu),9'” which is
simply a metal mesh (Figure 1-2 A), and porous reticulated vitreous carbon.'2' '3 Light is
transmitted through the electrode openings allowing one to probe changes in electronic
properties or structural features of a molecule brought about by a change in redox state.

The other design is a thin ﬁhn (Figure 1-2 B) supported on an optically

transparent substrate. Depending on the spectral region of interest, conductive thin ﬁlms,

I4, 15 16-18 19-21

such as metals , carbon materials and metal oxides, are used. The most
common OTE material is indium—doped tin oxide (ITO), which is deposited as a thin ﬁlm
on a transparent substrate, such as glass (visible), quartz (UV -visible), germanium or
silicon (infrared). Optically transparency results because of the thinness of the conducting
ﬁlm. Usually, a compromise between the electrical conductivity and the transparence has
to be made as these two parameters are inversely related to one another. A transparent
substrate can be used to increase the mechanical stability of the thin OTE.

For IR transmission spectroelectrochemical measurements, a gold or platinium
minigrid is the most commonly used electrode material. They are easily fabricated and
transparent over the whole electromagnetic spectrum. Modiﬁed gold minigrid electrodes
are used more often than the bare gold minigrid. The disadvantages of this electrode type
are the narrow potential window, the large background charging current and the effort
needed to modify the surface and the selection of mediators. Due to their geometry, they
are not suited to optical probing of surface localized processes such as adsorption and
chemical modiﬁcation. Also, their fragility limits the lifetime of the cell constructed

using them. A more stable OTE material is needed in order to enable more widespread

application of this informative technique.

1.3 Development of Optically Transparent Boron-Doped Diamond

Electrode

1.3.1 Boron-Doped Diamond Electrode

The use of electrically conductive diamond as an electrode material is a ﬁeld of
research that had its beginning back in the early 1990s.”24 Diamond is the hardest
natural material, consisting of a crystal lattice of tetrahedrally-arranged sp3-bonded
carbon atoms. This arrangement leads to six-membered rings in a chair conformation, as

shown in Figure 1-3.25

 

 

Figure 1-3 Schematic of bonding arrangement of sp3- bonded diamond. (adapted with
permission from reference 25)

It can be produced synthetically by two methods: (1) high-pressure-high-temperature
(HPHT) growth (>1400 °C, >50,000 bar) in the presence of a catalyst, and (2) low
pressure, low temperature (<1000 °C, 0.01-1 bar) chemical vapor deposition (CVD) from
a carbon-containing source gas mixture.26 CVD diamond is usually polycrystalline with

randomly-oriented crystallites and grain boundaries between the crystallites. These grain

boundaries consist of a mixture of sp2- and sp3- bonded carbon. Diamond is naturally an
insulating material, so in order for it to be electrically conducting, it must be doped.
Boron dopant atoms substitute for carbon atoms in the growing crystal lattice, which
produces p-type semiconductivity with an acceptor level 0.37 eV above the valence
band.27 Resistivities as low as 0.01 Q-cm have been reported for degenerately boron-
doped ﬁlms, making them useﬁtl as electrochemical electrodes.” 29 The surface carbon
atoms are stabilized by strong covalent bonds with hydrogen for ﬁlms deposited from
hydrogen-rich source gas mixtures, as illustrated in Figure 1-3.

The application of boron-doped diamond thin-ﬁlm electrodes in electroanalysis
has received much attention in recent years.”’ 28’ 30’ 3' Much of the work so far has
involved the characterization of the physical and electrochemical properties of boron-
doped diamond and its electrochemical response toward a variety of aqueous and
nonaqueous redox systems.32'36 From the results, it can be concluded that high-quality
diamond, which means minimal nondiamond carbon impurity content, low secondary
nucleation density, and hydrogen surface termination, offers signiﬁcant advantages over
other carbon and metal electrodes in terms of a wide potential window, lower and more
stable background current, fouling resistance, lower limits of detection, and better

‘ 7‘
robustness. 27’ 3° 3 38

1.3.2 Optically Transparent Boron-Doped Diamond Electrode

Another unique property of diamond is its optical transparency. Diamond
(undoped) has one of the widest optical windows of any material extending ﬁ'om the band

gap absorption edge (~225 mm) deep into the far infrared (~30 cm'l). Factors that

inﬂuence the optical throughput of diamond include the defect density, dopant type and
doping level, ﬁlm thickness and grain size (roughness). Adding boron dopant to improve
the electrical conductivity will reduce the optical throughput in isolated regions of the
spectrum. It is therefore essential to establish a balance between the electrical and optical
properties of boron-doped diamond in order to make the material a useful OTE. By
adjusting the CVD growth conditions, it is possible to manipulate and optimize this
balance.

The versatility of boron-doped diamond, both as an electrode and an optically
transparent material, makes it an obvious choice for spectroelectrochemical
measurements.39 In general, the diamond offers several advantages over other commonly
used OTEs, including (1) a wide optical window extending from near UV into far-IR
regions of the electromagnetic spectrum, (2) a wide working potential window and low
background current, (3) stability during both cathodic and anodic polarization even at
high current density and potential, (4) reproducibility of the electrical, optical and
electrochemical properties ﬁom batch-to-batch, and (5) resistantance to fouling because
of a nonpolar, hydrogen-terminated surface. 39

Our group has been developing optically transparent diamond electrodes for use
in the spectroelectrochemical measurements for some time now?“3 Both free-standing
diamond disks and diamond thin ﬁlm deposited on quartz or undoped Si have been
prepared, characterized and utilized in transmission spectroelectrochemical
measurements. Zak et a1. characterized a free-standing diamond disk in a thin-layer

-3/-4

spectroelectrochemical cell for UV-vis transmission measurements of Fe(CN)6 and

methyl viologen.40 Haymond et a1. applied the free-standing diamond disk to the study of

ferrocene in acetonitrile.42 Their works demonstrated that the boron-doped diamond is a
practical OTE material for reproducible spectroelectrochemical measurements both in
aqueous and nonaqueous media. However, due to the large thickness (380 um), the
transparency was limited. In general, the low transparency, the long deposition time and
laborious mechanical polishing needed to smooth the surface are limitations of this OTE
architecture. Stotter et al. reported on the preparation and characterization of a thin ﬁlm
of optically transparent boron-doped diamond deposited on quartz for UV-vis

spectroelectrochemical measurements.41

Absorption spectra of chlorpromazine and
Fe(CN)(,'3/4 were recorded at different potentials, and the changes were correlated with
alterations in redox state of the molecules. Stotter et al. also compared the electrical,
optical, and electrochemical properties of two optically transparent electrodes: diamond
and indium tin oxide (ITO).39 Boron-doped diamond has also been deposited on undoped
silicon and been used in preliminary mid-IR spectroelectrochemical studies of
cytochrome c. Such measurements allow one to probe the amino acid side chain and
peptide backbone vibrations as a function of the redox state of the protein.”

Compared to diamond deposited on quartz for use in the UV-vis region, work is
still needed to optimize the substrate preparation and deposition conditions for preparing
uniform, electrically conductive and IR transparent diamond thin ﬁlm. For use in the
mid-IR down to about 700 cm", undoped Si is a suitable substrate. Martin and Morrison
produced the only other report on IR spectroelectrochemical measurements with diamond.
They reported on the internal reﬂection IR spectroelectrochemical measurement of

carbon-oxygen ﬁmctional groups formed on the diamond surface during anodic

. . . . 4
polarization 1n ac1d.4

The long-term goal of this project is to fabricate, characterize and apply
transparent diamond electrodes for spectroelectrochemical measurements in the mid— and
far- IR. We desire to use this new OTE to elucidate structural changes in redox systems
as a function of the redox state. This thesis will describe: (1) the optimum substrate
preparation and deposition conditions required for preparing uniform, electrically
conductive and IR transparent thin ﬁlms of diamond on undoped Si, (2) the
characterization of the physical, electrical, optical and electrochemical properties of the
diamond/Si OTE, (3) the testing of the performance of a new thin-layer
spectroelectrochemical cell design that can house the diamond OTE, and (4) the

functioning of the OTE in IR transmission spectroelectrochemical measurements.

1.4 Outline of Thesis

This thesis describes the development and characterization of optically
transparent bom-doped diamond deposited on undoped silicon as well as its application
in IR transmission spectroelectrochemical measurements. Chapter 2 details the
experimental parameters and procedures used in the work including the methods,
instrumentation, and materials. The fabrication as well as the optical, electrochemical and
physical characterization of the diamond/Si OTE are discussed in Chapter 3. The
evaluation of these properties is critical for application of the diamond/Si OTE in
spectroelectrochemical measurements. In Chapter 4, the performance of the diamond/Si
OTE in the IR transmission Spectroelectrochemistry is described in which a specially
designed thin layer cell was used for the ﬁrst time. Finally, Chapter 5 summarizes the key

ﬁndings from the work and provides some future perspectives.

10

1.5 References

(1)

(2)
(3)
(4)

(5)

(6)

(7)

(8)
(9)

(10)
(11)
(12)
(13)
(14)
(15)
(16)

(17)
(18)
(19)

(20)

Heineman, W. R.; Hawkridge, F. M.; Blount, H. N. In Electroanalytical
Chemistry; Bard, A. J ., Ed.; Marcel Dekker: New York, 1984; Vol. 13, pp 1-113.

Heineman, W. R. Journal of Chemical Education 1983, 60, 305-308.
Kuwana, T.; Darlington, K. R.; Leedy, D. W. Anal. Chem. 1964, 36, 2023-2025.

McKinley, A. 8.; Kenny, C. F.; Martin, M. 8.; Ramos, E. A.; Gannon, A. T.;
Johnson, T. V.; Dorman, S. C. Spectroscopy Letters 2004, 3 7, 275-287.

Orcajo, 0.; Ventosa, E.; Martlnez, A.; Colina, A.; Heras, A.; Ruiz, V.; Lopez-
Palacios, J. Journal of Electroanalytical Chemistry 2006, 596, 95-100.

Palacios, R. E.; Fan, F.-R. F.; Bard, A. J .; Barbara, P. F. J. Am. Chem. Soc. 2006,
128, 9028-9029.

Bard, A. J.; Faulkner, L. R. In Electrochemical Methods: Fundamentals and
Applications, 2 ed.; John Wiley & Sons: New York, 2001, pp 680-735.

Heineman, W. R. Anal. Chem. 1978, 50, 390A-402A.

Murray, R. W.; Heineman, W. R.; O'Dom, G. W. Anal. Chem. 1967, 39, 1666-
1668.

Petek, M.; Neal, T. E.; Murray, R. W. Anal. Chem. 1971, 43, 1069-1074.
Blaedel, W. J .; Boyer, S. L. Anal. Chem. 1973, 45, 258-263.

Norvell, V. E.; Mamantov, G. Anal. Chem. 1977, 49, 1470-1472.

Sorrels, J. W.; Dewald, H. D. Anal. Chem. 1990, 62, 1640-1643.

Yildiz, A.; Kissinger, P. T.; Reilley, C. N. Anal. Chem. 1968, 40, 1018-1024.
Benken, W. V.; Kuwana, T. Anal. Chem. 1970, 42, 1114-1116.

