Table of contents

Conclusions reached by Urata and Watanabe (ibid., Vol. 4, L125, 1971) concerning the validity of Pradhan's impulse approximation are refuted.

Secondary electron velocity distributions have been obtained from the back surfaces of thin carbon foils during light ion bombardment. Small peaks in the spectra can be assigned to autoionization of transmitted ions or atoms in the beam, the energies of the electrons with respect to the beam being quite low (0.5-15eV). A simple method of measuring the beam velocity after the foil is described and the possibilities of the beam-foil technique for studying autoionization are evaluated.

The discrepancy between theory and experiment for the plasma broadening of the dipole-forbidden components of spectral lines is discussed. A calculation using a new formalism which rigorously includes the ion-dynamic perturbations is made. There is a significant improvement in the agreement with the experimental data.

Time dependent Hartree-Fock theory, in its coupled and uncoupled forms, is used to calculate the elastic singlet p-wave phase shifts for the scattering of electrons by H and He⁺. On comparison with the best available results it is concluded that the coupled scheme, which contains correlation to a least first order, is superior. Levinson's theorem is confirmed.

In atomic scattering calculations, certain difficulties stem from the use of inexact wavefunctions. These can be avoided, provided the projectile is distinguishable from the atomic electrons, by using modified forms of the atomic hamiltonian. A connection with the positron scattering work of Peterkop and Rabik (ibid., Vol. 4, 1440, 1971) is established.

In a recent paper by Frisch and Brissaud (ibid., Vol. 4, 1230, 1971) certain aspects of the unified classical path theory for Stark broadening are criticized; since it is felt that their results could be misleading. Some comments on their paper are offered.

In calculating the static Stark effect, when limiting the number of perturbing levels to those with the same principal quantum number, one introduces negligible errors on the peak separation between the allowed and forbidden lines. On the other hand, one may affect the profile in the region of the forbidden component. In the case of CsI(6G-5D), it is shown, by calculating the ionic profiles, that this limitation contributes to overestimate the maximum intensity of the (6G-5D) forbidden line and to reduce the intensity between the allowed (6F-5D) and the forbidden (6G-5D) lines.

The dissociation energy of the IO molecule has been estimated from the R-K-R curve; the result is in extremely good agreement with experiment.

The method adopted hitherto for measurement of the L₂ subshell fluorescence and Coster-Kronig yields omega ₂, f₂₃, in the region Z>or=90 is shown to be extremely sensitive to the value assumed for the intensity ratio L_alpha /L_beta of x-rays arising from L₃ vacancies; a recent result for f₂₃ is altered by a factor of 3 when the Hartree-Slater value of this ratio is used. More direct methods of studying L₂ subshell yields are proposed.

The electron impact excitation function of the 1048.2 AA and 1066.7AA lines resulting from the decay of the first excited state in argon have been measured from threshold to 180 eV. The data are compared with the plane and distorted wave Born approximations.

A core model, consisting of a (12-4) central potential displaced from the origin, is suggested as a representation of the interaction of polyatomic ions with neutral molecules. The diffusion collision integral, which describes ion mobility, is computed and tabulated as a function of temperature and core size. The addition of the core reduces the maximum in the mobility against temperature curve, and eventually reduces the mobility below its polarization limit at all temperatures. These results are in accord with limited available experimental data. Comparison is made with other models of ion-neutral interactions.

The exchange interaction between the ionic and the covalent states in the system of two atoms is considered at large internuclear distances. This interaction determines the nonadiabatic transition probabilities in the systems such as A*+B or A⁺+B^-. An asymptotically exact expression for this interaction is obtained accounting for the long range (polarization) interaction in the system.

Electric quadrupole multiplet strengths have been calculated for a large number of transitions in He and Li⁺ using analytical orbital wavefunctions of frozen core type. Very good agreement between length and velocity forms of the transition matrix element is obtained when the orbitals have the correct asymptotic form. The velocity form is far less sensitive to small changes in the orbitals than is the length form. The bound-state contributions to the quadrupole polarizabilities of the ions He through B⁺³ have been estimated from ¹S-¹D multiplet strengths, and it is concluded that the continuum contributions are very large.

