Combined theoretical and experimental deep-UV resonance raman studies of substituted pyrenes

The results of time-dependent density functional theory (TDDFT) calculations of resonance Raman intensities are combined with experimental deep-ultraviolet resonance Raman measurements at a single wavelength, i.e., 244 nm, in order to test the possibility to distinguish several very similar compounds. Pyrene and three of its substituted derivatives, in which a single hydrogen atom has been replaced by a halogen atom, are compared. The fixed 244 nm excitation wavelength overlapped with the same electronic transition of the four pyrenes. Ground-state calculations using the BP86 exchange-correlation functional were used to predict the Raman frequencies, whereas excited-state calculations have been carried out employing the "statistical averaging of (model) orbital potentials" (SAOP) potential within a linear-response TDDFT framework in combination with the short-time approximation of resonance Raman intensities. In view of the simplistic theoretical approach, we find a surprisingly good agreement between the simulated and measured resonance Raman spectra of pyrene and its substituted analogues in terms of frequencies and intensities, which shows that the calculations can be used reliably to interpret the experimental spectra. With this combined information, it is possible to find criteria to distinguish the compounds under investigation, although many features of their vibrational spectra are similar.     

On the accuracy of <Emphasis Type="Italic">ab initio</Emphasis> calculations of absolute band intensities in the IR spectra of molecules

Absolute intensities are calculated in the harmonic approximation for the IR spectrum of 18 hydrocarbons, oxygen- and nitrogen-containing organic compounds. The quantum chemical calculation is carried out in the 6-311G(3df,3pd) basis set. The calculated data are compared to the experimental values of absolute absorption intensities. It is found that calculations in the HF approximation substantially overestimate (on average by 87%) the integral absolute intensity of fundamental vibrations in the 575–4000 cm^?1 range. Most part of this overestimation falls on the stretching vibrations with large amplitudes (C-H and C=O bonds). When the MP2 method is used without electron correlation treatment, this overestimation decreases to 22% and becomes more uniform in the whole spectral region.     

Resonance Raman spectra of uracil based on Kramers-Kronig relations using time-dependent density functional calculations and multireference perturbation theory

The use of time-dependent density functional calculations for the optimization of excited-state structures and the subsequent calculation of resonance Raman intensities within the transform-theory framework is compared to calculations of Hartree-Fock/configuration interaction singles-type (CIS). The transform theory of resonance Raman scattering is based on Kramers-Kronig relations between polarizability tensor components and the optical absorption. Stationary points for the two lowest excited singlet states of uracil are optimized and characterized by means of numerical differentiation of analytical excited-state gradients. It is shown that the effect of electron correlation leads to substantial modifications of the relative intensities. Calculations of vibrational frequencies for ground and excited states are carried out, which show that the neglect of Duschinsky mixing and the assumption of equal wave numbers for ground and excited state are not in all cases good approximations. We also compare the transform-theory resonance Raman intensities with those obtained within a simple approximation from excited-state gradients at the ground-state equilibrium position, and find that they are in qualitative agreement in the case of CIS, but show some important differences in calculations based on density functional theory. Since the results from CIS calculations are in better agreement with experiment, we also present approximate resonance Raman spectra obtained using excited-state gradients from multireference perturbation theory calculations, which confirm the CIS gradients.     

“Spectroscopic” Calculations of CH Bond Dissociation Energies for Ethane, Propane, Butane, Isobutane, Pentane, Hexane, and Neopentane Using Fundamental Vibration Frequencies

The fundamental spectrum and the parameters of the potential function of a number of saturated hydrocarbon molecules are calculated in an anharmonic approximation. The calculation is performed by the variational technique using a minimal Morse-harmonic basis. The potential function is taken as the sum of the Morse function for CH bonds and the harmonic function for the skeletal and deformation vibrations. The initial approximation for the potential function is found by ab initio calculations in a 6-31G basis and refined by solving the inverse problem. The calculated CH bond dissociation energies depend significantly on the molecular structure and on the position of CH bonds in the molecule. These energies correlate well with the experimental cleavage energies of these bonds. The changes in the dipole moment of the molecule induced by vibrations were found by ab initio calculations in a 6-31G basis. The calculated IR transmission curves are in good agreement with the experimental curves.     

