Environ. Sci. Technol. 2004, 38, 831-837
Photolysis of Chloral under Atmospheric Conditions JOHN C. WENGER* Department of Chemistry and Environmental Research Institute, University College Cork, Cork, Ireland S T EÄ P H A N E L E C A L V EÄ † A N D HOWARD W. SIDEBOTTOM Department of Chemistry, University College Dublin, Dublin, Ireland KLAUS WIRTZ AND M O N T S E R R A T M A R T IÄ N R E V I E J O Centro de Estudios Ambientales del Mediterraneo, C. Charles R. Darwin 14, 46980 Paterna, Valencia, Spain JAMES A. FRANKLIN Solvay S.A., Brussels, Belgium
The photolysis of chloral under atmospheric conditions was studied at the large outdoor European Photoreactor (EUPHORE) in Valencia, Spain. The photodissociation rate coefficient, J(chloral), was measured directly under different sunlight conditions during April 1999. Values in the range of J(chloral) ) (4.61-6.11) × 10-5 s-1 were obtained, yielding an average value of J(chloral)/J(NO2) ) (6.15 ( 0.62) × 10-3. This corresponds to a photolysis lifetime of 4.5-6 h under conditions appropriate to the solar flux during summer months and confirms that atmospheric photolysis is the major degradation pathway for chloral. The overall quantum efficiency of photolysis under atmospheric conditions was determined to be 1.00 ( 0.05. The atmospheric photolysis of chloral produced phosgene, CO, and Cl atoms with molar yields of 0.83 ( 0.04, 1.01 ( 0.05, and 1.18 ( 0.06, respectively. The product yield data are consistent with a mechanism in which the primary photolysis channel produces a Cl atom and a CCl2CHO radical. The latter species is converted to the oxy radical OCCl2CHO, which decomposes by both C-C and C-Cl bond fission. A chemical mechanism for the photolysis of chloral by sunlight is proposed, and the atmospheric implications are discussed.
Introduction Chloral (CCl3CHO) is produced in the atmosphere from the hydroxyl radical initiated oxidation of methyl chloroform (1,1,1-trichloroethane) (1, 2). While methyl chloroform was formerly widely used as a solvent, it is now strictly regulated as an ozone-depleting substance under the Montreal Protocol. As a consequence, estimates of annual global emissions have fallen by approximately 95% over the last 10 yr (3). These calculations are borne out by observations of atmospheric background concentrations of methyl chloroform (4). However, recent field measurements indicate that there * Corresponding author e-mail:
[email protected]; telephone: +353 21 4902454; fax: +353 21 4903014. † Present address: Centre de Ge ´ ochimie de la Surface, CNRS and Universite´ Louis Pasteur, 28 Rue Goethe, F-67083 Strasbourg, France. 10.1021/es0300719 CCC: $27.50 Published on Web 12/30/2003
2004 American Chemical Society
appear to be continuing emissions of methyl chloroform from Europe (5). Thus, the atmospheric fate and impact of methyl chloroform is still of concern. To determine the full environmental impact of methyl chloroform, a detailed understanding of the atmospheric degradation pathways for its degradation product, chloral, is required. The major loss processes for chloral are expected to be photolysis by sunlight and reaction with hydroxyl (OH) radicals (6, 7). In addition, uptake by clouds has also been considered with the subsequent oxidation of chloral in cloudwater suggested as a major source of trichloroacetic acid observed in precipitation (8). Measurement of the rates of these competing loss processes yields lifetimes for each of the pathways and allows the overall atmospheric lifetime of chloral to be calculated. The photolysis lifetime of chloral in the atmosphere can be determined from the photodissociation rate coefficient, J, using the following expression:
J)
∫σ (λ)Φ(λ)F(λ) dλ T
Knowledge of the absorption cross-section (σ) at temperature (T), quantum efficiency (Φ), and solar flux intensity (F) are all required over the absorbing wavelength range (dλ). Absorption cross-section data for chloral are available in the literature (7, 9, 10). However, information on the quantum efficiency of the photolysis process is limited to the study of Talukdar et al. (7), who measured the quantum efficiency for production of O, H, and Cl atoms at 193, 248, and 308 nm. The lack of quantum efficiency data over the whole wavelength range relevant to the troposphere means that the photodissociation rate coefficient and hence the photolysis lifetime of chloral can only be estimated using the available data. To address this issue, the photolysis of chloral has been investigated using natural sunlight at the European Photoreactor (EUPHORE) in Valencia, Spain. The photodissociation rate coefficient has been measured directly and used to calculate the overall quantum efficiency of photolysis and the photolytic lifetime of chloral in the troposphere. The products of photolysis under atmospheric conditions have also been determined, and mechanisms for their formation are proposed. These results are discussed within the context of the atmospheric fate and impact of methyl chloroform.
