Investigations of the electrode-solution interface in microheterogeneous solutions involving surfactants

MISSING IMAGE

Material Information

Title:
Investigations of the electrode-solution interface in microheterogeneous solutions involving surfactants
Physical Description:
xii, 145 leaves : ill. ; 28 cm.
Language:
English
Creator:
Boyette, Stacey E., 1964-
Publication Date:

Subjects

Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1991.
Bibliography:
Includes bibliographical references (leaves 139-144).
Statement of Responsibility:
by Stacey E. Boyette.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001693356
notis - AJA5435
oclc - 25222045
System ID:
AA00003714:00001

Full Text









INVESTIGATIONS OF THE ELECTRODE-SOLUTION INTERFACE IN
MICROHETEROGENEOUS SOLUTIONS INVOLVING SURFACTANTS












By

STACEY E. BOYETTE


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


1991













ACKNOWLEDGEMENTS

Praise God for giving me the ability and the opportunity to pursue a higher

education. May I serve Him well.

Thanks go to my parents and my brother, Rodney, who have suffered and

celebrated with me from childhood to adulthood. Their love and support is

invaluable.

Thanks go to my friends scattered up and down the East coast who are too

many to name but too precious not to mention.

Special thanks go to the chemistry faculty at Georgia Southern University for

their friendship and encouragement throughout this past year. I look forward to

working with them next year.

Last of all, thanks go to the members, past and present, of the small but

powerful Toth group, especially S.A. Myers, Mike Freund, Allan Witkowski, Shi-Min

Zhu, Liakat Bodalbhai, and Dr. Anna Brajter-Toth.

The funding for my work came from a variety of sources, including U.S.

Army Chemical Research, Development, and Engineering Center, the Division of

Sponsored Research at the University of Florida, the National Institutes of Health,

and Battelle (Research Triangle Park, NC).














TABLE OF CONTENTS
page

ACKNOWLEDGMENTS .................................. ii

LIST OF TABLES ....................................... vi

LIST OF FIGURES ...................................... viii

ABSTRACT ............................................ xi

CHAPTERS

INTRODUCTION ................. ........... 1

Theories of Graphite Activity ..................... 1
Purpose of Proposed Work .................. 4
Surfactants and Micelle Systems in Electrochemistry ..... 5
Effects of Surfactants and Micelles on
Electrochemical Parameters ................ 5
Practical Considerations in Micellar
Electrochemistry ......................... 6
Methods of Quantitating Analyte-Micelle Interactions .... 7
Surfactant Adsorption at Solid-Liquid Interfaces ........ 9
Interfacial Parameters ...................... 9
Adsorption Mechanisms .................. ....... 9
Surfactant Molecules at Hydrophilic Surfaces .......... 9
Surfactant Molecules at Hydrophobic Surfaces ......... 10
Systems Studied ............................... 16
Probes ................................. 16
Surfactants .............................. 19
Electrodes .............................. 19

I. EXPERIMENTAL SECTION ..................... 23

Reagents .................................... 23
Electrochemical Apparatus and Procedures ........... 24
Electrode Preparation ...................... 24
Electrochemical Methods ................... 25
Fundamentals of Electrochemical Measurements .. 27














Determination of Equlibrium Binding Constants (Keq)
from Electrochemical Data ...................... 29
Chromatographic Apparatus and Procedures .......... 34
HPLC System ............................ 34
Choice of Detector Wavelength Maxima ........ 34
Procedures in Micellar Chromatography ......... 35
Micellar Chromatography Theory ............. 38

III. ELECTROCHEMICAL BEHAVIOR OF PROBE
MOLECULES ON GRAPHITE ELECTRODES ...... 42

Results and Discussion for Rough Pyrolytic Graphite .... 43
Probe Adsorption ......................... 43
Electron Transfer Kinetics ................... 53
Electrode Reactivity ....................... 53
Conclusions ............................. 61
Results and Discussion for Glassy Carbon ............ 61
Electrode Reactivity ....................... 61
Probe Adsorption ......................... 66
Electron Transfer Kinetics ................... 70
Conclusions ............................. 70

IV. RESPONSE OF HYDROPHILIC ELECTRODES IN
SURFACTANT SOLUTIONS ABOVE THE CRITICAL
MICELLE CONCENTRATION .................. 72

Effects of pH and Ionic Strength ................... 72
Deactivation of RPG Surface ................. 76
Electrode Reactivity ....................... 76
Probe Adsorption ......................... 81
Electron Transfer Kinetics ................... 82
Surface Model ................................ 83
Attractive Interactions Between Probe Molecules and
Adsorbed Surfactants .......................... 84
Electrode Reactivity ....................... 85
Probe Adsorption ......................... 86
Electron Transfer Kinetics ................... 87
Conclusions .................................. 88














V. BINDING CONSTANTS FOR PROBE-MICELLE
INTERACTIONS ............................. 89
Equilibrium Binding Constants (Keq) Determined
from HPLC Data ............................. 89
Comparison of Experimental and Literature Keq Values .. 94
Equilibrium Binding Constants (Keq) Determined
from Electrochemical Data ...................... 98

VI. RESPONSE OF HYDROPHILIC ELECTRODES IN
SURFACTANT SOLUTIONS BELOW THE CRITICAL
MICELLE CONCENTRATION .................. 102

Deactivation of RPG Surface ...................... 102
Electrode Activity ......................... 102
Electron Transfer Kinetics ................... 107
Surface M odel ................................ 109
Electrode Reactivity and Electron Transfer
Kinetics ............................... 110
Probe Adsorption ......................... 111
Conclusions .................................. 113

VII. RESPONSE OF HYDROPHOBIC ELECTRODES IN
SURFACTANT SOLUTIONS ................... 115

Probe Adsorption .............................. 116
Electrode Reactivity ............................ 116
Surface M odels ................................ 121
Electron Transfer Kinetics ........................ 124
Surfactant Adsorption ........................... 128
Keq Determined from Electrochemical Data ........... 131
Conclusions .................................. 133

VIII. SUMMARY AND FUTURE WORK ............... 135

BIBLIOGRAPHY ...................................... 139

BIOGRAPHICAL SKETCH ................................ 145














LIST OF TABLES
page


Table 1. SCAN RATE STUDIES: Slopes of log peak current vs
log scan rate plots at RPG electrodes in the presence
of surfactant concentrations above CMC; tdip = 0 min. ... 52

Table 2: Anodic to cathodic peak-to-peak potential separations
(AEp inmV) at RPG electrodes in the presence of
surfactant concentrations above CMC; v = 200 mV/s .... 54

Table 3: SCAN RATE STUDIES: Slopes of log peak current vs
log scan rate plots at polished GC electrodes in the
presence of surfactant concentrations above and below
CMC; t = 0 min. ............................ 69

Table 4: Anodic to cathodic peak-to-peak potential separations
(AEp in mV) at polished GC electrodes in the presence
of surfactant concentrations above CMC; v = 200 mV/s. .. 71

Table 5. Average anodic peak potentials (Epa) and normalized
average anodic peak currents (ia) for pH studies at
RPG electrodes of 0.54 M dopamine in phosphate buffer
(p = 0.5); v = 200 mV/s. ......................... 74

Table 6. Average anodic peak potentials (Ea) and normalized
average anodic peak currents (i.a) for ionic strength
studies at RPG electrodes of 0.54 M dopamine in phos-
phate buffer (pH 7); v = 200 mV/s .................. 75

Table 7. Micellar HPLC Data ............................. 91

Table 8: Probe-micelle association (binding) constants for: a)
the probes studied in this work, b) hydroquinone, c)
selected catechols, d) selected polar organic molecules,
and e) selected cyclic aromatic compounds (i.e., nonpolar
organic molecules). ............................. 95














Table 9: Average shifts in formal potential at RPG electrodes
over the range of exposure times from 0 to 10 min for
probes in surfactant concentrations above CMC and
equilibrium constants for the reduced and oxidized
forms of the probes with surfactant aggregates;
T = 25(1)OC; v = 200 mV/s. ..................... 99

Table 10: Anodic to cathodic peak-to-peak potential separations
(AEp in mV) at RPG electrodes in the presence of
surfactant concentrations below CMC; n = 200 mV. ...... 108

Table 11: SCAN RATE STUDIES: Slopes of log peak current
vs log scan rate plots at RPG electrodes in the
presence of surfactant concentrations below CMC. ...... 112

Table 12: Anodic to cathodic peak-to-peak potential separations
(AEp in mV) at polished GC electrodes in the presence
of surfactant concentrations below CMC; n = 200 mV. ... 125

Table 13: Average shifts in formal potential at polished GC
electrodes over the range of exposure times from
0 to 10 min for probes in the presence of surfactant
concentrations: a) above CMC and b) below CMC and
equilibrium constants for the reduced and oxidized
forms of the probes with surfactant aggregates;
T = 25( 1)C; v = 200 mV/s. ..................... 132













LIST OF FIGURES
page

Figure 1: Surface model for the adsorption of surfactant molecules
at hydrophilic surfaces, such as RPG in aqueous media,
for: a) low surface coverage, b) increased surface
coverage, and c) hemimicelle formation ............... 12

Figure 2: Surface model for the adsorption of surfactant molecules
at hydrophobic surfaces, such as GC in aqueous media,
for: a) low surface coverage, b) increased surface
coverage, and c) the unlikelihood of reverse hemimicelle
formation. ............................ 14

Figure 3: Structure of probe molecules at pH 7. ............... 18

Figure 4: Structure of surfactant molecules. .................. 21

Figure 5: Cyclic voltammograms of 0.2 mM DAPOL at RPG vs
SCE in pH 7 phosphate buffer (&p=0.5M) containing:
a) no surfactant, b) 3.080 mM CTAB, c) 9.36 mM SDS,
and d) 2.253 mM Triton X-100; exposure time to analyte
solution tdi = 0 min; v = 200 mV/s; 20pA scale;
electrode area (RPG) = 4.4 (1.2) x 10-2 cm2.......... 45

Figure 6. Cyclic voltammograms of 0.5 mM dopamine at RPG vs
SCE in pH 7 phosphate buffer (pi = 0.5 M) containing:
a) no surfactant, b) 3.043 mM CTAB, c) 9.26 mM SDS,
and d) 3.196 mM Triton X-100; exposure time to analyte
solution, tdip = 0 min; v = 200 mV/s; 20 iA scale;
electrode area (RPG) = 4.4 (1.2) x 10-2 cm 2......... 47

Figure 7. Cyclic voltammograms of 0.5 mM DOPAC at RPG vs
SCE in pH 7 phosphate buffer (i = 0.5 M) containing:
a) no surfactant, b) 3.147 mM CTAB, c) 9.71 mM SDS,
and d) 2.624 mM Triton X-100; exposure time to analyte
solution, ti = 0 min; v = 200 mV/s; 20 pA scale;
electrode area (RPG) = 4.4 ( 1.2) x 10-2 cm .......... 49













Figure 8. Cyclic voltammograms of 0.3 mM hydrdoquinone at RPG
vs SCE in pH 7 phosphate buffer (p = 0.5 M) containing:
a) no surfactant, b) 3.013 mM CTAB, c) 9.47 mM CTAB,
and d) 3.421 mM Triton X-100; exposure time to analyte
solution, td4 = 0 min; v = 200 mV/s; 20 pA scale;
electrode area (RPG) = 4.4 (t1.2) x 10-2 cm2 ..........

Figure 9: Normalized peak current (ipa/ACo) vs exposure time to
analyte solution (tdi) for the four probes at RPG
electrodes; pH 7 phosphate buffer (IL = 0.5 M);
v = 200 mV/s; probe concentrations: 0.2 mM DAPOL,
0.5 mM dopamine (DA), 0.3 mM hydroquinone (HQ),
and 0.5 mM DOPAC. ...........................


Figure 10:





Figure 11:




Figure 12:






Figure 13:


Normalized peak current (ipa/ACo) vs exposure time
(tdp) for the second oxidation peak (IIa) of DAPOL:
a) m the absence of surfactants, b) in SDS concen-
trations above CMC, c) in SDS concentrations below
CMC; phosphate buffer, pH 7, p = 0.5 M; v = 200 mV/s.

Cyclic voltammograms of 0.3 mM hydroquinone at RPG
(0.0382 cm2) and GC (0.0728 cm2) vs SCE in pH 7
phosphate buffer (Ip = 0.5 M); Exposure time to analyte
solution tdip = 0 min; v = 200 mV/s; 20 pA scale. ......

