Environ. Sci. Technol. 2006, 40, 3775-3781
Sources and Sinks of Hydroxyl Radicals upon Irradiation of Natural Water Samples D A V I D E V I O N E , * ,† GIANPAOLO FALLETTI,† VALTER MAURINO,† CLAUDIO MINERO,† EZIO PELIZZETTI,† MERY MALANDRINO,† ROBERTO AJASSA,‡ ROMEO-IULIAN OLARIU,§ AND CECILIA ARSENE§ Dipartimento di Chimica Analitica, Universita` di Torino, Via P. Giuria 5, 10125 Torino, Italy, Dipartimento di Scienze della Terra, Universita` di Torino, Via V. Caluso 37, 10125 Torino, Italy, and Department of Inorganic and Analytical Chemistry, “Al. I. Cuza” University of Iasi, Carol I Bd. 11, 700506 Iasi, Romania
Hydroxyl radical formation rates, steady-state concentration, and overall scavenging rate constant were measured by irradiation of surface lake water samples from Piedmont (NW Italy) and nitrate-rich groundwater samples from Moldova (NE Romania). Dissolved organic matter (DOM) was the main source and sink of •OH upon lake water irradiation, with [•OH] being independent of DOM amount. Water oxidation by photoexcited DOM is a likely •OH source in the presence of very low levels of nitrate and dissolved iron. Under different circumstances it is not possible to exclude other processes, e.g., DOM-enhanced photoFenton reactions. Under the hypotheses of no interaction and absence of mutual screening of radiation, nitrate would prevail over DOM as •OH source for a NO3-/DOM ratio higher than 3.3 × 10-5 (mol NO3-) (mg C)-1, DOM prevailing for lower values. Substantial DOM photolability was observed upon irradiation of nitrate-rich groundwater, mainly due to the elevated •OH generation rate. For the first time to our knowledge, evidence was also obtained of the photoformation of potentially toxic and/or mutagenic nitroaromatic compounds upon irradiation of natural lake water and groundwater samples, proportionally to the nitrate levels.
Introduction Photochemical processes are important pathways for the transformation of poorly biodegradable natural water components. They include direct photolysis, photosensitized transformation upon energy, electron, or hydrogen transfer involving photoexcited molecules, and reaction with photogenerated radical or transient species (1-4). Among the reactive species that can be photochemically generated in natural waters (•OH, 1O2, ROO•, CO3-• (5), Cl2-• and •NO2 (6)), • OH plays an important role because of the elevated reaction rate constants toward a very wide variety of organic and inorganic compounds (7). * Corresponding author phone: +39-011-6707633; fax: +39-0116707615; e-mail:
[email protected]. † Dipartimento di Chimica Analitica, Universita ` di Torino. ‡ Dipartimento di Scienze della Terra, Universita ` di Torino. § Department of Inorganic and Analytical Chemistry, “Al. I. Cuza” University of Iasi. 10.1021/es052206b CCC: $33.50 Published on Web 05/10/2006
2006 American Chemical Society
The steady-state concentration of •OH in the surface layer of many natural aquifers lies in the range 10-17-10-15 M (8-11), with the whole water column values being 1-2 orders of magnitude lower, depending on the column depth and the thickness of the photic zone (10, 11). Very elevated [•OH] values (10-14-10-12 M) have been observed in surface waters contaminated by acidic mine drainage (12). Major photochemical •OH sources have been identified as nitrate, nitrite, and DOM (the latter possibly also involved in photo-Fenton processes), with the relative role varying depending on the conditions encountered in different studies (8, 9, 13-20). Moreover, DOM and (at a lesser extent) carbonate and bicarbonate are agreed to play a major role as •OH sinks in freshwater (1, 9-11, 14, 15). A renewed interest into •OH-mediated processes in natural waters has arisen in recent years because of their impact on the transformation of pesticides and pharmaceuticals (13, 15, 16, 18, 21, 22). An overview of the cited literature also indicates an increasing role attributed over time to DOM as • OH source. Possible explanations are the increasing success of the attempts aimed at controlling NO3-/NO2- levels in surface aquifers, and a growing awareness of the role of DOM as a photochemical •OH source (23, 24). Despite these advances, the role of DOM as •OH source is still to be conclusively assessed (24). There is also a definite lack of quantitative knowledge on the relative role of nitrate and DOM as •OH sources. Such data would allow the definition of the relative contributions of the two species given the measured values of [NO3-] and TOC or NPOC. Also, the mechanism of •OH photoproduction by DOM deserves further investigation. Contribution to cover these gaps is the main purpose of the present work. The studied samples consisted of surface lake water from Piedmont (NW Italy), with the lake choice aimed at obtaining a wide variety of conditions, and groundwater from Letcani (near Iasi, Moldova, NE Romania). In the latter case the choice was motivated by the very high nitrate levels generally found in the zone, together with the fact that groundwater is commonly used for irrigation purposes and illuminated during basin storage and runoff. •OH formation rate, steady-state concentration, overall scavenging rate constants, and DOM photostability were also determined in the samples. For the first time to our knowledge, evidence was also obtained of aromatic nitration upon irradiation of natural water samples.
