Isotopic Compositions and Origins of O-Anomalous Grains
As in other presolar grain-rich primitive meteorites, most of the O-anomalous grains in Adelaide belong to group 1 (Fig. 1). These grains probably originated in O-rich low-mass (approximately 1.2–2.2 M⊙) red giant or asymptotic giant branch (AGB) stars with close to solar metallicity (Nittler et al. 1997, 2008); the first dredge-up after hydrogen burning increases the 17O/16O ratio in the stellar envelope where grains form, but has little effect on the 18O/16O ratio, which is largely determined by the initial metallicity of the star. The high abundance of group 1 grains is in good agreement with recent modeling work carried out by Gail et al. (2009), which shows that a large fraction of the O-rich dust contributed to our solar nebula originated in low-mass thermally pulsing AGB stars of close to solar metallicity. Their model also predicts that about half of the O-rich dust should come from more massive (approximately 4–8 M€) AGB stars, but grains with the isotopic signatures of hot bottom burning expected from such stars (Iliadis et al. 2008) are not present in the presolar grain population. At present there is no satisfactory resolution to this discrepancy. Gail et al. (2009) suggested nonhomogeneous distribution of stellar grains to the solar system as a possibility, whereas Nittler (2009) noted that the amount of hot bottom burning predicted to occur in intermediate mass stars of solar metallicity may be overestimated.
Dredge-up cannot account for the isotopic compositions of group 1 grains with very high 17O/16O ratios. Calculations by Boothroyd and Sackmann (1999) indicated a limit on the 17O/16O ratio of approximately 4 × 10−3 for low to intermediate mass stars of close-to-solar metallicity. However, recent calculations using a revised estimate of the solar metallicity and new reaction rates that are important for the CNO cycle suggest somewhat higher limits of approximately 6 × 10−3 (Gyngard et al. 2011). One of our grains (5d-17-o1) has a 17O/16O ratio of 33.8 × 10−4. Dilution from surrounding isotopically normal grains is known to have a significant effect on the isotopic compositions of presolar grains identified through raster ion imaging (Nguyen et al. 2007), and the effect is most pronounced for the smallest grains. Grain 5d-17-o1 has a diameter of only 180 nm and its true isotopic composition is almost certainly above the first dredge-up limit; calculations by Nguyen et al. (2007) suggest dilution by a factor of 2–3 for a grain of this size, which would imply a true isotopic composition of approximately 7–10 × 10−3. Vollmer et al. (2008) also identified several “extreme” group 1 grains with isotopic compositions similar to our grain. Nittler and Hoppe (2005) suggested that grains with very high 17O/16O ratios could have a nova origin. Early model predictions for novae of various stellar masses (e.g., José et al. 2004) were consistent with the O isotopic compositions of extremely 17O-enriched grains such as T54 (Nittler et al. 1997). However, updated calculations with new reaction rates indicate that the predicted O isotopic compositions do not match the grain data unless large amounts of mixing with solar composition matter is included (Gyngard et al. 2010, 2011). Isotopic data for other elements (e.g., Si, Mg) in a few extreme group 1 grains indicate similar large discrepancies (Vollmer et al. 2008; Gyngard et al. 2010, 2011). An alternate possibility is that mass loss to a binary companion could diminish the envelope of a main sequence star, with the result that the 17O brought to the envelope during the first dredge-up is less diluted, leading to more elevated 17O/16O ratios than normally expected from this process (Nittler et al. 2008). Clearly, additional data and modeling efforts are needed to understand the origin of these grains.
