Environ. Sci. Technol. 2005, 39, 6130-6137
Solar Photolysis of CH2I2, CH2ICl, and CH2IBr in Water, Saltwater, and Seawater CHARLOTTE E. JONES AND LUCY J. CARPENTER* Department of Chemistry, University of York, Heslington, York YO10 5DD, U.K.
Ultraviolet-visible absorption spectroscopy and purgeand-trap GC-MS were used to determine the rates and products of the photodissociation of low concentrations of CH2I2, CH2IBr, and CH2ICl in water, saltwater (0.5 M NaCl), and seawater in natural sunlight. Photoproducts of these reactions include iodide (I-) and, in salt- and seawater environments, CH2XCl (where X ) Cl, Br, or I). Thus, CH2ICl was produced during CH2I2 photolysis (with a molar yield of 35 ( 20%), CH2BrCl from CH2IBr photolysis, and CH2Cl2 from CH2ICl photolysis (in lower yields of 6-10%). Formation of these chlorine-atom-substituted products may be via direct reaction of Cl- with either (A) the isopolyhalomethane photoisomer or associated ion pair (e.g., CH2I+-I-) or (B) the initially produced CH2I• photofragment. Estimated quantum yields for photodissociation were 0.62 ( 0.09, 0.17 ( 0.03, and 0.26 ( 0.06 for CH2I2, CH2IBr, and CH2ICl, respectively, in 0.5 M NaCl, with only small differences from these values in water and seawater. The much higher quantum yield of CH2I2 photolysis compared to CH2IBr and CH2ICl photolysis may be explained by the higher yield of the isodiiodomethane photoisomer of CH2I2, resulting in reduced geminate recombination of the initially produced radical photofragments back to the parent molecule. We use a radiative transfer model with measured absorption cross-sections in saltwater to calculate seasonal values of CH2I2, CH2IBr, and CH2ICl photodissociation in surface seawater at midlatitudes (50° N) and show that a significant proportion of CH2ICl in surface seawater may arise from CH2I2 photodecomposition. We also suggest that surface seawater photolysis of CH2I2 over an 8 h period may contribute up to ∼10% of the surface seawater I- levels, with implications for the increased deposition of O3 to the surface ocean.
Introduction The surface ocean is an important source of reactive volatile organic iodine compounds (VOI) such as CH3I, CH2I2, CH2ICl, and CH2IBr to the marine boundary layer (MBL) (1, 2). The rapid photolysis of VOI in air and subsequent reaction of I with O3 to form the IO radical initiate various tropospheric ozone depleting cycles (3, 4). Indirect evidence for halogenrelated perturbations to the oxidative capacity of the * Corresponding author phone: +44 (0)1904 434588; fax: +44 (0)1904 432516; e-mail:
[email protected]. 6130
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 16, 2005
troposphere comes from observations of anomalous ozone (5, 6) and peroxy radical (7) concentrations. In addition to ozone destruction, it has been suggested that the photoproducts of biogenic iodine compounds such as CH2I2 may play a role in the formation of marine aerosol and cloud condensation nuclei (8). Because the surface ocean and MBL are fundamentally linked due to the continual exchange of species across the boundary, the photochemical processes occurring in the surface ocean directly affect emissions into the atmosphere. However, there are few studies on the aqueous-phase chemistry of iodomethanes in seawater/ saltwater. Gas-phase ultraviolet-visible photolysis of CH2I2, CH2ICl, and CH2IBr initially results in homolytic scission of one carbon-halogen bond to produce a halomethyl radical (CH2X•) and an iodine atom (I•). In the aqueous phase, the presence of a solvent cage results in the recombination of these photofragments to form CH2X-I isomers as shown in femtosecond pump-probe (9-12) and transient resonance Raman studies (13-17). The CH2X-I isomer has significant ion-pair (CH2X+-I-) character, and Tarnovsky et al. (12) report that in acetonitrile, ethanol, and methanol a fraction of CH2I-I may also undergo heterolytic cleavage to form CH2I+ and I-. However, Li et al. (17) and Kwok et al. (18) propose that in aqueous solution breakdown of the CH2I-I photoisomer is predominantly via a water-catalyzed OH insertion reaction to produce CH2(OH)2 and HI, with additional production of CH2ICl following CH2I2 photolysis in the presence of chloride ions (0.5 M NaCl) (19). Formation of the latter was suggested to be via reaction of the iodomethyl radicals with Cl- anions or Cl2-. Here we report photolysis rates and estimated quantum yields for the photodecomposition of CH2I2, CH2ICl, and CH2IBr by solar radiation in water, 0.5 M sodium chloride solution, and seawater. Molar yields of the CH2I2 polyhalomethane photoproduct and potential mechanisms for the formation of CH2XCl are presented, the latter with reference to previous work. We discuss the environmental implications of these data, particularly with reference to CH2ICl and I- formation from CH2I2 photodecomposition in the surface ocean.
