H AIL ”'5 lll‘ lllllllllllllllllllllllll r a If 0 I 8w s q 3 1293 00602 2598 LIERARY Michigan State Uni"er‘ity This is to certify that the thesis entitled Charge Transport Mechanisms in Layered Oxide and Polymeric Microsturctures: An Electro- chemical Approach presented by Randal King has been accepted towards fulfillment of the requirements for BhJ)- degree in Chemistry 9(2qu 3. flat/3V Major professor Date % ADgUSt M83 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution PLACE II RETURN BOX to romovo this checkout from your record. TO AVOID FINES return on or baton duo duo. DATE DUE DATE DUE DATE DUE UET 1““, 7.51599 ll— ff:— MSU In An Affinnotivo ActiorVEqual Opportunity Institution CHARGE TRANSPORT MECHANISMS IN LAYERED OXIDE AND POLYMERIC MICROSTRUCTURES: AN ELECTROCHEMICAL APPROACH By Randal Daniel King A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1989 ABSTRACT CHARGE TRANSPORT MECHANISMS IN LAYERED OXIDE AND POLYMERIC MICROSTRUCTURES: AN ELECTROCHEMICAL APPROACH By Randal Daniel King The complete electrochemical characterization of the mechanisms controlling charge transport in layered oxides and poly- pyridinium based polymers is established. Synthetic design of these microstructures on the microscopic level is systematically developed for the purpose of achieving specific functions. The nature of the electroactive sites in layered oxides (clays) is established and characterized. Redox ions incorporated into clays have, on the microscopic level, several possible locations: (i) electrostatically bound in the galleries between clay layers; (ii) electrostatically bound on the outer edges of the clay platelets; (iii) in voids around the clay particles; and (iv) adsorbed onto the outer edges of the clay layers as ion-pairs in excess of the charge exchange capacity (CBC) of the clay. Only the ions which are present in the voids of clay films display electroactivity. On the macroscopic level, clay platelets stack in random orientation near the electrode and a more highly ordered orientation away from the electrode. The observed electroactivity Randal Daniel King arises only from redox ions in voids created by the random face to edge stacking of clay platelets near the electrode. The amount of electroactivity from clay films can be increased by electro- polymerization of vinyl containing inorganic complexes. Significantly the redox ions in the galleries of clays can become activated by using intersalated clay to tailor the clay microstructure. Polymeric microstructures can also be manipulated with synthetic design which is established as a useful method for the controlling charge transport within modified electrode films. The rate of charge propagation of redox anions ion-exchanged in several poly-pyridinium films is observed to be dominated by electrostatic interactions involving cationic binding sites in the polymers. As the charge density of these binding sites increases so does their ability to interact with anions incorporated into their polymer matrix, thus slowing down the diffusion rate of the anion. In this way the rate at which anions diffuse in polycationic polymers soaking in nonaqueous solvents can be mediated with the synthetic design of cationic binding sites in polymers. This procedure is exploited by establishing a 2,2'-bipyridinium based polymer which possess an extremely high charge density binding pocket. Electrodes modified with 2,2'-bipyridinium films are developed into a novel electroanalytical sensor. This sensor displays unprecedented ability to extract anions from dilute solutions. Furthermore, many crucial features necessary to the successful development of electrochemical sensors in aqueous and nonaqueous Randal Daniel King solvents, which heretofore have eluded characterization, have been realized for 2,2'-bipyridinium polymer modified electrodes. ACKNOWLEDGEMENTS Many endeared friends have helped me survive the past five years. Of course, academically on the top of my list of friends to thank is Dan Nocera. I have learned a great deal about technical writing and oral presentations from Dan. His guidance towards receiving a PhD is greatly appreciated. In the early part of my research I also received useful advice and assistance from Thom Pinnavaia which is also appreciated. Additionally, I would like to acknowledge and express my gratitude to Bob Mussel] for working with me for many hours to become accustomed with electrochemical instrumentation and the Nocera lab. Of my friends I would like to thank for their support, my gratitude to my best friend, Michele is beyond words. She has stood by my side through my entire career at Michigan State, with continuous words of encouragement and mountains of affection. I also thank God for the wonderful gift of our daughter, Emily who has helped me maintain my sanity. I would also like to express my delight with working for a man who I not only can learn from but also enjoy his companionship at sporting events, picnics, tailgate parties, bars, and yes even A.C.S. meetings. The memories I will maintain from my graduate school experience throughout my life will be the fun times with my fellow students and Dan. I am certain that Bob, Mark, and Colleen were the direct cause of my taking five years to graduate. Although, they are also the reason I enjoyed myself. Recently our lab has acquired an assortment of friendly graduate students and Janice. They seem strange at first but after you get to know them you become certain they are good friends. Thank you all for your company in the lab and particularly to Janice for her delightful words of encouragement and Colleen for always being in a good mood. Finally, I would like to thank my Family, most importantly my parents for their continual support throughout my college career. I cannot express how grateful to my family I am for their constant encouragement and love over the past 10 years. vi TABLE OF CONTENTS Page LIST OF TABLES ................................................................................................ xi LIST OF FIGURES ............................................................................................... xiii I INTRODUCTION ...................................................................................... l I I EXPERIMENTAL ..................................................................................... 41 A. Preparation of Compounds ................................................... 41 1. Inorganic Metal Complexes ...................................... 41 a. General Procedures ......................................... 41 Synthesis of Fe(bpy)3(SO4), Fe(phen)3(SO4), and Fe(vbpy)3(SO4) ........ 41 c. Synthesis of (PPN)2Fe(bpy)(CN)4 ............... 41 d. Synthesis of Os(bpy)3(ClO4)2 ........................ 42 e. Synthesis of Ru(bpy)3(ClO4)2 ....................... 43 f. Synthesis of (NBu)4M06Cl14 .......................... 43 g. Synthesis of (NBu)4W(SBr14 ........................... 43 2. Organic Polymers and Monomers ......................... 44 a. General Procedures ......................................... 44 b. Synthesis of N-methylpolyvinyl- pyridinium hexafluorophosphate (PVPLMe/PFG') l ........................................... 45 c. Synthesis of {N,N'-bis[-3-(tri- methoxysilyl)propyl]-4,4'- bipyridinium}diiodide (PQ2+/21'),2 ....... 45 (1. Synthesis of 4-vinyl-4'-methyl- 2,2'-bipyridine (vbpy), 3 .............................. 46 vii Synthesis of 4-vinyl-4'-methyl- N ,N'-1 ,2-ethylene-2,2' bipyridinium (EVDQ2+),4 .......................................................... Synthesis of 4-vinyl-4'-methyl- N,N'-1,3-propylene-2,2'-bipyridinium (PVDQZ+),5 .......................................................... Synthesis of 4-vinyl-4'-methyl- N,N'-l,4-butylene—2,2'-bipyridinium (amt), 6 .......................................................... 3. Synthesis, Modification, Purification of Layered Oxides .............................................................. a. Sodium Montmorillonite (Wyoming) ........................................................... b. Synthetic Hectorite (Laponite) .................... c. Fluorohectorite .................................................. (1. Reduced Charge Montmorillonite .............. e. Intercalates .......................................................... f. Intersalated Montmorillonite ..................... 4 . Supporting Electrolytes ............................................. 5. Solvents ............................................................................ Methods and Procedures ...................................................... 1. Electrochemical Measurements ............................... 2. Preparation of Modified Electrodes ....................... 3. Spectroscopic Characterization and Instrumentation ........................................................... 4. Determination of Diffusion Coefficients .............. viii Page 46 47 43 48 48 48 49 49 49 49 50 50 5 1 5 1 5 1 53 54 C Construction and Cleaning of Working Electrodes ..... 1. PwnbflcChqmme ......................................................... a. Graphite Disc Electrodes ................................ b. Planer Graphite Electrodes ............................ 2. Platinum and Glassy Carbon Disk meumms ......................................................................... 3. Platinum Oxide Disc Electrodes .............................. 4. Preparation of Tin Oxide Electrodes ...................... I I I CLAY MODIFIED ELECTRODES .......................................................... .A. Bmignmmi ................................................................................. B. Results and Discussion ........................................................... 1. Nature of the Electroactive Sites ........................... 2. Elucidation of Clay Microstructural Control of Electroactivity .......................................... 3. Electrocatalytic Activity of Clay/Polymer films ..................................................... 4. Intersalated Clay Films .............................................. C Conclusion .................................................................................... IV POLYPYRIDINIUM BASED MODIFIED ELECTRODES ................. A. Bmkgomm ................................................................................. B. Results and Discussion ........................................................... L PVPi—NEIMhmmrhkflfidemnm ........................ 2 PQ2+ Polymer Modified Films ................................. 3. mod“ Polymer Modified Films ............................. 4 Comparison of EVDQ2+, PVDQ2+,BVDQ2+ Polymer Modified Films ............................................ ix Page 56 56 56 56 56 57 57 58 58 66 66 97 118 136 138 138 150 150 169 178 190 Page C Conclusion ..................................................................................... 204 V DEVELOPMENT OF AN ELECTROCHEMICAL SENSOR ............... 206 A. Background ................................................................................. 206 B. Results and Discussion ........................................................... 207 1. Development of an Electroanalytical Anionic Sensor for Nonaqueous Solvents ........... 207 2. Characterizatiion of Anionic Effects on the Microstructure of EVDQ2+ ................................. 213 C Conclusion ..................................................................................... 221 VI REFERENCES .............................................................................................. 222 LIST OF TABLES Page Conducting Silanes Used to Modify Oxide Surfaces ............... 5 Monolayers of Electroactive Organics Attached to Carbon Electrodes ................................................................................. 10,11 Class I Polmers with a Polyvinyl Backbone .............................. 16—18 Electroactive Organic Monomers Used to Form Class [Polymers ............................................................................................... 19,20 Class III Insulating Polyionic Polymers ..................................... 22,23 Time Required to Reach Maximum Current Response for Different Films Soaked in 0.2 mM 0s(bpy)32+ ............................................................................................... 81 Amount of Electroactivity From Different Clay Films of Varying Thickness After Soaking in Aqueous Fe(bpy)32+ Solutions Containing 0.1 M Na2804 ........................ 91 Amount of Electroactivity from 0.38 pm Montmorillonite Films Resulting from Incorporated or Polymerized Fe(vbpy)32+ ............................................................ 111 xi 10 11 12 13 Amount of Electroactivity from Different Clay Films after Soaking in 5.1 mM Fe(bpy)32+ ............................................ Diffusion Coefficients of W6Br142" in Acetonitrile Determined by Different Electrochemical Techniques .............................................................................................. Apparent Diffusion Coefficients and FWHM Values for EVDQ2+ Films With Varying Thicknesses ........................... Apparent Diffusion Coefficients and FWHM Values for PVDQ2+ Films With Varying Thicknesses ........................... Apparent Diffusion Coefficients for EVDQ2+ and FVDQ2+ Films w/wo Added W68r142' ......................................... xii Page 123 164 214 215 219 LIST OF FIGURES General reaction scheme for the reaction between organo-silanes and hydroxy functional groups on metal and semiconductor surfaces .............................................. Reactive pathways that are used to convert insulating Silanes, which are covalently attached to electrodes, into conducting films ................................................. Reaction schemes for attaching organic monolayers to carboxylic acid functional groups on carbon surfaces .......... Schematic depicting three classes of conducting polymers: (i) Class I possess the redox active site in the polymer backbone; (ii) Class II have redox material covalently linked to a moeity on the polymer chain; and (iii) Class III possess electrostatically bound redox ions within a charged polymer matrix ................................................................................. Ideal cyclic voltammogram obtained from a thin film of electroactive material attached to an electrode .............................................................................................. xiii Page 15 27 10 Page Ideal cyclic voltammogram obtained from W6Br142' freely diffusing through an acetonitrile solution containing 0.2 M TEAP ........................................................................ 33 Idealized structure of smectite clay with a hydrated cation between layers .................................................. 60 Model of a clay film on an electrode containing ion- exchanged polypyridyl metal complexes and electrolyte ................................................................................................. 64 Increasing cyclic voltammetric response from Os(bpy)32+ incoporating into a montmorillonite film coated on graphite in an aqueous solution containing 0.2 M sodium acetate and 0.2 mM 0s(bpy)32+ ................................................................................................. 68 Cyclic votammograms of a 80 % pre-exchanged montmorillonite film on pyrolytic graphite: (a) soaking in an aqueous solution containing 0.1 M N32804; and (b) electrode from above placed in an aqueous solution containing 0.1 M NaZSO4 and 0.2 mM0s(bpy)32+ ....................................................................................... 71 xiv 11 12 13 14 15 A reaction scheme devised to oxidize Os(bpy)32+ located in the galleries, with utilization of Fe(bpy)32+ as a redox charge shuttle agent ............................... Reversible cyclic votammogram upon the first oxidation scan of Fe(bpy)32+ ion-exchanged into a montmorillonite film containing pre—exchanged Os(bpy)32+ while soaking in an aqueous solution containing 0.1 M Na2S04 ..................................................................... Current responses obtained from Os(bpy)32+ exchanged into montmorillonite films from an aqueous solution containing 0.2 sodium acetate and 0.2 mM Os(bpy)32+: (a) 78 % pre-exchanged Os(bpy)32+; (b) 38 % pre-exchanged 0s(bpy)32+; (c) Na+-exchanged (humidified); (d) Na+-exchanged (air dried); and (e) reduced charge ........................................................ SEM images of Na+-exchanged montmorillonite films on pyrolytic graphite, (a) 1000 x and (b) 7800 x ................... The current response obtained as a function of time from various clay films adsorbed onto graphite soaking in 0.2 M sodium acetate and 0.2 mM Os(bpy)32+: (a) laponite; (b) montmorillonite; and (c) fluorohectorite ........................................................................................ XV Page 75 77 80 84 88 16 17 18 19 Page Model depicting locations of electroactive and non- electroactive polypyridyl metal complexes within a clay film ..................................................................................................... 96 2+ Growth of poly-Fe(vbpy)3 at 10 min intervals in montmorillonite adsorbed onto pyrolytic graphite while soaking in an acetonitrile solution containing 0.2 M TEAP and 2.0 mM Fe(vbpy)32+ ........................................... loo Cyclic voltammetric waves obtained from a poly- Fe(vbpy)32+/montmorillonite electrode in acetonin'ile containing 0.2 M TEAP: (a) upon scanning the electrodes potential from 0.00 V to +1.20 V, to -1.80 V, and back to 0.00 V vs. SCE; (b) after the reduction waves had been scanned once prior to measurement; and (c) after the oxidation wave had been cycled once prior to measurement ................ 102 Cyclic voltammogram of poly-Fe(vbpy)32+ adsorbed onto pyrolytic graphite while soaking in an acetonitrile solution containing 0.2 M TEAP: (a) upon scanning the electrodes potential from 0.00 V to +1.20 V, to -1.80 V, and back to 0.00 V vs. SCE; (b) after the reduction waves had been scanned once prior to measurement; and (c) after the oxidation wave had been cycled once prior to measurement ............... 105 xvi 20 21 22 23 24 Page A plot of the peak height ip'a (uA) vs. the scan rate (mV/sec) for the CV reproduced in Figure 18 .......................... 109 Cyclic votammograms scanned at 2 mV/sec, of a clay/poly-Fe(vbpy)32+ film soaking in an aqueous solution containing 0.2 M LiClO4: (a) after the addition of 4.2 mM Fe(CN)64' and; (b) before the addition of 4.2 mM Fe(CN)64” .......................................................... 115 Model depicting origins of electrocatalysis from a clay/poly-Fe(vbpy)32+ film on an electrode .............................. 117 The first two cyclic voltammograms of a Fe(bpy)32+ intersalated montmorillonite film on pyrolytic graphite immersed in a dichloromethane solution containing 0.2 M TBAP ........................................................................ 121 The X-ray diffraction patterns obtained from Fe(bpy)32+ intersalated films on pyrolytic graphite: (a) before the electrolysis of the film; and (b) after the electrolysis of the film in a dichloromethane solution containing 0.2 M TBAP ...................................................... 126 xvii 25 26 27 28 29 Page Cyclic voltammetric waves of a polymerized Fe(vbpy)32+/intersalated montmorillonite film adsorbed onto pyrolytic graphite while soaking in an acetonitrile solution containing 0.2 M TBAP ....................... 129 A plot of the peak height i1Ml (uA) vs. the scan rate (mV/sec) for the CV reproduced in Figure 25 .......................... 132 The X-ray diffraction patterns obtained from polymerized Fe(bpy)32+/intersalated montmorillonite films on pyrolytic graphite: (a) before the electrolysis of the film; and (b) after the electrolysis of the film in a acetnitrile solution containing 0.2 M TEAP ........................................................................ 134 Annihilation reaction between oxidized M6X14' cluster and reduced pyridinium ions to produce cluster in its emissive excited state or ground state .............. 141 Distance dependence of the differential bimolecular rate constant for the excited-state (es) and ground- state (gs) electron-transfer channels for the reaction between M06Cll4' and one electron reduced 4-cyano-N-methylpyridinium ........................................ 143 xviii 30 31 32 33 34 Page Reaction scheme for generation of excited state M6X14J cluster upon annihilation of oxidized cluster ion and a one-electron reduced poly- pyridinium based polymer ................................................................ 147 A 250 MHz NMR spectrum of N-methylpolyvinyl pyridinium hexafluorophosphate in CD3CN ................................ 153 Cyclic voltammograms of: (a) W6Br142‘ dissolved in acetonitrile containing 0.2 M TBAPF6 at a bare platinum electrode; (b) a PVP+—Me film with a surface coverage = 2.2 x 10’6 moi/cm2 absorbed onto a platinum electrode immersed in an acetonitrile solution containing 0.2 M TBAPF6; and (c) a PVP+—Me film ion-exchanged with W6Br142' immersed in an acetonitrile solution containing 0.2MTBAPF6 ......................................................................................... 157 A plot of peak height ip’a (uA) vs. the scan rate (mV/s) for the anodic wave of the CV reproduced in Figure 32c ................................................................................................. 159 The rate of WéBer‘ ion departure from an PVPi— Me film on a platinum electrode after potential steps to +1.3 V vs. SCE for increasing times ...................................................................................... 168 xix 39 4O 41 Page Cyclic voltammograms of a platinum electrode coated with a EVDQ2+ film (surface coverage = 1.7 x 10'8 mol/cmz) and dipped into an acetonitrile solution containing 0.2 M TEAP upon scanning the potential: (a) past the one electron reduction waves of EVDQ2+; (b) past the one electron reduction waves of EVDQ2+ after addition of 15 11M W6Br142‘ to the acetonitrile solution; and (c) past the oxidation of W6Br142' after incorporation into the EVDQ2+ film .............................................................................................. 185 Cyclic voltammograms of W6Br142‘ incorporated into a EVDQ2+ film adsorbed onto a platinum electrode (surface coverage = 1.7 x 10’8 mol/cmz) immersed in an acetonitrile solution containing 0.2 M TEAP recorded at (a) 10 V/s and (b) 33V/s ........................ 188 Cyclic votammogram of a PVDQ2+ film adsorbed onto a platinum electrode (surface coverage = 4.0 x 10'9 mollcmz) immersed in an acetonitrile solution containing 0.2 M TBAP ........................................................................ 193 xxi 42 43 44 Page Cyclic voltammetric waves obtained from a PVDQ2+ film coated onto a platinum electrode (surface coverage = 1.1 x 10'8 mol/cmz) dipped into an acetonitrile solution containing 0.2 M TEAP and 20 DM W6Br142': (a) upon scanning the electrodes potential from 0.00 V to +1.40 V to -1.10 V and back to 0.00 V vs. SCE; (b) after the reduction waves for the PVDQ2+ polymer had been scanned once prior to measurement; and (c) after the oxidation wave of W6Br142' had been cycled once prior to measurement .......................................................................... 195 Cyclic votammogram of a PVDQ2+ film on a platinum electrode soaking in an acetonitrile solution containing 0.2 M TBAPF6, 25 11M W6Brl42' and 50 11M IrC162' ................................................ 198 A plot of the apparent diffusion coefficients, Dapp for WéBer' incorporated into PVPt—Me, PQ2+, BVDQ2+, PVDQZ“, and EVDQ2+ polymeric films as a function of the number of binding sites ............................................................................................. 201 xxii 45 46 47 Page A plot of the distribution coefficients KD for W6Brl 42' incorporated into PVP+—-Me, PQ2+, BVDQ2+, PVDQ2+, and EVDQ2+ polymeric films as a function of the number of binding sites ..................................................... 203 Cyclic voltammogram of W6Br142‘ incorporated from an acetonitrile solution containing 0.2 M TEAP and 5 x 10'8 M W6Br142’ into a EVDQ2+ film adsorbed onto a platinum electrode (surface coverage = 4.1 x 10'9 mollcmz) ....................................................... 210 A plot of the cyclic voltammetric peak current for the oxidation of W6Brl42‘ incorporated into EVDQZ” films as a function of the concentration of W6Brl42' in the contacting acetonitrile solution for platinum electrodes with surface coverages of, (a) 2.1 x 10’9 moi/em2 and (b) 1.1 x 10‘9 moi/em2 ............................................ 