Mechanisms of the Aqueous Photodegradation of Polycyclic Aromatic

Nov 4, 2003 - Mechanisms of the Aqueous. Photodegradation of Polycyclic. Aromatic Hydrocarbons. MATTHEW P. FASNACHT AND. NEIL V. BLOUGH*...
0 downloads 0 Views 135KB Size
Environ. Sci. Technol. 2003, 37, 5767-5772

Mechanisms of the Aqueous Photodegradation of Polycyclic Aromatic Hydrocarbons MATTHEW P. FASNACHT AND NEIL V. BLOUGH* Department of Chemistry and Biochemistry, University of Maryland, College Park, Maryland 20742

The role of O2 and photoionization as well as the involvement of polycyclic aromatic hydrocarbon (PAH) cation radicals (P+) in the photodegradation of nine PAHs was examined. Photodegradation quantum yields for all PAHs increased with increasing O2 concentration, illustrating the key role of O2 in the photodegradation mechanism. In the presence of a series of electron donors (to P+), the photodegradation rate constants of most PAHs were largely unaffected at low O2 concentrations (e250 µM), indicating that P+ is not extensively produced. However, at higher O2 concentrations (up to 1.2 mM), the presence of the donors substantially lowered photodegradation rates for most PAHs, indicating that P+ is produced and is arising from O2 reaction with the excited singlet state. Because little P+ was detected at low O2 concentrations and, further, because degradation rates were not enhanced in the presence of N2O, we conclude that photoionization is unimportant. With some exceptions, photodegradation can proceed through reaction of O2 with both excited singlet and triplet states of the PAHs. Our results indicate that photodegradation via the excited singlet state occurs primarily through electron transfer to O2, whereas degradation via the triplet occurs predominately through a direct reaction of O2 with the PAH within the collision complex.

which can react further to form stable products (Scheme 1, eq 4) (8, 9, 14). Although a previous laser flash photolysis study indicated that naphthalene photoionization may be a monophotonic process in part (9), the photodegradation quantum yields of the PAHs do not correlate with ionization or oxidation potentials (2), implying that direct photoionization is unimportant. Further, we found previously that photodegradation quantum yields for a series of PAHs were independent of wavelength above 290 nm, inconsistent with a direct photoionization process. Electron transfer from the PAH to O2 following light absorption by a ground-state complex has been proposed for pyrene (Scheme 1, eqs 5 and 10) (8). The resultant chargetransfer complex ([P+-O2-]) may undergo solvent separation, forming P+ and O2- radicals as in photoionization (Scheme 1, eq 7), undergo charge recombination to regenerate the ground-state PAH and O2 (Scheme 1, eq 8), or react within the collision complex to form products (Scheme 1, eq 11). Following absorption of a photon (Scheme 1, eq 9), PAH excited singlet (1P*) and triplet states (3P*) may be diffusionally quenched by O2 (Scheme 1, eqs 12 and 13) to produce a [P+-O2-] complex identical to that formed via eqs 5 and 10 (Scheme 1). Finally, photodegradation could be initiated by the direct reaction of O2 with 3P* in a [3P*-3O2] complex (Scheme 1, eq 20) or in a [P-1O2] complex (Scheme 1, eq 21) formed by energy transfer from 3P* to 3O2 within the collision complex (Scheme 1, eq 17). Due to spin restrictions, the reaction of 3 O2 with 1P* to form products directly should not occur (15) but as indicated above could proceed through a [P+-O2-] complex (Scheme 1, eq 12). Here we examine the role of O2 and photoionization as well as the possible involvement of P+ in the aqueous photodegradation of PAHs. We find that quantum yields of PAH photodegradation increase with increasing O2 concentration and that photoionization does not appear to play a significant role in the photodegradation. Although the production of P+ is important for many PAHs under high O2 concentrations, most photodegrade in air-equilibrated solutions with little involvement (290 nm) can cause the direct photodegradation of PAHs to occur with quantum yields varying from 10-5 to 10-2 (24). Despite many studies, however, the mechanisms of direct photodegradation remain unclear. Three primary reactions have been proposed for the initial step in the photodegradation (Scheme 1), namely, photoionization (3, 9, 11), electron transfer to dioxygen (O2) (2, 6, 8), and direct reaction of the excited triplet state with O2 (2, 6). Photoionization involves the ejection of an electron from the PAH (P) following the absorption of a photon (Scheme 1, eq 1). In aerated solutions, this electron will react rapidly with O2 to form superoxide (O2-) (Scheme 1, eq 6) (13). The resultant PAH cation radical (P+) can react with water (or hydroxide ion) to form secondary (radical) intermediates,

