Photocatalytic Oxidation of Tabun Simulant-Diethyl Cyanophosphate

May 15, 2008 - Photocatalytic Oxidation of Tabun. Simulant-Diethyl Cyanophosphate: FTIR in Situ Investigation. P. A. KOLINKO* AND D. V. KOZLOV. Boresk...
0 downloads 0 Views 389KB Size
Download PDF
Environ. Sci. Technol. 2008, 42, 4350–4355

Photocatalytic Oxidation of Tabun Simulant-Diethyl Cyanophosphate: FTIR in Situ Investigation P. A. KOLINKO* AND D. V. KOZLOV Boreskov Institute of Catalysis, Novosibirsk 630090, Russia

Received September 16, 2007. Revised manuscript received March 14, 2008. Accepted March 14, 2008.

Gas phase photocatalytic oxidation of diethyl cyanophosphate vapor in a static reactor using TiO2 and modified TiO2 as the photocatalyst was studied with the FTIR in situ method. The transition metals Pt, Au, and Ag were used for TiO2 modification by the chemical and photochemical deposition methods as well as the mechanical mixture of TiO2 with manganese oxide to improve its adsorption and catalytic activity. Photocatalytic oxidation of diethyl cyanophosphate in a static reactor results in its complete mineralization with carbon dioxide, phosphoric and nitric acids, and water as the major final products. HCN was demonstrated to be the only toxic gaseous intermediate of diethyl cyanophosphate photocatalytic oxidation, formed as a result of diethyl cyanophosphate hydrolysis. Diethylphosphate and acetic and formic acids were registered as the surface intermediates. It was found that cyanhydric acid is oxidized slowly with the use of unmodified TiO2. The formation of surface cyanide complexes with Ag and Au ions could be responsible for the fast removal of HCN from the gas phase and its further photooxidation in the case of using TiO2 with deposited Au and Ag.

Introduction Photocatalytic oxidation (PCO) of organic compounds today is considered by many researchers as the only universal technique of air purification that is contaminated with low concentrations of dangerous organic species (1, 2). The modern approach to air purification using photocatalytic air cleaning devices becomes especially important in connection with the terroristic threat of the use of toxic gases and aerosols like chemical warfare agents (CWAs) (3, 4). The main feature of TiO2 based photocatalysis that makes it the most promising purification method is the high oxidative potential of photogenerated holes (∼+3V vs SHE (5)) under elatively mild UV radiation (λ < 360 nm). Interaction of photogenerated electron-hole pairs with the photocatalyst surface hydroxyl groups result in the formation of highly oxidative speciesshydroxyl radicals (OH · )swhich are believed to be the main oxidative particles in the PCO processes (6–8). The diethyl cyanophosphate (DECP) molecule is the simulant of the chemical warfare agent ethyl N,N-dimethylphosphoramidocyanidate (GA or tabun) (see the Supporting Information section for chemical structures). Tabun is an extremely toxic substance and is one of the world’s most dangerous military weapons. Because it fatally interferes with normal functioning of the mammalian nervous system, it is * Corresponding author phone: +7(383)3331617; e-mail: kolinko@ catalysis.ru. 4350

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 12, 2008

classified as a nerve agent. As a chemical weapon, it is classified as a weapon of mass destruction (9). In this way the information about the pathways of DECP transformation during PCO, its intermediates, and final product analysis will be helpful for successful development of a CWA photocatalytic utilization method. Despite DECP being the only analogue of tabun, it is still a very toxic and rarely used compound. This is the reason why information about thermodynamic properties and spectroscopic and reactivity data for DECP were not found in the literature. The formation of HCN vapor as the hydrolysis product is the common behavior of all phosphorcyanidates. Therefore the photocatalyst under investigation should provide effective adsorption and further decomposition of cyanhydric acid which is also the CWA (AC agent). Transition metals like silver, gold, and platinum were used for TiO2 modification to enhance its reactivity against HCN since they are proven to be electron trappers and therefore prevent the recombination of the photogenerated electronhole pairs on the surface of TiO2 based catalysts (10–12) thus increasing its photocatalytic activity. These metals (Ag and Au) are also known to form stable complexes with CN- anions which could enhance photocatalyst adsorptivity against HCN as well as addition of basic oxides (MgO) or hydroxides (NaOH) to the TiO2 samples. The kinetics of the gas phase PCO of DECP using pure and Au-, Ag-, and Pt-modified TiO2, the reactivity of HCN as the reaction intermediate, and the analysis of final products in the gas phase and on the photocatalyst surface were the subject of present investigation.

