Volume 115, Issue 24, 22 December 2001
Index of content:
- Theoretical Methods and Algorithms
Single-simulation determination of phase boundaries: A dynamic Clausius–Clapeyron integration method115(2001); http://dx.doi.org/10.1063/1.1420486View Description Hide Description
We present a dynamic implementation of the Clausius–Clapeyron integration (CCI) method for mapping out phase-coexistence boundaries through a single atomistic simulation run. In contrast to previous implementations, where the reversible path of coexistence conditions is generated from a series of independent equilibrium simulations, dynamic Clausius–Clapeyron integration (d-CCI) explores an entire coexistence boundary in a single nonequilibrium simulation. The method gives accurately the melting curve for a system of particles interacting through the Lennard-Jones potential. Furthermore, we apply d-CCI to compute the melting curve of an ab initio pair potential for argon and verify earlier studies on the effects of many-body interactions and quantum effects in the melting of argon. The d-CCI method shows to be effective in both applications, giving converged coexistence curves spanning a wide range of thermodynamic states from relatively short nonequilibrium simulations.
115(2001); http://dx.doi.org/10.1063/1.1421070View Description Hide Description
By taking the three-dimensional reaction system, the trajectory surface hopping method based on the Zhu–Nakamura theory is demonstrated to work much better than the old one and to be very promising to treat high-dimensional electronically nonadiabatic processes. The difference between the new and old survives even at high initial vibrational states and high energies.
Prediction of transition state barriers and enthalpies of reaction by a new hybrid density-functional approximation115(2001); http://dx.doi.org/10.1063/1.1415079View Description Hide Description
We present a new hybrid density-functional method which predicts transition state barriers with the same accuracy as CBS-APNO, and transition state barriers and enthalpies of reaction with smaller errors than B3LYP, BHandHLYP, and G2. The accuracy of the new method is demonstrated on 132 energies, including 74 transition state barriers and 58 enthalpies of reaction. For 40 reactions with reliable experimental barriers, the absolute mean deviations of the transition state barriers are 0.9, 1.0, 3.1, 3.5, and 3.6 kcal/mol for the new method and the CBS-APNO, G2, B3LYP, and BHandHLYP methods, respectively. The absolute mean deviations of the enthalpies of reaction for 38 reactions with reliable experimental enthalpies are 1.2, 1.4, 3.0, and 5.9 kcal/mol for the new method and the G2, B3LYP, and BHandHLYP methods, respectively. For the new method the maximum absolute deviations for the barriers and enthalpies of reaction are 2.6 and 5.6 kcal/mol, respectively. In addition, we present a simple scheme for a high-level correction that allows accurate determination of atomization energies. The accuracy of this scheme is demonstrated on the 55 atomization energies of the G2 test set [J. Chem. Phys. 94, 7221 (1992)].
115(2001); http://dx.doi.org/10.1063/1.1418442View Description Hide Description
Generalization of localized perturbation theory, which results with a method able to span the spin space correctly, is presented. This generalization is achieved by using a multiconfigurational (MC)wave function as the reference. This is the most comprehensive expansion used within MC–LMP2 approach to date, with, however, low computational cost [computational scaling with system size of the new method is Recently, we have reported the successful Jaguar2 (J2) model for calculating atomization energies. Within the MC–LMP2 framework, the J2 model for calculating heats of formation is based on the generalized valence bond–perfect pairing (GVB–PP) wave function. The J2 model was applied only to closed shell cases because of the perfect pairing (PP) restriction in the reference function. In order to describe other systems, the PP restriction needs to be lifted. This work describes efforts in that direction. The PP restriction can be lifted by a restricted configuration interaction (RCI) procedure applied to the GVB–PP wave function. In this paper, the equations describing the application of LMP2 theory to self-consistent RCI wave function are derived and explained. The RCI wave function is a “true” MC expansion as opposed to the GVB–PP, which uses only a single spin eigenfunction (SEF). We also present the self-consistent (SC) optimization of the RCI wave function. The SC–RCI–LMP2 is the first MC–LMP2 method where the spin space is spanned in the reference. This is important for describing the nondynamical correlation (near degeneracy) effects associated, for example, with bond breaking processes. The SC–RCI–LMP2 is an efficient method applicable to large systems; it is shown to reproduce the potential energy surfaces calculated by the complete active space–second order perturbation (CAS–SCF–PT2) method. This is demonstrated, for the first time, on some widely used test cases.
