ALBERT

Notes: The lowest few gerade Rydberg states of C2H2, C2HD, and C2D2 have been observed as two-photon resonances in the multiphoton ionization (MPI) spectra of these isotopic species. Analysis of the two-photon excitation spectra of these molecules, and of the accompanying MPI-photoelectron spectra for C2H2, enable characterization of the 1Σ+g and 1Δg Rydberg states arising from the electronic promotion π3u3pπ1u←π4u and reveal extensive vibronic interaction in this energy region. A search was made for higher gerade Rydberg states at energies up to 90 000 cm−1. Weak, diffuse signal is observed for the 4p complex. No ionization signal was observed in the energy range where the np (n=5–7) Rydberg states are expected to appear. This is taken as evidence for opening of a facile nonradiative decay channel at high excitation energies.

Notes: A pure rotational spectrum of SnH4 has been studied in the ν4 state by infared–microwave double resonance with a tunable diode laser. Twenty-seven double resonance signals have been observed for the 11 JJ–JJ type rotational transitions of the 5 Sn isotopic species. An infrared spectrum of the Coriolis-coupled ν2 and ν4 bands has been measured using a diode laser with Doppler-limited resolution. The ground-state scalar constants, B0, D0, and H0, have been determined by taking combination differences in the "forbidden'' ν2 band, which borrows intensity through a strong Coriolis interaction. The observed microwave and infrared data have been analyzed simultaneously by including the Coriolis interaction explicitly. More than 20 molecular constants have been determined accurately for the ν2/ν4 dyad of seven major isotopic species.

Notes: We report the first optical dephasing study of an inorganic impurity system possessing sharp, low frequency mode structure in its ground and excited state spectra. The total dephasing times T2 and population relaxation times T1 for the 1T2g and 1Eg d9s excited states of NaF:Cu+ are determined at a series of temperatures between 1.8 and 296 K. The T2 values are determined by extracting the Lorentzian components from one- and two-photon excitation line shapes. The T1 values caused by nonradiative decay rates are obtained by detecting very low quantum yield emission from the fast-relaxing excited states and applying the formula tnr=Qtr, where t's are radiative and nonradiative lifetimes and Q is the quantum yield. T1(1T2g)=4.6 ps and T1(1Eg )=2.0 ps at 8 K. Significantly, these are very similar to the T1 values calculated from lowest temperature Lorentzian linewidths by the relationship 1/T2=1/T′2 +1/(2T1). The T1 values stay approximately constant over the temperature range 1.8–45 K, while the linewidths rapidly increase indicating that pure dephasing dominates. Using ground and excited state information on the low frequency modes, we test optical Redfield theory and the nonperturbative harmonic theory for pure dephasing by pseudolocal phonons against data for this system which displays strongly anhamonic progressions. The nonperturbative theory fits the line broadening data to higher temperatures than optical Redfield theory for the least anharmonic excited state potential, 1T2g. Both theories underpredict the broadening with temperature of the extremely anharmonic 1Eg state. A simple anharmonic theory including scattering to overtone levels also fails to predict the observed linewidth temperature behavior, although it is demonstrated these processes should be occurring.

Notes: The relation between the fractal geometry of reactant trajectories and the rate of diffusion-controlled reactions is considered. A new possible mechanism for the effect of an external electric field on the rate of reactions is proposed on the basis of this consideration. The proposed mechanism predicts an increase in the rate constant with increasing electric field strength. Experiments in the literature which may be related with the proposed mechanism are surveyed. Finally directions in future extensions of the present study are suggested.

Notes: Collisional energy transfer between vibrationally highly excited CS2 molecules and 22 different collision partners has been measured directly by time resolved UV absorption spectroscopy after laser excitation. Experiments at temperatures up to 1200 K were performed in a reactor heated by cw CO2 laser radiation. Experiments at higher temperatures (up to 2400 K) with the collision partner Ar were done with pulsed laser excitation behind reflected shock waves. At all temperatures a marked dependence of the average energy 〈ΔE〉 transferred per collision on the excitation energy is noted. Also the molecular size of the deactivating molecule has a large influence on 〈ΔE〉. The temperature coefficients of 〈ΔE〉∝Tn are small in all cases. They vary between n=−0.6 for n-heptane and n=+1.0 for Ne.

Notes: A theoretical model for excitation trapping in poly-(N-vinylcarbazole) in the presence of excimer dissociation is presented. We assume a time-dependent trapping coefficient in the form k(t)=b+ct−1/2, which reflects both the diagonal disorder and quasi-one-dimensionality of carbazole hopping sites [Kauffmann et al., J. Chem. Phys. 85, 3566 (1986)]. The equations obtained for monomer and excimer fluorescence decays are given in terms of Laplace transforms. Respective equations for fluorescence decays have been obtained also in the case when excimer formation is time dependent. Equations for monomer and excimer quantum yields are also given.

Notes: We use random tempering formulas to create a basis set of explicitly correlated Gaussian-type geminals for the calculation of the second-order energy of lithium hydride. Like our earlier calculations on the beryllium atom, this technique matches results obtained by the full optimization of all nonlinear parameters but requires considerably less computational effort.

Notes: The potential energy curves for the 3Σ+u states of He2 which correspond to the interactions of He 1s2 1S with He 1s2s 3S and He 1s2s 3S with He 1s2s 3S are obtained from second-order configuration interaction calculations which themselves are based on a multiconfiguration self-consistent field calculation. The basis set includes functions obtained by optimizing the dipole, quadrupole, and octapole polarizability of the 3S atom and the dipole and quadrupole polarizabilities of the ground 1S atom. For the lowest 3Σ+u state we estimate the dissociation energy to be De =15 740±60 cm−1; we agree to within about 4 cm−1 with the observed vibrational energy level differences, and agree nicely with the position and depth of the short-range minimum deduced from spectra. We also obtain a potential hump at intermediate separations whose main features agree very well with those obtained from scattering experiments. The potential curve exhibits a van der Waals minimum about 3.82 cm−1 deep at about 11.6 a0. The excited 3Σ+u state has a deep metastable well at small internuclear separations as well as a well in the intermediate region which is about 3430 cm−1 deep.

Notes: An EPR study with optical detection is reported of the luminescent triplet state of the VO3−4 ion in single crystals of YVO4 and YPO4 doped with VO3−4 . The experiments in an external magnetic field and in zero field can be fitted by a spin Hamiltonian for an orbitally nondegenerate triplet state. Signals originate from four symmetry-related species; the directions of the principal axes of the four ZFS tensors indicate a lowering of symmetry of the VO3−4 ion in accordance with a static Jahn–Teller effect by which the tetrahedron tends to a trigonal pyramid on excitation. For YVO4 the ZFS parameters are: ||D||=0.58±0.04, ||E||=13.72±0.04 GHz, and for YPO4: ||D||=1.12±0.04, ||E||=12.98±0.04 GHz (z axis(parallel)V–O bond and y⊥ac or bc crystallographic plane). For the mixed crystal the lifetimes of the individual spin levels were determined: τx=0.35, τy=3.3, τz=1.4 ms.

Notes: RISM-polaron theory and simulation results of the primitive hard sphere model for an excess electron in simple fluids are used to interpret the recent path integral quantum Monte Carlo studies of an electron in supercritical helium and in xenon by Coker, Berne, and Thirumalai. It is shown that the different behaviors of the excess electron in these two different fluids are due primarily to differences in excluded volume effects. For xenon, due to the nature of the electron–solvent pseudopotential, this volume is relatively small and the excess electron remains extended for all fluid densities. In contrast, for helium, the random excluded volume is high leading to self-trapping or localization of the electron.

Notes: In previous work [J. Keizer, J. Chem. Phys. 82, 2751 (1985)] we used statistical nonequilibrium thermodynamics to predict a non-Nernstian component to the electromotive force (EMF) for half-reactions involving reactants at nonequilibrium steady states. In this paper we present a simple theory for calculating the nonequilibrium component of the EMF based on the elementary transport processes occurring in a continuously stirred tank reactor (CSTR). The calculations utilize the density–density correlation function, which is obtained from the statistical theory of nonequilibrium thermodynamics. This gives rise to an expression for the second partial derivatives of the generalized entropy, or sigma function, which is used to calculate generalized chemical potentials. The generalized chemical potentials are related to the EMF through a generalization of the Nernst equation. The calculations presented here depend on the residence time in the CSTR, reaction rate constants, feed line concentrations in the CSTR, and the diffusion constants of reactants and products. A characteristic diffusion length is used to represent the length scale below which turbulent mixing effects are not important. Calculations with the theory are carried out for several different reaction mechanisms, including A+B(arrow-right-and-left)C; A+B(arrow-right-and-left)C, D+E(arrow-right-and-left)B; A+B(arrow-right-and-left)2B; and A+B→C+D, A+D→C+E. Values of the nonequilibrium EMF depend on the mechanism as well as all of the transport parameters cited above. For a plausible choice of the diffusion length, corrections to the Nernst formula can be as large as 10–15 mV. Specific calculations for the reaction of Fe2+ with S2O2−8 are shown in the preceding paper to agree with experimental measurements on this system in a CSTR.

