Session D15 - Poster Session and Reception.
POSTER session, Friday evening, April 18
Congressional Hall, Renaissance

[D15.01] Photoexcitation and Predissociation Intensities of N_2 Singlet States

Direct determinations of the intensity of photoexcitation followed by predissociation and the branching between predissociation and photoemission for rovibrational levels of excited states of N_2 have been made. Laser excitation of a 3 keV beam of N_2 is used with identification of the dissociation fragments by a position- and time-sensitive detector.(C.W. Walter, P.C. Cosby, H. Helm, \itJ. Chem. Phys.) \bf99\rm, 3553 (1993).^,(C.W. Walter, P.C. Cosby, H. Helm, \itPhys. Rev. A) \bf50\rm, 2930 (1994). The states probed are b'^1\Sigma _u^+ (v=9-18), c'^1\Sigma _u^+ (v=3,4), c^1\Pi _u (v=3,4), e'^1\Sigma _u^+ (v=0), e^1\Pi _u (v=0), and o^1\Pi _u (v=3,4). Substantial predissociation is observed in each state, and dramatic variations in transition intensity are observed in some vibrational bands. The experimental results are compared with a perturbation model of the mixing between the valence and Rydberg states.(H. Helm, I. Hazell, N. Bjerre, \itPhys. Rev. A) \bf48\rm, 2762 (1993). The model reproduces many of the observed perturbations and reveals that b' character is the dominant factor in the transition strength for these states. This model extends to levels of N_2 beyond the experimental range, including the atmospherically important c' (v=0) state.

[D15.02] Quantum-Classical Comparison in Chirped Pulse Excitation and Dissociation of Diatomic Molecules

We study IR multiphoton dissociation of diatomic molecules using a chirped laser pulse. Nonlinear dynamical analysis of the chirped process shows that it can be described by bucket dynamics, in which a cluster of trajectories are trapped in a bucket which moves up steadily in phase space. This mechanism explains the high efficiency of a chirped pulse in lowering the threshold field strength for dissociation. We present comparisons between the quantum and classical results and discuss the effects of including the rotational degree of freedom.

[D15.03] Photofragmentation of CF_3Cl in the vicinity of the Cl K-shell resonance.

We have studied the photofragmentation of CF_3Cl for photon energies from resonance to 30 eV above threshold. The experiment was performed using a space focussed time-of-flight analyzer (TOF). Photon pulses were used to gate the coincidence electronics thus recording the time correlated arrival of the first and second ion fragments to arrive at the MCP. At resonance the line shapes in the time-of-flight spectra are split. In previous studies this has been attributed to the symmetry of the excited valence orbital. The profiles above resonance show no indication of being split. A discussion of the dominant fragmentation pathways and the energy dependence will be presented.

[D15.04] Floquet Formulation of Time-Dependent Density Functional Theory for Intense-Field Multiphoton Processes.

This abstract was not submitted electronically.

[D15.05] The Energy Spectra of Multiphoton Above-Threshold Dissociation of H_2^+ in Intense Laser Fields

This abstract was not submitted electronically.

[D15.06] \begincenter Bright light shines quickly. Electron, once Coulomb bound, has light now set free? \endcenter

The Strong Field Approximation (SFA) (H. R. Reiss, Prog. in Quantum Electron. \underline16), 1 (1992). method is applied to a hydrogenic atom interacting with an intense, short-pulse, circularly-polarized laser field. The field is taken to be in the form of a one-cycle wave-packet pulse whose amplitude has a Gaussian temporal dependence. Results from numerical calculations which determine the probability of ionization and the energies of the final-state electrons are presented. Transient probabilities that follow the progress in time of the electron show that, during the pulse, the electron has a large probability of reaching a Volkov state and then revisiting the atom. This is attributed to the interference between the terms of the interaction Hamiltonian. Unlike the final-state ionization probability in the long-pulse case, the probability in the short-pulse case exhibits dependence on the azimuthal angle. Spectral calculations of photoelectron energy present peaks seen in both long-pulse theory and in experiment.

[D15.07] Coulomb correction to the linear polarization SFA

The SFA (strong field approximation) is a nonperturbative analytical method for the treatment of atomic ionization by intense fields. The only approximation inherent in the SFA is that the state of the ionized electron is taken to be dominated by the laser field as opposed to the atomic binding potential. An inference of the SFA is that the ionized electron moves in a classical orbit in the laser field, with this orbit symmetrical with respect to the remnant ion. Coulomb correction can be introduced by evaluating the effect of the Coulomb field on the action function of the electron in its classical orbit. The correction in this form has been found for circular polarization of the laser field, and agrees closely with experiment. We examine the more difficult case of linear polarization, and give analytical and numerical results.

[D15.08] Femtosecond electron beam pulses produced by intense lasers

Very intense femtosecond laser pulses have been used to produce photoelectron spectra which can extend to energies of the order of one keV. However, these spectra are very broad, and those produced by linearly polarized lasers have most of the photoelectrons concentrated at the low end of the spectrum. It is shown that, using the phenomenon known as ``atomic stabilization'', it is possible to produce photoelectron spectra which peak strongly at several hundred eV, with relatively narrow energy and angular widths. These photoelectrons could then be used to produce an electron beam of femtosecond duration.

[D15.09] Effect of frequency chirping in resonant ionization by intense laser pulses

Via a non--perturbative method we study the population dynamics and photoelectron spectra of Cs atoms subject to intense chirped laser pulses. We include above threshold ionization spectral peaks. The frequency of the laser is near resonance with the 6s-7p transition. Dominant couplings are included exactly, weaker ones accounted for perturbatively. We calculate the relevant transition matrix elements, including spin--orbit coupling. The pulse is taken to be a hyperbolic secant in time and the chirping a hyperbolic tangent. This choice allows the equations of motions for the probability amplitudes to be solved analytically as a series expansion in the variable z=(\tanh(\pi t/T)+1)/2, where T is a measure of the pulse length. We find that the chirping changes the ionization dynamics and the photoelectron spectra noticeably, especially for longer pulses of the order of 10^4 a.u. The peaks shift and change in height, and interference effects between the 7P levels are enhanced or diminished according to the amount of chirping and its sign. The integrated ionization probability is not strongly affected.

[D15.10] A test of a simple semi-classical model of intense field ionization in the tunneling limit

In the high-intensity, tunneling limit, elastic rescattering of the ionized ``free'' electron by its parent ion core leaves a distinct signature on its energy and angular probability distributions. We show the dynamics of the ionization and laser-driven rescattering processes that occur in this regime can be accurately reproduced using a simple semi-classical model. Predictions from this model will be compared with recent high-precision measurements of short pulse, high intensity photoelectron spectra from helium.( B. Walker, B. Sheehy, K. C. Kulander and L. F. DiMauro, Phys. Rev. Lett.) 77 5031 (1996). This work was performed in part under the auspices of the U. S. Department of Energy at the Lawrence Livermore National Laboratory under contract No. W-7405-Eng-48 and in part at the Brookhaven National Laboratory under contract No. DE-AC02-76CH00016 with the U. S. Department of Enegry and supported by its Division of Chemical Sciences, Office of Basic Energy Sciences.

[D15.11] Hybrid Vacuum-Atmosphere Compressors for High Peak Power, Chirped Pulse Amplification Lasers

The theoretical and experimental results are presented for chirped pulse amplification compressors that contain only the final, high peak power grating in vacuum. Pulses of duration 25~fs have been compressed with the system and the influence of the optical aberrations from the vacuum/air interface in the compressor could not be detected. The hybrid vacuum-atmosphere compressor can be applied to a very broad range of wavelengths and pulse durations. The design greatly reduces the engineering, setup time, cost, and size of pulse compression in vacuum and will aid the future development and construction of laser systems with peak powers of ten terawatts and greater.

[D15.12] A Parallel Processing Method for Solving the Time Dependent Schrödinger Equation with Applications to Multiphoton Processes.

The time propagation of a large, spatially-extended wave packet is a challenging numerical problem since it requires a large number of spatial grid points and, consequently, a prohibitively large amount of computational time. In our approach the wave packet is sampled in small spatial domains. Then, each sample is independently propagated over one cycle of time, using the split operator and FFT methods. Our approach is suitable for parallel processing in a multi-processor computer, since each of the wave packet samples may be treated independently. However, our numerical tests were done on a single-processor computer for which the parallel processing is simulated by a multi-thread implementation of the algorithm. Numerical illustrations are presented for a variety of 1D time-dependent problems, including high harmonic generation.

[D15.13] Floquet States from Spectral Analysis of Wave Packets.

Numerical method for the solution of the (quasi-) stationary states of a time-independent Hamiltonian through the Fourier analysis of a time-dependent solution to Schrödinger's equation and its auto-correlation function [Feit, et al., J. Comp. Phys. 47, 412 (1982)] is extended to the case of a time-periodic Hamiltonian. This extension provides the positions, lifetimes, and channels functions associated with the Floquet states. One may also obtain the weights of the Floquet states in the wave packet. The effectiveness of this method is demonstrated for a one-dimensional model of an atom in a strong, monochromatic laser field by comparing with results from a time-independent integration of the corresponding Floquet system. The agreement is excellent. These methods are quite general, being applicable whenever a time-dependent solution to Schrödinger's equation is available.

[D15.14] The Autler-Townes Effect with a Modulated Pump: Simultaneously Forbidden Resonances

Systems subject to a strong modulating interaction can exhibit a new property in their response to a probe field. Under certain conditions an infinite number of resonances are simultaneously forbidden. This effect is investigated in the case of a three state system in which a strong field with a periodic frequency modulation Ømega couples a pair of excited levels while the complex Autler-Townes spectrum is probed via a weak field that connects one of the coupled states to the ground state. Under certain conditions a half-infinite comb of spectral lines, spaced by Ømega, simultaneously disappears from the Autler-Townes spectrum. These lines are positioned above or below a unique edge frequency, which is that of the probe transition in the absence of the strong field. This effect of simultaneously forbidden resonances (SFR) results from a special factorization property of the Floquet Hamiltonian that describes the Autler-Townes spectrum. The probe absorption line shape demonstrating the SFR effect is calculated, using the method of continued fractions. Possible experimental manifestations of SFR will be presented.

[D15.15] Multiphoton Microwave Processes in Rydberg Atoms.

Multiphoton microwave transitions of He and Na Rydberg atoms are studied using a new discrete-time picture of Floquet analysis. This approach increases the flexibility of Floquet calculations of multiphoton processes in strong radiation fields, particularly at low frequencies, by incorporating adiabatic states into the basis. The discrete-time approach yields more accurate ac-Stark shifts than calculations using truncated zero-field basis states. Calculations of the depopulation of the 11s state of Na reveal a strong correspondence between dc-field Stark maps and quasi-energy spectra and predict a new resonance feature associated with Landau-Zener transitions between dc-Stark levels.

