The cross sections of charge transfer and electron detachment in collisions of H$\sp{-}$ and D$\sp{-}$ with Na, K and Cs have been measured for collision energies ranging from 3 to 300 eV. Both charge transfer and electron detachment are significant electron-loss mechanisms for H$\sp{-}$(D$\sp{-}$); both processes exhibit velocity-dependent isotope effects for H$\sp{-}$ and D$\sp{-}$. $\sigma\sb{\rm cg}$(E) displays high energetic thresholds for Na and K (about 20 eV for H$\sp{-}$ + Na and 40 eV for H$\sp{-}$ + K) yet no obvious one for Cs. $\sigma\sb{\rm e}$(E) does not depend on the target as much as $\sigma\sb{\rm ct}$(E) and displays near zero-energy thresholds. The relative importance of charge transfer as an electron-loss mechanism decreases as the mass of the alkali-metal increases.

A quantitative theory of oscillatory spectra for atoms in a magnetic field is developed. When an atom is placed in a magnetic field, and absorption spectrum into states close to the ionization threshold is measured, it is found that the absorption as a function of energy is a superposition of many sinusoidal oscillations. Such interesting and surprising phenomenon are fully explained and described by the theory.;The theory is based on three approximations: (1) Near the atomic nucleus, the diamagnetic field is negligible. (2) Far from the nucleus, the wave propagates semiclassically. (3) Waves returning to the nucleus are similar to (cylindrically-modified) Coulomb-Scattering waves. Using these approximations, together with the simple physical picture of absorption process, formula is derived for the transition rate as a function of final states energy.;The main result is that the transition rate is equal to the sum of two very different kinds of contributions. The first is the averaged transition rate in the absence of the magnetic field, which is a smooth function of energy; the second is itself a sum over many oscillations. Each oscillation is closely associated with a band of wave, initially going out from the nucleus, propagating along a family of trajectories, and finally returning to the vicinity of the nucleus. Because in the center of the family of trajectories is a closed orbit going from the nucleus and returning to the nucleus, we say "a closed orbit makes an oscillatory contribution to the absorption spectrum".;Formulas and algorithms are derived and specified for the calculations of the spectrum from the initial quantum state, dipole polarization, and the properties of the closed classical orbits. Good agreements with experiments were found. Very detailed interpretations are obtained.

Total electron detachment and charge transfer cross sections, $\sigma\sb{\rm e}$(E) and $\sigma\sb{\rm ct}$(E), have been measured for collisions of the negative ions O$\sp{-}$, S$\sp{-}$, F$\sp{-}$, Cl$\sp{-}$, Br$\sp{-}$, I$\sp{-}$, Na$\sp{-}$, and K$\sp{-}$ with atomic hydrogen for laboratory energies ranging from 2 to 500 eV. For the systems F$\sp{-}$, Cl$\sp{-}$, Br$\sp{-}$, O$\sp{-}$ and S$\sp{-}$ + H, $\sigma\sb{\rm e}$(E) displays no barrier for associative detachment; the results are found to be adequately described by simple curve-crossing models based upon available intermolecular potentials, or by classical orbiting models which assume that the anion interacts with the H atom via an attractive potential of the form 1/R$\sp4.$ Analysis of $\sigma\sb{\rm e}$(E) for the system S$\sp{-}$ + H required the cross section for $\rm S\sp{-} + H\sb2\to e +\...$ to be experimentally determined, and these results resolved an apparent discrepancy in previous measurements. The measured detachment cross section for the Cl$\sp{-}$+ H is also found to be in agreement with a calculation for that system based on the effective range potential model. Unlike the other halogen anion-hydrogen systems, $\sigma\sb{\rm e}$(E) for I$\sp{-}$ + H is found to increase with increasing energy over the higher collision energies investigated. The cross section for charge transfer in collisions of O$\sp{-}$ S$\sp{-}$, F$\sp{-}$, Cl$\sp{-}$, Br$\sp{-}$ and I$\sp{-}$ with atomic hydrogen is found to be less than 1 A$\sp2$ over the entire range of laboratory energies investigated. A reasonable extrapolation of $\sigma\sb{\rm ct}$(E) for collisions of O$\sp{-}$ + H is found to agree with a previous measurement at a higher collision energy. For the collision systems K$\sp{-}$ and Na$\sp{-}$ + H, $\sigma\sb{\rm ct}$(E) is found to be much smaller than $\sigma\sb{\rm e}$(E). The measured detachment cross section for Na$\sp{-}$ + H is described using available potential energy curves and by assigning the anion state an average lifetime in the unstable region. A perturbed stationary state calculation of $\sigma\sb{\rm ct}$(E) for the reactant Na$\sp{-}$ is performed, and this calculation underestimates the observed cross section for charge transfer at low collision energies.

