Volume 102, Issue 24, 22 June 1995
Index of content:

Electron paramagnetic resonance lineshape analysis of the photoexcited triplet state of C_{ 60 } in frozen solution. Exchange narrowing and dynamic Jahn–Teller effect
View Description Hide DescriptionThe EPR lineshape of the photoinduced triplet state of C_{60} in frozen toluene solution was studied by pulsed EPRspectroscopy. Lineshape calculations of the triplet spectra were performed including dynamical exchange effects. The observed spectra in a glassy matrix are compatible with zero field splitting parameters ‖D _{1}‖ = 0.0114 cm^{−1} (12.2 mT) and ‖E _{1}‖ = 0.0005 cm^{−1} (0.5 mT). The temperature dependence of the powder lineshape was simulated using a dynamical exchange model, where the triplet principal axis jumps between all equivalent sites allowed by the D _{5d } symmetry for the lowest excited triplet state. The determined exchange rate turned out to be only weakly temperature dependent and suggests that the dynamic process is due to tunneling between Jahn–Teller distorted states rather than to real molecular jumps. In addition we have observed a different triplet state with zero field splitting parameters ‖D _{2}‖ = 0.0100 cm^{−1} (10.6 mT), ‖E _{2}‖ = 0.0015 cm^{−1} (1.6 mT) after annealing of the matrix. We attribute this to a C_{60} dimer or, alternatively, to crystal field effects.

Temperature dependence of the ^{14}N quadrupole coupling constant of isocyanomethane
View Description Hide DescriptionWe report NMRrelaxation timemeasurements of the ^{14}N quadrupole coupling parameter χ_{N} of isocyanomethane as a function of temperature in the solution‐state mixture of 6.4 mol % H_{3}CN ≡ C, 33.5 mol% d _{6}‐ethylene glycol, and 60.1 mol % d _{6}‐ethanol. We obtain values for the molecular correlation times using the T _{2}/T _{1} method, which is a function only of the spectral density functions. The results demonstrate a significant, and approximately quadratic, temperature dependence of χ_{N}, which shows a minimum value of about 55 kHz at 222 K; the largest value measured was 125 kHz at 161 K. The results are compared with, and discussed within the context of, more traditional methods. The value of χ_{N} for neat isocyanomethane in the solid phase was measured to be 26.3 kHz.

Study on the hydration structure of L‐xylo and D‐arabo ascorbic acid solutions by time domain reflectometry
View Description Hide DescriptionThe hydration structure of L‐xylo and D‐arabo ascorbic acids in aqueous solutions were investigated by a dielectric relaxationmeasurement over a wide frequency range from 10 MHz to 10 GHz from a standpoint on the difference of biological activity at 25 °C. In order to clarify the hydration structure the concentration dependence of dielectric relaxation was investigated not only in aqueous solution but in water–ethanol mixtures. Two kinds of dielectric relaxation processes were observed in each isomerism solution. The low frequency process is assigned to cooperative motions of ascorbic acid molecules and hydrated water. The high frequency process is assigned to reorientational motions of bulk water. From the results of the dehydration process out of the ascorbic acid surface by ethanol it is concluded that the amount of hydrated water of the L‐xylo ascorbic acid is more than that of the D‐arabo ascorbic acid.

Photoelectron spectroscopy of size‐selected transition metal clusters: Fe^{−} _{ n }, n=3–24
View Description Hide DescriptionA higher resolution magnetic bottle photoelectron spectrometer for the study of the electronic structure of size‐selected metalclusters is presented. The initial study on Fe^{−} _{ n } (n=3–24) is reported at a photon energy of 3.49 eV. The photoelectron spectra of these clusters exhibit sharp features throughout the size range. The spectra for Fe^{−} _{3–8} show large size dependence with many resolved features. The spectra for Fe^{−} _{9–15} exhibit some similarity with each other, all with a rather sharp feature near the threshold. An abrupt spectral change occurs at Fe^{−} _{16}, then again at Fe^{−} _{19} and Fe^{−} _{23}. These photoelectron spectral changes coincide remarkably with changes of the cluster reactivity with H_{2}. Extended Hückel molecular orbital (EHMO) calculations are performed for all the clusters to aid the spectral interpretations. The calculations yield surprisingly good agreement with the experiment for clusters beyond Fe_{9} when body‐centered cubic (bcc) structures are assumed for Fe_{9–15} and a similarly close‐packed structure with a bcc Fe_{15} core for the larger clusters. The EHMO calculations allow a systematic interpretation of the sharp photoelectron spectral features in Fe^{−} _{9–15} and reproduced the abrupt spectral change taking place from Fe^{−} _{15} to Fe^{−} _{16}. Most importantly, the reactivity changes of the clusters with H_{2} are successfully explained based on the detailed electronic structures of the clusters, as revealed from the photoelectron spectroscopy (PES) spectra and the theoretical calculations. The calculations also correctly predict the existence of magnetism in these clusters and yield reasonable values for the clustermagnetic moments.

