Elisa De Laurentiis

Photochemical Fate of Carbamazepine in Surface Freshwaters: Laboratory Measures and Modeling

It is shown here that carbamazepine (CBZ) would undergo direct photolysis and reaction with (•)OH as the main phototransformation pathways in surface waters. Environmental lifetimes are expected to vary from a few weeks to several months, and predictions are in good agreement with available field data. Acridine (I) and 10,11-dihydro-10,11-trans-dihydroxy-CBZ (V) are the main quantified phototransformation intermediates upon direct photolysis and (•)OH reaction, respectively. The photochemical yield of mutagenic I from CBZ is in the 3-3.5% range, and it is similar for both direct photolysis and (•)OH reaction: it would undergo limited variation with environmental conditions. In contrast, the yield of V would vary in the 4-8.5% range depending on the conditions, because V is formed from CBZ by (•)OH (9.0% yield) more effectively than upon direct photolysis (1.4% yield). Other important photointermediates, mostly formed from CBZ upon (•)OH reaction, are an aromatic-ring-dihydroxylated CBZ (VI) and N,N-bis(2-carboxyphenyl)urea (VII). Compounds VI and VII are formed by photochemistry and are not reported as human metabolites; thus, they could be used as tracers of CBZ phototransformation in surface waters. Interestingly, VI has recently been detected in river water.

Modelling the photochemical fate of ibuprofen in surface waters.

We show that the main photochemical processes involved in the phototransformation of anionic ibuprofen (IBP) in surface waters are the reaction with (•)OH, the direct photolysis and possibly the reaction with the triplet states of chromophoric dissolved organic matter ((3)CDOM). These conclusions were derived by use of a model of surface water photochemistry, which adopted measured parameters of photochemical reactivity as input data. The relevant parameters are the polychromatic UVB photolysis quantum yield (Φ(IBP) = 0.33 ± 0.05, μ±σ), the reaction rate constant with (•)OH (k(IBP,•OH)=(1.0 ± 0.3)⋅10(10) M(-1) s(-1)), the (1)O(2) rate constant (k(IBP,)( ¹O₂)= (6.0 ± 0.6)⋅10(4) M(-1) s(-1)), while the reaction with CO(3)(-•) can be neglected. We adopted anthraquinone-2-sulphonate (AQ2S) and riboflavin (Ri) as CDOM proxies and the reaction rate constants with the respective triplet states were k(IBP,3AQ2S)=(9.7 ± 0.2)⋅10(9) M(-1) s(-1) and k(IBP,3Ri) = 4.5⋅10(7) M(-1) s(-1). The reaction with (3)CDOM can be an important IBP sink if its rate constant is comparable to that of (3)AQ2S, while it is unimportant if the rate constant is similar to the (3)Ri* one. The photochemical pathways mainly lead to the transformation (oxidation and/or shortening) of the propanoic lateral chain of IBP, which appears to be significantly more reactive than the isobutyl one. Interestingly, none of the detected intermediates was produced by substitution on the aromatic ring.

Assessing the photochemical transformation pathways of acetaminophen relevant to surface waters: transformation kinetics, intermediates, and modelling.

This work shows that the main photochemical pathways of acetaminophen (APAP) transformation in surface waters would be direct photolysis (with quantum yield of (4.57 ± 0.17)⋅10(-2)), reaction with CO3(-·) (most significant at pH > 7, with second-order rate constant of (3.8 ± 1.1)⋅10(8) M(-1) s(-1)) and possibly, for dissolved organic carbon higher than 5 mg C L(-1), reaction with the triplet states of chromophoric dissolved organic matter ((3)CDOM*). The modelled photochemical half-life time of APAP in environmental waters would range from days to few weeks in summertime, which suggests that the importance of phototransformation might be comparable to biodegradation. APAP transformation by the main photochemical pathways yields hydroxylated derivatives, ring-opening compounds as well as dimers and trimers (at elevated concentration levels). In the case of (3)CDOM* (for which the triplet state of anthraquinone-2-sulphonate was used as proxy), ring rearrangement is also hypothesised. Photochemistry would produce different transformation products (TPs) of APAP than microbial biodegradation or human metabolism, thus the relevant TPs might be used as markers of APAP photochemical reaction pathways in environmental waters.

