Search

It has long been known that there are multiple solutions to the self-consistent Hartree-Fock equations. This can be problematic if careful attention is not given to the orbital occupation and electronic state in the converged wave function. The issues with convergence have been demonstrated through the calculation of potential energy curves for O2, F2, Cl2, Br2, LiF, NaCl, CaO, MgO, ScO, FeO, TiO, YO, and ZrO. Hartree-Fock (HF) calculations were used to compute the points on the potential energy surface, with dynamic electron correlation included through the use of the completely renormalized coupled cluster, including singles, doubles, and perturbative triples [CR-CC(2,3)]. Even in regions with little to no multireference character, as determined by the T1/D1 diagnostics, HF does not always converge to the ground electronic state. As HF provides the reference wave function for CR-CC(2,3), and other post-Hartree-Fock ab initio methods, treatment of electron correlation does not necessarily result in a smooth potential energy curve, especially if HF is unable to produce a smooth curve. Even the convergence rate of multireference methods can be affected as the initial orbitals that form the basis for multireference calculations are frequently obtained from HF calculations.

Thom H. Dunning, Jr., provides an outstanding example of success in science, computing, and scientific leadership. From his scientific contributions to knowledge to his developments and accomplishments that have enabled discovery, Thom’s influence upon science, computing, and the scientific community has been significant.

The ground state and excited state manifolds are computed for PrF(2+) and PmF(2+) at the CASSCF (n,8) level of theory where the active space spans the Ln 4f orbitals as well as the F 2pz orbital. Dynamical correlation is included using second-order multireference quasidegenerate perturbation theory (MCQDPT2). The spin-orbit multiplets for each of the excited states are resolved, and spin-orbit coupling constants are computed using the Breit-Pauli spin-orbit operator. Equilibrium geometries for each of the ground and excited states are computed, and the nature of the Ln-F bond is examined. Potential energy curves for the lowest four triplet states and lowest two quintet states are computed for PrF(2+), which split into 14 levels upon application of the spin-orbit Hamiltonian. Likewise, the lowest six quintet states are computed for PmF(2+) as well as the lowest triplet state and the lowest two septet states. These nine states split into 43 terms upon application of the spin-orbit Hamiltonian.

The ground and low-lying excited state potential energy curves of neodymium monofluoride were calculated using multireference (CASSCF) and single reference (EOM-CR-CCSD(T)) methods. Optimized bond lengths were obtained and accurate bond dissociation energies were computed. The EOM-CR-CCSD(T) method was used to determine the bond dissociation energy of lutetium monofluoride, and it is shown that core correlation is required to produce bond dissociation energies in agreement with experiment.

For molecules containing second row atoms, unacceptable errors have been found in extrapolating dissociation energies calculated with the standard correlation consistent basis sets to the complete basis set limit. By carefully comparing the convergence behavior of De(O2) and De(SO), we show that the cause of these errors is a result of two inter-related problems: near duplication of the exponents in two of the d sets and a lack of high-exponent functions in the early members of the sets. Similar problems exist for the f sets (and probably in higher angular momentum sets), but have only a minor effect on the calculated dissociation energies. A number of approaches to address the problems in the d sets were investigated. Well behaved convergence was obtained by augmenting the (1d) and (2d) sets with a high-exponent function and by replacing the (3d) set by the (4d) set and the (4d) set by the (5d) set and so on. To ensure satisfactory coverage of both the L and M shell regions, the exponents of the new d se...

Valence correlation consistent and augmented correlation consistent basis sets have been determined for the third row, main group atoms gallium through krypton. The methodology, originally developed for the first row atoms, was first applied to the selenium atom, resulting in the expected natural groupings of correlation functions (although higher angular momentum functions tend to be relatively more important for the third row atoms as they were for the second row atoms). After testing the generality of the conclusions for the gallium atom, the procedure was used to generate correlation consistent basis sets for all of the atoms gallium through krypton. The correlation consistent basis sets for the third row main group atoms are as follows: cc-pVDZ: (14s11p6d)/[5s4p2d]; cc-pVTZ: (20s13p9d1f )/[6s5p3d1f]; cc-pVQZ: (21s16p12d2 f1g)/[7s6p4d2 f1g]; cc-pV5Z: (26s17p13d3f2g1h)/[8s7p5d3f2g1h]. Augmented sets were obtained by adding diffuse functions to the above sets (one for each angular momentum present in the set), with the exponents of the additional functions optimized in calculations on the atomic anions. Test calculations on the atoms as well as selected molecules with the new basis sets show good convergence to an apparent complete basis set limit.

The potential energy curves of the rare gas dimers He2, Ne2, and Ar2 have been computed using correlation consistent basis sets ranging from singly augmented aug-cc-pVDZ sets through triply augmented t-aug-cc-pV6Z sets, with the augmented sextuple basis sets being reported herein. Several methods for including electron correlation were investigated, namely Moller—Plesset perturbation theory (MP2, MP3 and MP4) and coupled cluster theory [CCSD and CCSD(T)]. For He2 CCSD(T)/d-aug-cc-pV6Z calculations yield a well depth of 7.35 cm-1 (10.58 K), with an estimated complete basis set (CBS) limit of 7.40 cm-1 (10.65 K). The latter is smaller than the 'exact' well depth (Aziz, R. A., Janzen, A. R., and Moldover, M. R., 1995, Phys. Rev. Lett., 74, 1586) by about 0.2 cm-1 (0.35 K). The Ne2 well depth, computed with the CCSD(T)/d-aug-cc-pV6Z method, is 28.31 cm-1 and the estimated CBS limit is 28.4 cm-1, approximately 1 cm-1 smaller than the empirical potential of Aziz, R. A., and Slaman, M., J., 1989, Chem. Phys., 13...