Experimental Details Dilute (∼1 × 10-9 M) solutions of CH2I2, CH2ICl, and CH2IBr (all 99%, Aldrich) were prepared in deionized water (pH 7.65), 0.5 M sodium chloride solution (NaCl analytical grade, Fisher) (pH 7.85), and unfiltered seawater from the northeast coast of England (pH 7.90) and transferred to 250 mL UVtransparent airtight quartz flasks with no headspace. The deionized water and 0.5 M NaCl were degassed prior to use to remove volatile impurities such as chloroform. Experiments were conducted in a grassy area outside the University of York laboratories at 15 m above sea level (www.ordnancesurvey.co.uk) with the flasks clamped ∼50 cm over a wooden surface. Samples of 10 mL volume were extracted and analyzed for halocarbons using purge-andtrap gas chromatography-mass spectrometry (GC-MS) (Perkin-Elmer TurboMass) (1). Zero grade nitrogen gas (BOC) at 40 mL/min for 20 min was used to sparge volatile components from the water, and a Perkin-Elmer TurboMatrix automated thermal desorption (ATD) unit was used to preconcentrate and transfer analytes to the gas chromatograph. Separation was achieved using a 30 m × 0.32 mm internal diameter DB5 625 capillary column with CP grade helium carrier gas (BOC) with detection by electron impact 10.1021/es050563g CCC: $30.25
2005 American Chemical Society Published on Web 07/09/2005
ionization quadrupole mass spectrometry in selected ion mode. The system was calibrated for halocarbons using the dynamically diluted output of permeation tubes as described by Wevill and Carpenter (20), and system drift was also accounted for using a 100 ppbv ( 2% mixture of chlorinated halocarbons in nitrogen (National Physics Laboratory). The run-to-run reproducibility of samples using this method was (0.9%. The time interval between sampling depended upon the species being studied: CH2I2 photolyses very rapidly, so 10 mL samples were removed every 5-15 min, whereas CH2ICl photolysis occurred at a much slower rate, so samples were taken every few hours. During sample extraction, flasks were covered in aluminum foil to prevent photodecomposition and the timer was paused for the duration of the sampling. Once samples had been removed the headspace was refilled with solvent (pure water, 0.5 M NaCl, or seawater) to prevent losses due to volatilization. The flasks were then uncovered, and the experiment restarted. In some experiments, water and salt solutions were monitored simultaneously under the same light conditions to provide a direct comparison of photolysis rates and quantum yields in different solutions. All samples were analyzed within 24 h and stored in the dark in airtight syringes. Control experiments showed no loss or gain of halocarbons in the dark within 24 h. Production of I- during photolysis experiments was monitored by UV-vis absorption spectroscopy using two Schimadzu PharmaSpec 1700 UV-vis spectrophotometers. Higher concentrations of CH2I2, CH2IBr, and CH2ICl were required for these studies, and hence, these experiments were run separately from the GC-MS kinetic analyses. To derive photolysis rates from sets of measurements made over several hours, with changing zenith angles and cloud cover, we corrected each measured concentration change such that photolysis rates are relevant to the mean average intensity of solar irradiance over a given sampling period. The correction factor was calculated by dividing the calculated gas-phase dihalomethane photolysis rates for a particular sampling period by the average calculated gasphase photolysis rate for the entire experimental period. The gas-phase dihalomethane photolysis rates were calculated using a two-stream radiative transfer model (21), with absorption cross-sections from Roehl et al. (22) and Mo¨ssinger et al. (23) and TOMS O3 column data (http:// toms.gsfc.nasa.gov/ozone/ozone01.html). Cloud cover was accounted for by multiplying the calculated gas-phase dihalomethane photolysis rates by the ratio of measured/ modeled j(O1D) values. Modeled values were determined using the radiative transfer model, and measured j(O1D) values were based on the output from an on-site filter radiometer (24).
