Search

The kinetics of the removal of stearic acid (SA) islands by photocatalytic coatings is controversial, with some reporting that the islands fade as their thickness, h, decreases with the irradiation time, t, but maintain a constant area, a, −da/dt= 0, and others reporting that −dh/dt= 0 and −da/dt= −constant, i.e., the islands shrink, rather than fade. This study attempts to understand the possible cause for these two very different observations through a study of the destruction of a cylindrical SA island and an array of such islands, on two different photocatalytic films, namely, Activ self-cleaning glass, and a P25 TiO2coating on glass, which have established uniform and heterogeneous surface activities, respectively. In both cases, using optical microscopy and profilometry, it is shown that, irrespective of whether there is as a single cylindrical island or an array of islands, hdecreases uniformly with t, −dh/dt= constant, and −da/dt= 0, so that the SA islands just fade. However, in a study of the photocatalyzed removal of SA islands with a volcano-shaped profile, rather than that of a cylinder, it is found that the islands shrink and fade. A simple 2D kinetic model is used to rationalize the results reported in this work. Possible reasons for the two very different kinetic behaviors are discussed. The relevance of this work to self-cleaning photocatalytic films is discussed briefly.

Clearly discernible rings of high photocatalytic activity, i.e., rings of activation, RoA, are formed around Pt macro-sized (radius: ro≥ 1 mm radius) Pt islands, i.e., Pt “dots”, deposited on TiO2films, which have been covered with a visible pollutant, such as soot, or a dye, such as acid orange 7. The features of the RoA region appear very similar for soot- or dye-covered films. Thus, the radius of the RoA region, ra, appears to depend directly on ro, plus a constant, typically 0.22−0.68 mm. The rate of photocatalytic removal of the surface pollutant insidethe RoA is uniform and significantly greater, i.e., ca. 3.2–5×, than that outsidethe RoA region. RoA region formation is observed for (i) Pt dots which have been photodeposited or created through thermal reduction, (ii) Pt dots on sol–gel and commercial (paints and tiles) TiO2films, and (iii) Pd dots, but less clearly for Au dots. A simple mechanistic rationale for RoA region formation is presented, and the potential importance of this feature, with regard to improving the performance of commercial photocatalytic films, is discussed briefly.

The increasingly popular disrupted Langmuir–adsorption (DLA) kinetic model of photocatalysis does not contain an explicit function for the dependence of rate on the irradiance, ρ, but instead has a term αρθ, where, α is a constant of the system, and θ is also a constant equal to 1 or 0.5 at low or high ρ values, respectively. Several groups have recently replaced the latter term with an explicit function of the form χ1(−1 + (1 + χ2ρ)1/2), where χ1and χ2, are constants that can be related to a proposed reaction scheme. Here the latter schemes are investigated, and revised to create a more credible form by assuming an additional hole trapping step. The latter may be the oxidation of water or a surface saturated with O2–. Importantly, this revision suggests that it is only applicable for low quantum yield/efficiency processes. The revised disrupted Langmuir–adsorption model is used to provide good fits to the kinetic data reported for a number of different systems including the photocatalytic oxidation of nitric oxide (NO), phenol (PhOH), and formic acid (FA).

In this review, we highlight the role played by metal nanoparticles (NPs) in photocatalytic oxidation with titania as a support. This is presented in two parts, namely, partial photo-oxidation in which an organic sacrificial agent is oxidised in anaerobic conditions to produce hydrogen (photo-reforming), and photo-oxidative mineralisation of organics in aerobic conditions. We present some rules for such reactions that dictate which organic molecules can react readily, and which metals are likely to be useful for such reactions. Generally, the presence of metal NPs enhances enormously the ability of titania to yield hydrogen from photo-reforming, and a wide range of molecules can be used, including biomass. The metal NPs most used are those that are easily reduced, that is, the precious metals. The large enhancement in rate seen with metal for hydrogen production is not so extreme for the oxidation reactions, but is still significant. An important factor in all of this catalysis is the nature of the interaction between the metal NPs, which can play a multiplicity of chemical and electronic roles, and the photoactive support. A sharp dependency of rate on loading of metal is found, with maximum rates at ~0.5–2 wt% loading, depending on the metal used. The source of this dependency is the bifunctional nature of the system, in which the intimacy of both materials is crucial to performance. This rate variation is linked to the interface between the two, which is then linked to the size of the metal NPs. In fact, the rate is proportional to an area adjacent to the metal particles that we call the expanding photocatalytic area and overlap (EPAO) kinetic model. This model describes the dependence well. Rising rates with increasing coverage of particles is associated with increase in this total area but, at the maximum, these areas overlap and at higher loadings the available active area diminishes, reproducing the observed behaviour well.

