skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: A Numerical Study of On-Axis Dose Rate from Ta and W Bremsstrahlung Converter Targets

Technical Report ·
DOI:https://doi.org/10.2172/15016409· OSTI ID:15016409

The bremsstrahlung converter target in radiographic accelerators is not, in general, considered a high-technology piece of equipment. In its essential form it is merely a solid plate of high-Z metal, usually tungsten (W) or tantalum (Ta); electrons go in, X-rays come out [1]. However, there are some important factors to keep in mind for this kind of target system. One is a constraint on the target itself: the proper thickness of material. Too little material reduces the probability that an electron will have a significant nuclear collision before exiting the plate. Too much material has a number of effects: small-angle scattering will occur to such an extent that bremsstrahlung photons will not be pointed in the forward direction. Electrons which small-angle scatter away and then back to the forward direction will have moved to larger radii as they traverse the target, increasing the effective source size. Electrons ''backscattered'' from the target --primaries or secondaries ejected from the upstream surface after sufficient angular scatter--exert a defocusing force on the incoming beam due to increased space charge at fixed (or even slightly reduced) current. Finally, a sufficiently thick target will begin to self-attenuate the X-ray photons produced in the upstream portion of the plate. A second constraint is obvious but is harder to accommodate when designing a radiographic accelerator system. The angular distribution of the incoming electron beam will change the forward dose. Just as electrons which have undergone small-angle scatter will no longer produce forward dose, electrons which have large angles before they ever enter the target cannot produce forward dose. Accurate prediction of dose requires incorporating the effect of the initial angle of the electron coming into the target material. The further step of controlling the angular distribution--which means keeping it as close to zero as possible--is difficult since it tends to drive important beam parameters in directions we do not want (large spot size) or cannot achieve (very low emittance). In this report we characterize the bremsstrahlung performance of Ta and W converter targets over a range of electron energies (2-20 MeV) and angles. The studies are all performed with the MCNP radiation transport code [2], version 4b. A number of steps are involved in the process. First, we must construct the absorption properties of air given photons of various energies, so that the distribution of photons produced by the incoming electrons can be converted into a dose rate as commonly used in radiography. Then we study the dose rate of various electron energies in Ta and W, as a function of target thickness, and find that there is an optimum for a given energy. Following that, we describe how to incorporate angular dependence in the (rather obtuse) MCNP interface, and then study how the dose rate varies with the distribution of incoming angles. The major results and useful curve fits are summarized at the end. The appendix contains listings of useful scripts and MCNP templates. This report does not address the backscattered electron issue [3]. That problem is complex because it requires self-consistent treatment of the electrons in the external electric fields once they are ejected from the target. Suffice it to say that a thicker target produces more backscatter and therefore more defocusing. When choosing a target thickness, rather than selecting from the peak output given below, one may prefer to choose a (usually much thinner) target corresponding to the 95% output level, or even lower as desired. Of course, a target which must withstand multiple beam pulses has constraints driving the thickness in the opposite direction, in order to maintain sufficient line density during the hydrodynamic evolution of the material.

Research Organization:
Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Sponsoring Organization:
US Department of Energy (US)
DOE Contract Number:
W-7405-ENG-48
OSTI ID:
15016409
Report Number(s):
UCRL-TR-212255; TRN: US0502386
Resource Relation:
Other Information: PBD: 11 Feb 2005
Country of Publication:
United States
Language:
English