A posteriori corrections to systematic failures of standard density functionals: The dissociation of two-center three-electron systems115(2001); http://dx.doi.org/10.1063/1.1418439View Description Hide Description
The method we proposed recently [J. Chem. Phys., 114, 1447 (2000)] to a posteriori correct the unphysical dissociation behavior of radical homonuclear diatomic cations obtained in density functional theory calculations has been enlarged to nonsymmetric three-center two electrons systems. This approach, which is derived from Slater’s transition state technique, allows to remove most of the self-interaction energy error contained in the current exchange functionals. It has been shown that this is the main contribution to the overestimation of the bonding energy of systems with delocalized charges. Although approximate, the method yields a better agreement with experimental bonding energies than more sophisticated methods.
Prediction of electron paramagnetic resonance g values using coupled perturbed Hartree–Fock and Kohn–Sham theory115(2001); http://dx.doi.org/10.1063/1.1419058View Description Hide Description
A method for calculating the EPRg-tensor based on coupled perturbed Hartree–Fock (HF) and density functional theory(DFT) is presented. The one-electron molecular orbitals of a spin- unrestricted Slater determinant are calculated up to first order in the applied magnetic field. The g-tensor is evaluated as a mixed second derivative property with respect to the applied field and the electron magnetic moment. Thus, spin-polarization and spin–orbit coupling are simultaneously included in the calculation. The treatment focuses on orbitally nondegenerate molecules but is valid for a general ground state spin and, for the first time, it is possible to include hybrid density functionals in the treatment. The relativistic mass and diamagnetic gauge corrections are also considered. An implementation of the theory is described. Extensive numerical calculations for a series of small molecules are reported with the Hartree–Fock (HF) method, the local density approximation (LSD), the generalized gradient approximation (GGA) and hybrid density functionals such as B3LYP and PBE0 and large Gaussian basis sets. Detailed comparison with available ab initio and DFT calculations are made. The results indicate that the hybrid functionals offer little or no improvement over the GGA functionals for small radicals made of light atoms. For transition metal complexes the situation is different. The hybrid functionals give, on average, better results than the GGA functionals but significant disagreement between theoretical and experimental g-shifts still remain. Overall, the results indicate that the present method is among the most accurate so far developed models for the prediction of g values.
115(2001); http://dx.doi.org/10.1063/1.1421361View Description Hide Description
We propose a new method to compute reliable estimates for memory functions of dynamical variables from molecular dynamics simulations. The key point is that the dynamical variable under consideration, which we take to be the velocity of a fluid particle, is modeled as an autoregressive stochastic process. The parameters of this stochastic process can be determined from molecular dynamics trajectories using efficient algorithms that are well established in signal processing. The procedure is also referred to as the maximum entropy method. From the autoregressive model of the velocity autocorrelation function we compute the one-sided z transform of the discretized memory function and the memory function itself. Using liquid argon as a simple model system, we demonstrate that the autocorrelation function and its power spectrum can be approximated to almost arbitrary precision. The same is therefore true for the memory function, which is calculated within the same stochastic model.
