Cyclic Voltammetry Simulation and Analysis of Reaction Mechanisms David K. Gosser, Jr. Le ateDavid K. Gosser, Jr. CUNY, The City College of New York Chemistry Department 138th Street and Convent Avenue New York, NY 10031 This book is printed on acid-free paper. @ Library of Congress Cataloging-in-Publication Data Gosser, David K. Cyclic voltammetry : simulation and analysis of reaction mechanisms David K. Gosser, Jr. pom Includes bibliographical references and index. ISBN 1-56081-026-2 1. Voltammetry. 1. Title QDI116.V4G67 1994 541.37-de20 93-2440 CrP © 1993 VCH Publishers, Inc, ‘This work is subject to copyright. All rights reserved, whether the whole or part of the material is concerned, specifically those of translation, reprinting, re-use of illustrations, broadcasting, reproduction by photocopying, machine or similar means, and storage in data banks. Registered names, trademarks, etc., used in this book, even when not specifically marked as such, are not to be considered unprotected by law. Printed in the United States of America ISBN 1-$6081-026-2 VCH Publishers Printing History: 1098765432 Published jointly by VCH Publishers, Inc. VCH Verlagsgesellschaft mbH VCH Publishers (UK) Lid. 220 East 23rd Street P.O. Box 10 11 61 8 Wellington Court New York, New York 10010 69451 Weinheim, Germany Cambridge CBI 1HZ United KingdomTo YuyingPreface Cyclic voltammetry (CV) has been in the forefront of the study of electron transfer and its consequences. With the cyclic voltammetric method one can simultaneously activate molecules by electron transfer and probe subsequent chemical reactions. The cyclic voltammetric response curve thus provides information about electron transfer kinetics and thermodynamics as well as the consequences of electron transfer. This book introduces cyclic voltammetry and its application to the analysis of electrochemical reaction mechanisms. It also provides the experimentalist with a simulation-based approach for the analysis of cyclic voltammograms. Chapter 1 presents a brief summary of electrochemical principles. Emphasis is on a conceptual approach to the reduction potential and electrode kinetics. Chapter 2 introduces experimental and conceptual aspects of cyclic vol- tammetry. The relationship between electrode kinetics, chemical kinetics, and diffusion is explored, and the important concept of electrochemical reversibility is discussed. Chapter 3 is a survey of the use of CV for the study of reaction mechanisms in organic, organometallic, inorganic, and pharmacological chemistry. Chapter 4 introduces the method of simulation by explicit finite differences, the most commonly employed numerical method for CV analysis. Chapter 5 describes CVSIM, a general program for the simulation of cyclic voltammetric experiments. A general program for the simulation of double potential step experiments, DSTEP, is also provided. Chapter 6 describes CVFIT, a program that combines CVSIM with a viiviii PREFACE simplex driver to find the least-squares best fit between experimental and simulated cyclic voltammograms. CVSIM, CVFIT, and DSTEP are included on the diskette with this book. They require the use of a PC-compatible (80 x 86) computer. A math coproces- sor is not required, but it is recommended. I welcome questions or comments from those utilizing the simulation software (E mail : Gosser @ sci.ccny.cuny.edu). Acknowledgments Many thanks go to Philip Rieger, James Rusling, and Brenda Shaw, for their encouragement of my interest in simulation. Acknowledgment is due to Qindong Huang and Feng Zhang, my co-workers at City College, for their many contributions to the work represented here. Helpful discussions with Ron Birke, my colleague at City College, are gratefully acknowledged. 1 would like to thank the researchers who have used the CVSIM program, especially Dwight Sweigart, Jay Kochi, Carlo Nervi, and Christian Amatore, whose comments have been particularly helpful. Finally, I would like to thank Edmund Immergut, of VCH, for his valuable advice and assistance. David K. Gosser, Jr. May 1993Contents Useful Equations xi 1. The Reduction Potential and Electrode Kinetics 1.1 The Reduction Potential 1 1.2 Electrode Kinetics 12 References 24 . The Cyclic Voltammetric Experiment 27 2.1 An Overview 28 2.2 The Electrochemical Cell 30 2.3 Electrochemical Mechanisms: E&C Notation 35 2.4 Distortions of the Faradaic Response 56 2.5 Microelectrodes and Fast Scan Voltammetry 59 2.6 Potential Step Methods and Cyclic Voltammetry 60 2.7 Construction of a Fast Potentiostat 64 2.8 Determination of the Number of Electrons 68 References 68 A Survey of Electrochemical Mechanisms 71 3.1 The CE Mechanism 72 3.2 Multielectron Transfer 75 3.3 Protonations at Equilibrium 77 ix 1ee CONTENTS 3.4 Catalytic Mechanisms 80 3.5 The Reduction of Nitrobenzoic Acid 89 3.6 Reduction of the Nit'rosonium Cation and Its Complexes 92 3.7 Reactivity of 17-, 18-, and 19-Electron Tungsten Complexes 95 3.8 Mechanisms Involving Adsorption 97 References 102 4. The Simulation of Electrochemical Experiments 105 4.1 The Discretized Diffusion Equation 105 4.2 Evaluation of the Boundary Conditions 108 4.3 Dimensionless Units 109 4.4 Solution Chemical Kinetics 110 4.5 A Sample Simulation Program 110 References 114 CVSIM: A General Program for the Simulation of Cyclic Voltammetry Experiments 115 5.1 An Overview of CVSIM 115 5.2 Extensions of the Simulation Method 117 5.3 Accuracy of Simulations 120 5.4 Installation and Use of CVSIM and CVGRAF 123 5.5 Examples 125 References 135 a 6. CVFIT: Simplex Data Analysis with CVSIM 137 6.1 CVFIT: Simplex Data Analysis with CVSIM 137 6.2 Instructions for the use of CVFIT 138 6.3 A Prototype Analysis: The EC Mechanism 139 6.4 Some Final Comments on Simulation Analysis 145 References 148 Appendix: Summary of Instructions for CVSIM, CVGRAF, CVFIT, and DSTEP 149 Index 153Useful Equations AG® = — FE® = AGyia, + 5AGgoyy + constant Nernst Equation RT E=E-—| nF no Butler-Volmer Equation = =anF (p — £0] — (1a a)nF Og po" ant {exero[ a ERE | Cnet ap EE) k° = standard heterogeneous rate constant (cm/s) a = cathodic transfer coefficient 1 — a = anodic transfer coefficient, 8 Cyclic Voltammetry (CV) E, (reversible one-electron transfer) F}e i, = 0.4463 FA Fa C,,D¥2 v4? Ene + E, 50" = Foe + Epa 2 , JE, © 58mV xixii E,C, Mechanism RT RTC kebemRT =E_ chem E, = E — 0.280 + 5 mn me | E,as {adsorption mechanism} T, = surface coverage (mol/cm?) USEFUL EQUATIONSCHAPTER 1 The Reduction Potential and Electrode Kinetics The most striking feature of this process is the ease with which a doubly or trebly charged ion gives up a third or fourth electron. We know that this process takes place at the enormously high temperatures in the center of a star, but it is at first a little surprising to find it happening so readily at room temperature. Ronald W. Gurney, on the oxidation of Fe(II) to Fe(ti1), in Ions in Solution.‘ The primary event in most cyclic voltammetric (CV) experiments is the oxidation or reduction of a solution chemical species at an electrode. In this chapter we focus on physical and chemical interpretations of this process. 1.1 The Reduction Potential 1.1.1 The Fermi Level and the Reduction Potential Consider what happens when an electrode is placed in contact with a solution containing the oxidized and reduced species of a redox couple: for example, Fe(CN)3~ and Fe(CN)$~. Before immersion into the solution, electrons in the electrode occupy a continuous range of free energies, according to the conduction band model (Figure 1-1). The electron that is easiest to remove is at the top of the conduction band. This energy is termed the Fermi level energy. The free energy change associated with the removal of an electron from the electrode is equal and opposite the Fermi level free energy.* The Fermi level is also the electrochemical * The Fermi level is approximately equal (and opposite in sign) to the work function, the amount of energy required to remove an electron from bulk material to vacuum. Usually the Fermi level is expressed in units of electron volts (eV); the electron volt is defined as the energy required to move an electron across a potential of one volt. 12 CYCLIC VOLTAMMETRY Fermi Level =» Partially occupied valence Fully occupied valence FREE ENERGY —> Fully occupied tnner shel Figure 1-1. The conduction band model. Conduction is possible when partially filled valance electrons can be raised in energy by external ficlds. potential of the electron in the conducting material. In thermodynamics, the electrochemical potential is defined as the change in free energy per unit species with everything else kept constant: eG Hetec(€, conductor) = — e Ir p.... The units of the Fermi level free energy, usually expressed in electron volts, can be interconverted with other units that are more useful in other contexts (LeV x Ny = 96.485kJ/mol = 23.06 kcal/mol = 8066 cm~'). Now let us turn to the process in solution, the ferricyanide/ferrocyanide transition. The free energy change associated with the transfer of an electron to an oxidized species in solution can be viewed in the context of a thermodynamic cycle (Figure 1-2), The total free energy is seen to be composed of: 1. The free energy associated with transfer of the oxidized species from solution to vacuum. This is the negative of the solvation energy, —AG,yye(ox)- 2. The free energy associated with the transfer of an electron to the oxidized species in vacuum (electron attachment), AG,;... This is approximately the electron affinity (EA) of the oxidized species or the negative of the ionization potential (IP) of the reduced species. 