Experimental Section The photolysis of chloral by natural sunlight was investigated at EUPHORE during April 1999. EUPHORE consists of two large outdoor simulation chambers integrated into the Centro de Estudios Ambientales del Mediterraneo (CEAM) in Valencia, Spain. Technical information concerning the installation has been previously reported in the literature (11-14); hence, only details directly related to the present experiments are given here. The experiments were performed in a hemispherical reactor made of FEP (fluorinated ethylene propylene) foil with a volume of approximately 195 m3. The FEP foil is highly transparent to sunlight, with a transmission of greater than 75% over the wavelength range of 290-550 nm. A retractable steel housing surrounds the chamber and is used to control the time of exposure to sunlight. The floor of the reactor consists of aluminum panels covered with FEP foil and has a specially designed cooling system to compensate for heating of the chamber caused by solar radiation. A number of ports situated on the floor of the chamber are available for the introduction and sampling of reaction mixtures. The chamber VOL. 38, NO. 3, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
831
TABLE 1. Experimental Conditions and Results for the Sunlight Photolysis of Chloral at 298 ( 5 K and Atmospheric Pressurea date initial concn (ppbv)
J(chloral) (s-1) J(NO2) (s-1) J(chloral)/J(NO2) max theoretical loss rate (s-1) effective quantum efficiency molar yield of CCl2O molar yield of CO molar loss of cyclohexane molar yield of Cl atoms molar yield of cyclohexanone (from cyclohexane) molar yield of cyclohexanol (from cyclohexane)
April 6, 1999 [CCl3CHO] ) 1700 [c-C6H12] ) 2470 (6.11 ( 0.30) × 10-5 (9.27 ( 0.93) × 10-3 (6.30 ( 0.63) × 10-3 5.81 × 10-5 1.05 ( 0.05 0.84 ( 0.04 1.02 ( 0.04 1.17 ( 0.06 1.17 ( 0.06 0.36 ( 0.04 0.09 ( 0.02
April 7, 1999 [CCl3CHO] ) 1575 [c-C6H12] ) 27900 (4.61 ( 0.24) × 10-5 (8.01 ( 0.80) × 10-3 (6.00 ( 0.60) × 10-3 4.81 × 10-5 0.96 ( 0.05 0.82 ( 0.04 1.00 ( 0.05 1.19 ( 0.05 1.19 ( 0.05 0.34 ( 0.04 0.10 ( 0.03
a Except for J(NO ), quoted errors are twice the standard deviation arising from the least-squares fit of the data and include the uncertainty 2 in calibration and response factors. For J(NO2), the estimated error is 10% (see text).