A systematic study of the photoabsorption spectra of highly excited hydrogen and alkali atoms in electric fields is presented, based on the semiclassical closed-orbit theory. In most respects, hydrogen and alkali atoms behave similarly, because the excited alkali atoms have a single electron outside of a small ionic core, and the core only produces small shifts of energy levels and small phase shifts of scattered wave functions.;For hydrogen, the classical motion of the excited electron is regular and closed orbits can be enumerated. Above the zero-field ionization threshold, the system is rather simple. There is only one closed orbit, called the parallel orbit, which goes out from the Coulomb center along the electric field and later returns to the center. This orbit is unstable. Nevertheless, the orbit and its repetitions produce recurrences in time, that lead to oscillations in the absorption spectrum. Comparisons between theory and experiments show good agreement.;Below threshold, the parallel orbit becomes stable and, as the energy decreases, many other orbits bifurcate out of it. These closed orbits form orderly patterns, and the associated recurrences are most clear if the absorption spectrum is measured using a scaled-variables method and its Fourier transform, the recurrence spectrum, is computed. Bifurcations are readily observable in such spectra because they create new recurrences, and because at a bifurcation, observed recurrences are especially strong. We predicted the sequence of bifurcations, and the energies at which each would occur, in a paper published early in 1994. Recently, experimental measurements carried out at M.I.T. have confirmed these predictions.;Near a bifurcation, the original form of closed-orbit theory diverges, since a bifurcation is correlated with a focus of classical orbits. An improved closed-orbit theory is derived by using the uniform semiclassical approximation, and by extending the wave function from the real three dimensional space into a four dimensional space. In this extended space, the orbits of the electron near the nucleus are straight lines. These lines are arranged so that they form a cusped caustic, and furthermore they form cylindrical foci in two independent planes in the four-dimensional space. We derive a formula for the wave function associated with this cylindrically focused cusp, and make a new prediction of the behavior of the recurrence spectrum near a bifurcation. These predictions are compared with new experimental results. We find that the improved form of closed orbit theory accurately accounts for experimental measurements both globally and locally.

Absolute yields of negative ions and secondary electrons resulting from collisions of positive sodium ions (Na{dollar}\sp{lcub}+{rcub}){dollar} with polycrystalline aluminum (Al) and molybdenum (Mo) surfaces and the Mo (100) surface have been measured as a function of the oxygen (O) coverage, from none up to several monolayers, for impact energies, E {dollar}<{dollar} 500 eV. Negative oxygen ions (O{dollar}\sp{lcub}-{rcub}){dollar} are found to be the dominant sputtered negative ions and for the three surfaces at all O coverages and Na{dollar}\sp{lcub}+{rcub}{dollar} impact energies. The O{dollar}\sp{lcub}-{rcub}{dollar} and secondary electron yields share a common impact energy threshold at {dollar}\rm E\sb{lcub}th{rcub}\approx 50{dollar} eV, and both have a strong dependence on the oxygen coverage of the surface.;The kinetic energy distributions of the secondary electrons and sputtered O{dollar}\sp{lcub}-{rcub}{dollar} were determined as functions of O coverages and impact energies. The O{dollar}\sp{lcub}-{rcub}{dollar} distributions are characterized by a narrow, low energy peak (at {dollar}\sim{dollar}1-2 eV) followed by a low level, high energy tail. The secondary electrons have a narrow (FWHM {dollar}\sim{dollar} 1-2 eV) kinetic energy distribution, centered essentially at the same most probable kinetic energy as the ions. The shapes of the respective distributions and the most probable kinetic energies are essentially invariant with the impact energy, O coverage and the metal surface.;The results were analyzed in terms of conventional collision cascade model, but the calculation could not be fitted to the experimental results. An electronic excitation mechanism is proposed to augment the collision cascade and to provide a mechanism for secondary electron emission. In the model, adsorbed O, which resides on the surface essentially as O{dollar}\sp{lcub}-{rcub},{dollar} is collisionally excited into an (MO{dollar}\sp{lcub}-{rcub})\sp*{dollar} repulsive state, and as the O{dollar}\sp{lcub}-{rcub}{dollar} exits the surface along the surface potential energy curve, it can decay by emission of an electron to the metal or to the vacuum, or it can survive as an ion. The parameters of this model can be adjusted such that the calculated kinetic energy distribution, together with that of the collision cascade, can reasonably reproduce the experimental observations for the ions and provide a reasonable fit to the corresponding electron kinetic energy distributions as well.

We present the technique and results of our study of the redistribution of Rydberg states in barium by a half cycle pulse (HCP). A survey of previous research concerning the interaction of half cycle pulses with Rydberg atoms reveals a lack of experimental data in the area of state redistribution. This is primarily due to an inability to identify the redistributed states accurately using the currently available tool: selective field ionization (SFI). We present the limitations of SFI as a state identification tool. We then present a survey of isolated core excitation as a state identification tool, along with its limitations in dealing with a mixture of states. Combining SFI and ICE overcomes these limitations and allows for the accurate characterization of a mixture of Rydberg states, as found in HCP redistribution. We present our results for the HCP redistribution of a series of ns and nd initial states, for n from 30 to 40, at various HCP intensities. These results primarily involved single photon transitions, confining us to the linear regime. Analyzing the results of these redistributions with respect to the energy difference between initial and final states yielded information about the frequency structure, after transmission to the interaction region, of the HCP. This led us to postulate on the effects of diffraction on the HCP. We conclude with several suggestions for the improvement of the experimental system, as well as a direction for future research.

