Engineering topics
Publications and source records attributed to Huo, Winifred M..
Based on the Binary-Encounter-Bethe (BEB) model, the advantage of using relativistic effective core potentials (RECP) in the calculation of total ionization cross sections of heavy atoms or molecules containing heavy atoms is discussed. Numerical examples for Ar, Kr, Xe, and WF6 are presented.
A new N2-N2 rigid-rotor surface has been determined using extensive Ab Initio quantum chemistry calculations together with recent experimental data for the second virial coefficient. Rotational energy transfer is studied using the new potential energy surface (PES) employing the close coupling method below 200 cm(exp -1) and coupled state approximation above that. Comparing with a previous calculation based on the PES of van der Avoird et al.,3 it is found that the new PES generally gives larger cross sections for large (delta)J transitions, but for small (delta)J transitions the cross sections are either comparable or smaller. Correlation between the differences in the cross sections and the two PES will be attempted. The computed cross sections will also be compared with available experimental data.
Rotational energy transfer between two N2 molecules is a fundamental process of some importance. Exchange is expected to play a role, but its importance is somewhat uncertain. Rotational energy transfer cross sections of N2 also have applications in many other fields including modeling of aerodynamic flows, laser operations, and linewidth analysis in nonintrusive laser diagnostics. A number of N2-N2 rigid rotor potential energy surface (PES) has been reported in the literature.
We have applied the Z-matrix method, using a mixed basis of finite elements and Gaussians, to study e-H2 elastic and inelastic collisions. Special attention is paid to the quality of the basis set and the treatment of electron correlation. The calculated cross sections are invariant, to machine accuracy, with respect to the choice of parameters a, b, d, e as long as they satisfy Equation (3). However, the log derivative approach, i.e., the choice a = -e = 1, b = d = 0 appears to converge slightly faster than other choices. The cross sections agree well with previous theoretical results. Comparison will be made with available experimental data.
The present study adapts the Z-matrix formulation using a mixed basis of finite elements and Gaussians. This is a energy-independent basis which allows flexible boundary conditions and is amenable to efficient algorithms for evaluating the necessary matrix elements with molecular targets.
Computational study of electron-molecule collisions not only complements experimental measurements, but can also be used to investigate processes not readily accessible experimentally. A number of ab initio computational methods are available for these types of calculations. Here we describe a recently developed technique, the finite element Z-matrix method, Analogous to the R-matrix, method, it partitions the space into regions and employs real matrix elements. However, unlike the implementation of the R-matrix method commonly used in atomic and molecular physics, the Z-matrix method is fully variational. In the present implementation, a mixed basis of finite elements and Gaussians is used to represent the continuum electron, thus offering full flexibility without imposing fixed boundary conditions. Numerical examples include the electron-impact dissociation of N2 via the metastable A3Su+ state, a process which may be important in the lower thermosphere, and the dissociation of the CF radical, a process of interest to plasma etching. To understand the dissociation pathways, large scale quantum chemical calculations have been carried out for all target states which dissociate to the lowest five limits in the case of N2, and to the lowest two limits in the case of CF. For N2, the structural calculations clearly show the preference for predissociation if the initial state is the ground X1 Sg+ state, but direct dissociation appears to be preferable if the initial state is the A3Su+ state. Multi-configuration SCF target functions are used in the collisional calculation.
Electron-H2 Collisions have been studied using the finite-element R-matrix method. Our approach uses a mixed finite-element and Gaussian basis to describe the continuum electron. In the inner region, the multi-centered nature of the Gaussian basis provides an efficient representation in the regions of space near the nuclei, whereas the piecemeal and energy-independent nature of the FEM basis is particular well suited for R-matrix calculations. Fixed nuclei calculations have been carried out at 1.4 a(sub 0), the equilibrium internuclear distance of the ground state. Up to six target states are included: the X(sup 1)E+(sub g), b(sup 3)E+(sub u), a(sup 3)E+(sub g), B(sup 1)E+(sub u), c(sup 3)II(sub u), and C(sup 1)II(sub u) states, and configuration-interaction functions are used to describe the target states. The results will be compared with available theoretical and experimental data.
Low-energy electron impact excitation of the fundamental modes of CO2 has been extensively studied, both experimentally and theoretically. Much attention has been paid to the virtual state feature in the the (100) mode excitation and the (sup 2)II(sub upsilon) resonance feature around 3.8 eV, which is observable in all three fundamental modes. For the excitation of the (010) mode away from the resonance region, the Born dipole approximation was generally considered adequate. The present study employs the Born dipole approximation to treat the long range interaction and the Schwinger multichannel method for the short range interaction. The roles of the two interaction potentials will be compared.