Photochemical production of organic matter triplet states in water samples from mountain lakes, located below or above the tree line

The production of triplet states (T(*)) of chromophoric dissolved organic matter (CDOM), reacting with the probe molecule 2,4,6-trimethylphenol (TMP) was measured upon irradiation of water samples, taken from lakes located in a mountain area (NW Italy) between 1450 and 2750 m above sea level. The lakes are located below or above the tree line and surrounded by different vegetation types (trees, alpine meadows or exposed rocks). The most photoactive samples belonged to lakes below the tree line and their fluorescence spectra and CDOM optical features suggested the presence of a relatively elevated amount of humic (allochthonous) material. The lowest (negligible) photoactivity was found for a lake surrounded by exposed rocks. Its CDOM showed an important autochthonous contribution (due to in-lake productivity) and considerably higher spectral slope compared to the other samples, suggesting low CDOM molecular weight and/or aromaticity. Among the samples, CDOM photoactivity (measured as the rate of TMP-reactive T(*) photoproduction) decreased with changing vegetation type in the order: trees, meadows, rocks. It could be connected with decreasing contribution from catchment runoff and increasing contribution from autochthonous processes and possibly precipitation.

Photochemical processes involving the UV absorber benzophenone-4 (2-hydroxy-4-methoxybenzophenone-5-sulphonic acid) in aqueous solution: reaction pathways and implications for surface waters.

The sunlight filter benzophenone-4 (BP-4) is present in surface waters as two prevailing forms, the singly deprotonated (HA-) and the doubly deprotonated one (A(2-)), with pKa2 = 7.30 ± 0.14 (μ ± σ, by dissociation of the phenolic group). In freshwater environments, BP-4 would mainly undergo degradation by reaction with ·OH and direct photolysis. The form HA(-) has a second-order reaction rate constant with ·OH (k(·OH)) of (1.87 ± 0.31)·10(10) M(-1) s(-1) and direct photolysis quantum yield Φ equal to (3.2 ± 0.6)·10(-5). The form A(2-) has (8.46 ± 0.24)·10(9) M(-1) s(-1) as the reaction rate constant with ·OH and (7.0 ± 1.3)·10(-5) as the photolysis quantum yield. The direct photolysis of HA(-) likely proceeds via homolytic breaking of the O-H bond of the phenolic group to give the corresponding phenoxy radical, as suggested by laser flash photolysis experiments. Photochemical modelling shows that because of more efficient direct photolysis (due to both higher sunlight absorption and higher photolysis quantum yield), the A(2-) form can be degraded up to 3 times faster than HA(-) in surface waters. An exception is represented by low-DOC (dissolved organic carbon) conditions, where the ·OH reaction dominates degradation and the transformation kinetics of HA(-) is faster compared to A(2-). The half-life time of BP-4 in mid-latitude summertime would be in the range of days to weeks, depending on the environmental conditions. BP-4 also reacts with Br2(·-), and a rate constant k(Br2(·-),BP-4) = (8.05 ± 1.33)·10(8) M(-1) s(-1) was measured at pH 7.5. Model results show that reaction with Br2(·-) could be a potentially important transformation pathway of BP-4 in bromide-rich (e.g. seawater) and DOM-rich environments.

Optical and photochemical characterization of chromophoric dissolved organic matter from lakes in Terra Nova Bay, Antarctica. Evidence of considerable photoreactivity in an extreme environment.

Water samples from shallow lakes located in Terra Nova Bay, Antarctica, were taken in the austral summer season and characterized for chemical composition, optical features, fluorescence excitation-emission matrix (EEM) and photoactivity toward the generation of (•)OH, (1)O2, and (3)CDOM* (triplet states of chromophoric dissolved organic matter). The optical properties suggested that CDOM would be largely of aquagenic origin and possibly characterized by limited photochemical processing before sampling. Moreover, the studied samples were highly photoactive and the quantum yields for the generation of (3)CDOM* and partially of (1)O2 and (•)OH were considerably higher compared to water samples from temperate environments. This finding suggests that water in the studied lakes would have considerable ability to photosensitize the degradation of dissolved compounds during the austral summer, possibly including organic pollutants, also considering that the irradiance conditions of the experiments were not far from those observed on the Antarctic coast during the austral summer.

Phototransformation of the sunlight filter benzophenone-3 (2-hydroxy-4-methoxybenzophenone) under conditions relevant to surface waters.

