Application of Electron Linacs to Sub-Surface Science and other Environmental Sciences


D.P. Wells

Idaho State University


Compact electron linacs are widely used in industrial and medical settings because of their reliability, low cost, and highly-directed beams. We are investigating the application of such accelerators to sub-surface science and other environmental applications. We used bremsstrahlung radiation beams from these linacs to demonstrate accelerator-based X-ray fluorescence (AXRF) in soil-samples contaminated with lead, mercury, uranium and other heavy metals. X-ray fluorescence is a widely used nondestructive technique for identifying elements in thin samples. The advantage of AXRF over conventional XRF lies in the much greater penetrability of the primary photon beam because of higher photon energies. We show that high energy (36 MeV) bremsstrahlung beams produce large K-shell atomic fluorescence X-ray yields that allow much larger sample volumes than conventional XRF. Samples as thick as 15-20 g/cm2 can be assayed for heavy metals (Z>30) at the level of a few parts per million, depending upon atomic number. Radioactive, mixed, and hazardous waste can be rapidly assayed by AXRF in a few minutes. AXRF can also be employed in a number of other environmental field applications, such as soil analysis, ore assays, and well logging. We are currently working on the portability of such systems to allow for field applications. A major technological spin-off from this work has been accelerator-based, gamma-induced positron annihilation spectroscopy (AG-PAS). AG-PAS exploits the fact that positrons produced in materials from pair-production yield information about stress, strain, fatigue and defects in that material. This is the first sensitive, highly-penetrating, non-destructive probe that works in all materials and allows, for example, the identification of parts on an aircraft that are about to fail, prior to failure. This is a major technological spinoff with enormous potential for commercial, industrial and research applications.

This work was supported by the Inland Northwest Research Alliance under grant number ISU001.