The dielectronic recombination cross sections associated with the 2s→2p transition in the Li-like ions ${\mathrm{B}}^{2+}$, ${\mathrm{C}}^{3+}$, and ${\mathrm{O}}^{5+}$ have been calculated in the distorted-wave approximation. We compare calculated cross sections to selected doubly excited configurations of the type 2pnl in ${\mathrm{C}}^{3+}$ in pure LS coupling, intermediate coupling, and a configuration-average approximation. The LS-coupled results are about 33% low as compared to intermediate coupling for low and intermediate values of n and l, while the configuration-average approximation agrees quite well with the intermediate-coupled results for relatively low values of n or l, but tends to overestimate the intermediate-coupled cross sections at intermediate values of n or l. All three methods should agree at high values of n or l. In addition, we calculate the total cross sections associated with the 2s→2p excitation for all three ions using the configuration-average approximation, and we compare our results to merged electron-ion beam measurements of the cross sections for ${\mathrm{B}}^{2+}$ and ${\mathrm{C}}^{3+}$. The sensitivity of these cross sections to experimental conditions is also examined.

A theory has been formulated to characterize charge-transfer collisions in the presence of an external laser field. The molecular-state-expansion method is used to describe the scattering process within the impact-parameter formalism. Electron translation factors are included in the molecular-state expansion so that the scattering wave function satisfies the correct boundary conditions. The theory is applied to the process ${\mathrm{K}}^{+}$+Na→K+${\mathrm{Na}}^{+}$. In addition, we have made a detailed analysis of laser-assisted charge transfer for low-energy collisions. In this case, a Landau-Zener formula can be derived which shows that the cross section increases with decreasing incident energy. In general the laser coupling is dominant in the low-energy region, while the dynamical coupling becomes important as the collision energy increases.

The simple application of the radial frame-transformation theory to near-threshold rovibrational excitation of molecular hydrogen by electron impact results in integrated cross sections which show anomalous structures and which fluctuate about the exact laboratory-frame close-coupling results as the transformation radius is taken inside the short-range-interaction region. On the other hand, if the transformation point is pushed to an approximate infinity, according to conventional adiabatic-nuclei theory, the cross sections have to be adjusted by a kinematic correction factor to acquire the correct threshold behavior. The apparent aggravation of these deficiencies as the incident energy decreases suggests that the problem can be traced to the definition of the fixed-nuclei channel wave number, and consequently, to the mismatch of the different channel Green’s functions at the transformation point. To partially correct for these deficiencies, the frame transformation is applied within the spirit of the energy-modified adiabatic theory. In the near-threshold regime, it is shown that the anomalous structures practically vanish, and that the resulting cross sections display a smooth dependence on the transformation radius. Furthermore, with the transformation performed at infinity, the cross sections are in better agreement with the laboratory-frame results than are the conventional adiabatic-nuclei values.

A unified treatment of slow ion-atom and atom-atom collisions for many-electron systems is proposed. The time-dependent electronic wave function is expanded in terms of traveling atomic orbitals of the two collision centers at large internuclear separations and is matched to the solution in the inner region where it is expanded in terms of molecular orbitals without translational factors. It is shown that this method avoids the difficulties associated with the conventional perturbed-stationary-state approximation. Application to charge transfer in the ${\mathrm{H}}^{+}$+${\mathrm{He}}^{+}$ collision is illustrated.

Positronium (Ps) with kinetic energies from 10 to 500 eV is formed when keV-energy positrons are partially transmitted through a 50-Å carbon film in vacuum. A time-of-flight measurement shows that the number of triplet Ps atoms formed per incident positron and moving in the forward direction (θ=0) is approximately dN${=(10\mathrm{}}^{\mathrm{-}3}$ eV${)}^{1/2}$${\mathrm{E}}^{\mathrm{-}3/2}$dE dΩ, where dΩ is the detector solid angle in steradians and the Ps with kinetic energy in eV is in the range (E, E+dE). The total amount of Ps with kinetic energy greater than 10 eV is approximately 1/2% of the incident positron flux times the detector solid angle. A measurement of the Doppler shift of the singlet-Ps-annihilation photons shows that the fast Ps with E>20 eV has an angular distribution with ${\theta }_{\mathrm{rms}}$=[(15 eV)/E${]}^{5/8}$ approximately.

