Point-to-Point Kernel Analysis of an Experiment to Determine Wall Penetration of a Squarebased Concrete Structure by Gamma Radiation

Abstract

An analysis was made of shielding data obtained from an experiment designed to determine the wall attenuation properties against residual gamma radiation of a simple 12 ft × 12 ft × 8 ft concrete blockhouse for three different wall thicknesses (48, 93.7 and 139 lb/ft2). Roof thickness was chosen so that radiation scattered through the roof could be considered negligible. A plane source was simulated by successively positioning point isotropic Co60 and Cs137 sources on the ground surrounding the structure. The experiment was designed to obtain experimental evidence for the validity of certain aspects of L. V. Spencer's Theory of Structure Shielding against Fallout Radiation as presented in NBS Monograph 42. From analysis of a total of 9700 measurements, a point-to-point kernel was developed which permits calculation of the dose rate at any point within the structure from a point isotropic source on the ground outside. Doses calculated from the kernel differ less than 10 to 50 per cent from the experimental doses, except at extreme angles of incidence where the error can exceed a factor of 2. A computer program has been written which integrates the point-to-point kernel over infinite or finite rectangular areas. The kernel was integrated over an infinite area and results were compared with those obtained from Spencer's theory. It was found that the difference in reduction factor D/D0 from ‘experimental’ point-to-point kernel data and Spencer's theory did not exceed 40 per cent. Free-field (unshielded detector) dose rates obtained from integrating the point-to-point kernel over Co60 and Cs137 infinite-plane isotropic sources have been compared with those obtained experimentally by Rex-road and Schmoke. The difference was found to be 9 and 7 per cent respectively for the Co60 and the Cs137 source fields.