3. The free energy associated with transfer of the reduced species from vacuum to solution, AG,oiveirea)- We have: AGays = AGarar. + AGsoivitea) — AGsorv(ox) (a) The subscript abs denotes that the reference level for energy is established by the electron in vacuo (the “absolute” scale). If the oxidized and reduced species are at unit activity (at T= 298.15°K) then the free energy of reduction is aTHE REDUCTION POTENTIAL AND ELECTRODE KINETICS 3 © vac OX yoo ==> Red vac Ox, Red os Figure 1-2 Thermodynamic cycle for reduction in solution. standard free energy and has the usual relationship to the standard reduction potential (also referred to absolute scale), AG3,, = —nF Exp. The existence of the oxidized and reduced forms of a redox couple thus determine a kind of Fermi level in solution, [see Pleskov and Gurevich? for a statistical mechanical derivation]. The Nernst equation for the half-reaction, referenced to vacuum, is then given by*: RT [Red] — po RT, [Red] Exe = Ele Se "Tog Q) A Le Chatelier “concentration” effect raises or lowers the solution Fermi level by the relative concentrations of the reduced and oxidized forms. When an electrode is placed in a solution containing the oxidized and reduced species, the Fermi levels tend to equalize. This is achieved through electron transfer between the electrode and either of the redox species. Electron transfer from the reduced species to the electrode will raise the Fermi level of the electrode and lower that of the solution. Electron transfer from the electrode to the oxidized species will lower the Fermi level of the electrode and raise that of the solution. The total Fermi level energy is usually separated into a chemical part (the innate Fermi level of the metal) and an electrical part (energy as a result of charging). Thus we can write Float = FLinnae — Fe where ¢ is the electrical potential of the electrode. At equilibrium, the total free energy change must be zero for electron transfer: the Fermi levels are equal. In general, both Fermi levels might be expected to move to reach equilibrium. In practice, however, the electrode potential can be controlled (as in the CV experiment) and the solution must adjust (Figure 1-3). Alternatively, if the electrode potential is not controlled, the solution species can determine the electrode Fermi level. It is useful to emphasize the distinction between the electrical potential ¢, and the Fermi level or the electrode potential, E. The electrode potential consists of both the chemical and electrical energy associated with bringing an electron to * The concentration terms in Equation (2) should actually be written as activities4 CYCLIC VOLTAMMETRY a 2 le ew» (WT m Figure 1-3 Equilibration of the Fermi levels of the electrode and solution. The electrical potential }, of the electrode, is due to charging of the electrode. It is this process that changes the Fermi level of the electrode from its innate value. the electrode, whereas the electrical potential is that part of the energy due solely to electrostatic effects. 1.1.2 The Hydrogen Reference and the Absolute Potential The preceding discussion used the vacuum level as the reference, to emphasize the relationship between gas phase properties such as IP and EA. Electrochem- ical measurements, however, must utilize a solution-based reference. The primary reference is based on the hydrogen ion reduction in aqueous solution: Fiea.uni for which the potential is taken as E° = 0 volts. The potential of any other half- reaction can in theory be obtained by constructing a complete cell with the hydrogen reaction as the anodic (oxidation) reaction and another half-reaction as the cathodic (reduction) reaction (Figure 1-4). All redox reactions are thus described by their tendency to undergo reduction relative to the hydrogen ion. For instance, the standard potential for the reduction of Fe(CN)3~ to Fe(CN)$~ is E° = 0.356, and the reaction: Fe(CN)3 (aq) + H,(g) > Fe(CN)¢"(aq) + 2H *(aq) is favorable. vty $07 FHaGg.1um EL = 0 voltsTHE REDUCTION POTENTIAL AND ELECTRODE KINETICS 5 V2 Ha <> H* +0" Ox + © <=> Red Figure 1-4 Hypothetical determination of E. Note that a difference in electrode potentials is determined. Since the Fermi levels (FL) of the metals in contact must equilibrate, the difference in electrode potentials is transmitted exactly to a difference in lead electrical potentials, if the leads are of the same material. (FL—F)py2)—(FL—F $)pury = (FL—F$)cu2y— (FL —F b)eurry = F(b1 — $2) What is the relationship between the vacuum (or absolute) potential reference and a reference based on the hydrogen ion reduction? This is the tendency (i.c., the free energy change) for an electron to move from vacuum to reduce the hydrogen ion in aqueous solution. This quantity is very important in relating electrochemical measurements and gas phase measurements or spectroscopic measurements, and it has been measured to be about 4.42 V; however, values around 4.8 V have also been reported.?~ 5 Some physical insight into the origin of the absolute potential is obtained with a thermochemical cycle for hydrogen reduction (Figure 1-5). The individual steps are (1) desolvation of H*, (2) electron attachment in vacuo to give the hydrogen atom, and (3) formation of molecular hydrogen in the standard state. The system of standard reduction potentials is described with respect to aqueous solutions. Much electrochemical work involves nonaqueous solutions, and special consideration must be given to measurements in nonaqueous media.* In cyclic voltammetric studies it is common to report the reduction potential of a chosen standard, such as the ferrocinium ion, under the conditions of the experiment. The reduction potential of such a large ion is not expected to change appreciably with solvent. 1.1.3 Practical Reference Half-Cells In practice, electrode reactions other than hydrogen ion reduction are used to construct practical reference systems. Potentials determined using these half- cells can be related back to the hydrogen reference or absolute potential scale if6 CYCLIC VOLTAMMETRY 0,356 vs Hydrogen “AG : ic Feccnyg”* Hvac Wve 44S HH, “AG ev “AG oe 2 Gite eeccecee ec a Ho oa : eV - FE 0 Figure 1-5 Thermodynamic cycle for hydrogen ion reduction. desired. The saturated Ag/AgCl and Hg/Hg,Cl, (calomel electrode) are com- monly used secondary standards. The use of saturated solutions keeps the chloride concentrations constant and thus fixes the potential. AgCl(s) +e” =Ag(s)+ Cl“ (aq, sat KCl) E° = 0.199V Hg,Cl,(s)+2e> = 2Hg(l)+2CI-(aq, satKCl) — E° = 0.244V The potential of a complete cell is the difference in potential between the cathodic and anodic cell half-reactions: Eon = E, — Ey RT [Red] — po RT E, = E oF In Tox] og RT Rel a= a OF” TOx] where the concentration ratios indicate general reaction quotient terms. The relationship between the concentrations and the potential is given by the Nernst equation for the complete cell: RT E~E°-——in (0 ® where Q is the reaction quotient: aA + bBsD In this case a formal potential can be written which also takes into account E=E°12 CYCLIC VOLTAMMETRY the presence in enol groups of dissociable protons,'° with pK,, = 4.10 and pK,, = 11.79. The resulting expression for the formal potential is: RT 1 RT Fun Yee yO ceete ty EEE eee Eeerd eee eereeeeere eee [Pere na (Pe cent “lary + TAI + n.] oF” 1D] 2F where Fina = (H2A] + [HA~] + [A?7] 1.2 Electrode Kinetics Electron transfer reactions at the electrode may not be rapid enough to maintain equilibrium concentrations of the redox couple species near the electrode surface. It is therefore necessary to consider the kinetics of the electron transfer process. The rate equation for heterogeneous electron transfer (Equation 8) expresses the flux of electrons at the electrode surface (Figure 1-9): i SPA 7 FLONautuce — K-iLRedurtce ®) where A is the area of the electrode (cm). With a centimeter-based system, the heterogeneous rate constant is expressed in centimeters per second, and concentrations are given in moles per milliliter (M x 1073). 1.2.1 Transition State Theory for Electron Transfer The main task of electrochemical rate theory is to promote an understanding of the physical basis of the heterogeneous rate constants k, and k_,. Heterogeneous electron transfer can be envisioned with the aid of transition state diagrams. First consider the forward reaction, the transfer of an electron from the electrode to the oxidized species in solution adjacent to the electrode. A transition state diagram will in the first place reflect the reactant and the product free energy. The reactants are the oxidized species in solution and the electron in Figure 1-9 The flux of electrons at an electrode surface is described as moles of electrons per second through a surface area of one square centimeter.THE REDUCTION POTENTIAL AND ELECTRODE KINETICS 13 the electrode. We have seen that the free energy of the electron is equivalent to the electrode potential. Thus the reactant side of the transition state diagram (TSD) reflects the potential of the electrode. The product side of the TSD is the free energy of the reduced species in solution. The transition state for electron transfer must depend in some way on the physical and chemical changes that occur as a consequence of the electron transfer. We have already seen that solvation energy changes are a significant part of the process of electron transfer. The transition state is a nonequilibrium solvated state somewhere between the oxidized and ‘reduced forms. Often chemical or structural changes in the molecule itself occur as a consequence of electron transfer. In such cases, we might expect these changes to be reflected to some degree in the transition state as well, Transition state diagrams are sketched in Figure 1-10 for different electrode potentials. The effect of changing the potential is reflected partially in the energy of the transition state. The curves in the TSD series of Figure 1-10 are related to the free energy relationships that have often been observed by physical organic chemists for a homologous series of chemical reactions (Figure 1-11). Such homologous reaction series often follow linear free energy relationships (LFER), wherein in the following type of relationship between rate constants and equilibrium constants is observed: Inke/kecer,] = a In[keg/keqirets] (9a) In[k,/kecen.] = Bln[keg/keatrer] (9b) where the rate and equilibrium constants of a series of homologous reactions are compared to a reference reaction of the series. From the constants k,, = 1 [Keg and k;/k, = keq it can be easily shown that B = (1 — a). Em + Ox Red Figure 1-10 Transition state diagrams for reduction at different electrode potentials.14 CYCLIC VOLTAMMETRY The fundamental equation of transition state theory is: k = Zexp[—AG'/RT]. Together with the rate/equilibrium relationships this leads to linear free energy relationships for the forward and reverse reactions: AG} — AG, = a[ AG — AG ec] AG} — AGH, = BIAG' — AG ier] Inan LFER, the transition state is interpreted as resembling to some degree the reactants and products. The degree of resemblance is given by the coefficients « and p. The transition state resembles to some degree both the reactants and the products. The degree of resemblance is given by the coefficient in the LFER (Equation 9). Changes in the free energy of the reactants or the products are reflected as fractional changes in the transition state energy. In the series of organic reactions, the free energy is changed by substituent effects. 