212 xxiii CHAPTER I INTRODUCTION The investigation of modified electrodes has clearly been the most exciting and active area in electrochemistry over the past few decades.“6 Studies in this area have led to a clearer understanding of many aspects of electrochemistry including new insights about the electrode double layer and effects of adsorbed organics in the double layer, the effect of chemical modification of the electrode surface on heterogeneous electron transfer and diffusion rates, and a greater understanding of the role of distance, reorganizational energies, and electronic factors on homogeneous and heterogeneous electron transfer rates. These fundamental studies have had direct impact on the practical applications of organic and inorganic polymers in the 7-12 13-23 areas of semiconductors, electrocatalysis, 31-38 organic electronics,39'4o energy 51-61 synthesis,24'3o biochemical analysis, 41-50 conversion, and redox-chemical 62-65 electrochromic displays, sensors. The area of modified electrodes was initiated in 1973 when Hubbard and Lane"6 deliberately adsorbed monolayers of organic olefins onto platinum electrodes for the purpose of modifying the electrode double layer. The olefins were either intrinsically electroactive or possessed a free coordination site that was used to bind electroactive inorganic ions. These studies provided the first insights into unique issues in regard to modified electrodes and their double layers. Since these pioneering experiments, the field of modified electrodes has steadily progressed from organic monolayers towards the development of more complex microstructures on the electrode surface, including multilayer amorphous organic polymers and recently more highly ordered crystalline inorganic microstructures. Techniques for covalently attaching monolayers to electrode surfaces first utilized condensation reactions on reactive SnOz ‘57 Conducting silanes were coordinatively linked to this surfaces. wide—bandgap semiconductor through Sn-O-Si linkages following the general reaction scheme shown in Figure 1. The silanization reaction works well with virtually any alkoxy- or chloro-silane and on all electrodes possessing an OH functional group. Examples of conducting silanes, electrodes, and reduction potentials of the silanized polymer films are listed in Table 1. An alternative approach to the preparation of electroactive silanized microstructures is to modify the electrode with nonconducting silanes that are subsequently converted to conducting films through chemical reactions. Some of these silanes and their respective reactions are shown in Figure 2. Surface modification is not limited to hydroxylic functionalities but can include a variety of other organic groups as well. Miller et al.88 demonstrated that carboxylic acid groups present on the surfaces of carbon can be functionalized with organic monolayers according to the reaction scheme shown in Figure 3a. Reactions 3b and 3c are two additional general reactions which have since been reported.”'91 In almost all of these studies, the electroactivity of the film is generated by coordination of an inorganic complex to a functional group on the organic monolayer. Exemplary systems which have been reported in the literature, are presented in Table 2. Figure 1 General reaction scheme for the reaction between organo- silanes and hydroxy functional groups on metal and semiconductor surfaces. «comm...— m S x m I I 2 .GW I on: ~— .5 .=< .5 .2. .00 mm 60 \\ \ mo \ \x m + no >/mmAm \ x 4 e a l W \ axe/e \s \ + x N 5. Conducting Silanes Used to Modify Oxide Surfaces Table 1 b Silane' Electrode E°surf Ref. Fe[CpSi(OCH2CH3)3]2 Pt/PtO +0.60 6 8 , 69 Ge/GeO +0.45c 7 O FeCp2(|.t-SiC12) Au/AuO +0.43 6 9 CpFeCpSiC12 Si/SiO +0.08c 7 1 - 7 3 Pt/PtO +0.51 7 0 Au/AuO +0.43 7 4 Pt/PtO +0.40 7 5 "(CH2CI3HMCH2CHiy szFe COO(CHz)3Si(OCH3)3 Pt/PtO +0.80‘I 7 6 / . N—(CHfl3Sl(OCH3)3 \ SnOz +1.05 77,78 CH3 N / (bPY)2R '1 \ 2CH28iCl3 'Cp = cyclopentadienyl, bpy = bipyridine. bPotential in volts vs. SCE. cBroad waves. “Chemically irreversible wave. Figure 2 Reactive pathways that are used to convert insulating silanes, which are covalently attached to electrodes, into conducting films. 7 7 5 OSi(CH2)3N(CH2)2NH2 [3.5-(N02)Ph] 7 9 REC] ‘ > [3-(N02)Ph] 80,81 [4-(N02)Ph] 8 0 pl-anisyl CH3 N‘N 82 / Ph [(bpyCOOH)- Ru(bpy)2]2+ 3 3 o REOH > [pyRu(bpy)2N021+ [pyRu(bpy)2NO] s s I >~=< 1] 85,86 8 s xg—omcnzcnr-C) / _ / _CH3 87 RC1 G -Ru(bpy)2Cl 81,83 /\ Figure 2 Figure 3 Reaction schemes for attaching organic monolayers to carboxylic acid functional groups on carbon surfaces. ail§l m 953...“ some5232323833-2.2 u ooo ... 0.... as 0.... ejav as o§i smzm fume- \“ i m \\\ 68:6 o- a l m \\\ N68 .8930 n 2 Monolayers of Electroactive Organics Attached 10 Table 2 to Carbon Electrodes Electrode/-C(O)Rb E°surf" R- modifying agent Ist wave 2nd wave Ref. -(m-NH2)4TPP -1.50 -l.45 92,93 -Fe(p-NH2)4TPP -0.18 -1 . 18 93,94 -Co(m-NH2)4TPP +0.10 -0. 86 9 3 —Ni(p-NH2)4TPP -1.22 -l .80 93 ,94 +0.44 - 9 3 —NH—O—~ CpFeCp -0.10 - 9 5 —NH-CHf‘<::N—RU(EDTA) 11 Table 2 (cont'd.) Electrodel-C(O)R E° " surf R- modifying agent lst wave 2nd wave Ref +0.52 - 1 4 +0.20 - 9 6 -0.66 - 97 + + — NH-(Cth-N N- CH3 “Potential in volts vs. SCE. bTPP = tetraphenylporphyrin, Cp = cyclopentadienyl, EDTA = ethylenedinitrilotetraacetate. 12 The modification of electrodes greatly expanded in scope and breadth with the advent of electrodes modified with polymers.68'98'1°4 In contrast to the modified electrodes preceding polymer modification, polymer microstructures can be several hundred monolayers in thickness. There have been several methods developed to adhere polymers to electrode surfaces including physical adsorption, low solubility in the contacting solvent, and covalent attachment by using several different techniques. Dip coating involves soaking the electrode in a dilute solution of the polymer for one hour to several days,77'73v1°5'107 whereas droplet evaporation requires a solution of the polymer to be deposited directly onto the electrode surface followed by slow evaporation of the solvent or spin coating.75'1°3-109 Alternatively, the application of a radio frequency plasma discharge to gaseous organic monomers has been used to initiate polymerization directly on the electrode 89,103.110-112 surface. Polymer modified electrodes can also be prepared by oxidative or reductive deposition.1°1'113 This procedure takes advantage of changes in solubility which accompany electrochemically induced changes in ionic states. A different electrochemical technique utilizes free radical initiated polymerization of redox active organic monomers.““'116 Upon their reduction or oxidation, many organic compounds will polymerize (e.g. phenols, anilines, pyrrole, and olefins) and produce strongly adhering films on electrodes. For all of these techniques the polymer is believed to adhere to the electrode surface by some combination of physical adsorption and insolubility in the contacting solvent. Polymer films can be 13 covalently anchored to the electrode surface too by utilizing the same surface chemistry of hydroxyl groups with silanes as described in Figure 1. For this method the presence of water is required to initiate hydrolysis of the silane thereby resulting in polymer formation.‘58'69'75'109 Polymers used to modify electrode surfaces can be subdivided into the three classes displayed in Figure 4: electrochemically active (Class I); insulating which are converted into conducting by covalent coordination of a electroactive species (Class II); and charged insulating which incorporate electroactive ions through ion-exchange processes (Class 111). Class I polymers possess electroactive moieties built into the polymer backbone. The majority of investigations on Class I polymers to date have utilized polyvinyl backbones. Table 3 presents a selected listing of electrochemically active Class I polymers possessing a polyvinyl backbone and their respective reduction potentials. Additional Class I polymers which do not possess the vinyl backbone are listed in Table 4, along with their respective reduction potentials. Typically Class II polymers are converted to conducting by using a residue of the polymer as a ligand to covalently bind transition metal complexes. Class II polymers have been comprehensively reviewed in a recent article by Abruna.6 The first example of a metal complex coordinated to a polymer was by Oyama and Anson107 who coordinated Ru(EDTA)' and Ru(NH3)52+ to the pyridine and nitrile groups of polyvinylpyridine and polyacrylonitrile, respectively. Several other studies utilizing the coordination chemistry of ruthenium-based metal complexes with polyvinylpyridine have since been 14 Figure 4 Schematic depicting three classes of conducting polymers: (i) Class I possess the redox active site in the polymer backbone; (ii) Class II have redox material covalently linked to a moeity on the polymer chain; and (iii) Class III possess electrostatically bound redox ions within a charged polymer matrix. 15 Class I Polymer o - Electroactive Species L .- Coordination Site 0 - Electroactive Species C Class III Polymer @- Charged Binding Site 0 - Electroactive Species 16 Table 3 Class I Polymers with a Polyvinyl Backbone '(CHz'CHR)' Eosurfa MethOd or R lst Wave Preparation Ref. -1.60 Dip Coated 105,117 Nitrobenzene +0.16 Dip Coated 1 1 8 0 GI Jane“... Q. a. Acryloyl Dopamine -CpFeCp +0.40 Electro- l 01 a Ferrocene Deposited +0.42 Plasma 103a,119 +0.49 Droplet 1 2 0 Evaporation O _o_ S S\ - :8: ‘s2 Carboxytetrathiafulvalene 17 Table 3 (cont'd.). -(CH2-CHR)- E°surf" Method of R lst Wave Preparation Ref. -1.10b Droplet 121 Evaporation O + 4-vinyloxycarbonyl- l-methylpyridinium -0.45c Electro- 122 Polymerized CH3 \ / +N + N,N'-ethylene-4-methyl-2,2'- bipyridinium (EDQ2+) -1.88d Droplet 123 Evaporation 000 O 9,10-diphenylanthracene 18 Table 3 (cont'd.). -(CH2-CHR)- E°wrf' Method of R lst Wave Preparation Ref. -0.45° Droplet 124,125 Evaporation ‘Q—i— N:\_--/> <\—/:N+— CH3 N-methyl-N'-methylphenyl-4,4'- bipyridinium aPotential in volts vs. SCE. I’A second reduction wave appears at -2.10 V vs. SCE. cA second reduction wave appears at -0.92 V vs. SCE. dAn oxidation wave appears at +1.40 V vs. SCE. eThe second reduction wave was not reported. 19 Table 4 Electroactive Organic Monomers Used to Form Class I Polymers Electroactive E° “ surf Organic Monomer lst wave 2nd wave Ref. _ - +0070 0085" 126,127 1.‘ LINK) .. s _ EN" HNln 3,7-diaminophenothiazine +1.20” +1.54b 115b,128 N n pyrrole _ q +0.65c ' 129,130 CH2; 8 ln 3-methylthiophene 20 Table 4 (cont'd.). Electroactive Egon” Organic Monomer lst wave 2nd wave Ref. _ - +0.80c ' 131,132 CH3 |+ | .CH3 .. n N,N-dimethylaniline -0.55 -0.95 133,134 P OMe Me l I + — — + (i "L ?1(CH2)3-N N- (CH2)3?1" .. OMe MeO . n N,N'-bis(dimethoxysilyl)propyl-4,4'- bipyridinium (902+) -0.42 -0.86 1 2 2 I + — — + -S|i(CH2)2—Q"CH2' N'CHs OMe N-methyl-N'-4-(2-methoxysilyl)-ethylbenzyl~4,4'- bipyridinium (BVSi2+) l 5n 'Potential in volts vs. SCE. I’Chemically irreversible waves. cVery broad wave. 21 reported.‘35'l37 For the majority of the investigations in this area, the polyvinyl backbone is formed via electroreductive polymerization of vinyl containing ligands that coordinate the transition metal. In the initial studies utilizing this technique Murray, Meyer, and coworkers electroreductively polymerized vinylpyridine and vinylbipyridine ligands coordinated to iron, 114 osmium, and ruthenium. A variety of vinyl-ligands coordinated with metal complexes have been polymerized, including vinyl- phenanthroline, vinyl-terpyridine, and pyrrole-pyridines.l38’141 This general technique can be applied to virtually any transition metal complex that possesses a pyridine based ligand.”2'145 In the case of Class III polymers, electroactivity within Class III polymers is typically obtained by electrostatically binding redox active ions. The method of ion-exchanging redox species into polyionic films supported on electrodes was first introduced by Oyama and Anson in a now classic paper.146 In this study protonated polyvinylpyridine (PVP) was used as the support to electrostatically bind polyanionic ions such as IrCl62' and Fe(CN)63'. Several insulating polyionic polymers in conjunction with their incorporated ionic redox active counterparts are presented in Table 5. The unique ability of these films to indiscriminately incorporate a wide variety of ions make polyionic polymer modified electrodes a promising area for further investigations. Also it should be mentioned that some Class I and II polymers are polyionic and can electrostatically bind redox ions. In this case Class I and II polymers behave identically to Class III polymers. However, this differentiation in Class III polymers will not be made for the sake of simplicity. 22 Table 5 Class III Insulating Polyionic Polymers Electrostatically Polymer Bound Redox Ions Ref. -(CH2CH),,- Fe(CN)53'/4', Co(CN)63'/4' 147-149 I Ru(CN)63"", Mo(CN)g3'/4' N - - |+ IrC162 /3 R polyvinylpyridiniuma ll M0 CN 344-, W CN 3-/4- 150 -(CH-C-NH)n- ( )8 ( )8 (CH2), Fe(EDTA)"2'. Co(C20.)a3"“ +NH3 poly(L-lysine) - CH -CH - - - - - ( 2 )“ Fe(CN)63/4, zmrrrsf"4 151,152 +| Et-lI‘I-Et H poly—p-(diethylamino- methyl)styrene TPPS = meso-tetrakis- (4-sulfonatophenyl)porphine 23 Table 5 (cont'd.). Electrostatically Polymer Bound Redox Ions Ref. -(CH2CH)n- Os(bpy)33+/2+, Ru(bpy)33+/2+ 157,158 Cotbpy)33*’2". RutNH3)83*’2* 803' polystyrenesulfonate -(CF2-CF2);.—(CIF-CF2),- O CF3 'errc'r ('5 .1... is, is,- anionic perfluoropolymer (Nafion®) Os(bpy)33+/2+, Ru(bpy)33+/2+ 153 -15 6 Co 90.6 mV, whereas activity effects leading to attractive stabilizing interactions can yield EFWHM < 90.6 mV;”‘6’187 (ii) alternatively, Peerce and Bard163 have presented rather convincing evidence through digital simulations that multiple E° values in combination with activity effects can account for the observed deviations from the ideal EFWHM of 90.6 mV. The diffusional behavior sometimes observed for Class I and II polymers is analogous to that for freely diffusing ions; and the current in this case is described by ip = 2.65x105n3’2ADct1/2v1/2C* (5) where C" (moles/cm3) is the concentration of electroactive species in the bulk solution. Eq 5 predicts that a plot of peak current (ip) vs. the square root of scan rate (121/2) should be linear with a y- 31 intercept passing through the origin. The diffusion coefficient for charge transport Dct is obtained directly from the slope of this line. In contrast to the cyclic voltammograms corresponding to thin cell behavior (Figure 5), diffusive voltammetric waves exhibit the characteristic tail shown in Figure 6. The cyclic voltammograms have non-zero theoretical peak separations given by, RT 56.5 _ _— 0 IBM-E 1-2.2 - n mV (25 C) (6) 9.0 n—F Experimentally determined values of AED often exceed 56.5/n mV. Explanations for the larger peak separations are uncompensated film “9"“ slow heterogeneous electron transfer,189'190 and resistance, internal polymer effects (differences in swelling and or steric effects).163 The accepted method for determining E° values for both surface bound and diffusional waves has been to report the average of Ep,a and Ep.c measured from cyclic voltammetric studies. This procedure does not consider the possibility of slow kinetics which can cause unsymmetrical deviations in the locations of EM and Ep,c. Therefore, reported E° values determined by averaging Epfi and Ep,c for some polymer modified electrodes are not exact. In contrast to Class I and II polymers, charge propagation in Class III polymers is complicated by the fact that charge transport mechanisms can include molecular diffusion in addition to electron hopping. This in itself does not make the determination of diffusion coefficients any more difficult because the process of physical diffusion is indistinguishable from electron hopping, and the sum of 32 Figure 6 Ideal cyclic voltammogram obtained from W6Br142‘ freely diffusing through an acetonitrile solution containing 0.2 M TEAP. 33 p’° Pia Figure 6 34 the two processes simply appears as a new overall diffusion coefficient. The difficulty arises because the diffusion of ions electrostatically bound within polyionic films is much slower than that for their counterparts freely diffusing in solution. Thus, self electron exchange between adjacent electroactive sites, which is negligible in solution,191 becomes competitive with physical diffusion in the polymer. Several groups have proposedlM'ms'‘92;193 that the Dahms191 and Ruff194 treatment of freely diffusing ions in solution, C D .p = D0 + Dot (7) also applies to ions electrostatically bound in polymers. In eq 7, Dexp (cmzlsec) is the experimentally determined diffusion coefficient (often referred to as the apparent diffusion coefficient, Dapp), Do (cmzlsec) is the diffusion coefficient which would be measured in the absence of any contribution from self electron exchange (essentially the corresponding value to Dct for Class I and II polymers), and Det is the diffusion coefficient for electron self exchange. The electron transfer term Dct can be expressed in the following form, it]: 62C D __.El__ (8) et " 4 and Dexp can be rewritten as, 35 at 62C - L where kex (M‘ls'l) is the second order rate constant controlling electron self exchange, 6 (cm) is the electron transfer distance, and C (mol/cm3) is the concentration of the co-reactant. This equation demonstrates that ions with large kex will contribute most significantly to the overall charge transport process. This prediction, was confirmed with experiments conducted by Buttry and Anson)” Facci and Murray,196 and later by Bard and coworkers using a more 97 tenuous technique.1 The Buttry and Anson experiment employed two structurally similar ions, Ru(bpy)32+ and Co(bpy)32+ in Nafion®, possessing exchange rate constants ktx differing by several orders of magnitude. Although the values of Dexp for these two ions were shown to be significantly different, implicating electron self- exchange as the possible cause, the experimentally determined diffusion coefficient Dexp did not exhibit a linear dependence on the concentration as expected (eq 9) for a diffusional process governed by electron self-exchange. In a now classic paper, Buttry and Anson198 attributed this anomaly to the existence of hydrophilic and hydrophobic phases within the Nafion® film. Introduction of competing electron exchange in the hydrophobic phase, the hydrophilic phase, and across the two phase boundaries accounted for the charge transport properties of the film. This model for charge transport has experimentally been verified by using rotating disc 199-201 voltammetry. As an outgrowth of these studies, the overall charge transport for electroactive ions bound within polyionic films 36 is believed to be controlled by physical diffusion through the hydrophilic phase, electron or ion hopping through the hydrophobic phase (referred to as the Donnan domains), and/or cross-phase electron transfer or place exchange between the two phases. Cyclic voltammetric behavior for Class 111 films is analogous to that observed for Class I or II polymers. For thin films, if electrolysis of the trapped ions is completed during the time required to traverse the cyclic wave, then thin cell behavior is observed; and for thick films or fast scan rates diffusional behavior is observed. Rotating disc electrodes have been most extensively used to determine rates of electron transfer for electrocatalysis, and for theoretical studies designed to elucidate the limiting steps of the charge transport 202-207 process. It is obvious from the above discussion that the accurate determination of diffusion rates (Dct and D ) is crucial for the exp physical characterization of Class I, II, and III polymers. The electrochemical technique of choice for determining Dct and Dexp is chronoamperometry. It has an intrinsic advantage over cyclic voltammetry because (i) Dct can be determined for films which obey thin cell behavior (which is not the case for cyclic voltammetry, see eq 5) and (ii) uncompensated film resistance can be easily overcome through the application of a large overpotential.119 For this case, the current i(t) obtained from the polymer will obey the Cottrell equationl‘so'161 37 1/2 t . _nFAD C 10 1(0- —ul/2gtll2— ( ) where i(t) (amps) is the diffusionally controlled current obtained at time t (sec). This equation is only valid at short times, (i.e. < 50 ms) for Class I and II polymeric filmsl19'159'160’1‘57'177 and Class III films which have low redox loading.2°8"2l3 At longer times the concentration gradient within the film reaches the film-solution boundary, and the current is given by the finite diffusion relationship119 nFAD 1”c" k+1 2d2 i(t): 1,2—“—-k(2=01,2(-1)k1exp(}§:’,2-e>expl(n), ll) (11) CI. where k is an integral counter. The current time response for thin films of Class I polymers have been shown to conform to eq 11.119.159 Although chonoamperometry has been widely accepted as an accurate method for the determination of diffusion coefficients for all three classes of polymers, some discrepancies have been observed for diffusional currents. The most compelling data has been put forth by Majda and Faulkner,214 who determined the diffusion coefficient Dcxp for Ru(bpy)32+ in a Class III polymer (polystyrene sulfonate) by using a luminescence quenching technique. Values of Dup that were 40-80 times larger than those determined with chronoamperometry were measured. Because the luminescence 38 technique precludes counter-ion motion as the rate limiting step in the charge transport process, the discrepancy in Dexp for the two methods suggests that either charge transport of ions within polystyrene sulfonate is limited by these counter-ion motions, or that using a transient perturbation technique such as chronoamperometry is flawed in the determination of Dex Although the former p. explanation has been embraced, presumably due to the fact that the majority of the Dexp and Dct values been determined with chronoamperometry, recent studies by Faulkner et 01. clearly demonstrate this technique to be fraught with error. Faulkner and his coworkersns have recently determined, with a steady state experiment involving a rotating ring disc electrode, that Os(bpy)32+ electrostatically bound within a Nafion® film actually diffuses 100 times faster than previously thought. The results clearly show that diffusion coefficients determined by using chronoamperometry should only be used to compare a series of data and absolute values of Dexp measured by chronoamperometry are tenuous at best. As is evident from this chapter, charge transport mechanisms in polymer modified electrodes has been studied extensively and are fairly well understood. Investigations have primarily focussed on the intrinsic physical properties of charge propagation in polymers, rate determining steps, theoretical modeling, and applications with little regard on how charge transport properties of films are controlled by their microenvironment. The field of modified electrodes has matured to a level that researchers are now using the knowledge gained from existing polymeric systems to develop industrially useful devices. Although the amorphous nature of 39 polymers makes the characterization of their microstructures difficult, systems with the most potential with regards to new advances require the ability to control polymer microstructures, which in turn require a clear understanding of polymer microenvironments and existing guest-host interactions within the film.39"° The wide variety of covalent or electrostatic binding sites provide a great deal of flexibility for controlling the microstructure of amorphous materials. Alternatively, preformed highly ordered microstructures deposited onto electrode surfaces can also provide convenient probe systems for establishing charge transport] microenvironment relationships. Thus synthetically tailored polymer microstructures and the development of highly ordered microstructures on electrode surfaces will significantly broaden the scope of modified electrochemistry. This thesis presents fundamental studies that elucidate and clarify charge transport processes in polymeric and inorganic host matrices and utilizes this knowledge to design systems of practical applicability. Chapter III presents investigations that probe and define the charge transport processes within two-dimensional layered silicates (clays). By combining physical and electrochemical techniques, a unified