One group 3 grain and one group 4 grain are also present in our inventory. The group 3 grain is depleted in 17O, with solar 18O/16O and plots below the GCE line, whereas the group 4 grain is enriched in 18O, with solar 17O/16O (Fig. 1). The isotopic compositions of group 3 grains, which plot below the GCE line, cannot be explained by dredge-up in AGB stars unless they originated in stars with lower than solar 17O/18O ratios, which is unlikely since the solar 17O/18O ratio is already low relative to typical molecular clouds in the Milky Way (Wilson and Rood 1994). Nittler et al. (2008) suggested that such group 3 grains, along with 18O-rich group 4 grains, could have a supernova origin, as their isotopic compositions fall along a mixing line of 16O-rich material from the supernova interior and material from the He/C and He/N zones, and the H envelope. These authors noted, furthermore, that the fact that most group 4 (and some group 3) grains are broadly consistent with a single mixing line could indicate their origin from a single nearby supernova, possibly one triggering collapse of the molecular cloud from which the solar system originated. Group 4 grains normally make up about 10% of the presolar silicate and oxide population. However, in Adelaide the single group 4 grain represents only 2% of the O-anomalous grains identified. Yada et al. (2008) noted that group 4 grains were significantly more abundant in IDPs and Antarctic micrometeorites (approximately 35%) than in the overall presolar silicate and oxide population. If these grains were injected into the early solar nebula by a supernova, such differences could indicate formation of the respective parent bodies at different times or in different parts of the nebula (Yada et al. 2008).
Isotopic Compositions and Origins of C-Anomalous Grains
The majority of the C-anomalous grains in Adelaide have 12C/13C ratios between 10 and 100 and are likely to be mainstream SiC grains (although in the absence of additional isotopic data we cannot rule out the possibility that some of them are X or Z grains); the origin of these grains in C-rich AGB stars is well established, although their Si isotopic compositions remain unexplained (e.g., Zinner 2007). Our inventory also includes two A + B grains (Table 4), defined as those grains with 12C/13C ratios of less than 10 (Amari et al. 2001). These grains make up about 4–5% of the SiC population, but remain poorly understood. J-type carbon stars with low 12C/13C ratios are one likely source; A + B grains with enhanced abundances of s-process grains may also come from born-again AGB stars, such as Sakurai’s object (Amari et al. 2001). However, the large spread in 14N/15N ratios exhibited by these grains is not well understood (Zinner 2007).
Finally, the elemental compositions of four C-anomalous grains could not be determined; one of these is probably a mainstream grain, based on its 12C/13C ratio. The isotopic compositions of the other three grains, with elevated 12C/13C ratios, are consistent with X or Y grains. However, these groups each make up only about 1% of the SiC population, and it seems unlikely that we would find three such grains in this limited inventory of C-anomalous grains. Graphite grains have higher than solar 12C/13C ratios, but abundances of presolar graphite are lower than most other presolar grain types (Zinner 2007) and, therefore, like X or Y grains, we would not expect to find many graphite grains in our C-anomalous grain inventory. A final possibility is that these are carbonaceous grains of interstellar or cold molecular cloud origin, similar to grains with 13C depletions that have been found in some primitive meteorites and IDPs (Floss et al. 2004, 2006, 2010; Floss and Stadermann 2009c). However, these phases appear to be highly susceptible to thermal processing and are only found in the most primitive extraterrestrial materials. As discussed in more detail below, Adelaide appears to have experienced some metamorphism, which such grains are unlikely to have survived. With no additional elemental or isotopic information, it is difficult to say more about the nature of these three C-anomalous grains.
Complex Grains 3a-1-o2 and 5a-1-o2
The isotopic compositions of both complex grains (group 1) indicate that they formed in low-mass red giant or AGB stars with close to solar metallicity. Refractory Al-bearing oxides, such as the ones in 3a-1-o2 and 5a-1-o2, tend to form in stars with low-mass loss rates in the early AGB phase (e.g., Sloan et al. 2003; Maldoni et al. 2008) and are also the first solids expected to condense from an O-rich circumstellar gas (Lodders and Fegley 1999). Fe,Mg-oxides, like those present in 3a-1-o2, are also expected to form in O-rich AGB stars with low-mass loss rates (Gail and Sedlmayr 1999; Ferrarotti and Gail 2001). At higher mass-loss rates, silicate condensation is dominant, and olivine is the first phase to form (Ferrarotti and Gail 2001). As temperatures decrease following the initial condensation of corundum, this phase can react with the gas to form other phases, such as grossite, hibonite, or Al-bearing silicates, or can act as a nucleation site for the subsequent condensation of ferromagnesian silicates (Gail and Sedlmayr 1999; Demyk et al. 2000; Toppani et al. 2006).
