Answer to Question #10514 Submitted to "Ask the Experts"

I am doing some simulations with MCNPX (Monte Carlo N-Particle eXtended) to determine the response of a Geiger Müller (GM) detector to gamma rays with different energies. As expected, the response increases from 20 to 90 keV after which it starts decreasing. Surprisingly enough at around 350 keV the response starts significantly increasing. I could not come up with an explanation to this problem. Your comments on the problem are greatly appreciated.

It is not unusual to find the type of response you have described for GM tubes exposed to gamma radiation or x rays. GM tubes are typically fabricated of materials that have atomic numbers appreciably greater than that of air or soft tissue, the materials for which dose-related quantities are most often assessed. The increase in response at low energies is associated with the greatly enhanced photoelectric effect in the higher atomic number material compared to air or water, since the photoelectric cross section varies as the 4th to 5th power of the atomic number; some enhancement also results from the reduction in attenuation with increasing energy. The decrease in response beyond the 90 keV that you cite is associated with the decrease in photoelectric cross section with increasing energy, offset somewhat by reduction in photon attenuation in the detector wall. The attenuation effect tends to decrease with increasing energy, and this may produce enhanced response as energy continues to increase.

Additionally, the count rate of the detector varies with the magnitude of the effective interaction coefficient of the photons in the detector. The manner in which the attenuation coefficients for the detector materials change, particularly as compared to the manner in which they change for the medium for which dose is being calculated (e.g., soft tissue or air), has an impact on the response of the detector. If you review the interaction coefficients for the materials of interest you can rationalize some of the observed behavior. For example, if you have a steel-walled detector that is intended to provide an estimate of dose (rate) to soft tissue and you review values of the respective mass attenuation coefficients, you will see that in the interval from 100 keV to 500 keV the ratio of iron-to-soft tissue attenuation coefficients change from about 2.2 at 100 keV, to 1.3 at 150 keV, to 1.06 at 200 keV, to 0.94 at 300 keV, to 0.90 at 400 keV, and 0.88 at 500 keV. The implication of the reduced ratios is that the interaction cross sections are becoming more heavily dominated by the Compton scattering cross section as energy increases, and this cross section depends on electron density of the medium. The electron density, on a mass basis, does not change dramatically with atomic number but it is somewhat greater for low atomic number materials compared to higher atomic number materials. At energies beyond 1.022 MeV the pair production cross section will be more dominant in the higher atomic number detector (compared to tissue or air), often leading to appreciably enhanced response at high photon energies (several MeV).

It is then expected that the GM response will go through a minimum value, probably in the few hundred keV range, then increase and change rather slowly beyond that up to energies somewhat beyond 1 MeV, and then increase more noticeably for energies of several MeV.

You can see this kind of behavior in curves provided by manufacturers for some detectors. For example, see the curves for the Ludlum 44-38 steel-walled GM detector, and the thin window Ludlum 44-9 GM.

George Chabot, PhD, CHP