Notes: There has recently been a substantial increase in information on low-temperature phase transformations of ice and ice-like substances along with a rapid increse in molecular level information on the defect structure and activity of this subset of hydrogen-bonded solids. These data, some quantitative and some qualitative in nature, are examined from the viewpoint that the phase transformation mechanism depends on the availability of mobile orientational defects (Bjerrum L defects) within the new phase. Some of the data that seem particularly apt for establishing any dependence of the phase transformation on the presence of mobile L defects within the new phase, reflect a strong correlation between transformation rates/temperature and the availability of mobile L defects. One possible inference is that the integrity of a growing phase can be maintained only if defects responsible for orientational mobility at the interface can ultimately achieve equilibrium through recombination. The implications of such a dependence on mobile L defects are examined for a variety of systems. For example, one implication is that, at low temperature (〈160 K), the structure I clathrate hydrate of trimethylene oxide (TMO) should grow more rapidly than the structure II hydrate, since structure I hydrates are known to be relatively rich in mobile defects. New data for the clathrate hydrates of TMO are presented that confirm the preferential growth of the structure I hydrate from amorphous deposits containing water and TMO in ratios ranging from 7:1 to 〉17:1.

Notes: A theory on the attachment of very slow electrons to van der Waals clusters was developed on the basis of the electronic structure theory, and was applied to clarify the mechanism of the collisional electron transfer from a high-Rydberg atom to a CO2 cluster. The strong coupled electron–phonon model is found to afford a reasonable mechanism of the attachment. The equilibrium geometry of (CO2)N (2≤N≤13) clusters are determined and their vertical affinity levels are obtained by the DV-X α-transition state method. Using this information, as well as some plausible assumptions on the values of the coupling constants, the attachment cross section σ is evaluated as a function of the energy of the incident electron. The theory predicts the existence of the threshold cluster size for the attachment and a sharp decrease of σ with the energy, which are consistent with the experimental results.

Notes: The photodissociation of H2O in the first absorption band is studied from single rotational states of vibrationally excited water. A tunable IR laser is used to prepare single rotational states in the asymmetric stretch mode. The subsequent photodissociation at 193 nm favors product formation from these single prepared states. The formation of the OH product in different rotational, Λ-doublet, and spin states is analyzed for a series of initial rotational states of H2O. This is the first direct photodissociation studied on a state to state level. The product state distributions depend sensitively upon the prepared state in the parent molecule H2O and exhibit pronounced quantum structure. The experimental results are understood almost quantitatively in terms of theory. The photodissociation of water turns out to be a limiting case of a dissociation which is governed by transfer of parent motion to products. The experiment leads to a highly improved understanding for the selective population of Λ-doublet states.

Notes: Three-dimensional quantum mechanical calculations for vibrational predissociation of the Ne⋅⋅⋅Cl2 van der Waals complex are presented and compared with experiments. Lifetimes and final rotational state distributions were obtained for the two processes: (i) Ne⋅⋅⋅Cl2(X,v=1) →Ne+Cl2(X,v=0) and (ii) Ne⋅⋅⋅Cl2(B,v=11) →Ne +Cl2(B,v=10,9) where v denotes the vibrational quantum number of Cl2 and X and B specify electronic states of Ne⋅⋅⋅Cl2 which correlate with the X 1∑+0g and B 3∏+0u states of the free Cl2 molecule, respectively. The van der Waals interaction potential was taken to have the same form in the X and B states. At short distances, it is described by a sum of Morse pairwise potentials between the Ne atom and each of the Cl atoms, and between the neon atom and the center of mass of Cl2. At large distances the potential switches to an anisotropic van der Waals interaction with R−6 and R−8 dependence. The parameters were adjusted so that the T-shaped configuration the potential matched the one determined from scattering experiments. The initial quasibound state wave function of the complex was calculated variationally, while the final continuum wave functions were obtained by integration of the rotational close coupled Schrödinger equations. Finally, the lifetime and the final rotational state distribution were calculated using the Fermi golden rule. A line shape calculation verified the validity of the golden rule approximation for this system. The lifetimes obtained for the X and the B states differ by several orders of magnitude, the X state being the longest lived as observed experimentally. The calculated lifetimes and rotational distributions of the Cl2 fragments agree qualitatively with the experimental values. The rotational distribution is compared to that obtained from a decomposition of the initial quasibound state in terms of free rotor states.

Notes: Configuration interaction wave functions, potential energy curves, and dipole moment functions have been calculated for the four lowest 3Σ− and the three lowest 3Π states and 5Σ− states of NH. The electronic wave functions were constructed to give a balanced description of valence–Rydberg interactions. Two repulsive states have been identified as important photodissociation pathways. We present spectroscopic constants for the bound states and compare our results to other theoretical and to experimental work. The possible predissociation of the A 3Π state by the 1 5Σ− state is discussed.

Notes: Full configuration interaction ab initio studies of the 2p 3p 1P1 state of N+ are carried out in order to obtain the molecular electric field gradient at the nitrogen nucleus. A careful calibration of the basis set leads to a q value (−0.94 a.u.) obtained using a 〈13s9p5d2f/10s7p5d2f〉 basis set with an estimated accuracy of ±0.01 a.u. Combined with the experimental nuclear quadrupole coupling constant for this system, we predict the nuclear quadrupole moment of 14N to be Q(14N)=(20.7±0.4) mb.

Notes: The quadratic vibrational force field of propane has been obtained by scaling calculated ab initio force fields to fit well determined vibrational frequencies from the Raman and infrared spectra of gaseous propane-h8, propane-d8, 1,1,1,2,3,3,3-propane-d7, and the two rotational isomers of 1,1,2,2,3,3,3-propane-d7. Two different ab initio force fields were investigated, one using Hartree–Fock theory (HF/6-31G*) and the other including electron correlation effects by means of second-order Møller–Plesset perturbation theory (MP2/6-31G*). The scaled MP2/6-31G* force field gives a significantly better fit of the experimental frequencies, especially for those involving contributions from the CC stretching coordinate.

Notes: The self-diffusion coefficient, velocity autocorrelation function, and distribution of collision times for a two-dimensional nonoverlapping Lorentz gas were calculated using molecular dynamics simulation. The systems studied covered a range of densities, from a packing fraction (πNr2/L2) of 0.01 to 0.8. Self-diffusion coefficients were found to agree to all densities with kinetic theory predictions [A. Weijland and J. M. J. van Leeuwen, Physica 38, 35 (1968)] if the radial distribution function (rdf) was taken into account. The density dependence of the decay of the velocity autocorrelation function was qualitatively different from that predicted by kinetic theory. The distribution of collision times was nearly exponential for all but the highest density studied.

Notes: Based on a statistical thermodynamic theory, it has been predicted [J. Keizer, J. Chem. Phys. 82, 2751 (1985)] that at nonequilibrium steady states the electromotive force (EMF) of an electrochemical cell will differ from the local equilibrium value given by the Nernst equation. We describe here experiments designed to test this prediction for aqueous solutions of Fe2+ and Fe3+ in sulfate buffer. Using a continuously stirred tank reactor driven by a peristaltic pump, a feed solution containing Fe2+ and Fe3+ was mixed with a second feed solution containing the oxidant sodium peroxydisulfate Na2S2O8. The reaction leads to a steady nonequilibrium mixture, which at acidic pH in sulfate buffer is composed of Fe2+ and the ferric sulfate complexes FeSO+4 and Fe(SO4)−2. The EMF of this half-cell was measured vs a saturated calomel reference electrode as a function of residence time in the reactor. These potentials were compared to the Nernst potential calculated on the basis of the concentration ratio of Fe2+ to total Fe3+ at the steady states. The Nernst potential was reproducibly larger than the measured EMF by values that depended on the concentration ratio of Fe2+/Fe3+ in the feed solution and the residence time. The largest deviations were −1.8 mV, which occurred when the Fe2+/Fe3+ ratio was small and the residence time was about 40 s. We have ruled out streaming potentials, junction potentials, and incomplete mixing as the origin of this effect. We show that the dependence of the nonequilibrium portion of the EMF on feed concentrations and residence time is in good agreement with calculations based on methods that are described in the second paper in this series.

Notes: We have investigated how the rate of diffusion-controlled bulk recombination between electrons and positive ions is influenced by the mean free path of electrons. By use of the Monte Carlo method previously developed we have calculated the rate constant as a function of the mean free path. For mean free paths negligibly short compared with the Onsager length, the rate constant increases linearly with increasing mean free path, in agreement with the Debye theory of bulk ion recombination. However, as the mean free path further increases, the rate constant goes through a maximum and then decreases. The present result quantitatively explains recent experimental data on electron–ion recombination rate constants in liquid methane in which the mean free path of electrons is comparable with the Onsager length.

Notes: We outline a connection between recent mean-field theories of short-chain packing in micellar systems and earlier approaches developed for treating phase separation in polymer blends. These theories are easily unified on the basis of a common variational principle, thereby allowing a single route for deriving lateral pressures, local ordering, and thermodynamic properties. In this approach the search for conformational probability distribution functions is mapped into a constrained random walk problem, using the monomer propagator formalism first exploited by Edwards. As an application, the case of a compact (uniform density—"dry'') amphiphilic bilayer is studied in detail. We show that the "core'' free energy can be described via a single reduced variable relevant to both short and long chains: convenient scaling relations are proposed and discussed.

Notes: Desorption of an adatom from a solid surface through vibrational excitation by an IR laser is studied using a generalized master equation approach and the isomnesic approximation. A sharp transition from no desorption to almost instantaneous desorption is seen as the bound-to-continuum population transfer rate is varied. The critical rate increases linearly with laser power.

