AUGUST, 1973 THE ANALYST Vol. 98, No. 1169 Mass and Charge Transfer Kinetics and Coulometric Current Efficiencies Part VII.* Conditional Potentials, and Single-scan Voltammetry of Pure Vanadium(V) - Vanadium( IV) Systems in Various Media at Platinum Electrodes Pre-treated by Five Methodsf. BY E. BISHOP AND P. H. HITCHCOCKf (Chemistry Department, University of Exetev, Stockev Road, Exeter, E X 4 4QD) The limited previous work on vanadium is reviewed. Five methods of electrode pre-treatment have been selected and are described. The variation of the conditional potential of the vanadium(V) - vanadium (IV) system with hydrogen-ion concentration is reported. The experimental work is complicated by isopolymerisation reactions, of which a t least one is kineti- cally slow. The voltammetric reduction of vanadium (V) in saturated potas- sium sulphate - acetate buffer a t pH 4.0 is examined: the benefit of shifting solvent reaction potentials to more negative potentials is nullified by the direct reduction of un-ionised acetic acid.M sulphuric acid and in one of 2.0 M sulphuric acid. Possible adsorption effects are canvassed. The complex behaviour a t intermediate hydrogen-ion concen- trations is discussed, with illustrations drawn from a sulphuric acid medium of pH 2.0. The anodic oxidation of vanadium(1V) is briefly examined. It is concluded that the electrochemical behaviour of the vanadium system is strongly dependent on hydrogen-ion concentration and on the electrode pre-treatment. The electrode can be chemically oxidised in vanadium (V) solutions.The mechanism is a one-step one-electron process. No evidence could be found for reduction below the +4 oxidation state a t platinum. Vanadium(1V) cannot be oxidised without severe loss of current efficiency, nor reduced to vanadium(II1) a t platinum electrodes. IJNGANE~ could obtain little information from a polarographic investigation of the vana- dium(V) - vanadium(1V) system. The complex structure of the voltammetric waves at carbon electrodes in sulphuric and orthophosphoric acids defied interpretation,2 although the repro- ducibility was fairly good. Davis334 examined the reaction chronopotentiometrically and voltammetrically, and concluded that the reduction of vanadium(V) was accelerated by a lightly oxidised platinum surface (the oxygen bridge theory, since rejected5y6).Amperostatic determinations of vanadium(V) have been reported for copper(I),' iron(II),s t i t a n i u ~ n ( I I I ) , ~ ~ ~ ~ and tin( 11)5 as intermediates, Potentiostatic determination at a mercury cathode has been reported by Israel and Meites,ll but no report of reduction at platinum has appeared, although Davis12 remarked that the reaction was slow. A voltammetric study of the system has been made in a variety of media with a view to the determination of the mass and charge transfer kinetic parameters, and the selection of conditions for the potentiostatic determination of vanadium both alone and in mixtures with other steel alloying elements. EXPERIMENTAL A similar examination is made in a medium of 5 x General apparatus and procedures have been described earlier,l3 and electrode treatments and behaviour have been reviewed.14 The reagents were of the highest purity available (Aristar, P.V.S.or AnalaR grade) and were further purified if necessary. Sodium perchlorate was made from sodium carbonate and perchloric acid. Presented a t the Second SAC Conference, Nottingham, 1968. * For particulars of Part VI of this series, see reference list, p. 562; for Part VIII, see p. 563. Z Present address: Ever Ready Co. (G.B.) Ltd., Central Research Laboratories, St. Ann's Road, @ SAC and the authors. London, N15 3TJ. 553554 BISHOP AND HITCHCOCK: MASS AND CHARGE TRANSFER [Analyst, VOl. 98 Water-Throughout this series of papers, the terms water and distilled water refer to sterile, grease-free and surfactant-free water prepared in a special still,15 and having a total solids content of less than 10-l2 M, calculated as chloride.It contains dissolved carbon dioxide and oxygen, which are removed by methods described earlier.16917 Primary standard sodium carbonate solution, 0.05 M-This solution was made by direct weighing of P.V.S. sodium carbonate that had been dried at 280 "C for 2 hours. Primary standard sodium oxalate solution, 0.05 M-This solution was made by direct weighing of anhydrous sodium oxalate that had been dried