Materials. Acenaphthene (AC, 99%), anthracene (AN, 99%), benz[a]anthracene (BA, 99%), benzo[a]pyrene (BP, 97%), benzo[k]fluoranthene (BK, 98%), chrysene (CH, 98%), 9methylanthracene (MA, 98%) perylene (PE, 99.5%), potassium ferrocyanide(II) trihydrate (K4Fe(CN)6, 99%), and pyrene (PY, 99%) were obtained from Aldrich and used without further purification. Acetonitrile (ACN, HPLC grade), methanol (MeOH, HPLC grade), and sodium iodide (NaI, certified) were obtained from Fisher. 3-Amino-2,2,5,5 tetramethyl-1-pyrrolidinyloxy free radical (3ap, 99%) was obtained from Acros. Nitrogen (N2, UHP/zero grade), synthetic air (UHP/zero grade), and dioxygen (O2, 99%) were obtained from Air Products. Nitrous oxide (N2O) was obtained from Matheson. Water was obtained from a Millipore Milli-Q system. PAH stock solutions (0.5-10 µM) were prepared in ACN and stored in amber borosilicate vials at room temperature. Apparatus. Ultraviolet-visible absorption spectra were measured using a Hewlett-Packard 8452A diode array spectrophotometer. Steady-state fluorescence was acquired using an SLM Aminco Series 2 luminescence spectrometer (AB2). High-performance liquid chromatography (HPLC) separations employed a Waters Nova-PAK reversed-phase (C-18) column and the apparatus described previously (16). The

* Corresponding author phone (301) 405-0051; fax: (301) 3149121; e-mail: [email protected]. 10.1021/es034389c CCC: $25.00 Published on Web 11/04/2003

 2003 American Chemical Society

VOL. 37, NO. 24, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5767

SCHEME 1. Proposed Pathways of PAH Photodegradation

irradiation system was identical to that described previously (2). Briefly, light from a 300 W Xe arc was lamp passed through a Schott WG305 high-pass filter before entering the sample, which was contained in a 1 cm sealed cuvette. An Air Products gas proportioner with a total flow rate of ≈65 mL/min was used to obtain N2/air and N2/O2 gas mixtures. Conventional laser flash photolysis was performed using the apparatus described previously (17). Experimental Procedures. Identical to previous experiments (2), 10-20 µL of PAH stock solution was added to 4.0 mL of water, leading to solutions in the range of 0.7-25 nM. Experiments involving N2O contained no ACN or other organic solvent and were prepared by diluting saturated aqueous PAH solutions. Solutions were purged for 5-10 min using N2O, N2/air, or N2/O2 gas mixtures. O2 concentrations in the purged solutions were calculated from the fraction of O2 in air (χair ) 0.21) or the gas mixture (χ) and the concentration of O2 in air equilibrated water at 25 °C ([O2]air ) 250 µM) (eq 1).

[O2] ) (χ/χair)[O2]air

(1)

Under initial rate conditions (10-30% conversion), the initial fluorescence (Fo) and the fluorescence after appropriate irradiation times (F) were measured using identical wavelengths as in our previous study (2). These measurements were used to determine photodegradation rate constants from the slopes of plots of F/Fo versus time. Three or more experiments were performed for each experimental condition to determine the uncertainty (one standard deviation) in the rate constants. Using identical light fields and initial PAH concentrations, quantum yields at differing O2 concentrations (φO2) were calculated relative to air using eq 2 and the rate constants acquired at each O2 concentration (kair and kO2) and the quantum yields (φair) previously obtained in airequilibrated solutions (2, 3).