- Gas Phase Dynamics and Structure: Spectroscopy, Molecular Interactions, Scattering, and Photochemistry
115(2001); http://dx.doi.org/10.1063/1.1421072View Description Hide Description
The reaction has been investigated in a beam-gas arrangement, with characterization of ScO products by cw laser-induced fluorescence: absorption versus laser frequency over the band and fluorescence over the one. It leads to the direct determination of the nascent rotational state distribution in the level of ScO. This distribution is close to a Prior statistical one, with a well-characterized weak “surprisal,” indicating that a momentum constraint takes place during the reaction process. In the frame of this statistical distribution, a new accurate value for the dissociation energy of ScO is proposed: Spectroscopic data are reported for the band, up to
Electronic spectroscopy of jet-cooled CFCl: Laser-induced fluorescence, dispersed fluorescence, lifetimes, and C–Cl dissociation barrier115(2001); http://dx.doi.org/10.1063/1.1418732View Description Hide Description
The transition of jet-cooled chlorofluorocarbene (CFCl) has been measured by laser-induced fluorescence(LIF) excitation and dispersed fluorescence spectroscopy. Over 170 vibronic transitions were measured in the LIF spectrum, consisting of cold bands and hot bands of carbenes containing both and isotopes. Dispersed fluorescence spectroscopy was used both to map the ground-state vibrational levels and to provide confirmation of the vibronic identity of the emitting level. A predictor–corrector method was used to progressively assign almost all of the vibronic transitions, resulting in the positive assignment and measurement of almost every bound vibrational state within the Ã-state manifold. The vibrational structure is modeled well by a Morse potential with frequencies and for and 1235 cm−1, 397.0 cm−1, and 744.5 cm−1 for the same three vibrations in The standard diagonal and cross-anharmonicity constants for a three-coordinate Morse potential were also measured for each isotopic species. Dispersed fluorescence spectroscopy provided a map of ground-state vibrational levels up to about 4000 cm−1. Franck–Condon factors were modeled well by a simple, one-dimensional harmonic potential, and these were also used to confirm assignment of many transitions. The fluorescence lifetime of the excited vibronic states decreased markedly from a consistent 650 ns for most states, to <20 ns for the highest lying observed state. In addition, the Franck–Condon analysis indicates that higher lying members of progressions were missing in the LIF spectrum. This strongly indicated the presence of a nonradiative pathway that opens for energies above Analysis of the rotational structure of many transitions indicated that the molecule was not reaching the Renner–Teller intersection, where the and states are degenerate. We attribute the nonradiative channel to cleavage of the C–Cl bond directly on the state, in exact analogy with the observed process in CFBr. The height of the barrier, and the vibrational frequencies are all in reasonable agreement with recent ab initio values.
115(2001); http://dx.doi.org/10.1063/1.1419060View Description Hide Description
The pure rotational spectrum of the SrSH radical in its ground electronic and vibrational states has been measured using millimeter/submillimeter-wave direct absorption techniques. This work is the first observation of SrSH with rotational resolution. The spectrum of its deuterium isotopomer SrSD and SrS has been recorded as well. These species were created by the reaction of strontium vapor and in the presence of a dc discharge. SrS was also made with For SrSH and SrSD, eight rotational transitions were recorded, respectively, for which asymmetry components up to were measured; fine structure was also resolved in each component. Thirteen transitions of SrS in each of its 1, and 2 states have additionally been observed. These data have been analyzed and spectroscopic parameters determined for all three species, including spin-rotation terms for the strontium hydrosulfides. From an structure calculation, the bond angle in SrSH was determined to be 91.48(3)°, very close to that of and CaSH. This geometry indicates that SrSH is a covalently bonded molecule, as opposed to linear (and ionic) SrOH. The Sr–S bond length in SrSH was also found to be greater than that of SrS versus 2.441 Å), indicating a change in bond order. In addition, the spin-rotation interaction in SrSH and SrSD includes a small contribution from the off-diagonal term, resulting from the crossing of energy levels with Second-order spin-orbit coupling appears to make a significant contribution to the spin-rotation splitting, as well, which must arise from mixing of the and excited states.
Internal rotation in high-resolution ultraviolet spectra. I. Semirigid model of a top– frame internal motion115(2001); http://dx.doi.org/10.1063/1.1416874View Description Hide Description
An internal rotor model to fit and simulate rotationally resolved electronic spectra of molecules or complexes with an internal top of symmetry and a frame of symmetry or in the semirigid approximation is described. The rotation–internal rotation problem is solved independently for both electronic states by numerical diagonalization of the energy matrices, which are factored according to the symmetry properties of the Hamiltonian. Finally, selection rules and line strength expressions for the simulation of electronic spectra are given.
Internal rotation in high-resolution ultraviolet spectra. II. Spectrum and structure of the aniline–nitrogen van der Waals complex115(2001); http://dx.doi.org/10.1063/1.1416875View Description Hide Description
Rotationally resolved electronic spectra of the nitrogen complex of aniline have been observed. The spectra are split into two subbands due to internal rotation of The analysis of the rotational constants reveals that is located above the ring plane of aniline and in the symmetry plane of aniline in the equilibrium position. Barriers hindering internal rotation have been obtained from fitting experimental transitions frequencies using a semirigid top– frame internal rotation model. Upon excitation into the distance of to the ring decreases and the internal rotation barrier increases by a factor larger than 2. Possible reasons for this behavior are discussed.
