Multireference Single - And Double-excitation Configuration Interaction Research Articles

Trends in adsorption of open-shell atoms and small molecular fragments on the Ag(1 1 1) surface

The chemisorption of various atoms (C, N, O, Cl) and molecular fragments (OH, NH, CH, NH 2, CH 2) on the Ag(1 1 1) surface has been studied by employing the embedded cluster and multireference single- and double-excitation configuration interaction (MRD-CI) methods. Ground and excited states of the cluster–adsorbate systems have been computed and molecular orbitals (MOs) as well as electronic charge density distributions and Mulliken populations have been analyzed in order to extract general trends in chemisorption properties for different adsorbates. It has been found that the adsorbate–surface bond is energetically most favorable when a maximum of two electrons of the metal are shared with a given adsorbate. As a result atomic/molecular fragments with less than six valence electrons (N, CH, C) retain some open shells upon adsorption, whereas oxygen as well as chlorine isovalent species form a singlet ground state on the surface. All species considered except for Cl have mainly covalent bonding character to the surface, with an electronic charge of up to 1.0 transferred to the adsorbate from the silver cluster. It has been shown that the ionicity of the bond is strongly correlated with the electron affinity of the adsorbed species. Binding energies, equilibrium geometry and adsorbate location on the cluster have been computed and compared with available experimental data. In addition, the characteristic properties of chemisorption on Ag(1 1 1) and Pt(1 1 1) surfaces have been compared.

Multireference CI study of the potential curves and properties of positronic complexes of alkali hydrides

Ab initio multireference single- and double-excitation configuration interaction (MRD-CI) calculations have been carried out to study the manner in which the electronic structure of the series of neutral alkali hydride molecules is affected by the interaction of a lone positron. Four systems have been treated: LiH, NaH, KH and RbH. A new computer program has been constructed for this purpose that makes use of the Table-Direct-CI method for molecular calculations within the Born–Oppenheimer approximation. The main emphasis in the present work is on internuclear geometries with relatively large bond distances near the dissociation limit. Molecular charge density contour diagrams are presented to illustrate the bonding mechanism in each case. Trends in a variety of quantities such as positron affinities at both equilibrium molecular and separated atomic geometries, dissociation energies and positronium formation energies are computed and analyzed as the atomic number of the alkali atom is increased.

Positron Binding Energies for Alkali Hydrides

Ab initio multireference single- and double-excitation configuration interaction (MRD-CI) calculations are carried out to study the interactions of positrons with the members of the alkali hydride class of molecules. A new computer program has been constructed for this purpose that makes use of the Table-Direct-CI method for construction of the required Hamiltonian matrixes and electronic/positronic wave functions. The calculations indicate that the binding energy (positron affinity PA) of a single positron to these systems increases by an increment of 0.2-0.3 eV as the atomic number of the alkali atom is increased. It is found that the positron prefers a location in the more electronegative regions of such molecules, similarly as has been found in earlier calculations for the urea and acetone molecules. The positron orbital itself possesses a diffuse charge distribution with relatively small expectation values of the kinetic energy in all four systems considered. Each of the four positronic molecules is stable with respect to formation of either positronium (Ps) or HPs according to the present calculations. Relatively large changes in the equilibrium bond distance of the hydrides occur as a result of the positron interaction. The importance of bond dipole moments in producing the binding of positrons to molecules is discussed, as well as the role that the electronegativity of the constituent atoms plays in determining the magnitude of the PA for a given system.

Potential energy surface and intermolecular vibrations of O 2–H 2O

The potential energy surface (PES) of O 2–H 2O was studied by second-order Møller–Plesset perturbation and multi-reference single- and double-excitation configuration interaction (MRSDCI) methods with aug-cc-pVTZ basis sets. The in-plane mutual rotations of the monomers and the intermolecular stretching were also investigated by numerically solving the Schrödinger equation for nuclear motions with the three-dimensional finite element method. The lowest states for the conrotatory motion are actually degenerate, while those for the disrotatory one are split by 6.9 cm −1, which is consistent with the recent result of Fourier-transform microwave spectroscopy, where the C 2v structure was obtained.

Potential energy surface and intermolecular vibrations of O2?H2O

The potential energy surface (PES) of O 2 –H 2 O was studied by second-order Møller–Plesset perturbation and multi-reference single- and double-excitation configuration interaction (MRSDCI) methods with aug-cc-pVTZ basis sets. The in-plane mutual rotations of the monomers and the intermolecular stretching were also investigated by numerically solving the Schrödinger equation for nuclear motions with the three-dimensional finite element method. The lowest states for the conrotatory motion are actually degenerate, while those for the disrotatory one are split by 6.9 cm −1 , which is consistent with the recent result of Fourier-transform microwave spectroscopy, where the C 2v structure was obtained.

