Q-gases in a late-forming refractory interplanetary dust particle: A link to comet Wild 2

Abstract

We report the structure, chemical composition, O, Al-Mg, He, and Ne isotope systematics of an interplanetary dust particle, “Manchanito”. These analyses indicate that Manchanito solidified as refractory glass (with oxidized Fe but reduced Ti) in a chondrule-like formation environment more than 3.2 Myr after CAIs, after which it was exposed to Q-like noble gases in the dissipating solar nebula. Manchanito’s He and Ne isotopic composition and concentrations are similar to those measured in samples of comet Wild 2, from which we infer that Manchanito’s parent body was a comet. We propose that after formation and exposure to Q-like gases, Manchanito was transported to the outer Solar System where it came into contact with organics and volatile ices on its cometary parent body. Manchanito provides additional evidence that cometary solids have been subjected to energetic processing and large-scale transport in a wide range of environments in the Solar System.

1. INTRODUCTION

Microanalyses of dust returned by NASA’s Stardust mission from the coma of comet Wild 2 have revealed that most coarse-grained cometary particles >5 μm in size were primarily formed by high-temperature processes (Brownlee et al., 2012). Joswiak et al. (2017) reported the abundance of refractory objects (fragments of calcium-aluminum inclusions, amoeboid olivine aggregates, and aluminum-rich chondrules) in rocky cometary material (Stardust samples from Wild 2 and a giant cluster interplanetary dust particle of probable cometary origin) to be ~2%, similar to the abundance of refractory objects in primitive asteroids as sampled by the CV, CM, and CO chondrites (Hezel et al., 2008). Four high-temperature particles (a ferromagnesian silicate, a type II chondrule fragment, and two CAI-like particles) have been dated using the ²⁶Al-²⁶Mg chronometer (Matzel et al., 2010; Ogliore et al., 2012; Nakashima et al., 2015), and all were determined to form relatively late compared to similar objects in meteorites: at least ~1–3 Myr after the formation of CV CAIs. These measurements of Stardust samples and interplanetary dust have provided evidence that a comet’s nucleus is made of late-forming, high-temperature rocky material in intimate contact with extremely volatile ices. Cometary nuclei appear to reflect diverse formation environments – “fire and ice”.

Remote observations of comets have shown that organics are a significant component of comets as well (Chyba et al., 1989). Organic matter in interplanetary dust particles can be found as the “glue” that holds anhydrous grains together to form a single particle (Flynn et al., 2003), and can contain hydrogen and nitrogen isotopic anomalies that are evidence of low-temperature processing in the Solar System’s parent molecular cloud (Messenger, 2000; Busemann et al., 2006; Duprat et al., 2010).

Studies of light noble gas distributions in comet Wild 2 fragments from Track 41, extracted from Stardust aerogel cell C2044, revealed Ne isotope ratios within error of Ne compositions in “phase-Q” (Marty et al., 2008), a minor but gas-rich macromolecular organic phase ubiquitous in chondritic and achondritic meteorites (Busemann et al., 2000). Similar Q-like signatures in both an interplanetary dust particle (IDP) and comet Wild 2 fragments could suggest that the IDP originated in a comet, but given the widespread presence of Q-gases in meteorites this is insufficient evidence for a cometary provenance as opposed to origin from an asteroidal source. However, indigenous noble gases in IDPs, except those rapidly captured from Earth-crossing comet dust streams, are effectively masked by an abundant solar wind component implanted during their residence in space (Pepin et al., 2011).

Here we report detailed mineralogy, chemistry, O, Al-Mg, Ne, and He isotopic systematics of an unusual amorphous refractory particle from a giant cluster IDP, named “Manchanito”. The application of multiple techniques to a single small particle from an IDP can provide a more accurate understanding of the particle’s origin and subsequent processing than employing any single technique. A thorough understanding of Manchanito, in particular, will provide a better understanding of the relationship between interplanetary dust and rocky material in asteroids and comets. Giant cluster IDPs, possibly derived from comets (see, e.g., Joswiak et al., 2017), are the most promising samples to search for a link between interplanetary dust and bona-fide cometary material.

