Modeling the neutron yield of a therapeutic thermal neutron source driven with a repetitively pulsed electron linac

Summary form only given. Hampering the development of the medical application of thermal neutrons is the lack of a compact, low cost, high-flux thermal neutron generator. Presently, the only sources of high-flux thermal neutron beams are nuclear research reactors. These facilities are few in number and often lack the clinical environment necessary for medical research. For the therapeutic application of thermal neutrons to become a viable treatment modality, clinically deployable neutron generators need to be developed and tested. A thermal neutron generator driven by a repetitively pulsed electron linac would overcome these difficulties by being compact and relatively low cost. These devices have a long history in hospitals, such systems being found in virtually every major oncology department. Furthermore, the possibility exists for the conversion of these machines into neutron generators on either a full or part-time basis. The repetitively pulsed electron linac driven thermal neutron generator is based on the photonuclear dissociation of deuterium. Deuterium has a very low photonuclear threshold (2.225 MeV) and is available in a very convenient form, that of heavy water, which is also an excellent neutron moderator. Electrons of moderate energy (5 to 10 MeV) from the electron linac are directed onto a high Z target, generating bremsstrahlung. Very large photon fluxes can be generated in this manner. These photons impinge on the photoneutron target where the neutron generation takes place. The neutrons are subsequently moderated, filtered, and directed to the patient position through a beam port. This paper discusses the modeling of the neutron yield and neutron spectra within the photoneutron target. The Monte Carlo for N-Particle (MCNP) transport code is used for modeling the X-ray converter and the neutron generation within the photoneutron target. Current results are discussed.