Abstract

Boron-lined gaseous neutron detectors intrinsically suffer from the “self-absorption” effect and hence their neutron detection efficiencies might be nonconstant, when the threshold of them undergoes unexpected variation. To address this problem, we propose a maximum likelihood estimation-based threshold determining method, via which the pulse height spectra acquired under different high voltages could be rescaled to the same abscissa axis of deposited energies. Monte Carlo simulations were conducted to study the relationship between the precision of the threshold determination and the counting statistics, as well as the energy deposition spectrum, indicating that the grazing angle incidence neutron detector can achieve a fairly small relative standard variation for a spectrum with a modest total neutron counts. A prototype grazing angle incidence neutron detector has been constructed and tested under high voltages ranging from 500 V to 950 V. The experimental results show that, with the maximum likelihood estimation-based method, the threshold's relative standard deviations are less than 10% even with a total neutron counts as low as 100. As a result, the maximum relative variance of the counting plateau curve is 4.8% in the region of [500 V, 900 V] when the detector operates at a grazing angle of 1.8°. The results presented in this study indicate that combining the grazing angle incidence and the maximum likelihood estimation-based threshold determining method would be a promising way to achieve stable neutron detection efficiency for neutron detectors with the boron layer as the neutron convertor.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call