Index of content:
Volume 65, Issue 4, 15 August 1976

Intermolecular interactions in polymorphic p‐dichlorobenzene crystals: The α, β, and γ phases at 100 °K
View Description Hide DescriptionThe structures of the supercooled α and β phases of p‐dichlorobenzene are determined at 100 °K and a detailed comparison made with the γ‐phase structure previously reported at that temperature [Acta. Crystallogr. B 31, 911 (1975)]. The effects of temperature are investigated by a redetermination of the β‐phase structure at room temperature and by measurement of the thermal expansion coefficients for the α‐ and β‐phase crystals. The average atomic thermal displacements are found to be in the expected thermodynamic order. For chlorine we find Ū (β) =0.184 Å, Ū (α) =0.169 Å, and Ū (γ) =0.157 Å at 100 °K. Rigid body analysis of our data gives good agreement between calculated and observed librational lattice frequencies. New limits are placed on the density changes associated with the phase transitions [ΔV (α→β) =0.1±0.4% and ΔV (γ→α) =0.5±0.4%]. The effects of crystal field on the molecular geometry are limited to a slight out‐of‐plane displacement (0.045 Å) of the chlorine atoms in the γ phase. The possibility of such distortions in the high temperature α and β phases is masked by slight positional disorder (∼0.02 Å). The chlorine–chlorine interactions are found to be attractive and anisotropic. This indicates the necessity of including anisotropic terms in semiempirical atom–atom potential functions. All presently available functions, including Coulomb–Coulomb terms, are found to be inadequate for this reason.

Laser photoelectron spectrometry of PO^{−}, PH^{−}, and PH^{−} _{2}
View Description Hide DescriptionFixed‐frequency laser photoelectron spectrometry has been used to study the ions PO^{−}, PH^{−}, and PH^{−} _{2}. The electron affinities of PO, PH, and PH_{2} were determined to be (1.092±0.010), (1.028±0.010), and (1.271±0.010) eV, respectively. A bound excited state (ã ^{1}Δ) of PO^{−} was observed with a term energy of (0.556±0.010) eV. The vibrational frequencies of X̃ ^{3}Σ^{−} PO^{−}, ã ^{1}Δ PO^{−}, X̃ ^{2}Π PH^{−}, and ^{2} B _{1} PH_{2} (ν_{1}) were found to be (1000±70), (1020±80), (2230±100), and (2270±80) cm^{−1}, respectively. A Franck–Condon factor analysis of the PO^{−} and PH^{−} spectra gave the equilibrium internuclear separations r _{ e }(X̃ ^{3}Σ^{−} PO^{−}) = (1.540±0.010) Å and 1.414⩾r _{ e }(X̃ ^{2}Π PH^{−}) ⩾1.400 Å. The geometries of ^{1} A _{1} PH^{−} _{2} and ^{2} B _{1} PH_{2} were found to be nearly the same. The PH^{−}photoelectron spectrum includes transitions to the PH X̃ ^{3}Σ^{−} and ã ^{1}Δ states, allowing determination of the intercombination separation of (0.950±0.010) eV.

On microwave discharge sources of new chemical species for matrix‐isolation spectroscopy and the identification of charged species
View Description Hide DescriptionThe mechanism for trapping new chemical species by condensing the products of a microwavedischarge with inert matrices has been investigated. Variation of geometrical, electrical, and chemical parameters of the Ar, HCl, Cl_{2} system indicated that the major product species—HCl_{2} radical or anion—was formed under conditions where neither ions nor atomic species produced in the discharge were condensed in the matrix. The mechanism for forming the product species is vacuum ultraviolet photolysis of the sample during deposition with radiation from the microwavedischarge, since a coaxial orifice discharge tube provided photolysis and produced the product species, while studies with an off‐axis orifice discharge tube, which could not serve as a photolysissource, did not produce the product. The H atom–Cl_{2}reaction gave HCl using both discharge tubes, but the HCl^{ x } _{2} species was produced o n l y with the coaxial tube. Hence, this species requires vacuum ultraviolet light in addition to H and Cl atoms for its production, and its identification as HCl^{−} _{2} is strongly supported. Similar observations of Ar_{n}D^{+} and O^{−} _{3} with the coaxial tube, but not with the off‐axis tube, confirm their identification as charged species.