is filled to atmospheric pressure with purified dry air generated from ambient air using an air purification system. Homogeneous gas mixtures are obtained by the use of two mixing fans located inside the chamber. Between experiments, the chamber is flushed with clean air. The reaction chamber was equipped with a variety of instruments for chemical and physical analysis. Temperature and humidity inside the chamber were measured continuously using PT-100 thermocouples and a dew-point mirror system (Walz TS-2), respectively. The solar flux intensity was measured using a calibrated spectroradiometer (Bentham DM300) consisting of a double monochromator, photomultiplier, and silicon diode detector. Specially designed measurement heads with uniform sensitivity to the incident angles of the solar light were located inside the chamber to measure incident and reflected light. The heads were coupled through a quartz fiber bundle to the entrance optics of the monochromator such that the direct and reflected light beams passed geometrically separated through the double monochromator simultaneously. The solar flux intensity was recorded over the range of 290-520 nm with a spectral resolution of 1 nm fwhm. A full spectral scan took 420-430 s. The calibration of the spectroradiometer is based on a certified standard Wolfram tungsten lamp with a quoted precision of better than 5% in the relevant spectral range. Comparison of J(NO2) values calculated from the spectroradiometer data with those obtained by an independently calibrated J(NO2) filter radiometer (Schmitt Glashu ¨ tten, Germany) showed that values from these two instruments were within 10% of each other. The uncertainty in measurement of the absolute light intensity is thus less than 10%. A White mirror system installed inside the chamber and aligned with an optical path length of 553.5 m was used for in situ measurements by Fourier transform infrared (FTIR) spectroscopy. The FTIR spectrometer (Nicolet Magna 550) was positioned on a platform beneath the chamber and operated using an MCT-B detector. Infrared spectra were derived from the co-addition of 600 scans, collected over a 10-min period, and recorded using a resolution of 1 cm-1. The concentration of carbon monoxide was measured using a CO monitor (model 48C, Thermo Environmental Instruments Inc.). Additional chemical detection was provided by a gas chromatograph (Fisons 8000) equipped with flame ionization and photoionization detectors (FID and PID). The chromatograph was operated using a 30-m DB-624 fused silica capillary column (J&W Scientific, 0.32 mm i.d., 1.8 µm film). A trace gas analyzer (TGA, Fisons), which incorporates a cryogenic enrichment trap coupled to a flame ionization detector, was also used for chemical analysis of reactants and products. Air samples (200 cm3) were collected in a sampling loop at 120 °C and passed to a Tenax microtrap 832
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 3, 2004
cooled to -120 °C with liquid nitrogen. Injection onto the 30-m DB-1 chromatographic column (J&W Scientific, 0.25 mm i.d., 1.0 µm film) was achieved by rapid heating of the microtrap to 240 °C. The photolysis of chloral was studied using cyclohexane as a scavenger for Cl atoms. Chloral (99%, Fluka) was introduced into the chamber via a stream of purified air, while cyclohexane (Fluka, purity >99%) was added using a nebulizer. After allowing approximately 30 min for mixing of the compounds, photolysis was initiated by opening the protective housing, thereby exposing the chamber to sunlight. The temperature inside the chamber increased slightly as the experiments progressed but was always within the range of 293-303 K. Similarly, the dew point varied slightly between -40 and -38 °C. Chemical analysis was performed throughout the reaction. The reactants and products were quantified using calibrated reference infrared spectra and gas chromatographic sensitivity factors obtained by introducing known volumes of pure materials into the chamber. Complex infrared spectra were analyzed by successively subtracting the absorption features of the compounds using their calibration spectra. To compensate for sampling losses, thermal expansion of the gas mixture, and leakage through the FEP foil, the chamber was refilled continuously with a clean air stream of 1 L s-1, using a thermal mass-flow controller. The overall dilution rate was determined by adding about 20 ppbv (1 ppbv ) 2.46 × 1010 molecules cm-3 at 760 Torr and 298 K) of the unreactive tracer gas SF6 to the chamber and measuring its loss by FTIR spectroscopy throughout the course of the photolysis experiments. The derived correction factors were applied to determine the amounts of reactants consumed and products formed. Phosgene (CCl2O) was calibrated using an approximately 20% v/v solution in toluene (Fluka). The exact proportion of toluene present was determined by gas chromatography and FTIR spectroscopy, and the balance was assumed to be phosgene. Cyclohexanone (purity >99%) and cyclohexanol (purity >99%), products of the reaction of Cl atoms with cyclohexane, were also obtained from Fluka.
