Accurate and precise isotopic measurement of sub-nanogram sized samples of foraminiferal hosted boron by total evaporation NTIMS

We report a total evaporation negative ion thermal mass spectrometry (TE-NTIMS) technique that enables precise and accurate (+/- 0.7%o; 2 s.d.) measurements of boron isotope ratios. The fundamental advantage of TE-NTIMS is that the effect of instrumental mass fractionation is minimised and sample signal maximised by analysing samples to exhaustion. We can analyse as little as 300 pg of B, which enables repeat analyses of dissolutions of small numbers of foraminifera (as little as 0.1 mg or similar to 10 individual foraminfera). This represents a several fold reduction in the number of tests required compared to previous NTIMS studies and brings the amount of sample into line with other commonly used paleo-proxies. Standard addition experiments indicate that the B-11/B-10 ratio of the NIST SRM 951 standard is not biased by differing amounts of seawater or carbonate matrix and yield an B-11/B-10 within error of the certified value. We also show that our sample preparation induces no additional variations (e.g. blank contribution) beyond our analytical uncertainty. We obtain B-11/B-10 ratios for seawater within error of values obtained using plasma ionisation, positive and negative thermal ionisation mass-spectrometry. Our measurements of core-top G. sacculifer from three ocean basins yield delta B-11 within analytical error (23.3-24.3 parts per thousand) and fall within the range of published values. This study, however, further highlights significant interlaboratory biases in isotopic compositions of core-top foraminifera. Significantly, we show that our approach is not influenced by processing blank nor systematic differences in mass bias between measurements of sample and standard, which has yet to be documented for some other laboratories. (c) 2006 Elsevier B.V. All rights reserved.