Normalized peak current (ipa/ACo) vs exposure time
to analyte solution (tdi) for the four probes at
GC electrodes; pH 7 phosphate buffer (p = 0.5 M);
v = 200 mV/s; probe concentrations: 0.5 mM
dopamine (DA), 0.3 mM hydroquinone (HQ), 0.2 mM
DAPOL, and 0.5 mM DOPAC .....................

Normalized peak current (ip/ACo) vs exposure time
(tdi) for: (a) dopamine (DA), (b) DAPOL, (c) DOPAC,
and (d) hydroquinone (HQ) at RPG electrodes in phos-
phate buffer (pH 7, Ip = 0.5 M) and in the indicated
surfactant at concentrations above CMC; v = 200 mV/s. ..














Figure 13:






Figure 14:





Figure 14:






Figure 15:





Figure 15:


(continued) Normalized peak current (i /ACo) vs
exposure time (tdi) for: (a) dopamine (DA), (b) DAPOL,
(c) DOPAC, and (d) hydroquinone (HQ) at RPG
electrodes in phosphate buffer (pH 7, pI = 0.5 M)
and in the indicated surfactant at concentrations
above CMC; v = 200 mV/s .......................

Normalized peak current (ia/ACo) vs exposure time (tdP)
for: (a) dopamine (DA), (b) DAPOL, (c) DOPAC, and (d)
hydroquinone (HQ) at RPG electrodes in phosphate buffer
(pH 7, p = 0.5 M) and in the indicated surfactant at
concentrations below CMC; v = 200 mV/s ...........

(continued) Normalized peak current (i a/ACo) vs
exposure time (tdi) for: (a) dopamine (DA), (b) DAPOL,
(c) DOPAC, and (d) hydroquinone (HQ) at RPG
electrodes in phosphate buffer (pH 7, p = 0.5 M)
and in the indicated surfactant at concentrations
below CMC; v = 200 mV/s........................

Normalized peak current (ia/ACo) vs exposure time (ti,)
for: (a) dopamine (DA), (b) DAPOL, (c) DOPAC, and (d)
hydroquinone (HQ) at GC electrodes in phosphate buffer
(pH 7, p = 0.5 M) and in the indicated surfactant at
concentrations above and below CMC; v = 200 mV/s. ...

(continued) Normalized peak current (ia/ACo) vs exposure
time (tdi) for: (a) dopamine (DA), (b) DAPOL, (c) DOPAC,
and (d) hydroquinone (HQ) at GC electrodes in phosphate
buffer (pH 7, p = 0.5 M) and in the indicated surfactant
at concentrations above and below CMC; v = 200 mV/s.


80





104






106





118





120













Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy


INVESTIGATIONS OF THE ELECTRODE-SOLUTION INTERFACE IN
MICROHETEROGENEOUS SOLUTIONS INVOLVING SURFACTANTS

By

Stacey E. Boyette

August, 1991

Chairman: Dr. Anna Brajter-Toth
Major Department: Chemistry

The purpose of this research was to develop new modified carbon electrode

surfaces for the detection of small biological molecules. Carbon electrodes are

frequently used in electroanalytical chemistry, and rough pyrolytic graphite (RPG)

and glassy carbon (GC) electrodes were chosen for this work. Graphite electrodes,

like other solid electrodes, are prone to memory effects and frequently exhibit poor

sensitivity or poor reproducibility. Many methods have been developed to improve

the reactivity of graphite electrodes by direct modification of the electrode surface,

such as pretreatment methods, resurfacing techniques, and surface coating

procedures. In this work, a new method of indirect modification of the electrode

surface by manipulating the solution environment was developed.








Surfactants, at concentrations above and below CMC, were used to modify

analyte solutions. No systematic studies of the effects of surfactants on probe

responses have been reported, although the use of surfactants in electrochemistry

is not a new concept. The objective of this work was to provide a systematic study

which would allow predictions of the effects of a given surfactant on a particular

probe response.

The effects of surfactants on- the electrochemistry of dopamine,

hydroquinone, 2,6-diamino-8-purinol (DAPOL), and 3,4-dihydroxyphenylacetic acid

(DOPAC) were investigated at RPG and GC electrodes. All of these probes

undergo two electron/two proton oxidation in the same potential window and have

known adsorption properties on graphite surfaces. At pH 7, two of the probes,

hydroquinone and DAPOL, are neutral, while the other two probes, dopamine and

DOPAC, are charged. Dopamine is positively charged and DOPAC is negatively

charged. To investigate interactions at the electrode-solution interface, anionic

(sodium dodecyl sulfate (SDS)), cationic (hexadecyltrimethylammonium bromide

(CTAB)), and nonionic (Triton X-100) surfactants were used.

Surfactants adsorb at RPG and GC surfaces and attractive interactions

between probe molecules and adsorbed surfactants enhance probe responses at both

electrodes. Interactions between specific probe-surfactant pairs observed from the

electrochemistry were verified and quantified by micellar HPLC results.













CHAPTER I

INTRODUCTION

Theories of Graphite Activity


Graphite surfaces have found many applications in analytical chemistry.

Consequently, the characterization of graphite surfaces has become a major topic

of interest, especially with respect to the relationship between structure and

reactivity. Graphite electrodes, including glassy carbon and pyrolytic graphite, are

widely used in electroanalysis, although the specific parameters controlling electrode

activity are not well understood. Graphite electrodes generally have low residual

currents and are very useful over a large potential range; however, the apparent rate

of heterogeneous charge-transfer is slower for most compounds at carbon electrodes

than at metal electrodes [1-3]. Various pretreatment methods have been devised to

increase the electrochemical activity of graphite surfaces, including heating [2-6],

laser irradiation [7-12], exposure to radio frequency plasmas [13-15], polishing [3,5-

7,16-18], and chemical [19,20], and electrochemical [1,3,7,9,21-28] procedures. Many

explanations for the improvement in electrochemical response of carbon electrodes

following pretreatment have been formulated; some indicate changes in physical

properties of the carbon surface, such as surface roughness and edge plane density,

while others deal with the chemistry of the electrode surface. Although the








2

activation mechanism appears to be different for each pretreatment method, it is

evident that cleanliness of the electrode surface plays a crucial role in graphite

electrode activity [3,4,16,18,29]. Impurities are known to adsorb onto electrode

surfaces and contribute to the deactivation of the surface [8,10,16,18]. One common

effect of pretreatment methods is to remove contaminants and clean the electrode

surface as part of the activation process [4,5,7-10,12,16,18,21,25].

It is frequently proposed that electrode activation is due to an increase in

active sites [3,4,6,30], although what constitutes active sites remains undefined.

Electron Spectroscopy for Chemical Analysis (ESCA) studies of heat-treated [2],

highly polished [18], electrochemically pretreated [21], and RF plasma treated

[13,14] carbon electrodes have shown an increased oxygen concentration or oxygen-

to-carbon ratio on the electrode surface as compared to corresponding untreated

electrodes. Results such as these, along with electrochemical observations, at carbon

surfaces have led many researchers to implicate surface oxygen functionalities, such

as carboxyl and quinoidal groups, in the activation of carbon electrodes [13-

15,18,27,31-34]. However, it has been shown that heat treatments at high

temperatures eliminates surface redox functionalities while activating the electrode

surface [4,6]; thus, surface functionalities are not necessary for rapid electron

transfer. Consistent with increased oxygen concentrations at activated electrodes,

researchers have observed the formation of graphite oxide layers on

electrochemically pretreated carbon electrodes [1,25,28], which can selectively

activate the electrode surface based on permeability and/or ion-exchange properties.








3

Poon and McCreery [12] have demonstrated that laser treatment can remove the

effects of electrochemical pretreatment of glassy carbon and suggest that this implies

removal of an oxygen-rich surface film. Likewise, Hershenhart, McCreery, and

Knight [8] have reported that intense pulsed laser radiation of electrodes removes

polymeric or adsorbed films while creating an active surface.

Laser and other pretreatment methods may activate graphite surfaces by

increasing the edge plane density [3,9,11,22,29,35]. It has been reported that

graphite edge orientation is necessary for fast electron transfer [3,9]. Graphitic edge

planes are hydrophilic while the basal plane orientation is hydrophobic [3,36].

Indirect evidence for electrode activation due to an increase in edge planes can be

found in reported correlations between accelerated electron transfer rates and

increased hydrophilicity (and wettability) upon pretreatment [5,20,21,31,37]. Direct

evidence for graphite activation via increased edge plane density can be found in

Raman and electrogenerated chemiluminescence studies of highly ordered pyrolytic

graphite (HOPG) [9,22,29,35]. Raman studies of HOPG show a correlation between

the intensity of a band at 1360 cm"1, which is proportional to density of edge planes,

and electron transfer activation [9,22,29]. High intensity of the 1360 cm"1 band

corresponds to larger heterogeneous rate constants for laser and electrochemically

pretreated electrodes [9,22,29]. Electrogenerated chemiluminescence studies of

HOPG show enhanced electron-transfer kinetics at regions rich in edge plane

defects compared with regions of undisturbed basal plane HOPG [35]. The

electron-transfer rate constant for luminol oxidation differed by more than two








4

orders of magnitude between the two types of regions [35]. The activation theories

involving high surface oxygen concentrations and edge plane orientation are not in

direct conflict, since it appears that edge planes are the sites for the formation of

oxygen functionalities [3,33,36] and for the nucleation and rapid growth of graphite

oxides [23,34]. In light of the many different surface pretreatments that effectively

activate a variety of carbons, it seems obvious that no single mechanism or property

of carbon surfaces can explain all of the observed electrochemical behaviors [5,38].

More than likely it is a combination of effects that leads to the activation of

pretreated graphite electrodes to heterogeneous charge transfer [28,31,34,38].

Purpose of Proposed Work

The focus of the work presented in this report is not direct characterization

of active graphite electrodes, but rather exploitation of that activity for the detection

of small biological molecules. However, it is expected that the results of this work

will provide insight into the electron transfer and mass transfer processes occurring

at the electrode surface. High electrochemical reactivity for different types of

electron transfer systems such as outer-sphere systems including ferri/ferrocyanide

[2-4,9,10,16-19,22,25], ferrocene [18], and ruthenium and cobalt hexamines [1,19],

and inner-sphere systems including ascorbic acid [1-5,8,10,12,18,31,39], hydroquinone

[10,12,18,25,28], and catechols [1,5,9,10,12,18,24,27,28,38,39] such as dopamine, 3,4-

dihydroxyphenylacetic acid (DOPAC), and 4-methylcatechol, can be observed on

active (or activated) graphite surfaces. Sensitive electrochemical responses on

graphite electrodes are frequently accompanied by adsorption [1,7,24-28,40-42]. As








5

a result, the relationship between adsorption and electrochemical activity has

attracted much attention [6,7,27,41]. One way to investigate the relationship

between probe adsorption and electrode activity is to study the electrochemistry of

adsorbing compounds while controlling the electrode-solution interface. The

electrode-solution interface may be controlled and manipulated by either modifying

the electrode surface itself or by modifying the solution environment of the analyte.

In the work presented here, a new method combining both approaches of controlling

the electrode-solution interface is used; the electrochemistry of adsorbing

compounds on graphite surfaces is investigated in solutions containing surfactants

and micelles.

Surfactants and Micelle Systems in Electrochemistry

Effects of Surfactants and Micelles on Electrochemical Parameters

The use of surfactants and micelles in electroanalytical chemistry has been

well documented in several reviews [43-47]. The concentration and nature of

surfactant used in electroanalysis can affect the shape of the electrochemical wave,

the half-wave potential, electron transfer rates at the electrode, diffusion and

transfer coefficients, limiting currents, and the stability of intermediate species

[44,45]. Low concentrations of surfactants have been used for years as maxima

suppressors or selective masking agents in polarographic analysis [44,45,47-49]. At

higher concentrations, above the critical micelle concentration (CMC), micelles are

formed which have the ability to solubilize water-insoluble or sparingly soluble

species in aqueous media. Largely due to their solubilizing power, micelles have








6

been used in conjunction with ferrocene to form mediator-titrants [50,51], as

stabilizers for electrochemically generated ion radical intermediates [52-57], and to

improve the electrochemistry of organic compounds in aqueous media [58-60].

Ohsawa and Aoyagui [59] demonstrated that it is even possible to determine the

formal standard potential of a water-insoluble substance in aqueous media using

micellar electrochemistry in conjunction with solubility measurements.