Experimental Section Chemicals and Materials. The diethyl cyanophosphate (DECP) (>95%) (FLUKA), AgNO3 (high-purity grade), H2SO4 (high-purity grade), NaOH (analytical grade), EtOH (highpurity grade), MgO (high-purity grade), H2PtCl6, and HAuCl4 (REACHIM) were used as supplied without further purification. Distilled water purified with Ba¨rnsted “Easy pure II” ultra pure water system (Ω ) 18.2 Mohm/cm-1) was used in all experiments. TiO2 (Sachtleben Chemie, 100% anatase, SBET ) 347m2/g, average pore diameter 4.9 nm) was used as pure photocatalyst and as further starting agent for the synthesis of Au or Ag modified photocatalyst or for the synthesis of the photocatalyst with increased surface basicity (MgO or NaOH modified). Modification of the TiO2 Surface. TiO2 modification by NaOH was done according the following method. A 5 g of TiO2 was placed in a 150 mL round-bottom flask, and 100 mL of NaOH aqueous solution was added. The molar concentrations of NaOH solution was adjusted to 0.5, 1, and 2 M. Then the flask was thermostatted for 3 h at 60 °C. Then the TiO2 suspension was washed out by consequent centrifugation at 4500 rpm (∼3000 g) for 20 min as many times as was necessarily to achieve pH ∼ 7. The preparation of the TiO2:MgO mixed sample was done according to the following method. One gram of TiO2 and 1 g of MgO were thoroughly mixed in the agathic mortar and then sonicated with a 10 mL of water in an ultrasonic bath for 15 min. After that a part of th esuspension was deposited onto the CaF2 glass support as described below. Ag, Au, and Pt metals deposition were conducted by the TiO2 impregnation with AgNO3, HAuCl4, or H2PtCl6 corresponding water solutions with consequent reduction (chemical or photoassisted) according to the following methods. 10.1021/es7021818 CCC: $40.75