1.2.2 The Butler-Volmer Equation The situation in electrochemistry is unique. The free energy of the reactants is changed without any change in the overall reaction (without any chemical change in the reactants or products). It is not surprising that electrochemical kinetics have long been described by an LFER that is called the Butler—Volmer equation. If a reference rate is chosen as the rate when E = E°, the forward and reverse rates are equal (k, = k, = ko), and we have: AGorwara = AGEs + AF LE — E°] AGeverse = AGher — [1 — a] FLE — B°] Sometimes a is referred to as the cathodic or forward transfer coefficient and B = (1 — a) as the anodic or reverse transfer coefficient. From transition state theory the rate is exponentially related to the free energy of activation, so that i iF 1 F a: = ba {caeap| - aE - =| ~ Cero GME Booer (E- en] (10) which is the Butler~Volmer equation. The standard rate constant ky is a measure of the intrinsic energy barrier. The transfer coefficient « is generally believed to reflect the nature of the transition state, in the same sense that is used in physical organic chemistry. Small values (« > 0.5) are indicative of a product- like transition state, and large values (« > 0.5) are indicative of a reactant-like transition state. Figure 1-12 shows several electron transfer reactions with corresponding values of «. Anthracene!? has « = 0.55; the barrier for electron transfer consists largely of solvation reorganization energy, and the rate ofTHE REDUCTION POTENTIAL AND ELECTRODE KINETICS 15 fog 10K) Figure 1-11 Linear free energy relationship obtained for the addition of semicarbazide to substituted benzaldehydes. The slope, analogous to the transfer coefficient in electrochemistry, is 0.47. (Adapted from Reference 11.) COO +«- CO a = 0.55 Fe(CN),” + ¢ — Fe(CN)," a = 0.54 14 a ~ Sn -€ > >sn— B= 031 CHACH),Br + © — CHCH),+BrF a =03 a= 04 a=03 a = 0.78 + 7 NO, +€— NO, a= 05 Figure 1-12 Electron transfer reactions and transfer coefficients.16 CYCLIC VOLTAMMETRY electron transfer is fast. Both tetraalkyltin oxidation (6 = 0.29)'* and alkyl halide reduction (a = 0.3)'* occur with concurrent structural change (from tetrahedral to pyramidal for the alkyltins and bond breaking for the alkyl halides); these structural changes contribute to the barrier for electron transfer. Cyclooctatetraene (a = 0.4) goes from a tub shape to a partially flattened state upon reduction.!*-'® An interesting pharmaceutical, Artemisinin, has a value of a = 0.35: presumably the oxygen—oxygen bond in this endoperoxide is either lengthened or broken upon electron transfer. Methylcobalamin has a large transfer coefficient (« = 0.78). The reason for this is not as easily explained, but it can be noted that the electron trans- fer and following chemical reaction occur in two separate activated steps. It has been suggested that in this case the transfer coefficient will be large.'* Whereas NO} has a relatively facile reduction, with a = 0.5,'7 calculations based on solvent structural changes indicate that the electron transfer should be rather slow. This is indicative of a more complicated, “inner sphere” mechanism, in which the donor (the electrode) and acceptor (NO}) are coordinated in such a way as to reduce the barrier to electron transfer. 1.2.3 Outer and Inner Sphere Interactions Electron transfer, heterogeneous or homogeneous, can be classified as outer sphere or inner sphere, according to the extent of interaction between the electron donor and the electron acceptor (Figure 1-13). Electron transfer reactions that take place with weak interactions between reactants are outer sphere and those with strong interactions are inner sphere. Outer sphere reactions are characterized by the relative absence of steric effects on the rate of electron transfer, while inner sphere reactions are very sensitive to Figure 1-13 Schematic representation of outer and inner sphere electron transfer reactions.THE REDUCTION POTENTIAL AND ELECTRODE KINETICS 17 steric effects, which prevent the close encounter needed for full interaction. This observation has been suggested as the major experimental criterion for distingu- ishing inner from outer sphere electron transfer.'* 1.2.4 Marcus Theory The theory that has found the most application for characterizing the nature of outer sphere electron transfers is that developed by R. Marcus, who was awarded the Nobel Prize in 1992 for these contributions. As it is employed in most studies, the Marcus theory provides a method by which one can relate the rate of electron transfer to solvation reorganization changes, structural changes, and the overall free energy change. Focusing on the electron transfer act, theory for electron transfer presumes a preequilibrium factor for complex formation, and the following electron transfer: D+A<>[D,A] Kea [D,A]=[D*,A7] Keg It is the rate of the electron transfer step, k,,, that is discussed below. The “Marcusian” transition state diagram presents a reaction coordinate that is composed of solvent reorganization as one moves from reactant to product. Transition state diagrams of this sort can be applied to electrode reactions or homogeneous electron transfer (Figure 1-14). The free energy along the coordinate varies as a square of the deviation from the equilibrium position. The physical meaning is that small changes in the solvent medium around the ion 0.A 0 .A) Figure 1-14 Transition state diagram for electron transfer according to Marcus theory.18 CYCLIC VOLTAMMETRY can give rise to large changes in the electronic energy of the molecule. The intersection of the curves gives the transition state energy. When the driving force AG = 0, the intersection of the two curves occurs at an energy of AG), = 4/4. This is called the intrinsic free energy barrier, where A is related to the optical and static dielectric constants. 2 1 jae [= ee =| (ty a In polar solvents £.,, < E,, and the optical dielectric term dominates. The general expression for the activation energy is given by 7 AG +w}? é sa 12) aci[ 1+ nar | (12) where w is the work required to bring the Ox and Red to the electrode. Often the work term is neglected. Figure 1-15 shows the variation of the activation energy with driving force for a typical value of 4, 1.00eV. Inspection of Equation (12) (after expansion) shows that the free energy relationship, neglecting the quadratic term, is equivalent to the Butler—-Volmer equation, with the transfer coefficient set to 0.5. This is shown in Figure 1-15 to be true for small deviations from the reduction potential. We can also note that for the highly endoergic or exoergic regions, the transfer coefficient (the change in activation energy per change in driving force) is given by: aac? _1 [AG +w @AG? 2” | 8AG5 Several cyclic voltammetric studies have shown the expected change in the transfer coefficient with potential.'* The highly exoergic region, where the AG 25 ° 20 ° poet 2 StS Marcus. —>o0 zg 10 a 5 05 = oo g = os -10 Estee 1 2 3 Figure 1-15 Variation of activation energy with potential according to Marcus theory. The straight line corresponds to a linearized version of Marcus theory, equivalent to the Butler— Volmer equation.THE REDUCTION POTENTIAL AND ELECTRODE KINETICS 19 activation energy increases with driving force, is difficult to observe with electrochemical experiments* but has been observed in other contexts.'® The theory is similar for homogeneous electron transfer, except that because there are two ions involved, the value of J is expected to be larger. Also, the work terms are different as a result of the difference in work of bringing reactants and products together in solution compared to bringing a single ion to electrode. A term is sometimes included in reorganization energy to accommodate changes in vibrational energies for the molecule/ion transition. 1 zd Sidd? where f is the force constant and dd is a bond length change. These are sometimes called “inner sphere” contributions to the reorganization energy: they are not to be confused with an “inner sphere” interactions, as described earlier. 1.2.5 Kochi’s Free Energy Relationship for Electron Transfer The importance of inner sphere interactions to electron transfer was first recognized by H. Taube, who received the Nobel Prize for work in this area. The complexity of the problem, however, has prevented the development of a quantitative theory comparable to Marcus theory for outer sphere reactions. In practical terms, inner sphere interactions generally lower the activation barrier for electron transfer compared to outer sphere electron transfers. As a con- sequence, they may play important roles in electron transfer in biochemistry and in the catalysis of electrode reactions. A characteristic inner sphere reaction investigated by Kochi is the electron transfer of iridium hexachloride with tetraethyltin.'? A five-coordinate transition state around the tin is indicated, wherein the positive attraction in the ion pair stabilizes the transition state (Figure 1-16). In the case of inner sphere electron transfer, steric congestion prevents the close encounter needed to stabilize the TS, and lower rates of homogeneous electron transfer are observed. In contrast to outer sphere electron transfer, there is no obvious relationship between homogeneous and heterogeneous rates, as the interactions between D/A complexes in solution are generally divergent from electrode—molecule interactions. A new method to characterize inner sphere electron transfer reactions has recently been developed by Kochi’? and designated FERET (free energy | i \/ i Sn + —r—- —s—-a—r Figure 1-16 Typical inner sphere complex. * This is because the continuum of energy levels in the electrode (see reference 13a).20 CYCLIC VOLTAMMETRY relationship for electron transfer). The sequence of events for electron transfer between a general donor and acceptor can be written: Kon D+A=51D,A] [D, A] —s> [D*,A7] [D*,A~] > products (fast) The products can be either separated D* and A~ or a collapsed ion pair. For endergonic driving force, the observed rate constant is given by: Koos = keeKoalDI[A] (13) In the endergonic region, a cycle can relate the thermodynamics and kinetics of the electron transfer process (Figure 1-17): AG = AG? + w, — w, (14) The work terms w, and w, are associated with bringing the reactants or products together. For neutral reactants, w, can be neglected. Another cycle involves spectroscopic measurement of the optical analogue of electron transfer, the charge transfer transition energy, hvcy (Figure 1-18). The work term for inner sphere interaction of the ion-pair can be evaluated as follows: (D,A]"+ (D+, A-* wk = hycy — Pp + EAy + ©, (15) D+A Oo a [>a] D4 + As \ [> a] / Figure 1-17 Thermochemical cycle for thermal electron transfer.‘THE REDUCTION POTENTIAL AND ELECTRODE KINETICS 21 D+A Ns Dt +A / ™ [o: Figure 1-18 Thermochemical cycle for optical electron transfer. If a common acceptor is chosen, and a reference donor, then a comparative work term for the ion-pair can be defined: Aws = wi — wo" = Ahvcy — AIP (16) Finally, utilizing a combination cycle for thermal electron transfer and optical charge transfer (Figure 1-19), the relative free energy of activation for electron transfer can be expressed as follows: AG? = Abver + AGrow This equation, a generalized free energy relationship for electron transfer (FERET), has the advantage of consisting entirely of terms that can be evaluated from experiment. This approach has been utilized for a comparative study of the inner sphere reactions of nitronium ions at an electrode and with aromatic substrates in solution, providing insight into the activation process.'