2. SAMPLE PREPARATION AND INITIAL ANALYSES

Cluster IDP L2071-F1 (Fig. 1) was extracted from its collector flag, mounted on a Si chip, washed with hexane, and imaged by SEM-EDX using a Tescan Vega SEM with an 80 mm² Oxford silicon drift detector to qualitatively identify major elements in individual particles. Most particles in the cluster were consistent with Fe-Mg silicates and iron-sulfides. One particle in the cluster, Manchanito, contained abundant Al and relatively low Mg, and for this reason was a possible target for Al-Mg isotope systematics by ion probe as well as other in-depth analyses. A secondary electron image of Manchanito in its entirety and an EDS map are shown in Fig. 2.

Fig. 1.

(Left) Optical image of L2071-Cluster 2 from the Cosmic Dust Catalog, red arrow points to an arbitrary 8 × 10 μm particle. (Right) Optical image of a portion of L2071-Cluster 2 on a Si chip, with Manchanito circled.

Fig. 2.

(A) 30 keV secondary electron image of Manchanito shows a roughly euhedral morphology with fine material decorating the surface. (B) 30 keV EDS RGB map with Br (red), C (green), Ca (blue) shows that the fine material is carbonaceous and Br-rich.

Manchanito contains an enrichment of Br on a thin layer along its edge (Fig. 2). Contamination by Br has been seen on the surfaces of interplanetary dust particles (e.g. Rietmeijer, 1992, 1993; Flynn et al., 1996) and provides evidence that Manchanito spent significant time in the stratosphere. The Br layer, along with Manchanito’s association with a giant cluster particle and the noble gas concentrations and compositions described in this paper, provide strong evidence that Manchanito is of extraterrestrial origin.

We transferred Manchanito to an epoxy bullet where the particle was microtomed, creating twelve ~100-nm thin sections as well as a potted butt containing the rest of the particle. Manchanito was mounted with its long axis parallel to the slicing direction of the microtome, the same orientation as shown in Fig. 2.

3. METHODS

Following these preliminary analyses, we carried out the following analyses:

1. We analyzed microtome thin sections by transmission electron microscopy (methods described in Section 3.1).

2. We then measured X-ray absorption near edge spectra at beamline 11.0.2 at the Advanced Light Source, (ALS) at Lawrence Berkeley National Laboratory. We obtained spectra for O-K, Fe-L and Ti-L. (Section 3.1)

3. We measured O isotope abundances in the potted butt using the University of Hawaii Cameca ims 1280 ion microprobe. (Section 3.3)

4. We affixed two TEM grids containing thin sections of Manchanito onto an Au stub using C paint.

5. We measured Al-Mg isotopes on the potted butt and TEM grids with the UH Cameca ims 1280 ion microprobe. (Section 3.4)

6. We extracted Manchanito from the epoxy bullet using an FEI Stata 235 focused ion beam (FIB) and mounted the extracted material onto a Pt foil for He and Ne analyses.

7. We measured He and Ne concentrations and compositions at the University of Minnesota. The sample was entirely consumed during the He and Ne gas analyses. (Section 3.4)

3.1. Transmission electron microscopy

We studied the microtome thin sections using STEM/EDS on a Philips CM200 TEM operating at 200 keV, located at the National Center for Electron Microscopy at Lawrence Berkeley National Laboratory. We employed bright-field imaging, selected-area electron diffraction (SAED), and energy-dispersive X-ray spectroscopy (EDS) to investigate the composition and structure of Manchanito.

3.2. Scanning transmission X-ray microscopy

We performed soft X-ray near edge absorption spectroscopy at beamline 5.3.2.1 at the ALS in order to characterize the chemical environment of Ti and Fe in Manchanito. Spectra were extracted from stacks with two spatial dimensions and one energy dimension. For Fe-L and Ti-L, we fit pre-edges (background) with a line and then subtracted so that the pre-edge was flat. The post-edge for Fe-L was normalized to an intensity of one by fitting it with a second-order polynomial. Ti-L was normalized with a first-order polynomial.