High frequency (140 GHz) dynamic nuclear polarization: Polarization transfer to a solute in frozen aqueous solution
View Description Hide DescriptionDynamic nuclear polarization (DNP) transfers the large polarization of unpaired electrons to nuclei and thus significantly enhances the signal strength in nuclear magnetic resonance(NMR)spectroscopy. High frequency/field (140 GHz/5 T) DNP has been implemented in solid state NMR experiments using a nitroxide radical as the paramagnetic polarizing agent in a water:glycerol frozen solution. The ^{1}H and ^{13}C NMR signal strengths of both the solvent and an amino acid solute have been enhanced by a factor of 185, which represents a reduction of ≳10^{2} in sample size requirements or ≳10^{4} in signal acquisition time.

Photoionization spectroscopy of the In–N_{2} van der Waals complex
View Description Hide DescriptionA vibrationally resolved electronic spectrum is observed for the metal atom van der Waals complex In–N_{2}. Two electronic band systems are detected with mass resolved two‐color photoionizationspectroscopy. A lower energy system is observed slightly to the blue of the In ( ^{2} D←P) atomic asymptote. It is characterized by a progression in the In–N_{2} stretching mode with a frequency of ω^{’} _{ e }=76.7 cm^{−1}. The higher energy system is slightly to the blue of the In (^{4} P←^{2} P) asymptote. It also exhibits a progression in the In–N_{2} stretch with a frequency of ω^{’} _{ e }=87.7 cm^{−1}. Extrapolation of the vibrational progressions leads to determination of the excited statedissociation energies. Energetic cycles based on the electronic transition energies, excited statedissociation energies, and atomic asymptotes lead to a determination of the ground statedissociation energy of D ^{‘} _{0}=1519 cm^{−1} (0.188 eV). A single‐photon photoionization experiment determines the ionization potential to be 43 372 cm^{−1} (5.377 eV). This IP value, together with the atomic IP and the ground state neutral dissociation energy, yields a dissociation energy of D ^{‘} _{0}=4817 cm^{−1} (0.597 eV) for the In^{+}–N_{2} ion–molecule complex.

Far‐infrared spectra and two‐dimensional potential energy surface for the ring‐bending and ring‐twisting vibrations of 5,6‐dihydro‐4H‐thiopyran
View Description Hide Description5,6‐Dihydro‐4H‐thiopyran has been synthesized and its far‐infrared spectrum has been recorded. Eleven ring‐bending bands originating at 120.7 cm^{−1} and four ring‐twisting bands originating at 274.5 cm^{−1} were observed. Twelve sum and difference bands in the 383–397 and 148–166 cm^{−1} regions were also observed and these facilitated the construction of a detailed energy map including numerous excited vibrational states of the two coupled vibrations. The two‐dimensional potential energy surface, which satisfactorily fits the observed data, was determined to be V=9.48 ×10^{4} x ^{4}−4.13×10^{4} x ^{2}+1.37×10^{4}τ^{4}−1.82×10^{4}τ^{2}+1.10 ×10^{5} x ^{2}τ^{2}, where x and τ are the bending and twisting coordinates, respectively. The minima on the potential energy surface correspond to twisting angles of ±48° (half‐chair conformation). The lowest energy bent (boat) conformation corresponds to a saddle point 1500 cm^{−1} above the twisted conformation on the potential energy surfaces, and the barrier to planarity was estimated to be 6000 cm^{−1}. Both of these values have large uncertainties since the vibrational data only extend to 800 cm^{−1} above the potential surface minimum. The relatively low bending energy and high barrier to planarity can both be explained by the low force constant for the C–S–C angle bending.