model for electroactive clay films has been constructed. The nature of the electroactive sites can be manipulated by appropriate modifications of the clay film. With the information gained from these studies electrocatalytic activity has been observed from clay/polymer composites. In Chapter IV investigations of the electrochemical behavior of inorganic cluster ions electrostatically bound within polymeric 40 pyridinium and bipyridinium films in nonaqueous solvents are presented. The charge propagation rates of cluster ions are shown to be controlled by electrostatic interactions within the polymer microenvironment. By systematically tailoring the binding site of the polymer, specific control of the charge transport rates mediated by the polymer backbone and the electrostatic cluster sites has been achieved. The systems presented in this Chapter provide benchmarks for microstructure/electroactivity relationships. The structural design of the binding sites in bipyridinium microstructures of polymers is utilized to develop a novel electroanalytical sensor which is discussed in Chapter V. The sensor developed utilizes a 2,2'-bipyridinium based polymer which can detect extremely dilute concentrations of monoanions from both aqueous and nonaqueous solutions. Furthermore, electrolyte effects on the diffusion rates in these films support recently proposed charge propagation models dominated by migration effects. CHAPTER II EXPERIMENTAL A. Preparation of Compounds 1. Inorganic Metal Complexes a. General Procedures. Potassium hexachloroiridate K2IrCl6 and potassium ferrocyanide trihydrate K4Fe(CN)6-3H20, and hexaammineruthenium chloride [Ru(NH3)6]Cl2 were obtained from commercial sources (Aldrich Chemical Company and Alfa) and were used without further purification. Ferrocene (Cp)2Fe (Aldrich) was purified by sublimation. Other reagents were purchased from Aldrich unless otherwise specified and were used as received. All inorganic complexes were characterized by UV-vis spectroscopy and cyclic voltammetry. b. Synthesis of Fe(bpy)3(SO4),Fe(phen)3(SO4),and Fe(vbpy)3(SO4). Syntheses which relied on procedures similar to those previously reported were employed.216 An aqueous solution of ferrous sulfate was slowly added to an ethanolic solution containing 3.5 equivalents of polypyridyl ligand [2,2‘-bipyridine (bpy), 1,10- phenanthroline (phen), or 4-viny1-4'-methyl-2,2'-bipyridine (vbpy)]. The resulting mixture was stirred for 2 h. The red product was precipitated by addition of acetone, filtered, and washed with benzene to remove excess ligand. Highly crystalline solids of Fe(bpy)3(SO4), Fe(vbpy)3(SO4), and Fe(phen)3(SO4) were recovered from ethanol/acetone and methanol/acetone solution mixtures respectively. c. Synthesis of (PPN)2Fe(bpy)(CN)4. The procedure described by Schilt217 was followed with slight modifications. Stoichiometric amounts of ferrous ammonium sulfate hexahydrate 41 42 and 2,2'-bipyridine were added to distilled water and the solution was heated to a temperature just below boiling. A 15 molar excess of KCN was added to the warmed solution, which was stirred for 20 min. After 2 h red crystals of Fe(bpy)2(CN)2 formed. The product was filtered and washed several times with cold water, dissolved in hot water containing a 100 fold excess of KCN, and heated on a steam bath for 12 h. The dark orange-brown product K2Fe(bpy)(CN)4 was precipitated by the addition of an excess of acetone, filtered, washed with cold water, and recrystallized from ethanol. The bis(triphenylphosphoranylidene)ammonium(PPN+) salt was precipitated by addition of (PPN)C1 to an aqueous solution of KzFe(bpy)(CN)4. d. Synthesis of Os(bpy)3(ClO4)2- The procedure of Dwyer and Hogarth218 was followed with several minor changes. Stoichiometric amounts of 0804 (Alpha) and FeC12'6HzO were added to a solution of concentrated HCl and the mixture was heated for 2 h on a steam bath. After the solution changed color to dark red, the green complex KZOsClé was precipitated by addition of excess KCI. The solution was cooled and the precipitate was collected by filtration, washed several times with absolute ethanol, and dried. The precipitate was dissolved in glycerol containing an equal molar amount of 2,2'-bipyridine, and this solution was heated to 240 °C for 1 h. The solution was doubled in volume with distilled water and an excess of NaClO4 was added to precipitate Os(bpy)3(ClO4)2. The product was filtered, washed with cold water, and recrystallized from acetonitrile. 43 e. Synthesis of Ru(bpy)3(ClO4)2. Metathesis of Ru(bpy)3(C1)2, with NaClO4 yielded the perchlorate salt, which was recrystallized from aqueous solution. f. Synthesis of (NBu4)2Mo6Cll4. Molybdenum dichloride (Cerac Inc.) was dissolved in hot 6 M HCl, and the resulting solution was filtered. Addition of NBu4Cl immediately caused the yellow solid (NBu4)2M06C114 to precipitate, which was subsequently collected, washed several times with water and ethanol, and recrystallized from spectral grade (Burdick & Jackson) dichloromethane by slow evaporation. g. Synthesis of (NBu4)2W6Brl4. The procedure of Dorman and McCarley219 was followed with slight modifications. In a drybox, 15.0 g of WBr5 (Alpha), 0.72 g of Al metal (fine turnings obtained by shaving a 99.999 % pure aluminum rod with a tungsten carbide tip), 13.0 g of AlBr3 (Alpha), and 7.50 g of NaBr were added to a quartz reaction tube. The neck of the tube was cleaned and capped with a rubber septum. The tube was removed from the drybox and connected to a high vacuum manifold, evacuated for several hours, (until the pressure was below 1x10‘4 torr) and then flame sealed under dynamic vacuum. The contents were thoroughly mixed and the quartz tube was wrapped in asbestos and inserted in a steel pipe, which was placed into a high temperature furnace. The reaction vessel was heated to 200 °C for 3 h with the steel pipe being rotated at 20 min intervals. The temperature was then raised to 450 0C over a 3 h period, held at 450 0C for 9 h, and finally raised to 550 0C where it was held for 30 h (with the steel pipe being rotated at 30 min intervals). The tube was allowed to cool to room temperature 44 and wrapped in several layers of paper towel. The tube was carefully broken with a blunt object (Caution: violent explosions have occurred). Most of the glass was removed and the black fused solid was ground into a fine powder and added to a hot 9 M HBr/EtOH 50:50 mixture. The black powder was recollected and washed several times with 95 % ethanol, and then added to a 50:50 solution of EtOH and 9 M HBr. This mixture was heated with stirring for 12 h. The black powder and ground glass were removed by filtration. Addition of NBu4Br to the yellow EtOH/HBr solution produced the yellow (NBu4)2WGBrl4, which was recrystallized from spectral grade acetonitrile. 2. Organic Polymers and Monomers a. General Procedures. All reagents, including methylviologen (MV2+), were purchased from Aldrich unless otherwise specified, and were used as received. The organic polymers and monomers were characterized by proton NMR spectroscopy and by cyclic voltammetry. Elemental analysis was performed by Galbraith Laboratories. 45 b. Synthesis of N-methylpolyvinylpyridinium hexafluorophosphate (PVPL-Me/PF6’), l. —(' CHZ-CH-h— \ IN+ PFG. CH3 1 Stoichiometric amounts of polyvinylpyridine (PVP) and methyl iodide were added to methanol. The mixture was heated under nitrogen and in the dark to 60 0C for 10 h. The iodide salt of (PVPL —-Me) precipitated from solution and was collected by filtration and washed several times with cold ethanol. The hexafluorophosphate salt was prepared by the addition of NaPF6 to an aqueous solution of (PVP—Me)I. The white precipitate was washed with water and recrystallized from acetonitrile containing an excess of NaPF6. c. Synthesis of {N,N'-bis[-3-(trimethoxysilyl) propyl]-4,4'-bipyridinium}diiodide (PQ1+/2I'),2. (Mealssucnzlsfn: H :8? -(CH2)35i(0Me)3 21' 2. The procedure of Bookbinder and Wrighton220 was followed with slight modifications. A 75 fold molar excess of 1-iodo-3- 46 trimethoxysilypropane (Petrach Company) was added to dry acetonitrile containing 4,4'-bipyridine. The mixture was refluxed for 24 h, cooled to room temperature, filtered, and washed several times with cold acetonitrile. The light orange precipitate, (PQ2+) was used without further purification. (1. Synthesis of 4-vinyl-4'-methyI-2,2'-bipyridine (vbpy). 3. CH3 CH=CH2 The method of Abruna, Breikss, and Collum221 was followed with the modification that the final product (vbpy) was further purified by low temperature sublimation. e. Synthesis of 4-vinyl-4'-methyl-N,N'-ethylene- 2,2'-bipyridinium (EVDQ2+),4. Cll3 Cil=cn2 C8}? + + PFg' PFg' 4 Vinyl-bipyridine (vbpy) was added to freshly distilled 1,2- dibromoethane and heated under nitrogen and in the dark to 60 °C for 8 h. A light green precipitate was collected and the filtrate was 47 placed back into the reaction flask. This procedure was repeated five times producing high yields of the precipitate each time except the last. Solids collected from successive precipitations were combined and washed several times with ether. The hexafluorophosphate salt of EVDQ2+ was obtained by the addition of ammonium hexafluorophosphate to an aqueous solution of the light green precipitate. The light blue hexafluorophosphate salt was collected, washed with water, and dried in vacuo. The monomer was characterized by elemental analysis. f. Synthesis of 4-vinyI-4'-methyl-N ,N '-1,3- propylene-2,2'-bipyridinium (PVDQ’t),5. CH3 CH=CH2 / \ —N N + T PF,‘ U PF,‘ 5 The preparation of the hexafluorophosphate salt of (PVDQ2+) followed the same procedure as that described for EVDQ2++ except that 1,3-diiodopropane was substituted as the alkylating agent. 48 g. Synthesis of 4-vinyI-4'-methyl-N ,N ' - 1 ,4- butylene-2,2'-bipyridinium (BVDQ2*),6. cu3 cn=cu2 N N— + + PF,“ K 2 PF,’ 6 The preparation of the hexafluorophosphate salt of (BVDQ2+) followed the same procedure as that for EVDQ2+r except that 1,4- diiodobutane was substituted as the alkylating agent. 3. Synthesis, Modification, and Purification of Layered Oxides a. Sodium Montmorillonite (Wyoming). Sodium montmorillonite Nam“)[A13.23Fe0.42Mgo.,,’7](Si.,_87Alo_13)020(OH)4 was obtained from the Clay Minerals Repository (University of Missouri, Columbia, MO). The mineral was purified by sedimentation to collect the < 0.2 um fraction, and subsequently washed with HSO4' to remove carbonate and with dithionite to remove free iron oxides. The charge exchange capacity (CEC) of the montmorillonite was 80 meq of charge/100 g of clay. The platelet size was < 200 nm. b. Synthetic Hectorite (Laponite®). Laponite® Nao.22Li0.14[M35.64Li0.36](Si8.00)020(0H)4 was purchased from Laporte Industries Ltd. and used as received. The CBC and platelet size of 0 this Laponite were 55 meq/100 g and < 50 nm, respectively. 49 c. Fluorohectorite. Fluorohectorite Li1.60[Mg4.40Li1.60] (Sii;.()())02(,(F)4 was a synthetic product similar to that previously described by Barrer222 and was obtained from Corning Glass Works. The CEC and platelet size of this fluorohectorite were 190 meq/100 g and > 1000 nm, respectively. d. Reduced Charge Montmorillonite. A 5 mL suspension of a 2 % slurry of sodium montmorillonite was sealed in dialysis tubing with 2 mL of 1 M LiCl. After 24 h, the dialysis tubing was suspended in freshly distilled water, and over the next 48 h the water was changed every 8 h. The lithium-exchanged montmorillonite was removed from the dialysis tubing and deposited onto graphite electrodes, evaporated, and the resulting films were heated to 250 °C for 2 h. e. Intercalates. The desired inorganic cation was added to a aqueous suspension (2 % by weight) of clay. The amount of cation added depended on the desired loading and was calculated by using the CEC of the smectite clay used. f. Intersalated Montmorillonite. A 2.0 % by weight aqueous montmorillonite suspension was sonicated (this pretreatment of the clay sample is the crucial step of the synthesis) and rapidly added to an aqueous solution containing 3-5 CEC of FeL3(SO4) (L=bpy, vbpy), followed by stirring for 2 h. The amount of Fe(bpy)32+ in excess of bound material was determined by centrifuging the intersalated sample and quantifying the concentration of Fe(bpy)32+ in the supernatant by UV-vis Spectroscopy. The molar absorptivity was determined to be 7040 M ‘1 cm'1 from a 10 point Beer-Lambert plot with a correlation 50 coefficient of 0.9987. The determination of the Fe(bpy)32+ concentration of the supernatant (5.1 mM) was in accordance with the value calculated by assuming a Fe(bpy)32+ loading of 2 CEC (4.8 mM). 4. Supporting Electrolyte Tetrabutylammonium hexafluorophosphate (TBAPFG, Southwestern Analytical Chemicals) and tetrabutylammonium tetrafluoroborate (TBABF4, Aldrich) were dissolved in ethyl acetate containing MgSO4, filtered, and then recrystallized by the addition of ether. These salts were dried in vacuo at 90 °C for 10 h. Tetrabutylammonium (TBAP, Southwestern Analytical Chemicals) and tetraethylammonium perchlorate (TEAP, Southwestern Analytical Chemicals) were recrystallized from water and dried in vacuo at 30 °C for 12 h. Metathesis of potassium hexafluoroarsenate (TBAAsF6, Ozark-Mahoning) with NBu4Br yielded the tetrabutylammonium salt which was collected and dried in vacuo at 60 °C for 12 h. Lithium perchlorate (LiClO4, Fisher) was recrystallized from acetonitrile and dried in vacuo at 100 °C for 6 h. Sodium acetate (Aldrich) was recrystallized from twice distilled glacial acetic acid. All other supporting electrolytes were reagent grade and used as received. 5. Solvents Acetonitrile purchased from Burdick & Jackson Laboratories (distilled in glass grade), was subjected to four freeze-pump-thaw (fpt) cycles and then vacuum distilled onto 3-A activated molecular 51 sieves, dried for 6 h, distilled onto CaHz, and finally vacuum distilled into a sealed flask. Distilled water was passed through a column of activated charcoal and two columns of mixed-bed ion exchange resin. B. Methods and Procedures 1. Electrochemical Measurements. Cyclic voltammetry, chonopotentiometry, and bulk electrolysis experiments were conducted by using a Princeton Applied Research (PAR) Model 173 potentiostat, 175 universal programmer, 179 digital coulometer, and a Houston Instruments Model 2000 X-Y chart recorder. Chronoamperometric experiments were performed with a Bioanalytical Systems (BAS) 100A multianalyzer. The nonaqueous experiments were performed in a Vacuum Atmospheres HE-453 double-sided glove box. Electrochemical experiments employed a conventional H-cell equipped with a working electrode (pyrolytic graphite, glassy carbon, platinum, or antimony doped tin oxide coated glass), counter-electrode (Pt gauze), and a reference electrode (aqueous/saturated calomel and acetonitrile/silver wire). Potentials determined in acetonitrile were converted to the SCE reference scale using the ferrocenium/ferrocene couple of +0.31 V vs. SCE as an internal standard. 2 . Preparation of Modified Electrodes Clay modified electrodes were prepared by droplet evaporation of the desired clay suspension on the electrode surface. Best results 52 were obtained by maintaining the concentration of the clay suspension between 1-2 weight percent. The Na+-exchanged and Fe(bpy)32+ pre-exchanged experiments employed 0.31 mg/cm2 of clay on the electrodes; this coverage corresponds to a film thickness of 1.75 pm. In the experiments conducted for the purpose of comparison to intersalated montmorillonite, the amount of clay deposited on the electrode was 0.07 mg/cm2 corresponding to a film thickness of 0.40 um. The incorporated electroactive cations were ion-exchanged into the clay films by either soaking the clay modified electrode in an aqueous solution containing electroactive ions or by pre-exchanging polypyridyl metal complexes into the clay samples (as described in Section II.A.3) before the films were cast. The metal polypyridyl pro-exchanged and intersalated clay samples cast uniform strongly adhering films in the range of 0.1-0.8 CEC and 3-5 CEC, respectively. Polyvinylpyridinium films were cast by droplet evaporation of a 1% (PVP--Me)PF6 acetonitrile solution on the electrode surface. Poly-PQ2+ films were prepared according to the procedures of Wrighton et 01.220 with a few modifications. The best results were obtained by using a 1 mM concentration of monomer. It was necessary, to apply a potential step of -0.76 V vs. SCE for 3-10 minutes in order to observe film growth. Poly-EVDQ2+ and poly-PVDQ2+ were prepared by the methods of Willman and Murray.122 However, the preparation of stable films were best accomplished in the glovebox. Poly-Fe(vbpy)3+2/clay films were prepared by cycling montmorillonite films (NM-exchanged and intersalated) in an 53 acetonitrile solution containing Fe(vbpy)3+2 and TEAP following 114 published procedures. The polymerization was only successful if the electrolyte was TEAP. Attempts to produce the polymer by using other electrolytes (TBAP, TBAPF6, and LiClO4) failed. Electroactive inorganic anions were ion-exchanged into the polymeric films by soaking the modified electrodes in dilute solutions containing the anion of interest. 3. Spectroscopic Characterization and Instrumentation Electronic spectra were recorded on a Varian Associates Cary Model-2300 UV-visible spectrometer. NMR spectra were recorded on a Bruker Model WH 250 MHz instrument. The chemical shifts were measured with respect to the solvent signal. X-ray diffraction studies, used to determine the d(001) basal spacings, employed either Philips X-ray or Siemens Crystalloflex-4 Diffractometers. Both instruments were equipped with Ni filtered Cuka radiation sources. The samples were prepared by evaporating a l % aqueous suspension of intersalated clay onto either a glass slide or thinly shaved planar graphite. Scanning electron micrographs were recorded on a JEOL JSM 35C SEM at the Center for Electron Optics at Michigan State University. The samples were prepared by allowing an aqueous clay suspension to air dry onto pyrolytic graphite. The clay] coated graphite was mounted, and sputter coated with a thin layer of gold. Measurements were made by Dr. Stan Flegler. Film thicknesses of modified electrodes were determined by cutting films cast on SnO2 electrodes, with a razor blade and 54 measuring the height differential between the film and glass support with a Sloan Dektak Surface Profilometer. The molecular weights of EVDQ2+, PVDQ2+, and BVDQ2+ were determined by Fast Atom Bombardment Mass Spectrometry (FABMS) using a JEOL RX 110 double focusing mass spectrometer, which is housed in National Institute of Health/ Michigan State University Mass Spectrometry Facility. Samples of EVDQ2+, PVDQ2+ and PVDQ2+ were dissolved in glycerol matrices. 4. Determination of Diffusion Coefficients Polymer modified electrodes have been shown to obey semi- infinite linear diffusion at short times < 50 msec.“9'223'224 Accordingly, diffusion coefficients of multilayer modified electrodes can be determined by chronoamperometry by using the Cottrell equation (equation 10). The diffusion coefficient for charge propagation, D, (cm2/s) is obtained by using the slope of the line from a plot of current vs. time. The values for the constants in equation 10 were easily obtained; 11 is the number of electrons transferred during the redox event, F (coul/mole) is Faraday's constant, and A (cmz) is the area of the electrode. This latter value was also determined by chronoamperometry by using ferrocyanide as a standard because of it's well documented diffusion coefficient.225 The value for the concentration of electroactive material C“ (moles/cm3) was obtained by converting C“ into the form 1" Id, where I‘ (moles/cmz) is the surface coverage of the modifying layer, and (1 (cm) is the film thickness. By combining constants into a new term K 55 and changing the form of concentration, the Cottrell equation can be rewritten as 172 i(t)t1/2 = slope = @Tr; (12) The surface coverage 1' of electroactive polymeric films was determined by integrating the area under cyclic voltametric wave for a one electron reduction. Alternatively I“ of electroactive inorganic complexes ion-exchanged into both clay and polymeric films were ascertained by measurement of the number of coulombs passed on a slow (< 10 mV/sec) cyclic voltammetric scan. Film thicknesses for polymers were determined using surface profilometery. Errors in the calculation of diffusion coefficients arise primarily from inability to measure the thickness of fully hydrated films, because these films lose solvent during thickness measurements and tear easily. Therefore, we were obligated to measure thicknesses of dry films. Wrighton and coworkers220 have determined the error in diffusion coefficients calculated using the Cottrell equation resulting from dry film thickness measurements is i: 20 %. Thicknesses of clay films were determined from density measurements. 56 C. Construction and Cleaning of Working Electrodes 1. Pyrolytic Graphite Electrodes a. Graphite Disc Electrodes. Cylinders (0.476 cm in diameter) were machined from a block of pyrolytic graphite (Union Carbide). The cylinder axis was colinear with the c-axis of the graphite. The cylinders were sealed with heat shrink tubing onto the end of glass tubes containing copper wire. Electrical contact between the copper wire and graphite was made with mercury, which was confined within the tubing by parafilm wax. For each experiment a fresh surface was obtained by cleaving a thin layer of the graphite with a razor blade. b. Planar Graphite Electrodes. A thin plane of graphite was cleaved with a razor blade from a 2x1x1 cm3 block of graphite along the basal surface. Electrical contact between the graphite and a copper wire was made with conducting silver epoxy (Tra Con Corp). 2. Platinum and Glassy Carbon Disk Electrodes Electrodes were constructed by sealing a platinum cylinder (1.6 mm diameter) or a glassy carbon cylinder (3.2 mm diameter) into a Teflon shroud purchased from BAS. The Pt electrode was polished with 1 um diamond paste and with 0.05 um alumina, both purchased from BAS. The electrode was thoroughly washed with water and methanol. This same procedure was used for glassy carbon electrodes with the exclusion of the diamond paste polishing. 57 3. Platinum Oxide Disc Electrodes The preparation of platinum oxide electrodes utilized the Pt disc electrode (BAS). The Pt electrodes were pretreated by initially polishing according to the described procedures in Section II.C.2. The electrode was then soaked in 1 M sulfuric acid for 10 h. The electrode was removed from solution and potentiostated in an aqueous 0.5 M sulfuric acid solution at +1.9 V vs. SCE for 10 min, cycled between -0.15 and +1.2 V vs. SCE for 2 h, anodized at +1.1 V vs. SCE for 60 sec, and finally washed thoroughly with water and air dried. 4. Preparation of Tin Oxide Electrodes Antimony doped tin oxide coated glass (80 ohms/square) was generously donated by Pittsburgh Plate and Glass (PPG). The larger sheets of glass were annealed and cut into rectangles 9 x 15 mmz. Ohmic contact between the semiconductor surface and a copper wire was made by using conducting silver epoxy, which was further coated with insulating epoxy resin (Dexter Corp.). CHAPTER III CLAY MODIFIED ELECTRODES A. Background The modification of electrode surfaces with inorganic crystalline supports provides improved microstructural stability, the potential for increased heterogeneous catalytic activity, and well- defined microstructures. Ghosh and Bard's226 observation of electrochemical activity from transition metal complexes incorporated within treated clay adsorbed onto a platinum electrode represented the first report of an inorganic modified electrode. Since this seminal paper, a wide variety of inorganic modified electrodes have been reported including clays,227'231hydrotalcite,232’233 235 236,237 238 sepolite,234 bentonite, porous aluminum oxide, 44 sol gels, zeolites,239'243 layered Zr-phosphates,2 245447 2 polyoxometalates, 9 tungsten oxide bronzes, 48 and transition metal cyanides.24 Clays are inorganic supports possessing good thermal and chemical stability. The acidic clay interlayers (often called the gallery), make clays potentially useful as size and shape selective 250-252 catalysts, and catalyst supports. These properties in conjunction with the ability of clays to behave as cation-exchangers have led to numerous studies on clay modified electrodes, some of which have demonstrated their utility in the areas of organic 253 254.255 256,257 synthesis, energy conversion, 258.259 electrocatalysis, photocatalysis, 26° and chromatography. The idealized structure of a smectite clay is shown in Figure 7. These 2:1 aluminosilicates have structures consisting of an octahedral layer sandwiched between two inverted tetrahedral layers. The tetrahedral layers consist of a oxygen framework with silicon 58 59 Figure 7 Idealized structure of smectite clay with a hydrated cation between layers. 