Electrochemical/Electrospray-Mass Spectrometric Studies of the
Oxidation of Iodide and Cyanide at Gold and Platinum
Electrodes as well as Gas Phase Multiply-Charged
Fullerene C60 Anions.
Except where reference is made to the work of others, the work described in this
dissertation is my own or was done in collaboration with my advisory committee.
This dissertation does not include proprietary or classified information.
Tan Guo
Certificate of Approval:
William E. Hill
Professor
Department of Chemistry
Andreas Illies, Chair
Professor
Department of Chemistry
Vince Cammarata
Associate Professor
Department of Chemistry
Rik Blumenthal
Associate Professor
Department of Chemistry
Stephen L. McFarland
Acting Dean
Graduate School
Electrochemical/Electrospray-Mass Spectrometric Studies of the
Oxidation of Iodide and Cyanide at Gold and Platinum
Electrodes as well as Gas Phase Multiply-Charged
Fullerene C60 Anions.
Tan Guo
A Dissertation
Submitted to
the Graduate Faculty of
Auburn University
in Partial Fulfillment of the
Requirements for the
Degree of
Doctor of Philosophy
Auburn, Alabama
August 15, 2005
Vita
Tan Guo, son of Yuan, Suxia and Guo, Shenghui, was born on March 1, 1973,
in Xian, China P. R.. He graduated from Daqing High School in 1991, and then
he entered Harbin Institute of Technology in September 1991, and graduated with a
Bachelor of Engineer degree in chemistry in July 1995. After a two-year?s industry
employment working as an assistant engineer in Daqing Institute of Research and
Design, he entered Changchun Institute of Applied Chemistry, Chinese Academy of
Sciences, and graduated with a Master Degree of Science in Physical Chemistry in
July 2000. He entered Graduate School, Auburn University, in August 2000. He
married Xia Wang, daughter of Liu, Xuxiu and Wang, Fagui, on December 16, 2001.
iii
Dissertation Abstract
Electrochemical/Electrospray-Mass Spectrometric Studies of the
Oxidation of Iodide and Cyanide at Gold and Platinum
Electrodes as well as Gas Phase Multiply-Charged
Fullerene C60 Anions.
Tan Guo
Doctor of Philosophy, August 15, 2005
(M.S., Changchun Institute of Applied Chemistry, Chinese Academy of Sciences
2000)
(B.E., Harbin Institute of Technology, 1991)
136 Typed Pages
Directed by Dr. Andreas Illies
In this dissertation, we introduce a new spherical thin-layer flow cell to interface
electrochemistry with electrospray mass spectrometry. The successful applications of
the electrochemistry with electrospray mass spectrometry are reported.
A spherical thin-layer flow electrochemical cell, which is characterized by easy
access to the working electrode for cleaning and a high conversion efficiency, is de-
veloped. The cell has a relatively small dead volume from 5 ? 10?L. Calculated
Reynolds (Re) numbers predict that the flow pattern around the working electrode
is laminar. Both static cyclic voltammograms and hydrodynamic voltammograms at
different scan rates indicate that the cell design meets the requirements of a thin layer
flow cell. The multi-step oxidation of iodide and the oxidation of cyanide as well as
the stepwise reduction of C60 were studied.
iv
It was found that B(C6H5)?4 is a suitable internal standard for negative-ion stud-
ies in acetonitrile. The multi-step oxidation results of iodide demonstrate that the
apparatus is capable of these very challenging electrochemistry/electrospray mass
spectrometry experiments. With iodide at a platinum electrode, we observe well-
behaved oxidation to tri-iodide (I?3 ). Experiments on iodide at gold electrodes are
more complex, showing AuI?2 as well as I?3 . The AuI?2 mass spectrometric ion inten-
sity varies in a complex way throughout the applied electrochemical voltage range
studied; we propose that this variation involves the adsorption of I? on the gold
electrode surface.
With cyanide at a platinum electrode, we observe C5N?5 , which is due to the
oxidation of cyanide to cyanogen followed by the stepwise condensation of cyanogen
with cyanide. At the gold working electrode, the surface oxidation of the gold to gold
cyanide complex is observed. This is due to the high stability constant of the gold
cyanide complex in acetonitrile.
Gas phase observation of C?60, C3?60 and C4?60 anions generated at platinum elec-
trodes and detected by electrochemical/electrospray mass spectrometry is reported.
The anions are electrochemically generated from solutions of C60 dissolved in mix-
tures of toluene and acetonitrile. The gas phase observation of C3?60 and C4?60 , despite
the fact that they have negative electron affinities, is a result of a repulsive Coulombic
barrier to electron loss. These studies, which demonstrate the gas phase stability of
C3?60 and C4?60 , illustrate the promise of electrochemical/electrospray mass spectrome-
try for the studies of metastable anions.
v
Acknowledgments
The author would like to thank Dr.Illies and Dr. Cammarata for their help as
well as the committee members for their time and suggestion.
The author would also like to express his deepest appreciation to his wife and his
parents for their support and encouragement during the course of this dissertation.
vi
Style manual or journal used Journal of the American Chemical Society (together
with the style known as ?auphd?). Bibliograpy follows van Leunen?s A Handbook for
Scholars.
Computer software used The document preparation package TEXLatex
(specifically LATEX) together with the departmental style-file auphd.sty. Matlab
6.5, Adobe Acrobat 6.0, Microsoft Excel 2000, GSview 6.0, MassLynx DataBridge,
Origin 7.0.
vii
Table of Contents
List of Figures ix
List of Tables xi
1 Introduction and Literature Review 1
1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.1 Electrochemistry/Mass Spectrometry . . . . . . . . . . . . . . 5
1.2.2 Electrospray and Mass Spectrometry . . . . . . . . . . . . . . 8
1.2.3 Electrochemistry/Mass Spectrometry Flow Cells . . . . . . . . 15
1.2.4 Oxidation of Iodide and Cyanide at Platinum and Gold Electrodes 22
1.2.5 Fullerene C60 . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2 Cell Design for Electrochemistry Electrospray Mass Spectrom-
etry 33
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2.2 A New Electrochemical Flow Cell, Construction and Operation Details 35
2.3 Flow Pattern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
2.4 EC/ES-MS Experimental Conditions . . . . . . . . . . . . . . . . . . 42
2.5 Off-line Electrochemical Characterization of the Cell . . . . . . . . . 44
2.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
3 I? and CN? at Gold and Platinum Electrodes 56
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
3.2 Experimental Section . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
3.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . 61
3.3.1 Iodide at Platinum and Gold Electrodes . . . . . . . . . . . . 64
3.3.2 Cyanide at Platinum and Gold Electrodes . . . . . . . . . . . 80
3.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
4 Fullerene 90
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
4.2 Experimental Section. . . . . . . . . . . . . . . . . . . . . . . . . . . 92
4.3 Results and Discussion. . . . . . . . . . . . . . . . . . . . . . . . . . . 96
4.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Bibliography 113
viii
List of Figures
1.1 Schematic of the Electrospray Process Producing Positive Ions.The
Polarity of the Applied Potential Determines the Polarity of the Ions
Produced. [2] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.2 Dole, et al. Charge Residue Model. [65] . . . . . . . . . . . . . . . . . 12
1.3 Ion Evaporation Transition State and Initial State Models. [59] . . . . 14
1.4 Schematic Diagram of the DEMS Cell. [45] . . . . . . . . . . . . . . . 17
1.5 Thin-layer (a) and Wall-Jet (b) Flow cells . . . . . . . . . . . . . . . 19
1.6 Three Low-Lying Degenerate Lowest Unoccupied Molecular Orbitals
of C60. [36] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.1 Schematic of the Electrochemical Flow Cell . . . . . . . . . . . . . . . 36
2.2 Flow Pattern Around the Working Electrode . . . . . . . . . . . . . . 41
2.3 CyclicVoltammogramof [Fe(CN)6]3? in H2O at the Pt Electrode, V vs.
Ag/AgCl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
2.4 Peak Current (ip) vs. Square Root of Scan Rates (?12) . . . . . . . . . 48
2.5 Static Cyclic Voltammograms of [Fe(CN)6]3? at Various Scan Rates,
V vs. Ag/AgCl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
2.6 Hydrodynamic Cyclic Voltammograms of [Fe(CN)6]3? at Various Scan
Rates, V vs. Ag/AgCl . . . . . . . . . . . . . . . . . . . . . . . . . . 51
2.7 Nernstian Plot for the Scan Rate of 1 mV/s . . . . . . . . . . . . . . 53
3.1 Three Dimensional Plot of EC/ES-MS Mass Spectra B(C6H5)?4 as a
Function of Applied Electrochemical Cell Potential . . . . . . . . . . 63
3.2 Cyclic Voltammetry of I? at the Platinum Electrode, Scan Rate: 10
mV/s, V vs. Ag/AgCl . . . . . . . . . . . . . . . . . . . . . . . . . . 65
3.3 Hydrodynamic Cyclic Voltammetry of I? at the Platinum Electrode,
Flow Rate: 1 mL/hr, Scan Rate: 1 mV/s, V vs. Ag/AgCl . . . . . . 67
ix
3.4 ThreeDimensionalPlotofEC/ES-MSMassSpectraI? atthePlatinum
Electrode as a Function of Applied Electrochemical Cell Potential . . 68
3.5 Absolute Ion Intensity vs. Potential, I? at the Platinum Electrode . . 69
3.6 Hydrodynamic Cyclic Voltammetry and Current-Potential of I? at the
Platinum Electrode, V vs. Ag/AgCl . . . . . . . . . . . . . . . . . . . 71
3.7 Cyclic Voltammetry of I? at the Gold Electrode, V vs. Ag/AgCl . . . 73
3.8 Three Dimensional Plot of EC/ES-MS Mass Spectra I? at the Gold
Electrode as a Function of Applied Electrochemical Cell Potential . . 74
3.9 Absolute Ion Intensities vs. Potential, I? at the Gold Electrode . . . 75
3.10 Current-Potential of I? at the Gold and Platinum Electrodes . . . . . 78
3.11 Three Dimensional Plot of EC/ES-MS Mass Spectra of CN? at the
Platinum Electrode as a Function of Applied Electrochemical Cell Po-
tentials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
3.12 Schematic of the Formation of C5N?5 by Addition of C2N2 to CN?. [28] 82
3.13 Schematic of the Formation of C6N?6 and C12N2?12 During the Electro-
spray Process. [177,178] . . . . . . . . . . . . . . . . . . . . . . . . . 84
3.14 Mass Spectrum of the [TBA]CN oxidation at 2.0 V . . . . . . . . . . 85
3.15 Three Dimensional Plot of EC/ES-MS Mass Spectra CN? at the Gold
Electrode as a Function of Applied Electrochemical Cell Potential . . 87
4.1 Schematic Diagram of the Electrochemical Cell for the Controlled-
Potential Electrolysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
4.2 Cyclic Voltammogram of C60 in Toluene/Acetonitrile at the Scan Rate
100 mV/s, 20?C, V vs. Ag/Ag+ . . . . . . . . . . . . . . . . . . . . . 99
4.3 Near-IR Spectrum of Cn?60 (n=1,2) in Toluene/Acetonitrile with
[Bu4N]BF4 at 20?C, Curve (1): IR Absorbance of C?60, Curve (2): the
IR Absorbance of C2?60 . . . . . . . . . . . . . . . . . . . . . . . . . . 100
4.4 Three Dimensional Plot of EC/ES-MS Mass Spectra C60 as a Function
of Applied Electrochemical Cell Potential . . . . . . . . . . . . . . . . 103
4.5 C60 Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
4.6 Schematic Diagram of the Lowest Unoccupied Molecular Orbitals of
Cn?60 n=0,1,2,3 and the Shifting of the CRB Barrier Heights and the
Electron Affinities Due to the Stepwise Addition of Three Negative
Charges. The r Represents the Electron-Anion Distance . . . . . . . 110
x
List of Tables
4.1 Solubility of C60 in Various Solvents. . . . . . . . . . . . . . . . . . . 97
4.2 Surface Tension of C60 Solutions in Toluene . . . . . . . . . . . . . . 105
4.3 Surface Excess of C60 Solution in Toluene . . . . . . . . . . . . . . . . 106
xi
Chapter 1
Introduction and Literature Review
1.1 Introduction
The online combination of electrochemistry and mass spectrometry (EC/MS)
can be traced back as far as the 1970s. [1] A literature search reveals that research
applications of EC/MS have increased greatly in recent years, although this area has
never been a primary technique for either electrochemistry or mass spectrometry.
Electrospray mass spectrometry, a soft ionization method, [2] can be used to
transfer ionic species that are present in the solution phase into the gas phase with-
out significant fragmentation. This makes it possible to probe short-lived intermedi-
ates and products derived during electrochemical reactions in solution and analyze
them as a function of the working electrode potential. Experiments can be car-
ried out at close to real time by using this technique. [1,3,4] An electrochemical
flow cell with high electrochemical conversion efficiency is an important part of the
electrochemistry/electrospray mass spectrometry (EC/ES-MS) experimental appara-
tus. However, it has also been recognized that electrospray ionization sources act
as controlled-current electrolysis cells. [5] When an electrochemical cell is coupled
with an electrospray source, two problems commonly arise, namely electrochemical
cell potential interference due to the high voltage at the electrospray tip and severe
electrospray signal attenuation by the supporting electrolytes. [6] The objective of the
present study is to develop a new electrochemical flow cell and to apply the EC/ES-
MS method to several scientifically significant problems. These are briefly described
below and are presented in more detail in later chapters of this dissertation.
1
Iodide and Cyanide at Platinum and Gold Electrodes
Early investigations of the electrochemical oxidation of the iodide-iodine system
in different solvents at a platinum electrode mainly focused on the adsorption behav-
ior of the ions and the kinetics and mechanisms of the reactions. [7?13] The adsorption
behavior of iodide ions on a smooth electrode in aqueous solution was investigated
by Pirtskhalava. [10] Good agreement between the experimental and theoretical po-
lariographic characteristics of the iodide-iodine system in acetonitrile was reported
by Piccaradi, [11] and the electrochemical properties of iodide in different organic
solvents with a wide range of dielectric constants was reported by Iwamoto. [9] The
kinetics and mechanism of electrochemical oxidation of the iodide-iodine system were
investigated using a rotating platinum disk electrode by Giordano in two solvents,
acetonitrile and dimethylsulphoxide (DMSO). [14] The results showed that two po-
larographic waves were present in the system, and the difference between the half-
wave potentials of the two waves was related to the stability constant of tri-iodide ion
species in the solvents.
A number of studies have discussed the electrochemistry of gold(I) and its halide
complexes in both aqueous and non-aqueous solutions. [15?22] The structures of gold
iodide complexes, such as AuI?2 , AuI?4 , were investigated by Kissner, [23] and Bard
reported scanning tunneling microscopy images of iodide adsorbed on Au(111) sur-
faces from aqueous solutions. [18] The catalytic activity of halide ions towards the
electrochemical oxidation of gold electrodes in acetonitrile has also been reported. [24]
Gerischer examined the influence of water on the formation of an oxide layer on gold
in acetonitrile and found that the water was a reactant both in the formation of an
oxidation layer and reduction of the layer. [22]
2
The chemistry of gold cyanide complexes has been well studied. Early results
show that a monolayer of cyanide on the gold electrode plays a role in the gold
dissolution in the cyanide solution. [24?26] The formation of gold cyanide complexes
is primarily the result of the high stability constant.
Cyanide oxidation has attracted considerable attention because it is a general
waste treatment process to clear the contaminants in the industrial waste-water. The
oxidation of cyanide gives cyanogen, which is a reactive which can participate in
many reactions. [27] The mechanism of formation of C3N?3 and C5N?5 by reaction of
cyanogen with cyanide, which depends on the mole ratios of cyanogen and cyanide,
was proposed by Wiley. [28]
In this dissertation, we report our results on the electrochemistry of iodide as
well as cyanide at gold and platinum electrodes using EC/ES-MS in Chapter 3.
Fullerene C60.
The electron-accepting ability of C60 in solution has attracted much attention
from chemists since it was first discovered in the 1980s. [29?34] Molecular orbital
theory calculations reveal that C60 is a good electron acceptor and is capable of ox-
idation states up to negative six in solution. This is due to its triply degenerate
and energetically low-lying lowest unoccupied molecular orbitals (LUMO). [35,36]
The reversible stepwise reduction of up to six electrons in C60 was first demonstrated
by cyclic voltammetry following early molecular orbital calculations. [37] The field
of gas phase chemistry involving fullerene ions and its derivatives has experienced
tremendous growth during the last decade. Studies of C60 have provided very promis-
ing opportunities to explore exciting new areas of science. Positively multi-charged
gas phase fullerene ions with charge states ranging from 4+ to 1+ have been re-
ported. [38?40] However, observations of multiply-charged fullerene anions in the gas
3
phase are fairly rare. So far, only singly and doubly charged gas phase fullerene anions
have been reported. [41]
In the research presented in Chapter 4 of this dissertation, multiply-charged Cn?60
(with n up to 4) anions were studied using the EC/ES-MS. We observed C?60, C3?60 ,
C4?60 , but the di-anion, C2?60 , was not observed. Possible reasons for this are discussed.
4
1.2 Literature Review
1.2.1 Electrochemistry/Mass Spectrometry
Electrochemistry deals with the chemical action of electricity and the production
of electricity by chemical reactions. [42] This field encompasses a great variety of phe-
nomena. Electrochemical reactions are different from homogeneous chemical reactions
in that the amount of products formed at the working electrode is usually compar-
atively small. Many electrochemical techniques, such as linear scan voltammetry,
polarography, the pulse voltammetry method, and the controlled-current methods,
have been devised in order to obtain thermodynamic data about electrochemical reac-
tions, to generate products or unstable short-lived intermediates such as radicals, and
to detect trace amounts of metal ions and other organic or inorganic species. [42,43]
Though electrochemistry is well established as a field, the immediate detection of elec-
trochemical reaction products shortly after their formation on a working electrode has
been a difficult task.
Mass Spectrometry (MS) is an analytical tool that is primarily concerned with
the separation of molecular or atomic species according to their mass to charge ra-
tio. [44] Because mass spectrometry is capable of providing comprehensive informa-
tion on molecular masses of unknown species and short-lived intermediates directly,
as well as providing clues on molecular structures, the online combination of elec-
trochemistry and mass spectrometry holds the great promise to be a powerful new
approach. The main goal is to detect electrochemical reaction products, by-products,
or unstable intermediates shortly after their formation on a working electrode. The
initial combination of electrochemistry and mass spectrometry can be traced back as
early as the 1970s. [1]
5
Bruckenstein and Gadde first used an online combination of electrochemistry
with mass spectrometry in order to study gaseous products of fuel cells. [1] In their
work, gaseous electrochemical products were collected in a vacuum system, and then
the products were analyzed by electron impact ionization. Later, Heitbaum and
Wolter used a porous Teflon membrane covered by a 100 ?m layer of an electro-
catalyst that functioned as a working electrode as well as the interface to the mass
spectrometer. [3,45] To allow direct real time detection of volatile electrochemical
reaction products or intermediates, Heitbaum and Wolter significantly improved the
vacuum system by including two pumping stages. Due to the improved vacuum
design, this type of electrochemical and mass spectrometric interface was capable of
measuring rates of product formation. It was sensitive enough to detect desorption
products corresponding to a single monolayer of adsorbed species at porous electrodes.
To distinguish this technique from other methods, this method was called ?Differential
Electrochemical Mass Spectrometry? (DEMS).
DEMS is thus similar to membrane introduction mass spectrometry, which has
been widely used to detect trace amounts of non-polar molecules in water. [46] In
general, a thin, non-polar silicon or Teflon membrane is used to achieve the separation
of reactants from water. The non-polar membrane has the advantage of preventing
the diffusion of water into the mass spectrometer.
DEMS is ideally suitable to both the real time detection of electrochemical re-
action products and the study of the adsorbates on single crystal electrodes. [47]
With this method, analytes present in quantities of less than 1 nL can be easily de-
tected. [48] When coupled with electron impact ionization mass spectrometry, DEMS
can be used for quantitative measurements of electrochemical reaction rates and the
electrode surface coverage. The drawback of DEMS is that only volatile products can
6
be detected. Later, a technique incorporating rotating disc electrodes (RDE) coupled
with mass spectrometry was developed, this method allowed the study of fast elec-
trochemical reactions. [49] Using this rotating inlet system, the transfer efficiency can
be determined under controlled diffusion conditions. However, due to the difficulties
in interfacing the liquid environment of electrochemistry with the gas-phase environ-
ment of mass spectrometry, the combination of these two techniques has been limited
to studies of the volatile species, and has thus received little attention from chemists
working either in electrochemistry or mass spectrometry.
Over the past decade, several new ionization methods, including thermospray
(TS) [50?56] and electrospray (ES), [57?60] have been developed. This has made
interfacing the liquid environment of electrochemistry with the gas phase in the mass
spectrometer much more practicable. Hambitzer and Heitbaum [51,52] were the first
to study nonvolatile species in solution using electrochemistry coupled with thermo-
spray ionization in the 1980s. Brajter-Toth [56] and Yost [54,55] then expanded this
technique by combining it with high performance liquid chromatography (HPLC),
demonstrating a powerful new analytical method in the study of biological redox
reactions.
Other approaches have been taken by other groups. For example, Bartmess and
Phillips combined electrochemistry with fast atom bombardment ionization (FAB).
[61] In their work, the neutral analyte species were ionized by electrochemical redox
reactions on the FAB tip before sputtering the samples with a Xe atom beam. Other
techniques, such as coupling electrochemistry with mass spectrometry (particle beam
interface and atmospheric pressure chemical ionization (APCI)), along with variable
ionization-interfaces have also been attempted. [62]
7
The online combination of electrochemistry with electrospray mass spectrometry
(EC/ES-MS) became more common in the 1990?s. [63,64] This interface enables the
detection of nonvolatile solutes. It has been proven that EC/ES-MS is a powerful
technique for the direct on-line investigation of electrochemical reactions, it holds
promise for studies of the mechanisms of electrochemical reactions. As revealed by a
literature search, the number of reports of the online combination of electrochemistry
with electrospray mass spectrometry is increasing. It is the EC/ES-MS technique that
was used for the research for this dissertation. To trace the history of EC/ES-MS, it
is necessary to first look at the history of electrospray and mass spectrometry.