Main factors influencing the recoil energy distribution in the products of three‐atom reactions governed by long‐range forces and proceeding through long‐lived complexes
View Description Hide DescriptionWe propose a simplified version of the classical statistical theory of three‐atom reactions governed by long‐range forces. This version is based on a partial treatment of total angular momentumconservation. We focus the developments on the determination of the recoil energy distribution of reactions performed in supersonic crossed‐beam experiment. This distribution function is directly linked to the maximum values of the moduli of rotational and orbital angular momenta of the products consistent with their recoil energy and the total angular momentum disposal. Due to the simplicity of the present version, we can pinpoint the main factors which play a role in statistical reaction dynamics. It is shown that the shape of the translational energy distribution can be estimated essentially in terms of two mechanical parameters; the first one represents the available fraction of maximum angular momentum disposal in the products and the second gives a measure of the fraction of total angular momentum due to the rotation of the newly formed bond AB. Comparisons with exact calculations and experimental results are presented. They show that in addition to being able to determine propensity rules, our simple approach can even provide satisfying quantitative results.

Alignment probing of Rydberg states by stimulated emission
View Description Hide DescriptionThe possibility of probing the collisions of aligned Rydberg atoms by stimulated emission is assessed with studies of a polarized state and a new measurement of a collisional alignment effect in atomic Ca. The stimulated emission method uses a laser to dump the desired state to a lower level which subsequently fluoresces. The technique can be used to obtain populations and polarization dependent information. First, the method is tested by applying it to an aligned Ca(4s17d ^{1} D _{2}) state. Alignment curves are measured when the initial state is prepared with both parallel and perpendicular relative polarizations. The experimentally observed alignment compares well with that derived from theoretical considerations of a saturated stimulated transition. Second, a two‐vector collisional alignment experiment (initial state and relative velocity vector) is performed to study the energy transfer process Ca(4s7d ^{1} D _{2})+He→Ca(4s6f ^{1} F _{3})+He+ΔE=17.7 cm^{−1}, and alignment effects are measured by both stimulated emission and conventional direct fluorescence detection. A preference for the ‖m‖=1 and 2 initial states is observed in the relative cross sections. Essentially identical data are obtained with the two detection methods when elliptically polarized light is used for the stimulated emission detection method. The stimulated emission technique can provide alignment and population information of the final states, making it an excellent new tool for both three‐vector correlation experiments and state‐to‐state Rydberg transitions.

Orbital alignment cross sections by stimulated emission probing: The state‐to‐state Ca Rydberg process Ca(4s17d ^{1} D _{2})+Xe→Ca(4s18p ^{1} P _{1})+Xe
View Description Hide DescriptionThe initial state alignment effect vs relative velocity is measured for a state‐to‐state Ca Rydberg collisional energy transfer process. The stimulated emission detection method is used to determine the alignment effect for the n,l‐changing transition: Ca(4s17d ^{1} D _{2})+Xe→Ca(4s18p ^{1} P _{1}) +Xe+ΔE=−1.7 cm^{−1}. The rate of electronic energy transfer in this state‐changing collision is observed to vary with the direction of the Rydberg electron charge cloud relative to the collision axis. Both the expected cos(4β) and cos(2β) dependencies are observed. The alignment data are analyzed to obtain the relative cross sections for the individual Ca(^{1} D _{2}) magnetic sublevels. The values of the m‐sublevel cross sections σ^{0}:σ^{‖1‖}:σ^{‖2‖} are 1.13±0.02:1.11±0.02:0.83±0.02. Qualitative interpretations of the relative cross sections in terms of both molecular (van der Waals) Born–Oppenheimer potentials and the impulse approximation are presented.

Molecular dynamics of β‐carotene in solution by resonance enhanced optical Kerr effect
View Description Hide DescriptionThe orientational dynamics of β‐carotene in n‐alkane solutions is investigated by resonance enhanced optical Kerr effect. By use of this spectroscopic technique, it is possible to selectively investigate the relaxation of a probe molecule at a concentration level low enough to allow the observation of the averaged single‐molecule dynamics. For delay times longer than ∼20 ps all solutions show a single exponential decay, with a time constant depending on the viscosity, that is ascribed to the β‐carotene orientational relaxation. The dependence on viscosity of the measuredrelaxation times is compared with the predictions of different models. The purely hydrodynamic theories overestimate, by far, the solute effective volume and hence its orientational relaxation time; a much better agreement is obtained from two quasihydrodynamic models.