[D15.16] Scattering of Positrons and Electrons from Magnesium

We have used the model potential approach to calculate the total and elastic cross sections for scattering of positrons and electrons from magnesium for an extended range of impact energies. In this approach the interaction potential is partitioned into parameter-free static, correlation-polarization, exchange (for electron impact only) and absorption parts. The new absorption potential, developed by us specifically for positron scattering, is obtained using a modification of a method originally proposed in nuclear physics for nucleon-nucleon scattering. Our calculated cross sections show good agreement with the corresponding experimental cross sections wherever the experimental results are available.

[D15.17] Calculations of Positron Scattering from the Noble Gases and Alkaline Earths

We have applied the distorted-wave method to the scattering of positrons from both the noble gases and the alkaline earths. Specific calculations have been carried out for scattering cross sections for neon and magnesium. These include the dominant excitation channels as well as ionization. Elastic scattering cross sections have been obtained via the polarized-orbital method. The positronium formation channel has not been included. We will compare our results with existing experimental data for the total cross sections as well as with experimental excitation cross sections for neon.

[D15.18] Measurements of Elastic Differential Cross Sections for Positrons Scattered from Krypton and Xenon

We have extended to higher energies our measurements of relative elastic differential cross sections (DCS's) for positrons scattered at 30^o to 135^o from Kr and Xe. Previously we had measured relative elastic DCS's over an energy range from below the positronium (Ps) formation threshold of each atom to 20 eV for Xe and 60 eV for Kr, and had found evidence that the Ps formation channel was having an absorption effect upon the elastic scattering channel.( W.E. Kauppila et al., Can. J. Phys. 74, 474 (1996).)^,( L. Dou et al., Phys. Rev. A 46, R5327 (1992).) We find an unexpected "kink" in the positron-Xe DCS near 50^o for 30 eV that shifts towards smaller angles as the energy is increased. We find no evidence of a higher energy kink in Kr. We are extending our measurements to additional energies to trace the evolution of this behavior.

[D15.19] Positron annihilation with inner-shell electrons in atoms

Accumulation of large numbers of room-temperature positrons in a Penning trap facilitates a wide range of atomic physics experiments.(K. Iwata, et al., Phys. Rev. A, 51, 473 (1995).) One type of experiment is the measurement of \gamma-ray spectra from positron annihilation on molecules.(K. Iwata, et al., submitted to Phys. Rev. A, 1996.) These spectra are Doppler broadened due to the momenta of the annihilating electron-positron pairs. Due to the positron-nucleus repulsion, low-energy positrons cannot penetrate deep inside the atom so that they annihilate predominantly with the valence electrons. However, when the high-momentum regions of observed spectra from noble gas atoms are compared with theoretical calculations based on the static Hartree-Fock approximation, a small but measurable fraction of positrons are annihilating with inner-shell electrons for Kr and Xe. This is the first experimental study of inner-shell electron annihilation arising from isolated two-body positron-atom interactions.

[D15.20] The Hidden Crossing Method Applied to Positronium Formation

The hidden crossing method successfully describes reactions in ion-atom collisions(S. Yu. Ovchinnikov and E. A. Solovév, Comments At. Mol. Phys. 22) 69 (1988). (T. P. Grozdanov and E. A. Solovév, Phys. Rev. A 42) 2703 (1990). and electron-impact ionization of hydrogen. (J. H. Macek and S. Yu. Ovchinnikov and S. V. Pasovets, Phys. Rev. Lett. 74) 4631 (1995). (J. H. Macek and S. Yu. Ovchinnikov, Phys. Rev. A 54) 544 (1996). Using the hidden crossing method we computed the S- and P-wave cross sections for positronium formation for positron-hydrogen collisions in the Ore gap. We used the hyperspherical Hamiltonian of the form given by Zhou and Lin. (Y. Zhou and C. D. Lin, J. Phys. B: At. Mol. Opt. Phys. 27) 5065 (1994). The hidden crossing method provides a physical explanation of why the S-wave cross section for positronium formation is so small---the amplitudes corresponding to two different paths destructively interfere.

[D15.21] Momentum distributions of electrons in atomic few-body Coulomb problem: Positive energy Sturmian technique( Sponsored by the Office of Basic Energy Science, U.S. Department of Energy, under Contract No. DE-AC05-96OR22464 with Lockheed Martin Energy Research Corp.)

We propose a new technique for calculations of electron momentum distributions in electron-atom collisions. The ``positive energy Sturmian technique" is based on the following main steps: (i) time-dependent scaled transformation; (ii) Fourier transformation into the frequency domain; (iii) outgoing wave Sturmian expansions; and (iv) solution of converged coupled equations. The ``positive energy Sturmian technique" has been successfully applied in ion-atom collisions for calculations of electron-energy and angular-distributions.( S. Y. Ovchinnikov and J. H. Macek, Phys. Rev. Lett. 74) 4631 (1995).

[D15.22] Measurements of Total and Lower Limit-Positronium Formation Cross Sections for Positron and Electron-Magnesium Scattering

We have used a beam transmission technique to measure total cross sections (Q_T's) for positrons and electrons scattered by Mg atoms from 2 to 51 eV. In addition, lower limits on positronium (Ps) formation cross sections have been measured from 1 to 10 eV by detecting coincidences of 511 keV annihilation gamma rays produced by the decay of para-Ps and by the interaction of ortho-Ps with the walls of the scattering cell in which the Ps is formed. Mg is of particular interest for positron scattering because a positron-Mg bound state has been predicted recently(V.A. Dzuba, V.V. Flambaum, G.F. Gribakin, and W.A. King, Phys. Rev. A 52, 4541 (1995).), (2) Mg is the first alkaline earth target atom (for positrons) to be investigated experimentally, and (3) its Ps formation threshold (0.8 eV) is close to zero energy. Comparisons of our measured positron and electron Q_T's indicate that the positron Q_T's are higher than the corresponding electron Q_T's at all energies investigated below 40 eV.

[D15.23] Measurements of Positronium Formation Cross Sections for Positron-Kr, Xe Scattering

Our experimental approach(S. Zhou et al., Phys. Rev. Lett. 73, 236 (1994).) for measuring Ps formation cross sections (Q_Ps) involves passing a variable energy positron beam through a gas scattering cell and detecting the 511 keV annihilation gamma rays resulting from the decay of para-Ps and from the interaction of ortho-Ps with the walls of the scattering cell. It is found that the Q_Ps curves for both Kr and Xe rise rapidly from their formation threshold energies of 7.2 and 5.3 eV, reach maxima within about 10 eV of their thresholds and then decrease to become rather small (less than 10% of the peak heights) above 100 eV. At the maxima Q_Ps accounts for more than 50% of the total scattering cross sections. There is some evidence of possible small scale structure in the Q_Ps curves between 10 and 100 eV. The present results are consistent with the prior measurements of Diana et al.( L.M. Diana et al., in "Atomic Physics with Positrons", edited by J.W. Humberston and E.A.G. Armour (Plenum, New York and London, 1987), p. 55; and in "Positron Annihilation", edited by L. Dorikens-Vanpraet et al. (World Scientific, Singapore, 1989), p. 311.) from near threshold to 70 eV for Kr and from 15 to 100 eV for Xe.

[D15.24] Measurents of Positronium Formation Cross Sections for CO_2 and SF_6

Positronium (Ps) formation cross sections, Q_Ps's, have been measured for the scattering of positrons by CO_2 and SF_6 by monitoring the 2-gamma coincidences arising from the decay of para-Ps and from the interaction of ortho-Ps with the walls of the Al cell in which Ps is formed.(S. Zhou et al., Phys. Rev. Lett. 73, 236 (1994).) Our measured Q_Ps versus energy curve for SF_6 shows an unusual flat maximum from 20 to 50 eV, while for CO_2 we find a somewhat flat maximum and also clear evidence of structure near 12 eV. These measurements strongly suggest that two or more mechanisms are responsible for Ps formation for these molecules.

[D15.25] Measurements of Differential Cross Sections for Positrons Scattered from Molecules

We have now measured relative quasi-elastic differential cross sections (DCS's) for positrons scattered at 30^o to 135^o from CH_4, N_2, O_2, CO, N_2O, CO_2, SF_6 and C_2H_2 at energies extending from below the positronium (Ps) formation thresholds to well above them (up to 100 eV). For CH_4, N_2, O_2, CO, and SF_6 we find a well-defined minimum in the DCS versus angle data below the Ps formation threshold which becomes shallower as the energy increases and is no longer present above 20 eV, similar to absorption effects we have previously seen in positron-Ar scattering.(Steven J. Smith et al., Phys. Rev. Lett. 64, 1227 (1990).) In contrast, for N_2O, CO_2, and C_2H_2 we find only a very shallow minimum below the Ps formation threshold which also disappears above 20 eV. We speculate that the presence of the more pronounced structure for certain molecules may be related to their higher degree of spherical symmetry. For SF_6 we find additional unexpected structure that shifts towards smaller angles as the energy is increased.

[D15.26] Triply differential bremsstrahlung cross section for 100 and 140 keV electrons on Cu, Ag and TbF_3

We report new measurements of the bremsstrahlung triply differential cross section for 100 and 140 keV electrons on thin-film targets of Cu, Ag and TbF_3. The data for two geometries have been obtained simultaneously using a three detector multi-coincidence electron-photon spectrometer coupled to a 300 keV electron accelerator. In both cases the electron scattering angle is 45^o. The x-ray photon emission angles were either -45^o in the electron scattering plane or 90^o orthogonal to the scattering plane. The data will be compared with recent calculations of Shafer and Pratt(D. Shafer, private communication (1997)), and also with the Elwert-Haug and Bethe-Heitler predictions.

[D15.27] Absolute Triply Differential (e,2e) Cross Section Measurements for H and He with Comparison to Theory.

Absolute triply differential (e,2e) cross section measurements are presented for H (for incident energies, E_0, of 15.6 eV and 17.6 eV) and for He (for E_0 = 32.6 eV, 44.6 eV, 64.6 eV, and 104.6 eV) for equal-energy-sharing and the \theta_12 = \pi configuration. Results of distorted partial wave calculations agree with the measurements; those of convergent close coupling calculations agree with the relative angular distributions, but are lower than experiment by factors of up to 7. Relative experimental results for H for E_0= 14.6 eV show a qualitative change in shape, which agrees with theory. Implications of the absolute experimental results for the range of validity of the Wannier threshold law are presented.

[D15.28] Cadmium (e,2e) Energy Spectra in the Autoionizing Region

We will present an analysis of our comprehensive set of (e,2e) measurements in the Cd 4d^95s^25p autoionizing region, carried out for an incident electron-beam energy of 150~eV and scattering angles between 2^\circ and 18^\circ, corresponding to momentum transfer K=0.2 \to 1~au. The results are presented as the sum and difference of k_ej=\pm\hat k pairs of (e,2e) ejected-electron energy spectra(N.L.S. Martin, D.B. Thompson, R.P. Bauman and M. Wilson, Phys.Rev.Lett 72), 2163 (1994); Phys.\ Rev.\ A 50, 3878 (1994). for three special directions \hat k, and compared with plane wave Born calculations that include ejected-electron partial waves \ell=0\to7. It is found that the relative Born phases are incorrect for \ell=0,1,2 by amounts that are independent of scattering angle. The relative Born magnitudes are extremely good for \hat k=\hat K, but are extremely bad for the other two \hat k directions. With increasing scattering angle we observe a reduction in the ^3P_1/^1P_1 intensity ratio in the sum spectra, probably due to an increase in exchange scattering, and we also see a previously unobserved Cd 4d^95s^25p~J=3 autoionizing level.

