Improvements in routine uranium isotope ratio measurements using the modified total evaporation method for multi-collector thermal ionization mass spectrometry

A new version of the modified total evaporation (MTE) method for isotopic analysis of uranium samples by multi-collector thermal ionization mass spectrometry (TIMS), with high analytical performance and designed in a more user-friendly and routinely applicable way, is described in detail. It is mainly being used for nuclear safeguards measurements, but can readily be applied in other scientific areas like geochemistry. The development of the MTE method was organized in collaboration of several key nuclear mass spectrometry laboratories, namely the New Brunswick Laboratory (NBL), the Safeguards Analytical Laboratory (SAL, now SGAS-Safeguards Analytical Services) of the International Atomic Energy Agency (IAEA), the Institute for Transuranium Elements (ITU/JRC), and the Institute for Reference Materials and Measurements (IRMM/JRC), with IRMM taking the leading role. Due to the use of the total evaporation (TE) principle the measurement of the major ratio n((235)U)/n((238)U) is routinely being performed with an accuracy of 0.02%. In contrast to the TE method, in the MTE method the total evaporation process is interrupted on a regular basis to allow for correction for background from peak tailing, internal calibration of a secondary electron multiplier (SEM) detector versus the Faraday cups, peak-centering, and ion source re-focusing. Therefore, the most significant improvement using the MTE method is in the measurement performance achieved for the minor ratios n((234)U)/n((238)U) and n((236)U)/n((238)U). The n((234)U)/n((238)U) ratio is measured using Faraday cups only with the result that the (relative) measurement uncertainty (k = 2) is better than 0.12%, which is an improvement by a factor of about 5-10 compared to TE measurements. Furthermore, the IAEA requirement for the measurement performance, defined here as the sum of the (absolute) deviation of the measured from the true (certified) value plus the (absolute) measurement uncertainty (k = 2), for n((236)U)/n((238)U) ratio measurements is 1 x 10(-6), but the MTE method provides a measurement performance which is, depending on the ratio, by several orders of magnitude superior compared to this limit and to the TE method. For routine MTE measurements a detection limit of 3 x 10(-9) was achieved using an SEM detector for detecting the isotope (236)U. The MTE method is now routinely being used at all collaborating laboratories with the hope that more laboratories will implement this capability in the future as well. Additional applications for the MTE method are presented in this paper, e. g., for absolute Ca isotope measurements.