Variational methods have proven invaluable in theoretical physics and chemistry, both for bound state problems and for the study of collision phenomena. The application of the Schwinger variational (SV) method to e-molecule collisions and molecular photoionization has been reviewed previously. The present chapter discusses the implementation of the SV method as applied to e-molecule collisions. Since this is not a review of cross section data, cross sections are presented only to server as illustrative examples. In the SV method, the correct boundary condition is automatically incorporated through the use of Green's function. Thus SV calculations can employ basis functions with arbitrary boundary conditions. The iterative Schwinger method has been used extensively to study molecular photoionization. For e-molecule collisions, it is used at the static exchange level to study elastic scattering and coupled with the distorted wave approximation to study electronically inelastic scattering.
Variational methods have proven invaluable in theoretical physics and chemistry, both for bound state problems and for the study of collision phenomena. For collisional problems they can be grouped into two types: those based on the Schroedinger equation and those based on the Lippmann-Schwinger equation. The application of the Schwinger variational (SV) method to e-molecule collisions and photoionization has been reviewed previously. The present chapter discusses the implementation of the SV method as applied to e-molecule collisions.
The electronic transition moment functions between the B(sup 3)Pi(sub g) and the A(sup 3)Sigma(sup +, sub u), W(sup 3)Delta(sub u), B'(sup 3)Sigma(sup -, sub u) states of N2 are studied using the internally contracted multireference configuration interaction (ICMRCI) method based upon complete active space SCF (CASSCF) reference wave-functions. The dependence of the moments on both the one and n-particle basis sets has been investigated in detail. The calculated radiative lifetimes for the vibrational levels of B(sup 3)Pi(sub g) are in excellent agreement with the most recent measurement of Euler and Pipkin (1983)
At high altitudes and velocities equal to or greater than the geosynchronous return velocity (10 kilometers per second), the shock layer of a hypersonic flight will be in thermochemical nonequilibrium and partially ionized. The amount of ionization is determined by the velocity. For a trans atmospheric flight of 10 kilometers per second and at an altitude of 80 kilometers, a maximum of 1% ionization is expected. At a velocity of 12 - 17 kilometer per second, such as a Mars return mission, up to 30% of the atoms and molecules in the flow field will be ionized. Under those circumstances, electrons play an important role in determining the internal states of atoms and molecules in the flow field and hence the amount of radiative heat load and the distance it takes for the flow field to re-establish equilibrium. Electron collisions provide an effective means of transferring energy even when the electron number density is as low as 1%. Because the mass of an electron is 12,760 times smaller than the reduced mass of N2, its average speed, and hence its average collision frequency, is more than 100 times larger. Even in the slightly ionized regime with only 1% electrons, the frequency of electron-molecule collisions is equal to or larger than that of molecule-molecule collisions, an important consideration in the low density part of the atmosphere. Three electron-molecule collision processes relevant to hypersonic flows will be considered: (1) vibrational excitation/de-excitation of a diatomic molecule by electron impact, (2) electronic excitation/de-excitation, and (3) dissociative recombination in electron-diatomic ion collisions. A review of available data, both theory and experiment, will be given. Particular attention will be paid to tailoring the molecular physics to the condition of hypersonic flows. For example, the high rotational temperatures in a hypersonic flow field means that most experimental data carried out under room temperatures are not applicable. Also, the average electron temperature is expected to be between 10,000 and 20,000 K. Thus only data for low energy electrons are relevant to the model.
Rotationally inelastic transitions of N2 have been studied in the coupled state (CS) and infinite-order-sudden (IOS) approximations, using the N2-N2 rigidrotor potential of van der Avoird et al. For benchmarking purposes, close coupling (CC) calculations have also been carried out over a limited energy range. The CC and CS cross sections have been obtained both with and without identical molecule exchange symmetry, whereas exchange was neglected in the IOS calculations. The CS results track the CC cross sections rather well; between 113 - 219 cm(exp -1) the average deviation is 14%. Comparison between the CS and IOS cross sections at the high energy end of the CS calculation, 500 - 680 cm(exp -1), shows that IOS is sensitive to the amount of inelasticity and the results for large DELTA J transitions are subject to larger errors. It is found that the state-to-state cross sections with even and odd exchange symmetry agree to better than 2% and are well represented as a sum of direct and exchange cross sections for distinguishable molecules, an indication of the applicability of a classical treatment for this system. This result, however, does not apply to partial cross sections for given total J, but arises from a near cancellation in summing over partial waves. In order to use rigid-rotor results for the calculation of effective rotational excitation rates of N2 in the v=1 vibrational level colliding with bath N2 molecules in the v=0 level, it is assumed that exchange scattering between molecules in different vibrational levels is negligible and direct scattering is independent of Y. Good agreement with room temperature experimental data is obtained. The effective rates determined using the IOS and energy corrected sudden (ECS) approximations are also in reasonable agreement with experiment, with the ECS results being somewhat better. The problem with a degeneracy factor in earlier cross section expressions for collisions between identical molecules is pointed out and corrected.