[D15.29] Anti-Capture to the Continuum via a Double Collision Mechanism in (e,2e) Collisions

Recently, anti-capture to the continuum occurring via the (Thomas) double collision mechanism has been reported.(D. E. Golden, J. Xu, D. W. Mueller and J. Bernhard, Nucl. Inst. and Meth. in Phys. Res. B 99) 202 (1995).^,(D. E. Golden, Z. Xu, J. Bernhard and D. W. Mueller, J. Phys. B: At. Mol. Opt. Phys. 29) 3741 (1996). The doubly differential cross section was measured in coplanar geometry for electron-impact ionization of argon at a 1000 eV incident energy. The cross section is differential with respect to the solid angle of the ejected electron and scattered electron. The detector (split anode detector) measures the two outgoing electrons at approximately the same angle. Both a sharp dip---the signature of anti-electron capture to the continuum---and the Thomas peak were observed in the cross section. We computed the doubly differential cross section for electron-impact ionization of hydrogen. This was done by including in the Bethe-Born approximation a normalization factor that approximately incorporates electron-electron interactions in the final state. We considered hydrogen both in the ground state and the 2p excited state and were able to reproduce the main features of the measurements.

[D15.30] 100 eV Asymmetric (e,2e) Study of Molecular and Atomic Oxygen

Relative triple differential cross sections (TDCS) have been measured for the ionization of the O_2 1\pi_g, 1\pi_u, and 3\sigma_u molecular orbitals to give the ground state and the a and b excited states of the molecular ion. An incident energy of 100 eV (the peak of the total ionization cross section) was used and the scattering angle of the incident electron was typically 4 degrees. Ejected secondary electron energies from 3.5 to 11.5 eV were studied. The TDCS for all states are marked by a strong preference for ejection of the secondary electron at angles much larger than the momentum transfer angle, nearly perpendicular to the incident electron beam direction, although there are some variations between the TDCS for the various final states. In very recent experiments, an atomic oxygen source has been incorporated into the (e,2e) apparatus. Atomic oxygen is an open-shell atom with a triplet ground state. Its TDCS is of great interest. First (e,2e) results for atomic oxygen have been obtained at 100 eV and 6 eV ejected electron energy. The atomic oxygen measurements are considerably more difficult than those for permanent gases because of the smaller density of the target species and interference from undissociated parent gas molecules. However, it has been possible to obtain coincidence energy loss spectra which clearly show the ionization to the three final states of the OII ion.

[D15.31] Comprehensive Cross Section Results for Electron-Impact Excitation out of the 2^3S Level of He

We have used a hollow cathode discharge to measure the cross sections (apparent and direct) for electron-impact excitation out of the 2^3S metastable level of He into more than ten higher triplet levels for electron energies up to 20~eV. For n\geq 3 our results show a pattern of sharply peaked excitation functions into the ^3S and ^3P levels, and comparatively broad excitation functions into the ^3D levels. The 2^3P level, however, has a broad excitation function with a peak direct cross section of 112\times 10^-16\; cm ^2. Cross sections for excitation into three of these higher triplet levels (3^3S, 3^3D, \def\emline#1#2#3#4#5#6 \put(#1,#2)\specialem:moveto \put(#4,#5)\specialem:lineto

\parbox[b]78mm 4^3D) have been measured to larger electron energies using a He target formed via resonant charge exchange between a 1.6~keV He^+ beam and cesium vapor. There is excellent agreement between the absolute results generated by both experiments. Comparisons with theoretical calculations based on several different methods will be made. ~~~~~~~ Plot of Q~vs.~E \unitlength=1.00mm\specialem:linewidth .10pt\linethickness .10pt\beginpicture(40.00,34.00)\emline0.000.00140.000.002\emline40.000.00340.0034.004\emline40.0034.0050.0034.006\emline0.0034.0070.000.008\emline20.000.00920.001.6010\emline20.0034.001120.0032.4012\emline0.0017.00131.6017.0014\emline40.0017.001538.4017.0016\specialem:linewidth .10pt\linethickness .10pt\emline4.500.30175.705.0118\emline5.705.01196.9022.7220\emline6.9022.72217.2030.1722\emline7.2030.17237.7026.3724\emline7.7026.37259.3018.2726\emline9.3018.272711.3015.0928\emline11.3015.092913.3014.0330\emline13.3014.033115.3013.5632\emline15.3013.563317.8013.3934\emline17.8013.393519.8012.7936\emline19.8012.793721.8012.2938\emline21.8012.293923.8012.8440\emline23.8012.844125.8012.7142\emline25.8012.714329.8011.8944\emline29.8011.894533.8011.3446\emline33.8011.344739.9911.3348\emline39.9911.334939.998.2550\emline39.998.255137.407.7252\emline37.407.725333.408.1354\emline33.408.135529.408.0956\emline29.408.095725.408.0558\emline25.408.055923.408.7660\emline23.408.766121.409.2062\emline21.409.206319.409.4864\emline19.409.486517.409.8466\emline17.409.846715.4011.1968\emline15.4011.196913.9011.7470\emline13.9011.747111.9013.7072\emline11.9013.707310.6015.5174\emline10.6015.51759.5017.3476\emline9.5017.34778.3019.1578\emline8.3019.15797.1011.7680\emline7.1011.76815.902.5682\emline5.902.56834.700.1584\emline4.700.15853.500.2886\emline3.500.28873.000.1088\emline3.000.10894.200.0290\emline4.200.02915.400.6592\emline5.400.65936.603.4094\emline6.603.40957.8014.9996\emline7.8014.99979.0023.7998\emline9.0023.799911.0029.82100\emline11.0029.8210113.0031.01102\emline13.0031.0110315.0030.42104\emline15.0030.4210517.0030.23106\emline17.0030.2310719.0029.79108\emline19.0029.7910921.0028.84110\emline21.0028.8411123.0027.31112\emline23.0027.3111325.0026.63114\emline25.0026.6311527.0026.03116\emline27.0026.0311729.0025.42118\emline29.0025.4211931.0024.22120\emline31.0024.2212133.0023.85122\emline33.0023.8512335.0023.34124\emline35.0023.3412537.0022.75126\emline37.0022.7512739.0022.14128\emline39.0022.1412939.9921.99130\put(25.0 , 27,0)3^3D \put(25.0 , 15,0)3^3S \put(25.0 , 3,0)3^3P \endpicture

[D15.32] Relativistic Calculation of Superelastic Electron-Alkali Atom Scattering

Superelastic scattering of electrons from excited atoms is the time-reverse of inelastic scattering. However, since present inelastic experiments are unable to resolve the hyperfine-structure states of the excited atom whereas superelastic experiments usually do, the two processes are not necessarily the time-reverse of each other in practice. We derive a theoretical mechanism by which inelastic scattering amplitudes can be used for the calculation of superelastic scattering parameters for alkali atoms which have been laser-excited to a specific hyperfine-structure state of the (np) ^2P_3/2~multiplet where the magnitude of F=J+I is at its maximum (I is the nuclear spin). We have found that when using unpolarized electrons the Stokes parameters for this superelastic case are extremely similar to the analogous parameters for inelastic scattering even when relativistic effects are strong. We also introduce superelastic generalized Stokes parameters and give numerical values for these using a relativistic distorted-wave method. \vskip0.2truecm Currently at Drake University

[D15.33] Autoionization Resonances in Orientation and Alignment Parameters for Excited Ions after Electron Impact Ionization.

We have extended previous work for alignment and orientation parameters in \hboxsimultaneous ionization-excitation~[1] to include the effect of autoionizing resonances. The expected effect is similar to that suggested for photo\-ionization via autoionizing states~[2]. Special emphasis is given to the process e + He \longrightarrow e_scattered+e_ejected +He^+(2p), followed by He^+(2p) \longrightarrow He^+(1s) + \gamma where the scattered electron and the emitted photon are detected in coincidence. \par\medskip 1. R. Schwienhorst, A.~Raeker, K.~Bartschat and K.~Blum (1996), \par \phantom1. J.~Phys.~B~29, 2305 \par 2. V.V.~Balashov, N.M.~Kabachnik and V.S.~Senashenko (1983), \par \phantom2. Book of Abstracts ICPEAC XIII,~23 \par permanent address: Moscow State University

[D15.34] Electron scattering from HeII ions at intermediate energies

Electron collisional excitation cross sections and rate coefficients of HeII ions are of extreme interest in the modelling of astrophysical plasmas. They are required for reliable determination of the excitation and ionization in the solar corona, impulsive heating events in the solar transitions region, shock waves in the interstellar medium and in stellar atmospheres. Emission lines of HeII have been observed in a variety of solar and astrophysical objects at wavelengths below 350 ÅThe Extreme Ultra Violet Explorer (EUVE) spectra of Capella (HD 3402, G8 III + G0 III), the bright RS CV_n binary system, is dominated by HeII (\lambda 303 Åand high ionization stages of iron. Accurate knowledge of the electron collisional excitation rates for the n = 2 and n = 3 levels of HeII are required in the modelling of the electron-ion equilibration in non-radiative shocks associated with SN 1006.

Recently attention has focused on scattering at intermediate energies with emphasis on the n = 2 levels using the CCC formalism and the 2D-Rmatrix propagator method. In our work we have used the IERM approach of Burke, Scott and co-workers to obtain accurate cross sections for levels up to n = 3, as this has proved successful in dealing with electron - atomic hydrogen scattering at intermediate energies. A comprehensive set of results will be presented at the meeting.

[D15.35] Effective Collision Strengths for Electron-Impact Excitation of Mg VI

The multichannel R-matrix method is utilized in the present work to compute effective collision strengths for electron-impact excitation of the N-sequence ion Mg VI. The lowest twelve LS target eigenstates are included in the expansion of the total wavefunction, namely the eight n=2 states with configurations 2s^22p^3, 2s2p^4 and 2p^5, and the four n=3 states with configuration 2s^22p^23s. These target eigenstates are represented by configuration-interaction type wave expansions constructed from seven orthogonal one-electron basis orbitals, consisting of four `spectroscopic' (1s,2s,2p,3s) and three `pseudo' orbitals (øverline3p, øverline3d, øverline4s), the latter being included to allow explicitly for additional correlation effects.

The effective collision strengths for all possible forbidden, semi-forbidden and allowed fine-structure transitions are obtained by averaging the collision strengths, for a wide range of incident electron energies, over a Maxwellian distribution of velocities. Results will be presented for electron temperatures in the range logT(K) = 5.0 to logT(K) = 6.1, which are the temperatures of interest for many astrophysical applications. We will conclusively show that the effective collision strengths for many of the forbidden lines are significantly enhanced by autoionizing resonances converging to the target state thresholds.

