Observed discrepancies on the order of 40% between the previously computed radiation doses to factory workers in a Nagasaki factory and measured values estimated from genetic aberrations are investigated. A conjecture that these discrepancies can be resolved by including the shielding of workers by workbenches and tools is tested. A simple two-dimensional r - z model of the Nagasaki bomb explosion surrounding area with cylindrical shell representations of the factory walls and a smeared workbench is used to set an upper bound on the dose reduction as a function of workbench material. The model covers a 1353.5 m x 1500 m region, mostly air, discretized by almost 71,000 computational cells; energy is discretized by 46 neutron and 23 gamma groups. Angular anisotropy of scattering is represented by a P5 expansion, while the angular dependence of the angular flux is discretized along a 240-angle, biased quadrature set. A first collision source is computed by GRTUNCL and used in DORT to compute the fully collided flux in 64-bit arithmetic precision. All groups are converged in DORT to 10 -4 , and the sum of the converged flux and GRTUNCL's uncollided flux are folded with flux-to-dose conversion factors to generate pointwise maps of group dose rates. It is concluded that with this simplified model of the workbench, a workbench consisting of only 0.16 m of solid wood and 0.04 m of solid carbon steel is sufficient to reduce the total dose by 40%. These are reasonable material contents in a typical workbench, thus the conjecture is affirmed, and more detailed calculations of the total dose to individual workers in various exposure scenarios are expected to eliminate remaining discrepancies.
PDF
download
878987797
028-85220240
