Zero-point Energy Corrections Research Articles

A first principles study of the acetylene–water interaction

We present an extensive study of the stationary points on the acetylene–water (AW) ground-state potential energy surface (PES) aimed in establishing accurate energetics for the two different bonding scenarios that are considered. Those include arrangements in which water acts either as a proton acceptor from one of the acetylene hydrogen atoms or a proton donor to the triple bond. We used a hierarchy of theoretical methods to account for electron correlation [MP2 (second-order Moller–Plesset), MP4 (fourth-order Moller–Plesset), and CCSD(T) (coupled-cluster single double triple)] coupled with a series of increasing size augmented correlation consistent basis sets (aug-cc-pVnZ, n=2,3,4). We furthermore examined the effect of corrections due to basis set superposition error (BSSE). We found that those have a large effect in altering the qualitative features of the PES of the complex. They are responsible for producing a structure of higher (C2v) symmetry for the global minimum. Zero-point energy (ZPE) corrections were found to increase the stability of the C2v arrangement. For the global (water acceptor) minimum of C2v symmetry our best estimates are ΔEe=−2.87 kcal/mol (ΔE0=−2.04 kcal/mol) and a van der Waals distance of Re=2.190 Å. The water donor arrangement lies 0.3 kcal/mol (0.5 kcal/mol including ZPE corrections) above the global minimum. The barrier for its isomerization to the global minimum is Ee=0.18 kcal/mol; however, inclusion of BSSE- and ZPE-corrections destabilize the water donor arrangement suggesting that it can readily convert to the global minimum. We therefore conclude that there exists only one minimum on the PES in accordance with previous experimental observations. To this end, vibrational averaging and to a lesser extend proper description of intermolecular interactions (BSSE) were found to have a large effect in altering the qualitative features of the ground-state PES of the acetylene–water complex.

Inversion of stationary point levels in a flat transition structure region on account of vibrational energy

Characterization of the stationary points in a flat transition structure region of the potential energy surface for the HCOCl+Cl − reaction has been carried out at the MP2/6-311+G∗∗ level. Two stationary points are found: an intermediate with C S symmetry and all positive eigenvalues in the Hessian and a transition state with C 1 symmetry and only one negative eigenvalue in the Hessian. The MP2/6-311+G∗∗ electronic energies of these two only differ by δΔ E(=Δ E TS−Δ E Int)=0.01 kcal mol −1. The zero-point and thermal energy corrections to the electronic energy lead to an inversion of levels: δΔ E ZPE=−0.06 and δΔ H=−0.59 kcal mol −1, which becomes normal when the − TΔ S term is added to obtain δΔ G=+0.90 kcal mol −1 (+0.80 kcal mol −1 at the G2(+)MP2 level). This inversion arises solely from one less vibrational mode in the TS on the flat transition structure region.

The silaketenylidene (SiCO) molecule: Characterization of the X̃ 3Σ− and à 3Π states

The ground (X̃ 3Σ−) and first excited triplet (à 3Π) electronic states of carbonylsilene or silaketenylidene, SiCO, have been investigated systematically using ab initio electronic structure theory. The total energies and physical properties including equilibrium geometries, dipole moments, harmonic vibrational frequencies, and associated infrared (IR) intensities were predicted using self-consistent-field (SCF), configuration interaction with single and double excitations (CISD), coupled cluster with single and double excitations (CCSD), equation-of-motion (EOM) CCSD, CCSD with perturbative triple excitations [CCSD(T)] methods with a wide range of basis sets. The linear X̃ 3Σ− ground state of SiCO has a real degenerate bending vibrational frequency, whereas the à 3Π state of SiCO is subject to the Renner–Teller effect and presents two distinct real vibrational frequencies along the bending coordinate. The bending vibrational frequency of the à 3Π state was evaluated via the EOM-CCSD technique. At the highest level of theory with the largest basis set, cc-pVQZ CCSD(T), the adiabatic X̃–à splitting without the zero-point vibrational energy (ZPVE) correction (Te value) was determined to be 68.5 kcal/mol (2.97 eV, 23 900 cm−1) and the adiabatic splitting with the ZPVE energy correction (T0 value) to be 69.0 kcal/mol (2.99 eV, 24 100 cm−1), which are in excellent agreement with the experimental T0 value of 68.78 kcal/mol (2.983 eV, 24 056 cm−1). The theoretical ground state harmonic Si–C stretching frequency ω3=564 cm−1 is much less than the experimental estimate of 800 cm−1.