1.2.2 Electrospray and Mass Spectrometry
The idea of using electrospray ionization as a source for generating gas-phase
ions followed by mass spectrometry was first proposed by Dole in 1968. [65] In this
pioneering study, electrospray ionization was used to detect polymeric species, such
as polystyrenes. However, the observed gas phase ions were not themselves ionized in
the solution, and this made the results suspect. Pole extended this work by proposing
an electrospray ionization mechanism, the charge residue model, and this model has
become important in developing new applications of electrospray. [65]
Finally, in 1984, Yamashita and Fenn [57] introduced electrospray mass spec-
trometry. At approximately the same time, Aleksandrov and coworkers [66] reported
their independent development of electrospray mass spectrometry. Since then, the
development of electrospray mass spectrometry has revolutionized the application of
mass spectrometry for biochemical and pharmaceutical fields due to the fact that
many biochemical systems involve ions in solution.
8
Electrospray ionization is uniqueamongthe mass spectrometric methodsin terms
of both the principles on which it is based and the instrument required to perform
the experiments. It is an ionization (desolvation) technique by which ions present in
solution can be transferred into the gas phase, after which the gas phase ions can be
detected by mass spectrometry. Electrospray ionization results in very little fragmen-
tation and is one of the softest ionization methods available in mass spectrometry.
The overall process involved in the production of gas phase ions by electrospray
includes three major steps: [2] the production of charged droplets at the electrospray
needle tip; the shrinkage of relatively large charged droplets by solvent evaporation
and droplet fission; and the transfer of ions from the solution phase into the gas phase.
The latter is an extremely endothermic and endoergic process. Of the three steps,
only the first two are well understood.
Electrospray charged droplets are produced by pneumatic nebulization, which
is shown in Figure 1.1. Usually, a metal needle with a small inner diameter (ID)
(typically 0.1 mm ID and 0.2 mm OD) serves as the electrospray tip. A counter
electrode is located 1 - 3 cm from the needle and a high voltage of ?2.0 ? 5.0 kV
is applied to the metal needle, forming a high electric field in the air at the tip of
needle, vectorEc = 106 V/m. The sign of the potential depends on whether positive
or negative ions are to be detected. When an analyte is forced through the needle,
the electric field is sufficiently high to penetrate the solution, forming a double layer
on the surface of the Taylor cone. This disperses the emerging solution into a very
fine spray of charged droplets. These charged droplets have the same polarity as that
applied to the electrospray tip. The second stage involves solvent evaporation, during
which the droplets shrink and the surface charge increases. Eventually, at the Ray-
9
Spray
Mass Spectrometer
Liquid Flow
Metal
Capillary
High Potential
+ 2.0-3.6kV
Counter
Electrode
Nebulizing Gas
Flow
Figure 1.1: Schematic of the Electrospray Process Producing Positive Ions.The Po-
larity of the Applied Potential Determines the Polarity of the Ions Produced. [2]
10
leigh limit, equation 1.1 [67] describes how the coulombic repulsion overcomes the
droplet?s surface tension [68] and the droplet explodes.
qRy = pi(64??0R3)12 (1.1)
where ? is the surface tension; qRy is the surface charge of the droplet; R is the radius
of droplet and epsilon10 is the permittivity of vacuum.
The coulombic explosion forms a series of smaller, lower charged droplets. These
are the precursors of all the gas phase ions. The first stages of the electrospray
ionization process are relatively well understood. The later stages of the mechanism,
however, become increasing complex and are not well understood. It is these later
stages that ultimately gives rise to the gas phase ions.
Two mechanisms have been proposed to explain the final stage of the evaporation
of droplets remaining near the Rayleigh limit. [67] The first is the charged residue
model proposed by Dole, et al. in 1968, [65] shown in Figure 1.2. In this model, the
formation of gas phase ions depends on the development of extremely small droplets,
which contain only one naked analyte ion. The ions in solution are never emitted
from the charged droplet?s surface at any point during the whole process. Repeated
solvent evaporation followed by coulombic explosions ultimately produces droplets,
which contain only naked analyte ions. The formation of multiply charged ions is
dependent on the neutral analyte?s available surface sites and on the correct conditions
being present.
A detailed consideration of, and support for, this mechanism was presented by the
R?ollgen [69,70] and Kebarle groups. [2] They reported the observation of M(MX)n+
11
5-10nm
Figure 1.2: Dole, et al. Charge Residue Model. [65]
12
and (MX)nX? clusters from solutions containing (NaCl, KCl, NaBr and CsCl), where
the highest intensity was at n=1. These results agreed with the prediction of the
charged residue model.
The second possible mechanism is the ion evaporation model, shown in Figure
1.3, proposed by Iribarne and Thomson. [59,60] In principle, it is similar to the
charge residue model at the very beginning. The formation of the charged-droplets
follows the same sequence of evaporation and explosions when Rayleigh instabilities
occur. However, unlike the charge residue model, Iribarne and Thomson?s model
argues that prior to a charged droplet reaching the ultimate stage contemplated by
Dole, the electric field on the droplet surface becomes strong enough to overcome
solvation, and solute ions from the droplet surface are emitted into the ambient gas.
For charged droplet sizes ? 10 nm, the direct emission of ions from the solution thus
becomes possible. The emission process is certainly dominant over the coulombic
fission. Iribane and Thomson also derived equation 1.2 based on transition state
theory, which can be used to predict the rate of ion evaporation from the charged
small droplets.
kI = KBTh e??G?/KBT (1.2)
where KB is Boltzmann?s constant, h is Planck?s constant, and ?G? is the Gibbs free
energy of activation.
Experimental evidence supporting the ion evaporation model was presented by
Loscertales and Fernandez de la Mora, [71] who studied the electrospray mechanism
by producing a monodisperse cloud of charged droplets. The mechanism was verified
by measuring the charge, q, and diameter, r, of the residual particles remaining after
13
Initial state
   R
nm
Xm
0.6 nm
d
R
    d
Transition state
Figure 1.3: Ion Evaporation Transition State and Initial State Models. [59]
14
complete evaporation of the electrospray solvent. The experimental results showed
that the electric field E on the surface of the droplet is independent of r when the
droplets contain small monovalent dissolved ions. A consequence of this is that on a
clean surface, the rate of ion ejection from a charged droplet is simply related to the
rate of solvent evaporation. The measured ionization rates were well represented by
the results from the ion evaporation model.
It possible that the ion formation during the electrospray process involves con-
tributions from both solvent evaporation and ion ejection, and that the dominating
process depends on the chemical and physical characteristics of the solvent.
1.2.3 Electrochemistry/Mass Spectrometry Flow Cells
Several research groups have reported successful applications of EC/ES-MS.
[4,72?78] This combination has proven to be very effective for investigating elec-
trochemical processes. [55,63,79?84] The overriding strength of the method is that
it promises the full characterization of the details of electrochemical reactions. In a
recent paper, Permentier and Bruins reported the study of electrochemical oxidation
and the cleavage of small proteins, insulin and ?-lactalbumin, by EC/ES-MS. This
method is also feasible for use in analyzing molecules such as tyrosine and tryptophan.
These results indicated the possibility of developing a new technique with which to
study enzymatic and protein digestion. [85]
A difficulty in implementing EC/ES-MS is that the electrochemical cell is op-
erated relative to ground potential, while the electrospray tip operates at a high
potential (usually ?2.5?3.6 kV), and current leakage may occur between the two.
In addition, for experiments conducted in organic solvents, tetraalkylammonium salts,
15
often used as supporting electrolytes, can severely attenuate ES-MS signals; [6] hence
it is often necessary to use low concentrations of supporting electrolytes.
The central part of the EC/ES-MS setup is the electrochemical cell and the an-
alyte flow in the cell. In contrast to homogeneous chemical reactions, the amount of
product generated on the working electrode surface is relatively small in an electro-
chemical reaction. In order to generate sufficient products at working electrodes to
be detected by mass spectrometry, a flow cell must have a high electrochemical con-
version efficiency, well-defined hydrodynamics, and allow easy access to the working
electrode. The cell design should also take into account the optimal positions of the
reference electrode, counter electrode and working electrode.
A wide range of electrochemical cell designs have been developed for electro-
chemistry/mass spectrometry, these are discussed below.
Flow Cells for Electrochemical/Electron Impact Mass Spectrometry
A typical electrochemical cell used in the DEMS setup is shown in Figure 1.4.
[1,45] In this method, electron impact ionization mass spectrometry is used, hence
the DEMS cell is particularly suitable for studying electrochemical reactions that
generate gaseous or volatile species such as CO, CO2 and H2. A critical part of the
cell is the Teflon membrane used to separate the solution from the vacuum chamber.
The membrane, typically 75 nm thick, has a nominal pore width of 20 nm and a
porosity of 50%. The Teflon membrane is mechanically supported by a glass or
steel frit. An electrocatalyst layer, generally platinum, is deposited onto the Teflon
membrane, producing a catalyst layer with a typical thickness of 50?80 nm. The
hydrophobicity of the Teflon membrane stops the diffusion of aqueous solvents though
the membrane, whereas dissolved volatile products and gaseous species can penetrate
the pores and be detected by electron impact mass spectrometry. The critical pore
16
a b
c
Vacuum
1
23
4
Membrane Working Electrode
EC/MS Interface
Figure 1.4: Schematic Diagram of the DEMS Cell. [45]
Top: Membrane working electrode. a, Electrocatalyst; b, Teflon-membrane and c,
Pore size ? 0.02 ?m.
Bottom: EC/MS interface. 1, Glass Body; 2, Teflon-spacer; 3, Working Electrode
and 4, Steel frit.
17
size depends on the surface tension of the liquid and contact angle of the solution with
the Teflon membrane. [1,45] The size of the pores, calculated for water, is less than
0.8 ?m. A cell with this type of working electrode has a very small dead volume. The
response time of this cell for a electrochemical reaction is about 0.8 s. [48] Since the ion
intensity determined by electron impact mass spectrometry is directly proportional to
the incoming flow of the gas (Ji = K0JI), rates of product formation can be measured
by recording the ion current.
Tegtmeyer [49] developed a rotating porous electrode to be used as the inlet
system coupled with mass spectrometry. For an approximately 100 nm thin sputtered
platinum pore electrode, the transfer efficiency is greater than 0.9, this efficiency can
be obtained even at high rotating speeds. [49] The diffusion layer from the macroscopic
surface of the electrode to the Teflon membrane is comparable to the diffusion layer
in the bulk electrolyte.
Flow Cells for Thermospray and Electrospray Mass Spectrometry
Because thermospray and electrospray interfaces can transfer the analyte into the
gas phase and ionize nonvolatile species without significant fragmentation, many types
of electrochemical flow cells have been coupled with thermospray or electrospray mass
spectrometry. [4,72?78] Of the many geometries possible, three are most commonly
used: thin-layer, wall-jet, and tubular cells. The most widely used of these are the
flow cells based on the thin-layer and wall-jet configurations, shown in Figure 1.5. [42]
Here the flow channel is formed by applying pressure to a thin Teflon gasket, which
results in a very small dead volume (? 1 ?L). The working electrode is embedded in
a block in the cell. A high electrochemical conversion efficiency results from the thin
layer of sample solution flowing parallel to the planar working electrode surface.
18
Inlet
OutletOutlet
Working electrode
b
InletOutlet
Working electrode
a
Figure 1.5: Thin-layer (a) and Wall-Jet (b) Flow cells
19
The design of many electrochemical flow cells available commercially are based on
the thin-layer flow electrode. In order to achieve high conversion efficiency and high
electrolyte flow rates, most thin-layer flow cell designs have sacrificed the separation
of the counter and working electrodes. A simple two-electrode electrochemical flow
cell was used both by the Bond and Van-Berkel groups. [76,86] These EC/ES-MS
setups used two metal capillaries. The connection between the electrospray mass
spectrometer and the pump was made through these two metal capillaries, which also
served as the working and counter electrodes.
A three-electrode flow cell without separation of working and counter electrodes
was developed by Anne Brajter-Toth?s group. [79,80] A porous flow-through electrode
was used as the working electrode. The electrochemical process occurring on the
working electrode was characterized as a complex electrochemical reaction. The cell
design sacrifices the separation of counter, working, and quasi reference electrodes in
order to achieve a high cell conversion efficiency and high electrolyte flow rates. A
three-electrode flow cell with a 12 cm2 reticulated vitreous carbon graphite-working
electrode has also been introduced. [54,55,63]. The advantage of this cell design is
the ease of access to the working electrode. However, the design of the thin-layer flow
cell has a high ohmic loss, which results in the observation of no limiting current for
ferrocyanide reduction over the whole potential range measured.
To decrease the response time of this EC/ES-MS system, Cole and coworkers
placed an electrochemical cell at the very end of the tip of the electrospray nozzle. [87,
88] In this configuration, the auxiliary electrode of the electrochemical cell functions
also as the electrospray tip. However, this type of setup requires the potentiostat itself
to be connected to the electrodes, therefore making it subject to the high potential
at the electrospray tip, ?2.0 ? 4.5 kV. A similar design was adopted in ref [54],
20
where a stainless steel electrospray needle functioned as both the working electrode
of the three electrode cell and the electrospray tip. The potential is controlled by a
battery instead of a potentiostat due to the high potential at the electrospray tip.
A three-electrode thin layer flow cell without separation of working and counter
electrode compartments has also been used by van Berkel and co-workers for EC/ES-
MS.[4,74,76,84,90,91] Thisthinlayer flowelectrolytewas typicallyabout16?mthick.
[4,84,90,91] Due to the small dead volume of the cell, this design was characterized
by over 50% conversion efficiencies for flow rates below 50 ?L/min. [4] The response
time of this EC/ES-MS system was on the order of 1 s. However, a high ohmic drop in
the thin layer cell caused a considerable deviation in the voltammetric response from
reported data and the cyclic voltammetry measured in a conventional electrochemical
cell.
Hambitzer et al. first introduced electrochemical thin layer flow cells with sep-
aration of the working and auxiliary electrode compartments using a thermospray
technique as the interface to combine the electrochemistry and mass spectrometry
(TSIMS). [51,92] The separation of the individual compartments was achieved by
using glass frits. This flow cell design had a smooth, metal working electrode with
a relatively large surface area of about 1.5 cm2, and flow rates through the cell were
on the order of 1 mL/min. This cell design has another advantage, as it can simulta-
neously record voltammograms and the potential dependent abundance of products
and reactants of the electrochemical reaction.
In a recent paper, Modestov and Ovadia Lev introduced a new, radial, miniature,
three-electrode flow cell. [93] The cell is characterized by low resistive losses and high
conversion efficiency. The three-electrode compartments were separated to prevent
the counter electrode reaction by-products from entering the product stream reaching
21
the mass spectrometer. The authors developed a mathematical model describing
and theoretically predicting the convection diffusion in the radial cell over a wide
range of parameters. The performance of the cell was demonstrated by studies of
dimethylaminomethyl ferrocene (DMAMF) oxidation. The mathematical model was
validated using the simple single electron charge transfer reaction of DMAMF.
In chapter 2, we introduce a new spherical thin layer flow cell to interface elec-
trochemistry with mass spectrometry.
1.2.4 Oxidation of Iodide and Cyanide at Platinum and Gold Electrodes
The oxidation behavior of iodide on platinum and gold working electrodes in non-
aqueous solutions, covering a wide range of dielectric constants, has been extensively
studied since the 1970s. [7?12] The kinetics and mechanisms of iodide ? tri?iodide
in acetonitrile were studied using a rotating disk electrode from 0?30?C. The an-
odic and cathodic processes showed two well-defined waves, which were related to the
oxidation of iodide and tri-iodide. The limiting currents increased linearly with the
square root of the rotation speed of the platinum rotating electrode at a constant
solution composition. The results also showed the kinetics of the electrode reactions
involving iodine-iodide couples and iodide adsorption on platinum electrodes in both
dimethylsulfoxide (DMSO) and acetonitrile solutions. The formation of an adsorbed
iodide film on the platinum electrode was confirmed by a Tafel plot (4RT/F). The
diffusion coefficients (Di) of iodide and tri-iodide in acetonitrile and DMSO at 20?C
were determined to be 1.68?10?5 cm2/s and 1.55?10?5 cm2/s. [7] The rate deter-
mining step was found to be the electron transfer reaction on the platinum rotating
disk electrode by evaluating the kinetic parameters of the diffusion coefficients.
22
The early cyclic voltammetric studies of acetonitrile solutions containing iodine
and iodide and tri-iodide on platinum electrodes showed four waves, two for the anodic
reactions and the other two for the cathodic reactions. [13] For solutions containing
an excess of iodide, two anodic waves and only one cathodic wave were observed.
The two anodic waves were related to the oxidation of iodide and tri-iodide, while
the cathodic wave was due to the reduction of tri-iodide.
Formation of a iodide adsorption layer on the platinum electrode surface occurs
when the iodide concentration exceeds ? 5 mM. [9] The kinetics and mechanism of
formation, growth and dissolution of the iodide film were studied under both fixed po-
tential and open circuit conditions. [9] The oxidation behavior of an iodide film on the
platinum electrode was studied by a rotating ring disk electrode. [9] A hydrodynamic
modulation was used to study the oxidation behaviors under potentiostatic steady
state conditions at a constant rotation speed. The experimental results showed that
the disk current was controlled by the mass transport coupled with the equilibrium
of the iodide-iodine couple. The thickness of the adsorbed iodide film depended on
the flux of iodide at the film-solution interface. A consideration of the diffusion co-
efficient of iodide indicated that the iodide transport through the film followed the
Grotthuss chain transfer mechanism. Further calculations indicated that the iodide
film relaxation processes were slower than the relaxation of the diffusion layer.
A number of studies have discussed the detailed mechanisms of the initial stages
of the electro-oxidation of gold. [16?20,23] The absorption of anions, such as halides
or oxyanions ClO?4 , are known to play a major role in the initial stages of oxidation
of gold.
Oxidation of gold can be catalyzed by halides in acetonitrile. [23] The reaction
can be inhibited by traces of water. Experimental results have shown that the gold
23
electrode surface may be covered by a mono-ionic halide layer in the presence of even
a small amount of halide. The catalytic effect of halide in acetonitrile increases in
the order of coordinating ligands (Cl? < Br? lessmuch I?). A positive potential at the gold
electrode can promote a strong halide adsorption. [19,20,23]
It is well known [27] that cyanide can be oxidized by numerous chemical reac-
tions. Previous results show that the electrochemical oxidation of cyanide in aqueous
solution gives cyanogens. [94] Tri-cyanide is the most easily made pseudo-halogens.
A pseudo-halogen is similar to the halogens in their physical and chemical proper-
ties. [95] A well-known reaction of cyanogens is the nucleophilic addition. [27] The pos-
sible mechanism of stepwise formation of C3N?3 , C5N?5 and C7N?7 was proposed. [28].
Although C3N?3 was proposed as a pseudo-halogen ion, the structure of C3N?3 can
not be treated as the linear structure as of the tri-halogen ions. [96] This is because a
linear C3N?3 results in a very unstable structure which requires the centrally located
nitrogen atom to have a positive charge. The studies of the formation of C5N?5 and
C7N?7 in the non-aqueous solution suggested that the addition processes involved a
two stepwise cyanogens condensation with cyanide. [28]
In Chapter 3, we will study the oxidation behavior of iodide and tri-iodide as well
as cyanide at platinum and gold working electrodes using the EC/ES-MS method.
1.2.5 Fullerene C60
It was first discovered that vaporization of graphite by laser irradiation pro-
duced stable C60 and C70 clusters in 1985. [97] These clusters, known as fullerenes,
have attracted significant attention from scientists in chemistry, physics and mate-
rial sciences. This widespread interest in fullerenes is due, in part, to their chemical
24
properties. [98,99] One of the most characteristic chemical properties of C60 is its un-
usual ability to accommodate negative charges. The structure of C60 is described as
a truncated icosahedron composed of 32 faces, of which 12 are pentagonal and 20 are
hexagonal. [35] The closed-shell configuration consists of 60 pi-electrons in 30 bonding
molecular orbitals. Early theoretical molecular orbital calculations predicted that C60
would have three low-lying degenerate unoccupied molecular orbitals (LUMO) at the
T1u level, 2 eV above the highest occupied molecular orbitals (HOMO) Hu. Figure
1.6 shows a potential energy diagram for the orbitals. This electron configuration
makes C60 a good stepwise electron acceptor for up to six electrons. [36,100,101]
Electrochemical Generation of C60 Ions in the Solution Phase.
Shortly after the advent of the technique of producing gram quantities of pure
buckminsterfullerene (C60), attempts to confirm the theoretical predictions of C60
electron accepting ability were begun. [102?105] The methods commonly used for the
generation and study of C60 multi-charged anions in solution phase include: cyclic
voltammetry (CV), ultraviolet or visible light (UV) absorption, electron paramagnetic
resonance (EPR) and nuclear magnetic resonance (NMR). A literature search reveals
that the most significant discoveries about C60 multi-charged anions in solution phase
have been the result of finding the right cyclic voltammetric conditions, including the
cell temperature, the solvent, the working electrode material, the scan rate and the
supporting electrolyte. [37]
However, between 1990 and 1991, CV experiments recorded only five reversible
reduction peaks of C60. This is because of the limitations of the solvent potential
window. In 1992, three groups [104,106,107] reported the detection of the reversible
25
Hu       
T1u LUMO
HOMO
T1g LUMO + 1
Figure 1.6: Three Low-Lying Degenerate Lowest Unoccupied Molecular Orbitals of
C60. [36]
26
stepwise addition of six electrons to C60 by cyclic voltammogram in solution phase.
Qie?s group recorded the six successive, reversible reduction peaks using CV and
differential pulse voltammetry in a mixed-solvent system at ?10 ?C. The solvent
system consisted of toluene and acetonitrile in a 5 to 1 volume ratio. The potential
separation between two consecutive peaks was almost constant, at 450?50 mV. This
is consistent with the earlier theoretical predictions. [107]
Given the well-defined, distinct reduction potentials of fullerenes, it is natural
that classical electrochemical methods, such as controlled potential bulk electrolysis
and potential sweeps, are widely used to generate Cn?60 (n = 1?6). [102]
Cn?60 (n = 1?3) ions were electrochemically generated by square wave voltamo-
gram, then EPR was used to study the anions. [108,109] The results showed that a
dynamic Jahn-Teller distortion is operative in all three anions.