The angular distributions of the ${}_{}{}^1{}_{}{}^{}\mathrm{P}$, ${}_{}{}^3{}_{}{}^{}\mathrm{P}$, and ${}_{}{}^2{}_{}{}^{}\mathrm{P}$ lines of 2-MeV/amu Mg projectiles excited by collision with carbon and aluminum atoms in thin foils have been measured with a pivoted plane-crystal Bragg spectrometer. Both the polarizations of these lines and the reflection properties of the ammonium dihydrogen phosphate (ADP) diffraction crystal have been determined. The ${\sigma }_0$/${\sigma }_1$ cross-section ratios for populating the 2p(‖m‖=0) and 2p(‖m‖=1) states in the 1s2p He-like and the 2p H-like configurations were deduced from the polarization values and compared with the coupled-states calculation of Reading et al., assuming that the mechanism populating these states is the capture of electrons from the K shell of the target. The agreement between theory and experiment is very good for carbon excitation, but for aluminum excitation the theory significantly overestimates the cross-section ratio. The influence of foil tilting and cascade contributions on the polarization were investigated experimentally and taken into account in the analysis of the measurements.

We present photoionization cross sections for the ground and metastable (3s3p${)\mathrm{}}^3$P states of atomic magnesium, using ab initio many-electron wave functions in the LS-coupling scheme. The effects of initial-state correlation and channel coupling are incorporated in a straightforward manner with use of configuration-interaction methods. The selective use of complex basis functions obviates the need for explicit construction of continuum wave functions. This method also provides a convenient way of incorporating the effects of autoionization into atomic photoionization cross sections.

The four logarithmic mean excitation energies I(ν), for ν=-1, 0, 1, and 2, have been calculated analytically for hydrogen and helium. They appear in the expressions of total cross section, stopping power, and straggling effect for fast charged particles, and in the expression of the Lamb shift of atomic energy levels. The method is based on a generalization of a method by Dalgarno and Lewis [Proc. R. Soc. London, Ser. A 233, 70 (1956)]. The only input is the wave function of the initial state of the atom (in the present paper only the ground state is considered). For hydrogen the method is rigorous. For helium, Hartree-Fock-type wave functions were used, which is the only approximation of the present method. The accuracy of the method is essentially independent of the value of ν.

We demonstrate that the array of techniques employing complex basis functions, which have been broadly applied in the calculation of resonance positions and lifetimes, can also be used, with appropriate modification, in the practical computation of photoionization cross sections and nonresonant scattering amplitudes. We illustrate a procedure with which controlled complex distortions over a localized region of the energy spectrum can be realized in a finite basis-set calculation. This technique is used to directly obtain scattering amplitudes for potentials which are not exponentially bounded at large distances, in contrast to complex scaling techniques which have been shown to be divergent in such cases. Four numerical examples are presented to exhibit the utility of the local complex distortion approach in calculations of photoionization cross sections and scattering amplitudes. We also discuss this approach in the context of variational principles for continuum amplitudes and in the context of applications to many-electron systems.

The l,m-substate distribution in low-lying Rydberg manifolds (n≊10) following electron capture ${\mathrm{H}}^{+}$+H(1s)→H(n)+${\mathrm{H}}^{+}$ is calculated at high velocities (v>1 a.u.) in the continuum-distorted-wave (CDW) approximation. The standard CDW approximation is modified to account for final-state Stark mixing of the Rydberg manifold in the exit channel using the post-collision-interaction model. The influence of multiple-scattering contributions is analyzed and comparison is made with ${\sigma }_{\mathrm{lm}}$ predicted by the Born approximation. We find that the double-scattering contribution, closely connected with the classical Thomas process, becomes visible in the CDW approximation at surprisingly low nonasymptotic velocities.

A complete scattering-theoretic description of the ${}_{}{}^2{}_{}{}^{}\mathrm{\Sigma }_{\mathrm{u}}^{+}{}_{}{}^{}$ shape resonance in fixed-nuclei electron-${\mathrm{H}}_2$ scattering is given within the framework of the projection-operator formalism of Feshbach. Polarization and correlation effects are included via the many-body optical-potential approach using the two-particle-hole Tamm-Dancoff approximation. The calculations result in a well-defined and physically meaningful separation of the T matrix and the eigenphase sum into a smooth background term and a resonant term, which varies rapidly with energy and internuclear distance. T matrices, eigenphase sums, and Feshbach resonance parameters ${\epsilon }_d$, Γ(E), and Δ(E) are obtained for a range of internuclear distances extending from the equilibrium distance of ${\mathrm{H}}_2$ to 2.75 a.u. These fixed-nuclei data provide the basis for the ab initio calculation of cross sections for vibrational excitation and dissociative attachment in ${\mathrm{H}}_2$ beyond the local-complex-potential approximation, to be reported in a forthcoming paper.