The tunneling effects in the molecular inversion of aziridine are investigated. A full-dimensional potential was constructed based on ab initio results and used to study the effect of vibrational excitations on tunneling. Using the semiclassical approach that incorporates tunneling into classical trajectory calculations, it is found that excitations of all the modes except the reaction coordinate have negligible effects on tunneling. This allows accurate thermal rate calculations carried out by using a one-dimensional model. (This is in accord with the conclusions of the work by Rom et al. [Chem. Phys. Lett. 204, 175 (1993)] and by Smedarchina et al. [J. Chem. Phys. 102, 7024 (1995)].) A one-dimensional Weutzel–Kramers–Brillouin (WKB) formula was employed to obtain the level splitting and was found to be very accurate based on comparisons with the quantum results. Furthermore, the calculated thermal rates are in good agreement with the experimentally measured values. The results provide insight into the reaction mechanism and explanations for the experimental findings of non-Rice–Ramsperger–Kassel–Marcus (RRKM) behavior and incoherent tunneling.

Activation of the inhibitor of κB kinase subunit β (IKKβ) oligomer initiates a cascade that results in the translocation of transcription factors involved in mediating immune responses. Dimerization of IKKβ is required for its activation. Coarse-grained and atomistic molecular dynamics simulations were used to investigate the conformation-activity and structure-activity relationships within the oligomer assembly of IKKβ that are impacted upon activation, mutation, and binding of ATP. Intermolecular interactions, free energies, and conformational changes were compared among several conformations, including a monomer, two different dimers, and the tetramer. Modifications to the activation segment induce conformational changes that disrupt dimerization and suggest that the multimeric assembly mediates a global stability for the enzyme that influences the activity of IKKβ.

Using systematic sequences of the newly developed correlation consistent core-valence basis sets from cc-pCVDZ through cc-pCV6Z, the spectroscopic constants of the homonuclear diatomic molecules containing first row atoms, B–F, are calculated both with and without inclusion of 1s correlation. Internally contracted multireference configuration interaction (IC-MRCI) and singles and doubles coupled cluster (CCSD) theory with a perturbational estimate of connected triple excitations, CCSD(T), have been investigated. By exploiting the convergence of the correlation consistent basis sets, complete basis set (CBS) limits have been estimated for total energies, dissociation energies, equilibrium geometries, and harmonic frequencies. Based on the estimated CBS limits the effects of 1s correlation on D e (kcal/mol), r e (A), and ω e (cm−1) are: +1.1, −0.0070, +10 for B2; +1.5, −0.0040, +13 for C2; +0.9, −0.0020, +9 for N2; +0.3, −0.0020, +6 for O2; and −0.1, −0.0015, +1 for F2.

Neodymium monofluoride dication was studied as a model of the Nd-F bond in NdFx. Multiconfigurational self-consistent field (MCSCF) and second order multireference quasi-degenerate perturbation theory (MCQDPT2) methods were used with a variety of active spaces to elucidate the roles of the Nd 4f, 5d, and 6s orbitals. Spin-orbit coupling calculations were performed at the SO-MCQDPT2 level, and potential energy curves were obtained for the four lowest energy quartet states as well as for the four lowest doublet states and the lowest sextet state. Inclusion of spin-orbit coupling splits these states into 30 levels. Equilibrium bond lengths, dissociation energies, transition energies, and crossing points were determined.

Studies were focused on late 3d and 4d transition metal ion (Fe, Co, Ni, Cu, Ru, Rh, Pd, and Ag) mediated activation of dimethyl ether, to investigate the intrinsic catalytic properties of metals on C-O bond cleavage. A set of density functional (DFT) methods (BLYP, B3LYP, M06, M06-L, B97-1, B97-D, TPSS, and PBE0) with aug-cc-pVTZ were utilized, and the results were calibrated with CCSD(T)/CBS. The utility of CCSD(T)/CBS calculations for these systems was validated by MRCI/aug-cc-pVTZ calculations. Calculations showed an interesting energetic trend as a function of metal; earlier transition metals tend to give smaller reaction barriers and more exergonic reactions than later metals. This applies to both 3d and 4d systems. For the performance of DFT functionals, PBE0 gave the lowest root mean squared deviations (RMSDs) in terms of both reaction energies and barriers for both 3d and 4d systems, compared to the other functionals. Our studies found that the percentage of Hartree-Fock (HF) exchange plays an important role in the accuracy of DFT methods for these systems, and 26% HF exchange for 3d systems and 34% HF exchange for 4d systems gave the lowest RMSDs.

The family of correlation consistent polarized valence basis sets (cc-pVXZ) has been extended to include sextuple zeta sets (cc-pV6Z) for the atoms boron through neon. Potential energy functions have been calculated with these sets for the electronic ground states of N2 and HF using a number of correlated wave functions: MP2, MP3, MP4, CCSD, CCSD(T) and CAS+1+2. Spectroscopic constants have been calculated for each level of theory and have been compared with experiment. Combining these results with those of prior studies, complete basis set limits have been estimated for Ee, De and re. It is found that the cc-pV6Z basis sets yield dissociation energies that are within 0.6–0.8 kcal mol−1 (N2) and 0.1 kcal mol−1 (HF) of the estimated CBS limits. Adding core-core and core-valence contributions to the CCSD(T) CBS limits yields De's that are within 0.1 kcal mol−1 of the experimental values.

