Grantee Research Project Results
2000 Progress Report: Role of Reduced Sulfur Species in Promoting the Transformation of Triazines in Estuaries and Salt Marshes
EPA Grant Number: R826269Title: Role of Reduced Sulfur Species in Promoting the Transformation of Triazines in Estuaries and Salt Marshes
Investigators: Roberts, A. Lynn , Salmun, H.
Institution: The Johns Hopkins University
EPA Project Officer: Chung, Serena
Project Period: February 1, 1998 through January 31, 2001 (Extended to January 31, 2002)
Project Period Covered by this Report: February 1, 2000 through January 31, 2001
Project Amount: $304,163
RFA: Exploratory Research - Environmental Chemistry (1997) RFA Text | Recipients Lists
Research Category: Water , Land and Waste Management , Air , Safer Chemicals
Objective:
The objectives of this research are to: (1) determine the rates of abiotic reactions of triazines (and related species) with inorganic reduced sulfur nucleophiles; (2) examine the ability of azines to bind covalently to natural organic matter (NOM); and (3) examine the potential impact of reduced sulfur species on the fate of azines in anoxic coastal marine environments. We hypothesize that reactions in sulfidic hydrologic environments between well-defined reduced sulfur species (particularly polysulfide ions and aromatic thiolate moieties present in NOM and chloro-s-triazine herbicides, fungicides, and reactive dye compounds may provide a significant sink for such organic contaminants.Progress Summary:
This report covers the third year of a 3-year project. Objectives 1 and 2 have been completed and Objective 3 is approximately 70 percent completed. Because of a staffing issue, we have obtained a no-cost extension and will be continuing this project for a fourth year.
Rates of reaction of three commonly used chloro-s-triazine agrochemicals with polysulfide dianions and HS- species were determined in aqueous solution at 25?C. Experiments with atrazine were conducted by varying the identity and concentration of the nucleophile, as well as solution pH, to verify that reactions were first-order in the reactive nucleophile concentration.
To provide information pertinent to the fate of reactive dye compounds, reactions of HS- and Sn2- with seven halogenated pyridine, pyrimidine, and quinoxaline compounds also were investigated. A fluorinated analog of atrazine was synthesized, and its reactions were studied. For the reactions of chloroazines with HS- and Sn2-, a general trend was observed in which an increasing number of chlorine substituents resulted in increased reactivity. Moreover, ring aza nitrogens, as well as the presence of a fused ring system (bicyclic heterocycles), were found to confer increased reactivity.
Large differences in the reactivity of haloazines toward polysulfide dianions versus HS- were observed, with ratios of kSn2-/kHS- ranging from 75 to more than 30,000. For example, atrazine is recalcitrant to reaction with HS- , yet reacts readily with Sn2-. Cyanazine reacts with HS- at a slow but measurable rate over a 24-day period. Rate constants for reaction with polysulfides vary by seven orders of magnitude, and kHS- by at least five orders of magnitude. Products were identified by derivatization with methyl iodide or pentafluorobenzyl bromide, followed by GC/MS/EI analysis. Results indicate substitution of halogen by sulfur. The products, the first-order dependence on the concentration of the nucleophile, and the qualitative structure-reactivity trends observed, are all consistent with a nucleophilic aromatic substitution (SNAr) reaction mechanism.
A quantitative structure-activity relationship (QSAR) comparing chloroazine reactivity to calculated LUMO energies was developed in the hope it might provide a useful tool for predicting the environmental fate of untested azines. This QSAR indicated a weak correlation between LUMO energy and reactivity, no doubt reflecting its neglect of steric effects as well as the subtle effects of transition state structure on the rate of reaction. Better correlations were observed within individual classes of chloroazines than for chloroazines as a whole.