3.3. Oxygen isotope measurements by secondary ion mass spectrometry

We measured O isotopes in the potted butt of Manchanito, which was mounted in a specially designed holder that created a flat, conducting surface necessary for precise ion probe measurements (Westphal et al. 2011). We used a ~25 pA primary Cs⁺ beam focused to a ~2 μm spot size and a mass resolving power of ~5500 on ¹⁷O⁻ (sufficient to resolve ¹⁷O⁻ from ¹⁶OH⁻). We collected all three oxygen isotopes simultaneously: ¹⁶O⁻ on a Faraday cup, ¹⁷O⁻ on the monocollection electron multiplier, and ¹⁸O⁻ on a multicollection electron multiplier. Each measurement was 30 cycles with each cycle lasting 20 s. Miyakejima anorthite, mounted in a polished annulus around Manchanito, was used for an O isotope standard. We quantified instrumental mass fractionation from this sample-standard geometry by analyzing an analogously prepared potted-butt mount of San Carlos olivine. We found no significant instrumental mass-fractionation between the central and annulus-mounted San Carlos olivine. The count rate for ¹⁶O⁻ was ~7 × 10⁶ cps for Manchanito and ~5 × 10⁶ cps for the Miyakejima anorthite standard. We corrected our data for deadtime, detector backgrounds, and electron multiplier efficiencies. We corrected for a small mass-dependent drift (~2‰ per amu) of oxygen isotope ratios as a function of cycle (depth) within a measurement. Additionally, we determined the systematic error and instrumental mass fractionation of our standard measurements and applied these to the measurements of Manchanito.

3.4. Aluminum-magnesium isotope measurements by secondary ion mass spectrometry

For the Al-Mg measurements on the TEM sections, we used a primary ¹⁶O⁻ beam of ~10 pA focused to ~2 microns on the sample. We collected simultaneous measurements of ²⁴Mg⁺, ²⁵Mg⁺, and ²⁶Mg⁺ for 15 s followed by magnetic-field peak-jumping to ²⁷Al⁺ for 1 s. All species were measured on electron multipliers. The count rates of ²⁴Mg and ²⁷Al on Manchanito were, respectively, ~1000 and ~50,000 cps, and ~400 and ~150,000 cps on the plagioclase standard. We used a mass-resolving power of ~4800 which was sufficient to make the contribution of ²⁴MgH⁺ to measured ²⁵Mg⁺ negligible. We measured one spot each on 7 thin sections. Each measurement lasted between 100 and 300 cycles. We used Miyakajima plagioclase (An₉₅), microtomed into ~100-nm thin sections and placed onto a TEM grid (analogous to the Manchanito preparation) for a Mg isotope standard.

To improve the precision of our constraint of the initial ²⁶Al/²⁷Al in Manchanito, we also measured the potted butt for Al-Mg isotopes. We made this measurement after the O analyses described previously in order to avoid contaminating the O measurement with ¹⁶O from the primary ion probe beam used in the Al-Mg measurements. We mounted the potted butt in the same specially designed holder used for O isotope measurements (Westphal et al., 2011). An Al round holds the epoxy bullet at a fixed height. An Al annulus with polished standards sits on the round supporting an Au-coated, 500-nm-thick Si₃N₄ membrane which creates a flat, conducting surface. A 300-μm ion-milled hole in the window exposes the microtomed face of Manchanito at the same height as the polished standards, including Miyakajima plagioclase. We used a higher primary beam current for the potted measurement, ~25 pA, which increased the beam spot size to ~4 microns (still much smaller than the exposed face of Manchanito). For each of the 300 measurement cycles, we collected isotopes at three different magnetic fields: first ²⁴Mg⁺ on the monocollection electron multiplier for 4 s, then ²⁵Mg⁺ and ²⁷Al⁺ on the monocollection electron multiplier and a Faraday cup, respectively, for 10 s, then ²⁶Mg⁺ on the monocollection EM for 10 s. The count rate of ²⁴Mg⁺ and ²⁷Al⁺ on Manchanito was, respectively, ~15,000 cps and ~750,000 cps, and ~4,000 and ~1.5 × 10⁶ cps on the plagioclase standard. We corrected our data for deadtime and detector backgrounds.