® hver p+A—([D,A] —E[ + : [0 A Jeew Figure 1-19 Overall thermochemical cycle that establishes a free energy relationship for electron transfer.22 CYCLIC VOLTAMMETRY 1.2.6 Comparing Homogeneous and Heterogeneous Electron Transfer Rates The rates of heterogeneous and homogeneous reactions can be compared in the context of transition state theory: ~AGt b= Zero], a | where Z is a preexponential factor that includes a preequilibrium term for formation of the donor acceptor complex, and a frequency factor. For hetero- geneous reactions Z,., is on the order of 10*~5cm/s, and for homogeneous Teactions Zyom is on the order of 10! M~! s~ 1.20 1.2.7 The Double Layer and Electrode Kinetics In most electrochemical experiments, an inert salt is added (about 0.1-1 M) to increase the solution conductivity and to minimize mass transport by migration. The structure of the electrode—solution interface under such conditions has an effect on the kinetics of electrode processes. In general, an electrode has net charge that is related to its potential. Every electrode material has a potential of zero charge (pzc): the potential at which the net charge is zero. At potentials negative to the pze, the electrode is negatively charged, and at potentials positive to the pzc, the electrode has a net positive charge. Consider a negatively charged electrode in contact with an electrolyte solution: positive ions will be attracted to the electrode and negative ions repelled. Some ions may be immobilized at the surface. Thermal motion is balanced by electrostatic forces so that there will be an excess of positive ions near the electrode relative to the bulk solution (Figure 1-20, top). The details of the structure will depend on solvent properties (dielectric constant) and electrolyte charge and concentration. This is called the Gouy—Chapman-Stern model for the double layer (charged electrode and charged solution) at the electrode—solution interface. As a consequence of the structure of the double layer, the potential changes decays in a gradual manner as distance from the electrode (Figure 1-20, bottom). The electrical field (V/cm) can be very large across the electrode—solution interface. The potential decay will be steeper as the concentration of electrolyte is large and as the polarization is increased (the deviation of the potential from the potential of zero charge). The effect of the double layer on electrode kinetics is illustrated in Figure 1-21. A charged particle approaching the electrode requires work to move through the potential drop to reach a point near enough to the electrode to transfer an electron. This work term either increases or decreases the activation energy compared to the ideal case, where the potential drop is infinitely steep and there is no work associated with movement of the charged particle. Values of kg and « or f determined without taking into account double-layer effects are termed “uncorrected values.” The task of double-layer theory, insofar asTHE REDUCTION POTENTIAL AND ELECTRODE KINETICS 23 Figure 1-20 Sketch of the distribution of ions near a charged electrode (electrode double layer) and the corresponding potential drop. he Distance from electrode surface Figure 1-21 Origin of the double-layer correction. An ion must move across a poten- tial drop.24 CYCLIC VOLTAMMETRY electrode kinetics are concerned, is to determine the value of the potential drop from the OHP (closest approach) to the solution. The Butler-Volmer equation taking this correction (SE) into consideration is written: i —anF(E — E° — 5E) — 2,,6E shui tga {tcadexr eed nFA (1 — o)nF(E — E® — 6E) — sus? It —[Creal exo RT For charged species, the concentration terms in the Butler—Volmer equation must also be modified as follows: —zF(6E} Cus Coonenp| =F] The parameters necessary to make a correct model of the double layer are often not easily obtained. However, the uncorrected values of electrode kinetic parameters may retain significance under many circumstances. For instance, for a series of compounds studied under similar conditions, k° and « values can be compared. Double-layer effects are also minimized by large electrolyte concen- trations (0.1-5 M). The 5E term is smallest near the potential of zero charge (pzc), and largest at extreme potential values, far from the pzc of the electrodes. A good overview of double-layer theory is given in Bard and Faulkner's text?! and by Rieger.?? References Gurney, R. W. Ions in Solution. Dover: New York, 1962. . Pleskov, Y. Vs Gurevich, Y. Y. Semiconductor Photoelectrochemisiry. Consultants Bureau: New York, 1986. Antropoy, L. I. Theoretical Electrochemistry. Mir: Moscow, 1972. AY Koryta, J; Dvorak, J. Principles of Electrochemistry, Ist ed. John Wiley & Sons: New York, 1987. Goodisman, J. Electrochemistry: Theoretical Foundations, 1st ed. John Wiley & Sons: New York, 1987. Kebarle, P.; Chowdhury, S. Chem, Rev. 1987, 87, 514. Summerman, W.; Deffner, U. Tetrahedron 1975, 31, 593. Bordwell, F, G.; Cheng, J.-P.; Harrelson, J. A., Jr. J. Am. Chem. Soc. 1988, 110, 1129. Hupp, J. H.; Weaver, M. J. Inorg. Chem. 1984, 23, 3639. 10. Harris, D. Quantitative Chemical Analysis, 2nd ed. W. H. Freeman: New York, 1987. yrs 11. Leffler, J. E; Grunwalk, E, Rates and Equilibria of Organic Reactions. Dover: New York, 1963. 12. Kojima, H Bard, A. J. J. Am, Chem. Soc. 1975, 97, 6317.THE REDUCTION POTENTIAL AND ELECTRODE KINETICS 25 13. Kochi, J. K. Angew. Chem., Int. Ed. Engl. 1988, 27, 1227. 13a. Marcus, Rudolph A. Angewandte Chemie International Edition in English 1993, 32 1111 14. Saveant, J. M. Single Electron Transfer and Nucleophilic Substitution. Advances in Organic Chemistry, Vol. 26, Academic Press: London, 1990. 15. Allendorfer, R. D; Rieger, P.H. J. Am. Chem. Soc. 1965, 87, 2336. 16. Fry, A. J; Britton, W. E., Eds. Topics in Organic Electrochemistry, Ist ed.. Plenum Press: New York, 1986. 17. Lee, K. Y.; Kuchynka, D. J; Kochi, J. K. Inorg. Chem. 1990, 29, 4196. 18. Chem. Rev. 1992, 92 (3). 19. Kochi, J. K. Ace. Chem. Res. 1992, 25, 39. ; Logan, J; Newton, M. D, Sutin, N. J. Am. Chem. Soc. 1980, 102, 5798. 20. Brunschwig, B. 21, Bard, A. J., Faulkner, L. R. Electrochemical Methods. John Wiley & Sons: New York, 1980. 22. Rieger, P. H. Electrochemistry. Prentice-Hall: Englewood Cliffs, NJ, 1987.CHAPTER 2 The Cyclic Voltammetric Experiment In Chapter 1 we explored the fundamental relationship between the electrode potential and a redox couple in solution. It was also pointed out that if the potential of an electrode is controlled externally, the solution can be made to “adjust” by electron transfer to approach equilibrium with the electrode potential. In many electrochemical experiments, the solution initially has only one form of a redox couple present, and the electrode is initially set at a potential such that this form does not undergo electron transfer. This ensures that the experiment begins at zero faradaic current. The electrode potential is then changed to a position that favors electron transfer. The manner in which the potential is changed gives rise to a profusion of electrochemical “controlled potential” techniques. Cyclic voltammetry (CV) is a potential-controlled “reversal” electrochemical experiment. A cyclic potential sweep is imposed on an electrode and the current response is observed. Analysis of the current response can give information about the thermodynamics and kinetics of electron transfer at the electrode— solution interface, as well as the kinetics and mechanisms of solution chemical reactions initiated by the heterogeneous electron transfer. This chapter examines fundamental experimental and theoretical aspects of the CV experiment (Figure 21). 2728 CYCLIC VOLTAMMETRY Potential Current S Potential TIME | -——> Dynamic Potentiostat AE WE RE NoAr —» —— Figure 2-1 Overall view of the CV experiment. 2.1 An Overview A potentiostat system sets the control parameters of the experiment. Its purpose is to impose on an electrode (the working electrode) a cyclic linear potential sweep and to output the resulting current-potential curve. This sweep is described in general by its initial (E,), switching (E,), final (E,) potentials, and sweep (or scan) rate (v, in V/s). The potential as a function of time is: E= i; + vt (forward sweep) E = E, — vt (reverse sweep) More complicated sweeps are possible, such as the option of a second pontential. Multiple cycles are sometimes used, but in many instances these will not be more informative than a single cycle. The term linear sweep voltammetry (LSV) is used for a half-cycle CV. Figure 2-2 illustrates various possible cycles.THE CYCLIC VOLTAMMETRIC EXPERIMENT 29 a. co) Time | oe Time Cc. a a ~~ ‘Time eal / Time Figure 2-2 CV waveforms. Inexpensive analog instruments as well as microprocessor-controlled units are commercially available. The equipment can include useful features such as IR compensation, data smoothing, background subtraction, and automated display capabilities. Commercial instruments offer scan rates of up to several hundred volts per second. However, units that have very high scan rate capability (to 10° V/s) can be constructed at relatively low cost. The electrochemical reaction of interest takes place at the working electrode (WE). Electrical current at the WE due to electron transfer is termed faradaic current. An auxiliary, or “counter” electrode (AE) is driven by the potentiostatic circuit to balance the faradaic process at the WE with an electron transfer of opposite direction (e.g, if reduction takes place at the WE, oxidation takes place at the AE). The process at the AE is typically not of interest, and in most experiments the small currents observed mean that the electrolytic products at the AE have no influence on the processes at the WE. The faradaic current at the WE is transduced to a potential output at a selected sensitivity, expressed in amperes per volt, and recorded in a digital or analog form. The CV response is plotted as current versus potential. Figure 2-3a shows the shape of a CV current response for a typical reduction process. During the forward sweep the oxidized form is reduced, while on the reverse sweep the reduced form near the electrode is reoxidized. Chemical reaction30 CYCLIC VOLTAMMETRY (a) 200 150 2 100 g Ss Eo -20 ~100 -190 a Potential/V (b) ‘ 5 4 ect 1 ° “1 —- ‘13 a4 “18 x18 Potential/V Figure 2-3. CVs for (a) ferricyanide reduction and (b) methylcobalamin reduction. coupled to the electrode reaction can drastically affect the shape of the CV response. Compare the current response just described with that of the reduction of methylcobalamin (Figure 2-3b), where no reverse peak is observed. The absence of the reverse peak indicates that a following chemical reaction has removed the reduced species. 