at 105 "C for 2 hours. Primary standayd potassium dichi.omate solution, 0.01 6 67 M-This solution was made by direct weighing of potassium dichromate prepared by twice recrystallising AnalaR potassium dichromate from water, pulverising the crystals and drying them at 160 "C for 3 hours.Potassium permanganate solution, 0-02 Ki-Potassium permanganate was dissolved in water, the solution heated to boiling-point, simmered for 1 hour, cooled, filtered through a porosity 4 sinter and transferred into a conditioned amber Winchester bottle. A 25.00-ml aliquot of the 0-05 M sodium oxalate solution was transferred into a 250-ml conical flask containing 8.0 ml of 4.0 M sulphuric acid, the solution heated to between 80 and 90 "C and titrated with the permanganate solution to the first perceptible permanent pink colour. A blank of 8 ml of 4.0 M sulphuric acid in 50 ml of water was titrated in a similar way. In the early stages of the oxalate titration, the permanganate was added very slowly, so that the solution became colourless before the next drop of permanganate was added, until the accumulated manganese(I1) rendered the reaction fast.The standardisation was repeated before each period of use of the permanganate solution. A relative standard deviation on this and all other titrations of ,t0-04 per cent. was secured. Vanadium( V ) solutions-A solution of ammonium metavanadate was boiled with a small excess of potassium hydroxide until all the ammonia was removed. The solution was cooled and diluted to 0.1 M with water. Standardisation was effected by reducing a large aliquot of the solution with sodium sulphite and sulphuric acid, and the excess of sulphur dioxide was removed by refluxing the solution for 1 hour in a stream of nitrogen. The solution was cooled and diluted to its original volume with water.Aliquots of the vanadium(1V) solution were titrated with the standardised permanganate solution, with sodium diphenylamine- 4-sulphonate as the indicator. The 0.1 M potassium vanadate solution was used to prepare solutions of 0.05 M vanadium(V) in 2.0 M sulphuric acid, and 0-05 M vanadium(V) in 4.0 ni hydrochloric acid. Acetate bufer supporting electrolyte, pH 4-A 0.45 M solution of sodium acetate was saturated with potassium sulphate and then acidified with glacial acetic acid to pH 4, as indicated by glass and calomel electrodes and a 39A pH meter. ELECTRODE ACTIVATION PROCEDURES- Davis3** oxidised platinum electrodes chemically with a solution of silver( IT) oxide in 6.0 M nitric acid, and reduced them either potentiostatically, or by soaking them in iron(I1) sulphate solution.He obtained reproducible results if the oxidation immersion time was long enough. Ansonlg suggested that this method of oxidation could leave an adsorbed layer of silver on the electrode, which might have misled Davis to conclude that a thin platinum oxide film could accelerate oxidation - reduction processes. Iron(I1) in 1.0 M sulphuric acid was used by Anson18 for reducing platinum oxide films, and this method was found by Bishop and Riley17 t o be more reliable than electrolytic reduction in preparing platinum for the reduction of silver. Anson18 found that the oxide was relatively stable to iron(I1) in 1 . 0 ~ perchloric acid, and Kabanova has confirmed this observation.lg After many trials with diverse activation procedures, the following methods were selected for extensive testing.( a ) Chemical reduction with iron(l1)-The electrode was immersed in 0.5 M iron(I1) sulphate in 1.0 M sulphuric acid for 10 minutes at room temperature, followed by thorough washing with water. (b) Electrolytic cycling and reduction-The auxiliary electrode used in all electrolytic treatments was made of platinum. The working electrode in 1.0 M sulphuric acid was anodised a t 300 mA cm-2 for 30 s, followed by cathodisation at 300 mA cm-2 for 30 s; the anodisation and cathodisation were repeated once more, and the electrode was finally washed with distilled water.August, 19731 KINETICS AND COULOMETRIC CURRENT EFFICIENCIES. PART VII 555 (c) Chemical plus electrolytic strippiag and cathodisation-The electrode was immersed in fresh aqua regia a t 60 "C for 30 s, washed with distilled water, anodised in 11.6 M hydro- chloric acid at 208 mA cm-2 for 30 s, and again washed.Finally, the electrode was cathodised a t 230 niA cm-2 for 10 minutes in 1.0 M sulphuric acid and washed thoroughly. (d) Electrolytic reduction-The electrode was cathodised a t 300 mA cm-2 for 6 minutes in 1 . 