Accuracy of empirical formulas in evaluation of neutron dose equivalent inside the 60Co vaultsreconstructed for medical linear accelerators

Background: in southeast europe medical accelerators are sometimes placedin small vaults originally built for 60co treatment unit. in order to meetshielding requirements for high energy photon beams, the wall thickness hadto be increased. since the vaults are already limited in size, instead of addingmore concrete, materials with high-z elements were used. limited vault sizeand addition of high-z elements can contribute to the neutron doseequivalent for both medical personnel and patients. materials and methods:the most commonly used empirical equations for estimation of neutron doseequivalent at the maze door in the vault are by kersey and wu-mcginley. inorder to assess accuracy of these equations, monte carlo (mc) simulations ofvarious geometrical and compositional changes of vault were conducted.neutron ambient dose equivalent (h*n(10)) was observed when dimensionsof the vault walls were reduced gradually. results: the empirical equationsgave results with reasonable accuracy when vaults were of standard size.when the vault was decreased to the size of the usual 60co unit vault, the most commonly used equations showed significant difference in results (up to 90%) in comparison to mc simulations. mc simulations showed that introducing diffe rent materials in shielding can change the neutron doseequivalent in vicinity of accelerators. conclusion: for vaults limited in size, new simplified equation for neutron dose equivalent at the maze doors is presented, although performing a mc simulation of the specific vault is suggested.