DeAngelis, T. P.; Hurst, R. W.; Yacynych, A. M.; Mark, H. B.; Jr.; Wllliam R, H.;
Mattson, J. S. Anal. Chem. 1977, 49, 1395-1398.

Zak, J.; Porter, M. D.; Kuwana, T. Anal. Chem. 1983, 55, 2219-2222.
Anjo, D. M.; Brown, S.; Wang, L. Anal. Chem. 1993, 65, 317-319.

Armstrong, N. R.; Lin, A. W. C.; Fujihira, M.; Kuwana, T. Anal. Chem. 1976, 48,
741-750.

Ginley, D. 5.; Bright, C. MRS Bull. 2000,25, 15-18.

11

(21)

(22)

(23)

(24)

(25)

(26)

(27)

(28)

(29)

(30)

(31)

(32)

(33)

(34)

(35)

(36)

(37)

Donley, C.; Dunphy, D.; Paine, D.; Carter, C.; Nebesny, K.; Lee, P.; Alloway, D.;
Armstrong, N. R. Langmuir 2002, 18, 450-457.

Bachmann, P. K.; Leers, D.; Lydtin, H. Diamond and Related Materials 1991, 1,
1-12.

Liu, H.; Dandy, D. S. Diamond and Related Materials 1995, 4, 1173-1188.

Ashfold, M. N. R.; May, P. W.; Rego, C. A.; Everitt, N. M. Chemical Society
Reviews 1994, 21-30.

Fischer, A. Ph. D. dissertation, Michigan State University, East Lansing, MI,
2005.

Pankove, J. I.; Qiu, C. In Synthetic Diamond: Emerging C VD Science and
Technology; Spear, K. E., Dismukes, J. P., Eds.; John Wiley & Sons: New York,
1994.

Swain, G. M.; Anderson, A. B.; Angus, J. C. MRS Bull. 1998, 23, 56-59.

Xu, J .; Swain, G. M. Anal. Chem. 1998, 70, 1502-1510.

Granger, M. C.; Witek, M.; Xu, J.; Wang, J .; Hupert, M.; Hanks, A.; Koppang, M.
D.; Butler, J. E.; Lucazeau, G.; Mermoux, M.; Strojek, J. W.; Swain, G. M. Anal.
Chem. 2000, 72, 3793-3804.

Xu, J .; Granger, M. C.; Chen, Q.; Strojek, J. W.; Lister, T. E.; Swain, G. M. Anal.
Chem. 1997, 69, 591A-597A.

Rao, T. N.; Yagi, 1.; Miwa, T.; Tryk, D. A.; Fujishima, A. Anal. Chem. 1999, 71,
2506-2511.

Pleskov, Y. V.; Sakharova, A. Y.; Krotova, M. D.; Builov, L. L.; Spitsyn, B. V. J.
Electroanal. Chem. and Interfac. Electrochem. 1987, 228, 19-27.

Mermoux, M.; Roy, F.; Marcus, B.; Abello, L.; Lucazeau, G. Diamond and
Related Materials 1992, I, 519-524.

Strojek, J. W.; Granger, M. C.; Swain, G. M.; Dallas, T.; Holtz, M. W. Anal.
Chem. 1996, 68, 2031-2037.

Holt, K. B.; Bard, A. J.; Show, Y.; Swain, G. M. J. Phys. Chem. B 2004, 108,
15117-15127.

Mora, A. E.; Steeds, J. W.; Butler, J. E. Diamond and Related Materials 2002, 11,
697-702.

Rao, T. N.; Loo, B. H.; Sarada, B. V.; Terashima, C.; Fujishima, A. Anal. Chem.
2002, 74, 1578-1583.

12

(38)

(39)
(40)

(41)

(42)

(43)

(44)

Fujishima, A.; Rao, T. N.; Try, D. A. Proc.-Electrochem. Soc. 2000, 99-32, 383-
388.

Stotter, J.; Show, Y.; Wang, S.; Swain, G. Chem. of Mater. 2005, 1 7, 4880-4888.
Zak, J. K.; Butler, J. E.; Swain, G. M. Anal. Chem. 2001, 73, 908-914.

Stotter, J.; Zak, J.; Behler, Z.; Show, Y.; Swain, G. M. Anal. Chem. 2002, 74,
5924-5930.

Haymond, S.; Zak, J. K.; Show, Y.; Butler, J. E.; Babcock, G. T.; Swain, G. M.
Anal. Chim. Acta 2003, 500, 137-144.

Stotter, J.; Haymond, S.; Zak, J. K.; Show, Y.; Cvackova, Z.; Swain, G. M.
Interface 2003, 12, 33-38.

Martin, H. B.; Morrison, P. W. Electrochem. and Solid-State Lett. 2001, 4, E17-
E20.

13

CHAPTER 2

Experimental Methods

2.1 Diamond Thin Film Deposition

The diamond thin-ﬁlm electrode was deposited by microwave-assisted chemical
vapor deposition (1.5 kW, 2.54 GHz, Astex, Inc., Lowell, MA).1 A diagram of a typical

reactor is shown in Figure 2-1.

microwave
generator

 

 

 

 

 

 

 

    
 

quartz
chamber \\A ‘

...— substrate

 

 

gas inlet —>

rotary pump

Figure 2-1 A schematic of a microwave plasma-assisted CVD reactor used to grow
boron-doped diamond thin-ﬁlm electrodes.

Ultrahigh purity (99.999%) methane, hydrogen, and diborane were used as the

source gases. They entered the chamber at controlled rates regulated by mass ﬂow

14

controllers. The microwave plasma was then ignited inside a quartz cavity with the
ﬂowing source gases. The pressure of the chamber was regulated using a throttle exhaust
valve in series with a rotary pump.

Before growth, the nominally undoped Si (100) substrate (>100 Q-cm, 500 pm in
thickness, Virginia Semiconductor, Inc., Fredericksburg, VA) was mechanically
scratched for 5 min on a felt pad using a 100 nm diameter diamond powder (Diamond
Innovations, Worthington, OH) suspended in ultrapure water. The substrate was then
sonicated for 5 min each in ultrpure water, 2-propanol, acetone, 2-propanol, and ultrapure
water to remove polishing debris from the surface, especially from the scratches or
striations. The substrate was then dried and examined under an Olympus BX60M optical
microscope (Olympus America, Inc.) for cleanliness. Embedded diamond powder
particles as well as the scratching defects serve as the initial nucleation sites for diamond
growth.

The diamond ﬁlm was then deposited using a CH4/H2/BZH6 source gas mixture
consisting of 0.5% CH4/I-Iz with 2 ppm B2H6 added for boron doping, a total gas ﬂow of
200 sccm, a power of 1000 W, a system pressure of 45 torr, a substrate temperature of
~750 °C (estimated via an optical pyrometer), and a growth time of 4 h. At the end of the
deposition period, the CH4 and B2H6 gas ﬂows were stopped and the ﬁlm remained
exposed to an H2 plasma for an additional 10 min at 1000 W and 45 Torr. The plasma
power and pressure were then slowly reduced over a 5-min period to cool the samples to
a temperature below 300 °C in the presence of atomic hydrogen. The plasma power was
then turned off, and the ﬁlm was cooled to room temperature under a ﬂow of H2. This

post-growth annealing procedure in atomic hydrogen is needed to remove adventitious

15

nondiamond sp2 carbon impurity, to minimize dangling bonds, and to ensure full

hydrogen termination.

2.2 Material Characterization

A number of analytical methods were used to characterize the diamond/Si OTE
properties, including scanning electron microscopy, Raman spectrometry, four-point
probe electrical resistance measurements, IR spectroscopy, and electrochemical

measurements.

2.2.1 Scanning Electron Microscopy (SEM)

The ﬁlm morphology was probed by SEM using a ISM-6400 microscope (JEOL,
Ltd., Tokyo, Japan, Center for Advanced Microscopy at Michigan State University) with
a 20 kV accelerating voltage and a working distance of 15 mm. Due to the low
conductivity of the undoped silicon substrate, the diamond ﬁlm edges were connected to
the aluminum stub using conductive carbon tape. This reduced surface charging effects
and improved the image quality. The tape also served to afﬁx the electrode to the

aluminum stub.

2.2.2 Raman Spectroscopy

Raman spectroscopy was used to evaluate the microstructure of the diamond/Si
OTE. A Chromex Raman 2000 spectrograph (Chromex, Inc., Albuquerque, NM) was
employed that consisted of a diode-pumped, frequency-doubled CW Nd:YAG laser (500

mW at 532 nm, Coherent), a Chromex 500is spectrometer (f/4, 600 grooves/mm

l6

holographic grating, 50 um slit width) and a thermoelectrically-cooled 1024><256 element
charge-coupled device (CCD) detector was used to illuminate the sample surface. All
spectra were recorded at room temperature with an incident power density of ca. 500
kW/cm2 (100 mW at the sample and 5 pm diameter spot size) and a 10 5 integration time.
The spectrograph was ﬁrst calibrated using the spectrum of acetaminophen (tylenol) and
double checked with a high-pressure, high-temperature (HPHT), single crystal diamond
sample (ﬁrst-order phonon position = 1332 cm'l). A white-light spectrum was collected

using the same conditions and this served as the background.

2.2.3 X-Ray Diffraction Spectroscopy (XRD)

The crystallinity of diamond thin ﬁlm was investigated by powder XRD. Spectra
were obtained by scanning 20 from 20° to 100° using a Rigaku Rotaﬂex RTP300 RC

diffractometer equipped with a rotating anode and a Cu Ka radiation source (1.540 A).

2.2.4 Four-Point Probe Measurements

The electrical resistivity of the diamond/Si OTE was measured using an in-line
arrangement, four-point tungsten-tip probe connected to an HP 3748A multimeter
(Hewlett-Packard, Palo Alto, CA). The meter was operated in the four-wire resistance
measurement mode with a probe spacing of 0.1 cm. A constant current was applied
between the outer two probes and the voltage drop between the inner two probes was
measured. The resistance was determined from I-V data and converted to a resistivity
value according to equation,2

p (Q-cm) = 4.532 1 (cm) R (Q)

17

where l is the ﬁlm thickness in centimeters. Multiple measurements were made at

different locations on each electrode.3' 4

2.2.5 Optical Transmission Measurements

The diamond/Si OTE was rinsed with clean isopropanol (IPA) and dried with
nitrogen prior to making a transmission measurement. The alcohol was distilled and
stored over activated carbon prior to use.5 The electrode was positioned in the light path
using a small piece of double-sided adhesive tape. The sample was mounted vertically for
the IR transmission measurements with the diamond surface facing the incident light
beam. All IR spectra were recorded using a ATI Mattson Research Series 1 spectrometer
with an liquid nitrogen-cooled MCT detector. Typically, spectra were acquired using an

average of 100 scans at a 4 cm'1 resolution.