Preface 1. Three Millennia of Atoms and Molecules Klaus Ruedenberg and W. H. Eugen Schwarz 2. Pioneering Quantum Chemistry in Concert with Experiment Istvan Hargittai 3. George W. Wheland: Forgotten Pioneer of Resonance Theory E. Thomas Strom 4. The Free-Electron Model William B. Jensen 5. Michael J. S. Dewar: A Model Iconoclast Eamonn F. Healy 6. H. C. Longuet-Higgins: The Man and His Science Weston Thatcher Borden 7. The Golden Years at LMSS and IBM San Jose Paul S. Bagus 8. Quantum Chemistry Program Exchange, Facilitator of Theoretical and Computational Chemistry in Pre-Internet History Donald B. Boyd 9. Molecular Orbital Theory for Organic Chemists Andrew Streitwieser 10. John Pople: The Man and His Science Janet E. Del Bene Editors' Biographies Indexes Author Index Subject Index

The continuing search for broadly applicable, predictive, and unique potential functions led to the invention of the multi-state modified embedded atom method (MS-MEAM) (Baskes et al 2007 Phys. Rev. B 75 094113). MS-MEAM replaced almost all of the prior arbitrary choices of the MEAM electron densities, embedding energy, pair potential, and angular screening functions by using first-principles computations of energy/volume relationships for multiple reference crystal structures and transformation paths connecting those reference structures. This strategy reasonably captured diverse interactions between atoms with variable coordinations in a face-centered-cubic (fcc)-stable copper system. However, a straightforward application of the original MS-MEAM framework to model technologically useful hexagonal-close-packed (hcp) metals proved elusive. This work describes the development of an hcp-stable/fcc-metastable MS-MEAM to model titanium by introducing a new angular function within the background electron density description. This critical insight enables the titanium MS-MEAM potential to reproduce first principles computations of reference structures and transformation paths extremely well. Importantly, it predicts lattice and elastic constants, defect energetics, and dynamics of non-ideal hcp and liquid titanium in good agreement with first principles computations and corresponding experiments, and often better than the three well-known literature models used as a benchmark. The titanium MS-MEAM has been made available in the Knowledgebase of Interatomic Models (https://openkim.org/) (Tadmor et al 2011 JOM 63 17).

The total energies, atomization energies, molecular structures, and harmonic vibrational frequencies of SO2 have been determined using the newly revised correlation consistent basis sets, cc-pV(n+d)Z, with n=2–5. The impact of the additional tight d functions on the energies, structures, and frequencies as well as the convergence of these quantities to the complete basis set (CBS) limit is examined. Overall, the revised sets result in faster convergence to the CBS limit. The effect of the additional tight d function is particularly dramatic for the lower level, double zeta and triple zeta, basis sets.

The relativistic pseudopotential variant of the correlation consistent composite approach (rp-ccCA) has been extended to lower p-block-containing molecules. The lower p-block is defined here as elements indium through iodine (In-I) and thallium through astatine (Tl-At). The accuracy of rp-ccCA has previously been established in a study of energetic properties of molecules containing 4p elements and thermodynamic data of 4d-containing molecules. In the current study, the atomic ionization potentials (IPs) and electron affinities (EAs) of In-I and Tl-At, as well as dissociation energies and enthalpies of formation of lower p-block mono-hydrides, mono-oxides, and dimers were determined with rp-ccCA. The compiled data set is referred to as the LP80 data set. The mean absolute deviation of the IPs and the EAs from the experimental data were 2.50 and 3.06 kcal mol(-1), respectively. Additionally, the total atomization energy for lead clusters (Pb(n), n = 2-6) was determined. For the lead clusters, a decrease in the total atomization energy was observed with the inclusion of spin-orbit coupling. The spin-orbit coupling of the molecules was determined with multireference configuration interaction, while scalar relativistic effects were accounted for within the pseudopotentials. The results obtained by rp-ccCA for the lower p-block support the utilization of the method in future studies where accurate energetic, spectroscopic, and thermodynamic data are necessary.

To focus on the identification of potential alternative amine carbon capture compounds, CO(2) with methyl, silyl, and trifluoromethyl monosubstituted and disubstituted amine compounds were studied. Interaction energies of these CO(2)·amine complexes were determined via two methods: (a) an ab initio composite method, the correlation consistent composite approach (ccCA), to determine interaction energies and (b) density functional theories, B3LYP/aug-cc-pVTZ and B97D/aug-cc-pVTZ. Substituent effects on the interaction energies were examined by interchanging electron donating and electron withdrawing substituents on the amine compounds. The calculations suggested two different binding modes, hydrogen bonding and acid-base interactions, which arise from the modification of the amine substituents, echoing previous work by our group on modeling protein·CO(2) interactions. Recommendations have been noted for the development of improved amine scrubber complexes.