Penning ionization of He(2 ^{3} S)+Ar
View Description Hide DescriptionA new measurement of the differential cross sections for He(2 ^{3} S)+Ar has been carried out at 65 and 132 meV. A potential V _{0}−i V _{ i } is found which fits our data as well as the total ionization cross section results of Illenberger and Niehaus. Some sources of uncertainties inherent in our analysis are discussed. The first order semiclassical approximation, which has been employed by several groups, is critically examined. A guideline is proposed to define the region within which this approach can be safely adopted.

A semiclassical theory for spectral line broadening in molecules
View Description Hide DescriptionA semiclassical S‐matrix theory is developed and applied to spectral line broadening in linear molecules perturbed by atoms. This theory uses curved classical trajectories determined by the isotropic part of the atom–molecule interaction and the S‐matrix is treated to all orders in the interaction. Numerical calculations can be made rather easily even for high quantum numbers. The theory is least accurate for very low quantum numbers, but even then calculations agree to within 10% with close coupling results where comparisons could be made. Comparisons were also made with other theoretical approaches using model potentials and with experiment using a b i n i t i opotential surfaces.

A temperature dependent kinetics study of the reaction of OH with CH_{3}Cl, CH_{2}Cl_{2}, CHCl_{3}, and CH_{3}Br
View Description Hide DescriptionReported in this study are temperature dependent rate data for the reaction of OH with the partially halogenated methane species CH_{3}Cl, CH_{2}Cl_{2}, CHCl_{3}, and CH_{3}Br. The nominal temperature range covered was 245–375 K. The appropriate Arrhenius expressions are k _{A}= (1.84±0.18) ×10^{−12} exp[−(2181±70/R T)], k _{B}= (4.27±0.63) ×10^{−12} exp[−(2174±161/R T)], k _{C}= (4.69±0.71) ×10^{−12} exp[(2254±214/R T)], k _{D}= (7.93±0.79) ×10^{−13} exp[−(1766±116/R T)]. Units are cm^{3} molecule^{−1}⋅s^{−1}. No simple correlations between E _{act} and C–H bond strengths were found. The impact of these halogenated species on stratosphericozone is also discussed.

Low‐temperature vibrational relaxation in gaseous deuterium fluoride: Monomer and polymer deactivation effects
View Description Hide DescriptionDeuterium fluoride vibrational energy transfermeasurements have been performed at 198, 209, 232, 264, and 296°K using the laser excited vibrational fluorescence technique. Rate contants for the following processes have been determined: HF(v=1)+DF(v=0) →_{ k } _{ e } _{+k } _{ 12 } HF(v=0)+DF(v=1,0) ÷ΔE=1055,3962 cm^{−1}, DF(v=1)+DF(v=0) →_{ k } _{ 22 }2DF(v=0)+ΔE=2907 cm^{−1}, DF(v=1)+Ar→_{ k } _{ 2m }DF(v=0)+Ar+ΔE=2907 cm^{−1}. Experimental results taken at 198, 209, and 232°K exhibited rate enhancements for increasing DF pressure which is most easily explained as HF(v=1) and DF(v=1) quenching by (DF)_{ n }. Relaxation rates in this regime were best described by the empirical equation τ^{−1}=C exp(D P _{DF}), where C and D are temperature dependent coefficients and P _{DF} is the total DF pressure, [DF+(DF)_{ n }]. The variation τ^{−1} with exp(P _{DF}) cannot be reconciled in terms of a monomer–dimer model only. Higher order terms (n=4 and 6) are expected to dominate the collisional quenching of HF(v=1) and DF(v=1). The characterization of (DF)_{ n } and its effect upon these studies are discussed in detail. Measurements in undiluted DF were also performed at 200 and 296°K. The room temperature results clearly indicate a slower DF self‐quenching rate than measured in DF–argon mixtures. This relaxation time lengthening can be explained by assuming that vibrational and rotational relaxation occurs on a similar time scale in undiluted DF.

Wigner phase space method: Analysis for semiclassical applications
View Description Hide DescriptionWe investigate the suitability of the Wigner method as a tool for semiclassical dynamics. In spite of appearances, the dynamical time evolution of Wigner phase space densities is found n o t to reduce to classical dynamics in most circumstances, even as h→0. In certain applications involving highly ’coherent’ density matrices, this precludes direct h‐expansion treatment of quantum corrections. However, by selective resummation of terms in the Wigner–Moyal series for the quantum phase space propagation it is possible to arrive at a revised or renormalized classicallike dynamics which solves the difficulties of the direct approach. In this paper, we review the Wigner method, qualitatively introduce the difficulties encountered in certain semiclassical applications, and derive quantitative means of surmounting these difficulties. Possible practical applications are discussed.