Practical Considerations in Micellar Electrochemistry

Although micelles are considered to be relatively simple models for enzymatic

and membrane environments [45,55], they present a rather complex electrochemical

situation. Two fundamental processes must be considered when discussing

electrochemistry in micellar solutions: 1) interaction of analytes and reaction

products with micelles, and 2) interaction of surfactants with the electrode surface.

This situation becomes further complicated if the analyte (or reaction product)

adsorbs at the electrode surface. For example, in their studies of the reductive

electrochemistry of methyl viologen (MV2+) on glassy carbon, Kaifer and Bard [55]

found that the product of one-electron reduction of MV2+, the MV cation radical

(MV+-), adsorbs on glassy carbon in the presence of the negatively charged

surfactant, sodium dodecyl sulfate (SDS), at concentrations below the CMC due to

interaction with SDS molecules adsorbed on the electrode surface. However, at

SDS concentrations above the CMC, adsorption of MV'" is eliminated due to

solubilization of MV+" in SDS micelles in solution.








7

Solubilized species may reside in various regions of a micelle [44-46,61], some

penetrate far into the hydrophobic core, some remain near the micelle-solution

interface around the hydrophilic head groups, and others may orient themselves

somewhere between these two extremes. Although much of the literature indicates

a trend of preferential interaction of ions and ion radicals with oppositely charged

micelles [52-57,62], neither the stability nor the electrochemistry of these species in

micellar aggregates can be predicted based solely on electrostatic considerations [55].

The strength of interactions between electroactive species and micelles [54,62], the

structure [55,60,62] and size [60] of the micelle, as influenced by supporting

electrolyte and solubilized species, and the ability of the electroactive species to

move in and out of the micelles [44,53,54], will all influence the observed

electrochemical response.

Methods of Quantitating Analyte-Micelle Interactions

Several methods for quantitating the strength of interactions between analyte

molecules and micelles have been reported, including fluorescence and

phosphorescence quenching [63-66], electrochemistry in micellar media [44,59,67],

and micellar liquid chromatography [68,69]. Luminescence quenching techniques

are based on a correlation between the magnitude of quenching observed in the

presence of micelles and the entrance and exit rate constants for probes and/or

quenchers in micelles, which are related to the micelle-solute equilibrium association

constants (i.e., binding constants) [63-66]. Binding constants determined from

luminescence quenching data have been reported for a variety of systems including








8

neutral arenes in ionic micelles [63] and transition metal dications in anionic

micelles [65]. Electrochemical methods for determining binding constants between

probes and micelles depend on the relationship between shifts in half-wave

potentials (EM) in micellar solutions compared to those in aqueous solutions and

are determined by the extent of probe-micelle interactions [44,59,67]. The

mathematical relationship between EM and association constants of electroactive

species with micelles, developed by Ohsawa et al. [59,67], will be discussed later.

Micellar liquid chromatography techniques for determining binding constants

are based on the correlation between changes in mobile phase micelle concentration

and changes in probe retention, as measured by capacity factors [68-70]. Arunyanart

and Cline-Love have used micellar liquid chromatography to determine micelle-

solute binding constants for a series of uncharged aromatic solutes [68] and charged

aromatic compounds, such as organic acids and bases, which participate

simultaneously in other equilibria [69]. Foley [70] has compiled over 150 solute-

micelle association (binding) constants for a variety of neutral compounds and

phenylthiohydantoin-amino acids based on micellar liquid chromatography data. In

the work reported here, in addition to the electrochemical measurements, micellar

chromatographic techniques were used to investigate the extent of interaction

between the probe molecules and micelles.










Surfactant Adsorption at Solid-Liquid Interfaces

Interfacial Parameters

Aside from interactions between surfactants and probes, the interactions

between surfactants and the electrode surface must also be considered when dealing

with electrochemistry in surfactant media. A characteristic feature of surfactants is

their tendency to adsorb at solid-liquid interfaces [56,61]. Surfactant adsorption has

been studied to determine: 1) the concentration of surfactant at the interface, since

it is a measure of coverage; 2) the orientation of a surfactant at the interface, since

it indicates whether the interface becomes more hydrophilic or more hydrophobic

as a result of surfactant adsorption; and 3) energy changes, AG, AH, and AS, in the

system, resulting from adsorption, since they provide insight into the type and

mechanism of surfactant interactions at the interface [61].

Adsorption Mechanisms

Possible mechanisms by which surfactants may adsorb onto solid substrates

from aqueous solutions are: 1) ion-exchange; 2) ion pairing; 3) hydrogen bonds or

Lewis acid Lewis base reaction; 4) adsorption by ir- overlap; 5) adsorption by

London-van der Waals dispersion forces; and 6) hydrophobic bonding [61].

Surfactant Molecules at Hydrophilic Surfaces

Surfactants are amphiphilic molecules consisting of a hydrophobic portion

(tail group) and a hydrophilic portion (head group). If an electrode surface is

hydrophilic or has charge opposite to that of the surfactant head group, the

surfactant will bind with the hydrophilic head group oriented toward the surface








10

(Figure 1). At low surface coverage, the hydrocarbon tail groups may coil [61],

bend, or orient skewed to the electrode surface [71] to obtain the most energetically

favorable configuration (Figure l(a)). For alkylbenzene sulfonates on mineral oxide

surfaces, it has been reported that terminal carbons on tail groups having 10 carbons

or more may strongly interact with the surface in addition to the electrostatic

attraction between the charged head group and the surface [72]. As adsorption and

surface coverage increase on hydrophilic surfaces, head groups become more closely

packed and hydrocarbon chains align perpendicular, for the most part, to the

electrode surface (Figure l(b)) [71-74]. Once the binding sites on the surface are

saturated with surfactant molecules, adsorption may continue through the formation

of two-dimensional surfactant aggregates called hemimicelles (Figure l(c))

[61,71,72,75-78]. In the formation of hemimicelles, oncoming surfactant molecules

adsorb with reversed orientation and overlap with already adsorbed surfactants due

to the hydrophobic interactions between tail groups (normal hemimicelles) [61,71].

Surfactant Molecules at Hydrophobic Surfaces

If an electrode surface is hydrophobic; surfactant molecules will adsorb onto

the surface with the hydrophobic tail groups toward the surface and the head groups

toward the aqueous phase (Figure 2) [61]. At low surface coverage, surfactant

molecules may be parallel to the surface, slightly tilted from the surface, or L-shaped

(Figure 2(a)). As adsorption increases, surfactant molecules align more and more

perpendicular to the surface with the head groups oriented toward the solution

(Figure 2(b)) [61,78]. The formation of reverse hemimicelles of 1-decanol at the






























Figure 1: Surface model for the adsorption of surfactant molecules at
hydrophilic surfaces, such as RPG in aqueous media, for: a)
low surface coverage, b) increased surface coverage, and c)
hemimicelle formation.











a)










b)




c)m


c)





Ii "






























Figure 2: Surface model for the adsorption of surfactant molecules at
hydrophobic surfaces, such as GC in aqueous media, for: a)
low surface coverage, b) increased surface coverage, and c) the
unlikelihood of reverse hemimicelle formation.

















7


ill"'


7
^ 7


I


-ii


,ll


m


/


1111iiiiii


I


l J








15

heptane/carbon black interface, due to strong polar interactions between head

groups, has been reported by Zhu and Gu [79]. However, in aqueous solutions, the

formation of reverse hemimicelles is unlikely, especially for charged surfactants,

since head groups will not attract and will interact with the aqueous environment

(Figure 2(c)).

At surfactant concentrations above the critical micelle concentration, normal

micelles, three-dimensional aggregates with the head groups oriented outward from

a hydrophobic core (aqueous media), would also be expected to bind to hydrophilic

surfaces (in addition to hemimicelles), while reverse micelles, with tail groups

oriented outward from a hydrophilic core organicc media), would be expected to

bind to hydrophobic surfaces.

When surfactants and micelles adsorb at the solution-electrode interface,

changes occur in the interfacial tension, the double-layer structure, and the rate of

electron transfer at the electrode [43,44,47]. Several reports have appeared

[48,49,55-58,60,73,74,80,81] which discuss the formation of adsorbed layers on

electrode surfaces in surfactant and micellar solutions and the effects these films

have on electrochemical responses. The effects on surfactants of electrochemistry

have been studied for a wide variety of redox couples and surfactants at different

electrodes, in various potential windows, and for a variety of surfactant

concentrations in aqueous and nonaqueous media. The nonsystematic nature of

these studies makes it impossible to deduce any broad conclusions about

electrochemistry in the presence of surfactants. Systematic investigations of the








16
effects of surfactants on the electrical double layer as well as on the kinetics and

thermodynamics of electron transfer processes are needed before the predictions of

surfactant effects on the electrochemistry of a given probe will be possible.

Systems Studied

In this work, the electrochemical oxidation of four small biological molecules

was investigated at graphite electrodes in aqueous surfactant and micellar media.

Micellar liquid chromatography was performed to verify and quantify probe-micelle

interactions observed in the electrochemical work.

Probes

The four probes studied were dopamine (DA), 3,4-dihydroxyphenylacetic acid

(DOPAC), 2,6-diamino-8-purinol (DAPOL), and hydroquinone (HQ) (Figure 3).

These four probes were chosen on the basis of size, structure, biological significance,

and similarities in electrochemical behavior on carbon surfaces. All four probe

molecules have molecular weights below 250 g/mole. By choosing small molecules,

it was hoped that problems typical for large biological molecules, such as slow

heterogeneous electron transfer, would be avoided. These four probes are

structurally different and provide examples of a positively charged probe (DA), a

negatively charged probe (DOPAC), and two neutral probes (DAPOL and HQ) at

pH 7 (Figure 3). Each of the probes is of biological significance: dopamine is a

central nervous system neurotransmitter [20,24], DOPAC is a metabolite of

dopamine [20], DAPOL is a biological degradation product of 2,6-diaminopurine (a

growth inhibitor of bacterial and mammalian cells) [42], and hydroquinone































Figure 3: Structure of probe molecules at pH 7.














NH- C ONl

OH



Dopamine
3-hydroxytyramine
pKa= 8.92











S-OH

OH



DOPAC
3,4-dihydroxyphenylacetic acid
pKa1=4.22; pKa2=9.58; pKa3=12.15


NJ/

N N






DAPOL
2,6-diamino-8-purinol
pKal=2.0; pKa2=10











Ho- OH





Hydroquinone
1,4-dihydroxybenzene
pKa 1=9.91; pKa2=12.04








19
derivatives participate in electron transport in living organisms ranging from fungi

to vertebrates [82]. Therefore, the detection of these molecules is of interest to the

chemist and biochemist alike. Electrochemically, all four probes undergo a two

electron/two proton oxidation in the same potential window [2,18,25,42,82] and have

known adsorption properties on graphite [2,27,28,42].

Surfactants

The surfactants used to modify the electrochemical behavior of the probe

molecules were chosen based on their charges. Since we have anionic, cationic, and

neutral probes, we selected representative anionic, cationic, and nonionic surfactants

to investigate analyte-surfactant interactions and their effect on the observed

electrochemistry. Sodium dodecyl sulfate (SDS), cetyltrimethylammonium bromide

(CTAB), and Triton X-100 (Figure 4) were chosen as the anionic, cationic, and

nonionic modifiers, respectively.

Electrodes

The electrochemistry was studied at rough pyrolytic graphite (RPG) and

glassy carbon (GC) electrodes. RPG and GC provide two structurally [7,30,34,83,84]

and electrochemically (as prepared here) [5-7,16,21,30] different graphite surfaces

for the evaluation of probe responses and surfactant effects on probe responses.

RPG is a polycrystalline form of carbon manufactured by depositing carbon from

the vapor phase onto a substrate [34]. RPG is characterized by a high degree of

orientation of carbon layers and shows high electrical conductivity in the plane

parallel to the surface of deposition (edge plane) and low conductivity in the plane































Figure 4: Structure of surfactant molecules.












CH3(CH2)110SO- Na*


SDS
sodium dodecyl sulfate








CH3(CH2),1N+(CH3,)Br-


CTAB
cetyltrimethylammonium bromide






C(CH3)3CHC(CH3)2 -O(CH2CH0)9.H



Triton X-100
octyl phenoxy polyethanol








22

perpendicular to the surface of deposition (basal plane) [30,34]. Amorphous GC is

formed by the pyrolysis of phenolic polymers [83] and consists of tangled aromatic

ribbon molecules with strong intra-ribbon carbon-carbon bonds cross-linked by

highly strained (i.e., weaker) inter-ribbon carbon-carbon bonds [30,83]. Activated

graphite surfaces, such as the silicon carbide-polished RPG used here, exhibit

hydrophilic character [5,20,21,31,37] and high electrochemical reactivity [7,30].