2.2 The Electrochemical Cell 2.2.1 Cell Design A typical cell design for a cyclic voltammetric experiment is shown in Figure 2- 4a. The simplest approach is merely to have the three electrodes immersed in the solution in close proximity. A Luggin capillary (Figure 2-4b) further isolates the reference solution from the cell solution. At the outset of the experiment the cellTHE CYCLIC VOLTAMMETRIC EXPERIMENT 31 fa) (b) \ K Figure 2-4 (a) Typical electrochemical cell. (b) Luggin capillary reference. contains solvent, electrolyte, one or more principal electroactive species, and possibly added reagents that will undergo reactions with the electrolytic products. Before the experiment, it is necessary to remove dissolved oxygen, which has a cathodic signal that can interfere with observed current response. This is normally done by purging the solution with an inert gas such as N, or Ar. Removal of dissolved oxygen can also be effected with freeze—pump-thaw cycles on a vacuum line. Electrochemical experiments can be performed in glove bags, glove boxes, or in vacuum for electroactive species, that are sensitive to air or moisture. 2.2.2 Solvents The choice of solvent is determined by several factors, including conductance, solubility of electrolyte and electroactive substance, and reactivity with electro- lytic products. The solvent can also have important properties such as decreasing usually unwanted effects (e.g., adsorption of the electroactive species at the electrode). Because of the importance of the solvent in electrochemical processes, it is sometimes desirable to consider the physical and chemical properties of the solvent in some detail. Solvent properties relevant to electro- chemical experiments are listed in Table 2-1. The melting and boiling points define the useful temperature range for most solvents (with some variation due32 CYCLIC VOLTAMMETRY Table 2-1 Properties at Common Electrochemical Solvents Solvent Dielectric Ey Donor —_ Acceptor (mp, bp, °C) constant (Solvatochromatic) number number H,0 (0, 100) 78.3 10 548 Propylene 6496 0.491 carbonate (—545,241.7) Dimethyl 46.95 0.444 0.7 193 sulfoxide (18.5, 189) Dimethyl- 36.71 0.404 0.69 18.8 formamide 35.94 0.460 036 189 (438,81) Nitrobenzene 34.78 0.324 021 148 (5.8, 210) Methanol 32.66 0.790 (—97.7, 64.5) Ethanol 24.55 0.654 379 (=114.5,78.3) Acetone 20.56 0.355 0.44 12.5 (=94.7, 56.1) Source: Reference 1. to colligative effects). The dielectric constant is primarily an indicator of solvent polarity and solubilizing power. Other solvent properties, such as acceptor and donor numbers, indicate the ability of the solvent to participate in electron-pair donor-acceptor interactions. The spectral charge transfer energy (solvochroma- tic shift) of pyridinium-N-phenoxide betaine dye has also been used to characterize solvents.’ Solvent mixtures sometimes are used to average pro- perties. An example is the nonaqueous electrochemistry of methylcobalamin where mixtures of methanol and DMF achieve a balance between solubility and conductivity. Low temperature studies can be very useful in electrochemical investigations. Reactive species can be stabilized and reversible electrochemistry obtained. Solvent properties can change drastically with temperature. For example, DMF dimerizes at below —40°C and has a large potential window as well as a lower dielectric constant. In Fry and Britton’s handy review of solvents and electrolytes,” acetonitrile, ethanol, methanol, and methylene chloride are recommended as good oxidative (anodic) electrochemical solvents, while acetonitrile, DMF, and dimethyl sul- foxide (DMSO) are suggested for reductive (cathodic) electrochemistry. Acetonitrile is suggested as the best overall nonaqueous solvent on the basis of its electrochemical properties and its relative nontoxicity. The review of Fry and Britton also is a good place to start when looking for purification methods.THE CYCLIC VOLTAMMETRIC EXPERIMENT 33 Often, to minimize evaporation of the cell solvent (and consequent changes in concentrations), the purge gas (N2,, Ar) is passed through the same solvent used in the electrochemical cell. This step is particularly necessary for solvents with low vapor pressures. Mention should also be made of the increasing use of unconventional media in electrochemical experiments. Experiments in micellar solutions and micro- emulsions, for instance, can solubilize or concentrate reactants in micelles.? Cyclic voltammetric results have been obtained below the freezing point of the solvent: for instance, in frozen DMSO*® and in perchloric acid.°” 2.2.3 Electrolytes The electrolyte, added to enhance conductivity and to minimize double-layer and migration current effects, is chosen on the basis of solubility in a given solvent as well as inertness toward the electroactive substance and its electroly- sis products. There are of course many choices of electrolyte for use in aqueous solution. The tetraalkylammonium salts are the most commonly used non- aqueous electrolytes. Tetrabutylammonium tetrafluoroborate (TBATFB) and tetrabutylammonium hexafluorophosphate (TBAHFP) are recommended by Fry and Britton, who note that TBAHFP in acetonitrile has a particularly large useful potential range of +3.4 to —2.9 V (vs. SCE). 2.2.4 Electrodes Working Electrode (WE) Solid disk electrodes are most commonly employed in CV experiments (Figure 2-5). Platinum, glassy carbon, gold, silver, or amalgams* are often used. The use of carbon electrodes of various types has been reviewed.” The mercury drop electrode is useful for electrochemistry in aqueous solution at large negative Figure 2-5 Disk working electrode. * Recipe for a silver amalgam electrode: polish the Ag wire electrode (ca. 1 mm diameter) with fine alumonum oxide powder. Reduce at — 1.0 V(vs. SCE) for 3 minutes ih a cell containing nitric acid (pH 1.5) and a drop of mercury. While still holding the potential, touch the electrode surface to the Hg drop. Rinse several times with deionized water; immerse in deionized water for 50 minutes Polish with aluminum oxide powder and rinse again. The electrode can be used or months, like any other solid electrode.34 CYCLIC VOLTAMMETRY potentials because of the sluggishness of the hydrogen reduction compared to other electrodes. However, for nonaqueous work and for electrochemistry at positive potentials, the solid electrodes are often preferred. The WE should also have a facile electron transfer with the electroactive species. The factors that facilitate heterogeneous transfer are not always well understood, and careful pretreatment by polishing, sonicating, or holding or cycling of the WE potential of electrodes can markedly improve performance. Electrode fouling occurs quite easily, and frequent polishing is sometimes necessary. Deleterious effects such as IR drop and capacitive charging time are greatly reduced as the electrode radius is made smaller. For cyclic voltammetry, the result is that much higher sweep rates can be achieved with ultramicroelectrodes (radius < 100 um). Most pre- liminary studies at moderate scan rates utilize electrodes having a radius of approximately 0.2cm. Reference Electrode (RE) The most common reference electrode systems used in aqueous solutions are Ag/AgCl and the calomel electrode. If aqueous-based references are used in nonaqueous solution, however, large liquid junction is produced; and often More serious, aqueous contamination of the nonaqueous cell occurs. Thus this combination is not recommended. The use of an Ag/Ag* non-aqueous-based reference is suggested for nonaqueous electrochemistry. To avoid large junction potentials, the RE solvent should be as close in nature as possible to the cell solvent system. Often potentials are calibrated with a standard, such as ferrocene or cobaltocene. Suggested standards are listed in Table 2-2, along with reduction potentials and other properties. Construction of an Ag/Ag* reference for nonaqueous use is shown in Figure 2-6. Reference electrodes can drift with time and must be carefully maintained. Table 2-2 Reference Compounds for Cyclic Voltammetry Substances* E*(V) D x 10%(em?/s) Conditions? Fe(CN)g-/~ 0.2534. Ag/AgCl, Dy, = 0.76 pH 3.0 1MKa. D, O.1MKCI Ru(NH,)e*?*! 0.17 vs. SCE D, pH 7.0 Phosphate buffer Ferrocene 0.37 vs. SCE O6MTEAP in Fe(Cp),/Fe(Cp); * acetonitrile Tempo/Tempo*** 0.271 vs, Ag/0.1M Drea = 0.77 O.1MTBAP in AgNO, /acetonitrile acetonitrile * Cp, cyclopentadienyl. * TEAP, tetraethylammonium perchlorate; TBAP, tetrabutylammonium perchlorate, Source: * Baur, J. E, and Wightman, R. M., J. Electroanal. Chem., 1991, 305, 75. +*Summerman, W. and Definer, U., Tetrahedron, 1975, 31, 593.THE CYCLIC VOLTAMMETRIC EXPERIMENT 35 a ++—— 0,001 M to 0.01 M AgNO, \ { in non-Aq. solvent 1 | | ptt mum <——— Glass frit or vycor plug Figure 2-6 Nonaqueous reference electrode. 2.2.5 Potential Window and Background Subtraction Taken together, the components of a solvent—electrolyte-WE system have a certain potential range in which useful information can be collected: the potential “window.” Outside this region, solvent or electrolyte electrolysis reaches a level that obscures the signal. The potential window is greatly affected. by the purity of solvents and electrolytes. Distillation and drying can extend the potential window significantly. As with many other experiments, it is wise to collect “background” spectra— cyclic voltammograms with only solvent and electrolyte present. The potential window of the solvent—electrolyte system can be established, and any obvious impurities can be identified before they cause havoc in the interpretation of results with the electroactive species present. In quantitative studies, the background can be subtracted from equivalent experiments with the electroact- ive substance present. A particularly effective method proposed by Wightman and Wipf utilizes a flow injection cell to obtain background and experimental currents under the same conditions.* Normally, concentrations of the electroactive substance in the 0.5-10mM range are utilized in mechanistic studies. Strategies to improve detection limits may be needed under some circumstances, such as signal averaging.° Such methods can improve detection limits dramatically, down to the nanomolar region. 