[D15.36] Anomalous Behavior in Zero-Angle Electron Differential Cross-Sections for Mg 3 \; ^1S \rightarrow 3 \; ^1 D and Cd 5 \; ^1S \rightarrow 5 \; ^1 D Excitation.^\ast

Anomalies such as measured by Sakai et al.^1 in zero-angle electron differential cross sections (DCSs) for optically forbidden transitions can be predicted with reliability through a recent forward scattering function^2. Through the combined use of experimental data^1 (higher impact energy, E) and theoretical data (lower E), the energy minimum for the He 1 \; ^1S \rightarrow 2 \; ^3 S transition is determined more precisely to be at 193 eV. A cancellation of the second-order and first-order effects, dominant at higher and lower E, respectively, is responsible for the energy minimum in the DCSs^3. For \Delta L = 2 energy minima in the zero-degree DCSs are predicted for the first time for Mg 3 \; ^1S \rightarrow 3 \; ^1 D and Cd 5 \; ^1S \rightarrow 5 \; ^1 D transitions to be at 28.9 eV and about 20.0 eV, respectively. ^\astSupported in part by AFOSR and DoE, Division of Chemical Sciences, Office of Basic Energy Sciences, Office of Energy Research. ^1 Y. Sakai et al., Phys. Rev. A \underline43, 1656 (1991) ^2 Z. Felfli et al., NIM B \underline119, 337 (1996) ^3 W. M. Huo, J. Chem. Phys. \underline60, 3544 (1974)

[D15.37] Accurate Determination of Small-Angle Electron Excitation Differential Cross Sections: Aid to Measurements.^\ast

In a appropriate representation^1, the variation of the generalized oscillator strength (GOS) with momentum transfer squared, K^2, is linear at small-scattering angles and fixed electron impact energy, for both optically allowed and forbidden transitions in atoms, ions and molecules. This represents a breakthrough for small-angle electron differential cross section (DCS) measurements, particularly for impact energies near threshold where the linear behavior can include angles as high as 20^o or more. The accurate determination of the generally poorly known or determined small-angle electron DCSs from their more reliable larger scattering angle measurements can now be effected and hence, accurate integral cross sections obtained. Experimental and theoretical data are presented as illustrative examples, focusing mainly on the experimentally difficult to access threshold energy region where the results are dramatic. ^\astSupported in part by AFOSR and DoE, Division of Chemical Sciences, Office of Basic Energy Sciences, Office of Energy Research. ^1 Z. Felfli and A. Z. Msezane, Phys. Rev. Lett. (submitted)

[D15.38] R-Matrix with Pseudo-State Calculations for Electron Scattering from Beryllium and Boron Atoms and Ions.

Accurate cross sections results for electon scattering from Be and B atoms and ions are very important for modeling plasma processes. Based on the R-Matrix with Pseudo-States (RMPS) approach~[1], we have performed calculations for neutral beryllium and boron targets, as well as for Li-like Be^+ and B^2+ ions. The results are compared with those obtained in other approaches, such as a first-order distorted-wave (DWBA) model~[2], the convergent-close coupling (CCC) approach~[3], and standard close-coupling calculations with only discrete target states included in the expansion. We find that only CCC and RMPS calculations produce reliable results in the ``intermediate energy regime'' for impact energies from and up to several times the ionization threshold. Predictions from simpler models sometimes deviate from the RMPS results by more than a factor of three. \par\medskip 1. K.~Bartschat, E.T.~Hudson, M.P.~Scott, P.G.~Burke and V.M.~Burke (1996), J.~Phys.~B~29, 115 \par 2. R.E.H.~Clark and J.~Abdallah J (1996) Physica Scripta T62,~7 \par 3. I. Bray and D.V.~Fursa (1996), private communication

[D15.39] The IRON Project: Electron excitation of Be-like iron

The international IRON Project aims to produce accurate cross sections for excitation of ions of Fe and other elements which are crucial in many areas of astrophysics and technology. We have used an R-matrix code\footnote[1]K.A. Berrington, W.B. Eissner and P.H. Norrington 1995, Comp. Phys. Comm. 92 290 for the inelastic scattering of electrons from Be-like Fe XXIII. Here we give preliminary results for collision strengths obtained using an LS coupled target with 54 terms. Our calculation has been done for energies above 85 Ry with respect to the ground term. We compare our data with those of other theoretical calculations. A discussion of how to compute the contribution from high partial waves is presented\footnote[2]M.C. Chidichimo and K. Rohlf 1996, Z. Phys. D 38 133. We also show how the use of Born limits\footnote[3]A. Burgess, M.C. Chidichimo and J.A. Tully 1997, J. Phys. B in press together with the graphics program OMEUPS\footnote[4]A. Burgess and J.A. Tully 1992, Aamp;A 254, 436 allows collision strengths at high energies to be estimated without having to do a partial wave expansion. Future work will extend the present calculations to lower energies where the Breit-Pauli Hamiltonian will be used.

[D15.40] Progress Toward an Absolute Measurement of Electron Impact Excitation of Si^2+

We are in the process of measuring the electron impact excitation (EIE) absolute rate coefficients for Si^2+(3s^2\ ^1S - 3s3p\ ^1P) and Si^2+(3s3p\ ^3P^o - 3p^2\ ^3P) for energies near threshold. A beams modulation technique with inclined electron and ion beams is being used. The radiation from the excited ions at \lambda 120.6 nm and \lambda 130.0 nm is detected using an absolutely calibrated optical system that subtends slightly over \pi steradians. The population of the Si^2+(3s3p\ ^3P^o) metastable state in the incident ion beam is determined by making an absolute measurement of the intensity of the \lambda 189.2 nm light from the beam due to radiative decay of the metastable state(G. P. Layfatis and J. L. Kohl, Phys. Rev. A 36), 59 (1987).. Because of the high density of overlapping resonances above threshold, the cross sections should exhibit a complex energy dependence(D. C. Griffin et al)., Phys. Rev. A 47, 2871 (1993).. We expect to resolve some of this structure. Research progress and the experimental apparatus will be described. There are no previous measurements of EIE in a Mg-like system, nor of EIE in a metastable ion. This work was supported in part by NASA Supporting Research and Technology Program in Solar Physics grant NAGW-1687 and NASA Training Grant NGT-51081.

[D15.41] Integrated Stokes Parameters for Impact Excitation of Noble Gases by Spin-Polarized Electrons.

We have applied the Breit-Pauli R-matrix method~[1] to model electron impact excitation of the [np^5(n+1)p] states in noble gases. Besides the total and differential cross sections, we have calculated the polarization of the light emitted in the subsequent optical decay to the [np^5(n+1)s] states, integrated over the projectile scattering angle. The results will be compared with recent experimental data~[2--4]. \par\medskip 1. K.A.~Berrington, W.B.~Eissner and P.H.~Norrington, Comp. Phys. \phantom1. Commun~92, 290 (1995) 2. C.~Noren and J.W.~McConkey, Phys. Rev. A53, 3253 (1996) 3. T.J.~Gay, J.E.~Furst, K.W.~Tramtham and W.M.K.P.~Wijayaratna, \phantom3. Phys. Rev. A53, 1623 (1996) 4. D.H.~Yu, P.A.~Hayes and J.F.~Williams~(1997), \phantom4. private communication

[D15.42] Use of Fourier Transform Spectroscopy in the Study of Electron-Impact Excitation of Rare Gas Atoms

We measure cross sections for electron-impact excitation of Ne, Ar, and Xe using the optical method. In these gases, the 2p (Paschen's notation) levels are populated by both direct excitation and cascade transitions from higher nd and ns levels. These predominantly IR transitions can be detected with the Fourier Transform Spectroscopy technique. By subtracting the IR cascade contribution from the apparent excitation cross sections determined from the emission intensities of the 2p \rightarrow 1s transitions, we obtain direct cross sections for the 2p levels for incident electron energies from 10 to \mbox300\ eV. Previous investigations of Ar and Xe have revealed a pressure dependency in the 2p \rightarrow 1s apparent cross sections in the mTorr regime. We confirm this dependence, and find that the 2s \rightarrow 2p and 3d \rightarrow 2p apparent cross sections exhibit similar pressure variation. We attribute this to resonance radiation reabsorption. Subtraction of the cascade from the apparent cross section removes the pressure dependency in the direct cross sections.

[D15.43] Cross Sections for Electron-Impact Excitation of the 5F He Levels

The measurement of electron-impact excitation cross sections for the 5F levels in He using the optical method poses experimental challenges, as the decay channels for this state emit entirely in the IR spectrum. Using the technique of Fourier Transform Spectroscopy, we have detected these emissions and determined the apparent excitation cross sections over a range of incident electron energies between 10 and \mbox300\ eV. We find that the shape and magnitude of these excitation functions vary greatly with target gas pressure. The origin of this pressure dependence is the collisional transfer excitation that occurs between the 5^1 P and 5F levels. The rate at which the 5F optical cross sections increases with pressure is directly related to the collisional transfer excitation cross section. Measuring the pressure dependence of the 5F optical cross section allows us to determine the 5^1P \rightarrow 5F transfer cross section.

[D15.44] On the Theory of Electron Induced Ionization of Hydrogen

Several approximations to the three-body Coulomb wave function have been proposed in recent years and used in approximate calculations for cross sections for the ionization of hydrogen by electron impact. It is shown here how to use these approximate wave functions to define three-body Coulomb wave operators, and it is proved that these wave operators are identical to the standard ones defined long ago by J. D. Dollard. This allows one to define a class of distorted-wave transition operators that have the same on-energy-shell values as the exact transition operator and for which the approximate amplitudes previously computed are first Born approximations.

[D15.45] Time Dependent Electron Impact Excitation and Ionization of Hydrogen

Study of electron hydrogen collisions will be reported using an ab-initio time dependent simulation. A fully correlated two electron wavefunction is represented utilizing a Hylleraas expansion and a spline basis. The initial wavefunction is then propagated in time to allow the collision process to take place. As the scattering collision evolves and the wavepackets approach asymptotic regions, an analysis of the system is effected by taking projections onto the eigenstates of the target hydrogen atom. Partial cross sections for excitation and ionization can be obtained from these projections. These results will be compared to other theoretical approaches and experiment.

[D15.46] Three-Body Models of Electron-Hydrogen Ionization

The field of atomic ionization by electron impact, or (e,2e) as it is called, is now nearly three decades old. Although the importance of electron-electron correlations in the (e,2e) problem has been emphasized since the first (e,2e) experiment was reported in 1969, only recently have theoretical models included these correlations. Theoretical studies using ``three-body'' models (models with explicit dependence on the electron-electron coordinate) are hampered by the difficult six-dimensional integral for the scattering amplitude. ``Two-body'' models, on the other hand, retain only electron-ion interactions and the standard techniques of partial-wave analysis greatly simplify the integral. In contrast, the wave functions used in three-body models give rise to two-center integrals, which are not efficiently evaluated using partial-wave methods. Therefore, a different method was sought. We have found that direct six-dimensional numerical quadrature is efficient. As a result, we now have the capability of calculating (e,2e) triply-differential cross sections using three-body wave functions of arbitrary complexity. Results for hydrogen will be presented using recently reported three-body wave functions.