0 ~ sulphuric acid and then washed thoroughly. ( e ) Electrolytic cycliag and oxidation-After two cycles of anodisation - cathodisation as in method ( b ) , the electrode was anodised for a third time at 300 mA cm-2 for 30 s and thoroughly washed. Unless otherwise stated, the electrodes were used directly after the last wash; any hydrogen present would be removed chemically on immersion in the electrolyte.RESULTS AND DISCUSSION First, in order to make best use of pattern theory,2O the conditional potential of the vanadium(V) - vanadium( IV) system and its dependence on hydrogen-ion concentration were determined. Then the voltammetric behaviour of vanadium(V) over a range of hydro- gen-ion concentrations was examined with respect to the diverse electrode treatments outlined above. In order to abbreviate the presentation, three conditions will be discussed, acetate buffer at pH 4.0, strongly acidic medium, 2.0 M stllphuric acid, and an intermediate state of pH 2.0 in sulphuric acid medium. Unless otherwise stated, electrodes were used for a single scan and then re-activated before further use. THE CONDITIONAL POTENTIAL OF THE VANADIUM(V) - VANADIUM(IV) SYSTEM- The zero-current equilibrium potential was found to obey the relationship, .... .. RT [VV] F [VIV] - * E,, = EL + - In ~ where EL is the conditional potential, which depends on the hydrogen-ion concentration of the medium, as shown in Fig. 1. For solutions 0.1 M or more in sulphuric acid, the acid content was determined by titration against primary standard sodium carbonate solution. 1.1 2 1 0 - I -2 -3 -4 -5 -6 Log (hydrogen-ion concentration/mol I-') Fig. 1. Variation of the conditional potential of the vanadium (V)- vanadium(1V) system with hydrogen-ion concentration (S.H.E. denotes standard hydrogen electrode). e, Values measured in sulphuric acid media; X , values measured in acetate buffer of pH 4.0; and I, range of values reported by Davis3 for sulphuric acid medium For less acidic solutions, the pH was measured potentiometrically by using glass and calomel electrodes calibrated with pH standards.The zero-current potential was independent of the method of electrode activation if sufficient time was allowed for the potential to become stabilised. Methods (b) and (c) produced electrodes that became stabilised faster than electrodes treated by other methods; but even for these two methods the stabilisation time varied, from 10 minutes to several hours, indicating very small exchange currents.556 BISHOP AND HITCHCOCK: MASS AND CHARGE TRANSFER [A.italyst, Vol. 98 VOLTAMMETRY OF VANADIUM (V)- It was quickly evident that freshly acidified vanadate solutions gave erratic results compared with aged solutions.This behaviour is caused by kinetically slow isopolymerisa- tion, probably21 the step at pH 7.0 to 6.8: Reaction PH 2 [V0413- + 2 H+ + [V,O7I4- + H20 . . .. . . 12 to 10.6 (2) 2 H4[V501,]3- + 6 H+ $ 5 V205 + 7 H,O . . .. . . 2.2 (5) V,O, + 2 H+ s 2 [VO,]+ + H20 . . .. . . (1.0 (6) 2 [V2O7l4- + 4 H+ + H2[\74013]4- + HZO . . . . 9.0 to 8-9 (3) 5 H2[V40,3]4- + 8 H+ + 4 H4[V5016]3- + H20 . . . . 7.0 to 6.8 (4) Israel and Meitesll obtained evidence from ultraviolet spectra that ammonium metavanadate dissolved in sulphuric acid slowly changed form. A recent review22 indicates that slow steps are involved in the polymerisation reactions, but that the situation is more complicated than equations (2) to (6) suggest, because the degree of polymerisation depends also on the vanad- ium concentration.To eliminate the external complication of polymerisation equilibria, all voltammetric work was carried out with solutions that had been appropriately acidified and stored for several weeks. The voltammetric behaviour of vanadium(V) in saturated potassium sulpliate solution containing sulphuric acid in the range lod5 to 2.0 M was examined; slight differences in electrode pre-treatment caused significant changes in the voltammograms. Solutions of pH about 2 gave very erratic results. In none of the experiments performed in this study was any evidence found for the reduction of vanadium below the +4 oxidation state. Anson5 and Davis3 both found chronopotentiometric evidence for the formation of vanadium( 111) at freshly reduced platinum.They also found that the electrode quickly became