Conformational analysis. Part 33.1 An NMR, solvation and theoretical investigation of conformational isomerism in N,N-dimethylfluoroacetamide and N,N-dimethyl-α-fluoropropionamide

The solvent and temperature dependence of the 1H and 13C NMR spectra of N,N-dimethylfluoroacetamide (DMFA) and N,N-dimethyl-α-fluoropropionamide (DMFP) are reported and the 5JCF, 1JCF and 4JCF couplings analysed by solvation theory. Density function theory (DFT) at the B3LYP/6-311+G(d,p) level with ZPE (zero point energy) corrections was used to obtain the conformer geometries. In DMFA, the DFT method gave only two minima for the cis (F–C–CO, 0°) and gauche (F–C–CO, 140.6°) rotamers. The trans rotamer was not a minimum in the energy surface. Assuming only the cis and gauche forms, the observed couplings when analysed by solvation theory gave the energy difference (Ecis − Eg) of 2.5 kcal mol−1 in the vapour phase, (cf. the ab initio value of 2.3 kcal mol−1) decreasing to 0.87 kcal mol−1 in CCl4 and to −1.29 kcal mol−1 in DMSO. In DMFP the ab initio calculations gave three minima; the cis (F–C–CO, 30.4°), gauche-1 (F–C–CO, 144.7°) and gauche-2 (F–C–CO, −124.1°) rotamers with (Ecis − Eg2) equal to 2.5 kcal mol−1 and (Eg1 − Eg2) equal to 0.3 kcal mol−1. The observed couplings were analysed by solvation theory assuming one “average” gauche conformer to give (Ecis − Eg(AV)) equal to 2.1 kcal mol−1 in the vapour phase, decreasing to 0.83 kcal mol−1 in CCl4 and to −1.11 kcal mol−1 in DMSO.

An ab initio study of water clusters in gas phase and bulk aqueous media: (H2O)n,n=1-12

Geometries of several clusters of water molecules including single minimum energy structures of n-mers (n=1–5), several hexamers and two structures of each of heptamer to decamer derived from hexamer cage and hexamer prism were optimized. One structural form of each of 11-mer and 12-mer were also studied. The geometry optimization calculations were performed at the RHF/6-311G* level for all the cases and at the MP2/6-311++G** level for some selected cases. The optimized cluster geometries were used to calculate total energies of the clusters in gas phase employing the B3LYP density functional method and the 6-311G* basis set. Frequency analysis was carried out in all the cases to ensure that the optimized geometries corresponded to total energy minima. Zero-point and thermal free energy corrections were applied for comparison of energies of certain hexamers. The optimized cluster geometries were used to solvate the clusters in bulk water using the polarized continuum model (PCM) of the self-consistent reaction field (SCRF) theory, the 6-311G* basis set, and the B3LYP density functional method. For the cases for which MP2/6-311++G** geometry optimization was performed, solvation calculations in water were also carried out using the B3LYP density functional method, the 6-311++G** basis set, and the PCM model of SCRF theory, besides the corresponding gas-phase calculations. It is found that the cage form of water hexamer cluster is most stable in gas phase among the different hexamers, which is in agreement with the earlier theoretical and experimental results. Further, use of a newly defined relative population index (RPI) in terms of successive total energy differences per water molecule for different cluster sizes suggests that stabilities of trimers, hexamers, and nonamers in gas phase and those of hexamers and nonamers in bulk water would be favored while those of pentamer and decamer in both the phases would be relatively disfavored. © 2001 John Wiley & Sons, Inc. Int J Quant Chem 81: 90–104, 2001

Ab initio study of the formation of B3H7 derivative from B3H8− anion protonation

Formation of B3H7 species from the protonation of B3H8− has been studied by ab initio calculations at the HF and correlated MP2 levels using extended basis set 6-31+G∗∗. Two elementary processes, B3H8−+H+→B3H9 and B3H9→B3H7+H2, have been investigated. The calculated absolute values of B3H8−+H+→B3H9 protonation energy are very high (above 314kcalmol−1). Loss of molecular hydrogen from B3H9 proceeds with an activation barrier above 12kcalmol−1, leading to a B3H7 one-bridged structure. The activation barrier of reverse reaction is found non-negligible when the zero-point energy correction is added. The calculated energy and activation barriers of B3H7 rearrangement are small, confirming the fluxional behaviour for this species. However, a mechanism of B3H7 formation is proposed. The inclusion of a solvent reaction field using water as a solvent (ε=78.54) has a small influence on the energies and molecular properties.