Theoretical Simulation of Vibrational Sum‐Frequency Generation Spectra from Density Functional Theory: Application to p‐Nitrothiophenol and 2,4‐Dinitroaniline

The molecular orientation of adsorbed molecules forming self‐assembled monolayers can be determined by combining vibrational sum‐frequency generation (SFG) measurements with quantum chemical calculations. Herein, we present a theoretical methodology used to simulate the SFG spectra for different combinations of polarizations. These simulations are based on calculations of the IR vectors and Raman tensors, which are obtained from density functional theory computations. The dependency of the SFG vibrational signature with respect to the molecular orientation is presented for the molecules p‐nitrothiophenol and 2,4‐dinitroaniline. It is found that a suitable choice of basis set as well as of exchange‐correlation (XC) functional is mandatory to correctly simulate the SFG intensities and consequently provide an accurate estimation of the adsorbed molecule orientation. Comparison with experimental data shows that calculations performed at the B3LYP/6‐311++G(d,p) level of approximation provide good agreement with experimental frequencies, and with IR and Raman intensities. In particular, it is demonstrated that polarization and diffuse functions are compulsory for reproducing the IR and Raman spectra, and consequently vibrational SFG spectra, of systems such as p‐nitrothiophenol. Moreover, the investigated XC functionals reveal their influence on the relative intensities, which show rather systematic variations with the amount of Hartree–Fock exchange. Finally, further aspects of the modeling are revealed by considering the frequency dependence of the Raman tensors.     

Solid-state modeling of the terahertz spectrum of the high explosive HMX

The experimental solid-state terahertz (THz) spectrum (3-120 cm(-1)) of the beta-crystal form of the high explosive octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine (HMX) has been analyzed using solid-state density functional theory calculations. Various density functionals (both generalized gradient approximation and local density approximation) are compared in terms of their abilities to reproduce the experimentally observed solid-state structure and low-frequency vibrational motions. Good-to-excellent agreement between solid-state theory and experiment can be achieved in the THz region where isolated-molecule calculations fail to reproduce the observed spectral features, demonstrating a clear limitation of using isolated-molecule calculations for the assignment of THz frequency motions in molecular solids. The deficiency of isolated-molecule calculations is traced to modification of the molecular structure in the solid state through crystal packing effects and the formation of weak C-H...O hydrogen bonds.     

Quantum Monte Carlo study of the reaction: Cl+CH3OH-->CH2OH+HCl

A theoretical study is reported of the Cl+CH3OH-->CH2OH+HCl reaction based on the diffusion Monte Carlo (DMC) variant of the quantum Monte Carlo method. Using a DMC trial function constructed as a product of Hartree-Fock and correlation functions, we have computed the barrier height, heat of reaction, atomization energies, and heats of formation of reagents and products. The DMC heat of reaction, atomization energies, and heats of formation are found to agree with experiment to within the error bounds of computation and experiment. M?ller-Plesset second order perturbation theory (MP2) and density functional theory, the latter in the B3LYP generalized gradient approximation, are found to overestimate the experimental heat of reaction. Intrinsic reaction coordinate calculations at the MP2 level of theory demonstrate that the reaction is predominantly direct, i.e., proceeds without formation of intermediates, which is consistent with a recent molecular beam experiment. The reaction barrier as determined from MP2 calculations is found to be 2.24 kcal/mol and by DMC it is computed to be 2.39(49) kcal/mol.     

Statistical thermodynamics of 1-butanol, 2-methyl-1-propanol, and butanal

The purpose of the present investigation is to calculate partition functions and thermodynamic quantities, viz., entropy, enthalpy, heat capacity, and Gibbs free energies, for 1-butanol, 2-methyl-1-propanol, and butanal in the vapor phase. We employed the multi-structural (MS) anharmonicity method and electronic structure calculations including both explicitly correlated coupled cluster theory and density functional theory. The calculations are performed using all structures for each molecule and employing both the local harmonic approximation (MS-LH) and the inclusion of torsional anharmonicity (MS-T). The results obtained from the MS-T calculations are in excellent agreement with experimental data taken from the Thermodynamics Research Center data series and the CRC Handbook of Chemistry and Physics, where available. They are also compared with Benson's empirical group additivity values, where available; in most cases, the present results are more accurate than the group additivity values. In other cases, where experimental data (but not group additivity values) are available, we also obtain good agreement with experiment. This validates the accuracy of the electronic structure calculations when combined with the MS-T method for estimating the thermodynamic properties of systems with multiple torsions, and it increases our confidence in the predictions made with this method for molecules and temperatures where experimental or empirical data are not available.     

