Trends in R-X Bond Dissociation Energies (R ) Me, Et, i-Pr, t-Bu; X ) H, CH 3 , OCH 3 , OH, F): A Surprising Shortcoming of Density Functional Theory Ekaterina I. Izgorodina, † Michelle L. Coote,* ,† and Leo Radom* ,‡ Research School of Chemistry, Australian National UniVersity, Canberra, ACT 0200, Australia, and School of Chemistry, UniVersity of Sydney, Sydney, NSW 2006, Australia ReceiVed: April 19, 2005; In Final Form: June 21, 2005 The performance of a variety of high-level composite procedures, as well as lower-cost density functional theory (DFT)- and second-order perturbation theory (MP2)-based methods, for the prediction of absolute and relative R-X bond dissociation energies (BDEs) was examined for R ) Me, Et, i-Pr and t-Bu, and X ) H, CH 3 , OCH 3 , OH and F. The methods considered include the high-level G3(MP2)-RAD and G3-RAD procedures, a variety of pure and hybrid DFT methods (B-LYP, B3-LYP, B3-P86, KMLYP, B1B95, MPW1PW91, MPW1B95, BB1K, MPW1K, MPWB1K and BMK), standard restricted (open-shell) MP2 (RMP2), and two recently introduced variants of MP2, namely spin-component-scaled MP2 (SCS-MP2) and scaled-opposite-spin MP2 (SOS-MP2). The high-level composite procedures show very good agreement with experiment and are used to evaluate the performance of the lower-level DFT- and MP2-based procedures. The best DFT methods (KMLYP and particularly BMK) provide very reasonable predictions for the absolute heats of formation and R-X BDEs for the systems studied. However, all of the DFT methods overestimate the stabilizing effect on BDEs in going from R ) Me to R ) t-Bu, leading in some cases to incorrect qualitative behavior. In contrast, the MP2-based methods generally show larger errors (than the best DFT methods) in the absolute heats of formation and BDEs, but better behavior for the relative BDEs, although they do tend to underestimate the stabilizing effect on BDEs in going from R ) Me to R ) t-Bu. The potentially less computationally expensive SOS-MP2 method offers particular promise as a reliable method that might be applicable to larger systems. 1. Introduction The bond dissociation energy (BDE) is an important funda- mental concept in chemistry. It is used as a measure of the strength of a chemical bond and is defined as the enthalpy change for the dissociation reaction: Relative values of BDEs are also extremely important in chemistry. For example, the difference between the R′-X and R-X BDEs is effectively the enthalpy change for the X-transfer reaction: Moreover, when X ) H and R′ ) CH 3 , the enthalpy change for reaction 2 is defined as the radical stabilization energy (RSE) for the radical R • and is often used as a measure of radical stability. The accurate prediction of BDEs and RSEs has numerous applications, including the identification of sites for potential free-radical attack in peptides, the assessment of the effectiveness of antioxidants, and the study of chain-transfer processes (such as long-chain branching) in free-radical po- lymerization. However, for useful practical applications in large polymer-related or biologically related systems to be feasible, it is necessary to identify reliable low-cost methods to calculate these quantities. Density functional theory (DFT) is now widely used as a computational chemistry tool and is found to provide reasonable accuracy at modest computational cost for a wide range of chemical systems. 1 Although there are several types of calcula- tion (notably reaction barrier heights 2,3 and heats of formation 4-6 ) for which popular DFT methods such as B3-LYP are known to show substantial errors in some cases, the calculation of relative BDEs (such as RSEs and the enthalpies of abstraction reactions) is not normally regarded as problematic. For instance, Brinck et al. 7 concluded that, although the absolute BDEs were unreliable, the B3-LYP method was suitable for the prediction of the effect of substituents on the C-H BDEs in substituted methanes, C-O BDEs in peroxyl radicals, and O-H BDEs in hydroperoxides. We have also noted that B3-LYP underesti- mates C-H BDEs but, through a systematic cancellation of errors, generally produces reasonable values of RSEs. 8 In a number of recent assessment studies, Truhlar and co-workers have indicated that methods such as B3-LYP, though inadequate for the prediction of barrier heights, nonetheless perform well for the enthalpies of hydrogen-abstraction reactions and also for bond dissociation energies. 2 We have also shown that the B3-LYP method provides reasonable performance for the enthalpies of hydrogen-atom-abstraction reactions involving substituted carbon-centered radicals, and the associated C-H BDEs of the closed-shell reactants and RSEs of the open-shell reactants. 9 Finally, Chen and Bozzelli 10 have noted that the B3- LYP method models very well the relative heats of formation * Corresponding authors. E-mail: (M.L.C.) mcoote@rsc.anu.edu.au; (L.R.) radom@chem.usyd.edu.au. † Australian National University. ‡ University of Sydney. R-X f R • + • X (1) R′-X + R • f R-X + R′ • (2) 7558 J. Phys. Chem. A 2005, 109, 7558-7566 10.1021/jp052021r CCC: $30.25 © 2005 American Chemical Society Published on Web 08/04/2005