Laboratory-derived rate constants for haloazines were extrapolated to an environmentally relevant concentration of hydrogen sulfide and polysulfides (reported for salt marsh sediment porewaters at pH 7.1; Boulegue, 1982). The resulting half-lives range from < 1 minute (for anilazine) to approximately 4 days (for atrazine). The half-life for the latter compound is much shorter than the reported uncatalyzed hydrolysis half-life of 1800 years at a pH of 6.97 (Plust, et al., 1981). For purposes of comparison, exchange coefficients for solutes between the water column and the sediment porewaters underlying the Chesapeake Bay are on the order of 0.1-1 m/d; characteristic times (for a 5-m water column) are thus on the order of 5-50 days. The mean residence time of water in the Chesapeake Bay is ~ 50 days. Abiotic reactions with reduced sulfur nucleophiles?particularly polysulfides?present in sediment porewaters could therefore exert a significant effect on the fate of azine agrochemicals or reactive textile dyes in this estuary.
To test the ability of laboratory-derived rate constants to predict haloazine fate in the much more complicated milieu of natural waters, several samples from a natural sulfidic lake water were fortified with sulfur and the pH was adjusted to produce varying concentrations of polysulfides. These modified waters were then allowed to react with atrazine. A sharp increase in reactivity was observed on increasing polysulfide concentration. The observed atrazine transformation products in each case were consistent with chlorine displacement. The observed transformation rates compared very favorably to those predicted by the laboratory-derived second-order rate constants and the measured concentrations of S[Sn2-] in each solution. Similar results were obtained for cyanazine, except that a mixture of transformation products (indicative of hydrolysis of the nitrile moiety in addition to displacement of chlorine) was observed.
During the coming year, we intend to seek evidence of whether abiotic SNAr reactions with reduced sulfur nucleophiles within sediment porewaters could affect contaminant fate in the Chesapeake Bay ecosystem. To address this issue, we intend to measure vertical profiles for atrazine and other haloazine agrochemicals (as well as their environmental transformation products) within the Chesapeake Bay. An apparent decrease in atrazine concentration over and above that anticipated from dilution via mixing with seawater, coupled with observation of transformation products, would provide evidence of such processes.
Our ability to obtain the detailed spatial resolution needed will be facilitated by our recent acquisition of a large volume injector for our GC/MS. Method development for the parent herbicides is well under way. Preliminary tests have been conducted in which herbicides spiked into 100-mL samples of natural waters (containing high concentrations of dissolved organic carbon) have been concentrated via solid phase extraction. Samples are eluted and exchanged into 1 mL of an appropriate solvent, and then 100 µL aliquots have been analyzed by GC/MS. Detection limits are on the order of 0.5 ppt, many orders of magnitude below concentrations (~ 0.03-4.3 ppb) reported from Chesapeake Bay surface water samples.
To allow us to quantify potential environmental degradation products, the sulfur substituted product of atrazine reaction with polysulfides has been synthesized and pentafluorobenzylated. This derivative has been purified (by column chromatography), has been characterized by MS and NMR, and has been dried so that a reference material exists. During the coming months, we will develop methods for recovering trace concentrations of mercaptoatrazine from natural waters, based on solid-phase extraction/derivatization methods.
To complement our laboratory and field investigations with modeling studies, a three-dimensional hydrodynamic model, INTROGLLVHT (developed by J.E. Edinger Associates, Inc.), is being adapted. INTROGLLVHT is a comprehensive software package used for hydrodynamic and water quality modeling. Unlike the previously tested box-model, this model accounts for horizontal advection, and both horizontal and vertical diffusion, all of which are important processes in the Bay. The software simulates the reaction between atrazine and polysulfides based on a specified pseudo first-order rate constant, based on a laboratory-derived second-order rate constant and an assumed polysulfide concentration. This model has been used to simulate circulation in reservoirs, small bays, estuaries and other enclosed water bodies. Tests of this model reveal that INTROGLLVHT adequately describes the physical and chemical processes that characterize the Chesapeake Bay. Modeled salinity and temperature fields, and flow patterns coincide with published characteristics of the Upper Bay. Salinity profiles generated from model runs show increasing salinity with depth and a strong pycnocline between approximately 6 m and 17 m. Temperature profiles show decreasing temperature with depth and, as expected, the temperature gradient is not as steep as the salinity gradient. Superimposed on the tidal motion, mean flow characteristics are typical of the estuary: surface waters flow toward the south and bottom waters toward the north, or landward, constituting the gravitational circulation.
