The hydroxyl radical (HO•) is a photochemically generated species that is important in the attenuation of organic contaminants in sunlit natural surface waters and in advanced oxidation processes (AOPs) used to treat drinking and wastewater. The steady state concentration of HO•, ([HO•]ss), in photochemical systems is typically measured indirectly using probe compounds. In this work, we investigate the use of caffeine as a photochemical probe for the detection of HO• in the presence of dissolved organic matter (DOM) solutions derived from allochthonous and autochthonous precursors using both the first order and initial rates approach to kinetic analysis to determine [HO•]ss in aqueous solution. Values for [HO•]ss are found to agree within a factor of two of values measured using terepthalic acid (TPA), a well-established probe for the detection of HO•, in the identical DOM solutions. Additionally, we find that caffeine is more selective for HO• than previously believed since it does not appear to undergo reaction with long-lived reactive transients as was formerly presumed.
The presence of nitrate in wastewater effluent could play an important role in the formation of HO•, and therefore also in the fate of contaminants in sunlit receiving waters. We investigate the interplay between HO• formation derived from both effluent dissolved organic matter (EfOM) and nitrate (NO3-). The assumption of little to no interaction between NO3- and EfOM during HO• production (RHO•), needed to determine the contribution of DOM to hydroxyl radical formation in waters containing NO3- without
extracting EfOM from the wastewater, is invalid. We based our findings on measurements of RHO• and [HO•]ss using benzene and caffeine as HO• probes in both whole wastewater effluent and solutions of EfOM isolated by solid phase extraction from the same effluent sample. The RHO• value measured for the isolate in the absence of NO3- (0.074 ± 0.009 (SE) nM s-1) is 1.5 fold higher than that found for bulk, non-isolated EfOM (0.049 ± 0.026 (SE) nM s-1), which was calculated by subtracting the estimated contribution of NO3- to overall HO• production from the RHO• obtained for the Southerly whole wastewater. This difference calculated this way is not significantly different (a=0.05, p=0.39). We find that we are able to reproduce the RHO• found with benzene for Southerly wastewater effluent (0.199 ± 0.022 nM s-1) using a NO3-ROH• curve generated in a Southerly isolate solution (0.178 ± 0.011 nM s-1). Similarly, [HO•]ss values measured in Southerly whole wastewater effluent using caffeine as a probe compound (1.32 ± 0.096 fM) were comparable to measurements made in Southerly EfOM (1.50 ± 0.091 fM). We revise the ratio at which NO3- and DOM contribute equally in solution from the value reported by Vione et al. (2006) of 3.3 x 10-5 [(mol NO3-) (mg-C)-1]) upwards to 8.1 ± 1.5 x 10-5 [(mol NO3-) (mg-C)-1] based on the average of the ratios determined for the three DOM isolates used in this study. Values for [HO•]ss in IHSS isolate solutions and whole wastewater effluent determined by benzene and caffeine are in good agreement, which supports the use of caffeine as a credible HO• probe compound, subject to caveats detailed within this work, in both isolate solutions and whole water containing NO3-.
Organic UV-filters are key ingredients found in sunscreens, cosmetics and plastic goods. Concerns have been raised about potential ecological and human health effects of certain organic UV filters that are currently FDA approved for use in the United States. Here, we investigate the photochemical fate of two of these compounds, oxybenzone and sulisobenzone. Both oxybenzone and sulisobenzone have previously been detected in surface waters, seawater, and treated wastewater effluent. Enhanced photodegradation of oxybenzone and sulisobenzone was observed under simulated solar irradiation in solutions of International Humic Substance Society standards (Pony Lake fulvic acid and Suwannee River Natural Organic Matter), filtered wastewater effluent (Southerly Wastewater Treatment Plant in Lockbourne, OH), and Scioto River water (Columbus, OH). Quenching experiments with isopropanol revealed that the main pathway for degradation appears to be reaction with the hydroxyl radical (HO•). Observed degradation rates were 2-3 times slower than estimates calculated using literature reported second-order rate constants and measured hydroxyl radical steady-state concentrations for SRNOM, PLFA and Scioto waters. The Southerly sample, however, exhibited nearly identical expected and observed rate constants, which we take to indicate the presence of unidentified reactive species that can react with oxybenzone and sulisobenzone. Values obtained in this work were used to calculate second-order rate constants for oxybenzone and sulisobenzone with the hydroxyl radical, as well as to estimate environmental half-lives for these compounds. Near surface 24-hour averaged half-lives of 3.0 and 4.0 days were calculated for oxybenzone and sulisobenzone,respectively. When extrapolated to an environmentally representative water column, these same 24-hour averaged half-lives increased to 2.4 and 3.5 years, respectively.