Reaction dynamics of chlorine atom with methane: Dual-level ab initio analytic potential energy surface and isotope effects

An analytic potential energy surface for the Cl+CH4⇌HCl+CH3 reaction in C3V symmetry has been obtained by fitting to 1136 energy points from a dual-level MP2/SAC (Mo/ller–Plesset second order perturbation/scaling all correlation) calculation using the 6-311G(2d,d,p) basis set. A zero-point energy correction is made to account for all modes not explicitly treated with the time-independent quantum scattering rotating line umbrella (RLU) model, which is used for the dynamics calculations. The effective potential gives a vibrationally adiabatic ground-state barrier height of 3.36 kcal/mol and an endothermicity (0 K) of 1.19 kcal/mol for the Cl+CH4 reaction, and 4.43 kcal/mol and 2.29 kcal/mol, respectively, for Cl+CD4. Thermal rate constants, tunneling and kinetic isotope effects have been investigated in detail. Calculated differential cross sections for Cl+CD4→DCl+CD3, with reactants and products in their vibrational ground states, show that the DCl product is strongly backward scattered. Further, ground state Cl+CD4 reacts to give the CD3 product predominantly unexcited at collision energies ranging from 0.15 eV to 0.25 eV. Generally, good agreement with experimental measurements and previous theoretical work is obtained.

Kinetic Monte Carlo study of the effect of hydrogen on the two-dimensional epitaxial growth of Ni(100)

The effects of hydrogen on Ni~100! submonolayer homoepitaxy have been investigated by kinetic Monto Carlo ~KMC! simulations of the Ni growth process with and without the presence of a small hydrogen impurity. The two-dimensional KMC simulations are based upon rate constants for a set of uncorrelated Ni site-to-site hopping mobilities with and without the presence of hydrogen atoms which are found to act as a catalyst. The rate constant activation energies are determined by classical-potential total-energy calculations with semiclassical zero-point energy corrections for the hydrogen atom. This set of KMC simulations extends preliminary work on the activation energies which were reported along with their connection to available experiments in Ref. 1. We find that fast diffusion of H atoms occurs on the flat Ni~100! surface and the presence of these highly mobile H atoms is found to have significant effects on the mobility of lone Ni adatoms, and therefore also on the Ni islands which form during the epitaxial growth. The H atoms increase the mobility of lone Ni adatoms across the flat Ni~100! terrace by stabilizing the transition state for site-to-site hopping of the Ni adatom. While the effects of the H atom on the Ni mobility are primarily catalytic, the kinetically determined Ni island morphologies differ substantially as a function of H atom concentration over time periods which are long on the deposition time scale, and therefore the morphology differences can become frozen in place. Even quite low H atom coverage, by acting to increase lone Ni atom mobility, tends to decrease Ni island density which results in a corresponding increase in the average Ni island size. This suggests that the H atoms are acting as an antisurfactant in the Ni~100! homoepitaxy at low H coverage. @S0163-1829~99!06339-0#

Complexes and saddle point structures, vibrational frequencies and relative energies of intermediates for CH2Br + HBr «-» CH3Br + Br

Ab initio molecular orbital theory was used to study parts of the reaction between the CH2Br radical and the HBr molecule, and two possibilities were analysed: attack on the hydrogen and attack on the bromine of the HBr molecule. Optimized geometries and harmonic vibrational frequencies were calculated at the second-order Moller-Plesset perturbation theory levels, and comparison with available experimental data was favourable. Then single-point calculations were performed at several higher levels of calculation. In the attack on the hydrogen of HBr, two stationary points were located on the direct hydrogen abstraction reaction path: a very weak hydrogen bonded complex of reactants, C···HBr, close to the reactants, followed by the saddle point (SP). The effects of level of calculation (method + basis set), spin projection, zeropoint energy, thermal corrections (298K), spin-orbit coupling and basis set superposition error (BSSE) on the energy changes were analysed. Taking the reaction enthalpy (298K) as reference, agreement with experiment was obtained only when high correlation energy and large basis sets were used. It was concluded that at room temperature (i.e., with zero-point energy and thermal corrections), when the BSSE was included, the complex disappears and the activation enthalpy (298K) ranges from 0.8kcal mol-1 to 1.4kcal mol-1 above the reactants, depending on the level of calculation. It was concluded also that this result is the balance of a complicated interplay of many factors, which are affected by uncertainties in the theoretical calculations. Finally, another possible complex (X complex), which involves the alkyl radical being attracted to the halogen end of HBr (C···BrH), was explored also. It was concluded that this X complex does not exist at room temperature.