We have applied this model to the Upper Chesapeake Bay. A significant fluvial input of atrazine enters this region. The detailed information needed to set up the model includes bathymetry, latitude, average wind speed, initial temperature and salinity, and the inflow rate, location, and concentration of atrazine within tributary rivers, groundwater, and runoff. Location of the tidal boundary, mean tidal height, tidal amplitude, tidal period, and salinity, temperature, and atrazine concentration at the tidal boundary are needed. In addition, it also is necessary to estimate the Chezy coefficient (which amounts to parameterizing the vertical mixing due to turbulent processes in the estuary), surface heat exchange coefficient, equilibrium temperature of heat exchange, and pseudo first-order rate constant for the reaction between atrazine and polysulfides.
To date, 23 simulations have been run, most of which addressed the sensitivity of the model to changes in bathymetry, groundwater input, the assumed pseudo first-order rate constant for atrazine reaction with polysulfides, wind speed, tidal height, and initial conditions of salinity, temperature, and atrazine concentration. Variation of these characteristics produced expected changes in the modeled flow, salinity, and atrazine concentration. The initial conditions and heat exchange data were chosen to simulate the late spring and summer months because most of the atrazine enters the Bay in the late spring. At present, we are in the process of conducting a detailed analysis of these results.
Our simulations consider that the potential reaction between atrazine and polysulfides may only take place under anoxic conditions, which are widespread in the summer. Data input correspond to average spring-summer values of salinity, temperature, flow rate, and atrazine concentrations. Model results yield predicted concentrations that are highest at the mouth of the Susquehanna River and decrease approximately 90 percent by the time the tidal boundary is reached, about 110 km south from the river. Vertical profiles show a constant value of atrazine in the first 6 m and a gradual decline in concentration with depth. Applying a pseudo first-order rate constant to the bottom depths of the Bay results in significantly reduced atrazine concentrations compared to those predicted in the absence of abiotic reactions. Future simulations will vary the concentration of polysulfides (and thus the pseudo first-order rate constant) with depth. Average daily and monthly results also will be generated and compared with averages taken over longer periods of time. We also will explore the possibility of including an exchange coefficient in the model to simulate the effect of azine diffusion into underlying sediment porewaters (with subsequent reactions with reduced sulfur nucleophiles therein). The results of these simulations should guide future sampling of atrazine concentrations in a water column of Chesapeake Bay.
Future Activities:
We will seek to determine vertical distributions of azine agrochemicals at several stations in the Upper Chesapeake Bay, as well as likely herbicide degradation products, along with relevant water quality parameters (dissolved oxygen, chloride, and concentrations of reduced sulfur nucleophiles). Field measurements will be complemented by numerical simulations of azine fate. This work will allow us to assess the extent to which important removal processes may exist for azine agrochemicals and fiber-reactive dyes.Journal Articles:
No journal articles submitted with this report: View all 23 publications for this projectSupplemental Keywords:
chemical transport, environmental chemistry, hydrology, modeling, Chesapeake Bay, agriculture, textile dyes, marine, estuary, herbicides, toxics, waste treatment, triazine, fiber-reactive dyes, agrochemicals., Scientific Discipline, Air, Waste, Ecosystem Protection/Environmental Exposure & Risk, Ecology, Chemistry, Fate & Transport, Engineering, Chemistry, & Physics, Biology, fate and transport, waste treatment, toxicology, estuaries, salt marshes, sulfur, kinetic models, Triazines, agriculture, halogenated hydrocarbons, water quality, herbicidesProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.