2.3 Electrochemical Mechanisms: E & C Notation In this section we examine the CV response for the simplest electrochemical mechanisms. The mechanisms are described by the E&C notation: E represents a one-electron transfer at the electrode, while C refers to a solution chemical36 CYCLIC VOLTAMMETRY reaction that is coupled to an electrode reaction. For instance, the EC mechanism refers to an electrode reaction followed by a chemical reaction. Ox +e7 =Red E°",k,0 Red— Products k.hem Subscripts are used to provide additional information. The subscript “r” stands for reversible (meaning that both forward and reverse processes are fast enough to maintain equilibrium or “Nernstian” conditions at the surface), and “i” represents irreversible (only the forward reaction is significant); “i” and “r” are limiting cases of “gq,” or quasi-reversible (meaning that both the forward and reverse processes take place but are not fast enough to be considered at equilibrium). Thus, in an E,C, mechanism the electrode reaction is fast and reversible and the chemical reaction is irreversible. After “ECE” the E&C nomenclature breaks down: ECE can mean several different things (and one cannot be sure what an “ECECCE” mechanism refers to !), At this point the mechanism needs to be further defined—for instance, by a reaction sequence or schematic drawing that indicates all possible chemical and electrochemical events considered. We note that every reductive mechanism has an oxidative analogue. For instance, the oxidative EC is: Red —e” = Ox Ox — Products Thus, all the theoretical results quoted for a reductive mechanism can be immediately applied to the analogous oxidative process. The only difference is in the sign of the current and the direction of the potential sweep. 2.3.1 The “E,” Mechanism The simplest possible mechanism is the one-electron oxidation or reduction of a solution chemical species at the WE. Consider a typical experiment in which 1mM of an electroactive species in an electrolyte solution undergoes a facile, reversible one-electron reduction with the following parameters: Experimental Settings E, = 0.0, E, = —0.5, E; = 0.0, v = 1 V/s Area of electrode A = 0.01 cm? Cell temperature T = 298 K. Electroactive Species Parameters Ox + e7 = Red, E” = —0.25V (ie., vs. a reference) C,, = 1.00mM, C,.4 = 0.0mM Diffusion coefficients D,, = D,-g = 1 x 107>cm?/s,THE CYCLIC VOLTAMMETRIC EXPERIMENT 37 The voltage sweep and current response for this experiment are shown in Figure 2-7. The data can be represented as a current-time curve, or, since the potential is linearly related to time for each half-cycle, as a current—potential curve (the usual representation). Note that reduction current is taken as positive, the cathodic sweep goes from left to right. For any CV, the direction of the initial sweep should be indicated or at least apparent (the initial sweep potential is always set, if possible, at a potential where zero faradaic current occurs). () > 3 3 [o> e a Thee Time (b) 10 8 Pee 2 fo4 2 bee ge g -2 é -6 -8 ° 250 300 750 000 Time/mSec © a 6 eee | zg =e 4 q a 00 0.1 02 0.3 04 =0.8 Potential (V) Figure 2-7 (a) Potential waveform. (b) Current—time and (¢) current—potential representations.38 CYCLIC VOLTAMMETRY Consider what happens near the working electrode in the foregoing example {refer to Figure 2-8, which shows selected points along the CV with the corresponding concentration profiles of Ox and Red). As the potential is swept past the reduction potential, the oxidized form is converted to the reduced form in proportions consistent with the Nernst equation. Because the potential sweep is fast (relative to the rate of diffusion of material to the electrode), the oxidized species is depleted near the electrode. As a consequence of this depletion, cyclic voltammograms have a peak shape, in contrast to the familiar polarographic sigmoidal wave. Sigmoidal CVs can be indicative of a catalytic regeneration of reactant near the electrode. The peak resulting from the reduction process is called the cathodic peak current; by convention, it is positive, and it occurs 28.5 mV negative to the E° (at 25°C). At the switching potential, the direction of the potential sweep is reversed. A negative anodic peak current is observed as the potential is swept past the E° (the electron transfer is in the opposite direction). The anodic peak occurs about 28.5mV positive of the E° (the exact value depends very slightly on the switching potential). Note that the major perturbation on the initial concentrations of Ox and Red lies within the mean square diffusion length Xp = (2D x time)", which is about 45ym for this experiment—large with respect to the double-layer thickness. Current/j Amps 0.0 -0.1 -0.2 -0.3 -04 -0.5 Potential/V Figure 2-8 The current response and corresponding concentration profiles (A—F) for a reversible CV. Concentrations of the oxidized (solid circles) and reduced form (open circles) are shown as a function of distance from the electrode.THE CYCLIC VOLTAMMETRIC EXPERIMENT Concentration/mM Distance/m Concentration/mM Distance/um Figure 2-8 (Continued) 3940 Concentration/mM Concentration/mM Distance/um 10 20 30 Distance/um Figure 2-8 (Continued) CYCLIC VOLTAMMETRY 40THE CYCLIC VOLTAMMETRIC EXPERIMENT a Concentration/mM Concentration/mM 1.0 08 0.6 0.4 0.2 0.0 0 10 20 30 40 Distance/zm “0 10 20 30 40 Distance/um Figure 2-8 (Continued)42 CYCLIC VOLTAMMETRY Current/Amps 0.0 -01 -0.2 -0.3 -04 -0.5 Potential/V Figure 2-9 Three half-cycles. Note the slight diminution in the second cathodic peak current. The initial sweep peak current for a reversible electron transfer is: 12 = 0.4463nF A [=| CygD'2 yt? () At 25°C this is: i, = 2.686 x 105n??C,,D!?v/74 (2) The reverse peak current is a function of switching potential. A reversible CV is a diffusion-controlled process; that is, the rate of electron transfer is controlled by the rate of supply of material to the electrode by diffusion. Multiple sweeps give similar results (Figure 2-9 shows a CV with three half- cycles), with slightly smaller currents (as the initial condition of [Ox] and [Red] near the electrode is nearly but not quite the same for each successive cycle). While for simple one-electron transfer no information is to be gained by scanning more than a single cycle, if chemical reactions are coupled to the electron transfer, more than a complete cycle may be informative. The peak potential (for the forward seeep) is*: RT = Eo E,=E' 1.105| £7 (3) * Assuming that the diffusion coefficients of the oxidized and reduced species are equal. Simulation methods using individual diffusion coefficients give the exact result,THE CYCLIC VOLTAMMETRIC EXPERIMENT 43 Criterion for the E, Mechanism Several criteria can be utilized to confirm a single, reversible, electron transfer. 1, The difference in cathodic (E,,.) and anodic (E, ,) peak potentials is around 57-60 mV (depending on the switching potential): AE, = abs[E,,. — Ep] ¥ 58mV In actual experiments the expected 58 mV is rarely observed because of small distortions due to solution resistance effects and electronic or math- ematical “smoothing” of data. The result is that AE, is often 60-70mV for reversible electron transfer. For reversible multielectron transfer, the peak separation is 60/nmV. 2. The difference between the initial sweep peak and half-peak potentials of the forward sweep is 56mV/n. 3. The shifted ratio of the cathodic to anodic currents is unity: i,,./it, = 1. In the “shifted ratio,” the anodic peak current is measured from a baseline that is moved to a value that can be predicted from the decaying portion of the cathodic peak. In this part of the cyclic voltammogram the current can be predicted by assuming an inverse square root dependence of the current of the time. The baseline is assumed to be the current that would be obtained if the forward sweep were continued for the same amount of time that it takes to reach the reverse peak. This procedure is illustrated in Figure 2-10. The absolute ratio (with zero current as a baseline) depends on switching potential and can be simulated (see Chapter 5). 4, The forward scan peak current should be proportional to the square root of the scan rate. This criterion is used to distinguish “diffusion-controlled” processes from processes featuring the adsorption of the electroactive species onto the electrode (in which case a linear current-scan rate relationship is observed), A plot of the log i, versus log v is linear, with a slope of 0.5 for a diffusion peak and a slope of 1 for an adsorption peak. Intermediate values of the slope are sometimes observed, suggesting a “mixed” diffusion-adsorption peak, Adsorption mechanisms are discussed in more detail in Chapter 3. The Reduction Potential and Number of Electrons . The reduction potential is given by: = Foe + Ene 2 If AE, has been determined to be 60 mV, and the reduction has been shown to be diffusion controlled, then the number of electrons n = 1, and the diffusion coefficient can be calculated from Equation (1). The geometric electrode area can be found from Equation (1), or the effective area can be calibrated with a standard with known D and n (e.g., ferrocene). Even if the diffusion coefficient can be only roughly estimated, a reliable value of n can be found. This is Ee” (4) N44 CYCLIC VOLTAMMETRY 3 See & = & — = -8- laa creer eeeeeeee peered eer eeeeeseer eres eeeesee eee ee oo 01 ~-02~OSSC OSC Potential (V) Current/pamps 00 02-04-06 0.8 Potential/V Figure 2-10 Characteristics of the E, mechanism: (a) peak potentials and currents and (b) measurement of the shifted peak current ratio. because the current has a square root dependence on D but is proportional to n3?. Conversely, if n is known, the D,, can be determined. The determination of the number of electrons involved in an electrochemical mechanism is crucial but not always as straightforward as the discussion above might suggest. This is because the electron transfer is not reversible. The number of electrons can be determined utilizing potential step experiments (see Section 2.8). Bulk electrolysis (exhaustive electrolysis of all the electroactive species present) is sometimes helpful, but here one must be careful in interpre- tation because time scales differ widely. Thin-layer bulk electrolysis brings the time scale of the electrolysis more in line with the CV time scale.'