Rotational inelastic transitions of N2 have been studied in the coupled state (CS, also called centrifugal sudden) and infinite-order-sudden (IOS) approximations, using the N2-N2 rigid-rotor potential of van der Avoird et al. For benchmarking purposes, close coupling (CC) calculations have also been carried out over a limited energy range and for even j - even j collisions only. Both the CC and CS cross sections have been obtained with and without exchange symmetry, whereas exchange is neglected in the IOS calculations. The CS results track the CC cross sections rather well. At total energies between 113 to 219 cm(exp -1) the average deviation is 14%. The deviation decrease with increasing energy, indicating that the CS approximation can be used as a substitute at higher energies when the CC calculations become impractical. Comparison between the CS and IOS cross sections at the high energy end of the CS calculation, 500 - 680 cm(exp-1), shows significant differences between the two. In addition, the IOS results exhibits sensitivity to the amount of inelasticity and the results for large DELTA J transitions are subjected to bigger errors. At total energy 113 cm(exp -1) and above, the average deviation between state-to-state cross sections calculated with even and odd exchange symmetries is 1.5%.
For the calculation of electron molecule collision cross sections R-matrix methods automatically take advantage of the division of configuration space into an inner region (I) bounded by radius tau b, where the scattered electron is within the molecular charge cloud and the system is described by an correlated Configuration Interaction (CI) treatment in close analogy to bound state calculations, and an outer region (II) where the scattered electron moves in the long-range multipole potential of the target and efficient analytic methods can be used for solving the asymptotic Schroedinger equation plus boundary conditions.
Current interest in clusters stems from their role as novel materials as well as a possible extension of cluster results to bulk systems. Experimental investigations on clusters have been carried out using laser spectroscopy, microwave spectroscopy, heavy-particle collisions, as well as electron collisions with earlier experimental work on electron attachment and ionization having been reviewed previously. Recently, Mark and coworkers studied the decay channels of cluster ions following electron impact ionization. Rauth et al. reported the formation of the superhalogen ion SF7(-) and other nonstoichiometric cluster ions in their study of electron attachment to SF6 clusters. Kresin et al. measured the absolute electron-impact depletion cross section of metal clusters Na8, Na(20), and Na(40). They found that the inelastic scattering cross section increased with cluster size and was considerably greater than the hard sphere collision cross sections. They hypothesized that electron attachment and collision-induced fragmentation were the dominant physical processes responsible for this effect. For the two smaller clusters, they also found a sharp increase in the cross section near threshold. Most theoretical studies of clusters have been devoted to their electronic structures, vibrational relaxation, and predissociation while investigations of electron scattering from clusters has been lacking. In view of this, we recently undertook an ab initio study of electron scattering from small Be clusters and BeCO. Beryllium was chosen because it is readily amenable to ab t'nitio calculations. Moreover, the electronic structure of Be clusters has been studied extensively, showing that the Be-Be bond is relatively weak in comparison with a normal chemical bond. Our investigation focuses on how the cross sections change with cluster size and geometry. The range of energy studied, 0.05 - 5.0 eV, is chosen because of the ubiquitous resonance in the low-energy scattering of Be. Hence it can be determined if it is possible, as a consequence of the weaker bonds in the Be clusters, to identify the atomic origin of the cluster resonance.
Using the N2-N2 intermolecular potential of van der Avoird et al. rotational energy transfer cross sections have been calculated using both the coupled state (CS) and infinite order sudden (IOS) approximations. The rotational energy transfer rate constants at 300 K, calculated in the CS approximation, are in reasonable agreement with the measurements of Sitz and Farrow. The IOS approximation qualitatively reproduces the dependence of the rate constants on the rotational quantum numbers, but consistently overestimates their magnitudes. The treatment of exchange symmetry will be discussed.
Explore the source record for details and available documents.