LOW ⁶⁰FE ABUNDANCE IN SEMARKONA AND SAHARA 99555

Chondrules are quenched droplets of magma that formed within a few million years of the formation of the Solar System (Kita & Ushikubo 2012; Scott & Krot 2014). After incorporation in chondrite parent-bodies, they were subjected to thermal metamorphism and aqueous alteration, processes that could disturb ⁶⁰Fe-⁶⁰Ni systematics, so care must be exercised in selecting the most pristine samples for study. Semarkona is a LL3.00 chondrite (Quirico et al. 2003; Grossman & Brearley 2005), meaning that it was minimally modified by parent-body aqueous alteration or metamorphism (the degree of aqueous alteration increases from type 3–1 while the degree of metamorphism increases from type 3–6) The Smithsonian Institution provided a fragment of Semarkona of approximately ∼370 mg, from which 14 chondrules were hand-picked under a binocular microscope In order to identify the chondrules with relatively high Fe/Ni ratios, small areas were polished using 1200 grit abrasive paper. The internal areas thus exposed allowed us to measure the chemical compositions of the chondrules using energy dispersive spectroscopy on a JEOL JSM-5800LV scanning electron microscope operated in low vacuum mode to prevent charging. The surfaces were not well polished and the samples were not carbon coated, so the chemical analyses were of limited quality but sufficient for our purposes. Out of the 14 starting chondrules, 6 had relatively high Fe contents and were selected for Ni isotopic analysis. The bulk chondrules (∼2–14 mg) were first rinsed with acetone to get rid of possible surface contamination. To avoid the risk of accidental sample loss, recover the maximum amount of Ni for high-precision isotopic analysis (a low ⁶⁰Fe/⁵⁶Fe ratio implies that ⁶⁰Ni-excess should be barely resolvable in chondrules), and preserve the bulk nature of the measurements, no further characterization or fragmentation was done on these samples, which were directly digested.

Angrites are a group of basaltic achondrites that record some early igneous activity (Mittlefehldt et al. 2002; Mittlefehldt 2003) According to their textural characteristics, angrites are divided into two subgroups. Fine-grained angrites, such as D'Orbigny and Sahara 99555, are characterized by quenched textures indicative of rapid cooling (Mittlefehldt et al. 2002). Coarser-grained angrites, such as Angra dos Reis and NWA 4801, experienced more protracted cooling histories (e.g., Nyquist et al. 2009; Kleine et al. 2012). High-resolution chronometers including ²⁶Al-²⁶Mg, ⁵³Mn-⁵³Cr, ¹⁸²Hf-¹⁸²W, and Pb-Pb have been applied to date angrites (Lugmair & Galer 1992; Nyquist et al. 1994, 2003; Lugmair & Shukolyukov 1998; Baker et al. 2005; Amelin 2007, 2008; Markowski et al. 2007; Connelly et al. 2008a, 2008b; Spivak-Birndorf et al. 2009), providing a means of testing the concordance between different extant and extinct radiochronometers. Due to its quenched texture and rapid cooling, Sahara 99555 is a well-suited anchor for early Solar System chronology. It is mainly composed of Ca-rich olivine (∼31–42%), Al-Ti rich pyroxene (∼24–28%), and anorthitic plagioclase (∼33–39%; Mikouchi et al. 2000; Mikouchi & McKay 2001). A 500 mg sample of Sahara 99555 purchased from L. Labenne was crushed into powder in an agate mortar. The fragmented samples were separated into two parts: one was processed with a hand magnet followed by sieving to separate the grains into three silicate size fractions (100–166 μm, 166–200 μm, and >200 μm); the other was processed with a hand magnet and split according to density (below or above 3.10 g cm⁻³) using sodium polytungstate solution. The mineral fractions were not characterized but Tang & Dauphas (2012) applied the same procedure to D'Orbigny angrite and the low-density fraction was relatively rich in anorthite while the high-density fraction was rich in olivine and pyroxene. For density separation, the fragmented sample was placed in the separatory funnel and heavy liquid was then added. The funnel was left for 10 minutes until dense grains (>3.10 g cm⁻³) sank to the bottom while light grains(<3.10 g cm⁻³) remained suspended in the liquid. Silicate fractions with different densities were collected onto pieces of weighing paper, rinsed with Millipore Milli-Q water, and dried in an oven.

To assess data quality and make sure that no analytical artifact was present, terrestrial standards were processed and analyzed together with the meteorite samples.