The use of electron paramagnetic resonance matrix spin flip satellite lines as a tool in the structural determination of paramagnetic centers in disordered solids
View Description Hide DescriptionIt is shown that measurement of the first moment of one proton spin flip satellite with respect to the EPR main line, when combined with other methods, leads to a determination of the effective number of near neighbor protons and an effective distance to these protons in disordered solids. The method is applied to existing data for hydrogen atoms trapped in irradiatedacids and results in n _{eff}(H_{2}SO_{4}) =3.1, r _{eff}(H_{2}SO_{4}) =2.06 Å and n _{eff}(H_{3}PO_{4}) =3.4, r _{eff}(H_{3}PO_{4}) =2.30 Å. In the case of one near neighbor proton, it is shown that the satellite line is always narrower than the main line. For Gaussian shaped lines, an explicit expression is given for the main and satellite linewidth difference in terms of the hyperfine coupling parameters.

The screened INDO (INDO/S) model: Application to photoelectron spectrum of benzonitrile
View Description Hide DescriptionA screened INDO model, which we previously developed for analysis of optical spectra and excited‐state properties of organic compounds, is applied to the prediction of photoelectron spectra. Results are quite satisfactory for benzene, and an assignment based on Koopmans’ theorem allows complete assignment, including previously unassigned peaks, in the benzonitrile molecule.

Dynamics of adsorption: Model calculations for several interaction potentials
View Description Hide DescriptionA model potential to describe the interaction of a diatomic molecule with a rigid solid surface has been derived previously. Using this interaction potential, classical trajectory calculations have been carried out to study the dynamics of adsorption on a solid surface. Several forms of the interaction potential were employed so as to observe the effect of different types of potential surfaces on adsorptiondynamics, and both H_{2} and HD were used in these model calculations in order to study the isotope effects on adsorption. Sticking probabilities for the different cases are reported and the validity of the rigid surfacemodel is discussed.

Relative quenching rates of NO(B ^{2}Π, v′=9)
View Description Hide DescriptionThe quenching rates of NO(B ^{2}Π, v′=9) have been measured for He, H_{2}, CO, CF_{4}, N_{2}, and CO_{2}, relative to NO self‐quenching rate, by noting the reduction in fluorescence intensity of the β (v′=9) bands. The values obtained are 1.0:0.12:0.13:0.13:0.14:0.17:0.18 for NO, He, H_{2}, CO, CF_{4}, N_{2}, and CO_{2}. Absolute quenching rates are also estimated to be close to hard‐sphere collision rate constants and the quantum yield of the β (v′=9) emission to be of the order of 10^{−2}.

Laser‐fluorescence study of the reactions of alkaline earth atoms with BrCN: Spectroscopic observation of the alkaline earth monocyanides
View Description Hide DescriptionThe reactions of M+BrCN, where M=Ca, Sr, and Ba, have been studied by laser‐induced fluorescence. The dominant reaction pathway is the formation of the alkaline earth monocyanides (MCN); for the Sr and Ba reactions weak features due to products of the other possible pathway, MBr and CN, are also observed. This study represents the first spectroscopic observation of the MCN molecules. The dominant features of their excitation spectra are broad ’’quasicontinua’’ several hundred angstroms wide which occur at the same wavelengths as the spectra of the homologous alkaline earth monohalides (MX). Thus, the electronic structure of the MCN molecules is qualitatively similar to that of MX, which has been adequately explained by an ionic model, as discussed previously by Dagdigian, Cruse, and Zare. Radiative lifetimes of the MCN electronic states have been determined by direct observation of the fluorescence decay after pulsed laser excitation. The lifetimes are: CaCN: B state 40.8±1.5 nsec, C state 165±38 nsec; SrCN: B state 51.2±6.2 nsec, C state 104.4±6.3 nsec; and BaCN: C state 229±13 nsec, where the quoted uncertainties represent three standard deviations. The MCN and MX B state lifetimes are similar in magnitude, but the MCN C state lifetimes are considerably longer than the homologous MX C ^{2}Π lifetimes. One possible interpretation is that the monocyanide excited C states, unlike those of the halides, are not simply described by one electron excitations into nonbonding metal atom orbitals but also include some excited state CN^{−} character.

