Isotopic Constraints on Earth System Processes. Группа авторов

diffused out of an initially homogeneous grain. However, the associated lithium isotopic fractionation calculated using the relative diffusivity of ⁶Li and ⁷Li from the laboratory experiments (i.e., β _Li = 0.27) resulted in a profile with a shape and amplitude that is nothing like the measured data. The lack of significant isotopic fractionation across this grain suggests that the zoning was most likely the result of crystallization from a melt of evolving composition. Had it not been for the isotopic data, one might have incorrectly assumed that the lithium zoning was due to diffusion and used it to make a spurious calculation of the cooling rate of the host rock. In fact, the cooling rate of the host rock must have been extraordinarily fast (> 100°C/hr) as there was not enough time for any significant diffusion of lithium down the concentration gradient.

Figure 1.14 Results of a diffusion calculation for the evolution of the lithium abundance and isotopic fractionation of a grain with an initial 6 ppm lithium concentration (dashed line) and 1 ppm at each edge. The panel on the left shows the calculated lithium concentration profile that developed after a time such that it resulted in a reasonably good fit to the measured lithium concentration data shown as unfilled circles. The panel on the right compares the measured lithium isotopic fractionation across the grain (black circles with two sigma error bars) with the lithium isotopic fractionation calculated assuming δ⁷Li = 0‰ as the initial and boundary condition and with the relative diffusion coefficients of the lithium isotopes calculated using β_Li = 0.27.

      1.5.3. Lithium Isotopic Fractionation by Diffusion in Olivine

      Richter et al. (2017) ran laboratory experiments similar to those discussed in the previous section whose purpose was to document isotopic fractionation as lithium diffused into olivine. They found that lithium diffusion in olivine also resulted in large isotopic fractionations that were fit by model calculations with β _Li = 0.4 ± 0.1. Richter et al. (2017) used the same model calculations to fit lithium concentration and isotopic fractionation data that had been measured by Xiao et al. (2015) across a natural olivine grain from a peridotite xenolith from the Eastern North China Craton. The Xiao et al. (2015) lithium isotopic fractionation data can be fit using β _Li = 0.30 when lithium was assumed to occupy two sites in the olivine or with β _Li = 0.36 for lithium in just one site. The Xiao et al. (2015) data were not sufficiently detailed to distinguish whether lithium was in one or two sites, but the fact that the values β _Li that fit the isotope data in both cases are quite similar to those of the laboratory experiments is evidence that the diffusion of lithium did have a significant role in producing the zoning of the olivine grain studied by Xiao et al. (2015).

      1.5.4. Fe‐Mg zoning and Fe and Mg Isotopic Fractionation in Olivine

Sio et al. (2013) reported iron and magnesium isotopic fractionations in a zoned olivine phenocryst from Kilauea Iki Lava Lake in Hawaii. An important aspect of this study was that it demonstrated that in the case of olivine, where the diffusive flux of one element (e.g., Mg) is balanced by an equal and opposite flux of another element (e.g., Fe), the isotopic fractionations are anti‐correlated. Two years later, Oeser et al. (2015) used femtosecond laser ablation and a multi‐collector inductively coupled plasma mass spectrometer to make a detailed study of the iron and magnesium isotopic fractionations across a number of zoned olivine xenoliths and phenocryst from various continental volcanic settings. Fig. 1.15 shows an example from their study of the Fe–Mg zoning and the anti‐correlated isotopic fractionation of iron and magnesium. Oeser et al. (2015) reported average values of β _Mg = 0.08 ± 0.01 and β _Fe = 0.14 ± 0.05 based on model calculations that fit the data from six olivine grains that showed the expected correspondence between the Fe–Mg zoning and the anti‐correlated iron and magnesium isotopic fractionations. Richter et al. (2016) reported similar magnesium isotopic fractionations in three olivine grains from Martian meteorite NWA 817, which were fit with β _Mg = 0.13, 0.10, and 0.08, respectively. A subsequent experimental study of Fe–Mg exchange in olivine by Sio et al. (2018) determined a value of β _Fe = 0.09 ± 0.05.

Figure 1.15 Chemical concentration (open circles) and isotopic fractionation (black circles with ±2 sigma error bars when larger than the symbols) measured across one edge of an olivine grain from the Massif Central, France. Note the anti‐correlation of the iron and magnesium isotopic fractionations between 0 and 400 μm.

      These data are from Oeser et al. (2015).

      The examples described in this section highlight the role of isotopic data as a “fingerprint” of the extent of diffusive mass transport in zoned igneous minerals pyroxene and olivine. The role of the laboratory experiments is to calibrate the “fingerprint,” which can then be used to determine the extent that diffusion is responsible for a given instance of zoning of a natural pyroxene or olivine grain. This isotopic “fingerprint” is especially important when considering whether to use the mineral zoning to determine the thermal history of the host rock.

1.6. ISOTOPE FRACTIONATION BY EVAPORATION FROM SILICATE MELTS

The chemical and isotopic composition of meteorites has had a very prominent role in cosmochemistry in that it provided, among other things, the age of the solar system and the best estimate of the bulk composition of the solar system for all but the most volatile elements. Refractory calcium‐aluminum rich inclusions (CAIs), like the one shown in Fig. 1.16, are particularly important components of primitive chondritic meteorites in that they are the oldest dated materials to have formed in the solar system and, having been present at the creation, they are unique recorders of processes and conditions that prevailed in the early proto‐planetary solar nebula.

      A compelling qualitative narrative has been developed regarding the origin and evolution of the Type B CAIs. Type B CAIs, or their precursors, are condensates from a cooling gas of solar composition as evidenced by their being made up of four major minerals – spinel (MgAl₂O₄); melilite (a solid solution between gehlenite, Ca₂Al₂SiO₇, and åkermanite, Ca₂MgSi₂O₇); a Ca‐pyroxene (CaMgSi₂O₆); and anorthite (CaAl₂Si₂O₈) – that are predicted by thermodynamic calculations to be the early condensed minerals from a cooling solar‐composition gas (Grossman 1972). The thermodynamic calculations indicate that the materials that condensed at about 1125°C and became the precursors of the CAIs were solids. The obvious igneous texture of CAIs, like the one shown in Fig. 1.16, is evidence that at some point they must have been melted to a very high degree. In order for the Type B CAIs to have partially melted to the degree required to crystallize large euhedral melilite grains, they must have been reheated to about 1450°C (Stolper, 1982; Stolper & Paque, 1985). A very important characteristic of many CAIs is that they have distinctive oxygen, magnesium, and silicon isotopic compositions. The oxygen isotopic composition of the major minerals in CAIs fall along what appears to be a mixing line between a very ¹⁶O‐rich reservoir and a reservoir with oxygen isotopic composition close to that of Earth and other inner solar system materials. The origin