Molecular structure of the octatetranyl anion, C8H-: a computational study

The equilibrium molecular structure of the octatetranyl anion, C8H(-), which has been recently detected in two astronomical environments, is investigated with the aid of both ab initio post-Hartree-Fock and density functional theory (DFT) calculations. The model chemistry adopted in this study was selected after a series of benchmark calculations performed on molecular acetylene for which accurate gas-phase structural data are available. Geometry optimizations performed at the CCSD/6-311+G(2d,p), QCISD/6-311+G(2d,p), and MP4(SDQ)/6-311+G(2d,p) levels of theory yield for C8H(-) an interesting polyyne-type structure that defies the chemical formula displaying a simple alternation of triple and single carbon-carbon bonds, [:C[triple bond]C-C[triple bond]C-C[triple bond]C-C[triple bond]CH](1-). In the optimized geometry of C8H(-), as one proceeds from the naked carbon atom on one side of the chain to the CH unit on the opposite side of the chain, the short (formally triple) carbon-carbon bonds decrease in length from 1.255 to 1.213 A whereas the long (formally single) carbon-carbon bonds increase (albeit only slightly) in length from 1.362 to 1.378 A (CCSD results). In striking contrast, both MP2 and DFT (B3LYP and PBE0) calculations fail in reproducing the pattern of the carbon-carbon bond lengths obtained with the CCSD, QCISD, and MP4 methods. The structures of three shorter n-even chains, C(n)H(-) (n = 2, 4, and 6), along with those of four n-odd compounds (n = 3, 5, 7, and 9) are also investigated at the CCSD/6-311+G(2d,p) level of theory.     

Second hyperpolarizability (gamma) of singlet diradical system: dependence of gamma on the diradical character

The dependence of the second hyperpolarizability (gamma) on the diradical character (y) for singlet diradical systems is investigated using a model compound, the p-quinodimethane (PQM) molecule with different both-end carbon-carbon (C-C) bond lengths, by several ab initio molecular orbital and density functional theory methods. The diradical character based on UHF calculations indicates that at equilibrium geometry PQM is in a singlet ground state and primarily exhibits a quinoid structure, whereas the diradical character increases when increasing both-end C-C bond lengths. At the highest level of approximation, that is, using the UCCSD(T) method with the 6-31G+diffuse p (zeta = 0.0523) basis set, the longitudinal static gamma of PQM presents a maximum value for intermediate diradical character (y approximately 0.5) while the gamma values are larger for intermediate and large diradical character (y approximately 0.5-0.7) than for small diradical character (y < 0.2). This feature suggests that the gamma values of singlet diradical systems in the intermediate and somewhat strong correlation regimes are significantly enhanced as compared to those in the weak correlation regime. These results are substantiated by a complementary study of the variation in gamma upon twisted ethylene.     

Calculation of enthalpies of hydrogenation of hydrocarbons

Sets of hydrogen molecule equivalents have been developed which permit the calculation of hydrogenation of different types of carbon-carbon bonds from ab initio total energies (3-21G and 6-31G* basis sets, and, to a more limited extent, for MP2/6-31G* data) of reactants and products. The calculated enthalpies of hydrogenation are in good agreement with experiment for unstrained molecules, with average errors on the order of 2 kcal/mol. The 6-31G* equivalents allow the enthalpies for strained molecules to be calculated accurately, but the 3-21G equivalents do not. The equivalents for both basis sets have been tested by calculating the enthalpies of hydrogenation of carbon-carbon bonds in nitrogen- and oxygen-containing organic molecules, free radicals, and classical carbocations. The results are in good agreement with experiment in most cases.     