Grain 5a-1-o2 consists of a forsterite-like silicate and a hibonite-like grain. Although the hibonite-like grain probably formed before the silicate, it is not clear from the texture of this compound grain (Fig. 3) whether it acted as a nucleation site for the silicate, or whether the two subgrains formed independently. The homogeneous O isotopic composition of the compound grain does argue for aggregation of the two subgrains in the same stellar environment, and indicates that, despite the difference in elemental compositions, there was no substantial change in the O isotopic composition of the gas during condensation of these components. Grain 3a-1-o2 is an aggregate of Al2O3 and CaAl4O7 (grossite), together with Fe,Mg-bearing oxides (Fig. 5). The CaAl4O7 partially surrounds the Al2O3 and may have formed by reaction of this grain with the stellar gas as temperatures dropped (Lodders and Fegley 1999). The Fe,Mg-oxides, in turn, partially surround the CaAl4O7, forming a sequence of increasingly less refractory grains from the upper right to the lower left of the grain as shown in Fig. 5. This complex grain shows heterogeneity in its O isotopic composition, with less anomalous O in the Fe,Mg-oxides than in the Al-bearing oxides. Despite this heterogeneity, it is likely that all the sub-components also formed in the same stellar environment from a gas with a uniform O isotopic composition. As discussed more extensively below, the presence of Fe-rich rims around both of these complex grains suggests infiltration of Fe into the grains due to secondary alteration. The less anomalous O isotopic compositions of the Fe,Mg-oxides are probably due to partial re-equilibration of the grains with isotopically normal O from the surrounding matrix during this process. Dilution of the most anomalous region (area 1 in Table 1) in complex grain 3a-1-o2 with 45% solar composition oxygen can reproduce the O isotopic compositions of the least anomalous region (area 3 in Table 1); 33% dilution is required to reproduce the O isotopic composition of area 2.
Presolar Grain Abundances
Kakangari, as noted above, has upper limits of approximately 4–5 ppm for both SiC and presolar silicates and oxides. For Adelaide, the calculated abundance of all O-anomalous grains in our inventory is 70 ± 10 ppm; the oxide abundance is 17 ± 6 ppm and the silicate abundance is 53 ± 8 ppm. Davidson et al. (2010) obtained a significantly higher abundance of about 250 ppm for O-anomalous grains in Adelaide. However, their estimate was based on only seven grains in 2600 μm2 area; as discussed below, presolar grain abundances in this meteorite are very heterogeneous from one matrix area to another and the grains identified by Davidson et al. (2010) appear to be from a presolar grain-rich matrix area. The data from this study, with a higher number of grains identified and a much larger area measured, are more representative of Adelaide as a whole. The abundance we obtain for Adelaide is lower by a factor of 2–3 than the presolar silicate and oxide abundances of other highly primitive carbonaceous chondrites, which are generally between 150 and 200 ppm (Fig. 6). Abundances in IDPs are even higher, with minimum values of about 375 ppm for the most primitive IDPs (Messenger et al. 2003; Floss et al. 2006; Busemann et al. 2009). The CR2 chondrite NWA 852 also has relatively low O-anomalous grain abundances, approximately 75 ppm (Fig. 6). Leitner et al. (2012) suggested that NWA 852 may have lost some presolar grains due to the aqueous alteration it has experienced, and that this meteorite might represent a link between nearly pristine CR3 chondrites like QUE 99177 and MET 00426 (Floss and Stadermann 2009a) and other CR chondrites, which have lower presolar grain abundances (e.g., Nagashima et al. 2004). Floss and Stadermann (2009a) suggested that the presolar silicate/oxide ratio may be a measure of the degree of alteration experienced by a meteorite, with high silicate/oxide ratios indicative of less processing since oxide grains are expected to be more resistant to secondary processes than silicate grains. This ratio is quite variable in the presolar grain-rich meteorites that have been analyzed to date, with values up to about 22 for IDPs and the CR3 chondrites (Floss and Stadermann 2009a). Leitner et al. (2012) calculated a theoretical value of about 23 for the silicate/oxide ratio of dust from AGB stars, suggesting that the presolar silicate and oxide abundances observed in the CR3 chondrites and IDPs reflect their initial abundances in the regions of the solar nebula where the parent bodies of these samples formed. NWA 852, however, has a very low silicate/oxide ratio of about 2, consistent with destruction of some silicates through aqueous alteration (Leitner et al. 2012). The presolar silicate/oxide ratio of Adelaide is also very low, about 3; this together with the lower abundance of O-anomalous grains, suggests that secondary processing has destroyed some fraction of the presolar silicate inventory of Adelaide.