To provide an efficient means to address nondynamical correlation, an empirical correction for single and double hybrid density functionals has been formulated. The correction utilizes the highest doubly occupied, lowest unoccupied, and singly occupied generalized Kohn-Sham orbitals and, for potential energy curves, a few additional strongly correlated orbitals. Utilizing proposed nondynamical correlation corrections, hybrid BLYP predicted potential energy curves that fit well to those obtained by ab initio multireference methods for the torsional rotation of ethylene and the automerization of cyclobutadiene. The empirical nondynamical correlation corrections, calculated at the transition structures, also reduce the overall errors of B2K-PLYP for the DBH24 and BH76 barrier height data sets. The proposed empirical correction can efficiently enhance the utility of both single and double hybrid density functionals for chemical situations with moderate to significant nondynamical correlation.

Rising global temperatures require innovative measures to reduce atmospheric concentrations of CO(2). The most successful carbon capture technology on Earth is the enzymatic capture of CO(2) and its sequestration in the form of glucose. Efforts to improve upon or mimic this naturally occurring process will require a rich understanding of protein-CO(2) interactions. Toward that end, extensive all-atom molecular dynamics (MD) simulations were performed on the CO(2)-utilizing enzyme phosphoenolpyruvate carboxykinase (PEPCK). Preliminary simulations were performed using implicit and explicit solvent models, which yielded similar results: arginine, lysine, tyrosine, and asparagine enhance the ability of a protein to bind carbon dioxide. Extensive explicit solvent simulations were performed for both wild-type PEPCK and five single-point PEPCK mutants, revealing three prevalent channels by which CO(2) enters (or exits) the active site cleft, as well as a fourth channel (observed only once), the existence of which can be rationalized in terms of the position of a key Arg residue. The strongest CO(2) binding sites in these simulations consist of appropriately positioned hydrogen bond donors and acceptors. Interactions between CO(2) and both Mn(2+) and Mg(2+) present in PEPCK are minimal due to the stable protein- and solvent-based coordination environments of these cations. His 232, suggested by X-ray crystallography as being a potential important CO(2) binding site, is indeed found to be particularly "CO(2)-philic" in these simulations. Finally, a recent mechanism, proposed on the basis of X-ray crystallography, for PEPCK active site lid closure is discussed in light of the MD trajectories. Overall, the results of this work will prove useful not only to scientists investigating PEPCK, but also to groups seeking to develop an environmentally benign, protein-based carbon capture, sequestration, and utilization system.

The correlation consistent composite approach (ccCA) is a thermochemical model devised to achieve energetics (i.e. enthalpies of formation, ionization potentials, and proton affinities) akin to those obtained using electronic correlation methods such as CCSD(T) at the complete basis set limit, but at much greater computational efficiency. While the approach has proven great utility, the prediction of features of potential energy surfaces, bond formation and bond breaking, and excited states can warrant a multireference wavefunction-based approach. This report illustrates the development of MR-ccCA, a multireference analog of ccCA, which is based upon a CASPT2 reference. The successful application of MR-ccCA has been demonstrated in the study of potential energy curves of C 2 , spectroscopic constants of C 2 , N 2 , and O 2 , and properties of silicon compounds and a few other first row and second row molecules.

Theoretical studies, using the multireference correlation consistent composite approach (MR-ccCA), have been carried out on the ground and lowest lying spin-forbidden excited states of a series of silicon-containing systems. The MR-ccCA method is the multireference equivalent of the successful single reference ccCA method that has been shown to produce chemically accurate (within ±1.0 kcal mol(-1) of reliable, well-established experiment) results. The percentage contributions of the SCF configurations to complete active space self-consistent field wave functions together with the Frobenius norm of the t(1) vectors and related D(1) diagnostics of the coupled-cluster single double wave function with the cc-pVTZ basis set have been utilized to illustrate the multi-configurational characteristics of the compounds considered. MR-ccCA incorporates additive terms to account for relativistic effects, atomic spin-orbit coupling, scalar relativistic effects, and core-valence correlation. MR-ccCA has been utilized to predict the atomization energies, enthalpies of formation, and the lowest energy spin-forbidden transitions for Si(n)X(m) (2 ≤ n + m ≥ 3 where n ≠ 0 and X = B, C, N, Al, P), silicon hydrides, and analogous compounds of carbon. The energetics of small silicon aluminides and phosphorides are predicted for the first time.

The mechanism of the reverse water-gas shift reaction (CO(2) + H(2) → CO + H(2)O) was investigated using the 3d transition metal complexes L'M (M = Fe, Mn, and Co, L' = parent β-diketiminate). The thermodynamics and reaction barriers of the elementary reaction pathways were studied with the B3LYP density functional and two different basis sets: 6-311+G(d) and aug-cc-pVTZ. Plausible reactants, intermediates, transition states, and products were modeled, with different conformers and multiplicities for each identified. Different reaction pathways and side reactions were also considered. Reaction Gibbs free energies and activation energies for all steps were determined for each transition metal. Calculations indicate that the most desirable mechanism involves mostly monometallic complexes. Among the three catalysts modeled, the Mn complex shows the most favorable catalytic properties. Considering the individual reaction barriers, the Fe complex shows the lowest barrier for activation of CO(2).

Computational studies of nickel carbides, particularly Ni(2)C, are scarce. A systematic density functional theory study is reported for Ni(2)C, along with NiC and Ni(3)C, to understand the stability and electronic structure of nickel carbides of varying stoichiometry. A comprehensive study was executed that involved 28 trial structures of varying space group symmetry for Ni(2)C. An analysis of the electronic structure, geometry and thermodynamics of Ni(2)C is performed, and compared with that for Ni(3)C and NiC as well as several defect structures of varying composition. It is found that the most stable ground state arrangement of Ni(2)C exists within a simple orthorhombic lattice and that it has metallic character. The calculated formation energies (kcal mol(-1)) of NiC, Ni(2)C, and Ni(3)C are 48.6, 7.9 and 6.4, respectively.