Results and Discussion Determination of Photodissociation Rate Coefficient and Effective Quantum Efficiency. The photolysis of chloral was performed at EUPHORE under spring-time conditions in April 1999. A summary of the initial conditions and results from experiments carried out on two different days are provided in Table 1. The experiments were conducted for at least 4 h during the middle of the day. The loss of chloral was monitored using FTIR spectroscopy and the trace gas analyzer. To avoid interference from the reaction of Cl atoms with chloral, cyclohexane (c-C6H12) was used as a scavenger
FIGURE 1. Concentration-time profiles of the reactant and products during the photolysis of chloral at EUPHORE: [CCl3CHO] ) 1700 ppbv: [c-C6H6] ) 2470 ppbv. Also shown is the photolysis rate of NO2 during the experiment.
FIGURE 2. Photolytic loss rate of chloral. for Cl atoms. The rate coefficient for the reaction of Cl with cyclohexane is approximately 50 times larger than that for the reaction with chloral, k(Cl + chloral) ) 7.1 × 10-12 cm3 molecule-1 s-1 and k(Cl + cyclohexane) ) 3.08 × 10-10 cm3 molecule-1 s-1 (15, 16). The concentration of cyclohexane was always greater than that of chloral, and thus all Cl atoms should be effectively scavenged by cyclohexane. The product yields reported in Table 1 for experiments where there was a 10-fold change in the concentration of added cyclohexane are in good agreement, indicating that the Cl atoms were efficiently trapped by c-C6H12. The experiments yielded essentially similar results, and only the experimental data obtained on April 6, 1999, are graphically presented here. The concentration-time profiles for chloral and the photolysis products (phosgene and carbon monoxide) are shown in Figure 1. Also shown is the photolysis rate of NO2, J(NO2), which is a conventional measure of the solar light intensity. J(NO2) was calculated from the solar
flux intensity measurements of the spectroradiometer and reported data on the absorption cross-section and quantum efficiency for the photolysis of NO2 by sunlight (17). The loss rate of chloral in this experiment was determined to be (6.11 ( 0.30) × 10-5 s-1 from the gradient of the line of best fit through all the points shown in Figure 2. The wall loss and uptake of chloral by water in the reactor was found to be negligible on the time scale of the experiments and, therefore, did not influence J(chloral). Consequently, it can be safely assumed that photolysis is the only loss process and that J(chloral) ) (6.11 ( 0.30) × 10-5 s-1 under the prevalent sunlight conditions. The values of J(chloral) determined for the experiments shown in Table 1 differ slightly because the sunlight intensity was different on both days, as reflected by the J(NO2) values. The atmospheric lifetime of chloral due to photolysis was calculated from 1/J(chloral). In the experiments performed in this study, J(chloral) was in the range of (4.61-6.11) × 10-5 VOL. 38, NO. 3, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
833
FIGURE 3. Yield plot for products detected during the photolysis of chloral. s-1, corresponding to photolysis lifetimes of 4.5-6 h. Obviously the photolysis rate depends on the intensity and spectral distribution of the sunlight; consequently, it is more useful to quote the photolysis rate relative to a conventional measure of solar light intensity, such as J(NO2). For experiments carried out in this work, an average value of J(chloral)/J(NO2) of (6.15 ( 0.62) × 10-3 was obtained. The experiments performed at EUPHORE (latitude 39.5° N) measured the photolysis rate during the middle of the day, under relatively clear sky conditions in the spring-time. The maximum photolysis rate occurs at solar noon in the summer months, where J(NO2) is typically about 1 × 10-2 s-1 (11-14). Under these conditions, J(chloral) ) 6.1 × 10-5 s-1, which yields a lifetime due to photolysis of