Our calculated SiC abundance in Adelaide is 8 ± 3 ppm; this estimate includes only those grains that were positively identified as SiC (Table 4). Davidson et al. (2010) determined an abundance of 23 ppm, based on a single grain; within errors this is consistent with our estimate. This abundance is within the range of SiC abundances determined for a series of carbonaceous chondrites based on noble gas measurements (Huss et al. 2003), but is about an order of magnitude lower than recent estimates based on NanoSIMS ion imaging measurements (up to approximately 100 ppm; Floss and Stadermann 2009c; Bose et al. 2012; Leitner et al. 2012). Presolar SiC is more resistant to secondary processing than some other presolar phases, particularly silicates. For example, Davidson et al. (2009) showed that SiC abundances do not vary significantly with the degree of aqueous alteration in CR chondrites, whereas presolar silicate abundances are affected (Nagashima et al. 2004; Leitner et al. 2012). However, Huss et al. (2003) argued that thermal metamorphism does affect SiC abundances and have used differences in abundance to estimate the degree of nebular processing experienced by different chondrite groups. If this is the case, then the low SiC abundance of Adelaide suggests that it has experienced significant thermal processing. As noted above, NanoSIMS ion imaging studies, for the most part, indicate higher SiC abundances than the noble gas studies of Huss and Lewis (1995) and Huss et al. (2003); however, these investigations have focused only on meteorites of low petrologic type. Similar NanoSIMS measurements on meteorites of higher petrologic type could help determine whether SiC is, indeed, destroyed at higher temperatures or has simply lost some of its noble gas inventory, as suggested by Davidson et al. (2009).
Presolar grain abundances are also highly variable within Adelaide; three different matrix regions (7, 3, and 5) have O-anomalous grain abundances that vary from about 30 ppm (one grain, 8000 μm2) to about 95 ppm (20 grains, 7300 μm2) to about 175 ppm (24 grains, 6400 μm2), respectively. Areas 3 and 5, with the highest abundances of O-anomalous grains, also contain the most C-anomalous grains (five grains each). In other matrix regions the number of grains identified is low, but the amount of area measured (≤∼2000 μm2) is not sufficient to provide statistically meaningful abundance estimates. Examination of the matrix areas clearly shows that the difference in presolar grain abundances is correlated with grain size (Fig. 7). The matrix in area 5 is fine-grained, with grain sizes typically one micrometer or less, whereas area 7 is significantly more coarse grained. Similar heterogeneities in presolar grain abundances have been observed in other meteorites, for example QUE 99177 (Floss and Stadermann 2009a), but the difference is more pronounced in Adelaide. Moreover, unlike Adelaide, the matrix material in QUE 99177 did not show any obvious petrographic differences between matrix areas with different presolar grain abundances. The coarser grain size of the matrix from area 7 suggests that this material has experienced recrystallization and grain growth, probably in response to elevated temperatures. This process is likely also responsible for the low presolar grain abundance in this area; the single presolar grain that was identified is an oxide grain and is the largest presolar grain found in our study (Table 1). These features probably account for its survival relative to presolar silicates, which are generally smaller and more labile. The ungrouped C chondrite Ningqiang also shows a strongly heterogeneous distribution of O-anomalous grains, with larger average grain sizes in the matrix areas that are presolar grain-depleted (Zhao et al. 2011b).