Nanosecond and femtosecond probing of the dynamics of the UV-photodissociation of perfluoroethyliodide115(2001); http://dx.doi.org/10.1063/1.1418743View Description Hide Description
The ns photodissociation of perfluoroethyliodide at 266 nm has been studied by using the resonant two-photonionization (R2PI) technique. Recoil anisotropy parameters as well as average translational energy of the I atoms in the fine structure states and have been determined. The main contribution (99%) to the absorption at 266 nm was found to be caused by a parallel transition to the state which gives mainly excited-state atoms The ground-state atoms were found to appear mainly (88%) from the primarily excited state via curve-crossing and to a lesser extent (12%) from direct absorption by a perpendicular transition to the and states. The fs pump–dump technique in combination with ns R2PI probing of the fragments and and time-of-flight mass spectrometry have been applied to probe the early stage dynamics of the molecule on the excited statepotential energy surface (PES). The evolution time of the excited molecule to the point where the energy gap between the excited state and the ground-statepotential energy surfaces drops to a value of about 12 440 cm−1 was found to be 52±13 fs. This time corresponds to about 0.8 Å extension of the C–I bond distance. The molecular dynamics simulation with DFT calculated ground-state PES and PES with the shape calculated for methyl iodide found in the literature gives reasonable agreement with the experimental result for the evolution time.
Variation with the intermolecular distance of properties dependent on the electron density in hydrogen bond dimers115(2001); http://dx.doi.org/10.1063/1.1420749View Description Hide Description
The variation with the intermolecular distance of features in hydrogen bond (HB) dimers dependent on the electron density ρ(r) are studied in four complexes representative of weak/medium HB interactions. Topological properties, energy densities and integrated atomic properties are obtained with ρ(r) of dimers at optimized structures obtained upon fully relaxing the geometry of monomers. The dependence of A–H⋯B bondproperties on intermolecular distances allows to characterize the nature of the interaction as monomers move nearer from infinite separation. At long distances the interaction is only electrostatic while for separations about 1 Å larger than the equilibrium distance quantum effects arising from ρ(r) begin to dominate. In the immediate neighborhood of the interaction is mainly led by the stabilization of the H-donor due in turn to energy lowerings in A and B atoms associated to polarization effects. The mutual penetration of electron densities of donor and acceptor monomers provokes a considerable reduction of atomic volumes for H and B atoms which reveals in the form of redistribution rather than transfer of charge. This range of distances exhibits noncovalent bond features but shortly after, when monomers approximate a few tenths of Å below characteristics typical of covalent interactions begin to appear while the rate of change of all the ρ(r)-dependent properties increases rapidly.
115(2001); http://dx.doi.org/10.1063/1.1421378View Description Hide Description
Anion femtosecondphotoelectron spectroscopy (FPES) has been used to monitor intramolecular electronic relaxation dynamics following the excitation of the electronic transition in The time-dependent photoelectron spectra provide a detailed picture of the relaxation dynamics in which the initially excited level evolves into highly vibrationally excited in its ground electronic state. The spectra show evidence for a two-step relaxation mechanism: internal conversion (IC) to vibrationally excited and states, occurring on a time scale of followed by IC from these intermediate states to highly vibrationally excited levels in the ground state with a time constant of
115(2001); http://dx.doi.org/10.1063/1.1419059View Description Hide Description
The possibility of electron binding to 9-acridinamine (9-AA) was studied at the second order Møller–Plesset perturbation theory level with the aug-cc-pVDZ basis set augmented with a diffuse set that has proven appropriate in earlier studies of weakly bound anions. It was found that both the amino and imino tautomers of 9-AA bind an excess electron to form stable anions. The vertical electron attachment energies corresponding to the amino and imino form were calculated to be 20 and 41 cm−1, respectively. It was found that while the imino 9-AA tautomer forms a typical dipole-bound anion, the electron binding energy for the amino tautomer calculated at the electrostatic Koopmans’ theorem level appears to be cancelled when the correlation correction to the dipole moment of the neutral is taken into account at the MP2 level. Therefore, the stability of the latter anion may be caused only by additional electron correlation effects, which are dominated by dispersion interactions. For this reason, we suggest that this anion may be termed a dispersion-bound anion.
