Cosmogenic Be-10 and Al-26 sample preparation at the University of Wollongong

The University of Wollongong (UOW) cosmogenic Be-10 and Al-26 sample preparation laboratory has been in operation since the start of 2017. As primarily a feeder laboratory to ANSTO' s Centre for Accelerator Science, our sample preparation procedures have been optimised with consideration to the setup of ANSTO' s 6MV SIRIUS accelerator, and aim to achieve a balance between sample throughput as well as Be and Al target purity. A comparatively small number of samples (n = 68) have also been prepared for measurement at the Australian National University 14UD accelerator. Be-10/Be-9 ratios of procedural blanks measured on SIRIUS have fluctuated over time with the median for the year 2017 being 5.16 x 10(-16 )(IQR = 4.11 x 10(-16) to 7.16 x 10(-16), n = 18), increasing to 1.62 x 10(-15) (IQR = 1.05 x 10(-15) to 2.17 x 10(-15), n = 31) for 2018 and 2019, a period coinciding with elevated boron levels in our samples, and finally decreasing to 1.15 x 10(-15 )(IQR = 8.63 x 10(-16) to 1.60 x 10(-15), n = 34) for 2020 and 2021. In contrast, Al-26/Al-27 ratios of procedural blanks measured on SIRIUS have shown a slight but continuous improvement over time with the median for 2017 of 1.70 x 10(-15) (IQR = 8.19 x 10(-16) to 5.25 x 10(-15), n = 12) decreasing to 1.07 x 10(-15) (IQR = 8.5 x 10(-16) to 1.53 x 10(-15), n = 13) for 2021. Median Be-10/Be-9 relative uncertainty of procedural blanks analysed on SIRIUS is 18 % (IQR = 15 % to 22 %; n = 86) whereas the median Al-26/Al-27 relative uncertainty of procedural blanks is higher at 60 % (IQR = 41 % to 100 %; n = 56), statistic resulting from most blanks yielding low Al-26 counts (median = 2; IQR = 1-4.5). Average (BeO-)-Be-9 output relative to standard is between similar to 70 % - 80 % for samples analysed on SIRIUS (n = 895) and similar to 130 % for samples analysed at ANU (n = 68). Al-27(-) output relative to standard is lower for samples pressed into cathodes at UOW (similar to 60 %, n = 432) and analysed on SIRIUS than for those pressed at ANSTO (similar to 90 %, n = 119). Average Al-27(-) output relative to standard for samples analysed at ANU is similar to 80 % (n = 62). Be-10 and Al-26 measurements of various laboratory intercomparison materials prepared at UOW between 2017 and 2022 yield results in agreement with consensus values confirming that our chemistry procedures are robust and in line with those elsewhere.