Figure 7. Reflected light images of matrix areas 7 and 5 in Adelaide (A5 and A7) and matrix area 1 in Kakangari (K1). Note that the scale is the same in all three images, but that the NanoSIMS raster imaging areas (dark squares) are 10 × 10 μm2 in Adelaide and 20 × 20 μm2 in Kakangari. Area 7 in Adelaide has a coarser grain size and significantly lower presolar grain abundances than area 5; grain size is coarsest in the matrix of Kakangari.
Download figure to PowerPoint
Elevated Fe Abundances in Adelaide Presolar Silicates
One of the early surprises after the initial discovery of presolar silicates was that the vast majority of these grains had higher Fe contents than expected based on equilibrium condensation calculations (e.g., Lodders and Fegley 1999) and astronomical observations (e.g., Waters et al. 1996; Demyk et al. 2000), which predicted Mg-rich silicates. Whether these high Fe contents are a primary signature of the presolar grains or the result of secondary alteration has been extensively discussed in the literature (e.g., Floss and Stadermann 2009a; Vollmer et al. 2009; Bose et al. 2010a; Nguyen et al. 2010). Although secondary alteration may have played some role (e.g., Bose et al. 2010a), the ubiquitous presence of Fe-rich presolar silicates in meteorites that have experienced very little secondary processing (e.g., Floss and Stadermann 2009a; Nguyen et al. 2010) argues for a primary origin for much of the Fe in these grains. Iron isotopic measurements can provide a definitive test and, indeed, several presolar grains from Acfer 094 have nonsolar Fe isotopic compositions (Mostefaoui and Hoppe 2004; Floss et al. 2008; Vollmer and Hoppe 2010; Ong et al. 2012), indicating primary Fe in these grains. A number of other presolar silicates have solar Fe isotopic ratios (e.g., Vollmer and Hoppe 2010; Ong et al. 2012), but the significance of these results is not clear, as the measurements are difficult to carry out and errors are typically large.
Figure 8 shows the range of Fe contents in presolar silicates from different primitive chondrites; abundances typically range up to about 30 atom%. The presolar silicates in Adelaide, however, have even higher Fe contents, with abundances up to about 40–45 atom%. Presolar silicates in Acfer 094 are also very Fe-rich, with a range greater than those of Adelaide. However, the median Fe content of Adelaide is significantly higher (26 atom%) than that of Acfer 094 (14 atom%) or the other primitive carbonaceous chondrites (12–13 atom%). In addition, Fig. 2 shows that in Acfer 094 most of the grains have compositions that are similar to either olivine or pyroxene, or fall between the two; this is also true for the presolar silicates in most other meteorites (e.g., Floss and Stadermann 2009a; Vollmer et al. 2009; Bose et al. 2010b; Nguyen et al. 2010). The situation is very different for the presolar silicates in Adelaide: four grains have compositions similar to olivine and one grain has a composition similar to pyroxene, but all other grains in Adelaide have (Mg + Fe)/Si ratios that are significantly higher than that of olivine (Fig. 2). Cation/O ratios are also significantly elevated in many of the presolar silicates from Adelaide (0.66–1.35, with most >0.80; Table 2) compared to a range of 0.59–0.97 for presolar silicates from other primitive meteorites (e.g., Floss and Stadermann 2009a; Vollmer et al. 2009).