60 O— Hydroxyl Groups .— Oxygen Atoms Figure 7 61 occupying the tetrahedral sites. The various smectites are differentiated from one another by the cations residing in the octahedral sites. As an example, montmorillonite, which has aluminum ions occupying 2/3 of the octahedral sites, acquires an overall negative charge on the clay layers by isomorphous substitution of Mg2+ for M“, which is Stoichiometric and uniform throughout the clay. The resulting negative charge is counterbalanced by hydrated cations located between the silicate layers. The charge equivalence is quantified by the cation exchange capacity (CEC) which is defined as the molar equivalence of charge per quantity of clay. Ghosh and Bard's226 initial work established that Ru(hpy)32+, Fe(bpy)32+, and Fe(Cp)2+ could be incorporated into clay/PVA films and that these ions could participate in electron transfer with the underlying platinum electrode. The scope of clay modified electrodes 228 was expanded with subsequent studies by Liu and Anson who reported that metal polypyridyl complexes, Ru(NH3)62+, and Ru(NH3)5py2+ could be ion-exchanged into free standing clay films on pyrolytic graphite electrodes. Interestingly, polypyridyl metal complexes and meyhylviologen (MV2+) bind in clays at a lOO-fold excess over the concentration of the contacting solution. Whereas the polypyridyl metal complexes require many hours to be extracted from clay films, 227,228 Ru(NH3)62+ is leached from films within minutes. In accordance with these observations the experimentally determined diffusion coefficient of Ru(NH3)62+ is a factor of 300 larger than that for polypyridyl metal complexes within montmorillonite films.261 62 These results have been interpreted as evidence of the importance of hydrophobic interactions between the clay layers and electrostatically exchanged organophilic cations in the binding process. Further insight into the origins of electroactivity of clay films is provided by coulometric experiments. Quantitative studies show that only a fraction (15-30 %) of the total complex confined in clay films is electrochemically active.227'228 Moreover, electrodes displaying activity will lose electroactive ions from the film when soaked in an aqueous solution containing only electrolyte, yet the film retains the color of the inorganic ion. The degree of electroactivity in general correlates with the external surface area of the clay, thereby suggesting that externally adsorbed complex is the primary contributor to charge transport in clay-modified electrodes.261 Moreover, Fe(CN)63' and other negative ions (C2042', Mo(CN)84', I') readily penetrate clay films and their diffusion is strongly dependent on the electrolyte concentration and pH of the solution.227'23m"231 Because these ions should be repelled by the permanent negative charge of the clay layers, the presence of electroactive channels between the clay particles has been proposed?” The above results have lead to the general model for electroactive clay films shown in Figure 8. Ions reside in clay galleries, micropores, and on the clay particle surfaces with most of the ions exchanged in the clay film being electrochemically silent. Organophilic cations, bound tightly within the clay film, are for the most part electrochemically inactive due to their immobility while 63 Figure 8 Model of a clay film on an electrode containing ion-exchanged polypyridyl metal complexes and electrolyte. ////1/E/'€?t}°}e/ /////// 64 0+ 0+ 0+ 0+ ML02+ X‘ X— 9 O MLn2+ 0 2+ X" MLn Q X" 0+ / 3 2+ 2* X MLn NW6 Jr MLn2+ X- 47 C x‘ {)9 o 0 0+ 0 2+ + MLn MLn2 CLAY FILM Figure 8 SOLUTION 65 purely inorganic ions are very mobile, with the degree of electrochemical activity correlating to the external surface area of the clay. Because ions reach the electrode surface by diffusing through channels around clay particles, the swelling of the clay plays an important role in the observed electrochemical activity. Although the qualitative aspects of the electroactivity in clay films have begun to emerge, the quantitative nature of the electroactive sites and charge transport mechanisms have yet to be determined. Ion-exchanged electroactive ions M+ have potentially three distinct locations within the film (Figure 8): (i) voids or channels around the clay platelets; (ii) outer edges of the clay layers; and (iii) the galleries between clay layers. Because one objective of clay studies is to derive electrocatalytic activity from ions between clay interlayers, the elucidation of the electroactive sites within the clay film is of paramount importance, particularly in lieu of the initial studies on clay-modified electrodes which have indicated that ions rigorously bound between clay layers might be electrochemically silent. Conventional electrochemical approaches utilized to date have not provided sufficient insight into many of these crucial issues of clay modified electroactivity. Therefore, unique studies designed to probe the clay's microstructure and its effects on the observed electrochemical activity of bound redox ions were undertaken. With the information garnered from these studies, clay-modified electrodes can begin to be systematically designed. 66 B. Results and Discussion 1. Nature of the Electroactive Sites. All previously reported studies on clay modified electrodes utilized the Na+-exchanged form of the clay to cast films on conducting metal oxide, metal, or carbon surfaces. Because Na+- smectites are totally delaminated in aqueous suspension the formation of continuous, crack-free films capable of imbibing electroactive species is facilitated. For instance, a graphite electrode coated with Na+-exchanged montmorillonite (0.31 mg/cmz), immersed in solution of M(bpy)32+ (M = Fe, Ru, 08), produces cyclic voltammograms exhibiting well-defined M(bpy)33+/2+ waves. As previously observed,227'228 the intensity of the wave increases monotonically with electrode immersion time, and the functional dependence of the peak current on scan rate is characteristic of a diffusional process. Figure 9 shows the waves obtained at 10 min intervals for a Na+-exchanged montmorillonite film in a 0.2 mM Os(bpy)32+ solution containing 0.1 M Na2804 when the electrode potential is continuously cycled between 0.00 and +0.90 V vs. SCE. After soaking the electrode for 2 h, during which the electrode becomes golden upon incorporation of Os(bpy)32"' in the film, the voltammetric response reaches a plateau. Direct transfer of the electrode to solutions containing pure electrolyte causes the peak currents to be attenuated only slightly. With continued soaking, however, the electrode response decreases, indicating that the electroactive dipositive cation is displaced from the film. Significantly, although the current response is virtually eliminated 67 Figure 9 Increasing cyclic voltammetric response from Os(bpy)32+ incoporating into a montmorillonite film coated on graphite in an aqueous solution containing 0.2 M sodium acetate and 0.2 mM os(hpy)32+. l \ llllllJllli 0 0.2 0.4 0.6 0.8 1 .0 ENolts vs. SCE Figure 9 69 after soaking the electrode for 24 h in pure electrolyte solution, the film retains the deep golden color of Os(bpy)32+ . As mentioned above these results indicate that only a fraction of the total ML32+ cations that are present in the clay film are electroactive, a fact recognized by all earlier workers. However, the extent to which ML32+ ions at the exchange sites of the clay contribute to the voltammetric response heretofore has remained an unresolved question of fundamental significance. One simple yet incisive approach to this issue is to examine the behavior of films formed from pre-exchanged Os(bpy)32+-montmorillonite. Accordingly, montmorillonite electrode films in which 25, 50, 80 and 100 % of the Na+ ions had been replaced by Os(bpy)32"’ were prepared. In this way, Os(bpy)32+ is present on exchange sites both within the intracrystalline galleries and on the external surfaces. No evidence for an Os(bpy)33+/2+ wave is observed for any of these electrodes when immersed in solutions containing only pure electrolyte. Figure 10a shows the electrochemical response of an electrode coated with 80 percent Os(bpy)32+-exchanged montmorillonite after 2 h of continually scanning the electrode potential at 50 mV/sec between 0.00 and +0.90 V vs. SCE. An Os(bpy)_,,2+ wave is not detected even at high current sensitivities. Electrodes coated with pre-exchanged Ru(bpy)32+-, Fe(bpy)32"'-, Fe(phen)32+-, and MV2+-montmorillonite films similarly failed to yield voltammetric responses. It is clear from these results that ML?’2+ and MV2+ cations electrostatically bound in the galleries and external surfaces of smectite clays are rigorously electroinactive. 70 Figure 10 Cyclic votammograms of a 80 % pre-exchanged montmorillonite film on pyrolytic graphite: (a) soaking in an aqueous solution containing 0.1 M NaZSO4; and (b) electrode from above placed in an aqueous solution containing 0.1 M Na2804 and 0.2 mM 0s(bpy)32+. 71 MNODIC .1, A \ 1111111111 O 0.2 0.4 0.6 0.8 EIVolts vs. SCE Figure 10 72 Pro-exchanged ML32+-montmorillonite films are similar to Na+- montmorillonite in their ability to incorporate redox active cations. Dipping a 80 % Os(bpy)32+-exchanged montmorillonite coated electrode into 0.2 mM solutions of Os(bpy)32+ yields the appropriate cyclic voltammetric wave. As in the case of Na+-exchanged films, the current response is attenuated when the electrode is transferred to pure electrolyte solution. Figure 10b shows the limiting cyclic voltammetric response of the film in pure electrolyte solution. The cyclic voltammogram has the same shape, potential characteristics, and limiting current response obtained with Na+-exchanged montmorillonite films; however, the peak current of the Os(bpy)32+- exchanged film grows at a rate five times as fast as that obtained for Na+-exchanged montmorillonite films. Thus, while ML32+ and MV2+ ions within clay galleries do not communicate with the electrode surface, dipping the pre-exchanged ML32+-clay electrode in solutions of electroactive ions does promote electrochemical activity. Although ML32+ cations at exchange sites do not communicate with the electrode surface, one might expect freely diffusing ions incorporated in clay films to transfer electrons to the electrostatically bound surface cations. Like the function of Ru(NH3)62+ as a charge transport ion between the electrode and immobile cobalt 62 mobile tetraphenylporphyrin in NafionQ-coated polymer films,2 electroactive ions are potentially capable of shuttling charge between ions incorporated in the clay gallery and the electrode surface. In this context, the large self-exchange rate constants,263 appropriate half-wave potentials of the 3+/2+ couples, and structural similarities of the osmium and iron bipyridyl complexes establish Fe(bpy)32+ as a 73 convenient probe of oxidation-reduction processes in Os(bpy)32+- exchanged clay films. The charge shuttle scheme referred to here is graphically shown in Figure 11. A graphite electrode coated with Os(bpy)32+-exchanged montmorillonite (80 % of the charge exchange capacity) was placed into a solution of Fe(bpy)32*, rinsed once in water, placed in a cell containing 0.05 M NaZSO4 as the electrolyte, and scanned between 0 and +1.10 V vs. SCE. As shown in Figure 12, the initial cyclic voltammogram of the electrode gave a well-defined, of value was independent of scan rates . 2+ . . . . revers1ble Fe(bpy)3 wave w1th a peak current ratio, lp’a/lp’c unity. The initial ip a/ between 2.0 and 200 mV/sec. This result is not consistent with a 1p,c charge shuttle mechanism. If charge were being shuttled between the clay gallery and electrode, then Fe(bpy)33+ formed by oxidation at the electrode would be reduced by Os(bpy)32+ at the clay interface. Therefore, cyclic voltammograms of Fe(bpy)33+/2+ would exhibit behavior characteristic of a catalytic EC' mechanism (i.e.,i > 1m ip 0)?“ Under no conditions, even when the Fe(bpy)32+ solution concentration is increased to 0.05 M, is current flow attributable to Os(bpy)32+ oxidation observed. The incorporation ratei obtained for pre-exchanged clay is five times faster than for the Na+-exchanged form. This indicates that the method of preparation of the clay affects it's electrochemical properties. Because the gallery ions are electrochemically inert, the only significant difference between these samples is that the pre- exchanged clay is swollen to a greater degree than the Na+- exchanged form. The effect swelling has on the incorporation rate of Os(bpy)32+ into various forms of montmorillonite was determined 74 Figure 11 A reaction scheme devised to oxidize Os(bpy)32+ located in the 2+ galleries, with utilization of Fe(bpy)3 as a redox charge shuttle agent. 75 2 23E >86 n N. m. >30 n NM [0 i +N0A>Qflv®l +8 Eeoo $3.388 +mm§o£ou /////9901138l3////// 76 Figure 12 Reversible cyclic votammogram upon the first oxidation scan of Fe(bpy)32+ ion-exchanged into a montmorillonite film containing pre-exchanged Os(bpy)32+ while soaking in an aqueous solution containing 0.1 M NaZSO4. 77 IIOpA l O 0.2 0.4 0.6 0.8 'E/Volls vs. SCE Figure 12 Lo 78 with the preparation of five montmorillonite films: a) 78% pre- exchanged Os(bpy)32+, b) 38% pre-exchanged Os(bpy)32+, c) Na+- exchanged placed in 100% humidity for 24 h, d) Na+-exchanged air dried for 4 h and e) reduced charge (collapsed) films. As previously mentioned, the 78% pro-exchanged film should swell the most, followed by the 38% pre-exchanged film, and then the humidified and air dried films. The collapsed clay film should not swell. The current response obtained at various times for the five clay samples while soaking in an aqueous solution containing 0.2 mM Os(bpy)32+ and 0.2 M Na+-acetate is shown in Figure 13. The incorporation rates parallel the degree of swelling for these films. The films maximum loading and time required to attain them are listed in Table 6. These results, which demonstrate that the preparation and resulting swelling of clay films have a pronounced effect on the incorporation rates of ML32+ metal complexes are further supported by 261,265 investigations with Al, Zr, Fe, and Si pillared clays. Increased 2+ diffusion rates for Fe(bpy)3 in pillared clay films over Na“- exchanged clay have been reported. In addition pillared clays incorporate polypyridyl cations from nonaqueous solutions whereas Na+-exchanged clay films do not. Presumably the large fixed d- spacing of pillared clays resulting from the molecular props in their clay galleries causes the films to behave as if they were fully swollen.265 6‘ have As previously mentioned Bard and coworkers2 suggested that the electrochemical response obtained for anionic complexes arises from the presence of microchannels which are filled by solution containing the redox active anion. Such microchannels 79 Figure 13 Current responses obtained from Os(bpy)32+ exchanged into montmorillonite films from an aqueous solution containing 0.2 sodium acetate and 0.2 mM Os(bpy)32+: (a) 78 % pre-exchanged Os(bpy)32+; (b) 38 % pre-exchanged Os(bpy)32+;(c) Na+- exchanged (humidified);(d) Na+-exchanged (air dried); and (e) reduced charge. 80 12.0“ . I a . b 10.0- c I cl 0 < 8.0 3. ‘5“ 0 O 3,": 6.0“ ' I = U ° I 4.0" I O I 2.0- , ° " I O O . . I . O I . e 0.0%— . r . T . l I 0.0 15.0 30.0 45.0 60.0 75.0 90.0 Time, min Figure 13 81 Table 6 Time Required to Reach Maximum Current Response for Different Films Soaked in 0.2 mM Os(bpy)32+ S a m p l e Montmorillonite Film t(im a x )/min im 1. x In A a 80% pre-exchanged 4 2 1 1.9 with 0s(bpy)32+ b 38% pre-exchanged 9 0 1 1.4 with 0s(hpy)32+ c Na+-exchanged 1 3 0 1 3 .2 humidified d Na+-exchanged 3 60 10.6 air dried 4 h e reduced charge 3 00 0.60 (collapsed) 82 would also contribute to the activity observed for polypyridyl complexes and methylviologen cations. To obtain more direct evidence for microchannels, we have investigated the texture of montmorillonite films deposited on graphite by scanning electron microscopy. The low-magnification SEM image shown in Figure 14a reveals the characteristic waffle-like surface formed by the deposition of clay into interconnected domains of ridges and valleys. At higher magnification (Figure 14b) one can detect grain boundaries formed by the imperfect stacking of tactoids with a rag-like texture. These grain boundaries may represent the channels where redox active ions in solution can invade the film. Despite the above evidence for microchannels in clay films, the simple solution-filling of such channels cannot be the sole mechanism responsible for the electroactivity of clay films. It has been shown that the effective concentration of electroactive ML32+ cations in clay films (10'1 M) is substantially higher than the concentration of cations in the soaking solution (10'3 M). This suggests that the cations are being concentrated in the film by interactions with the clay surface. Also, Yamagishi and Aramata have demonstrated enantioselectivity for oxidation of racemic Co(phen)32+ at a SnO2 glass electrode coated with A-Ru(phen)32"’-montmorillonite.96'253 This latter result also establishes that the redox active polypyridyl complex communicates with the clay surface. Yet, we must keep in mind that ions bound to external surfaces, are rigorously electroinactive. How can these facts be reconciled? It is proposed that the electroactive ML32+ and MV2+ cations in Clay modified electrodes in part are bound as ion pairs to the 83 Figure 14 SEM images of Na+-exchanged montmorillonite films on pyrolytic graphite, (a) 1000 x and (b) 7800 x. 84 Figure 14 85 surfaces of the clay particles which border the microchannels. Although anions should normally be repelled by the surfaces of smectite clays, polypyridyl complexes are known to be exceptionally effective in promoting anion binding through ion pair formation.2“‘268 The driving force for ion pairing is facilitated by strong physical adsorption of the polypyridyl complexes to the organophilic surface and by the ability of the large organocation to shield the accompanying anion from the negative charge of the clay. 256 Methylviologen cations and related organic dye cations which have also been shown to be electroactive in clay films,258 share many of the same properties of polypyridyl complexes in promoting ion pair formation. The distinction between inactive, electrostatically bound complexes at exchange sites and active complexes bound by ion pairing is that in the former case the counter anion is the clay itself, whereas in the latter case the counterions are the anions of the electrolyte. Thus the electroactive cations are those which bind to the clay surface in excess of the cation exchange capacity by an ion pairing mechanism. Support for ion pairing is provided by the fact that peak currents {are dependent on the anion of the supporting electrolyte. For instance, the limiting peak current for a Os(bpy)33+/2+ wave with Na2804 as electrolyte is 1.4 to 3.0 times as large as the currents observed with NaC2H3O2 as the electrolyte over the ionic strength range 0.05-0.30 M. Related electrolyte effects have been noted by Bard and his co-workers.2‘51'265 In order to determine the relative importance of ion pair formation on basal surfaces and edge surfaces, the cyclic 86 voltammetric properties of clay films with different morphological properties have been investigated. Specifically, we have examined laponite and fluorohectorite, synthetic clays with particle sizes larger and smaller than that of montmorillonite. The clay platelet sizes decrease along the series fluorohectorite (> 1000 nm), montmorillonite (< 200nm), and laponite (< 50 nm). Decreasing the platelet size should increase the edge surface area greatly without altering the total basal surface area within the oriented film. Thus, the edge surface area of laponite should be more than twenty times as large as the edge surface area of fluorohectorite. As Figure 15 illustrates, the incorporation rate and limiting current response for Os(bpy)32+ in Na+-smectite films is correlated with clay particle size. The limiting currents after 24 h are 24.3 11A for laponite, 12.6 11A for montmorillonite, and 4.6 11A for fluorohectorite. To make certain that these differences were not associated with the diffusion of ions through these films, the apparent diffusion coefficients (D ) for the 31’? three films using slow scan cyclic voltammetry were determined by methods described by Anson et 01.228 The Dapp values obtained, as expected were very similar for all three clay films, (5.5 x 10’“ cm2/sec for laponite, 3.3 x 10’11cm2/sec for montmorillonite, and 2.8 x 10'11cm2/sec for fluorohectorite) and were also in good agreement with the D of Os(bpy)32+ (3.5 x 10'11 cm2/sec) in a aPP montmorillonite film.228 The same relative results are obtained for clays pre-exchanged with Os(bpy)32+. The results thus far establish that ML32+ and MV2+ cations electrostatically bound to the exchange sites of smectite clay do not 87 Figure 15 The current response obtained as a function of time from various clay films adsorbed onto graphite soaking in 0.2 M sodium acetate and 0.2 mM Os(bpy)32+: (a) laponite; (b) montmorillonite; and (c) fluorohectorite. 88 20- IS?- _ 0 <1 \ ezmmmso 60 90 V min 30 Figure 15 89 undergo electron transfer with the electrode surface or with freely diffusing cations contained in the microchannels. This is a result of an unfavorable exchange equilibrium with counterions required to maintain charge neutrality after electron transfer has occurred. In order to oxidize metal polypyridyl complexes electrostatically bound to clays one of two processes must accompany the electron transfer step: ejection of a metal complex from the clay or uptake of a negative counterion. Polypyridyl complex ions bind very strongly to montmorillonite, as reaction 13 is known to proceed essentially to completion. (2 Nat) =1» (1413+) + [ML32+] + [2 Na+]soln (13) clay soln clay Thus, the desorption of ME,“ from montmorillonite upon electrolytic oxidation of ML32+ to ML33+, as expressed in equations 14 and 15, is unfavorable and electron transfer is impeded. (2 ML32+)clay'."[Iqa‘hlsoln -e (ML33+)clay+(Na+)clay+[ML32+]soln (l4) 2 -26‘ . 3 2 (3 ML3 +)clay (2 MLB +)clay + [ML3 +]soln (15) The alternative mechanism, counterion uptake is thermodynamically a highly unfavorable process. A large driving force would be required to overcome the energy required for the uptake of a negative ion into negative clay layers. Because electron-self CXchange between ML32+ and ML33+ has a driving force 90 approximately equal to zero, (even the cross reaction between Os(bpy)32+ and Fe(bpy)33+ only has a driving force ~ 0.2 eV) the system does not possess enough energy to overcome this barrier. Thus, of the three possible locations for ML32+ ions within the clay microstructure (Figure 8), only those ions freely diffusing in voids around clay particles are electrochemically active while electrostatically bound ions are electrochemically inert, presumably due to unfavorable thermodynamics required to maintain charge neutrality during the redox event. 