around 4.5 h. This is in good agreement with the calculated values of J(chloral) ) 7.0 × 10-5 s-1 estimated by Rattigan et al. (9) and J(chloral) ) (3.0-5.0) × 10-5 s-1 calculated by Talukdar et al. (7). The calculations of Rattigan et al. (9) and Talukdar et al. (7) were based on atmospheric conditions appropriate to the solar flux during summer at 30° N and assumed a quantum efficiency of unity for the photodissociation of chloral across the complete atmospheric absorption range. However, from the experiments reported in this work, the effective quantum efficiency for photodissociation of chloral by sunlight can be determined by simply calculating the ratio of the measured photolysis rate to the maximum theoretical loss rate. The latter was calculated from the solar flux intensity measurements of the spectroradiometer, reported absorption crosssection data (7, 9), and assuming a quantum efficiency of unity across the atmospheric absorption range of chloral. The data from the present series of experiments show that, within experimental error, the theoretical loss rate is the same as the measured loss rate and yields an average value of 1.00 ( 0.05 for the effective quantum efficiency. Thus, the assumptions of Rattigan et al. (9) and Talukdar et al. (7) that Φ ) 1 appear to be justified. Photolysis Products. The major carbon-containing products from the photolysis of chloral in air were phosgene and carbon monoxide. Yield plots for CCl2O and CO were linear, thus indicating that both compounds are primary products of chloral photolysis. Figure 3 shows the data from the experiment described in the first column of Table 1. The wall loss and uptake of cyclohexane, phosgene, and other reaction 834
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 3, 2004
products by water in the reactor were found to be negligible on the time scale of the experiments and, therefore, did not influence the product yields. There is considerable loss of the Cl atom scavenger (cyclohexane) as shown in Figure 4. Assuming that all of the Cl atoms are effectively scavenged by cyclohexane, then a plot of the loss of chloral versus the loss of cyclohexane allows determination of the Cl atom yield from chloral photolysis (Figure 5). The Cl atom-initiated oxidation of cyclohexane yields cyclohexanone, cyclohexanol, and a ring-opening product but no CO (18, 19). Cyclohexanone and cyclohexanol were detected with yields similar to those reported in previous studies (18, 19) (Figure 6). There are four possible reaction pathways for the photolysis of chloral under atmospheric conditions (7, 10);
CCl3CHO + hν f CCl3 + HCO
∆H°(298 K) ) 69 kcal mol-1, λ e 415 nm (1) f CCl2CHO + Cl
∆H°(298 K) ) 71 kcal mol-1, λ e 403 nm (2) f CCl3H + CO
∆H°(298 K) ) -9 kcal mol-1 (3)
f CCl3C(O) + H
∆H°(298 K) ) 87.5 kcal mol-1, λ e 327 nm (4) Information on the mechanisms for photodissociation of chloral and other halogenated aldehydes is relatively limited. However, Talukdar et al. (7) have recently reported the primary quantum yields for O, H, and Cl atoms from the photolysis of CCl3CHO by detecting them via atomic resonance fluorescence following pulsed excimer laser photolysis at 308 nm and 298 K. The quantum yield of Cl atoms was 1.3 ( 0.3, while the yields of O and H atoms were negligible. As pointed out by Talukdar et al. (7), there is no energetically available channel for O atom formation and the lack of production of H atoms clearly demonstrates that reaction 4 is negligible. They presented compelling evidence that Cl atoms were generated in the primary photolytic process and concluded that, since it is not energetically favorable to produce directly more than one Cl atom from the photolysis at 308 nm, then the primary yield of Cl atoms at this