 2008 American Chemical Society

Published on Web 05/15/2008

(i) Photoassisted Reduction/Impregnation (Ag/TiO2 and Au/ TiO2). A total of 2.8 mL of 0.1 M AgNO3 or 0.6 mL of 0.25 M HAuCl4 in 0.05 M HCl water solutions was added to 50 mL of the 3 g TiO2 water suspension under stirring. The AgNO3: TiO2 and HAuCl4:TiO2 molar ratios were chosen to obtain 1 wt % of deposited metal. Then the suspension was irradiated by a high pressure Xe lamp with the incident UV light intensity of 5.77 mW/cm2 (λ < 400nm) for 3 h. The initial white color of the suspension became deep-brown at the end of the irradiation process. Thus obtained catalysts were washed out by consequent centrifugation for 5 times with subsequent drying at 120 °C for 2 h. (ii) Chemical Reduction/Impregnation (Ag/TiO2, Au/TiO2, and Pt/TiO2). Chemical reduction of all three metals was conducted by the 3 times molar excess of NaBH4 against corresponding metal. The detailed chemical deposition method is described in our previous works (13). In few words, the procedures of TiO2 impregnation with AgNO3, HAuCl4, and H2PtCl6 and of final washing out and drying are the same as for the photoassisted method but the reduction was conducted by NaBH4. The AgNO3:TiO2, HAuCl4:TiO2, and H2PtCl6:TiO2 molar ratios were chosen to obtain 1 wt % of deposited Ag or Au and 0.5 wt % of deposited Pt. Nobel metals content in prepared photocatalysts was measured by fluorescent X-ray method with the VRA-30 X-ray fluorimeter. FTIR in Situ Experiments. All TiO2 samples for FTIR in situ experiments were prepared as follows. The TiO2 water suspension treated with ultrasound during a 15 min time interval was uniformly deposited onto the CaF2 support (diameter 20 mm and thickness 1 mm) and dried at room temperature. The TiO2 density on obtained samples was approximately equal to 1 mg/cm2. These samples were placed in the thermostatic reactor mounted in the cell compartment of FTIR spectrometer (Bruker, Vector 22) which was described previously (14). The main feature of the reactor is the possibility of measuring either IR spectra of the gas phase or the TiO2 catalyst surface with adsorbed species during the PCO process. All IR spectra were measured in the 900-4000 cm-1 region with the 2 cm-1 resolution and the accumulation of 16 scans. A 0.5 µL sample of DECP was usually injected into the reactor before PCO. Since the vapor pressure and volatility of DECP are low at ambient conditions, the 80 °C temperature was used in all PCO experiment to decrease the time which was necessary to achieve initial equilibrium and to expand the concentration range. The high pressure Hg lamp DRSH-1000 (Russia) was used as the UV light source. Sample irradiation was conducted by condensed light passed through a water filter (λ ∼ 365 nm, W ) 20 mW/cm2). The measurement of light intensity was conducted by the microvolt meter F-136 (Russia) equipped with the semiconductor light intensity detector calibrated with the actinometer. Surface Species Analysis. The detection of final products and intermediates of the DECP PCO adsorbed on the photocatalyst surface was conducted by the analysis of water rinsed from the photocatalyst. Two methods were used for rinsing water analysis: GC/MS with SATURN 2000 (Varian) and ion chromatography with Metrohm 861 (Advanced Compact IC) for the detection of organic species and inorganic ions, respectively. The rinsing of surface species was done according to the following method. A total of 3 mg of deposited photocatalyst was gently transferred from the CaF2 glass support into the 20 mL glass beaker, and then the 10 mL of deionized water was added. Then the suspension was sonicated for the 5 min time interval and left for 1 h for surface species desorption.

FIGURE 1. IR spectra of DECP: (1) experimental and (2) theoretical. The amount of injected DECP was 1 µL; reactor volume was 300 mL.

Results and Discussion FTIR Study. Since the DECP transformation was under investigation it was useful to register the IR spectrum of gaseous DECP at the start of the study and to make the peak assignment. For the higher confidence the same spectrum was calculated using the quantum-mechanical approach B3LYP in the 6-31G* basis with the Gaussian program. Both spectra are presented in the Figure 1. Although the absorption bands positions differ in the experimental and theoretical IR spectra it is still possible to make peak assignments: 2886, 2922, and 2947 cm-1 bands correspond to the valence C-H vibrations in the -CH2and -CH3 fragments; the 2211 cm-1 band corresponds to the valence CtN vibration in the -CtN group; 1310 cm-1 corresponds to ν(P ) O) vibration; and the 1039 cm-1 band corresponds to the δ(P-O-C) vibration. It will be demonstrated below that the DECP molecule easily reacts with the water to form gaseous HCN. The ν(-CtN) vibration wavenumber for the gaseous HCN molecule is equal to 2097 cm-1 (15) and differs from 2211 cm-1 observed for the DECP IR spectrum. It indicates that 2211 cm-1 absorption band corresponds namely to the P-CtN group like in the phosphorus cyanide P(CN)3 where ν(CtN) ) 2204 cm-1 (16). There were observed only the 3310 cm-1 absorption band in the IR spectrum of the gas phase after the addition of the 0.5 µL of the liquid DECP into the reactor and subsequent evaporation and adsorption on the TiO2 surface. The 3310 cm-1 absorption band corresponds to the valence H-CN vibration in the gaseous cyanhydric acid molecule (17) which is forming as a product of DECP hydrolysis. At the same time only 2980, 2936, 2905, and 2872 cm-1 absorption bands were observed in the IR spectrum of the TiO2 surface which correspond to the valence C-H vibrations and the 2211 cm-1 band corresponding to the valence CtN vibration in the DECP molecule was not observed (Figure 2). The fast complete DECP hydrolysis with the formation of gaseous HCN was observed in all experiments when the photocatalyst sample was placed in the reactor. The DECP molecule is hydrolyzing according to the following equation: (C2H5O)2P(O)CN + H2O f (C2H5O)2P(O)OH + HCN The obtained phosphoric acid ester is practically nonvolatile and is quantitatively adsorbed on the TiO2 surface, whereas cyanhydric acid is completely desorbed into the gas phase, giving the 3310 cm-1 absorption band. The low adsorption ability of pure TiO2 against HCN was used for the calibration of the FTIR spectrometer to obtain the quantitative informaVOL. 42, NO. 12, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4351