Electrosynthesis and electrodoping of C60n? (n = 1?3) films were studied using
an electrochemical quartz crystal microbalance in acetonitrile solution. [110] The re-
sults indicated that C60 films could be electrodoped with Bu4N+ cations from solution
upon electroreduction at appropriate potentials. [110]
Fullerides and fullerene cations have electronic absorptions throughout the visible
and ultraviolet spectral regions. The Cn?60 (n = 1 ? 5) ions have distinctive near-
infrared (NIR) spectra, as reported by Lawson, [111] Heath and coworkers, [112,113]
and Baumgarten and coworkers. [31]
In a recent paper, Bruno and coworkers reported the first electrochemical re-
versible generation of C602+ and C603+ ions in ultra-dry dichloromethane (DCM) so-
lutions. [114] The experiments were carried out using supporting electrolytes contain-
ing low nucleophilic and high oxidation resistant counteranions, such as [Bu4N]AsF6.
These were used because of their high electrochemical oxidation potential (as high
27
as 4.7 V vs SCE). [115?117] The CV curve they recorded showed two reversible one-
electron oxidation peaks, at the E1/2 = 1.27 and 1.71 V (vs. Fc+/Fc), that were
attributed to the oxidations of C60 to C60+ and C602+, respectively.
Chemical Generation of Fullerides
Given the fact that nearly 450 mV separates the stepwise reductions of C60,
an alternative method for generating Cn?60 anions is to use chemical reducing agents
that are specific for a desired reduction rate and state. In this manner, a careful
stoichiometric control of strong reducing agents is usually required in order to produce
soluble C60 anions or salts, because the solvents that dissolve C60 and its anions often
do not dissolve the C60 salts. [118,119]
Due to the powerful reducing abilities of alkali metals (Li, Na, K, Rb, Cs), they
have been successfully used in organic solutions or liquid ammonia to chemically
generate C60n? (n up to 5) anions. [120,121] However, this method requires an in
situ monitoring of the reduction state to stop the reduction at the level of reaction
desired. Successful generation of C60n? (n up to 6) anions using this method is
reported. [122,123],
Cn?60 anions were produced by simply sonicating a C60 solution in the presence
of an excess of lithium, sodium metal or potassium naphthalide. The -3 state of
C60 was generated by a sonication of C60 tetramethylethylenediamine solution with
an excess of potassium. [121?124] Alkali metals are active and difficult to handle,
but reducing agents of complexants, such as crown ethers or cryptands, are relatively
inert and easier to handle, so they provide an attractive alternative choice for reducing
C60. [125]
28
A complexed alkali metal is usually characterized by a low solubility in the sol-
vents into which C60 anions dissolve, thus crystallization of the salt in the solution is
easily accomplished, and further reductions can be halted by separating the products
from the excess of alkali. Using this method, Na(dibenzo?18?crown?6)+ salts of
C60n? for n ( 1, 2, and 3 ) were successfully isolated from THF solutions. [126?128]
The highest charged fulleride anion so far isolated using this method is C604?. The
complexant 2.2.2-cryptand was used as the reducing agent. [129]
Generation of Cn?60 (n = 1?6) using alkali metals is superior to using electro-
chemical methods for some purposes, because no supporting electrolytes are needed
during the reduction. The drawback is that it is difficult to control the stoichiometry,
and products must be quickly isolated from the excess of alkali metal in order to avoid
over-reduction.
An ingenious way to control the stoichiometry and monitor the C60 reduction
state is to use an electron carrier to control the electron transfer. [130] In this method,
the desired level of C60 reduction can be determined by the added carrier molecules.
By using cryptand as an electron carrier without sonication, C60n? (n=4) anions have
been successfully generated with an excess of alkali metal in THF. [123] Huang and co-
workers reported the generation of C60n? (n = 1?3) salts using an electron carrier,
1-methylnaphthalene, as well as stoichiometric control of alkali metal. [123?125]
Other attempts to generate C60 [130] have made use of the coordination and
organometallic compounds, such as Cr(TPP) (tetraphenylporphyrinato) and cobal-
tocene, as reducing agents. These compounds have precisely defined redox potentials,
which makes them well-suited as reducing agents for the generation of fullerenes an-
ions. [130?134] Schmelzeisn-Redeker et al. [70] reported generation of C60n?(n=1)
salt by treatment of [Cr(TPP)(THF)2] with C60 in toluene/THF solvent mixture.
29
The redox potential can be shifted into the appropriate range by adding THF, and
the Cr(III) can be stabilized via coordination.
Observation of C60 Ions in Gas phase
Gas Phase C60 Cations. Studies of gas phase fullerene C60 cations by using
the mass spectrometry have been a major subject of research of C60 in the last decade.
Ionization of gas phase neutral C60 has been accomplished by various methods such
as electron impact, [39] collisions with fast atoms, [136,137] neutral particles, [138]
ions with high charge states as well as cluster ions, [139] photons with energies up to
340 eV, and surface collision [140]. It was found that C60 needs to capture at least
45 eV of energy in order to allow the ionization and subsequent dissociation of C2
loss. [38]
Much of experimental work of the electron-fullerene interaction has been devoted
to the study of the formation of singly and multi-charged C60 cations. M?ark and
coworkers [141] reported the first measurements of cross-section functions for the
production of the various parent Cz+60 and the most abundant fragment ions C60 by
single electron collision. The electron energies range from threshold to up to 1000 eV.
A crossed-beam apparatus was used in order to achieve the accurate determination
of partial ionization cross-sections. The results show both an unusually large parent
ion cross section and irregularities in the production of multiply charged parent and
fragment ions.
Theelectron-impactinducedfragmentationoffullerenewasalsostudiedbySalzborn
group. [39] The results show that two different processes dominate the cross section
for the loss of C2 fragment. At low electron energies, the electron-fullerene interaction
leads to the direct excitation of the fullerene. When the electron energies are greater
30
than 100 eV, the dissociation of the fullerene is due to an unsuccessful ionization. It
is the higher energy part of the cross section that shows a dependence on the charge
state, q, of the precursor ion.
The dissociation of fullerene ions Cz+60 (z=1,2,3,4) upon the excitation undergoes a
series of C2 losses, equation (1.3), only even-numbered fragment of ions are observed.
Cz+60 ?? Cz+60?2m + C2m (1.3)
Therearethreedifferentcoolingmechanismsproposedfortheexcitedfullerenecations,
i.e. radiative cooling, [142] the release of neutral or charged particles, [143,144] and
the evaporation of electrons. [145]
Gas Phase C60 Anions. Multiply charged fullerene anions (C60) are very
common in condensed phases, and their existence and stability have been proven, but
their observation in the gas phase has been problematic. C60? is readily detected
in negative ion mass spectrometry, [31,146?149] however, like other multiply-charged
anions (MCAs), reports of observations of gas phase multiply charged fullerene anions
are fairly rare.
Gas phase C2?60 and C2?70 [150,151] have been observed by Fourier transform mass
spectrometry. In this method, a probe with a fullerene film coated tip was placed
in the source chamber of a Fourier transform/ion cyclotron resonance (ICR) mass
spectrometer. A laser was used to desorb the C60 film. In order to distinguish the
signal of the second harmonic of the monoanion C?60 from the signal of the dianion C2?60 ,
C?60 was selectively ejected from the ion trap by single-frequency resonant irradiation
at the ICR orbital frequency of C?60. [150] At a backgroud pressure of 10?5 Pa, the
31
lifetime of C2?60 is about 10?2 second. The observed abundance ratios of M2?/M? are
between 0.02-0.20. [150]
An attempt was also made to generate gas phase C2?60 by attaching two successive
electrons to the fullerene neutral in the gas phase. [150] In this method, C60 was
thermally desorbed to interact as a vapor with a low-energy electron beam. [150]
Only C?60 was obtained, so attachment of a second electron to C2?60 was unsuccessful
in this method. Thus, this proves that the laser desorption technique ejects C?60 and
C2?60 directly from the metal surface. [150] Dianionic adducts of C60 with F47,48 [152]
and (CN)2,4,6 [41] attached are also observed.
A literature search reveals that the observation of gas phase Cn?60 (n > 2) to date
has not yet been reported. A consideration of the electron affinities (EAs) of C60
provides a simple rationale for this fact. [153] The EA of C60 is 2.65 eV, [154] whereas
the pseudo-potential calculations indicate the second EA of C?60 is in the range of
0.1-0.4 eV. [150] The fact that C602? is only observed under extremely energetic
conditions, while dianions of higher fullerenes are much more common in negative ion
spectrometry, has been attributed to the more positive EAs for the larger molecules
and the ability of the larger fullerenes to mitigate coulombic repulsion. [155,156]
In Chapter 4, C60 reduction is studied by using the EC/ES-MS technique. Gas
phase observation of C?60, C3?60 and C4?60 anions generated at platinum electrodes and
detected by electrochemistry/electrospray mass spectrometry is reported.
32
Chapter 2
Cell Design for Electrochemistry Electrospray Mass Spectrometry
Abstract
In this chapter, we introduce a new spherical thin layer flow cell with a relatively
small volume, experimental conditions suitable for electrochemistry electrospray mass
spectrometry using the cell are also described. The flow pattern through the cell is
estimated by calculating the Reynolds (Re) number. The cell is characterized by easy
access to the working electrode for cleaning and a high electrochemical conversion
efficiency. The performance of the spherical thin-layer flow cell is demonstrated in
this chapter by off-line electrochemistry.
2.1 Introduction
One of the major advantages of the combination of electrochemistry electrospray
mass spectrometry (EC/ES-MS) is to simultaneously record the voltammetric and
mass spectrometric responses, thus mechanisms of electrochemical reactions can be
probed by this method. In order to successfully achieve this goal, a good design for
an electrochemical flow cell is necessary.
Various home-built and commercial flow cells, which are based on thin-layer flow
electrodes, have been coupled with electrospray mass spectrometry to study different
types of electrochemical reactions under a variety of conditions. [4,74,75,77,79,80,
85,91] A challenge of designing a flow cell is to provide easy access to the working
electrode for cleaning, thus meeting the requirements for different electrochemical re-
actions. At the same time the cell should have a small dead volume and a predictable
33
electron conversion efficiency. A good flow cell design also should fulfill the transi-
tion from a qualitative description of the composition of cell e?uent to quantitative
analysis.
No matter what the purpose of the cell, when designing a new flow cell, some cri-
teria should be considered for optimizing the cell design: the design and construction
of a electrochemical flow cell should meet the basic requirements of the desired level
of precision required for the electrochemical measurements, optimum electrode geom-
etry, the chemical reactivity of the system, the working temperature of the system
and the scale of the system. The cell material also should be resistant to chemical
attack of the solution loaded in the cell and it should be reasonable easy to change
solutions and to clean the cell between the measurements.
When a thin-layer flow cell is coupled with electrospray mass spectrometry, the
following additional criteria should be considered in order to meet the special chal-
lenges posed by EC/ES-MS studies.
1. The degree of conversion of the analyte in the cell should be easily estimated
or calculated.
2. The cell should be connected to the electrospray source by a ?bridging? line
which enables fast substance transfer and protection of the electrochemical
equipment from current leaks from the electrospray needle.
3. The reference and counter electrodes should be separated from the working
electrode to prevent the mixing of products of the reactions in the two com-
partments.
34
4. Since electrospray ionization sources have been recognized as controlled-current
electrolytic cells, [72,75,157?160] operation of the electrostatic sprayer necessi-
tates passage of currents in the test liquid. The interference of these currents
on the liquid sample composition needs to be minimized relative to effect of
electrochemical flow cell.
2.2 A New Electrochemical Flow Cell, Construction and Operation De-
tails
Part of the work presented in this dissertation was to develop a new low-cost
electrochemical flow cell with a predictable conversion efficiency. In order to make the
new flow cell configuration meet the special challenges posed by EC/ES-MS studies,
our first concern was to build an easy access working electrode for cleaning which
could meet the requirements for the various electrochemical reactions to be studied.
This goal was achieved by using high purity gold or platinum balls as the working
electrode.
The working electrodes were made from gold (99.9999%) or platinum (99.9999%)
wires (1 mm OD). The gold or platinum working electrodes themselves were fabricated
by melting a ? 2 mm diameter ball in a reducing H2/O2 flame at one end of the wire.
Then the gold or platinum ball was annealed in the H2/O2 reducing flame for 30 mins
prior to use. A freshly annealed electrode was always used for all off-line and on-line
electrochemical experiments. The annealed electrodes have polycrystalline surfaces
with surface areas between 0.124 and 0.264 cm2. The surface areas were measured
by the reduction of K3[Fe(CN)6] in aqueous solution. Teflon heat shrink tube was
used to seal the reminder of the gold/platinum wire, allowing only the ball-shaped
electrode to be exposed to the solution.
35
C
To ESI Probe
A     C
F
G
2. Top view of electrochemical cell
A B
15mm
D
E
 1. Cross section of electrochemical cell
Figure 2.1: Schematic of the Electrochemical Flow Cell
36
The design of the electrochemical cell is shown in Figure 2.1. A schematic dia-
gram of the electrochemical working electrode and ion collection region are shown in
the top of Figure 2.1. The flow cell body is made from a Pyrex glass tube with two
coaxial cylindrical chambers, D and E. Chamber D (ID 2.5 mm) is for the working
electrode inlet, and E with a larger ID (4 mm) is for a cup-shaped glass tube outlet.
The working electrode is placed at one end and the cupped collection assembly lead-
ing to the mass spectrometer at the other end. The collection assembly is composed
of a fused silica capillary and a convex thick-walled Pyrex capillary tube. A and B
show the placement of the working electrode, the cupped collection assembly and the
flow around the working electrode into the capillary.
A 800-900 mm long untreated silica capillary (ID 0.1 mm, Supelco) was used
to connect the electrochemical flow cell and electrospray probe. The assembly, B
in Figure 2.1, is composed of a fused silica capillary, which is glued into a convex
thick-walled Pyrex capillary tube using 24-hour-cure Torr Seal (Varian). This epoxy
prevented leaks and solvent attack against the glue. The advantage of using a 800
- 900 mm long untreated silica capillary (ID 0.1 mm) as a duct line connecting the
electrochemical flow cell and the electrospray probe is that the silica capillary electri-
cally isolates the electrochemical cell from the electrospray tip potential, this results
in a reduction of current leaks from the electrospray needle.
A Swagelok fitting and Teflon tube (ID 0.25 mm) was used to connect the silica
capillary to a homemade electrospray probe. The design of homemade probe is based
on the Micromass factory-built electrospray probe. All other connections were made
either by O-rings or Supelco septa (Diameter 11 mm).
Two types of home-made reference electrodes were used, aqueous (Ag/AgCl) and
nonaqueous (Ag/Ag+). The choice depended on the experimental conditions and the
37
requirements for the desired level of precision of the electrochemical measurements.
The reference electrode was placed over the working electrode upstream of the working
electrode (Figure 2.1, part C). To separate the reference electrode from the working
electrode, the reference electrode was connected to the working electrode chamber
using a double salt bridge with a fine Vycor glass frit. The entire electrochemical cell
was maintained at room temperature.
A Cole Palmer Series 74900 syringe pump with a flow rate of 1.0 mL/hr was used
to provide the constant mobile phase and produce the spray. The syringe pump was
connected to the cell by a Teflon tube (0.25 mm ID) with a Swagelok fitting.
The cell volume is determined by the cup-shape collector (Figure 2.1 part B). A
well made ?cup? has a volume of 5 - 10 ?L. The gold or platinum ball, serving as a
working electrode, was placed into the center of the ?cup? without touching the inner
wall of the ?cup?. The ?cup? functions as the collector, which collects the products
generated at the surface of the working electrode. The flowing solution between the
sphericalworkingelectrodesurfaceandtheinnerwallofthe?cup?servesasaspherical
flow thin-layer, and the products formed on the electrode surface are guided into the
800 mm untreated silica capillary to the electrospray probe. In this configuration, the
cell dead volume of the spherical flow cell can be limited to a relatively small volume,
thus meeting the requirements posed by the EC/ES-MS experiments.
A piece of platinum mesh, used as counter electrode, was connected directly
to the working electrode compartment. This configuration maintains a small cell
volume and at same time provides reasonably easy assembly of the electrochemical
cell. In order to isolate possible counter electrode reactions from the flow stream
into the electrospray source, the counter electrode was placed outside the cup-shaped
38
collector and upstream from the working electrode. A drawback of this configuration,
however, is that an uneven current distribution could develop.
The electrochemical cell body was always cleaned using a solution mixtures of
sulfuric acid (98%) and H2O2 (30%) (piranha solution) [161] and dried in a vacuum
oven at 80 ?Cfor at least 72hoursprior touse. Theworking electrodes werecleaned by
rinsing sequentially with distilled water, hydrofluoric acid 49%, piranha solution, [161]
distilled water, HPLC grade methanol and fresh acetonitrile. The electrodes were then
stored in acetonitrile in a glove bag under an argon atmosphere. The assembly of the
electrochemical cell was carried out in a glove bag under argon and all solutions were
degassed for at least 15 minutes with dry Grade 5.0 Ar gas before use.
2.3 Flow Pattern
Since the way solutions pass over a working electrode can greatly affect the
electrochemical processes happening on the electrode surface, [162] it is important for
us to understand the solution flow pattern around the spherical working electrode. To
study the flow around a spherical working electrode, a Reynolds (Re) number, [163]
equation(2-1),was used to estimate the various fluid properties at different velocities.
The Re is defined as:
Re = ?v d? (2.1)
The specific values for our cell using acetonitrile are: ? is the density of acetoni-
trileat20 ?C, 0.776?103 kg/m3; v isthemeanvelocityofacetonitrile, 1.26?10?5 m/s
- 3.44?10?2 m/s; d is the diameter of working electrode, 2.0?10?3 m and ? is the
viscosity of acetonitrile, 3.5?10?4 kg?m?1 ?s?1
39
Fast flow rates can substantially decrease the residence times of the analyte in
the electrospray, and reduce unwanted electrochemical reactions at the electrospray
needle. [5] Fast flow rates, however, can also cause an increase in the charged droplet
sizes in the electrospray, therefore affecting the transfer of solvated ions from the
liquid phase to the gas phase. To maintain a high conversion efficiency, the source
temperature must be adjusted to match the operating flow rate. In this dissertation,
various flow rates, ranging from 1 - 5 mL/hr, were used to meet the requirements of
the various electrochemical reactions studied.
For mathematical simplicity, the electrolyte flow in the cell is assumed to be a
fluid passing a sphere. [163] The electrolyte has the slowest flow rate when it is flowing
through the upstream part of the spherical electrode. As the velocity of electrolyte is
increased, the cross section area is decreased when solution passes the spherical thin-
layer channel. The velocity reaches a maximum of 3.44?10?2 m/s at the entrance
of the silica capillary where the smallest cross section area, (7.85 ? 10?4 mm2), is
encountered.
Based upon above analysis, the calculated Re number is less than 1, we, thus,
assume fully developed laminar flow [163] when the electrolyte is flowing through the
spherical thin-layer channel as shown in the top part of Figure 2.2.
If a fluid with a sufficient high flow rate passing through the cell, the Re number
is high (> 10), and then a weak turbulence could develop in the downstream part of
the working electrode, [163] shown in bottom of Figure 2.2. In this case, products
formed on the working electrode surface could react further on the electrode surface
due to the turbulence. Nevertheless, a laminar flow is developed at a flow rate of 1
mL/hr, which is the most often used flow rate in this dissertation.
40
  Au/Pt
Silica Capillary 
to Espray Probe
Silica Capillary
to Espray Probe
 Au/Pt
Figure 2.2: Flow Pattern Around the Working Electrode
41
2.4 EC/ES-MS Experimental Conditions
Solvents Used. In classical electrochemical measurements, the choice of solvent
is determined by the electrochemical system to be studied. Non-aqueous solvents usu-
ally offer a wider usable potential range. These solvents include the aprotic solvents,
such as: dimethylformamide, dimethyl sulfoxide, dichlorobenzene, trichloromethane,
propylene carbonate and acetonitrile. [43] These solvents have the tendency of sim-
plifying the electrochemical reactions and they have the ability to stabilize radicals
or other intermediates produced during electron transfer process. [43] When choosing
a suitable solvent, some important criteria, such as the voltage limits, dielectric con-
stant, dipole moment, protic character, and solubility of reactants and product must
be considered.
The most important of these properties are the voltage limits and dielectric con-
stant of the solvent. The voltage limits define the accessible potential range available
for the electrochemical oxidation or reduction process. A large dielectric constant, in
general, promotes the dissociation of ionic solutes, thus decreasing the cell resistance.
Since electrolytic conductivity is always necessary, a dielectric constant value larger
than 10 is usually necessary to avoid experimental difficulties. [43] A low cell resistance
minimizes the iR drop resulting in a better electrode potential measurements.
EC/ES-MS, however, imposes more restrictions on the solvents than conventional
electrochemical experiments. Of most concern is how the solvent used in the electro-
chemical system affects the electrospray process, such as the formation of charged-
droplets and the transfer of solvated ions into gas phase. Surface tension is also an
important factor because it determines the charged droplet sizes that remain near the
Rayleigh instability range, equation (1-1).
42
Experimental results show that the electrospray current is a function of the
conductivity ? of the solution, equation (2-2). [6]
I = H?n (2.2)
where: n = 0.2?0.4, ? is the conductivity of solution and H is a constant.
The dielectric constant is not directly related to the ion transfer process from the
solution phase to the gas phase, but it can affect the conductivity of the solution
which is related to the formation of charged-droplets. Other physical properties of
the solvents, such as the boiling point, are not directly related to the transfer of ions
from the charged-droplets to the gas phase, but a solution with high boiling point
could affect the sizes of charged-droplets at the operating source temperature.
In this dissertation, the most often used solvents are acetonitrile and acetoni-
trile/toluene mixtures. Acetonitrile is a relatively inert solvent, and is resistant to
both oxidation and reduction. It, also, is an excellent solvent for many compounds.
Its dielectric constant of 37 allows good conductivity, minimizing the iR drop in the
electrochemical cells. [43] Standard procedures are available to easily dry and purify
acetonitrile and dissolved oxygen can easily be removed by simply bubbling a dry
clean inert gas, such as Ar, N2, through the solvent. [164]
Supporting Electrolytes. Since most organic solvents used for studying elec-
trochemical reactions are nonconductors, it is necessary to add inert soluble support-
ing electrolytes to the solutions. A concentration of 0.1 M is often used. [43] There
are several ways in which the supporting electrolyte affects the solvent medium. The
primary function is to regulate the cell resistance and mass transport by electrical
migration. It also may determine the structure of the double layer, and be involved in
43
the formation of ion-pairs with the electroactive species. The supporting electrolyte
may also limit the potential window due to its redox properties. The most often used
supporting electrolytes in aprotic solvents are the tetraalkylammonium salts such as,
[Et4N]ClO4, [Et4N]BF4, [Et4N]ClO4, [n?Bu4N]BF4. [42,43,165,166]
A high concentration of supporting electrolyte must, however, be avoided when
studying an electrochemical reaction using the EC/ES-MS technique. The tetraalky-
lammonium salts are soluble in many aprotic solvents, and are resistant to oxidation
and reduction, but these ions usually have high surface activities compared with the
analytes. Hence, they tend to form clusters in gas phase during the electrospray
process. Formation of clusters suppresses and changes mass spectrometric signals.