2.2.6 Electrochemical Characterization

Cyclic voltammetry was the main method used to characterize the electrochemical
properties of the OTEs.6’ 7 All the electrochemical measurements were performed at room
temperature with a computer-controlled potentiostat (Model 650A, CH Instruments, Inc.,

Austin, TX) in a standard three-electrode conﬁguration, as shown in Figure 2-2.8

18

 

Figure 2-2 Design of the single-compartment, three-electrode electrochemical cell with
(a) the diamond ﬁlm working electrode, (b) O-ring seal, (0) nickel foil and rubber spacer,
((1) reference electrode inside a glass capillary tube with a cracked tip, (e) carbon rod
counter electrode, and (1) input for nitrogen purge gas. (adapted with permission from
reference 8)

The working electrode was clamped to the bottom opening of a single-
compartment, glass electrochemical cell. A Viton® O-ring (i.d. 0.5 cm), placed between
the cell opening and the electrode surface, deﬁned the electrode area exposed to the
solution (0.2 cmz). Contact was made to the working electrode by pressing a neoprene
rubber spacer and a piece of nickel foil against the entire ﬁlm surface outside the O-ring.
The O-ring was sonicated for 10 min in ultrapure water, rinsed with distilled IPA, and
dried with a steam of N2 before use. The electrode was then soaked in clean IPA for 20

min before any measurements.5’ 9 After rinsing thoroughly with distilled water, the cell

was ﬁlled with the electrolyte solution of interest. Nitrogen was bubbled through the

solution for 15 min to displace dissolved oxygen and the solution was blanketed by the
gas during the entirety of a measurement. The cell was housed in a Faraday cage for
electrical shielding.

In aqueous media, a home-made Ag/AgCl reference electrode (4 M NaCl
saturated with AgCl) was positioned in close proximity to the working electrode. A large-
area carbon rod served as the auxiliary electrode and was positioned normal to the
working electrode. A Viton® O-ring was used for the aqueous measruements. No iR
compensation was applied to the voltammetric data. In nonaqueous media (ferrocene), a
ChemRaz® (Ace Glass) O-ring was employed. A Ag wire, encased in Teﬂon®, served as
the quasi-reference electrode, and a Pt foil was used as the auxiliary electrode. A low
analyte concentration (0.1 mM) and iR compensation were used in the voltammetric
measurements of ferrocene in an effort to minimize ohmic effects in the wave shape.
Digital compensation of the uncompensated solution resistance was accomplished via a

positive-feedback method.

2.3 Spectroelectrochemical Measurements

FTIR spectra were measured using a Bruker Equinox 55 spectrometer with a
MCT detector. All spectra presented are an average of 128 scans recorded at 4 cm'1
resolution. The electrode potential was controlled with an Omni-90 analog potentiostat

(Cypress Systems, Inc.) in the constant potential mode.

2.3.1 Thin-Layer Spectroelectrochemical Cell

20

A new, custom-built transmission thin-layer cell, as shown in Figure 2-3,10 was
used for the 1R spectrolectrochemical measurements. A ﬁlm spacer of desired thickness
(7.5 pm for F e(CN)5'3M and 75 pm for ferrocene in this work) compressed between the
working OTE and the 1R window (Can, 9 mm dia. x 5 mm and Si, 9 mm dia. x 500 um
(used together with spacer» determined electrochemical cavity. The cell was equipped
with an integral, miniature Ag/AgCl reference electrode for aqueous solution and a
platinum wire quasi-reference electrode (QRE) for nonaqueous solution. A platinum wire
ring outside the sealed cavity was used to establish electric contact with the working
electrode. Another Pt wire, encircling the rear of the IR window several times, formed the

auxiliary electrode. Analyte solution was injected into the cell cavity with a syringe.

 
   
  
  
  
 
   
 
   

reference
electrode

sample

auxillary
electrode

hv

1%"

working
electrode

 

IR/UVNIS
window

'«Ewg‘x’ 7;“:"1"’.““1‘4».i-.=.. -.-. .-
= 1 IA" .

Figure 2-3 Diagram of the IR transmission thin—layer spectroelectrochemical cell.

21

2.3.2 Potential Step Method

Spectroelectrochemical measurements were made by stepping the potential from a
value where the redox system was stable (e. g. reduced form) to values where the redox
reaction (e. g., oxidation) occurred. An absorbance spectrum was collected at each
potential after a equilibration period (c. g. 1 min) at each potential, as depicted in Figure

2-4.

+

 

 

Potential (V)

 

 

 

 

 

Time (see)

Figure 2-4 Potential-time proﬁle for the recording of spectroelectrochemical
measurements. An absorbance spectrum was recorded at each potential as indicated by
the arrows.

Absorbance difference spectra were then generated by subtracting the spectrum
for the unreacted redox system from the spectrum obtained at each potential of the redox
reaction. Small absorbance changes that occur during the transition between redox states

are not always obvious. They are identiﬁable when a difference absorbance spectrum is

generated. Figure 2-5 shows an idealized difference spectrum with negative going bands

22

being characteristic of the initial state A, and positive going bands reﬂective of product

stateB.‘l

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A B
i :> i
Wavenumberlcm'1 Wavenumborlcm'1
l l
1
Difference
3 J\
<
t iv
Wavenumberlcm'1

 

 

 

Figure 2-5 Principle of absorbance difference spectroscopy.

2.3.3 Temperature Control

For the IR spectroelectrochemical measurements, the spectroelectrochemical cell
was loaded with the sample and mounted onto a support inside a sealed, thermostated
chamber. The temperature was controlled by Lakeshore model 340 temperature controller
using 200 Ohm thermistors (YSI Inc., Dayton, OH) to suppress the thermal cont1ibutions.
A constant stream of dry nitrogen gas was used to control the temperature and to purge
water vapor/CO; from chamber. The gas was ﬁrst cooled by passage through a heat

exchange in a solid C02 (dry ice)/ethanol bath followed by heating with 50 Ohm

23

resistance wire inside a glass transfer line. A thermistor suspended in the vicinity of, but
not attached to, the thin-layer spectroelectrochemical cell was used for temperature
measurement and control. Separate sensors one at the gas inlet and another imbedded in
the body of the cell were used to optimize the control parameters for temperature stability
(S 0.02 0C) and response (I z 20 min). This setup allowed for quick purging of the
sample compartment and precise temperature control. The temperature was regulated at

19 :t 0.01 °C in this work.

2.4 Chemicals

All chemicals were reagent grade quality, or better, and were used without
additional puriﬁcation. Solutions of 0.1 mM potassium ferrocyanide (Aldrich),
hexaammineruthenium (III) chloride (Aldrich), methyl viologen (Aldrich), potassium
hexachloroiridate (IV) (Aldrich) in 1 M potassium chloride (CCI) were prepared fresh
daily with ultrapure water (~18 MQ-cm) from a commercial water puriﬁcation system
(Bamstead E-pure). Ferrocene (Aldrich) was dissolved in 0.1 M tetrabutylammonium
tetraﬂuoroborate (TBABF4) (99%, Aldrich) plus acetonitrile (ACN). The ACN was
puriﬁed by distillation followed by storage over 3 A molecular sieves. The molecular
sieves were activated by heating at 300 °C for 15 h followed by cooling in a dessicator.
After storage over the activated sieves (5-10% w/v) for several days, ACN was
transferred to a clean amber bottle and allowed to stand over a second batch of molecular
sieves.12 All glassware was cleaned in a three-step manner: KOH-ethanol bath,

alconox/ultrapure water solution, and ultrapure water rinse.

24

2.5

(1)

(2)

(3)

(4)
(5)
(6)
(7)

(8)

(9)

(10)

(11)
(12)

References

Swain, G. M. In Electroanalytical Chemistry; Bard, A. J ., Rubinstein, I., Eds.;
Marcel Dekker, 2004; Vol. 22, pp 182-277.

Schroder, D. K. Semiconductor Material and Device Characterization, 2 ed.;
John Wiley & Sons: New York, 1990.

Stotter, J.; Zak, J .; Behler, Z.; Show, Y.; Swain, G. M. Anal. Chem. 2002, 74,
5924-5930.

Stotter, J.; Show, Y.; Wang, S.; Swain, G. Chem. of Mater. 2005, I 7, 4880-4888.
Ranganathan, S.; Kuo, T.-C.; McCreery, R. L. Anal. Chem. 1999, 71 , 3574-3580.
Nicholson, R. S. Anal. Chem. 1965, 37, 1351-1355.

Bard, A. J .; Faulkner, L. R. In Electrochemical Methods: Fundamentals and
Applications, 2 ed.; John Wiley & Sons: New York, 2001.

Granger, M. C.; Witek, M.; Xu, J .; Wang, J.; Hupert, M.; Hanks, A.; Koppang, M.

D.; Butler, J. E.; Lucazeau, G.; Mermoux, M.; Strojek, J. W.; Swain, G. M. Anal.
Chem. 2000, 72, 3793-3804.

Show, Y.; Witek, M. A.; Sonthalia, P.; Swain, G. M. Chem. Mater. 2003, 15, 879-
888.

Dai, Y.; Proshlyakov, D. A.; Swain, G. M. In 58th Pittsburgh Conference on
Analytical Chemistry and Applied Spectroscopy: Chicago, IL, 2007.

Zscherp, C.; Barth, A. Biochem. 2001, 40, 1875-1883.
Fry, A. J. In Laboratory Techniques in Electroanalytical Chemistry : Second

Edition, Revised and Expanded; Kissinger, P. T., Heineman, W. R., Eds; Marcel
Dekker: New York, 1996.

25

CHAPTER 3
Optical, Electrical, Physical and Electrochemical

Properties of the Boron-Doped Diamond Electrodes

3.1 Introduction

Optically transparent boron-doped diamond electrodes were prepared by
microwave plasma-assisted chemical vapor deposition (CVD), as described in Chapter 2.
An ideal OTE would possess these optical and electrochemical properties: a wide optical
window, a low and stable background current over a wide working potential window,
good activity towards redox systems without the need for pretreatment, chemical stability
and fouling resistance.

Diamond has the widest optical window of any known material, ranging from the
fundamental absorption edge at 0.22 pm to the far-IR.l We are speciﬁcally interested in
the optical transparency of diamond OTE in the infrared region of the electromagnetic
spectrum. In the IR region, the optical transparency of diamond OTE is affected by
several factors that are listed below.1

First, absorption by chemical impurities and defects, such as adventitiously
incorporated nitrogen and nondiamond sp2 bonded carbon, result in the throughput loss
due to the loss of the lattice symmetry. For example, substitutional nitrogen leads to a
peak at 1100 cm'12’3, and the CH stretch for spZ-bonded carbon has characteristic peaks

between 3000 cm" and 3200 cm'”’ 5.

26

Second, absorption by the boron acceptor band lowers the optical throughput.
Absorbance peaks at 2460, 2800 and 2930 cm'1 are characteristic of substitutionally
inserted boron and these peaks are associated with transitions from the ground state to the
ﬁrst, second and third excited states, respectively 2’ 3’ 6'8. These absorption bands broaden
considerably at high boron concentrations (>1019 B cm'3) 2.