deactivated towards the second step at +0.25 V. In most of the runs in the present study ramp speeds of &lo0 mV min-l were used, and the time required for the electrode to deactivate towards vanadium(II1) formation was exceeded before the potential for reduction of vanadium(1V) was reached. The cause of this deactivation is not clear, but adsorption of impurities or polymeric vanadium species on the electrode seems likely. REDUCTION AT pH 4- A single reduction wave appeared for reduced electrodes [treatment (b) or (c)] in both supporting electrolytes, saturated potassium sulphate solution adjusted to pH 4 with sulphuric acid or the saturated potassium sulphate - acetate buffer.In Fig. 2, for the acetate buffer medium, curves 1 and 2 represent the background currents for reduced ( b ) and oxidised (e) electrodes, respectively. The oxidised electrode showed a maximum at about 0.7 V corresponding to reduction of the oxide film on the electrode. Addition of vanadate to a concentration of 1.09 x loA3 M gave a wave that is superimposed on curves 1 and 2 to give the composite curves 3 and 4. Chemically reduced electrodes (a) gave curves similar to 1 and 3, but the curves were not as reproducible. After curve 2 had been recorded, it was found possible to reproduce it without reactivating the electrode, so a single scan did not deactivate the electrode, and the supporting electrolyte was free from impurities that could be adsorbed.An activated electrode remained stable for 30 minutes or more in the vanadium solution, whether on open circuit or used for scanning, so the whole system could be judged to be free from readily adsorbable impurities.14 The shape of the vanadium wave at pH 4.0 is independent of the medium, but the buffer introduced a second wave, which was identified as resulting from the direct reduction of un-ionised acetic acid. 2 CH3COOH + 2e- $ 2 CH3COO- + H, . . .. - * (7) The observation of Laitinen and I l ~ l t h o f f , ~ ~ that in the presence of a large excess of sodium acetate the limiting current of the acetate wave was proportional to the total con- centration of acetic acid, was confirmed. The existence of the acetic acid wave nullified the benefit of shifting the background reactions towards more negative potentials by raising the pH.On the other hand, hydrogen ions are consumed in the reduction, and sulphuricAugust, 19731 KINETICS AND COULOMETRIC CURRENT EFFICIENCIES. PART VII I 1 557 0.9 0.6 0.3 0.0 -0.3 Electrode potential versus S.H.E./V Fig. 2. Reduction of vanadium(V) in acetate buffer saturated with potassium sulphate a t pH 4.0: 1, supporting electrolyte alone, electrode treatment (b) ; 2, supporting electrolyte alone, electrode treatment (e) ; 3, supporting electrolyte + 1.09 x M vanadium(V), electrode treat- ment ( b ) ; and 4, supporting electrolyte + 1-09 x M vanadium(V), electrode treatment ( e ) . Ramp speed - 100 mV min-1 acid at pH 4, although adequate for voltammetry, would not provide a sufficient supply unless it was supplemented at intervals during a coulometric determination.The use of a buffer solution in which the salt predominates would ameliorate the situation, but no suitable system could be found. The height of the vanadium(V) reduction wave was propor- tional to concentration over the range 0.3 to 3.5 x M. The mass-transfer rate constant was smaller in acetate buffer than in sulphuric acid, because of the higher viscosity of the buffer medium. In both media the limiting current of vanadium reduction decreased with decreasing stirring speed. 0 6 I 1 a E -.- + Lo > S .- 1 -2 0.9 0.6 0.3 0.0 Electrode potential versus S.H.E./V Fig. 3. Reduction of vanadium(V) in 2.0 M sulphuric acid: 1, 2.0 M sulphuric acid, electrode treatment (b) ; 2, 2-0 M sulphuric acid, electrode treatment (e) ; 3, 2.0 M sulphuric acid + 1.13 x M vanadium(V), electrode treatment ( b ) ; 4, 2.0 M sulphuric acid + 1-13 x 10-3 M vanadium(V), electrode treatment ( e ) ; 5, constructed by subtracting curve 2 from curve 4, to compensate for oxide reduction; and 6, recorded immediately after curve 3 was recorded, without intervening activation of the electrode.Ramp speed - 100 mV min-1558 BISHOP AND HITCHCOCK: MASS AND CHARGE TRANSFER [Analyst, VOl. 98 REDUCTKON IN 2.0 M SULPHURIC ACID- Voltammetric curves in this medium (Fig. 3) are more complicated than those at pH 4, but less complex and more reproducible than those at pH 