RISM-SCF study of the free-energy profile of the Menshutkin-type reaction NH 3 +CH 3 Cl→NH 3 CH 3 + +Cl − in aqueous solution

The free-energy profile for the Menshutkin-type reaction NH3 + CH3Cl → NH3CH3 + + Cl− in aqueous solution is studied using the RISM-SCF method. The effect of electron correlation on the free-energy profile is estimated by the RISM-MP2 method at the HF optimized geometries along the reaction coordinate. Solvation was found to have a large influence on the vibrational frequencies at the reactant, transition state and product; these vibrational frequencies are utilized to calculate the zero-point energy correction of the free-energy profile. The computed barrier height and reaction exothermicity are in reasonable agreement with those of experiment and previous calculations. The change of solvation structure along the reaction path is represented by radial distribution functions between solute-solvent atomic sites. The mechanisms of the reaction are discussed from the view points of solute electronic and solvation structures.

Influence of the Methyl Groups on the Structure, Charge Distribution, and Proton Affinity of the Retinal Schiff Base

The effect of the methyl substitutions at different carbon atoms of the polyene chain of the retinal Schiff base has been studied on the structure, charge distribution, and proton affinity (PA) in a Schiff base model with the same number of conjugated double bonds (six conjugated double bonds including the Schiff base group). The methyl groups were added to the nitrogen atom and/or different carbon atoms along the main chain, and the results were compared to those of the retinal Schiff base. To study the effect of the polarization functions on hydrogen atoms, all of the structures were studied using 6-31G* and 6-31G** basis sets at the Becke3LYP (B3LYP) level of theory. For each optimized model, the PA was then calculated on the basis of the difference of the total energy of the protonated and unprotonated species. The influence of the zero-point energy corrections on the PA values was also examined at the same level of theory. The results show that, for all species, the application of the larger basis se...

A Theoretical Study of the Reaction of ClONO2 with HCl on Ice

The direct ClONO2 + HCl → Cl2 + HNO3 reaction on ice, implicated in polar stratospheric ozone depletion, is studied quantum chemically on a model ice lattice comprising nine water molecules. The reaction path is calculated at the HF/(HW*,3-21G) level, using the Hay−Wadt effective core potential for Cl. At these geometries, energies are recalculated at the MP2/(SBK+*,6-31+G*) level, with the Stevens−Bash−Krauss effective core potential for Cl. HCl is found to be ionized in the reactant complex. The calculated reaction internal energy barrier, including zero-point energy correction, is 6.4 kcal/mol. The reaction mechanism involves proton transfer in the ice lattice, accompanied by nucleophilic attack of Cl- on the Clδ+ in ClONO2; the lattice is an active participant in the reaction. Implications for heterogeneous atmospheric chemistry are discussed.

Theoretical analysis of the internal rotation in aminoborane and borylphosphine

Using a recently proposed orbital deletion procedure and the block-localized wavefunction method, the rotational barriers in H2BNH2 and H2BPH2 are analyzed in terms of conjugation, hyperconjugation, steric effect and pyramidalization. With the zero-point energy corrections, the π-binding strengths in the planar H2BNH2 and H2BPH2 are both around 20 kcal/mol at the HF level using the 6-311+G** basis set. With the deactivation of the π atomic orbitals on the boron atom and the evolution from a planar structure to a 90°-twisted structure, the steric repulsion between the B‐H and the N‐H or P‐H is relieved and moreover, the negative hyperconjugation from the lone electron pair or pairs on the nitrogen or phosphorus atoms to the antibonding orbital χ* B H2 of the BH2 group stabilizes the twisted structure by 7.4(8.8) or 4.0(5.0) kcal/mol at the HF/6-31G*(6-311+G**) level. However, the repulsive interaction between the lone pair(s) and the two BH σ bonds is so prominent that the overall steric effect contributes 20.3(22.9) and 19.3(19.8) kcal/mol to the rotational barriers in H2BNH2 and H2BPH2 with the 6-31G*(6-311+G**) basis set. The present techniques and analyses may also give some clues to justify the parameterization in the empirical molecular mechanics methods.