2. Elucidation of Clay Microstructural Control of Electroactivity. Having established that electrochemical responses arise from excess complex around clay platelets, important issues include: are the electrochemically active ions equally distributed throughout the clay film?; and which of these ions have electrochemical access to the electrode? The first issue of interest is the determination of the percentage of material outside the clay platelets that is electrochemically active. Shown in Table 7 is the number of coulombs (after subtracting background coulombs) for the oxidation of Fe(bpy)32+ measured by slow scan cyclic voltammetry for various montmorillonite film thicknesses on graphite electrodes. The samples were prepared by soaking the clay modified electrodes in several different concentrations of Fe(bpy)32+ until equilibrium was established, with subsequent transfer to an aqueous solution containing 0.1 M NaZSO4. It is important to maintain a common 91 Table 7 Amount of Electroactivity From Different Clay Films of Varying Thickness After Soaking in Aqueous Fe(bpy)32* Solutions Containing 0.1 M Na2SO4 Montmorillonite Thickness Concentration Coulombs Film (um) Fe(L)32+ (tnM)‘I x 10“"-c Na+-exchanged 1.27 0.2 0.93‘I 0.76 0.2 l.01° 0.76 2.0 1.35 0.38 2.0 1.18 1.27 5.0 3.31 0.38 5.0 3.42f 1.27 10.0 3.61 Pre-exchanged 80% of 1.27 0.2 0.973 CEC with Fe(hpy)32+ 1.27 5.0 3.21 0.76 2.0 1.56h 0.38 5.0 3.11h 'L = bipyridine. I’The number of coulombs under reduction wave after subtracting background. cStandard deviation for three separate identical experiments: “3: 0.08, er 0.09, '3: 0.26, 8r 0.11. llThe clay was pre-exchanged with 88 % of the CEC. 92 electrolyte with identical concentrations for each system studied, because as mentioned in Section III.B.1, the anion of the electrolyte plays an intimate role in the ion-exchange process. The amount of electrochemically active Fe(bpy)32+ incorporated into montmorillonite at a given concentration of Fe(bpy)32+ is independent of film thickness, within the error of the experiment. The amount of electroactive Fe(bpy)32+ in these films monotonically increases with the concentration of Fe(bpy)32+ in the soaking solution and reaches a maximum value at as 5 mM. Importantly the clay films which were prepared by pro-exchanging 80 % and 88 % of the CEC with Fe(bpy)32+ behave similarly to the Na+-exchanged form. These results are consistent with those listed in Table 6 and illustrated in Figure 13. The fact that the observed electroactivity does not increase with thicker films suggests that only the Fe(bpy)32+ near the electrode is electrochemically active. This conclusion is further supported by electroactivity / morphology relationships of the clay microstructure on the electrode. The casting of montmorillonite films from aqueous suspensions onto flat substrates results in 90 % of the clay platelets being oriented parallel to the surface, where the remaining 10 % make up so-called defect zones.269 In the defect zone clay particles are jumbled and owing to the random stacking of clay platelets, cavities and voids are more likely to form. In conjunction with our previous results, this implies that the observed electroactivity should increase with the creation of more defect zones in the clay microstructure. One approach to creating defect sites is to impart disorder in the microstructure by using the electrode surface 93 as a template. Increased roughness of the electrode surface should enhance the production of additional defect zones by inducing edge to face stacking of the clay platelets. This would be most pronounced near the electrode surface, while normal stacking of the clay platelets occurs with increased distance from the electrode. Within this framework the electrochemical responses of several electrode microstructures can be understood. Pyrolytic graphite electrodes which possess a visibly rough surface, with rocking curves of 6.41°, incorporate a larger quantity of electroactive Fe(bpy)32+ than smooth surface electrodes. For example 0.3 pm montmorillonite films on Sn02,227 and Pt231 display current densities of 30 uA/cm2 and 60 llA/cm2 respectively, whereas the same films on pyrolytic graphite have current densities in the range of 100-120 uA/cmz. In addition the incorporation rates for ML32+ into clay films on Pt, Sn02, and glassy carbon are very dependent on the thickness of the clay film 2+ and incorporate ML3 ions at a rate between 3-7 times slower (depending on the film thickness) than for identical films on pyrolytic graphite. The common practices of rapidly heating clay films on smooth electrodes and the use of polyvinylalcohol (PVA) to increase the measured electrochemical activity consistent with increasing disorder of the clay microstructure. The former procedure prevents formation of parallel aligned platelets, thereby increasing the amount of edge to face stacking in the clay microstructure270 and the latter procedure prevents platelet alignment by requiring the platelets to conform to the amorphous polymer structure. 94 These results permit the model for clay electroactivity to be modified as shown in Figure 16. This model, which is consistent with all of our results to date, separates the clay film into two regions: one which electrochemically communicates with the electrode surface and one that does not. As discussed above, the electroactive ML32+ ions are located in defect zones (previously referred to as channels or voids) throughout the clay film. But only those ions in Region I with access to the electrode surface either by direct contact or charge shuttling through open pathways display electroactivity. At further distances away from the electrode surface the amount of parallel stacking of clay platelets increases resulting in the more ordered microstructures represented by Region II. This increased order reduces the basal volume and eliminates electrochemical pathways to the electrode surface thereby resulting in electrochemically isolated ML32+ ions. Substrates with smooth surfaces have a smaller quantity of the defect zones, at least near the electrode surface, and thus Region I for these electrodes will not extend as far away from the electrode surface as for pyrolytic graphite. In the context of these results, one final note of interest here bears on the common practice in the literature of reporting the amount of electroactivity as a function of the CEC for the clay sample. Because the amount of ML32+ incorporated into clay does not depend directly on film thickness and at some distance from the electrode ions are electroinactive, normalizing the electroactivity observed from exchanged redox ions to the quantity of clay on the electrode is problematic. On the basis of this work, it is advised that activities 95 Figure 16 Model depicting locations of electroactive and non-electroactive polypyridyl metal complexes within a clay film. 2// ma/ 97 should be reported as charge/cm2 where the charge is measured from electrolysis experiments. In our studies we report only the charge because our electrode area remains constant 0.178 cmz. In light of the complexities involved with reporting consistent data from clay modified electrodes, the necessity to perform control experiments is of paramount importance. 3. Electrocatalytic Activity of Clay/Polymer Films The model described in Figure 16 shows the electroactive ML32+ ions incorporated into clay films to be confined to a small volume in the microstructure. The isolated ions in Region II should become electroactivated if a pathway to the electrode is established. This might be accomplished by the formation of a conductive polymer network throughout the clay film. In this manner, electron/hole transport from the electrode surface to Region 11 may be achieved. The preparation of conductive polymers in clay films by electrochemical methods has been established with the formation of dimers of methylviologen (MV2+)271 and later tetrathiafulvalenium (TTFl’)272 in montmorillonite. Heterogeneous 273,274 and electrochemical initiated polymerization of pyrrole aniline275 has since been achieved. Our effort of electrochemically accessing the isolated ions in Region II with conductive polymers was initiated with the reductive polymerization of transition metal complexes containing vinyl bearing ligands, which have been ion-exchanged into the clay. This method of polymerization was first described by Meyer, Murray and 114 coworkers. The polymerization proceeds via a radical-radical 98 coupling process involving pairs of vinyl groups.276'277 The incorporation of Fe(vbpy)32+ (vbpy = 4-methyl-4'-vinyl-2,2'- bipyridine) into montmorillonite films of Na+-exchanged and 80 % pro-exchanged Fe(vbpy)32"’ from an acetonitrile solution containing 2 mM Fe(vbpy)32+ was achieved with limited success. Upon transfer of these electrodes to acetonitrile solutions containing only electrolyte the incorporated complex was completely leached from the film on the first cycle. However, continuously cycling the potential of montmorillonite modified electrodes between -1.70 and +1.35 V vs. SCE, while soaking in an acetonitrile solution containing 0.2 M tetraethylammonium perchlorate (TEAP) and 0.5 mM Fe(vbpy)3(ClO4)2, promotes the incorporation of metal complex as shown in Figure 17. These data were recorded at 10 min intervals for a duration of l h. Film growth continued for several hours thereafter providing oxygen was rigorously excluded from solutions. The electrode was washed extensively with acetonitrile and transferred to an acetonitrile solution containing 0.2 M TEAP. Figure 18a shows the CV for an electrode whose potential was cycled sequentially from 0.00 to -1.85 V, to +1.45 V and back to 0.00 V vs. SCE. The anodic wave at +0.97 V vs. SCE. arises from the 2+/3+ couple of the metal center, whereas the two cathodic waves at -1.46 and -l.58 V originate from the sequential one-electron reduction of the vinylbipyridine ligands. The ip’a/ip’c = 1.25 for the oxidation wave is greater than the expected value of 1.00. Conversely the opposite behavior is observed for the cathodic waves which display /' < 1.00. As shown in Figure 18, ip'a/i 113.0 ratios W111 be unlty ‘lm p.c upon consecutively cycling (two times) the potential between 0.00 99 Figure 17 Growth of poly-Fe(vbpy)32+ at 10 min intervals in montmorillonite adsorbed onto pyrolytic graphite while soaking in an acetonitrile solution containing 0.2 M TEAP and 2.0 mM Fe(vbpy)32+. 100 S 053...— HOm .m.» m=o>\m . .e 9... oz... .3. N _ m. . . 101 Figure 18 Cyclic voltammetric waves obtained from a poly- Fe(vbpy)32+/montmorillonite electrode in acetonitrile containing 0.2 M TEAP: (a) upon scanning the electrodes potential from 0.00 V to +1.20 V, to -1.80 V, and back to 0.00 V vs. SCE; (b) after the reduction waves had been scanned once prior to measurement; and (c) after the oxidation wave had been cycled once prior to measurement. 102 N. I. u m... 2 use... mUm .u.» n=e>\m 90° “6°. ‘ . 103 and +1.45 V. Similarly ip clip a = 1.00 for both reduction waves (Figure 18c) after cycling the potential two times between 0.00 and -1.85 V. In an effort to elucidate the origins for i on the first 13.8 I inc scan, an identical experiment was conducted at a bare graphite electrode. The potential sequence was 0.00 to -1.80 V, to +1.20 V, and back to 0.00 V vs. SCE. Figure 19a shows a prewave spike before the oxidation wave and the first reduction wave. The prewaves are eliminated (Figure 19b and 19c) by performing the parallel experiments used to obtain the CV's shown in Figure 18b and 18c. These prewaves are clearly stable and associated with one another; once one spike is cycled it does not reappear until after the other is activated. Thus these waves represent the anodic and cathodic branches of the same redox couple. Although the origin of these prewaves obtained at the bare electrode has not unequivocally been established possible explanations have been presented. Observed prewaves at bare electrodes have been attributed to isolated sites of damaged polymer, which are either to dilute or too immobile to undergo direct electron transfer with the electrode.176 Recently Guarr and Anson have proposed prewaves to arise from the partial reduction of ligands which have been protonated causing the metal complex to have a significantly more negative III/II redox couple.145 Although the clay/polymer composite modified electrode is 81m11ar to the polymer at the bare electrode With regards to 1WI at 1p.c’ the absence of the prewave spikes for the former suggests the clay microenvironment is altering the polymerization process. Further support for clay mediated polymerization comes from observations 104 Figure 19 Cyclic voltammogram of poly-Fe(vbpy)32+ adsorbed onto pyrolytic graphite while soaking in an acetonitrile solution containing 0.2 M TEAP: (8) upon scanning the electrodes potential from 0.00 V to +1.20 V, to -1.80 V, and back to 0.00 V vs. SCE; (b) after the reduction waves had been scanned once prior to measurement; and (c) after the oxidation wave had been cycled once prior to measurement. 105 NO 1. n m... 3 2:2... wow .a> 832m e0: “0°. b d. a 106 that the growth rate of the polymer at the bare graphite electrode is 2.5-3.0 times faster in acetonitrile solutions containing equal concentrations of Fe(vbpy)32+ than that for Na+-exchanged modified electrode. Because the radical initiated polymerization process requires Fe(vbpy)32+ from solution to approach and react with reduced polymer already formed on the electrode, unfavorable electrostatic interactions between the positively charged diffusing ion and the positively charged polymer must be overcome for polymerization to occur. The negatively charged clay layers may electrostatically facilitate this process. The nature of the defect site for clay mediated polymerization will be different than that of a bare electrode. It is important to note that the inability to observe prewave spikes from clay/polymer microstructures does not preclude their existence. Indeed, the fact that ipa at ip c for the clay/polymer composite suggests that isolated defect sites are present in this system. The prewaves arising from these defect sites may be shifted in potential such that they fall under the waves of the bulk polymer. Clay/polymer films are very stable to oxidation in acetonitrile solutions containing electrolyte, but rapid loss of activity occurs if the films are reduced over several cycles. For example the film whose CV is shown in Figure 18 loses only 6 % of its activity, upon cycling the electrode potential past the oxidation wave continuously for 2 h and films soaked (48 h) in acetonitrile solutions containing 0.2 M TEAP, which are not cycled, demonstrate no measurable loss of activity. Conversely, activity is lost within a few cycles over the reduction waves of the polymer. The cause for the degradation of 107 these films upon reduction in fresh acetonitrile solutions is not clear. These acetonitrile solutions are not dry and water, which is required to assist in swelling of the films, can be reduced at the high negative potentials needed for the reduction of the Fe(vbpy)32+ polymer. It is possible that the reduced polymer is attacked by OH' produced from the reduction of H20. Another interesting result is that the oxidation wave stops growing after about 3 h of cycling the potential of the clay modified electrode in 0.5 mM Fe(vbpy)32+. However, the reduction waves continue to grow and shift to more negative potentials. This result implies the presence of a chemically induced polymerization initiated by cathodic currents. The current response for the clay/poly-Fe(vbpy)32+ film indicates thin cell behavior. Figure 20 displays a linear dependence of current with scan rate, for the clay/poly-Fe(vbpy)32+ film. Additional support for thin cell behavior is provided by the symmetrical shape of the cyclic voltammogram for the Fe(vbpy)32+ oxidation wave (Figure 18c). The peak currents of the reduction waves and Ep 6 for the metal-centered oxidation wave appear at identical potentials for the polymer at bare graphite and within clay. However, E for the oxidation wave of the polymer incorporated in Pr“ clay films is shifted 40 mV positive of the Ep’a for the polymer on graphite. This shift in potential could result from interactions of the positively charged polymer with negative sites within the clay producing an energy barrier which must be overcome for electron transfer to occur. Peak separations of 81 mV for the oxidation wave of Fe(vbpy)32"’ (100 mV/sec) and a FWHM of 125 mV are greater than predicted and most likely result from resistive effects. 108 Figure 20 A plot of the peak height ip a (DA) vs. the scan rate (mV/sec) for the CV reproduced in Figure 18. 109 Current, 11A l l 1 1 l 0 100 150 200 250 Scan Rate, mV/sec Figure 20 110 Although thin film behavior does not unequivocally establish the Fe(vbpy)32+ to be polymerized, it is noteworthy that clay films incorporating comparable loadings of electrolyzed Fe(bpy)32+ (30 % of CEC for Fe(bpy)32+ clay film active / 34 % of CEC for Fe(vbpy)32+ clay film active) demonstrate diffusional behavior (ip is linear with 1/2 the square root of scan rate 12 and diffusional tail is observed in CV). Thus thin cell behavior appears to be specific to poly- Fe(vbpy)32+ attached to the electrode. Moreover electroactive ions do not leach from this film even after days of soaking in water and/or acetonitrile solutions containing electrolyte. These results strongly suggest that Fe(vbpy)32+ is polymerized in the clay. With the formation of a polymerized film established, the possibility of extending electroactivity into Region II can be investigated. The polymerization of Fe(vbpy)32"’ in Na+-exchanged and 88 % Fe(vbpy)32"' pre-exchanged montmorillonite films on graphite was performed in acetonitrile solutions containing 2.0 mM Fe(vbpy)37-+ and 0.2 M TEAP. Table 8 lists the number of coulombs of charge accessed during a slow voltammetric scan from the cathodic portion of the oxidation wave of polymerized Fe(vbpy)32+ in clay that has been transferred to acetonitrile solutions containing 0.2 M TEAP. Almost identical values of 2.2 and 2.1 x 10"4 coulombs for the polymerized films indicate that the gallery height does not greatly affect electroactivity, a result which supports earlier conclusions. For the purpose of comparison Table 8 also lists the coulombs obtained from Na+-exchanged and 88 % Fe(vbpy)32+ pre- exchanged montmorillonite films identically prepared to those above; polymerized films were soaked in 2.0 mM solutions of Fe(vbpy)32+ 111 Table 8 Amount of Electroactivity from 0.38 pm Montmorillonite Films Resulting from Incorporated or Polymerized Fe(vbpy)32+ . Solution Fe(L)32+ Solution Coul. Coulombs Film Extracted froma Measured in x 104 c Na+-exchanged polymerized acetonitrile acetonitrile 2.2 soaked acetonitrile acetonitrile - acetonitrile water 0.3 water water 1.2 Pre-exchanged 88 % crzc‘l polymerized acetonitrile acetonitrile 2.1 soaked acetonitrile acetonitrile - acetonitrile water 0.7 water water 1.3 'L = vinyl-bipyridine, and the solution contains 2 mM Fe(vbpy)32+. l’The soaking solution contains just electrolyte. coulombs under reduction wave after subtracting background. exchanged ion is Fe(vbpy)32+. cThe number of dThe 112 for 8 h and transferred to either acetonitrile or aqueous solutions containing 0.2 M TEAP and 0.2 M LiClO4, respectively. Significantly diminished responses from these films results from the poor swelling ability of clays in acetonitrile. Comparison of the electroactivity from Fe(vbpy)32"’ polymerized in clay films as opposed to incorporation by soaking clearly shows that polymerization enhances swelling. For additional comparison, these same Na+-exchanged and 88 % Fe(vbpy)32+ pre-exchanged montmorillonite films were soaked in aqueous solutions containing 2.0 mM Fc(vbpy)32+ and 0.2 M LiClo4 for 8 h and then transferred to aqueous solutions containing 0.2 M LiClO4. The coulomb data listed in Table 8 demonstrate that even films soaked in aqueous solutions, which completely eliminates swelling effects, have a diminished response over the polymerized clay films. The electroactivity data in Table 8 corresponds to 0.26 of the CEC for the films soaked in aqueous Fe(vbpy)32+ solutions and 0.48 of the CEC for the films polymerized in acetonitrile Fe(vbpy)32"' solutions. The results are clear. The polymerized films display almost twice the electroactivity over films which extract ions from solution. We therefore conclude that polymerized Fe(vbpy)32+ in clay films effectively extends Region I further away from the 2+ electrode by accessing previously isolated Fe(vbpy)3 in Region 11. Nevertheless all of the Fe(vbpy)32+ from Region II is not activated in the clay/polymer films because doubling the film thickness does not increase the amount of activated Fe(vbpy)32+. Thus Region I activity can be enhanced by creating a mode for charge propagation into Region II by polymerizing Fe(vbpy)32+ in the clay film, although much of Region II is still electroinactive. 113 Clay/poly-Fe(vbpy)32+ films exhibit unique electrocatalytic behavior. A clay/polymer modified electrode was placed in an aqueous solution containing 0.2 M LiClO4 and 4.2 mM Fe(CN)64' and its potential was cycled from - 0.10 to + 1.20 V vs. SCE at 2 mV/s. The resulting voltammogram (Figure 21a) demonstrates the characteristic S-shaped voltammetric response expected from electrocatalytic systems. Presumably the Fe(vbpy)32+ polymer is oxidized during the forward oxidation scan and before Fe(vbpy)33+ can be reduced at the electrode on the reverse cycle, it is reduced by Fe(CN)64' to generate Fe(CN)63' and Fe(vbpy)32"’, which in turn becomes reoxidized at the electrode. The large excess of Fe(CN)64' present in solution coupled with the slow scan rate enables a steady state response to develop which is manifested in the S shape of the wave. The efficiency of this catalytic cycle is evidenced by the magnitude of the difference in current responses from the clay/poly- Fe(vbpy)32+ in the presence (Figure 21a) and absence (Figure 21b) of added Fe(CN)64". In order to observe catalytic behavior the Fe(CN)64’ must necessarily come in contact with Fe(vbpy)33+. Therefore, Fe(CN)64‘ must either penetrate through the clay/polymer film or react at the edges of the film in Region I and/or Region II (Figure 22). The former pathway would result in a buildup of Fe(CN)63" within the clay/polymer film. Yet Figure 21b shows no evidence of Fe(CN)63'/4' in the film (the redox couple for Fe(CN)63'/4' appears at as +0.3 V) suggesting that catalysis is occurring at the edge of the clay/polymer film in either Region I or Region 11 (Figure 22). Because electroactivity from clay films arises only from ions near the electrode in Region I, and isolated ions further from the electrode in 114 Figure 21 Cyclic votammograms scanned at 2 mV/sec, of a clay/poly- Fe(vbpy)32+ film soaking in an aqueous solution containing 0.2 M LiClO4: (a) after the addition of 4.2 mM Fe(CN)64' and; (b) before the addition of 4.2 mM Fe(CN)64'. 115 Ill) uA E/Volts vs. SCE Figure 21 116 Figure 22 Model depicting origins of electrocatalysis from a clay/poly- Fe(vbpy)32+ film on an electrode. Fe(CN)6" FC(CN)63- 117 &\ \\\\. 