Time and space dependent energy distributions of atoms from surfaces
View Description Hide DescriptionTime and space dependent energy distributions for atoms effluxing from a surface are obtained analytically using the harmonic soft‐cube model. The effluxing atoms consist of a desorbed fraction and a reflected fraction. The tangential energy distribution of the desorbed fraction is assumed to have equilibrated with the surface, while that of the reflected fraction is assumed unchanged. The normal energy distribution of the desorbed fraction depends on the adsorbed atom energy distribution obtained here from the governing master equation without the steady state approximation. The normal energy distribution of the reflected fraction is obtained from the soft‐cube model for the collision dynamics. As expected, due to the gas–surface potential well, the desorbed fraction of the flux from the surface has a noncosine distribution. The reflected fraction has a lobular distribution about the specular angle. Time dependent scatteringpatterns are obtained as a function of the interacting atomic species, the incident energy distribution, and the gas and surface temperatures. Comparison with experimental desorption and adsorption rates and with scatteringpatterns from clean surfaces is possible.

Quantization for lattice‐gas models of water
View Description Hide DescriptionSeveral recent studies have shown that some of the unusual properties of liquid water can be reproduced with a classical lattice‐gas model, wherein the host lattice is body‐centered cubic. In this paper we quantize both rotational and translational motion in those models, using suitable hopping operators. Several alternative forms are possible for the rotational kinetic energy; we provide comparisons for each with the experimental spectrum. Variational calculations have been performed for (H_{2}O)_{2}, (D_{2}O)_{2}, and (T_{2}O)_{2}ground states on the lattice to estimate hydrogen bond destabilization by zero‐point motion. Finally, expressions have been developed for thermodynamic‐property and distribution‐function quantum corrections that should be useful in classical lattice−gas simulations of water via computer.

Valence bond calculation of nuclear spin–spin coupling constants. I. The distinguishable electron method
View Description Hide DescriptionThe uncoupled valence bond ’’distinguishable electron’’ method (DEM) is extended to the nuclear spin–spin coupling property. Provision is made in the theory for the triplet spectrum to have a different zero order potential from that of the ground state singlet. The case of a strong interaction with a low lying triplet is treated explicitly. As an example, we carry out a calculation on HD in the simple screening approximation using identical singlet and triplet screening constants. For this zero‐order potential the one electron DEM equations are exactly soluble. Furthermore, the major effect due to splitting of the screening constants is readily determined. Our value for the HD coupling constant is in good agreement with experiment.

Energy flux of hot atoms
View Description Hide DescriptionThe process in which hot atoms collide with thermal atoms of a gas, transfer kinetic energy to them, and produce additional hot atoms is investigated. A stochastic method is used to obtain numerical results for the spatial and time dependent energy flux of hot atoms in a gas. The results indicate that in hot atom systems a front followed by an intense energy flux of hot atoms may develop.

The heats of transition of solid ethane
View Description Hide DescriptionThe existence of a solid–solid transition in ethane is now generally accepted. A value for the heat of transition, 2437±35 J/mol at 89.774 K, together with a new value for the heat of fusion, 279±6 J/mol at 90.337 K, is reported in this work. The sum of the new values is in excellent agreement with the earlier measurements, provided that the heat required to raise the temperature of the solid between transitions is included. The transition temperatures are in agreement with the most recent determinations of other authors.

Rotational diffusion of spheroidal molecules
View Description Hide DescriptionBerne’s formalism for calculation of the correlation function of the permanent dipole moment of a sphere of linear molecules imbedded in its own medium is applied to spheroidal molecules. It is found that distinct parts of the correlation function are associated with the individual dipole moment components but the constraint imposed by the geometry does not prevent an interaction between components which affects the dynamic response. The calculation provides an algorithm for the ellipsoidal case (generalization of the Perrin model) and gives limiting cases relevant to dielectricexperiments to which the ellipsoidal calculation must conform.

The electronic structure of molecules by a many‐body approach. I. Ionization potentials and one‐electron properties of benzene
View Description Hide DescriptionThe ionization potentials of benzene are studied by an a b i n i t i o many‐body approach which includes the effects of electron correlation and reorganization beyond the one‐particle approximation. The calculations confirm the assignment of the photoelectron spectrum experimentally proposed by Jonsson and Lindholm: 1e _{1g }(π), 2e _{2g }, 1a _{2u }(π), 2e _{1u }, 1b _{2u }, 1b _{1u }, 2a _{1g }, 1e _{2g } in order of increasing binding energy. To definitely establish the ordering of the ionization potentials in the second band, which has been very controversial, the corresponding vibrational structure has been calculated. A number of one‐electron properties are calculated in the one‐particle approximation and compared to experimental work and other theoretical calculations.