Third, scattering losses reduce the optical transparency. Scattering occurs because
of the surface roughness of the polycrystalline ﬁlms or scattering centers in the ﬁlms. " 9

Fourth, reﬂectance losses reduce the light throughput due to the high refractive
index of diamond, which is ca. 2.4 in the IR region (500 to 5000 cm"). 1 According to the
Fresnel equation (0 = 0°), a reﬂectance loss of 17% is calculated for the diamond/air
interface, and a 3% loss is calculated for the diamond/silicon interface. The calculated
17% loss at the diamond/air interface is in good agreement with the 80% transparency of
a free-standing diamond disc. (see Figure 3-1)

Fiﬁh, ﬁlm thickness, which affects the transparency according to Beer’s law
relationship between light path and adsorption.

Through optimization of the deposition conditions, one attempts to minimize the
incorporated impurities, defects, surface roughness and ﬁlm thickness. Most importantly,
a balance must be reached between the boron-doping level required for the desired
electrical conductivity and the maintenance of sufﬁcient optical transparency}11

The adventitious incorporation of spZ-bonded carbon impurity in diamond ﬁlms
can also reduce the light throughput due to the absorption.4’ 5 This impurity (speciﬁcally

at the surface) can also affect the electrochemical properties of the electrode by reducing

the working potential window and increasing the background current 12’ '3. Bennett et al.

27

described how the ratio of CH4/H2 in the source gas mixture affects the amount of spz-
bonded nondiamond carbon impurity incorporated into ﬁlms grown by microwave
CVD.l3 A CH4/Hz ratio of 0.5% provided a relatively rapid diamond grth rate with a
low level nondiamond carbon incorporation.” In general, the higher the CH4/H2 ratio, the
greater the level of nondiamond carbon. Controlling the boron doping level is very
important for maximizing the optical transparency while maintaining sufﬁcient electrical
conductivity. A relatively smooth ﬁlm can be obtained if a high instantaneous nucleation
density exists during the diamond deposition. This can be achieved by substrate
pretreatment (mechanically polishing) before growth. The ﬁne scratching of the substrate
surface produces a high initial density of nucleation sites, which leads to the grth of a
relatively smooth and continuous ﬁlm in a short period of time. The growth time can be
adjusted to control the ﬁhn thickness. Postgrth annealing in the presence of atomic
hydrogen is also essential for removing any adventitious spz-bonded nondiamond carbon
impurity and hydrogen terminating the surface. The hydrogen-terminated surface has
been reported to resist fouling and generally exhibits the most rapid electron-transfer
kinetics for redox couples.“ ‘4 Table 3-1 lists the controllable growth parameters and
their effect on the electrical and optical properties of diamond.

Table 3-1 Effect of growth parameters on the properties of the diamond/Si OTE.

 

Growth Parameter Diamond OTE Property

 

CH4/H2 ratio1 3 spz-bonded nondiamond impurity

 

. Balance between the electrical
Boron dop1ng level

conductivity and optical transparency

 

 

 

 

Mechanical polishing of the substrate Surface roughness
Growth time Film thickness
Hydrogen termination Fouling resistance

 

 

28

 

This chapter describes the characterization of the optical, electrical, physical and
electrochemical properties of boron-doped diamond thin ﬁlm. Optical transparency in the
IR region is the reason for selecting a particular set of growth conditions, as described
herein. Four-point probe measurements were performed to determine the electrical
resistivity of the ﬁlm. The ﬁlm morphology and microstructure of the diamond thin ﬁlm
were characterized by scanning electron microscopy (SEM) and Raman spectroscopy,
respectively. X-ray diffraction (XRD) analysis was used to verify the crystallinity of the

electrode. The electrochemical properties of the diamond/Si OTE were investigated by

-3/.4 3/+2
cyclic voltammetry using both in aqueous ( Fe(CN)6 , methyl viologen, Ru(NH3)6+

we
and IrCl6 , all in 1 M KCl ) and nonaqueous ( ferrocene in TBABF4/CH3CN) redox

molecules.

3.2 Results and Discussion

3.2.1 Optical Properties of Diamond/Si OTE-IR

Figure 3-1 shows IR transmission spectra for several different types of boron-
doped diamond electrodes: (A) a CVD “white” (type IIa) diamond disc, (B) undoped and
polished Si substrate, (C) boron-doped diamond thin ﬁlm deposited on the undoped Si
using 2 ppm B2H6 and a 4-h grth time, (D) boron-doped diamond thin ﬁlm deposited
on the undoped Si using 5 ppm B2H6 and a 4-h growth time, and (E) boron-doped

diamond thin ﬁlm deposited on the undoped Si using 2 ppm B2H6 and a 10-h grth time.

29

100

 

 

 

    

°\° 80' \ A‘ /—' A A
w p
o
g 60 - C
'5
g 40 -
l-
D

20

 

 

ﬂ/r/fvf/‘TE

0 A g l
3000 2500 2000 1500 1000
1

 

 

Wavenumberlcm'

Figure 3-1 IR transmission spectra for different types of boron-doped diamond
electrodes. The ﬁlms are described in the text.

The CVD “white” diamond (Figure 3-1 A) is relatively free of structural defects

and chemical impurities and the one-phonon mode is prohibited due to the high degree of

lattice symmetry in this type Ila diamond.15 The absorption bands in the 2666-1333 cm-‘
region are attributed to the two-phonon absorption and they are intrinsic to all diamond.4’
5 The undoped Si substrate (Figure 3-1 B) possesses about 75% transmittance throughout
the entire mid-IR with the exception of some weak absorption below 1200 cm]. The
reduced light throughput is attributed to be relatively constant reﬂectance losses over the
entire wavenumber range.9 When boron is introduced into the diamond lattice to make it
conductive, the transparency decreases, as shown in Figure 3-1 C, D and E. The

introduction of boron breaks the lattice symmetry, resulting in the broad absorption
-1
centered at 1290 cm in the one-phonon vibronic region.8’ 9’ '6 For highly doped ﬁlms, a

-1
broad continuum develops above 2000 cm as a consequence of the interaction of the

30

boron centers, as evident in Figure 3-1 C and D. This continuum leads to a reduction in
light throughput to less than 30% for the 4-h ﬁlm and less than 1% in the 10-h ﬁlm.
The 4-h ﬁlm has sufﬁcient conductivity and optical throughput, rendering it

useful for spectroelectrochemical measurements. Presently, our moderately boron-doped

-1
ﬁlms transmit approximately 40-60% of the light between 2000 and 700 cm .
Irnportantly, the optical transparency of the diamond OTE is stable after extended use in

electrochemical measurements, which is a requirement for difference

spectroelectrochemical measurements.

 

15

.3
N

 

Absorp. coeff. (cm'1)

 

 

L 4“

1000 1500 ‘ 2000 A 2500 3000
Wavenumber (cm'1)

 

o A n . I

Figure 3-2 IR absorption spectrum of boron-doped diamond/Si OTE deposited for 4 h
using 2 ppm B2H6.

Figure 3-2 shows the IR absorption spectrum of boron-doped diamond/Si OTE
deposited for 4 h using 2 ppm B2H6. This ﬁgure is generated from the Beer’s Law using

diamond/Si OTE transmission spectrum and total thickness of the diamond + Si substrate

31

as the pathlength. This format is convenient for comparing materials with different

thicknesses.

3.2.2 Electrical Properties of Diamond/Si OTE

The electrical resistivity of the diamond/Si OTE was obtained by a four-point
probe measurement. Six measurements were made at different locations on each
electrode in order to obtain an average value of the resistivity. Figure 3-3 shows the
nominal electrical resistivity for ﬁve different diamond OTEs grown the same way. The
electrical resistivity of the diamond/Si OTE is comparable from ﬁlm to ﬁlm, all in the 0.1
to 0.4 Q-cm range. Importantly, these ﬁlms possess the 40-60% transparency below 2000

cm'l, as shown in Figure 3-1.

 

+

P
w
I

Resistivity (Q—cm)

    

    

 

 

 

 

 

 

#609 I #610" #611 I #613 ‘ #514
Diamond/Si OTE

Figure 3-3 Four-point probe electrical resistivity measurements for ﬁve different
diamond] Si OTEs deposited for 4 h using 2 ppm B2H6.

32

3.2.3 Physical Properties of Diamond/Si OTE

Representative scanning electron micrographs of the diamond/Si OTE grown for
4 h using 2 ppm B2H6 are shown in Figure 3-4. Figure 3-4A clearly reveals the ﬁlm is
well faceted and polycrystalline, and covers the substrate with no visible cracks or voids.
The crystallite size ranges ﬁ‘om 0.1 tol 11m. Figure 3-4B shows a side-view image of the

diamond layer. The ﬁlm is dense and homogeneous with a thiclmess of 1.5 pm.

Diamond ﬁlm

Silicon substrate

 

Figure 3-4 Scanning electron micrographs of a diamond/Si OTE: (A) top view and (B)
side view. The ﬁlm was deposited for 4 h using 2 ppm B2H6.

33

The microstructure of the diamond was probed by Raman spectroscopy. This
technique is commonly employed to characterize the microstructure of both sp2- and sp3-
bonded carbon materialsn’ ‘8 The ﬁrst-order diamond phonon line intensity, width and
position are all sensitive to the ﬁlm quality, speciﬁcally the crystalline size, boron-doping
level, intrinsic stress and defect density.”19 A typical Raman spectrum for a ﬁlm

deposited for 4 h using 2 ppm B2H6 is presented in Figure 3-5.

 

 

 

 

 

 

0.4

A 0.3 1- Diamond

=1

3 ' C-— S]

33:; 0.2 e \

5 N

3 sp2
0'1 - W K
0.0 k % ‘

 

400 ‘ 800 A 1200 1600 ' 2000
Raman Shift (cm'1)

Figure 3-5 Raman spectrum for a diamond/Si OTE deposited for 4 h using 2 ppm B2H6.

The characteristic one-phonon diamond line is present at 1332.7 cm'1 with a full
width at half maximum (FWHM) of 11 cm'l. This line is asymmetric with an upward
shift in the intensity on the high wavenumber side. This asymmetry is attributed to a
Fano-type interference, which is caused by quantum interaction between the phonon and

a continuum of electronic transitions around the same energy involving the boron

34

impurity band.20 The F WHM of the diamond phonon peak is, to a ﬁrst approximation,
inversely related to the phonon lifetime and is a measure of the defect density. As stated,
the FWHM of the one-phonon line for the diamond ﬁlm is 11 cm"; a value that is
comparable with the ca. 8 cm'1 typically observed for polycrystalline diamond ﬁlms
grown on Si measured in the same spectrograph.12 The weak scattering intensity between
1500 and 1600 cm'1 is associated with amorphous or spz-bonded nondiamond carbon
impurity. With visible excitation, the Raman cross-section for spz-bonded carbon
(graphite as the model) is 50 times larger than that for diamond.18 Therefore, the weak
intensity indicates that there is very little sp2 carbon impurity in the ﬁlm. The sharp peak
at 522 cm‘1 and the broad peak centered at 960 cm"1 are associated with the ﬁrst-order and

second-order phonon modes for the Si substrate.2

Figure 3-6 presents the XRD spectrum for a diamond ﬁlm grown on undoped Si
for 4 h with 2 ppm B2H6. Reﬂections for the diamond crystalline orientation of (111),
(220), and (311) (ASTM 6-0675) are observed and are marked along with the Si (400)
(ASTM 27 -1402) substrate peak, with the dashed lines, respectively. The broadness of Si
(400) peak is likely due to the formation of Si-C at the interface. The spectrum also
reﬂects the high crystalline quality of the diamond as there are no peaks for spz-bonded
carbon impurity (C1002 graphite peak at 26.50 degrees or amorphous carbon peak at 33.96
degrees).21 Table 3-2 compares the peak positions and relative intensities with these for a
diamond standard from the American Society for Testing and Materials (ASTM). The
relative intensity ratio of the (111) and (220) peaks can suggest the change of crystalline
orientation of the diamond due to the incorporation of the nondiamond impurity.22 The

relative intensity ratio of the two peaks for diamond/Si OTE is 100/24 and is in good

35

agreement with the expected ratio of 100/25 for cubic diamond, which means relatively

low nondiamond impurities in the ﬁlm.