Results and Discussion Photodissociation Rates. Aqueous-phase photodecomposition rates of CH2I2, CH2IBr, and CH2ICl were determined from concentration measurements made following exposure to solar radiation at 54° N, 1° W, between April and October 2004. Figure 1 shows typical first-order decay plots for CH2IBr in water and 0.5 M NaCl. Photolytic lifetimes were found to vary significantly as a function of the strength of solar irradiance present at ground level but were in line with the differences in zenith angle between different experiments, as shown by the ratios of measured aqueous photolysis rates to modeled gas-phase photolysis rates (which vary predominantly as a function of zenith angle) in Table 1. To facilitate a direct comparison of the relative dihalomethane photolysis rates, the measured aqueous rates for each species were normalized and corrected for 12:00 p.m.,
FIGURE 1. Photolytic decay of CH2IBr measured simultaneously in pure water (open tilted squares) and 0.5 M NaCl solution (filled circles) on 08/12/04. Photolysis rates are statistically equivalent (see Table 1). July 1, 2004. To minimize errors and give an accurate comparison of the aqueous-phase photolysis lifetimes of the three dihalomethanes, data sets from photolysis measurements made on two separate days were combined. This was achieved by making a correction to each data point based on modeled photolysis rates to account for the difference in solar photon flux between the two days. Table 2 summarizes the aqueous photolysis rates determined in 0.5 M NaCl. The errors reported are the limits ((Cb) about the estimated value of the slope corresponding to a confidence level of 90%, calculated by the formula Cb ) tp,v sb, where tp,v is the critical value corresponding to the 90% confidence level and sb is the standard deviation of the slope. As in the gas phase, the rate of aqueous photolysis was found to increase in the order CH2ICl < CH2IBr < CH2I2. The lifetime of CH2I2 with respect to photodecomposition in salt solution was on the order of 10 min during summer, whereas CH2IBr and CH2ICl had lifetimes of 4.5 and 9 h, respectively. In each case absorption of radiation at wavelengths present in the troposphere is thought to lead to cleavage of the C-I bond. The absorption maximum of CH2I2 occurs at ∼290 nm, whereas the CH2IBr absorption maximum is blue-shifted to ∼260 nm. Wavelengths of less than 290 nm are not present in the lower troposphere, so photolytic cleavage of the C-I bond by solar radiation at ground level is less efficient in CH2IBr than in CH2I2. The CH2ICl absorption maximum occurs at a wavelength similar to that of CH2IBr but has a lower maximum cross-section (∼8.8 × 10-19 cm2 molecule-1 in 0.5 M NaCl) compared to that of CH2IBr (∼1.9 × 10-18 cm2 molecule-1 in 0.5 M NaCl), and as a result CH2ICl has the longest photolysis lifetime in the troposphere of the three dihalogenated compounds. Figure 2 compares gas-phase absorption cross-sections of the dihalomethanes with liquidphase (0.5 M NaCl) absorption cross-sections measured in this study. Tabulated values of wavelength-resolved absorption cross-sections for the dihalomethanes in 0.5 M NaCl are shown in Appendix 1. Since organic iodine compounds are not sufficiently soluble in aqueous solution, initial halocarbon standards were made up in 2 mL of methanol prior to being dissolved in 250 mL salt solution to achieve the relatively high concentrations required for UV-vis analyses. Compared to those in the gas phase, absorption crosssections of the dihalomethanes in 0.5 M NaCl solution are slightly blue-shifted and reduced in intensity. The lowest energy absorption band (∼290 nm for CH2I2, ∼260 nm for CH2IBr and CH2ICl) corresponds to the n(I) f σ*(C-I) transition (11). It follows that in aqueous solution the nonbonding electrons will be reduced in energy due to hydrogen-bonding interactions with the solvent. The σ* orbital will also be lower in energy in solution, since solvent VOL. 39, NO. 16, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6131
TABLE 1. Comparison of Aqueous Photolysis Rates Measured on Different Days between April and October 2004 and Average Modeled Gas-Phase Values
CH2I2 CH2I2 CH2IBr CH2IBr CH2ICl CH2ICl
solvent
date and time (GMT)
measured aqueous j(CH2IX)/s-1
aqueous photolysis lifetime
modeled gas-phase j(CH2IX)/s-1
ratio j(aqueous)/ j(gas phase)
water 0.5 M NaCl seawater 0.5 M NaCl water 0.5 M NaCl seawater 0.5 M NaCl water 0.5 M NaCl seawater 0.5 M NaCl
04/19/04 09:00-10:30 10/21/04 09:15-10:15 08/12/04 08:00-16:30 10/09/04 08:30-14:00 06/02/04 07:00-17:00 10/07/04 08:00-15:00
1.20E-3 ( 1.69E-4 1.25E-3 ( 2.95E-4 5.50E-4 ( 9.40E-5 6.47E-4 ( 2.20E-4 4.02E-5 ( 5.89 E-6 3.54E-5 ( 6.79E-6 1.83E-5 ( 6.80E-6 2.27E-5 ( 9.00E-6 2.17E-5 ( 1.50E-6 2.10E-5 ( 8.30E-6 1.12E-5 ( 3.80E-6 1.24E-5 ( 3.50E-6
13.9 min 13.3 min 30.3 min 25.8 min 6.9 h 7.9 h 15.2 h 12.2 h 12.8 h 13.2 h 15.2 h 12.2 h
4.72E-3
0.25 0.26 0.25 0.29 0.13 0.11 0.11 0.13 0.25 0.25 0.26 0.29
TABLE 2. Comparison of Photolysis Rates and Lifetimes (to the Nearest Minute or Hour) for CH2I2, CH2IBr, and CH2ICl in 0.5 M NaCl Solution Normalized for 12:00 p.m., July 1, 2004
halocarbon
j(experimental)/s-1 normalized to 12:00 p.m., July 1
photolysis lifetime
CH2I2 CH2IBr CH2ICl
1.90E-3 ( 2.92E-4 6.01E-5 ( 8.94E-6 3.14E-5 ( 6.98E-6
9 ( 1 min 4.5 h ( 40 min 9(2h
