Trace‐element concentrations in single circumstellar silicon carbide grains from the Murchison meteorite

— Concentrations of the trace elements Mg, Al, Ca, Ti, V, Fe, Sr, Y, Zr, Ba and Ce were determined by ion microprobe mass spectrometry in 60 individual silicon carbide (SiC) grains (in addition, Nb and Nd were determined in 20 of them), from separate KJH (size range 3.4–5.9 μm) of the Murchison carbonaceous meteorite, whose C-, N- and Si-isotopic compositions have been measured before (Hoppe et al., 1994) and provide evidence that these grains are of stellar origin. The selected SiC grains represent all previously recognized subgroups: mainstream (20 < 12C/13C < 120; 200 < 14N/15N; Si isotopes on slope 1.34 line), grains A (12C/13C < 3.5), grains B (3.5 < 12C/13C < 10), grains X (15N excesses, large 28Si excesses) and grains Y (150 < 12C/13C < 260; Si isotopes on slope 0.35 line). Data on these grains are compared with measurements on fine-grained SiC fractions. Trace-element patterns reflect both the condensation behavior of individual elements and the source composition of the stellar atmospheres. A detailed discussion of the condensation of trace elements in SiC from C-rich stellar atmospheres is given in a companion paper by Lodders and Fegley (1995). Elements such as Mg, Al, Ca, Fe and Sr are depleted because their compounds are more volatile than SiC. Elements whose compounds are believed to be more refractory than SiC can also be depleted due to condensation and removal prior to SiC condensation. Among the refractory elements, however, the heavy elements from Y to Ce (and Nd) are systematically enriched relative to Ti and V, indicating enrichments by up to a factor of 14 of the s-process elements relative to elements lighter than Fe. Such enrichments are expected if N-type carbon stars (thermally pulsing AGB stars) are the main source of circumstellar SiC grains. Large grains are less enriched than small grains, possibly because they are from different AGB stars. The trace-element patterns of subgroups such as groups A and B and grains X can at least qualitatively be understood if grains A and B come from J-type carbon stars (known to be lacking in s-process enhancements shown by N-type carbon stars) or carbon stars that had not experienced much dredge-up of He-shell material and if grains X come from supernovae. However, a remaining puzzle is how stars become carbon stars without much accompanying dredge-up of s-process elements.