Table of contents

Gryzinski's (1965;72) free-fall atomic model is tested by being used in classical calculations on electron capture in fast collisions between protons and hydrogen atoms. It is found to be unsatisfactory.

The scattering of electrons by cesium atoms is calculated in the two state close coupling approximation for incident electron energies below 5 eV. The shape of the calculated total cross section agrees well with experiment, in contrast to earlier work, but lies some 20% higher. The occurrence of a narrow ³F shape resonance at 1.7 eV and cusps at the 6p_1/2 and 6p_3/2 thresholds are discussed.

It is shown that the capture process is unimportant in ionizing collisions of electrons with helium atoms, in the range of validity of Born's approximation, except in predictions of the triple differential cross section in regions of energy and angle where the exchange contribution is small.

The authors point out a major discrepancy between the Faddeev-Watson multiple scattering expansion and the Born multiple scattering expansion for the elastic scattering of electrons and positrons by atomic hydrogen and give arguments in favour of the Born approach.

The electron loss and capture cross sections for multiply charged He, N, Ne and Ar ions at velocities V from 3*10⁸ to 10⁹ cm s^-1 in helium, nitrogen, neon and argon have been measured. It has been found that due to a higher average velocity of outer electrons in Ne atoms the dependences of these cross sections on nuclear charge of target atoms at V approximately 10⁹ cm s^-1 become nonmonotonic: the values sigma _i'i+1 in neon become smaller and sigma _i,i-1 greater than in nitrogen and argon.

A method for determining the number of inner shell vacancies associated with unresolved inelastic energy loss spectra is discussed. This consists of using atomic total energy calculations together with the data on total ionization to reproduce the energy losses observed in scattering experiments. The method is applied to existing data for Ar-Ar collisions at various energies. Several features in the data can be interpreted in terms of a molecular orbital model for the collision.

The experimental technique of measuring cross sections for momentum transfer between ions and atoms by observing hydromagnetic waves is greatly simplified if standing rather than propagating waves are used. Measurements are reported of Ar⁺-Ar momentum transfer cross sections in the energy range 4-7 eV using standing magnetoacoustic waves.

The effect on a simple two-level atom of an electromagnetic field having an arbitrary distribution of photon occupation numbers is considered. It is shown that atomic energy levels shift of the order of magnitude of the Bloch-Siegert shift can be produced by interference between different states of the field. The interference term is similar to a phase term which appears in the calculation of the classical Bloch-Siegert shift. It is predicted that in general this term will average to zero, thus permitting interference effects between states of the field to be neglected.

Anharmonic couplings via Fermi resonance are included in the strong coupling theory of the v_o-H...o modes of acetic acid dimers. The effects of these resonance are discussed, and another possible origin for the difference in the spectra of (CD₃COOH)₂ and (CH₃COOH)₂ is suggested.

The one-dimensional form of the correspondence principle for strongly coupled states of Percival and Richards (1970) is modified to include, approximately, the perturbation of the bound particle orbit. The modification will only be significant when the expectation value of the interaction potential depends on the state of the bound system. The excitation of a linear harmonic oscillator by a time dependent potential q²F(t) and the collinear collision of an atom and an anharmonic diatomic molecule are considered as examples. The results obtained for both these systems are in good agreement with the exact quantum mechanical transition probabilities even for low order transitions.

Relativistic orbitals have been used to derive the matrix elements for the photo-detachment of electrons from atoms and ions. The asymmetry parameter beta is shown to deviate from the values predicted by the non-relativistic theory, particularly for levels split by spin-orbit coupling. These effects will be largest at the Cooper minima of photoionization spectra.

In the distorted wave approximation, radial functions F_i for the colliding electron are calculated using a central potential V_i(r). The Kohn variational method is then used to calculate the R matrix; this is the matrix, used in quantum defect theory, which varies slowly as a function of the energy. For closed channels the function F_i are calculated at the eigen-energies for the central-potential problem, and suitably normalized. The resonance structures are calculated using methods of quantum defect theory. Illustrative results are obtained for the 2s²¹S-2s2p³P transition in C III and in O V, and compared with results obtained from exact solutions of coupled integro-differential equations. It is concluded that, for the calculation of rate coefficients for the excitation of more highly ionized positive ions: (i) it is necessary to take resonances into account and; (ii) good results can be obtained using the distorted wave method.

The author reports a theoretical study on the single direct ionization of He by He⁺ impact at moderate energies. Following Barat et al (1972) this process is considered as arising from a molecular autoionization of the (1s sigma _g2p sigma ²_u)² Sigma ⁺_g diabatic state of He₂⁺ as it crosses the boundary of the continuum (He₂⁺+e^-). First the molecular orbitals involved in the problem are expanded in terms of Slater type orbitals centred centred at the midpoint of the internuclear distance of He₂⁺. The resulting wavefunctions together with a Coulomb description of the ejected electron lead to an expression of the autoionization width as sums over hypergeometric functions ₂F₁. This width is subsequently involved in the coupled equations for the nuclear motion which are solved in a distorted wave-(WKB)-semi classical approximation. The computed oscillatory energy distribution of the ejected electrons is in satisfactory agreement with the experimental results of Barat et al.