dipole effects will cause lengthening and therefore weakening of the C-I bond. Our measured absorption cross-sections suggest that, in aqueous media, the reduction in energy of the nonbonding orbital is greater than that of the σ* orbital, so the net effect is that the energy of the n(I) f σ*(C-I) transition is increased and therefore absorption crosssections are blue-shifted relative to gas-phase values. The decreased intensity of these bands relative to the equivalent gas-phase cross-sections may be due to alterations in the spatial orientation of the n orbital brought about by solvent interactions, thus limiting overlap with σ* and hence reducing the probability of a transition between these orbitals. We note that our reported cross-sections are in good agreement with previously published absorption maxima of these dihalomethanes in the condensed phase, including CH2I2 absorption measured in acetonitrile (12), CH2ICl absorption in cyclohexane (25), and CH2IBr absorption in acetonitrile (10). Apparent Quantum Yields. Quantum yields of dihalomethane photolysis indicate the proportion of the radical photofragments that react with other species, either directly or via the photoisomer, or escape the cage, rather than undergo geminate recombination back to the parent molecule. Apparent quantum yields for aqueous photodecomposition of the three dihalomethanes were estimated on the basis of the modeled values of the aqueous photolysis rates, eq 1,
∫φ ∫φ λ2
j(aqueous(measured)) ) j(aqueous(modeled))
λ1
λ2
λ1
exptlσexptlθexptl
(1)
modσmodθmod
where φexptl is the apparent (experimental) quantum yield, φmod is 1, σ is the absorption cross-section (cm2 molecule-1), and θ is the photon flux (cm-2 s-1). Because the measured absorption cross-sections of dihalomethanes in 0.5 M NaCl solution were entered into the model and used to calculate j(aqueous(modeled)) (see the Photodissociation Rates sec6132
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 16, 2005
2.21E-3 3.13E-4 1.73E-4 8.52E-5 4.32E-5
tion), σexptl and σmod are equivalent. Making an assumption that φ is constant across the solar wavelengths, the ratio of measured to modeled j(aqueous) gives φexptl, on the basis of the model photon flux. Since the model assumes a cloudless sky, the ratio of measured to modeled j(O1D) values was used to make a correction to the quantum yield to account for cloud cover during our experiments. Our approach assumes equal photon fluxes throughout the quartz flasks used in the experiments. In practice, there may have been some attenuation of light in the solution within the ∼3 cm radius of the quartz flasks, but this will also be the case in seawater, so these calculations should be taken as an “apparent” quantum yield in the top few centimeters of seawater. The dihalomethane lifetimes determined at different zenith angles resulted in the same (within error) quantum yield, indicating that the approach used was valid and can be used to determine surface oceanic photolysis rates at a range of zenith angles. Quantum yields for CH2I2, CH2IBr, and CH2ICl in water, 0.5 M NaCl, and seawater are given in Table 3. Errors in quantum yields were 7-35%. For CH2IBr and CH2ICl it is clear that the majority of the initially produced photofragments undergo cage-induced geminate recombination, with only ∼15-25% escaping the cage or reacting to form other species. However, we observe a substantially higher quantum yield for CH2I2, with the majority of the initial photofragments (∼60%) reacting to form products rather than recombining to form the parent molecule. In each case there was no statistically significant difference between quantum yields in the different solvent environments, which suggests that the chemical complexity of seawater does not dramatically alter its solvent properties compared to those of pure water (26). Therefore, any enhanced solvent cage effect which may be present in seawater does not appear to have a significant impact on the photolysis lifetimes of the dihalomethanes. Justification for the difference in the quantum yield of CH2I2 compared to the other two dihaloalkanes will be addressed in the Potential Mechanisms section, following an explanation of potential mechanistic pathways. Identification of Photoproducts. Photolysis of CH2IX (where X ) Cl, Br, or I) in all solvents was found to lead to formation of iodide (I-) and, in salt and seawater environments, production of CH2XCl (where X ) Cl, Br, or I). Thus, CH2ICl was produced during CH2I2 photolysis, CH2BrCl from CH2IBr photolysis, and CH2Cl2 from CH2ICl photolysis. Formation of I- was identified by its absorption band at 225 nm. Production and decay of the dihalomethanes could also be estimated by their UV absorption spectra, but their absorption cross-sections are weak compared to that of I-, so we used GC-MS for quantitative analyses. Figure 3 shows production of I- in salt solutions (0.5 M NaCl) containing
Note that the short-wavelength absorption band present in the CH2IBr solution cross-sections may be due to the n(Br) f σ*(C-Br) transition of CH2IBr which occurs at ∼210 nm (10); however, since the absorption maximum occurs at 290 nm) suggest that CH2ICl would in fact be a significant product of CH2I2 photodecay in the surface ocean, with a typical molar yield of ∼35% (see Figures 4 and 5). 6134
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 16, 2005
Guan et al. propose the mechanism for formation of CH2ICl from CH2I2 is by direct reaction of Cl- with the initially formed CH2I• photofragment, in competition with the isomer formation, and that the I radical is scavenged by Cl- ions in a similar way via the following reaction:
I + Cl- f IClICl- + Cl- f Cl2- + IThis pathway is proposed due to their observations of Cl2-
SCHEME 1. Fate of CH2I2 Photofragmentsa
a k recom ) rate constant for geminate recombination to the parent molecule, kes ) rate of escape from the solvent cage, and kiso ) rate constant for isomerization.
