The Performance in Fluidized Bed Reactors of Photocatalysts Immobilized onto Inert Supports.
Roberto L Pozzo, Alberto E. Cassano y Miguel A.Baltanás
INTEC (UNL / CONICET) – Guemes 3450 Santa Fe, Argentina
Introduction and scope
Much attention has been paid in recent years to the study of the photo-oxidation of air and water pollutants, catalyzed by semiconductor metal oxides. In particular, it has been shown that many organic compounds undergo rapid mineralization in redox type reactions induced by electron-hole pairs generated when near-UV radiation excites the band gap of the semiconductors.
Two modes of catalyst availability are currently favored for these type of photocatalytic processes; a) as a finely divided slurried powder, for the aqueous media, or b) as an immobilized coating on support materials in a fixed or fluidized bed configuration, both for air [1-3] and water [4-9] decontamination systems. Focusing on the aqueous media and from an engineering point of view, the immobilized catalyst (usually TiO2, in anatase form) is preferred, to avoid downstream treatments (i.e., particle-fluid separation/catalyst recycling).
Because of the small particle size of the powdered catalyst generally synthesized by the industry (30-200 nm) the cost requirements of any downstream separation stage impose severe economic constraints but, on the other hand, the efficiency of slurry type reactors appears to be generally higher than those using immobilized photocatalysts [4-8].
Common basis to establish significant, quantitative comparisons are still lacking. So, it is of interest to be able to contrast the performance of both modes of catalyst availability (powdered and immobilized) under controlled and comparable conditions. With this goal in mind, special attention has been paid by us to assure that both types of reacting system were operating as a fully irradiated photoreaction space (FIP reactor [10]), and with identical catalyst structure and texture.
Experimental Approach and Methods
Two reacting configurations: Suspended bed (SB), for powder, and fluidized bed (FB), for immobilized, supported TiO2, were used. A physical deposition method was developed to fix the catalyst (Degussa P25: ~75% anatase; Sg @ 50 m2 g-1; dp = 30-70 nm), to compare the same active material under both arrangements. The FB reactor was selected to maximize mass transport rates. Oxalic acid was chosen as the model substrate since it is easily decomposed, degrades without stable intermediates and it is not volatile.
The experimental set-up consisted of a recirculating, isothermal batch reaction system. The photoreactor was a multitubular device: 3 concentric annuli and a tubular black light lamp (15 W, emitting mostly at 360 nm) placed at its axis. The central annulus (radial space: 7.5 mm) was the photoreactor, the outer one was an actinometric space to measure the exiting radiation, outgoing from the reaction vessel. The inner annulus was used as an IR filter.
The catalyst was fixed on quartz sand (dp = .25 mm) by dry mixing, followed by humectation and evaporation in vacuo and calcining at 400ºC. The deposited amount, digested with (NH4)2SO4, was evaluated by UV spectroscopy; its distribution onto the surface of the sand was analyzed by SEM. Three degrees of sand coverage were tested: 100% (One ‘nominal’ monolayer, about 1 mg of TiO2/g sand), 60% and 40%, respectively. Also, three levels of expansion of the FB: 7.0, 3.8 and 2.3 times the unexpanded bed level were used. Thus, the concentration of catalyst in the reaction vessel in some of the tested configurations could be adjusted to be the same and, also, equal to the corresponding concentration used for comparison in the slurry (250 ppm/liter). In total, six different combinations of sand coverage and bed expansion were tested. A standard aqueous solution, 50 ppm of oxalic acid di-hydrate (Carlo Erba RSE, 99.9%), was used throughout the work as a model reactant. The reaction was followed by TOC; actinometric measurements were conventional.
Results and discussion
For performance comparison purposes, an apparent quantum efficiency (h ) was defined as the ratio between the initial decomposition reaction rate of the oxalic acid and the apparent captured power (the difference between incident and outgoing radiation power per unit of reacting volume). Both type of reaction systems were found to operate as FIP devices. The mechanical stability of the catalyst was satisfactory and no detachment of TiO2 was detected during or after the experimental runs. Under these conditions it was found that the photocatalytic performance of the immobilized catalyst was significantly poorer (about 5 to 6 times lower) than that obtained with the powdered form
Several factors may contribute to this lower performance of the anchored catalyst operating in the FB: a) reduction of available specific surface area resulting from the binding with the supporting surface of the quartz sand, b) significant radiation extinction (absorption and scattering) by the support, or c) catalyst agglomeration (surface clumping, observed by SEM) during fixation, among others.
It was also found that h is fairly sensitive to bed expansion: the higher the bed expansion, the higher the efficiency. However, for any given bed expansion the increment in reaction rate is just mild and the apparent quantum efficiency is rather insensitive to the surface coverage, within the used range of sand coverage.
On the basis of these results, and aiming to elucidate the reasons for such a significant loss of chemical activity upon fixation, a new program is now in progress. It focuses on the following aspects: a) effects upon h of significant increments (5 to 10 times) of the catalyst coverage degree; b) effect of pH of the wetting solution on catalyst agglomeration; c) effect of support pretreatments (e.g., surface smoothening with concentrated hot nitric acid) on the optical properties (radiation extinction) of the FB and on the characteristics of the supported photocatalyst layer.
References
Profile of the research team.
This piece of research work is part of a two-pronged effort, currently in progress in the Chemical Reaction and Reactors Engineering area of INTEC, which is aimed to contributing to the abatement of water pollution using advanced oxidation technologies (AOT). Photocatalytic processes, using preferably titanium dioxide immobilized onto supports, as well as photochemical processes employing UV with and without hydrogen peroxide are under scrutiny.
In the first case slurry reactors, fixed and fluidized beds are employed, while several configurations of radiation emitting system and reactor are utilized in both the photocatalytic and the photochemical approaches. The reactor geometries range from mono- to bi- or three-dimensional radiation field(s).
Measurement of the optical characteristics and properties of the devices and materials and/or obtaining intrinsic kinetics is combined with computational modeling of the radiation fields, software development, model validation and process simulation(s). Both, priority pollutants or ‘real world’ samples from rivers, aquifers or process plants are considered.
The research group, directed by Prof. Dr. Alberto E. Cassano, is integrated by a team of CONICET/UNL researchers (11), professionals (3) and technicians (4), as well as several graduate students. Photocatalytic and photochemical reaction engineering represents about 65% of the research work of this Area of INTEC. The fields of expertise of these team of researchers encompasses chemical reaction and reaction engineering (homogeneous and heterogeneous), photoreactor and photocatalytic reactor modeling and simulation, molecular dynamics, heat and mass transfer and catalysts development and characterization.
Research is mostly financed by multinational or national government agencies (Interamerican Development Bank, SECyT, CONICET, ANPCyT) and Universidad Nacional del Litoral. Also, bilateral or multilateral research projects between the INTEC group and EU countries are now in progress. The group seeks and welcomes international cooperation as an efficient means for a mutual strengthening of material and human resources, technical capabilities, scientific knowledge and achievements, and likelihood of technological impact.
Facilities
The photoreactor engineering group occupies approximately 180 m2 of laboratory space and 100 m2 of offices at INTEC. Air-conditioning and other special services (compressed air, vacuum, carrier gases) are available. INTEC has access to the computational and instrumentation facilities of the Regional Center for Research and Development of Santa Fe (CERIDE) and to the National Center of Catalysis (CENACA), both located in our city. The group owns the following dedicated, specific instruments and equipment:
Contact: Prof. Dr. Alberto E. Cassano
Fax: 54 (342) 455 9185 e-mail: acassano@arcride.edu.ar