The first gas propagation experiment on ATA is planned to be conducted in a 1-foot diameter tank of up to 10 m length. The primary objectives are to measure beam parameters at injection to determine whether the desired beam conditioning is achieved, and to observe how such conditioned beams propagate in air and neon.

The philosophy of these tests is to measure the motion of a low current, small diameter electron beam in the accelerator before running high current. By using low current, we can study particle motion in the applied fields without any extra complications associated with the self-forces of high currents. With the steering magnets off, we have measured the transverse drift of the probe beam. Also, we have used the probe beam to optimize the current in the steering magnets to compensate for the drift. There have been concurrent efforts to locate the source of the error field which is presumed to cause the drift. So far, the source has not been established but the search is continuing.

Careful selection of the best diffraction geometry matched to the sample, with use of high resolution two-dimensional detectors and a mask integration procedure will allow the collection of statistically accurate data for many proteins and crystals with a sample size of at least 1 mm/sup 3/. Accurate data and improved refinement techniques will allow the determination of all atoms including hydrogens, hydrogen-deuterium exchange, water of hydration, and solvent water densities. The developments in the above described neutron techniques have been gradually used in the analysis of myoglobins, gramicidin, trypsin, and crambin. 16 references, 11 figures, 2 tables.

In anticipation of current and future needs for the Particle Beam Program and other programs at the Lawrence Livermore National Laboratory, we are continuing efforts in the development of high-repetition-rate magnetic pulse compressors that use ferromagnetic metallic glasses, both in the linear and very high saturation rates. These devices are ideally suited as drivers for linear induction accelerators, where duty factor or average repetition rates (hundred of hertz) requirements exceed the parameters that can be achieved by pulse compression using spark gaps. The technique of magnetic pulse compression has been with use for several decades, but relatively recent developments in rapidly quenched magnetic metals of very thin cross sections, has led to the development of state-of-the-art magnetic pulse compressors with very high peak power, repetition rates, and reliability. This paper will describe results of recent experiments and the relevant electrical and mechanical properties of magnetic pulse compressors to achieve high efficiency and reliability.

Experimental measurements are reported of the time development of compositional changes which occur deep in a Cu-40 at. % Ni alloy during sputtering at elevated temperatures with 5-keV Ar ions. Large effects are observed due to radiation-induced segregation (RIS), i.e. the preferential transport of nickel atoms in the direction of the defect fluxes. Results are compared with calculated composition profiles obtained using the phenomenological model of Lam and Wiedersich. Qualitative agreement between the model calculations and experimental results is found. Quantitative differences between the model and experiment preclude a reliable estimate of RIS effects in the very near-surface layers. However, the fact that the calculations underestimate the RIS contribution very deep in the specimen while still predicting a substantial RIS contribution in the near-surface region suggests that RIS can play an important role in determining near- surface compositional changes during sputtering at temperatures where defects are mobile.

The viscous flow of fluids and the plastic flow of solids, such as metals, are interesting from both the practical and the theoretical points of view. Atomistic molecular dynamics simulations provide a way of visualizing and understanding these flows in a detailed microscopic way. Simulations are necessarily carried out at relatively high rates of strain. For this reason they are ideally suited to the study of nonlinear flow phenomena: normal stresses induced by shear deformation, stress rotation, and the coupling of stress with heat flow, for instance. The simulations require appropriate boundary conditions, forces, and equations of motion. Newtonian mechanics is relatively inefficient for this simulation task. A modification, Nonequilibrium Molecular Dynamics, has been developed to simulate nonequilibrium flows. By now, many high-strain-rate rheological studies of flowing (viscous) fluids and (plastic) solids have been carried out. Here I describe the new methods used in the simulations and some results obtained in this way. A three-body shear-flow exercise is appended to make these ideas more concrete.