The Metallurgy and Processing of Plutonium and Its Alloys Are Topics of Ongoing Research

Figure 1. "Telluride" simulation: Filling of a chalice with molten copper after 0.5 second.

Figure 2. Solid/liquid zones during copper chalice solidification after 425 seconds.

Figures provided by K. Lam, ESA Division

Plutonium's dual roles as a nuclear fuel and as a weapons material require a thorough understanding of its properties and the relationship of its properties to the processing methods used to manufacture components. Further, the challenges presented by the needs of stockpile surveillance require the ability to design new processing methods that produce completely functional systems but without the means for verification by the testing that was available in past years. Still further, the task of stockpile stewardship requires the ability to accurately assess the physical state, the properties, and the functionality of older (i.e., "aged") components.

These needs have motivated new lines of research in areas related to process modeling, in particular casting; the physical metallurgy of aged alloys, including oxidation; and the mechanical properties of new and aged alloys. The present article will focus on new efforts to design casting process methods and very briefly mention a few issues related to the mechanical behavior of plutonium.

Plutonium is, as a metal of the actinide group, active in nearly every way. The pure metal has six allotropes; properties such as density and the coefficient of thermal expansion vary significantly from phase to phase. The low symmetry of the alpha, beta, and gamma phases is, in fact, associated with several coefficients of thermal expansion, most of which are positive as they are for the more common metals. In contrast, the thermal expansion of the delta phase is negative, which can present problems during casting as noted below. The allotropic phase transformations also lead to significant changes in density, which again can lead to problems during processing. Mechanical properties are vitally important for component performance, and these too depend sensitively on phase as well as on alloy content and microstructure. Thus, it is critical to control phase and alloy content, and to insure phase and microstructural stability in aged materials. Some issues related to the process simulation of casting and mechanical properties will serve here as examples.

Of the six allotropic phases, alpha is the hardest and least ductile, and delta is the softest and most ductile. In terms of the material mechanical properties, system performance depends on achieving optimized strength and formability, which is related to ductility; for this reason the delta phase is preferred. The design of alloys and process controls are therefore required to stabilize the delta phase and to control its microstructural characteristics such as grain size and shape. Typical delta stabilizing solutes are Al, Ce, or Ga. The solidification processes that occur during casting and cooling with these solutes, for example plutonium with aluminum, would involve the general progression of states as follows: from liquid (L) to liquid/epsilon (L/e), to epsilon (e), to epsilon/delta (e/d), to delta (d).

Problems that arise through these steps include segregation of the solute (also known as "coring"), grain size control, and a range of mechanical phenomena associated with the density changes that occur during the solid-state phase changes. For example, volume expansions during the phase change from epsilon to delta will, in particular, lead to internal stresses that can cause distortion of the component and even degradation of the tooling. The prevention of solute segregation requires, at the very least, homogenization treatments, which can be expensive and difficult to perform adequately. Depletion of alloying elements in regions of the microstructure can, of course, lead to nonuniform phase structure. Accurate design of the tooling, the starting alloy chemistry, and the imposed thermal history is therefore required. This, in turn, requires detailed modeling of the hydrodynamics of fluid flow in the tools, analysis of the evolving thermal and solute fields in the tools and the part, and analysis of the mechanical stresses throughout the solidifying part and on the tools.

At present the process modeling efforts involve the development and implementation of a new casting simulation tool called "Telluride" and a thermomechanical simulation tool for analyzing the stresses and deformations that occur during casting. Telluride employs robust, high-resolution volume algorithms for incompressible fluid flow, volume tracking of interfaces, and solidification physics. The governing (i.e., field) equations implemented in Telluride include mass, momentum, energy, and species conservation. These are supplemented by data derived from phase diagrams and from models for permeability and back-diffusion. In particular, simulations using Telluride will analyze the motion of freezing interfaces and the solid-state transforming interfaces in addition to analyzing solute diffusion. These simulations will provide complete thermal, phase, and compositional fields as functions of time and, of course, as functions of the imposed boundary conditions. The boundary conditions will include tool characteristics and imposed external temperature conditions; these are, in fact, process design variables. Figure 1 and Figure 2 illustrate an example of a Telluride simulation of the solidification of metal shape; this particular example is that of a copper "chalice."

The thermomechanical analysis tool involves the development of an accurate constitutive theory for the solid-state deformations, which themselves involve concurrent phase transformations. Thus, within this constitutive theory there must be algorithms for predicting the kinetics of phase transitions as functions of time and temperature and for quantitatively describing the deformations that result. For the immediate future this theory will be implemented within the computer code "Chad," which currently has the capability of performing stress analysis.

Both Chad and Telluride are high-performance parallel codes. The expected outcome of the development of these simulation tools is the ability to accurately predict the microstructure and stress states that result from the casting process; this predictive capability will, in turn, allow for more optimized designs of the casting process itself.

Still other efforts involve the development of diagnostic tools to characterize the mechanical properties of new and aged plutonium alloys as well as to provide a comprehensive documentation of both quasistatic and truly dynamic deformation behavior. Small-scale measurement techniques involving microindenters will be evaluated as to their suitability for characterizing strength on small samples extracted from new or aged systems. These will be supplemented by a host of more precise tests aimed at documenting dynamic mechanical response and correlating mechanical behavior with chemistry and microstructure. Mechanical behavior will be studied as functions of temperature and the rate of deformation.

Robert Asaro, who contributed this article, is Professor of Applied Mechanics and Materials Science at the University of California, San Diego.

NMT | LANL | DOE
Phone Book | Search | Help/Info

L O S  A L A M O S  N A T I O N A L   L A B O R A T O R Y
Operated by the University of California for the US Department of Energy