Differential scattering measurements of ionization in He-He collisions are made in the keV energy range using the energy loss technique. Combining the results with those obtained from the ejected electron spectra measurements of Gerber et al., the ionization process is discussed in the framework of the quasimolecular picture of the collision. Two-step mechanisms are proposed involving a primary radial or rotational coupling either to a molecular state which autoionizes during the collision or to a state which leads on dissociation to the formation of He^- in an auto-ionizing state. Calculations taking account of rotational coupling effects between the ¹ Sigma _g, ¹ Pi _g and ¹ Delta _g states of He₂ are in satisfactory agreement with experiment for energy E and scattering angle theta when E theta exceeds 3 keV deg.

The transition amplitude corresponding to multiphoton absorption resulting in simultaneous ionization and excitation of the helium atom in the presence of an intense laser beam has been evaluated by using the method of Reiss (1970). Results show that reduced transition amplitude varies linearly with initial value of the intensity but for high intensity the slope of the curve decreases, showing an important nonlinear effect. It is found that at the intensity increases the maximum number of photons taking part in the process also increases. Another important feature found is the dominance of the multiphoton ionization process over the multiphoton excitation process at high intensities.

The continuous oscillator strengths for photoionization of the 1¹S state of helium are evaluated, in the dipole length and velocity formulations for transitions which leave the He⁺ ion and a 1s, 2s or 2p state. The calculations are performed with two sets of wave-functions; calculation A uses frozen core Hartree-Fock bound and free state functions; calculation B uses a 56-parameter bound state representation with frozen core Hartree-Fock free state functions. Comparison with the results of Jacobs and Burke (1972) indicates the energy ranges in which more accurate free state wave functions should be used.

The first Born approximation is used to calculate cross sections for excitation or de-excitation to discrete states from fast metastable hydrogen atoms passing through atomic helium. Full account is taken of double transitions and the calculations are performed using length and velocity formulation values of the Born helium matrix elements. Close coupling values in the impact parameter theory for the specific transitions 2s to nl(nl=1s, 2p, 3s, 3p) in hydrogen with the target helium atom unexcited indicate that coupling to excited states and to the hydrogen continuum is not important for these particular destruction processes for impact energies greater than 10 keV. Addition of the present Born approximation results to those of Bell and Kingston (1971) for the electron loss contribution provide total destruction cross sections which are a factor of about 2.5 lower than the experimental data (5-30 keV) of Gilbody et al (1971). For electron loss, the recent data of Hughes and Choe suggests that the first Born approximation may be satisfactory for impact energies greater than 100 keV.

The first Born approximation is used to calculate total cross sections for excitation of the 2s and 2p states of hydrogen by collision with helium. Full account is taken of double excitation processes and the calculations are performed using both length and velocity formulation values of the helium matrix elements. An impact parameter close coupling treatment of the particular process in which the helium atom remains unexcited is carried out with the inclusion of pseudo excited hydrogenic states. The first Born approximation values are in good accord with experiment for energies greater than 10 keV and 30 keV for 2s and 2p excitation, respectively, and, for the lattice case, are in excellent agreement with experiment in the region 40-120 keV where the cross section is almost constant due to the interplay of non-excitation and excitation of the target. At low impact energies, the pseudo-state calculation is an improvement upon previous 4-state calculations but still fails to agree both in shape and in magnitude with experimental data.

The second order diagonalization method is used to study proton-helium scattering. The results are compared for 1¹S-2¹S and 1¹s-2¹P cross sections with various other theoretical calculations and with the experiment of Park and Schowengerdt (1969). For higher levels, values are compared with experiments. A very good agreement is found for n¹S and n¹P states (n<or=5). The discordance for the n¹D states is discussed. Polarizations of n¹P-2¹S, n-1D-2¹P and 4¹F-3¹D lines are also presented.

The excitation of helium by 20-100 keV He⁺ and He beams has been investigated by the spectroscopic analysis of radiation emitted in the spectral range 3900-6000 AA during the spontaneous decay of the excited products of single collisions. Observations have been made at 54.5 deg with respect to the primary beam to permit a Doppler shift separation of target and projectile emissions. Absolute cross sections for the excitation of 4¹S, 4¹D, 4³S and 4³D levels have been determined. Cross sections for excitation by He⁺ and He impact have been compared with a relative accuracy of +or-12%. Measurements have also been arried out with He beams containing different known fractions of metastable atoms. These data have been used to determine approximate cross sections at 70 keV for the processes: He(2³S)+He to He(4³S or 4³D)+He and He(2³S)+He to He+He(4³S or 4³D).

Electron-loss cross sections sigma ₀₁ for ground-state lithium atoms traversing hydrogen, helium, nitrogen, neon, argon, krypton, and xenon within the range of impact energies from 20 to 500 keV have been measured by beam attenuation in a transverse electrostatic field. These experimental cross sections for loss of the loosely bound valence electron in neutral lithium are compared with the theoretical estimates, based upon the semi-empirical classical impulse approximation of Bates and Walker (1966). Further, the present cross sections for lithium atoms are compared with published experimental electron-loss cross sections for metastable hydrogen and helium atoms. On the basis of the theory mentioned above, similarities between these cross sections are expected. The overall agreement between the cross sections presented and the theoretical estimates, as well as the experimental electron-loss sections for metastable hydrogen and helium atoms is reasonable. A more detailed comparison is, however, made difficult by the fact that results from different laboratories in some cases show poor agreement.