We saw no significant isotopic variation as a function of depth over the course of any of the measurements of Miyakajima plagioclase. We calculated electron multiplier efficiencies for ²⁵Mg and ²⁶Mg such that the weighted mean of the standard measurements was consistent with only mass-dependent fractionation. The standard measurements exhibited instrumental mass fractionation of ±2‰ per amu (1σ, assuming a linear law) within our set of standard measurements, taking into account the statistical uncertainty of the analyses. We assume that this amount of instrumental mass fractionation is also present in the measurements of Manchanito, and the unaccounted for mass-dependent fractionation is the intrinsic mass fractionation of Manchanito.

Following Park et al. (2017), the excess ²⁶Mg in delta notation, δ²⁶Mg*, is calculated by:

$${\delta }^{26}\text{Mg}*=\left(\frac{\frac{{\left(\frac{{}^{26}\text{Mg}}{{}^{24}\text{Mg}}\right)}_{\text{sample}}}{{\left(\frac{{}^{26}\text{Mg}}{{}^{24}\text{Mg}}\right)}_{\text{standard}}}}{{\left(\frac{{\left(\frac{{}^{25}\text{Mg}}{{}^{24}\text{Mg}}\right)}_{\text{sample}}}{{\left(\frac{{}^{25}\text{Mg}}{{}^{24}\text{Mg}}\right)}_{\text{sample}}}\right)}^{1/\beta }}-1\right)1000$$ (1)

where we have assumed an exponential law for mass fractionation, with an exponent of β. We choose β = 0.5128 based on vacuum evaporation experiments of CAI-like melts by Davis et al. (2015), though the results of our analyses are not sensitive to the choice of β.

The relative sensitivity factor, RSF = (²⁷Al/²⁴Mg)_SIMS/(²⁷Al/²⁴Mg)_true, for An₉₅ is similar to melilite glass as measured previously on the UH ion probe. Since Manchanito’s chemical composition is relatively close to anorthite, we assume that the RSF for Manchanito is the same as the RSF that we measured on the standard Miyakajima plagioclase (0.93 ± 0.07 (1σ)). The ²⁷Al/²⁴Mg ratios measured in the anorthite standard showed variation larger than would be expected by counting statistics. This is likely due to the variable ionization efficiency among different measurements of these two elements in the ion probe. Unlike the isotope ratios, the measured Al/Mg ratios in Miyakajima plagioclase showed a systematic variation as a function of measurement cycle (depth). We quantified this variation as a function of measurement cycle by using the suite of standard measurements and their statistical uncertainties. We assumed that the Manchanito measurements were subjected to the same systematic uncertainties in their relative sensitivity factor. We also assumed that the Al/Mg ratio of Manchanito is accurately represented by our measurements of the potted butt and microtome slices.

We calculated the initial ²⁶Al/²⁷Al ratio in Manchanito using a York regression (Mahon, 1996) of δ²⁶Mg* vs. ²⁷Al/²⁴Mg and associated uncertainties in these quantities.

3.5. He and Ne isotope measurements

Most of the adhering epoxy was removed from the potted Manchanito sample by FIB milling. He and Ne abundances released in analysis of a large sample of the epoxy itself were comparable to the processing system blank without the epoxy, so any of this material present on the sample contributed little if any noble gases. The sample was loaded in a gas extraction furnace and baked at ~150 °C for three days. It was next heated in two 10-s steps at ⩽200 °C in a further effort to remove background gases associated with surface contamination. No noble gases above blank levels were released in these two steps. Stepped heating to higher temperatures, as described in Pepin et al. (2011), was then applied to the sample. A total of three 10-s heating steps were performed, at approximately 400, 650, and 790 °C. During these three individual heating steps, cumulative ⁴He and ²⁰Ne abundances were monitored. When sufficient helium and neon had evolved from the sample for accurate isotopic determinations, step heating was halted and the accumulated gas immediately admitted to the mass spectrometer for full isotopic analysis.