We have studied the electron-impact excitation of the 2 ${}^1$S and 2 ${}^1$P states of helium in the Coulomb-Born approximation. Closed-form expressions for scattering amplitudes have been obtained with Fourier decomposition of the interaction potential and then the use of accurate correlated wave functions. The results of the present calculations have been compared with other available theoretical calculations and experimental measurements.

The SU(3) dynamical evolution of three-level systems at two-photon resonance induced by two strong linearly polarized monochromatic fields is studied exactly by means of the semiclassical many-mode Floquet theory (MMFT) recently developed by the authors. Within the rotating-wave approximation (RWA), Hioe and Eberly have recently shown that the eight-dimensional SU(3) coherent vector S→ characterizing the time evolution of three-level systems can be factored into three independent vectors of dimensions three, four, and one, at appropriate two-photon resonance conditions. In practice, however, if the laser-atom interactions occur away from the two-photon resonance, or if the RWA is not valid, etc., this Gell-Mann–type SU(3) dynamical symmetry will be broken. It is shown in this paper that instead of solving the time-dependent generalized Bloch equations, the SU(3) dynamical evolution of the coherent vector S→ as well as various symmetry-breaking effects can be expediently studied by the use of the MMFT and expressed in terms of a few time-independent quasienergy eigenvalues and eigenvectors. Furthermore, we have extended the generalized Van Vleck (GVV) nearly degenerate perturbation theory to an analytical treatment of the two-mode Floquet Hamiltonian. This reduces the infinite-dimensional time-independent Floquet Hamiltonian into a 3×3 effective Hamiltonian, from which useful analytical properties of the SU(3) coherent vector can be easily obtained. The combination of the MMFT and the GVV method thus greatly facilitates the study of the dynamical evolution. Pictorial comparison of the exact and the RWA results of the time evolution of the eight-dimensional coherent vector under several different physical conditions is presented and discussed at length.

By using a pulsed hollow-cathode discharge, metastable atoms are produced in the (2$p^5$3s3p)${}^4$$D_{7/2}$ and (2$p^5$3s3p)${}^4$$S_{3/2}$ levels of Na i. A tunable laser is used to excite these atoms to levels in the 2$p^5$3s3d and 2$p^5$3s4s configurations, and thereby to establish a partial Grotrian diagram for the quartet system. Several possible Na i xuv laser systems are described.

High-resolution measurements of Ne K Auger electrons have been performed using multiply charged ${}_{}{}^{20}{}_{}{}^{}\mathrm{Ne}$ projectiles in collisions with He, ${\mathrm{CH}}_4$, and Ar at impact energies of 70 and 100 MeV. Kinematic line-broadening effects were substantially reduced by observing the electrons at 0° with respect to the beam axis. For the He-gas target the outer-shell electronic configuration of the incident ions is essentially retained during K-shell ionization. Thus, B-, Be-, and Li-like electron configurations were selectively produced for incident ions of ${\mathrm{Ne}}^{4+}$, ${\mathrm{Ne}}^{5+}$, and ${\mathrm{Ne}}^{6+}$, respectively. For target gases ${\mathrm{CH}}_4$ and Ar, Li-like lines are found to be strongly excited, independent of the initial charge state of the projectiles.

We describe a method for evaluating the consistency of oscillator strengths in the neutral and ionized species of an atom. The method is based on the constant number density of a dense vapor. In our procedure the vapor is excited and then ionized within less than 1 μs by resonant laser irradiation. The column densities $N_i$$l_i$ of the levels that are appreciably occupied during the excitation and ionization are sampled by absorption, emission, and/or anomalous-dispersion measurements in a set of suitable spectral lines. Given correct oscillator strengths $f_i$ for these lines, the column density ${\mathcal{J}}_i$ $N_i$$l_i$ (which is actually determined from the measured quantities $N_i$$f_i$$l_i$) will remain constant while excitation and ionization take place. In barium—with diagnostics based on accepted f values in the literature—the column density appeared reduced by a factor of 2 during the initial portion of the excitation, i.e., while only the low-lying levels of neutral barium were predominantly excited. Upon complete ionization, the column density took again the value derived before the laser irradiation, when all atoms had been in the ground state. From this behavior we concluded that the literature oscillator strengths for absorption lines involving the neutral ground state and the ground state and low-lying levels of ${\mathrm{Ba}}^{+}$ were consistent on a relative scale, but that the f values for absorption transitions from the low-lying metastable levels and from the resonance level of neutral barium were too large by a factor of 2. We recommend a set of correction factors to be applied to the existing data in the literature. Our conclusions are supported by the results of a critical analysis of recent oscillator strength and lifetime measurements for neutral barium.

