Environ. Sci. Technol. 2003, 37, 3448-3453
Photocatalytic Oxidation of Gaseous 2-Chloroethyl Ethyl Sulfide over TiO2 IGOR N. MARTYANOV AND KENNETH J. KLABUNDE* Department of Chemistry, Kansas State University, Manhattan, Kansas 66506
Photocatalytic oxidation of gaseous 2-chloroethyl ethyl sulfide (2-CEES, ClCH2CH2SCH2CH3) over TiO2 illuminated with UV light and maintained at 25 or 80 °C in air has been investigated. 2-CEES was found to suffer progressive oxidation to yield ethylene (CH2CH2), chloroethylene (ClCHCH2), ethanol (CH3CH2OH), acetaldehyde (CH3C(O)H), chloroacetaldehyde (ClCH2C(O)H), diethyl disulfide (CH3CH2S2CH2CH3), 2-chloroethyl ethyl disulfide (ClCH2CH2S2CH2CH3), and bis(2-chloroethyl) disulfide (ClCH2CH2S2CH2CH2Cl) as the main primary intermediates, and water (H2O), carbon dioxide (CO2), sulfur dioxide (SO2), surface sulfate ions (SO42-), and hydrogen chloride (HCl) as the final products. Trace concentrations of gaseous 2-chloroethanol (ClCH2CH2OH), ethanesulfonyl chloride (CH3CH2SO2Cl), ethyl thioacetate (CH3CH2SC(O)CH3), and considerable amounts of acetic acid (CH3C(O)OH), crotonaldehyde (CH3CHCHC(O)H), methyl acetate (CH3C(O)OCH3), and methyl formate (CH3OC(O)H) were also detected in the gas phase during the photooxidation conducted at 80 °C. Increase in temperature from 25 to 80 °C accelerates formation of gaseous ethanol, acetaldehyde, chloroacetaldehyde, diethyl disulfide, 2-chloroethyl ethyl disulfide, and bis(2-chloroethyl) disulfide but suppresses ethylene and chloroethylene production at initial stages of the process. Some aspects of the possible reaction mechanism leading to this wide array of intermediates and final products are discussed.
Introduction Bis(2-chloroethyl) sulfide ((ClCH2CH2)2S), the main component of mustard gas (HD), was obtained as a pure substance at the end of the 19th century. The possibility of using this compound as a chemical warfare agent was recognized soon and was followed by HD first use on a battlefield at Ypres in 1917. Since that time, a number of simple and cost-effective methods of HD production have been developed (1), and large quantities of this agent have been accumulated. A search for a means of HD detoxification was started very soon after HD was used for a first time as a chemical weapon. In the beginning, the main efforts were focused upon HD decontamination in battlefield conditions (2). Later, however, the potentially devastating effects of chemical warfare agents (CWAs) on civilian populations and the environment have been recognized. CWAs were banned for production, and a search for appropriate methods of detoxification (3, 4) of existing stockpiles has been launched. Bis(2-chloroethyl) sulfide, the main component of HD (5), is a liquid boiling at about 220 °C. Having toxicity similar * Corresponding author phone: (785)532-6849; fax: (785)532-6666; e-mail:
[email protected]. 3448
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 15, 2003
to phosgene or hydrocyanic acid, mustard gas because of its low volatility affects a contaminated region for a long time and is especially dangerous in aerosol form. Toxic and vesicant properties of bis(2-chloroethyl) sulfide are believed to be related to its ability of rapid penetration through a membrane into a living cell followed by water hydrolysis releasing HCl. Since hydrolysis of bis(2-chloroethyl) sulfide was found to proceed through a transient cyclic ethylenesulfonium ion intermediate (6), the presence of sulfur in the molecule is important for enhancing vesicant properties of this compound. Accordingly, appropriate methods of HD detoxification may include removal of chloride from bis(2chloroethyl) sulfide, for example, in a process of nucleophilic substitution, HCl elimination, or destruction of bis(2chloroethyl) sulfide via a deep oxidation. High toxicity of HD is a serious obstacle for a search of methods of HD detoxification. In research laboratories its simulants, the closest simulant, 2-chloroethyl ethyl sulfide (half mustard, 2-CEES, ClCH2CH2SCH2CH3), or even diethyl sulfide (CH3CH2SCH2CH3) have often been used for performing experiments. One of the promising methods for detoxification of organic pollutants their deep photocatalytic oxidation over TiO2 illuminated with UV light. In the presence of atmospheric concentrations of oxygen and at room temperature, organic molecules were found to suffer a progressive oxidation up to the complete mineralization into CO2, H2O, and inorganic acids. Attempts to use a TiO2-mediated photocatalytic process for CWAs destruction were successful (7-10). In particular, photocatalytic oxidation of 2-CEES and 2-chloroethyl methyl sulfide on TiO2 in acetonitrile suspension has been studied (7). Corresponding sulfoxides and sulfones were found to be the reaction intermediates. The initial stages of the photocatalytic oxidation of gaseous diethyl sulfide over TiO2 have also been investigated (9). A number of partial oxidation products have been identified. The present study was aimed at expanding these investigations. In this respect, two aspects of this work should be noted: (i) The use of photoactive outdoor-situated TiO2 coatings is quite attractive from a practical point of view. Indeed, being exposed to sunlight such coatings demonstrate selfcleaning properties, oxidizing adsorbed organic pollutants with final nonvolatile inorganic ions being removed with rainwater. In this study, a film made of TiO2 powder was employed to oxidize a HD simulant. Despite its poor adhering properties, this film allowed the study of a few important aspects of the reaction while avoiding complications from a binding agent. (ii) Since the presence of the chlorine moiety can considerably alter the physical and chemical properties of the molecule and actually is responsible for the toxic properties of mustard gas, all experiments were carried out with the closest HD stimulant, gaseous 2-CEES. Special emphases were given to identification and evaluation of the lifetimes of toxic gaseous intermediates as compounds posing a major threat to humans.