Conformational analysis, Part 32.† NMR, solvation and theoretical investigation of conformational isomerism in 3-fluorobutan-2-one and 3,3-difluorobutan-2-one

The solvent and temperature dependence of the 1H and 13C NMR spectra of 3-fluorobutan-2-one (FB) and 3,3-difluorobutan-2-one (DFB) are reported and the 4JHF, 1JCF and 2JCF couplings analysed using ab initio calculations and solvation theory. The solvent dependence of the IR spectra (carbonyl band) was also measured. In FB, ab initio theory at the 6-31G**/MP2 level gives only two energy minima for the cis (F–C–CO 22°) and trans (F–C–CO 178°) rotamers. The gauche rotamer was not a minimum in the energy surface. Assuming only the cis and trans forms, the observed couplings when analysed by solvation theory lead to the energy difference (Ecis – Etrans) between the cis and trans rotamers of 3.7 kcal mol–1 in the vapour phase, decreasing to 2.5 kcal mol–1 in CCl4 and to 0.1 kcal mol–1 in DMSO. In all solvents used the trans rotamer is more stable than the cis. The vapour state energy difference compares very well with that calculated [3.67 kcal mol–1 including a zero-point energy correction (ZPE)]. In DFB ab initio calculations at this level and also at (6-311G**/MP2 and ZPE) gave only one minimum in the potential energy surface corresponding to the cis rotamer (C–C–CO 0°). The 1H and 13C NMR data, 4JHF, 1JCF and 2JCF couplings do not change with solvent confirming that there is only one rotamer in solution for DFB, in agreement with the ab initio calculations.

Ab initio/RRKM approach toward the understanding of ethylene photodissociation

The optimized structures and harmonic frequencies for the transition states and intermediates on the ground state potential energy surfaces of ethylenes, including C2H4, C2D4, D2CCH2, and cis- and trans-HDCCDH, related to the molecular and atomic hydrogen elimination channels of photodissociation in VUV were characterized at the B3LYP/6-311G(d,p) level. The coupled cluster method, CCSD(T)/6-311+G(3df,2p), was employed to calculate the corresponding energies with the zero-point energy corrections by the B3LYP/6-311G(d,p) approach. Ethylidene was found to be an intermediate in the 1,2-H2 elimination channel. The barrier for the 1,1-H2 elimination was computed to be the lowest (4.10–4.16 eV), while the 1,2-H2 elimination and H loss channels have barriers of a similar height (4.70–4.80 eV). The rate constant for each elementary step of ethylene photodissociation at 193 and 157 nm was calculated according to the RRKM theory based on the ab initio surfaces. The rate equations were subsequently solved, and thus the concentration of each species was obtained as a function of time. The concentrations at t→∞ were taken for calculating branching ratios or yields. In accord with previous experimental findings, the calculated branching ratio for the 1,1-H2 elimination process is higher than that for the 1,2-H2 elimination, and the atomic elimination channel is predicted to be favored at increasing excitation energy when competing with the molecular elimination. The significant discrepancies between theoretical and experimental results in the magnitude of the yields and their dependence on the wavelength for the molecular elimination channels suggest the dynamics of either 1,2-H2, or 1,1-H2 elimination, or both channels may be nonstatistical in nature.

Ab initio MO study on the global minimum structure of C 3H −3 anion: propynl-1-yl versus allenyl anions

The geometrical structures of the C 3H − 3 anion are surveyed at the coupled-cluster doubles (CCD) level of theory with the aug-cc-pVDZ basis set. To clarify the CCD geometries, the stable two isomers — propynl-1-yl 1 and allenyl 2 anions — are further optimized at the coupled-cluster singles, doubles (triples) (CCSD(T)) level of theory both with the aug-cc-pVDZ and aug-cc-pVTZ basis sets. The final energies are calculated at the CCSD(T) and the complete active space self-consistent field (CASSCF) multi-reference internally contracted CI (MRCI) levels of theory with the aug-cc-pVTZ basis set. At the MRCI level of theory including both the corrections due to the cluster energies (MRCI+Q) and the zero-point vibrational energies, the allenyl anion 2 is about 1.3 kcal mol −1 lower in energy than the propynl-1-yl anion 1. These results contrast with the previous theoretical estimates, where the propynl-1-yl anion 1 is 2–3 kcal mol −1 lower in energy than the allenyl anion 2. The activation energies of the intramolecular hydrogen transfer in the 1→ 2 conversion reactions are 63.5 kcal mol −1 at the MRCI+Q level of theory with the aug-cc-pVTZ basis set including the zero-point energy corrections. The adiabatic electron affinity of the planer propargyl (H 2CCCH) radical, which is the global minimum of the C 3H 3 radical, is calculated to be 0.976 eV (after correction for the zero-point energy changes) at the CCSD(T) level of theory with the aug-cc-pVTZ basis set. The present electron affinity is in fairly good agreement with the experimental one (0.893 eV) observed by Oakes and Ellison.