No description available.

No description available.

Measurements of total cross sections for collision induced dissociation (CID), proton abstraction, and charge transfer have been made for the reactants H+3+H 2, He, and Ar CH+4+CD 4,H2 , and Ar, and CF+3 F++CF4,H2 and Ar for laboratory collision energies ranging from a few to 400 eV. Isotopic substitutions of the target and projectile have been made where possible to investigate any isotope effects, and in some cases to more clearly identify product ions. The purpose of this investigation is to expand the limited database of collisional processes pertinent to hydrogen, methane, and Freon discharges and their numerical modeling.;Cross sections for CID are observed to be relatively constant and CID is an important process over the energy range studied for the methane and Freon experiments. Cross sections for proton abstraction are, for the most part small ( ≤ 10 A2), and this process is important only at the lowest collision energies. The newly formed ion produced from proton abstraction for all reactants studied often has sufficient internal energy such that it may autodissociate. Charge transfer is observed for higher impact energies and cross sections for this reaction do not exceed 15 A 2. In general, production of secondary ions is observed at or near the energetic thresholds required for ground state reactants. The role of internal energy contained in the primary ion beam and its effects on the measured cross sections presented here will be addressed.;In addition to total cross sections, kinetic energy distributions have been measured for H+, H2+, and H 3+ ions present in a low pressure hydrogen discharge. These measured ion energy distributions are compared to predicted values made by a recently developed Monte Carlo simulation which necessarily incorporates the cross section measurements presented here for the H3 + + H2 system, among others. Complete agreement between theory and experiment is achieved only if measured cross sections for select CID reaction channels are arbitrarily increased by a factor of 2--3. Possible justifications for this modification, in addition to other inadequacies and further improvements that should be made to this numerical model are elaborated on. The importance and implications of all the measurements presented in this work to hydrogen, methane, and Freon discharge modeling and the yet to be realized goal of complete characterization of a molecular discharge will also be discussed.

The objectives of this dissertation are to advance and broaden the traditional average eye modeling technique by two extensions: 1) population-based and personalized eye modeling for both normal and diseased conditions, and 2) demonstration of applications of this pioneering eye modeling.The first type of representative eye modeling can be established using traditional eye modeling techniques with statistical biometric information of the targeted population. Ocular biometry parameters can be mathematically assigned according to the distribution functions and correlations between parameters. For example, the axial dimension of the eye relates to age, gender, and body height factors. With the investigation results from the studies of different population groups, population-based eye modeling can be established. The second type of eye model includes the optical components of the detailed corneal structure. Many of these structures, especially the corneal topography and wavefront aberration, are measured directly from the human eye. Therefore, the personalized eye models render the exact clinical measure and optical performance of the eye. In a sense, the whole eye, other than the identity of the individual, is quantified and stored in digital form for unlimited use for future research and industrial applications. The presentation of this dissertation is: Chapter 1 describes the background of the research in this area, the introduction of eye anatomy, and the motivation of this dissertation work. In Chapter 2, a comprehensive review of the contemporary techniques of measuring ocular parameters is presented and is followed by the review of literature and then the statistical analysis of the ocular biometry parameters. The goal of this chapter is to build a statistical base for population-based schematic eye modeling research. The analysis includes the investigation of the correlations between ocular parameters and ocular refraction, subject age, gender, ethnicity, and accommodation conditions. In Chapter 3, the tools and methods that are used in our optical eye modeling are introduced. The operation of the optical program ZEMAX is discussed. The detail of the optical eye modeling procedure and method of optical optimization, which is utilized to reproduce desired clinical measurement results, are described. The validation functions, which will be used to evaluate the optimization results, are also addressed. Chapter 4 includes the discussion of the population-based eye modeling and the personalized eye modeling. With the statistical information and the clinical measurements presented in Chapter 2 and the computation method described in Chapter 3, the two types of eye modeling technologies are demonstrated. The procedure, difficulty, and validation of eye modeling are included. The considerations of optical opacities, irregular optical surface, multiple reflection, scattering, and tear film breakup effects are discussed and the possible solutions in ZEMAX are suggested. Chapter 5 presents eye modeling applications of the simulations of ophthalmic instrument measurements. The demonstrated simulation results are retinoscopy and photorefraction. The simulation includes both normal eye model and diseased eye model. The close conformity between the simulation results with the actual clinical measurements further validates the eye modeling technique. The ophthalmic simulation application provides the potential for medical training and instrument development. The summary of the dissertation is given in Chapter 6.