FIGURE 4. Production of CH2ICl following photolytic breakdown of CH2I2. Unfilled tilted squares represent the CH2I2 concentration, times signs represent the uncorrected CH2ICl concentration, and filled circles correspond to the CH2ICl concentration after a correction to account for losses due to photolysis.
FIGURE 5. Typical correlation between CH2I2 photodecomposition and CH2ICl production.
FIGURE 6. Iodide production from CH2I2 photolysis. Tilted squares correspond to I- production in water and circles to production in 0.5 M NaCl solution, from the same initial CH2I2 concentration. in Raman spectra. An alternative or additional mechanism may be that Cl- attacks the photoisomer or associated ion pair (e.g., CH2I+-I-) or, following heterolytic decay of the CH2I-I isomer (as suggested to occur by Tarnovsky et al. (12)), the Cl- ions react directly with CH2I+, leaving I- in solution. We found that significantly lower concentrations of Iwere produced from CH2I2 photolysis in salt solution compared to pure water (see Figure 6). This is in agreement with the observations of Guan et al. (19), and is consistent with a lower conversion of CH2I2 to I- in salt solution, where a fraction of the I atoms are “tied up” in the less photolabile CH2ICl molecule. After one lifetime the amount of I- produced following CH2I2 photolysis in 0.5 M NaCl is only ∼60-70%
of the I- concentration produced in pure water. Simultaneously with the decrease in [I-], Guan et al. also found a decrease in [H+] in salt solution compared to water, but by differing amounts, suggesting that formation of these ions is not simply via the equimolar HI elimination occurring in pure water. As the molar yield of H+ declined more than that of I-, this suggests that I- is also formed from an additional source as well as by OH insertion of the isomer, and this is consistent with either of the proposed reaction pathways with Cl- ions. The equivalent isodihalomethane photoisomers have also been identified following CH2IBr and CH2ICl excitation (CH2Br-I and CH2Cl-I, respectively) in organic solvents (10, 11, 14, 15). The efficiency of photoisomerization of these dihalomethanes is significantly reduced compared to that of CH2I2, with quantum yields of photoisomerization of ∼70%, 25%, and 9%, respectively, for CH2I2, CH2IBr, and CH2ICl (11). As noted in the Apparent Quantum Yields section, we observed a large difference in our measured CH2I2 quantum yield compared to those of CH2IBr and CH2ICl (see Table 3). Assuming that isomer formation is essentially irreversible in aqueous solution, as it is in methanol and ethanol (12), and CH2I-I does not break down to re-form CH2I• and I•, then the quantum yield of photolysis may be expressed in terms of the rate constants in Scheme 1. The relative isophotoisomer yields observed by Tarnovsky et al., thought to be due to the differing thermal stabilities of the three photoisomers (with CH2I-I most stable and CH2Cl-I least stable), may go some way to explaining the differences in our quantum yields. The CH2IBr and CH2ICl quantum yields measured in this work are significantly lower than that of CH2I2, and the CH2I-I yield of ∼70% is in line with our CH2I2 photolysis quantum yield of ∼0.62. However, the photoisomer yields determined by Tarnovsky et al. also imply that we should expect a lower CH2ICl quantum yield compared to that of CH2IBr. Thus, the fraction of isomer formation cannot fully explain our observations, since our CH2ICl photolysis quantum yield is greater than that of CH2IBr, and in addition our CH2ICl photolysis quantum yield is significantly greater than the yield of the CH2Cl-I isomer. The quantum yield of photolysis is also regulated by the efficiency of solvent cage escape of the primary radicals (CH2X• and I•). Since CH2Cl• is the smallest halomethyl radical, it follows that the probability of cage escape of the CH2ICl photofragments will be increased relative to that of the photofragments of CH2IBr, leading to an increased CH2ICl photolysis quantum yield. Modeling and Environmental Implications. Our experimentally determined absorption cross-sections and quantum yields for photolysis (both in 0.5 M NaCl solution) were utilized in conjunction with zenith angles and O3 column data from 2004, together with the two-stream radiative transfer model (see the Experimental Details) to predict the seasonal variation in the rate of surface ocean photolysis for the three dihalomethanes at 50° N, 0° E (see Figure 7). Photolysis rates were computed for every hour on the 15th day of each month, and a 24 h average was calculated. VOL. 39, NO. 16, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6135