The chemistry was performed under clean laboratory conditions at the Origins Lab of the University of Chicago. Optima grade HF, reagent grade acetone, and double distilled HCl and HNO₃ were used for digestion and column chromatography. Millipore Milli-Q water was used for acid dilution.

Tang & Dauphas (2012, 2014) provide details on the procedures of sample digestion and chemical separation. Semarkona chondrules and Sahara 99555 fractions weighing 2–130 mg were digested in 5–20 ml HF-HNO₃ (in a 2:1 volume ratio) in a Teflon beaker placed on a hot plate at ∼90°C for 2 days. The solution was subsequently evaporated to dryness and re-dissolved in a 5–20 mL mixture of concentrated HCl-HNO₃ (in a 2:1 volume ratio) until all the sample powder was completely digested. The solutions were dried down and the residues taken back in solution with a minimum amount of concentrated HCl (∼11 M) for loading on the first column. In order to obtain sufficiently clean Ni for isotopic measurements, chemical separation of Ni from matrix elements and isobars was done in three steps of chromatography that are described below.

U/TEVA cartridge (Horwitz et al. 1992) was used for the first step of chemistry to get rid of Ti, Co, Zr and Fe. The column (2 mL volume, 2.5 cm length, 1 cm diameter) was pre-cleaned with 10 mL water, 15 mL 0.4 M HCl, 15 mL 4 M HCl and was then conditioned with 10 mL of concentrated HCl. The sample solution was loaded onto the column in 5–10 mL 10 M HCl. The load solution was collected in clean Teflon beakers and an additional 10 mL of concentrated HCl was passed through the resin and collected in the same beaker. This eluate contained Ni together with Na, Mg, Ca, and other matrix elements. After drying down, the Ni elution cut from the first column chemistry was re-dissolved in 5 mL of a mixture of 20% 10 M HCl-80% acetone (by volume) and loaded onto a Teflon column containing 5 mL (40 cm length, 0.4 cm diameter) of pre-cleaned Bio-Rad AG50-X12 200–400 mesh hydrogen-form resin, previously conditioned with 10 mL 20% 10 M HCl-80% acetone. After loading the sample solution and rinsing with 30 mL 20% 10 M HCl-80% acetone mixture to eliminate Cr and any remaining Fe, Ni was collected by eluting 150 mL of the HCl-acetone mixture into a jar containing 30 mL of water to dilute HCl and stabilize Ni in the eluate. In those conditions, Mg, Na, Ca, and other matrix elements were retained on the resin (Strelow et al. 1971) The collected Ni solution was evaporated at moderate temperature (<90°C) under a flow of N₂ to avoid the formation of organic complexes with acetone and accelerate evaporation. After evaporation, the Ni fraction was dissolved in 1 mL of aqua regia (1:3 HNO₃:HCl) to remove any organic residue formed during evaporation. This HCl-acetone column was repeated five times to ensure thorough separation of major rock forming element Mg from Ni, two elements that are notoriously difficult to separate. Zinc is a significant interference on low abundance isotope ⁶⁴Ni. It was removed using a third column filled with 1 mL (2 cm length, 0.8 cm diameter) Bio-Rad AG1W-X8 anionic ion exchange resin in 8 M HBr medium (Moynier et al. 2006). Nickel was eluted in 8 mL 8 M HBr, whereas Zn was retained on the resin.

The entire procedural blank was ∼35 ng, which is negligible compared to the amounts of Ni in the samples. The nickel yield of the entire procedure was 90–100%.

All measurements were performed at the Origins Laboratory of the University of Chicago using a Neptune MC-ICPMS equipped with an OnTool Booster 150 (Pfeiffer) interface jet pump. Jet sampler and X skimmer cones were used. The samples were re-dissolved in 0.3 M HNO₃ and introduced into the mass spectrometer with Ar + N₂ using an Aridus II desolvating nebulizer at an uptake rate of ∼100 μL minutes⁻¹. The instrument sensitivity for ⁵⁸Ni was 100 V/ppm. One analysis consisted of 25 cycles, each acquisition lasting for 8.4 s. During a session, each sample solution was measured 13 times bracketed by SRM 986. A small isobaric interference from the least abundant isotope of iron, ⁵⁸Fe, on the most abundant isotope of nickel, ⁵⁸Ni, was corrected by monitoring ⁵⁷Fe (the correction is always smaller than 0.5 $\varepsilon$ ). All isotopes were measured using Faraday cups with 10¹¹ Ω resistance amplifiers. Background was subtracted using an on-peak zero procedure. Internal normalization was used to correct mass-dependent isotopic fractionation by fixing ⁶¹Ni/⁵⁸Ni to 0.016720 or ⁶²Ni/⁵⁸Ni to 0.053389 (Gramlich et al. 1989) using the exponential law (Maréchal et al. 1999). Nickel-64 is reported only for samples that gave ∼15 μg Ni because below this level, an isobaric interference from $^{48}{\rm T}{{{\rm i}}^{16}}{{{\rm O}}^{+}}$ can affect the results (Tang & Dauphas 2012).