Moreover, as we noted above, the two complex grains found in Adelaide have Fe-rich rims that seem to suggest diffusion of Fe into the grains. We see similar rims around some of the individual presolar grains in Adelaide. Figure 9 shows two such presolar silicates with clear rims of Fe penetrating the grains, compared with a presolar silicate from the MET 00426 CR3 chondrite, where no such rim is present. Taken together, our observations suggest that at least some fraction of the Fe in the presolar silicates from Adelaide is due to secondary alteration. The Fe-rich rims indicate infusion of Fe into the grains from the surrounding matrix, resulting in elevated Fe contents and, consequently, elevated Mg + Fe/Si and cation/O ratios compared to most presolar silicates from other primitive chondrites. Diffusion, strictly defined, is an exchange process, but in these grains we see a clear net addition of Fe into the grains, suggesting a kinetic process rather than equilibrium diffusion. Although we do not have structural information on these grains, it is likely that many of them are amorphous and this may also facilitate infusion of Fe into the grains.
Figure 9. False-color Mg and Fe distribution maps of presolar silicates from Adelaide and the MET 00426 CR3 chondrite.
Download figure to PowerPoint
Secondary Processing in Adelaide and Kakangari
Our study shows that both Adelaide and Kakangari have lower presolar grain abundances than Acfer 094 and ALHA77307 and, therefore, seem to be less primitive. The matrix of Kakangari is Mg-rich and is dominated by structurally and mineralogically distinct aggregates of crystalline enstatite and olivine, with lesser amounts of albite, anorthite, spinel, troilite, and Fe,Ni metal; amorphous silicates are essentially absent (Brearley 1989). Matrix textures and enstatite microstructures indicate high-temperature (>1000 °C) processing with rapid cooling, but melting does not seem to have occurred. Moreover, Berlin et al. (2007) showed that chondrules as well as other components in Kakangari have experienced wide-spread reduction, which probably took place both in the nebula at the time of chondrule formation and later on the parent body. The matrix mineralogy of Kakangari probably formed through high-temperature reaction of fine-grained amorphous or partially crystalline dust, in a thermal event similar to the one that formed chondrules. Such processing will destroy presolar silicates, as well as other presolar grains (e.g., Huss and Lewis 1995; Huss et al. 2003), accounting for the paucity of presolar grains in this meteorite.
Although ungrouped, Adelaide shows similarities to the CO3 chondrite ALHA77307, such as matrix abundance, mineralogy, and chondrule size; both meteorites are also highly un-equilibrated (Brearley 1991). However, there are also important differences. Olivines in the matrix of Adelaide are significantly more fayalitic than the Mg-rich ones typically seen in ALHA77307 (Brearley 1993) and bulk analyses of the Adelaide matrix show that it is more Fe-rich than chondrite matrices in general. In addition, although both meteorites contain amorphous material in their matrices, the matrix of Adelaide is more crystalline than that of ALHA77307, suggesting that it has experienced some thermal annealing (Brearley 1991). Adelaide is also highly weathered, with extensive replacement of metal and sulfides by goethite (Davy et al. 1978). Terrestrial weathering has been shown to remove Mg and Si from ordinary chondrites, but Fe abundances appear to be unaffected (Bland et al. 1998). In contrast thermal metamorphism can result in increased Fe contents in originally Mg-rich grains (Jones and Rubie 1991). Annealing of Adelaide matrix material is, therefore, probably responsible for the elevated Fe contents in its presolar silicates, with Fe permeating into the grains from the surrounding Fe-rich matrix. This process does not require that all Fe in the presolar silicates is secondary. Indeed, as noted earlier, the fact that Fe contents are high in presolar silicates from all primitive meteorites argues for incorporation of at least some Fe into the grains during formation in their stellar environments (Floss and Stadermann 2009a; Vollmer et al. 2009; Bose et al. 2010a, 2012; Nguyen et al. 2010).
