Amorphous hydrogenated boron carbide for solid-state direct-conversion thermal neutron detection

Abstract

Amorphous hydrogenated boron carbide (a-BC:H) is one of a very limited number of neutron-sensitive materials. It has been studied over the last three decades for solid-state thermal neutron detection due to its high resistivity, moderate bandgap, high neutron absorption cross-section, and high stability under harsh physical and chemical environment. However, its success has been hindered by its poor and/or not well understood charge transport properties as well as fabrication challenges. This study focuses on obtaining thick and stable a-BC:H films using plasma-enhanced chemical vapor deposition (PECVD) and optimizing their charge transport properties for potential application in solid-state direct-conversion thermal neutron detection. We have investigated the effect of single-carrier transport and low charge carrier mobility, which are expected in a-BC:H films, on detection efficiency and spectral performance of a thin-film B4C detector using numerical Monte Carlo calculations. Experimentally, we have used space-charge-limited current (SCLC) analysis to extract the charge carrier mobility in a-BC:H thin films. To better describe the extracted charge carrier mobility, we have presented an extension of SCLC theory to include negative field-dependence in mobility as well as a theory to check self-consistency of the extracted mobility. Toward obtaining stable thick films and optimizing the charge transport metrics, we have deposited multiple series’ of a-BC:H thin films using an ortho-carborane precursor with varying PECVD process parameters, including substrate temperature, RF power, process pressure, carrier gas flow rate, and partial precursor flow. Films grown using higher substrate temperature and higher RF power demonstrated higher charge carrier mobility. Partial precursor flow was found to correlate significantly with film growth rate and film properties. The precursor flow rate was found to be very sensitive to the precursor bubbler temperature, which allowed us to fine-tune the precursor flow and produce films with predictable growth rate and film properties. With optimized growth conditions, we demonstrated a 3 μm thick and stable film deposited on copper foil exhibiting a carrier mobility value of 8×10–6 cm2⋅V–1⋅s–1 and a resistivity value of ∼1012 Ω⋅cm. With this film thickness and carrier mobility, our Monte Carlo calculations suggested that a neutron detector using an integration time of ≥5 μs can detect neutrons for a carrier lifetime >10 μs, and the intrinsic detection efficiency saturates to ∼10% for a carrier lifetime ≥100 μs.