Neutron beam monitoring for single-event effects testing

Abstract

The research described in this thesis aims to develop a neutron beam monitoring system using silicon photodiodes as the detecting elements for accelerated testing of electronics against neutron-induced single event effects (SEEs). The system can measure the transmission along the neutron beam line where several devices are tested simultaneously, allowing beam degradation by upstream experiments to be taken into account.

In signal processing, the pulse events from the output of the sensor are extracted by using a matched filter. This technique allows the pulse events to be detected effectively at a low false event rate. The pulse arrival times are acquired using the nonparalyzable counting system, which allows the interaction rates to be determined either indirectly based on the decay constant of the pulse interval distributions or directly on the basis of detection rates and sensor pulse widths. The optimum of sensor pulse width has been investigated in order to achieve the maximum of probability of detection with adequate energy resolution. A series of calculations have been undertaken to verify the correct operation of the detection software and to investigate system performance variations.

Results from irradiations at various neutron facilities where SEEs experiments can be carried out, such as Los Alamos Neutron Science Center (LANSCE), Tri-University Meson Facility (TRIUMF) and The Svedberg Laboratory (TSL) neutron beam, are presented and analyzed. The measurement of transmission in the neutron beam can be made via the measured pulse height spectra, pulse interval distributions and responses of the sensor irradiated. As the beam can be scattered, absorbed or enhanced by upstream devices, there are likely to be fluctuations in the transmission along the neutron beam, which can also be characterized by the beam monitoring system.

The recommended protocols for beam monitoring used at each neutron facility are investigated. At a low interaction rate, the protocols depend on the neutron fluence provided by the facility and the response of the sensor. At a high interaction rate, the probability of detection of the sensor should be determined first, and then the protocols are based on the probability of detection, neutron fluence and the response. In this thesis, the protocols of the beam monitoring system for use at ISIS ChipIr are also predicted on the basis of the pulse interval distributions and the mean of detected energy.

Based on the work undertaken in the project and presented in this thesis, suggestions are put forward for improving the monitoring system. Geant4 techniques can be used to model pulse height spectra so as to enable direct comparisons between theoretical simulations and experimental results. Furthermore, the protocols for beam monitoring for use at ChipIr as recommended in this thesis will be verified when neutron beams become available in the future.