Notes: The electrical double layer is studied using a cluster perturbation theory which incorporates the effect of images and fluctuations due to ion correlations. These equations are solved numerically and ion density profiles are determined for a wide range of surface charge and dielectric discontinuity. Two choices of the perturbation potential, corresponding to either the total electrostatic interaction or the polarization energy, are examined for a 1–1 electrolyte. Both theories exhibit excellent agreement with Monte Carlo results at low surface charge density and electrolyte concentration, but as the surface charge is increased for any dielectric discontinuity, the former theory overestimates the role of fluctuations, while the latter theory does better for intermediate values of the wall charge.

Notes: The stochastic trajectory method has been applied to the scattering of CO from an LiF(100) surface. At low surface temperature TS, the trajectories of the gas molecule exhibited multiple collisions with the surface. The degree of rotational and translational energy accommodation could be related to the residence time at the surface. The residence time of the molecule on the surface was in turn related to a desorption rate constant which had an Arrhenius form with an activation energy which was about one third of the interaction potential well depth. At high TS most of the trajectories exhibited only one gas–surface collision. In this scattering regime we used stochastic sensitivity analysis (SSA) to obtain first and second order sensitivity coefficients which described how the final rotational and translational energies were coupled to TS and to the initial rotational and translational energies. At low initial translational energies EiT, we found that the most important effect on the final rotational energy of increasing EiT was the increase of the accommodation of energy between the surface modes and the rotational modes of the molecule. The direct coupling of the translational to rotational modes became dominant only at higher EiT. The energy parameters found at high TS with the SSA were also found to yield the per collision rate of energy accommodation at low TS.

Notes: We present a combined experimental/theoretical study of the coverage and frequency dependence of surface (enhanced) resonance Raman scattering [S(E)RRS] of cobalt phthalocyanine (CoPc) on CaF2 roughened silver films. The experimental spectra indicate a rather strong coverage dependence at very low coverage for excitation at or close to the molecular resonance frequency, with a peak at 0.07 monolayer (ML) followed by a rapid decrease above that. This coverage dependence differs strongly with observations on smooth films, where a much weaker dependence is observed. At very low coverage on the rough Ag films S(E)RRS enhancements comparable to SERS are observed. To model this coverage dependence, we consider the electromagnetic interactions between the adsorbate and substrate, with the substrate modeled as a metal spheroid and the adsorbate as a layer with variable coverage. Two models for this layer are considered, an effective medium model in which the layer is taken to have a coverage dependent dielectric constant, and a coupled dipole model in which both the molecules and metal are taken to be polarizable dipoles. The dependence of field enhancement and S(E)RRS intensity is studied for these two models as a function of frequency, coverage, and Stokes shift. It is found that although there are differences between the two models, the coverage dependence is similar in both, with peak intensities at about 0.1 ML for reasonably prolate spheroids. These models also demonstrate that the drop in intensity above 0.1 ML arises from damping of the plasmon resonance by the adsorbed layer. Interadsorbate interactions are found to play a role in determining the coverage dependence of the S(E)RRS intensity that is secondary to this damping. The large enhancements seen below 0.1 ML suggest that excited state quenching by the surface is not important for this nonfluorescent molecule.

Notes: The 193 nm dissociation of CS2 has been studied under supersonic molecular beam conditions, with the atomic S products detected by vacuum ultraviolet laser-induced fluorescence. These experiments measure the branching ratio between singlet and triplet channels to be S(3P)/S(1D) =2.8±0.3. Doppler spectroscopy on the S(3P2) and S(1D2) products was used to study the dynamics of each channel separately. The energy release in each channel was similar, with the CS fragments formed vibrationally inverted with about 50% of the available energy in CS rotation and vibration. The recoil velocity anisotropy in each channel was also measured, found to be the same for both, and was consistent with a 2 ps lifetime for the 1Σ+u(1B2) excited state of CS2.

Notes: Time-of-flight mass spectra are presented for the S1←S0 origin region of jet-cooled 1,2-, 1,3-, and 1,4-diethylbenzene, and 2-, 3-, and 4-ethyltoluene. The spectra for the diethylbenzenes exhibit two origins which are assigned to syn and anti conformations of the ethyl groups relative to the aromatic ring. The existence of two origins in the 1,3-diethylbenzene spectrum, and only one in the 3-ethyltoluene spectrum, strongly implies that the stable orientation of the ethyl groups is with the β-carbon atom of the ethyl group projecting perpendicular to the plane of the aromatic ring. The size and shape of the potential barrier to rotation of the ring methyl group is obtained by treating the methyl group as a one-dimensional rigid rotor and fitting the calculated energy levels to observed features in the spectra. The results (for the para isomer B=5.20 cm−1, V6=31.00 cm−1; for the meta isomer, B=5.40 cm−1, V3=78.00 cm−1; and for the ortho isomer, B=5.50 cm−1, V3=89.00 cm−1) indicate that the methyl group experiences an increased barrier to rotation in the order para〈meta〈ortho isomer.

Notes: Resonance ion dissociation (RID) spectra are reported for cations of naphthalene and 2-methylnaphthalene cooled in a supersonic beam. Discrete vibronic resonances were observed in the ultraviolet region of both ions. A discrete red system was also observed for 2-methylnaphthalene that is subject to strong degenerate vibronic interaction with an underlying quasicontinuum associated with a lower energy electronic transition. This leads to effective power broadening of the RID spectra at moderately low laser intensities. The red two-photon dissociation process of 2-methylnaphthalene was successfully modeled by classical rate equations applied to a four level system, consisting of a ground state, the directly excited state, a lower energy excited state, and a final dissociative state.

Notes: Laser excitation of matrix isolated IF has been used to characterize three previously unobserved electronic states. Excitation at wavelengths longer than 535 nm resulted in emission from the metastable A' 3Π(2) state. Vibrational analysis of the A'–X system gave an approximate value of Te(A')=13 250 cm−1. The radiative lifetime of IF (A') was found to be about 11 ms. Excitation of IF at wavelengths shorter than 475 nm produced a long-lived (τ=5 ms) fluorescence in the 720–1500 nm region. This emission originated from a low-lying electronic state (T00(approximately-equal-to)19 040 cm−1) and terminated on the v‘=9 to 22 levels of the ground state. Laser excitation spectra, recorded by scanning the laser wavelength and monitoring this emission system, showed a progression of bands from 474 to 415 nm. Franck–Condon arguments show that the absorbing state is not the emitting state. The excitation spectra provided a T00 value of 21 100 cm−1 and a vibrational constant of 500 cm−1 for the absorbing state.

Notes: Absolute Raman trace scattering cross sections have been measured for gaseous propane-h8, propane-d8, 1,1,1,2,3,3,3-propane-d7, and the two rotational isomers of 1,1,2,2,3,3,3-propane-d7. A set of ∂α¯/∂Sj intensity parameters were determined from the experimental cross sections and compared with the ones obtained from quantum chemistry calculations. The anomalous intensity ratio observed for the CH stretching bands of the two conformers of CHD2CD2CD3 is explained by an unanticipated difference in the ∂α¯/∂rCH value for the two types of methyl CH bonds.

Notes: Expressions constructed by applying statistical mechanics to a molecular model of polyion–counterion rod assemblies are used to find values for two phenomenological parameters in the stochastic theory of Szabo, Haleem, and Eden for the transient electric birefringence of rod-like polyions. These parameters are the polyion electric polarizability and the decay rate of fluctuations in macroion dipole moment. The range of system parameters over which the model of Szabo et al. applies is delineated.

Notes: The two-photon resonant absorption of Kr2(1Σ+g) has been measured via (2+1) multiphoton ionization studies in the energy range 91 750–92 420 cm−1. Two vibrational progressions are observed. One of them is attributed to the atomic 4p5(2P3/2)5p[5/2]2 state, implying D'0 =220 cm−1. A long progression is assigned as correlating with the 4p5(2P3/2)5p[3/2]2 state, implying D0 ≥1480 cm−1.

Notes: Angle-integrated, rotationally resolved photoelectron spectra were determined for three-photon resonant, four-photon (3+1) ionization of H2 via the C 1Πu, v'=0–4 intermediate levels using a magnetic bottle electron spectrometer. The results confirm and extend our previous angle-resolved results obtained using a hemispherical electron energy analyzer. As predicted by both Franck–Condon factors and recent ab initio Hartree–Fock level calculations, ionizing transitions in which the vibrational level of the resonant intermediate state is preserved in the ion are most probable; however, large deviations from theoretical expectations are observed for photoionization of all intermediate levels. These deviations take the form of a progressive broadening of the ionic vibrational distribution as the vibrational level of the resonant intermediate state is increased. The rotational branching ratios also exhibit systematic variations with both the vibrational level of the C 1Πu state and the vibrational level of the ion. Photoelectron signal from ionization of electronically excited atomic hydrogen produced in the photodissociation of H2 C 1Πu is also observed.

Notes: Polarized fluorescence from the A 2Πu and b 4Σ−g states of O+2 following photoionization of O2(X 3Σ−g) in the 17–21 eV region has been employed to determine population alignment in the O+2 photoion. The measured A-channel polarization exhibits pronounced structure which correlates well with the known positions of (autoionizing) Rydberg states, whereas polarization signal in the b channel has broad and less prominent features. The measured b-channel polarization values (ranging from −0.02 to −0.05) are consistent with single-channel calculations, while the A-channel polarization measurements (which range from 0 to 0.09) are of opposite sign to the previously reported single-channel calculations. The results of more involved calculations show that this sign reversal in the A channel can be attributed to the presence of intravalent 3Σ−u states which autoionize into the A channel in the near-threshold region.