Unactivated graphite surfaces, such as the alumina-polished GC used here, exhibit

hydrophobic character [5,37,85] and low electrochemical reactivity [5-7,16,21,30].

Presented here is a systematic study of the effects of surfactant and micellar

systems on the electrochemical behavior of four structurally different probes. The

results provide new insights into the role of surfactant, micelle, and electrode surface

interactions in the electrochemistry of small adsorbing biological molecules and

provide information which may lead to improved electrochemical analysis of these

species.













CHAPTER II

EXPERIMENTAL SECTION

Reagents

A pH 7 phosphate buffer of ionic strength 1.0 M was prepared from

anhydrous dibasic sodium phosphate, Na2HPO4 (Mallinckrodt, Inc.) and monobasic

sodium phosphate, NaH2PO4 x H20 (Mallinckrodt, Inc.; Fisher Scientific Co.) in

deionized-distilled water. The pH of the phosphate buffer was measured using a

Corning pH Meter 130. Adjustments of pH of the buffer solution were made by

adding small amounts of either phosphoric acid, H3PO4 (Mallinckrodt Chemical

Works), or sodium hydroxide, NaOH (Fisher Scientific Co.), as needed. The pH 7

phosphate buffer (u=1.0) was diluted to ionic strengths of 0.5 M for use in

electrochemical experiments and 0.05 M for use in micellar HPLC and UV

spectroscopy experiments. All experiments were run at room temperature, 25

(1)C.

The probe compounds, 3-dihydroxytyramine hydrochloride (dopamine)

(Sigma Chemical Co.), 3,4-dihydroxyphenylacetic acid (DOPAC) (Sigma Chemical

Co.), 2,6-diamino-8-purinol hemisulfate monohydrate (DAPOL) (Aldrich Chemical

Co.), and hydroquinone (HQ) (Chem Service, Westchester, PA), and the surfactants,

sodium lauryl sulfate (SDS) (Fisher Scientific Co.), hexadecyltrimethylammonium








24
bromide (CTAB) (Sigma Chemical Co.), and Triton X-100 (Sigma Chemical Co.),

were used as received. The pK, values for the probe molecules are:

Probe p-- pK gK REF.

dopamine 8.92 [86]

DOPAC 4.22 9.58 12.15 [87]

DAPOL 2.0 10.0 [7]

hydroquinone 9.91 12.04 [88]



The CMC's and aggregation numbers for SDS, CTAB, and Triton X-100 in water

at 25(1)C are:

SDS CTAB Triton X-100

CMC (mM): 2.25' [55] 1.3 [47] 0.24 [89]

Agg. #: 62 [47] 78.0 [47] 140.0 [90]


'determined in 50 mM NaCl.

Electrochemical Apparatus and Procedures

Electrode Preparation

Rough pyrolytic graphite (RPG) and glassy carbon (GC) electrodes were

made by sealing a ca. 0.05 cm2 piece of RPG or a ca. 3 mm diameter rod of GC in

glass tubing with epoxy (Hysol Aerospace and Industrial Products Division). GC

electrodes were initially polished on 600-grit silicon carbide paper (Mark V

Laboratory) using a Buehler Ecomet I Polisher-Grinder before being polished by








25
hand to a glassy finish with gamma alumina suspensions of particle sizes finer than

0.1 pm (Gamal, Fisher Scientific) on an alpha A polishing cloth (Mark V

Laboratory). The electrodes were resurfaced prior to each measurement. After the

final polish, GC electrodes were ultrasonicated for 5-10 min in doubly distilled

water, and gently wiped with a Kimwipe. RPG electrodes were resurfaced on 600-

grit silicon carbide paper using the Buehler Ecomet I Polisher-Grinder, rinsed with

deionized-distilled water, and gently wiped with a Kimwipe.

Electrochemical Methods

Electrode areas were determined using chronocoulometry with a Bioanalytical

Systems electrochemical analyzer, BAS 100, with a Houston Instruments DMP-40

Series Digital Plotter (Bausch & Lomb). The reference electrode was an Saturated

Calomel Electrode (SCE) (Fisher Scientific Co.) and the counter electrode was a 0.8

cm square Pt ribbon (Sargent). The potential was stepped from 0.400 to 0.0 V vs

SCE at a pulse width of 250 ms. Electrode areas were determined in solutions of

4.0 mM potassium ferricyanide, K3Fe(CN)6 (Fisher Scientific Co.), in 1.0 mM

potassium chloride, KCI (Fisher Scientific Co.), using a diffusion coefficient of Do

= 7.63 x 10-6 cm2/s [91]. Solutions were deaerated for at least 5 min with nitrogen

and kept under nitrogen atmosphere during chronocoulometric experiments.

Measurements were taken immediately after placing resurfaced RPG and GC

electrodes in solution. The electrode areas determined by chronocoulometry were

(3.5 5.2) x 102 cm2 for RPG electrodes and (4.0 7.5) x 10.2 cm2 for GC electrodes.








26
Cyclic voltammetry was performed using an EG & G Princeton Applied

Research Model 173 Potentiostat/Galvanostat, EG & G PARC 175 Universal

Programmer, and a Houston Instruments Omnigraphic 2000 Recorder (Bausch &

Lomb). The working electrodes were RPG and GC electrodes, the counter

electrode was a Pt wire electrode, and the reference electrode was an SCE with a

4% agar/KCl salt bridge. All probe solutions consisted of a single analyte dissolved

in a pH 7 phosphate buffer (I = 0.5 M) plus one of the following: 1) a single

surfactant at concentrations below the CMC and 2) a single surfactant at

concentrations above the CMC. Electrode response as a function of time, tdip, was

studied by varying the length of time that the electrode was exposed to solution at

an open circuit before activating the cell. Cyclic voltammograms (CVs) were taken

at exposure times of 0, 2, 5 and 10 min at a scan rate of 200 mV/s in the potential

window -0.5 V to +0.75 V, encompassing the formal potentials for all probes which

lie between -0.12 V and +0.40 V at both RPG and GC electrodes. Voltammetric

peak currents obtained from exposure time studies were normalized for electrode

area and probe concentration. Electrode response as a function of scan rate was

investigated at scan rates of 5, 20, 50, 100, 200, and 500 mV/s in the potential

window -0.5 V to +0.75 V for all probes on RPG and for DAPOL on GC.

Measurements were made at scan rates of 10, 20, 50, 100, and 200 mV/s in the same

potential window as above for DA, DAPOC, and HQ on GC. Scan rate studies

were run at tip = 0 min unless otherwise stated. All sample solutions were

deaerated for at least 5 min with nitrogen and kept under nitrogen atmosphere








27
during electrochemical experiments. All potentials are reported vs SCE and at room

temperature, 25( 1)*C.

Fundamentals of Electrochemical Measurements

Electrochemical parameters, such as cyclic voltammetric anodic peak currents

(ipa), separation of cathodic and anodic peak potentials (AEp, where AE = Ep -

Ep,, with E, and Epa representing the cathodic and anodic peak potentials,

respectively), and slopes of log ipa anodicc peak current) vs log v (scan rate), were

measured at different electrode exposure times to the analyte solution, tdp, in order

to study electrode reactivity, electron transfer kinetics, and surface interactions.

Peak current measurements at different electrode exposure times to the

analyte solution are of interest since changes in peak current (i.e., sensitivity) with

exposure time are indicative of analyte adsorption onto the electrode surface

[7,8,92]. Decreases in peak current (i.e., deactivation of the electrode surface) with

increasing exposure time are associated with adsorption, such as physisorption, i.e.,

relatively weak, long-range adsorption, of a probe onto the electrode surface and

chemisorption, i.e., strong, specific adsorption, which is indicated by an initial

increase in peak current with exposure time [7,8,92].

Measurements of AEp values provide insight into the processes controlling

heterogeneous electron transfer (i.e., diffusion or adsorption) and the reversibility

of those processes. According to theory for reversible (fast) heterogeneous electron

transfer [92], AEp values for a diffusion controlled process will be 59/n (mV), where

n is the number of electrons transferred per mole, and AEp for an adsorption








28
controlled process will be 0. If the electron transfer kinetics are slow (i.e., quasi-

reversible or irreversible), larger than predicted AEp values will be obtained (i.e.,

AEp > 59/n (mV) for a diffusion controlled process and AEp > 0 (mV) for

adsorption controlled process). The probe molecules studied here undergo 2e'/2H+

oxidation [5,7,42,82], so that AEp values of approximately 30 mV are expected if the

oxidation is reversible and strictly diffusion controlled.

Slope values for plots of log ipa anodicc peak current) vs log v (scan rate)

indicate to what degree probe molecules adsorb at the electrode surface. According

to theory for a diffusion controlled electron transfer process [92], the equation

defining cyclic voltammetric peak current for a reversible system is:


ip = (2.69 x 105)n3/AD v n2C* (1)
and for an irreversible system is:


ip = (2.99 x 105)n(ana)l/ACDo/2v1'2 (2)
where n is the number of electrons per mole oxidized or reduced, A is the area of

the electrode (cm2), Do is the diffusion coefficient (cm2/sec), v is the scan rate (V/s),

Co' is the bulk concentration of the electroactive species (mol/cm3), a is the transfer

coefficient, and na is the number of electrons in the rate-determining step. The

equation defining peak current for an adsorption controlled electron transfer process

[92] for a reversible system is:
S 4RT_ A (3)









and for an irreversible system is:


SnanaF2Avr, (4)
ip 2.718RT
where To is the surface excess of the electroactive species (mol/cm2), F is the

Faraday constant, R is the ideal gas constant, and T is the absolute temperature (K).

Based on these equations, plots of log ipa vs log v should be linear and have slopes

values of 0.5 for diffusion (since ip a v2) and 1.0 for adsorption (since ipa a v)

independent of electrochemical kinetics.

Determination of Equilibrium Binding Constants (Kq)
from Electrochemical Data

Differences between the formal potentials observed for probe responses in

phosphate buffer and those observed in micellar media are indicative of differences

in the strength of binding of the reduced and oxidized probes with surfactant

aggregates. The formal potential for a probe, E', is calculated from the equation:


E' = Epc +Epa (5)
2
where Ec and Epa represent the cathodic and anodic peak potentials, respectively.

The equilibria for binding of the reduced and oxidized forms of a probe to a

surfactant aggregate can be expressed as:


Ox(aq) + M Ox-M KO (6)

Red(aq) + M Red-M KR (










and the electrochemical reactions are:


Ox(aq) + e- Red(aq) E o0 (8)


Ox-M + e- Red-M E0M (



where Ox and Red represent the oxidized and reduced forms of the probe,

respectively, Ox-M represents the oxidized form of the probe bound to a surfactant

aggregate, Red-M represents the reduced form of the probe bound to a surfactant

aggregate, and M represents a surfactant aggregate which may be a micelle or a

hemimicelle. In the case presented here, probe molecules are expected to interact

with surfactant molecules adsorbed onto the electrode surface in the form of

hemimiclles. A scheme similar to the one reported by Kaifer and Bard [55] can be

used to more clearly illustrate the relationship between the equilibria expressed in

equations 6-9:


Ox(aq) +e- ,- Red(aq)
+M 11 # +M

Ox-M + e- Red-M



At a given potential, E, the Nernst equation for the redox reaction in phosphate

buffer (denoted by aq) and the redox reaction in micellar media (denoted by M) can

be written:


RT [Oxa ]
E = E 0', + ( RT)ln( aq (10)
nF [Reda]













RT
E = E' + In
SnF


equations (10) and (11) can be combined to yield:


E oM E o, =


RT In
nF


[Oxq]
[Redq]


SRT
nF


In [Ox-M]
[Red-M]


which can be rearranged as follows:


S_ RT
E O'M -E E q RTIn
nF





o, RT
E O' E O' = In
nF


E M E q =


RT
nF


[OXq] [Red-M]
[Red] [Ox-M] '


[Oxaq [Red-M] [M]
Ox-M] [Redq] [M]


[Oaq][M] [Red-M]
[Ox-M] [Redq][M]


and


[Ox-M]
[Red-M]


(11)


(12)


(13)






(14)


and


(15)








32
The equilibrium constants for the binding of the probe forms to a surfactant

aggregate are defined as:


KR = [Red -M]
[Redaq][M]




K [Ox-M]
[OXaq][]


for the reduced form and





for the oxidized form


which can be substituted into equation (15) to obtain the relationship between

changes in formal potential and binding constants:


SRTiKR
E o'M E o, KR
nF Ko


(18)


The probe molecules studied in this work undergo a two electron/two proton

oxidation and reduction, so that equation (18) becomes:


RT K1R[H+ 2
E 'M E' 2F K [H ]2
aq 2F Ko[H ]


(19)


Since the pH is constant throughout the redox reaction, the proton concentration

terms can be neglected yielding:


E O' E RT In KR (20)
a 2F KO


(16)


(17)








33
Differences between the formal potentials observed for probe responses in

phosphate buffer, E0~q, and those observed in micellar media, EO', provide a means

of evaluating the strength of interaction of the oxidized probe with surfactant

aggregates relative to the strength of interaction of the reduced probe with

surfactant aggregates. The probe molecules studied in this work are purchased in

the reduced form, so that the Kq calculated from HPLC data, which will be

discussed in this chapter in the Micellar Chromatography Theory Section, are the

binding constants for the reduced form of the probe, KR. Shifts in formal potential,

EO Eo'q, obtained from the electrochemical data indicate the strength of probe

interactions with surfactant aggregates in the form of hemimicelles, while

equilibrium constants obtained from HPLC data indicate the strength of probe

interactions with micelles. The strength of probe binding with a surfactant is

thermodynamically independent of the structure of the aggregate, however the

number of surfactant molecules in an aggregate may affect the amount of probe that

interacts with the aggregate. Assuming that the aggregation number for a

hemimicelle is the same as that for a micelle, shifts in formal potential, EO& EO,

and the binding constants for the reduced probe, KR, can be used to calculate the

binding constant for the oxidized probe, Ko, from equation (25). Since Ko can only

be calculated relative to KR in this work, the absolute numerical values of Ko and

KR are less important than the relative values of Ko and KR for comparing strengths

of interactions of the oxidized and reduced probes with surfactant aggregates.










