The H + OCS potential-energy surface (PES) was used to evaluate the performance of density functional theory by comparing the results to ab initio calculations at the QCISD(T)//UMP2 and UMP2 levels using the aug-cc-pVTZ and 6-311+G(2df, 2p) basis sets. The two major reaction paths on this PES involve formation of OH(2Π) + CS(1Σ) (reaction I) and SH(2Π) + CO(1Σ) (reaction II). Experimental and QCISD(T)//UMP2/aug-cc-pVTZ activation barriers for (II) and reaction enthalpies for (I) and (II) were compared to values calculated by several density functionals (BLYP, B3LYP, B3PW91, BPW91, BP86, and B3P86) using the aug-cc-pVTZ basis set. All DFT/aug-cc-pVTZ predictions, except for the B3LYP prediction of the enthalpy of reaction I, were outside the range of experimental uncertainty. B3LYP predictions were in closest agreement with the experimental values and QCISD(T) predictions. B3LYP, BPW91, and B3PW91 predictions of the rate-limiting barrier to reaction II are within 3.5 kcal/mol of the QCISD(T) prediction, and all DFT values are below that of the QCISD(T). Reaction enthalpies for (I) and (II) were calculated using the BHandHLYP density functional and the 6-311+G(2df,2p) basis set. These predictions were closer to experiment and QCISD(T) values than any other DFT calculations, and the predicted enthalpy for reaction I is within the range of experimental values. The second portion of the study compared B3LYP and BLYP predictions of the 12 transition states and 6 stable intermediates within this PES with previously reported QCISD(T)//UMP2/6-311+G(2df,2p) predictions. The complexity of this surface allows for the evaluation of barrier heights for 28 reactions involving hydrogen addition, elimination, isomerization, migration, and radical diatomic elimination. With the exception of five reactions, all B3LYP barrier heights are within 3.7 kcal/mol of the QCISD(T) predictions and in several cases are in as good or better agreement than the UMP2 predictions. In addition, all but one of the B3LYP barriers lie below the QCISD(T) values. The most significant differences between the ab initio and DFT predictions were in the saddle points for radical elimination or addition. BLYP/6-311+G(2df, 2p) failed to find the two transition states associated with SH elimination from the cis- and trans-HSCO species. B3LYP located the saddle point for SH elimination from cis-HSCO, but its prediction of a saddle-point structure for SH elimination from trans-HSCO has an energy (without zero-point corrections) lower than that of the products. These transition states were subsequently optimized using the BHandHLYP functional and the 6-311+G(2df,2p) and 6-31G** basis sets. The geometries of these saddle points were in better agreement with UMP2/6-311+G(2df,2p) predictions than were the BLYP and B3LYP predictions. The BLYP predictions are in overall worse agreement with the QCISD(T) results than are the B3LYP predictions.

Ab initio calculations were performed to investigate reaction mechanisms for formation and decomposition of the six-membered ring C3N3H3, known as sym-triazine. MP2 geometry optimizations with QCISD(T) energy refinements for critical points on the potential energy surface were calculated with the 6-31G**, 6-311++G** and cc-pVTZ basis sets. Good agreement is found for MP2 geometries and frequencies of sym-triazine and HCN when compared with the corresponding experimental values. Two decomposition mechanisms of sym-triazine, the concerted triple dissociation (sym-triazine → 3 HCN) and the stepwise decomposition (sym-triazine → H2C2N2 + HCN → 3 HCN) were investigated. All calculations show that the lowest energy decomposition mechanism is the concerted triple dissociation. Our best calculations predict the zero-point-energy-corrected barrier to decomposition to be 81.2 kcal/mol. The calculated reaction enthalpy is 35.5 kcal/mol, 7.7 kcal/mol lower than experiment. Intrinsic reaction coordinate calculations leading from the transition state of the concerted triple dissociation reaction to three HCN molecules led to a minimum on the potential energy surface. The corresponding structure is a cyclic (HCN)3 cluster. The temperature-corrected formation enthalpy of the cluster is −8.7 kcal/mol relative to three isolated HCN molecules. The zero-point-corrected barrier to formation of sym-triazine from the cluster is 58.1 kcal/mol. QCISD(T) energy refinements did not differ significantly from the MP2 results.

Stable points and transition states on the potential energy surface (PES) for sym-triazine (C3N3H3) have been calculated by using nonlocal density functional (NDFT) methods. Two decomposition mechanisms for sym-triazine are investigated. The first is a concerted triple dissociation of the sym-triazine ring to form the HCN products. Three-fold symmetry is maintained along the reaction path for this mechanism. The second is a stepwise decomposition mechanism involving the formation of an intermediate dimer species. The NDFT results, including structures, relative energies, harmonic vibrational frequencies, and corresponding eigenvectors, are compared with previously reported ab initio calculations. These results include critical points located and characterized through normal mode analyses at the MP2 level. QCISD(T) energy refinements of the MP2 critical points are used for the comparison of DFT predictions. Basis set size dependence is also examined. The nonlocal density functionals used are the exchange functional of Becke and the correlation energy functional of Perdew (BP86), Becke's exchange and the correlation energy functional of Lee, Yang, and Parr (BLYP), Becke's three-parameter hybrid exchange functional with the LYP correltation energy functional (B3LYP), and the Becke exchange with Perdew and Wang's 1991 gradient-corrected correlation functional (BPW91). Basis sets used are 6-31G**, 6-311++G**, and cc-pVTZ. The reaction endothermicity predicted by B3LYP and BPW91 are in closer agreement with experiment than the QCISD(T) and MP2 predictions using the largest basis set. B3LYP predictions are within 1.1 kcal/mol of experiment. BPW91, BP86, and BLYP frequencies agree most closely with experimental values for sym-triazine and HCN. DFT eigenvectors corresponding to vibrational modes for critical points on the PES compare well with MP2 predictions for most modes, indicating similarity in force fields and, therefore, atomic motion for the vibrations. Geometries predicted by all methods are in excellent agreement with experimental values for sym-triazine and HCN. All methods predict that the concerted triple-dissociation mechanism is the low-energy decomposition pathway for sym-triazine. DFT predictions of energies along the reaction path for the concerted triple-dissociation reaction are in qualitative agreement with MP2. All DFT methods predict structures of species along the reaction path that are in quantitative agreement with MP2 predictions.

Density functional calculations show that H2 desorption from Si(100)−2×1 via a ‘‘prepaired’’ state is consistent with energetic and dynamic measurements. The corresponding adsorption process is discussed and comparisons are made to earlier theoretical studies.

To establish guidelines for using small silicon-containing molecules to model silicon surfaces, a series of substituted silanes have been studied using post-HartreeFock and density functional theories. The two theories differ in their detailed predictions, but similar trends are found with both methods. S i S i and Si-H bond lengths vary slightly among the compounds studied. Mulliken charges on the substituted silicon atom are altered significantly, but Mulliken charges on the other atoms remain unaffected. The S i S i bond energy decreases by about 0.5 kcal/mol with each silyl group that replaces hydrogen. Force constants change by a few percent. The most dramatic effect of substitution is that the energy of Si-H bonds at the substituted Si decrease by about 3 kcal/mol with each successive replacement of hydrogen by silyl groups. Of the properties calculated, only the S i S i bond strength correlates with substituent electronegativity. The effects of model structure on surface ionization potentials have also been determined for comparison to earlier work. In general, substitution a t a surface site alters some model properties, but more distant substitutions have little effect.