Counts

500

 

400 -
300 -
200 -

100 -

Diamond(1 1 1)

..L

 

 

 

A l a l l A

Dlamond(220)
5114901

    

0
O
i
e

l
. a

Diamond(31 1)

 

‘

 

I 1 . I

 

 

 

20

30 4050

60 7

26 (degrees)

‘ so 90 A100

Figure 3-6 Powder X-ray diffraction spectrum of a diamond/Si OTE deposited for 4 h
using 2 ppm B2H6.

Table 3-2 Comparison between the measured XRD data for a diamond/Si OTE and
values from ASTM (6-0675) cubic diamond standard.

 

 

 

 

 

Measured ASTM (6-0675)
Plane 20 d-spacing Relative 20 d-spacing Relative
(degree) (A) Intensity (degree) (A) Intensity
Diamond (111) 44.402 2.0386 100 43.915 2.0600 100
Diamond (220) 75.718 1.2551 24 75.302 1.2610 25
Diamond (311) 91.858 1.0721 13 91.495 1.0754 16

 

3.2.4 Electrochemical Properties of Diamond OTE

Background cyclic voltammetric i-E curves are very informative about the boron-

doped diamond ﬁlm quality. The working potential window, magnitude of the

36

background current, the presence of nondiamond carbon impurity at the surface and the
electrical conductivity are all revealed from such measurements. Figure 3-7 shows a
background cyclic voltammetric i-E curve in 1 M KCl for a diamond/Si OTE deposited
for 4 h using 2 ppm B2H6. The curve is stable with cycling and is largely featureless over
the entire potential range, which is typical for ﬁlms devoid of appreciable quantities of
spz-bonded carbon impurity at the surface.12 The anodic current at 1.3 V is due to the
onset of chlorine evolution and the cathodic current at -1.6 V is attributed to hydrogen
evolution. The background current for diamond is a factor of 5-10 less than a comparably
sized piece of glassy carbon. This leads to improved signal-to-background ratios in

electroanalytical measurements with the former.

 

40>-

 

Current (11A)

-20 -

40-

 

 

l i I n I A l a I n l A l

-2.o -1.5 .1.0 41.5 0.0 0.6 1.0 1.5
Potential (V vs log/A901)

 

Figure 3-7 Cyclic voltammetric i-E curve in 1 M KCl for a diamond/Si OTE deposited
for 4 h using 2 ppm B2H6. Scan rate = 0.1 V/s.

Cyclic voltammetry was used to investigate the electrochemical response of the

diamond OTE for different aqueous and nonaqueous redox systems. Representative

37

+3l+2

-3/4
cyclic voltammetric i-E curves for 0.1 mM Fe(CN)6 , methyl viologen, Ru(NH3)6

em
and IrCl6 , all in l M KCl, and or 0.1 mM ferrocene in 0.1 M TBABF4/CH3CN, are

presented in Figure 3-8 and Figure 3-9. The measurements were made with a diamond/Si
OTE grown for 4 h with 2 ppm B2H6. There are two reasons for selecting these redox
systems: 12 (1) the known sensitivity or insensitivity of the heterogeneous electron-
transfer rate constant for each to the electronic properties, surface microstructure, and

surface chemistry of sp2 and sp3 carbon electrodes, and (2) the wide potential range (~
.3/4
+1.1 to -1.3V vs. Ag/AgCl), the E" values for the redox systems span. Fe(CN)6 is a

surface-sensitive redox system that undergoes electron transfer through a more inner-
sphere electron-transfer pathway. '4’ 23 The electrode reaction kinetics for this couple are

highly sensitive to the diamond surface termination but relatively insensitive to

-3

nondiamond sp2 carbon impurity.” 24 For methyl viologen, Ru(NH3)6+3/+2 and IrCl6-2/ ,
outer-sphere redox systems, the electrode kinetics are relatively insensitive to the surface
microstructure, surface oxides, nondiamond carbon impurity and other defects.M The
kinetics are mainly inﬂuenced by density of electronic states of the electrode at the

formal potential of the redox couple. 14’ 23’ 25 They are more useful for probing the

diamond ﬁlms’ electronic density of states over a large potential range.

38

10

 

 

 

 

, Methyl Vologen Ru(NH3);3“2 Fe(CN)6 “'1’“

i 5 '
E .
2 ° ”
h
a n
0

.5 .

.10 - - - - - -
-1 .2 -0.8 4).. 4 0.0 0. 4 0.8 1.2

Potential (V vs. AglAgCl)
-3/4
Figure 3-8 Cyclic voltammetric i-E curves for 0.1 mM Fe(CN)6 , methyl viologen,

3/ 2 -2/-3
11004113); + and IIC16 , all in 1 M KC1, at a diamond/Si OTE deposited for 4 h using
2 ppm B2H6. Scan rate = 0.1 V/s. Electrode geometric area = 0.2 cm2.

15

 

10-

Current (11A)

 

 

 

0.2 A 0.4 I 0.6 ‘ 0.8 1 1.0
Potential (V vs. Ag-QRE)

Figure 3-9 Cyclic voltammetric i-E curve for 0.1 mM ferrocene in 0.1 M

TBABF4/CH3CN at a diamond/Si OTE deposited for 4 h using 2 ppm B2H6. Scan rate =
0.1 V/s. Electrode geometric area: 0. 2 cm.

39

The reproducibility of the electrochemical responses was evaluated using ﬁve

different diamond/Si OTEs. A summary of the data is presented in Table 3-3. AEp for

-3/4
Fe(CN)6 is 74 :I: 6 mV (RSD = 8 %) at 0.1 V/s, suggesting the diamond/Si OTE has a

+3/+2 -2/-3
clean, hydrogen-terminated surface. AEp for methyl viologen, Ru(NH3)6 and IrCl6

are63:t3mV(RSD=5%),74i6mV(RSD=8%)and65i4mV(RSD=6%),
respectively, at 0.1 V/s. This observation suggests the diamond/Si OTE has a high density
of electronic states over a wide range of potentials, sufﬁcient to support rapid electrode-
reaction kinetics. AEp for ferrocene is 68 i 9 mV (RSD = 13 %), indicating the diamond/
Si OTE is active for a nonaqueous redox system. For all these redox systems, the forward
scan peak current, ip°", varied linearly with the scan rate”2 between 50 and 500 mV/s.
This indicates the reaction rate is limited by semi-inﬁnite linear difﬁision of the analyte to
the surface. The ip°"/ i6red and Qp""/Qpmd ratios were near 1.0 for these scan rates, as

expected for a reversible system, and the values were stable with cycling.

Table 3-3 Summary of cyclic voltammetric data for diamond/Si OTEs deposited for 4 h
using 2 ppm BzH6. The scan rate was 0.1 V/s.

 

 

 

 

 

 

AEP Ep/Z l-pox ipox/ipred
Redox system (mV) (mV) (pA)
-314
Fe(CN)6 74 i 6 279 d: 2 6.2 :t 0.3 0.98 i 0.01
Methyl 63 e 3 -681 e 1 6.0 e 0.1 1.01 e 0.03
vrologen
+31+2
Ru(NH3)6 74 i 6 -171:h1 6.8 :t 0.1 0.97 :t 0.01
we
IrCl6 65 i4 777d: 5 5.0:l:0.1 1.00:1:0.01
Ferrocene 68 :I: 9 551 i 5 11.7 i 0.3 1.03 d: 0.02

 

 

 

 

 

 

 

40

In cyclic voltammetry, the applied potential is varied linearly with time. The

faradaic peak current is given by
i, = (2.69x105)n3’2AD“2C* 0‘”
and the charging current, i6, that always ﬂows, is given by:26
ic = ACdv
Combinging these two equations yields the following ratio:

. Ill
1, Cdu

i =(2.69x105)n3’2D“2C*

P

 

A low C* and high 1) values, some distortion will occur to the wave shape as ic becomes
relatively more important. This effect will set the limit of maximum useful scan rate and
miminum useful concentration.26

Uncompensated ohmic resistances (R6) in cyclic voltammetric experiment is
manifested in large and distorted AEp values for reversible redox couples. Ohmic
resistance can originate from several sources: solution resistance, contact resistance, and
electrode resistance. The solution resistance can be minimized by using a high
concentration of supporting electrolyte and positioning the reference electrode as close to
the working electrode as possible. We attempted to minimize the contact resistance by
using graphite or Ag paste on the backside contact with the electrode. The electrode
resistance is dominated by the boron-doping level.

The practical effect of Ru is to shift the reduction peak toward more negative
potentials and the oxidation peak towards more positive potentials resulting in larger peak

1/2

potential separations (AEP). Since the current increases with C* and v , the larger the

concentration or the scan rate, the more Ep will be shifted if there is signiﬁcant

41

uncompensated resistance. 26 If ohmic resistance is signiﬁcant, then AEp will increase
with the solution concentration of the redox molecule. If slow-electron transfer kinetics
are dominating AEp, then no change is expected with increasing analyte concentration.

Figure 3-10 shows cyclic voltammetric i-E curves for 0.05, 0.1, 0.5, and 1 mM

+3/+2

-3/-4 -2/-3
Fe(CN)6 , Ru(NH3)6 , IrCl

6 , and methyl viologen, all in 1 M KCl, at a

diamond/Si OTE deposited for 4 h with 2 ppm B2H6. The curves were recorded at 0.1 V

+3/+2

-2/-3
, IrCl , and methyl

-3/4
/s. Table 3-4 summarizes the ip,at (Fe(CN)6 ) or i6,c (Ru(NH3)6 6

viologen) and AEP values for the redox systems at different concentrations. It is clear that
AEP for all the redox systems increases with increasing analyte concentration.
Comparison measurements made with a diamond/Si electrode deposited for 10 h with 10
ppm B2H6 revealed an unchanging AEp with increasing analyte concentration. 27 The fact
that AEP increases for the lower doped diamond/Si OTE indicates the uncompensated

resistance is associated with the electrode.

42

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

00- 00
30. A m.