φO2 ) (kO2/kair)φair

(2)

To obtain rate constants for the quenching of P+ by electron donors, laser flash photolysis (LFP) experiments were performed using air-equilibrated solutions of PAHs in 75% water/25% ACN (0.05 < A355 < 0.8). The solutions (∼3 mL), which contained PAH and varying concentrations of electron donor within a 1 cm cell, were mixed with a magnetic stirrer, while transient absorption waveforms were obtained from a limited number of laser pulses (usually 10 pulses, 355 nm excitation) and then averaged. This protocol was employed to ensure that significant levels of products did not accumulate within the cell during the course of an LFP experiment. P+ was detected at the absorption peak maximum (AN, 720 nm; MA, 690 nm; BA, 400 nm; BP, 550 nm; PY, 445 nm). The rate constant for PY photodegradation under highintensity laser irradiation (355 nm) was acquired in air5768

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 24, 2003

equilibrated solutions (75% water/25% ACN) in the presence and absence of 50 µM 3ap. Using a repetition rate of 1 Hz, 1000 laser light pulses (355 nm) were directed onto a 1 cm cuvette. Solutions were mixed continually with a magnetic stirrer, and the PY (initial concentration ∼11 µM) was measured using a diode-array spectrophotometer after every 100-300 laser pulses.

Results Role of O2. To examine the role of O2 in PAH photodegradation, quantum yields (φ) of PAH loss were determined across a wide range of O2 concentrations (Figure 1) using a low-intensity polychromatic light source. For all PAHs examined, φ increased with increasing O2 concentration. As O2 concentrations approached zero, the quantum yields of BP, PE, BK, and PY also reached or approached zero whereas the quantum yields of AN, BA, AC, and MA remained significantly above zero. Note that deoxygenation by purging with N2 is not expected to remove all O2 rigorously; thus, small residual amounts of O2 are likely present in these N2purged solutions. Uncharacteristically, the quantum yield of AN loss was observed to increase upon N2 purging. Our results agree with previous studies of BA, BP, and PY but not with studies of MA (3, 4, 8). Role of P+. During PAH photodegradation, P+ could be formed through either photoionization (Scheme 1, eq 1) or electron transfer to O2 (Scheme 1, eq 7). However, the involvement of P+ in the photodegradation of PAHs at solar wavelengths and intensities has yet to be established. To test for the presence of P+, we employed three sacrificial electron donors (D), 3ap, NaI (18), and K4Fe(CN)6. At sufficiently high concentrations of D, any P+ formed will be reduced to P and will act to protect the PAH from photodegradation (eq 3).

P+ + D f P + Dox

(3)

Thus, by examining the decrease in φ in the presence of these donors, the contribution of P+ in the photodegradation can be quantified. To acquire rate constants for the reaction of these donors with P+, LFP was employed using high laser powers to generate P+ through a biphotonic process (14). Transient absorption spectra collected 2 µs after laser excitation of PAHs in air-equilibrated solutions (75% water/25% ACN) are likely due to a combination of P+ and residual 3P*. 3P* is rapidly quenched by O2, so that by 10 µs, the transient absorption spectra are consistent with those reported previously for P+ (Figure 2) (19, 20). After 5 µs, the decay of absorption was markedly biphasic (Figure 3 and Supporting Information, Figure S1). We attribute this behavior to the decay of P+ to a long-lived intermediate. Consistent with this view, addition of the donors decreased both the lifetime of the short-lived component and the amplitude of the long-lived component.

FIGURE 1. Dependence of PAH photodegradation quantum yields on O2 concentration in the absence (O) and presence (4) of 50 µM 3ap. Dioxygen concentrations were varied using different gas mixtures (N2/air and N2/O2). Irradiations employed a Xe lamp and a WG305 high-pass filter. These results agree with the simple competitive reaction scheme (eq 4)

Thus, to obtain P+ decay rate constants, transient absorption waveforms at times >4 µs (in the presence of donors) were fit to the sum of two exponentials, with the fast component representing the decay of P+ (kfast; inset Figure 3). Bimolecular rate constants for the reduction of P+ by the three donors, obtained from the slope of the dependence of P+ decay (kfast) on donor concentrations (inset Figure 3), were very large, ranging from 108 to 1010 M-1 s-1 for five of the nine PAHs examined in this study (Table 1; eq 3). Because the one-electron oxidation potentials of the five PAHs examined by LFP fall approximately within the same range as the other four (2), we expect the P+ of all nine PAHs to react rapidly with these donors, consistent with the results reported below. Rate constants for the quenching of AN+ and PY+ by Iacquired in this study (75% water/25% ACN) were also very large (∼109 M-1 s-1; Table 1) but were about an order of magnitude lower than those previously reported for methanol (18). To test directly whether these donors could act to protect PAHs from photodegradation via eq 3, the rate constant of PY photodegradation was measured in the presence and absence of 50 µM 3ap while employing high laser powers to produce P+ biphotonically. In the presence of 3ap, the observed loss of PY (k ) 0.0049 ( 0.0008 min-1) was 62% lower than that in the absence of 3ap (k ) 0.013 ( 0.001 min-1). Consistent with protection via eq 3, we calculate that 55% of the photodegradation should have been quenched