2. The reduction wave showed a limiting current region that varied slightly in height but very greatly in the rising portion of the wave with the method of electrode activation. The chemical treatment, (a), gave much less reproducible curves than the electrolytic treatments and was therefore abandoned.Treatments (b) and (c) gave very similar voltammograms but method (b) gave rather more reproducible results, although even then a 5 per cent. difference had to be accepted as being satisfactory. Scan 1 of the background with a reduced [treatment ( b ) ] electrode showed that the electrolyte and electrode were clean. Scan 2 is of an oxidised electrode [treatment (e)] and shows the oxide reduction peak. When the oxidised electrode is used to reduce vana- dium(V), the oxide reduction peak is superimposed on the vanadium wave, as in curve 4. By subtracting curve 2 from curve 4, curve 5 was obtained and showed that the rate of reduction of vanadium(V) remained very slow until some of the oxide on the platinum surface had been reduced, but when about 80 per cent.of the oxide had been reduced, the electrode behaved like a pre-reduced electrode. The observation that the rate of reduction of vanadium(V) decreased at an oxidised platinum surface is in agreement with the findings of Anson and King.5 It is not, however, easy to explain why the electrode in this experiment behaves as if it were fully reduced, when curve 2 indicates that oxide remains on the electrode. The curves in Figs. 2 and 3 were drawn at the same ramp speed, yet the two oxide reduction curves differ; that in Fig. 3 is less sharp and shows tailing on the cathodic side: the tailing could be diminished by using a slower ramp speed.Curve 3 in Fig. 3 shows that reduction of vanadium(V) at a pre-reduced electrode is not simple: a minimum appears at about 0.6 V. This behaviour occurred whenever the hydrogen-ion concentration of the solution was such that vanadium(V) reduction could be observed at potentials more positive than 0-6 V. Although the conditional potential of the vanadium(V) - vanadium(1V) system could be changed by 0.4 V or more by changing the hydrogen-ion concentration, there was very little change in the potential at which the minimum occurred, which is close to the potential of the platinum oxide reduction maximum, so that it seemed possible that the two might be related. The chemical effect of a 10-3 M solution of vanadium(V) in 2.0 M sulphuric acid on a reduced electrode prepared by method (b) was investigated by immersing the fresh electrode in this solution for 2 minutes, washing it well with distilled water, transferring it into a pure 2.0 M sulphuric acid solution and making a cathodic scan.The scan showed a small maximum at 0.65 to 0.70 V, thus indicating light oxidation of the electrode surface. Repetition of this procedure with 2.0 M sulphuric acid alone showed that the oxidation occurred chemically in the vanadium(V) solution and not during the washing or transfer. Soaking the electrode for 15 s in the vanadium(V) solution produced a detectable amount of oxide. Although a fully reduced electrode can be immersed in the vanadium(V) solution for scanning, the electrode surface was lightly oxidised before the scan was started.The electrode had to be carefully positioned, as did the Luggin capillary, then connections had to be made to the ramp generator and the circuit tested. This process took 15 to 30 s to complete. The chemical oxidation of the working electrode, reduction of the oxide at about 0-6 V and the minimum in the vanadium(V) reduction wave at pre-reduced electrodes [methods (b) and (c)] appear to suggest that light oxidation of the electrode does facilitate the vanadium reduction as compared with a fully reduced electrode. If this be so, then vanadium(V) in 2.0 M sulphuric acid is reducible at three different rates, depending on the amount of oxide on the electrode surface. A heavy oxide film [treatment (e)] must, on the evidence of curve 5 in Fig.3, permit only very slow reduction of vanadium(V). The light oxidation by chemical attack by vanadium(V) permits a relatively fast reduction, while a fully reduced electrode (formed when the potential becomes negative to 0.6 V) must reduce vanadium(V) at an intermediate rate. This effect was, in part, confirmed by lowering the zero-current potential of the vanadium solution by addition of vanadium(IV), so that chemical oxidation did not occur. Curve 1 in Fig. 4 shows no minimum at 0.6 V in the reduction wave, and both k and cc are obviously decreased. Curve 3 shows the normal behaviour in the initial absence of vanadium(1V) under otherwise the same conditions. Curve 2 is the immediate reverse scan after curve 1 and shows the large hysteresis.August, 19731 KINETICS AND COULOMETRIC CURRENT EFFICIENCIES.PART VII 0.04 I 1 I CI c 6 0.03 559. 