[D15.47] The Electron-Impact Ionization Cross Sections of Be-like Ions

Electron-Impact ionization cross section for the Be atom and positive ions in the Be isoelectronic sequence are calculated in the two-potential distorted-wave approximation. Single-differential and total cross sections are calculated in a relativistic formulation. Relativistic effects are studied by calculating the cross sections in the non-relativistic limit. The exchange effects are included by antisymmetrizing the transition amplitudes. The transition amplitudes, represented by 84 diagrams after antisymmetrization, are grouped into parts analogical to the ionization and excitation processes of H-, He-, and Li-like ions. The cross sections of highly charged ions are studied and the scaling rule of the cross section is investigated.

[D15.48] Electron-Impact Total Ionization Cross Section of Rb.

The Binary-Encounter-Dipole (BED) model(Y.-K. Kim and M.E. Rudd, Phys. Rev. A 50), 3954 (1994). has been applied to electron-impact ionization of Rb. The BED cross section is in good agreement with a recent experimental data.(R.S. Schappe et al., Phys. Rev. Lett. 76), 4328 (1996). The BED theory combines a modified Mott cross section with the high-incident energy behavior of the Born cross section. The required continuum f-values were calculated from Dirac-Fock continuum wave functions with a core polarization potential.(J. Migda\l ek and W.E. Baylis, J. Phys. B 11), L497 (1978). The cut-off radius of the matching dipole transition operator was adjusted to reproduce the position of the known minimum in the photoionization cross section.(H. Suemitsu and J.A.R. Samson, Phys. Rev. 28), 2752 (1983). The contributions of the 4p \rightarrow 4d,\ 5s, and 5p autoionizing excitations were included using the plane-wave Born approximation. We also present f-values for the 5s \rightarrow np_1/2,\ np_3/2 transitions for high n near the ionization threshold.

[D15.49] Electron Impact Ionization of Cadmium for Energies near Autoionizing States

The process of electron impact ionization of atoms, normally called (e,2e), has been studied for three decades. Most of this work has concentrated on energies far from autoionization resonances. However, for electron energies corresponding to an autoionization resonance, it is not possible to distinguish electrons which have been directly ejected from the atom from those which were first excited to an autoionizing level and subsequently ejected. As a result, these two processes will interfere and the interference can potentially be sensitive to different physical effects than the normal (e,2e) process far from resonances. We will examine (e,2e) for cadmium in the energy range of autoionizing levels. First order plane-wave and distorted-wave results will be compared with recent measurements.

[D15.50] Electron Impact Ionization of D_2O and OD

Despite the importance of H_2O and the OH radical in many areas of application, the collisional data base for these species is scarce, in particular for the OH radical. Electron impact ionization cross section data available in the literature are especially controversial with a factor of 4 discprepancy between various measurements of the total H_2O ionization cross section (and no ionization cross section data are available for the OH radical). We report measurements of absolute partial and total ionization cross sections for the D_2O molecule and the OD free radical from threshold to 200 eV using the fast-beam technique. The deuterated target species were used for experimental convenience. Our data are compared with other available experimental data and with the results of recent cross section calculations.

This work was supported by the Planetary Atmospheres Program of NASA.

[D15.51] Absolute partial cross sections for electron-impact ionization of CH_4, H_2O, and D_2O from threshold to 1000 eV.

Absolute partial cross sections for the production of both parent and fragment ions from electron-impact ionization of CH_4, H_2O, and D_2O are reported for electron energies from threshold to 1000 eV. The product ions are mass analyzed using a time-of-flight mass spectrometer and detected using a position-sensitive detector whose output demonstrates that all product ions are completely collected. The overall uncertainty in the absolute cross section values for singly charged parent ions is \pm3.5% and is slightly greater for fragment ions. Previous cross section measurements are compared to the present results.

[D15.52] Electron Attachment at Near Zero Scattering Energy

A new apparatus has been developed for measuring electron attachment cross sections at very low scattering energies. Electrons are generated with \approx 1 meV resolution by laser ionization of atomic xenon. Tunable ultraviolet radiation at \approx 276 \: nm, produced through nonlinear mixing, is frequency tripled in a pulsed supersonic jet of xenon. The resulting vacuum ultraviolet photon energy is at, or slightly larger than, the Xe^+ (^2P_1/2) ionization potential. Photoelectrons, continuously tunable between 0 and 100 meV, interact with target molecules in a 4\pi steradian scattering arrangement. The vacuum ultraviolet wavelength is accurately calibrated by observing the autoionizing Rydberg series leading to the Xe^+ (^2P_1/2) threshold. This has allowed the determination of the nd Rydberg series up to quantum levels 30% closer to the ionization threshold than had previously been measured. The details of electron attachment to SF_6 at energies below 10 meV are currently being investigated. This work was carried out at JPL/Caltech, and supported through agreement with NASA.

[D15.53] Electron Attachment to BCl_3

I have used a numerical optimization algorithm(W.L. Morgan, Phys. Rev. A \textbf44), 1677 (1991); J. Phys. D \textbf26, 209 (1993). to derive low energy electron attachment cross sections for BCl_3 from published attachment rate coefficient data. The low energy cross section peaks at about 10 Åbetween 0.1 and 0.2 eV. The energy is in agreement with Olthoff's(J.K. Olthoff, Ph.D. Dissertation, University of Maryland (1985).) unpublished relative measurements. Due to the size of this cross section, its position relative to the BCl_3 vibrational excitation thresholds, and to the Ramsauer minimum in the Ar momentum transfer cross section, its effects on the drift velocities measured(D.L. Mosteller, Jr., \textitet al.), J. Appl. Phys. \textbf74, 2247 (1993). in Ar/BCl_3 mixtures is expected to be significant. This will be discussed using Boltzmann calculations of the drift velocities in such mixtures.

[D15.54] VUV Photon Emission Following Electron Impact on Ar Clusters. *

A pulsed electron beam crossed with a pulsed cluster beam has been used to study the excitation of Ar clusters. Photons were detected using a CsI-coated channel electron multiplier. Use of a variety of filters enabled the emissions from Ar ions, neutral atoms and molecules to be identified. Measurements of excitation functions revealed a number of excitation mechanisms occuring in, or on the surface of, the clusters.

* Research supported by the Natural Sciences and Engineering Research Council of Canada.

[D15.55] Dissociative Excitation of Normal and Heavy Water by Electron Impact.*

Pulsed electron beam and time-of-flight techniques, combined with a novel surface matrix detector, have been used to study the break up of water molecules yielding metastable O(singlet S) fragments. An energy range from threshold to 400 eV has been explored. Fragment kinetic energies peak around 0.2, 0.27eV for H2O, D2O respectively. The energy distributions display long tails extending to a few eV.A number of dissociation channels have been identified.

* Research supported by the Natural Sciences and Engineering Research Council of Canada.

[D15.56] O(^1D) from the Dissociative Recombination of O_2^+

The dissociative recombination of the low vibrational levels of the O_2^+ ground state leading to O(^1D) atoms is dominated by three routes: B ^3\Sigma_u^-, ^1\Delta_u, and f'^1\Sigma_u^+. We report new large scale ab initio calculations for the potential curves and new calculations of the electron capture widths, cross sections and rates. The results for the equilibrium separations agree to three significant figures with experiment. In order to compare our free electron capture widths to experiment, we have calculated predissociation widths for the lowest Rydberg states. The width for f^1\Sigma_u^+ is 1698cm^-1 at R=2.5a_o in agreement with the experimentally derived width(B. R. Lewis et al., Phys. Rev. A 52, 2717 (1995)) of 1500-2000cm^-1. For B^3\Sigma_u^-, the width at R=2.2819a_o is 4319cm^-1 and is in agreement with an experimentally derived R independent value(B. R. Lewis et al.,J. Chem. Phys. 102, 6631 (1995)) of 4038cm^-1. Using Multichannel Quantum Defect Theory,(S. L. Guberman and A. Giusti-Suzor, J. Chem. Phys. 95, 2602 (1991)) it is shown that indirect recombination reduces the rate constants for generating O(^1D) below those for the direct process only.(S. L. Guberman, Planet. Space Sci. 36, 47 (1988))

[D15.57] Measurement of iron L-shell dielectronic recombination (DR) rates for X-ray photoionized nebulae: Fe XVIII and Fe XIX \Delta n=0 DR

In an X-ray photoionized nebula (XPN) the ionization structure is driven by photoionization balanced by radiative recombination (RR) and DR; and the electron temperature at which an ion is formed is far below the temperature at which the ion would exist in coronal equilibrium. (Kallman et al.), Astrophys. J. 465, 994 (1996). (Arnaud, M. and Raymond, J., Astrophys. J. 398), 394 (1992). As a result of the low electron temperature in an XPN, X-ray line emission is produced by RR and DR and not by electron impact excitation.^\ 1 For the iron L-shell ions (Fe XVII to Fe XXIV) which are a dominant source of line emission in the 0.8-2.0 keV bandpass, DR proceeds via \Delta n=0 core excitations. To address the needs for interpreting spectra from present and upcoming X-ray astronomy satellite missions, we have initiated a program to measure the \Delta n=0 DR rate coefficients for the iron L-shell ions. We will present our first results on DR of Fe XVIII and Fe XIX.

[D15.58] Very-high-n Stark wavepackets generated by an electric field step.^*

Wavepackets comprising a coherent superposition of high-n Stark states have been created by rapid application of a DC field to potassium np atoms with n \sim 388. Their evolution is examined using a short half-cycle probe pulse that is applied following a variable time delay and which ionizes a fraction of the excited atoms. The survival probability exhibits pronounced oscillations (quantum beats) that are associated with the time evolution of the wavepacket. Interestingly, even in the present regime where several different Stark manifolds overlap, a single dominant beat frequency is observed. Quantum mechanical calculations show that this can be explained in terms of energy level statistics. Classical trajectory Monte Carlo simulations also reproduce well the observed beats without recourse to any adjustable parameters, further demonstrating classical-quantum correspondence in high-n systems. Damping of the beats is slow suggesting it might be possible to detect dephasing induced by collisions. \vskip 12pt ^*Research supported by the NSF, the R. A. Welch Foundation, and the U.S. DOE, OBES, Div. of Chem. Sciences, under Contract DE-AC05-96OR22464 with ORNL managed by LMERC and by the NSF.

[D15.59] Single-Shot Detection of Wavepacket Evolution

We have developed a detector for monitoring the time-dependent evolution of dynamic states in atoms using a single electromagnetic pulse pair. The detector utilizes pump and probe beams which cross through a sample of atoms. The pump pulse excites a wavepacket in each atom. The final state of the atom after the probe pulse depends on the evolution of the non-stationary state during the time interval between the two pulses. However, because the beams cross at a non-zero angle, the relative delay between the pump and probe pulses varies linearly across the interaction region. The spatial distribution of the ions formed directly by the probe or through subsequent photo- or field ionization is imaged using a phosphor screen and CCD camera. In this way, wavepacket evolution can be monitored over a range of times in a single shot. This detector greatly reduces the time required to make time-domain measurements using low repetition rate laser systems. The resolution of the detector is determined by the angle between the pump and the probe beams. We have achieved resolutions of <1 fsec in a nearly colinear geometry and <1 psec for counter-propagating beams using the same apparatus.