FIGURE 3. CO2 accumulation during the 0.5 µL DECP PCO in the static reactor (TiO2 (9), Ag/TiO2 (b), Au/TiO2 (2)) on TiO2. T ) 353 K; relative humidity ≈ 0%. The horizontal dashed line corresponds to the theoretical 100% conversion.

FIGURE 2. IR spectra of the TiO2 surface (A) and gas phase (B) during the PCO of the 0.5 µL DECP aliquot. Numerals on graph A correspond to (1) the surface spectrum of TiO2; (2) the surface spectrum after the adsorption equilibrium was stated; and (3-6) the surface spectra at 10, 30, 60, and 210 min of surface irradiation, respectively. Numerals on graph B correspond to (1) the gas phase spectrum after the adsorption equilibrium was stated and (2-6) the gas phase spectra at 10, 30, 40, 60, 210 min of photooxidation, respectively. Experimental conditions: T ) 353 K, relative humidity ≈ 0%, sample irradiation by the high pressure Hg-lamp (W) ) 20 mW/cm2. tion about the HCN gas phase concentration during the DECP PCO. A 0.5 µL DECP aliquot produced 305.8 ppm of gaseous HCN after hydrolysis in the reactor of 300 mL volume at 80 °C temperature. The 3310 cm-1 peak area was found to be 0.478 cm-1 in the 3220-3410 cm-1 region for this HCN concentration (Figure 2B, spectrum 1), giving the reactor constant cr ) 639.7 ppm · cm. The unknown HCN vapor concentration was calculated further from the reductive Lambert’s absorption law [C ) crD, where D ) absorption band area (cm-1); cr ) reactor constant for certain species and wavelength range (ppm · cm); C ) gas concentration to be desired (ppm)]. IR spectra with numbers greater than 1 (Figure 2) correspond to the PCO of the products of DECP hydrolysis: diethyl ester of the phosphoric acid (diethylphosphate or DEP) and of the cyanhydric acid. It is clear from the TiO2 surface IR spectra that the complete PCO of diethylphosphate occurs because its absorption bands disappear from the spectra for the 210 min period with the simultaneous accumulation of carbon dioxide (CO2) which can be seen from the gas phase IR spectra (2350 cm-1 absorption band). 4352