In addition, a solution with a high concentration of supporting electrolyte can act
as a conductor between the electrochemical cell and the high potential electrospray
needle. This causes potential voltage floating of the electrochemical cell as well as
significant current leakage from the electrospray needle. Another reason for avoid-
ing a high concentration of supporting electrolyte in EC/ES-MS is because a high
concentration of supporting electrolyte can significantly suppress mass spectrometric
signals of the analytes. [6]
2.5 Off-line Electrochemical Characterization of the Cell
The performance of the spherical thin-layer flow cell was characterized by off-line
static and hydrodynamic cyclic voltammograms of the rapid one electron reversible
reaction, [Fe(CN)6]3?/4?, using solvents and electrolytes applicable to EC/ES-MS.
Solvents and Chemicals. Electrochemical grade potassium hexacyanoferrate
(K3[Fe(CN)6],99%) was purchased from Fluka (St. Louis, MO), and potassium chlo-
ride (KCl) was obtained from Aldrich Chem. Co. All chemicals were used as received.
44
Aqueous solutions of 1 mM K3[Fe(CN)6] were prepared in 0.1 M KCl. Deionized water
with a measured resistance of 18.6 M? was used to make all the aqueous solutions.
Electrochemical Methods. A Model BAS Epslion-EC potentiostat with data
system version 1.31.65NT and firmware version 1.30a (Bioanalytical Systems Inc.)
were used for off-line EC/ES-MS experiments. A platinum-mesh electrode was used
as the counter electrode. A Ag/AgCl reference electrode was connected to the cell
through a double salt bridge. All potentials are quoted versus aqueous Ag/AgCl and
all the measurements were done twice. Origin version 7.0 (Microcal Software, Inc.)
was used to plot the Nernstian curves and linear-fits.
Measurements of the Gold Electrode Surface Areas. In order to be qual-
itative about the electrochemical measurements, it is important to know the working
electrode surface areas. The gold working electrode surface areas were measured by
using an electrochemical method, linear sweep voltammetry. [42] A fast one electron
reversible redox reaction, was used to produce cathodic and anodic peaks, equation
(2.3), [167].
[Fe(CN)6]3? + e? ?? [Fe(CN)6]4? (2.3)
Figure 2.3 shows a typical cyclic voltammogram of 1 mM [Fe(CN)6]3? at a gold
electrode in a 0.1 M KCl aqueous solution. The separation of anodic and cathodic
peaks is 65 mV (theoretical value is 59 mV). [42] The two peaks are symmetrical in
terms of their heights and shapes. It is evident that the redox reaction is a one-
45
-1.E-05
-8.E-06
-6.E-06
-4.E-06
-2.E-06
0.E+00
2.E-06
4.E-06
6.E-06
8.E-06
1.E-05
-0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6
Potential (V)
Cu
rre
nt 
(A
)
Figure 2.3: CyclicVoltammogramof [Fe(CN)6]3? in H2O at the Pt Electrode, V vs.
Ag/AgCl
46
electron reversible reaction under these experimental conditions. Therefore, the ca-
thodic or anodic peak height, ip, is linearly related to the square root of the scan
rates at the [Fe(CN)6]3? concentrations, equation (2.4). [167]
ip = (2.69?105)n32AD120 C0??12 (2.4)
In this equation, A is area in cm2, D0 is diffusion coefficient in cm2 ?s?1, C0 is con-
centration in mol?cm?3, and ? is scan rate in V?s?1, n is the number of electrons.
The diffusion coefficient (D0) is taken as 7.63 ? 10?6 cm2/s. [168]
With this equation, the working electrode surface area can be derived by plotting
the ip vs. ?1/2 as shown in Figure 2.4. The slope of the linear ip vs. ?1/2 dependence
in Figure 2.4 yields the working electrode surface area. In our experiments, the
calculated working electrode surface areas range from 0.124 to 0.264 cm2.
Results and Discussion. Static cyclic voltammograms of 1 mM [Fe(CN)6]3?/4?
at voltammetric scan rates of 1-4 mV/s are shown in Figure 2.5. Peak-shaped voltam-
mograms are observed at all scan rates. This indicates that rate of electron transfer
exceeds the rate of mass transfer on the working electrode surface under these ex-
perimental conditions, which is not expected at the slow scan rate of 1 mV/s. The
potential differences of the two peaks, however, are much greater than 59 mV, and
increase as a function of the scan rate, these result from the high cell resistance.
The well-defined anodic and cathodic peaks are observed even at a scan rate of
1 mV/s. This may be caused by the small distance between the spherical working
electrode and the ?cup? wall. Given a scan time on the order a hundred seconds, one
can estimate that the thickness of the diffusion layer is about several hundred
47
s50s48 s52s48 s54s48 s56s48 s49s48s48 s49s50s48 s49s52s48 s49s54s48 s49s56s48 s50s48s48
s56
s49s48
s49s50
s49s52
s49s54
s49s56
s50s48
s50s50
s48
s46
s48
s55
s52
s49
s65
s40s109s86s47s115s41
s49s47s50
Figure 2.4: Peak Current (ip) vs. Square Root of Scan Rates (?12)
48
-4.E-05
-3.E-05
-2.E-05
-1.E-05
0.E+00
1.E-05
2.E-05
3.E-05
4.E-05
-0.15 -0.05 0.05 0.15 0.25 0.35 0.45 0.55
Potential (V)
Curre
nt 
(A)
10uA
4: 4mV/s
3: 3mV/s
2: 2mV/s
1: 1mV/s
1
2
3
4
Figure 2.5: Static Cyclic Voltammograms of [Fe(CN)6]3? at Various Scan Rates, V
vs. Ag/AgCl
49
?m, [169] which could be greater than the size of spherical channel between the
electrode and the cell wall.
Figure 2.6 shows hydrodynamic cyclic voltammograms of 1 mM [Fe(CN)6]3?/4?
at a flow rate of 1 mL/hr. The scan rates range from 1 mV/s to 6 mV/s with a
step increase of 1 mV/s. When the flow rate is greater than 2 mL/s, the rate of
electron transfer rate is greater than the mass transfer rate, and thus, peak-shaped
hydrodynamic cyclic voltammograms are observed. At low scan rates, 1 mV/s and 2
mV/s, constant limiting currents were observed at high over-potentials. This indicates
that the electron transfer rate is controlled by the mass transport of analyte in the
cell.
In order to be more quantitative about the shapes and sizes of the hydrodynamic
cyclic voltammograms obtained at scan rates of 1 mV/s and 2 mV/s, a logarithmic
analysis of the current i relative to the potential (E) is carried out for laminar flow
developed at a flow rate of 1 mL/hr. The redox reaction of [Fe(CN)6]3?/4? is a
rapid one electron reversible reaction. This electron transfer process has a typical
Nernstian response. [170] With only [Fe(CN)6]3? initially present at the millimolar
level in solution, an equation in terms of electrode potential is written as follow: [170]
E = E1/2 + RTnF log iL ?ii (2.5)
Where iL is the limiting current and F is the Faraday constant 96485.3 C/mol.
When a system conforms to equation (2-5), a plot of E vs. log[(iL ?i)/i] will
give a straight line with a slope of 59.1/n mV (here n = 1) or 2.3RT/(nF). In our
experiments, the measured limiting currents at scan rates of 1 mV/s and 2 mV/s are,
respectively, 3.5846?10?5 A and 3.7482?10?5 A,
50
s48s46s50 s48s46s49 s48s46s48 s48s46s49 s48s46s50 s48s46s51 s48s46s52 s48s46s53 s48s46s54
s86
s49s109s86s47s115
s50s109s86s47s115
s51s109s86s47s115
s52s109s86s47s115
s53s109s86s47s115
s54s109s86s47s115
s50s48s32s117s65
Figure 2.6: Hydrodynamic Cyclic Voltammograms of [Fe(CN)6]3? at Various Scan
Rates, V vs. Ag/AgCl
51
Logarithmic plots of E vs. log[(iL ?i)/i] in the low over potential range are
shown in Figure 2.7. The data corresponds to a scan rate of 1 mV/s. Linear fits of
the curve give slopes ranging from 63.31 mV to 68.8 mV (theoretical 59.1 mV) This
indicates that Nernstian behavior of [Fe(CN)6]3?/4? occurs at the working electrodes
at low-over potentials.
The cell conversion efficiency is defined as the ratio of the number of moles of
analyte reacted on the working electrode over the number of moles of analyte flowing
through the cell per unit time. It is very important factor to determine the quality
of the design of a flow cell.
The amount of analyte, [Fe(CN)6]3?, pumping through the flow cell at the flow
rate of 1.0 mL/hr, which is the flow rate that is mostly used in this dissertation, is
2.778 ? 10?10 mol/s. At the steady state region, the constant limiting current is
35.85 ?A and 37.48 ?A at the flow rates of 1 mV/s and 2 mV/s, respectively. The
amount of electron transferred at the working electrode per unit time is calculated as
3.716 ? 10?10 mol/s. The calculated results show that number of moles electrons
transferred per unit time is slightly greater than the number of moles of analytes
flowing through the cell per unit time. This indicates that the flow cell has high
electrochemical conversion efficiency as well as that some analyte could diffuse to the
back of the ball-shaped working electrode, shown in Figure 2.2, and thus participate
the electrochemical reactions.
52
s48s46s54 s48s46s56 s49s46s48 s49s46s50 s49s46s52 s49s46s54 s49s46s56 s50s46s48 s50s46s50
s48s46s50s50
s48s46s50s52
s48s46s50s54
s48s46s50s56
s48s46s51s48
s48s46s51s50
s48s46s51s52
s48s46s51s54
s69s32s40s86s41
s108s111s103
s49s48
s40s32s40s105
s76
s32s45s32s105s41s32s47s32s105s32s41
s32s69s32s118s115s46s32s108s111s103
s49s48
s40s32s40s105
s76
s32s45s32s105s41s32s47s32s105s32s41s44s32s115s99s97s110s32s114s97s116s101s58s32s49s109s86s47s115
s32s108s105s110s101s97s114s32s102s105s116s58s32s97s32s115s108s111s112s101s32s111s102s32s54s51s46s51s49s32s116s111s32s54s56s46s56s49s32s109s86s32
Figure 2.7: Nernstian Plot for the Scan Rate of 1 mV/s
53
2.6 Conclusions
A spherical thin-layer flow electrochemical cell, characterized by easy access for
replacing the working electrode and a high conversion efficiency, was developed. The
cell has a relatively small cell volume, 5 - 10 ?L. Calculated Re numbers predict that
the flow pattern around the working electrode is Laminar flow. Both the static cyclic
voltammograms and hydrodynamic voltammograms at different flow rates indicate
that the cell design meets the requirements of a thin layer flow cell.
The drawbacks and the advantages of the cell design are summarized as follow:
Drawbacks:
? The cell design sacrifices the separation of counter and working electrode in
order to achieve a small dead volume.
? The cell design consists of too many hand-made components, which makes dif-
ficult to assembly.
? The counter electrode was placed the upstream of the working electrode, which
may cause an uneven cell current density distribution.
Advantages:
? The cell design has easy access for cleaning the working electrode.
? The cell has a high conversion efficiency.
? A well-made ?cup? volume is about 5 - 10 ?L and functions as a collector.
? At low flow rates, between 1 - 5 mL/hr the fluid around the working electrode
has well-developed Laminar flow.
54
? Counter electrode reactions are separated from the working electrode reactions
by placing the counter electrode upstream of the working electrode.
The cell just described was used in the experiments to be described in Chapter
3 and 4. A 800 mm silica capillary was used in the on-line EC/ES-MS experiments
to connect the electrochemical cell to the electrospray probe. This silica tube isolates
the high potential at the tip preventing current leakage from the electrospray needle.
55
Chapter 3
I? and CN? at Gold and Platinum Electrodes
Abstract
We report results on the electrochemistry of I? and CN? at gold and platinum
electrodes using an electrochemical flow cell described in Chapter 2 coupled to an
electrospray mass spectrometer. We demonstrate that our apparatus is capable of
these very challenging electrochemical/electrospray experiments and that B(C6H5)?4
is a suitable internal standard for negative-ion studies in acetonitrile. With I? at
a platinum electrode, we observe well-behaved oxidation to I?3 . Experiments on I?
at gold electrodes are more complex, showing AuI?2 as well as I?3 . The AuI?2 mass
spectrometric ion intensity varies in a complex way throughout the applied electro-
chemical voltage range studied; we propose that this variation involves the adsorption
of I? on the gold electrode surface. In experiments on CN? at the gold electrode,
we observe Au(CN)?2 at 0 V. This arises from the surface oxidation of the freshly
annealed gold electrode. In the platinum electrode case, the oxidation of cyanide
yields cyanogen. Addition of one mole of CN? to two moles of C2N2 gives the C5N?5 .
3.1 Introduction
We chose to study the iodide system because it undergoes a well-understood
multi-step oxidation, and the iodide absorption onto electrode surfaces is well under-
stood. [7,8] Studies of the electrochemical oxidation of iodide to iodine in different
solvents at platinum electrodes have focused on the kinetics and mechanisms of the
reactions and the adsorption of ions on the electrode surfaces. [7?13,15] Agreement
56
between experimental and theoretical polarographic characteristics of the oxidation
of iodide in acetonitrile was reported by Guidelli and Piccaradi, [12] while the elec-
trochemical properties of iodide in different organic solvents [10] with a wide range
of dielectric constants were reported by Iwamoto. A two-step oxidation of iodide to
iodine was proposed. The kinetics and mechanism of the electrochemical oxidation
of iodide to iodine have also been investigated in acetonitrile and dimethylsulfoxide
with a rotating platinum-disk electrode by Arvia. [7,8] The results showed two polaro-
graphic waves, and the difference between the half-wave potentials of the two waves
was related to the stability constant of tri-iodide in the solvent. The adsorption of
iodide on a smooth electrode in aqueous solutions has been investigated by Bagotskv,
et al. [11] It was found that the adsorption formed a stable adlayer that dramatically
influenced the surface reactivity.
Numerous studies have focused on the electrochemistry of gold and its halide
complexes in both aqueous and non-aqueous solutions. [16?22] Metallic gold can be
dissolved as gold complexes in acetonitrile with strongly coordinating ligands by both
chemical or electrochemical oxidation. [23] The stepwise stability constants for AuI
and AuI?2 complexes in acetonitrile, determined by Fenske, are given below. [24] These
stability constants indicate that AuI and AuI?2 are stable in acetonitrile.
K1 = AuI[Au+][I?] logK1 = 17.1 (3.1)
K2 = [AuI
?
2 ]
[AuI][I?] logK2 = 6.7 (3.2)
In acetonitrile, halide ions have been found to be adsorbed on smooth gold sur-
faces, [11,15] especially when a positive potential is applied to the gold. The halide
57
adlayer plays an important role in subsequent electrochemical reactions. It has been
reported that halide adsorption on the gold electrodes shifts the oxidation potential
for the oxidation of iodide to iodine to more negative values. [24] Scanning tunneling
microscopy images of iodide adsorbed on gold(111) surfaces from aqueous solution
have been reported by Bard. [18] The atomic images of the iodide adlayers provide
direct evidence for the organization of the adlayer.
The chemistry of gold cyanide complexes has been known since the 17th century.
The treatment of heterogeneous gold and electrolyte (cyanide, halide) interface as
a model system for studying other interfacial processes has been a subject of many
investigations.
Previous results, obtained using surface structure sensitive techniques, [25,26],
show that the process of gold dissolution first starts from the formation of monolayer
of AuCN on the gold electrode in the presence of O2.
The adsorption of CN? ions on gold single crystals was monitored by visible-
infrared difference frequency generation (DFG). [26] The results showed a significant
effect of the surface orientation on the adsorption strength of cyanide. The interface
between polycrystalline gold electrodes and the cyanide aqueous solution as a function
of the potential and cyanide concentration was studied by polarization-modulated
Fourier transform infrared reflection absorption spectroscopy (FT-IRRAS). [25]
In situ atomic scale scanning tunneling microscopy (STM) images of gold(111)
films undergoing dissolution in aqueous cyanide solutions have been reported. [183]
The STM images provided direct evidence that the complete disruption of gold(111)
surface took place in less than 1 min at high concentrations of cyanide ( > 10?4 M)
in the presence of oxygen.
58
Electrochemical oxidation of cyanide at platinum electrodes has been studied
by various electrochemical methods in aqueous solution. [172] The results suggest
that the cyanide oxidation is a non-reversible one-electron transfer process that is
characterized by diffusion control at the platinum electrode. The oxidation product
is cyanogen.
Cyanogen is an active species and can be involved in many reactions in both
aqueous and non-aqueous solutions. The reaction of cyanogen with cyanide is one
of the least studied areas of cyanogen chemistry. [27] A reversible reaction between
cyanogen and cyanide yields C3N?3 , but, to date, there is no direct evidence to prove
the exitance of this species in both aqueous and non-aqueous solutions. The proposed
reactive anion C3N?3 is an important intermediate in the process of formation of many
cyano-carbons. The structure of C3N?3 was studied by using the MNDO molecular
orbital method, [96] the result shows that C2v is the most stable structure. Addition
of two cyanogens to cyanide ion yields C5N?5 in a step-wise mechanism. [28]
In this chapter, we present our results of electrochemical studies of iodide and
cyanide oxidations at gold and platinum electrodes using the EC/ES-MS method.
3.2 Experimental Section
Experimental Chemicals and Solutions. Potassium iodide (KI, 99.0+%),
sodium tetraphenylborate (Na[B(C6H5)4] ? 99.5%), potassium cyanide (KCN, 99%),
potassium chloride (KCl, 99.8%), and acetonitrile (CH3CN, HPLC grade) were ob-
tained from Fisher. Tetrabutylammonium cyanide ([Bu4N]CN > 95%) was from
Aldrich. All salts were used as received.
A 0.1 mM K3[Fe(CN)6] in 0.1 M KCl aqueous solution was used to measure
the areas of the working electrodes with K3[Fe(CN)6] and KCl acting respectively
59
as probe molecules and supporting electrolyte. [43] Acetonitrile was dried over CaH2
under N2 gas for at least 4 hours prior to use. [164] The solutions were thoroughly
degassed with high-purity dried argon (Grade 5.0) for at least 10 min before starting
each experiment. [Bu4N]CN was dried at least 48 hours under vacuum at 30 ?C before
use.
Instrumentation and Electrodes. A VG-Trios-2000 quadrupole mass spec-
trometer with VG electrospray source was used in these experiments. Here we report
results on anions; the mass spectra were acquired in negative-ion mode, and 50 scans
at a scan rate of 1 scan/s. The ES tip voltage was ? 2.89 kV, the source temperature
was 60 ?C, and the cone voltage was 20 V. Since the residence time of solutions in the
sampling area of the electrochemical cell is 18 - 30 seconds, the time delay between
two consecutive potentials was 60 seconds.
A Bioanalytical System Inc. CV-27 potentiostat was used to apply potentials
to the homemade three-electrode electrochemical cell, Figure 2.1, and to record the
cell current. The potential ranged from 0.0 to 1.9 V with potential steps of 0.1 V. A
Bioanalytical System Inc. Epsilon-EC potentiostat was used to obtain conventional
cyclic voltammograms for comparison purposes. Matlab version 6.5 was used to pro-
cess all mass spectral data and plot the 3-D figures. Origin version 7.0 and Microsoft
Excel 2000 was used to plot other figures.
Working Electrodes and the Electrochemical Setup. A schematic diagram
of the electrochemical working electrode and ion collection region is shown in Figure
2.1. The platinum-mesh was used as the counter electrode. A Ag/AgCl reference
electrode was connected to the cell through a double salt bridge. All potentials are
quoted versus aqueous Ag/AgCl. The connections to the cell were made through
Viton O-rings or through Supelco Thermogreen LB-2 septa. A Cole Palmer Series
60
7490 syringe pump with a flow rate of 1.0 mL/hr was used to provide the constant
mobile phase and produce the spray.
It was found that current leakage could be avoided by using an 800 - 850 mm
long (ID 0.1 mm) untreated silica capillary connecting the electrochemical cell with
the electrospray probe tip. The entire electrochemical cell was maintained at room
temperature.
The organic stream in the cell and the duct line connecting to the ESI source
fulfills several functions: it minimizes the liquid conductivity in the transfer line from
the flow cell to the ESI probe due to its small cross section area (0.785 mm2) and
relatively long distance. Therefore it disconnects the high voltage of the ESI needle
and the working electrode. An organic solute and low concentration of electrolytes
also quenched chemical reactions between the oxidation products and the reactants,
thus reducing compositional changes in the duct line.
The working electrodes were made from gold (99.9999%) or platinum (99.9999%)
wires (OD 1 mm). Before use, the electrodes were cleaned by rinsing sequentially with
distilled water, hydrofluoric acid 49%, piranha solution, [161] distilled water, HPLC
grade methanol and fresh acetonitrile. The electrodes were then stored in acetonitrile
in a glove bag under an argon atmosphere. The assembly of the electrochemical cell
was carried out in a glove bag under Ar.
All the electrochemical measurements were done twice.
3.3 Results and Discussion
Internal Standard. High conductivity of the electrolytic solution is usually pre-
ferred in conventional electrochemical measurements to decrease the potential drop
61
between the working and reference electrodes. However in the EC/ES-MS experi-
ments, the conductivity of the electrolyte solution is kept low for two reasons.
A low solution conductivity minimizes interfering currents due to the high potentials
at the electrospray tip. Low electrolyte concentrations also avoid attenuation of
the electrospray signals. [6] An organic solute and low concentration of supporting
electrolyte also discourages the possible chemical reactions between the products with
the supporting electrolyte, thus reducing compositional changes in the silica capillary.
As stated before, the reason we choose the iodide system is that it undergoes a
well-understood multi-step oxidation, and iodide absorption onto the electrode sur-
faces. The formation of solvated K+ and I? ions is readily achieved in acetonitrile
because of its high dielectric constant (36). Therefore, the iR drop can be minimized
without using a high concentration of supporting electrolytes under the experimental
conditions. The internal standard NaB(C6H5)4, acting also as a supporting elec-
trolyte, in the acetonitrile solution was used at low concentrations (10?4 ? 10?5 M).