3+ Fetvbpyh“ FC(VbPY)3 F<‘=(pr)’)32+ Region I Region II Fe(CN)64‘ Figure 22 118 Region II are electroinactive, necessarily implies communication between Fe(CN)64‘ and Fe(vbpy)33+ to occur at Region I edges. Thus as graphically displayed by the model proposed in Figure 22, electrocatalysis from the clay/poly-Fe(vbpy)33+ film is believed to arise from the edges of Region I. 4. Intersalated Clay Films With the studies described heretofore, electroactivity from ions in clay galleries has not been realized. As discussed in Section [11.8.], inactivity of ions in the the clay galleries results from unfavorable thermodynamics arising from the inability to maintain charge neutrality during electrolysis. The requirement of mobile negative ions within the clay galleries can, in principle, be achieved with intersalated clays. The general structure of montmorillonite intersalated with Fe(bpy)32+ is given below by structure 6. Negatively Charged Clay Layer Ft‘=(bP)’)32+ Fe(bpy)32+ S042- 1160313932+ SO42' Fe(bpy),2+ so.” 2+ 2+ Fe(bpy)3 Fe(bpy)3 Negatively Charged Clay Layer 6 An intersalated clay, prepared by adding the clay to 5 CEC of a polypyridyl metal complex in the presence of dianions of the form 119 X042’ (X = S, Mo, W), is distinguished by its incorporation of twice the CEC in the clay galleries. One CEC balances the charge of the clay layers while the additional CEC is present as an ion pair between the metal complex and the X042' counterion. The additional 3 CEC is external to the clay gallery (i.e. on surface layers and in voids around the clay platelets) and is required to maintain a favorable equilibrium for intersalate formation. This excess loading of the galleries is reflected by increased d-spacings. The d-spacing obtained for montmorillonite films with l CEC loading of Fe(bpy)32+ is 18.4 A whereas ion incorporation from solutions of 5 CBC yields intersalates with d-spacings of 29.5 A. The sulfate ions of intersalates are mobile and can be easily displaced with different anions.268 Because intersalated clays should allow for total activity of all redox cations within the clay film, the unfavorable thermodynamics associated with the maintenance of charge neutrality should be circumvented. Accordingly electrochemical studies were undertaken with the goal of observing electroactivity from ions electrostatically bound between clay layers. Intersalated modified electrodes were prepared by air drying a dilute suspension of intersalate onto a pyrolytic graphite electrode. Figure 23 shows the first two scans of an intersalated modified electrode immersed in a dichloromethane solution containing 0.2 M tetrabutylammonium perchlorate (TBAP). The quantity of Fe(bpy)32+ oxidized on the first scan, as determined by coulometry, corresponds to ~ 3 CBC of the montmorillonite film. However, the amount of metal complex electrolyzed upon the subsequent scan is greatly reduced and continues to decrease with continued scanning 120 Figure 23 The first two cyclic voltammograms of a Fe(bpy)32+ intersalated montmorillonite film on pyrolytic graphite immersed in a dichloromethane solution containing 0.2 M TBAP. 121 g fi—fi IluA \\ 111 11111_L11_1 0.2 0.4 0.6 0.8 1.0 1 .1 E/VOItS vs SCE Figure 23 122 of the oxidation wave. To ensure that this result was not caused by the collapse of the clay in dichloromethane, the same experiment was conducted in a saturated aqueous solution of sodium sulfate. Almost identical results were obtained. Although the increased electrochemical response of the intersalates is encouraging, current attributed to a totally activated intersalated film is not observed. The total number of coulombs of intersalates should correspond to 5 CEC. Noting that of the 5 CEC in intersalated films, 2 CBC are in the clay galleries and 3 CBC are external to the galleries, the observation of current corresponding to 3 CBC may reflect electroactivity from only Fe(bpy)32+ outside of the clay particles. In this case, the ultimate incorporation of Fe(bpy)32+ into an 88% pro-exchanged Fe(bpy)32+ montmorillonite film soaked in a 5.1 mM solution (the concentration of the 3 CEC excess Fe(bpy)32+ in intersalate suspensions as determined in Section II.A.3.f) should be the same as that for the intersalate because the 2+ total amount of electroactive Fe(bpy)3 incorporated in clay films is independent of the d-spacing. As shown in Table 9 the number of coulombs measured on the 1St and 20”1 scans of an 88% pre- exchanged Fe(bpy)32+ film, which has been soaked in 5.1 mM Fe(bpy)32+ for 3 h and transferred to a dichloromethane solution containing 0.1 M TBAP, corresponds to the Na+-exchanged film and not to the intersalate. The values in Table 9 were determined for the reduction rather than the oxidation of Fe(bpy)32+ because the former excludes charge arising from the oxidation of water. As one additional control experiment, 3 CBC of Fe(bpy)32+ was added to an 88 % pre-exchanged montmorillonite slurry before the film was cast. 123 Table 9 Amount of Electroactivity from Different Clay Films after Soaking in 5.1 mM Fe(bpy)32” # of coulombs Montmorillonite Film 1St scan 20th scan Intersalated + 5 CBC 13.5 x 104 ' 3.0 x 104 ' of Fe(bpy)32+ 4.4 x 104 b 2.5 x 104 b Pre-exchanged with 88% 1.4 x 10'4 b 1.0 x 10'4 b CEC of Fe(bpy)32+ Na+-exchanged 1.3 x 10'4 b 0.9 x 10‘4 b Pre-exchanged with 88 % 3.9 x 104 ' 1.5 x 104 ' CEC of Fe(bpy)32+ + 3 CBC 3.7 x 10*4 b 1.2 x 104 b of Fe(bpy)32+ “ “The number of coulombs under oxidation wave after subtracting background. l’The number of coulombs under reduction wave after subtracting background. “This film was not soaked in a 5.1 mM Fe(bpy)32+ solution. 124 Initial electroactivation is ~3 times that obtained for pro-exchanged films (Table 9) soaked in Fe(bpy)32+. Nevertheless, the activity on the first oxidation for the intersalate is ~10 times this value. In addition, by the 20“I scan the responses from both pro-exchanged films are basically the same whereas the intersalate activates about 3 times more Fe(bpy)32+. Thus excess activity is clearly observed for intersalates. The decrease in electroactivity from intersalates after the first scan suggests significant and irreversible perturbation of the microstructure upon passing current into the film. This could mean that Fe(bpy)33+ is not stable when exchanged into clay galleries. Previous experiments by Oyama, Oyama and Ansonzm'279 and Rudzinski and Bard261 have demonstrated that structural iron located in the clay octahedral sites is redox active. Could the Fe(bpy)33+ produced upon oxidation of the intersalate react with the clay structure thereby leading to its degradation in the clay gallery? UV-vis spectroscopy in combination with bulk electrolytic experiments suggest otherwise. Electronic absorption spectra show only the presence of Fe(bpy)32+ and Fe(bpy)33+ and CV's display only the reversible 3+/2+ couple. Therefore, instability of intersalates under electrolytic conditions must result from a different source. X-ray patterns of the intersalated clay sample on graphite before and after electrolysis are enlightening. Figure 24a shows the expected diffraction profile for an intersalated clay on pyrolytic graphite; the appropriate d-spacing of 29.4 A is observed.2“7'268 The pattern changes significantly upon electrolysis of a single oxidative scan. As shown in Figure 24b, the order of the clay film is almost 125 Figure 24 The X-ray diffraction patterns obtained from Fe(bpy)32+ intersalated films on pyrolytic graphite: (a) before the electrolysis of the film; and (b) after the electrolysis of the film in a dichloromethane solution containing 0.2 M TBAP. 126 06—. 0d 90 0.5 3 2:2...— mu woo. mon— 06 o.m 06 0.9 gm laud 1 cm..- 10b... 1094 I In on F 10m... eurelou Allsuetul 127 completely lost; only the sharp peak corresponding to crystalline Fe(bpy)32+, present in the regions outside of the clay particles, is present. This x-ray data suggests that upon electrolysis the intersalate is collapsing to a monolayer with the rapid ejection of Fe(bpy)33+ from the galleries. Significantly, the diffraction pattern in Figure 24b indicates that the clay platelets are no longer stacked in an ordered manner. Increased activity of intersalate on the 20th scan is consistent with our previous data which suggests the manifestation of disorder in increased electroactivity owing to a larger basal volume for electroactive ions. Collapse of the intersalate will create large void volumes which will be occupied by the ejected ions. The greater volume of voids in Region I should allow the incorporation of more cations and hence greater electroactivity. These results clearly show that in order to sustain electroactivity from intersalated species, the metal complex must be immobilized within the galleries as an intersalate to prevent ejection of the ion upon electrolysis. This can be achieved with the electrochemical polymerization of a clay film intersalated with Fe(vbpy)32+. A Fe(vbpy)32+ intersalated montmorillonite film was cast on graphite, placed into a acetonitrile solution containing 2.0 mM Fe(vbpy)32+ and 0.2 M TEAP, and the oxidation and reduction waves were cycled for 1 h to induce polymerization. Figure 25 reproduces the oxidation and reduction waves for this film in acetonitrile containing 0.2 M TEAP. The amount of material oxidized as determined by coulometry corresponds to 1.55 CEC of the clay. The current response vs. scan rate for this intersalated clay/polymer film shows thin cell behavior at slow scan rates but negative deviation at 128 Figure 25 Cyclic voltammetric waves of a polymerized Fe(vbpy)32+/intersalated montmorillonite film adsorbed onto pyrolytic graphite while soaking in an acetonitrile solution containing 0.2 M TEAP. 129 u.~ m.e mu 2:3..— m—Um .m.» 3.3:”.— > x w 0.0 0.0 no- _ _ _' <1 98 l. 0.... 31120113 O ' ’ DO ’. .-~..... g { lueuno 186 The oxidation wave of the cluster becomes electrochemically irreversible at slow scan rates. In Figure 40, the wave clearly shows irreversible behavior at 10 V/s. The enhanced reversibility at faster scan rates is possibly due to irreversible chemical reaction of the oxidized cluster which is circumvented by faster scan rates. However, W6Br142' is completely reversible at a bare platinum electrode in acetonitrile ( ip’a = 1p.c and ipC‘ is constant from cyclic voltammetry; it“2 is constant for chornoamperometry (short times); and ‘tr = “(f/3 for reverse Chronopotentiometry) and we have no evidence that decomposition of the monoanion is promoted by the polymer film. Alternatively, the presence of prewave humps on both the first reduction wave of EVDQ2+ and the oxidation wave of W6Br142' (Figure 39) provide a clue as to the origins of the electrochemical irreversibility. this unusual behavior. Similar to the prewave spikes described in Section III.B for Fe(vbpy)32+, the prewaves in Figure 39 are associated with one another and correspond to the anodic and cathodic components of a redox couple. This behavior is consistent with the oxidized cluster becoming trapped in an inaccessible region of the polymer and therefore being unable to return to the electrode (ip a >' 1p’c). By 1ncreas1ng the scan rate, the W6Br14' ion does not have enough time to diffuse into the isolated domalns resultlng 1n 1pm = 11”. On thls bas1s the poss1blllty exists that the redox sites of the polymer might be able to shuttle charge to the isolated W6Br14'. If this does occur we would expect that scanning the reduction waves of the polymer subsequent to scanning the oxidation wave of the cluster should result in an enhanced response from the cathodic portion of the polymer 187 Figure 40 Cyclic voltammograms of W6Br142' incorporated into a EVDQ2+ film adsorbed onto a platinum electrode (surface coverage = 1.7 x 10'8 mol/cmz) immersed in an acetonitrile solution containing 0.2 M TEAP recorded at (a) 10 V/s and (b) 33V/s. 3. «Laura 0 _+ 0.. + n0... . 0.0 + 188 udq 4* Aqiqqq «W «1 _ O 0000 00000.0. OOIOOOCQOOO 00.... <1 QSH coo-coco. Noun: 9190“ o 189 reduction wave. It is noteworthy that the polymer reduction wave is reversible if the oxidation wave of cluster is not scanned. This is expected because isolated oxidized cluster is not present for the polymer to shuttle charge to, unless W6Brl4' is produced. Similarly the anodic portion of the oxidation wave for the cluster should be significantly diminished upon its second scan without cycling the polymer reduction waves because the oxidized cluster in the isolated regions of the polymer can not communicate directly with the electrode. Without activation of the polymer, isolated cluster ions can not be reduced. Indeed these are the observed results. Enhanced response from the polymer reduction wave is observed with attenuated responses of the cluster oxidation wave. In a typical experiment, the number of coulombs lost on subsequent scans of the cluster oxidation wave appear almost identically as an increased response for the cathodic current of the polymer wave. The ratio of these experimentally determined differences in the cluster and polymer waves is 1.18 1.04 (for three separate films). This value is close to unity and suggests that the polymer mediates electron transfer to isolated W6Br14‘. Therefore, while W6Brl42' is tightly bound within the polymer, W6Br14' freely diffuses. The apparent diffusion coefficient of W 6Br142' incorporated within EVDQ2+ films determined by using the procedure described in Section II.B.4 is 1.8 x 10'11 cm2/s (accurate determination of the diffusion coefficient of W6Br14' was impossible because of its rapid exodus to the isolated regions of the polymer). This value is significantly smaller than the value obtained for the 190 PO“ system (8.7 x 10'11 cm2/s) suggesting even tighter binding for W6Br142' in the 2,2' over the 4,4' bipyridinium films. 4. Comparison of EVDQ2+,PVDQZ*, and BVDQ2+ Polymer Modified Films The charge density of the cationic binding site increases along the series PVPt—Me, PO“, and EVDQ2+ and a corresponding decrease in the diffusion coefficient of W6Br142‘ is observed. The ability to control the diffusion rates of ions within polymeric films by specifically designing polymers with varying charge density binding sites is a novel and potentially useful concept. However, in addition to the binding site, other significant differences in the polymer microstructures exist. Therefore, in an effort to unequivocally assess the role of the binding site in determining charge transport in the pyridinium and bipyridinium films it was of interest to investigate the diffusion rates of cluster in structurally similar polymers with binding sites of varying charge density. The 2,2'-bipyridinium systems provides the opportunity to undertake such studies. By increasing the length of the linkage between the nitrogen groups, the effective charge density of the dication binding site can be altered. The interplanar angle of the pyridine rings in N,N'-ethylene-2,2'- bipyridinium (EDQ2+) is 9°, whereas for N,N'-1,3-propylene-2,2'- bipyridinium (PDQ2+) is 60° and for N,N'-1,4-butylene-2,2'- bipyridinium (BDQ2+) is 80°.337 Therefore, polymers formed from the vinyl containing analogs of these monomers should possess progressively weaker binding pockets as the interplanar angle 191 increases, while maintaining structural uniformity in the overall polymer microstructure. The physical properties of 4-vinyl-4'-methyl-N,N'-propylene- 2,2'-bipyridinium (PVDQ2+)S and 4-vinyl-4'-methyl-N,N'-butylene- 2,2'-bipyridinium (BVDQ2+) 6, synthesized as described in Sections II.A.2.f and II.A.2.g, were identical to EVDQ“. FABMS exhibited parent peaks which correspond to their prospective molecular weights with a water of hydration. The same polymerization procedure for EVDQ2+ was employed with the exception that the I‘ salts were used. The cyclic voltammogram of PVDQ2+ on a platinum electrode with a surface coverage of 4.0 x 10'9 mol/cm2 in acetonitrile containing 0.1 M TBAP is shown in Figure 41. The cathodic peak potentials are Ep’l = -0.76 V and Ep,2 = —l.04 V vs. SCE. The shift in the first reduction wave to more negative potentials as compared to EVDQ2+ is attributed to the decreased delocalization of the x-system arising from the torqued pyridinium rings of PVDQ“. The second reduction wave is relatively insensitive to changes in structure as evidenced by their values only differing by 20 mV. The two reduction waves for PVDQ2+ (Figure 41) display reversible thin cell behavior (AE = 20 mV, FWHM = 90 mV) and peak currents are directly proportional to the scan rate. A PVDQ2+ modified platinum electrode (surface coverage = 1.06 x 10'8 mol/cmz) readily incorporates cluster from an acetonitrile solution containing 20 nM wésrlf and 0.2 M TBAPF6 as shown in Figure 42a. The oxidation wave does not have a diffusional tail and peak currents are directly proportional to scan rate. Ion-exchanged 192 Figure 41 Cyclic votammogram of a PVDQ2+ film adsorbed onto a platinum electrode (surface coverage = 4.0 x 10'9 mol/cmz) immersed in an acetonitrile solution containing 0.2 M TEAP 193 2.5 11A l l l l l I l -1.2 -1.0 -0.8 -O.6 -0.4 -0.2 0.0 EJVolts vs. A0 QRE Figure 41 194 Figure 42 Cyclic voltammetric waves obtained from a PVDQZ“ film coated onto a platinum electrode (surface coverage = 1.1 x 10'8 mol/cm“) dipped into an acetonitrile solution containing 0.2 M TEAP and 20 11M W6Br142': (a) upon scanning the electrodes potential from 0.00 V to +1.40 V to -1.10 V and back to 0.00 V vs. SCE; (b) after the reduction waves for the PVDQ2+ polymer had been scanned once prior to measurement; and (c) after the oxidation wave of WéBer' had been cycled once prior to measurement. 195 ooé mud end a. 8:3... mac m< a, 2m mud . 8.0 _ _ mud- 196 cluster in PVDQ2+ shows no evidence of a return wave even at scan rates as fast as 50 V/s. In this regard the PVDQ2+ behaves similarly to EVDQ2+ in that W6Br14' diffuses to isolated sites within the PVDQ2+ microstructure, as shown in Figure 42a. Scanning the oxidation wave two consecutive times results in a decreased electroactive response (Figure 42b). Subsequent scans of the reduction waves of the PVDQ2+ polymer film show an enriched response of the cathodic portion of the polymer reduction wave on the first scan (ip c > ' 11),! 011 first scan, i = p c i1) 8 on second scan). If the increased response of the cathodic portion of the reduction wave results from the polymer film shuttling charge to isolated cluster ions, then the differences in electroresponses for the oxidation of cluster and reduction of polymer from Figure 42 should correspond to the same number of coulombs. Integration of these waves establishes that this is clearly the case (the ratio for the number of excess coulombs for polymer reduction divided by the number of lost coulombs for the oxidation of cluster = 1.04 i .03). This model is further substantiated by charge shuttling experiments. Reactivation of isolated oxidized cluster ions which cannot directly communicate with the electrode surface, is not necessarily related to polymer but can be accomplished by a variety of freely diffusing shuttle reagents. The reduction of isolated cluster ions by IrCl63' is shown in Figure 43. IrCl63' initially generated at the electrode is in evidence by a cathodic component of the IrCl62‘l3‘ couple. Of course in a charge shuttle mechanism the anodic component of the couple is not observed because the IrC163" diffuses to isolated W6Brl4', reduces it, and the IrCl62' generated by this 197 Figure 43 Cyclic votammogram of a PVDQ2+ film on a platinum electrode soaking in an acetonitrile solution containing 0.2 M TBAPF6, 25 11M W6Br142‘ and 50 M er162'. 198 E/V vs. Ag ORE Figure 43 199 reaction diffuses back to the electrode and becomes rereduced whereupon it diffuses to another oxidized cluster thereby establishing a catalytic cycle. This mechanism is borne out by the absence of an oxidation wave for IrCl62'. These results establish that W6Br14’ ions freely diffuse within the PVDQ2+ microstructure and become confined in an isolated region of the film which can only be accessed by the use of intrinsic and extrinsic charge mediators. The reduction wave for BVDQ2+ polymerized onto platinum produces one very broad reduction wave at approximately -O.95 V vs. SCE. The first reduction wave is evidently shifted far enough negative that it engulfs the second reduction wave. Thin films of BVDQ2+ formed onto latinum extract cluster ions from a acetonitrile P solution containing 10 uM W6Br142 . W6Brl42' exchanged into BVDQ2+ and PVDQ2+ display almost identical The oxidation waves for behavior. Apparent diffusion coefficients for W6Br142'electrostatically bound within PVDQ2+ and BVDQZ”, determined as described in Section II.B.4 are 3.0 x 10'11 cm2/s and 4.1 x 10'11 cmzls respectively. The Dapp values for PVDQ2+ and BVDQ2+ are more similar to each other than for EVDQ2+ because the interplanar angles are more closely correlated. The apparent diffusion coefficients Dam) of cluster in l>v1>+—Me, PQZ", EVDQ“, PVDQZ”, and mm“ are graphically summarized in Figure 44. These data clearly demonstrate that the diffusion rate of exchanged W6Br142' ion decreases with increasing, charge density of the binding sites. Moreover this conclusion is further evidenced by determinations of the distribution coefficient KD (the concentration of redox ion in the 200 Figure 44 A plot of the apparent diffusion coefficients, D for W6Br142’ aPP incorporated into PVPt—Me, Po“, BVDQZ”, PVDQZ“, and EVDQZ” polymeric films as a function of the number of binding sites. 201 -8.0_ + PVPMe+ -8.5.._ “"2 -9.0_ In ”9.. E 3 -95; a“ o 8" .J -10.0_. O H)“ A BVDQ2+ -10.5— 0 PVDQ2+ I] EVDQ2+ l l 1 2 Number of Binding Sites Figure 44 202 Figure 45 A plot of the distribution coefficients KD for W63r142' incorporated into PVPL—Me, PQ2+, woo“, woo”, and EVDQ“ polymeric films as a function of the number of binding sites. L03 Kn (W6Bl'142.) 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 203 + PVPMe+ O PQ2+ A liVDQ2+ O PVDQ2+ [j EVDQ2+ l l 1 2 Number of Binding Sites Figure 45 204 polymer film divided by the concentration in the soaking solution)109 for W6Br142' incorporated into the polymers. As shown in Figure 45 the KD values for the five polymers investigated monotonically increase as a function of binding site charge density in a parallel manner to the apparent diffusion coefficients lending further evidence to our conclusion that in acetonitrile, electrostatic binding in ionic polymers is directly related to the charge density of the binding site.. C. Conclusion These studies demonstrate that the diffusion rates of ions exchanged into polyionic polymer films can be influenced by the design of the polymer microstructure. The binding pocket of PVP+— Me requiring the juxtaposition of two pyridinium rings per cluster ion is structurally floppy and W6Br142' is mobile. The diffusion coefficient of cluster is suppressed by several orders of magnitude by bringing the individual positive charges of the PVPt—Me sites into a single pocket in PQ2+. By localizing the charge even further into the 2,2' position of the bipyridinium ring the ion-pairing is further enhanced. As originally inferred from structural considerations, the diffusion coefficients of W6Br142‘ decrease along the the series EVDQ“, PVDQZ”, and EVDQ2+ owing to the tilt angle between the bipyridinium rings. This correlation of the diffusion rates among the pyridinium and bipyridinium series, and parallel behavior of KD and Dapp strongly suggests that the primary perturbation on the charge 205 transport in these films is the nature of the binding site and that the macroscopic polymer structure plays at most only a secondary role. In each polymer system investigated the W6Br142’ ions are immobile in tight ion-pairs, however, the oxidation of cluster or reduction of polymer binding sites to +1 reduces ion-pair interactions significantly, such that W6Brl4' becomes mobile. Hence fixed site distance for ecl reaction between oxidized cluster and reduced polymer will not be maintained. On this basis the eel yields for these cluster/polymer systems are expected to be similar to cluster/pyridinium ions in solution. This is the case. Thus in order to accomplish fixed distance electron transfer in polymer microstructures, reactant pairs will have to be covalently liked to each other to ensure they do not diffuse apart. This can be accomplished by preassembling the system. For example, M06C113LL' modified with the bidentate ligand L,L' will enable W6Brl3' to be coordinated. Production of oxidized tungsten bromide cluster and reduced molybdenum chloride cluster in the charge rectification sense will establish an ecl annihilation reaction at closest contact. CHAPTER V DEVELOPMENT OF AN ELECTROANALYTICAL SENSOR A. Background One of the major challenges facing analytical chemists is the ability to quantify trace and ultratrace quantities of materials from multicomponent samples. The best-known electroanalytical technique is stripping analysis,339 whereby a trace metal is electrolytically deposited onto an electrode thus satisfying the dual purpose of concentration and isolation of the analyte from the complex matrix. The analysis is conducted by electrochemically stripping, in one potential step, the deposited metal from the electrode surface, resulting in a enhanced current response from the electrode over that which would be observed if no analyte was deposited onto the electrode surface. This technique is virtually limited to metal ions as the analyte and mercury electrodes because most organics and inorganic complexes cannot be electrodeposited and solid electrodes become easily fouled.339 The area of voltammetry following nonelectrolytic concentration has been established as a technique with tremendous 4“ The most potential for trace analysis of electroactive analytes.3 recent development in this field has been the use of modified electrodes. The methods developed to confine analytes within polymers on electrodes have been: (i) complexation which utilizes functional groups on the polymer backbone to coordinate the analyte; (ii) covalent attachment of the analyte to the polymer backbone; and (iii) electrostatic binding of the analyte within a polyionic 340 polymer. Of these techniques electrostatic binding of analytes is the most versatile because specific reactions between the analyte 206 207 and polymer are not required. The analysis of multicharged electroactive analytes has been achieved in many systems."6 Although there has been a report in the literature claiming to detect A g"’ at concentrations as low as 1 x 10'11 M341 the majority of the polyionic sensors have detection limits in the range of 1 x 10'“ to l x 10'9 M with linear calibration curves over approximately 2 orders of concentration.342'348 The desired features for an electroanalytical sensor for analyte are: (i) rapid equilibration between contacting solution and polymer; (ii) low detection limits with large concentration ranges of linear current responses; (iii) efficient operation over a large pH range; (iv) detectability of monocharged species; (v) ability to reversibly establish the sensor (i. 8. reverse ion-exchange can be accomplished); (vi) operation in both aqueous and nonaqueous solvents; and (vii) the ability to discriminate the desired analyte from complex mixtures. The sensing systems developed heretofore achieve only a few of these goals. Therefore the development of new sensors is of major interest. Described herein is a novel electroanalytical sensor which realizes the above goals. B. Results and Discussion 1. Development of an Electroanalytical Anionic Sensor for use in Nonaqueous Solvents The EVDQ2+ polymer films extract and concentrate anions from very dilute solutions. Therefore, these films are viable candidates as electrochemical sensors for anion detection in nonaqueous solutions. 