Notes: Sixty-six lines in the ν4 band of H3O+ have been measured using diode laser velocity modulation spectroscopy. 36 have been assigned to the a–a component of the spectrum and 30 to the s–s component. The observed spectra were fit to a standard symmetric top Hamiltonian including the effects of l-type doubling. The band origin of the s–s transitions is at 1625.947(5) cm−1, whereas that of the a–a component lies at 1638.533(3) cm−1. The effect of the asymmetric bending vibration on the inversion splitting is in qualitative agreement with previous theoretical results, but the experimental band origins deviate by about 60 cm−1 from calculated values. Coriolis perturbations are found to have an important effect on the rotational constants of the ν4 and ν2 levels.

Notes: High resolution fluorescence and fluorescence excitation spectra of all-trans-2,4,6,8,10,12,14-hexadecaheptaene have been obtained in n pentadecane at 10 K. The most prominent vibronic features, like those in spectra of shorter polyene hydrocarbons, are due to combinations of symmetric carbon–carbon stretching vibrations. These in turn are dominated by a double bond mode whose frequency increases from 1555 to 1782 cm−1 upon electronic excitation (1 1Ag→2 1Ag). This 227 cm−1 increase, the largest yet observed in polyene spectroscopy, can be explained by vibronic coupling between the ground and first excited singlets. The possible role of vibrational mixing due to mutual polarizibilities between carbon–carbon bonds also is discussed.

Notes: We have measured the absolute emission cross section value for the O i(3p 3P–3d 3D0, λ=11 287 A(ring)) and the (3p 5P–4s 5S0, λ=11 299 A(ring)) multiplets excited by electron impact on O2. The former cross section has a value of 3.05×10−19 cm2±15% at 100 eV, while the latter's value is 4.14×10−19 cm2±15%. The quintet emission at λ=11 299 A(ring) is the only deexcitation channel for the 4s 5S0 state, and, therefore, represents the total excitation cross section for that state. However, the λ=11 287 A(ring) branch from the O i(3d 3D0) state only constitutes ∼25% of the total 3d 3D0 cross section, with the major branch from this state being the transition to the ground state observed at λ=1027 A(ring). The branching ratio obtained in this study is in good agreement with the theoretical calculation of Pradhan and Saraph (ω˜11 287=0.28), but our absolute σ(λ=11 287 A(ring)) value is substantially smaller than the result reported recently by Schulman et al.

Notes: The determination of a quantum correction factor for the transitional modes of a unimolecular dissociation/free radical recombination reaction having a transition state of varying looseness is described. The quantum correction factor for the high pressure canonical rate constant is calculated via Monte Carlo path integral evaluation of partition function ratios, and is applied to the recombination reaction 2CH3→C2H6.

Notes: Supersonic molecular beams of ammonia seeded with various rare gases are investigated in the expansion region using one-dimensional coherent anti-Stokes Raman scattering (CARS). A Raman spectrum of the ν3 vibration of the NH3 monomer with resolved rotational structure and inversion splitting is reported. Spectroscopy of this vibration is ideally suited to determine the rotational temperature in the beam. In the spectral region between 3100 and 3500 cm−1, broad structures are observed which can be attributed to clusters. These structures arise at strong stagnation conditions and increase when going from pure NH3 expansions to NH3/He and NH3/Ne mixtures. Analysis of the employed beams by mass spectrometry suggests that large clusters containing ∼5–50 NH3 molecules are responsible for the observed features. A narrow-peaked structure, red shifted by 3.5 cm−1 from the ν1 Q branch, is attributed to small clusters or NH3 molecules residing on the outside of the cluster whose H atoms are free and not bound to the complex.

Notes: A semiclassical calculation is presented that fully accounts for the angular momentum disposal in photodissociation of triatomic molecules. Rotational distributions are calculated for O2(3Σ−g) produced by the visible photolysis of ozone, O2(1Δg) by the UV photolysis of ozone, and OH by the 157 nm dissociation of water, to illustrate the effects of parent internal motion on fragment rotational distributions in the dissociation of C2v geometry molecules. A simple, but realistic, impulsive model of the energy release is used to describe the dissociation dynamics. The calculations are carried out for parent molecules at room temperature, as well as at the low temperatures characteristic of molecular beams. The contributions to the diatomic fragment rotational distribution from both parent triatom rotation and zero-point bending vibration are computed. Comparison of the calculated distributions with experimentally measured distributions indicates that the spread in rotational and bending vibrational angular momenta of the parent molecule can account for all or nearly all of the spread in final J of the diatomic photofragment. However, the rotational distributions of the diatomic photofragment reveal a strong vector correlation between the diatom angular momentum produced by the dissociative energy release, and the angular momentum associated with the in-plane rotation. The correlation is such that only half of all the photofragment states allowed by energy and angular momentum conservation are actually produced with appreciable probability. Of two energy degenerate photofragment states, corresponding to breaking of one or the other nominally equivalent bonds in the AB2 molecule, the one with the smaller orbital angular momentum/recoil linear momentum is strongly favored. This is explained by larger Franck–Condon overlap in the photoexcitation for the state of lower recoil angular momentum. The correlation involves selection of which of the two nominally equivalent bonds will break in the photodissociation of an AB2 triatom of C2v geometry, and thus represents a symmetry breaking mechanism in such a photodissociation.

Notes: In an endeavor to establish a suitable ab initio basis set for the description of the structure and vibrational spectra of phosphorus chlorides, the effect of adding center-of-bond (cb) s and p functions to 3-21G* basis sets was explored. To ensure generality of 3-21G*cb results were compared with various simpler bases for a range of molecules containing second row elements. While geometries and vibrational frequency patterns are well reproduced using the 3-21G* basis, dipole moments are less adequately reproduced. By addition of s and p orbitals at the bond centers there is a marked improvement in the polar properties with the other properties being little affected. Use of only s orbitals in this context is, if anything, detrimental. The results are applied to a study of the simplest of phosphorus chlorides, PCl3. Among the extant literature force fields, field A of Cazzoli is identified as being near correct. As with this field, all the ab initio fields show that the experimentally estimated first order Coriolis constant of ζ4 is too negative by ∼0.2.

Notes: Full configuration-interaction (FCI) calculations have been performed for the X 1Σ+g, a 3Πu, and b 3Σ−g states of C2 and the X 3Σ−g and A 3Πu states of Si2 in valence double-zeta plus polarization one-particle basis sets. Of the various approximate correlation treatments compared to the FCI, the CASSCF/MRCI method is found to be far superior in reproducing the FCI separations between the low-lying states. Hence this procedure was taken to near the one-particle basis set limit using [5s4p3d2f1g] carbon and [6s5p3d2f1g] silicon atom Gaussian basis sets contracted using the newly devised atomic natural orbital (ANO) method. The computed C2 separations in the large basis set are in excellent agreement with experiment. The ground state of Si2 is definitively determined to be X 3Σ−g, but the A 3Πu state is only 440±100 cm−1 higher in energy. For SiC, the A 3Σ− state is nearly 4000 cm−1 above the X 3Π ground state. Our best estimates of the spectroscopic parameters for the X 3Π state of SiC are re=1.719 A(ring), ωe=962 cm−1 and De=4.4 eV.

Notes: We extend recent works of Berry, Simon, and others on the evolution of adiabatic wave functions in parameter spaces with nontrivial global geometry, to show the interesting ways in which wave functions can acquire nonintegrable phase (commonly termed Berry's phase, geometric phase, or topological phase) upon transport along paths in the parameter space. We emphasize the case of arbitrary paths on the Born–Oppenheimer potential energy surfaces (the parameter space of the electronic states in an isolated molecule) of the linear plus quadratic E⊗e Jahn–Teller system. It is found that these surfaces are degenerate not only at the origin but also at three other, equivalent points, which lie on a radius ρ=2k/g. Here k and g are the linear and quadratic vibronic coupling constants, respectively. This radius is then shown to mark a sharp transition between Jahn–Teller behavior, characterized by half-odd-integral vibronic angular momentum, and Renner–Teller behavior, which has integral angular momentum. Finally, we examine the conditions necessary for adiabatic evolution in the E⊗e system, and the observable consequences of the geometric phase.

Notes: The construction of orthogonal operators to represent correlation crystal field (CCF) effects in the f N configuration of lanthanide and actinide ions is discussed. The approach of B. R. Judd [J. Chem. Phys. 66, 3163 (1977)] is extended to make all the operators orthogonal. The spin-correlated crystal field (SCCF) parametrization that has been successful in removing several long-standing anomalies from crystal field fits involves only two of the two-electron orthogonalized operators. We calculate values of the orthogonal CCF parameters from recent ab initio results. On the basis of these calculations the sixth rank SCCF operators are predicted to be very important. Fits using various combinations of the CCF parameters are reported for Ho3+ and Gd3+in LaCl3. For these systems we only obtain consistent results for the parameters associated with the sixth rank SCCF.