A multireference analog of the correlation consistent composite approach (MR-ccCA) based on complete active space with second-order perturbation theory (CASPT2) has been utilized in an investigation of the ground and valence excited states of C(2), N(2), and O(2). The performance of different second-order multireference perturbation theory methods including second-order n-electron valence state perturbation theory, second-order multireference Møller-Plesset, and second-order generalized van Vleck perturbation theory has been analyzed as potential alternatives to CASPT2 within MR-ccCA. The MR-ccCA-P predicts spectroscopic constants with overall mean absolute deviations from experimental values of 0.0006 Å, 7.0 cm(-1), and 143 cm(-1) for equilibrium bond length (r(e)), harmonic frequency (ω(e)), and term values (T(e)), respectively, which are comparable to the predictions by more computationally costly multireference configuration interaction-based methods.

Density functional (Xα, BLYP, BPW91, B3LYP, and B3PW91), MP2, and coupled cluster with singles, doubles, and quasiperturbative triple excitations (CCSD(T)) methods in combination with LANL2DZ, cc-pVxZ, cc-pVxZ-PP, or cc-pVxZ-DK (where x = D (double), T (triple) Q (quadruple), and 5 (quintuple)) basis sets have been employed in computations of excited states of zinc and cadmium dimers and monomers. The spectroscopic constants r(e), D(e), T(e), ω(e), ω(e)x(e), and B(e) are compared among group 12 dimers in their ground and low-lying phosphorescent excited states. The weak metallophilic bonding in the ground state requires MP2 or CCSD(T) description, whereas the excimeric bonding is well described by CCSD(T), BLYP, B3LYP, BPW91, and B3PW91. Consistent with our previous work on Hg(n) species, absorption and phosphorescence transition energies that agree well with experimental results can be obtained with CCSD(T) and B3PW91 methods in combination with the correlation consistent basis sets of triple-ζ or higher level. This is the case for monomers and *Cd(2) and *Zn(2) lowest triplet excimers (in either the (3)Σ(u)(+) or (3)Π(g) state). The CCSD(T)/aug-cc-pV5Z-PP combination results in values that are only 150 and 240 cm(-1) (1% errors) removed from the experimental values for the (1)S → (3)D atomic transitions for Zn and Cd, respectively. Such spectroscopic constants provide the most accurate theoretical values known to date for bonding and photophysical parameters characteristic of Zn(n) and Cd(n) species.

Scale factors for (a) low (

A QM/QM approach incorporating the correlation consistent composite approach (ccCA) into the ONIOM multilayer methodology has been implemented. This new multilayer composite scheme, ONIOM-ccCA, enables the accurate prediction of thermochemical properties for systems containing dozens-and potentially hundreds-of atoms. ONIOM-ccCA is used to predict the C-H bond dissociation energies of 18 anthracene and fluorene analogues, containing up to 43 atoms, to within 1.2 kcal mol(-1) of experimental values. Several density functional and basis set combinations are evaluated for use as the low level QM layer. The mean absolute deviation (1.2 kcal mol(-1)) using the most accurate ONIOM-ccCA method, ccCA:B3LYP/cc-pVTZ, is significantly lower than that of an earlier reported multilayer composite scheme [2.4 kcal mol(-1), Li, M.-J.; Liu, L.; Fu, Y.; Guo, Q.-X. J. Phys. Chem. B, 2005, 109, 13818]. Thus, through use of ONIOM-ccCA, accurate thermochemical calculations are now feasible for sizable molecular systems of chemical or biological interest.

Gas-phase enthalpies of formation (Δ H f(g) °) have been determined for 40 nitrogen-containing compounds at 298 K. Three ab initio composite methods have been compared in their abilities to quantitatively determine Δ H f(g) °; the G3, G3(MP2), and correlation consistent Composite Approach (ccCA) methodologies. The ccCA method resulted in a mean absolute deviation (MAD) of 1.1 kcal mol −1 when compared to available experimental values. The comparable G3(MP2) method resulted in a MAD of 1.8 kcal mol −1 , while the G3 method resulted in a MAD of 1.2 kcal mol −1 . As a result of their comparable accuracies, the ccCA and G3 methods have been utilized to predict the Δ H f(g) ° of five energetic but highly endothermic tetrazine-containing compounds with potential applications as insensitive high explosives.

The correlation consistent Composite Approach, ccCA [ Deyonker , N. J. ; Cundari , T. R. ; Wilson , A. K. J. Chem. Phys. 2006 , 124 , 114104 ] has been demonstrated to predict accurate thermochemical properties of chemical species that can be described by a single configurational reference state, and at reduced computational cost, as compared with ab initio methods such as CCSD(T) used in combination with large basis sets. We have developed three variants of a multireference equivalent of this successful theoretical model. The method, called the multireference correlation consistent composite approach (MR-ccCA), is designed to predict the thermochemical properties of reactive intermediates, excited state species, and transition states to within chemical accuracy (e.g., 1 kcal/mol for enthalpies of formation) of reliable experimental values. In this study, we have demonstrated the utility of MR-ccCA: (1) in the determination of the adiabatic singlet-triplet energy separations and enthalpies of formation for the ground states for a set of diradicals and unsaturated compounds, and (2) in the prediction of energetic barriers to internal rotation, in ethylene and its heavier congener, disilene. Additionally, we have utilized MR-ccCA to predict the enthalpies of formation of the low-lying excited states of all the species considered. MR-ccCA is shown to give quantitative results without reliance upon empirically derived parameters, making it suitable for application to study novel chemical systems with significant nondynamical correlation effects.

