RBS and NRA Study of RuO2-SnO2-TiO2 Film Electrodes

In the present work some compositional features of film electrodes based on ternary mixtures of the dioxides of ruthenium, tin and titanium, have been studied by Rutherford backscattering spectrometry (RBS) and nuclear reaction analysis (NRA). The results have been correlated with the electrochemical properties. The electrode films have been prepared by pyrolysis (400°C) of ternary mixture RuCl3.3H2O - SnCl2 - Ti diisopropoxide bis-2,4-pentane-dionate. Two series of samples were prepared, adding different amounts of titanium precursor at the constant Ru/Sn ratio of 3/7 and 1, respectively. RBS was used to follow the concentration of the three main metal ions and oxygen. The results indicating the presence of significant amounts of residual carbon in other oxide electrodes, suggested to study the presence of residual carbon by NRA also in films based on ternary mixtures. Some information on the surface morphology of electrodes was carried out in situ by cyclic voltammetry (CV) and Deuterium exchange, followed by NRA. The electrochemical characterization was further developed through a study of chlorine evolution kinetics (quasi-steady polarization curves) and service life experiments. The RBS analysis confirmed the nominal concentration values (from precursor solutions). The amount of carbon, found by 12C (d,p)13C nuclear reaction, was larger than 1E17 atoms cm-2 in all samples. This high values suggest that this component has to be distributed across all the film thickness, rather than localized at its surface. An indirect confirmation of this hypothesis comes from the observed constancy of C content, after different electrochemical experiments. It has also been found that the amount of residual carbon increases with increasing the titanium concentration. This is somewhat expectable, because of the large amounts of organics in the titanium precursor. However, this cannot be the only factor. In fact, large amounts of C were also found in films prepared from precursors not containing organic anions. More in detail, in both the series of samples the residual carbon content increases linearly with the titanium concentration. The Ru/Sn ratio does not seem to have any influence on this parameter. The amount of residual carbon in the two 30:70 and 50:50 RuO2-SnO2 binary mixtures is also with good approximation the same (about 1E17 atoms cm-2). In this respect, both these oxides exhibit the same behavior, which can have important impli-cations in surface texture and microstructure of the oxide film. Cyclic voltammograms of all of the prepared electrodes did not show well defined anodic/cathodic peaks, at variance with what observed for two-component electrode films where only titanium dioxide was added to the group VIII metal-oxide component (1). Under these conditions all of the voltammetric areas, anodic and cathodic, were assumed to be of non-faradaic character. On the basis of literature data, from voltammetric charges an estimate of the effective roughness factor of the electrode films was attempted. A roughness factor of about 70-80 was found, independent from the titanium dioxide concentration. In TiO2-containing binary electrode films, like RuO2-TiO2 and IrO2-TiO2, the behavior was quite different. Much larger roughness factors were observed, ranging between 150 and more than 300. In other papers a possible correlation between carbon content and porosity had been tentatively hypothesized, in relation to the high defectivity of the areas of the film where the impurity is stored (2,3). The results of the present work seem to contradict this idea. In fact, all the studied samples contain a large amount of carbon and anyway increasing with the titanium concentration. As a complement to CV data, a series of experiments of deuterium exchange was carried out on the electrode films, after polarization at different potentials, in 1 N D2SO4/D2O solutions. The 2H(d,p)3H nuclear reaction was used to follow the changes of D concentration. After a ten minutes polarization at a potential of 0.10 V (vs. SCE) amounts of D of the order of 1E15 could be measured. Almost one order of magnitude larger values have been measured for the above mentioned binary systems (1,2). This result can be taken as a support to the CV evidence, indicating low roughness factors for the two series of three-component films. The uptaken D+ amount normalized to the Ru+Sn amount was found to be independent, as well as the roughness factor, from the titanium concentration in the electrode film. Polarization of the electrodes at 0.10 V for twenty min. more, did not result in a significant increase of the deuterium concentration. The electrode film thickness did not affect the amount of uptaken deuterium either. The set of CV and D+ exchange data indicate that, possibly, the presence of tin dioxide in the mixture facilitates dissolution of RuO2 and TiO2, limiting segregation phenomena. Independent microstructural data obtained by the present authors on the SnO2-TiO2 system, indicated, in fact, complete miscibility across all the phase diagram. References 1 G. Battaglin, A. De Battisti, A Barbieri, A. Giatti and A. Marchi, Surf. Sci. 251/252, 73(1991) 2 A. De Battisti, G. Battaglin, A. Benedetti, J. Kristof, J. Liszi, Chimia 49, 3(1995) 3 A. De Battisti, L. Nanni, G. Battaglin, Ch. Comninellis, in V. Barsukov and F. Beck, Eds., “New Promising Electrochemical Systems for Rechargeable Batteries”, NATO ASI Series, Series 3-High Technology, vol. 6, Kluwer Ac. Publ., Dordrecht, 1996, p.197.