Chromatographic Apparatus and Procedures

HPLC System

A Macintosh computer-controlled Rainin gradient HPLC system (Rainin

Instruments Co., Inc.) with ultraviolet-visible detection was used to perform micellar

chromatography. A Rheodyne sample injection valve (20 til sample loop), a

Spectroflow 757 Absorbance Detector (Applied Biosystems Inc.), and a Fisher

Recordall Series 5000 chart recorder (Houston Instruments) were included in the

HPLC set-up. The columns used for these experiments were reversed-phase

Microsorb 5 pm C18 columns (15 cm x 4.6 mm i.d.) with guard columns (1.5 cm x

4.6 mm i.d.) (Rainin Instruments Co., Inc.). A precolumn (15 cm x 4.6 mm i.d.)

packed with silica gel (Adsorbosil Silica, 200/425 Mesh) was located between the

pumps and the sample injector to saturate the mobile phase with silica and minimize

dissolution of analytical column packing. A filter between the injector and the

analytical column was used to protect the guard column and the analytical column

from precolumn silica particles which may be drawn out by mobile phase. All

experiments were performed at a flow rate of 1.0 ml/min with the column at room

temperature, 25(+1)*C.

Choice of Detector Wavelength Maxima

Ultraviolet (UV) spectra for each probe were recorded on a Tracor Northern

TN-6500 diode array spectrophotometer to determine absorbance maxima and the

stability of probe solutions. HPLC detector wavelength settings were chosen based

on these spectra. Solutions used to record UV spectra contained a single probe in








35
pH 7 phosphate buffer (0.05 M) and a single surfactant at a concentration above

CMC. The phosphate-SDS buffer solutions contained 0.050 M SDS and the

phosphate-CTAB solutions contained 6.0 mM CTAB. Solutions were prepared just

prior to initial data acquisition and were studied over a period of time up to 5 hrs.

From the UV spectra obtained, a detector wavelength of 280 nm was selected for

the detection of dopamine, DOPAC, and HQ and a wavelength of 246 nm was

selected for the detection of DAPOL.

Procedures in Micellar Chromatography

Micellar mobile phase solutions were prepared by dissolving SDS and CTAB

in deionized-distilled water. SDS mobile phases which contained 0.020, 0.040, 0.060,

and 0.080 M SDS were prepared. CTAB mobile phases which contained 2.00, 4.01,

6.00, 8.00, and 10.1 mM CTAB were prepared. All mobile phases were filtered

through 0.45 tim Nylon 66 membrane filters (Alltech Associates, Inc.). SDS mobile

phases were filtered three times before use and the CTAB mobile phases were

filtered twice to ensure that all sizable particles were removed from solution. HPLC

sample solutions were prepared fresh daily and contained a single probe in pH 7

phosphate buffer (0.05 M) and either 0.050 M SDS or 6.0 mM CTAB.

SDS and CTAB micellar HPLC was performed on two separate columns.

Initial attempts at performing micellar HPLC with mobile phase surfactant

concentrations slightly above CMC were plagued by irreproducible retention times

for consecutive probe solution injections. Retention times continually increased for

injections of probes known to interact with the C18 stationary phase. These changes








36
in retention times were interpreted as indicating adsorption of surfactant molecules

from the mobile phase onto the stationary phase which eventually leads to complete

coverage (i.e., saturation) of the stationary phase with surfactant molecules. In

chromatography, analyte molecules partition between the mobile phase and the

stationary phase. In the case presented above, the stationary phase is constantly

being altered by surfactant adsorption onto the column, while the properties (e.g.,

surfactant concentration) of the mobile phase remain constant. As the amount of

surfactant adsorbed on the column increases, the retention times of probes that

interact with this surfactant increase due to an increase in "surfactant" stationary

phase sites. In the case of probes that interact with the C18 stationary phase,

retention times decrease as surfactant adsorption increases due to a decrease in

available C18 stationary phase sites. Gradual increases in column pressure during

the above experiments also indicated that the column was being "loaded" with

surfactant molecules. As surfactant molecules adsorb onto the stationary phase, the

inside diameter of the column is decreased, causing an increase in column back

pressure. In order to obtain reproducible results, it was decided that columns

should be saturated with the mobile phase surfactant prior to experiments. The

column to be used with SDS mobile phases was saturated with SDS molecules by

flushing the column with a 0.060 M SDS solution in 10% methanol until

reproducible retention times for HQ injections were obtained (over 120 mL of this

solution were passed through the column). The column to be used with CTAB

mobile phases was saturated with CTAB molecules by flushing the column with a








37

10.07 mM CTAB aqueous solution until reproducible retention times for DAPOL

injections were obtained (over 120 mL of this solution were passed through the

column). HQ and DAPOL are not strongly retained on the SDS and CTAB

columns, respectively, so that injections of these samples provide a quick check for

column saturation.

Solubility experiments showed that CTAB was soluble in methanol and

sparingly soluble in water while SDS was soluble in water and insoluble in methanol.

Thus, column loading with CTAB occurs on a reasonable scale in terms of time and

materials in an aqueous solution, while it is necessary to add methanol to the SDS

"loading" mobile phase to ensure efficient saturation of the column with SDS. While

using the columns daily, the SDS column was rinsed with and stored in methanol

overnight and the CTAB column was rinsed with and stored in water overnight.

The SDS column was not flushed with water between experiments since this

procedure would remove some of the loaded SDS from the column. The CTAB

column was not flushed with methanol between experiments for the same reason.

It is usually undesirable to store columns in water for extended periods of time,

especially when buffers are used, since this provides an environment favorable for

the growth of microbes. However, storing columns in water that are used daily does

not seem to degrade column performance over a two to three week experimental

period. When the work on both columns was complete, the SDS column was

flushed with large amounts of water and the CTAB column was flushed with large

amounts of methanol in an effort to remove some of the surfactant molecules from








38
the columns. Total removal of the surfactants from the columns was not possible.

Methanol was used for long term storage of both columns.

Micellar Chromatography Theory

In micellar HPLC, changes in probe retention with changes in mobile phase

surfactant concentration can be used to determine the strength of the interactions

between probes and micelles. The calculation of equilibrium constants for analyte-

micelle interactions from micellar HPLC data depends on a linear relationship

between the reciprocal of the capacity factor, 1/k', and the strength of the mobile

phase, as determined by the concentration of surfactant (i.e., micellar aggregates)

in the mobile phase, [Mm] [68,70]. As discussed in the previous section, the

stationary phase in micellar chromatography is covered with surfactant molecules.

Therefore, a probe that interacts with the surfactant will partition between the

stationary phase and the mobile phase due to the presence of surfactant molecules

in both phases. Since the column is saturated with surfactant, the concentration of

surfactant molecules on the stationary phase will remain effectively constant

throughout the experiment, while the mobile phase surfactant concentration may be

changed by changing mobile phase concentration. Changes in probe retention with

changes in mobile phase strength (i.e., concentration) are due to an increase or

decrease in the number of sites of interaction (i.e., micelles) in the mobile phase.

The magnitude of the change in retention time (in the form of 1/k') with increasing

mobile phase strength (i.e., molar concentration of surfactant in the mobile phase,

[Mm]) provides a measure of probe-micelle interaction.








39
Three reversible equilibria are involved in reversed-phase micellar

chromatography [68,70]: 1) interaction between probe in bulk mobile phase, Pm,

and stationary phase sites, S,, to form the complex PS,; 2)interaction between probe

in bulk mobile phase, Pm, and surfactant molecules in micellar aggregates in the

mobile phase, Mm, to form the complex PMm; and 3) transfer of probe in a complex

with a micelle, PMm, to stationary phase sites to form the complex PS,.



K1 (21)
Pm + Ss PS,





K2 (22)
Pm + Mm PMm





K3 (23)
PMm + Ss PS, + Mm

The subscripts denote the location of a species in the mobile phase (m) and the

stationary phase (s). K, and K2 are the respective binding constants for the probe

with the stationary phase and the mobile phase, while K3 describes the exchange

equilibrium for the probe between the mobile and stationary phases.








40
The capacity factor is a measure of probe retention and is given by k' = (tR -

t.)/t, where tR is the retention time for a given probe and t. is the retention time

of the solvent. The capacity factor can be expressed in terms of the above equilibria

as:


k' -- [Ss]K1
(1 + K2[Mm)
where 4 is the phase ratio (i.e., the ratio of the volume of stationary phase to the

volume of the mobile phase in the column) and [Mm], expressed in moles/Liter, is

calculated from the equation [Mm] = [surfactant] CMC [68,70]. Notice that K3 is

absent from equation (24). It is assumed that probes bind to stationary phase sites

as shown in equation (21). The interaction between probe-micelle complexes and

the stationary phase as shown in equation (23) is considered to be negligible [68,70].

Therefore, equilibrium (23) is not significant in the retention mechanism of probe

molecules. Equation (24) can be linearized as:


[Mm]K2 1
1/k [MM + 1 (25)
O[SJ]K, [Sjkl


A plot of 1/k' vs [Mm] should yield a straight line with a slope of K2/I[S,]K and an

intercept of 1/1[Sj]K1 from which K2 can be calculated. K2 (eqn. 22) is defined as

the binding constant for the interaction between the analyte and a single surfactant

molecule that is part of a micellar aggregate in the bulk mobile phase. Therefore,

the equilibrium constant for the interaction between the analyte and the entire








41
micelle, Kq, is calculated by multiplying K2 by the aggregation number for the given

surfactant.

K = K x Agg. # (26)













CHAPTER III

ELECTROCHEMICAL BEHAVIOR OF PROBE MOLECULES ON
GRAPHITE ELECTRODES



The focus of this work is the study of surfactant effects on the reactivity of

graphite electrodes for small biological materials. In order to elucidate the effects

of surfactants on the electrochemical behavior of analyte molecules, it was necessary

to first characterize the response of the analytes in the absence of surfactants.

Therefore, the electrooxidation of dopamine (DA), DOPAC, DAPOL, and

hydroquinone (HQ) was investigated at rough pyrolytic graphite (RPG) and glassy

carbon (GC) electrodes in pH 7 phosphate buffer.

The electrochemical results obtained here confirm other reports that these

molecules adsorb from pH 7 buffered aqueous solutions on activated graphite

electrode [2,7,24-28,42] and do not adsorb on unactivated graphite electrodes [5,7].

Cyclic voltammetric experiments on RPG electrodes yielded sharp, symmetrically

shaped peaks, small separations between cathodic and anodic peak potentials (AEp),

and slopes of log ia vs log v significantly above 0.5, all of which indicate probe

adsorption on the electrode [92]. Cyclic voltammetric experiments on GC electrodes

yielded broad peaks, large AEp values, and slopes of log ip, vs log v slightly below

0.5, all of which indicate slow electron transfer with no adsorption. The








43

electrochemical behavior of the probe molecules on these two surfaces is discussed

in further detail below.