A method is described for calculating the relative absorption oscillator strengths of spectral lines of an element from the gradients of their working curves in emission spectral analysis. A mathematical expression for the gradient of the working curve is derived taking into account the atomic self- absorption of the spectral lines in the cooler region of the discharge and assuming a Doppler broadening line profile for both emission and self-absorption. Local thermal equilibrium is assumed in the absorption region in the arc fringe, and the principle of detailed balance is applied to that region for evaluating the relative population of the lower excited states of the atoms. Correlation between theory and published experimental results for the average gradient of the working curves for 14 lines of Fe I lead to calculation of their relative oscillator strengths as well as to some informations of the average temperatures of the emission and absorption regions. The good agreement between the calculated gf values and those determined by other investigators emphasizes the validity of the assumption of local thermal equilibrium in the arc fringe.

The representation of spectroscopic transition amplitudes as a sum of multipolar interactions with the radiation field in the presence of spin-orbit coupling is investigated. It is found that electric dipole, magnetic dipole and electric quadrupole interactions can be invoked if matrix elements are taken between molecular states that are eigenstates of a Hamiltonian which includes spin-orbit coupling. A minor new term that contributes to singlet-triplet transition intensities is introduced.

Self-consistent field calculations have been made of the potential energy curves for three states of HCl⁺, using a basis set believed to give energies very close to the Hartree- Fock limit. The curve for the ⁴ Pi state, dissociating to H(²S) and Cl⁺(³P), crosses that of A² Sigma ⁺ state. Since these states may interact by spin-orbit coupling, a predissociation may occur in the A² Sigma ⁺ state. Calculations of the magnitude of this effect for both HCl⁺ and DCl⁺ gives results compatible with observed emission and photoelectron spectra.

A new method of experimental data processing is proposed for the analysis of interactions between two levels. From the energies W(J) and with the help of a computer, it gives the values of the deperturbed constants and the dispersion. It is useful when only one of the states is observed, and particularly if the energy levels near the perturbation centre are not known or if they are observed before or after this centre only. From expressions derived from W(J) and its derivatives, the various coefficients are estimated by an exploration method in a field limited by the greatest and the lowest values taken by the quantities ¹/₂(T_A+T_B) and ¹/₂(B_A+B_B). The knowledge of the coefficients involves that of ¹/₂(T_A-T_B) and ¹/₂(B_A-B_B). The quantities T_A, T_B, B_A, B_B being known, the interaction parameter H_AB is calculated from W(J). At last, when possible, the coordinates of the perturbation centre are obtained by using symmetry property of curves. Two examples are given: the C² Delta _5/2 state of AsO and the upper level of the ² Sigma -² Sigma transition of LuO.

Calculations have been performed for the dielectric second virial coefficient B_epsilon of the polar gases CF₃H and CH₂:CF₂ by using Buckingham and Pople's theory (1955). For this purpose, the contributions of a large number of non-spherical interaction terms have been included. The results of the calculations have helped in clarifying some of the inconsistent results obtained by previous workers.

The authors have measured the lifetimes of the first two vibronic levels of the b³ Sigma ⁺ state of carbon monoxide using the inelastic electron-photon coincidence technique. Their respective values are 53.6+or-0.3 and 69.1+or-2.1 ns. In addition it is found that the lifetime for the v=1 level decreases with increasing pressure, showing a large cross section for collisional transfer of excitation. It is considered that the lifetime of the v=1 level may be affected by the perturbations known to exist in its rotational levels.

The red degraded band of MgO at 2637.5 AA has been assigned as the O-O band of the electronic transition F¹ Pi -x¹ Sigma of MgO. The initial state F¹ Pi is a new electronic state of MgO and is located at 37879 cm^-1 (T₀₀). A number of intensity perturbations have been observed. The rotational constants of the F¹ Pi state are (in cm^-1): B₀=0.5590; D₀=1.42*10^-6; q=0.0001. A few weak bands give a tentative vibrational frequency of 710 cm^-1 for the F¹ Pi state.

The spectrum of the SrI molecule has been excited in a high frequency discharge source and photographed in the fourth order of a plane grating spectrograph at a reciprocal dispersion of 1.8 AA mm^-1. The existence of a new band system in the region lambda 3350- lambda 3560 AA has been established and the analysis has been carried out. The nature of the electronic transitions involved have been discussed.

During the last two or three years many people have studied the importance of electronic correlations for the scattering of X-rays and fast electrons by atoms. The present authors have calculated the increments to the mean values of the operators which appear in the scattering phenomena using the atomic Bethe Goldstone equations which give the total correlation energies range from 98.5 to 100.3% of the empirical correlation energy. The results are compared with those obtained by several authors. It is concluded that it is necessary to take account of the electron correlations for comparison between theory and experiment.