1 I 1 I 1 1 1 1 1.2 1.0 0.8 0.6 0.4 0.2 0;O -0.2 Electrode poten t i a I versus S. H. E ./V Fig. 4. Reduction of vanadium(V) in 0-1 M sulphuric acid: 1, 0.1 M sulphuric acid + 1.0 x M vanadium(V) + 1.0 x M vanadium(IV), electrode treatment ( b ) ; 2, reverse scan at + 100 mV min-1 immediately after recording curve 1; and 3, 0-1 M sulphuric acid + 1.0 x 10-4 M vanadium(V), electrode treatment (b). Ramp speed -100 mV min-l POSSIBLE ADSORPTION EFFECTS IN THE REDUCTION OF VANADIUM(V) IN 2.0 M SULPHURIC ACID- The hysteresis just mentioned is not confined to solutions containing vanadium( IV). Reverse scans are not shown in Fig. 3 in order to avoid confusion; reverse scans in the context have little significance, but such scans were performed and also showed hysteresis, It appeared possible that a known trace impurity in the sulphuric acid may become adsorbed on the elect rode.Aristar sulphuric acid diluted to 10-0 BI was cleaned by electrosorption, as will be des- cribed in a later paper, and the solution used to prepare the supporting electrolyte. Back- ground and vanadium(V) reduction scans showed no significant difference from those in Fig. 3, so adsorption of impurities could be eliminated. Nevertheless, it is possible that adsorption of vanadium species could be responsible for hysteresis and for the difference between scan 3 in Fig. 3 with the electrode reduced by method (b) and scan 6 performed immediately afterwards with the same electrode and without re-treatment.This conclusion was supported by an experiment in which an electrode used to record a curve similar to scan 3 in Fig. 3 was subjected to treatment (d) and then scanned in the same solution. The result was a curve very similar to curve 6 in Fig. 3, and showed that simple cathodisation was not so effective as treatments (b) and (c) in producing an active electrode. The difference cannot be due to any form of platinum - oxygen structure produced during pre-treatment, because treatment (c) does not produce an oxidised surface during the anodic treatment, but merely strips platinum as the soluble hexachloroplatinic(1V) acid. The difference between treatments (b) and (c) and treatment (d) lies in the ability of ( b ) and (c) to clean the electrode by anodic stripping.Deactivation of electrodes prepared by methods (b) and (c) cannot be attributed to reduction products of the electrode reaction, because a freshly activated electrode does become deactivated on standing at zero current in the vanadium(V) solution. It is possible that the chemically produced film may contain vanadium species, perhaps in a mixture of oxidation states. In 2.0 M sulphuric acid a kinetically slower curve than scan 6 in Fig. 3 could not be obtained either by allowing the electrode to stand in the solution or by continually scanning the reduction wave, which showed conclusively that on deactivation the electrode quickly attains a stable and reproducible form. REDUCTION I N SULPHURIC ACID AT pH 2- The behaviour of vanadium(V) at the variously treated electrodes in sulphuric acid solutions of 2.0 M down to 0-1 M concentration was found to be similar to that discussed for 2.0 M acid, but with further decrease in acid concentration changes occurred in the voltam- mograms.Electrodes treated by methods (b) and (c) again showed a minimum, as in 2.0 M sulphuric acid, but at pH 2 the height of the wave before the minimum was not reproducible, and could In illustration the behaviour at pH 2 will be considered.560 BISHOP AND HITCHCOCK: MASS AND CHARGE TRANSFER [Analyst, Vol. 98 change by 50 per cent. from one experiment to the next, but was always lower than the height of the main wave formed at potentials cathodic to the minimum. The variability of the pre-wave is probably due to small differences in the amount of chemical oxidation suffered by the platinum surface before the scan was recorded. As the hydrogen-ion concentration decreases, the oxidising power of the