Experimental Section Reagents, Materials, and Instrumentation. The adopted reagents, their degrees of purity, and the supplier data are reported in the Supporting Information. Natural water samples were vacuum filtered on 0.22 µm membranes (cellulose esters, Schleicher and Schuell) and kept refrigerated. Whenever possible, analyses and irradiation experiments were carried out in the same week of the sampling. Ions in natural water were determined by ion chromatography with conductivity suppression and detection (see Supporting Information), with the exception of HCO3-/CO32which were quantified as inorganic carbon with a Shimadzu TOC-5000 total organic carbon analyzer. pH was measured with a combined glass electrode connected to a Metrohm 713 pH meter. The concentration values of H2CO3, HCO3-, and CO32- were derived from the inorganic carbon data, the solution pH, and the acidic constants of the carbonate system corrected for the ionic strength of the solutions according to the Guntelberg equation (25). Organic carbon was quantified as non-purgeable organic carbon (NPOC) with the Shimadzu VOL. 40, NO. 12, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3775
TOC-5000, after acidification of the sample with concentrated HClO4 and a 20 min air purge to eliminate CO2. Fe(III) and Fe(II) were quantified spectrophotometrically, the former at 480 nm as complex with SCN-, and the latter at 510 nm as complex with 1,10-phenanthroline. The instrument used was a Varian Cary 100 Scan UV-Vis spectrophotometer, also adopted to measure full natural water spectra. Total Fe was determined by ICP-AES (Supporting Information). Hydrogen peroxide was measured spectrophotometrically at 352 nm upon oxidation of iodide (26). The seasonal trend of nitrate evolution in groundwater was monitored spectrophotometrically (Supporting Information). Quantification of •OH. Irradiation experiments were carried out under a Solarbox device (CO.FO.ME.GRA., Milan, Italy), equipped with a 1500 W Philips xenon lamp and a glass filter, to simulate the sunlight spectrum under summertime fair weather conditions. The lamp emission power was 22 W m-2 in the wavelength interval 290-400 nm (see Supporting Information for further details). The transformation reaction of benzene into phenol (95% yield, (27)) was used to quantify photoformed •OH. Benzene is a suitable compound for this investigation because it undergoes fast reaction with •OH (rate constant 7.8 × 109 M-1 s-1, (7)), does not absorb lamp radiation (absence of direct photolysis to phenol was checked), and does not alter either the pH or the buffer capacity of the solutions. Phenol formation was monitored rather than benzene disappearance, as benzene volatilization would significantly bias benzene degradation but not phenol formation (28). Further details are reported as Supporting Information. The initial formation rates of phenol were calculated as reported in refs 29 and 30.
Results and Discussion Validation of the System. The first issue in the present context is to ascertain the selectivity toward •OH of the reaction transforming benzene into phenol. Benzene is little reactive toward 1O2 (31), but significant phenol formation was observed upon irradiation of benzene in the presence of Aldrich humic acids. Reactive species produced upon humic acid irradiation are •OH, excited triplet states, 1O2, and aquated e- (1, 2, 4). Even in this case phenol is likely to form upon selective reaction between benzene and •OH because addition of 2-propanol decreased phenol initial formation rate by the amount expected from a competition between benzene and 2-propanol for reaction with •OH. This finding also indicates that 2-propanol is unlikely to modify the yield of phenol formation from benzene and •OH, therefore suggesting that organic compounds might have a limited impact on the reaction (also see Supporting Information). Moreover, the formation of phenol from benzene upon natural water irradiation was compared in the presence and in the absence of 0.085 M 2-propanol. The alcohol at the adopted concentration value has been reported to inhibit •OH-induced reactions, but to have little to negligible effects on the processes promoted by excited triplet states or 1O2 (21). The observed quenching of phenol photoformation upon irradiation of the samples in the presence of 2-propanol indicates that the transformation reaction of benzene into phenol is selective toward •OH. Irradiation of Lake Water Samples. Each sample was analyzed for various parameters (ions, NPOC, pH) and irradiated to obtain the initial •OH formation rate ROH, the steady-state [•OH], and the overall scavenging rate constant ΣikSi[Si]. The latter is the sum of the products between the concentration values of the •OH scavengers ([Si]) and their second-order rate constants for reaction with •OH (kSi). The lake water data are reported in Table 1. The results concerning ROH, [•OH], and ΣikSi[Si] will be discussed separately. 3776
