[PPS10-09] Al−Mg systematics for partial melting of an Allende type B CAI, Golfball
Ca-Al-rich inclusions (CAIs) are the oldest objects formed in our Solar System [1]. Recent high-precision Al−Mg mineral isochron studies using secondary ion mass spectrometry (SIMS) revealed detailed distributions of initial 26Al/27Al values, (26Al/27Al)0, for individual CAIs in CV chondrites [e.g. 2−4]. Igneous and non-igneous CAIs show similar variations in (26Al/27Al)0, which range from ~5.2 to ~3.4 × 10−5 [4]. These variations in (26Al/27Al)0 suggest that CV CAI formation processes continued for at least ~0.4 Myr at the very beginning of the Solar System. However, at the same time, they also raise a possibility of heterogeneous distribution of 26Al in the CAI-forming region [e.g. 3, 5]. In this study, we present Al−Mg systematics of relict minerals and later-crystallized minerals from multiple partial melting events in an Allende type B CAI, Golfball [6−8]. Such an approach would enable to constrain age differences between multiple melting events, regardless of the possibility of heterogeneous distribution of 26Al.
The Golfball CAI has a type B CAI bulk composition [6] and a unique structure [7]: a fassaite-rich mantle enclosing a melilite-rich core. Most of melilite crystals (Åk30−70) occur as laths at the inclusion rim and in the core or as blocky shapes poikilitically enclosed in the fassaite grains in the core. In the core, gehlenitic melilite grains (Åk5−12) are enclosed in strongly zoned (Åk15−70) overgrowths with small compositional discontinuity at the boundary. These gehlenitic melilite grains (Åk5−12) clearly could not have formed from a melt with bulk composition of Golfball or that of its core [6, 9], indicating they are relict grains that survived later melting event(s). High-precision in situ Al−Mg systematics with SIMS [3] was applied for the gehlenitic relict grains and the other melilite crystals as well as spinel, a liquidus phase. An Al−Mg isochron regression line for the data of gehlenitic relict grains and spinel gives (26Al/27Al)0 = (4.41 ± 0.22) × 10−5, while that for the other melilite crystals gives (26Al/27Al)0 = (4.41 ± 0.20) × 10−5. These results indicate that partial melting event(s) for the Golfball CAI occurred in very short order, probably shorter than a few tens of Kyr, after a precursor formation.
References: [1] Connelly et al. (2012) Science 338, 651−655. [2] MacPherson et al. (2012) EPSL 331−332, 43−54. [3] Kawasaki et al. (2019) EPSL 511, 25−35. [4] Kawasaki et al., under review. [5] Bollard et al. GCA 260, 62−83. [6] Simon and Grossman (2004) GCA 68, 4237−4248. [7] Simon et al. (2005) MaPS 40, 461−475. [8] Itoh et al. (2009) MaPS 40, Suppl., A116. [9] Beckett et al. (1999) 30th LPSC #1920.
The Golfball CAI has a type B CAI bulk composition [6] and a unique structure [7]: a fassaite-rich mantle enclosing a melilite-rich core. Most of melilite crystals (Åk30−70) occur as laths at the inclusion rim and in the core or as blocky shapes poikilitically enclosed in the fassaite grains in the core. In the core, gehlenitic melilite grains (Åk5−12) are enclosed in strongly zoned (Åk15−70) overgrowths with small compositional discontinuity at the boundary. These gehlenitic melilite grains (Åk5−12) clearly could not have formed from a melt with bulk composition of Golfball or that of its core [6, 9], indicating they are relict grains that survived later melting event(s). High-precision in situ Al−Mg systematics with SIMS [3] was applied for the gehlenitic relict grains and the other melilite crystals as well as spinel, a liquidus phase. An Al−Mg isochron regression line for the data of gehlenitic relict grains and spinel gives (26Al/27Al)0 = (4.41 ± 0.22) × 10−5, while that for the other melilite crystals gives (26Al/27Al)0 = (4.41 ± 0.20) × 10−5. These results indicate that partial melting event(s) for the Golfball CAI occurred in very short order, probably shorter than a few tens of Kyr, after a precursor formation.
References: [1] Connelly et al. (2012) Science 338, 651−655. [2] MacPherson et al. (2012) EPSL 331−332, 43−54. [3] Kawasaki et al. (2019) EPSL 511, 25−35. [4] Kawasaki et al., under review. [5] Bollard et al. GCA 260, 62−83. [6] Simon and Grossman (2004) GCA 68, 4237−4248. [7] Simon et al. (2005) MaPS 40, 461−475. [8] Itoh et al. (2009) MaPS 40, Suppl., A116. [9] Beckett et al. (1999) 30th LPSC #1920.