Improved lower bounds to the energy levels of the 1s ns ¹S states of helium through n=6 have been calculated with analytical orbital wavefunctions using an extension of the methods of Cohen and Feldmann (see abstr. A51946 of 1970). For the excited states, these bounds are all within 1.25% of the observed levels, and the error reduces to 0.15% for the highest levels. These results confirm the reliability of the frozen core wavefunctions for excited states.

The variation of the transition moment, R_e(r), with the internuclear separation for the comet-tail band system of CO⁺ has been studied by the method developed by Jain and Sahni (1968). The expression for R_e(r) has been used to calculate the Einstein A coefficients, the band oscillator strengths and the absolute band strengths for this system.

The Lambda doubling constants of the ² Pi states of OH, BeH, CH and NO are calculated from Hartree-Fock wavefunctions to test the validity of Van Vleck's hypothesis of 'pure precession'. Excellent agreement with experiment is found for OH and BeH while for CH the discrepancies are relatively small. In the case of NO, however, no agreement is found and it is concluded that the hypothesis only holds when the molecular situation resembles that in atoms.

Variational calculations are presented from the n³P₂-1¹S₀ magnetic-quadrupole decay rates, with n=2,5, for members of the helium isoelectronic sequence up to Z=10. The magnetic-quadrupole decay rates are compared with the competing n³P₂-2³S₁ electric dipole decay rates to determine the importance of the magnetic- quadrupole transitions in the radiative decay of the n³P₂ states.

Hartree-Fock calculations of the s-electron density at the nucleus have been made for a selection of electron configurations in sm I-IV and Eu I-IV chosen for their interest in interpreting observed isotope and isomer shifts. Theoretical screening ratios and core relaxation effects are discussed and compared with those derived by empirical means.

The fine structure of the A ³ Pi and X ³ Sigma ^- states of ND, NH and OH⁺ has been investigated. An iteration procedure based on numerical matrix diagonalization takes the ³ Pi -³ Sigma ^- interaction completely into account, and contributions from singlet states are considered to second order. Centrifugal distortion terms are included in the off-diagonal ³ Pi -³ Sigma matrix elements through a quantum mechanical approach, and these terms are found to yield the correct explanation for the Lambda doubling anomalies encountered for high rotational quantum numbers. The dependence of the coupling constant A on the rotational quantum number observed by previous investigators is removed, even for the highest rotational quantum numbers (J=30), through the inclusion of higher order terms and the spin-rotation interaction. The A ³ Pi state of AsD and AsH is briefly commented upon, and the relation A=A₀+2A_JJ(J+1) is found to explain certain anomalies in the multiplet splitting. For the ³ Sigma ^- state the unified treatment of the ³ Pi -³ Sigma ^- interaction leads to some further conclusions regarding the origin of the splitting.

The author is concerned with the calculation of the perturbed orbital wavefunction chosen to give the full polarizability of the ground and excited states of a complex atom. A new numerical technique is described and is tested on the hydrogen atom, where all polarized wavefunctions are analytic and known exactly. The accuracy obtained covers at least four significant places. Applications of these pseudo-states to the modified close coupling expansion are considered elsewhere.

The usefulness of Harris' equation (see abstr. A38796 of 1968) in the semi-empirical LCAO-MO approximations are discussed. On some bases the discussion leads to the fact that in such procedures it is of little importance.

By approximating Wigner 3-j and 6-j coefficients for the coupling of large angular momenta, simplified expressions have been obtained for angular distributions in the excitation of rotational states in resonant electron molecule reactions. These simplifications have given rise to a useful classification of rotational transitions of linear molecules into branches, each of which has a characteristic energy loss spectrum and angular distribution.

A theoretical model is presented for the interaction between an alkali atom in the first excited ²P state and a ground state rare gas atom, which is effective in causing transitions among the alkali fine structure states. The main difference from previously used models is a more detailed treatment of short range effects including deviations from straight line paths in the semiclassical treatment. An application to the case of Rb interacting with He, Ne, Ar, and Kr gives results for the depolarization and inelastic cross sections which are in reasonable accord with experiment.