Approximately 20% of the original sample solutions were kept as safety aliquots and for Fe/Ni ratio measurements by MC-ICPMS using both standard bracketing and standard addition techniques. The Fe/Ni ratios measured by standard addition agree well with Fe/Ni ratios measured by simple sample-standard bracketing. Fe/Ni ratios in terrestrial standards were all within 3% of their reference values, demonstrating the accuracy of our measurements.

The nickel isotopic compositions of Semarkona chondrules (LL3.00) were analyzed and give an initial ⁶⁰Fe/⁵⁶Fe ratio of (539 ± 3.27) × 10⁻⁹ at the time of chondrule formation. Telus et al. (2013, 2014) measured Fe and Ni distribution in chondrules by synchrotron X-ray fluorescence and found that late-stage fluids had mobilized these elements from the matrix to deposit them as iron and nickel oxide/hydroxide in chondrule fractures. Such mobilization has little bearing on the inferred low ⁶⁰Fe abundance in Semarkona chondrules measured in this study for the following three reasons.

• (i)  
  In chondrites of metamorphic grade as low as 3.10, Telus et al. (2014) found that 100% of the chondrules were affected by Fe-Ni mobilization. However, in Semarkona (type LL3.00, one of the most pristine meteorite samples available), they report that approximately 70% of the chondrules analyzed did not display any evidence for mobilization (the probability to have a random Semarkona chondrule free of Fe-Ni mobilization is $p=0.7)$. The probability that $k$ out of $n=8$ random Semarkona chondrules be free of Fe-Ni mobilization is,

  $$\begin{eqnarray}&&P\left( k \right)=\left( \begin{array}{ccccccccccccccc} n \\ k \\ \end{array} \right){{p}^{k}}{{\left( 1-p \right)}^{n-k}}.\end{eqnarray} \tag{ 2 }$$

  The calculated probabilities are $P\left( 0 \right)\sim 0\%$, $P\left( 1 \right)\sim 0\%$, $P\left( 2 \right)\sim 1\%$, $P\left( 3 \right)\sim 5\%$, $P\left( 4 \right)\sim 13\%$, $P\left( 5 \right)\sim 25\%$, $P\left( 6 \right)\sim 30\%$, $P\left( 7 \right)\sim 20\%$, and $P\left( 8 \right)\sim 6\%$. At 95% confidence level, the majority (k > 4) of the chondrules analyzed here were not affected by Fe-Ni mobilization. Given that the chondrules were pre-screened to select those with high Fe/Ni ratios, the sample set is biased (p is probably higher than 0.7) and k > 4 is a conservative estimate.
• (ii)  
  In bulk measurements, the effect of Fe-Ni mobilization would be to lower the Fe/Ni ratios and $\varepsilon$⁶⁰Ni values toward chondritic values Such physical admixture of Fe-Ni in chondrules through fractures could have added some scatter to the data points and could have brought the samples toward the chondritic value but overall, the points should have moved along the isochron line. Indeed, mixing between two components in the $\varepsilon$⁶⁰Ni versus Fe/Ni diagram is a straight line, so the isochronous behavior of bulk chondrule measurements should have been preserved at some level.
• (iii)  
  The Semarkona chondrules most likely to have been affected by Fe-Ni addition should be among those that display low Fe/Ni ratios in bulk. Chondrules with high Fe/Ni ratios such as SC-13-6 (Fe/Ni ratio ∼24× chondritic) provide the most leverage to define the slope of the isochron and are least likely to have been affected by Fe/Ni mobilization. For reference, an initial ⁶⁰Fe/⁵⁶Fe ratio of 4 × 10⁻⁷ in Semarkona chondrules (Mishra & Goswami 2014; Mishra & Chaussidon 2014) should have been associated with excess ⁶⁰Ni of +6.3 in chondrule SC-13-6 (Equation (1)) while a value of +0.051± 0.043 was measured.