Notes: A zero-free-parameter ligand field model is used to account for the observed low-lying (E〈10 000 cm−1) electronic states of LaF. The electronic structure of LaF is represented as the effect of a nonpolarizable point F− ligand on the two valence electrons of a free La+ atomic ion. Free-ion configuration interaction (CI) effects (represented by a parametric fit of the free La+ energy levels to Fk,Gk Slater–Condon electrostatic and ζ spin–orbit parameters) and ligand-driven CI (treated by Bk0 ligand field radial integrals evaluated using Hartree–Fock La+ orbitals) are included in the model. A series of calculations is described, from the simplest "primitive LFT model'' which includes the three lowest-lying and most important configurations (6s2, 5d6s, 5d2) to the most elaborate "balanced s-polarized LFT model'' which includes six configurations (6s2, 5d6s, 5d2, 6s6p, 5d6p, and 6p2) in order to account properly for s–p polarization effects. The inclusion of ligand-driven CI (metal-centered orbital polarization) effects shows that simple electrostatic atomic-ion-in-molecule models are capable of accounting for the electronic structure of many highly polar diatomic molecules having more than a single metal-centered valence electron.

Notes: Hartree–Fock cluster calculations have been carried out for the ground 3d10 and excited 3d94s configurations of the Cu+ ion in a NaCl host. Special emphasis has been given to providing an accurate representation of the Coulomb potential due to the remainder of the lattice. Configuration coordinate curves were determined for the symmetric displacement of the nearest-neighbor Cl− ions and are compared to recent Xα calculations. The Hartree–Fock equilibrium Cu–Cl distance was found to be 5.327 bohr, slightly shorter than the calculated nearest-neighbor distance of 5.353 bohr for the pure NaCl host. Comparison of the Hartree–Fock and Xα ground and excited state energies, shows that the Xα approximation reverses the ordering of the 3T2g and 1Eg states, overestimates the equilibrium nearest-neighbor distance, and predicts the a1g vibrational frequency to be about twice the Hartree–Fock value. Using the Franck–Condon factors found with the ab initio potential energy curves, the calculated bandwidths for the 1,3T2g states are found to be in excellent agreement with the low-temperature absorption spectra.

Notes: Accurate calculations for the H−(H2O) complex with extended basis sets are reported at the restricted Hartree–Fock (RHF) through the fourth-order Møller–Plesset (MP) perturbation levels of theory. In the equilibrium geometry of the H−(H2O) complex the H− anion is found to lie almost along one of the H–O bond directions. The H–H− distance proved to be very sensitive to electron correlation effects: it is 1.8 and 1.4 A(ring) at the RHF and MP2 levels, respectively. The interaction energy between H− and H2O at the MP4 level including conterpoise corrections for basis set superposition error, depending upon the basis set used, is found to range from 16.2 to 16.9 kcal/mol, and the electron correlation is responsible for one-third of this value. The enthalpy of formation of H−(H2O) is estimated to be from −15.2 to −16.0 kcal/mol compared with the experimental value of −17.3±1.2 kcal/mol. The vibrational frequencies of H−(H2O) are also reported. The H−(H2O)2 complex is also studied by using a polarized double zeta basis set. The geometry of the H−(H2O)2 complex is partly optimized at the MP2 level and the energetic effects of the addition of the second water to H−(H2O) are analyzed.

Notes: The self-diffusion coefficient Ds is calculated to first order in volume fraction for a water-in-oil microemulsion model. Assuming spherical droplets, the interaction consists of hard-core repulsions, an attractive square-well potential due to overlapping surfactant tails, and hydrodynamic interactions. The latter are described by high-order series expansions outside the range of the attractive potential, whereas simplifying assumptions are made inside the range of overlapping. Within this model, expressions for Ds are derived in terms of the depth V0 of the attractive potential, the ratio of the range of overlapping to the radius of the hard cores, and a parameter describing the hydrodynamic interaction. Assuming a linear relation between V0 and the droplet radius the initial slope of Ds vs the volume fraction becomes more negative when V0 increases. A similar behavior is found if V0 increases at constant droplet radius. These results are compared to those obtained without hydrodynamic interactions.

Notes: The Raman spectra and phase diagram of fluorine were studied in a diamond-anvil cell up to 6 GPa over the temperature range 10–300 K. The sample slowly reacted with the diamond anvils to form CF4. The vibron frequencies in α-F2 and β-F2, as well as the lattice modes in α-F2, were determined as a function of pressure. No new phases were discovered. The α–β phase boundary can be fit with an equation of the Simon form: Pαβ (GPa)=−0.385(10) +4.80(9)×10−4 Tαβ1.75 (K). The melting curve was established to lie between limiting curves: Puppermelt (GPa)=−0.107+1.01×10−4T1.75melt (K) and Plowermelt (GPa)=−0.140+1.32×10−4 T1.75melt (K).

Notes: An analytical form of the HNC pair correlation function for the primitive ionic solution is proposed. This form interpolates between the EXP and MSA approximations. For 2:2 electrolytes it gives very good agreement to the thermodynamics and pair correlation function of the HNC.

Notes: The optical reflectivity of fluid interfaces near the critical point is analyzed in the light of recently improved estimates of the surface tension and measurements of the bulk correlation length. In contrast to earlier analyses, it becomes possible to discriminate between the two leading phenomenological theories of fluid interfaces, the intrinsic structure theory originated by Maxwell and van der Waals and the capillary wave theory as developed by Buff, Lovett, and Stillinger. For one of the two systems studied, cyclohexane–methanol, capillary wave theory is found to fit the reflectivity data well, while intrinsic structure theory is in strong disagreement with the reflectivity data. A simple combination of the two theories does not significantly improve the performance of capillary wave theory. For the second system studied, sulfur hexafluoride, neither capillary wave theory nor intrinsic structure theory fits the reflectivity data well, while their combination gives partial agreement with experiment.

Notes: Previous Monte Carlo simulations of the liquid–vapor interfaces of simple metals, which predict stratification of the liquid–vapor transition zone extending three atomic diameters into the bulk, suffer from a limitation arising from the use of a local electroneutrality approximation. This approximation is embodied in the assumption that, for the purpose of calculating the change in potential energy accompanying a change in configuration, there is exact coincidence of the electronic and ionic density profiles. In this paper we describe a generalized Monte Carlo simulation which avoids the local electroneutrality approximation by directly incorporating the Lang–Kohn treatment of the jellium-vacuum interface into the simulations. We report the results of generalized Monte Carlo simulations of slabs of sodium at 100 and 200 °C and cesium at 100 °C. The new simulations predict structure in the liquid–vapor interface very similar to that predicted using the local-electroneutrality approximation. The geometry of the slabs used in the new simulations allows us to make more careful studies of the transverse pair correlation functions in the interface than does the geometry of the clusters used in previous simulations of simple alkali metals. We find that in the peaks of the oscillations of the longitudinal density profile, the transverse pair correlation function resembles that of a fluid which is less dense than that of the corresponding homogeneous liquid, except for a slight amplification of the height of the innermost peak. Thus the longitudinal density oscillations predicted by the model do not lead to transverse structure factors resembling those of a higher density liquid, as had previously been expected. Despite the stratification, outer layers of the interface are not packed in a crystalline lattice, as is indicated by the isotropy of the transverse structure factor.

Notes: We study the stochastic differential equation for the motion of rod-like polymers in the isotropic and liquid crystalline phase for simple shear flow and elongational flow. We find a simple interpretation of the approximation given by Doi and we are able to treat the exact equations numerically. Comparing the results with the Doi approximation we find qualitative agreement.

Notes: An approach to the kinetic theory of a dilute solution of bead-spring rings with hydrodynamic interaction is outlined. When the cyclic symmetry of the ring is exploited, the analysis of Hookean-spring rings is greatly simplified. Exact expressions for the time constants of a Hookean-spring ring with an arbitrary number of beads are obtained. The only effect of ring closure is to alter the spectrum of time constants. The form of the constitutive equation for a dilute solution of rings is the same as that for a dilute solution of Rouse chains. The results of Wiest et al. for Hookean-spring rings without hydrodynamic interaction, which were obtained by using a different approach, are recovered as a special case.

Notes: The chemisorption of hydrogen on both the Ir(111) and Pt(110)-(1×2) surfaces has been examined under ultrahigh vacuum conditions with thermal desorption mass spectrometry, LEED, and contact potential difference measurements. No ordered adsorbate superstructures were observed on either surface at any fractional coverage and at surface temperatures from 100 to 700 K, and the (1×2) reconstruction of the Pt(110) surface was stable in all cases. Hydrogen adsorbs dissociatively on the Ir(111) surface, the adsorption reaction described by second-order Langmuir kinetics with an initial probability of adsorption of 7×10−3. The rate parameters describing the second-order desorption reaction of hydrogen from the Ir(111) surface are weakly dependent on coverage between fractional coverages of 0.1 and 0.3, and are given by Ed (approximately-equal-to)12.6 kcal mol−1 and k(2)d (approximately-equal-to)2×10−6 cm2 s−1. Beyond a fractional coverage of 0.3, however, both rate parameters decrease with increasing coverage. Hydrogen adsorbs dissociatively on the Pt(110)-(1×2) surface into two distinct β2 and β1 adstates, and the ratio of the saturation densities of these two states, β2:β1, is 1:2. Adsorption into the higher binding energy β2 adstate is described by first-order Langmuir kinetics with an initial probability of adsorption of 0.46, whereas adsorption into the β1 adstate is described by second-order Langmuir kinetics with an "initial'' probability of adsorption of 0.022. The rate parameters describing the desorption reaction of hydrogen fromthe Pt(110)-(1×2) surface are strongly dependent on the coverage. In the coverage regime characteristic of the β2 adstate (θ≤0.32) the rate parameters are approximately symmetric about one-half of saturation of this state. Specifically, from the values for the zero-coverage limit of Ed (approximately-equal-to)18 kcal mol−1 and k(2)d (approximately-equal-to)10−4 cm2 s−1, the parameters first increase to maximum values of Ed (approximately-equal-to)26.5 kcal mol−1 and k(2)d (approximately-equal-to)0.3 cm2 s−1 at θ=0.15, and subsequently decrease approximately to the values for the zero-coverage limit at θ=0.32 In the coverage regime characteristic of the β1 adstate (θ〉0.32), the activation energy decreases continuously with increasing coverage from a value of Ed (approximately-equal-to)17 kcal mol−1 at θ=0.35, whereas the preexponential factor remains essentially constant with a value of 3×10−4 cm2 s−1. The contact potential difference for hydrogen on Pt(110)-(1×2) increases continuously with coverage to a value of 0.17 eV at θ=0.30. As the coverage increases further, however, it decreases continuously approaching a value of −0.50 eV at saturation. Probable binding states for the β2 and β1 adstates on the Pt(110)-(1×2) surface are inferred from both the adsorption and desorption kinetics and the contact potential difference measurements. Comparisons of the results obtained on the (111) and (110)-(1×2) surfaces of both iridium and platinum suggest strongly that local surface structure (e.g., "step'' sites vs terrace sites) has a profound influence on the kinetics of adsorption of hydrogen on these surfaces. Surface structure apparently also has a profound influence on the desorption kinetics of hydrogen via the mediation of adatom–adatom interactions. Whereas both attractive and repulsive interactions are clearly manifest within the β2 adstates on the (110)-(1×2) surfaces, only repulsive interactions are apparent on the (111) surfaces and for the β1 adstates on the (110)-(1×2) surfaces.