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 12, 2008

At the same time the HCN concentration remains practically unchanged (does not oxidize) because the corresponding absorption band intensity was almost constant (Figure 2B, 3310 cm-1 absorption band). The kinetics of the CO2 accumulation during the DECP PCO is shown in Figure 3. The concentration of CO2 was calculated from the corresponding peak area in the 2280-2400 cm-1 range of gas phase IR spectra during PCO. The same reducted Lambert‘s absorption law form as for HCN was used with the cr ) 236.6 ppm · cm for the 2280-2400 cm-1 range of CO2 absorption band. The DECP conversion ratio into the CO2 was found to be near 85% after the 300 min of irradiation. The remaining carbon was obviously in the form of gaseous HCN which is supported by the simple consideration that cyanhydric carbon is 20% of the total DECP carbon atoms. The low oxidative ability of TiO2 against HCN molecule could be explained by the fact that HCN is acid and it should be weakly adsorbed on the acidic TiO2 surface (18). The apparent-quantum yield (amount of quanta of light used for oxidation) was calculated according to the formula φ ) n[W/(IS)], in which W ) the oxidation rate of the DECP (molecules/cm2), I ) light intensity (quanta/cm2), S ) illuminated area (cm2), and n ) amount of charge carriers (electron-hole pairs) participating in oxidation of one substrate molecule. For DECP, n was taken to be 24 in the case of leaving the HCN molecule unchanged according to the following equation: hν, TiO2

(C-3H3-C-1H2-O)2P(O)CN + 6O2 98 4C+4O2 + H3PO4 + HCN + 3H2O (n ) 24) The origin of n is the total change of carbon atoms oxidation numbers between products and reagents. During the estimation of apparent-quantum yield, the assumption was done that every change of oxidation number by the value of 1 requires one electron-hole pair and therefore one UV quantum. The apparent-quantum yield of the DECP PCO was calculated for the first 60 min linear period of the kinetic curve, and it turned out to be equal to ∼24%. DECP Photocatalytic Oxidation on Modified TiO2. It was observed during the first 300 min of DECP PCO that only a small amount of HCN disappeared from the gas phase whereas diethylphosphate was completely oxidized. The first problem to investigate was if this low HCN disappearance rate could be attributed to (1) TiO2 photooxidation, (2) self-oxidation by the air oxygen, or (3) leakage from the reactor or adsorption on reactor’s parts. Three prolonged kinetics of HCN vapor disappearance in Ar

FIGURE 4. Kinetic curves of HCN vapor removal in the process of 0.5 µL DECP PCO in the static reactor (TiO2 (2), AgNO3/TiO2 (b), Ag/TiO2 (9), Au/TiO2 (()). T ) 353 K; relative humidity ≈ 0%. atmosphere and in O2 atmosphere (in dark and under the UV irradiated TiO2) were measured. In all cases, rates of HCN removal were equal to WHCN ) 0.14 ppm/min and were independent of the presence or absence of irradiation or air oxygen. It means that in our case a slow HCN disappearance is explained by its leakage from the reactor or adsorption on the reactor’s parts and to a smaller extent by its PCO. As a result of the weak influence of unmodified TiO2 photocatalyst on the HCN disappearance, some ways of photocatalyst modification were studied to force the TiO2 photoactivity against HCN. The idea was to enhance the HCN vapor removal rate at the expense of increasing TiO2 adsorptivity and photocatalytic reactivity. To enhance the TiO2 adsorptivity, NaOH and MgO addition was used as the first preparation technique. NaOH water solutions of 2 M molar concentrations were used for TiO2 impregnation, and MgO was used for mechanical mixture with TiO2. Both ways of TiO2 modification resulted in complete HCN removal from the gas phase due to its adsorption on the TiO2 surface in the form of surface cyanide (CNads-) which was supported by the surface IR spectra. Unfortunately, adsorbed HCN was not further oxidized for the observed 300 min time period as well as CO2 formation; as a result, DEP PCO was lower than if using unmodified TiO2 (see the Supporting Information section). The lower photocatalytic activity of the base treated TiO2 was expected since it was shown in our previous studies (14) that stabilization of acidic PCO intermediates occurs as well as CO2 formation which in turn occupies the photocatalyst sample surface and decreases its photoactivity. To enhance the TiO2 adsorptivity against HCN and at the same time to increase its general photoactivity, the transition metals (Pt, Au, and Ag) modification was applied as the second preparation technique. Silver nitrate salt (AgNO3) was used by two ways: (1)impregnation of TiO2 without further reduction and subsequent drying and (2) impregnation of TiO2 with subsequent reduction (chemical or photoassisted) and the following washing out of water soluble compounds and drying. Unlike the case of using base modified TiO2, the HCN vapor formation occurs during the DECP adsorption and subsequent hydrolysis. The obtained cyanhydric acid then undergoes the PCO (Figure 4). Photocatalyst samples Ag/TiO2 and Au/TiO2 obtained by the photoassisted reduction have demonstrated the DECP mineralization ratio near 95% for the 300 min of PCO reaction (Figure 3), and they also possess the highest activity in the HCN vapor decomposition process (Figure 4). The complete removal of HCN during the PCO of 0.5 µL DECP was observed