[D15.60] Electron dynamics in a strong electric field

We present calculations for the dynamics of an electron wave packet in a strong electric field. The wave packet is created by a pulsed laser interacting with an alkali atom in its ground state. The wave packet has an average energy between the classical ionization threshold in an electric field and zero energy. In this energy region, the wave packet exhibits quite complicated time dependence. To compare with recent experiments(G.M.~Lankhuijzen and L.D.~Noordam, Phys.~Rev.~Lett.~76), 1784 (1996), the wave packet is probed by measuring the time dependent flux into a detector a large distance down field. We connect the behavior of the wave packet to som closed orbits in the field. There are several interesting features of these wave packets, including a peculiar confining effect for the escaping electrons and standing wave type patterns down field.

[D15.61] A New Approach to the Production of Coherent Atomic States.

We report a new method for production of coherent atomic states. The method is based on perturbation of atoms in excited Stark states by nanosecond electric field pulses. These pulses destroy the energy eigenstates of the time independent hamiltonian, creating instead a coherent superposition of them. The process that generates the wave packet is described and our initial measurements indicating the presence of the state are presented. In addition, we describe the relationship between the motion of the wave packet and the classical description of the atom it represents.

[D15.62] Coherent States Composed of Stark Eigenfunctions of the Hydrogen Atom.

It is well known that for the non-relativistic hydrogen atom it is possible to separate and solve the Schrödinger equation in parabolic as well as spherical coordinates. The eigenfunctions obtained in these coordinate systems are each a suitable basis set in the absence of an electric field, but only the parabolic states, the Stark eigenfunctions, retain their character in the presence of a weak field. The properties of coherent superpositions of these Stark states are investigated and the motion of the wave packet described. It is shown that a properly constituted superposition will mimic classical motion. It is also shown that the constant energy separation between adjacent Stark states of a given principal quantum number, produce wave packets that execute periodic motion. In general, they split into distinct "clumps" of probability, but revive to form a single packet after each period. It is, however, possible to construct a packet that maintains its shape for all time. A discussion of the classical equations of motion for this system may be viewed at the www site for the UMSL Atomic Physics Lab.

[D15.63] Ramsey Interference of ``Stroboscopic'' Rydberg States of Diatomic Molecules

In some regions of the spectra of diatomic molecules, the rotational energy intervals and the Rydberg energy intervals are commensurate; therefore, the period of the electronic motion is commensurate with the period of the nuclear rotational motion. These regions contain manifolds of entangled states with the same total angular momentum, but very different relative amounts of rotational and electronic orbital energy and angular momentum. The angular momentum coupling for these states is Hund's case (a), and when the electron revisits the core, the scattering is regular. The regularity and intensity of the spectrum is known as the ``stroboscope effect." We report progress on Ramsey fringe spectroscopy of stroboscopic states of Na_2 molecules.

[D15.64] Ionization of Very-High-n Rydberg Atoms by Multiple Half Cycle Pulses in the Short-Pulse Regime.^*

The ionization of potassium np Rydberg atoms with n \sim 388 by a train of electric field pulses whose duration (\sim 2 ns) is short compared to the classical orbital period of the excited electron (\sim 9 ns at n \sim 388) is investigated experimentally and the data compared with the results of classical trajectory Monte Carlo (CTMC) calculations. Ionization probabilities measured as a function of the amplitude, number (typically 10 to 50), and frequency of the pulses agree well with theory on an absolute scale without any adjustable parameters. It is observed that the peak field required to induce ionization decreases as more pulses are applied. More importantly, the Rydberg atom survival probability exhibits a broad maximum for pulse repetition frequencies near twice the classical orbital frequency, indicating a dynamic stabilization of the Rydberg atoms. Work is now under way to understand the physical origin of this behavior. \vskip 12pt ^*Research supported by the NSF, the R. A. Welch Foundation, and the U.S. DOE, OBES, Div. of Chem. Sciences, under Contract DE-AC05-96OR22464 with ORNL managed by LMERC and by the NSF.

[D15.65] Time-Gated Photoionization Spectroscopy of Cesium Rydberg Wave packets in an Electric Field.

The electron emission of atomic cesium after optical excitation with a picosecond laser pulse is measured with picosecond time resolution using an atomic streak camera. By scanning the frequency of the exciting laser and measuring the electron flux in certain time gates, time gated photoionization spectra are obtained. We found that for cesium Rydberg wave packets in an electric field of 1.0 kV/cm the electron ejection probability is enhanced when the radial and angular oscillation periods of the excited wave packet are commensurable. The experimental data is confirmed by quantum calculations of the time resolved electron emission.

[D15.66] Coherent Phase Control of the Photoionization and Photodissociation of HI and DI

Recently we reported the coherent phase control of the competition between ionization and dissociation of HI molecules.\footnote L. Zhu, V. Kleiman, X. Li, S. Lu, and R. J. Gordon, Science 270, 77 (1995). Control was achieved by absorption of three photons of frequency ømega_1 and one photon of frequency ømega_3 = 3 ømega_1 just above the first ionization threshold. The detected products were HI^+ and I^+. We now address two key aspects of the control mechanism. First, by exciting HI with a pulse of 266 nm radiation preceding the control pulses, we ruled out the possibility that the modulated I^+ signal was produced from I atoms generated by 2 ømega_1 photodissociation of HI. Second, the phase lag \Delta \phi of the two ion signals was measured as a function of the excitation energies of HI and DI in the vicinity of the 5s\sigma molecular Rydberg state. For both isotopes \Delta \phi reached a minimum at the peak of the one-photon absorption profile. The data indicate that at the center of the profile a discrete state is populated, so that the transitions to the two continua are indirect. In this region the matrix elements for the transition operators can be written as the products of absorption and decay terms, and the phase dependence of the branching ratio cancels out. Far from the center of the profile the continua are excited directly, and the initial phase information is retained.

[D15.67] Reflection on a modulated evanescent wave mirror : from diffusion to diffraction

We present a study of normal incidence reflection of atoms on an evanescent wave mirror. In this experiment, a cloud of cold atoms is released from a MOT 17-mm above a prism on which a 3W laser beam at total reflection creates an evanscent wave.

When the laser is detuned very far from the atomic resonance (i.e. when the spontaneous emission rate is low), the reflection is expected to be specular. Nevertheless, we observed significant momentum diffusion after one bounce (A. Landragin, G. Labeyrie, C. Henkel, R. Kaiser, N. Vansteenkiste, C.I. Westbrook and A. Aspect, Opt. Lett. 21, 1591 (1996)). Theoretical studies (C. Henklel, K. Molmer, R. Kaiser, N. Vansteenkiste, C.I. Westbrook and A. Aspect, to be published in Phys. Rev. A (feb. 1997)) showed that this could be infered to spatial modulation of the atomic mirror created by interferences between the evanescent wave and the light diffused by the prism.

In order to verify these assumptions, we studied normal reflection when a standing evanescent wave of well controlled visibility is created on the dielectric. This was done by retroreflecting a small fraction of the incident light back to the prism. We observed diffraction (C. Henkel, J.Y. Courtois and A. Aspect, J. Phys. II (France) 4,1955 (1994)) (P.L. Gould, G.A. Ruff and D.E. Pritchard, Phys. Rev. Lett. 56, 827 (1986)) even for a very small visibility of the standing evanescent wave (only 0.005 % of the light returning to the prism). Furthermore, the experimental signal fits well with theoretical models in which the contribution of the Van der Waals potential to the atomic mirror in the vincinity of the dielectric is taken into account (A. Landragin, J.Y. Courtois, G. Labeyrie, N. Vansteenkiste, C.I. Westbrook and A. Aspect, Phys. Rev. Lett. 77, 1464 (1996)).

[D15.68] Atom Guiding Through Hollow-Core Optical Fibers

We use evanescent laser light to guide Rb atoms through hollow-core optical fibers. Previously, we demonstrated that light coupled into the cladding region surrounding the hollow core of a fiber will reflect atoms from the inner wall through dipole forces. Atoms can then be transported through the fiber with low loss and little heating. We will report progress on increasing the guided atomic flux by serveral orders of magnitude by loading the fiber with the output of a recently developed low velocity atomic beam source. This source of precooled atoms will facilitate the creation of single transverse mode atoms in micron sized hollow-core fibers, analagous to single mode optical fibers. Applications to atomic interferometry will be discussed.

[D15.69] Conical emission from a self-focused laser beam

We report observations of conical emission from a single self-focused laser beam, or filament, propagating with nearly constant diameter through a strontium (Sr) vapor. The experimental arrangement consists of a 5 cm long vapor cell with a Sr density of \sim 10^14\;cm^-3 in \sim1 Torr of Ar buffer gas, and a 4 ns Nd:YAG pumped dye laser tuned to the blue of the 5s^2\; ^1\!S_0 - 5s5p\; ^1\!P_1 (460.7 nm) transition. Emission angles from the vapor cell are imaged into height on the input slit of a 1.5 m monochromator. The exit plane of the monochromator is imaged onto a CCD camera providing angle versus wavelength data. The incident beam diameter and power are matched to the minimum diameter stationary filament solution for our experimental conditions. (M. L. Dowell, B. D. Paul, A. Gallagher, and J. Cooper, Phys. Rev. A 52), 3244 (1995) At the critical self-focusing detuning, where diffraction and self-focusing are balanced, a faint cone can be observed. Once the detuning is decreased further, the filament goes through oscillations in its size as it propagates through the cell, and the cone becomes brighter. Quantitative data on the dependence of cone intensity on the various experimental parameters will be presented.

[D15.70] Z-Scan Measurement of Nonlinear Optical Properties of Sulfur Rich Materials

The search for materials exhibiting large third order susceptibilities has been of importance to many disciplines. Semiconductor physics including nonlinear propagation in fibers, optical switching, and self focusing effects have been of recent interest, as well as the study of conjugated polymers which have application in electronics and fast optical devices. With the discovery of large hyperpolarizabilities in simple organic and organometallic compounds, interest has moved towards synthesis of macroscopic nonlinear materials based on molecular conjugation. We report here the use of the Z-scan technique to measure the nonlinear susceptibility and absorption of a series of sulfur rich materials in an effort to characterize some of the molecular structures necessary to generate a strong \chi^3 material.

[D15.71] Temporal Behavior of Pulsed Optically Pumped Lasers.

We discuss the temporal behavior of optically pumped laser (OPL) emission that is produced when K_2 vapor is pumped with a pulsed dye laser. K_2 is contained in a heat-pipe oven. The temporal behavior of the OPL emission with and without the addition of a weak probe laser tuned to the OPL emission transition is considered. The probe laser temporally overlaps and leads the pump laser into the oven. Results of numerical simulations are included.

