It is well known that the trace-elements content present in archaeological pottery can give information on the artefacts provenance. Their absolute determination can be obtained by using XRF spectrometers. However, portable X-ray tubes are subjected to intensity and energy fluctuations that could induce errors in the concentration determinations. In the present work a new portable XRF spectrometer (the BSC-XRF system), based on a X-ray tube with a stability control system, has been used for the non-destructive quantitative determination of some trace elements (Rb, Sr, Y, Zr and Nb) in 50 fine ceramic sherds. The BSC-XRF system, designed and realised at the LNS/INFN LANDIS laboratory of Catania (Italy) in collaboration with the CNR-Agenzia 2000 Project (Italy), consists of a portable X-ray tube coupled to an ancillary Si-PIN detector for controlling the stability of the beam (both in energy and in intensity) and of a second Si(Li) detector for acquiring the characteristic X-ray spectra emitted by the sample being analysed. The BSC-XRF system as well as the energy and intensity stability control method is presented and discussed. The absolute concentration of Rb, Sr, Y, Zr and Nb has been determined in 50 pottery sherds by using a method based on a multi-linear regression approach developed at the LNS/INFN LANDIS laboratory. The analytical procedure has been previously tested by using different petrologic standards; the average deviation between certified composition and the XRF quantitative data is less than 10%. It could be observed that the XRF technique is a surface method and a quantitative determination is meaningful only when the material being analysed is homogenous. In order to test the homogeneity of the material composing the analysed samples, a small portion (less than 1 gr) of 10 sherds has been powdered, carefully homogenised and pressed to a disk-shaped pellet. The powdered pellets have been measured with the BSC-XRF system and quantitatively analysed by using the above mentioned method. The deviations between the non-destructive and destructive XRF data are 6 % for Rb, 10 % for Sr, 25 % for Y and Zr and 13% for Nb. Finally, in order to classify the 50 artefacts in typological classes, a multivariate cluster analysis has been performed. Results are presented and discussed.
Non-destructive quantitative determination of trace elements in fine ceramics by using a portable Beam Stability Controlled XRF spectrometer (BSC-XRF)
Romano FP;
2004
Abstract
It is well known that the trace-elements content present in archaeological pottery can give information on the artefacts provenance. Their absolute determination can be obtained by using XRF spectrometers. However, portable X-ray tubes are subjected to intensity and energy fluctuations that could induce errors in the concentration determinations. In the present work a new portable XRF spectrometer (the BSC-XRF system), based on a X-ray tube with a stability control system, has been used for the non-destructive quantitative determination of some trace elements (Rb, Sr, Y, Zr and Nb) in 50 fine ceramic sherds. The BSC-XRF system, designed and realised at the LNS/INFN LANDIS laboratory of Catania (Italy) in collaboration with the CNR-Agenzia 2000 Project (Italy), consists of a portable X-ray tube coupled to an ancillary Si-PIN detector for controlling the stability of the beam (both in energy and in intensity) and of a second Si(Li) detector for acquiring the characteristic X-ray spectra emitted by the sample being analysed. The BSC-XRF system as well as the energy and intensity stability control method is presented and discussed. The absolute concentration of Rb, Sr, Y, Zr and Nb has been determined in 50 pottery sherds by using a method based on a multi-linear regression approach developed at the LNS/INFN LANDIS laboratory. The analytical procedure has been previously tested by using different petrologic standards; the average deviation between certified composition and the XRF quantitative data is less than 10%. It could be observed that the XRF technique is a surface method and a quantitative determination is meaningful only when the material being analysed is homogenous. In order to test the homogeneity of the material composing the analysed samples, a small portion (less than 1 gr) of 10 sherds has been powdered, carefully homogenised and pressed to a disk-shaped pellet. The powdered pellets have been measured with the BSC-XRF system and quantitatively analysed by using the above mentioned method. The deviations between the non-destructive and destructive XRF data are 6 % for Rb, 10 % for Sr, 25 % for Y and Zr and 13% for Nb. Finally, in order to classify the 50 artefacts in typological classes, a multivariate cluster analysis has been performed. Results are presented and discussed.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
