Quantitative EPMA of Nano-Phase Iron-Silicides in Apollo 16 Lunar Regolith

Until recently, quantitative EPMA of phases under a few microns in size has been extremely difficult. In order to achieve analytical volumes to analyze sub-micron features, accelerating voltages between 5 and 8 keV need to be used. At these voltages the normally used K X-ray transitions (of higher Z elements) are no longer excited, and we must rely of outer shell transitions (L and M). These outer shell transitions are difficult to use for quantitative EPMA because they are strongly affected by different bonding environments, the error associated with their mass attenuation coefficients (MAC), and their proximity to absorption edges. These problems are especially prevalent for the transition metals, because of the unfilled M5 electron shell where the Lα transition originates. Previous studies have tried to overcome these limitations by using standards that almost exactly matched their unknowns. This, however, is cumbersome and requires accurate knowledge of the composition of your sample beforehand, as well as an exorbitant number of well characterized standards. Using a 5 keV electron beam and utilizing non-standard X-ray transitions (Ll) for the transition metals, we are able to conduct accurate quantitative analyses of phases down to ~300nm. The Ll transition in the transition metals behaves more like a core-state transition, and unlike the Lα/β lines, is unaffected by bonding effects and does not lie near an absorption edge. This allows for quantitative analysis using standards do not have to exactly match the unknown. In our case pure metal standards were used for all elements except phosphorus. We present here data on iron-silicides in two Apollo 16 regolith grains. These plagioclase grains (A6-7 and A6-8) were collected between North and South Ray Craters, in the lunar highlands, and thus are associated with one or more large impact events. We report the presence of carbon, nickel, and phosphorus (in order of abundance) in these iron-silicide phases. Although carbon is an especially difficult measurement, (with contamination from the lab environment, sample, and vacuum system being a large problem) we found that the iron-silicide phases contain a few weight percent carbon. X-ray mapping shows carbon to be concentrated within the silicide blebs. We conducted sample reference (i.e. baseline) carbon measurements in standards mounted in the same block as the sample, to establish a contamination baseline then any carbon measured above this baseline was assumed to be real. This finding seems to indicate that while the iron-silicide phases formed in the reducing conditions of the lunar surface, these conditions were not low enough to form the phases on their own and needed the presence of carbon to reduce them down to the much lower reducing conditions were native silicon is stable. The source of the carbon and nickel found in the iron-silicides is most likely form an impactor, rather than from the lunar surface.