Appendix 1. Absorption Cross-Sections for Dihalomethanes in 0.5 M NaCl Solution 1020σ/cm2
FIGURE 7. Seasonal cycle of the 24 h averaged surface ocean photolysis rate constants of CH2I2, CH2IBr, and CH2ICl at 50° N, 0° E. Open circles correspond to j(CH2I2), filled tilted squares represent j(CH2IBr), and times signs represent j(CH2ICl) values. We have found from our experiments in the absence of light that other potential chemical sinks of dihalomethanes in seawater, such as hydrolysis and chloride ion substitution reactions (which are known to be the main chemical fate of CH3I in the ocean (26, 29)), occur at a negligible rate in comparison with the rate of photolysis in surface waters. Therefore, the abiotic breakdown of dihalomethanes in the surface ocean is controlled almost exclusively by photolysis. The photolytic formation of CH2ICl from CH2I2 may go some way to explain CH2ICl levels in the surface ocean. Measurements of these halocarbons in surface waters of the Northwest Atlantic (30) and Southern (31) Oceans show that the molar CH2I2/CH2ICl ratio decreases toward the surface, ranging from 20-35 at 50 m depth to 10-16 at 5 m, which supports this hypothesis. Whether CH2I2 is completely photolyzed in the top few meters of the ocean depends on its residence time, which is a function of ocean turbulence, mixed layer depth, and the attenuation coefficient of the ocean waters. Assuming an attenuation coefficient at 300 nm of 0.01 cm-1 (32), the 24 h averaged photolysis lifetime in July at 50° N at 5 m depth is around 2 days, i.e., competitive with volatilization to the atmosphere. The production of I- following dihalomethane photolysis in the surface ocean may have implications for the oxidation capacity of the lower troposphere. Dry deposition of O3 to the surface waters of the ocean is known to be enhanced by I- (33, 34). Current model predictions consistently underestimate the rate of O3 deposition, and although they do take into account O3 depletion by reaction with I-, they are based on typical surface seawater concentrations of I- ((100-400) × 10-9 mol dm-3) (35, 36). This work suggests that significant I- production can occur via dihalomethane photolysis in the surface ocean, so the role of the I-/O3 reaction in regulating O3 deposition to the surface waters may be more substantial than previously thought. Given our observation of 1.65 mol of I- generated/mol of CH2I2 photolyzed, and assuming a (constant) surface water CH2I2 concentration of 0.1 nM (31) and negligible destruction of the produced I- beneath the microlayer, CH2I2 photolysis over an 8 h period in summertime at midlatitudes may contribute between 2% and 8% of surface seawater I- levels.
Acknowledgments C.E.J. acknowledges the Natural Environment Research Council for funding her studentship. We are grateful to Dr. James Lee, University of York, for a loan of the filter radiometer. 6136
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 16, 2005
λ/nm
CH2I2
CH2IBr
CH2ICl
390 385 380 375 370 365 360 355 350 345 340 335 330 325 320 315 310 305 300 295 290 285 280 275 270 265 260 255 250 245 240 235 230 225 220 215 210 205
0.00 0.00 0.00 0.00 0.00 0.00 1.07 4.00 8.43 16.57 28.30 46.20 69.43 98.23 130.20 161.07 190.93 216.37 236.60 248.47 254.60 254.00 243.47 217.83 178.50 137.20 108.80 102.50 111.97 121.80 121.97 115.30 119.77 148.33 203.57 299.07 460.90 385.30
0.31 0.38 0.50 0.53 0.57 0.60 0.60 0.63 1.19 1.79 2.80 4.56 6.45 8.93 12.64 17.48 23.46 30.79 40.53 53.40 70.60 93.05 118.77 146.38 170.94 186.67 187.01 169.69 140.79 110.06 91.60 96.60 141.26 244.81 392.39 514.75 538.90 372.45
0.41 0.58 0.30 0.30 0.44 0.36 0.32 0.70 0.88 0.97 1.39 2.12 3.05 4.30 5.73 7.42 9.58 12.76 17.11 22.86 31.11 42.50 55.67 69.52 81.03 87.65 87.36 80.64 69.98 57.62 45.77 35.38 27.00 21.80 21.88 37.98 106.86 145.77
