This report summarizes the work that has been performed at Argonne National Laboratory on the development of an analytical procedure to analyze TRUEX process solvents; these solvents are composed of a bifunctional organophosphorus extractant.

A fully coupled solution algorithm for pressure-linked fluid flow equations earlier found to be rapidly convergent in laminar flows has been extended to calculate turbulent flows. The governing mean flow equations are solved in conjunction with a two-equation (k - epsilon) turbulence model. A number of two-dimensional recirculating flows have been computed and it is shown that the calculation procedure is rapidly convergent in all the cases. The calculations have been compared with published experimental data; their agreement is in accord with other published experiences with the (k - epsilon) model in similar flows.

A robust and efficient numerical scheme has been developed for the solution of the finite-differenced pressure linked fluid flow equations. The algorithm solves the set of nonlinear simultaneous equations by a combination of Newton's method and efficient sparse matrix techniques. In tests on typical recirculating flows the method is rapidly convergent. The method does not require any under-relaxation or other convergence-enhancing techniques employed in iterative schemes. It is currently described for two-dimensional steady state flows but is extendible to three dimensions and mildly time-varying flows. The method is robust to changes in Reynolds number, grid aspect ratio, and mesh size. This paper reports the algorithm and the results of calculations performed.

Techniques for isolating and solidifying the nuclear wastes in the storage tanks at the Western New York Nuclear Service Center plant have been examined. One technique involves evaporating the water and forming a molten salt containing the precipitated sludge. The salt is allowed to solidify and is stored in canisters until processing into a final waste form is to be done. Other techniques involve calcining the waste material, then agglomerating the calcine with sodium silicate to reduce its dispersibility. This option can also involve a prior separation and decontamination of the supernatant salt. The sludge and all resins containing fission-product activity are then calcined together. The technique of removing the water and solidifying the salt may be the simplest method for removing the waste from the West Valley Plant.

Type 308 CRE stainless steel weld specimens were subjected to metallographic and fractographic analysis after failure in elevated temperature (593 degrees C) creep-fatigue tests. The failure mode for specimens tested under continuous-cycle fatigue conditions were predominantly transgranular. When the test cycle was modified to include a hold time at the maximum tensile strain, the failure mode became predominantly inter-phase. Sigma phase was observed within the delta-ferrite regions of the weld. However, the presence of sigma phase did not appear to affect the failure mode.

The quantities of krypton-85 that can be released to the environment from nuclear energy production are to be limited after 1983 by Federal regulations. Although procedures for collecting the krypton-85 released in the nuclear fuel cycle have been developed to the point that they are commercially available, procedures for terminal disposal of the collected gas are still being examined for their feasibility. In this work, the possibilities of underground disposal of krypton-85 by several techniques were evaluated. It was concluded that (1) disposal of krypton-85 as a solution in water or other solvents in deep wells would have the major disadvantages of liquid migration and the requirement of extremely large volumes of solvent; (2) disposal as bubbles entrained in cement grout injected underground presents the uncertainty of gaseous migration through permeable solid grout; (3) disposal by injection into abandoned oil fields would be favored by solubility of krypton in residual hydrocarbons, but has the disadvantages that such fields contain numerous shafts offering avenues of escape and also that the fields may be reworked in the future for their hydrocarbon residues; (4) underground retention of krypton-85 injected as a gas may be promising, given the right lithology, through entrapment in interstices …

On August 6--10, 1990, Argonne National Laboratory hosted a Theory Institute in Automated Reasoning. The institute was organized by the Mathematics and Computer Science Division and was supported by special funding from Argonne's Physical Research Program Administration. The focus of the Institute was on the obstacles confronting the effective automation of reasoning. The objective was to lay the groundwork for formulating a theory governing the interrelationship of representation, inference rule, and strategy. Here we summarize the activities that took place during the week-long Institute. We also present an evaluation of the progress achieved-progress that includes the solution of challenge questions, the increasing use of both our database of problems and our automated reasoning program OTTER, and the discovery of new used for OTTER.

In both mathematics and logic, many theorems exist such that each can be proved in entirely different ways. For a striking example, there exist theorems from group theory that can be proved by relying solely on equality and (from the viewpoint of automated reasoning) the use of paramodulation, but can also be proved in a notation in which equality is totally absent and the inference rule is condensed detachment (captured with a single clause and the rule hyper-resolution). A study of such examples immediately shows how far from obvious is the problem of producing a proof in one system even in the presence of a proof in another; such problems can be viewed as ones of translation, where the rules of translation and the translation itself are frequently difficult to obtain. In this report, we discuss in detail various techniques that can be applied by the automated reasoning program OTTER to address the translation problem to obtain a proof in one notation and inference system given a proof in a completely different notation and inference system. To illustrate the techniques, we present a full treatment culminating in a successful translation'' of a proof of a theorem from many-valued sentential calculus. …

Weight losses of marble and limestone samples exposed to outdoor environments at field sites in the eastern United States have been monitored in studies initiated in 1984. The procedures are described, and the results are tabulated and discussed. A rate of marble loss approximately equivalent to 16 micrometers of surface recession per year was found in North Carolina, and losses of this order were also observed in New Jersey, New York, and Washington, DC. Limestone weight losses were much higher than for marble in the first year; loss of extraneous materials from the porous limestone appeared to be a likely contributor to the overall loss. The rate of limestone loss diminished in the second year, though it continued to be higher than for marble. Exposures are continuing in a planned 10-yr program of tests.