solution decreases as shown by the conditional potentials in Fig.1. It has been ~ h ~ ~ n ~ ~ p ~ ~ that the amount of oxide formed on platinum depends on the time of contact with the oxidising solution and the potential of the solution. The amount of oxide formed at pH 2 seems more time dependent than in 2.0 M sulphuric acid. Attempts to verify this observation by coulometric stripping of electrodes that had been soaked for different periods in vanadium solutions were frustrated by the errors introduced by charging currents. An electrode activated by method (b) gave as its first and second scans curves 1 and 2 in Fig.5. The deactivation after the first scan was greater than that for a single scan in 2.0 M sulphuric acid. Curve 3 shows the effect of leaving a similar electrode in the vanadium solution for 80 minutes at zero-current potential before starting the scan. In 2-0 M sulphuric acid a single scan gave full deactivation of the electrode; at pH 2.0 the deactivation was greater, and the act of scanning accelerated it. It seems likely from equation (5) that at pH 2.0 colloidal vanadium(V) oxide could form and stick on the electrode, where it could be reduced to an insoluble species such as an oxovanadium(1V) polyvanadate(V), which would interfere with mass transfer. 09 0.6 0.3 00 Electrode potential versus S.H.E./V Fig.5. Reduction of vanadium(V) in sulphuric acid a t pH 2.0. All solutions 1.6 x M vanadium(V) in 5 x Methods used to activate the electrodes : 1, method (b) ; 2, no reactivation after curve 1, used immediately; 3, method ( b ) , then left a t zero current in the vanadium solution for 80 minutes before being recorded; 4, method ( d ) ; and 5, method (e). Ramp speed -85 mV min-l M sulphuric acid adjusted to pH 2.0. The effect of the cathodic activation (d) applied to a used electrode is shown in curve 4 i n Fig. 5, and is similar to the behaviour in 2.0 M sulphuric acid. Unlike the latter behaviour, however, an electrode having once been activated and used could be reactivated by method (d), so that the following scan showed a faster reaction than it would have shown if no additional activation had been applied.Treatment (e) produced the same shape of curve as in 2 . 0 ~ sulphuric acid, with a potential shift of the leading edge corresponding to the shift in with the change in hydrogen-ion concentration. For an activated electrode, the true limiting current at 0-1 V was proportional to the concentration of vanadium(V) in the solution. While the height of the current maximum at 0.75 V increases with increasing vanadium(V) con- centration, the relationship is not proportional.August, 19731 KINETICS AND COULOMETRIC CURRENT EFFICIENCIES. PART VII 561 As the pH of the vanadium(V) solution in sulphuric acid increased from 2 to 3 the shape of the voltammograms remained similar to those at pH 2, except for the expected shift to more negative potentials.Furthermore, the electrodes showed less deactivation by the vanadium(V) solutions. When the pH increased to 3.5, the platinum oxide reduction maximum and the vanadium reduction wave no longer coincided and the curves produced resembled those in Fig. 2. At about this hydrogen-ion concentration, the oxidation potential of the vanadium solution dropped below that at which platinum can be extensively oxidised, thus simplifying the voltammograms. The decrease in deactivation of the electrode as the pH increased from 2 to 4 can be explained by the reversal of reaction (5) to produce soluble species such as the tetrahydrogenpentavanadate (V) ion. VOLTAMMETRY OF VANADIUM(IV) SOLUTIONS- The high conditional potential and the low charge-transfer speed of the vanadium(V) - vanadium(1V) process forced the oxidation of vanadium(1V) to such a high anodic potential that no limiting current region could be isolated: the wave merged with the solvent oxidation wave.A typical scan in acetate buffer medium is shown in Fig. 6, and as the hydrogen-ion concentration increased in sulphuric acid media the merger of vanadium and solvent waves became more complete until in 2.0 M sulphuric acid the presence of a vanadium wave could scarcely be detected. Davis26 found that a chronopotentiometric transition time for vana- dium(1V) in 1 . 