°THE CYCLIC VOLTAMMETRIC EXPERIMENT 45 2.3.2 Electrochemical Reversibility Let us consider the electrochemical rate equation and its limiting case of very fast rates, the Nernst equation: i nFA ke ‘RT. a Ran Q= oxo| ee E | If the rate of electron transfer is slow with respect to the time scale of the experiment, then non-Nernstian concentrations will exist at the electrode surface. The qualitative effect is to shift the peak wave to more negative potentials in the case of reduction and to more positive potentials in the case of an oxidation. The proximity to equilibrium is termed “reversibility.” Increasing the scan rate is equivalent to increasing the rate of diffusion of Red away from the electrode (for a reductive, Ox + e~ = Red, process). At faster scan rates, steeper concentration gradients are produced, increasing the rate of diffusion. Red will diffuse away from the electrode faster as the scan rate is increased. The diffusion of Red away from the electrode is a process that competes with the oxidation of Red. Thus for the E mechanism, the greatest reversibility is observed at slow scan rates. In this case reduct potentials are most accurately measured at slower scan rates, where reversibi is greatest {and distortions least). Three cyclic voltammograms with various degrees of reversibility are shown in Figure 2-11. In the reversible region the CV response Ke Con — keCrea Current (Amps) Potential (V) Figure 2-11 _ Effect of the heterogeneous rate constants, with the transfer coefficients set to 0.5: circles (reversible), dashed lines (k°=0.01cm/s), squares (k°=0.001cm/s), and solid line (k° =0.0001cm/s).46 CYCLIC VOLTAMMETRY is called Nernstian, because the forward and reverse electron transfers are fast and occur simultaneously. In the irreversible region, the cathodic wave and anodic peaks are well separated, and in each case electron transfer occurs in one direction. The intermediate region is “quasi-reversible.” Qualitatively, the demarcation point separating quasi-reversible from irreversible has been reached when there is no (potential) overlap between the cathodic and anodic peak shapes. The following criteria have been suggested for evaluation of reversibility, with D = 10-5cm?/s and T = 298 K!!: Reversible: k° > 0.3 v"/? cm/s Quasi-reversible: k° > 2 x 10-5 v'/?. cm/s Totally irreversible: k° <2 x 1075 v'/? cm/s The experimentalist’s attitude toward reversibility or lack of it is shaped by his or her goals. When one desires to obtain information about either reduction potentials or the rates and mechanisms of following chemical reactions, a greater degree of reversibility of electron transfer allows for the use of a wide range of scan rates, and interpretations are simpler. However, studies of the fundamental act of electron transfer are often accomplished with totally irreversible systems. In these cases, the transfer coefficient a and the forward rate of electron transfer are easily measured. The transfer coefficient, being in the exponent of the Butler-Volmer equation, and reflecting the symmetry of the barrier to electron transfer, is simply given by: = 1857 RT 6) mE, — EyalF where E,,2 = is the half-peak potential, and n, is the number of electrons in the rate-determining step. In most circumstances n, = 1. Note that the symmetry of the barrier is reflected in the “steepness” of the cathodic wave (Figure 2-12). The anodic transfer coefficient B is determined by the same equation applied to an irreversible anodic wave. For quasi-reversible systems, the reduction potential can be obtained by curve-fitting methods.'? If the peaks are symmetrically shaped, then « ~ 0.5, and E° © (Ep,. ~ Ep..)/2- 2.3.3 The EC Mechanism Upon oxidation or reduction, an electroactive species may be activated toward bond cleavage, as in the case of methycobalamin reduction (Figure 2-13). The addition of an electron activates the molecule toward dissociation of a methyl radical. The rate constant of the following reaction is 600s~' at —30°C ina DMF/propanol solvent mixture. The cathodic transfer coefficient « for methyl- cobalamin reduction is 0.78. At even lower temperatures, the anodic peak grows in as a result of the lowering of the dissociation rate. Generally speaking, the EC mechanism is recognized by a diminished reverseTHE CYCLIC VOLTAMMETRIC EXPERIMENT a7 10 8 s Current (jAmps) -8 04 0.2 0.0 -02 -04 -06 -0.8 1.0 -12 Potential (v) Figure 2-12 Effect of transfer coefficients for a heterogeneous rate constant k° =0.0001, with a scan rate of 1V/s. fa) (b) CHs Figure 2-13 (a) Methylcobalamin. (b) Mechanism of methylcobalamin reduction in nonaqueous solution.48 CYCLIC VOLTAMMETRY peak or the lack of one. Both the chemical and electrode reactions may have different degrees of reversibility, which affect the waveshape in characteristic ways. However, at the outset, we will consider the case of a reversible electron transfer followed by an irreversible chemical reaction. In this (limiting) case, the CV peak potential will be determined by the E° of the electroactive species and the rate of the following chemical reaction. Consider the case of a reductive EC mechanism. Let us keep the parameters that describe the experimental con- ditions and properties of the electroactive species the same as in the “E” example, adding only a following chemical reaction. Experimental Settings E, = 00, F, = —0.5,E, = 0.0 v =1V/s Area of electrode A = 0.01 cm? Cell temperature T = 298K Electroactive Species Parameters Ox + e— = Red, E” = —0.25V Red = Products k.hem Co, = 100mM, C,eg = 0.0mM Diffusional coefficients D,, = D,.4 = 1 x 10° 5 cm?/s The following chemical reaction removes the reduced form from solution as it is produced near the electrode. The effect is to shift the peak to more positive values* and to increase the peak current slightly (Figure 2-14). These effects can be understood qualitatively on the basis of the Nernst equation (removing Red makes the reaction more favorable and increases the net rate of reduction). Increasing the scan rate has the effect of making the CV more “chemically reversible,” until, at high scan rates, the chemical reaction is “frozen” on the time scale of the experiment and the CV is completely reversible. The main effects of the E,C, mechanism on the CV waveshape are summarized next. Characteristics of E,C,; Mechanism . The ratio of cathodic to anodic peak current is a function of chemical rate constant and scan rate. The cyclic voltammogram can be analyzed to give an estimate of the chemical rate constant. * At some point, however, the electrode reaction will be pushed into the “endergonic” region, and the electron transfer can no longer be considered reversible, Also, at potentials negative to E°, the backward electron transfer can be small. For a more general treatment of the EC mechanism, see Chapter 6.THE CYCLIC VOLTAMMETRIC EXPERIMENT 49 = 1000 5" Current/« Amps 0.0 01 02 08-04-05 Potential/V Figure 2-14 CVs for E,C, mechanism for a series of increasing solution chemical rate constants at constant scan rate. The same effect would be observed with a constant chemical rate constant and decreasing scan rate. 2. The peak potential of the forward scan (when k.remRT/nFv > 4) is given by: RTL KehemRT = "Ton hem E, 80 + — nFY Set) (6) Several methods have been utilized to determine the rate of the following chemical reaction from a series of CVs at different scan rates. The simplest involves a comparison of i,,, and if,. The cathodic peak current is measured from the zero current baseline, while the anodic current baseline is established by the current at which the potential is switched. The experimental peak current ratios can then be compared to a previously calculated theoretical “working” curve to find the rate constant (for a first-order or pseudo-first-order reaction. '3 Parker has emphasized the use of working curves based on derivative cyclic voltammetry, which discriminates to some degree against capacitive back- ground current.!* Our preference in analyzing the EC mechanism is a simulation-based method in which the following chemical rate constant is determined by double potential step chronoamperometry. Subsequently, CV simulation analysis can be utilized to determine E° and « and k° for the quasi-reversible case. Generally, we are not fortunate enough to be presented with the E,C, limiting case. A simplified two-dimensional “zone” diagram illustrates the situation (Figure 2-15). Along one dimension are changes in the heterogencous rate constant, and along the other dimension are changes in the homogeneous rate constant. Cyclic voltammograms for several limiting cases are shown. 1. Irreversible. The electrode kinetics are slow and the chemical kinetics are slow—the CV is irreversible. The forward peak shifts —59 mV per tenfold change in scan rate.50 CYCLIC VOLTAMMETRY Heterogeneous Rate Homogeneous Rate Figure 2-15 Zone diagram for he EC mechanism. Typical shapes of voltammograms are given for limiting cases. 2. Reversible. The electrode kinetics are fast and the chemical kinetics are slow. The peak potential is not a function of scan rate, and the CV appears as in the E, mechanism. . The electrode kinetics are fast and the chemical kinetics are fast. The peak potential shifts —30mV per tenfold change in scan rate. 4. Irreversible. The electrode kinetics are slow and the chemical kinetics are fast. we A more detailed analysis can locate intermediate regions. Reversibility in the zone diagram has an inverse square root dependence on scan rate in the “heterogeneous” axis and an inverse scan rate dependence in the “homog- eneous” axis. The former case obtains because increasing the scan rate is equivalent to increasing the rate of diffusion, while in the latter case increasing the scan rate gives less time for the chemical reaction to occur. Thus, all else being equal, one can move around in the zone diagram by changing the scan rate. Klinger and Kochi have introduced a quantitative measure of reversibility in electrochemistry as the deviation of the surface concentrations from the values that would exist under Nernstian conditions.'* Reversibility thus defined is a function of position along the cyclic voltammogram (i.e., potential). This is because the heterogeneous rates are modulated several orders of magnitudeTHE CYCLIC VOLTAMMETRIC EXPERIMENT 51 (through the Butler-Volmer equation) over the entire cyclic voltammogram. A more detailed discussion of this is deferred until Chapter 6. 