Vibrational dynamics of hydrogen‐bonded HCN complexes with OH and NH acids: Computational DFT systematic study

Vibrational properties (band position, infrared [IR], and Raman intensities) of C?N stretching mode were studied in 65 gas phase hydrogen‐bonded 1:1 complexes of HCN with OH acids and NH acids using density functional theory (DFT) calculations at the B3LYP‐6‐311++G(d,p) level. Furthermore, general characteristics of the hydrogen bonds and vibrational changes in acids OH/NH stretching bands were also considered. Experimentally observed blue shift of the C?N stretching band promoted by hydrogen bonding, which shortens the triple bond length, is very well reproduced and quantitatively depends on the hydrogen bond length. Both IR and Raman ν(C?N) band intensities are enhanced, also in good agreement with the experimental results. IR intensity increase is a direct function of the hydrogen bond energy. However, the predicted Raman intensity raise is a more complex function, depending simultaneously on characteristics of both the hydrogen bond (C?N bond length) and the H‐donating acid (polarizability). With these two parameters, ν (C?N) Raman intensities of the complexes are explained with a mean error of ±2.4%. © 2006 Wiley Periodicals, Inc. Int J Quantum Chem, 2007     

Amide I vibrational circular dichroism of polypeptides: generalized fragmentation approximation method

Fragment analyses of vibrational circular dichroic response of dipeptides were carried out recently [Choi and Cho, J. Chem. Phys. 120, 4383 (2004)]. In the present paper, by using a minimal size unit peptide containing two chiral carbons covalently bonded to the peptide group, a generalized fragmentation approximation method is discussed and applied to the calculations of infrared-absorption and vibrational circular dichroism (VCD) intensities of amide I vibrations in various secondary structure polypeptides. Unlike the dipole strength determining IR-absorption intensity, the rotational strength is largely determined by the cross terms that are given by the inner product between the transition electric dipole and the transition magnetic dipole of two different peptides. This explains why the signs and magnitudes of VCD peaks are far more sensitive to the relative orientation and distance between different peptide bonds in a given protein. In order to test the validity of fragmentation approximation, three different segments in a globular protein ubiquitin, i.e., right-handed alpha-helix, beta-sheet, and beta-turn regions, were chosen for density-functional theory (DFT) calculations of amide I vibrational properties and the numerically simulated IR-absorption and VCD spectra by using the fragmentation method are directly compared with DFT results. It is believed that the fragmentation approximation method will be of use in numerically simulating vibrational spectra of proteins in solutions.     

Ab initio study of the one- and two-photon circular dichroism of R-(+)-3-methyl-cyclopentanone

One- and two-photon circular dichroism spectra of R-(+)-3-methyl-cyclopentanone, a system that has been the subject of recent experimental studies of (2+1) resonance-enhanced multiphoton ionization circular dichroism, have been calculated with an origin-invariant density functional theory approximation in the region of the lowest electronic excited states, both for the gas phase and for a selection of solvents. A polarizable continuum model is used in the calculations performed on the solvated system. Two low-lying conformers are analyzed, and a comparison of the intensities and characteristic features is made with the corresponding two-photon absorption for each species, also for the Boltzmann-averaged spectra. The effect of the choice of geometry, basis set, and exchange-correlation functional is carefully analyzed. It is found that a density functional theory approach using the Coulomb attenuating method variant of Becke's three-parameter exchange and the Lee-Yang-Parr correlation functionals with correlation-consistent basis sets of double-zeta quality can reproduce the experimental electronic circular dichroism spectra very well. The features appearing in experiment are characterized in terms of molecular excitations, and the differences in the response of each state in the one- and two-photon processes are highlighted.     

Structural and conformational properties and intramolecular hydrogen bonding of (methylenecyclopropyl)methanol, as studied by microwave spectroscopy and quantum chemical calculations

The microwave spectra of (methylenecyclopropyl)methanol (H(2)C=C(3)H(3)CH(2)OH) and one deuterated species (H(2)C=C(3)H(3)CH(2)OD) have been investigated in the 20-80 GHz spectral range. Accurate spectral measurements have been performed in the 40-80 GHz spectral interval. The spectra of two rotameric forms, denoted conformer I and conformer IX, have been assigned. Both these rotamers are stabilized by intramolecular hydrogen bonds formed between the hydrogen atom of the hydroxyl group and the pseudo-pi electrons on the outside of the cyclopropyl ring, the so-called "banana bonds". The carbon-carbon bond lengths in the ring are rather different. The bonds adjacent to the methylene group (H(2)C=) are approximately 7 pm shorter that the carbon-carbon bond opposite to this group. It is found from relative intensity measurements of microwave transitions that conformer IX, in which the hydrogen bond is formed with the banana bonds of the long carbon-carbon bond, is 0.4(3) kJ/mol more stable than conformer I, where the hydrogen bond is formed with the pseudo-pi electrons belonging to the shortest carbon-carbon bond of the ring. The microwave study has been augmented by quantum chemical calculations at the MP2/6-311++G, G3 and B3LYP/6-311++G levels of theory.     