[D15.72] Study of Threshold Condition for the Interference-based Suppression of 3-Photon Resonant Ionization Signals(RIS) in C_2H_2 Due to Laser Bandwith Considerations

This abstract was not submitted electronically.

[D15.73] Experimental Work on Quantum Computation at Los Alamos National Laboratory

There has been heightened interest in quantum computation since 1994, when it was shown (P.W. Shor, Proc. 35th FOCS, S. Goldwasser ed. (IEEE, Los Alamitos, 1994)) that a quantum computer could efficiently factor large numbers. A scheme was proposed (J. I. Cirac and P. Zoller, Phys. Rev. Lett. 74, 4094 (1995)) for performing quantum logic gates with cold trapped ions; ions cooled to their motional ground state in a linear trap may be used as the individual registers and entangled states between these ions can be constructed. At Los Alamos, we have initiated a program to test the feasibility of this proposal. We have designed and built a linear RFQ ion trap, which we have used to trap Ca+ ions. We have constructed a laser system for cooling the ions, and have used it to image ions in the trap. We will discuss the cooling of the ions to their Doppler limit, and describe our plans to further cool them to the vibrational ground state in the near future.

[D15.74] Quantum Computing in Fock Space Systems

Fock space system (FSS) has unfixed number (N) of particles and/or degrees of freedom. In quantum computing (QC) main requirement is sustainability of coherent Q-superpositions. This normally favoured by low noise environment. High excitation/high temperature (T) limit is hence discarded as unfeasible for QC. Conversely, if N is itself a quantized variable, the dimensionality of Hilbert basis for qubits may increase faster (say, N-exponentially) than thermal noise (likely, in powers of N and T). Hence coherency may win over T-randomization. For this type of QC speed (S) of factorization of long integers (with D digits) may increase with D (for 'ordinary' QC speed polynomially decreases with D). This (apparent) paradox rests on non-monotonic bijectivity (cf. Georg Cantor's diagonal counting of rational numbers). This brings entire aleph-null structurality ("Babylonian Library" of infinite informational content of integer field) to superposition determining state of quantum analogue of Turing machine head. Structure of integer infinititude (e.g. distribution of primes) results in direct "Platonic pressure" resembling semi-virtual Casimir efect (presure of cut-off vibrational modes). This "effect", the embodiment of Pythagorean "Number is everything", renders Godelian barrier arbitrary thin and hence FSS-based QC can in principle be unlimitedly efficient (e.g. D/S may tend to zero when D tends to infinity).

[D15.75] Implementation of Qubits by Coaxial RF Squids

Coaxially-centered,narrowly spaced, inductively coupled RF squids are investigated as the physical basis for qubits. In the case of two such squids we denote the states by (nS,mA) where nS( mA) is the oscillator level of the symmetric(antisymmetric) normal mode. One could,e.g.,assign qubit 0 to the state (0,0) and qubit 1 to (0,1). Calculations indicate that excited states of the antisymmetric mode , (0,n), are radiatively long-lived compared to the (n,0) states by factor of (spacing between loops/wavelength) squared. For the implementation of logic gates, a method employing coaxial alignment of several RF squids with concommitant RF irradiation is presently under investigation. For these qubits the (N)AND variety of gates appear easier to realize than does the XOR type.

[D15.76] POVM Receivers for Quantum Cryptography

Positive operator valued measures (POVMs) are finding increasing use in quantum cryptography.(A. Peres, Quantum Theory: Concepts and Methods, Kluwer Academic Publishers, Boston (1993).) I present quantum circuit analyses of two recently proposed (H. E. Brandt, J. M. Myers, and S. J. Lomonaco Jr., "New Results in Entangled Translucent Eavesdropping in Quantum Cryptography," in Photonic Quantum Computing, S. P. Hotaling and A. R. Pirich, Editors, Proc. SPIE 3076 (1997)) all-optical designs for POVM receivers to be used in conjunction with Bennett's B92 protocol for key distribution based on two photon polarization states. Expectation values for the POVM operators are calculated and shown to be faithfully realized by the POVM detectors.

[D15.77] Determinstic Single Photon Generation in Cavity QED

We present an efficient scheme of generating a single-photon state transmitted out of an optical cavity. For realistic cavity QED parameters, we show that the scheme can produce a single-photon pulse with a probability exceeding 99% in a user-specified time interval. By recycling the system, the scheme can be used to create a stream of single-photon pulses in which binary information can be encoded.

[D15.78] Demonstration of Quantum Memory and Creation of Massive EPR Pairs.

This abstract was not submitted electronically.

[D15.79] Source for slow atoms without laser cooling

Many atomic physics experiments use laser cooling to provide a source of slow atoms. We are investigating the use of magnetic fields to separate out the low velocity atoms (\leq 200 m/s) in an atomic beam. In theory, this method should produce a slow-atom flux which is competitive with laser cooling methods. The use of permanent magnets reduces cost and complexity. The magnets create a potential which guides slow atoms along a circular arc, while fast atoms are lost from the atomic beam. The maximum velocity of the guided atoms is set by the radius of the circle and the strength of the magnets. Preliminary results yield a guided slow-atom flux of \sim 10^5 atoms/s. This number is lower than expected, due apparently to a non-Maxwellian distribution of slow atoms in the beam caused by collisions with fast atoms.

[D15.80] The Recoil Regime: A New Domain of Laser Cooling

We have begun exploration of the domain of laser cooling where the ratio of the recoil frequency ømega_r to the natural width \gamma no longer satisfies ømega_r/\gamma\equiv\varepsilon\ll1. For polarization gradient cooling (lin\perplin in our case) on the J=1\rightarrow2 transition 2\,^3S\rightarrow2\,^3P of metastable He, the 1-D cooling limit is \approx25\,\hbarømega_r, whereas the smaller Doppler limit is 7\,\hbar\gamma/20\approx13\,\hbarømega_r. In the usual cases of laser cooling, the Doppler limit is much larger (\approx25\times larger for ^85Rb where \varepsilon\approx1/1600). For this transition in He, \varepsilon\approx1/40, still small compared with unity, but large enough to produce this unusual circumstance. In combination with our calculations, we have found that the lin\perplin cooling helps the Doppler cooling instead of the other way around. The steady-state presence of atoms in velocity space outside the linear region of the damping force results in velocity distributions with quite non-Gaussian shapes. We are beginning experiments on the 2\,^3S\rightarrow3\,^3P transition of He* at \lambda=389~nm, where \varepsilon\approx1/5 to push further into this unexplored, recoil-dominated domain.

[D15.81] Experimental and computational analysis of Doppler cooling and the magneto-optic trap.

Cold, dense clouds of cesium atoms have been produced in an undergraduate research lab using Doppler cooling and a vapor cell magneto-optic trap (MOT). Measurements of temperature, density, and filling of a MOT will be reported. In conjunction with this experiment, Windows-based software has been developed to simulate laser cooling and trapping. The program graphically depicts the position and velocity of a variable number of atoms in two dimensions as they are cooled and trapped. The algorithm accounts for absorption, spontaneous and stimulated emission, Doppler and Zeeman shifts, and radiation trapping. The user can adjust time step, atom velocity, laser intensity and detuning, and magnetic field gradient. The program can demonstrate optical molasses, a MOT, and a dark MOT. Simulations can be recorded and played back in the same graphical environment, along with movies from the actual experiment. The simulation is a useful educational tool for explaining and investigating Doppler cooling and the magneto-optic trap. The program allows anyone to obtain computational results and compare them with published experimental results. A beta version of Cool Simulation is downloadable off the World Wide Web.

[D15.82] Trap loss in photoassociative spectroscopy.

Measurements of photoassociative spectroscopy in a gas of lithium atoms at ultralow temperatures have been carried out (E.R.I Abraham, N.W.M. Ritchie, W.I. McAlexander, and R.G. Hulet, J. Chem. Phys. 103), 7773 (1995); E.R.I Abraham, W.I. McAlexander, J.M. Gerton, R.G. Hulet, R. Côté, and A. Dalgarno, Phys. Rev. A 53, R3713 (1996).. In the experiments, photons are absorbed into high vibrational levels of excited electronic singlet and triplet states of Li_2 (R. Côté, A. Dalgarno, Y. Sun, and R.G. Hulet, Phys. Rev. Lett. 74), 3581 (1995).. The excited vibrational levels decay by spontaneous emission into the vibrational continuum and the discrete vibrational levels of the ground electronic states (R. Côté and A. Dalgarno, to be published in Phys. Rev. A.). Decay into the vibrational continuum can produce atoms with sufficient kinetic energy to escape the trap, while spontaneous emission into discrete levels produces molecules no longer trapped. We explore the trap loss from spontaneous emission as a function of the trap depth as well as the excited vibrational levels considered. The analysis is conducted for the singlet and triplet transitions for both ^6Li and ^7Li isotopes.

[D15.83] Magneto-Optic Trapping of Radioactive ^38mK and ^37K for Precision Nuclear Experiments

We report on our progress to test weak interaction symmetries in \beta-decay experiments on radioactive potassium atoms in a magneto-optic trap. Ions produced in TRIUMF's on-line isotope separator, TISOL, are neutralized and trapped in a vapor-cell type MOT. The atoms are ejected from this collection trap using a spatially narrow, near-resonant push laser and transferred to a second MOT in a chamber containing nuclear detectors. The use of the atom transfer and a second trap provides a backing-free, localized source of atoms and has allowed us to observe \beta ^+ particles in coincidence with Ar^0 and Ar^+ recoils. Use of a position sensitive microchannel plate for recoils and a position sensitive \beta-spectrometer will allow us to deduce the neutrino momenta. Supported in part by NSERC.

[D15.84] Long-Range Interaction of ^39K(4s) + ^39K(4p) Asymptote by Photoassociative Spectroscopy of Ultracold ^39K Atoms

Analytical expressions for the sixteen adiabatic Hund's case (c) long-range potential curves are derived including the higher order dispersion forces and the effects of retardation. Six Hund's case (c) long-range molecular states including the "pure long-range" 0_g^- state are observed with rotational resolution by photoassociative spectroscopy in a high density magneto-optical trap. The long-range potential constants of the 4s + 4p interaction are precisely determined using the measured molecular binding energies of the 0_g^- state. The long-range potential curve of the 1_g state calculated with the experimentally determined long-range constants is in excellent agreement with the long-range portion of the RKR potential curve of the 1^1\Pi _g state determined by conventional molecular spectroscopy. This confirms the internal consistency of the theoretical model used in this work. The radiative lifetime of the K 4p state derived from the resonant dipole-dipole interaction constant C_3^\Pi also agrees excellently with a recent fast-beam measurement.

[D15.85] Impact of the ^87Rb singlet scattering length on suppressing inelastic collisions

A computation of ^87Rb - ^87Rb spin-exchange rate constants as functions of the two-body singlet scattering length a_s demonstrates that the inelastic collision rate is suppressed over a small range of the possible values for a_s. A two-channel model relates this inelastic suppression to an interference phenomenon, manifested in the near coincidence of the singlet and triplet scattering lengths. This mechanism explains the diminished rates measured in recent ``double trap'' experiments. Combining information extracted from these rates and from previous scattering length measurements allows us to place bounds on the ^87Rb singlet scattering length (74-102 a.u.).

