The application of neutron activation analysis to the determination of element release from simulated nuclear waste forms during leaching is described for several different glasses. Potential neutron irradiation effects are discussed, and it is shown, by a series of leach tests on activated and non-activated glass samples, that neutron irradiation has no discernible effect on the release of silicon and cesium during leaching. The radioisotopes best suited for analysis with this method and their associated detection limits are identified, and the method's applicability to waste forms other than glass is discussed.

This evaluation of radiation effects in silicate glasses caused by ionization is based on our own investigations, on material collected in our files (reports, articles, and notes), and on a computer literature search through recent issues of Physics Abstracts and Chemical Abstracts (and the apparently pertinent references which appeared). Some of the recent results, available heretofore only in internal correspondence, are presented in some detail. It is concluded that research into the behavior of silicate glasses generally will be required before the specific effects in the radioactive waste storage glasses can be properly understood and evaluated. Two particular neglected areas of investigation are targeted for immediate concern: a kinetic analysis of annealing data and the acquisition of data on effects of irradiation at controlled elevated temperatures.

This report covers the research, development, and management activities of the programs involving high-performance lithium-aluminum/iron sulfide batteries at Argonne National Laboratory (ANL) and at contractors' laboratories during the period October 1980 through September 1981. These batteries, which are being developed for electric-vehicle propulsion and stationary energy-storage applications, consist of vertically oriented prismatic cells with one or more inner positive electrodes of FeS or FeS2, facing negative electrodes of lithium-aluminum, and molten LiCl-KC1 electrolyte.

This report describes how solar system tanks fail to stratify, a new solar system control strategy that allows stratification, a one-dimensional analytical model of thermally stratified tanks, experimental measurement of thermal stratification in tanks, correlation of experimental measurements with empirical constants in the analytical model, and a procedure for designing thermally stratified tanks. Failure to stratify is explained in terms of the critical Richardson number. The key to the new control strategy is to avoid a Richardson number that decreases during solar collection. The analytical model is an approximate solution based on assumptions that (1) the solution is a function of elevation and time, only, (2) plug flow exists, (3) flow rate is constant, (4) the cross-sectional areas of the tank and tank wall are constant, (5) there is a step change of inlet temperature, (6) there is heat transfer between the tank wall and the water, and (7) thermal losses from the tank are negligible. Empirical constants in the theory are determined by adjusting them until the best least-squares fit with experimental data is obtained and correlating the constants with the Fourier and Richardson numbers. The new control strategy allows tanks to stratify and reduces the average collector operating …

Newton's method plays a central role in the development of numerical techniques for optimization. In fact, most of the current practical methods for optimization can be viewed as variations on Newton's method. It is therefore important to understand Newton's method as an algorithm in its own right and as a key introduction to the most recent ideas in this area. One of the aims of this expository paper is to present and analyze two main approaches to Newton's method for unconstrained minimization: the line search approach and the trust region approach. The other aim is to present some of the recent developments in the optimization field which are related to Newton's method. In particular, we explore several variations on Newton's method which are appropriate for large scale problems, and we also show how quasi-Newton methods can be derived quite naturally from Newton's method.