208 For example the unprecedented detection of a dianion (W6Br142‘) at concentrations as low as 5 x 10'8 M in acetonitrile by cyclic voltammetry is shown in Figure 46. The detection limit for W6Br142 with a film of EVDQ2+ (surface coverage 2.1 x 10'9 moi/cm“) is 3 x 10'8 M as determined from the anodic peak current of the cyclic voltammogram and would presumably be even greater by using a more sensitive technique such as differential pulse polarography. Moreover, the anodic peak current for the detection of W6Br142’ is linear over 3.5 orders of magnitude (Figure 47a). Unfortunately the peak current is dependent on the surface coverage of the film as evidenced by Figure 47b which shows that a film with a surface coverage of 1.1 x 10'9 mol/cm2 has a limit of detection of l x 10'7 M and displays linear responses only over 2.5 orders of magnitude. The sharp rise in peak currents in Figure 47b corresponds to currents which would be attained from a bare electrode at the same concentration indicating the film is not altering the electroresponse at higher concentrations. Films thicker than 4.5 x 10‘9 mol/cm2 also display diminished ability to detect low concentrations of W6Brl42'. Film coverages between 2.0 x 10’9 mol/cm2 and 4.0 x 10'9 mol/cm2 yielded the greatest sensitivity and accompanying linear range of detection for the cluster anions. The fact that the magnitude of peak currents are affected by the film coverage of EVDQ2+ is a drawback because these films must be calibrated before they can be used to quantify electroactive material in solution. Importantly, the high sensitivity for anion binding and detection have also been observed for anions other than cluster including Fe(CN)63', Fe(CN)4bpy', Irle', 1', and Br'. These films have tremendous potential for the 209 Figure 46 Cyclic voltammogram of W6Br142' incorporated from an acetonitrile solution containing 0.2 M TEAP and 5 x 10'8 M W6Br142' into a EVDQ2+ film adsorbed onto a platinum electrode (surface coverage = 4.1 x 10'9 mol/cmz). 210 3 2:3,... mac 2 2, >5 0.... 0.... 0.0+ ogpoue O 111811110 211 Figure 47 A plot of the cyclic voltammetric peak current for the oxidation of W6Br142' incorporated into EVDQ2+ films as a function of the concentration of W6Br142’ in the contacting acetonitrile solution for platinum electrodes with surface coverages of, (a) 2.1 x 10'9 moi/cm2 and (b) 1.1 x 10‘9 mol/cmz. Current (11A) 212 I 6.0 5.0 " -Log (Concentration M) WgBrMZ' Figure 47 213 electrochemical detection of trace quantities of anions in nonaqueous solvents. 2. Characterization of Anionic Effects on the Microstructure of EVDQ2+ Seldom is the analysis of a pure sample required and most often extraction and detection of analyte from a complex mixture is required. For EVDQ2+ polymer films, anions other than the analyte, have the most potential to interfere with the function of the film as a sensor. Thus the examination of anion effects on the charge transport rates of EVDQ2+ modified electrodes was undertaken. The Dct for the reduction of EVDQ2+ and PVDQ2+ could not be determined because potential steps large enough to overcome resistive effects can not be achieved. The potential was stepped past the reduction wave for several seconds to ensure complete reduction of the film, and after the potential was stepped back to 0.0 V vs. SCE the current response as a function of time was measured. The charge transfer diffusion rates Dct for EVDQ2+ and PVDQ2+ films soaking in acetonitrile solutions containing different electrolytes are given in Tables 11 and 12, respectively. These Dct values are clearly sensitive to the supporting electrolyte. The Dct values are determined for the oxidation of +1 polymers to +2, which necessarily requires accompanying cation ejection or anion uptake. The role cations play in the charge neutrality process can in part be understood by comparison of the Dct values for the perchlorates (Tables 11 and 12). The rate of charge transport decreases along the series LiClO4 > TBAP 2 TEAP. Clearly no obvious 214 Table 11 Apparent Diffusion Coefficients and FWHM Values for EVDQ2+ Films With Varying Thicknesses rm, x 109 0.2 M Fwanmv l)ct x 1011 moles/cm2 ' Electrolyte in,c 1p“ cmzlsec“ 5.6 LiClO4 90 110 3.61 5.6 TBAP 91 114 2.61 14.3 TBAP 120 153 2.80 5.6 TEAP 90 115 2.54 12.9 TEAP 115 144 2.50 14 3 TEAP 110 140 2.45 5.6 TBAPF6 130 146 1.49 12.9 TBAPF6 140 170 1.50 14.3 TBAPF6 151 179 1.41 14.3 TBABF4 . 162 190 0.57 14.3 TBAAsF6 185 236 0.43 “The film coverage 1" CO” was determined using the electrochemically determined area of the Pt electrode. I’Full width at half maximum of the cyclic voltammetric peak for the first one electron reduction of EVDQ“. “Dct is the apparent diffusion coefficient from EVDQ2+ films. 215 Table 12 Apparent Diffusion Coefficients and FWHM Values for FVDQ2+ Films With Varying Thicknesses rem x 109 0.2 M FWHMme Dct x 1011 2 ~ 2 moles/cm “ Electrolyte lp’c {pa cm /sec“ 9.7 1.1004 120 146 2.34 12.0 1.1004 117 150 2.69 3.5 TBAP 90 115 1.69 9.7 TBAP 135 166 1.60 12.0 TBAP 140 160 1.65 2.4 TEAP 90 124 1.37 9.7 TEAP 151 183 1.35 10.0 TEAP 146 180 1.41 2.4 TBAPF6 124 152 0.48 3.5 TBAPF6 130 155 0.48 9.7 TBAPF6 176 204 0.53 1.0 TBAFF6 180 210 0.41 3.5 TBABF4 155 176 0.32 9.7 TBABF4 220 255 0.39 9.7 TBAAsF6 240 306 0.31 aThe coverage I“corr was determined using electrochemically determined area of the Pt electrode. I’Full width at half maximum of the cyclic voltammetric peak for the first one electron reduction of EVDQ“. “D ct is the apparent diffusion coefficient from EVDQ2+ films. 216 trend between the size of the cation and Dct rates exists. These observations implicate anion uptake as the process responsible for maintaining charge neutrality. The Dct values for the anions in Tables 11 and 12 follow the order ClO4‘ > PF6‘> BF4' > AsF6'. With the exception of BF4', this series parallels hard/soft acid/base formalisms with the hardest base ClO4' (smallest least polarizable anion) producing the fastest Dct rate and the softest base AsF6’ (largest most polarizable anion) producing the slowest Dct rate. Accordingly, ClO4' would be expected to form the strongest ion-pair with the positively charged sites in EVDQ2+ films and AsF6' the weakest. These data seemingly contradict conventional thinking which predicts that the anion movement will slow and become the rate limiting step for charge transport as anion binding increases; thus the smallest Dct values should be expected intuitively for C104' and the largest for AsF6‘. How can this disparity be explained? Electron hopping between fixed sites attached to electrodes require an accompanying displacement of counterions to maintain charge neutrality. Recently Saveant has established that steady state currents obtained from immobilized sites on an electrode should be 349 Furthermore, Andrieux and independent of counterion mobility. Saveant suggest that the current responses from transient techniques (like chronoamperometry) should increase as the mobility of the counterion decreases.350 Charge propagation is driven by a potential gradient established in the polymer film by an applied potential at the electrode. When counterions are mobile the potential gradient, which drives charge propagation through fixed polymer redox sites by site-to-site electron hopping, is diminished by the counterion 217 movement. However, when counterions are immobilized the potential gradient can not be satisfied by ion movement. Therefore, for transient potential steps the electron hopping rate and Dct will be driven by a larger potential field and thus charge transport is dominated by migration increasing the observed propagation rate. As ion mobility decreases, the potential gradient becomes larger and thus charge propagation rate increases. This increase in observed current from transient techniques, like chornoamperometry used in this study, results in larger calculated Dct values (see eq 10, Section I). To our knowledge Tables 11 and 12 lists the first data that are consistent with this model. The anion's increasing ability to form tighter ion-pairs with positively charged polymers parallels increasing Dct values. This implies that the currents measured for C104‘ have the largest contributions from migration and AsF6‘ lowest, resulting in larger calculated Dct values for CIO4'. Saveant's model is further supported by the fact that for a given supporting electrolyte EVDQ2+ which has the tightest binding pocket exhibits larger Dct values than PVDQ“. Admittedly the Li+ and BE,“ data on Tables on 11 and 12 are not in line with Saveant's model. The fact that Li+ and BF 4' are the smallest and least polarizable cation and anion studied respectively, may imply that more complex relationships between these counterions and polymer motion exists. Of course the Saveant model will only apply for charge app of W6Br142' because the monoanion can freely diffuse. Thus the charge propagation along the polymer redox sites and not for the D propagation rate for cluster will always contain a diffusional component. Currents arising from the physical diffusion of ions as 218 the mode of charge propagation as opposed to site-to-site charge hopping can not be dominated by migration (diffusion and migration are competing processes). Therefore, the Dapp's of W6Br142' should obey intuitive diffusional behavior i.e. tighter binding equals slower diffusion. The charge transport rates, Dct were determined for both EVDQ2+ and PVDQ2+ as a function of increasing W6Br142" concentration in acetonitrile containing 0.2 M TBAPF6 and are listed in Table 13. On the basis of the Saveant model we would expect an increase in Dct values from the films with tighter ion-pairs. As shown in Figures 44 and 45, presumably due to its higher charge W6Br142' binds more efficiently to the 2,2'-bipyridinium sites as compared to monoanionic electrolytes. The Dct value for EVDQ2+ increases from 1.50 x 10'11 cm2/s with only TBAPF6 present to 4.39 x 10‘11 cm2/s when 5 11M WéBrMZ' is added to the solution. Similar behavior is observed for PVDQ2+. These data clearly support the Saveant model. Eventually the Dct values of polymer decrease with increasing concentration of W6Br142’ (however it should be noted that even with the observed decrease, Dc 's are faster than that 2- 1 measured when no W6Brl4 is present. As each 2,2'-bipyridinium site becomes bound by cluster, it is likely that the neighboring binding sites become diametrically opposed to each other across polymer chains in an effort to reduce W6Brl42‘—W6Brl42’ interactions. Thus would perturb the polymers ability to establish proper orientation required for site-to-site electron hopping to occur. Interestingly, the FWHM of the polymer reduction waves was noticed to parallel Dct values in that as D decreases, FWHM Ct 219 Table 13 Apparent Diffusion Coefficients for EVDQ2+ and PVDQ2+ Films w/wo Added whit-142- Conc. of Added rem x 109 0.2 M D c, x 1011 WGBruz' uM moles/cm2 " Electrolyte cmzlsec" EVDQ2+ 0 1.3 TBAFF6 1.50 5 1.3 TBAPF6 4.39 10 1.3 TBAFF6 3.29 50 1.3 TBAFF6 2.19 FVDQ2+ 0 1.0 TBAPF6 0.41 5 1.0 TBAPF6 2.33 10 1.0 TBAFF6 1.33 50 1.0 TBAPF6 0.79 “The film coverage 1'60” was determined using the electrochemically determined area of the Pt electrode. bDct is the apparent diffusion coefficient. 220 increases. The FWHM is controlled by interactions between neighboring charged sites to the extent that repulsive interactions are manifested in increasing FWHM. The predicted value for FWHM of 90 mV is only observed for thin films using ClO4' as the electrolyte. For a given electrolyte in Tables 11 and 12 the FWHM becomes larger for thicker films. This observation is expected because repulsive interactions between neighboring chains become more likely. These repulsive interactions between chains will be mediated by ion-pairing. Anions which bind tighter are more efficient at neutralizing charge and reducing repulsive interactions between polymer redox sites than weaker binding anions. On this basis the former would be expected to possess smaller FWHM values. Moreover, as shown in Tables 11 and 12, the FWHM is always larger for the anodic wave then the cathodic wave. Because anion uptake has been established as the process by which charge neutrality is maintained upon oxidation of reduced polymer, it is likely that anion ejection controls charge neutrality for reduction of the polymer. The reduction of the film should occur rapidly because anions already in the film are simply ejected. However, reoxidation involves the uptake of anions thereby requiring swelling of the polymer to permit rentry of the anion into the film. This is a thermodynamically more difficult process producing equilibrium effects which cause multiple E° values. Thus broadened waves are observed for the anodic component of the reduction wave. 221 C. Conclusion This is the first reported example of a electrochemical sensor based on modified electrodes which can detect extremely dilute concentrations of monoanions in nonaqueous solutions. The characterization of electrolyte effects on the charge propagation rates support charge propagation driven by migration. In addition preliminary experiments demonstrate that these films work equally well with aqueous solutions over wide pH ranges. Further investigations of aqueous solutions are ongoing in our laboratory. Ultimately development of practical sensors requires further development with emphasis on stability, impurity fouling, and low cost. However, the fundamental function of these polymer modified electrodes as novel sensor materials has been established. CHAPTER VI REFERENCES 10. 11. 12. (a) Murray, R.W. Acc. Chem. Res. 1980,13, 135-141. (b) Murray, R.W.; Bard, A.J. (Ed.) ”Electroanalytical Chemistry"; M. Dekker: New York, 1984, 13, 191-392. Snell, K.D.; Keenan, A.G. Chem. Soc. Rev. 1979, 8, 259-282. Miller, J.S. (Ed.) "Chemically Modified Surfaces in Catalysis and Electrocatalysis"; American Chemical Society: Washington DC, 1982, pg 1-301. Albery, W.J.; Hillman, A.R. Ann. Rev. C. R. Soc. Chem. Lond. 1981, 377-385. Faulkner, L.R. Chem. Eng. News 1984, Feb 27, 28-45. Abruna, H.D. Coord. Chem. Rev. 1988, 86, 135-189. Harrison, D.J.; Wrighton, M.S. J. Phys. Chem. 1984,88, 3103- 3107. (a) Fan, F.-R.F.; Wheeler, B.L.; Bard, A.J.; Noufi, RN. J. Electrochem. Soc. 1981,128, 2042-2045. (b) Fan, F.-R.F.; Hope, G.A.; Bard, A.J. J. Electrochem. Soc. 1982, 129, 1647- 1649. (a) Noufi, R.; Tench, D.; Warren, L.F. J. Electrochem. Soc. 1980, 127, 2709-2713. (b) Noufi, R.; Frank, A.J.; Nozik, A.J. J. Am. Chem. Soc. 1981.103, 1849-1850. Rosenblum, M.D.; Lewis. N.S. J. Phys. Chem. 1984, 88, 3103- 3108. Nanthakumar, A.; Armstrong, N.R. J. Electroanal. Chem. 1988, 248, 349-362. Gningue, D.; Horowitz, G.; Garnier, F. J. Electrochem. Soc. 1988, 92, 5221-5229. 222 13. 14. 15. 16. 17. l8. 19. 20. 21. 223 (a) Collman, J.P.; Denisevich, P.; Konai, Y.; Marrocco, M.; Koval, C.; Anson, F.C. J. Am. Chem. Soc. 1980, 102, 6027-6036. (b) Collman, J.P.; Anson, F.C.; Bencosme, C.S.; Durand, R.R.; Kirch, R.P. J. Am. Chem. Soc. 1983.105, 2710-2718. (c) Collman, J.P.; Kim, K. J. Am. Chem. Soc. 1986, 108, 7847-7849. Evans, J.F.; Kuwana, T.; Henne, M.T.; Royer, G.P. J. Electroanal. Chem. 1977, 80, 409-416. (a) Chang, C.K.; Liu, H.-Y.; Abdalmuhdi, I. J. Am. Chem. Soc. 1984, 106, 2725-2726. (b) Liu, H.-Y.; Abfdalmuhdi, I.; Chang, C. K.; Anson, F.C. J. Phys. Chem. 1985, 89, 665-670. (a) Sullivan, B.P.; Meyer, T.J. J. Chem. Soc. Chem. Commun. 1984, 1244-1245. (b) O'Toole, T.R.; Margerum, L.D.; Westmoreland, T.D.; Vining, W.J.; Murray, R.W.; Meyer, T.J. J. Chem. Soc. Chem. Commun. 1985, 1416-1417. (a) Cabrera, C.R.; Abruna, H.D. J. Electroanal. Chem. 1986, 209, 101-107. (b) Guadalupe, A.R.; Usifer, D.A.; Potts, K.T.; Hurrell, H.C.; Mogstad, A.-E.; Abruna, H.D. J. Am. Chem. Soc. 1988, 110, 3462-3466. (a) Stalder, C.J.; Chao, S.; Summers, D.P.; Wrighton, M.S. J. Am. Chem. Soc. 1983, 105, 6318-6320. (b) Stalder, C.J.; Chao, S.; Wrighton, M.S. J. Am. Chem. Soc. 1984, 106, 3673-3675. Lieber, C.M.; Lewis, N.S. J. Am. Chem. Soc. 1984, 106, 5033- 5034. Cosnier, S.; Deronzier, A.; Moutet, J.-C. J. Electroanal. Chem. 1986.207, 315-321. Bettelheim, A.; Ozer, D.; Harth, R.; Murray, R.W. J. Electroanal. Chem. 1988.246, 139-154. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 224 Rault-Berthelot, J.; Mabon, G.; Sinonet, J. J. Electroanal. Chem. 1988.240, 355-359. Jiang, R; Dong, S. J. Electroanal. Chem. 1988, 246, 101-117. (a) Firth, B.E.; Miller, L.L.; Mitani, M.; Rogers, T.; Lennox, J.; Murray, R.W. J. Am. Chem. Soc. 1976,98, 8271-8272. (b) Firth, B.E.; Miller, L.L. J. Am. Chem. Soc. 1976, 98, 8272-8273. (a) Komori, T.; Nonaka, T. J. Am. Chem. Soc. 1983, 105, 5690- 5691. (b) Abe, S.; Nonaka, T.; Fuchigami, T. J. Am. Chem. Soc. 1983, 105, 3630-3632. (c) Komori, T.; Nonaka, T. J. Am. Chem. Soc. 1984, 106, 2656-2659. (a) Yamagishi, A.; Aramata, A. J. Chem. Soc. Chem. Commun. 1984, 452-453. (b) Yamagishi, A.; Aramata, A. J. Chem. Soc. Chem. Commun.1985, 449-452. Tse, D.C.S.; Kuwana, T. Anal. Chem. 1978,50, 1315-1318. Fijihira, M.; Yokozawa, A.; Kinoshita, H.; Osa, T. Chem. Lett. 1982, 1089-1092. Hazard, R.; Jaouannet, S.; Tallec, A. Tetrahedron 1982,38, 93- 102. Salmon, M.; Bidan, G. J. Electrochem. Soc. 1985, 132, 1897- 1899. (a) Landrum, H.C.; Salmon, R.T.; Hawkridge, F.M. J. Am. Chem. Soc. 1977, 99, 3154-3158. (b) Castner, J.F.; Hawkridge, F.M. J. Electroanal. Chem. 1983, 143, 217-232. 32. 33. 34. 35. 36. 37. 38. 39. 225 (a) Eddowes, M.J.; Hill, H.A.O.; J. Am. Chem. Soc. 1979, 101, 4461-4464. (b) Albery, W.J.; Eddowes, M.J.; Hill, H.A.O.; Hillman, A.R. J. Am. Chem. Soc. 1981, 103, 3904-3910. (c) Allen, P.M; Hill, H.A.O.; Walton, 0.]. J. Electroanal. Chem. 1984, 178, 69-85. (a) Chao, S.; Wrighton, M.S. J. Am. Chem. Soc.1987,109, 5886-5888. (b) Chao, S.; Simon, R.A.; Mallouk, T.E.; Wrighton, M.S. J. Am. Chem. Soc. 1988, 110, 2270-2276. (a) Taniguchi, I; Murakami, T; Toyosawa, K.; Yamaguchi, H.; Yasukouchi, K. J. Electroanal. Chem. 1982, 131, 397-401. (b) Taniguchi, I; Toyosawa, K.; Yamaguchi, H.; Yasukouchi, K. J. Electroanal. Chem. 1982, 140, 187-193. (a) Dhesi, R.; Cotton, T.M.; Timkovich, R. J. Electroanal. Chem. 1983, 154, 129-139. (b) Cotton, T.M.; Kaddi, D.; Iorga, D. J. Am. Chem. Soc.1983, 105, 7462-7464. Elliot, C.M.; Martin, W.S. J. Electroanal. Chem. 1982, I37, 377- 385. Ye, 1.; Baldwin, R.P. Anal. Chem. 1988, 60, 2263-2268. (a) White, B.A.; Murray, R.W. J. Am. Chem. Soc. 1987, 109, 2576-2581. (b) Oliver, B.N.; Egekeze, J.O.; Murray, R.W. J. Am. Chem. Soc.1988,110, 2321-2322. (a) Chidsey, C.E.D.; Murray, R.W. Science 1986,23], 25-31. (b) Reed, R.A.; Geng, L.; Murray, R.W. J. Electroanal. Chem. 1986.208, 185-193. (c) Geng, L.; Reed, R.A.; Longmire, M.; Murray, R.W. J. Phys. Chem. 1987, 91, 2908-2914. (d) Wilbourn, K.; Murray, R.W. J. Phys. Chem. 1988,92, 3642- 3648. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 226 (a) Wrighton, M.S. Science 1986,231, 32-37. (b) Chao, S.; Wrighton, M.S. J. Am. Chem. Soc.1987, 109, 2197-2199. (c) Natan, M.J.; Mallouk, T.E.; Wrighton, M.S. J. Phys. Chem. 1987, 91, 648-654. ((1) Shu., C.-F.; Wrighton, M.S. J. Phys. Chem. 1988, 92, 5221-5229. Fruge, D.R.; Fong, G.D.; Fong, F.K. J. Am. Chem. Soc. 1979, 101, 3694-3697. White, H.S.; Abruna, H.D.; Bard, A.J. J. Am. Chem. Soc.1982, 124, 265-271. (a) Noufi, R.N. J. Electrochem. Soc. 1983, 130, 2126-2128. (b) Rajeshwar, K; Kaneko, M.; Yamada, A.; Noufi, R.N. J. Phys. Chem. 1985, 89, 806-811. (a) Dubois, D.L.; Turner, LA. J. Am. Chem. Soc. 1982, 104, 4989-4990. (b) Dubois, D.L. Inorg. Chem. 1984,23, 2047- 2052. (a) Sammells, A.F.; Ang, P.G.P. J. Electrochem. Soc.1984, 131, 617-619. (b) Cook, R.L.; Sammells, A.F. J. Electrochem. Soc. 1985, 132, 2429-2431. Yamamoto, T. J. Chem. Soc. Chem. Commun. 1981, 187-188. Kaufman, J.H.; Chung, T.-C.; Heeger. A.J.; Wudl, F. J. Electrochem. Soc.1984, 131, 2091-2093. Simon, R.A.; Malleuk, T.E.; Daube, K.A.; Wrighton, M.S. Inorg. Chem. 1985, 24, 3119-3126. Hagemeister, M.P.; White, H.S. J. Phys. Chem. 1987, 91 , 150- 154. Holdcroft, S.; Morrison; Funt, L.B. J. Electroanal. Chem. 1988, 245, 191-199. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 227 Jasinski, RJ. J. Electrochem. Soc. 1977, 124, 637-641. Bruinink, J.; Kregting, C.G.A. J. Electrochem. Soc.1978, 125, 1397-1401. Cieslinski, R.C.; Armstrong, N.R. J. Electrochem. Soc. 1980, 127, 2605-2610. Reichman, B; Fan, F.-R.F.; Bard, A.J. J. Electrochem. Soc.1980, 127, 333-338. Akahoshi, H.; Toshima, S.; Itaya, K. J. Phys. Chem. 1981,85, 518-522. Garnier, F.; Tourillon, G.; Gazard, M.; Dubois, J.C. J. Electroanal. Chem. 1983, 148, 299-303. Thackeray, J.W.; White, H.S.; Wrighton, M.S. J. Phys. Chem. 1985,89, 5133-5140. (b) Smith, D.K.; Lane, G.A.; Wrighton, M.S. J. Phys. Chem. 1988, 92, 2616-2628. Viehbeck, A.; DeBerry, D.W. J. Electrochem. Soc.1985, 132, 1369-1375. Elliott, C.M.; Redepenning, 1.6. J. Electroanal. Chem. 1986, 197, 219-232. Yashima, H.; Kobayashi, M.; Lee, K.-B.; Chung, D.; Heeger, A.J.; Wudl, F. J. Electrochem. Soc. 1987, 134, 46-52. Sugimoto, T.; Nagatomi, T.; Ando, H.; Yoshida, Z.-I. Angew. Chem. Ed. Engl. 1988,27, 560-561. (a) Thackeray, J.W.; White, H.S.; Wrighton, M.S. J. Phys. Chem. 1985,89, 5133-5137. (b) Thackeray, J.W.; Wrighton, M.S. J. Phys. Chem. 1986,90, 6674-6679. (c) Chao, S.; Wrighton, M.S. J. Am. Chem. Soc.1987, 109, 6627-6631. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 228 Marrese, C.A.; Blubaugh, E.A.; Durst, R.A. J. Electroanal. Chem. 1988.243, 193-201. Kasem, K.K.; Abruna, H.D. J. Electroanal. Chem. 1988, 242, 87- 96. Wang 1.; Zadeii, J. Anal. Chim. acta. 1987, 188,187-194. (a) Lane, R.F.; Hubbard, A.T. J. Phys. Chem. 1973, 77, 1401- 1410. (b) Lane, R.F.; Hubbard, A.T. J. Phys. Chem. 1973, 77, 1411-1421. Moses, P.R.; Wier, L.; Murray, R.W. Anal. Chem. 1975,47, 1882-1886. Wrighton, M.S.; Austin, R.G.; Bocarsly, A.B.; Bolts, J.M.; Haas, 0.; Legg, K.D.; Nadjo, L.; Palazzotto, M.C. J. Electroanal. Chem. 1978,87, 429-433. Wrighton, M.S.; Palazzotto, M.C.; Bocarsly, A.B.; Bolts, J.M.; Fischer, A.B.; Nadjo, L. J. Am. Chem. Soc. 1978, 100, 7264- 7271. Bolts, J.M.; Wrighton, M.S. J. Am. Chem. Soc.1978, 100, 5257- 5262. Bolts, J.M.; Bocarsly, A.B.; Palazzotto, M.C.; Walton, E.G.; Lewis, N.S.; Wrighton, M.S. J. Am. Chem. Soc. 1979,10], 1378- 1385. Bocarsly, A.B.; Walton, E.G.; Bradly, M.G.; Wrighton, M.S. J. Electroanal. Chem. 1979, 100, 283-306. Wrighton, M.S.; Bolts, J.M.; Bocarsly, A.B.; Palazzotto, M.C.; Walton, E.G. J. Vac. Sci. Technol.1978, 15, 1429-1433. Fisher, A.B.; Kinney, J.B.; Staley, R.H.; Wrighton, M.S. J. Am. Chem. Soc. 1979,10], 7863-7864. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 229 Nakahama, S.; Murray, R.W. J. Electroanal. Chem. 1983, 158, 303-322. Simon, R.A.; Ricco, A.J.; Wrighton, M.S. J. Am. Chem. Soc. 1982, 104, 2031-3034. Ghosh, P.; Spiro, T.G. J. Am. Chem. Soc. 1980, 102, 5543-5549. Ghosh, P.; Spiro, T.G. J. Electrochem. Soc. 1981, 128, 1281- 1287. Moses, P.R.; Murray, R.W. J. Electroanal. Chem. 1977, 77, 393- 399. Murray, R.W.; Leyden, D.E.; Collins, W. (Ed.) "Silayted Surfaces" Gordon and Breach: New York, 1980. Finklea, H.C.; Abruna, H.D.; Murray, R.W. Adv. Chem. Ser. 1980, 184, 253-256. Diaz, A. J. Am. Chem. Soc. 1977, 99, 5838-5840. Abruna, H.D.; Meyer, T.J.; Murray, R.W. Inorg. Chem. 1979, I], 3233-3240. Abruna, H.D.; Walsh, J.L.; Meyer, T.J.; Inorg. Chem. 1981,20, 1481-1486. Kuo, K.; Moses, P.R.; Lenhard, J.R.; Green, D.C.; Murray, R.W. Anal. Chem. 1979,51, 745-748. Diaz, A.F.; Kanazawa, K.K. IBM J. Res. Dev. 1979, 23, 316-321. Moses, P.R.; Murray, R.W. J. Am. Chem. Soc. 1976,98, 7435- 7436. Watkins, B.F.; Behling, J.R.; Kariv, 13.; Miller, L.L. J. Am. Chem. Soc. 1975,97, 3549-3550. Umana, M.; Rolison, R.; Nowak, R.; Daum, P.; Murray, R.W. Surf. Sci. 1980, 10], 295-310. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 230 Fujihara, M.; Matsue, T.; Osa, T. Chem. Lett. 1977, 361-366. Koval, C.A.; Anson, F.C. Anal. Chem. 1978,50, 223-229. Lennox, J.C.; Murray, R.W. J. Electroanal. Chem. 1977, 78, 395- 401. Lenhard, J.R.; Rocklin, R.; Abruna, H.; Willman, Kuo, K., Nowak, R.; Murray, R.W. J. Am. Chem. Soc. 1978, 100, 5213-5215. Rocklin, R.D.; Murray, R.W. J. Electroanal. Chem. 1979, 100, 271-282. Oyama. N.; Anson, F.C. J. Am. Chem. Soc. 1979, 10], 1634- 1635. Kuwana, T. J. Electroanal. Chem. 1978,88, 299-303. Tse, D.C.S.; Kuwana, T.; Royer, G.P. J. Electroanal. Chem. 1979, 98. 345-353. (a) Miller, L.L.; Van De Mark, M.R. J. Am. Chem. Soc. 1978, 100, 639-640. (b) Van De Mark, M.R.; Miller, L.L. J. Am. Chem. Soc. 1978, 100, 3223-3225. (c) Miller, L.L.; Van De Mark, M.R. J. Electroanal. Chem. 1978, 88, 437-440. Doblhofer, K.; Nolte, D.; Ulstrup, J. Ber. Bunsenges. Phys. Chem. 1984,88. 345-351. Pham, M.-C.; Lacaze, P.C.; Dubois, LE. J. Electroanal. Chem. 1978,86, 147-157. (a) Merz, A.; Bard, A.J. J. Am. Chem. Soc. 1978, 100. 3222- 3223. (b) Itaya, K.; Bard, A.J. Anal. Chem. 1978, 50, 1487- 1489. (a) Tachikawa, H.; Faulkner, L.R. J. Am. Chem. Soc. 1978, 100, 4379-4385. (b) Tachikawa, H.; Faulkner, L.R. J. Am. Chem. Soc. 1978.100. 8025-8026. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 231 (a) Nowak, R.; Schultz, F.A.; Umana, M.; Abruna, H.D.; Murray, R.W. J. Electroanal. Chem. 1978, 94, 219-225. (b) Murray, R.W. ”Symposium on Silated Surfaces,” Midland, MI. May, 1978. Dautartas, M.F.; Evans, J.F.; Kuwana Anal. Chem. 1978.5], 104-110. Kerr, 1.B.; Miller, L.L.; Van De Mark, M.R. J. Am. Chem. Soc. 1980, 102, 3383-3390. Rocklin, R.D.; Murray, R.W. J. Phys. Chem. 1981.85, 2104- 2112. Oyama, N.; Anson, F.C. J. Am. Chem. Soc. 1979, 101 , 739-741. Oyama, N.; Shigehara, K.; Anson, F.C. Inorg. Chem. 1981,20, 518-522. Schneider, 1.; Murray, R.W. Anal. Chem. 1982,54, 1508-1515. (a) Dautartas, M.F.; Evans. J.F. J. Electroanal. Chem. 1980, 109, 301-312. (b) Dautartas, M.F.; Mann, K.R.; Evans, J.F. J. Electroanal. Chem.1980,]]0, 379-386. (a) Rolison, D.R.; Umana, M.; Burgmayer, P.; Murray, R.W. Inorg. Chem. 1981, 20. 2996-3002. (b) Facci, 1.; Murray, R.W. Anal. Chem. 1982,54, 772-777. Heider. G.H.; Gelbert, M.B.; Yacynych, A.M. Anal. Chem. 1982, 54, 322-324. Rolison, D.R.; Unana, M.; Burgmayer, P; Murray. R.W. Inorg. Chem. 1981, 20, 2996-3002. Abruna, H.D.; Denisevich, P; Umana, M.; Meyer. T.J., Murray, R.W. J. Am. Chem. Soc. 1981, 103, 1-5. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 232 (a) Diaz, A.F.; Logan. 1.A. J. Electroanal. Chem. 1980. III, 111- 114. (b) Kanazawa, K.K.; Diaz, A.F.; Geiss. R.H.; Gill. W.D.; Kwak, J.F.; Logan. 1.A.; Rabolt, J.F.; Street, G.B. J. Chem. Soc. Chem. Commun.1979, 854-855. Pham, M.-C.; Lacaze, P.-C.; Dubois, 1.-E. J. Electroanal. Chem.1979, 99. 331-340. Kerr, 1.B.; Miller, L.L. J. Electroanal. Chem. 1979. 101, 263- 267. Fuki, M.; Degrand. C.; Miller, L.L. J. Am. Chem. Soc. 1982, 104, 28-33. Daum, P.; Murray, R.W. J. Phys. Chem. 1981, 85, 389-396. Schroeder, A.H.; Kaufman, F.B.; Patel, V.; Engler, E.M. J. Electroanal. Chem. 1980. 113, 193-208. Takiguchi, T.; Nonaka, T.; Fuchigami, TJ. Electroanal. Chem. 1985.195, 177-181. Willman. K.W.; Murray. R.W. J. Electroanal. Chem. 1982, 133, 211-231. Fan, F.-R.F.; Mau, A.; Bard, A.J. Chem. Phys. Lett. 1985, 116, 400-404. (a) Ohsaka, T.; Sato, K.; Matsuda, H.; Oyama, N. J. Electrochem. Soc. 1985.132, 1871-1879. (b) Oyama, N., Ohsaka, T.; Yamamoto, H.; Kaneko, M. J. Phys. Chem. 1986.90. 3850- 3856. Burgmayer, P.; Murray, R.W. J. Electroanal. Chem. 1982.135, 335-342. Mcquillan, 1.A.; Reid, M.R. J. Electroanal. Chem. 1985, 194, 237-245. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 233 Mengoli, G.; Muscani, M.M.; Zotti, G. J. Electroanal. Chem. 1984, 175, 93-104. Diaz. A.F.; Martinez. A.; Kanazawa, K.K. J. Electroanal. Chem. 1981.130. 181-187. Waltman. R.1.; Bargon, 1.; Diaz A.F. J. Phys. Chem. 1983, 87, 1459-1463. Tourillon, G.; Garnier, F. J. Electroanal. Chem. 1984, 161, 407- 414. Oyama, N.; Ohsaka, T.; Shimizu, T. Anal. Chem. l985,57. 1526- 1532. Ohsaka. T.; Okajima, T.; Oyama. N. J. Electroanal. Chem. 1986. 215, 191-207. Lewis, A.; White, H.S.; Wrighton, M.S. J. Am. Chem. Soc. 1984, 106, 6947-6952. Gaudiello, 1.G.; Ghosh, P.K.; Bard. A.J. J. Am. Chem. Soc. 1985, 107, 3027-3032. Scott, N.S.; Oyama, N.; Anson, F.C. J. Electroanal. Chem. 1980, 110, 303-310. Wan, G.-X.; Shigehara, F..; Tsuchida, E.; Anson, F.C. J. Electroanal. Chem. 1984, 179. 239-250. Calvert, J.M.; Meyer. T.J. Inorg. Chem. 1981, 20, 27-33. Potts, K.T.; Usifer, D.; Guadalupe, A.; Abruna, H.D. J. Am. Chem. Soc. 1987, 109, 3961-3967. Ellis, C.D.; Margerum, L.D.; Murray. R.W.; Meyer, T.J. Inorg. Chem. 1983. 22, 1283-1291. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 234 (a) Bidan, G.; Deronzier, A.; Moutet, 1.-C. J. Chem. Soc. Chem. Commun.1984, 1185-1186. (b) Cosiner, S.; Deronzier, A.; Moutet, 1.-C. J. Phys. Chem. 1985. 89, 4895-4897. Eaves, 1.G.;Munro. H.S.; Parker. D. J. Chem. Soc. Chem. Commun. 1985. 684-685. Margerum, L.D.; Meyer, T.1 .; Murray, R.W. J. Electroanal. Chem. 1983, 149, 279-285. (a) Ikeda, T.; Leidner, C.R.; Murray, R.W. J. Am. Chem. Soc. 1981, 103, 7422-7424. (b) Pickup, P.G.; Kutner, W.; Leidner, C.R.; Murray, R.W. J. Am. Chem. Soc. 1984, 106, 1991-1998. (c) Leidner, C.R.; Murray, R.W. J. Am. Chem. Soc. 1985, 107, 551-556. Vining, W.J.; Surridge, N.A.; Meyer, T.J. J. Phys. Chem. 1986, 90, 2281-2283. Guarr, T.F.; Anson, F.C. J. Phys. Chem. 1987, 9]. 4037-4043. Oyama, N.; Anson, F.C. J. Electrochem. Soc.1980, 127, 247- 250. (a) Oyama, N; Shimomura, K.; Shigehara, K.; Anson, F.C. J. Electroanal. Chem. 1980, 112, 271-280. Oyama. N; Ohsaka, T.; Ushirogouchi, T. J. Phys. Chem. 1984, 88, 5274-5280. Niwa, K.; Doblhofer, K. Electrochim. Acta. 1986, 3], 553-549. (a) Ohsaka, T.; Anson, F.C. J. Phys. Chem. 1983, 87, 640-647. (b) Anson, F.C.; Ohsaka, T.; Saveant. 1.-M. J. Am. Chem. Soc. 1985, 107, 4883-4892. Sumi,K.; Anson, F.C. J. Phys. Chem. 1986. 20, 3845-3850. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 235 Crouch. A.M.; Ordonez, 1.; Langford, C.H.; Lawrence, M.F. J. Phys. Chem. 1988. 92, 6058-6065. (a) Henning, T.P.; White, H.S.; Bard, A.J. J. Am. Chem. Soc. 1981. 103. 3937-3938. (b) Krishnan. M.; White, 1.R.; Fox, M.A.; Bard, A.J. J. Am. Chem. Soc. 1983, 105, 7002-7003. (a) Buttry, D.A.; Saveant. 1.-M.; Anson, F.A. J. Phys. Chem. 1984,88, 3086-3091. (b) McHatton, R.C.; Anson, F.C. Inorg. Chem. 1984. 23, 3935-3942. (a) Martin C.R. Anal. Chem. 1982, 54, 1639-1641. (b) Penner. R.M.; Martin, C.R. J. Electrochem. Soc. 1985, 132. 514-515. Lieber, C.M.; Schmidt, M.H.; Lewis, N.S. J. Am. Chem. Soc. 1986, 108. 6103-6108. (a) Majda. M; Faulkner, L.R. J. Electroanal. Chem. 1984, 169, 97-112. (b) Jones. E.T.T.; Faulkner, L.R. J. Electroanal. Chem. 1987.222, 201-222. Vining. W.J.; Meyer, T.J. Inorg. Chem. 1986, 25. 2023-2033. Nowak, R.1.; Schultz, F.A.; Umana, M.; Lam, R.; Murray, R.W. Anal. Chem. 1980,52, 315-321. Oyama. N.; Anson. F.C. J. Electroanal. Chem. 1980, 127, 640- 647. Daum, P.; Murray, R.W. J. Electroanal. Chem. 1979, 103, 289- 294. Kaufman, F.B.; Engler, E.M. J. Am. Chem. Soc. 1979. 10], 547- 549. Peerce, P.J.; Bard. A.J. J. Electroanal. Chem. 1980. 114, 89-115. Laviron, E. J. Electroanal. Chem. 1980, 112, 1-9. 165. 166. 167. 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 236 Andrieux, C.P.; Saveant, 1.M. J. Electroanal. Chem. 1980, 1]], 377-381. (a) Ikeda, T.; Leidner, C.R.; Murray. R.W. J. Electroanal. Chem. 1982. 138, 343-365. (b) Facci, 1.8.; Schmehl, R.H.; Murray, R.W. J. Am. Chem. Soc. 1982, 104, 4959-4960. Shigehara, K.; Oyama, N.; Anson, F.C. J. Am. Chem. Soc. 1981. 103, 2552-2558. Lenhard, 1.R.; Murray. R.W. J. Am. Chem. Soc. 1978.100, 7870-7875. Peerce, P.1.; Bard, A.J. J. Electroanal. Chem. 1980, 112, 97-115. Angerstein-Kozlowska, H.; Klinger, 1.; Conway, B.E. J. Electroanal. Chem. 1977. 75, 45-60. (a) Brown, A.P.; Anson, F.C. J. Electroanal. Chem. 1978. 92, 133-145. Wrighton, M.S.; Austin, R.G.; Bocarsly, A.B.; Bolts, J.M.; Haas, 0.; Legg, K.D.; Nadjo, L.; Palazzotto, M.C. J. Am. Chem. Soc.1978, 100, 1602-1603. Sharp, M.; Peterson. M.; Edstrom, K. J. Electroanal. Chem. 1979,95, 123-130. Deronzier, A.; Latour. J-M J. Electroanal. Chem. 1987.224, 295-301. Oyama, N.; Anson, F.C. J. Am. Chem. Soc.1979. 10], 3450- 3456. Kaufman, F.B.; Schroeder, A.H.; Engler E.M.; Kramer, S.R.; Chambers, 1.Q. J. Am. Chem. Soc. 1980, 102, 483-488. Denisevich, P.; Abruna, H.D.; Leidner, C.R.; Meyer, T.J.; Murray. R.W. Inorg. Chem. 1982.21, 2153-2161. 178. 179. 180. 181. 182. 183. 184. 185. 186. 187. 188. 189. 190. 191. 237 Leddy, 1.; Bard, A.J.; Maloy. 1.T.; Saveant. J.M. J. Electroanal. Chem. 1985. 187, 205-227. Oyama, N.; Ohasaka, T.; Yamamoto, H.; Kaneko, M. J. Phys. Chem. 1986, 90, 3850-3856. (a) Chambers. 1.Q.; Inzelt, G. Anal. Chem. 1985.57, 1117- 1121. (b) Inzelt, G.; Bacskai. 1.; Chambers. 1.Q.; Day, R.W. J. Electroanal. Chem. 1986,20], 301-314. Peerce, P.1.; Bard, A.J. J. Electroanal. Chem. 1980, 108, 121- 125. Samuels, G.T.; Meyer, T.J. J. Am. Chem. Soc. 1981. 103, 307- 312. Laviron, E. J. Electroanal. Chem. 1979. 105, 25-35. Kaufman, F.B.; Schroeder, A.H.; Engler, E.M.; Kramer, S.R.; Chambers. 1.Q. J. Am. Chem. Soc. 1980. 102, 483-488. Henning, T.P.; Bard, A.J. J. Electrochem. Soc.1983. 130, 613- 621. (a) Brown, A.P.; Anson, F.C. Anal. Chem. 1977, 49, 1589-1595. (b) Brown, A.P.; Koval, C.; Anson, F.C. J. Electroanal. Chem. 1976, 72. 379-387. Smith, D.F.; Willman, K.; Kuo, K.; Murray. R.W. J. Electroanal. Chem. 1979, 95, 217-227. Kuo, K.N.; Murray. R.W. J. Electroanal. Chem. 1982, I31, 37- 60. Laviron, E. J. Electroanal. Chem. 1980, 112, 11-23. Faci, 1.; Murray, R.W. J. Phys. Chem. 1981. 85, 2870-2873. Dahms, H. J. Phys. Chem. 1968, 72, 362-364. 192. 193. 194. 195. 196. 197. 198. 199. 200. 201. 202. 238 Laviron, E.; Roullier. L.; Degrand, C. J. Electroanal. Chem. 1980, 112, 11-23. (a) Andrieux, D.C.; Saveant, J.M. J. Electroanal. Chem. 1978. 93. 163-168. (b) Andrieux, D.C.; Dumas-Boouchait, J.M.; Saveant. J.M. J. Electroanal. Chem. 1980, 114, 159-163. (a) Ruff, I. Electrochim. Acta. 1970. 15, 1059-1061. (b) Ruff. 1.; Korosi-Odor, I. Inorg. Chem. 1970, 9. 186-188. (c) Ruff, 1.; Friedrich. V. J. Phys. Chem. 1971. 75, 3297-3309. Buttry, D.A.; Anson, F.C. J. Electroanal. Chem. 1981, 85, 2870- 2873. Facci, 1.; Murray, R.W.; J. Phys. Chem. 1981, 85, 2870-2873. White, H.S.; Leddy, 1.; Bard. A.J. J. Am. Chem. Soc.1982, 104, 4811-4817. (b) Martin, C.R.; Rubenstein, 1.; Bard, A.J. J. Am. Chem. Soc.1982, 104, 4817-4824. Buttry, D.A.. Anson, F.C. J. Am. Chem. Soc. 1983, 23, 685-689. Anson. F.C.; Saveant, J.M.; Shigehara, K. J. Phys. Chem. 1983. 87. 214-219. (a) Andrieux, C.P.; Dumas-Bouchiat, J.M.; Saveant, J.M. J. Electroanal. Chem. 1982,13], 1-35. (b) Andrieu. C.P.; Saveant, J.M. J. Electroanal. Chem. 1982.134, 163-166. (c) Andrieu. C.P.; Saveant, J.M. J. Electroanal. Chem. 1982, I42, 1- 30. Anson, F.C.; Saveant, J.M.; Shigehara, K. J. Am. Chem. Soc. 1983.105, 1096-1106. (a) Andrieux, C.P.; Dumas-Bouchiat, J.M.; Saveant. J.M. J. Electroanal. Chem. 1981. 123, 189-243. (b) Andrieux, C.P.; Saveant, J.M. J. Electroanal. Chem. 1984, 17], 65-93. 203. 204. 205. 206. 207. 208. 209. 210. 211. 212. 213. 214. 215. 239 (a) Oyama, N.; Anson, F.C. Anal. Chem. 1980.52, 1192-1198. (b) Tsou, Y.M.; Anson, F.C. J. Phys. Chem. 1985, 89. 3818- 3823. Leddy. 1.; Bard, A.J. J. Electroanal. Chem. 1985. 189, 203-219. Feldman, 8.1.; Ewing, A.G.; Murray, R.W. J. Electroanal. Chem. 1985. 194. 63-81. Ohsaka, T.; Okajima, T.; Oyama, N. J. Electroanal. Chem. 1986, 215, 191-207. Sharp, M. J. Electroanal. Chem. 1987,230, 109-124. Facci, 1.; Murray, R.W. J. Electroanal. Chem. 1981, 124, 339- 343. (a) Rubinstein, I; Bard, A.J. J. Am. Chem. Soc.1980,102, 6641-6642. (b) Kaifer, A.B.; Bard, A.J. J. Phys. Chem. 1986, 90, 868-873. Montgomery. D.D.; Anson, F.C. J. Am. Chem. Soc.1985. 107, 3431-3436. (a) Oyama, N.; Ohsaka, T.; Kaneko, M.; Sato. K.; Matsuda, H. J. Am. Chem. Soc. 1983. 105. 6003-6008. (b) Ohsaka, T.; Okajima, T.; Oyama. N. J. Electroanal. Chem. 1986.206, 191- 207. Doblhofer, K.; Braun, H.; Lange, R. J. Electroanal. Chem. 1986. 206, 93-100. Lindholm. B. J. Electroanal. Chem. 1988.250. 341-354. Majda. M; Faulkner. L.R. J. Electroanal. Chem. 1984, 169, 77- 95. Chen, X.; Peixin, H.; Faulkner, L.R. J. Electroanal. Chem. 1987, 222, 223-242. 240 216. (a) Stalnaker, N.D.; Solenberger, 1.C.; Wahl, A.C. J. Phys. Chem. 1977,81, 601-605. (b) Burstall. F.H.; Nyholm. R.S. J. Chem. Soc. 1952, 3570-3579. 217. Shilt, A.A. J. Am. Chem. Soc.1960, 82, 3000-3005. 218. Dwyer, F.P.; Hogarth. 1.W. Inorg. Synth.1957,5, 206-207. 219. Dorman, W.C.; McCarley, R.E. Inorg. Chem. 1974, 13, 491-493. 220. Bookbinder, D.C.; Wrighton, M.S. J. Electrochem. Soc.1983, 130, 1080-1087. 221. Abruna, H.D.; Breikss, D.B.; Collum, D.B. Inorg. Chem. 1985, 24, 987-988. 222. Barrer, R.M.; Jones. D.L. J. Chem. Soc. A. 1970, 1531-1537. 223. Daum, P.; Lenhard, 1.R.; Rolison, D.R.; Murray, R.W. J. Am. Chem. Soc. 1980.102, 4649-4653. 224. Oyama. N.; Anson F.C. J. Electrochem. Soc. 1980, 127, 640-647. 225. Adams, R.N. ”Electrochemistry at Solid Electrodes"; M. Dekker: New York, 1969. 226. Ghosh, P.K.; Bard, A.J. J. Am. Chem. Soc.1983, 105, 5691- 5693. 227. Ege. D.; Ghosh, P.K.; White. 1.R.; Equey, 1.-F.; Bard. A.J. J. Am. Chem. Soc.1985, 107. 5644-5652. 228. Liu, H.Y.; Anson, F.C. J. Electroanal. Chem. 1985, 184, 411-417. 229. King, R.D.; Nocera, D.G.; Pinnavaia, T.J. J. Electroanal. Chem. 1987, 236, 43-53. 230. (a) Fitch, A.; Lavy-Feder, A.; Lee, S.A.; Kirsh, M.T. J. Phys. Chem. 1988,92, 6665-6670. (b) Fitch, A.; Faust. C.L. J. Electroanal. Chem. 1988.257, 299-303. 231. 232. 233. 234. 235. 236. 237. 238. 239. 240. 241. 241 Inoue, H.; Haga, S.; Iwakura, C.; Yoneyama, H. J. Electroanal. Chem. 1988, 249, 133-141. Itaya, K.; Chang. H.-C.; Uchida, I. Inorg. Chem. 1987, 26, 624- 626. Shaw, B.R.; Creasy, K.E. J. Electroanal. Chem. 1988.243. 209- 217. (a) Hernandez, L.; Hernandez, P.; Sosa, Z. Z. Anal. Chem. 1988. 33]. 525-527. (b) Hernandez, L.; Hernandez, P.; Sosa Ferrera, Z. Z. Anal. Chem. 1988,33]. 756-759. (c) Hernandez, L.; Hernandez, P.; Lorenzo, B.; Sosa Ferrera, Z. Analyst. 1988, 113, 621-623. Hernandez, L.; Hernandez, P.; Blanco, M.H.; Lorenzo, E.; Alda, E. submitted to Z. Anal. Chem. (a) Miller, C.1.; Majda. M. J. Electroanal. Chem. 1986.207. 49- 72. (b) Miller. C.1.; Majda, M. Anal. Chem. 1988. 60, 1168- 1176. Miller, C.1.; Widrig. C.A.; Charych, D.H.; Majda, M. J. Phys. Chem. 1988,92, 1928-1936. Lundgren, L.A.; Murray, R.W. J. Electroanal. Chem. 1988.227, 287-294. Murray. C.G.; Nowak, R.1.; Rolison, D.R. J. Electroanal. Chem. 1984, 164, 205-210. De-Vismes, B.; Bedioui, F.; Devynck, 1.; Bied-Charreton, C. J. Electroanal. Chem. 1985, 187, 197-202. (a) Gemborys, H.A.; Shaw, B.R. J. Electroanal. Chem. 1986.206, 95-107. (b) Sargeant, 1.A.; Tirhado, M. J. Electrochem. Soc. 1988, 135, 869-876. 242. 243. 244. 245. 246. 247. 248. 249. 250. 251. 252. 253. 242 Murr, E.N.; Kerkeni. M.; Sellami, A.; Taarit, Y.B. J. Electroanal. Chem. 1988, 246, 461-465. (a) Li, Z.; Mallouk. T.E. J. Phys. Chem. 1987, 91, 643-648. (b) Li. Z.; Wang, C.M.; Persaud. L.; Mallouk. T.E. J. Phys. Chem. 1988, 92, 2592-2597. Lee, H.; Kepley, L.1.; Hong, H.-G.; Akhter. S.; Mallouk. T.E. J. Phys. Chem. 1988,92, 2597-2601. (a) Keita. B.; Nadjo, L.; Haeussler, 1.P. J. Electroanal. Chem. 1988.243, 481-491. (b) Keita. B.; Nadjo, L. J. Electroanal. Chem. 1988, 240, 325-332. Renneke, R.F.; Hill, C.L. J. Am. Chem. Soc. 1986, 108. 3528- 3529. Mizumo, N.; Watanabe, T.; Misono, M. J. Phys. Chem. 1985, 89, 80-85. (a) Kulesza, P.1.; Faulkner, L.R. J. Am. Chem. Soc. 1988, 110. 4905-4913. (b) Kulesza, P.1.; Faulkner, L.R. J. Electroanal. Chem. 1988, 248, 305-320. Itaya, K.; Uchida, 1.; Neff, V.D. Acc. Chem. Res. 1986, 19, 162- 168. Thomas, J.M.; Whittingham, M.S. and Jacobson, A.J. (Eds) "Intercalation Chemistry" Academic Press: New York, 1982, Chpt. 3. Fripiat, 1.1.; 1e11, A.N.; Poncelot. G.; Andre. J. J. Phys. Chem. 1965,69, 2185-2197. Pinnavaia, T.J. Science 1983.220, 365-371. Yamagishi, A.; Aramata. A. J. Electroanal. Chem. 1985,] 9], 449-452. 254. 255. 256. 257. 258. 259. 260. 261. 262. 263. 264. 265. 243 (a) Kanbara, T.; Yamamoto, T.; Tokuda, K.; Aoki, K. Chem. Lett. 1987, 2173-2176. (b) Yamamoto, T.; Kanbara, T. Inorg. Chim. Acta.1988,142. 191-193. (c) Kanbara, T.; Yamamoto, T. Denki. Kagaku. Oyobi. Kogyo. Butsuri. Kagaku. 1989.57, 91- 92. Aramata, A.; Yamagishi, A. Denki. Kagaku. Oyobi. Kogyo. Butsuri. Kagaku. 1987,55, 422-427. Ghosh, P.K.; Mau-A.W.-H.; Bard. A.J. J. Electroanal. Chem. 1984. 169, 315-317. (a) Rusling. J.F.; Shi, C.-N.; Suib, S.L. J. Electroanal. Chem. 1988, 245. 331-337. (b) Shi, C.; Rusling. J.F.; Wang. Z.; Willis, W.S.; Winiecki, A.M.; Suib, S.L. Langmuir 1989.5, 650-660. Kamat, P.V. J. Electroanal. Chem. 1984. 163. 389-394. (a) Fan, F.-R.F.; Liu. H.-Y.; Bard, A.J. J. Phys. Chem. 1985, 89. 4418-4420. (b) Enea. O; Bard, A.J. J. Phys. Chem. 1986. 90, 301-306. Kotkar, D.; Joshi, V.; Ghosh. P.K. Proc. Indian Natl. Acad. A 1986,52, 736-743. Rudzinski, W.E.; Bard, A.J. J. Electroanal. Chem. 1986, 199, 323-340. Buttry. D.A.; Anson, F.C. J. Am. Chem. Soc. 1984, 106, 59-64. Chan, M.-S.; Wahl, A.C. J. Phys. Chem. 1978, 82, 2542-2549. Bard A.J.; Faulkner, L.R. "Electrochemical Methods. Fundamentals and Applications"; Wiley: New York, 1980, pg 431. Itaya, K.; Bard, A.J. J. Phys. Chem. 1985, 89, 5565-5568. 266. 267. 268. 269. 270. 271. 272. 273. 274. 275. 276. 277. 278. 244 Traynor, M.F.; Mortland, M.M.; Pinnavaia, T.J. Clays Clay Miner. 1978.26, 318-324. Loeppert, R.H.; Mortland, M.M.; Pinnavaia. T.J. Clays Clay Miner. 1979,27, 201-208. Kadkhodayan, A.; Pinnavaia, T.J. J. Mol. Catal. 1983. 21, 109- 112. Cebula, D.J.; Thomas, R.K.; Middleton. S.; Ottewill, R.H.; White. 1.W. Clays Clay Miner. 1979,27, 39-52. Van Olphen, H. "An Introduction to Clay Colloid Chemistry"; Wiley: New York, 1977, pg 27. White. 1.R.; Bard, A.J. J. Electroanal. Chem. 1986, I97. 233- 244. Carter. M.T.; Bard. A.J. J. Electroanal. Chem. 1987.229. 191- 214. (a) Fan, F.-R.F.; Bard. A.J. J. Electrochem. Soc. 1986. 133. 301- 304. (b) Castro-Acuna, C.M.; Fan, F.-R.F.; Bard, A.J. J. Electroanal. Chem.1987,234, 347-353. Rudzinski, W.E.; Figueroa, C.; Hoppe, C.; Kuromoto, T.Y.; Root, D. J. Electroanal. Chem. 1988.243, 367-378. Inoue, H.; Yoneyama, H. J. Electroanal. Chem. 1987 , 233, 291- 294. Calvert. J.M.; Schmehl, R.H.; Sullivan, B.P.; Facci. 1.8.; Meyer, T.J.; Murray. R.W. Inorg. Chem. 1983,22, 2151-2162. Denisevich, P.; Willman, K.; Murray, R.W. J. Am. Chem. Soc. 1981, 103, 4727-4737. Oyama, N.; Anson, F.C. J. Electroanal. Chem. 1986, 199, 467- 470. 279. 280. 281. 282. 283. 284. 285. 286. 287. 288. 289. 290. 291. 292. 293. 245 Oyama, N. Appl. Chem. Resour.1987.29, 138-141. Faulkner, L.R. Int. Rev. Sci: Phys. Chem., Ser. Two 1975, 9, 213-263. Pighin. A. Can. J. Chem. 1973,51, 3467-3472. Itoh, K.; Honda. K.; Sukigara, M. J. Electroanal. Chem. 1980, 110, 277-284. Turro, N.1. "Modern Molecular Photochemistry"; Benjamin/Cummings: Menlo Park, CA, 1978. (a) Tokel, N.E.; Bard. A.J. J. Am. Chem. Soc. 1972, 94, 2862- 2863. (b) Wheeler, B.L.; Nagasubramanian, G.; Bard, A.J.; Schechtman, L.A.; Dininny, D.R.; Kenney. M.E. J. Am. Chem. Soc. 1984, 106, 7404-7410. Abruna, H.D. J. Electrochem. Soc. 1985, 132, 842-849. Luong. J.C.; Nadjo, L.; Wrighton, M.S. J. Am. Chem. Soc. 1978. 100, 5790-5785. . Vogler, A.; Kunkely, H. Angew. Chem. Int. Ed. Engl.1984,23, 316-317. Ouyang. 1.; Zietlow, T.C.; Hopkins. M.D.; Fan. F.-R.F.; Gray. H.B.; Bard, A.J. J. Phys. Chem. 1986, 90, 3841-3844. Bonafede. S.; Ciano. M.; Bolletta, F.; Balzani, V.; Chassot, L.; Zelewsky, A. J. Phys. Chem. 1986, 90, 3836-3841. Wallace, W.L.; Bard, A.J. J. Phys. Chem. 1979, 83, 1350-1357. Itoh, K.; Honda. K. Chem. Lett. 1979, 99-102. Kim, 1.; Fan, F.-R.F.; Bard, A.J.; Che, C.-M.; Gray, H.B. Chem. Phys. Lett. 1985, 121 543-546. Che, C.-M.; Atherton. SJ.; Butler, L.G.; Gray, H.B. J. Am. Chem. Soc. 1984, 106, 5143-5145. 294. 295. 296. 297. 298. 299. 300. 301. 302. 303 304. 305. 306. 246 Liu, D.K.; Brunschwig, BS; Creutz, C.; Sutin, N. J. Am. Chem. Soc. 1986.108. 1749-1755. Mussell, R.D.; Nocera, D.G. J. Am. Chem. Soc.. 1988, 110, 2764- 2772. Rubinstein, 1.; Bard, A.J. J. Am. Chem. Soc.1980, 102, 6641- 6642. Rubinstein, 1.; Bard, A.J. J. Am. Chem. Soc.. 1981, 103. 5007- 5012 Buttry, D.A.; Anson, F.C. J. Am. Chem. Soc.. 1982, 104, 4824- 4829. Ghosh, P.K.; Bard, A.J. J. Electroanal. Chem. 1984, I69, 113- 128. Xun, Z.; Bard, A.J. J. Phys. Chem. 1988, 92, 5566-5569. Abruna, H.D.; Bard, A.J. J. Am. Chem. Soc.. 1982, 104, 2641- 2642. Fan, F.-R.F.; Man, A.; Bard. A.J. Chem. Phys. Lett. 1985, 116, 400-404. Anson. F.C.; Ni. C.L.; Saveant, J.M. J. Am. Chem. Soc. 1985, 107, 3442-3450. Fan, F.-R.F.; Reichman, B.; Bard, A.J. J. Am. Chem. Soc.. 1980, 102, 1488-1492. Heller, A.; Miller, B.; Lewerenz, H.1.; Bachmann, K.J. J. Am. Chem. Soc.. 1980.102, 6555-6556. Bookbinder, D.C.; Bruce, 1.A.; Dominey. R.N.; Lewis, N.S.; Wrighton, M.S. Proc. Natl. Acad. Sci. UAS 1980.77, 6280— 6284. 307. 308. 309. 310. 311. 312. 313. 314. 315. 316. 317. 318. 319. 320. 247 Dominey. R.N.; Lewis, N.S.; Bruce. 1.A.; Bookbinder, D.C.; Wrighton, M.S. J. Am. Chem. Soc.. 1982, 104, 467-482. Bruce. 1.A.; Murahashi, T. Wrighton, M.S. J. Phys. Chem. 1982, 86. 1552-1563. Dabestani, R.; Wang, X.; Bard. A.J.; Campion. A.; Fox, M.A.; Webber, S.E.; White. J.M. J. Phys. Chem. 1986, 90, 2729-2732. Shimidzu, T.; Ohtani, A.; Iyoda. T.; Honda, K. J. Electroanal. Chem. 1987.224, 123-135. Patil, A.G.; Ikenoue, Y.; Wudl, F.; Heeger, A.J. J. Am. Chem. Soc.. 1987.109, 1858-1859. Reynolds, 1.R.; Sundaresan. N.S.; Pomerantz, M.; Basak. S.; Baker, C.K. J. Electroanal. Chem. 1988.250, 355-371. (a) Oyama, N.; Anson, F.C. J. Electrochem. Soc. 1980. 127, 249- 250. (b) Shigehara, K.; Oyama, N.; Anson, F.C. Inorg. Chem. 1981,20, 518-522. Oyama, N.; Sato, K.; Matsuda, H. J. Electroanal. Chem. 1980, 115. 149-155. Cox, 1.A.; Kulesza, R]. J. Electroanal. Chem. 1983, 154, 71-78. Belanger, D. J. Electroanal. Chem. 1988.251, 55-62. Braun, H.; Storck, W.; Doblhofer, K. J. Electroanal. Chem. 1983. 130, 807-811. Zumbrunnen, H.R.; Anson, F.C. J. Electroanal. Chem. 1983, 152. 111-124. Doblhofer, K.; Braun, H.; Storck, W. J. Electroanal. Chem. Soc 1983, 130, 807-811. Bovey, F.A. "High Resolution NMR of Macromolecules"; Academic Press: New York, 1972. 321. 322. 323. 324. 325. 326. 327. 328. 329. 330. 331. 332. 333. 334. 335. 248 Mortimer, R.1.; Anson, F.C. J. Electroanal. Chem. 1982, 138, 325-341. Bruce, 1.A.; Wrighton. M.S. J. Am. Chem. Soc. 1982. 104, 74- 82. Miller, T.A.; Prater, B.; Lee, 1.K.; Adams, R.N. J. Am. Chem. Soc.. 1965,87, 121-122. Kuwana, T.; Bublitz, D.E.; Hoh, G. J. Am. Chem. Soc.. 1960, 82, 5811-5817. Osa, T.; Kuwana, T. J. Electroanal. Chem. 1969. 22, 389-406. Winograd, N.; Kuwana, T. J. Am. Chem. Soc. 1971. 93, 4343- 4350. Samec, Z.; Nemec, C.I. J. Electroanal. Chem. 1971,31, 161-173. Bookbinder, D.C.; Wrighton, M.S. J. Am. Chem. Soc.. 1980, 102, 5123-5125. Lewis, N.S.; Wrighton, M.S. Science 1981.21], 944-947. Bookbinder. D.C.; Lewis, N.S.; Bradley, M.G.; Bocarsly, A.B.; Wrighton, M.S. J. Am. Chem. Soc.. 1979, 10], 7721-7723. Dominey. R.N.; Lewis, T.J.; Wrighton, M.S. J. Phys. Chem. 1983, 87. 5345-5354. Ryan, T.M.; Day: R.1.; Cooks, R. Anal. Chem. 1980, 52, 2054- 2057. Ghosh, P.K.; Spiro, T.G. J. Am. Chem. Soc.. 1980, 102. 5543- 5549. Ghosh, P.K.; Spiro, T.G. J. Electrochem. Soc. 1981. 128, 1281- 1287. Willman, K.W. Thesis; University of North Carolina .1981. 336. 337. 338. 339. 340. 341. 342. 343. 344. 345. 346. 347. 348. 349. 249 Haque. R.; Coshow. W.R.; Johnson, L.F. J. Am. Chem. Soc.. 1969, 91, 3822-3827. (a) Spotswood. T.M.; Tanzer, C.I. Aust. J. Chem. 1967, 20. 1213-1225. (b) Spotswood. T.M.; Tanzer, C.I. Aust. J. Chem. 1967, 20, 1227-1234. Wang, 1.; Mahmoud, 1.S. J. Electroanal. Chem. 1986.208, 383- 392. Wang. 1. ”Stripping Analysis: Principles, Instrumentation, and Applications"; VCH Publishers: Deerfield Beach, Fla., 1985. Wang, 1.; Bard. A.J. (Ed.) "electroanalytical Chemistry"; M. Dekker: New York, 1988, 17, pg 1-88. Cheek, G.T.; Nelson, R.F. Anal. Lett. 1978, 393-402. Whiteley, L.D.; Martin C.R. Anal. Chem. 1987,59, 1746-1751. Guadalupe, A.R.; Abruna, H.D. Anal. Chem. 1985,57, 142-149. Murray, R.W.; Ewing, A.G.; Durst, R.A. Anal. Chem. 1987, 59, 379A-385A. Wang 1.; Greene. B.; Morgan, C. Anal. Chim. Acta. 1984, 158, 15-22. Kalcher, K. Analyst 1986. 111, 625-630. Ohsaka, T.; Okajima, T.; Oyama, N. J. Electroanal. Chem. 1986, 200. 159-178. Nagy, G.; Gerhardt, G.A.; Oke, A.F.; Rice, M.B.; Adams, R.N.; Moore, R.B. 111.; Szentirmay. M.N.; Martin, C.R. J. Electroanal. Chem. 1985, 188, 85-94. (a) Saveant, J.M. J. Electroanal. Chem. 1987.238, 1-8. (b) Saveant, J.M. J. Electroanal. Chem. 1988.242, 1-21. 250 350. Andrieux, C.P.; Saveant, J.M. J. Phys. Chem. 1988.88. 6761- 6767.