115(2001); http://dx.doi.org/10.1063/1.1414347View Description Hide Description
The and electronic states of have been studied using a previously developed method [Carter et al., Mol. Phys. 98, 1967 (2000)] suitable for triatomic molecules showing three-state (Renner-Teller+vibronic) interactions.Ab initio three-dimensional diabatic potential energy surfaces for the and states have been computed at the multireference configuration interaction level of theory, using extended Gaussian basis sets. Additional computations were done to determine the barrier to isomerization over the three surfaces and the spin–orbit constant for the state. Energies, spin–orbit splittings, and rotational constants have then been calculated for all rovibronic levels of and symmetry up to 5800 for and 4800 for Assignments based on plots of vibrational wave functions are also provided. These computations have been finally used to revise previous interpretations of photoelectron spectra.
115(2001); http://dx.doi.org/10.1063/1.1412251View Description Hide Description
Spectroscopic constants and energy levels of the ground state and two low-lying excited states, and of ironcarbide FeC have been calculated by the ab initio multireference singles and doubles configuration interaction (MR-SDCI) molecular orbital method taking relativistic and spin–orbit coupling effects into account. Predicted rotational constant (0.6697 cm−1) and spin–orbit coupling constant (−181 cm−1) for the state agree well with experimental values. The first state which is in isoconfiguration with the state is predicted to lie at 3528 cm−1 above the state with and values of 0.6861, 923, and 7 cm−1, respectively. The lowest state described by one electron excitation from nonbonding 1δ orbital to antibonding 4π orbital should be located at 7248 cm−1 above the state with and values of 0.6268, 850, and 5 cm−1, respectively. Thus, considering the coincidence in the predicted excitation energy and spectroscopic constants for the state, the recently observed state at 3460 cm−1 above the state by Aiuchi et al. [Chem. Phys. Lett. 309, 229 (1999)], though tentatively assigned to the state, should be reassigned to the state.
- Condensed Phase Dynamics, Structure, and Thermodynamics: Spectroscopy, Reactions, and Relaxation
Crystallization and glass formation processes in methylcyclohexane: Vibrational dynamics as a possible molecular indicator of the liquid–glass transition115(2001); http://dx.doi.org/10.1063/1.1420490View Description Hide Description
We present the result of Raman studies on vibrational dynamics and static properties of the rocking mode of methylcyclohexane as a function of temperature and cooling rate. We have found that vibrational dynamics as well as the static properties are very sensitive indicators to specify phases and phase transitions at the molecular level. It was found that methylcyclohexane in the undercooled liquid phase may form distinct thermodynamic states that strongly depend on the quenching rate. We have identified the characteristic temperature that is quenching rate dependent which is the no-return-point between crystallization and amorphization. Below this temperature the undercooled methylcyclohexane exists in either the liquid state that is out of equilibrium and represents nonergodic behavior that leads to glass forming or in the metastable equilibrium ergodic state that leads to crystallization.
The production and decay kinetics of ClOO in water and freon-11: A time-resolved resonance raman study115(2001); http://dx.doi.org/10.1063/1.1418733View Description Hide Description
The production of ClOO following OClO photolysis in water and fluorotrichloromethane (freon-11) is investigated using time-resolved resonance Raman (TRRR) spectroscopy. Stokes spectra are obtained as a function of time following OClO photoexcitation using pump and probe wavelengths of 390 and 260 nm, respectively. Scattering assignable to ClOO is observed, and appears with a time constant of 27.9±4.5 ps in water and 172±30 ps in freon-11. The ClOO intensity decays with a time constant of ∼398±50 ps in water and 864±200 ps in freon-11. Although the production and decay kinetics are solvent dependent, the quantum yield for ClOO production is similar between water and freon-11. Femtosecond pump–probe studies designed to monitor the evolution in optical density at 390 and 260 nm following OClO photoexcitation are also presented. These studies demonstrate that geminate recombination of the primary photoproducts is less efficient in freon-11 relative to water. This result taken in combination with the solvent invariance of the ClOO-production quantum yield indicates that ClOO is not formed via geminate recombination. Instead, the results presented here suggest that OClO photoisomerization results in the production of ClOO. Finally, the vibrational energy content of ClOO upon internal conversion to the ground state is studied through comparison of the ClOO Raman and absorption cross sections to those predicted using computational methods. These studies suggest that ground-state ClOO is produced with minimal excess vibrational energy. The results presented here provide new insight into the mechanism of ClOO formation following OClO photoexcitation.