Notes: Various approaches to the translational diffusion of polymers in a dilute solution are discussed for the Zimm model. The whole discussion is carried out for an arbitrary definition of the center of a polymer, for all homogeneous steady flows, and for constant external forces. It turns out that the result for the diffusion coefficient differs from the well-known Kirkwood expression, however, the numerical difference is less than 2%. We point out which approximations in the various derivations of the Kirkwood expression for the diffusion coefficient lead to this difference. The expression for the diffusion coefficient given in this paper can be generalized to arbitrary bead-spring models. Furthermore, we give the configurational distribution function for the Zimm model with time-dependent external forces in an arbitrary homogeneous flow field.

Notes: From a Landau theory the static scattering intensity distribution I(q) of microemulsions is obtained. As essential ingredient we have included a negative gradient term in the free energy expression. The form of I(q)∼(a2+c1q2+c2q4)−1 yields for a2〉0, c1〈0, and c2〉0 a single broad scattering peak and a q−4 decay at large q, both properties experimentally observed for a variety of microemulsions containing comparable amounts of water and oil. The peak originates from the modulation in the corresponding space correlation function given by γ(r)=(d/2πr)⋅e−r/ξ⋅sin(2πr/d). It is shown that the scattering intensity relation describes experimental literature data remarkably well, using only three fit parameters.

Notes: A variable time step algorithm has been implemented for solving the stochastic equations of motion for gas–surface collisions. It has been tested for a simple model of electronically inelastic collisions with an insulator surface in which the phonon manifold acts as a heat bath and electronic states are localized. In addition to reproducing the accurate nuclear dynamics of the surface atoms, numerical calculations have shown the algorithm to yield accurate ensemble averages of physical observables such as electronic transition probabilities and total energy loss of the gas atom to the surface. This new algorithm offers a gain in efficiency of up to an order of magnitude compared to fixed time step integration.

Notes: Spectroscopic observation of the highest vibrational level of HD+(v=21) in its ground electronic state is reported. Calculations of constants from the resolved hyperfine structure allow unambiguous determination of the Fermi contact parameters for the levels 17,1 and 21,0, showing that in the level 21,0 the electron distribution is asymmetric toward the lower dissociation limit H++D, with an average electron density of 0.094 at the proton and 0.902 at the deuteron.

Notes: Direct measurements of dynamic charge localization in tris (2,2′-bipyridyl) ruthenium (II) have been performed in fluid solution for the first time. Using picosecond Raman spectroscopy it has been determined that charge localization occurs on a time scale of (very-much-less-than)30 ps for low-viscosity solvents. In high-viscosity solvents such as glycerol at room temperature, the charge localized species Ru(bpy)2(bpy)− appears on a time scale of 100 ps. It is clear from these data that solvent reorganization is one of the dominant mechanisms in electron trapping.

Notes: We report the successful application of an optically pumped molecular beam laser for efficient and selective population transfer from v″=0 to high lying vibrational levels of Na2 in the electronic ground state. More than 50% of the molecules in a thermally populated rotational level of v″=0 can be transferred to a state v″(very-much-greater-than)1 with typically ≥75% of the molecules excited out of the initial state found in the predetermined end vibrational level and approximately ≤25% distributed over other levels according to the Franck–Condon factors. The applicability of this technique to the study of collision processes involving vibrationally highly excited levels is discussed. This work also contributes in general to a better understanding of optically pumped dimer lasers.

Notes: We have analyzed the Mössbauer spectra of several iron proteins in the regime of intermediate fluctuation rates of the magnetic hyperfine interaction using the dynamic line shape model of Clauser and Blume. All Mössbauer data were recorded in applied fields so that the electron Zeeman interaction is much larger than the hyperfine splittings. We modeled the transitions between the eigenstates of the electronic spin Hamiltonian by an effective spin–phonon coupling V(1)=V0∑′i,j=x,y,zSi Sj and described the phonons in the Debye approximation. With V0 as the only adjustable parameter the model reproduced the major temperature dependence of the spectral shapes for spins from S=1 to S=5/2. We found V0 to lie in the narrow range 3 K ≤V0/k≤19 K, which is of the order of the zero-field splittings, but does not correlate with it. The smallest values of V0 were observed in the four-coordinate iron–sulfur proteins. Refinements and implications of the model are discussed.

Notes: Analytic recurrence relations for the calculation of the Yun Dunham coefficients are established, thereby allowing to generate all the set of the Yun from the Yu0 coefficients. Then, expressions for diagonal radial matrix elements of any power of the internuclear displacement from equilibrium in an arbitrary vibration–rotation level of the 1Σ-state diatomic molecule follow in terms of Yu0 only. As an illustrative application, all the set of Yun coefficients previously published are found again by using the computer algebraic manipulation language reduce. A number of diagonal matrix elements of CO, HBr, and HCl have been symbolically computed and compared with previous available results.

Notes: Nondiffusive triplet energy transport is observed in a p-dibromobenzene single crystal doped with p-dichlorobenzene (DCB) at variable concentrations. Phosphorescence kinetics of DCB monomers and aggregates are strongly nonexponential and can be described assuming a stretched exponential exp[−(t/τ)β], where β varies from 1 at low concentrations to 0.28±0.05 at concentrations above 0.05 mol/mol.

Notes: The binary reaction of CH+3 with NH3 as well as their ternary association have been investigated in the regime from 0.047 to 0.2 eV, KEcm, using the Innsbruck selected ion drift tube (SIDT). In this energy regime the association rate coefficient declines from 4×10−26 to 6×10−27 cm6 s−1 and the lifetimes of the collision complex (CH+3−NH3)* range from ∼10−7 to ∼10−8 s.

Notes: The effect of the finite time for resonance Raman scattering on Raman depolarization ratios is evaluated. Partial depolarization of the resonance Raman emission will occur if the scatterers can rotate during the time interval between the absorption and emission steps. Equations are derived which express the depolarization ratio in terms of Fourier transforms of the usual time-dependent vibrational overlap multiplied by a rotational correlation function. Particularly simple results are obtained in the limit of severe inhomogeneous broadening of the electronic transition. The effect of the finite resonance scattering time on the line shape of the resonance Raman transition is also discussed.

Notes: A one degree of freedom, periodically forced Hamiltonian is shown to describe much of the relevant classical dynamics observed in earlier studies of a two degree-of-freedom model for HeI2 fragmentation. The simplicity of the forced oscillator model permits easy, but thorough numerical investigation of the quantum dynamics of fragmentation using an accurate wave packet propagation method. Quantum dynamics results are compared with classical dynamics results based on both the usual quasiclassical procedure for selecting initial conditions and on the Wigner distribution. The exact Wigner distribution as a function of time is obtained from the quantum dynamics results and very interesting, unexpected parallels with the classical phase space structure are evident.

Notes: The cross sections for the Rb(7S→5D), Rb(5D→7S), and Na(4D→5S) collisional transfers induced by O2 have been measured in absolute values as a function of the collision energy using a crossed beam apparatus. The experimental data have been compared with the predictions given by a multiple curve-crossing model where the symmetries of the states formed by the colliding alkali-metal atom–O2 system as well as the molecular orientation are considered. Good agreement is observed between experimental and calculated cross sections. This tends to show that in spite of its approximations, a multiple curve-crossing model would be useful to obtain fairly accurate informations about collisional processes involving alkali atoms with as much as 4 eV electronic excitation.

Notes: Accurate ab initio calculations are performed to study the structures and energies of small carbon clusters (Cn, n=2–10). The effects of polarization functions and electron correlation are included in these calculations. Significant odd–even alternation is found in the nature of the cluster geometries with the odd-numbered clusters having linear structures and many of the even-numbered clusters preferring cyclic structures. Energetically, odd-numbered clusters (up to C7) are found to be more stable than the adjacent even-numbered clusters. Ionization potentials are calculated and used in conjunction with the cluster energies to explain the fragmentation behavior of small carbon cluster ions.

