On the accuracy of total-IBA

Abstract “Total-IBA” implies the synergistic use of multiple IBA techniques. It has been claimed that Total–IBA inherits the accuracy of the most accurate IBA technique used. A specific example is now given of this where (in vacuo) EBS/PIXE of a glass sample uniform in depth is validated against absolutely calibrated EPMA of the same sample. The EPMA results had a mass closure gap of 2.0 ± 0.6 wt%; the full PIXE analysis determined the composition of this missing 2 wt%. The PIXE calibration was against a single certified glass sample, with uncertainties per line ~10%. Benchmarking also demonstrates ~10% underestimation of the Si scattering cross-section at proton energies ~3 MeV. But the Total-IBA determination of the silica content had a low standard uncertainty of about 2%. This is due to the strong constraints of both the chemical prior and also the mass closure properties of the EBS. Irradiation-induced sodium migration in this soda-lime glass is explored.