Band 3, Heft 2

Photocatalytic degradation of formic acid (HCOOH) in solution was monitored using titanium dioxide (TiO 2 ) films supported on Pyrex and soda-lime glasses (SLG) calcined with two different calcination methods: programmed and short time (10 min) with temperature 100 and 700 °C for both methods. As the calcination temperature and duration increased, the photocatalytic activities using TiO 2 films supported on SLG decreased, whereas the photocatalytic activities of TiO 2 films supported on Pyrex were not much affected by calcination temperature and duration. TiO 2 films supported on both Pyrex and SLG showed reduced photocatalytic activities after -50 days.

The long term behaviour of suspended photocatalysts (Degussa P25) is studied in two endurance tests. Model waste water with an acid dye (Special Brilliant Blue FFR) or 4-chlorophenol as main contaminant have been used. Titanium dioxide is recovered from the effluent by pressure filtration and reused in the feed. A change in suspension behaviour of the titanium dioxide was observed over the period of the endurance tests (approx. 400 h). The photocatalytic activity decreased in the same period by approx. 55 % in the case of chlorophenol and approx. 25 % in the case of the dye. Intermediates from the chlorophenol degradation have been identified. The result of the tests are fitted to a physical model of the process. They are also useful for improved reactor design

During the last decade the use of titanium dioxide has been the focus of water purification studies for photocatalytic degradation of organic compounds. However, the possible use of this technology for water disinfec­tion has been essentially unexplored. In our study U.V. black light (365 nm) at 5.5 mW cm -2 irradiation intensity was used on different TiO 2 concentrations (g/L) in batch reactor. Using this irradiation intensity the most effective inactivation of E. coli CN 13 was obtained at 1 g/L suspension of TiO 2 (viable numbers reduction of 5 orders of magnitude in 5 min was observed). Under the same experimental conditions, MS2 bacteriophage viable count was reduced by 4 orders of magnitude, also in 5 min. B. fragilis bacteriophages have been well documented to be highly resistant to UV irradiation (245 nm) compared to enteric viruses and other bacteriophages. A B. fragilis bacterio­phage isolated from sewage was also tested for inactivation, and the results showed a drop of 2 orders of magnitude in 5 min. Tested for longer periods of irradiation, E. coli CN 13 was inactivated completely after 60 min (initial concentration 1O 6 CFU/mL), MS2 dropped to 1O 2 PFU/mL after 60 min (initial concentration 1O 9 PFU/mL) and B. fragilis phage dropped to 0 after 60 min (initial concentration 1O 4 PFU/mL). The Deinococcus radiophilus bacterium well known to be highly resistant to ionizing radiation (λ-radiation) was also subjected to TiO 2 photo- catalytic inactivation in water. The observed reduction in viable count of Deinococcus radiophilus under similar conditions applied to E. coli CN 13 , was approximately one order of magnitude. However addition of O 2 into the experimental environment increased the inactivation of Deinococcus radiophilus by four orders of magnitude in 60 min (first 30 min no inactivation was observed, but in the next 30 min the viable count dropped by four logs). The TiO 2 photocalytic inactivation shows promising results for inactivation of microorganisms in drinking water, though further studies are required to optimize this system according to various engineering aspects.

Clean and sterile water is scarce in developing tropical countries and the goal of universal access to water and sanitation has not yet been achieved. Standard water treatment techniques are often too expensive both in capital investment and operation and maintenance for use in developing countries. Since most technology, historically, has been developed in the temperate zone, it may not be the most appropriate for the tropical climate. The solar photochemical process has proven effective against organic chemical pollutants, including agricultural pesticides, which pose a growing threat to water sources in the tropics. Research conducted at the University of Florida has shown this technology to be effective for bacterial decontamination, which is often the primary concern in water supplies of developing countries. The use of solar energy as the primary driver serves to make this technology economically viable. Screening experiments were conducted to investigate the effect of pH and concentration on the simultaneous disinfection and detoxification of water using TiO 2 photocatalysis. Preliminary results indicate that this process is feasible for appropriate technology applications. Benzene, toluene and xylene were mineralized to levels below detection limits at a concentration of 0.01% TiO 2 after 4 h of irradiation at 29 W/m2. Disinfection of water contaminated with Escherichia coli, Pseudomonas aeruginosa and Serratia marcescens proceeded to completion after four hours at TiO 2 concentrations of 0.01%.

Titania-coated silica microspheres containing between 0.5 and 3.0 theoretical layers of TiO 2 have been prepared by homogeneous precipitation of TiCl 3 . All these xTiO 2 -SiO 2 materials were characterized by X-ray diffraction, UV-Vis, FT-Raman and XPS. Calcination of silica-supported precipitates leads to the formation of titanium dioxide. TiO 2 species are distributed on the surface as small anatase crystals and possess a titania-like behaviour with a much better performance than bulk TiO 2 for the photodegradation of phenol. Crystal size of titania particles is critical for optimum performance of the catalyst since it determines the appropriate surface exposure of titanium sites.

Photocatalytic degradation of 2-, 3-, and 4-chlorobenzaldehydes (2-ClBzA, 3-ClBzA and 4-ClBzA) in aqueous solution was investigated in the presence of TiO 2 fine suspension upon irradiation with UV light λ = 254 nm. Oxidative degradations of the substrates were followed by absorption changes at their corresponding absorption maximum as well as with HPLC measurements. The formation of chloride ions and formaldehyde have been taken as indicators for the substrate decomposition. Quantum yield values for reactant disappearance and product formation and a probable reaction mechanism are presented.