FIGURE 2. Transient absorption spectra acquired following 355 nm excitation of PAHs in air-equilibrated solutions (75% water/25% ACN). The legend indicates the time elapsed following the laser pulse. VOL. 37, NO. 24, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5769

FIGURE 4. Dependence of the photodegradation rate constants of AN and PY on 3ap concentrations in air-equilibrated aqueous solution using a Xe lamp and a WG305 high-pass filter. FIGURE 3. Transient absorption waveforms measured in the quenching of AN+ and PY+ by 3ap (0, 30, and 90 µM), NaI (0, 9, and 69 µM), and Fe(CN)64- (0, 20, and 30 µM for AN and 0, 10, and 40 µM for PY). PAHs in air-equilibrated solutions (75% water/25% ACN) were excited at 355 nm and detected at 720 nm for AN and 445 nm for PY. (Inset) Dependence of the rate constant for P+ decay on donor (D) concentration.

TABLE 1. Rate Constants for P+ Reduction by 3ap, I-, and Fe(CN)64- a kq (M-1 s-1) × 10-9 PAH

3-ap

AN BA BP MA PY

0.85 ( 0.05 0.93 ( 0.05 0.9 ( 0.2 1.2 ( 0.2 0.55 ( 0.04

NaI 2.4 ( 0.3 0.15 ( 0.01 1.1 ( 0.2

Fe(CN)643.4 ( 0.4 14.0 ( 0.5 2.3 ( 0.2 3.6 ( 0.2

a Measured using LFP. Error reported is the standard error of the slope in the plots of rate constant versus donor concentration (Figure 3 inset).

if all PY+ proceeded to products (Scheme 1, eqs 2 and 3), based on the measured PY+ lifetime (43 µs) and the rate constant of PY+ reduction by 3ap under these conditions (Table 1, Figure 3). To determine the contribution of P+ to the photodegradation of PAHs in air-equilibrated aqueous solutions, rate constants were measured in the presence and absence of the donors under our standard reaction conditions (lowintensity irradiation, Xe lamp). At low donor concentrations (5-50 µM), only that portion of the photodegradation which proceeds through P+ will be eliminated; these donor concentrations are too low to quench 1P* (τ < 250 ns (21)) or to compete with O2 to quench 3P* (22). For AN and PY, only small reductions in the rate constant were observed at concentrations of 3ap up to 180 µM (Figure 4). At concentrations of 3ap above 10 µM, the rate constants were independent of 3ap concentration, indicating that sufficient 3ap was present to reduce P+ quantitatively (eq 3). 5770

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 24, 2003

Similar results were obtained with the other donors and PAHs (Table 2). Rate constants remained the same for most PAH/donor combinations; only for BA, BK, and AC were the rate constants reduced by more than 30%. With one exception (PY and Fe(CN)64-), similar reductions in the rate constants were observed for all three donors, providing further evidence that this inhibition is produced by the reduction of P+ to P (eq 3). The independence of the photodegradation rate constants on donor concentration (Table 2), which was also observed for AN and PY with 3ap (Figure 4), indicates that P+ is reduced quantitatively by each of these donors over this concentration range (5-50 µM). To ensure that the greater inhibition measured for BA in the presence of 3ap was not a measurement artifact, rate constants were measured using both steady-state fluorescence and HPLC with fluorescence detection and were found to be identical (See Supporting Information, Table S1). The low concentrations of donors needed to reduce P+ quantitatively (Figure 4 and Table 2) implies that P+ lifetimes are considerably longer at the very low PAH concentrations used in these experiments. To test this possibility, the decay of P+ was examined by LFP using different initial PAH concentrations. Consistent with the data presented in Figure 4 and Table 2, lowering PY concentrations by 10-fold led to a ∼10-fold increase in PY+ lifetime. This behavior may result from P-P+ dimer formation at higher PAH concentrations, as reported previously for naphthalene (14). To examine the combined roles of O2 and P+ in the aqueous photodegradation of PAHs, the dependence of the quantum yields on O2 concentration was measured in the presence of 50 µM 3ap (Figure 1). For all PAHs except PE, quantum yields were lowered substantially in the presence of 3ap at high O2 concentrations (>500 µM) but were not significantly affected at low O2 concentrations. Interestingly, the anomaly seen with AN under N2-purged conditions (see above) was not observed in the presence of 3ap. These results show that P+ strongly contributes to PAH photodegradation at high O2 concentrations but not at low O2 concentrations (Figure 1 and Table 2), including air-equilibrated solutions (O2 ) 250 µM). Interestingly, PE showed no evidence of P+ formation over the range of O2 concentrations examined here.

