Mobile Edition
Print
Get the Mag
Weekly UpdatesABSTRACT
We have studied deactivation of titanium dioxide (Ti[O.sub.2]) photocatalyst by oxidation of polydimethylsiloxane and silicone sealant off-gas in a recirculating batch reactor. Polydimethylsiloxane vapor is a model indoor air pollutant. It does not adsorb strongly on Ti[O.sub.2] in the dark, but undergoes oxidation when the ultraviolet (UV) photons are also present. Commercial silicone (room-temperature vulcanizing) sealant off-gas is an actual indoor air pollutant subject to short-term spikes in concentration. It does adsorb on the Ti[O.sub.2] surface in the dark, but UV photons also catalyze its oxidation. The oxidation of the Si-containing vapors was monitored using a Fourier transform infrared spectroscope equipped with a gas cell. Subsequent to each incremental exposure, a hexane oxidation reaction was performed to track the titania catalyst's activity. The exposures were repeated until substantial deactivation was achieved. We have also documented the regenerative effect of washing the catalyst surface with water. Surface science techniques were used to view the topography of the catalyst and to identify the elements causing the deactivation. Procedural observations of interest in the context of our recirculating batch reactor include the following: the rate of oxidation of hexane was used to assess the activity of a photocatalyst sample; hexane is an appropriate choice of a probe molecule because it does not adsorb in the dark and it undergoes photocatalytic oxidation (PCO) completely, forming C[O.sub.2]; and hexane does not deactivate the photocatalyst surface.
INTRODUCTION
Titanium dioxide (Ti[O.sub.2]) particles irradiated with sufficiently energetic photons ([lambda] <385 nm) produce hydroxyl radicals and other reactive oxidative species at the air particle interface. Oxidation at near room temperature ensues. Photocatalytic oxidation (PCO) is, therefore, a process to be evaluated in the context of indoor air quality (IAQ) improvement. (1)
According to the Environmental Protection Agency (EPA), sources of indoor air pollutants include: paints, paint strippers, and other solvents; wood preservatives; aerosol sprays; cleansers and disinfectants; moth repellents and air fresheners; stored fuels and automotive products; hobby supplies; dry-cleaned clothing, as well as outdoor pollutants introduced by ventilation. EPA studies of human exposure to air pollutants indicate that indoor air levels of many pollutants may be 25 times, and on occasion more than 100 times, higher than outdoor levels. (2) These levels of indoor air pollutants are of particular concern because it is estimated that most people spend as much as 90% of their time indoors. PCO of these pollutants to less harmful intermediates, water, and carbon dioxide has therefore acquired increased significance.
Bioaerosols and other particles in suspension in air may cause deactivation of photocatalysts. Chemical pollutants in solution in air may also adsorb on and deactivate the irradiated surface. Deactivation of the photocatalyst by common indoor pollutants would undermine the commercial utility of this technology. This study focuses on photocatalyst deactivation by exposure to chemical pollutants.
Cao et al. (3) reported severe deactivation during the PCO of toluene, which was attributed to formation of partially oxidized intermediates (e.g., benzaldehyde and benzoic acid) on active sites. Activity was restored after exposure to high temperature (420[degrees]C). Mendez-Roman et al. (4) made similar observations. Vorontsov et al. (5) reported deactivation during oxidation of acetone at elevated temperatures. This was attributed to thermal oxidation of intermediates produced by the PCO. Deactivation during ethanol PCO has also been reported. (6) These references site loss of photocatalytic activity as a result of oxidation of hydrocarbons and suggest partially oxidized intermediates are blocking active sites on the catalyst. These intermediates are slow to oxidize, but not refractory. Therefore, continued exposure to oxidizing conditions restores some or all of the photocatalytic activity.
Other investigators assert that stable, refractory oxides and mineral acids form on and bind to the catalyst surface, thereby blocking catalytic sites. Further exposure to oxidizing conditions will not lead to recovery of activity. Sun et al. (7) studied the effect of oxidizing octamethyltrisiloxane on Ti[O.sub.2] film photocatalyst. The authors report the proposed deposition of a layer of silica on the titania surface, which reduces photocatalytic activity. The silica layer is removed by dilute base. In this work, oxidation occurs slowly over a span of tens of hours. This is a key difference from the present study. In the recirculating batch reactor described below, we observed complete oxidation of polydimethylsiloxane (PDMS) and commercial silicone sealant off-gas in less than 1 hr, as evidenced by disappearance of reactants and appearance of carbon dioxide (C[O.sub.2]). Differences in feed concentration or light intensity do not account for the discrepancy in reaction rates, raising the possibility of transport limitations in the referenced work. Also, the rate of oxidation of the pollutant itself was used as a measure of catalytic activity. In the present study, we assessed photocatalytic activity by using hexane as a probe molecule.