Literature Cited (1) Carpenter, L. J.; Malin, G.; Liss, P. S.; Kupper, F. C. Novel biogenic iodine-containing trihalomethanes and other short-lived halocarbons in the coastal East Atlantic. Global Biogeochem. Cycles 2000, 14, 1191-1204. (2) Carpenter, L. J.; Liss, P. S.; Penkett, S. A. Marine organohalogens in the atmosphere over the Atlantic and Southern Oceans. J. Geophys. Res. 2003, 108, 4256-4269. (3) McFiggans, G.; Plane, J. M. C.; Allan, B. J.; Carpenter, L. J.; Coe, H.; O’Dowd, C. A modeling study of iodine chemistry in the marine boundary layer. J. Geophys. Res. 2000, 105, 14371-14385. (4) Vogt, R.; Sander, R.; Von Glasow, R.; Crutzen, P. J. Iodine chemistry and its role in halogen activation and ozone loss in the marine boundary layer: A model study. J. Atmos. Chem. 1999, 32, 375-395. (5) Dickerson, R. R.; Rhoads, K. P.; Carsey, T. P.; Oltmans, S. J.; Burrows, J. P.; Crutzen, P. J. Ozone in the remote marine boundary layer: A possible role for halogens. J. Geophys. Res. 1999, 104, 21385-21395. (6) Galbally, I. E.; Bentley, S. T.; Meyer, C. P. Mid-latitude marine boundary-layer ozone destruction at visible sunrise observed at Cape Grim, Tasmania, 41°S. Geophys. Res. Lett. 2000, 23, 38413844. (7) Kanaya, Y.; Yokouchi, Y.; Matsumoto, J.; Nakamura, K.; Kato, S.; Tanimoto, H.; Furutani, H.; Toyota, K.; Akimoto, H. Implications of iodine chemistry for daytime HO2 levels at Rishiri Island. Geophys. Res. Lett. 2002, 29, 1212-1215. (8) O′Dowd, C. D.; Jimenez, J. L.; Bahreini, R.; Flagan, R. C.; Seinfeld, J. H.; Hameri, K.; Pirjola, L.; Kulmala, M.; Jennings, S. G.; Hoffmann, T. Marine aerosol formation from biogenic iodine emissions. Lett. Nat. 2002, 417, 632-636. (9) Tarnovsky, A. N.; Alvarez, J.-L.; Yartsev, A. P.; Sundstro¨m, V.; A° kesson, E. Photodissociation dynamics of diiodomethane in solution. Chem. Phys. Lett. 1999, 312, 121-130.
(10) Tarnovsky, A. N.; Wall, M.; Gustafsson, M.; Lascoux, N.; Sundstro¨m, V.; A° kesson, E. Ultrafast study of the photodissociation of bromoiodomethane in acetonitrile upon 266 nm excitation. J. Phys. Chem. A 2002, 106, 5999-6005. (11) Tarnovsky, A. N.; Wall, M.; Rasmusson, M.; Sundstro¨m, V.; A° kesson, E. In Femtochemistry and Femtobiology, Ultrafast Dynamics in Molecular Science; Douhal, A., Santamaria, J., Eds.; World Science: Singapore, 2002; pp 247-260. (12) Tarnovsky, A. N.; Sundstro¨m, V.; A° kesson, E.; Pascher, T. Photochemistry of diiodomethane in solution studied by femtosecond and nanosecond laser photolysis. Formation and dark reactions of the CH2I-I isomer photoproduct and its role in cyclopropanation of olefins. J. Phys. Chem. A 2004, 108, 237249. (13) Zheng, X.; Phillips, D. L. Solvation can open the photoisomerization pathway for the direct photodissociation reaction of diiodomethane: Transient resonance Raman observation of the isodiiodomethane photoproduct from ultraviolet excitation of diiodomethane in the solution phase. J. Phys. Chem. A 2000, 104, 6880-6886. (14) Kwok, W. M.; Ma, C.; Parker, A. W.; Phillips, D.; Towrie, M.; Matousek, P.; Zheng, X.; Phillips, D. L. Picosecond time-resolved resonance Raman observation of the iso-CH2Cl-I and iso-CH2ICl photoproducts from the “photoisomerization” reactions of CH2ICl in the solution phase. J. Chem. Phys. 2001, 114, 75367543. (15) Kwok, W. M.; Ma, C.; Phillips, D.; Parker, A. W.; Towrie, M.; Matousek, P.; Phillips, D. L. Picosecond time-resolved resonance Raman observation of Iso-CH2Br-I following A-band photodissociation of CH2BrI in the solution phase. Chem. Phys. Lett. 2001, 341, 292-298. (16) Kwok, W. M.; Ma, C.; Parker, A. W.; Phillips, D.; Towrie, M.; Matousek, P.; Phillips, D. L. Picosecond time-resolved resonance Raman study of CH2I-I produced after ultraviolet photolysis of CH2I2 in CH3OH, CH3CN/H2O and CH3OH/H2O solutions. J. Phys. Chem. A 2003, 107, 2624-2628. (17) Li, Y.-L.; Zhao, C.; Kwok, W. M.; Guan, X.; Zuo, P.; Phillips, D. L. Observation of a HI leaving group following ultraviolet photolysis of CH2I2 in water and an ab initio investigation of the O-H insertion/HI elimination reactions of the CH2I-I isopolyhalomethane species with H2O and 2H2O. J. Chem. Phys. 2003, 119, 4671-4681. (18) Kwok, W. M.; Zhao, C.; Guan, X.; Li, Y.-L.; Du, Y.; Phillips, D. L. Efficient dehalogenation of polyhalomethanes and production of strong acids in aqueous environments: Water-catalyzed O-Hinsertion and HI-elimination reactions of isodiiodomethane (CH2I-I) with water. J. Chem. Phys. 2004, 120, 9017-9032. (19) Guan, X.; Du, Y.; Li, Y.-L.; Kwok, W. M.; Phillips, D. L. Comparison of the dehalogenation of polyhalomethanes and production of strong acids in aqueous and salt (NaCl) water environments: Ultraviolet photolysis of CH2I2. J. Chem. Phys. 2004, 121, 83998409. (20) Wevill, D. J.; Carpenter, L. J. Automated measurement and calibration of reactive volatile halogenated organic compounds in the atmosphere. Analyst 2004, 129, 634-638. (21) Hough A. M. The calculation of photolysis rates for use in global tropospheric modelling studies; AERE Report R-13259; HMSO: London, 1988; p 53.