TABLE 2. PAH Photodegradation Rate Constant Ratios (k′/k) in the Absence (k) and Presence (k′) of Electron Donors in Aqueous Air-Equilibrated Solutionsa k′/k 3ap

a

k′/k Fe(CN)64-

PAH

5 µM

50 µM

5 µM

50 µM

5 µM

50 µM

AN PY BA BP MA PE CH AC BK

0.76 ( 0.08 0.87 ( 0.02 0.41 ( 0.07 0.7 ( 0.1 0.82 ( 0.07 0.9 ( 0.1 0.7 ( 0.1 0.52 ( 0.06 0.1 ( 0.2

0.74 ( 0.07 0.85 ( 0.03 0.38 ( 0.06 0.7 ( 0.1 0.80 ( 0.07 0.8 ( 0.1 0.8 ( 0.2 0.52 ( 0.05 0.4 ( 0.2

0.7 ( 0.2 1.5 ( 0.4 0.3 ( 0.2 0.84 ( 0.08 0.83 ( 0.04 1.1 ( 0.2

0.9 ( 0.1 2.5 ( 0.8 0.3 ( 0.2 0.85 ( 0.07 0.84 ( 0.04 1.0 ( 0.2

0.78 ( 0.08 0.9 ( 0.1 0.39 ( 0.09 0.85 ( 0.06 0.99 ( 0.08 1.0 ( 0.2 0.8 ( 0.2

0.6 ( 0.1 0.9 ( 0.1 0.39 ( 0.07 0.75 ( 0.07 0.88 ( 0.08 0.9 ( 0.2 0.8 ( 0.2

0.4 ( 0.3

0.3 ( 0.3

The experiments were performed using a Xe lamp and a WG305 high-pass filter.

TABLE 3. Aqueous PAH Photodegradation Rate Constants When Purged with N2O and N2a k (min-1)

a

k′/k I-

PAH

N2O purged

N2 purged

AN BA MA AC

0.12 ( 0.01 0.044 ( 0.004 0.154 ( 0.004 0.011 ( 0.001

0.15 ( 0.03 0.07 ( 0.01 0.223 ( 0.002 0.0086 ( 0.0003

A Xe lamp and a WG305 high-pass filter were used.

Role of Photoionization. Because photoionization requires the formation of P+, the absence of a significant contribution of this intermediate at low O2 concentrations argues against this process for most PAHs (Figure 1). Nevertheless, to test further for photoionization, particularly for AN, BA, AC, and MA, which exhibited significant degradation in N2-purged solutions, photodegradation rate constants were also acquired in solutions saturated with N2O (but absence of ACN; see Experimental Section). If photoionization is important, the hydrated electron produced via this process will react with N2O to form the hydroxyl radical (‚OH) (eq 5) (23); because the PAH is the only available reactant, this radical will react with the PAH (eq 6), producing an increase in the photodegradation rate constant relative to N2-purged solutions.

N2O + e- + H2O f ‚OH + OH- + N2 .

OH + PAH f products

(5) (6)

Contrary to this prediction, rate constants of AN, BA, and MA decreased in N2O-purged solutions as compared to N2purged solutions, presumably due to slightly better deoxygenation with N2O (Table 3). Although the rate constant of AC did increase slightly, it did not double as expected from eqs 5 and 6. The results obtained from these experiments as well as from the P+ experiments indicate that photoionization contributes very little to the photodegradation of PAHs at irradiance levels comparable to those of surface solar radiation.