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 12, 2006
• OH Sources. The initial formation rate of •OH under the adopted irradiation device (22 W m-2 UV radiation in solution) varied in the range 10-12-10-11 M s-1. As a comparison, the outdoor UV intensity at 45°N in the cloudless reference day July 15, 2004 peaked at solar noon at 29 W m-2. Lamp irradiation intensity being constant, the total outdoor energy density measured in the reference day (7.5 × 105 J m-2) corresponds to 9.5 h irradiation time under the lamp. Among the water components that can produce •OH upon irradiation, one is certainly the nitrate ion (32)
NO3- + hν +H+ f •OH + •NO2
[Φ1305nm ) 0.01]
(1)
The contribution of nitrate to •OH photogeneration was evaluated by irradiation of synthetic solutions containing different concentration values of nitrate (from 5 × 10-5 to 4 × 10-4 M), and 3 × 10-3 M benzene as the only •OH scavenger in the system. A calibration curve was derived of •OH generation rate vs nitrate concentration under the adopted irradiation conditions, that is ROH ) 1.7 × 10-7 s-1 [NO3-]. From the concentration values of nitrate in the lake water samples it is possible to derive the percentage of •OH photoformation accounted for by nitrate photochemistry, reported in Table 1. It ranges from negligible (corresponding to nitrate levels below detection limit) to around 40%, indicating that nitrate is not the prevailing source of •OH in the studied samples. It is also possible to subtract the nitrate contribution from the total •OH formation rate, thus obtaining the fraction of ROH unaccounted for by nitrate, [ROH - ROH/NO3-]. Figure 1 shows the correlation between [ROH - ROH/NO3-] and nonpurgeable organic carbon (NPOC). It is r2 ) 0.711, while the minimum r2 value for statistical significance for 7 data points according to the Pearson’s correlation at 95% confidence level is 0.57. The statistical significance of the correlation between [ROH - ROH/NO3-] and NPOC therefore suggests that DOM is likely to contribute substantially to the photochemical generation of •OH in the lake water samples under study. Possible pathways that can photochemically yield •OH in the presence of DOM are water oxidation by photoexcited DOM (reactions 2 and 3 (33)), DOM-enhanced photo-Fenton processes (reactions 4-7, where L1 is an organic ligand (34, 35)), and possibly the photochemical generation of nitrite (36).
DOM + hν f DOM*
(2)
DOM* + H2O f DOM-H• + •OH
(3)
FeIII - L1 + hν f Fe2+ + L1+•
(4)
Fe2+ + O2 f Fe3+ + O2-•
(5)
2 O2-• + 2 H+ f H2O2 + O2
(6)
Fe2+ + H2O2 f Fe3+ + •OH + OH-
(7)
The H2O2 precursor HO2•/O2-• can also be formed via reaction between •OH and aromatic compounds, because of hydrogen abstraction carried out by O2 on the radical intermediates of these reactions (2). To get better insight into DOM-assisted •OH photogeneration, a further sample was taken from the Lago Grande in Avigliana (Av. Grande) on June 30, 2005. Analysis yielded (9.22 ( 1.41) × 10-6 M NO3- and 6.29 ( 0.10 mg C L-1 NPOC. The initial formation rate of •OH was (3.13 ( 0.32) × 10-11 M s-1, to which nitrate would contribute a mere 5%. Upon irradiation of lake water alone (not spiked with benzene),
TABLE 1. Data Concerning the Studied Lake Water Samplesa
VOL. 40, NO. 12, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
Av. Piccolo
Av. Grande
Viverone
Orta
Meugliano
Mucrone
Rosset
sampling date:
05 Jul 2004
12 Jul 2004
19 Jul 2004
26 Jul 2004
30 Aug 2004
06 Sep 2004
13 Sep 2004
F-, M Cl-, M NO3-, M PO43-, M SO42-, M CO32-, M HCO3-, M H2CO3, M Na+, M NH4+, M K+, M Mg2+, M Ca2+, M NPOC, mg C L-1 pH ROH, M s-1 c %•OH(NO3-)d [•OH], Mc ΣikSi[Si], s-1 e %scav (HCO3- + CO32-)f t1/2(acet), equiv. summer daysg
(1.21 ( 0.06)E-5 (1.82 ( 0.03)E-4 (9.25 ( 0.27)E-5
