Personnel Hazards from Medical Electron Accelerator Photoneutrons "

PERSONNEL HAZARDS FROM MEDICAL ELECTRON ACCELERATOR PHOTONEUTRONS" R. C. McCall, T. M. Jenkins, R. A. Shore and P. D. LaRiviere+ Stanford Linear Accelerator Center, CA, U.S.A. and Varian Associates? INTRODUCTION For medical accelerators, neutron penetration through the room entry door is the major personnel hazard. Most therapy accelerator rooms are designed with at least a rudimentary maze to avoid the use of massive doors. Often, however, the maze may be similar to those shown in Fig. 1. In Fig. 1, scale outline drawings of some medical electron accelerator rooms are shown where the authors have made neutron measurements outside the doors which were of different thicknesses and compositions. The results are tabulated in Table I. It should be noted that there can be significant dose equivalents (H) at the door when a maze is inadequate, and that all three componentsfast neutron, thermal neutron, and neutron capture y-rayscan be equally important. Also, these capture Y-rays are very penetrating; (TVL"5-7 cm of lead). SIMPLE METHODS OF CALCULATING MAZE EFFECTIVENESS For a good review of neutron penetration of mazes, the authors suggest Chapter 4 by Selph of Ref. 1. Most of the extensive work on mazes is not directly applicable to medical electron accelerators, however, for various reasons. Monte Carlo or albedo computer calculations have been shown to correctly calculate neutron maze penetration. We have explored several simpler methods of predicting neutron penetration of a maze which do not rely upon computer codes or difficult calculations. Method 1 is an albedo method based upon the work of French and Wells (2), and is described as follows: On a room drawing, the portion of the walls, floor and ceiling that could be directly irradiated by neutrons from the accelerator, and then scatter the neutron directly to the door, are outlined, and their areas determined. An effective center, P, is chosen for each. The incident and reflected angles are measured from these points. Next, the dose albedo ad (21, is used; ad = ct(Eo) COS~/~~~ cos 0 Eq. 1 where B. and 0 are the incident and reflected angles, respectively, measured from the normal to the wall. For the range of neutron spectra from medical accelerators, a single value for a(E,) of 0.11 can be used for concrete. Next, H is assumed to propagate according to the inverse square law for the distances, Ra and Rb, from the accelerator source to P and from P to the door, respectively. H at the door then is the sum of the individual contributions from each of the n illuminated areas; that is, H =? Ho An ad, iaX Rb; Eq. 2