Plasmas generated by irradiating targets with {approx}20 kJ of laser energy are routinely created in inertial confinement fusion research. X-ray spectroscopy provides one of the few methods for diagnosing the electron temperature and electron density. For example, electron densities approaching 10{sup 24} cm{sup -3} have been diagnosed by spectral linewidths. However, the accuracy of the spectroscopic diagnostics depends on the population kinetics, the radiative transfer, and the line shape calculations. Analysis for the complex line transitions has recently been improved and accelerated by the use of a database where detailed calculations can be accessed rapidly and interactively. Examples of data from Xe and Ar doped targets demonstrate the current analytic methods. First we will illustrate complications that arise from the presence of a multitude of underlying spectral lines. Then, we will consider the Ar He-like 1s{sup 2}({sup 1}S{sub 0}) - 1s3p({sup 1}P{sub 0}) transition where ion dynamic effects may affect the profile. Here, the plasma conditions are such that the static ion microfield approximation is no longer valid; therefore in addition to the width, the details of the line shape can be used to provide additional information. We will compare the data to simulations and discuss the possible pitfalls involved …

Results of using GLO (Global Local Optimizer), a general purpose nonlinear optimization software package for investigating multi-parameter problems in science and engineering is discussed. The package consists of the modular optimization control system (GLO), a graphical user interface (GLO-GUI), a pre-processor (GLO-PUT), a post-processor (GLO-GET), and nonlinear optimization software modules, GLOBAL & LOCAL. GLO is designed for controlling and easy coupling to any scientific software application. GLO runs the optimization module and scientific software application in an iterative loop. At each iteration, the optimization module defines new values for the set of parameters being optimized. GLO-PUT inserts the new parameter values into the input file of the scientific application. GLO runs the application with the new parameter values. GLO-GET determines the value of the objective function by extracting the results of the analysis and comparing to the desired result. GLO continues to run the scientific application over and over until it finds the ``best`` set of parameters by minimizing (or maximizing) the objective function. An example problem showing the optimization of material model is presented (Taylor cylinder impact test).

This report consists of vugraphs from presentations at the meeting. The papers covered the following topics: (1) APS as a proton source; (2) target status for NSNS (National Spallation Neutron Source); (3) spallation neutron source in Japan; (4) liquid LiBi flow loop; and (5) research and development plans for high power tests at the AGS.

The American National Standards Institute (ANSI) defines systematic error as An error which remains constant over replicative measurements. It would seem from the ANSI definition that a systematic error is not really an error at all; it is merely a failure to calibrate the measurement system properly because if error is constant why not simply correct for it? Yet systematic errors undoubtedly exist, and they differ in some fundamental way from the kind of errors we call random. Early papers by Eisenhart and by Youden discussed systematic versus random error with regard to measurements in the physical sciences, but not in a fundamental way, and the distinction remains clouded by controversy. The lack of a general agreement on definitions has led to a plethora of different and often confusing methods on how to quantify the total uncertainty of a measurement that incorporates both its systematic and random errors. Some assert that systematic error should be treated by non- statistical methods. We disagree with this approach, and we provide basic definitions based on entropy concepts, and a statistical methodology for combining errors and making statements of total measurement of uncertainty. We illustrate our methods with radiometric assay data.

For the bcc-U-Ti alloy system, H and D solubility measurements have been made on 12 alloy specimens ranging in composition from pure U to pure Ti and temperature range bounded by 900 K to 1,500 K. The results are described by a model within a standard error of 3%.