Discussion Although it has been suggested that photoionization plays a role in the aqueous photodegradation of PAHs (3, 9), the results from this study as well as from previous work (2) argue strongly against this mechanism. First, if photoionization was an important process, substantial amounts of P+ should be generated independent of O2 concentration. However, little or no production of P+ was observed at low O2 concentrations for most PAHs (with one exception, see

above). Second, photodegradation rate constants were not enhanced in the presence of N2O, which will react with hydrated electron to form ‚OH and thus accelerate the degradation. Third, all PAHs examined previously exhibited wavelength-independent photodegradation quantum yields (2), which would not be expected for a direct photoionization mechanism; instead, quantum yields should increase with decreasing wavelength due to the greater probability of photoionization at higher photon energies. Fourth, quantum yields do not correlate with PAH ionization or oxidation potentials (2). Taken together, these results lead us to conclude that photoionization is unimportant in the aqueous air-equilibrated photodegradation of PAHs. Although the uncharacteristic behavior of AN in the presence and absence of 3ap in N2-purged solutions (Figure 1) could be interpreted as due to photoionization, the results from the N2O experiment contradict this view. At this time, we have no good explanation for the anomalous behavior exhibited by this compound under N2-purging. At O2 concentrations greater than 250 µM, increasing amounts of P+ are generated for all PAHs examined except PE (Figure 1). Because the triplet lifetimes of PAHs are exceedingly long (>100 µs) (21) and the rate constants for triplet quenching by O2 are very large (g109 M-1 s-1) (24), the triplet state will be quenched quantitatively at very low O2 concentrations and thus cannot account for the enhanced P+ production observed at high O2 concentrations. Instead, this P+ production must arise from electron transfer via diffusional quenching of the excited singlet state by O2. Thus, these results indicate that PAH photodegradation proceeding via the singlet state occurs predominately through electron transfer to O2 (Scheme 1, eqs 12 and 7). In contrast, at low O2 concentrations (e250 µM), where reaction with the triplet tends to dominate (due to the much longer triplet lifetimes), little P+ formation is observed, indicating that photodegradation via the triplet proceeds primarily through a different pathway, most likely through a direct reaction within the collision complex (Scheme 1, either eq 20 or 21). For some PAHs, finite quantum yields were observed in N2-purged solutions (Figure 1). We attribute this result to our inability to exclude traces of O2 rigorously from these solutions as well as to the very long triplet lifetimes and large rate constants for triplet quenching by O2 (see above). For PAHs such as MA, CH, AC, and AN, which exhibit significant quantum yields at low O2 concentrations (e250 µM), the triplet state appears to contribute substantially to the photodegradation under aerated conditions (Figure 1). In contrast, PAHs such as BP, BK, PY, and PE, whose quantum yields do go to zero under N2-purging, appear to react primarily through the singlet state (however, see further ref 25). The absence of a significant triplet contribution to PY photodegradation is consistent with previous work (8), but the large triplet contribution to MA photodegradation is not VOL. 37, NO. 24, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5771

(3). Although the photodegradation of PY and BP proceeds almost exclusively through the excited singlet state, not all of this degradation proceeds through the P+ intermediate (Figure 1), suggesting a competitive branching between solvent separation (Scheme 1, eq 7) and reaction within the collision complex (Scheme 1, eq 11). Because of its low quantum yield for intersystem crossing (0.01) (21), it is not surprising that PE undergoes little photodegradation under very low O2 concentrations. It is surprising however, that contrary to the other PAHs that react through the excited singlet state, PE shows no evidence of photodegradation through a P+ intermediate. A recently completed, detailed kinetic analysis of these and additional data provides a quantitative assessment of the involvement of the excited singlet and triplet states in the photodegradation as well as the relative contributions of electron transfer versus direct reaction from each state (25). This work has shown that dilute aqueous solutions of PAHs photodegrade through reaction(s) of O2 with the excited states of these compounds. Depending on the PAH and O2 concentration, degradation can proceed via reaction with the excited singlet, triplet, or both states. Under aerobic conditions, we find that the contribution of P+ to the degradation is small or nonexistent for most of the PAHs examined. This result is consistent with our previous work with natural waters (2), where little inhibition of photodegradation was observed in waters containing potentially high concentrations of P+ donors, such as Br- or I- in seawater and organic matter in freshwaters. Further, we can predict from our results that the type of reaction products observed should vary with O2 concentration and with the mechanism of PAH reaction (e.g., singlet versus triplet). Future work will address this issue, while other recent work has been directed to analyzing these results within a quantitative kinetic framework (25).