Potential barriers in atomic interactions often lead to shape resonances in which atoms are held together by the barrier. The importance of these resonances in thermal atomic collision processes is assessed. Radiative association is discussed as an example, and numerical results for CH formation suggest that the presence of a potential barrier might enhance the association rate.

Cross sections for ionization of positive ions with configuration 1s²2s²2p^q by electron impact are calculated in the no-exchange Coulomb-Born approximation, contributions from ejection of both 2s and 2p electrons being included. Reduced cross sections for one and twice ionized systems, and for the limiting case of infinite nuclear charge are evaluated, enabling cross sections for other systems to be obtained by interpolation. Comparison with experiment is made where possible, and agreement is good in all cases at high incident energies. In some cases there is evidence that the effects of autoionization, not included in this work could be important at low energies.

Fredholm and Pade approximations taken to the second order in the interaction potentials are used to evaluate elastic and inelastic cross sections for a schematic model scattering problem involving a pair of coupled channels with square well potentials. It is found that resonances occurring in the elastic cross section which arise from the presence of bound states in the second channel are described quite well by the second order Fredholm approximation provided that the potentials are not too strong, while the (1,1) Pade approximation as well as the first and second Born approximations entirely fail to take account of these resonances.

The interatomic potential V(R) and hyperfine splitting constant a(R) for a hydrogen atom in interaction with a helium atom are calculated as functions of the internuclear distance R(3<or=R<or=12a₀) by a generalized valence bond procedure. Using these configurations, a separate perturbation calculation of the coefficients of the leading (R^-6) terms in the asymptotic expansion of V(R) and a(R) yields values in good agreement with previous estimates. Comparison with previous calculations of V(R) indicates that further investigation is required to establish accurate values in the vicinity of the potential minimum. The function a(R) is also found to exhibit a shallow minimum, and it is emphasized that calculations of a(R) can be inaccurate if the decision on which terms to include in the wavefunction is based on energetic considerations alone. The computed hyperfine shift at 50 degrees C is 10% smaller than the most recent experimental value, the agreement becoming closer with increasing temperature. Values obtained for the temperature derivative of the shift are consistent with the available experimental data.

The authors have calculated the polarization fraction of the Lyman- alpha radiation emitted in e^--H(1s) collisions using the Glauber, Born and various Vainshtein approximations, and have compared these predictions with observation. It is found that the Born, Glauber and all forms of the Vainshtein yield results which are very close to each other and to the data in the energy range 30 eV<E_a<700 eV. The comparatively small differences between the various approximations are detailed in the text; also, the reason for the close agreement of the predicted polarization fractions-although the computed 1s-2p total and differential cross sections are not so close-is explained. With respect to the Glauber approximation, the importance of computing the phase integral along a direction perpendicular to the momentum transfer q at each value of q, is stressed.

A semiquantal theory is applied to the examination of the excitation and ionization produced by the impact between two hydrogen atoms with incident energies E in the range 0.2<or=E<or=2000 keV. Simultaneous transitions in both the target and the projectile atoms are explicitly acknowledged. The cross sections calculated from the simple semiquantal formulas exhibit remarkable harmony with previous quantal calculations. Also, in certain cases, the velocity distribution of the orbital electron can be neglected and the hydrogen-hydrogen excitation cross sections reduce to simple algebraic expressions while the ionization cross sections involve the numerical evaluation of a one-dimensional integral only.

The distortion method, derived previously, is briefly summarized for the case of partial waves. further, it is applied with practical trial distortion functions to the s-wave electron-atomic hydrogen scattering and to the scattering of particles by the exponential potential. A comparison with other methods is made. The author's method, with a simple trial function, is found to be preferable both for its results and its simplicity. The variational principle for the so called distortion matrix element is derived and the identity which is a counterpart to the ordinary Kato identity is applied to improve the results. The results are very good, especially for the exponential potential scattering, and require considerably less labour than the Kohn procedure.