Notes: X-ray diffraction measurements up to momentum transfer q=16 A(ring)−1 were performed at atmospheric pressure on both the high-density form of amorphous ice obtained by pressurization of crystalline ice Ih at 77 K up to 2 GPa and the low-density form obtained by further heating. These two amorphous phases exhibit significant differences at the level of the second- and third-nearest neighbors which involve a decrease of the O–O–O angle on compression at high pressure. Moreover, it is found that the low-density form is quite similar to that obtained by vapor deposition on cooled substrates at 77 K.

Notes: A simple model is presented that describes the complex behavior of the magnetic field dependence of the thermal conductivity of a polyatomic gas as a function of the Knudsen number. The model treats the solid surface as one component of a gas mixture, and a first-order Chapman–Enskog solution is adequate to account for the thermomagnetic effects in single gases. The results provide an interpolation scheme between the continuum and Knudsen regimes, in the form of scaling rules in which the Knudsen number appears in the coefficients and arguments of the functions that describe the limiting cases. Good agreement is obtained with the available experimental data, and leads to the conclusion that different surface accommodation coefficients are needed for the translational and internal energies of the gas molecules. This conclusion is consistent with analysis of independent field-free conductivity data, and with independent vibrating-surface measurements.

Notes: The dynamics of interacting molecules involve the translation and simultaneous rotation of each molecule in the ensemble. This means that the frame of reference defined by the principal molecular moments of inertia, frame (1,2,3), is a noninertial frame with respect to the laboratory frame (x, y,z). Furthermore, a rotating frame of reference (1,2,3)' can be generated from (1,2,3) a by translation for each molecule from the center of mass of the origin of (x, y,z). The equations relating velocities, accelerations, and their derivatives in one of these frames to those of another are described and then used to calculate the cross correlation functions.

Notes: Neutron diffraction measurements have been performed on liquid equiatomic potassium–lead alloy at T=870, 900, and 930 K to determine the influence of temperature on the structure. A well-defined first sharp diffraction peak (FSDP) at Q=0.96 A(ring)−1, similar to that found in glassy materials, was obtained in the total structure factors S(Q) at the three temperatures. While the position of the peak remained unaffected by temperature changes, its magnitude decreased with increasing temperature. The other features of S(Q) were almost unaltered by temperature, as expected. The FSDP can be related to the presence of chemical bonds or clusters whose stability decreases with increasing temperature. A molecular model based on the random packing of K4Pb4 structural units is shown to give a remarkable correspondence with the measured structure factors.

Notes: The phase-space region occupied by a dense fluid is characterized by significant multiparticle correlations, and by small density fluctuations. On the basis of this characterization, a statistical-mechanical formula is constructed, to represent the entropy of an equilibrium dense fluid. For the example of fluid sodium, the statistical entropy agrees well with the measured thermodynamic entropy.

Notes: The structure and stability of aluminum clusters containing up to six atoms have been studied using correlated wave functions and extended basis sets. The lowest energy structure is planar for Al4 and Al5, but three dimensional for Al6. The icosahedral, hcp, fcc, and two planar structures of Al13 were considered at the SCF level. The lowest energy structure is the icosahedron, but the planar structures are fairly low lying even in this case. A simplified description using two- and three-body interaction potentials is found to agree well with the ab initio structures and binding energies.

Notes: Formal solutions to the exact BGY equations of the single occupancy model of classical anharmonic solids have been developed and approximate solutions have also been constructed on the basis of the "consistent'' closure relation. The partition function of the system is determined by the normalization condition of the singlet function ρ1 and the total distribution function ρN through the use of the linked hierarchical relationship between distribution functions. It is explicitly proved that the free energy expression, the virial pressure, and the energy equation are all consistent at all orders of truncation approximation. A new method to calculate solid-phase entropy in Monte Carlo experiments is also proposed based on the free energy we developed and applied to inverse 12th-power systems to give a numerical example.

Notes: Recent dielectric relaxation measurements on poly(vinyl acetate), PVAc, by Nozaki and Mashimo (NM) have demonstrated temperature and frequency dependences for well-annealed samples near the glass transition temperature Tg which cannot be explained by conventional theories or concepts. NM concluded that a new concept has to be introduced to interpret the dielectric behavior of PVAc near Tg. We demonstrate that the coupling model of relaxation is able to provide this new concept and can quantitatively describe the dielectric behavior at all temperatures. NM have carried out dielectric relaxation measurements between T=84.77 and 26.85 °C on a PVAc sample with Tg=31 °C. The samples were carefully annealed at each temperature to ensure that it is in equilibrium. NM found that the asymmetric dielectric loss curves become broader as temperature is lowered toward Tg with the most significant changes occurring near Tg. Furthermore the frequency of maximum loss fm was found to be described by the WLF equation (expected to be valid on the basis of free volume models) only at high temperatures. Extrapolation of the WLF curve to the glass transition region deviates significantly from the measurements. All of these phenomena are expected by the coupling model. The model predicts a relation between the shape of the dielectric loss curves and properties of the relaxation time. We demonstrate that the temperature dependence of fm is quantitatively predicted at all temperatures, including the deviations from WLF behavior near Tg. The same coupling model was previously shown to resolve several other problems in polymer physics near Tg.

Notes: A tunable alexandrite laser was used to produce O2(b 1∑+g) in the gas phase, with quantitative measurement of O2(b 1∑+g) and O2(a 1Δg) densities achieved by detection of the b→X fluorescence and dimol emission signals, respectively. A simple kinetics analysis indicates that collisional deactivation of O2(b 1∑+g) by ground state oxygen molecules occurs with unit production of O2(a 1Δg). These results are in accord with previous assumptions regarding collisional deactivation of O2(b 1∑+b) based on spin conservation.

Notes: The characteristics of three types of collisional interactions in molecular collisions (dipole–dipole, dipole–quadrupole, and dipole–induced dipole) have been studied by a novel laser spectroscopic method. The rotational relaxation rate constants of CH3CN against various kinds of collision partners are measured. It has been found that the relaxation rate constant critically depends on the energy level structure of collision partners for the long range interaction while it is less sensitive to the level structure for the short range interactions.

Notes: A theoretical determination of the Jahn–Teller coupling constants of d5 ions in tetrahedral clusters is performed in the framework of the angular overlap model and the quasimolecular approximation. It is shown that the angular model correctly predicts the trend of the preponderant coupling of the 4T1(G) and 4T2(G) levels of Mn2+ in MnS4 and MnSe4 to the E-vibrational effective mode. Finally, the limitations of this model are discussed. In particular, it is shown that more elaborated covalent models are needed to interpret the Jahn–Teller coupling coefficients of the 4E and 4A1 levels of d5 configurations and to determine the influence of the crystal on the clusters.

Notes: A molecular orbital model has been elaborated to determine the orbit–lattice coupling coefficients to strains of symmetry E of the orbital triplet states of d5 ions in tetrahedral symmetry. The wave functions and energies of the monoelectronic molecular orbitals have been determined for a tetrahedral molecule MnX4 in a crystal and for a slightly distorted molecule of symmetry D2d corresponding to a strain Eθ. It is shown that in D2d symmetry, the splitting of the monoelectronic molecular orbitals e and t2 is due to the variations of the group overlap integrals and the variation of the crystal electrostatic field with respect to their values in cubic symmetry. The orbit–lattice coupling coefficients which are independent of the spin–orbit interactions, are obtained by diagonalizing the Tanabe–Sugano matrices for the orbital triplet states and then using the molecular orbitals in D2d symmetry as zero-order wave functions in a first-order perturbation scheme. The proposed model for the orbit–lattice coupling coefficients of the orbital triplet states 4T1 and 4T2 at lower energy for Mn2+ in ZnS and ZnSe gives theoretical values which differ at most by 30% from the experimental values. It is shown that three fifths of the theoretical values for the orbit–lattice coupling coefficients are due to the variations of the group-overlap integrals, the remaining part of the theoretical values being due to the variations of the crystal electrostatic field describing the Coulomb's interaction in the molecular model.

Notes: Absolute Raman trace scattering cross sections have been measured for gaseous cyclohexane-h12, cyclohexane-d12, and the axial and equatorial isomers of cyclohexane-d11. A set of ∂α¯/∂Sj intensity parameters were determined from the experimental cross sections. The quadratic vibrational force field required for this analysis was obtained by scaling an ab initio force field to fit well determined vibrational frequencies. The unexpected intensity ratio observed for the CH stretching bands of the two conformers of d11 is explained by a difference in the ∂α¯/∂rCH value for the axial and equatorial CH bonds. The transferability of Raman intensity parameters is considered in light of recent results.

Notes: High resolution spectra of the [(2Γ3+ν3)←1Γ1] transition from the absorption band systems near 800 nm in PuF6 have been recorded. Rotational lines and some fine structures have been resolved. The results support a spherical-top configuration despite the possibility of Jahn–Teller distortion.

Notes: The electron impact spectrum of ethane has been examined in a region that includes ionization out of the inner valence shell. One diffuse structure and a progression of ten vibrational bands have been found in a 4 eV range below and to some degree overlapping the 2A2u ion threshold. Evidence indicates that the observed transitions belong to the symmetry forbidden Rydberg series (2a2u)2→(2a2u, npσ or npπ).

