Quantum Chemical Prediction of Redox Reactivity and Degradation Pathways for Aqueous Phase Contaminants: An Example with HMPA JENS BLOTEVOGEL, † THOMAS BORCH,* ,†,‡ YURY DESYATERIK, § ARTHUR N. MAYENO, | AND TOM C. SALE ⊥ Departments of Soil and Crop Sciences, Chemistry, Atmospheric Science, Chemical & Biological Engineering, and Civil & Environmental Engineering, Colorado State University, Fort Collins, Colorado 80523 Received March 1, 2010. Revised manuscript received June 11, 2010. Accepted June 23, 2010. Models used to predict the fate of aqueous phase contaminants are often limited by their inability to address the widely varying redox conditions in natural and engineered systems. Here, we present a novel approach based on quantum chemical calculations that identifies the thermodynamic conditions necessary for redox-promoted degradation and predicts potential degradation pathways. Hexamethylphosphoramide (HMPA), a widely used solvent and potential groundwater contaminant, is used as a test case. Its oxidation is estimated to require at least iron-reducing conditions at low to neutral pH and nitrate- reducing conditions at high pH. Furthermore, the transformation of HMPA by permanganate is predicted to proceed through sequential N-demethylation. Experimental validation based on LC/ TOF-MS analysis confirms the predicted pathways of HMPA oxidation by permanganate to phosphoramide via the formation of less methylated as well as singly and multiply oxygenated reaction intermediates. Pathways predicted to be thermo- dynamically or kinetically unfavorable are similarly absent in the experimental studies. Our newly developed methodology will enable scientists and engineers to estimate the favorability of contaminant degradation at a specific field site, suitable approaches to enhance degradation, and the persistence of a contaminant and its reaction intermediates. Introduction Due to the vast number of anthropogenic chemicals in the environment and their potential adverse health effects, environmental fate predictions have become an indispen- sable component of risk assessment today (1). Furthermore, theoretical models that predict unknown physicochemical properties or biogeochemical reactivity are used to preassess newly developed compounds in order to identify environ- mental effects prior to large-scale production and com- mercialization (2). Currently, standard risk assessment strategies most frequently involve predictions based on quantitative structure-activity relationships (QSARs) (3). These approaches are attractive since they generate results with minimal computational costs once the relationship and descriptor values have been determined. However, the utility of QSARs is constrained as they rely on experimental databases and are often only valid for narrow ranges of conditions that do not necessarily cover the widely varying redox settings of natural and engineered environments. Thermodynamic properties ultimately govern the per- sistence or degradability of a compound in the environment. For many contaminants of concern, however, information on Gibbs free energies of formation or reaction in aqueous phase is unavailable, leading researchers to estimate fun- damental thermodynamic properties (4, 5) to assess the favorability of transformations or to predict degradation pathways (6-8). In recent years, increases in computational speed have enabled thermodynamic predictions based on quantum chemical calculations for molecules or systems comprising up to a few hundred atoms (9). These approaches allow for accurate investigation of chemicals without any previous knowledge on the substance. Furthermore, quantum chemical models can be used to generate kinetic information, such as activation energies and reaction rates (10, 11). This strategy has been widely applied to identify reaction mech- anisms and primary degradation pathways when multiple pathways are possible (12-16). The fate of a contaminant in groundwater, however, will also depend on the redox conditions in an aquifer (17, 18). Consequently, in many in situ remedial approaches, the redox conditions are manipulated by the delivery of strong oxidants or reductants in order to enable or enhance the degradation of redox-sensitive contaminants. For inorganic compounds, readily applicable thermodynamic equilibrium models are already available (ref 19 and references therein). Strategies to assess the redox reactivity of organic contaminants, on the other hand, are very limited. Several quantum chemical methods used for calculation of standard redox potentials have been developed (ref 20 and references therein). One of the most popular approaches is the application of the Born-Haber cycle, in which the free energy of a half reaction is determined from the free energy of reaction in gas phase and solvation free energies of the oxidized and reduced species (21). This approach, however, requires the inclusion of at least two explicit hydration shells to yield accurate predictions, which substantially increases the computational cost (22). Namazian and co-workers (23, 24) calculated standard redox potentials for organic compounds based on the quantum chemically predicted free energy of reaction and a known experimental redox potential for one of the constituent half reactions, thus eliminating the need of explicit solvation by using less costly implicit solvation models. However, no attempts have so far been made to include the influence of pH on redox reactivity in predictive models and use this information to elucidate the conditions necessary for contaminant degradation. In this article, we present a novel quantum chemical approach to predict the fate of aqueous phase contaminants in natural and engineered environments. The calculations are based on density functional theory (DFT) which allows for a practical balance between accuracy and computational efficiency, compared to the accurate but more expensive * Corresponding author phone: (970)491-6235; fax: (970)491-0564; e-mail: thomas.borch@colostate.edu. Corresponding author address: Department of Soil and Crop Sciences, 1170 Campus Delivery, Colorado State University, Fort Collins, Colorado 80523-1170. † Department of Soil and Crop Sciences. ‡ Department of Chemistry. § Department of Atmospheric Science. | Department of Chemical & Biological Engineering. ⊥ Department of Civil & Environmental Engineering. Environ. Sci. Technol. 2010, 44, 5868–5874 5868 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 15, 2010 10.1021/es1006675  2010 American Chemical Society Published on Web 07/07/2010