FIGURE 4. Concentration-time profile of the reactant, cyclohexane, cyclohexanone, and cyclohexanol during the photolysis of chloral. wavelength is unity. Thus, the authors conclude that the major photolytic pathway for chloral is C-Cl bond dissociation (reaction 2). Talukdar et al. (7) observed phosgene and CO as the major products following multipulse laser photoylsis of CCl3CHO at 248 nm in the presence of O2. The absence of CCl3H in the products confirms that the molecular channel for photodissociation of chloral; reaction 3 is unimportant. These results are supported by the work of Ohta and Mizoguchi (20), who detected CCl2O, CO, and CO2 as the products of the photolysis of CCl3CHO in oxygen over the wavelength range of 300-400 nm. In both these product studies, which were carried out in the absence of a Cl atom scavenger, it is clear that the products are largely generated in a chain reaction (7):
CCl3CHO + hν f CCl2CHO + Cl
(2)
Cl + CCl3CHO f CCl3C(O) + HCl
(5)
CCl3C(O) + M f CCl3 + CO + M
(6)
CCl3 + O2 + M f CCl3O2 + M
(7)
2CCl3O2 f 2CCl3O + O2
(8)
CCl3O f CCl2O + Cl
(9)
It is evident that the reported chloral photooxidation studies provide little information on the yield of products arising from the CCl2CHO radical generated in the primary photolytic process. Talukdar et al. (7) suggested that the CCl2CHO radical generates CCl2O quantitatively in the following reaction sequence:
CCl2CHO + O2 + M f O2CCl2CHO + M 2O2CCl2CHO f 2OCCl2CHO + O2 OCCl2CHO + M f CCl2O + HCO HCO + O2 f CO + HO2
(10) (11) (12) (13)
The average molar yield of CO from the present series of experiments is close to unity, while the yields for CCl2O and Cl (as determined by the loss of cyclohexane) are 0.83 ( 0.04
and 1.18 ( 0.06, respectively. These results indicate that if the major photodissociation pathway generates a Cl atom and a CCl2CHO radical (reaction 2), as shown by the work of Talukdar et al. (7), then CCl2CHO does not react quantitatively to form CCl2O. In fact, the radical OCCl2CHO formed in reaction 11 can decompose via either carbon-carbon or carbon-chlorine bond fission:
OCCl2CHO + M f CCl2O + HCO f ClC(O)CHO + Cl
(12) (14)
C-C bond cleavage would produce equal amounts of CO, CCl2O, and Cl with yields of unity, whereas C-Cl bond breaking generates chloroglyoxal, ClC(O)CHO, and Cl atoms with molar yields of 1.00 and 2.00, respectively. It is proposed that C-C bond fission of the OCCl2CHO radical accounts for the production of phosgene with a molar yield of 0.83 and leads to the formation of CO (0.83) following the reaction of HCO with O2. The remaining 17% of the OCCl2CHO radicals undergo C-Cl cleavage to generate chlorogyloxal (ClC(O)CHO) and a chlorine atom. This proposed mechanism predicts molar yields of CCl2O (0.83), CO (0.83), ClC(O)CHO (0.17), and Cl (1.17) as compared to observed yields of CO (1.0) and Cl (1.18). Chloroglyoxal was not observed among the products, although the predicted low yield would make it difficult to detect with the available instrumentation. It is possible that chloroglyoxal could react with Cl atoms or undergo photolysis in the chamber. There is no rate coefficient data available for the reaction of Cl with chloroglyoxal. However, if it is assumed that chloroglyoxal has similar reactivity to glyoxal, k(Cl + glyoxal) ) 3.8 × 10-11 cm3 molecule-1 s-1 (21), then under the conditions employed in the experiments, the reaction of Cl with chloroglyoxal will be negligible as compared to the reaction of Cl with cyclohexane. Regarding the possibility of photolysis, the dialdehyde functionality of chloroglyoxal means that it is susceptible to photolytic degradation in the chamber. A sample of ClC(O)CHO was not available; however, photolysis of the structurally similar compound oxalyl chloride (ClC(O)C(O)Cl) has been measured at EUPHORE and found to have a photolysis lifetime of around 60 min (22). By analogy, we propose that ClC(O)CHO degrades to a VOL. 38, NO. 3, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
835
FIGURE 5. Loss rate of cyclohexane during the photolysis of chloral.
