SIZE MATTERS: HOW SMALL IS TOO SMALL FOR ISOTOPE RATIO ANALYSIS? (Invited Presentation)
For oxygen isotope analysis of d18O in silicates (18O/16O ~2 x 10-3), precision of ±0.3‰ (2 SD) is attained for 10 micron SIMS spots and ±2‰ for sub-micron spots (IMS-1280, Page et al. 2007) that are within an order of magnitude of counting statistics. The precision limit for d18O is approximated by N0.5 for number of atoms counted of the minor isotope by SIMS (106 atoms of 18O ~ 1‰). For sulfur, analysis of d34S in pyrite (34S/32S ~ 4.4 x 10-2) will attain similar precision with samples 20 times smaller for a similar ion yield. Thus for materials at terrestrial abundance, useful analytical volumes approach physical limits and smaller spots should be limited to extraterrestrial or experimental samples where extreme isotope ratios can be studied and high precision is not required. Likewise for lead, analysis of 207Pb/206Pb can been extended to far smaller volumes because radioactive decay of U causes greater variability, ratios are closer to unity, and precisions over 10% can be geologically informative. The ability by APT to locate and identify single atoms by mass requires decisions on how small a population must be analyzed to be meaningful. For APT 207Pb/206Pb ratios in zircon, the precision of both isotopes and their backgrounds must be considered and results are most significant for older Precambrian zircons, or if Pb is concentrated in clusters or other features. For example, in the 2.5 Ga ARG zircon, 207Pb/206Pb = 0.17 ± 0.065 (2 SD) by APT (vs. 0.1648 by SIMS) based on counting 1326 atoms of clustered Pb (Valley et al. 2015). The high spatial resolution afforded by in situ analysis provides fundamental new information for the study of small, precious and zoned materials from Earth and elsewhere.