‘5 3
‘E 0 '5 0

g 8
3 ‘ 3 .

U 0
out .30.
ﬂp .mp

00 02 0.4 0.6
WWVSAOMOCD
100
m" c '
40- 100-
5 20+ 3: 50-
E o. E '
E . t °-
o -20- o .
I m.
.40» ’
m, -100-
04 ‘ 0.6 A 0.8 i 1.0 ‘ 12 443124 4.0; on A 0.69-0.4‘02
marvvspdm) mearvvsAdAaa)

-3/4
Figure 3-10 Cyclic voltammetric i-E curves for 0.05, 0.1, 0.5, and 1mM (A) Fe(CN)6 ,

+3/ 2 we
(B) Ru(NH3)6 + , (C) IrCl(5 , and (D) methyl viologen, all in 1 M KCl, with a

diamond/Si OTE deposited for 4 h with 2 ppm B2H6. Scan rate = 0.1 V/s. Electrode
geometric area = 0.2 cm2.

43

Table 3-4 Concentration dependence of the cyclic voltammetric data at a diamond/Si

OTE.

 

 

 

 

 

Anal yte Concrerrrbt/rl'atron rip; 53; 12:11:); 1116:2111:
Fe(CN)6'3M 0.05 306 238 68 2.624
0.1 306 239 67 6.525
0.5 318 225 93 32.26
1 332 209 123 62.45
Ru(NH3)6+3’+2 0.05 -133 -203 70 3.495
0.1 -132 -204 72 6.977
0.5 -117 -222 105 33.85
1 -101 -242 141 64.38
11062” 0.05 816 754 62 3.411
0.1 812 745 67 6.755
0.5 821 733 88 24.85
MV 0.05 -651 -712 61 3.323
0.1 -648 -713 65 6.128
0.5 -636 -719 83 37.41
1 -600 -731 131 79.67

 

+3/+2 -2/-3

-374
Table 3-5 summarizes the ip,al (Fe(CN)6 ) or i6,c (Ru(NH3)6 , IrCl6 , and

methyl viologen) and AEp values for the redox systems at different scan rates in 0.1 mM

+3/+2

-3/-4 -2/-3
Fe(CN)6 ,Ru(NH3)6 ,IrCl

6 , and methyl viologen, all in 1 M KCl, using a

diamond/Si OTE deposited for 4 h with 2 ppm BZH6. The scan rate was 0.1 V/s. Similarly,
AEP for each of the redox system was unchanged at low scan rates (< 0.1 V/s) and

increased with increasing high scan rate.

44

Table 3-5 Scan rate dependence of the cyclic voltammetric data at a diamond/Si OTE.

 

 

 

 

 

Scan rate E _a E ,6 AE i ,a ori ,6
Analyte V/s 1’va mpV rn\llJ p 11A p
Fe(CN)6'3"4 0.05 304 240 64 4.61
0.1 306 239 67 6.525
0.3 312 234 78 11.21
0.5 316 230 86 14.24
Ru(NH3)6+3/+2 0.05 -134 , -204 70 4.911
0.1 -132 -204 72 6.977
0.3 -125 -213 88 11.96
0.5 -119 -219 100 15.20
Il‘Clé-ZH 0.05 81 1 746 65 4.721
0.1 812 745 67 6.755
0.3 817 741 76 11.82
0.5 819 736 83 15.23
MV 0.05 -650 -713 63 4.260
0.1 -648 -713 65 6.128
0.3 -645 -717 72 11.17
0.5 -638 -717 79 15.02

 

The concentration and scan rate dependence studies indicate that the
uncompensated resistance (R6) effects of the voltammetric response at the diamond OTE

can be neglected at low concentrations (< 0.1 mM) and low scan rates (< 0.1 V/s).

3.3 Conclusions

Optically transparent boron-doped diamond ﬁlm has been reproducibly deposited
on an undoped silicon substrate using microwave-assisted chemical vapor deposition.
Substrate pretreatment by mechanical polishing and complete cleaning introduces a high
density of surface scratches and embedded diamond particles, with both functioning as

initial nucleation sites for diamond growth. This is the key to the formation of a smooth

45

and continuous ﬁlm in a short growth time. The deposition conditions were adjusted to
maximize both the electrical conductivity and optical transparency.

The optical, physical, electrical and electrochemical properties of the diamond/Si
OTE were described. SEM, Raman spectroscopy and XRD results conﬁrm the presence
of a thin, high quality boron-doped diamond ﬁlm. The diamond/Si OTE transmits
approximately 40-60% of IR light (2000-700 cm'l). The ﬁlm resistivity (0.1-0.4 Q-cm) is
larger than the typical boron-doped polycrystalline ﬁlms deposited on Si (0.01 Q-cm)
previously reported on by our group.28 This is due to a lower carrier concentration in the
diamond/Si ﬁlm due to a lower boron doping level. Even so, the resistivity is low enough
to allow for relatively rapid electron-transfer reaction kinetics for several redox systems.
The electrochemical properties were observed to be similar to those reported for high
quality boron-doped diamond thin-ﬁlm electrodes. This includes a wide working

potential window of nearly 3 V in aqueous solution, a low and stable background current

-3/4
and good electrochemical responsiveness for aqueous (Fe(CN)6 , methyl viologen,

+3/+2 -273
Ru(NH3)6 and IrCl6 , all in 1 M KCl) and nonaqueous (ferrocene in 0.1 M

TBABF4/CH3CN) redox systems. These results further indicate that the diamond/Si OTE
consists of a clean, mostly hydrogen-terminated sp3-bonded carbon surface, with no
measurable contribution from the electrochemical signal by adventitious spz-bonded

carbon impurity or carbon-oxygen functional groups.

46

3.4

(1)
(2)

(3)

(4)

(5)

(6)

(7)

(8)
(9)

(10)
(11)

(12)

(13)

(14)

(15)

(16)

References

Zaitsev, A. M. Optical Properties of Diamond; Springer-Verlag: Berlin, 2001.

Ager 111, J. W.; Walukiewicz, W.; McCluskey, M.; Plano, M. A.; Landstrass, M. 1.
App]. Phys. Lett. 1995, 66, 616-618.

Clark, C. D.; Mitchell, E. W. J.; Parsons, B. J. In The Properties of Diamond;
Field, J. E., Ed.; Academic: New York, 1979, pp 26-33.

McNamara, K. M.; Williams, B. E.; Gleason, K. K.; Scruggs, B. E. J. App]. Phys.
1994, 76, 2466-2472.

Pankove, J. I.; Qiu, C. In Synthetic Diamond: Emerging C VD Science and
Technology; Spear, K. E., Dismukes, J. P., Eds.; John Wiley & Sons: New York,
1994, pp 401-418.

Zhang, R; Xie, E.; Yang, B.; Cai, Y.; Chen, G. Mater. Lett. 1994, 19, 115-118.

Ogasawara, A.; lnushima, T.; Shiraishi, T.; Ohya, S.; Karasawa, S.; Shiomi, H.
Diam. Rel. Mater. 1997, 6, 835-838.

Mort, J .; Machonkin, M. A.; Okumura, K. Appl. Phys. Lett. 1991, 58, 1908-1910.

Stotter, J.; Haymond, S.; Zak, J. K.; Show, Y.; Cvackova, Z.; Swain, G. M.
Interface 2003, 12, 33-38.

Zak, J. K.; Butler, J. E.; Swain, G. M. Anal. Chem. 2001, 73, 908-914.

Stotter, J.; Zak, J.; Behler, Z.; Show, Y.; Swain, G. M. Anal. Chem. 2002, 74,
5924-5930.

Granger, M. C.; Swain, G. M. J. Electroanal. Chem. 1999, 146, 4551-4558.

Bennett, J. A.; Wang, J .; Show, Y.; Swain, G. M. J. Electrochem. Soc. 2004, 151,
E306-E313.

Granger, M. C.; Witek, M.; Xu, J .; Wang, J .; Hupert, M.; Hanks, A.; Koppang, M.
D.; Butler, J. E.; Lucazeau, G.; Mermoux, M.; Strojek, J. W.; Swain, G. M. Anal.
Chem. 2000, 72, 3793-3804.

Enckevort, W. V. In Synthetic Diamond: Emerging C VD Science and Technology;
Spear, K. E., Dismukes, J. P., Eds.; John Wiley & Sons: New York, 1994, pp 322-
325.

Haymond, S. Ph. D. Dissertation, Michigan State University, East Lansing, 2002.

47

(17)

(18)
(19)

(20)

(21)

(22)
(23)

(24)

(25)

(26)

(27)

(28)

Yarbrough, W. A.; Messier, R. Science (Washington, D. C., 1883-) 1990, 247,
688-696.

Knight, D. S.; W. B. White, 385 (1989). J. Mater. Res. 1989, 4, 385-393.

Ushizawa, K.; Watanabe, K.; Ando, T.; Sakaguchi, I.; Nishitani-Gamo, M.; Sato,
Y.; Kanda, H. Diamond and Related Materials 1998, 7, 1719-1722

Gonon, P.; Gheeraert, E.; Deneuville, A.; Fontaine, F.; Abello, L.; Lucazeau, G.
Journal of Applied Physics 1995, 78, 7059-7062.

Kinoshita, K., Ed. Carbon: Electrochemical and Physicochemical Properties;
John Wiley & Sons: New York, 1988.

Bennett, J. Ph. D. dissertation, Michigan State Universtiy, East Lansing,MI, 2006.
Fischer, A. E.; Show, Y.; Swain, G. M. Anal. Chem. 2004, 76, 2553-2560.

Yagi, I.; Notsu, H.; Kondo, T.; Tryk, D. A.; Fujishima, A. J. Electroanal. Chem.
1999, 473, 173-178.

Show, Y.; Witek, M. A.; Sonthalia, P.; Swain, G. M. Chem. Mater. 2003, I5, 879-
888. .

Bard, A. J.; Faulkner, L. R. In Electrochemical Methods: Fundamentals and
Applications, 2 ed.; John Wiley & Sons: New York, 2001, pp 228-238.

Wang, 8.; Research Proposal, 2005 .

Holt, K. B.; Bard, A. J.; Show, Y.; Swain, G. M. J. Phys. Chem. B 2004, 108,
15117-15127.

48

CHAPTER 4
IR Transmission Spectroelectrochemical Measurements
With an Optically Transparent Boron-Doped Diamond

Electrode

4.1 Introduction

As discussed in Chapter 1, IR spectroelectrochemistry is a useﬁrl technique for
probing the molecular structure of reactants and products in redox reactions. I11 the work

described herein, the redox system F e(CN)6'3/ '4

was used to evaluate the performance of
the diamond OTE and the thin-layer transmission cell. This redox system undergoes a
reversible, one-electron redox reaction:

Fe(CN)63' + e' .4 Fe(CN)64' 15° = 0.361 vs. NHE’
This system is useful for mid-IR measurements because the CN vibrational frequency is
sensitive to the oxidation state of the iron (Fe(CN)6'3 (Max = 2116 cm") and Fe(CN)6'
4( 1,6,. = 2039 cm")).