FIGURE 6. Yield plots for the cyclohexane oxidation products detected during the photolysis of chloral.
certain extent during the time frame of the chloral photolysis experiments to generate the additional amounts of CO and Cl atoms that are observed. A summary of the proposed degradation scheme for the photolysis of chloral under atmospheric conditions is given in Figure 7. Atmospheric Implications. The degradation of methyl chloroform in the atmosphere occurs mainly by reaction with hydroxyl radicals in the troposphere (lifetime around 6 yr) to give chloral in almost quantitative yields. Chloral is readily photolyzed by sunlight with a quantum efficiency close to unity. Under relatively clear sky conditions at ground level and a latitude of 39.5° N, the photodissociation rate coefficient was determined to be (4.61-6.11) × 10-5 s-1, which corresponds to a photolysis lifetime of 4.5-6 h. The average value of J(chloral)/J(NO2) ) (6.15 ( 0.62) × 10-3 determined in this work can be used to calculate the photolytic lifetime of chloral under a range of atmospheric conditions. The other possible atmospheric loss processes for chloral include gas-phase reaction with OH, O3, and NO3 and uptake 836
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 3, 2004
by clouds followed by wet deposition. The atmospheric lifetime of chloral due to reaction with hydroxyl radicals is 10-15 d (7, 23), while the reactions with O3 and NO3 are of negligible importance (7). General circulation model (GCM) calculations suggest that rainout lifetimes of highly soluble gases such as chloral depend mainly on their mode of input into the troposphere. When a species is formed fairly uniformly within the troposphere, as is the case for chloral, the average lifetime is of the order of 15-20 d (24). Thus, it is clear that photolysis is the major degradation pathway for chloral in the atmosphere. This effectively rules out the possibility that atmospheric chloral is the major source of trichloroacetic acid found in precipitation and is in agreement with the most recent review of the scientific literature on this subject by McCulloch (25). The photolysis of chloral produces CCl2O, CO, and Cl atoms. Kindler et al. (24) have concluded from a modeling study that tropospheric removal of CCl2O is predominantly via wet deposition with a lifetime of around 70 d. The major
FIGURE 7. Proposed degradation scheme for the photolysis of chloral under atmospheric conditions. fate of CO is reaction with OH radicals, while the chlorine atoms can readily undergo reaction with hydrocarbons present in the atmosphere.
Acknowledgments The authors are grateful for the financial support of this work by the European Commission (Projects Radical, ENV4-CT970419, and HALOBUD, ENV4-CT97-0393). The Spanish authors in addition acknowledge the financial support by Generalidad Valenciana, Bancaja, and Ministerio de Ciencia y Tecnolo´gia (1998-1661-CE). Finally the authors would like to thank Tanya Kelly (UCD) for her assistance in the production of this manuscript.
Literature Cited (1) Nelson, L.; Shanahan, I.; Sidebottom, H. W.; Treacy, J.; Nielsen, O. J. Int. J. Chem. Kinet. 1990, 22, 577-590. (2) Platz, J.; Nielsen, O. J.; Sehested, J.; Wallington, T. J. J. Phys. Chem. 1995, 99, 6570-6579. (3) Midgely, P. M.; McCulloch, A. In Handbook of Environmental Chemistry, Vol. 4, Part E, Reactive Halogens in the Atmosphere; Fabian, P., Singh, O., Eds.; Springer-Verlag: Berlin, 1999; pp 155-190.