We report the resolution of the Ba (6$p_j$${\mathrm{nd}}_j$${)}_{J=1\mathrm{}}$ and (6$p_j$${\mathrm{nd}}_j$${)}_{J=3\mathrm{}}$ autoionizing levels and their mixing with the (6$p_j$${\mathrm{ns}}_{1/2}$${)}_{J=1\mathrm{}}$ and (6$p_j$${\mathrm{ng}}_j$${)}_{J=3\mathrm{}}$ states by a three-step laser excitation approach. For the 6$p_{1/2}$${\mathrm{nd}}_j$ levels we are unable to discern a difference between J=1 and J=3. For the 6$p_{3/2}$nd levels we are able to identify the two J=3 states, but only one J=1 feature, and measure their quantum defects. Furthermore we are able to measure the mixing of the (6$p_{3/2}$${\mathrm{nd}}_j$${)}_{J=1\mathrm{}}$ states with the (6$p_{3/2}$${\mathrm{ns}}_{1/2}$${)}_{J=1\mathrm{}}$ states and determine that the analogous mixing of the (6$p_{3/2}$${\mathrm{nd}}_j$${)}_{J=3\mathrm{}}$ states with the (6$p_{3/2}$${\mathrm{ng}}_j$${)}_{J=3\mathrm{}}$ states is not observable, nor is any mixing of the 6$p_{1/2}$${\mathrm{nd}}_j$ states with either 6$p_{1/2}$${\mathrm{ns}}_{1/2}$ or 6$p_{1/2}$${\mathrm{ng}}_j$ states.

L-shell Coster-Kronig transition rates are calculated for Ar through a formulation which includes the effects of final-state channel mixing. The interchannel interaction changes individual transition rates by as much as 84%. The relative term intensities from the present calculation agree better with experiment than calculations which do not include channel mixing. The discrepancies previously found between measured and calculated total $L_1$-$L_2$,3${\mathrm{M}}_1$ Coster-Kronig transition rates and the ratios of $L_1$-$L_2$,3${\mathrm{M}}_1$ to $L_1$-$L_2$,3${\mathrm{M}}_2$,3 and $L_1$-$L_2$,3${\mathrm{M}}_1$${(\mathrm{}}^1$${\mathrm{P}}_1$ ) to $L_1$-$L_2$,3${\mathrm{M}}_1$${(\mathrm{}}^3$${\mathrm{P}}_1$ ) intensities are largely removed when effects of channel coupling are taken into consideration.

The partial cross section, the satellite branching ratio, and the angular-distribution asymmetry parameter for simultaneous photoionization and excitation to the n=2 states of the ${\mathrm{He}}^{+}$ ion have been measured in the 67.5–90-eV photon-energy range. In the nonresonance regions (67.5 eV≤hν≤69.5 eV and 75 eV ≤hν≤90 eV), the asymmetry-parameter values have been used to infer the ratio of the 2p cross section to the 2s cross section. These results indicate that the ${\mathrm{He}}^{+}$(n=2) satellite is predominantly 2p near threshold, in agreement with the experimental and most of the theoretical results reported to date. In the region below the ${\mathrm{He}}^{+}$(n=3) threshold (69.5 eV≤hν≤73.0 eV), the effects of a series of autoionizing Rydberg levels on the n=2 cross section, branching ratio, and asymmetry parameter have been measured, this being the first detailed measurement of the angular distribution of a satellite over an autoionization resonance. In addition, qualitative information concerning the total cross section and the 1s partial cross section has been obtained for the first member of this series (3s3p), disagreeing with previous experimental and theoretical results for the total cross section, but in agreement with recent photoemission measurements of the 1s cross section. The present results suggest that the qualitative shapes of the total and 1s cross sections over the 3s3p resonance are similar to the profile of the n=2 cross section for this resonance. To illustrate quantitative methods for the interpretation of autoionization phenomena, the derivation from the resonance data of several parameters defining the autoionization process is described.

The K series of muonic x rays of iron has been measured from ${\mathrm{FeCl}}_3$, ${\mathrm{Fe}}_2$${\mathrm{O}}_3$, ${\mathrm{K}}_3$Fe(CN${)}_6$, and aqueous solutions of Fe(${\mathrm{NO}}_3$${)}_3$ and ${\mathrm{K}}_3$Fe(CN${)}_6$. Significant differences in the x-ray-intensity pattern were observed and related to the structure of the outer electrons of Fe in these compounds. The influence of the ionic radius and ionic charge and of hydrogen in aqueous solutions is discussed. The importance of open d- or f-electron configurations is demonstrated.