0 ~ sulphuric acid could not be measured on account of the simultaneous oxidation of water. Media with pH higher than 4 have not been examined, but it is possible that the change of E; with pH would be greater than the shift in solvent oxidation potential and a separate wave might be attainable at pH 6.The curve in Fig. 6 was obtained with cathodised electrodes [treatment ( b ) , (c) or (41, but the working potential is such that an anodic film must form on the electrode. A heavily oxidised electrode [treatment (e)] was found to give an even slower oxidation, and a rather erratic performance. t 1-5 1.0 0.5 0.0 Electrode potential versus S.H.E./V Fig. 6. Anodic oxidation of vana- dium(1V). Electrode activated by method (!), acetate buffer saturated with potas- sium sulphate and adjusted to pH 4.0, 2.78 x M vanadium(IV) and 0-21 x Ramp speed + 100 mV rnin-l M vanadium(V). CONCLUSIONS The voltammetric behaviour of the vanadium(V) - vanadium( IV) system at platinum electrodes is highly dependent on the composition of the supporting electrolyte and hydrogen- ion concentration, and on the pre-treatment of the electrodes. It is further complicated by participation of the electrode surf ace in chemical and electrochemical reactions, and by562 BISHOP AND HITCHCOCK direct reduction of the weak acid in buffer media.The values of the mass and charge transfer kinetic parameters will be discussed in a later paper after the behaviour in the presence of other transition-metal compounds has been described. There is no evidence in the voltam- metry of pure vanadium(V) and vanadium(1V) solutions that the reaction is other than a single-step one-electron process, although there must at least be a following chemical step, either reaction of VO, with hydrogen ion to form V02+ or reaction of V03+ with solvent to give VO,+.Isopolymers must be involved at higher pH values, and what appears to be a one electron - one vanadium atom process may be a five electron - one tetrahydrogen- pentavanadate(V) ion reaction, certainly by an electron tunnelling process, followed by decomposition or rearrangement of the polymer. A one electron - one tetrahydrogenpenta- vanadate(V) ion reaction is unlikely in view of the similar mass-transfer rates in 2.0 M sulphuric acid and in acetate buffer. Although the reproducibility cannot be considered excellent, the voltammograms of the reduction of vanadium(V) in 2.0 M sulphuric acid (Fig. 3) and in acetate buffer at pH 4 (Fig.2) show sufficiently well defined limiting current regions to permit the choice of control potentials for potentiostatic reduction of high efficiency. Vanadium(1V) cannot be oxidised without severe loss of current efficiency, nor reduced to vanadium(II1) at platinum electrodes. We express our sincere gratitude to Imperial Chemical Industries Limited for a research grant 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. coveiing a period of 3 years. REFERENCES Lingane, J. J., J . Amer. Chew. Soc., 1945, 67, 182. Miller, F. J., and Zittel, H. E., J . EEectroanalyt. Chem., 1964, 7, 116. Davis, D. G., Analyt. Chem., 1959, 31, 1461. -, J - Electroanalyt. Chem., 1960, 1, 73. Anson, F. C., and King, D. M., Analyt. Chem., 1962, 34, 362. James, S. D., J . Electrochem. SOG., 1967, 114, 1113. Meier, D. J., Myers, R. J.. and Swift, E. H., J . Amer. Chem. SOC., 1949, 71, 2340. Furman, N. H., Reilley, C. N., and Cooke, W. D., Analyt. Chem., 1951, 23, 1665. Kennedy, J. H., and Lingane, J. J., Ibid., 1958, 18, 240. Lingane, J. J., Analytica Chim. Acta, 1956, 15, 465. Israel, Y., and Meites, L., J . Electvoanalyt. Chem., 1964, 8, 99. Davis, D. G., Ibid., 1960, 1, 73. Bishop, E., and Hitchcock, P. H., Analyst, 1973, 98, 465. Bishop, E., and Sutton, J. R. B., Analytica Chim. Acta, 1960, 22, 690. Bishop, E., and Riley, M., Analyst, 1973, 98, 305. -- , Ibid., 1973, 98, 416. Anson, F. C., Analyt. Chem., 1961, 33, 934. Kabanova, 0. L., Russ. J . Phys. Chem., 1961, 1219. Bishop, E., Analyst, 1972, 97, 761. Remy, H., “Treatise on Inorganic Chemistry,” Elsevier Publishing Company, Amsterdam, 1956, Pope, M, T., and Dale, B. W., Q. Bev. Chem. Soc., 1968, 22, 527. Laitinen, H. A., and Kolthoff, I. M., J . Phys. Chem., 1941, 45, 1061. Kolthoff, I. M., and Tanaka, N., Analyt. Chem., 1954, 26, 632. Nicholson, R. S., and Shain, I., Ibid., 1965, 37, 178. Davis, D. G., J . Electvoanalyt. Chem., 1960, 1, 73. NOTE-References 13, 14 and 20 are to Parts V, VI and 111, respectively, of this series. I , Ibid., 1973, 98, 475. -- Volume 11, p. 100. Received January 24th, 1973 Accepted March 26th, 1973