Theoretical study on lower electronic states of the FeSi molecule

The low lying electronic states of the FeSi molecule have been investigated by performing complete active space self-consistent field (CASSCF) and multireference single and double excitation configuration interaction (MRSDCI) calculations. Although the classification of the ground state was not established, the results of the MRSDCI calculations suggest that the ground state of FeSi is 3Δ, and the lowest excited state is 5Π whose transition energy is 0.36 eV including the Davidson-type Q correction. The R e given by MRSDCI with Q correction is 2.23 Å for 3Δ and 2.19 Å for 5Π. The spectroscopic constants and dipole moments of the low lying 5Δ and 3Π states as well as the 3Δ and 5Π states are also evaluated. The bonding nature of FeSi in the 3Δ state is discussed in comparison with the FeC molecule.

Theoretical study of the TiC molecule: clarification of the ground state

The electronic structure of the TiC molecule was examined using three types of multi-reference single- and double-excitation configuration interaction (MRSDCI) calculations with highly extended basis sets. Multi-reference coupled pair approximation (MRCPA) was applied after the MRSDCI calculations that included core—valence and core—core correlation in addition to the valence correlation (v-c-c CI). From the results of the most accurate calculation with MRCPA (v-c-c CPA) it was concluded that a 1Σ+ state is the ground state, despite previous calculations that suggested a 3Σ+ state with a 1Σ+ state lying only slightly above it. The 1Σ+ state is more highly correlated than the 3Σ+ state, and it was found that use of a size-consistent method is necessary to predict the relative stability accurately. The study evaluated the spectroscopic constants and considered the effect of the core (Ti 3p) correlation on these parameters. By taking the core correlation effect into account, the estimation of the dissociation energy (D e) was improved dramatically; D e obtained through v-c-c CPA was 4.457eV for the 1Σ+ ground state, which agrees well with the experimental value (the latest being 4.35 ± 0.31 eV).

GRECP/5e‐MRD‐CI calculation of the electronic structure of PbH

AbstractThe correlation calculation of the electronic structure of PbH is carried out with the generalized relativistic effective core potential (GRECP) and multireference single‐ and double‐excitation configuration interaction (MRD‐CI) methods. The 22‐electron GRECP for Pb is used and the outer core 5s, 5p, and 5d pseudospinors are frozen using the level‐shift technique, so only five external electrons of PbH are correlated. A new configuration selection scheme with respect to the relativistic multireference states is employed in the framework of the MRD‐CI method. The [6, 4, 3, 2] correlation spin–orbit basis set is optimized in the coupled cluster calculations on the Pb atom using a recently proposed procedure, in which functions in the spin–orbital basis set are generated from calculations of different ionic states of the Pb atom and those functions are considered optimal that provide the stationary point for some energy functional. Spectroscopic constants for the two lowest‐lying electronic states of PbH (2Π1/2, 2Π3/2) are found to be in good agreement with the experimental data. © 2002 Wiley Periodicals, Inc. Int J Quantum Chem, 2002

Ab initio study of spectroscopic and radiative characteristics of ion-pair states of the Cl2 molecule

Electronic structure and radiative characteristics of low-lying ion-pair states of Cl2 converging to the Cl+(3P, 1D)+Cl−(1S) limits are studied. Ab initio calculations of potential energy curves for the valence and ion-pair states and dipole moments for transitions between them are carried out employing the multireference single- and double-excitation configuration interaction (MRD-CI) method, including spin–orbit coupling. It is shown that the lowest two pairs of the Ω=0u+, 1u ion-pair states arise from an avoided crossing between the Σu−3 and Πu3 parent Λ−S states, which leads to notably anharmonic shapes of the corresponding potential curves and their mixed Λ−S nature. This causes significant radial coupling, resulting in the strongly perturbed character of the 0u+ and 1u states observed experimentally. In contrast, their gerade counterparts run parallel to one another and exhibit much less perturbation. Spectroscopic properties of the computed adiabatic curves are in very good agreement with the available experimental data. Dipole moments have been calculated for parallel ion-pair–valence state transitions and radiative lifetimes have been obtained for the adiabatic ion-pair states. A reanalysis of the experimental bound–free emission spectra from the D0u+(3P2) state [N. K. Bibinov et al., Chem. Phys. 254, 89 (2000)] is given.