(4) Montzka, S. M.; Butler, J. H.; Elkins, J. W.; Thompson, T. M.; Clarke, A. D.; Lock, L. T. Nature 1999, 398, 690-694. (5) Krol, M. C.; Lelieveld, J.; Oram, D. E.; Sturrock, G. A.; Penkett, S. A.; Brenninkmeijer, C. A. M.; Gros, V.; Williams, J.; Scheeren, H. A. Nature 2003, 421, 131-135. (6) Sidebottom, H.; Franklin, J. Pure Appl. Chem. 1996, 68, 17571769. (7) Talukdar, R. K.; Mellouki, A.; Burkholder, J. B.; Gilles, M. K.; Le Bras, G.; Ravishankara, A. R. J. Phys. Chem. 2001, 105, 51885196. (8) Jordan, A.; Frank, H. Atmos. Environ. 1999, 33, 4525-4526. (9) Rattigan, O. V.; Wild, O.; Jones, R. L.; Cox, R. A. J. Photochem. Photobiol. A: Chem. 1993, 73, 1-9. (10) Rattigan, O. V.; Wild, O.; Cox, R. A. J. Photochem. Photobiol. A: Chem. 1998, 112, 1-7. (11) Becker, K. H., Ed. The European Photoreactor EUPHORE; Final Report of the EC-Project, Contract EV5V-CT92-0059; Wuppertal, Germany; 1996. (12) Klotz, B.; Sørensen, S.; Barnes, I.; Becker, K. H.; Etzkorn, T.; Volkammer, R.; Platt, U.; Wirtz, K.; Martı´n-Reviejo, M. J. Phys. Chem. 1998, 102, 10289-10299. (13) Magneron, I.; The´venet, R.; Mellouki, A.; Le Bras, G.; Moortgat, G. K.; Wirtz, K. J. Phys. Chem. A 2002, 106, 2526-2537. (14) Volkammer, R.; Platt, U.; Wirtz, K. J. Phys. Chem. A 2001, 105, 7865-7874. (15) Scollard, D. J.; Treacy, J. J.; Sidebottom, H. W.; Balestra-Garcia, C.; Laverdet, G.; LeBras, G.; MacLeod, H.; Te´ton, S. J. Phys. Chem. 1993, 97, 4683-4688. (16) Aschmann, S. M.; Atkinson, R. Int. J. Chem. Kinet. 1995, 27, 613-622. (17) DeMore, W. B.; Sander, S. P.; Golden, D. M.; Hampson, R. F.; Kurylo, M. J.; Howard, C. J.; Ravishankara, A. R.; Kolb, C. E.; Molina, M. J. In Chemical Kinetics and Photochemical Data for Use in Stratospheric Modelling; JPL Publication 97-4; Jet Propulsion Laboratory: Pasadena, CA, 1997. (18) Aschmann, S. M.; Chew, A. A.; Arey, J.; Atkinson, R. J. Phys. Chem. A 1997, 101, 8042-8048. (19) Platz, J.; Sehested, J.; Nielsen, O. J.; Wallington, T. J. J. Phys. Chem. A 1999, 103, 2688-2695. (20) Ohta, T.; Mizoguchi, I. Int. J. Chem. Kinet. 1980, 12, 717-727. (21) Niki, H.; Maker, P. D.; Savage, C. M.; Breitenbach, L. P. Int. J. Chem. Kinet. 1985, 17, 547. (22) Wenger, J.; Mellouki A.; Le Bras, G.; Treacy, J.; Sidebottom, H. In The European Photoreactor EUPHORE Report 2000; Barnes, I., Sidebottom, H., Eds.; Bergische Universitat Wuppertal: Wuppertal, Germany, 2001. (23) Barry, J.; Scollard, D. J.; Treacy, J. J.; Sidebottom, H. W.; Le Bras, G.; Poulet, G.; Te´ton, S.; Chichinin, A.; Canosa-Mas, C. E.; Kinnison, D. J.; Wayne, R. P.; Nielsen, O. J. Chem. Phys. Lett. 1994, 221, 353-358. (24) Kindler, T. P.; Chameides, W. L.; Wine, P. H.; Cunnold, D. M.; Alyea, F. N.; Franklin, J. A. J. Geophys. Res.-Atmos. 1995, 100 (D1), 1235-1251. (25) McCulloch, A. Chemosphere 2002, 47, 667-686.
Received for review June 2, 2003. Revised manuscript received November 10, 2003. Accepted November 13, 2003. ES0300719
VOL. 38, NO. 3, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
837
