1Kei Shimizu,1ConelM. O’D. Alexander,1Erik H.Hauri,2Adam R.Sarafian,1Larry R.Nittler,1Jianhua Wang,3Steven D.Jacobsen,4Ruslan A.Mendybaev
Geochimica et Cosmochimica Acta (in Press) Link to Artiel [https://doi.org/10.1016/j.gca.2021.03.005]
1Earth and Planets Laboratory, Carnegie Institution for Science, Washington, DC
2Science and Technology Division, Corning Incorporated, Corning, NY
3Dept. of Earth and Planetary Sciences, Northwestern University, Evanston, IL
4Dept. of Geophysical Sciences, University of Chicago, Chicago, IL
The partial pressures and isotopic compositions of volatiles present during chondrule formation can be constrained by the highly volatile element or HVE (H, C, F, Cl, and S) abundances and isotopic compositions in chondrules. Here we present the results of high spatial resolution and low background secondary ion mass spectroscopy (SIMS) analyses of the HVE concentrations and H isotopic compositions in type I and II chondrules in primitive ordinary chondrites Semarkona (LL3.00) and Queen Alexandra Range (QUE) 97008 (L3.05), and the primitive carbonaceous chondrite Dominion Range (DOM) 08006 (CO3.00). The HVEs in the chondrules primarily reside in the mesostases, in which the HVE contents and H isotopic compositions vary significantly (H2O: 8–10,200 ppm, CO2: 2.4–1170 ppm, F: 0.3–30 ppm, Cl: 0.07–175 ppm, S: 0.38–4400 ppm, δD: 77–15,000‰). To dissolve such HVE contents in a silicate melt requires significantly higher total pressures (up to 1900 bars), and in some cases requires anomalous gas compositions (CO dominated), compared to those expected from canonical conditions of chondrule formation (∼10-3 bars, H2+H2O dominated). Rather, the enrichments of H2O, CO2, Cl, and F in the mesostases at the edges of some chondrules suggest that there were secondary influxes of HVEs into the chondrule mesostases from the surrounding matrix during parent body processes. Consistent with this, melt inclusions sealed in olivine phenocrysts have significantly lower HVE contents than the mesostases in contact with the surrounding matrix material. Further, the calculated diffusion distances of H2O in silicate glasses under the relevant conditions are comparable to the radii of the chondrules. The high δD values in the mesostases could have been generated through isotopic Rayleigh fractionation as a result of the loss of very D-poor H2 generated from Fe metal oxidation by H2O in the parent bodies. Based on these results, we hypothesize that the bulk of the HVEs in the chondrules are secondary in origin. However, a small portion of the HVEs in chondrules could be primary, as there are low but measurable amounts of HVEs in the melt inclusions that are sealed in phenocrysts. Further, measured S contents in some chondrule mesostases agree with those predicted in a sulfide saturated silicate melt based on an experimentally calibrated thermodynamic model. We constrain the upper limits of primary HVEs in the chondrules based on the lowest measured HVE contents to minimize the effects of the secondary influx of HVEs (type I H2O: 7–11 ppm, CO2: 0.3–0.6 ppm, F: 0.1–0.2 ppm, Cl: 0.01–0.03 ppm, S: 0.3–60 ppm, and type II H2O: 50–85 ppm, CO2: 0.4–3 ppm, F: 0.04–2 ppm, Cl: 0.04–2 ppm, S: 190–260 ppm).