The National Geothermal Information Resource (GRID) of the Lawrence Berkeley Laboratory is chartered by the U.S. Department of Energy (DOE) to provide critically evaluated data and other information for the development and utilization of geothermal energy. Included are both site dependent and site independent information related to resource evaluation, electrical and direct utilization, environmental aspects, and the basic properties of aqueous electrolytes. The GRID project is involved in cooperative agreements for the interchange of information and data with other organizations. There are currently three U.S. data centers working to implement the collection and exchange of information on geothermal energy research and production: the DOE Technical Information Center (TIC), Oak Ridge, the GEOTHERM database of the U.S. Geological Survey in Menlo Park, and the GRID project. The data systems of TIC, GEOTHERM and GRID are coordinated for data collection and dissemination, with GRID serving as a clearinghouse having access to files from all geothermal databases including both numerical and bibliographic data. GRID interfaces with DOE/TIC for bibliographic information and with GEOTHERM for certain site-dependent numerical data. The program is organized into four principal areas: (1) basic geothermal energy data; (2) site-dependent data for both electrical and direct utilization; (3) environmental aspects, …

The orbit-averaged kinetic equation suitable for studying collisional transport in an axisymmetric mirror machine is derived and reduced to a three-dimensional equation in phase space in the case of ..nu../sub c/ << ..omega../sub b/ << ..cap omega.. where ..nu../sub c/ is the collision frequency, ..omega../sub b/ is the bounce frequency, and ..cap omega.. is the gyrofrequency. We found that the gyrophase dependence of the distribution function, ignored in previous Fokker-Planck calculations, is responsible for the diffusion of guiding centers and should be retained when the Larmor radius is comparable to the plasma scale length. The complete Fokker-Planck equation for a square-well magnetic mirror is given as a special case to clarify the approximations and assumptions involved in the radial Fokker-Planck code of Futch. The quasilinear transport code of Matsuda and Berk is also discussed relative to the orbit-averaged Fokker-Planck equation.

Incomplete LU and incomplete-Cholesky conjugate gradient methods are becoming widely used in both laser and magnetic fusion research. In my original presentation of these methods, the problem of what to do if a pivot (L/sub ii/U/sub ii/) becomes very small or zero was raised and only partially answered by the suggestion that it be arbitrarily set to some non-zero value. In what follows it will be shown precisely how small the pivot can become before it must be fixed and precisely what value it should be set to in order to minimize the error in LU. Numerical examples will be given to show that not only does this prescription improve incomplete LU-conjugate gradient methods , but exact LU decomposition carried out with this prescription for handling small pivots and followed by a few linear or conjugate gradient iterations can be much faster than the permutations of rows and columns usually employed to circumvent small pivot problems.

The space and domestic water solar heating system designated SIMS Prototype System 1 has been evaluated at the MSFC Solar Heating and Cooling Test Facility. The test system used 720 ft2 (gross) of Solar Energy Products Air Collectors, a Solar Control Corporation SAM 20 Air Handler with Model 75-175 control unit, a Jackson Solar Storage tank with Rho Sigma Mod 106 controller, and 20 tons of rock storage. The test data analysis performed by IBM evaluates the system performance and documents the suitability of SIMS Prototype System 1 hardware for field installation.

Single-charge and row-charge subsurface cratering experiments were performed to learn how close-spacing enhances single-crater dimensions. Our first experimental phase established cratering curves for 60-lb charges of the chemical explosive. For the second phase, to be described in a subsequent report, the Row-cratering experiments were designed and executed. This data report contains excavated dimensions and auxiliary data for the single-charge cratering experiments. The dimensions for the row-charge experiments will be in the other report. Significant changes in the soil's water content appeared to cause a variability in the excavated dimensions. This variability clouded the interpretation and application of the cratering curves obtained.

Winding of a Nb--Ti test coil at the Lawrence Livermore Laboratory is nearly complete. The conductor in this coil operates in a maximum field of 7.5 T and provides the 2-T field required by the Mirror Fusion Test Facility. Nb/sub 3/Sn multifilamentary conductors, made using the ''bronze'' technique, appear capable of providing the higher fields needed by commercial reactors.

The production of UF/sub 6//sup -/ by reaction of a collimated stream of fluorine gas with a resistively heated uranium wire was studied at temperatures from 870 to 1020/sup 0/C and pressures less than 10/sup -3/ torr. At these temperatures below the uranium melting point, the formation of UF/sub 3/ intermediate on the uranium surface resulted in low UF/sub 6//sup -/ yields. The kinetic energy of the UF/sub 6//sup -/ ion was on the order of thermal energies. The work function of uranium was measured to be 4.20 +- 0.14 eV.