Published: October 03, 2011 r2011 American Chemical Society 3725 dx.doi.org/10.1021/ct200418e | J. Chem. Theory Comput. 2011, 7, 3725–3732 ARTICLE pubs.acs.org/JCTC Insights into the Solvation and Mobility of the Hydroxyl Radical in Aqueous Solution Edelsys Codorniu-Hernandez and Peter G. Kusalik* Department of Chemistry, University of Calgary, 2500 University Drive NW, Calgary, T2N1N4, Alberta, Canada ABSTRACT: A detailed description of the local solvation structure and mobility of hydroxyl radicals (OH*) in aqueous solution near ambient conditions is provided by Car Parrinello molecular dynamics simulations. Here, we demonstrate that for HCTH/120 and BLYP functionals, smaller systems (i.e., 31 3 H 2 O OH*) are contaminated by system size effects, being biased for the presence of a three-electron two-centered hemibond structure between the oxygen atoms of a water molecule and the radical. Radial and spatial distribution functions of relatively large 63 3 H 2 O OH* systems reveal the existence of a 4-fold coordinated “inactive” OH* structure with three H-bond donating neighbors and a strongly coordinated H-bond accepting neighbor. The local hydration structure around the radical exhibits more H-bond ordering than has been predicted by recent simulations employing classical force fields. Local structural fluctuations can end with spontaneous H-transfer reactions from the nearest H-bond donor water molecule, facilitated by the formation of an “active” OH* state, resembling the proton transfer mechanism of hydrated OH (i.e., slight polarization of the (H 3 O 2 )* complex). A comparison of the free energy barriers for the H-transfer reaction obtained by both DFT functionals and for both system sizes is also provided, demonstrating that this can be a very rapid process in water. 1. INTRODUCTION The hydroxyl radical (OH*) has posed significant challenges to theoretical and experimental studies due to its high reactivity and very short lifetime. 1 However, it is still the target molecule of numerous investigations owing, in part, to its biological and atmospheric significance as well as its crucial role in industrial applications. 2,3 Described as the atmospheric “vacuum cleaner”, this radical is responsible for many of the reactions that remove volatile organic compounds from the air. 4,5 It oxidizes approxi- mately 83% of annual methane emissions, making OH* the most important processor of greenhouse gases. 4,5 Serious ailments such as cancer and Parkinson’s disease have also been related to OH*, 6 and it is recognized as the most reactive of the so-called reactive oxygen species (ROS). The only means to protect impor- tant cellular structures from its action is the use of antioxidants because, contrary to other oxidants, OH* cannot be eliminated by an enzymatic reaction. This makes it a very dangerous compound to an organism, but interestingly, OH* is also essential to the body’s natural defense mechanisms. 6 Water apparently has a crucial role in OH* chemistry. Recent studies have pointed out its effect on modulating OH* reactivity, providing a clear stabilization of transition states and higher reactivity via hydrogen bonding. 7,8 High level ab initio calcula- tions of gas phase OH*(H 2 O) n clusters 9 28 have also been conducted. Their focus has been on the possible existence of a H 2 O OH* complex which is speculated to influence strongly the diffusion and oxidative capacity of the radical. Only a few studies have considered the solvation of OH* in liquid water, 29 34 aimed at demonstrating an expected OH* ability to diffuse in water via hydrogen exchange analogous to the proton-exchange reaction in the case of OH . 30,35 Prior to our work, 36 this key reaction had not been directly detected by either experimental or theoretical studies due to the large challenges posed by OH* to both fields. In addition, a recent spectroscopic observation 37 made during the irradiation of OH in aqueous solutions has been attributed to ultrafast H-transfer reactions from neighbor- ing water molecules to OH*. These authors 37 had attempted to compare their proposed mechanism with previous Car Parrinello MD simulations 31 in which a three-electron two-center hemi- bond structure between the oxygen atom of the radical and the oxygen atom of one water molecule was found to be a particularly stable structure. 30,31 In the presence of the hemibond, the supposed diffusion mechanism of OH* in liquid water via a hydrogen exchange reaction is effectively impeded. 30 Afterward, Vandevondele et al. 33 claimed that BLYP and all GGA func- tionals overestimate the hemibond structure and, using self- interaction corrected methods, reported that OH* acts as a good hydrogen bond donor but accepts fewer than two hydrogen bonds on average. However, the previous apparent inability of Car Parrinello MD to determine a OH* H-transfer reaction 29 34 seems inconsistent with the low reaction barrier (around 4.2 kcal/mol) 27,38 predicted for the hydrogen transfer reaction in the gas phase, which is in good agreement with the available experimentally derived data. 11 As we have shown in a very recent paper, 36 Car Parrinello molecular dynamics simulations of a larger (63 3 H 2 O OH*) system provide a different picture with respect to previous simulations using the smaller system (31 3 H 2 O OH*). 29 34 Here, we present a comparison between these two systems, providing a detailed description of the solvation and electronic behavior of OH* in aqueous solution and demonstrating that smaller systems are contaminated by system size effects. Both the HCTH/120 and BLYP density functionals are employed. Im- portant insights into the features of the H-transfer reaction (OH* + H 2 OT H 2 O + OH*) are also provided. Received: June 18, 2011