FIGURE 5. IR spectra of the Ag/TiO2 (photodeposited) surface during the 0.5 µL DECP PCO: (1) surface spectrum Ag/TiO2; (2) surface spectrum after the adsorption; (3-9) surface spectra corresponding to 20, 40, 60, 180, 260, 315, and 420 min of photooxidation, respectively. Experimental conditions: T ) 353 K; relative humidity ≈ 0%. for less than 100 min for the most active Ag/TiO2 sample and for near 150 min for AgNO3/TiO2 and Au/TiO2 samples. Average HCN disappearance rates were equal 1.7, 1.8, and 1 ppm/min for Ag/TiO2, AgNO3/TiO2, and Au/TiO2 samples, respectively, and were about 10 times higher than the leakage rate of 0.14 ppm/min (Figure 4). In principal the AgNO3/TiO2 and Ag/TiO2 samples should be very similar since the in situ reduction of silver nitrate supported on the titanium dioxide occurs when the adsorption equilibrium was stated and the sample illumination turned on. The difference of PCO rates for these samples as well as apparent-quantum yields could be explained by the different photoassisted reduction conditions because the AgNO3/TiO2 was reducted in situ in the dried atmosphere in the presence of organic species whereas Ag/TiO2 was priliminary reducted in the water suspension without any organic admixtures. The different conditions could result in different sizes of silver particles on the TiO2 surface and a different ratio of Ag+/Ag0 atoms (19). Detailed data of DECP oxidation rates and conversions on the samples preparation methods are listed in the Supporting Information section. In general it could be stated that the samples synthesized by a photoassisted reduction method demonstrate a higher activity in the DECP mineralization reaction unlike the samples received by chemical reduction. In our opinion, the given phenomenon is explained by the fact that the photoassisted reduction process occurs on the same active sites of the catalyst as the PCO occurs whereas chemical reduction leads to the formation of metal particles on the entire surface of the catalyst (20). An interesting behavior of the FTIR surface spectra of the Ag/TiO2 photocatalyst synthesized by the photoassisted AgNO3 reduction was observed (Figure 5). The 2162 cm-1 absorption band immediately appeared after the DECP adsorption equilibrium was established (spectrum 2, Figure 5). This band was constantly shifting to the higher wavenumber values after the PCO reaction was started, transforming into the 2202 cm-1 absorption band (spectra 3-6, Figure 5). Finally the 2202 cm-1 absorption band almost disappeared as the PCO reaction completed (spectrum 9, Figure 5). According to Nakamoto (21), the valence CtN vibration wavenumber is dependent on the coordination number of the central atom. The higher the coordination numberis, the lower the absorption band wavenumber is, which is explained VOL. 42, NO. 12, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4353