Different concentrations of NaB(C6H5)4 were evaluated and a concentration of 0.02
mM appeared optimum.
Figure 3.1 shows a three-dimensional plot of the absolute ion intensity versus
mass to charge ratio and applied electrochemical potential for 0.02 mM NaB(C6H5)4
in acetonitrile using a platinum electrode. Data were collected in 0.1 V increments
from 0.0 to +1.9 V. These data clearly show the intensity of the tetraphenylborate
anion, B(C6H5)?4 , does not change with applied potential. A similar result (not
shown) was found for B(C6H5)?4 at gold electrodes. These results demonstrate that
B(C6H5)?4 is a suitable internal standard for experiments at both platinum and gold
electrodes; hence all mass spectra were normalized to the B(C6H5)?4 intensity at m/z
319. Sodium tetraphenylborate also has the advantages of good solubility and good
62
127
319
381
451
0
0.5
1
1.5
1.9
2.0e8
m/zV
Absolute Intensity
Figure 3.1: Three Dimensional Plot of EC/ES-MS Mass Spectra B(C6H5)?4 as a
Function of Applied Electrochemical Cell Potential
63
voltage limits in acetonitrile, and the peak at m/z 319 lies in a convenient mass range.
The other advantage of using NaB(C6H5)4 as an internal standard is the ions did not
appear to suppress the electrospray signal when present in micromolar quantities in
the electrospray solution. [6,88]
3.3.1 Iodide at Platinum and Gold Electrodes
Platinum Working Electrodes. The cyclic voltammogram of a 0.5 mM KI
and 0.02 mM NaB(C6H5)4 solution in acetonitrile at scan rate 10 mV/s is shown in
Figure 3.2. Two anodic peaks and one cathodic peak are recorded in Figure 3.2. The
first and second anodic peaks correspond, respectively, to the oxidation of I? and I?3 ,
consistent with the literature. [7] The reactions are illustrated by following equations:
2I? ?? I2 + 2e? Wave I (3.3)
I2 + I? ?? I?3 (3.4)
2I?3 ?? 3I2 + 2e? Wave II (3.5)
I?3 + 2e? ?? 3I? (3.6)
Only one cathodic peak corresponding to the reduction of I?3 (reaction 3.6) is
observed, since in acetonitrile the equilibrium constant for reaction (3.4) is ? 4?106
at 20 ?C. [7] Thus, most I2 is rapidly removed by reaction (3.4) and it is not available
for re-oxidation in the reverse of reaction (3.3) or (3.5). This result can be contrasted
with aqueous results where the equilibrium constant is smaller (K = 768) and one
anodic peak and two cathodic peaks are observed. [7]
64
-2.40E-05
-1.90E-05
-1.40E-05
-9.00E-06
-4.00E-06
1.00E-06
6.00E-06
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00
Potential (V)
Cu
rre
nt 
(A
)

?
?

?
Figure 3.2: Cyclic Voltammetry of I? at the Platinum Electrode, Scan Rate: 10
mV/s, V vs. Ag/AgCl
65
Figure 3.3 shows the hydrodynamic cyclic voltammogram of a 0.5 mM KI and
0.02 mM NaB(C6H5)4 solution in acetonitrile. The scan rate is 1 mV/s, and the
flow rate is 1 mL/hr. In contrast to the static cyclic voltammetry (Figure 3.2) where
two anodic peaks, assigned to oxidation of I? and I?3 respectively, are observed, the
hydrodynamic cyclic voltammetry (Figure 3.3) has only one clear anodic peak which
can be assigned to the oxidation of I?. No clear peak is observed for the oxidation
of I?3 in Figure 3.3. When the potential is scanned in the positive direction, the
anodic current increases with increasing potentials, Figure 3.3. This is because the
flow solution in the cell can bring the analyte, I?, to the working electrode surface
and remove the product, I?3 . A steady state current is observed at potentials greater
than 1.6 V. This indicates that convection dominates at high potentials range (1.6 -
1.9 V), rather than an electron transfer or a diffusion.
Figure 3.4 shows the results of a normalized EC/ES-MS experiment of the oxi-
dation of KI in acetonitrile at a platinum electrode at potentials from 0.0 V to +1.9
V in 0.1 V increments. The peak at m/z 381 corresponds to I?3 and first appears at
? +0.3 V; it increases slowly until ? +1.6 V, beyond which it increases more rapidly
and then plateau. This is in agreement with the pattern recorded by hydrodynamic
cyclic voltammetry, Figure 3.3, showing a plateau when the potential is greater than
1.6 V. The peak at m/z 319 (B(C6H5)?4 ) appear relatively constant throughout the
voltage range. The small peak at m/z 293 is due to KI?2 , which is believed to be
the cluster of [(KI)?I]?. This peak, originating from I?, is also relatively constant
over the entire voltage range. These results clearly demonstrate the formation of I?3
resulting from the oxidation of I?.
ThedatafromFigure3.4arealsoshowninFigure3.5wheretheabsoluteintensity
of the (B(C6H5)?4 ), I? and I?3 are plotted versus the applied electrochemical potential.
66
-2.80E-06
-8.00E-07
1.20E-06
3.20E-06
5.20E-06
7.20E-06
9.20E-06
1.12E-05
1.32E-05
1.52E-05
1.72E-05
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00
Potential (V)
Curren
t (A)
Figure 3.3: Hydrodynamic Cyclic Voltammetry of I? at the Platinum Electrode, Flow
Rate: 1 mL/hr, Scan Rate: 1 mV/s, V vs. Ag/AgCl
67
127
293319
381
451
0
0.5
1
1.5
1.9
1
5
9
m/zV
Relative Intensity
Figure 3.4: Three Dimensional Plot of EC/ES-MS Mass Spectra I? at the Platinum
Electrode as a Function of Applied Electrochemical Cell Potential
68
s48s46s48 s48s46s50 s48s46s52 s48s46s54 s48s46s56 s49s46s48 s49s46s50 s49s46s52 s49s46s54 s49s46s56 s50s46s48
s48s46s48
s53s46s48s120s49s48
s55
s49s46s48s120s49s48
s56
s49s46s53s120s49s48
s56
s50s46s48s120s49s48
s56
s50s46s53s120s49s48
s56
s65
s98
s115
s111
s108
s117
s116
s101
s32
s73
s110
s116
s101
s110
s115
s105
s116
s121
s80s111s116s101s110s116s105s97s108s32s40s86s41
s32s73
s45
s32s66s40s67
s54
s72
s53
s41
s52
s45
s32s73
s51
s45
Figure 3.5: Absolute Ion Intensity vs. Potential, I? at the Platinum Electrode
69
This figure shows an important trend; the absolute intensities of (B(C6H5)?4 ) and
I? decrease as the potential is increased and I?3 is formed. The decreasing intensity
of the I? could be interpreted as a decrease resulting from the conversion to I?3 .
However, such an explanation would not account for the steady decrease in the
B(C6H5)?4 intensity which is constant in the control experiment. A more plausible
interpretation is that there is an increased competition for the overall charge as the
electrospray droplets form ions. Hence, the steady decrease in the (B(C6H5)?4 ) in-
tensity and the I? intensity as I?3 is formed may be the result of the electrospray
ionization process and not an electrochemical phenomena.
The increase in the I?3 current is rapid from +0.3 V to +0.5 V and slower there-
after. The slower increase beyond +0.5 V is most likely due to increased mass transfer
from increasing migration of I? towards the electrode. Migrational effects are likely
due to the low electrolyte concentration and can be substantial. [173]
The decreased concentration of I? and increased concentration of I?3 would result
in a smaller net mass transfer due to the slower diffusion coefficient of I?3 [D(I?),
1.68?10?5 cm2/s; D(I?3 ), 1.55?10?5 cm2/s]. [7]
The steady-state current-voltage curve for I? at a platinum electrode, which is
plotted with the hydrodynamic cyclic voltammogram, is given in Figure 3.6. This data
was collected simultaneously with the spectra shown in Figure 3.4. The figure shows
that the current rises from baseline to a mass-transfer-controlled limit at ? 0.9 V.
Note also that cell current drops slightly in the mass-transfer-controlled region. The
decrease in current occurs at ? 1.3 V and is accompanied by the increased intensity
of tri-iodide as shown in Figures 3.4 and 3.5. The steady-state current-voltage curve
is consistent with the hydrodynamic cyclic voltammogram.
70
s48s46s48 s48s46s50 s48s46s52 s48s46s54 s48s46s56 s49s46s48 s49s46s50 s49s46s52 s49s46s54 s49s46s56 s50s46s48
s48
s50
s52
s54
s56
s49s48
s49s50
s49s52
s49s54
s67
s117
s114
s114
s101
s110
s116
s32
s40
s117
s65
s41
s80s111s116s101s110s116s105s97s108s32s40s86s41
s32s72s121s100s114s111s100s121s110s97s109s105s99s32s67s86
s32s67s117s114s114s101s110s116s45s80s111s116s101s110s116s105s97s108s32
Figure 3.6: Hydrodynamic Cyclic Voltammetry and Current-Potential of I? at the
Platinum Electrode, V vs. Ag/AgCl
71
The concentration changes are confirmed by a color change of the solution around
the working electrode due to the characteristic brown color of I?3 . The current-
potential curve in Figure 3.2 shows two waves at ? +0.4 V and ? +1.0 V. These are
in agreement with the reported current-potential curve of I? in acetonitrile solution
studied using a rotating-platinum-disk electrode. [7]
Gold Working Electrode. The cyclic voltammogram of KI in acetonitrile at
gold electrodes is shown in Figure 3.7. The cyclic voltammogram shows a similarity
to the voltammogram at platinum electrodes, two anodic peaks, corresponding to
(3.3) and (3.5), and one cathodic peak corresponding to reaction (3.6). However, the
EC/ES-MS spectra of KI at gold electrodes in acetonitrile, (shown in Figure 3.8) is
quite different.
The difference between Figures 3.4 and Figure 3.8 is clearly evident. As seen in
Figure 3.8, the mass peak that changes with potential is AuI?2 m/z 451 at ? +0.3 V.
This peak increases with increasing potential and then starts to decrease at ? +0.6
V at which potential m/z 381 (I?3 ) begins to increase. The intensity of the peak at
m/z 381 reaches a maximum at a potential of ? +1.7 V where it starts to gradually
decrease. This decrease is also seen in the I? peak at m/z 127. Finally, the peak at
m/z 451, corresponding to AuI?2 , starts to increase in intensity at ? +1.7 V. These
data indicate that I?3 accumulates around the gold electrode in the potential range
of +0.8 to +1.5 V. At less positive potentials AuI?2 is formed, and at more positive
potentials, AuI?2 is again formed.
The data from Figure 3.8 are also shown in Figure 3.9 where the absolute intensi-
ties of B(C6H5)?4 , I?, I?3 and AuI?2 ions are plotted versus the applied electrochemical
potential. A slightly decrease of the absolute intensities of B(C6H5)?4 versus applied
potential was also recorded in Figure 3.8.
72
-2.5E-05
-2.0E-05
-1.5E-05
-1.0E-05
-5.0E-06
0.0E+00
5.0E-06
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00
Potential (V)
Curr
en
t (A)
Figure 3.7: Cyclic Voltammetry of I? at the Gold Electrode, V vs. Ag/AgCl
73
127
319
381
451
0.5
1
1.5
1.9
1
10
18
m/z
V
Relative Intensity
Figure 3.8: Three Dimensional Plot of EC/ES-MS Mass Spectra I? at the Gold
Electrode as a Function of Applied Electrochemical Cell Potential
74
s48s46s48 s48s46s50 s48s46s52 s48s46s54 s48s46s56 s49s46s48 s49s46s50 s49s46s52 s49s46s54 s49s46s56 s50s46s48
s48s46s48
s53s46s48s120s49s48
s55
s49s46s48s120s49s48
s56
s49s46s53s120s49s48
s56
s50s46s48s120s49s48
s56
s50s46s53s120s49s48
s56
s51s46s48s120s49s48
s56
s51s46s53s120s49s48
s56
s52s46s48s120s49s48
s56
s52s46s53s120s49s48
s56
s53s46s48s120s49s48
s56
s53s46s53s120s49s48
s56
s65
s98
s115
s111
s108
s117
s116
s101
s32
s73
s110
s116
s101
s110
s115
s105
s116
s121
s80s111s116s101s110s116s105s97s108s32s40s86s41
s32s73
s45
s32s66s40s67
s54
s72
s53
s41
s52
s45
s32s73
s51
s45
s32s65s117s40s73s41
s50
s45
Figure 3.9: Absolute Ion Intensities vs. Potential, I? at the Gold Electrode
75
The absolute intensities of B(C6H5)?4 decrease as the potential is increased and
I?3 and AuI?2 are formed. As stated in platinum electrode case, these could be in-
terpreted as increased competition for the overall charge as the electrospray droplets
form ions. The observation of the same steady decrease of the B(C6H5)?4 intensity at
gold electrode proves that this is the result of the electrospray ionization process and
not an electrochemical phenomena.
The current-potential curve of I? at the gold electrode is also shown in Figure
3.10. Clearly, the current-potential curves at platinum and gold electrodes are very
similar from 0 V to about 1.4 V. However, above 1.4 V the current at the gold
electrode begins to rise again; this is in the same region where the AuI?2 reappears,
as seen in Figure 3.8.
There are three waves in the current-potential curve at the gold electrode. These
correspond to:
Au+ I? ?? Au(s)?I?(adsorbed) (3.7)
Au(s)?I?(adsorbed) ?? AuI(s) +e? (3.8)
AuI(s) + I? ?? AuI?2 (3.9)
2I? ?? I2 +2e? (3.10)
I2 + I? ?? I?3 (3.11)
Au ?? Au+ + e? (3.12)
Au+ + 2I? ?? AuI?2 (3.13)
76
A plausible interpretation for our observation of AuI?2 in acetonitrile at m/z 451
at low potentials is shown in equations (3.8) and (3.9). Equation (3.8) describes
the surface oxidation of the gold iodide adlayer to form AuI on the electrode surface.
That reaction is followed by the dissolution of the AuI as AuI?2 (equation 3.9) driven
by the high stability constant of AuI?2 . A similar mechanism of gold oxidation with
I? in water with (0.1%) acetonitrile was proposed by Kissner. [23]
The standard gold electrode potential (E?Au/Au+) vs. NHE at 20?C in acetonitrile
is 1.511 V (1.312 V vs. Ag/AgCl), [20] so it is not surprising to see that the AuI?2 at
m/z 451 increases at about +1.3 V in Figure 3.8. At high potentials gold atoms are
electrochemically oxidized to gold(I), following equations (3.11) and (3.12) above.
As previously stated, the current-potential curves of I? at platinum and gold
are different (shown in Figure 3.10). The gold electrode current does not reach a
diffusion-limited region at the most positive potentials. This can be explained by
the oxidation of gold to Au+ that is kinetically limited by the surface area of the
electrode and not solution diffusion-limited. A second point is at lower potentials,
the current-potential curves do not differ appreciably. Thus, the limiting current due
to Au-I(adsorbed) oxidation must be small relative to I? and I?3 oxidation.
The other question that we need to address here is the observation of different
absolute intensities of I? between platinum and gold electrodes, shown in Figure 3.4
and 3.8.
The absolute intensity of I? is relatively constant at all applied potentials at
the platinum electrode, shown in Figure 3.4. In the gold electrode case, the absolute
intensity of I? is constant when the surface oxidation of gold electrode is occurring, it
increases when the formation of I?3 (oxidation of I?) at gold electrode surface begins at
77
s48s46s48 s48s46s50 s48s46s52 s48s46s54 s48s46s56 s49s46s48 s49s46s50 s49s46s52 s49s46s54 s49s46s56 s50s46s48
s48
s50
s52
s54
s56
s49s48
s49s50
s49s52
s49s54
s49s56
s67
s117
s114
s114
s101
s110
s116
s32
s40
s117
s65
s41
s80s111s116s101s110s116s105s97s108s32s40s86s41
s32s65s117s32s101s108s101s99s116s114s111s100s101
s32s80s116s32s101s108s101s99s116s114s111s100s101
Figure 3.10: Current-Potential of I? at the Gold and Platinum Electrodes
78
potentials between 0.9 - 1.4 V. The intensity decreases again when the AuI?2 dominates
the spectra, Figure 3.8.
It is easy to understand that the absolute intensity of the I? is constant at
potentials between 0 and 0.7 V. This is because the surface oxidation of a freshly
annealed gold surface has a diffusion-controlled rate determining step. The reaction
rate is limited by the rate of AuI?2 diffusion away the electrode or the I? diffusing into
the electrode double layer, [167] so the absolute intensity of I? is relatively constant
during these processes. [174]
The absolute intensities of I? begins to differ between platinum and gold elec-
trodes at potentials between 0.9 - 1.4 V. This is not expected under the same experi-
mental conditions, of which only oxidation of I? at both gold and platinum electrode
surface is occurring in this potential range, shown in Figures 3.4 and 3.8. The differ-
ence between the trends might indicate a different rate-determining step in the same
electrochemical reactions. [175]
Figures 3.5 and 3.9 show that initial absolute intensities of I? are approximately
equal. The later upward trend of absolute intensities of I? at a gold electrode at poten-
tial ranging from 0.9 - 1.4 V could indicate a slow electron-transfer process dominates
the heterogenous interface of gold-electrolyte. This causes I? to accumulate around
the working electrode due to the high stability constant of I?3 in acetonitrile, and
the relatively long residue time (18 s) of solution in the ?cup? collector at potentials
between 0.9 - 1.5 V.
The experiments on iodide at gold and platinum electrodes may be summarized
as follows. With gold electrodes, oxidation of gold in the presence of I? takes place at
low potentials as a surface oxidation process. This is followed by oxidation of I? from
solution at higher potentials (this does not involve adsorption). At high potentials
79
gold atoms are electrochemically oxidized to gold(I), following equations (3.13) and
(3.14) above. In contrast, at the platinum electrode, the I? is adsorbed on the active
sites of platinum electrode, [11] but the oxidation of platinum requires much higher
potentials so I? is oxidized to I?3 .
3.3.2 Cyanide at Platinum and Gold Electrodes
Platinum Working Electrode. Cyanide online oxidation was carried out in
acetonitrile solutions at different concentrations of [Bu4N]CN. The concentrations
ranged from 0.2 mM to 8 mM. Experimental results show that a 0.5 mM [Bu4N]CN
acetonitrile solution gives the expected oxidation products. Figure 3.11 shows the
results of a normalized EC/ES-MS experiment of the oxidation of CN? from 0.0 V
to +2.1 V in 0.1 V increments. All peaks are normalized to cyanide at m/z 26.
The peaks at m/z 26, 41 and 87 correspond to CN?, CH3CN? and BF?4 , respec-
tively. A tiny peak at m/z 59 can be assigned to [CH3CN?H2O]? possibly arising
from the presence of trace amounts H2O in the acetonitrile. It is not surprising to
see these clusters in the spectra since electrospray often results in clustering. It is
important to note that these peaks are present over the whole potential range and
appear relatively constant. Two voltage dependent peaks are observed, one at m/z
130 and the other one at m/z 156. The peak at m/z 130 first appears at 1.6 V; it
increases slowly until 2.1 V, which was the highest potential applied. This peak can
be assigned to C5N?5 . The results shown in Figure 3.11 clearly demonstrate that this
peak is potential dependent. The mechanism for the formation of C5N?5 in acetonitrile
is shown in Figure 3.12. [28]
In the first step, the cyanogen is formed by oxidation of cyanide. [28] The addition
of cyanide with the cyanogen gives the unstable tri-cyanide (C3N?3 ) species. C3N?3
80
26 41
59
87
130
156
180 0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
22.1
0.5
1
V
m/z
Relative Abundance
130 
156 
Figure 3.11: Three Dimensional Plot of EC/ES-MS Mass Spectra of CN? at the
Platinum Electrode as a Function of Applied Electrochemical Cell Potentials
81
C
C N
C N
N
C2N2 C
N
N
C
C
C
C
N
N
C
N
C
C
C N
C
N
NN
C5N5-
CN- CN. + e
2CN. N C C N
N C C N + CN- N C
C
C
N
N
N
C3N3-
rearrangement
Figure 3.12: Schematic of the Formation of C5N?5 by Addition of C2N2 to CN?. [28]
82
has a stability constant of 0.31 in acetonitrile. [176] A further addition of cyanogen
by the tri-cyanide, followed by the rearrangement, shown in Figure 3.12, leads to the
formation of C5N?5 . [28] The observation of C5N?5 at high potential could indicate that
a relatively high concentration of cyanogen must be present resulting in the formation
of C5N?5 as showing in Figure 3.12. [28]
The second potential dependent peak at m/z 156 seen in Figure 3.11 is not
as intense as the peak at m/z 130, but it does grow as the potential is increased.
We suggest that this small peak may be assigned as C6N?6 or C12N2?12 , [177,178]
which is the trimer of cyanogen. The proposed structure of C6N?6 , shown in Figure
3.13, has been identified by X-ray in solid state. [177,178] Observation of negative
charged C6N6 in our experiment is interesting because C6N6 is a neutral species and
potentials applied at the platinum electrode are positive. This could indicate that
neutral C6N6 is formed at the platinum electrode and is reduced to negative C6N?6
in the ESI capillary. It has been shown that the C6N?6 is an unstable radical and
can immediately dimerize at a ring carbon to form [C12N12]2?. [178] Unfortunately
we were not able to prove this peak by the isotope pattern because of the weak signal
and low resolving power of the single quadrupole mass spectrometer.
Comparing these CN? results with those of iodide at platinum, no peak corre-
sponding to platinum cyanide complexes is observed at any potential studied. Figure
3.14 shows a mass spectrum at 2.0 V and shows the absence of a peak at m/z 247.
The absence of a complex could be as the result of the low concentration of CN?. [172]
Sawyer, et al. [172] reported that the cyanide oxidation can be inhibited by increasing
the concentration of cyanide, this is because the platinum electrode is oxidized rather
than cyanide at high potential.