Acknowledgments This work was supported by grants to N.V.B. from the Office of Naval Research (N00014-95-10201 and N00014-99-10034) and the United States Environmental Protection Agency through the Science to Achieve Results (STAR) program. We thank the Dan Falvey Lab (University of Maryland, College Park) for help with the LFP experiments.

Supporting Information Available LFP experiments showing BP+ and MA+ quenching by 3ap, NaI, and K4Fe(CN)6 as well as a comparison of steady-state fluorescence and HPLC detection methods are provided.

5772

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 24, 2003

Literature Cited (1) Neff, J. M. Polycyclic Aromatic Hydrocarbons in the Aquatic Environment; Applied Science Publishers: London, 1979. (2) Fasnacht, M. P.; Blough, N. V. Environ. Sci. Technol. 2002, 36, 4364-4369. (3) Zepp, R. G.; Schlotzhauer, P. F. In Polynuclear Aromatic Hydrocarbons; Jones, P. W., Leber, P., Eds.; Ann Arbor Science Publishers: Ann Arbor, MI, 1979; pp 141-158. (4) Mill, T.; Mabey, W. R.; Lan, B. Y.; Baraze, A. Chemosphere 1981, 10, 1281-1290. (5) Fukuda, K.; Inagaki, Y.; Maruyama, T.; Kojima, H. I.; Yoshida, T. Chemosphere 1988, 17, 651-659. (6) Sigman, M. E.; Zingg, S. P.; Pagni, R. M.; Burns, J. H. Tetrahedron Lett. 1991, 32, 5737-5740. (7) Sigman, M. E.; Chevis, E. A.; Brown, A.; Barbas, J. T.; Dabestani, R.; Burch, E. L. J. Photochem. Photobiol., A 1996, 94, 149-155. (8) Sigman, M. E.; Schuler, P. F.; Ghosh, M. M.; Dabestani, R. T. Environ. Sci. Technol. 1998, 32, 3980-3985. (9) Vialaton, D.; Richard, C.; Baglio, D.; Paya-Perez, A.-B. J. Photochem. Photobiol., A 1999, 123, 15-19. (10) Lehto, K.; Vuorimaa, E.; Lemmetyinen, H. J. Photochem. Photobiol., A 2000, 136, 53-60. (11) Miller, J. S.; Olejnik, D. Water Res. 2001, 35, 233-243. (12) McConkey, B. J.; Hewitt, L. M.; Dixon, D. G.; Greenberg, B. M. Water Air Soil Pollut. 2002, 136, 347-359. (13) Blough, N. V.; Zepp, R. G. In Active Oxygen in Chemistry; Foote, C. S., Valentine, J. S., Greenberg, A., Liebman, J. F., Eds.; Chapman and Hall: New York, 1995; pp 280-333. (14) Steenken, S.; Warren, C. J.; Gilbert, B. C. J. Chem. Soc., Perkin Trans. 2 1990, 335-342. (15) Turro, N. J. Modern Molecular Photochemistry; University Science Books: Sausalito, CA, 1991. (16) Kieber, D. J.; Blough, N. V. Anal. Chem. 1990, 62, 2275-2283. (17) Pochon, A.; Vaughan, P. P.; Gan, D.; Vath, P.; Blough, N. V.; Falvey, D. E. J. Phys. Chem. A 2002, 106, 2889-2894. (18) Koike, K.; Thomas, J. K. J. Chem. Soc., Faraday Trans. 1992, 88, 195-200. (19) Shida, T.; Iwata, S. J. Am. Chem. Soc. 1973, 95, 3473-3483. (20) Shida, T. Electronic Absorption Spectra of Radical Ions; Elsevier: Amsterdam, 1988. (21) Dabestani, R.; Ivanov, I. N. Photochem. Photobiol. 1999, 70, 10-34. (22) Murov, S. L. Handbook of Photochemistry, 2nd ed.; M Dekker: New York, 1993. (23) Thomas-Smith, T. E.; Blough, N. V. Environ. Sci. Technol. 2001, 35, 2721-2726. (24) Abdel-Shafi, A. A.; Wilkinson, F. J. Phys. Chem. A 2000, 104, 5747-5757. (25) Fasnacht, M. P.; Blough, N. V. Aquat. Sci. 2003, 65, 349-355.

Received for review April 24, 2003. Revised manuscript received September 28, 2003. Accepted October 1, 2003. ES034389C