Lower presolar grain abundances in Adelaide compared to other primitive carbonaceous chondrites (Fig. 6) and dilution of the O isotopes in some parts of complex grain 3a-1-o2 (Fig. 4) are also consistent with thermal processing. However, if the lower presolar grain abundances are due to parent body thermal metamorphism, we would expect to see uniform depletions throughout the meteorite. Instead, as noted earlier, the distributions of both O-anomalous and C-anomalous presolar grains are highly heterogeneous among different matrix areas in Adelaide, with most of the presolar grains present in only two of the nine matrix areas studied. One possibility is that the extensive terrestrial weathering experienced by Adelaide has destroyed some of the presolar silicates. This type of alteration would be consistent with the heterogeneous distribution of these grains, as weathering is likely to act variably in different parts of the meteorite. However, presolar SiC abundances also differ among the different matrix areas, but are probably resistant to terrestrial alteration, as these grains survive even the harsh acid treatments traditionally used to separate them from their host meteorites (Amari et al. 1994). Moreover, there are no obvious petrographic features in our thin section indicative of localized weathering, such as the preferential occurrence of alteration veins near the presolar grain-poor matrix areas.
Instead, as noted earlier, the primary difference between matrix areas rich in presolar grains and those poor in presolar grains is grain size (Fig. 7). Areas poor in presolar grains have a coarser average grain size than areas rich in presolar grains, probably as a result of recrystallization and grain growth during heating. The close proximity of matrix areas with different grain sizes, and different presolar grain abundances, suggests that this thermal processing was highly variable. Short-lived high temperature events, like those thought to be responsible for the formation of chondrules (Bally et al. 2005), will also process fine-grained dust in the nebula (Nuth et al. 2005). Such events may be highly localized, with different effects on different parcels of gas and dust, leading to grain coarsening and the destruction of presolar grains in some material, but not in others. Brearley (1991) noted that most matrix components in ALHA77307 are also found in Adelaide, only in very different modal abundances, and suggested that sampling of poorly mixed material could account for the observed differences. Similar accretion of materials processed to different degrees may be responsible for the variability observed in Adelaide matrix material. Based on its paucity of presolar grains (one oxide grain in 8000 μm2 of area 7 measured), the degree of heating and recrystallization experienced by the most coarse-grained matrix area in Adelaide approaches that of the matrix material in Kakangari, in which no presolar grains were found; comparison of the matrix in Adelaide with that in Kakangari supports this suggestion (Fig. 7).
A similar process of short-lived high temperature events may also be responsible for infiltration of Fe from the surrounding matrix, as observed in the presolar silicates and complex grains. However, this would require that heating occurred after aggregation of millimeter- to centimeter-sized clumps of material in order for Fe to enter the presolar grains from surrounding Fe-rich material. Such an environment would be highly oxidizing, due to the local enhancement of dust-to-gas ratios and could be responsible for the overall Fe-rich nature of Adelaide matrix material compared to other chondrites.
Based on the differences between Adelaide and Kakangari, and other primitive chondrites with more presolar grains, it is clear that one of the defining characteristics common to primitive meteorites with high presolar grain abundances is the presence of amorphous silicates. Both Acfer 094 and ALHA77307 are highly un-equilibrated and contain abundant fine-grained Fe- and Si-rich amorphous material in their matrices (Brearley 1993; Greshake 1997). The CR3s QUE 99177 and MET 00426 are very primitive members of the CR chondrite group; unlike other CRs, they contain few crystalline phyllosilicates and their matrices are dominated by fine-grained amorphous silicates (Abreu and Brearley 2010). The new CO3 chondrite LAP 031117 and the CR chondrite GRV 021710 have similarly high presolar grain abundances (Haenecour and Floss 2011; Zhao et al. 2011a), but to date there have not been any detailed investigations of their matric mineralogies. Chizmadia and Cabret-Lebrón (2009) have suggested, on the basis of olivine compositions, that LAP 031117 is very primitive, like ALHA77307. GRV 021710 was classified as a CR2 chondrite (Russell et al. 2005), but a type 3 classification may be more appropriate.