Analysis of Pu by Virtual-standard WDS-EPMA. Results of an Interlaboratory Round-robin Test

The understanding of the behaviour of nuclear fuels under different operating conditions requires accurate wavelength-dispersive electron-probe microanalysis (WDS-EPMA) of materials that contain actinide elements. Conventional WDS-EPMA is based on the measurement of k-ratios, i.e. the ratio of the x-ray intensity emitted by the element of interest in the unknown to that emitted by the same element in a reference standard. The advantage of using such approach is that several instrumental, spectroscopic and atomic parameters that are generally poorly known cancel out from the quantification equations and are no longer required. Moreover, the effect of the approximations adopted to describe quantities such as the depth-distribution of ionizations or the cross section for inner-shell ionization is largely reduced. As a result, the accuracy of concentrations obtained by WDS-EPMA using reference standards can be as good as 2%, even when x-ray attenuation effects are important. However, for some actinide elements such as Plutonium (Pu) or Americium (Am), reference standards are difficult to obtain. In these situations, the use of standardless methods of analysis [1,2], which use calculated intensities to replace measurements on the standards, would be an ideal alternative. However, the accuracy of such methods is limited by the poor knowledge of the different instrumental, spectroscopic and atomic parameters required, as mentioned above. Significant benefits would be gained if measurements on standards could be transferred among different WDS instruments; this would allow the creation of a set of ¿virtual standards¿ which could be shared among different laboratories. The use of virtual standards requires accurate knowledge of instrumental parameters (e.g. the spectrometer efficiency and solid angle of detection) and spectroscopic parameters (e.g. the line shape), but knowledge of atomic parameters is not required. This is important for the actinide elements, which involve the measurement of M xrays, where atomic parameters such as the fluorescence yield, the Coster-Kronig and super-Coster-Kronig transition probabilities or the cross section for M-subshell ionization are affected by large uncertainties (>30%).

MERLET C.;  LLOVET X.;  AUFORE L.;  BREMIER Stephan;  DESCHANELS X.;  DUGNE O.;  LAMONTAGNE J.;  RESTANI R.;  ROCHE C.;  TRIBET M.;  VAN RENTERGHEM W.; 

2009-12-21

CAMBRIDGE UNIV PRESS

JRC56281

1431-9276,   

http://journals.cambridge.org/action/displayJournal?jid=MAM,    https://publications.jrc.ec.europa.eu/repository/handle/JRC56281,   

10.1017/S1431927608081737,