The MRSDCI/CIS study of excited electronic states of the SF 2 radical

The vertical ( T v) and adiabatic ( T 0) excitation energies for singlet electronic excited states of the SF 2 radical have been calculated by using the multireference single and double excitation configuration interaction (MRSDCI) method and aug-cc-pVTZ basis sets augmented by Rydberg functions. The MRSDCI T v calculations indicate that the X 1 A 1 , 1 1 A 2 , 1 1 B 1 , 2 1 B 1 , 2 1 A 2 , 2 1 A 1 , 3 1 B 1 , 4 1 B 1 , 3 1 A 1 , and 1 1 B 2 states are the 10 lowest-lying singlet states. Based on the MRSDCI//CIS T 0 calculations (using CIS optimized geometries for excited states), the A, B, C, E, F, G, H, and I states of SF 2 are assigned to 1 1 B 1 , 2 1 B 1 , 3 1 B 1 , 2 1 A 2 , 2 1 A 1 , 3 1 A 1 , 4 1 B 1 , and 1 1 B 2 , respectively.

GRECP/MRD-CI calculations of spin-orbit splitting in ground state of Tl and of spectroscopic properties of TlH

The generalized relativistic effective core potential (GRECP) approach is employed in the framework of multireference single- and double-excitation configuration interaction (MRD-CI) method to calculate the spin-orbit splitting in the 2Po ground state of the Tl atom and spectroscopic constants for the 0+ ground state of TlH. The 21-electron GRECP for Tl is used, and the outer core 5s and 5p pseudospinors are frozen with the help of the level shift technique. The spin-orbit selection scheme with respect to relativistic multireference states and the corresponding code are developed and applied in the calculations. In this procedure both correlation and spin-orbit interactions are taken into account. A [4,4,4,3,2] basis set is optimized for the Tl atom and employed in the TlH calculations. Very good agreement is found for the equilibrium distance, vibrational frequency, and dissociation energy of the TlH ground state (Re=1.870 Å, ωe=1420 cm−1, De=2.049 eV) as compared with the experimental data (Re=1.872 Å, ωe=1391 cm−1, De=2.06 eV). © 2001 John Wiley & Sons, Inc. Int J Quant Chem 81: 409–421, 2001

Neon Matrix Isolation Electron Spin Resonance and Theoretical Studies of the Various Isotopomers of the CH3Mg Radical

The 12CH3Mg, 13CH3Mg, 12CH325Mg, 12CD325Mg, 13CH325Mg and 13CD325Mg radicals have been isolated in an inert neon matrix at 4.3 K and their electronic structure probed, for the first time, using matrix isolation electron spin resonance (MI-ESR) spectroscopy. These radicals were formed from the reaction of laser-ablated magnesium metal and an appropriately labeled derivative of acetone or methyl iodide. The spin Hamiltonian parameters, g⊥ = 1.9999(4), A⊥(25Mg) = −184(1) MHz, A⊥(13C) = 128(2) MHz and A⊥(H) = 7(1) MHz were determined from an exact diagonalization analysis of the experimental spectra and estimates were derived for A∥(25Mg) = −197(10) MHz and A∥(13C) = 180(20) MHz assuming g∥ = 2.0023. A model for the bonding in the CH3Mg radical is derived using this hyperfine data. Comparisons are made between the CH3Mg radical and other related magnesium and monomethylmetal radicals, MgH, MgOH, CH3Cd, CH3Zn, and CH3Ba. Theoretical nuclear hyperfine coupling constants for the CH3Mg radical were evaluated using Hartree−Fock single and double excitation configuration interaction (HFSDCI), multireference single and double excitation configuration interaction (MRSDCI) and density functional theory (DFT) ab initio calculations. While these theoretical methods yielded values for Adip(25Mg) and Adip(13C) in agreement with the experimental values, the calculated Aiso(25Mg) value was low by 4% (HFSDCI) and 15% (MRSDCI). Whereas the calculated Aiso(13C) values were low by 50% (HFSDCI) and 32% (MRSDCI). Unrestricted DFT calculations using the B3PW91 and B3LYP functionals yielded values of Aiso(25Mg) low by approximately 15% for both functionals and values of Aiso(13C) in agreement with experiment for UB3LYP and low by 10% for UB3PW91. The discrepancy between the calculated and experimental values of Aiso(13C) for the CI results is attributed to the limited reference space resulting in an overestimation of the ionic character in the bonding of the CH3Mg radical.

