Several studies have shown that addition of NaF into the aqueous dispersion of TiO 2 (Degussa P25) can result in significant enhancement in the photocatalytic degradation (PCD) of organic pollutants, ascribed to the enhanced production of free OH radicals in solution as a result of fluoride displacement of surface hydroxyl groups. In this work, we have observed different results of NaF addition for the PCD of phenol over synthetic TiO 2 in aqueous suspension under UV light irradiation (λ g 320 nm). Upon the addition of NaF, the rate of phenol PCD was only increased with anatase, but it was decreased with rutile under similar conditions. In the presence of AgNO 3 , however, the fluoride-induced rate enhancement of phenol PCD could be observed with both anatase and rutile, ascribed to the increased rate of scavenging the conduction band electrons. As the catalyst sintering temperature was increased, the amount of fluoride adsorption on TiO 2 was decreased, but the degree of PCD rate enhancement due to NaF addition as observed with anatase was first increased and then decreased, the trend of which was similar to that in the absence of NaF. The result reveals that the excess fluoride ions present in the suspension play some positive role to the phenol PCD, which is hardly interpreted by previous mechanism of surface fluorination. Moreover, as initial concentration of fluoride and initial pH of suspension were increased, the degree of rate enhancement was increased and decreased, respectively, which also could not be ascribed solely to the change in fluoride adsorption. Possible interference from catechol and hydroquinone intermediates and the fluoride-induced enhancement in the production of OH radicals in solution are analyzed. A new mechanistic model is proposed, involving enhanced desorption of surface bound OH radicals from irradiated TiO 2 , by fluoride ions present in the Helmholtz layers, through a fluorine hydrogen bond.

No abstract

Photoinduced organic oxidation with iron (hydr)oxides in aqueous suspension has been argued with respect to two principal mechanisms: (a) photoinduced ligand-to-metal charge transfer within a surface complex and (b) semiconductor photocatalysis. In this work, the photodegradation of azo dye orange II with UV light (lambda > or = 320 nm) in the aerated aqueous suspensions of haematite, maghemite, magnetite, goethite, lepidocrocite, and feroxyhite at an initial pH of 6.5 has been examined. The results showed that (1) all of the catalysts were effective at initiating dye photodegradation but the iron oxides appeared to be more active than the iron hydroxides; (2) the photodissolution of different iron phases and the dye photolysis in the dissolved iron solutions were very slow; (3) the initial rate of dye loss was proportional to the initial amount of adsorption, implying dye photodegradation on the catalyst surface; and (4) upon addition of H2O2, AgNO3, and NaF to the suspension, the rate of dye photodegradation was significantly enhanced with all the catalysts. In the presence of H2O2, less than 50% of the total rate enhancement was ascribed to the photo-Fenton reaction in solution and the dark Fenton reactions in solution and on the catalyst. In the presence of AgNO3, about 1 mole of silver particles was produced by consuming 3 moles of the dye substrate. In the presence of NaF, hydroxyl radicals were detected by an ethanol scavenger, whereas such radicals were not found in the absence of NaF. Moreover, under visible-light irradiation (lambda > or = 450 nm), the dye degradation was much slower than that under UV irradiation, but the reaction was also accelerated by the addition of NaF and AgNO3. The results suggest that mechanism b, not mechanism a, is operative for dye photodegradation occurring on the iron (hydr)oxides. A detailed discussion of all possible pathways is given in the text.