Ollis and co-workers produced most of the literature related to deactivation of photocatalysts operating in the gas phase. (8-11) They observed that air pollutants containing Si of N atoms generate irreversible deposits containing those atoms along with nonvolatile carbon. This type of deactivation does not respond to oxidative methods of regeneration and other methods of catalyst regeneration and/or renewal must be explored. Provided titania surfaces are sufficiently robust, water and aqueous solutions can be used for regeneration purposes. (12)
As discussed above, pollutant molecules that contain heteroatoms (atoms other than C, H, and O) are likely candidates to lead to the formation of refractory oxides. This is particularly true for Si, which is present in the compounds undergoing oxidation in this study. One such pollutant is the polysiloxane vapor emitted from commercial silicone sealants otherwise known as room temperature vulcanizing (RTV) multipurpose sealants. This is a product that is used widely in residential and commercial buildings, often in significant quantities. Cast resin countertop silicone sealant contains up to 20 g/L volatile organic compound (VOC) whereas High-performance silicone has up to 250 g/L VOC. The rate of vapor evolution (off-gas) is high immediately after the product is used and decreases with time. Therefore, this vapor might be present in significant concentrations in an indoor environment.
[FIGURE 1 OMITTED]
In this work, we exposed a Ti[O.sub.2]-coated monolith to a model silicone sealant off-gas (PDMS vapor) and to an actual silicone sealant off-gas in a well-mixed, recirculating batch reactor. Both vapors underwent oxidation. We measured the effect of incremental exposures and oxidations on the photocatalyst activity with respect to the oxidation of a probe molecule, hexane.
EXPERIMENTAL APPROACH
Deactivation of the Ti[O.sub.2] catalyst was explored by exposing an irradiated titania surface to a potential poison introduced into the air contained in a recirculating batch reactor. The potential poisons were oxidized, hypothetically having a residual effect on photocatalyst activity. We selected PDMS as a model pollutant and silicone sealant off-gas for more realistic experimentation. The exposure was incremental. An individual increment of pollutant was mixed with the air via recirculation while the ultraviolet (UV) light source was shielded. As noted below, PDMS did not adsorb on the catalyst in the dark, but a significant fraction of the RTV off-gas did adsorb in the dark over time in the recirculating batch reactor. The reactor, which is described in more detail below, is equipped with a light shield that blocks all photons. Upon removal of the shield, the concentration of the deactivating agent decreased with time, accompanied by an increase in the concentration of C[O.sub.2] as oxidation of the carbonaceous material proceeded. This process can be followed qualitatively by reviewing the spectrum in Figure 1, which shows the disappearance of PDMS accompanied by the appearance of C[O.sub.2].
Before and between each exposure to the deactivating agent, the activity of the catalyst was ascertained by oxidizing a probe molecule in air. During the course of the work, we explored the use of five species as probe molecules: hexane, acetone, ethanol, trichloroethylene (TCE), and toluene. Acetone, ethanol, and toluene showed marked adsorption on the Ti[O.sub.2] catalyst "in the dark." This was noted by the disappearance of the compounds' Fourier transform infrared (FTIR) spectral peaks during recirculation as soon as they came in contact with the Ti[O.sub.2] catalyst, i.e., before the shield was removed. Adsorption to the walls of the recirculating batch reactor was not significant, as the peaks were steady when the monolith reactor was bypassed. Evolution of C[O.sub.2], however, did not commence until the light shield was removed. Nimlos et al. (13) and Falconer et al. (14) showed that ethanol forms acetaldehyde, which either desorbs or oxidizes through at least two parallel pathways. We also observed gas-phase intermediates during oxidation of ethanol. Intermediate formation consistent with the literature was also seen during the oxidation TCE. (15)
[FIGURE 2 OMITTED]
Hexane was chosen as the probe molecule because it does not form gas-phase intermediates. Like PDMS, it also did not adsorb to the catalyst in the dark. Hexane undergoes complete, quantitative PCO (producing 6 C[O.sub.2] molecules) without deactivating the catalyst, unlike for example acetone. (4) Figure 2 illustrates the change in the infrared (IR) absorption profile during the oxidation of hexane in air. We used rate of disappearance of the reactant species and the rate of evolution of C[O.sub.2] as a measure of the activity of the catalyst. Figure 3 is a plot of the C[O.sub.2] peak area from a time series of spectra. The slope indicated provides a numerical basis for calculating a reaction rate and evaluating and catalyst activity.
FEED