By exposing samples to various irradiance levels from a calibrated thermal radiation source, the ignition responses of blackened alpha-cellulose and cotton cloth with and without fire-retardant additives were compared. Samples treated with retardant compounds which showed the most promise were then isothermally pyrolyzed in air for comparisons between the pyrolysis rates. Alpha-cellulose samples containing a mixture of boric acid, borax, and ammonium di-hydrogen phosphate could not be ignited by irradiances up to 4.0 cal cm/sup -2/ s-1 (16.7 W/cm/sup 2/). At higher irradiances the specimens ignited, but flaming lasted only until the flammable gases were depleted. Cotton cloth containing a polymeric retardant with the designation THPC + MM was found to be ignition-resistant to all irradiances below 7.0 cal cm/sup -2/ s/sup -1/ (29.3 W/cm/sup 2/). Comparison of the pyrolysis rates of the retardant-treated alpha-cellulose and the retardant-treated cotton showed that the retardant mechanism is qualitatively the same. Similar ignition-response measurements were also made with specimens exposed to ionizing radiation. It was observed that gamma radiation results in ignition retardance of cellulose, while irradiation by neutrons does not.

Estimates were made of the shielding thicknesses required at various points around the ISABELLE ring. Both hadron and muon requirements are considered. Radiation levels at the outside of the shield and at the BNL site boundary are kept at or below 1000 mrem per year and 5 mrem/year respectively. Muon requirements are based on the Wang formula for pion spectra, and the hadron requirements on the hadron cascade program CYLKAZ of Ranft. A muon shield thickness of 77 meters of sand is indicated outside the ring in one area, and hadron shields equivalent to from 2.7 to 5.6 meters in thickness of sand above the ring. The suggested safety allowance would increase these values to 86 meters and 4.0 to 7.2 meters respectively. There are many uncertainties in such estimates, but these last figures are considered to be rather conservative.

About 40 underground nuclear explosions detonated at the Nevada Test Site (NTS) were chosen for analysis of their spectra and any relationships they might have to source parameters such as yield, depth of burial, etc. The sample covered a large yield range (less than 20 kt to greater than 1 Mt). Broadband (0.05 to 20 Hz) data recorded by the four-station seismic network operated by Lawrence Livermore Laboratory were analyzed in a search for unusual explosion signatures in their spectra. Long time windows (total wave train) as well as shorter windows (for instance, P/sub n/) were used as input to calculate the spectra. Much variation in the spectra of the long windows is typical although some gross features are similar, such as a dominant peak in the microseismic window. The variation is such that selection of corner frequencies is impractical and yield scaling could not be determined. Spectra for one NTS earthquake showed more energy in the short periods (less than 1 sec) as well as in the long periods (greater than 8 sec) compared to those for NTS explosions.

In the study of ductile fracture it is useful to simulate fracture on the computer under plane stress conditions. In general, this is a three dimensional problem. Presented here is a method for adapting a two dimensional elastic-plastic computer program to calculate problems in plane stress as well as plane strain geometry. A simulation of a tension test of a flat aluminum plate pulled to failure is calculated with the modified two dimensional program. The results are compared with a fully three dimensional calculation. Finally a comparison is made with an experiment to demonstrate the effectiveness of the computational methods for studying fracture of work hardening materials.