The results presented here thus reaffirm our earlier conclusion (Tang & Dauphas 2012) that ⁶⁰Fe was present at a low level in the chondrule-forming region, ⁶⁰Fe/⁵⁶Fe = (5.39 ± 3.27) × 10⁻⁹.

The mineral separates in the Sahara 99555 angrite give an initial ⁶⁰Fe/⁵⁶Fe ratio of (1.96 ± 0.77) × 10⁻⁹ (Figure 1(b)), which is in excellent agreement with the initial value of (1.8 ± 0.5) × 10⁻⁹ reported by Quitté et al. (2010) in the same meteorite. The weighted average of those two values is (1.85 ± 0.42) × 10⁻⁹.

The Ni isotopic compositions of mineral separates from the D'Orbigny meteorite, a quenched angrite like Sahara 99555, were measured independently by Quitté et al. (2010), Spivak-Birndorf et al. (2011) and Tang & Dauphas (2012). The initial ⁶⁰Fe/⁵⁶Fe ratios reported by these three studies are (4.1 ± 2.6) × 10⁻⁹, (2.81 ± 0.86) × 10⁻⁹, and (3.42 ± 0.58) × 10⁻⁹, respectively. The three values agree and the weighted average is (3.26 ± 0.47) × 10⁻⁹.

The calculated initial ⁶⁰Fe/⁵⁶Fe ratio of D'Orbigny is significantly higher than the value measured in Sahara 99555. The age difference between these two angrites is given by Δt₂₋₁ = In(r₁/r₂)/λ with an associated error of $\sigma ({\rm \Delta }{{t}_{2-1}})=\sqrt{\sigma _{{{r}_{1}}}^{2}/r_{1}^{2}+\sigma _{{{r}_{2}}}^{2}/r_{2}^{2}}/\lambda ,$ where r denote the initial ⁶⁰Fe/⁵⁶Fe ratio in either Sahara 99555 or D'Orbigny, and λ is the half-life of ⁶⁰Fe (2.62 Myr). The calculated age difference between D'Orbigny and Sahara 99555 is +2.1 ± 1. Myr.

The relative chronology of formation of D'Orbigny and Sahara 99555 can be compared with independent estimates obtained using various dating techniques. Internal ²⁶Al-²⁶Mg isochrons in Sahara 99555 and D'Orbigny give initial ²⁶Al/²⁷Al ratios of (4.50 ± 0.54) × 10⁻⁷ and (3.97 ± 0.26) × 10⁻⁷, respectively (Spivak-Birndorf et al. 2009; Schiller et al. 2010). These two values are almost indistinguishable, meaning that the two objects crystallized within ∼0.2Myr of each other Similarly, ¹⁸²Hf-¹⁸²W internal isochrons in Sahara 99555 and D'Orbigny give initial ratios of (6.83 ± 0.14) × 10⁻⁵ and (7.15 ± 0.17) × 10⁻⁵ (Kleine et al. 2012), corresponding to a time difference between D'Orbigny and Sahara 99555 of +0.6 ± 0.4 Myr. Pb-Pb ages have also been reported for D'Orbigny (4563.37 ± 0.25 Myr; Brennecka and Wadhwa, 2012) and Sahara 99555 (4563.64 ± 0.14 Myr, Connelly et al. 2008b; Larsen et al. 2011). These absolute Pb-Pb ages should be regarded with caution because inter-laboratory calibration for this dating system is lacking, yet the ages are indistinguishable. All available evidence thus suggests that D'Orbigny and Sahara 99555 crystallized at approximately the same time.

The difference between ⁶⁰Fe-⁵⁶Fe and other decay systems cannot be explained by a difference in closure age because the cooling rates of quenched angrites estimated from diffusion profiles and petrographic textures are rapid, 7–50 °C/hr (Mikouchi & McKay 2001). This means that it would take 1–6 days for the samples to cool by 1000 °C, which covers more than the span of closure temperatures for the systems discussed above. This is instantaneous in regard to planetary timescales, so one would expect all extant and extinct chronometers to be closed to isotopic exchange at approximately the same time.