2.3.4 Scan Rate and the Role of Diffusion We have seen that for the E, mechanism, the CV waveshape becomes more irreversible with higher scan rates. This happens because increasing the scan rate is equivalent to increasing the rate of diffusion of the reduced material from the electrode. As a consequence, the diffusion process competes with the back electron transfer. In fact we can write the E mechanism more generally as follows: A OXsu, +O Red, Redyyp—**—> Reda This is shown schematically in Figure 2-16. We have already examined in detail the origin of the forward and reverse electron transfer rate constants. In this scheme a diffusion rate constant kp is introduced. To understand the origin of kp, we take a closer look at diffusion. Cyclic voltammetric experiments are performed under conditions of so-called semi-infinite linear diffusion. This means that the electrode dimensions are larger than the thickness of the diffusion layer (Figure 2-17) and that there is a bulk solution far enough from the electrode that the concentrations of all species remain unchanged throughout the experiment. The symmetry of diffusion under these conditions means that we need consider only diffusion perpendicular to the electrode. No net diffusion takes place in the y or z directions, parallel to the electrode surface. Diffusion is described by Ficks’ first law of diffusion (Equation 7), which states that the number of particles diffusing through a cross-sectional area per kaitt > Kchem Figure 2-16 Schematic of the EC mechanism.32 CYCLIC VOLTAMMETRY ANE | Zt ® ZA Z\s | ZN ZAlS| | | ZN Al | | AlS| | | Z\4 | Zs | AA i ZN Z\4 i ANE | (b) Figure 2-17 Representations of (a) linear and (b) spherical diffusion. unit time (the flux, J) is proportional to the concentration difference across the selected area. The proportionality constant, which describes the inherent mobility of the particles, is called the diffusion coefficient (D) and is expressed in square centimeters per second. aC, I IO (7) Taking the derivative of J with respect to distance indicates how the flux changes with distance (the difference between particles coming in and going out) and as a consequence describes the net accumulation of particles at a point (Ficks’ second law). Fick’s laws follow from the random nature of the motion of particles.* For linear diffusion, the net displacement of particles by diffusion has a square root * See Bard and Faulkner (Electrochemical Methods) or Rieger (Electrochemistry) for a more detailed look at diffusion. References 11 and 18, respectively.THE CYCLIC VOLTAMMETRIC EXPERIMENT 53 dependence on time and on the diffusion coefficient. The average net displace- ment of a particle by diffusion can be shown to be: AX = (2DAt)'? 8) Equation (8) can be used to estimate the diffusion layer thickness in a CV experiment. Let us now return to the original observation of this section, namely that diffusion rate is related to scan rate and competes with the reverse rate of electron transfer. The concentration gradients of the oxidized species (at the cathodic peak) is shown in Figure 2-18. The surface flux of both species is indicated, and by conservation of material we have: i Seep, RFA J red We can take the irreversible case to make an estimate of the diffusion rate constant k¥. The peak current is given by: i aF eee SAY yepiec = arg 7 04958 [=] v2DU2C. = Inca The terms multiplying C,, can be taken as the diffusion rate constant, which has the same units as the heterogeneous rate constant k°(a = 0.5 and T = 25°C). kp = 2.18(vD)"? (9) For D = 10-5 cm?/s, we can estimate limiting cases of reversibility, as when one rate constant is two orders of magnitude larger than the other. In this way we arrive at the “quasi-reversible” region ranging from 0.5 v'/? (the reversible Gradient = + <— Flux Concentration/mM 701820 Distance/um 30 Figure 2-18 Concentration gradient near the electrode. The tangent at a point determines the direction and rate of diffusion.54 CYCLIC VOLTAMMETRY case) to 5 x 10° v!/? (the irreversible case), in good agreement with the regions suggested earlier. 2.3.5 Competition between Heterogenous and Homogenous Reactions It is also useful to be able to compare heterogeneous and homogeneous rate constants. Irreversibility in the electrode reaction can be induced by the following chemical reaction by removing the reduced species at a rate faster than the reverse electron transfer. The heterogeneous rate constant can be converted to an equivalent homogeneous rate constant by introducing a third dimension, which converts the surface reaction rate (molcm™?s~*) to a volume reaction (mol L~!s~1), A reasonable choice is the diffusion layer thickness at one second (the time unit of the rate constant), shown in Figure 2-19. Thus, the effective volume for the surface reaction is 1cm x 1cm x (2D x 1s)'/? cm. The heterog- eneous rate can be converted to an equivalent homogeneous rate by multiplying by (2D)~'?. A factor of 1000 takes the result from cubic centimeters to liters. We have, finally, _10°kpey vol (2D)? For D = 1075 cm?/s, ky = 2.2 x 10° Kner Homogeneous reactions must be rather fast to compete with heterogeneous reactions. Even though cyclic voltammograms may appear to be irreversible (no reverse peak is apparent) the competition between heterogeneous and homog- eneous processes makes itself apparent in more subtle ways (see Chapter 6). (10) 1 cm Figure 2-19 Volume of a heterogeneous reaction.THE CYCLIC VOLTAMMETRIC EXPERIMENT 55 The three components of an electrochemical mechanism (disregarding adsorption) are electrode reaction, diffusion, and chemical reaction. These rates of these processes can be compared in a semiquantitative manner. As a consequence of the effect of scan rate on the CV result for both homogeneous and heterogeneous processes, theoretical presentations often employ dimensionless clusters. For instance, for the E, mechanism, the factor: ary ef EZ] "won controls the result, and for the E,C,; mechanism the controlling factor is: KeremRT Fv As the mechanisms become more complicated, the number of dimensionless factors needed to characterize the CV response also increases. While this approach is powerful in showing the combined effect of all parameters at once, it can become extremely abstract. 2.3.6 Visualizing the Meaning of Reversibility with Cprof The program Cprof (provided on the diskette) simulates and graphs the concentration profiles near the electrode for a reductive EC mechanism. Ox+e7 = Red EM, k° (a = 0.5) Red = Product Kerem In this way it is possible to visualize the effect of heterogeneous rate constants and rates of following chemical reactions on the concentrations of the oxidized species, the reduced species, and the product of the following chemical reaction. Cprof is for illustrative purposes only and has not been optimized. The scan rate should be set to at least 1 volt/sec, and the rate constant of the following chemical reaction can be practically set from 0 to 500s ~'. It is suggested that the reader simulate a reversible CV using the following parameters: E=-.20 k°=1cms"* Initial Potential = 0.0 V Final Potential = —0.3V Scan rate = 1 Vs~* Kerem = 0.08~* Following this, simulations can be performed varying the k° (making it smaller) and k,hem (making it larger) to visualize the meaning of reversible, quasi- reversible, and irreversible.56 CYCLIC VOLTAMMETRY 2.4 Distortions of the Faradaic Response There are several experimental realities that tend to distort the observed CV waveshape. Through a combination of experimental and theoretical methods, it is often possible to alleviate these problems to the extent that a reliable analysis becomes possible. An electrical circuit model for the electrochemical cell illustrates the origin of these problems (Figure 2-20). The electrode—solution interface is modeled as a capacitor in series with a resistance. The formation of the double layer as potential is changed gives rise to an analogue of an electrical capacitor. So far we have considered in detail only the “ideal” or faradaic current due to electrode processes. A complete model of the CV experiment must include the current due to capacitive charging as the potential changes. A resistance element models the resistance between the working electrode and the reference electrode. The potentiostat is controlling and monitoring the potential of the working electrode with respect to the reference electrode. To the extent that there is an JR drop across the solution, the potential of the WE will be misrepresented by the monitored potential. 2.4.1 IR Drop Theoretical treatments of cyclic voltammetry usually assume a cyclic linear potential sweep at the working electrode. However, the solution resistance causes a potential drop to exist across the working electrode and the reference electrode. Thus, the potential at the working electrode is really the applied potential plus the solution JR drop: Ewe = Expp + IReot (This equation is sometimes written with R, to represent “uncompensated” resistance, indicating the possibility of partial instrumental compensation of the solution resistance). Consider the effect on a typical CV. For positive (cathodic) current, the actual WE potential is less negative than the applied (measured) potential. The effect is WE RE ‘ ‘ t t Ca Raacson Figure 2-20 Equivalent circuit for an electrochemical cell.THE CYCLIC VOLTAMMETRIC EXPERIMENT 57 to shift peak potentials in a negative direction. For negative (anodic) current, the shift is in the positive direction, Thus for a reversible electron transfer, AE, > 58 mV in an experiment that has even a relatively small IR drop. Typical experiments involving 5 4A of current in nonaqueous solution with resistance of approximately 1000 can result in peak separation about 70mV. Use of the equation E gicorrecied) = Epymeas) + 1Reotarion (11) is a reasonable way to correct measured peak potentials in solution with moderate resistance problems. Note that the correction for 1R drop tracks the current (the anodic and cathodic peak currents differ in sign and in magnitude). The effect of IR drop can be included in simulation treatments. It is best to make reasonable effort to minimize the solution IR drop by: Optimizing the electrolyte concentration. Placing WE and RE close together. Using the feedback compensation method, available as a automated option in some commercial units. Using smaller electrodes. 2.4.2 Capacitive Current Another distortion of the CV from the ideal waveshape arises because the electrode—solution interface acts as a capacitor in series with the solution resistance. The most familiar capacitive current occurs in a potential step (to a potential at which no faradaic current occurs). In this case, the current is given by: p= SE ence (12) where the total capacitance is Ca = C/om? x area and AE is the size of the potential step (V), t is time, and R is resistance. A plot of In(i) versus time has a slope of —(RC,,)~' and an intercept In(AE/R). In this way the capacitance and resistance can be measured (Figure 2-21). The C/em~? (double layer capacitance) for most electrodes falls in the region 10-20 yF/cm?. For a cyclic linear sweep, the situation is more complex. The double-layer capacitance is in general a function of potential. However, the general features can be described ith constant capacitance. In this case, the capacitive current rises (starting at E,), with a characteristic time that levels off. Upon the reverse in scan direction, the capacitive current changes sign. The plateau current is equal to the i,,, = Cy x v (Figure 2-22), while the faradaic current is proportional to the square root of the scan rate. Thus, the distorting effect of capacitive current is largest at higher scan rates (Figure 2-23). The ratio of faradaic to capacitive current is independent of electrode size.