The reverse water-gas shift reaction CO(2) + H(2) --H(2)O + CO has been investigated using a set of homogeneous catalyst models L'M(I) (L' = beta-diketiminate, C(3)N(2)H(5)(-); M = Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn). The thermodynamics of prototypical reaction pathways were simulated at two levels of theory: B3LYP/6-311+G(d) and B3LYP/aug-cc-pVTZ. The modeled catalytic reaction has been considered in the following steps: coordination of CO(2) by the catalyst to generate a carbon dioxide complex, L'M(CO(2)); scission of L'M(CO(2)) to yield L'M(CO) and L'M(O); L'M(O) hydrogenation to form L'M(H(2)O). The final products, H(2)O and CO, were obtained from the dissociation of L'M(H(2)O) and L'M(CO). All of the reactants, intermediates, and products were modeled, where different possible conformers and multiplicities were identified and considered as potential minima. The reaction enthalpy DeltaH, of all steps for each catalyst as a function of transition metal have been determined. The Mn and Fe catalysts show more thermodynamically accessible pathways than the other catalyst models studied. The overall reaction enthalpy has been determined not only by B3LYP/6-311+G(d) and B3LYP/aug-cc-pVTZ but also via a more rigorous ab initio electron-correlation-based approach, the correlation consistent Composite Approach (ccCA).

We report an atomistic simulation study of the behavior of nanocomposite materials that are formed by incorporating single-walled carbon nanotubes SWCNTs, with three different diameters, and a multiwalled carbon nanotube MWCNT into a single-crystal nickel matrix. The interactions between carbon and nickel atoms are described by a modified embedded atom method potential. Mechanical properties of these nanocomposite materials are predicted by atomistic calculations and compared with that of fcc nickel and pristine CNTs. Our simulations predict that all Ni/CNT composites studied in this work are mechanically stable. Their elastic properties depend on the volume fraction and diameter of embedded CNTs. The single-crystal Young’s modulus E11 of Ni/SWCNT composites exhibit a large increase in the direction of CNTs alignment compared to that of a single-crystal nickel. However, a moderate but gradual decrease is seen for E22 and E33 in the transverse directions with increase in CNT diameters. As a consequence, Ni/SWCNTs show a gradual decrease for the polycrystalline Young’s, bulk and shear moduli with the increasing CNT diameters and volume fractions. These reductions, although moderate, suggest that enhancement of mechanical properties for polycrystalline Ni/SWCNT nanocomposites are not achievable at any CNT volume fraction. The Ni/MWCNT composite with high CNT volume fraction shows the highest increase in E11. Unlike the E22 and E33 for Ni/ SWCNTs, there is a significant increase in the E22 and the E33 for Ni/MWCNT. As a result, polycrystalline Ni/MWCNT composites show slight increase in the elastic properties. This suggests that nickel nanocomposites with enhanced mechanical properties can be fabricated using large volume fractions of larger diameter MWCNTs. Depending on type, alignment and volume fraction, Ni/CNT composites show varying degrees of elastic anisotropy and Poisson’s ratio compared to pure Ni. Simulation predicts strong adhesion at the Ni/CNT interface and a significant interfacial stress transfer between CNT and Ni matrix.

The rising atmospheric concentration of CO(2) has motivated researchers to seek routes for improved utilization, increased mitigation, and enhanced sequestration of this greenhouse gas. Through a combination of bioinformatics, molecular modeling, and first-principles quantum mechanics the binding of carbon dioxide to proteins is analyzed. It is concluded that acid/base interactions are the principal chemical force by which CO(2) is bound inside proteins. With respect to regular secondary structural elements, beta-sheets show a marked preference for CO(2) binding compared to alpha-helices. The data also support the inference that while either or both oxygens of CO(2) are generally tightly bound in the protein environment, the carbon is much less "sequestered." First principles and more approximate modeling techniques are assessed for quantifying CO(2) binding thermodynamics.

The geometries, enthalpies of formation (DeltaH(o)(f)), separations of electronic states, electron affinities, gas-phase acidities, and bond dissociation energies associated with the alkali metal monoxide anions (MO(-)), monoxide radicals (MO(*)), and hydroxides (MOH) (M = Li, Na, and K) have been investigated using single-reference and multireference variants of the WnC procedures. Our best estimates of the DeltaH(o)(f) values for the ground states at 298 K are as follows: 8.5 ((3)Pi LiO(-)), 48.5 ((2)Pi LiO(*)), -243.4 ((1)Sigma(+) LiOH), 34.2 ((3)Pi NaO(-)), 86.4 ((2)Pi NaO(*)), -190.8 ((1)Sigma(+) NaOH), 15.1 ((1)Sigma(+) KO(-)), 55.9 ((2)Sigma(+) KO(*)), and -227.0 ((1)Sigma(+) KOH) kJ mol(-1). While the LiO(*) and NaO(*) radicals have (2)Pi ground states, for KO(*), the (2)Sigma(+) and (2)Pi electronic states lie very close in energy, with our best estimate being a preference for the (2)Sigma(+) state by 1.1 kJ mol(-1) at 0 K. In a similar manner, the ground state for MO(-) changes from (3)Pi for LiO(-) and NaO(-) to (1)Sigma(+) for KO(-). The (1)Sigma(+) state of KO(-) is indicated by the calculated T(1) diagnostic and the SCF contribution to the total atomization energy to have a significant degree of multireference character. This leads to a difference of more than 100 kJ mol(-1) between the single-reference W2C and multireference W2C-CAS-ACPF and W2C-CAS-AQCC estimates for the (1)Sigma(+) DeltaH(o)(f) for KO(-).