NMR versus molecular modelling: menthone and isomenthone

ABSTRACT: trans-2-Isopropyl-5-methylcyclohexanone (menthone) may be considered as locked in the diequatorial conformer. However, the cis isomer (isomenthone) may exist in two major conformations with the isopropyl group either equatorial or axial. No measure of the equilibrium mix has been determined previously. Proton NMR allows such a determination at room temperature via the vicinal couplings between protons on carbons 5 and 6. Combined with modelling results, the calculated value with the isopropyl axial was 79%, of the total in good agreement with an earlier prediction. Two different conformer searching programs were employed to determine the various conformer populations. It was found that molecular mechanics estimates of the Boltzmann distributions varied considerably from the experimental results. These differences were not due to substituent electronegativity effects or to hydrogen bonding with the deuterochloroform solvent. Semiempirical PM3 and low-level ab initio calculations of the conformer energies also failed to account adequately for the conformer populations. DFT calculations (Becke3LYP/6–31G*) with zero-point energy corrections gave results most consistent with the experimentally determined values (67.5% axial isopropyl conformer). © 1988 John Wiley & Sons, Ltd.

Calculation of electronic affinity and vertical detachment energy of the water dimer complex using the density functional theory

Though the electron attachment on the water dimer has been observed, ab initio calculations performed at the Hartree-Fock and post Hartree-Fock levels do not succeed in predicting a bounded (H2O)2− anion. It is shown that the hybrid density functional approach (B3LYP) yields results in reasonable agreement with experiment, provided a systematic optimization of the basis sets in the variational sense. Neglecting the zero point energy (ZPE) correction, the adiabatic electron affinity (EA) is calculated to be 21.2 meV (30±2 experimentally) whereas the vertical detachment energy (VDE) is overestimated by about 10 meV (55.7 against 45). The ZPE correction has been estimated from the frequencies calculated at the harmonic level. It improves noticeably the EA and the VDE which corrected values are 35.6 and 41.9 meV, respectively. The sign and magnitude of the EA and VDE isotopic shifts between the hydrogenated and deuterated species are correctly predicted. The analysis of the electron density difference and of the spin density shows that the electron attachment occurs for one half between the two H2O molecules, the remaining being located outside in the dipole moment direction.

Ab initio and density function theory computational studies of the CH 4 + H → CH 3 + H 2 reaction

The activation barrier for the CH 4 + H → CH 3 + H 2 reaction was evaluated with traditional ab initio and Density Functional Theory (DFT) methods. None of the applied ab initio and DFT methods was able to reproduce the experimental activation barrier of 11.0–12.0 kcal/mol. All ab initio methods (HF, MP2, MP3, MP4, QCISD, QCISD(T), G1, G2, and G2MP2) overestimated the activation energy. The best results were obtained with the G2 and G2MP2 ab initio computational approaches. The zero-point corrected energy was 14.4 kcal mol −1. Some of the exchange DFT methods (HFB) computed energies which were similar to the highly accurate ab initio methods, while the B3LYP hybrid DFT methods underestimated the activation barrier by ~3 kcal mol −1. Gradient-corrected DFT methods underestimated the barrier even more. The gradient-corrected DFT method that incorporated the PW91 correlational functional even generated a negative reaction barrier. The suitability of some computational methods for accurately predicting the potential energy surface for this hydrogen radical abstraction reaction was discussed.

Critical test of PM3-calculated proton transfer activation energies: a comparison with ab initio and AM1 calculations

Intra- and intermolecular proton transfer activation energies are calculated by means of the semiempirical PM3 method and compared with ab initio results obtained at MP2/6–31G ∗ or MP2/6–31G ∗∗ level. It is shown that corrections for zero-point vibrational energy (ZPVE) and thermal contributions do not affect the correlation between semiempirical and ab initio results. The linear relationship obtained suggests that PM3, in spite of some weaknesses, can be employed with a certain confidence to predict proton transfer activation barriers of large molecules or to obtain semi-quantitative trends for large numbers of interrelated systems. It is also shown that PM3 and AM1 results are of comparable quality for intramolecular proton transfers, whereas PM3 is to be preferred for intermolecular reactions, in particular when water molecules are involved.