The main advantage of using a thin-layer cell design for spectroelectrochemical
measurements is the short time required (typically 20-120 s) for complete electrolysis.
This requires a cell with a large electrode area (A) to solution volume (V) ratio.2 One
method for achieving a large A/V ratio is to conﬁne a very small volume (i.e., a few 1.1L)

into a thin layer (< 100 um) adjacent to an electrode surface. As long as the solution

thickness, 1, is less than the diffusion layer thickness on the time scale of measurement,

49

1 << JZDt , ﬁnite diffusion will exist and the solution can be complete electrolyzed. For

ideal thin-layer behavior, the cyclic voltammetric peak current, ip, is equal to the
following:

1,. = (9.39x105)n2vVC.‘
It can be seen that the peak current, varies linearly with the scan rate, v, (V/s) and the
analyte concentration, C (mol/cm3). The faradaic charge, Q, is independent of the scan

rate according to the following expression:3

Q = nFN0 = nFVC0 *
From the faradaic charge, one can calculate the cell volume, V. The electrochemical
performance of the thin-layer spectroelectrochemical cell was evaluated using these two
relationships.
Thin-layer spectroelectrochemical studies are often conducted at a ﬁxed potential.
For a reversible system, the ratio of the concentration of the oxidized and reduced forms,
[Ox]/[Red], in the thin-layer cell is determined by the applied potential, E, according to

the Nemst equation,

E=E"'

 

0.059 [0x]
+ —log .
n [Re d]

In this expression, 11 is the number of electrons transferred per mole, and E0 is the formal

reduction potential of the couple. The concentration of Ox and Red can be related to the
absorbance through Beer’s law:

A=ebC.

50

In which, a is the molar extinction coefﬁcient (M'lcm'1), b is the optical pathlength (cm),
and C is the analyte concentration (M). If the calculation is based upon one species, the

molar extinction coefficient, 8, and the pathlength, b, can be cancelled:

E = E"' +wlog A” .
n Ared

 

One could generate a Nernst plot form the spectra at different potentials, from which the

n and E0 ' can be extracted.

After characterizing the thin-layer transmission cell, the ferrocene/ferricenium
redox couple was selected because of its usefulness in far-IR spectroelectrochemical
measurements. This redox couple is routinely used as a model analyte in nonaqueous
electrochemistry due to the highly reversible, well-behaved kinetic responses at a variety
of electrode materials.2’ 4 Ferrocene undergoes one-electron oxidation to the stable

ferricenium cation:

- - + e' 150:0.31 st. SCE‘

Another reason for selecting this redox couple is the C-H bending modes perpendicular to
the plane of the cyclopentadienyl ring in the far-IR region (ferrocene (kmax = 823 cm")
and ferricenium ()‘rmax = 857 cm'l).5'7

Characterization of the diamond OTE in the thin-layer transmission cell with
Fe(CN)6'3‘/'4 and the ferrocene provided means for evaluating the performance of the

diamnd OTE in the mid- and far- IR. The far-IR work is a prerequisite for our future

51

studies that will involve investigating bonding interactions of the metalloproteins

between the metal sites and their ligands in the far-IR (< 1000 cm").

4.2 Results and Discussion

Spectroelectrochemical measurements in the IR region were performed in the
specially designed thin-layer cell, as described in Chapter 2. Characterization of the thin-
layer cell was initially performed. Figure 4-1 (A) presents a background-corrected cyclic
voltammetric i-E curve for 1 mM Fe(CN)64' in 1 M KCl. A low potential scan rate of 1
mV/s was employed to minimize ohmic distortion of the voltarnmogram. As expected,
the oxidation and reduction peak currents are approximately Gaussian in shape with no
evidence of any difﬁrsion or tailing. The ij,,""/i,,,"’“l ratio and onx/Qp’ed ratio are both near 1.
The if" value linearly with the scan rate (R= 0.9999) and the average of the oxidation
and reduction peak potential, Em, is 0.285 V. AEP is 0.032 V, larger than the 0 V
expected for ideal thin-layer behavior when the electrode kinetics are fast compared to
the scan rate. The larger than expected AEp is due to the ohmic resistance effects in the
cell. The peak current increases linearly with the scan rate (oxidation current R=0.9999,
reduction current R=0.9998), as shown in Figure 4-1 (B), while the faradaic charge is
independent of the scan rate. From the average charge, Q = 110 11C, the cell volume was

calculated to be 1.14 uL according to V = Q/nFCbulk. The volume calculated from the thin

layer area (9.5 mm dia.) and thickness (7.5 um) was 0.53 11L. The difference between the
electrochemical volume and theoretical volume might due to the unevenly distributed

spacer and the volume around the window.

52

 

Current (11A)

 

A L A

0.0 0.1 0.2 ‘ 0.3 A 0.4 A 0.5 0.6

l L

 

 

 

 

 

 

 

Potential (V vs. AgIAgCI)

4 ' B
A
‘5 2 -
g .
s o-
o b
x

.4 .

1 2 3 4 5
Scan Rate (mV/s)

Figure 4-1 Cyclic voltammetric data for the diamond/Si OTE with 7.5 pm spacer. (A)
Background-corrected thin-layer cyclic voltammetric i-E curve for 1 mM Fe(CN)6'4 in 1
M KCl. The scan rate was 1 mV/s. (B) Plots of the oxidation (square) and reduction
(triangular) peak current vs. scan rate for 1 mM Fe(CN)64 in 1 M KCl. Data are shown
for scan rates from 1 to 5 mV/s.

53

A series of difference absorbance spectra for 10 mM Fe(CN)6'4 in 1 M KCl is
presented in Figure 4-2. The potential was stepped ﬁ'om 0 (fully reduced) to values up to
0.50 V (fully oxidized) and a spectrum was collected 60 seconds after equilibration at
each potential. The difference spectrum at each potential was generated by subtracting
the reference spectrum obtained at 0 V for the fully reduced form of the couple. As the
potential was stepped more positive, the oxidation of ferrocyanide to ferricyanide occured.
The positive-going peak at 2116 cm'1 is due to the Fe(CN)6'3 generated in the cell. No
changes in the absorbance spectra were seen beyond 0.5 V, indicating complete
electrolysis of ferrocyanide. In good agreement with published data, the CN stretching
mode at 2039 cm'1 shifts to 2116 cm'1 upon oxidation to fenicyanide. 8'10 The molar
extinction coefﬁcients for these two transitions are obviously different. The greater the
positive charge on the central metal, the less the metal can back bond electrons into 1tm
orbitals of CN ligands. The peak shift to the higher frequency for Fe+3 state means that
there is less back bonding of metal electrons, a higher bond order and a larger CN bond

energy.

54

 

 

 

   

 

 

 

001 _ 2116 CHI-1
,6 . 1 i,
0 0.00
1
>2 , 0.1V
9 -0.01 -
0
0 .
5 -0 02
e I
3 .
'2 -0.03 i. 0.5V
< .
-0.04 - 2039 cm"
2200 2100 2000 1900

Wavenumber (cm'1)

Figure 4-2 IR spectroelectrochemical difference absorbance spectra of Fe(CN)6'3/‘4 at

the diamond/Si OTE with a Can window. The optical pathlength was 7.5 pm. The
analyte solution was 10 mM Fe(CN)6“1 in 1 M KCl, The potential was stepped from 0 to
0.1 V, 0.2 V, 0.25 V, 0.3 V, 0.35 V, 0.4 V, and 0.5 V vs. Ag/AgCl. Each spectrum was
generated by subtracting the reference spectrum obtained at 0 V.

Combining the Nemst equation and Beer’s law, a plot can be generated from the
difference spectral data, that enables one to determine values of n and E" '. The
calculations can be based upon one species, then the molar extinction coefﬁcient, 8, and

the pathlength, b, are constant and can be cancelled. The modiﬁed Nemst equation is as

follow:

0.059 A0,
log .
n A

E=E°'+

 

 

red

in which, A0, = AA ,Am, = A —AA

Where AA is the difference absorbance at a given potential, Amax is the maximum

difference absorbance observed when Eapp, (>> E0.) = 0.5 V.

55

Figure 4-3 presents Nemst plots for F e(CN)6'3/ 7’ at two different wavenumbers. A
linear relationship between the potential E and log (Aon/AAmd) is seen for both
indicating that equilibrium was reached and complete electrolysis occured during the
measurement. The E” ' value of 0.294 V, determined from the y-axis intercept, is in good
agreement with the value of 0.285 V obtained from cyclic voltammetry. The slope of the
plot yields the number of electrons being transferred per mole in the redox reaction, in

this case 1.

56

 

 

 

 

 

 

 

 

A ,1 i
1.; 0.4 . 2039 cm //-/
S /"
a 1'"
as //
> 0.3 l- /I,/
>
r: /
.g /./ R = 0.9964
g // slope = 0.0609 Vldec
ac: 0.2 _ A/ Intercept = 0.294 V
/
11.5 -1.0 05 0.0 0.5 1.0 1.5 2.0
109(on’Ared>
A _ .1 / ’/’
5 0-4 2116 cm / t
a) ,I/
s ///
g / [,y
'9 0.3 - ,y/
e /
g ,2” R = 0.9984
g /,./ slope = 0.0594 Video
8 0.2 _ / 1”] Intercept = 0.297 V
/
.2.0 .1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0
109(on’Ared)

Figure 4-3 Nernst plots for 10 mM F e(CN)6'4 in 1 M KCl. (A) the Fe(CN)6'4 absorption
at 2039 cm”1 and (B) the Fe(CN)6'3 absorption at 2116 cm'l. Plots were generated from
spectroelectrochemical data obtained with a diamond/Si OTE as presented in Figure 4-2.

57

Figure 4-4 shows the forward (oxidized minus reduced, dashed line) and reverse
(reduced minus oxidized, solid line) IR difference absorbance spectra measured for 10
mM Fe(CN)64 in 1 M KCl. The spectra were collected at 0.5 V and 0 V after 1 min
equilibration time, respectively. The fact that these spectra are mirror images of one
another indicates the reaction is reversible and complete with the thin-layer cell with little

signal loss during the measurement in the custom-built cell-

 

 

 

 

 

 

°-°4 ’ Red-Ox
0.02 -
8
‘§ 0.00 f
s F
< 1‘ 1
<1 -0.02 - 1 ,'
. ‘. 1
‘ I
.0.“ '- ‘v' OX'ROd
2200 2100 2000 1900
Wavenumber (cm'1)

Figure 4-4 IR spectroelectrochemical absorbance spectra for Fe(CN)6'4 (solid line) and
electrogenerated F e(CN)6'3 (dashed line) at the diamond/Si OTE with a Can window.
The optical pathlength was 7.5 pm. The analyte solution was 10 mM Fe(CN)6‘4 in 1 M
KCl, Oxidized minus reduced (dashed line) and reduced minus oxidized (solid line)
spectra were generated from the oxidized and reduced spectra collected at 0.5 and 0 V
during a forward and reverse potential step for 60 s.

