Conversion of AMS 10Be/Be ratios to reliable exposure ages (normalisation against standards and 10Be concentrations to age) is directly linked to the absolute 10Be half-life. However debate persists regarding its value, which in turn, dictates nominal values for (most) 10Be/Be AMS standards. The history behind this uncertainty is convoluted with conflicting details. Effectively upto 1990, the accepted 10Be half-life was 1.51Ma and nominal values for all 10Be/Be “standards” were contingent on this half-life. In 1991, NIST (Gaithersburg, USA) issued SRM-4325, with a certified 10Be/Be ratio of 26,800(±5%)x10-15 and a new 10Be half-life of 1.34±0.07 Ma. In 1993, Middleton et al, based on “absolute” measurements of SRM-4325, concluded its ratio to be 30,600x10-15 and that the 10Be half-life of 1.51Ma thus needed no correction. Hence it would seem that reliable ages can be assured if one makes a self-consistent choice in paired half-life and SRM ratio. However for ages >200ka and 10Be concentrations in ice and sediments, the absolute half-life value is critical - and equally, so is the reference (SL-HL) value for the 10Be production rate. At ANTARES we conducted a survey of AMS standards to confirm relative values against SRM-4325 and attempted to repeat an absolute (normalisation independent) ratio measurement. Preliminary results, within constraints of uncertainties in currents, transmissions, fractionation measurements etc suggests that NIST may have got it right. A lesser known problem centres on the accuracy of Al assays via ICP-AE in HF solutions of dissolved quartz for 26Al exposure dating. Multiple assays from independent aliquots and from repeats over extended periods of the same quartz powder and dissolved quartz-HF solutions (using the same procedures and ICP-AE facility) show occasionally inconsistent and unacceptable reproducibility well beyond the 2% single analytical error. A variety of non-HF, non-quartz solutions (blanks, reagents, ICP standard Al dilution series etc) indicated an acceptable precision and accuracy for the ICP-AE facility. To investigate this further we obtained form NIST a quartz glass powder (#SRM-165a) with a recommended Al concentration of 312ppm±5% (1ó). Initial tests on this powder (and control solutions) with variations in the HClO4-HF fuming process appear to reproduce the spread causing depressed Al-ICP values. We suspect that insoluble and hence intractable Al oxide maybe formed during fuming resulting in reduced Al ICP concentrations. The consequence is resulting ages which mimic burial or complex exposure histories when artificially reduced 26Al ages are compared to unaffected 10Be ages.
See more of Poster Session II
See more of The 10th International Conference on Accelerator Mass Spectrometry (September 5-10, 2005)