State‐selected vibrational relaxation rates for highly vibrationally excited oxygen molecules
View Description Hide DescriptionThe state‐selected vibrational relaxation rates in O_{2}+O_{2} collisions, with one O_{2} molecule in a highly vibrationally excited state, have been calculated from first principles. The vibrationally close‐coupled, rotationally infinite order sudden approximation has been used to treat the collision dynamics and a potential energy surface based on high quality ab initio calculations, which include the variation of the O_{2} vibrational coordinates, has been developed. The calculated relaxation rates are in good agreement with those obtained from experiment for 8≤v<26 but fail to reproduce the sharp increase observed experimentally for v≥26 indicating the onset of a new vibrational relaxation mechanism.

Light intensity effects on diffusion‐influenced fluorescence quenching in a hard‐sphere liquid: Molecular dynamics simulation and the many‐body Smoluchowski equation approach
View Description Hide DescriptionMolecular dynamics simulation of a model fluorescence quenching reaction A*+B→B (A* is fluorophore and B is quencher molecule) in a hard‐sphere liquid where the fluorophore is excited for the first time or reexcited shortly after a bimolecular quenching process is carried out. The effects of light intensity on the temporal behavior of the fluorescence quenching kinetics is obtained by a summed form of an appropriate convolution integral using the simulation data. The convoluted results are compared with the recently developed general theoretical framework for the quenching kinetics where the exciting light pulse has a short but finite duration. The theory is based on hierarchy of phenomenological kinetic equations involving reactant molecule distribution functions. The alone effect of potential of mean force is examined and the radiation boundary condition is considered. Improvements over the simplest version of the Smoluchowski theory has been found. Considering the error introduced due to truncation of repeated excitation of A at the level of first repeated excitation (after a bimolecular process) the agreement between theory and simulation is excellent under certain limiting time profile of the exciting light pulse.

Is slow thermal isomerization in viscous solvents understandable with the idea of frequency dependent friction?
View Description Hide DescriptionThermal Z/Eisomerization of substituted azobenzenes and N‐benzylideneanilines takes place slowly after fast photoinduced E/Zisomerization. Its rate constantk _{obs} is smaller than about 10^{3} s^{−1} because of a high reaction barrier of about 50 kJ/mol. The pressure dependence of k _{obs} measured in solvents as glycerol triacetate can well be understood in the framework of the transition state theory(TST) at low pressures. At high pressures, however, k _{obs} begins to steeply decrease as the pressure increases, to be more exact, as the solventviscosity η increases with the pressure, and the reaction enters the non‐TST regime. Since the η‐induced decrease of k _{obs} at high pressures is slower than η^{−1}, it cannot be described by the Kramers theory which regards the reaction as the barrier surmounting by Brownian motions regulated by frequency independent friction. Next, it was adjusted to the Grote–Hynes theory incorporating the idea of frequency dependent friction. The situation of k _{obs} mentioned earlier enabled us to derive, without adjustable parameters, the correlation time τ_{sc} among random forces for friction due to solvent microscopic motions in the generalized Langevin equation on which the theory is based. At η∼10^{7} Pa s, we obtained τ_{sc}∼1 ms. It is too long to justify the theory, since such a long‐time correlation cannot be realized among random forces exerting on the isomerizing moiety with an angstrom dimension. It will also be shown that τ_{sc} must be so long unphysically as to be at least much longer than 1 ps even if k _{obs} at low pressures is adjusted to the theory.

Rate expressions for excitation transfer. III. An ab initio study of electronic factors in excitation transfer and exciton resonance interactions
View Description Hide DescriptionA detailed theory for electronic aspects of electronic excitation (energy) transfer (EET) for sandwich dimers was derived in paper II of this series [J. Chem. Phys. 101, 10 521 (1994)]. In II, the electronic transfer matrix element for EET was evaluated, then simplified to various levels of approximation. The results of ab initio molecular orbital calculations on an ethene sandwich dimer are reported here in order to test and quantify the theory of II. The calculations were undertaken using a STO‐6G basis set and localized molecular orbitals, with separations of 4, 5, and 6 Å between the molecules. It is demonstrated that the important electronic factors contributing to EET are the Coulombic interaction (for direct singlet–singlet transfer) and, for both singlet–singlet and triplet–triplet EET, orbital overlap‐dependent interactions. The dominant orbital overlap‐dependent terms arise from through‐configuration interaction, which involves successive one‐electron transfers mediated via bridging ionic configurations, first presented in II. The results confirm that the Dexter‐type exchange interaction is small in comparison.

