The Attenuation of Gamma Rays at Oblique Incidence

Abstract

The high cost and the heavy weight of barriers required for protection against high-energy roentgen or gamma rays warrant a detailed study of the factors involved in obtaining an economical solution to the protection problem. One factor which has had no detailed experimental or theoretical evaluation is the attenuation of radiation obliquely incident upon radiation barriers. It is usually assumed (1) that radiation obliquely incident on a barrier is attenuated to the same extent as it would be when normally incident on a barrier whose thickness is equal to the slant distance through the oblique barrier. This assumption, however, may lead to a significant underestimate of the barrier thickness required to produce a specified attenuation. In most protection problems, scattered radiation makes an important contribution to the transmission of the barrier (2). When radiation is normally incident, the scattered radiation which emerges from the barrier must always travel a distance greater than that traveled by the unscattered radiation. In oblique incidence, however, some of the scattered radiation travels a much shorter distance than the unscattered radiation. In view of this, the assumption mentioned above may lead to estimates of barrier thickness that are too low. The results given in this report indicate the errors that may be introduced. For purposes of discussion and presentation of results, it will be convenient to compare the attenuation experimentally determined for obliquely incident radiation with that computed from the attenuation curves obtained for normal incidence, assuming a barrier thickness equal to the slant thickness through the oblique barrier. The experimental arrangement developed for the above problem was also used to determine the transmission characteristics of combination barriers which have special shielding properties. For example, since lead readily absorbs much of the radiation scattered in a low atomic number material, a relatively thin layer of lead on the emergent face of such material will greatly increase the attenuation of the combination. Although such effects can be predicted from theoretical considerations, no previous attempt has been made to evaluate them experimentally. Qualitative Predictions A detailed theoretical study of the attenuation of obliquely incident radiation is now in progress at this laboratory and hence no attempt will be made to give a complete analysis of the problem. However, a simple qualitative discussion employing only single scattering is included in an appendix and will be helpful in visualizing the difference between the attenuation of normally and obliquely incident radiations. It is shown that for a given slant thickness one should expect for thin barriers a decrease in transmission for increasing angles of incidence and, conversely, for thick barriers an increase in transmission with increasing angles of incidence.