An ab Initio Molecular Orbital Prediction of the Spectroscopic Constants of [FORMULA][F][A][AC]X[/AC][AC][d5][/AC][/A][HSP SP="0.167"][ZW][SUP]1[/SUP]A[INF]1[/INF][/F][/FORMULA] M[CLC]g[/CLC]C[TINF]2[/TINF

Spectroscopic constants for 1A1 MgC2 have been calculated at the multireference single- and double-excitation configuration interaction (MR-SDCI) + Davidson correction (Q) level with the correlation-consistent polarized valence quadruple zeta (cc-pVQZ) basis sets to estimate rotational transition frequencies. The MR-SDCI + Q potential energy function predicts the rotational constants A0, B0, and C0 to be 51,794, 11,494, and 9379 MHz and the centrifugal distortion constants ΔJ, ΔJK, ΔK, δJ, and δK to be 0.014, 0.21, -0.023, 0.0027, and 0.14 MHz, respectively. The C-C and Mg-C internuclear distances in the T-shaped equilibrium geometry have been found to be 1.275 and 2.012 A, respectively. MgC2 consists of an Mg+ ion and a (CC)- moiety with dipole moment of 7.9 D. The ν1 (C-C stretching), ν2 (Mg-C2 stretching), and ν3 (rocking) vibrational frequencies have been estimated to be 1704, 595, and 456 cm-1, respectively. These results indicate that MgC2 is a rigid molecule, in contrast to SiC2 which is known to show a large-amplitude motion. Rotational transition frequencies have been estimated based on these spectroscopic constants.

The formation of SiH+, PH+, and SH+ by radiative association

Cross sections for the radiative association of Si+, P+, and S+ with atomic hydrogen to form the diatomic molecular ions SiH+, PH+, and SH+, respectively, are presented. The results are obtained using fully quantum-mechanical techniques and utilize ab initio molecular data calculated with the multireference single- and double-excitation configuration-interaction (MRD-CI) method. Thermal rate coefficients are presented and applications of the results to diffuse interstellar clouds are discussed.

Neon and argon matrix ESR and theoretical studies of the 12CH3Cd, 12CD3Cd, 13CH3Cd, 12CH3111Cd, and 12CH3113Cd radicals

Electron spin resonance (ESR) studies are reported for the first time on the various isotopomers of the CH3Cd radical isolated in neon and argon matrices. The radicals were generated in neon matrices by the reaction of laser-ablated cadmium metal and various methyl precursors, and in argon matrices by x-irradiation. The neon matrix values measured were g⊥=1.9491(1), A⊥(H)=17.0(1) MHz, A⊥(D)=2.6(1) MHz, A⊥(13C)=163(3) MHz, and A⊥(111Cd)=−3083(3) MHz, and estimates were derived for A∥(13C)=230(50) MHz and A∥(111Cd)=−3486(5) MHz. The argon matrix values measured were g⊥=1.952(1), A⊥(H)=16(1) MHz and A⊥(111Cd)=−3301(3) MHz, and an estimate was derived for A∥(111Cd)=−3704(5) MHz. The ESR experimental Adip(111Cd) values for the neon and the argon matrices agree with the reported gas-phase value [J. Chem. Phys. 101, 6396 (1994)]. The matrix ESR Aiso(111Cd) values show small shifts compared with the gas-phase results (5% greater for the neon matrix and 12% greater for the argon matrix). At 4.3 K in the neon matrices, additional ESR lines assigned to tunneling phenomena were observed. The radical geometry obtained from ab initio calculations was consistent with that reported from the various experimental results. Multireference single and double excitation configuration interaction (MRSDCI) calculations of the hyperfine interactions gave values that were consistently below the experimental values of Aiso and Adip for 111Cd, H, and 13C. MRSDCI calculations for the CdH radical showed an analogous trend.

Investigation of vibrational states of the ArHCl+ cation in the electronic ground state

Abstract The potential energy surface (PES) of the electronic ground state of the ArHCl + molecular ion is calculated by the multireference single- and double-excitation configuration interaction (MRD-CI) technique. An analytical representation of the potential energy data is obtained in the form of a power series. The vibrational eigenvalue problem is treated with high accuracy on the basis of the variational method using a direct product basis contracted in two steps (truncation/recoupling method). Unexpectedly for such a weakly bound system the spectrum turns out to be of a rather regular nature up to the first dissociation energy, allowing for an assignment of quantum numbers to a major part of the vibrational states. Furthermore, the wave functions are analysed together with several geometrical and energetic characteristics of the various vibrational states.

