Standardisation of 85Kr

As part of a BIPM key-comparison of (85)Kr, a primary standardisation using internal gas proportional counting was performed. The supplied activity for the comparison was approximately 40 MBq and a two-stage dilution was required to reduce the activity concentration to a level suitable for gas counting. A new gas-handling rig was constructed for performing the dilutions. The dilutions, however, introduced significant uncertainties in the final result, so that additional methods suitable for measurement at higher activity levels were also used. A series of dilution ampoules with activities of 6 and 1 MBq were prepared in the new NPL gas-handling rig using inactive krypton as a diluent. Internal gas proportional counting was performed on each of the 1 MBq dilution ampoules. The proportion of the activity transferred to the counting system was estimated using pressure and volume data and the total ampoule activity calculated. Counting losses below the threshold were assumed to be 2%. The effect of changing the composition of the counting gas by inclusion of krypton was evaluated and found to not significantly change the gas gain, i.e. losses below the noise threshold (approximately 120 eV) remained essentially constant. The proportional counters were assumed to be 100% efficient with an uncertainty of 0.5% (k=1). Both 1 and 6 MBq ampoules were assayed by gamma-spectrometry using HPGe and NaI(Tl) detectors. This method resulted in an activity value with a smaller uncertainty than the primary method. Activity values for the three methods employed were consistent within the uncertainty of measurement.