Experimental Section Reagents and Materials. 2-CEES, ethylene, 2-chloroethanol, acetaldehyde, chloroacetaldehyde, diethyl disulfide, ethanesulfonyl chloride, ethyl thioacetate, acetic acid, crotonaldehyde, methyl acetate, methyl formate, methyl sulfoxide (DMSO), and titanium oxysulfate-sulfuric acid complex hydrate TiOSO4‚xH2SO4‚xH2O from Aldrich; ethanol and isopropyl alcohol from Fisher; and TiO2 P25 from Degussa (50 m2/g) were used as received. Distilled water was ad10.1021/es0209767 CCC: $25.00
2003 American Chemical Society Published on Web 06/24/2003
ditionally cleaned prior to its use with a water purification system from Millipore Corporation. Kinetic Experiments. A standard rectangular cell (1-cm path length, 5 mL total volume) was used in kinetic photooxidation experiments. This cell was made of UV-grade quartz and equipped with a septum/screw cap allowing multiple samplings. TiO2 was deposited on a 4-cm2 cell wall from TiO2/acetone suspension (40 g/L) so that the final density of TiO2 adhered to the cell wall was 4 mg/cm2. The layer of TiO2 was dried at room temperature and then at 70 °C overnight. After being dried, the cell was closed and illuminated for 2 h with the full arc of a UV lamp to remove the organic impurities. Final concentrations of CO2, and H2O received after TiO2 purification were taken as a new zero level. Later, illumination was interrupted, the required amount of liquid 2-CEES was injected through the septum, the cell was put into a temperature-stabilized holder, and agitation was started. Starting at that time, 10 µL of gaseous mixture was periodically taken for GC-MS analysis. After 2-CEES concentration stabilized, illumination was turned on. Periodically (every 5-10 min) the illumination was interrupted and the current composition of reaction gaseous mixture was determined. Quantitative and Qualitative Analysis. A gas chromotograph equipped with a mass selective detector (GC-MS QP5000 from Shimadzu) was employed for qualitative and quantitative analysis. The following procedure was used for qualitative identification and quantification of analyzed gaseous reaction products. About 10 µL of gaseous mixture to be analyzed was injected into the GC-MS injection port maintained at 200 °C. Immediately after injection, the temperature of the column (phase XTI-5, Restek Corp.) started to follow this program: 1 min at 40 °C, ramp at 40 °C/min to 280 °C, and 3 min at 280 °C. The separated products were identified by comparison of experimental and reference mass spectra, by following characteristic masses when separation of reaction products was not complete and comparison of the retention times of reaction product with the retention times of pure compounds. Quantification of the reaction products was based on calibration curves determined with pure compounds. 2-Chloroethyl ethyl disulfide and bis(2-chloroethyl) disulfide were not commercially available. For quantification purposes, it was assumed that ionization probabilities of the abovementioned disulfides were similar to the ionization probability of diethyl disulfide. To identify nonvolatile compounds, which cannot leave the TiO2 surface during the photocatalytic reaction, the illumination at some point was interrupted. Isopropyl alcohol, DMSO, or benzene (0.2 mL) was injected in the cell through the septum followed by TiO2 suspension agitation for 30 min. About 4 µL of the TiO2 suspension was later analyzed with GC-MS working in the same mode as described above but with an appropriate solvent cutoff time. IR spectroscopy (Nexus 670 FTIR from Nicolet Instrument Corp.) was employed as an additional tool for characterization of final reaction products. Identification of final gaseous specious was done in situ in a 1-cm path length rectangular cell (5 mL total volume) made of IR-type quartz transparent in the 4000-2100 cm-1 region. For identification of final nonvolatile products absorbing in the 2100-400 cm-1 region, the cell was opened, used TiO2 was scraped off the cell wall and pelletized with KBr (5 wt % of TiO2) followed by IR spectrum recording. All IR spectra were an average of 128 scans recorded at 1 cm-1 resolution. Illumination and Optical Spectra Measurements. A light source from Oriel Instruments equipped with a 1000W He(Xe) lamp was employed for illumination of TiO2. A combination of two filters, nos. 57396 and 59062 both from Oriel Instruments, Inc., was used for receiving mild UV light (UVA,
FIGURE 1. Optical spectra of gaseous 2-CEES (ca. 28 µM), powdered TiO2 (Degussa P25), and filter combination used for kinetic experiments. 320 nm < λ