The performance of 44 density functionals used in conjunction with the correlation consistent basis sets (cc-pVnZ where n = T and Q) has been assessed for the gas-phase enthalpies of formation at 298.15 K of 3d transition metal (TM) containing systems. Nineteen molecules were examined: ScS, VO, VO(2), Cr(CO)(6), MnS, MnCl(2), Mn(CO)(5)Cl, FeCl(3), Fe(CO)(5), CoH(CO)(4), NiCl(2), Ni(CO)(4), CuH, CuF, CuCl, ZnH, ZnO, ZnCl, and Zn(CH(3))(2). Of the functionals examined, the functionals that resulted in the smallest mean absolute deviation (MAD, in parentheses, kcal mol(-1)) from experiment were B97-1 (6.9), PBE1KCIS (8.1), TPSS1KCIS (9.6), B97-2 (9.7), and B98 (10.7). All five of these functionals include some degree of Hartree-Fock (HF) exchange. The impact of increasing the basis set from cc-pVTZ to cc-pVQZ was found to be slight for the generalized gradient approximation (GGA) and meta-GGA (MGGA) functionals studied, indicating basis set saturation at the triple-zeta level. By contrast, for most of the generalized gradient exchange (GGE), hybrid GGA (HGGA), and hybrid meta-GGA (HMGGA) functionals considered, improvements in the average MAD of 2-3 kcal mol(-1) were seen upon progressing to a quadruple-zeta level basis set. Overall, it was found that the functionals that include Hartree-Fock exchange performed best overall, but those with greater than 40% HF exchange exhibit significantly poor performance for the prediction of enthalpies of formation for 3d TM complexes. Carbonyl-containing complexes, a mainstay in organometallic TM chemistry, are demonstrated to be exceedingly difficult to describe accurately with all but 2 of the 44 functionals considered. The most accurate functional, for both CO-containing and CO-free compounds, is B97-1/cc-pVQZ, which is shown to be capable of yielding results within 1 kcal mol(-1) of high-level ab initio composite methodologies.

The correlation consistent composite approach (ccCA) was applied to the prediction of reaction barrier heights (i.e., transition state energy relative to reactants and products) for a standard benchmark set of reactions comprised of both hydrogen transfer reactions and nonhydrogen transfer reactions (i.e., heavy-atom transfer, SN2, and unimolecular reactions). The ccCA method was compared against G3B for the same set of reactions. Error metrics indicate that ccCA achieves "chemical accuracy" with a mean unsigned error (MUE) of 0.89 kcal/mol with respect to the benchmark data for barrier heights; G3B has a mean unsigned error of 1.94 kcal/mol. Further, the greater accuracy of ccCA for predicted reaction barriers is compared to other benchmarked literature methods, including density functional (BB1K, MUE=1.16 kcal/mol) and wavefunction-based [QCISD(T), MUE=1.10 kcal/mol] methods.

The family of correlation consistent polarized valence basis sets has been extended in order to account for core-core and core-valence correlation effects within the third-row, main group atoms gallium through krypton. Construction of the basis sets is similar to that of the atoms boron through argon, where either the difference between core-correlated and valence-only correlation energies were calculated via configuration interaction (CISD) computations on the ground electronic states of the atoms (named cc-pCVnZ) or the sets were optimized with respect to the core-valence correlation energy and a small weight of core-core correlation energy (cc-pwCVnZ). Due to the correlation of 3d orbitals, added shells of higher angular momentum exponents compared to the valence sets are necessary to describe the core region. The pattern of added core-correlating functions is (1s1p1d1f) for double-zeta, (2s2p2d2f1g) for triple-zeta, (3s3p3d3f2g1h) for quadruple-zeta, and (4s4p4d4f3g2h1i) for quintuple-zeta. Atomic and molecular results show good convergence to the CBS limit, with the cc-pwCVnZ sets showing improved convergence compared to the cc-pCVnZ ones for molecular core-valence correlation effects. After testing the basis sets on the homonuclear diatomics Ga2-Kr2 with coupled cluster wave functions, it is concluded that a treatment of core-valence correlation effects is essential for high-accuracy ab initio investigations of third-row-containing molecules. Though the basis sets are optimal for 3s3p3d correlation, preliminary atomic and molecular results show the basis sets to be efficient with respect to 3d-only correlation, and these potentially could be used with 3d-only correlation for more qualitative studies on larger species.