Results and Discussion for Rough Pyrolytic Graphite

In Figures 5a, 6a, 7a, and 8a, representative cyclic voltammograms are shown

which illustrate the peak shapes indicative of adsorption [92] and give a qualitative

look at AEp and sensitivity of probe responses on RPG. Cyclic voltammograms of

DAPOL (Figure 5a) show two oxidation peaks. It has been reported [84] that the

first oxidation peak (Ia) is controlled by a combination of diffusion and adsorption,

while the second oxidation peak (II) is controlled by relatively strong adsorption

(AGRed = -8.3 kcal/mol) of the reactant. Quantitative results and discussion of the
electrochemistry of DAPOL in this report will concern the first oxidation peak

unless otherwise specified, since the peaks observed for the other three probes are

controlled by the same combination of diffusion and adsorption as the processes at

this peak. Dopamine (Fig. 6a), DOPAC (Fig. 7a), and hydroquinone (Fig. 8a)

exhibit one sharp oxidation peak and the corresponding reduction peak in the

potential window studied.

Probe Adsorption

Slope values of log ipa vs log v plots for the probe molecules on RPG in pH

7 phosphate buffer (0.5M) are shown in Table 1. Slope values for all four probe

molecules are significantly larger than 0.5 (Table 1), the theoretical value for strictly

diffusion controlled electron transfer, indicating contributions from adsorption to

probe responses. As shown in Table 1, the responses of DAPOL and hydroquinone









|t







.dII


U2#
la
d 813
Igs .




5 *.a


'C II
ois










0*O
sa^ +1







4: II
he v
9 szi ^-



1* 11
CLf ^d



o ars



Pv 5


"E
O^O

ga





11
U^
I^lip
M ~o
&' c
MS g

sE|






























































































C
fca
i i-----


"-
0
u









CS
4..

a Q
3^0"

-ri-
11 8 +i
V- Ca
II
.0 ^

Sso


z i^ 8

asl
1
aD T







o t
a f



iis
^UII
Sam


g 11
a8



aE
" 1.s
2c -i



Oa <
' 2.&




vd
S
Er





























































































0r
'I











9F1
:S









ci s


U*









to E











46
11 1
08











ag l
ag







t o



a '

dt4 S

ge ||

lOf
a1


(^^

t ^

/4E
'~c




















































































h
P




I
O












oeb





4 1




h^E



scs|


W0
m

o8
ae


2 "a





| "-



o .r





-3


*8
B '
*3M






























































































0



















Table 1. SCAN RATE STUDIES: Slopes of log peak current vs log scan rate plots
at RPG electrodes in the presence of surfactant concentrations above CMCa; tp =
0 min.

Solution Dopamineb DAPOLb Hydroquinoneb DOPACb

Bufferc 0.62(0.01) 0.72(0.01) 0.69(0.02) 0.65(0.02)

SDSa 0.56(0.02) 0.64(0.02) 0.44(0.01) 0.46(0.01)

CTABa 0.51(0.03) 0.59(0.03) 0.58(0.02) 0.57(0.02)

Triton X-100 0.47(0.01) 0.57(0.01) 0.46(0.01) 0.45(0.01)




aSurfactant concentrations above CMC: 10 mM SDS, 3mM CTAB, 3mM Triton X-
100.
bProbe concentrations: 0.5 mM dopamine, 0.2 mM DAPOL, 0.3mM hydroquinone,
0.5 mM DOPAC.
CAll solutions in phosphate buffer, pH 7.0, ji = 0.5 M.








53

(HQ) exhibit the strongest adsorption character (i.e., largest slope values of log ip

vs log v plots) on RPG.

Electron Transfer Kinetics

The AEp values for the probes under same conditions as in Table 1 are shown

in Table 2. The AEp values for the probe molecules, all of which undergo 2e"/2H+

oxidation, [5,7,42,82] are very close to 30 mV (Table 2), implying reversible,

diffusion controlled electron transfer. However, the slopes of log ipa vs log v plots

discussed above show that adsorption is involved in the electrochemical processes

for these probes on RPG. The fact that AEp values are larger than 0 (the

theoretical value for fast, adsorption controlled electron transfer) indicates that the

observed electron transfer is a quasi-reversible (i.e., slower than theoretical) process.

Very large AEp values would indicate very slow (i.e., irreversible) electron transfer.

As shown in Table 2, DAPOL exhibits the fastest electron transfer kinetics of all the

probes at tip = 0 min (i.e., AEp = 14(4) mV), while the kinetics for DOPAC are

also very fast (i.e., AEp = 25(5) mV).

Electrode Reactivity

When anodic peak currents, ipa, of the probes are normalized for the

geometric electrode area and for the probe concentration (Figure 9), different

normalized responses are obtained for different probe molecules even though the

oxidation of all probes is a two electron process and the kinetics of oxidation of all

probes are similar and relatively fast, which can be concluded from similar peak-to-

peak separations shown in Table 2. A number of reports [1,18,19,36,38] have shown
















Table 2: Anodic to cathodic peak-to-peak potential separations (AEp in mV) at
RPG electrodes' in the presence of surfactant concentrations above CMCb; v = 200
mV/s.
tdp Surfactantb Dopamine DAPOLd Hydro- DOPAC
(min) quinone
0 N 32 (8) 14 (4) 30 (4) 25 (5)
2 O 43 (5) 26 (4) 32 (2) 22 (2)
5 N 35 (5) 30 (0) 30 (4) 28 (12)
10 E 40 (0) 30 (8) 40 30
0 C 68 (4) -- 50 (0) 33 (5)
2 T 75 (4) -- 53 (5) 37 (5)
5 A 87 (5) -- 70 (8) 37 (9)
10 B 85 -- 70 30
0 S 50 (0) -- 123 (5) 263 (5)
2 D 50 (0) -- 130 (0) 290 (8)
5 S 53 (5) -- 148 (8) 308 (6)
10 50 -- 140 350
0 Triton 128 (6) -- 190 (8) 310 (36)
2 X-100 130 (29) -- 190(22) 325 (23)
5 123 (5) 183 (5) 360 (16)
10 150 ,-- 210 320

aAll solutions in phosphate buffer, pH 7.0, pi = 0.5 M.
bSurfactant concentrations above CMC: 10mM SDS, 3 mM CTAB, 3 mM Triton X-
100.
Ctdi exposure time of electrode to analyte solution at an open circuit.
dAdE values for the first oxidation peak of DAPOL (Peak I).





























Figure 9: Normalized peak current (ip,/ACo) vs exposure time to analyte
solution (tdip) for the four probes at RPG electrodes; pH 7 phosphate
buffer (p = 0.5 M); v = 200 mV/s; Probe concentrations: 0.2 mM
DAPOL, 0.5 mM dopamine (DA), 0.3 mM hydroquinone (HQ), and
0.5 mM DOPAC.







































DAPOL

DA

HQ
DOPAC


0 1 2 3 4


tdip (min.)


1.00








57
a relationship between the charge of a molecule and its electrochemical reactivity

on highly polished and electrochemically treated graphite surfaces. Kovach, Deakin,

and Wightman [1] studied the electrochemistry of anionic and cationic species at

electrochemically oxidized carbon fiber electrodes and found that cations adsorb

onto the electrode and yield larger than predicted (i.e., by geometric area) currents

while anions showed no adsorptive behavior and yielded smaller than predicted

currents. These observations were explained on the basis of an insulating surface

oxide layer formed on the electrode during electrochemical treatment which has

cation-exchange capabilities [1]. Saraceno and Ewing [38] studied the electron

transfer reactions for a series of catechols at electrochemically treated ultrasmall

carbon ring electrodes and observed charge-selective enhancement of oxidation

rates. Voltammetric measurements for cationic catechols were found to be less

reproducible but more Nernstian than those obtained for anionic catechols [38]. To

explain these results, Saraceno and Ewing [38] suggested a surface model involving

charge-specific sites on the electrode which are somehow involved in the electron

transfer process, perhaps as specific adsorption or ion-exchange sites. In the studies

presented here, charge does not appear to be an important factor since both the

cationic and the anionic probes exhibit similar (almost within standard deviation)

AEp and ipa values on polished RPG electrodes (Table 2, Figure 9). In fact, the

neutral probes, DAPOL, exhibits stronger adsorption (Table 1) and higher sensitivity

(Figure 9) than both the cationic and the anionic probes. Hance and Kuwana [6]

have reported that the method of electrode pretreatment largely determines the








58
extent of probe adsorption. However, pretreatment procedures used here are the

same for all experiments and do not provide an explanation for differences in

sensitivity of response from probe to probe. Differences in diffusion coefficient

values cannot explain the differences in normalized peak currents since the diffusion

coefficients for the probes are very similar.

Probe Do x 106 (cm2/s) Conditions Ref.
DA 6.05 (0.25) 0.1 M (pH 7.4) [93]
phosphate buffer
DOPAC 5.93 (0.17) 0.1 M (pH 7.4) [93]
phosphate buffer
DAPOL 6.04 (0.02) 0.5 M (pH 7.0) [94]
phosphate buffer
HQ 8.5 0.1 M KCI, 0.0005 M [95]
H2SO4
7.4 0.1 M KNO3 [95]



A plausible explanation for differences in probe responses on RPG is that the

adsorbing molecules pack differently on the electrode surface, i.e., effectively

different areas are available for the reactions of different probes. This explanation

is consistent with the observation of different slope values of log ipa vs log v for each

of the four probes (Table 1) and different kinetics (Table 2).

Peak currents of all probes change with the time of exposure of the electrode

to the analyte solution (tdip) (Figure 9). The responses of all probes, with the

exception of the second oxidation peak of DAPOL, are initially high and/or increase

initially but decrease when exposure time exceeds 5 min. The decreasing responses

of dopamine (DA) and DOPAC level off (i.e., reach steady state) between exposure








59
times of 5 and 10 min. while the responses of DAPOL and hydroquinone (HQ) are

still decreasing at exposure times of 10 min. (Figure 9). Kepley and Bard [28]

studied the deactivation of electroactivated graphite and concluded that deactivation

occurs due to extensive electrochemical reduction of an active surface graphite oxide

layer formed during electrochemical pretreatment. The loss of activity observed in

our studies was faster than the reductive deactivation of graphitic oxide observed by

Kepley and Bard [29]. The loss of activity in our studies is also faster than the

deactivation of laser pretreated GC electrodes observed by Poon and McCreery [10]

and of polished GC electrodes as observed by Hu, Karweik, and Kuwana [16], where

adsorption of impurities was blamed for deactivation. Changes in sensitivity with

time indicate that the effective surface area available for reaction of probes is

changing with time. Subtraction of background current measured with time in the

absence of analytes did not eliminate the observed changes in sensitivity indicating

that degradation of response is directly related to the presence of analyte in

solution. The shape of the ip vs tdp curves (Figure 9), the magnitude of the change

in sensitivity with time (Figure 9), and the electron transfer kinetics (Table 2)

depend on the probe confirming that the differences in probe response are due to

different available surface areas (i.e., different probe packing) for reactions of

different probes. It is likely that probe packing at the electrode surface blocks a

fraction of the active area. The exact shape of the ipa vs tdi curves (Figure 9) also

depends on the ionic strength and pH of solution and attempts were made to keep

the solution conditions constant.








60
The observed decrease in effective electrode area (i.e., loss of activity) at

RPG electrodes is relatively fast as shown in Figure 9 and such deactivation of

graphite has been associated with adsorption [7,8]. Deactivation of RPG occurs

without significant effects on the electrochemical kinetics (as measured by AEp) of

the analyte which remain relatively fast (Table 2). Similar behavior has been

observed on other types of graphite where the rate of electron transfer remained

immune to considerable adsorption [7]. Large changes in sensitivity with time, such

as those observed for DAPOL and hydroquinone, indicate that physical adsorption

similar to the intercalation observed by Kepley and Bard [28] on electroactivated

graphite may be responsible for the observed loss of activity. For physisorption,

long-range interactions between the electroactive species and the surface affect the

electrochemical response by increasing the concentration of the species at the

electrode surface and perturbing the potential distribution (i.e., distribution of ions)

near the electrode surface [92]. Intercalation is a type of physisorption where a

species becomes incorporated into the electrode surface environment perhaps by

uptake into an oxide layer on the surface or into defect sites in the electrode

surface. In physisorption, as observed for the first DAPOL oxidation peak and for

each of the other probes, the adsorbed electroactive species "blocks" the active sites

on the electrode from the electroactive species in the bulk solution, thus decreasing

the response of the electrode to the bulk analyte, while the electrochemical kinetics

of the adsorbed species remain fast (Table 2). The slight decrease in electron

transfer kinetics observed with time for DAPOL and DOPAC in the absence of








61
surfactants (Table 2) confirms that active sites are being blocked by adsorbed

probes.

