Preliminary results on the development of a fast-neutron tomography system for void fraction distribution measurements

There has been a continued desire to include more physical phenomena and mechanistic treatment in the prediction of Critical Heat Flux (CHF) for decades. A major impediment to these efforts is the lack of local measurements at CHF location, as typical instrumentation employed in these full-scale experiments only involves wall temperature measurements and at best a few point measurements of void fraction. There is a need to have subchannel (or better) void measurements to improve our phenomenological understanding of CHF and for the validation of computer codes. Several test facilities are currently deploying subchannel level void measurements using computed tomography including recent gamma-ray based systems at the University of Michigan. While such systems are very promising, the lack of contrast when imaging through thick pressure vessels may prevent their application at some full-scale R&D facilities. Fast neutron sources are particularly advantageous in this regard since the interaction cross sections for fast neutrons with structural materials is small and scattering cross sections with liquid water are substantial. Pioneering work has previously demonstrated the potential for fast-neutron CT methods but limitations on source neutron flux and detector efficiencies limited their industrial development/deployment. There have been recent advancements in small portable fast neutron generators capable of fluxes up to 1x10¹⁰ n/s, ultra-high efficiency neutron-photon pixelated convertor screens, and low light sensing Silicon Photomultipliers (SiPMs) and associated electronics. Portable neutron sources and advanced detectors such as these may allow for the deployment to existing full scale facilities. The goal of this research is to design and build a prototype fast neutron tomography system capable of imaging CANDU full-scale CHF experiments with subchannel (or better) resolution. We have performed MCNP neutron transport calculations which demonstrate that there is a strong potential for measuring subchannel level void fraction within an acceptable time frame. This work will present the design, MCNP calculations and some of the preliminary results using advanced SiPMs and supporting electronics for void fraction measurements in a CANDU bundle.