[D15.86] Triplet s-wave resonance in ^6Li collisions and scattering lengths of ^6Li and ^7Li

Using two-photon photoassociative spectroscopy of the \mboxa^3\Sigma_u^+ ground-state of ^6Li_2, the triplet s-wave scattering length of ^6Li is determined to be -2160 \pm 250\ a_o, where a_o is the Bohr radius. This scattering length is the largest known for any atomic species and its magnitude signifies a near-threshold resonance. This resonance may enable unique investigations of large collisional frequency shifts and may lead to a BCS-type phase transition in a degenerate gas of ^6Li. This experiment is the final one in a series made to determine the scattering lengths of lithium. The previous measurements have been reanalyzed using new long-range coefficients and a coupled channel calculation. The most current values for the singlet and triplet scattering lengths of ^6Li and ^7Li as well as the mixed isotope ^6Li^7Li are presented. This inclusive set can be used to determine the low-energy scattering length of any binary collision in lithium.

[D15.87] Photoassociation of metastable Ne using a bright atomic beam

We present experimental results of an atomic beam brightener, along with theoretical studies of photoassociation spectroscopy in this beam. The atomic beam of metastable neon atoms is first collimated via transverse Doppler cooling, then Zeeman-slowed along the beam axis, and finally compressed in a 2-D magneto-optical trap. Experimental results for each individual stage, and for the combination as a whole, will be presented. One goal of this project is to study atomic properties of the rare gases via photoassociation spectroscopy involving intrabeam cold collisions. We have determined molecular states for the case of colliding neon partners, and have calculated optical potentials to model ionization( Our optical potentials are calculated using methods detailed by S. S. Op de Beek, J. P. J. Driessen, H. C. W. Beijerinck, and B. J. Verhaar, J. Chem. Phys. 106) 182 (1997). . Numerically calculated photoassociation spectra based on these results will be presented. This work is supported by the Foundation for Fundamental Reseach on Matter (FOM) and the Royal Dutch Academy of Sciences (KNAW).

[D15.88] Control of ultracold atomic collisions with resonant light

We present a detailed semianalytic theory of cold atom collisions in the presence of one or more laser fields tuned near molecular resonances. This theory expands upon a previous development(J. L. Bohn and P. S. Julienne, Phys. Rev. A 54), R4367 (1996). by accounting explicitly for light-induced shifts and line broadenings apparent at high laser intensities. We use this theory to determine realistic laser settings for guiding the outcome of cold collision processes. One aspect of such control is the production of ensembles of ultracold molecules, which can then be trapped in FORT experiments.

[D15.89] Quantum defect theory of atomic collisions and molecular vibrational spectra

A quantum defect theory of atomic collisions and molecular vibrational spectra is presented. Based on the exact solutions of the Schrödinger equation for a 1/r^6 potential, the theory provides a systematic interpretation of molecular bound states and atom-atom scattering properties and establishes the relationship between them. Two applications will be discussed in some detail. The first is the analytic description of energy dependences of cold-atom collision cross sections. The second is the prediction of scattering length and C_6 coefficient from experimental measurements of binding energies. Results for systems including ^7Li_2 and ^23Na_2 will be presented.

[D15.90] Direct Observation of Fine Structure Exchange Predissociation of the 0_u^+ State of ^39K_2 by Photoassociative Spectroscopy.

We have directly observed for the first time the predissociation of the long-range 0_u ^+ state (dissociating to the 4^2S_1/2 + 4^2P_3/2 limit) of ^39K_2 by ionization detection of the atomic fragment of ^39K (4^2P_1/2) in a high density magneto-optical trap. The predissociation is presumably caused by the curve crossing at short-range between the b^3\Pi _u (Ømega = 0^+) and the A^1\Sigma _u^+ states which dissociate asymptotically to the 4^2S_1/2 + 4^2P_3/2 and the 4^2S_1/2 + 4^2P_1/2 limits, respectively. Through the fine structure exchange predissociation, the atomic fragments gain sufficient kinetic energy to escape from the trap. Therefore, the free-bound resonant excitations to this 0_u^+ long-range state of ^39K_2 result in significantly larger trap loss than excitations to other long-range molecular states.

[D15.91] Progress towards a cesium atomic fountain clock

We have been developing a fountain of laser--cooled cesium atoms for use as an atomic clock. Our design largely follows that of the fountain built at LPTF in Paris. In our fountain, chirp--slowed atoms are first collected in a Magneto--Optic Trap (MOT) and then cooled to a few \muK in optical molasses. The cooled atoms are then launched vertically into a "moving molasses" by shifting the frequencies of the vertical cooling beams. The atoms then travel through a microwave cavity tuned to the 9.2 GHz cesium hyperfine frequency for a first Ramsey pulse. After roughly 0.5 seconds of free flight under the influence of gravity, the atoms fall back through the microwave cavity and into an optical state--detection region which detects the number of atoms making the F=3 \rightarrow F=4 transition. The increased Ramsey interaction time improves the short--time precision as compared to traditional atomic beam experiments, while many systematic shifts which limit the accuracy of an atomic beam clock are reduced by the low atomic velocity and the retrace of the atomic trajectory through the microwave cavity. We will discuss the progress towards a working fountain being assembled in our laboratory.

[D15.92] Driven Equations in Time-Dependent and Time-Independent Quantum Mechanics

Driven or inhomogeneous equations have been used to great advantage in problems involving the photodissociation of molecules^1-3. This idea has been extended to treat time-independent scattering and the evolution of an initial wavepacket in a time-dependent potential. By formulating these problems as inhomogeneous partial differential equations, it is possible to incorporate the required boundary or initial-value conditions in a very convenient fashion. For the time-independent scattering problem, this is equivalent to a direct calculation of the scattered wave part of the Schroedinger wavefunction where the imposed boundary conditions may be shown to be independent of the scattering event. This reformulation of the problem enables the user to directly extract the scattering information from the asymptotic form of the inhomogeneous wavefunction. For the time-dependent Schroedinger equation, the inhomogeneous equation formulation, coupled with a novel DVR basis set expansion in both the space and time coordinates, reduces the entire problem to the solution of a set of sparse, complex linear algebraic equations. ^1 Y. B. Band, K. F. Freed and D. J. Kouri, J. Chem Phys. 74, 4380(1980) ^2 K. C. Kulander and J. C. Light, J. Chem. Phys. 73, 4337(1980) ^3 B. I. Schneider and H. S. Taylor, J. Chem. Phys. 77, 379(1982)

[D15.93] Density-Matrix Descriptions of Optical Interactions in Quantized Electronic Systems

Density-matrix descriptions have been developed to investigate the influence of relaxation phenomena during resonant and non-resonant radiative transitions of quantized electronic systems, including atomic systems and quantum-confinement systems (e. g., semiconductor microstructures). Radiative and collisional relaxation phenomena have been treated using Liouville-space projection-operator techniques. Both time-independent (resolvent-operator) and time-dependent (equation-of-motion) formulations have been developed. The self-energy operator that occurs in both formulations provides the basis for a self-consistent determination of the non-equilibrium electronic-level populations and the spectral-line shapes. The effects of quantum confinement on the non-equilibrium kinetics and the optical spectra are investigated for quantum confinement in one, two, and three dimensions.

[D15.94] Unitary Integration: A Numerical Technique Preserving the Structure of the Quantum Liouville Equation

Liouville's equation for an~N-level atomic system interacting with radiation has a non-trivial kinematic structure; the quantities~Tr\,\rho^n,~n=1,\dots,N remain constant in time, independent\/ of the Hamiltonian. These invariants are physically significant; the qualitative character of the solution depends on their existence. A generic numerical method, by its very nature, will not in general preserve these invariants. We present a numerical technique, ``Unitary Integration,'' that exactly\/ preserves these invariants to all orders\/ in the time step. This method, which evolves the density matrix via\/ unitary transformations, is the quantum mechanical analog of symplectic integrators commonly used in classical mechanics. Although the evolution is approximate (as in any numerical method), the structure of Liouville's equation is preserved exactly since the time advance map is constructed so as to yield a unitary transformation. For Hamiltonians possessing dynamical symmetry this method can be used to construct integrators that preserve all invariants arising from this symmetry as well as the generic~Tr\,\rho^n invariants. We provide examples, comment on applications and discuss numerical stability and efficiency.

[D15.95] Parafermionic Extension of the Jaynes Cummings Model

The interaction between an atom with p+1 levels and a one-mode electromagnetic field is simulated by a Jaynes Cummings model consisting of a parafermion of order p coupled to a boson. The symmetry algebra of the system is a deformed u algebra, possessing a Schwinger realization in terms of one parafermion and one boson. The representation theory of the deformed u(2) algebra is constructed. Mean values as well as time averaged mean values of observable quantities are calculated.

[D15.96] A Neural Network Method To Analyze Data Compression In Dna And Rna Sequences

Neural network computations on RNA sequences are used to demonstrate that date compression is possible in these sequences. The result implies that a certain discrimination should be achievable between structured vs. random regions. The technique is illustrated by computing the compressibility of short RNA sequences such as tRNA. The method should be valuable in measuring the information content of DNA, including noncoding DNA, which has been shown to display certain properties resembling natural language attributes.

[D15.97] Correlation: How to Understand Complex Systems in Terms of Simple Ones

Correlation is interdependency. Correlation may describe how systems interact with one another or it may indicate the presence of underlying patterns in complex systems. Electron correlation in atomic systems arises from the Coulomb interaction between electrons in atoms. Electron correlation may be detected in observations of binding energies and transitions probabilities. Methods for dealing with the dynamics of electron correlation in simple atomic collisions are both relatively new and varied(J.H.McGuire, Electron Correlation Dynamics in Atomic Collisions), Cambridge University Press, 1997.. There is also evidence for correlation in larger molecular systems ranging from the distribution of H_2O molecules in water to codons in DNA. Mathematically statistical evidence for correlation and long range ordering has been found in much larger and more complex systems(H.E. Stanley, invited talk at this meeting.).

[D15.98] A set of fluid equations for the selfconsistent description of transport and fluctuations in plasmas

A complete set of equations that describes the evolution of a toroidal plasma in two time scales is developed. The plasma is described as a multifluid system, but actually, the ion dynamics is the dominant one. These equations include collisional and non-collisional viscosities as well as diamagnetic effects. The fast scale is related to the evolution of fluctuations produced by plasma instabilities driven by the magnetic field curvature (ballooning modes), and depend on the equilibrium profiles of the plasma parameters. These in turn evolve according to the slow transport scale, which is determined by the properties of the turbulent fluctuations. This set of two-time-scales equations is appropriate to study the formation and evolution of a thermal barrier, resulting by the quenching of fluctuations produced by changes in the equilibrium velocity profile (shear). From the complete equations, a simpler model is derived which is amenable of direct numerical solution, and is used to self-consistently simulate the so-called L-H transition observed in toroidal plasmas.

Part D of program listing