The use of an acid violet 7 (AV7) smart ink to assess the activity of photocatalytic paint is demonstrated. A linear correlation is established between the change in oxidized dye concentration, as measured by diffuse reflectance, and the change in the green component of the RGB color values, obtained using a portable hand-held scanner, suggesting that such tests can be monitored easily using an inexpensive piece of hand-held office equipment, as opposed to an expensive lab-based instrument, such as a diffuse reflectance UV/vis spectrophotometer. The bleaching of the AV7 follows first order kinetics, at a rate that is linearly dependent upon the UVA irradiance (0.30–3.26 mW cm–2). A comparison of relative rate of bleaching of the AV7 ink with the relative rate of removal of NOx, as determined using the ISO test (ISO 22197-1:2007), established a linear relationship between the two sets of results and the relevance of this correlation is discussed briefly.

The observed adsorption of methylene blue, MB+, on P25 titania over a range of pH values (pH 2–11) is reported and the results fitted to a charge distribution, multisite complexation, i.e. CD-MUSIC, model, modified for both monomer and aggregated dye adsorption and (for aggregates) co-adsorption of hydroxyl ions. The rate of bleaching of MB+, photocatalysed by titania using UV light, increases over the pH range 4–11 in much the same way the amount of adsorbed MB+, [MB+]ads, increases, indicating that the rate-determining step involves direct hole oxidation of adsorbed MB+. In contrast, at pH 2, the rate of UVA dye photobleaching appears unduly high but is due to an additional reductive, as well as the usual oxidative, bleaching processes. The initial rate of bleaching of MB+via a visible light-driven, dye-sensitised process, measured at different pH's correlates with the associated CD-MUSIC-calculated values of [MB+]ads, indicating that only dye molecules that are in direct contact with the surface can undergo photodegradation via a dye-sensitised route.

Three commercial exterior photocatalytic paints, from STO, Boysen and Graphenstone and one In-house paint, are tested using the NO ISO; FT-IR analysis shows that all paints contain an acrylate binder, but the commercial paints also contain siloxane and/or silicate binders. All exhibit little or no photocatalytic activity (for NO removal) without weathering. The Graphenstone and In-house paints are only activated by ‘standard’ weathering, i.e. a continuous cycle of UV irradiation (0.76 mW cm-2, 8 h at 50 °C) plus condensation (4 h at 60 °C). In contrast, the STO and Boysen paints are also activated by ‘condensation-only’ weathering by water droplets at 60 °C for ca. 56 h. This is the first report of the efficacy of the latter type of weathering. The activity vs weathering time profiles for the STO photocatalytic paint show that it is almost exclusively activated by condensation-only weathering, which is fast (60% maximum activity within 8 h) and produces a film that is active and durable. In contrast, the In-house paint is only activated by standard weathering, which is relatively slow (only 25% maximum activity after 24 h) and involves the photocatalytic oxidation of the acrylate binder; as a consequence the In-house paint film is not very durable. Irrespective of the method of activation, all the photocatalytic paints exhibit little, or no, NOxremoval activity towards the end of the 5 h irradiation period of the NO ISO, due to the accumulation of surface HNO3, which promotes the conversion of NO to NO2.

We report the first results in which ALD-TiO2layers electronically couple silicon to an overlying catalyst coating while inhibiting corrosion during brine splitting, a reaction that is capable of generating not only a fuel (H2) but also a disinfectant (Cl2or NaOCl). An n-type silicon photoanode, and a p+-silicon anode are protected by a 1.7 nm coating of amorphous TiO2and 2 nm coating of Ir. The p+-Si/TiO2/Ir anode is able to effect the oxidation of chloride (0.5 M H2SO4, 3.5 M NaCl) in the dark with a low overpotential compared to that for oxidation of water (0.5 M H2SO4). The 0.294 V overpotential difference occurs despite the difference in standard redox potentials, ΔE = (E°(Cl2/Cl−) - E°(O2/H2O)), of 130 mV. Under 1 sun irradiation a photovoltage of ca. 0.566 V is achieved by comparing the illuminated n-Si/TiO2/Ir photoanode with the dark p+-silicon anode. Although the n-Si/TiO2/Ir photoanode is only one half of an eventual tandem cell needed for photosynthetic brine splitting, its ABPE for chloride oxidation is 1.42%, ca. 28 times that for water oxidation. The illuminated n-Si/TiO2/Ir photoanode remained stable at 1 mA cm−2during a six-day chronopotentiometry test.