83
C
N N
C C
N
CC N
C
N
NC NCN3
C
N N
C C
N
CC N
C
N
N
C
N N
C C
N
CC N
C
N
Ne
TCT
C
NN
CC
N
C
C
N
C
N
N
C
NN
CC
N
C C
N
C
N
N
TCT TCT
electrospray
+
TCT2
2-
Figure 3.13: Schematic of the Formation of C6N?6 and C12N2?12 During the Electrospray
Process. [177,178]
84
0.50
50.50
100.50
150.50
200.50
250.50
10 60 110 160 210 260
Mi
llio
ns
m/z
Int
en
sit
y
26
41 87 130 156
59
Figure 3.14: Mass Spectrum of the [TBA]CN oxidation at 2.0 V
85
Gold Working Electrode. The molecular structures are similar for both
Au(CN)?2 and AuI?2 , consisting of linear two-coordinate complex ions. [179] Au(CN)?2
is among the most stable two-coordinate complexes of the transition metal ions, with
the stability constant 1037 M?2 in aqueous solution. [180,181]
The following trend of stability constants of gold(I) complexes in acetonitrile was
reported. [24]
NCO? < NCS? < Cl? < Br? < I? lessmuch CN?
The above given order (I? lessmuch CN?) indicates that Au(CN)?2 is a very stable
complex with a stability constant of 1035 M?2 in acetonitrile solution. [19]
In this section, we present further results dealing with the surface reaction of
a fresh annealed gold electrode in non-aqueous potassium cyanide solution by using
EC/ES-MS. Figure 3.15 shows the results of a normalized EC/ES-MS experiment
at gold electrodes in the presence of 1 mM potassium cyanide in acetonitrile. The
potential ranges from 0.0 V to -1.0 V. The peak at m/z 249 corresponds to Au(CN)?2
and first appears with the strongest peak at 0 V with the potentiostat connected to
the cell; it decreases slowly with decreasing potential until ? ?1.0 V. The peak at
m/z 319 assigned to B(C6H5)?4 appears relatively constant over the entire voltage
range. These results clearly demonstrate the formation of Au(CN)?2 resulting from
the surface oxidation of a freshly annealed gold electrode.
A plausible interpretation for our observation of Au(CN)?2 at 0 V at a gold
working electrode is shown in equations (3.17), (3.18) and (3.19). [182] Equation
(3.17) describes the surface oxidation of the gold cyanide adlayer to form AuCN on
the electrode surface. Reaction (3.17) is followed by the dissolution of the AuCN as
Au(CN)?2 (equation 3.18 and 3.19).
86
150
249
319
450
?1
?0.8
?0.6
?0.4
?0.2
?0.1
0
0
0.5
1
1.4
m/z
V
Relative Intensity
Figure 3.15: Three Dimensional Plot of EC/ES-MS Mass Spectra CN? at the Gold
Electrode as a Function of Applied Electrochemical Cell Potential
87
Au+ CN? ?? AuCN?(ads) (3.14)
AuCN?(ads) ?? AuCN(ads) + e? (3.15)
AuCN(ads) + CN? ?? Au(CN)?2 (3.16)
The dissolution is driven by the high stability constant of Au(CN)?2 . Even with
potentiostat set at 0 V, a current can flow through the cell. The trace amount of O2
and water in the acetonitrile could also make the dissolution of the surface gold take
place readily. [183] By applying more negative potentials at the working electrode,
the surface oxidation step (equation 3.18) slows down, causing the decrease in the
intensity of the peak at m/z 249.
3.4 Conclusions
The experimental results presented in this chapter very clearly demonstrate the
power of coupling electrochemical experiments to mass spectrometry. These chal-
lenging experiments can provide a wealth of detailed information on electrochemical
processes. We demonstrated that B(C6H5)?4 is a suitable internal standard for neg-
ative ion in EC/ES-MS studies in acetonitrile. Experiments on I? at a platinum
electrode resulted in well-behaved oxidation to I?3 . In the CN? case, the oxidation of
CN? gives cyanogen, addition of two moles of cyanogen to one mole of CN? leads to
the formation C5N?5 . The peak at m/z 156 could be assigned to C6N?6 or C12N2?12 .
With I? at gold electrodes we observed AuI?2 as well as I?3 . The AuI?2 mass
spectrometric ion intensity varies in a non-linear way throughout the applied elec-
trochemical voltage range studied. We suggest that this variation results from the
88
competition between I? adsorbed on the gold electrode surface and I? solution. In
the CN? case, we observed surface oxidation of the gold electrode at 0 V applied
at working electrode with a potentiostat connected to the cell. However, as more
negative potentials are applied, the AuI?2 intensity clearly decreases, indicating that
the ion is formed at the gold electrode.
89
Chapter 4
Fullerene
Abstract
Reduction of C60 to the 4- state in a mixture of acetonitrile/toluene at 20 ?C
was studied by using cyclic voltammetry at a scan rate of 100 mV/s. Cn?60 (n=1,2),
generated by using the bulk electrolysis method, was confirmed by ultraviolet spec-
troscopy. Gas phase observation of C?60, C3?60 and C4?60 anions generated at platinum
electrodes and detected by electrochemistry/electrospray mass spectrometry is re-
ported. The anions were electrochemically generated from solutions of C60 dissolved
in toluene/acetonitrile. The gas phase observation of C3?60 and C4?60 , despite the fact
that they have negative electron affinities, is a result of a coulombic repulsion barrier
to electron loss. These studies, which demonstrate the gas phase kinetic stability of
C3?60 and C4?60 , illustrate the promise of electrochemistry/electrospray mass spectrom-
etry for the study of metastable anions.
4.1 Introduction
One of the most interesting aspects of fullerene chemistry is the ease with which
these remarkable molecules accommodate negative charges. It has been said that
the electron-accepting ability of C60 is its most characteristic chemical property. [146]
The fact that C60 has three low-lying degenerate unoccupied molecular orbitals allows
stepwise electrochemical [29,41,121,123,184] or alkali metal reduction [185?191] to
produce fulleride anions with up to six negative charges. [146,192] Salts of C60 with up
to three negative charges have been isolated [14,128,193] and the ease of generating
90
C2?60 has lead to its use as a reagent in the synthesis of disubstituted fullerenes. [194,
195]
Although it is clear that these multiply charged fullerides exhibit remarkable
stability in solution, their observation in the gas phase has been problematic. Thus,
while C?60 is readily detected in negative ion mass spectrometry [31,34,146,196,197]
and C2?60 has been observed by laser desorption mass spectrometry, [151] gas phase
detection of fullerides with more than two negative charges has proved elusive. A
consideration of the electron affinities (EAs) of C60 provides a simple rationale for
this fact. Thus, while the first EA is quite positive at 2.66 eV [198] and the second EA
has been calculated to be positive, higher EAs are calculated to be negative. [153] The
fact that C2?60 is only observed under extremely energetic conditions while dianions
of higher fullerenes are much more common in negative ion spectrometry has been
attributed to more positive EAs for these molecules and the ability of the larger
fullerenes to mitigate coulombic repulsion. [41,152,156,199,200] Dianionic adducts of
C60 with F47,48 and (CN)2,4,6 attached are also readily observed.
It is certainly true that negative EAs will render multiply charged fulleride anions
metastable and difficult to generate by electron attachment. However, multianions
often exhibit a coulombic barrier to electron loss that can temporarily stabilize Mn?
anions with negative nth EAs. [201?203] Since negatively charged C60 anions are read-
ily generated electrochemically, it seemed to us that electrochemistry/electrospray
mass spectrometry (EC/ES-MS) [30,76,88,159] would be an ideal technique for the
detection of multiply charged C60 anions in the gas phase. Although both electrospray
mass spectrometry (ES-MS) and EC/ES-MS have been used to study C60 anions in
the gas phase, neither C3?60 nor C4?60 has been detected in these studies. We now report
EC/ES-MS studies of C60 in which these metastable polyanions are observed.
91
4.2 Experimental Section.
Chemicals. C60 was obtained from MER Corporation and used as received.
Acetonitrile (MeCN), toluene, and 49% hydrofluoric acid were purchased from Fisher
Scientific Co.
Electrochemicalgradetetrabutylammoniumtetrafluoroborate([Bu4N]BF4, 99.5%),
tetrabutylammonium perchlorate ([Bu4N]ClO4, 99.5%), sodium tetraphenylborate
(NaBPh4, 99.5%), CaH2 (99.0%) were obtained from Aldrich Chemical Co. and
used as received. Potassium tetraphenylborate (KBPh4) was purchased from Fluka.
All solutions were degassed with dried grade 5.0 Ar gas at least 10 mins.
Tetrabutylammonium tetrafluroborate, ([Bu4N]BF4) and tetrabutylammonium
perchlorate were dried at least 72 hours in a vacuum oven at 40 ?C prior to use.
Acetonitrile was distilled over CaH2 under N2 at least 6 hours before use. Toluene
was dried over sodium under N2 at least 10 hours before use.
Cyclic Voltammetry. Cyclic voltammetric measurements were performed with
an Epsilon-EC potentiostat (Bioanalytical System Inc. with data system version
1.31.65NT.) in a 5:1 v/v deaerated toluene/acetonitrile mixture. The solvent mixture
contained 0.1M [Bu4N]BF4 as supporting electrolyte. A conventional three-electrode
cell with a platinum working electrode and a platinum mesh as the counter electrode
was used. All potentials were measured versus a non-aqueous Ag/Ag+ reference
electrode at room temperature, 20 ?C. (The values were converted to values vs. SCE
by adding 0.29 V. [165])
A non-aqueous reference electrode Ag/Ag+ was prepared according to a standard
procedure. [43] A silver wire, cleaned by sonication in an acetone solution, was im-
mersed in a dried acetonitrile solution containing 0.01 M AgNO3 and the supporting
92
electrolyte. The concentration of supporting electrolyte usually was the same as the
concentration used in the electrolysis solution. The reference electrode solution was
contained in a glass tube (ID 5 mm), which was constructed using Pyrex tube with
a porous Vycor plug sealed to one end with heat shrink tubing, this tube was then
inserted into a 7 mm OD glass salt bridge (E porosity), which was dipped into the
C60 solution.
All measurements were done twice.
Controlled-Potential Electro-reduction C60. A homemade bulk electrolysis
cell, shown in Figure 4.1, was used to perform the experiments of controlled-potential
electro-reduction of a C60 acetonitrile suspension. The acetonitrile suspension was
prepared by adding 0.5 mL saturated C60 toluene solution in 7 - 8 mL acetonitrile.
The acetonitrile suspension in the working electrode compartment contained 4 mM
[Bu4N]BF4 as supporting electrolyte. In order to minimize the cell resistance, the
solution in the counter electrode compartment contained 20 mM [Bu4N]BF4 as sup-
porting electrolyte. The working electrode compartment has a volume of 8 mL, which
is separated from the counter electrode reaction by a fine ceramic frit. The cell body,
platinum mesh working electrode and glass salt bridges were dried in a vacuum oven
at 80 ?C for at least 72 hours prior to use.
The controlled-potential electrolysis was performed in a glove-bag with high pu-
rity dried N2 flowing through it. During the electrolysis, a Teflon-coated stirring bar
was used to agitate the solution in the dark. The solution was continuously purged
by dried Ar (Grade 5.0) to minimize possible O2 contamination. The concentration
of C60 anions in the acetonitrile solution was estimated as 0.23 - 0.26 mM.
The controlled-potentials, corresponding to generation of Cn?60 (n=1,2,3), were
applied at the working electrode sequentially. The calculated elapse time was ? 50
93
Ar Gas
Counter Electrode
Pt Mesh
Pt   Mesh
Working Electrode
Salt Bride
Reference Electrode(Ag/Ag+)
Teflon Cap
Fine Ceramic Frit
Figure 4.1: Schematic Diagram of the Electrochemical Cell for the Controlled-
Potential Electrolysis
94
mins for each stepwise electrochemical reduction, and the actual experimental elapsed
time was 1 hour. The Cn?60 (n=1,2,3) solution obtained at different potentials was then
transferred anaerobically via a 500 ?L gas tight syringe to a UV spectrometer and
the electrospray mass spectrometer.
Visible Near IR Spectro-electrochemistry. The Vis-NIR spectra were
recorded on Hewlett-Packard 5452A and 8453 diode array spectrophotometers which
were thermostated at 298 K.
EC/ES-MS Instrumental. Studies of gas phase Cn?60 polyanions by EC/ES-
MS requires finding a suitable solvent medium in which the desired quantity of solid
C60 can be dissolved. The solvent also needs to support the potential range to be
electrochemically studied and needs to to be suitable for electrospray ionization.
In this dissertation, we choose a toluene/acetonitrile solvent mixture and use the
?slow? method [212,215] to prepare the C60 solution. It was found that a 5:1 mix-
ture of toluene and acetonitrile with a low concentration [Bu4N]BF4 as a supporting
electrolyte was a satisfactory solvent for the C60 EC/ES-MS experiments.
AllEC/ES-MSdatawerecollectedwithaVG-Trios-2000quadrupoleelectrospray
mass spectrometer. The mass spectra were acquired in negative-ion mode co-adding
50 scans at a scan rate of 0.5 scan/s. The ES tip voltage was ? 2.89 kV, the source
temperature was 60 ?C for the toluene/acetonitrile solvent mixture and the cone
voltages were typically 20 V. The solutions were pumped with a syringe pump at a
flow rate of 5.0 mL/hour.
The electrochemical cell used for EC/ES-MS is shown in Figure 2.1. As de-
scribed, the cupped collection assembly leads to the mass spectrometer through a
fused silica capillary. The working electrode potentials were applied using a Bioan-
alytical System Inc. CV-27 potentiostat. A platinum-mesh counter electrode and
95
a Ag/AgCl reference electrode (connected to the cell through a double salt bridge)
were used. All potentials are quoted vs. aqueous Ag/AgCl. Literature values for C60
reduction potentials [192] have been converted to potentials vs. Ag/AgCl. Before
use, the working electrodes were cleaned by rinsing sequentially with distilled water,
49% hydrofluoric acid, piranha [161] solution, distilled water, HPLC grade methanol
and fresh acetonitrile.
4.3 Results and Discussion.
The solubility of C60, up to now, has been determined in up to 150 solvents.
[210,211,213?217] Table 4.1 lists the solubility of C60 in some inexpensive aromatic
solvents. Toluene is used in this dissertation, this is because it is a relatively easy to
obtain a clean and dry solvent, and is stable over a wide potential range. [42,43]
Despite the good solubility of C60 in toluene, a disadvantage of using pure toluene
as solvent is its low dielectric constant, shown in Table 4.1. The low dielectric con-
stant will cause a significant ohmic drop unless high concentrations of supporting
electrolytes are added. This problem can be successfully solved by use of a solvent
mixture, toluene-acetonitrile.
The properties of solubility, aggregation and electrochemical reduction of C60
in the solvent mixtures, toluene-acetonitrile, have been studied extensively. [212,
217?219] Earlier research results [212,215] show a somewhat unique property of the
fullerene?s (C60, C70) when dissolved in a toluene-acetonitrile mixtures. As the com-
position of the solvent mixtures varies, the C60 solution undergoes dramatic solva-
tochromic changes, and the changes are strongly dependent on the C60 concentration.
96
Solvents Toluene Dichlorobenzene Dichloromethane Benzonitrile Acetonitrile
Solubility
(mg/mL) 3.2[1]/2.8[2] 27[2] 0.26[2] 0.41[2] 0[2]
Dielectric Constant 2.5[3]) 8.93[3] 25.2[3] 37.5[3])
Surface Tension 25.97[4] 27.2[4] 19.10[5]
mN/m2
Table 4.1: Solubility of C60 in Various Solvents.
(1):reference [214]. (2): reference [213]. (3): reference [209]. (4): reference [207], at 313 K. (5): reference [208]
97
The solvatochromism is attributed to the formation of C60 clusters (dimers and poly-
mers) in the solvent mixtures at a room temperature. The clusters were generated
by either ?slow? or ?fast? addition of acetonitrile to toluene solutions of C70 or C60.
Cyclic Voltammetry. Typical cyclic voltammograms of a 0.95 mM C60 solution
in 5:1 (v/v) toluene-acetonitrile at a platinum electrode (0.10 M [Bu4N]BF4) is shown
in Figure 4.2. The CV was scanned at a rate of 100 mV/s, it proceeds an initial
forward scan until the potential is switched at -3.20 V (vs Ag/Ag+), shown in Figure
4.2. Fourclearlywell-definedreduction/oxidationpeaksareassignedtothegeneration
of 1-, 2-, 3- and 4- anions of C60. Electrochemical reversibility is evident from Figure
4.2.
The four reduction peaks with cathodic peak potentials(E(pc)) measured relative
to Ag/Ag+ are at -1.02, -1.47, -1.96, -2.50 V. In all cases, the potential separation
between any two successive peaks is consistent with the reported value, 450 ? 50
mV. [37,106,107]
CV data clearly show that Cn?60 (n=1,2,3,4) can be generated under the experi-
mental conditions present in this dissertation.
UV-vis-near-IR Spectroscopy. The fullerene anion, C?60, was prepared by
controlled-potential electro-reduction at -1.2 V (vs. Ag/Ag+) in a C60 acetonitrile
suspension. The elapsed time for electrolysis was 60 minutes. The reduction of C60
to C?60 was confirmed by the conversion of the solution from a brown suspension to a
solution with apple-red color. The solution was then anaerobically transferred to an
UV cell by a 0.5 mL gas tight syringe. Figure 4.3 curve 1 shows the UV spectrum
98
-2.5E-05
-1.5E-05
-5.0E-06
5.0E-06
1.5E-05
2.5E-05
3.5E-05
4.5E-05
5.5E-05
6.5E-05
-3.50-3.00-2.50-2.00-1.50-1.00-0.500.00
Potential (V)
Cu
rre
nt 
(A
)
Figure 4.2: Cyclic Voltammogram of C60 in Toluene/Acetonitrile at the Scan Rate
100 mV/s, 20?C, V vs. Ag/Ag+
99
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
350 450 550 650 750 850 950 1050 1150
Wavelength (nm)
Ab
so
rba
nc
e
1
2
Figure 4.3: Near-IR Spectrum of Cn?60 (n=1,2) in Toluene/Acetonitrile with
[Bu4N]BF4 at 20?C, Curve (1): IR Absorbance of C?60, Curve (2): the IR Absorbance
of C2?60
100
of the C?60 in acetonitrile, it has a strong, characteristic near-IR peak at 1072 nm, (epsilon1
= 15000 - 20000 M?1 ?cm?1). This peak has been assigned to the allowed t1u to t1g
transition, Figure 1.6. [111,112]
Further electrolysis at -1.8 V led to conversion of C?60 to C2?60 with a new peak
shown in Figure 4.3 curve 2. The absorption peak at 948 nm is in good agreement
with the early reports. [31,206,220?224] The acetonitrile solution color changed from
apple-red to dark-red. The solution was then transferred anaerobically via a 500 ?L
gas tight syringe to either a UV spectrometer or the electrospray mass spectrometer.
Although the C2?60 was observed in UV-Vis spectroscopy, we were not able to observe
the C2?60 by electrospray mass spectrometry.
As the charge on C60 increases, the anions become increasingly sensitive to oxy-
gen, water and the solvent purity. Dry-box and vacuum line (O2,H2O < 1ppm) tech-
niques are needed for preparation and transportation of the higher charged C60 an-
ions solutions. [108,128,225?229] Attempts by us to generate C3?60 were carried out
by further electrolysis at -2.2 V, but we were not able to confirm C3?60 by UV-Vis
spectroscopy.
EC/ES-MS studies of C60. C60 online stepwise electro-reductions were carried
out at both platinum and gold electrodes with different supporting electrolytes as well
as different concentrations of tetrabutylammonium tetrafluoroborate ([Bu4N]BF4),
tetrabutylammoniumperchlorate([Bu4N]ClO4)andsodiumtetraphenylborate(NaBPh4).
The concentration of supporting electrolytes range from 1 mM to 8 mM in a solvent
mixture, (5:1 by v/v) toluene-acetonitrile. The toluene solution of C60 was prepared
first, and then the acetonitrile was slowly added drop by drop using a gas-tight syringe
while stirring continuously.
101
It was found that a 5:1 mixture of toluene and acetonitrile with 4 mM [Bu4N]BF4
as a supporting electrolyte and 0.05 mM NaBPh4 as an internal standard was a satis-
factory solvent for the C60 EC/ES-MS experiments. Solutions with this composition
were loaded into the electrochemical flow cell, Figure 2.1, and the applied poten-
tial was decreased incrementally from 0.0 to -3.4 V. The solution was continuously
sprayed into the mass spectrometer. Figure 4.4 shows a three-dimensional plot of the
observed mass spectra as a function of the applied potential on the cell. The peaks
have all been normalized to the intensity of the internal standard. It can be seen
that small amounts of C?60, C3?60 , and C4?60 are detected at 0.0 V. The intensities of all
of these anions increase as the voltage is made more negative, with C4?60 showing the
most dramatic increase in intensity. In none of these experiments were we able to
see evidence of C2?60 . In addition, a peak at m/z = 761, corresponding to the adduct
of C?60 and acetonitrile, was observed. Another peak at m/z = 254 was observed at
potentials between -2.75 and -3.00 V, this peak was assigned as the adduct of C3?60
and acetonitrile ([C60 ?CH3CN]3?).
First, the peaks observed at 0.0 V will be discussed. Since it is known that
electrospray sources function as a type of electrochemical cell, [197] it is not surprising
that weak -peaks for the fulleride anions are observed when spraying C60 solutions
from the cell at 0.0 V applied potential. Another probable reason for observation of
the fulleride anions at 0 V might be the low dielectric constant of toluene and the
surface activity of C60. Early work reported measurements of the surface tension of
toluene solutions of C60 as a function of temperature and concentration, [207] it was
found that C60 was surprisingly surface active (Table 4.2); at the largest concentration
studied (Table 4.3) C60 occupied nearly a monolayer on the surface. In the ion
evaporation mode of the electrospray ionization, [2] the mass-analyzed ion intensity
102
100 180 240
319
720761
?3
?2.75
?2.5
?2.25
?2
?1.75?1.55
?1
0
0
1
2
3
4
V
m/z
Relative Intensity
254 
Figure 4.4: Three Dimensional Plot of EC/ES-MS Mass Spectra C60 as a Function
of Applied Electrochemical Cell Potential
103
ne-
ES-MS
n-
n = 1, 3, 4
Figure 4.5: C60 Reduction
104
Solution
Concentration 0 0.208 0.313 0.625 1.250 2.083
mM
Surface Tension, ? (mN/m)
273 K 30.72 30.18 29.73 29.70 28.99
303 K 27.35 27.14 27.06 26.78 26.35 25.85
313 K 25.97 25.75 25.64 25.21 25.00
Table 4.2: Surface Tension of C60 Solutions in Toluene
? Reference [207]
105
Solution
Concentration 0.208 0.313 0.625 1.250 2.083
mM
Surface Excess, ?12 d?/dC
(mol/cm)
273 K 1.136 1.704 3.406 6.814 11.357 1.238
303 K 0.609 0.913 1.826 3.652 6.085 0.712
313 K 0.406 0.609 1.218 2.436 4.059 0.475
Table 4.3: Surface Excess of C60 Solution in Toluene
? Reference [207]
106
IA of an analyte ion depends on the surface activity of the analyte ion. [6]
When a charged droplet is formed during the electrospray process, it contains
about 10 C60 (C60 radius: 7?A) units if the radius of the charged-droplet is 10 nm. [2]
On the basis of above results, it is clear that most surface sites of a charged droplet
will be occupied by solvated C60 molecules because of its surface activity in the solvent
mixture, toluene/acetonitrile.