The response of the second oxidation peak of DAPOL (II) increases rapidly

with exposure times up to tp = 2 min where the rate of increase decreases yielding

a characteristic specific adsorption (i.e., chemisorption) isotherm (Figure 10a) [92].

In chemisorption, the electroactive species is preconcentrated at the electrode

surface in a tightly bound layer [92] so that approaching probe molecules interact

with the electrode surface relatively uninhibited by the presence of the adsorbed

species. Therefore, sensitivity of response is sustained at a high level even in the

presence of adsorption.

Conclusions

In summary, probe molecules adsorb at RPG electrodes with different

efficiencies and yield sensitive responses that are characteristic of quasi-reversible,

mixed adsorption-diffusion controlled heterogeneous electron transfer.

Results and Discussion for Glassy Carbon

Electrode Reactivity

Figure 11 shows representative cyclic voltammograms of hydroquinone (HQ)

on RPG and GC electrodes. Qualitative comparison of these curves, in terms of

peak shape, sensitivity of response, and AE, can be generalized to describe the

response of all probes on RPG and GC surfaces. For example, oxidation and

reduction peaks obtained on GC are smaller and much broader than those observed

on RPG and AEp values obtained on GC are much larger than those observed on




























Normalized peak current (i /ACo) vs exposure time (tdip) for
the second oxidation peak (Ia) of DAPOL: a) in the absence
of surfactants, b) in SDS concentrations above CMC, c) in SDS
concentrations below CMC; phosphate buffer, pH 7, i = 0.5
M; v = 200 mV/s.


Figure 10:



















4.00-


E

-i~s
.j.


2.00


1.00


ti, (mlii.


















000
r-


a
m e






0 0
1J


0
eI

s-i
*oS
So


at



so













Pa


C4





























0


0


0


e
o
(u
~---------1








66
RPG for all probe molecules. Quantitatively, normalized anodic peak currents

(Acm'211) on GC electrodes are approximately one-fourth as large as those

obtained on RPG electrodes (Figures 9 and 12). This result is consistent with

previous reports that GC, as prepared here, is a less active surface than RPG

[7,30,96]. Probe responses on GC show very little, if any, change with time (Figure

12) indicating no slow adsorption of these probes on GC. Also, the second

oxidation peak of DAPOL on RPG corresponding to strongly adsorbed DAPOL is

not observed on GC. Scan rate studies (Table 3) confirm that the responses of all

probes at GC electrodes are diffusion controlled.

Probe Adsorption

Slope values for the plots of log ipa vs log v of the probe molecules on GC

in pH 7 phosphate buffer (0.5 M) are shown in Table 3. Scan rates ranged from 5-

500 mV/s for DAPOL and from 10-200 mV/s for DA, DOPAC, and HQ. Slope

values for all probes are below 0.5, the theoretical value for diffusion controlled

processes, confirming that the probes do hot adsorb on GC electrodes. The

equations used to relate peak current and scan rate (equations 1-4) are based on the

assumptions that the entire electrode surface is active and that the heterogeneous

electron transfer kinetics of the analyte molecule are fast or do not change within

the scan rate window [92]. The fact that the slope values are less than theoretically

predicted for diffusion controlled processes indicates that one or all of these

assumptions fails (i.e., only a portion of the entire electrode area is active and/or the

heterogeneous electron transfer kinetics of analytes are sufficiently slow to cause




























Figure 12: Normalized peak current (ipJACo) vs exposure time to analyte
solution (tip) for the four probes at GC electrodes; pH 7 phosphate
buffer (p = 0.5 M); v = 200 mV/s; probe concentrations: 0.5 mM
dopamine (DA), 0.3 mM hydroquinone (HQ), 0.2 mM DAPOL, and
0.5 mM DOPAC.





















- 0O-
Iwc





o.-


o20-


1C di(ii


/DA
HGL
"DAPOL
"DOPAC


0.0 I I I I

4(j^






















Table 3: SCAN RATE STUDIES: Slopes of log peak current vs log scan rate plots
at polished GC electrodes in the presence of surfactant concentrations abovea and
below CMC; td = 0 min.
Solution Dopaminec DAPOLC Hydroquinonec DOPAC'
Bufferd 0.40 (0.01) 0.48 (0.01) 0.45 (0.03) 0.44 (+0.03)
SDSa 0.40 (0.02) 0.47 (0.01) 0.50 (+0.01) 0.47 (0.01)
SDSb 0.42 (+0.01) 0.48 (+0.00) 0.44 (0.03) 0.47 (0.01)
CTABa 0.39 (+0.02) 0.46 (0.01) 0.40 (0.01) 0.45 (+0.01)
CTABb 0.42 (+0.02) 0.45 (+0.01) 0.40 (0.01) 0.49 (0.01)

aSurfactant concentrations above CMC; 10mM SDS, 3mM CTAB, 3mM Triton X-
100.
bSurfactant concentrations below CMC; 1mM SDS, 0.3mM CTAB, 0.106mM Triton
X-100.
CProbe concentrations: 0.5mM dopamine, 0.2mM DAPOL, 0.3mM hydroquinone,
0.5mM DOPAC.
dAll solutions in phosphate buffer, pH 7.0, p = 0.5 M.








70
deviations from linearity in the plots of log ip vs log v at fast scan rates). Slow

kinetics are confirmed by AEp values as discussed below.

Electron Transfer Kinetics

The AEp values for the probes on GC in pH 7 phosphate buffer (0.5 M) are

shown in Table 4. The AE, values for DAPOL are not listed, since the reduction

peak for DAPOL on GC is not well defined. The AEp values for DA, DOPAC, and

HQ are much larger than 30 mV, the expected value for a two electron, diffusion

controlled process, indicating slow electron transfer kinetics for the probes at GC.

DOPAC exhibits the slowest kinetics of all probes on GC. These results are

consistent with the results of Deakin et al. [5] who studied the oxidation of

substituted catechols at alumina-polished (i.e., unactivated) and heat treated (i.e.,

activated) GC electrodes. Deakin et al. [5] obtained voltammetric peak shapes and

reduced heterogeneous rate constants indicative of electrochemically quasi-reversible

processes and found that responses for DOPAC were more irreversible than those

for other catechols studied.

Conclusions

In summary, GC is a less active surface than RPG yielding responses for all

probes which exhibit poor sensitivity and are characteristic of slow heterogeneous

electron transfer with no contribution from adsorption.



















Table 4: Anodic to cathodic peak-to-peak potential separations (AE, in mV) at
polished GC electrodes' in the presence of surfactant concentrations above CMCb;
v = 200 mV/s.
tdp Surfactantb Dopamine DAPOLd Hydro- DOPAC
(min) quinone
0 N 133(21) -- 310(29) 366(69)
2 O 191(5) -- 300(18) 333(57)
5 N 150(22) -- 280 390
10 E 150(22) -- 315 350
0 C 285(7) 147(9) 175(0)
2 T 277(5) -- 130(0) 150(0)
5 A 270 -- 130 160
10 B 275 -- 130 170
0 S 163(5) -- 363 (9) 537 (9)
2 D 150(0) -- 355 (4) 547 (5)
5 S 160 -- 370 560
10 160 -- 365 550

aAll solutions in phosphate buffer, pH 7.0, g, = 0.5 M.
bSurfactant concentrations above CMC; 10mM SDS, 3 mM CTAB, 3 mM Triton X-
100.
td exposure time of electrode to analyte solution at an open circuit.
p values cannot be obtained reduction peak not observed on GC.













CHAPTER IV

RESPONSE OF HYDROPHILIC ELECTRODES
IN SURFACTANT SOLUTIONS ABOVE THE
CRITICAL MICELLE CONCENTRATION

In electroanalysis it is desirable to control the electrode-solution interface so

that the most sensitive and most precise measurements possible are obtained. The

electrode-solution interface may be manipulated (i.e., controlled) by either modifying

the electrode surface or by modifying the probe solution. The electrode surface may

be altered by using various pretreatment methods, by coating the electrode surface

with a polymer, by depositing thin films or monolayers on the electrode surface, or

by varying the form of the electrode material (e.g. solid carbon vs carbon paste

electrodes). Solution modification may be achieved by varying the pH of solution,

varying the ionic strength of solution, using solution additives, or changing solvents

completely. In the work presented here, the electrode-solution interface was

modified using surfactants as solution additives.

Effects of pH and Ionic Strength

Throughout this work it was observed that anodic peak currents and peak

potentials varied for different preparations of a given probe solution in pure

phosphate buffer. All probe solutions were thought to be similarly prepared, i.e.,

with the same probe concentration, same pH, and same ionic strength, however,

various batches of phosphate buffer (prepared by different laboratory personnel)

72








73
were used to make up probe solutions. Slight variations in phosphate buffer

composition can lead to deviations in probe response, therefore, the pH and ionic

strength of the buffer were investigated as possible sources of the observed

variations in probe responses. A narrow range of pH and ionic strength values was

investigated since all buffer preparations were similarly prepared to be approximately

pH 7 and ionic strength 1.0 M. Anodic peak current and peak potential

measurements were found to be sensitive to both pH (Table 5) and ionic strength

(Table 6). In the pH and ionic strength ranges studied, definite trends in the change

of peak potential with increasing pH and ionic strength were observed, while no

patterns for the change in peak current with"increasing pH or ionic strength could

be discerned (Tables 5 and 6). For dopamine, anodic peak potentials become

slightly less positive (i.e., shift by approximately -15 mV) as ionic strength is

increased from 0.1 M to 1.0 M while marked shifts in the negative direction (i.e.,

approximately -45 mV per pH unit) are observed with increasing pH (in the range

from pH 6-8). The observation of a negative shift in anodic peak potentials with

increasing pH is consistent with the Nernst equation for a 2e'/2H' oxidation which

predicts a possible potential shift of up to -60 mV per pH unit for reversible

electron transfer. An effort was made to use phosphate buffer of the same ionic

strength and the same pH throughout the course of this work, however there appear

to be slight variations in buffer composition from one batch to another. Variations

in buffer composition will affect electrochemical results, such as formal potentials

and ipa vs tp plots as shown in Figure 9.


















Table 5. Average anodic peak potentials (Epa) and normalized average anodic peak
currents (ip/ACo) for pH studies at RPG electrodes of 0.54 mM dopamine in
phosphate buffer (IL = 0.5); v = 200 mV/s.

pH td, (min) Epa (mV) ip, (Acm'2M21)
6.0 0 170(8) 1.85(0.35)
2 180(0) 2.38(0.06)
5 177(5) 2.21(0.49)
10 190 2.32
7.0 0 127(5) 2.16(0.29)
2 135(7) 2.33(0.32)
5 137(5) 2.46(0.16)
10 130 2.56
7.4 0 100(4) 2.47(0.19)
2 113(5) 3.06(0.24)
5 107(2) 2.39(0.29)
10 120 1.52
8.1 0 78(2) 1.93(0.31)
2 70(0) 2.36(0.28)
5 81(5) 2.67(0.16)
10 92(2) 1.79(0.29)





















Table 6. Average anodic peak potentials (E,8) and normalized average anodic peak
currents (ip/ACo) for ionic strength studies at RPG electrodes of 0.54 mM
dopamine m phosphate buffer (pH 7); v = 200 mV/s.

;(M) tdp (min) Epa (mV) ipa (Acm'M)
0.1 0 12(+6) 1.61(0.13)
2 17(9) 1.51(0.23)
5 17(5) 1.45(0.14)
10 20 1.46
0.5 0 -6.7(5) 1.54(0.10)
2 1.7(2) 1.82(0.37)
5 -1.2(2) 1.24(0.31)
10 0 1 1.56
1.0 0 -9.2(8) 1.30(0.12)
2 -1.3(4) 1.40(0.30)
5 -5.0(4) 1.55(0.20)
10 0 1.22









Deactivation of RPG Surface

In this study, different surfactants were added to probe solutions at

concentrations above the critical micelle concentration (CMC) in order to modify

the electrode-solution interface in a controlled manner. Representative cyclic

voltammograms for the responses of probe molecules in the presence of surfactant

at RPG are shown in Figures 5b-d, 6b-d, 7b-d, and 8b-d. Effects on probe responses

as observed from anodic peak currents (ip) in micellar solutions include: 1)

decrease in the magnitude of the electrochemical response of all probes compared

to that in phosphate buffer alone (Figures 5-8; Figure 13); 2) effective elimination

of the changes in response with time (Figure 13); and 3) increase in the

reproducibility (i.e., precision) of peak current (ip) measurements (as indicated by

error bars) (Figure 13).

