January 16, 2023 Dataset Open Access

Interlaboratory Comparison – PIXE for Heritage Science Experimental protocol (Metadata RADIATE TA Proposal 21002468-ST)

Pacheco, Claire; Berthet, Jean-Paul; Chiari, Massimo

Proposal title "ILC – PIXE for HS Experimental protocol". Measurements carried out at the external beam set-up for cultural heritage of the INFN LABEC accelerator laboratory, from 5 till 9 July 2021. Hands-off mode.

In the paradigm of open science, researchers in heritage science (HS) dream of being able to access instrumental data acquired on a particular object held by a foreign museum or on a corpus of archaeological objects that he / she will never have the opportunity to directly study. Can the comparison of PIXE datasets acquired on two separate beamlines be done safely? Are the measurement uncertainties the same from one facility to the other? These are the questions of users wishing, for example, to compare PIXE results obtained on Neolithic obsidians for provenance issues.

An inter-laboratory comparison (ILC) is then necessary and is being organized on an international scale. The dual objective of the study is the following: 1) the establishment of reference values of the components of a glass which is natural obsidian; 2) the comparison of the results obtained by different laboratories and the measurement uncertainties.

Five Japanese obsidian samples coming from the same block and a NIST 610 glass certified reference were analyzed by PIXE technique at the INFN LABEC ion beam laboratory using an extracted proton beam of 3.0 MeV to check reproducibility and repeatability of PIXE analysis and to establish reference values of the components of a glass such as natural obsidian. For the obsidians, the detected elements above the limits of detection were: Na, Mg, Al, Si, Cl, K, Ca, Ti, Mn, Fe, Zn, Ga, Rb, Sr, Y, Zr, N band Pb). Major elemental concentrations were actually transformed into concentrations of the prevalent oxides, whereas for trace elements they were not.

To test repeatability several measurements (from 6 to 8) were made on the very same area on each sample; this measurement batch on each sample has been repeated for five days to test reproducibility, simulating a new run for each batch (for instance shutting down and restarting the accelerator each day, using different extracted beam conditions...). Glass reference standards (NIST 1412 and NIST 620) were analysed as well each day for quality assurance purposes.