Most likely, the low ⁶⁰Fe/⁵⁶Fe ratio measured in Sahara 99555 versus D'Orbigny reflects terrestrial contamination in the former. Sahara 99555 was found in the Sahara desert. As pointed out by Crozaz et al. (2003), Floss et al. (2003), and Amelin (2008), Sahara 99555 shows more evidence of terrestrial alteration than other angrites, including mobilization of rare earth elements. The sample that we studied was partly covered with some desert-weathering product that we physically removed but Crozaz et al. (2003) and Floss et al. (2003) showed that chemical alteration also affected Sahara 99555 samples with fresh appearances. The ⁶⁰Fe/⁵⁶Fe ratio of (3.26 ± 0.47) × 10⁻⁹ from D'Orbigny internal isochrons may thus provide the best estimate of the ⁶⁰Fe/⁵⁶Fe ratio at the time of crystallization of the quenched angrites (Quitté et al. 2010; Tang & Dauphas 2012).

Tang & Dauphas (2012) were able to constrain the ⁶⁰Fe/⁵⁶Fe ratio at Solar System birth to (115 ± 0.26) × 10⁻⁸ using internal isochrons in Gujba CB chondrite, D'Orbigny angrite, and unequilibrated ordinary chondrites as well as bulk rock isochrons for angrites and HED meteorites. The new results presented here for Semarkona chondrules and Sahara 99555 allow us to refine the initial ⁶⁰Fe/⁵⁶Fe ratio. Table 2 gives the ages relative to CAIs and initial ⁶⁰Fe/⁵⁶Fe ratios for all MC-ICPMS measurements for which resolvable ⁶⁰Ni-excess could be resolved, that is to say Semarkona chondrules, bulk HEDs, bulk angrites, and mineral separates of D'Orbigny and Sahara 99555. Those data relate to the ⁶⁰Fe/⁵⁶Fe ratio at the time of CAI formation using the free decay equation,

$$\begin{eqnarray}&&{{\left( ^{60}{\rm Fe}{{/}^{56}}{\rm Fe} \right)}_{t}}={{\left( ^{60}{\rm Fe}{{/}^{56}}{\rm Fe} \right)}_{{{t}_{{\rm CAI}}}}}{{e}^{-\left( t-{{t}_{{\rm CAI}}} \right)}},\end{eqnarray} \tag{ 3 }$$

where t and t_CAI are the formation ages of the samples considered and CAIs, respectively. The ⁶⁰Fe/⁵⁶Fe ratios inferred from ⁶⁰Ni-⁵⁶Fe/⁵⁸Ni isochrons are plotted in Figure 2 and all the results obtained thus far point to an initial ⁶⁰Fe/⁵⁶Fe ratio of (1.01 ± 0.27) × 10⁻⁸ and a homogeneous distribution of ⁶⁰Fe (see Table 2).

Zoom In Zoom Out Reset image size

The results from the present study reaffirm and strengthen our earlier conclusion that the abundance of ⁶⁰Fe in the early Solar System was low. This contrasts with recent in situ studies of chondrules from unequilibrated ordinary chondrites that report high initial ⁶⁰Fe/⁵⁶Fe ratios. Telus et al. (2014) made the case that most chondrites other than Semarkona have been affected by Fe-Ni mobilization. For this reason, we focus our comparison on the results obtained by SIMS on chondrules from the Semarkona meteorite. The initial ⁶⁰Fe/⁵⁶Fe ratios reported by Mishra & Goswami (2014) and Mishra & Chaussidon (2014) correspond to an initial ⁶⁰Fe/⁵⁶Fe ratio at Solar System birth of ∼70 × 10⁻⁸. This is almost two orders of magnitude higher than the initial ⁶⁰Fe/⁵⁶Fe ratio obtained by MC-ICPMS (∼10⁻⁸, Tang & Dauphas 2012; this study). This cannot be due to ⁶⁰Fe heterogeneity because Tang & Dauphas (2012) did not detect isotopic anomalies in ⁵⁸Fe, which is produced in stars together with ⁶⁰Fe. Furthermore, the same sample types, Semarkona chondrules, were measured by both SIMS and MC-ICPMS. Telus et al. (2013) also reported SIMS measurements of 3 Semarkona chondrules and found barely detectable ⁶⁰Fe in only one chondrule ⁶⁰Fe/⁵⁶Fe ∼(1.4± 1.2) × 10⁻⁷, while the other two chondrules only provided upper-limits. One such chondrule has an upper-limit of ⁶⁰Fe/⁵⁶Fe <5.1 × 10⁻⁸, which is significantly lower than the values reported by Mishra & Goswami (2014) and Mishra & Chaussidon (2014; Figure 3).