On the other hand, by comparing the physical and chemical properties of toluene
and acetonitrile, Table 4.1, one can immediately see that approximately same surface
tension (? 20 mN?m?2) but remarkably different dielectric constants. The dielectric
constants of toluene and acetonitrile are, respectively, 2.4 and 38.8 at 60 ?C. A solvent
with high dielectric constant usually is capable of holding more charges than a solvent
with a low dielectric constant. [230] A solvent with high dielectric constant tends to
form solvated ions rather than ion-pairs. [42]
Since C60 is an extremely hydrophobic cage-shaped molecule, it is probable that
the C60 will have non-polar toluene molecules surrounding it rather than an acetoni-
trile molecule when dissolved in the solvent mixture. Acetonitrile has been proven to
be an excellent electrospray solvent for transferring ions from solution phase to gas
phase, [5] but a different story might be true for toluene/acetonitrile (5/1 by v/v)
mixtures. When a solution of C60 passes through the electrospray needle, the presence
of high negative electric field at the tip renders the negative charges on the surface of
the droplets. [2] Since most available surface sites are occupied by the solvated C60,
the negative charge on the surface will be mostly taken by C60 molecules because of
the electron-accepting ability of C60 and the low electric constant of toluene. Further-
more, a charged C60 cage allows for the attachment of acetonitrile molecules. This
could explain the peak m/z = 761 at zero applied potential at the flow cell.
107
The fact that the peaks in Figure 4.4 increase with increasing applied negative
working electrode voltage provides evidence for their electrochemical formation in
solution and subsequent transport into the gas phase, Figure 4.5. That we observe
no C2?60 in these experiments is puzzling since the gas phase dianion has been re-
ported. [151] The reasons that no C2?60 was observed are unknown. Given the ease
with which this species is generated electrochemically, it must be formed initially in
our experiments. It could be possible that C2?60 is rapidly removed by the unknown
chemical reactions, but it is difficult to imagine a reaction exclusive to C2?60 that would
not also remove other negatively charged C60 species.
As described in Chapter 1, C60 has been considered a good candidate for prepara-
tion of gas-phase multiply-charged anions, [31,34,146,151,196,197] but, to date, only
C2?60 and C2?70 have been observed by using electrospray and laser desorption meth-
ods. [151] In this dissertation, we report for the first time observation of multiply
charged Cn?60 (n=3,4) in the gas phase by using EC/ES-MS. Observation of gas-phase
Cn?60 (n=3,4) by using the EC/ES-MS method is scientifically important for both
experimental and theoretical chemistry.
The available techniques for generating, observing and studying long-lived mul-
tiply charged anions can be categorized as follows: laser desorption of electrophilic
species from a metal surface; [150] electrospray of an aqueous solution containing di-
anion salts into a vacuum [201,231,232] followed by photodetachment photoelectron
spectroscopy; electron capture of neutral clusters of eletrophilic molecules; [233] elec-
tron capture by single molecules followed by dimerization to form dianions in the gas
phase [234] and formation of multiply charged anions by collision induced dissociation
of a high energy monoanion. [235] We demonstrate that the EC/ES-MS method can
108
be categorized as another experimental method for generating and detecting multiply
charged anions in the gas phase by mass spectrometry.
During the past decades, extensive knowledge of physical chemistry has been
built on experimental and theoretical studies of single charged anions in the gas phase.
[236,237] But turning to gas-phase multiply charged negative anions (MCAs), both
theoretical and experimental information on them are rather insufficient. [236,237]
MCAs are very common in the condensed phases because they are stabilized
by solvation in solution or counter ions in solids, but they are very rare in the gas
phase. This is because the difficulty of formation of the multiply-charged anions in
the gas phase. A neutral atom or molecule holds its electrons in their orbitals by the
electrostatic interaction between the electrons and the positively charged core. Once a
neutral molecule binds more than one electron, the dynamic stability (attachment or
detachment of electrons) of the gas-phase multiply-charged anions is determined by a
coulombic repulsive barrier (CRB) between the anion and the electron. [202] The CRB
is formed by the interaction of the long-range Coulomb repulsion among the excess
charges and the short-range interaction of electrons in the LUMO. [202] Both the
physical and chemical properties of the anions are determined by the CRB. In some
cases, the CRB can trap electrons in molecules which have negative EAs. [203,204]
The EA is positive if energy is released during the process of the electron attachment.
EA can be high if the incoming electron enters a vacancy of a compact shell and can
interact strongly with the nucleus. In most cases, the process of electron attachment
on anions is a endothermic, the attaching electron is inevitably repelled by the charge
already present in the anion.
Figure 4.6 schematically shows the different EAs and CRB barriers shifting due
to the stepwise addition of negative charges to C60. Cn?60 (n=2,3) have negative EAs.
109
0 eV EA
2 > 0
EA1 > 0
EA3 <  0
EA4 <  0
2-
3-
1-
Po
ten
tia
l E
ne
rg
y
r (C60n-          e- )
CRB
El
ect
ro
n A
ffi
nit
y
t1u
Figure 4.6: Schematic Diagram of the Lowest Unoccupied Molecular Orbitals of Cn?60
n=0,1,2,3 and the Shifting of the CRB Barrier Heights and the Electron Affinities
Due to the Stepwise Addition of Three Negative Charges. The r Represents the
Electron-Anion Distance
110
The CRB barrier height increases with the number of the attached negative charges
increasing due to the strong mutual Coulomb repulsion of their excess charges.
The negative EAs indicate that Cn?60 (n=3,4) holds excess electrostatic energy,
therefore they are metastable in the gas phase. The common method of generating
multiply charged C60 anion is to transfer the solid C60 into gas phase followed by the
stepwise attachment of electrons. Due to the large CRB barriers for Cn?60 (n=2,3)
attachment of electron to the anions forming Cn?60 (n=3,4) in the gas phase has not
been possible.
In the EC/ES-MS method, however, Cn?60 (n=3,4) is generated in solution where it
is stabilized by solvation and then sprayed into the gas phase. Once the Cn?60 (n=3,4),
formed in solution, are transferred into the gas phase, they are immediately trapped
inside the CRB barrier, which acts like electrostatic ?well?. Both the large barrier
height and the large molecular size of C60 play important roles in the stabilization
of Cn?60 (n=3,4) and in preventing electron detachment. Since electrostatic energy is
stored in these multiply charged anions, they can be treated as an electrostatic energy
storage medium in gas phase. [203]
This observation of Cn?60 (n=3,4) with negative binding energy in gas phase
greatly enriches the knowledge of the field of gas-phase multiply charged anions in
physical chemistry. These results will allow further studies on these and other gas
phase multiply charged anions with negative binding energy such as C70.
4.4 Conclusions
The gas phase observation of C?60, C3?60 and C4?60 in these experiments demonstrate
that these metastable polyanions may be observed in the gas phase by first generating
them in solution and spraying into the gas phase. The fact that C3?60 and C4?60 can be
111
observed despite the fact that they have negative EAs is attributed to a coulombic
repulsion barrier to electron loss. As in any EC/ES-MS technique, the method is
successful only for those fulleride anions that are soluble. The reason that C2?60 was
not detected in these experiments is unknown, however we suggest the observation
of no C2?60 could be due to the unknown chemical reactions. The high concentration
of charge in these fulleride polyanions is expected to render them strong nucleophiles
and the method promises to become a convenient way of measuring their gas phase
reactivities.
112
Bibliography
[1] Bruckenstein, S.; Gadde, R. R. J. Am. Chem. Soc. 1971, 93, 793-794.
[2] Kebarle, P.; Peschke, M. Analytica Chimica Acta 2000, 406, 11-35.
[3] Wolter, O.; Heitbaum, J. Ber. Bunsengens. Phys. Chem. 1984, 88, 6-10.
[4] Deng, H.; Van Berkel, G. J. Electroanal. 1999, 11, 857-865.
[5] Pramanik, B. N.; Ganguly, A. K.; Gross. M. L. Applied electrospray mass spec-
trometry New York : Marcel Dekker, 2002.
[6] Tang, L.; Kebarle, P. Anal. Chem. 1993, 65, 3654-3668.
[7] Macagno, V. A.; Giordano, M. C.; Arv??a, A. J. Electrochimica Acta 1969, 14,
335-357.
[8] Arv??a, A. J.; Giordano, M. C.; Podest?a, J. Electrochimica Acta 1969, 14, 389-
403.
[9] Swathirajan, S.; Bruckenstein, S. J. Electroanal. Chem. 1983, 143, 167-178.
[10] Iwamoto, R. Anal. Chem. 1959, 31, 955-955.
[11] Bagotsky, V. S.; Vassilyev, Y. B.; Weber, J.; Pirtskhalava, J. N. J. Electroanal.
Chem. 1970, 27, 31-46.
[12] Guidelli, R.; Piccardi, G. Electrochimica Acta 1967, 12, 1085-1095.
[13] Popov, A. I.; Geske, D. H. J. Am. Chem. Soc. 1958, 80, 1340-1352.
[14] Bhyrappa, P.; Paul, P.; Stinchcombe, J.; Boyd, P. D. W.; Reed, C. A. J. Am.
Chem. Soc. 1993, 115, 11004-11005.
[15] Svensson, P. H.; Rosdahl, J.; Kloo L. Chem. Eur. J. 1999, 5, 305-311.
[16] Lane, R. F.; Hubbard, A. T. J. Phys. Chem. 1975, 79, 808-815.
[17] Sedlmaier, H. D.; Plieth, W. J. J. Electroanal. Chem. 1984, 180, 219-230.
[18] McCarley, R. L.; Bard, A. J. J. Phys. Chem. 1991, 95, 9618-9620.
[19] Goolsby, A. D.; Sawyer, D. T. Anal. Chem. 1968, 40, 1978-1983.
[20] Ahrland, S.; Nilsson, K.; Persson, I.; Yuchi, A.; Penner-Hahn, E. Inorg. Chem.
1989, 28, 1833-1838.
113
[21] Cahan, B. D.; Villullas, H. M.; Yeager, E. B. J. Electroanal. Chem. 1991, 306,
213-238.
[22] Wagner, D.; Gerischer, H. J. Electroanal. Chem. 1989, 258, 127-137.
[23] Kissner, R. J. Electroanal. Chem. 1995, 385, 71-75.
[24] Roulet, R.; Lan Quang, N.; Mason, W. R.; Fenske, G. P. Helv. Chim. Acta 1973,
56, 2405-2418.
[25] Kunimatsu, K.; Seki, H.; Golden, W. G.; Gordon, J. G.; Philpott, M. R. Lang-
muir 1988, 4, 337-341.
[26] Abderrahmane Tadjeddine, A.; Le Rille, A. Electrochimica Acta 1999, 45, 601-
609.
[27] Brotherton T. K.; Lynn, J. W. Chem. Rev. 1959, 59, 841-883.
[28] Wiley, D. W.; Webster, O. W.; Blanchard, E. P. J. Org. Chem. 1976, 41,
1889-1893.
[29] Kukolich, G. S.; Huffman, D. R. Chem. Phys. Lett. 1991, 182, 263-265.
[30] Schell-Sorokin, A. J.; Mehran, F.; Eaton, G. R.; Eaton, S. S.; Viehbeck, A.;
O?Toole, T. R.; Brown, C. A. Chem. Phys. Lett. 1992, 195, 225-232.
[31] Baumgarten, M.; Guegel, A.; Gherghel, L. Adv. Mater. 1993, 5, 458-461.
[32] Baumgarten, M.; Gherghel, L. Appl. Magn. Reson. 1996, 11, 171-182.
[33] Hiraoka, K.; Kudaka, I.; Fujimaki, S.; Shinohara, H. Rapid Commun. Mass
Spectrom. 1992, 6, 254-256.
[34] Fullagar, W. K.; Gentle, I. R.; Heath, G. A.; White, J. W. J. Chem. Soc. Chem.
Commun. 1993, 6, 525-527.
[35] Disch, R. L.; Schulman, J. M. Chem. Phys. Lett. 1986, 125, 465-466.
[36] Bochvar, D. A.; Gal?apern, E. G. Dokl. Akad. Nauk SSSR 1973, 209, 610-612.
[37] Xie, Q.; Perez-Cordero, E.; Echegoyen, L. J. Am. Chem. Soc. 1992, 114, 3978-
3980.
[38] Smith, P.; Mark, T. D. Chem. Phys. Lett. 1993, 213, 202-206.
[39] Hathiramani, D.; Aichele, K.; Arnold, W.; Huber, K.; Salzborn, E. Phys. Rev.
Lett. 2000, 85, 3604-3607.
[40] Robert, J.; Doyle, Jr.; Ross, M. M. J. Phys. Chem. 1991, 95, 4954-4956.
[41] Khairallah, G.; Barrie, P. Chem. Phys. Lett. 1997, 268, 218-222.
114
[42] Bard, A.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applica-
tions Application, 2000, 2nd ed, New York: John Wiley & Sons, Inc.
[43] Sawyer, D. T.; Sobkowiak, A.; Roberts, J. L. Electrochemistry For Chemists 2nd
ed, New York: John Wiley & Sons, Inc. 1995,
[44] Schalley, C. A. Modern Mass Spectrometry 2003, Berlin, New York: Springer,
c2003.
[45] Wolter, O.; Heitbaum, J. Ber. Bunsengens. Phys. Chem. 1984, 88, 2-6.
[46] Kotiaho, T.; Lauritsen, F. R.; Choudury, T. K.; Cooks, R. G. Anal. Chem. 1991,
63, 875A-883A.
[47] Hartung, T.; Schmiemann, U.; Kamphausen, I.; Baltruschat, H. Anal. Chem.
1991, 63, 44-48.
[48] D?ulberg, A. In Institut F?ur Physikalische Chemie, Universit?at Witten-Herdecke:
Witten-Herdecke, 199, page 164
[49] Tegtmeyer, D.; Heindrichs, A.; Heitbaum, J. Ber. Bunsenges. Phys. Chem.1989,
93, 201-206.
[50] Arpino, P. Mass Spectr. Rev. 1990, 9, 631.
[51] Hambitzer, G.; Heitbaum, J. Anal. Chem. 1986, 58, 1067-1070.
[52] Hambitzer, G.; Heitbaum, J.; Stassen, I. J. Electroanal. Chem. 1998, 447, 117-
124.
[53] Stassen, I.; Hambitzer, G. J. Electroanal. Chem. 1997, 440, 219-228.
[54] Volk, J. K.; Lee, M. S.; Yost, R. A.; Brajter-Toth, A. Anal. Chem. 1988, 60,
720-722.
[55] Volk, J. K.; Yost, R. A.; Brajter-Toth, A. Anal. Chem. 1992, 64, 21A-26A,30A-
33A.
[56] Volk, J. K.; Yost, R. A.; Brajter-Toth, A. Anal. Chem. 1989, 61, 1709-1717.
[57] Yamashita, M.; Fenn, J. B. J. Phys. Chem. 1984, 88, 4451-4459.
[58] Yamashita, M.; Fenn, J. B. J. Phys. Chem. 1984, 88, 4671-4675.
[59] Iribarne, J. V.; Thomson, B. A. J. Chem. Phys. 1976, 64, 2287-2290
[60] Thomson, B. A.; Iribarne, J. V. J. Chem. Phys. 1979, 71, 4451-4459
[61] Bartmess, J. E.; Philips, L. R. Anal. Chem. 1987, 59, 2012-2014.
115
[62] Huang, E. C.; Wachs, T.; Conboy, J. J.; Henion, J. D. Anal. Chem. 1990, 62,
713A-725A.
[63] Volk, K. J.; Yost, R. A.; Brajter-Toth, A. J. Electrochem. Soc. 1990, 137, 1764-
1771.
[64] Volk, K. J.; Yost, R. A.; Brajter-Toth, A. Anal. Chem. 1989, 474, 231-243.
[65] Dole, M.; Mack, L. L.; Hines, R. L.; Mobley, R. C.; Ferguson, L. D.; Alice, M.
B. J. Chem. Phys. 1968, 49, 2240-2249.
[66] Aleksandrov, M. L.; Gall, L. N.; Krasnov, V. N.; Nikolaev, V. I.; Pavlenko, V.
A.; Shkurov, V. A. Dokl. Akad. Nauk SSSR, 1984, 277, 379.
[67] Reyleigh, L. Philas. Mag. 1882 14, 184.
[68] Reid, R. C.; Prausnitz, J. M.; Sherwood, T. K. The properties of Gases and
Liquids. 3rd ed., 1977. New York: McGraw-Hill. The surface tension of a luiquid
is defined as the force per unit length exerted in the plane of the liquid?s surface.
Te unit if surface tenion is dyn(cm)?1.
[69] R?ollgen, F. W.; Bramer-Wegne, E.; Buttering, L. J. Phys. Colloq. 1984, 45,
supplement 12, C9-297.
[70] Schmelzeisn-Redeker, G.; Buttering, L.; R?ollgen, F. W. Int. J. Mass Spectrom.
Ion Processes 1989, 90, 139.
[71] Loscertales, I. G.; Fern?andez De La Mora, J. J. Chem. Phys. 1995, 103, 5041-
5080.
[72] Pretty, J. R.; Duckworth, D. C.; Van Berkel, G. J. Anal. Chem. 1997, 69, 3544-
3551.
[73] Pretty, J. R.; Deng, H.; Goeringer, D. E.; Van Berkel, G. J. Anal. Chem. 2000,
72, 2066-2074.
[74] Deng, H.; Van Berkel, G. J. Anal. Chem. 1999, 71, 4284-4293.
[75] Van Berkel, G. J.; Asano, K. G.; Kertesz, V. Anal. Chem. 2002, 74, 5047-5056.
[76] Zhou, F.; Van Berkel, G. J. Anal. Chem. 1995, 67, 3643-3649.
[77] Zhang, T.; Palii, S. P.; Eyler, J. R.; Brajter, A. Anal. Chem.2002, 74, 1097-1097.
[78] Zhang, T.; Brajter-Toth, A. Anal. Chem. 2000, 72, 2533-2540.
[79] Regino, M. C. S.; Weston, C.; Brajter-Toth, A. Anal. Chim. Act. 1998, 369,
253-262.
[80] Regino, M. C. S.; Brajter-Toth, A. Anal. Chem. 1997, 69, 5067-5072.
116
[81] Hambitzer, G.; Heitbaum, J. Anal. Chem. 1998, 70, 838-842.
[82] Gun, J.; Modestov, A.; Lev, O.; Poli, R. Eur. J. Inorg. Chem. 2003, 12, 2264-
2272.
[83] Diehl, G.; Karst, U. Anal. and Bioanalytical Chem. 2002, 373, 390-398.
[84] Kertesz, V.; Dun, N. M.; Van Berkel, G. J. Electrochimica Acta 2002, 47, 1035-
1042.
[85] Permentier, H. P.; Bruins, A. P. J. Am. Soc. Mass Spec. 2004, 15, 1707-1716.
[86] Bond, A. M.; Colton, R.; D?Agostino, A.; Downnard, A. C.; Traeger, J. C. Anal.
Chem. 1995, 67, 1691-1695.
[87] Xu, X.; Lu, W.; Cole, R. B. Anal. Chem. 1996, 68, 4244-4253.
[88] Lu, W.; Xu, X.; Cole, R. B. Anal. Chem. 1997, 69, 2478-2484.
[89] Zhang, T.; Palii, S. P.; Eyler, J. R.; Brajter-Toth, A. Anal. Chem. 2002, 74,
1097-1103.
[90] Pretty, J. R.; Van Berkel, G. J. Rapid Comm. Mass. Spectrom. 1998, 12, 1644-
1648.
[91] Kertesz, V.; Van Berkel, G. J. Electroanal. 2001, 13, 1425-1430.
[92] Hambitzer, G.; Heitbaum, J.; Stassen, I. Anal. Chem. 1998, 70, 838-842.
[93] Modestov, A. D.; Gun, J.; Mudrov, A.; Lev, O. Electroanal. 2004, 367-378.
[94] Austad, T.; Esperas, S. Acta. Chem. Stand. A28, 1974, 892-896.
[95] Sharpe, A. G. Inorganic chemistry New York : Wiley, c1992. 3rd ed.
[96] DeKOCK, R. L.; Caswell, D. S. Inorganica Chimica Acta 37, 1979, L469-L470.
[97] Kroto, H. W.; Heath, J. P.; O?Brien, S. C.; Curl, R. F.; Smalley, R. E. Nature
1985, 318, 162.
[98] Dresselhaus, M. S.; Dresselhaus, G.; Eklund, D. C. Sciences of Fullerernes and
Carbon Nanotubes, 2000 A John Wiley Sons, Inc. 1st ed.
[99] Kadish, K. M.; Ruoff, R. S. Fullerenes Chemistry, Physics, and Technology,
2000, A John Wiley Sons, Inc. 1st ed.
[100] Haddon, R. C.; Brus, L. E.; Raghavachari, K. Chem. Phys. Lett. 1986, 125,
459-464.
[101] Haymet, A. D. J. Chem. Phys. Lett. 1985, 122, 421-424.
117
[102] Kr?atschmer, W.; Lamb, L. D.; Fostiropoulus, K.; Huffman, D. R. Nature 1990,
347, 354-358.
[103] Johnson, R. D.; Meijer, G.; Bethune, D. S. J. Am. Chem. Soc. 1990, 112,
8983-8984.
[104] Taylor, R.; Hare, J. P.; Abdul-Sada, A. K.; Kroto, H. W. J. Chem. Soc. Chem.
Commun. 1990, 115, 1423-1425.