Notes: The second-order perturbation theory of convex-molecule fluids with pair interactions of the generalized Kihara type is considered. The double Yukawa potential function and the Laplace transform of the hard-sphere correlation function in the PY approximation are used to evaluate a part of the perturbation integral. The perturbation theory was applied to model systems of the LJ molecules and Kihara rod-like particles; calculated thermodynamic functions agree well with the corresponding Monte Carlo data and results from the Weeks–Chandler–Andersen theory. Characteristic parameters of several fluids with spherical and rod-like molecules were evaluated from the saturated liquid properties. Comparison with the corresponding parameters of the multicenter LJ potential is given.

Notes: The magnetic susceptibilities of single crystals of NaMg[Cr(oxalate)3]⋅9H2O are reported over the temperature interval 40 mK–4.2 K. Large anisotropy is observed due to the large zero-field splitting of the 4A2 ground state; the parameters that fit the data are 2D/k =−2.45(3) K, g(parallel)=1.985(5), g⊥=2.035(10), and an exchange correction of zJ/k =0.006(1) K.

Notes: CO2 laser–microwave double resonance (LMDR) spectroscopy with an intense electric field was applied to the NH3 molecule. Using infrared transitions in the ν2 band for pumping, 76 and 37 LMDR signals originating from 25 and 16 inversion transitions in the ground vibrational state were observed for 14NH3 and 15NH3, respectively. The applied electric field was calibrated with a precision of 10 ppm by a new method using the infrared–infrared double resonance signals of PH3, which have been calibrated against the OCS LMDR signals. The observed signals were analyzed to yield the precise dipole moment with rotational dependence, which was expanded as μ0+μJJ(J+1) +μKK2+⋅⋅⋅. The coefficients determined are, with one-standard-deviation uncertainties in the parentheses, μ0=1.471 932(7) D, μJ=1.8736(34)×10−4 D, and μK =−3.5550(47)×10−4 D for 14NH3, and μ0=1.471 964(11) D, μJ =1.9244(77)×10−4 D, and μK =−3.5903(98)×10−4 D for 15NH3. The absolute accuracy of the dipole moment is, however, limited to 50 ppm as determined by the uncertainty of the OCS dipole moment. The effective dipole moments for the J,K=1, 1 and 1,0 levels of 14NH3 calculated from the present coefficients are consistent with the results of a molecular beam electric resonance experiment. The μJ and μK coefficients are interpreted in terms of Watson's θαβγ constants, resulting in a good agreement between the observed and theoretical values. The consistency between the two isotopic species is also confirmed.

Notes: Photoionization from hydrogen fluoride has been studied with synchrotron radiation in the 30–130 eV photon energy range. Relative photoelectric cross sections have been obtained for the 1π (ionization energy of 16.1 eV) MO, the 3σ MO (19.9 eV ionization energy), and in some cases the inner 2σ valence MO (I.E.=39.6 eV). Comparison is made with other measurements such as quasiphotoionization results and with one-center calculations. Angular distribution asymmetry parameters have also been measured for the 1π and 3σ MO's and compared with those of Ne2p. In the inner valence region, satellites have been observed; this is in accord with most theoretical calculations which predict the breakdown of the one-particle description in this region.

Notes: It was recently suggested that vibrational excitation of van der Waals molecules could lead to prereaction instead of predissociation. A mechanism for vibrational prereaction based on tunneling of a light atom is proposed and tested. The effect of van der Waals wells on the reactivity of collinear MuD2, ClHBr, and ClHCl systems is studied. We find that vibrational prereaction is sensitive to the ratio of tunneling and vibrationally nonadiabatic interaction. If tunneling dominates, prereaction will take place. This is the case for the ClHBr and ClHCl systems. The interplay between reaction probability and photodissociation cross sections is studied. We conclude that vibrational prereaction will lead to an increase in reaction probability and an increase in the formation of products in photodissociation. This study suggests that vibrational prereaction could be observed at least in principle in light atom transfer systems.

Notes: We have studied the kinetics of the lowest excited singlet state of molecular nitrogen, N2(a′ 1Σ−u,v=0), in a discharge flow reactor. The metastables were generated in a hollow cathode dc discharge through molecular nitrogen highly diluted in argon, and detected by VUV flourescence of the forbidden N2(a′ 1Σ−u–X 1Σ+g) band system. Observations of the equilibrium between the N2(a 1Πg) and N2(a′ 1Σ−u) states in our reactor indicate that the metastable has a radiative lifetime (approximately-greater-than)23+11−6ms, assuming a radiative lifetime of 80+40−20μs for N2(a 1Πg). Rate coefficients for quenching by NO, CH4, CO, and N2O are approximately gas kinetic, those by H2, O2, and CO2 roughly tenth gas kinetic, and that by molecular nitrogen (1.9±0.5)×10−13 cm3 molecule−1 s−1. The interaction between N2(a′) and CO produced CO(A 1Π–X 1Σ) emission in about 21+10−8% of the quenching events.

Notes: We consider in the present paper an extension of numerical path integral methods for use in computing finite temperature time correlation functions. We demonstrate that coordinate rotation techniques extend appreciably the time domain over which Monte Carlo methods are of use in the construction of such correlation functions.

Notes: Monte Carlo calculations and gas phase mass spectrometry studies have been carried out on the system Ca+(H2O)n. The parameters used in the calculations to describe the interaction energy between Ca+ and a water molecule are deduced from quantum mechanical ab initio determinations. The clusters formation energies through the reaction Ca+(H2O)n−1+H2O (arrow-right-and-left) Ca+(H2O)n have been calculated for n=1 to 10 and determined from gas phase mass spectrometry for n=1 to 5. For n=6 to 10, theoretical work shows that, though the most stable structures have six water molecules in a first shell surrounding Ca+, the statistical distribution increases from 5.28 to 5.63, due to the presence of many stable configurations with five water molecules in the first shell. A comparison is done with the system H3O+(H2O)n.

Notes: The most stable structures for the alkali-like clusters M−3–M−8 are calculated within the framework of the simple Hückel model. The Hückel geometries are, on average, slightly "less compact'' than those of the neutral and cation clusters, a phenomenon which may be related to the additional electronic kinetic energy of the anions. Cluster compactness is quantified by an estimation of "soft sphere'' volumes, which also allows for a comparison of classical and experimental polarizabilities. The Hückel model gives electron affinities which compare favorably with the experimental results for Cu2–Cu8. To our knowledge, the Hückel results in this paper represent the first systematic search for the stable structures of small alkali-like anion clusters.

Notes: Electron propagator theory is applied to calculating vertical electron affinities of the title molecules. Third order and outer valence approximation quasiparticle calculations with and without effective core potentials are compared for molecules with fourth period atoms. Discrepancies between the two sets of results are small and stimulate applications to fifth period analogs. Good agreement with experiment is obtained for all the molecules; a prediction for the still unmeasured electron affinity of SbH2 is made.

Notes: A scaled fundamental equation is presented for the thermodynamic properties of carbon dioxide in the critical region. The equation is constructed by combining earlier experimental pressure data of Michels and co-workers with new specific heat data obtained by one of the authors and represents the thermodynamic properties of carbon dioxide in the critical region at temperatures from 301.15 to 323 K and at densities from 290 to 595 kg/m3.

Notes: Some fundamental aspects of crystallization of two dimensional (2D) crystals are examined, including an analysis of the fluctuations in step positions, computer simulation results for the step density and growth rates, and analysis for predicting crystal shapes. A discussion is given of whether nucleation can occur on the 1D surface of a 2D crystal. During the course of the paper it is shown that a more appropriate description, for any infinite 1D surface, is of a step density related to the equilibrium value and a growth rate which is proportional to the density and speed of these steps. The speed depends essentially linearly on the net thermodynamic driving force for crystallization. (This description is not appropriate for a 2D surface since the equilibrium step density can then be zero.) Nevertheless, one can define a criterion, depending both on the density of steps and on the extent of the fluctuations in step positions, for a regime of growth with some similarities to nucleation (1D nucleation). Kinetic (nucleation) theories for polymer crystallization, using a rate equation approach, have implicitly assumed the conditions for 1D nucleation to hold. Even when the conditions do not hold, the results of this rate equation approach are in reasonable agreement with the simulation results (though a rather curious discrepancy of a factor of 21/2 occurs since the rate equation approach omits a growth mediated approach to equilibrium). The agreement in general trend is shown to be the fortuitous consequence, in the rate equation approach, of neglecting both the enhancement of step annihilations caused by fluctuations and the creation of pairs of steps at cavities. If the latter effect is neglected but not the former, anomalous results can be obtained which violate thermodynamic principles. Growth rate results from simulation are obtained as a function of the obliquity of the growth face with respect to the principal directions in the lattice. These growth rate anisotropy data are used to predict the shapes of crystal outlines. Comparisons between these results andexperimental observations on shapes of polymer crystals are made, and a preliminary account is given of how the step free energy per length of step decreases with increasing temperature. In general, the total step free energies tend to decrease as the crystal thickness increases for polyethylene.

Notes: Using the theory of normal forms, we investigate the effects of mixing in a continuous flow stirred tank reactor (CSTR) for a reaction model exhibiting oscillatory behavior in the vicinity of a degenerated bifurcation point (here, a Takens–Bogdanov point). In addition we show without specification of a particular reaction system that, as long as reaction rates remain much slower than the inverse mixing time, incomplete mixing introduces a new bifurcation parameter for nonpremixed feeding conditions, whereas premixed feeding conditions merely lead to a renormalization of flow rate.