The effect of a specifically adsorbed ion, phosphate, on the electrochemical response and adsorption properties of nanocrystalline TiO2 is examined. Phosphate is known to affect the ζ potential, as measured by electrophoretic mobility, by changing the charge of the oxide surface. The adsorption of a cationic probe molecule, thionine, onto TiO2 was monitored with an in-situ cell using UV−vis spectroscopy. The adsorption of the cationic dye molecule was found to be governed by changes in the ζ potential, whether the ζ potential was modified by pH or by changes in phosphate concentration. Onset potential measurements were used to estimate the flat-band potential of a Ti/TiO2 electrode. The flat-band potential results for the electrode showed a nearly Nernstian response to changes in the pH for a broad pH range. The addition of phosphate had no effect on the onset potential or on the shape of the photocurrent/potential curve. Flat-band potentials determined by Mott−Schottky analysis in the absence of phosphate were Nernstian only for pH 3−7, matching the pH dependence of the electrophoretic mobility results. With the addition of phosphate, impedance spectroscopy results showed additional space charge capacitance, peaking at potentials 150 mV positive of the flat-band potential. UV irradiation also resulted in an additional space charge capacitance. For both cases, the additional space charge capacitance was accompanied by a decrease in the resistance of the electrode, as shown in Nyquist plots. The change in film conductivity is believed to affect the space charge layer capacitance. Similarly, a decrease in film resistance was also seen with lower pH values. Currently, this change in TiO2 film conductivity with surface acidity is being investigated in our laboratory for application in fuel cell electrolytes.

A photoelectrocatalytic reactor containing titania-coated titanium electrodes was employed to degrade solutions of formic acid (2 mmol dm -3 ) in 0.01 mol dm -3 NaCl. Reaction rates were increased above that observed for a purely photocatalytic experiment by operating at applied potentials of at least +1.0 V (versus saturated calomel electrode). The kinetics of photodegradation at +1.0 V was modeled effectively using a Langmuir-Hinshelwood-Hougen-Watson expression. Unexpected results were obtained when only the background electrolyte was passed through the reactor. During initial recirculation of this solution with no UV illumination and no applied potential, the pH increased from 6.5 to 9, suggesting ion exchange of chloride ions with hydroxyl ions from the catalyst surface. However, when UV illumination was initiated with an applied potential, the pH decreased to 3.5-4.2, depending on the magnitude of the potential. The cause of this behavior is not known, although there are several explanations. Addition of formic acid to this system buffered the pH near 3, producing the highest rate of degradation at an applied potential of +1.0 V. When the formic acid test solution was adjusted to higher initial pH values, the reaction rate was unaffected until the pH increased above 5, at which point the rate decreased.

Processes that occur in the TiO2-photocatalysis of binary aqueous solutions containing model photolytes with different affinity for the TiO2 surface (methanol and oxalic and salicylic acids) are analyzed from the photoelectrochemical response of TiO2 films under bias in a time window of 1-100 s. Long-lived oxidized intermediates produced upon illumination at 0.6 VSCE are detected by cathodic sweep run in the dark after irradiation. The main conclusion derived from this work is that a scheme of competitive kinetics describes only those cases in which one of the components is weakly or nonadsorbed on TiO2, whereas for two photolytes with high affinity for the surface cooperative effects may occur. The methanol-oxalate system is quantitatively modeled by considering that oxalate forms surface complexes with different reactivity and a parallel pathway for hole transfer to -OH and adsorbed oxalate. In this case as well as for electrolytes containing methanol and salicylate photooxidation of methanol (with low affinity for the surface) via intermediates formed by reaction with trapped holes (-*OH) is partially or fully suppressed. For electrolytes containing oxalic and salicylic acids in which both components chemisorb on TiO2 the photoelectrochemical response depends on preadsorption, the photooxidation pathways deviates those of single component systems, and there is remotion of salicylate adsorbed byproducts assigned to cooperative effects.

No abstract