[105] Niyazimbetov, M. E.; Evans, D. H.; Lerke, S. A.; Cahill, P. A.; Henderson, C.
C. J. Phys. Chem. 1994, 98, 13093-13098.
[106] Ohsawa, Y.; Saji, T. J. Am. Chem. Soc. 1992, 114, 3978-3980.
[107] Zhou, F.; Jeboulet, C.; Bard, A. J. Am. Chem. Soc. 1992, 781-782.
[108] Khaled M. M.; Carlin, R. T.; Trulove, P. C.; Eaton, G. R.; Eaton, S. S. J. Am.
Chem. Soc. 1994, 116, 3465-3474.
[109] Olsen, S.; Bond, A. M.; Compton, R. G.; Lazarev, G.; Mahon, P. J.; Marken,
F.; Raston, C. L.; Tedesco, V.; Webster, R. D. J. Phys. Chem. A 1998, 102,
2641-2649.
[110] Koh, W.; Dubois D.; Kutner W.; Thomas Jones, M.; Kadish, K. M. J. Phys.
Chem. 1993, 97, 6871-6879.
[111] Lawson, D. R.; Feldheim, D. L.; Foss, C. A.; Dorhout, P. K.; Elliott, C. M.;
Martin, C. R.; Parkinson, B. J. Electrochem. Soc. 1992, 139, L68-L71.
[112] Fullagar, W. K.; Gentle, I. R.; Heath, G. A.; White, J. W. J. Chem. Soc.,
Chem. Commun. 1993, 525-527.
[113] Heath, G. A.; McGrady, J. E.; Martin, R. L. J. Chem. Soc., Chem. Commun.
1992, 1272-1274.
[114] Bruno, C.; Doubitski, I.; Marcaccio, M.; Paolucci, F.; Paolucci, D.; Zaopo, A.
J. Am. Chem. Soc. 2003, 125, 15738-15739.
[115] Garcia, E.; Kwak, J.; Bard, A. J. Inorg. Chem. 1988, 27, 4377-4382.
[116] Ceroni, P.; Paolucci, F.; Paradisi, C.; Juris, A.; Roffia, S.; Serroni, S.; Cam-
pagna, S.; Bard, A. J. J. Am. Chem. Soc. 1998, 120, 5480-5487.
[117] Bard, A. J.; Garcia, E.; Kucharenko, S.; Stretlets, V. V. Inorg. Chem. 1993,
32, 3528-3531.
[118] Ruoff, R. S.; Tse, D. S.; Malhotra, R.; Lorents, D. C. J. Phys. Chem. 1993, 97,
3379-3383.
118
[119] Sivaraman, N.; Dhamodaran, R.; Kaliappan, I.; Srinivasan, T. G.; Rao, P. R.
V.; Mathews, C. K. J. Org. Chem. 1992, 57, 6077-6079.
[120] Haddon, R. C.; Hebard, A. F.; Rosseinsky, M. J.; Murphy, D. W.; Duclos, S.
J.; Lyons, K. B.; Miller, B.; Rosamilia, J. M.; Fleming, R. M.; Kortan, A. R.;
Glarum, S. H.; Makhija, A. V.; Muller, A. J.; Eick, R. H.; Zahurak, S. M.; Tycko,
R.; Dabbagh, G.; Thiel, F. A. Nature 1991, 350, 320-322.
[121] Bausch, J. W.; Prakash, G. K. S.; Olah, G. A. J. Am. Chem. Soc. 1991, 113,
3205-3206.
[122] Ederl?e, Y.; Mathis, C. Macromolecules 1997, 30, 4262-4267.
[123] Chi, Y.; Bhonsle, J. B.; Canteenwala, T.; Huang, J.-P.; Shiea, J.; Chen, B.-J.;
Chiang, L. Y. Chem. Lett. 1998, 465-466.
[124] Janiak, C.; M?uhle, S.; Hemling, H.; K?ohler, K. Polyhedron 1996, 15, 1559-1563.
[125] Chen, J.; Huang, Z. E.; Cai, F. F.; Shao, Q. F.; Chen, S.-M.; Ye, H.-J. J. Chem.
Soc. Chem. Commun. 1994, 2177-2178.
[126] Stinchcombe, J.; Pe?onicaud, A.; Bhyrappa, P.; Boyd, P. D. W.; Reed, C. A. J.
Am. Chem. Soc. 1993, 115, 5212-5217.
[127] Stinchcombe, J.; Pe?onicaud, A.; Bhyrappa, P.; Boyd, P. D. W.; Reed, C. A. J.
Am. Chem. Soc. 1994, 116, 6484-6484.
[128] Boyd, P. D. W.; Bhyrappa, P.; Paul, P.; Stinchcombe, J.; Bolskar, R. D.; Sun,
Y.; Reed, C. A. J. Am. Chem. Soc. 1995, 117, 2907-2914.
[129] Sun, Y.; Reed, C. A. J. Chem. Soc. Chem. Commun. 1997, 747- 748.
[130] Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877-910.
[131] Chen, J.; Huang, Z. E.; Cai, R. F.; Shao, Q. F.; Ye, H. J. Solid State Commun.
1995, 95, 233-237.
[132] Chen, J.; Cai, R. F.; Huang, Z. E.; Shao, Q. F.; Chen, S. M. Solid State
Commun. 1995, 95, 239-243.
[133] Chen, J.; Cai, F. F.; Shao, Q. F.; Huang, Z. E.; Chen, S. M. J. Chem. Soc.
Chem. Commun. 1996, 1111-1112.
[134] Shabtai, E.; Weitz, A.; Haddon, R. C.; Hoffman, R. E.; Rabinovitz, M.; Khong,
A.; Cross, R. J.; Saunders, M.; Cheng, P.-C.; Scott, L. T. J. Am. Chem. Soc.
1998, 120, 6389-6393.
[135] Pe?onicaud, A.; Hsu, J.; Reed, C. A.; Koch, A.; Khemani, K. C.; Allemand,
P.-M.; Wudl, F. J. Am. Chem. Soc. 1991, 113, 6699-6700.
119
[136] Walch, B.; Cocke, C. L.; V?opel, R.; Salzborn, E. Phys. Rev. Lett. 1994, 72,
1439-1479.
[137] Christian, J. F.; Wan, Z.; Anderson, S. L. J. Chem. Phys. 1993, 99, 3468-3479.
[138] Takayama, M. Int. J. Mass Spectrom. Ion Process. 1991, 121, R19-R25.
[139] Farizon, B.; Farizon, M.; Gaillard, M. J.; Gobet, F.; Carr?e M.; Buchet, J. P.;
Scheier, P.; M?ark, T. D. Phys. Rev. Lett. 1998, 81, 4108-4111.
[140] Yeretzian, C.; Hansen, K.; Beck, R. D.; Whetten, R. L. J. Chem. Phys. 1993,
98, 7480-7484.
[141] D?unser, B.; Lezius, M.; Scheier, P.; Deutsh, H.; M?ark, T. D. Phys. Rev. Lett.
1995, 74, 3364-3367.
[142] Andersen, J. U.; Brink, C.; Hvelplund, P.; Larsson, M. O.; Bech Nielsen, R. B.;
Shen, H. Phys. Rev. Lett. 1996, 77, 3991-3994.
[143] Scheier, P.; D?user, B.; M?ark, T. D. Phys. Rev. Lett. 1995, 74, 3368-3371.
[144] O?Brien, S. C.; Heath, J. R.; Curl, R. F.; Smalley, R. E. J. Chem. Phys. 1988,
88, 220-230.
[145] Campbell, E. E. B.; Ulmer, G.; Hertel, I. V. Phys. Rev. Lett. 1991, 67, 1986-
1988.
[146] Reed, C. A.; Bolskar, R. D. Chem. Rev. 2000, 100, 1075-1120.
[147] Fullagar, W. K.; Gentle, I. R.; Heath, G. A.; White, J. W. J. Chem. Soc. Chem.
Commun. 1993, 6, 525-527.
[148] Hampe, O.; Neumaier, M.; Blom, M. N.; Kappes, M. N. Chem. Phys. Lett.
2002, 354, 303-309.
[149] Kozlovski, V.; Brusov, V.; Sulimenkov, I.; Pikhtelev, A. Rapid Commun. Mass.
Spectrom. 2004, 18, 780-786.
[150] Hettich, R. L.; Compton, R. N.; Ritchie, R. H. Phys. Rev. Lett. 1991, 67,
1242-1245.
[151] Limbach, P. A.; Schweikhard, L.; Cowen, K. A.; McDermott, M. T.; Marshall,
A. G.; Coe, J. V. J. Am. Chem. Soc. 1991, 113, 6795-6798.
[152] Zhou, F.; Van Berkel, G.; Donovan, B. T. J. Am. Chem. Soc. 1994, 116, 5485-
5486.
[153] Luo, J.; Peng, L. M.; Xue, Z. Q.; Wu, J. L. J. Chem. Phys. 2004, 120, 7998-
8001.
120
[154] Yang, S. H.; Pettiette, C. L.; Conceicao, J.; Cheshnovsky, O.; Smalley, R. E.
Chem. Phys. Lett. 1987, 139, 233-238.
[155] Ehrler, O. T.; Weber, J. M.; Furche, F.; Kappes, M. M. P. Phys. Rev. Lett.
2003, 91, 113006/1-113006/4.
[156] Khairallah, G.; Peel, J. B. Chem. Phys. Lett. 1998, 296, 545-548.
[157] Balades, A. T.; Ikonomou, M. G.; Kebarle, P. Anal. Chem. 1991, 63, 2109-2114.
[158] Van Berkel, G. J.; McLuckey, S. A.; Glish, G. L. Anal. Chem. 1992, 64, 1586-
1593.
[159] Van Berkel, G. J.; Zhou, F. Anal. Chem. 1995, 67, 2916-2923.
[160] Van Berkel, G. J.; Asano, K. G.; Granger, M. C. Anal. Chem. 2004, 76, 1493-
1499.
[161] Licht, S.; Cammarata, V.; Wrighton, M. S. J. Phys. Chem. 1990, 94, 6133-6140.
[162] Bard, A.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Appli-
cations, New York: Wiley, c2001, 2nd ed, page 333.
[163] Schlichting, H. Boundary-Layer Theroy, 1999, John Wiley & Sons, Inc. 9th ed.
[164] Coetzee J. F.; Cunningham G. P.; McGuire D. K.; Padmanabhan G. R. Anal.
Chem. 1962,34, 1139-1143.
[165] Mann, C. K.; Banes, K. K. Electrochemical Reactions in Nonaqueous Systems,
1970 John Wiley & Sons, Inc. 2nd ed.
[166] Sawyer, D. T.; Sobkowiak, A.; Roberts, J. L. Electrochemistry For Chemists,
2nd ed. 1995, John Wiley & Sons, Inc. p.318.
[167] Bard, A.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Appli-
cations, New York: Wiley, c2001, 2nd ed. p231.
[168] Von Stackelberg, M.; Pilgram, M.; Toome V. Z. Elektrochem. 1953, 57, 342-350.
[169] Bard, A.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Appli-
cations, New York: Wiley, c2001, 2nd ed. p.147
[170] Bard, A.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Appli-
cations Application, New York: Wiley, c2001, 2nd ed. p.31
[171] Bard, A.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Appli-
cations, New York: Wiley, c2001, 2nd ed. p.165
[172] Sawyer, D. T.; Day, R. J. J. Electroanal. Chem. 1963, 5, 195-207.
[173] Ciszkowska, M.; Osteryoung, J. G. Anal. Chem. 1995, 67, 1125-1130.
121
[174] The surface oxidation rate could be determined by the I? diffusion becasue of
the observation of the constant absolute intensity of I?.
[175] The available surface area of Au electrode could limit the Au oxidation rate,
but this can not explain that the approximately equal absolutely intensity of I?
observed with both Au and Pt electrode at 0 V in Figure 3.2 and 3.7.
[176] Beck, M. T.; G?asp?ar V.; Goel, D. P. Inorganica Chimica Acta 1979, 33, L147-
L148.
[177] Del Sesto, R. E.; Arif, A. M.; Miller, J. S. Chem. Commum. 2001, 2730-2731.
[178] Rico, E.; Del Sesto, R. E.; Arif, A. M.; Novoa, J. J.; Iwona Anusiewicz, I.;
Skurski, P.; Simons, J.; Dunn, B. C.; Edward, M.; Eyring, E. M.; Joel, S.;
Miller, J. S. J. Org. Chem. 2003, 68, 3367-3379.
[179] Holleman, A. F.; Wiberg, E. Inorganic Chemistry Academic Press, 1995, 3ed.
[180] Puddephatt, R. J. The Chemistry Of Gold 1978, Elsevier Scientific Publishing
Company, London, st ed.
[181] Sharpe, A. G. The Chemistry of Cyano Complexes of the Transition Metals
1976. Academic Press: London
[182] Kirk, D. W.; Foulkes, F. R. J. Electrochem. Soc. 1980, 127, 1993-1998.
[183] McCarley, R. L.; Bard, A. J. J. Phys. Chem. 1992, 96, 7410-7416.
[184] Buffinger, D. R.; Ziebarth, R. P.; Stenger, V. A.; Recchia, C.; Pennington, C.
H. J. Am. Chem. Soc. 1993, 115, 9267-9270.
[185] Haufler, R. E.; Conceicao, J.; Chibante, L. P. F.; Chai, Y.; Byrne, N. E.;
Flanagan, S.; Haley, M. M.; O?Brien, S. C.; Pan, C.; Xiao, Z.; Billups, W. E.;
Ciufolini, M. A.; Hauge, R. H.; Margrave, J. L.; Wilson, L. J.; Curl, R. F.;
Smalley, R. E. J. Phys. Chem. 1990, 94, 8634-8636.
[186] Loy, D. A.; Assink, R. A. J. Am. Chem. Soc. 1992, 114, 3977-3978.
[187] Ohsawa, Y.; Saji, T. J. Chem. Soc. Chem. Commun. 1992, 10, 781-782.
[188] Allemand, P. M.; Koch, A.; Wudl, F.; Rubin, Y.; Diederich, F.; Alvarez, M.
M.; Anz, S. J.; Whetten, R. L. J. Am. Chem. Soc. 1991, 113, 1050-1051.
[189] Cox, D. M.; Behal, S.; Disko, M.; Gorun, S. M.; Greaney, M.; Hsu, C. S.; Kollin,
E. B.; Millar, J.; Robbins, J.; Robbins, W.; Sherwood, R. D.; Tindall, P. J. Am.
Chem. Soc. 1991, 113, 2940-2944.
[190] Dubois, D.; Kadish, K. M.; Flanagan, S.; Haufler, R. E.; Chibante, L. P. F.;
Wilson, L. J. J. Am. Chem. Soc. 1991, 113, 4364-4366.
122
[191] Dubois, D.; Kadish, K. M.; Flanagan, S.; Wilson, L. J. J. Am. Chem. Soc.
1991, 113, 7773-7774.
[192] Echegoyen, L.; Echegoyen, L. E. Accts. Chem. Res. 1998, 31, 593-601.
[193] Paul, P.; Xie, Z.; Bau, R.; Boyd, P. D. W.; Reed, C. A. J. Am. Chem. Soc.
1994, 116, 4145-4146.
[194] Allard, E.; Riviere, L.; Delaunay, J.; Dubois, D.; Cousseau, J. Tetrahedron Lett.
1999, 40, 7223-7226.
[195] Allard, E.; Delaunay, J.; Cheng, F.; Cousseau, J.; Orduna, J. J. Org. Lett.
2001, 3, 3503-3506.
[196] Hampe, O.; Neumaier, M.; Blom, M. N.; Kappes, M. N. Chem. Phys. Lett.
2002, 354, 303-309.
[197] Kozlovski, V.; Brusov, V.; Sulimenkov, I.; Pikhtelev, A.; Dodonov B. Rapid
Commun. Mass. Spectrom. 2004, 18, 780-786.
[198] Wang, X.-B.; Ding, C.-F.; Wang, L.-S. J. Chem. Phys. 1999, 110, 8217-8220.
[199] Ehrler, O. T.; Weber, J. M.; Furche, F.; Kappes, M. M. Phys. Rev. Lett. 2003,
91, 113006/1-113006/4.
[200] Jin, C.; Hettich, R. L.; Compton, R. N.; Tuinman, A.; Kovacs-Derechskei, A.;
Marynick, D. S.; Dunlap, B. I. Phys. Rev. Lett. 1994, 73, 2821-2824.
[201] Wang, L.-S.; Wang, X.-B. J. Phys. Chem. A 2000, 104, 1978-1990.
[202] Sheller, M. K.; Compton, R. N.; Cederbaum, L. S. Science 1995, 270, 1160-
1166.
[203] Wang, X.-B.; Wang, L.-S. Nature 1999, 400, 245-248.
[204] Electron Affinity: EA is the energy that released when a electron attaches to a
gas phase atom, molecule or ion. Usually an EA is positive if energy is released
during the process of the electron attachment. Electron affinity can be great if
the coming electron enters vacancy in a compact shell and can interact strongly
with the nucleus. In most cases, the process of electron attachment on anions is
a endothermic, the coming electron is inevitably repelled by the charge already
present in the compact shell.
[205] Koh, W.; Dubois, D.; Kutner, W.; Jones, M. T.; Kadish, K. M. J. Phys. Chem.
1992, 96, 4163-4165.
[206] Mehran, F.; Schell-Sorokin, A. J.; Brown, C. A. Phys. Rev. B 1992, 46, 8579-
8581.
123
[207] Xu, Q.; Liu, H.; Frisch, H. L.; William E.; Broderick, W. E. J. Chem. Phys.
1993, 98, 9648-9650.
[208] Weast R. C.; Astle, M. J. CRC Handbook of Physics and Chemistry 1982 63rd
Ed, CRC Press, Boca Raton, Florida.
[209] Sawyer, D. T.; Sobkowiak, A.; Roberts, J. L. Electrochemistry For Chemists,
2nd ed, 1995, John Willy & Sons, Inc. p.332
[210] Sun, Y.; Bunker, C. Nature 1993, 365, 398.
[211] Sun, Y.; Ma, B.; Bunker, C.; Liu, B. J. Am. Chem. Soc. 1995, 117, 12705
-12711.
[212] Bezmelnistin, V. N.; Eletskii, A. V.; Stepanov, E. V. J. Phys. Chem. 1994, 98,
6665-6667.
[213] Ruoff, R. S.; Tse, D. S.; Malhotra, R.; Donald, C.; Lorents, D. C. J. Phys.
Chem. 1993, 97, 3379-3383.
[214] Kazuhiro, K.; Hirose, T.; Moriuchi, K.; Ken, H.; Araki, T.; Tanaka, N. Anal.
Chem. 1995, 67, 2556-2561.
[215] Ghosh, H. N.; Sapre, A. V.; Mittal, J. P. J. Phys. Chem. 1996, 10, 9439-9443.
[216] Ying, Q.; Maracek, J.; Chu, B. Chem. Phys. Lett. 1994, 219, 214-218.
[217] Ajie, H.; Alvarez, M. M.; Anz, S. J.; Beck, R. D.; Diederich, F.; Fostirooulos,
K.; Huffman D. R.; Kr?Atschmer, W.; Rubin, Y.; Schriver, K. E.; Sensharma, K.;
Whetten, R. L. J. Phys. Chem. 1990, 40, 8630-8633.
[218] Jehoulet, C.; Bard, A. J. J. Am. Chem. Soc. 1991, 113, 5456-5457.
[219] Olsen, S. A.; Bond, A. M.; Compton, R G.; Lazarev. G.; Mahon P. J.; Marken,
F.; Rason, C. L.; Tedesco, V.; Webster, D. J. Phys. Chem. A 1998, 102, 2641-
2649.
[220] Kato, T.; Kodama, T.; Shida, T.; Nakagawa, T.; Matsui, Y.; Suzuki, S.; Shiro-
maru, H.; Yamauchi, K.; Achiba, Y. Chem. Phys. Lett. 1991, 180, 446-450.
[221] Greaney, M. A.; Gorun, S. M. J. Phys. Chem. 1991, 95, 7142- 7144.
[222] Kato, T.; Kodama, T.; Oyama, M.; Okazaki, S.; Shida, T.; Nakagawa, T.;
Matsui, Y.; Suzuki, S.; Shiromaru, H.; Yamauchi, K.; Achiba, Y. Chem. Phys.
Lett. 1991, 186, 35-39.
[223] Thomas, R. N. Anal. Chim. Acta 1994, 289, 57-67.
[224] Catalan, J.; Saiz, J. L.; Laynez, J. L.; Jagerovic, N.; Elguero, J. Angew. Chem.
Int. Ed. Engl. 1995, 34, 105-107.
124
[225] Fukuzumi, S.; Suenobu, T.; Hirasaka, T.; Arakawa, R.; Kadish, K. M. J. Am.
Chem. Soc. 1998, 120, 9220-9227.
[226] Fukuzumi, S.; Nakanishi, I.; Maruta, J.; Yorisue, T.; Suenobu, T.; Itoh, S.;
Arakawa, R.; Kadish, K. M. J. Am. Chem. Soc. 1998, 120, 6673-6680.
[227] Tanaka, T.; Kitagawa, T.; Komatsu, K.; Takeuchi, K. J. Am. Chem. Soc. 1997,
119, 9313-9314.
[228] D?Souza, F.; Choi, J. P.; Kutner, W. J. Phys. Chem. B 1998, 102, 4247-4252.
[229] Cliffel, D. E.; Bard, A. J. J. Phys. Chem. 1994, 98, 8140-8143.
[230] Dielectric Constant
[231] Yang, X.; Wang, X-B.; Niu, S.; Pickett, C. J.; Ichye, T.; Wang, L-S. Phys. Rev.
Lett. 2002, 89, 163401-1(4).
[232] Yang, X.; Wang, X-B.; Wang, L-S. J. Phys. Chem. 2002, 106, 7607-7616.
[233] Leiter, K.; Ritter, W.; Stamatovic, A.; M?ark, T. D. Int. J. Mass Spectrom. Ion
Processes 1986, 68, 341-346.
[234] Dougherty, R. C. J. Chem. Phys. 1969, 50, 1896-1897.
[235] Maas, W. P. H.; Nibbering, N. M. M. Int. J. Mass Spectrom. Ion Proc. 1989,
88, 257-266.
[236] Kalchei, J.; Sax, A. F. Chem. Rev. 1994, 94, 2291-2318.
[237] Hotop, H.; Lineberger, C. J. Phys. Chem. Ref. Data (a) 1975, 4, 539, (b) 1985,
14, 731
125