FIGURE 6. Kinetic curves of DEP (9), NO3- (2), and PO43- (1) during the 0.5 µL DECP PCO in the static reactor on the TiO2. T ) 353 K; relative humidity ≈ 0%. by the increase of the negative charge of the central atom and the corresponding decrease of CtN vibration force constant. In our case it means that at the beginning of the PCO, the large amount of adsorbed HCN gives the highly coordinated complexes [Ag(CN)3]2- or [Ag(CN)2]- (21). During the PCO, the amount of adsorbed HCN decreases and the coordination number of silver ion also decreases, giving the AgCN. The formation of labile complex species on the photocatalyst surface could explain the increased adsorptivity and photocatalytic activity of modified TiO2 against HCN. DECP PCO Products Analysis and Mass Balance. The dynamic of the surface species concentrations during the DECP PCO was investigated by conducting the multiple PCO experiments in the same conditions using the same amount of the photocatalyst and the same initial amount of the DECP to be oxidized. All experiments were interrupted at different reaction times varying from 0 min (immediately after adsorption) to 350 min of PCO reaction, and then the rinsing water analyses were done. Kinetic curves of the final product amounts as well as DEP, which is the initial surface species to be oxidized, are shown in Figure 6. The dashed horizontal line corresponds to the 100% DECP hydrolysis with the formation of the surface DEP, and it indicates that DECP is completely hydrolyzed before starting the PCO reaction since the detected initial DEP quantity is equal to theoretical and is equal to 2.5 × 10-6 mol. It is clear from Figure 6 that phosphate (PO43-) and nitrate (NO3-) anions are the main final surface products of DECP PCO. The final quantity of PO43-(ads) is almost equal to the initial quantity of DEP, indicating its complete mineralization. At the same time the final quantity of NO3-(ads) is equal to only 25% of the initial DECP quantity. It means that although the main way of HCN disappearance with the use of unmodified TiO2 was found to be leakage, nevertheless the slow HCN photooxidation occurs with NO3- to be the final product of PCO. Two surface intermediate acetic and formic acids were identified during the DECP PCO (Figure 7). The behavior of both kinetic curves is similar with the maximum at 25 min of PCO reaction. Both intermediates are the products of the partial photooxidation of (-O-C2H5) fragment. They are the known intermediates of the ethanol PCO, and their formation and transformation are described in the literature in detail (13, 22).

Acknowledgments Authors are grateful to Dr. A. Vorontsov for the help in the conduction of catalysts surface species analysis. This work 4354

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 12, 2008

FIGURE 7. Kinetic curves of acetic (solid line) and formic (dot line) acids on TiO2 during the 0.5 µL DECP PCO in the static reactor. T ) 353 K; relative humidity ≈ 0%. was financially supported by NATO (“Science for Peace”, Grant 981461) and ISTC (Grant 3305). D.V. appreciates support from the “Russian Science Support Foundation”.

Supporting Information Available Chemical structures of CWA tabun and its simulant diethyl cyanophosphate (DECP); kinetic curves of HCN vapor disappearance in the static reactor indicating that HCN does not undergo PCO if pure unmodified TiO2 is used; and a table with detailed information about the activities of TiO2 based photocatalysts against DECP vapor PCO. This material is available free of charge via the Internet at http:// pubs.acs.org.

