Assessment of PGAA capability for low-level measurements of H in Ti alloys

The accuracy of low-level hydrogen measurements with prompt gamma-ray activation analysis (PGAA) depends on identifying and accounting for all background H signals, including interfering signals. At the cold-neutron (CN)PGAA facility at the NIST Center for Neutron Research, the sources of background H signals were investigated in the context of titanium-based matrices containing low-levels of H (<300 mg H per kg Ti) with the measurements of prepared standards (mixtures of polyvinyl chloride and titanium oxide) and Ti alloy (Ti6Al4V) samples. The sensitivity ratio, defined as the ratio of the H signal to the Ti signal per unit mass ratio of H in Ti, was determined (1) with the measurements of prepared standards and (2) based on partial gamma-ray production cross sections and full-energy detection efficiencies. The resulting calibrations from these two approaches agreed within experimental uncertainty. A series of Ti alloy NIST Standard Reference Materials (SRMs) previously certified for the H content (SRMs 2452, 2453, 2453a, 2454) were used as test cases, with the mass fractions determined based on the sensitivity ratios derived from method 1 and method 2, respectively. The results agreed with the certified values within experimental uncertainties, validating the analysis performed on the new instrument with newly-prepared standards at low H mass fractions (method 1), and with the standard-independent analysis (method 2). Various sample mounting improvements were made to lower the background H signal. Spectral interferences near the H peak were identified as potential sources of bias and as a limiting factor in the detection limit of H in Ti alloy samples.