Electrode Reactivity

The normalized peak currents for all probes in the presence of high

surfactant concentrations (i.e., above CMC) are smaller and more constant with time

than those obtained in the absence of surfactants (Figures 9 and 13). One exception

to this behavior is the response of the second oxidation peak (II) of strongly

adsorbing DAPOL in the presence of SDS micelles which is more sensitive, at short

exposure times, than the response in the absence of surfactant and which changes

with time before levelling off around tdip = 2 min to a value lower than that

observed in the absence of surfactant (Figure 10a and Figure 10b). The response









5'.9


-so%
II
^d
.0

II
?*@


1. 0


E5














.5
is .



S, a






O0.0
UN I




'8
ol
&0l tc



a 88



Av
on
eK S




.IS o






oO a,
'ijb
til
3 '0 *


r**
Oas
*gM7
^*a .3


















































(Wpu=/vA


s
0D
en


78












c!
a





o
*S




r











t.







B
E' J
.#











c II











00
")


s da


tt


























g%
PO
I-s










a-



IEI









































































































q








81
of DAPOL peak II levels off around tdp = 2 min in SDS micellar solutions while

the response in the absence of surfactants is still increasing at tip = 2 min indicating

that adsorption equilibrium is reached more quickly in the presence of SDS. In the

presence of CTAB and Triton X-100, the second oxidation peak of DAPOL

becomes distorted and can no longer be measured. These results indicate that time

dependent physisorption and chemisorption of analytes on RPG are inhibited in the

presence of micelles, with the exception of the second oxidation peak of DAPOL in

the presence of SDS. The unique effect of SDS on the response of the strongly

adsorbing molecule, DAPOL, suggests attractive interactions between the probe and

the surfactant, since the probe saturates the electrode surface (i.e., reaches

adsorption equilibrium) more quickly in the presence of surfactant than in the

absence of surfactant. The elimination of changes in probe responses with time in

micellar media (Figure 13) indicates that the activity of the electrode surface is

constant with time and that probe adsorption is less prominent in the presence of

micelles in solution. The dramatic increase in precision of measurement (in most

cases (Figure 13)), in micellar media compared to that in pure phosphate buffer

indicates that the electrode surface is more reproducible for probe response in the

presence of high surfactant concentrations.

Probe Adsorption

The effects of high surfactant concentrations (i.e., above CMC) on the

electrochemistry of the probes at RPG electrodes can also be seen from changes in

the dependence of peak currents (ip) on scan rate (v) (Table 1) and from changes








82
in peak-to-peak separation (AEp) (Table 2). The slopes of log ip. vs log v for all

probes in micellar media (Table 1) are significantly lower than those obtained in pH

7 phosphate buffer (0.5 M) indicating a decrease in adsorption. This observation

is in agreement with the conclusion drawn from peak current measurements with

time that probe adsorption is inhibited in the presence of micelles in solution, with

the exception of DAPOL peak II, in the presence of SDS. Some slope values fall

below the theoretical value of 0.5 for diffusion controlled processes indicating either

that less than the entire electrode area is active or that the heterogeneous electron

transfer kinetics are slow and cause deviations from linearity in log ipa vs log v plots

at fast scan rates.

Electron Transfer Kinetics

The AEp values (Table 2) for the probes in micellar media are larger than

those observed in pure phosphate buffer. Such an increase in AE, values can

indicate a decrease in the rate of heterogeneous electron transfer or a decrease in

probe adsorption on the electrode [92]. For cases in which responses are strictly

diffusion controlled, increases in AEp values clearly indicate a decrease in

heterogeneous electron transfer kinetics. However, for cases in which probe

adsorption is involved it becomes difficult to discern whether increases in AEp values

are due to decreases in probe adsorption or electron transfer kinetics. For the

systems studied here, it has been shown (from slope values of log ipa vs log v) that

probe adsorption is decreased in the presence of micellar solutions, thus explaining

the increases in AEp values. However, it also seems likely (based on the slope values








83
for log ip vs log v that were below 0.5) that the rate of electron transfer may be

diminished to some extent in the presence of high surfactant concentrations.

Surface Model

Several possible explanations for the observed decreases in electrode activity,

in standard deviation of peak current measurements, in probe adsorption, and in

electron transfer kinetics in micellar media must be considered. All of the above

results imply that the RPG electrode surface is deactivated in solutions containing

surfactant molecules at concentrations above CMC. Surfactants commonly adsorb

at solid-liquid interfaces [56,61], and in this case it appears that surfactants adsorb

at the electrode surface and effectively reduce the surface area available for the

adsorption of probe molecules. The likelihood of surfactant adsorption is supported

by the observations that the electrode surface conditions are more constant with

time and are more reproducible (for a given exposure time) in the presence of

surfactants than in pure phosphate buffer.

Surfactants may adsorb at solid surfaces in the form of surfactant monomers,

micelles, or hemimicelles. Activated graphite surfaces are known to have

hydrophilic character [5,20,21,31,37], which would favor adsorption of surfactant

monomers with the hydrophilic head groups oriented toward the electrode surface

and the tail groups oriented toward the solution (Figure la and b). The

hydrophilicity of RPG would also favor the adsorption of surfactants onto the

electrode surface in the form of normal micelles (i.e., three-dimensional aggregates

with the hydrophilic head groups oriented outward from a hydrophobic core).








84
However, there have been no reports of micelle adsorption at solid surfaces

indicating affinity of micellar aggregates for the solution (i.e., aqueous) phase.

Several reports [61,71,75-78] have suggested the adsorption of surfactants onto

hydrophilic surfaces in the form of hemimicelles (Figure Ic). Hemimicelles (normal)

are two-dimensional aggregates which form due to hydrophobic interactions between

surfactant hydrocarbon chains, so that surfactant head groups are oriented toward

both the electrode and the solution with a hydrophobic region in the center (Figure

Ic) [61,71,72]. Evidence for the adsorption of surfactants in the form of

hemimicelles, such as attractive interactions between probe molecules and adsorbed

surfactants and time-dependent surfactant effects at concentrations below CMC, will

be discussed later in this text.

Attractive Interactions Between Probe Molecules and Adsorbed Surfactants

The observed electrochemical responses for the probes in the presence of

surfactants at concentrations above CMC imply that attractive interactions occur

between specific probe-surfactant pairs. For example, enhanced peak currents

(Figure 13) for a given probe in the presence of one surfactant compared to those

observed in the presence of other surfactants, e.g., dopamine in the presence of SDS

(Figure 13a), indicate attractive interactions between the probe molecule and a

specific surfactant. These attractive interactions may be due to electrostatic effects

resulting from the net charges on the probe and surfactant molecules, electrostatic

effects resulting from induced dipoles on the neutral probe molecules and the net

charge on the surfactant molecules, and/or hydrophobic effects.










Electrode Reactivity

Evidence for attractive interactions is shown by an increased response of

negative DOPAC in the presence of the positive surfactant, CTAB (compared to

that in negatively charged SDS solutions) (Figure 13c) and by a similar increased

response for positive dopamine in the presence of the negative surfactant, SDS

(compared to that in CTAB solutions) (Figure 13a). Repulsive interactions were

ruled out as a possible explanation for the observation of less sensitive responses for

charged probes in the presence of like charged surfactants compared to those

obtained in solutions of oppositely charged surfactants based on probe responses in

the presence of neutral Triton X-100 (Figure 13a and c). Low responses similar to

those obtained for negative DOPAC in the presence of negative SDS and for

positive dopamine in the presence of positive CTAB are observed for these probes

in the presence of neutral Triton X-100 (Figure 13a and c). These results indicate

that the decrease in response for probes in the presence of like charged surfactants

is not due to repulsive interactions between probes and adsorbed hemimicelles but

is likely caused by deactivation of the graphite surface, as suggested above.

Increased responses for neutral hydroquinone (HQ) in the presence of

positive CTAB (Figure 13d) and for neutral DAPOL (Peak I) in the presence of

negative SDS (Figure 13b) may suggest attraction between probe and surfactant

molecules due to an induced dipole in the neutral probe molecules or hydrophobic

interactions between probe molecules and adsorbed surfactants. The response of

the second oxidation peak of DAPOL is very poorly defined in the presence of








86
CTAB and Triton X-100 while it is well-defined and quantifiable in the presence of

SDS. Based on the fact that Triton X-100 is the largest surfactant, i.e., has the most

hydrophobic character, and that neutral probes show suppressed responses in the

presence of Triton X-100, it appears that enhanced responses for neutral probes in

the presence of adsorbed surfactants may be due to electrostatic rather than

hydrophobic interactions. The effect of surfactant chain length may also be

important.

Probe Adsorption

The effects of attractive interactions between specific probes and surfactants

on electrochemical responses can be seen from larger slope values for log ip, vs log

v plots (Table 1) for probes in the presence of one surfactant compared to the slope

values obtained in the presence of other surfactants. From slope values of log ip

vs log v (Table 1), it appears that adsorption of HQ and DOPAC is eliminated in

the presence of negative SDS, whereas dopamine and DAPOL exhibit adsorption

character in this medium. The same type of observations can be made from slopes

in positive CTAB micellar media, where adsorptive interactions for dopamine are

negligible (i.e., slope of log ipa vs log v = 0.51) and HQ, DOPAC, and DAPOL

adsorption remains significant (Table 1). Triton X-100 deactivates the RPG

electrodes for all probes except DAPOL, the response of which exhibits adsorption

characteristics on RPG in all media studied here (Table 1). The variations in slope

values of log ip vs log v with micellar media confirm attractive interactions for








87
dopamine and DAPOL with adsorbed SDS and for DOPAC and HQ with adsorbed

CTAB.

Electron Transfer Kinetics

The effects of attractive interactions on heterogeneous electron transfer rates

are evident from smaller AEp values (Table 2) for probes in the presence of one

surfactant compared to the AE, values obtained in the presence of other surfactants.

Small AEp, similar to those obtained in the absence of surfactants, show that the

heterogeneous electron transfer kinetics of DOPAC are essentially unaffected by the

presence of CTAB at concentrations above CMC and that the kinetics of dopamine

are not significantly altered by the presence of SDS at concentrations above CMC.

The kinetics of dopamine are the least sensitive of all of the probe kinetics to the

presence of surfactants at concentrations above CMC (Table 2), since the AE,

values for dopamine in all media are relatively small. Also, AE, values for

dopamine in the presence of SDS and CTAB are relatively small compared with the

AE, values for HQ and DOPAC in the presence of these surfactants indicating that

dopamine interacts to some extent with positively charged CTAB as well as with

negatively charged SDS. Attractive interactions of dopamine with SDS and CTAB

are confirmed by slope values of log ipa vs log v for dopamine in the presence of

SDS and CTAB (Table 1) which are above 0.5, indicating adsorption like behavior

in the presence of the adsorbed surfactants, while the slopes for other probe-

surfactant combinations fall below 0.5, indicating no interaction of probes with the

electrode surface. The most dramatic decrease in electron transfer rates (i.e., largest
X








88
increase in AEp's) for all probes is observed in the presence of neutral Triton X-100

(Table 2), and low slope values of log ipa vs log v similar to those obtained in like

charged surfactants are obtained for all probes in the presence of Triton X-100

(Table 1). Large AEp values and small slope values of log ipa vs log v for all probes

in the presence of Triton X-100 verify that repulsive interactions between probes

and surfactants are not the source of the observed decrease in probe responses in

micellar solutions since the nonionic surfactant alters the electrochemical kinetics

and the adsorption behavior of the probes to the same degree as the surfactant with

the same charge as the probes. Triton X-100 is much larger than SDS and CTAB,

and chain length may contribute to the observed surfactant effects.

Conclusions

The results discussed above suggest that the situation at the RPG electrode

surface in pH 7 phosphate buffered micellar media is the following: 1) surfactants

adsorb at the electrode surface, most likely in the form of hemimicelles, and

effectively deactivate the electrode; 2) specific probes interact with adsorbed

surfactants and are attracted to the RPG surface; and 3) the probes which

attractively interact with adsorbed surfactants exhibit responses indicative of fast

electron transfer and adsorption behavior.