A simple resonance theory of three-body ion-atom association reactions is presented. The reaction is considered as proceeding through the formation of a long lived orbiting resonance complex between the atom and the ion. The population of these quasibound states is estimated assuming thermal equilibrium. A stable molecular ion may then be formed upon deactivation of the complex by collision with a third body. Various simplifying approximations to the potential curves and surfaces are employed. For example, the orbiting complexes are determined by assuming that the long-range R^-4 interaction, which is calculated from the atomic polarizabilities, is not important. Furthermore, the deactivation cross sections for the relevant complexes are estimated from the corresponding atomic 'sizes'. A simple analytical formula for the three-body rate constant is thus derived. Reasonable agreement with experiment is obtained for He⁺ in He and fair agreement for other light systems.

The 3s² 3pⁿ-3s 3pⁿ⁺¹ and 3pⁿ-3p^n-1 3d transitions in Ni XI to XVII and Co X to XVII are classified as a result of the analysis of the spectra emitted from a theta-pinch plasma containing small percentages of nickel or cobalt carbonyl. The nickel lines account for a number of lines in the solar spectrum between 140 AA and 300 AA. The observations check the identity of six coronal forbidden lines.

High resolution spectral profile measurements have been made of cw amplified spontaneous emission from xenon at 3-5 mu m, and of radiation from a non-resonant feedback laser.

The authors report an application of the pulsed magnetic field technique to the study of resonant inter-atomic collisions on the decay of alignment ( gamma ₂) of the 6³P₁ state of mercury. Resonance fluorescence scattered at right angles from atoms near the front face of a cell is observed; the depolarizing effect of multiple scattering is thereby reduced, allowing studies of same-isotope collisions to proceed to higher density. For this geometry the effect of wall collisions on gamma ₂ is not large. Magnetic scanning of the lamp is employed, firstly, to determine the isotopic abundance in the cell, and, secondly to enable study of both 'same-isotope' and 'different-isotope' collisions using the same lamp and cell. The results for different-isotope collisions confirm those obtained by others. For Hg¹⁹⁸-Hg¹⁹⁸ collisions, the contribution to gamma ₂ is found to be (4.7+or-0.5)*10^-9 N cm²s^-1, in satisfactory agreement with the theoretical value.

Hyperfine structure interval in the 4d⁵5s ⁷S₃ atomic ground state of ⁹⁵Mo and ⁹⁷Mo have been measured by the methods of atomic beam magnetic resonance. A theory of quadrupole interactions in half filled atomic shells, known to be successful in several other cases, has been used to obtain the nuclear quadrupole moments from the B values with the results Q(⁹⁵Mo)=-0.075(25) barns and Q(⁹⁷Mo)=+0.15(4) barns. The atomic beam was produced by collimating atoms evaporating from a 2 mm diameter molten ball of molybdenum suspended on the end of a 0.3 mm diameter molybdenum wire.

The pressure broadening and shift of the 1.15 mu m spectral line of Ne²² have been measured and used to determine the constants in the Lennard-Jones interatomic potential. The value of the attractive van der Waals constant is about twice the predicted value, which is consistent with the results of another investigation involving other lines; the repulsive constant determined from the measurements agrees with a calculated value.

Intensity measurements on bands of the second positive system of nitrogen have been made over a wide range of conditions in the position column of a nitrogen discharge. The measurements have been used to obtain relative population of the vibrational levels of the C³ Pi _u state. Calculated excitation probabilities are examined for some of the possible excitation processes contributing to these populations. As a result of using these in conjunction with the measurements, it is suggested that in a positive column, the main process for populating the C³ Pi _u state is direct excitation by electrons from the ground state (x¹ Sigma _g), the part played by other processes being very small.

The vibration-rotation spectrum of ammonia gas in the visible region of the spectrum was photographed on a 1.5 m concave grating spectrograph. A White type multiple reflection tube with an effective path length of 0.5 Km was used. In this region ten new absorption features were observed and vibrationally analyzed. A reassignment of vibrational quantum numbers of six previously reported bands are made. An identification table of the forty-two vibration-rotation bands now known is given. The set of binary anharmonicity constants for overtone and combination bands was determined from the data of this table.

A systematic study of screening doublets in X-ray satellite (nondiagram lines) spectra is made. Many new screening doublets are reported for the first time. The screening doublets already reported in the literature are confirmed and finally established. A consolidated report of screening doublets is given from 1931 to the present time.