The removal, from aqueous solution, of uranyl ions by adsorption in the absence and presence of the chelant EDTA, and the photocatalytic removal by reduction of uranium from U:EDTA solutions is reported. Uranyl (UO 2 2+ ) adsorption at pH6 approaches saturation at about 40 ppm UO 2 (NO 3 ) 2 . Adsorption is negligible at pH 2, and increases to a maximum at pH = 7-8, then declines at pH 10. The increased adsorption is consistent with a progres­sively higher deprotonation of TiO 2 surface sites, whereas the decrease at pH 10 corresponds to partial conversion of positive uranyl specie, UO 2 2+ , to a negatively charged hydroxy complex, UO 2 (OH) 3 - . Added EDTA competes for uranyl with the surface sites, and a monotonic diminution of adsorbed uranium is observed with increasing EDTA level. At 1:1 U:EDTA ratio, the adsorbed metal increases monotonically from zero at pH 2 to a maximum at pH 10 (more basic conditions were not explored), a result reflective of the stability of the U:EDTA complex. Under illumination in the absence of dissolved oxygen, the uranium (VI) in solution (free and complexed) decreases indicating a reductive process for the metal. The apparent first order rate constant for disappearance is observed to decrease with increasing pH, i.e. the rate constant is maximum at pH 2 and smallest at pH 10. Centrifugation of catalyst and exposure of supernatant for 24 h to air results in regaining the original uranyl (VI) concentration at pH near 2 and 5; in contrast, one day exposure to air provides no reoxidation of supernatants at pH 7 or 10. These results indicate that uranyl is photocatalytically reduced to a lower, positive oxidation state to form a positive specie which adsorbs appreciably only under neutral/ basic conditions, above the isoelectric point of TiO 2 .

Three Advanced Oxidation Technologies (UV/H 2 O 2 , UV/TiO 2 and Electron Beam) have been evaluated using the Electrical Energy per Order (EE/O) figure-of-merit. These processes have been tested on two aqueous treatment systems: the bleaching of methylene blue and the decay of phenol. In general, as expected, for a given process, the EE/O increased as the concentration of the contaminant increased. However, when the three processes were compared for the same system at the same concentration, the EE/Os for the UV/H 2 O 2 and Electron Beam processes were very similar, but the EE/O for the UV/TiO 2 was much higher than that for the other two processes. The UV/H 2 O 2 process has the disadvantage that there is an added cost of chemicals (principally H 2 O 2 ) and the E-beam process has the disadvantage that there is a constant power required to run the apparatus even when the beam is off. Nevertheless, the UV/TiO 2 system has an inferior performance relative to the other two processes; the principal reason is probably the low quantum yield (~0.04-0.08) for the generation of hydroxyl radicals on the surface of TiO 2 particles.

Gas-phase photocatalytic oxidation of different classes of volatile organic compounds (VOCs), including alkanes, ketones, alcohols, chlorinated compounds and aromatic compounds, was investigated using an annular thin film reactor. For all organic compounds tested, catalytic deactivation was not observed, except for toluene (506 ppmv). For this compound, maximum destruction (87%) was maintained during the first 60 min of irradiation, dropping steadily due to catalyst deactivation to 20% after 150 min. The use of ozone as auxiliary agent in the photocatalytic oxidation of toluene and pyridine was also tested. For toluene, when O 3 is co-injected into the reactor, deactivation of TiO 2 is no longer detectable, and after 150 min of irradiation, 72% of degradation was obtained. When exposing pyridine to this combined treatment, destruction rates raised from 16 to 68%.

As the bench-scale photoreactors are upscaled to progressively larger units, heat and mass transfer considerations become increasingly important. Powerful analytical and computational techniques are available to augment experimental data and aid process optimization and scale up. In this paper, the analytical and computational techniques available for the design of vapor-phase photocatalytic reactors are discussed. First, the Graetz- Nusselt-Leveque problem in annuli is analyzed and its application to the design of the photocatalytic reactors described. Then, the analytical predications are compared to experimental flow reactor data. Finally, results from a Computational Fluid Dynamics program simulating a flow field within an annular baffled photoreactor are given and discussed. These techniques are particularly useful as they simplify the design and scale-up of vapor-phase photocatalytic reactors.

Trichloroethylene conversion in gas-solid photocatalysis ranks as one of the fastest reactions for air purification and treatment. The development of a complete kinetic model for this conversion is important in that it represents, under some conditions, a likely upper limit to conversion photoefficiency. Both reactant concentration and light intensity are known to influence reaction rate, but only two literature reports have explored a sufficiently wide range of concentrations to find that the intensity dependence is coupled to, and not independent of, the reactant concentration. We develop here a simple model which rationalizes this experimental behavior based upon the familiar competition between species for photo-produced holes (h + ) and electrons (e - ). First order intensity dependence of rate is associated with good photoefficiency, whereas half-order dependence provides low efficiencies. Thus, this coupling of concentration and intensity effects serves as a reactor design warning: The kinetic order in intensity, and therefore the process economics with respect to photoefficiency, for other pollutant conversions may also be found to depend upon concentration, not just intensity.