Literature Cited (1) Alberici, R. M.; Jardim, W. F. Photocatalytic destruction of VOCs in the gas-phase using titanium dioxide. Appl. Catal. B: Environ. 1997, 14, 55–68. (2) Benoit-Marguie, F.; Wilkenhoner, U.; Simon, V.; Braun, A. M.; Oliveros, E.; Maurette, M. T. VOC photodegradation at the gassolid interface of a TiO2 photocatalyst. Part 1: 1-Butanol and 1-butylamine. J. Photochem. Photobiol. A: Chem. 2000, 132, 225– 232. (3) Dominguez, C.; Garsia, J.; Pedraz, M. A.; Torres, A.; Galan, M. A. Photocatalytic oxidation of organic pollutants in water. Catal. Today 1998, 40, 85–101. (4) Kozlova, E. A.; Smirniotis, P. G.; Vorontsov, A. V. Comparative study on photocatalytic oxidation of four organophosphorus simulants of chemical warfare agents in aqueous suspension of titanium dioxide. J. Photochem. Photobiol. A: Chem. 2004, 162, 503–511. (5) Strehlow, W. H.; Cook, E. L. Compilation of energy band gaps in elemental and binary compound semiconductors and insulators. J. Phys. Chem. Ref. Data 1973, 2 (1), 163–200. (6) Fujishima, A.; Hashimoto, K.; Watanabe, T. TiO2 photocatalysis fundametals and applications; BKC: Tokyo, 1999. (7) Munuera, G.; Rives-Arnau, V.; Saucedo, A. A. Photo-adsorption and photo-desorption of oxygen on highly hydroxylated TiO2 surfaces. Part 1. Role of hydroxyl groups in photo-adsorption. J. Chem. Soc., Faraday Trans. 1979, 75, 736–747. (8) Gerischer, H.; Heller, A. The role of oxygen in photooxidation of organic molecules on semiconductor particles. J. Phys. Chem. 1991, 95, 5261–5267. (9) Available athttp://en.wikipedia.org. (10) Lee, W.; Gao, Y. M.; Dwight, K.; Wold, A. Preparation and characterization of titanium(IV) oxide photocatalysts. Mater. Res. Bull. 1992, 27 (6), 685–692. (11) Carlson, T.; Griffin, G. L. Photooxidation of methanol using vanadium pentoxide/titanium dioxide and molybdenum trioxide/titanium dioxide surface oxide monolayer catalysts. J. Phys. Chem. 1986, 90, 5896–5900. (12) Fernndez-Garca, M.; Martnez-Arias, A.; Hanson, J. C.; Rodriguez, J. A. Chem. Rev. 2004, 104, 4063–4104.

(13) Vorontsov, A. V.; Dubovitskaya, V. P. Selectivity of photocatalytic oxidation of gaseous ethanol over pure and modified TiO2. J. Catal. 2004, 221, 102–109. (14) Kozlov, D. V.; Vorontsov, A. V.; Smirniotis, P. G.; Savinov, E. N. Gas-phase photocatalytic oxidation of diethyl sulfide over TiO2: kinetic investigations and catalyst deactivation. Appl. Catal. B: Environ. 2003, 42, 77–87. (15) Allen, H. C.; Tidwell, E. D.; Plyler, E. K. Infrared spectra of hydrogen. cyanide and deuterium cyanide. J. Chem. Phys. 1956, 25, 302–307. (16) Goubeau, J.; Haeberle, H.; Ulmer, H. The vibration spectrum of the phosphortricyanids. Z. Anorg. Allg. Chem. 1961, 311, 110– 116. (17) Smith, A. L. The coblentz society desk book of infrared spectra, 2nd ed.; The Coblentz Society: Kirkwood, 1982. (18) Kozlov, D. V.; Panchenko, A. A.; Bavykin, D. V.; Savinov, E. N.; Smirniotis, P. G. Influence of humidity and acidity of the titanium dioxide surface on the kinetics of photocatalytic oxidation of volatile organic compounds. Russ. Chem. Bull. 2003, 52, 1100–1105.

(19) Chan, S. C.; Barteau, M. A. Preparation of highly uniform Ag/ TiO2 and Au/TiO2. Supported nanoparticle catalysts by photodeposition. Langmuir 2005, 21, 5588–5595. (20) Henglein, A. Reactions of organic free radicals at colloidal silver in aqueous solution. Electron pool effect and water decomposition. J. Phys. Chem. 1979, 83, 2209–2216. (21) Nakamoto, K. Infrared Spectra of Inorganic and Coordination Compounds; John Wiley and Sonns: New York, London, 1966. (22) Nimlos, M. R.; Wolfrum, E. J.; Brewer, M. L.; Fennell, J. A.; Bintner, G. Gas-phase heterogeneous photocatalytic oxidation of ethanol: pathways and kinetic modeling. Environ. Sci. Technol. 1996, 30, 3102–3110. (23) Sauer, M. L.; Ollis, D. F. Photocatalyzed oxidation of ethanol and acetaldehyde in humidified air. J. Catal. 1996, 158, 570–582.

ES7021818

VOL. 42, NO. 12, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4355