First-principles T-matrix calculations of double-ionization energy spectra of atoms and molecules

Strong electron correlation plays an important role in the determination of double ionization energy, which is required for removing or adding two electrons, particularly in small-sized systems. Starting from the state-of-the-art GW approximation, we evaluate the particle-particle ladder diagrams up to the infinite order by solving the Bethe-Salpeter equation of the T-matrix theory to calculate the double-ionization energy spectra of atoms and molecules (Be, Mg, Ca, Ne, Ar, Kr, CO, C(2)H(2), Li(2), Na(2), and K(2)) from first principles. The ladder diagrams up to the infinite order are significant to calculations of double-ionization energy spectra. The present results are in good agreement with available experimental data as well as the previous calculations using, e.g., the configuration-interaction method.     

1H and 13C NMR chemical shifts and spin-spin coupling constants in trans- and cis-decalins

The NMR parameters characterizing the spectra of trans- and cis-decalins were determined from theoretical calculations and experimental spectra. The calculated values of the shielding constants are in good agreement with the measured chemical shifts, with a small but noticeable difference in accuracy for the bridgehead atoms. Of all the spin-spin coupling constants, only most of (1)J(C,C) and (1)J(C,H) values could be extracted from the spectra, and the corresponding computed values are in good agreement with experiment. It appears that the applied density functional theory (DFT) approach overestimates slightly the J(C,C) coupling and underestimates the differences between one-bond (1)J(C,H) coupling constants. For all these constants [J(C,C), J(C,H) and J(H,H)] through one to three bonds, which could not be obtained experimentally, the predicted values are in good agreement with the general rules relating spin-spin coupling to the number and spatial arrangement of the intervening bonds.     

Computer-aided spectral assignment in nuclear magnetic resonance spectroscopy

The proton-carbon correlation spectra, HMBC (heteronuclear multiple bond correlation) and HMQC (heteronuclear multiple quantum correlation), respectively, provide direct and remote connectivity information with high sensitivity. Their combination enables carbon-carbon proximity relationships to be deduced, which are formally identical to those produced by a fictitious INADEQUATE-2D experiment, where correlations would be established exclusively between atoms linked by one or two bonds. The CASA program uses these relationships, as well as DEPT spectra and elementary chemical-shift considerations to assign the ¹³C spectrum of a compound if its structure is known or assumed. If the structure conflicts with the experimental data, no assignment is produced. The CASA program serves as an aid to either spectral assignment or structural elucidation.     

Vibrational spectroscopy and quantum chemical studies of 1,6,7,12,13,18-hexaazatrinaphthylene and related compounds

The spectral properties of 1,6,7,12,13,18-hexaazatrinaphthylene (HATN) and a number of related compounds are modeled using density functional theory, B3LYP. The calculations predict the frequencies with mean absolute deviation of 6 cm(-1) and there is little improvement on going to basis sets larger than 6-31 G(d). The substituent effects on the observed spectra are modeled effectively in both frequency shifts and relative intensities. The electronic properties may be predicted using TD-DFT and these are in very good agreement, in terms of transition energies and intensities, with the experimental data.     

Structural, electronic, and optical properties of CaCO3 aragonite

Density functional theory ab initio calculations of the structural parameters, electronic structure, carriers effective masses, and optical absorption of the CaCO₃ aragonite polymorph were performed within the local density and generalized gradient approximations, local density approximation (LDA) and generalized gradient approximation (GGA) respectively. A good agreement between the calculated lattice parameters and experimental results was obtained. Both the LDA and GGA results for CaCO₃ aragonite exhibit very close indirect and direct energy gaps, and the computed effective masses are heavy and anisotropic. Two optical absorption regimes related to distinct electronic transitions are predicted by the calculations.