The correlation consistent composite approach (ccCA) is a model chemistry that has been shown to accurately compute gas-phase enthalpies of formation for alkali and alkaline earth metal oxides and hydroxides (Ho, D. S.; DeYonker, N. J.; Wilson, A. K.; Cundari, T. R. J. Phys. Chem. A 2006, 110, 9767). The ccCA results contrast to more widely used model chemistries where calculated enthalpies of formation for such species can be in error by up to 90 kcal mol-1. In this study, we have applied ccCA to a more general set of 42 s-block molecules and compared the ccCA DeltaHf values to values obtained using the G3 and G3B model chemistries. Included in this training set are water complexes such as Na(H2O)n+ where n = 1 - 4, dimers and trimers of ionic compounds such as (LiCl)2 and (LiCl)3, and the largest ccCA computation to date: Be(acac)2, BeC10H14O4. Problems with the G3 model chemistries seem to be isolated to metal-oxygen bonded systems and Be-containing systems, as G3 and G3B still perform quite well with a 2.7 and 2.6 kcal mol-1 mean absolute deviation (MAD), respectively, for gas-phase enthalpies of formation. The MAD of the ccCA is only 2.2 kcal mol-1 for enthalpies of formation (DeltaHf) for all compounds studied herein. While this MAD is roughly double that found for a ccCA study of >350 main group (i.e., p-block) compounds, it is commensurate with typical experimental uncertainties for s-block complexes. Some molecules where G3/G3B and ccCA computed DeltaHf values deviate significantly from experiment, such as (LiCl)3, NaCN, and MgF, are inviting candidates for new experimental and high-level theoretical studies.

Systematic truncation of the correlation consistent basis sets has been investigated in first and second row atoms and molecules to better understand basis set requirements for density functional theory, particularly the need for high angular momentum functions, as well as to understand possible computational cost savings that could be achieved by using reduced basis sets. The truncation scheme employed follows that recently introduced for ab initio methods [B. Mintz et al., J. Chem. Phys. 121, 5629 (2004)]. Properties examined in the current study include geometries, ionization potentials, electron affinities, and dissociation energies. In general, this investigation shows that a degree of truncation of higher angular momentum functions is possible with limited impact upon energetic properties, and does result in useful CPU time savings. However, not all properties investigated have the same level of dependence upon high angular momentum functions, and, thus, careful selection of truncated basis sets should be made.

Density functional theory (B3LYP) and ab initio theory [second-order Moller-Plesset perturbation theory (MP2) and coupled-cluster theory including single, double, and quasiperturbative triple excitations (CCSD(T))] have been used in combination with the standard and augmented correlation consistent basis sets (cc-pVnZ and aug-cc-pVnZ, where n = D, T, and Q) to investigate potential new noble gas compounds. Two classes of molecules were studied: XNgCCNgX and XNgCCX, where Ng = Kr and Ar and X = F and Cl. These molecules were characterized by finding the ground-state structures and calculating the relative energies, charge distributions, and vibrational frequencies. In addition, transition-state structures were also determined and decomposition pathways were identified through intrinsic reaction coordinate calculations.

Preface 1. Explicityly Correlated Basis Functions for Large Molecules 2. Uniform Density Limit of Exchange-Correlation Energy Functionals 3. Self-Consistent Hartree-Fock-Wigner Calculations: A Two-Electron-Density Functinoal Theory 4. New Alternatives for Accurate Electronic Structure Calculations of Potential Energy Surfaces Involving Bond Breaking 5. Bond Breaking in Quantum Chemistry: A Comparison of Single- and Mulit-Reference Methods 6. Breaking the Curse of the Non-Dynamical Correlation Problem: the Spin-Flip Method 7. Economical Description of Electron Correlation 8. Correlation Consistent Basis Sets with Relativistic Effective Core Potentials: The Transition Metal Elements Y and Hg 9. Multilevel Methods for Thermochemistry and Thermochemical Kinetics 10. A Nonlocal Energy Functional Derived from the Fluctuation-Dissipation Theorem 11. The Protonation Site of Aniline Revisited: A "Torture Test" for Electron Correlation Methods 12. Equilibrium Structure of the Silicon Trimer Author Index Subject Index

An effective approach for the determination of lanthanide energetics, as demonstrated by application to the third ionization energy (in the gas phase) for the first half of the lanthanide series, has been developed. This approach uses a combination of highly correlated and fully relativistic ab initio methods to accurately describe the electronic structure of heavy elements. Both scalar and fully relativistic methods are used to achieve an approach that is both computationally feasible and accurate. The impact of basis set choice and the number of electrons included in the correlation space has also been examined.

Computing the enthalpies of formation for alkali metal and alkaline earth metal oxides (M(x)O) and hydroxides [M(OH)(n)] using the Gaussian-n (Gn) and Weismann-n (Wn) ab initio model chemistries is difficult due to an improper treatment of core-valence electron correlation effects. Using a new model chemistry called the correlation consistent Composite Approach (ccCA), enthalpies of formation were determined for eight different alkali/alkaline earth metal oxides and hydroxides. Unlike the Gn and Wn model chemistries, which must be modified to properly account for core-valence electron correlation, the standard implementations of the ccCA provide acceptable results, and all enthalpies of formation obtained with the ccCA are within the accepted range of recommended values.

Spectroscopic and bonding properties of Hg(n) oligomers and *Hg(n) exciplexes are determined by rigorous theoretical treatments. Reliable values that agree well with experimental data have been computed for the luminescence energies and other molecular spectroscopic parameters by making a careful selection of theoretical methods and basis sets. The calculations clarified the assignments for several phosphorescence bands in the mercury vapor based on calculated energies and other parameters that quantify the large excited-state distortion in the emitting states. Both the weak ground-state mercurophilic bonding and the stronger covalent bonding in the triplet and quintet excited states studied are found to be cooperative, which is important for fundamental and applied research for luminescent and magnetic materials that have spectral behavior similar to that of Hg(n) systems.