Orbital‐invariant second‐order many‐body perturbation theory on parallel computers: An approach for large molecules
View Description Hide DescriptionThe equations for the second‐order many‐body perturbation theory [MBPT(2)] energy are derived in an orbital‐invariant representation, analogous to that obtained with the method of self‐consistent electron pairs of Meyer. This formulation is well suited to take advantage of the localized nature of interactions in large chemical systems in order to reduce the computational effort required to study them. This formulation of the MBPT(2) method also lends itself to implementation on parallel computers. We describe a scalable implementation in which the key data are distributed across the parallel computer rather than being replicated. Portability to both shared‐ and distributed‐memory computer architectures is provided through the use of a subroutine library implementing a ‘‘global array’’ programming model. We demonstrate that this approach is scalable even for relatively small chemical systems.

Linear dependencies among basis set products and near locality of some nonlocal operators
View Description Hide DescriptionLinear dependencies among basis‐set products and the decomposition of the matrices of some one‐electron operators into local and nonlocal components have been investigated for several common quantum chemical basis sets and for a harmonic oscillator basis. For the first ten atoms and some simple diatomics, the kinetic energy, Fock, and density matrices were investigated. It was found that, for the basis sets used, these operators are all nearly local, in the sense that their matrices could be reproduced as the matrices of simple multiplicative‐function operators, even though there are significant numbers of linear dependencies among the products of the basis functions. SCF eigenfunctions for these systems were found to have no linear dependencies among products. Basis sets of one‐dimensional harmonic oscillator eigenfunctions were found to give a kinetic energy matrix that is not nearly local.

Analytical energy gradients and geometry optimization in the divide‐and‐conquer method for large molecules
View Description Hide DescriptionBased on the divide‐and‐conquer method in the density‐functional theory, an efficient approach is developed to compute analytically the energy gradients with respect to the nuclear coordinates. Tests performed show that both energy gradients and optimized molecular geometry converge to the corresponding results of the Kohn–Sham method when the nearest neighbor contributions are increased.

Effective and intermediate Hamiltonians obtained by similarity transformations
View Description Hide DescriptionA simple similarity transformation is used to derive equations for effective and intermediate Hamiltonians in a lucid way. Effective and intermediate Hamiltonians based on the wave operator formalism provide only a subset of all eigenvalues while the similarity transform technique divides the eigenvalue problem into two subproblems that can be solved separately. This means that the complete spectrum of the Hamiltonian remains well defined and this proves to be advantageous in the formal analysis and may be useful in many applications. Moreover both left and right hand eigenvectors of the transformed Hamiltonian can be obtained and this allows a convenient evaluation of properties. Rayleigh–Schrödinger and Brillouin–Wigner perturbation expansions of the intermediate Hamiltonians are discussed and a comparison is made of the various possible schemes.

Protonated hydrochlorous acid (HOClH^{+}): Molecular structure, vibrational frequencies, and proton affinity
View Description Hide DescriptionProtonated hydrochlorous acid (HOClH^{+}) has been examined theoretically. Equilibrium geometries have been optimized and harmonic vibrational frequencies obtained for each of the parent and protonated structures at various levels of theory employing second‐order Mo/ller–Plesset perturbation interaction theory (MP2), singles and doubles excitation configuration interactiontheory (CISD), and coupled‐cluster theory (CCSD). Our study has found that protonation of the oxygen of HOCl is favored over protonation at the chlorine site. Protonation of the oxygen leads to a pyramidal structure of C _{ s } symmetry. There is a planar C _{ s } structure which is the inversion transition state. The inversion barrier is 3.2 kcal mol^{−1}. The proton affinity of hypochlorous acid, HOCl, is found to be 153.1 kcal mol^{−1} at 0 K.