Density functional and ab initio study of the ketene and methylketene radical cations

The present study for the ketene and methylketene radical cations involves geometry, conformation, and hyperfine structure. The geometry and conformation (for the methylketene cation) investigations were based on the B3LYP/6-311G(d,p) and MP2/6-311G(d,p) calculations, and the isotropic proton hyperfine coupling constants (a(H)) were calculated using various density functional theory (DFT) methods and ab initio UHF, UMP2, and multireference single and double excitation configuration interaction (MRSDCI) methods with large basis sets. The ketene cation is predicted to have a C2v geometry. The barriers to methyl group rotation in the methylketene cation is predicted to be very low (<1 kcal/mol) and the cis (staggered) conformation is the most stable. On the rotation, the sum of the B3LYP/6-311G(d,p) a(H) values on the three β protons in the methylketene cation is almost a constant approximately equal to three times the experimental β coupling value. All the three ab initio methods predict too small a(H) values for the β protons. The B3LYP (also B3P86 and B3PW91) calculations predict a(H) values for the α and β protons in the two radicals in excellent agreement with experiment and the DFT methods do not underestimate the β couplings. The UMP2(FULL)/6-311G(d,p) hyperfine structure calculations predict a(H) values for the α protons in excellent agreement with experiment and the UMP2 hfcc results are overall better than the MRSDCI results. As an ab initio method for hyperfine structure calculations, the UMP2 method is suggested.

Ab Initio Structure and Energetics for the Molecular and Dissociative Adsorption of NH3 on Si(100)-2 × 1

We predict the structures and detailed energetics for the dissociative adsorption of NH3 to form NH2 and H adsorbed on a single Si dimer on the Si(100)-2 × 1 surface at the MRSDCI (multireference single and double excitation configuration interaction) level of theory. We predict that this dissociation involves two steps: (i) barrierless molecular chemisorption of NH3 followed by (ii) activated N−H bond cleavage of NH3(a) to form NH2(a) + H(a). While the second step involves a barrier, its relatively small height renders the overall reaction barrierless. The extremely high adsorption exothermicity (∼75 kcal/mol) results in a very high desorption barrier. These results can explain the experimentally determined high sticking probability of NH3, the observation of NH3(a) at low temperatures, and the observed stability of NH2(a) and H(a) on the Si(100) surface up to ∼600 K. Additionally, our CASSCF level (complete active space self-consistent-field) calculated geometries for the dissociatively adsorbed specie...

Bound states and time-dependent dynamics of the N2H+ molecular ion in its ground electronic state. I. 2D treatment

The ground-state potential energy surface (PES) for linear arrangements of the N2H+ molecular ion is numerically computed by the multireference single- and double-excitation configuration interaction (MRD-CI) technique. An analytical representation of the potential energy function is obtained by fitting a power series in the Simons–Parr–Finlan coordinates to the numerical data. For investigating the intramolecular dynamics we describe the nuclear motion by a Gaussian wave packet located initially in the strong interaction region of the PES. The vibrational eigenvalue spectrum is calculated by Fourier transforming the time autocorrelation function. The spectrum is then analyzed statistically in the light of random matrix theory (RMT) to understand the nature of the intramolecular dynamics. We examine the short-range correlation in the spectrum through the nearest neighbor level spacing distribution P(s) and the long-range correlation through Δ3 and Σ2 statistics. The spectrum in the time domain is analyzed by computing the ensemble averaged survival probability 〈〈P(t)〉〉. The above four quantities obtained from the spectrum are compared with the distribution predicted for regular, irregular, and mixed (intermediate) spectra by the RMT. We find the system is of mixed type and the fractional irregularity is 0.7±0.05. In order to reveal a possible correspondence to the classical dynamics, we have carried out the spectral analysis of the dynamical variables for classical trajectories over a wide range of internal energies. In addition the classical dynamics of proton collisions with N2 molecules has also been preliminarily studied on the same PES, in particular the dependence of the final vibrational action nf on the initial vibrational phase φi of N2 and, furthermore, the Poincaré surface-of-section superimposed with the zero-order separatrix; we find a large number of trapped trajectories.

Accuracy of the MRSD-CI technique in computing the X 2∑ + ScO and TiN hyperfine tensors

The hyperfine tensors of X 2∑ + ScO and TiN are computed using the multi-reference single and double excitation configuration interaction (MRSD-CI) method. The accuracy of this method is studied as the number of double excitations and reference configurations in the CI wave function is increased. The computed b F ( 45 Sc) and b F ( 47 Ti) parameters are within 92 and 93% of the experimental values respectively while those of c( 45 Sc) and c( 47 Ti) are within 97 to 99%. Thus the MRSD-CI technique seems to be a feasible tool for predicting the hyperfine parameters in this class of diatomics.