The UV filter benzophenone-3 (BP3) has UV photolysis quantum yield ΦBP3=(3.1±0.3)·10(-5) and the following second-order reaction rate constants: with (•)OH, k(BP3,(•)OH)=(2.0±0.4)·10(10) M(-1) s(-1); with the triplet states of chromophoric dissolved organic matter ((3)CDOM*), K(BP3,(3)CDOM*)=(1.1±0.1)·10(9) M(-1) s(-1); with (1)O2, k(BP3,(1)O2)=(2.0±0.1)·10(5) M(-1) s(-1), and with CO3(-•), k(BP3,CO3(-•))<5·10(7) M(-1) s(-1). These data allow the modelling of BP3 photochemical transformation, which helps filling the knowledge gap about the environmental persistence of this compound. Under typical surface-water conditions, direct photolysis and reactions with (•)OH and (3)CDOM* would be the main processes of BP3 phototransformation. Reaction with (•)OH would prevail at low DOC, direct photolysis at intermediate DOC (around 5 mg C L(-1)), and reaction with (3)CDOM* at high DOC. If the reaction rate constant with CO3(-•) is near the upper limit of experimental measures (5·10(7) M(-1) s(-1)), the CO3(-•) degradation process could be somewhat important for DOC<1 mg C L(-1). The predicted half-life time of BP3 in surface waters under summertime conditions would be of some weeks, and it would increase with increasing depth and DOC. BP3 transformation intermediates were detected upon reaction with (•)OH. Two methylated derivatives were tentatively identified, and they were probably produced by reaction between BP3 and fragments arising from photodegradation. The other intermediates were benzoic acid (maximum concentration ~10% of initial BP3) and benzaldehyde (1%).

Assessing the occurrence of the dibromide radical (Br2−) in natural waters: Measures of triplet-sensitised formation, reactivity, and modelling

The triplet state of anthraquinone-2-sulphonate (AQ2S) is able to oxidise bromide to Br(•)/Br(2)(-•), with rate constant (2-4)⋅10(9)M(-1)s(-1) that depends on the pH. Similar processes are expected to take place between bromide and the triplet states of naturally occurring chromophoric dissolved organic matter ((3)CDOM*). The brominating agent Br(2)(-•) could thus be formed in natural waters upon oxidation of bromide by both (•)OH and (3)CDOM*. Br(2)(-•) would be consumed by disproportionation into bromide and bromine, as well as upon reaction with nitrite and most notably with dissolved organic matter (DOM). By using the laser flash photolysis technique, and phenol as model organic molecule, a second-order reaction rate constant of ~3⋅10(2)L(mg C)(-1)s(-1) was measured between Br(2)(-•) and DOM. It was thus possible to model the formation and reactivity of Br(2)(-•) in natural waters, assessing the steady-state [Br(2)(-•)]≈10(-13)-10(-12)M. It is concluded that bromide oxidation by (3)CDOM* would be significant compared to oxidation by (•)OH. The (3)CDOM*-mediated process would prevail in DOM-rich and bromide-rich environments, the latter because elevated bromide would completely scavenge (•)OH. Under such conditions, (•)OH-assisted formation of Br(2)(-•) would be limited by the formation rate of the hydroxyl radical. In contrast, the formation rate of (3)CDOM* is much higher compared to that of (•)OH in most surface waters and would provide a large (3)CDOM* reservoir for bromide to react with. A further issue is that nitrite oxidation by Br(2)(-•) could be an important source of the nitrating agent (•)NO(2) in bromide-rich, nitrite-rich and DOM-poor environments. Such a process could possibly account for significant aromatic photonitration observed in irradiated seawater and in sunlit brackish lagoons.

Photochemical generation of photoactive compounds with fulvic-like and humic-like fluorescence in aqueous solution

The irradiation of L-tryptophan, L-tyrosine and 4-phenoxyphenol in aqueous solution produced compounds with similar fluorescence properties as humic substances, and with absorption spectra that were significantly extended into the UVA and visible regions compared to the starting compounds. The irradiated systems had photosensitizing properties, as proven by the photodegradation of 2,4,6-trimethylphenol and furfuryl alcohol (probes of excited triplet states and (1)O2, respectively). The described photochemical processes could constitute an additional pathway for the formation of humic substances in clear and shallow water bodies, which would be added to the complex network of reactions involving dissolved organic matter.

Modelling lake-water photochemistry: Three-decade assessment of the steady-state concentration of photoreactive transients (OH, CO3- and 3CDOM∗) in the surface water of polymictic Lake Peipsi (Estonia/Russia)

Over the last 3-4 decades, Lake Peipsi water (sampling site A, middle part of the lake, and site B, northern part) has experienced a statistically significant increase of bicarbonate, pH, chemical oxygen demand, nitrate (and nitrite in site B), due to combination of climate change and eutrophication. By photochemical modelling, we predicted a statistically significant decrease of radicals ·OH and CO3(-·) (site A, by 45% and 35%, respectively) and an increase of triplet states of chromophoric dissolved organic matter ((3)CDOM(∗); site B, by ∼25%). These species are involved in pollutant degradation, but formation of harmful by-products is more likely with (3)CDOM(∗) than with ·OH. Therefore, the photochemical self-cleansing ability of Lake Peipsi probably decreased with time, due to combined effects of climate change and eutrophication. In different environments (e.g. Lake Maggiore, NW Italy), ecosystem restoration policies had the additional advantage of enhancing sunlight-driven detoxification, suggesting that photochemical self-cleansing would be positively correlated with lake water quality.