We have no explanation for the discrepancy between the MC-ICPMS and SIMS studies but we note that all SIMS estimates are based on measurements of a single sample type (i.e., pyroxene in chondrules from unequilibrated ordinary chondrites) and that most of the SIMS isochrons are defined by points that have error bars that largely overlap with zero, the significance of the isochron arising from a few data points with small errors (Figure 3) In contrast, the estimate from MC-ICPMS is derived from isochrons measured on a variety of samples formed at different times (bulk HEDs—Vesta, SNCs—Mars, bulk angrites, D'Orbigny and Sahara 99555 angrites, Gujba, NWA 5717, and Semarkona chondrules) and the significances of the regressions are in most cases very well established (Figure 1(b); Tang & Dauphas 2012). The isochron for Semarkona chondrules presented here (Figure 1(a)) is barely resolvable from zero but the corresponding initial ⁶⁰Fe/⁵⁶Fe ratio is much lower than the initial ratio expected based on SIMS data. Care was put in the SIMS studies to avoid analytical artifacts and it is not clear how much improvement can be made on that front but other instruments exist or are being developed that may provide a direct comparison with SIMS measurements, such as resonant ionization mass spectrometry, atom probe, or megaSIMS. At present, the weight of evidence supports a low and uniform abundance of ⁶⁰Fe in the early Solar System.

Zoom In Zoom Out Reset image size

The low initial ⁶⁰Fe/⁵⁶Fe ratio is consistent with background abundances in the Galaxy with no compelling need to invoke late injection from a nearby star (Tang & Dauphas 2012). Indeed, the average ⁶⁰Fe/⁵⁶Fe ratio in the ISM at Solar System birth inferred from γ-ray astronomy (Wang et al. 2007) is (2.8± 1.4) × 10⁻⁷, which is 30 times higher than the initial ratio in the early Solar System (Tang & Dauphas 2012). For comparison, the average ²⁶Al/²⁷Al ratio in the ISM at Solar System birth inferred from γ-ray astronomy (Diehl et al. 2006, 2010) is (3.0± 0.8) × 10⁻⁶, which is 17 times lower than the early Solar System ratio (Lee et al. 1976). It thus appears that ²⁶Al and ⁶⁰Fe in meteorites have different origins (Tang & Dauphas 2012).

Several scenarios can be considered to incorporate freshly made ²⁶Al without adding too much ⁶⁰Fe. Adjusting the timescale between nucleosynthesis and Solar System formation does not help because ²⁶Al has a shorter half-life than ⁶⁰Fe, so any delay would cause the ²⁶Al/⁶⁰Fe ratio to decrease, making the problem worse. The possible scenarios to explain the high ²⁶Al/⁶⁰Fe ratio of the early Solar System include (1) supernova (SN) explosion with fallback of the inner layers, so that only ²⁶Al is efficiently ejected while ⁶⁰Fe is trapped in the stellar remnant (Meyer & Clayton 2000) but this is unlikely because one would need to have a lot of fallback (a cutoff in the C/O-burning layer) to prevent ⁶⁰Fe from escaping (Takigawa et al. 2008), (2) interaction of a SN with an already formed cloud core, so that only the outer layers are efficiently injected while the inner layers are deflected (Gritschneder et al. 2012), and (3) ejection of ²⁶Al as winds from one or several massive stars (Arnould et al. 1997, 2006; Gaidos et al. 2009; Tatischeff et al. 2010; Gounelle et al. 2012; Young 2014). The last scenario is appealing because it could be a natural outcome of the presence in the Solar System forming region of one or several Wolf–Rayet (W–R) stars, as such stars shed their mass through winds rich in ²⁶Al whereas ⁶⁰Fe is ejected at a later time following the SN explosion (Tang & Dauphas 2012). W–R winds would have carved ²⁶Al-rich bubbles in molecular cloud material, which could have subsequently been incorporated in the molecular cloud core that formed the Solar System. Semi-analytic approaches have been used recently to assess the feasibility of such a scenario (Gounelle & Meynet 2012; Young 2014) but it remains to be seen whether high ²⁶Al/low ⁶⁰Fe regions can exist because SN explosion is the main factor that drives the mixing between W–R product and surrounding cloud material, so that the ²⁶Al-rich W–R material may be contaminated with ⁶⁰Fe-rich SN ejecta. Existing semi-analytic models aimed at explaining the high ²⁶Al/⁶⁰Fe ratio in meteorites have not addressed this critical aspect of the process, which will require high-resolution modeling of the interactions of stellar winds and ejecta with the surrounding medium.