(22) Roehl, C. M.; Burkholder, J. B.; Moortgat, G. K.; Ravishankara, A. R.; Crutzen, P. J. Temperature dependence of UV absorption cross sections and atmospheric implications of several alkyl iodides. J. Geophys. Res. 1997, 102, 12819-12829. (23) Mo¨ssinger, J. C.; Shallcross, D. E.; Cox, R. A. UV-Vis absorption cross-sections and atmospheric lifetimes of CH2Br2, CH2I2 and CH2BrI. J. Chem. Soc., Faraday Trans. 1998, 94, 1391-1396. (24) Junkermann, W.; Platt, U.; Voltzthomas, A. A photoelectric detector for the measurement of photolysis frequencies of ozone and other atmospheric molecules. J. Atmos. Chem. 1989, 8, 203227. (25) Kwok, W. M.; Phillips, D. L. Early time photodissociation dynamics of chloroiodomethane in the A-band absorption from resonance Raman intensity analysis. J. Chem. Phys. 1996, 104, 9816-9832. (26) Zika, R. G.; Gidel, L. T.; Davis, D. D. A comparison of photolysis and substitution decomposition rates of methyl iodide in the ocean. Geophys. Res. Lett. 1984, 11, 353-356. (27) Lin, X.; Guan, X.; Kwok, W. M.; Zhao, C.; Du, Y.; Li, Y.-L.; Phillips, D. L. Water-catalyzed O-H insertion/HI elimination reactions of isodihalomethanes (CH2X-I, where X ) Cl, Br, I) with water and the dehalogenation of dihalomethanes in water-solvated environments. J. Phys. Chem. A 2005, 109, 981-998. (28) Lee, S. J.; Bersohn, R. Photo-dissociation of a molecule with 2 chromophoressCH2IBr. J. Phys. Chem. 1982, 86, 728-730 (29) Zafiriou, O. C. Reaction of methyl halides with seawater and marine aerosols. J. Mar. Res. 1975, 33, 75-81. (30) Moore, R. M.; Tokarczyk, R. Volatile biogenic halocarbons in the Northwest Atlantic. Global Biogeochem. Cycles 1993, 7, 195210. (31) Palmer, C. J.; Wevill, D. J.; Jones, C. E.; Carpenter, L. J. Measuring reactive halogen species in the marine boundary layer; what does their distribution tell us about their sources, transfers and sinks? Poster at 1st International SOLAS Science Conference, Halifax, Nova Scotia, 2004. (32) Wells, N. G. The absorption of solar radiation in the ocean. The Atmosphere and Ocean, 2nd ed.; John Wiley: New York, 1997; pp 84-88. (33) Schwartz, S. E. Factors governing dry deposition of gases to surface water. In Precipitation scavenging and atmospheresurface exchange; Schwartz, S., Slinn, S., Eds.; Hemisphere Publishing Corp.: Washington, DC, 1992. (34) Garland, J. A.; Elzerman, A. W.; Penkett, S. A. The mechanism for dry deposition of ozone to seawater surfaces. J. Geophys. Res. 1980, 85, 7488-7492. (35) Chang, W.; Heikes, B. G.; Lee, M. Ozone deposition to the sea surface: chemical enhancement and wind speed dependence. Atmos. Environ. 2004, 38, 1053-1059. (36) Campos, M. L. A. M.; Farrenkopf, A. M.; Jickells, T. D.; Luther, G. W. A comparison of dissolved iodine cycling at the Bermuda Atlantic Time-series Station and Hawaii Ocean Time-series Station. Deep Sea Res. 1996, 43, 455-466.
Received for review March 22, 2005. Revised manuscript received June 8, 2005. Accepted June 14, 2005. ES050563G
VOL. 39, NO. 16, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6137