58

The absorbance-potential (voltabsorptometry) curve is another method for
demonstrating complete electrolysis within the thin-layer cell.2’ 1" 12 Figure 4-5 shows
absorbance-potential curve recorded during one complete cyclic voltammetric cycle. The
scan was initiated from 0 V and reversed at 0.6 V at a scan rate of 2 mV/s. Each point
was determined from the absorbance intensity at the two wavenumbers recorded during
32 spectral scans from 2010 cm'1 to 2150 cm'1 at 4 cm'1 resolution. Since the IR spectra
were collected over a range of wavenumbers during the scanning, the ﬁnal difference
absorbance at each potential was actually an average over a small range of potentials.
There is no change in absorbance until 0.2 V at which point there are increases or
decreases with measuring positive potential due to the oxidation of ferrocyanide to
ferricyanide. Finally, a constant level is reached at potentials positive of 0.45 V. The
reverse scan produces a near-mirror image of the forward scan with little hysterisis. The
mirror images indicate that exhaustive electrolysis occurs in the cell and that the redox
reaction is reversible. The full magnitude of the absorbance change at both wavenumbers
is comparable with the data recorded during the potential step measurements shown in

Figure 4-4.

59

 

. 2116 cm" /
0.00 /

A Absorbance

-o.02 \ \

.o,o4 . 2039 cm"

 

 

L L 4; 1 L

0.0 0.1 0.2 ‘ 0.3 i 0.4 ‘ 0.5 0.6
Potential (V vs. AglAgCl)

 

Figure 4-5 Difference absorbance-potential curves for Fe(CN)6'3/'4 at the diamond/Si
OTE with a Can window. The optical pathlength was 7.5 um. The measurement was

made for a solution of 10 mM Fe(CN)6'4 in 1 M KC1. The data were collected at a scan
rate of 2 mV/s. The IR spectra were collected within a range of wavenumbers during the
scanning, the ﬁnal difference absorbance at each potential was actually an average of a
small range of potential.

The potential-dependent thin layer far-IR difference spectra for 20 mM ferrocene
in 0.1 M TBABFi/CH3CN are presented in Figure 4-6. Similarly, the potential was
stepped from 0.1 to 0.9 V in increments of 0.1 V and a spectrum was collected 60
seconds after equilibration at each potential. Each difference spectrum was generated by
subtracting the reference spectrum obtained for the fully reduced form at 0.1 V from the
spectrum obtained for the oxidation product at each potential. As the potential was
stepped positive, a peak due to the OH bending mode perpendicular to the plane of
cyclopentadienyl ring of ferrocene at 823 cm'1 became more negative and the same mode

for the ferricenium ion appeared at 857 cm". The quantitative conversion of the redox

couple is in good agreement with the results that have been reported.7’ ‘3 This experiment

60

is an important step toward our ultimate goal of applying the spectroelectrochemical

method to measure the low-frequency (< 1000 cm'l) metal-ligand vibrations in

 

 

 

 

 

metalloproteins.
0.06
3 0.04
C!
“F
g 0.02
O
8
a 0.00
'E
o
B -0.02
<
<1
-0-04 823 cm"
900 L 850 A 800
Wavenumber (cm'1)

Figure 4-6 IR spectroelectrochemical difference absorbance spectra of ferrocene at the
diamond/Si OTE with a Si window. The optical pathlength was 75 pm. The analyte
solution was 20 mM ferrocene in 0.1 M T BABF4/CH3CN, The potential was stepped
from 0.1 to 0.9 V vs. Pt-QRE in increments of 0.1 V. Each spectrum was generated by
subtracting the reference spectrum for the fully reduced form at 0.1 V.

4.3 Conclusions

The spectroelectrochemical performance of the diamond/Si OTE was evaluated in
a new, custom-built transmission thin-layer cell. The OTE possessed good electrical
conductivity and transparency for transmission spectroelectrochemical measurements.
Well-behaved, thin-layer voltammetry was observed for the Fe(CN)6'3/“1 redox couple
with an in that increased linearly with scan rate and a Q that was independent of scan rate.

The CN stretching vibration for ferrocyanide at 2039 cm'1 shifted to 2116 cm'1 upon

61

oxidation to ferricyanide in good agreement with the literature. A linear Nemst plot was
observed with a slope of ca. 59 mV, consistent with 1 electron being transferred per mole
of analyte. The observation of the perpendicular C-H bending mode shiﬁ from 823 cm'1
to 857 cm'1 showed the quantitative conversion of the ferrocene to the ferricenium ion.
These results indicate that the thin layer cell design combined with the properties of the
diamond/Si OTE enable rapid reproducible and quantitative spectroelectrochemical

responses in the mid— and far- IR.

62

4.4

(1)

(2)

(3)

(4)
(5)
(6)

(7)
(8)
(9)
(10)
(11)
(12)
(13)

References

Bard, A. J .; Faulkner, L. R. In Electrochemical Methods: Fundamentals and
Applications, 2 ed.; John Wiley & Sons: New York, 2001.

Haymond, S.; Zak, J. K.; Show, Y.; Butler, J. E.; Babcock, G. T.; Swain, G. M.
Anal. Chim. Acta 2003, 500, 137-144.

Bard, A. J .; Faulkner, L. R. In Electrochemical Methods: Fundamentals and
Applications, 2 ed.; John Wiley & Sons: New York, 2001, pp 228-238.

Haymond, S.; Babcock, G. T.; Swain, G. M. Electroanalysis 2003, 15, 249-253.
Lippincott, E. R.; Nelson, R. D. Spectrochimica Acta 1958, 10, 307-329.

Rosenblum, M. Chemistry of the Iron Group Metallocenes: ferrocene,
ruthenocene, osmocene; Interscience: New York, 1965.

Lin, X.; Liu, Y.; Wang, E. Acta Chimica Sinica 1990, 48, 1080-1084.
Noguchi, T.; Inoue, Y. J. Biochem. 1995, 118, 9-12.

Pharr, C. M.; Grifﬁths, P. R. Anal. Chem. 1997, 69, 4673-4679.
Pharr, C. M.; Grifﬁths, P. R. Anal. Chem. 1997, 69, 4665-4672.

Niu, J .; Dong, S. Electrochimica Acta 1995, 40, 823-828.

Zak, J .; Porter, M. D.; Kuwana, T. Anal. Chem. 1983, 55, 2219-2222.

Ma, Y.; Zheng, J.; Zhu, S. Journal of Inorganic Chemistry 1999, 15, 61-67.

63

CHAPTER 5

Key Findings and Perspectives

5.1 Conclusions

The goal of the present work was to develop and characterize the new optically
transparent boron-doped diamond thin-ﬁlm deposited on undoped silicon. Such a
material would be useful to elucidate structural changes in molecules associated with a
change in the redox state using IR transmission spectroelectrochemical methods.

Chapter 3 demonstrated that optically transparent boron-doped diamond ﬁhn can
been reproducibly deposited on an undoped silicon substrate using microwave-assisted
chemical vapor deposition. Detailed optical, physical, electrical and electrochemical
properties of the diamond/Si OTE were described. The deposition conditions were
adjusted to maximize both electrical conductivity and optical transparency. SEM, Raman
spectroscopy and XRD results conﬁrmed the presence of a thin, high quality boron-doped
diamond ﬁlm. The diamond/Si OTE transmits approximately 40-60% of IR light (2000-
700 cm'l). The low ﬁlm resistivity (0.1-0.4 Q—cm) and good electrode activity produced
relatively rapid electron-transfer reaction kinetics for several test redox systems. The
electrochemical properties were found to be similar to those reported for high quality
boron-doped diamond thin-ﬁlm electrodes. These properties included a wide working

potential window of nearly 3 V in aqueous solution, a low and stable background current

-3/4
and good electrochemical responsiveness for aqueous (Fe(CN)6 , methyl viologen,

64

3/+2 -2/-3
Ru(NH3)6+ and IrCl6 , all in 1 M KCl) and nonaqueous (ferrocene in 0.1 M

TBABF4/CH3CN) redox systems. These results supported the notion that the diamond/Si
OTE consists of a clean, mostly hydrogen-terminated sp3-bonded carbon surface, with no
measurable contribution to the electrochemical signal from adventitious spz-bonded
carbon impurity or carbon-oxygen functional groups.

In Chapter 4, the mid- and far- IR spectroelectrochemical performance of the
diamond/Si OTE in a new, custom-built transmission thin-layer cell was reported on. The
initial testing with Fe(CN)6'3M solutions showed well-behaved thin-layer voltammetry,
with an ip that increased linearly with scan rate and a Q that was independent of scan rate.
The CN stretching vibration mode for ferrocyanide at 2039 crn'l shifted to 2116 crn'l
upon oxidation to ferricyanide in good agreement with the literature. A linear Nemst plot
was observed with a slope of ca. 59 mV, consistent with 1 electron being transferred per
moleof analyte. The observation of the C-H bending mode perpendicular to the plane of

1 to 857 cm'l showed the

cyclopentadienyl ring of ferrocene shift from 823 cm”
quantitative conversion of the ferrocene to the ferricenium ion. The results showed that
the diamond/Si OTE possesses good electrical conductivity and optical transparency

enabling rapid, reproducible and quantitative spectroelectrochemical measurements in the

mid- and far- IR.

65

5.2 Future Research Directions

5.2.1 Metalloproteins

One-third of all proteins are "metalloproteins", which are chemical combinations
of protein atoms (carbon, nitrogen, oxygen, hydrogen, sulfur) with metal ions such as
iron, calcium, copper, and zinc. Hemoglobin, for example, that carries oxygen in the
bloodstream, is an iron-containing metalloprotein. The metal ions in metalloproteins are
critical to the protein's function, structure, and stability. In fact, numerous essential
biological functions require metal ions, and most of the metal ion functions involve
metalloproteins. Thus, metalloproteins make life on Earth possible and the ability to
understand and ultimately control the binding and activity of protein metal sites is of
great biological and medical importance.

Understanding the structure and ﬁrnction of metalloproteins requires knowledge
of the structure and bonding at the metal center. These bonding interactions (e.g., M-OH,
M=O, M-HZO, etc) can be probed with mid- and far- IR electromagnetic radiation.
Therefore, spectroelectrochemical measurements with the diamond/Si OTE have the
potential to provide insight into the molecular structure of reaction intermediates and
products in redox protein and enzyme processes. Future work will target several
important metalloproteins: myoglobin, hemoglobin, cytochrome c and cytochrome c

oxidase.

66

5.2.2 Far-IR Spectroelectrochemical Measurements of Redox
Metalloproteins

There are a number of obstacles to making mid- and far-IR
spectroelectrochemical measurements: strong water absorbance, limitations in the optical
windows, detector sensitivity and temperature control problems. The broad water lattice
vibrations below 1000 cm'1 are very sensitive to minute changes even for small
pathlength (6 pm) or sampling conditions. The low signal-to-noise ratio because of the
decreased light intensity in the far-IR region and the smaller vibrational mode extinction
coefﬁcients makes it hard to select the window material and detector with the right
sensitivity. The baseline distortions due to the imprecise temperature control can result in

the thermal bands. Each of these issues will be systematically addressed in future work.

67

llillilljiliiiliji’ell(iiiI