Higher temperature heating of Manchanito was carried out the next day, with a total of five 10 s pulses from approximately 800 to 1400 °C. These subsequent heating steps released no accumulated gas above blank levels. Measured amounts of gases released in the initial three step heating were far above system blanks, by factors of approximately 8.5 and 32 for ³He and ⁴He respectively, and 12, 5.0, and 8.3 for ²⁰Ne, ²¹Ne and ²²Ne respectively (Table 4).

Table 4.

Manchanito He and Ne data.

	3He/4He [×10−4]	4He [×10−12 cc]	20Ne/22Ne	21Ne/22Ne [×10−2]	20Ne [×10−12 cc]	20Ne contenta [ccSTP/g]	4He/20Ne
Blank	8.2(55)	1.68(50)	7.2(20)	4.8(18)	0.77(19)	-	2.18(84)
Manchanito*	2.61(26)	54.7(11)	10.53(51)	2.84(41)	9.45(37)	0.0065(16)	5.79(25)

Numbers in parentheses represent uncertainties in the last two digits of the listed numbers.

^aestimated Manchanito mass = 1.45 ± 0.36 ng (see text).

^*corrected for blank.

4. RESULTS

4.1. Mineralogy

Manchanito is entirely amorphous with a refractory composition. It is composed mostly of Ca- and Al-rich silica with detectable but minor amounts of Fe, Mg, and other elements (see Table 1). Its size is ~15 × 8 × 8 μm. Man chanito’s elemental composition does not closely match any minerals commonly found in meteorites. Comparing to amorphous phases found in meteorites, its composition most closely matches Al- and Ca-rich silica glass mesostasis found in chondrules and Al-rich chondrules, though it has more Ca than has been reported in these phases (Table 1). Amorphous silicates in the matrices of primitive chondrites are not similar in composition to Manchanito as they are enriched in Fe and Mg and contain only minor Al and Ca (Table 1). Despite its amorphous structure, Manchanito has an apparent euhedral shape. One TEM-EDS measurement from a thin section near Manchanito’s edge showed a more silica-rich composition with lower amounts of Al and Ca, signficantly different from Manchanito’s bulk composition by SEM-EDS reported in Table 1. Manchanito’s composition may not be homogeneous throughout its entire volume, though SEM-EDS maps (4–30 keV) show that most of the particle is relatively homogeneous in its bulk. SEM maps of the microtomed sample do not show strong zoning or other large elemental variations throughout the grain. Manchanito also has a C-rich coating on one side of the particle (Fig. 2) which is likely similar to the organic coatings frequently seen on interplanetary dust particles (Flynn et al., 2003; Flynn et al., 2013) (see Table 2).

Table 1.

Manchanito composition measured by SEM-EDS compared to chondrule mesostasis (“Mes.”) from selected chondrules from MET 00426 (CR2) & MET 00526 (L/LL3.05) (Berlin, 2010), glass in an Al-rich chondrule and bulk Al-rich chondrules from Semarkona (LL3.00) (MacPherson and Huss, 2005), and the mean values of amorphous silicates in the matrix of MET 00426 (Guillou et al., 2014) and QUE 99177 (CR2) (Guillou et al., 2015). “det.” = below detection limit, “–” = not reported.

Oxide Weight %
Sample	Na2O	MgO	Al2O3	SiO2	SO2	K2O	CaO	TiO2	FeO
Manchanito	0.78	0.39	14.99	57.08	det.	det.	25.93	0.75	det.
MET 00526 Ch1 Mes.	1.18	5.7	19.36	56.83	0.13	0.05	15.39	0.9	0.52
MET 00426 Ch1 Mes.	1.89	4.12	23.54	55.50	0.08	0.13	12.19	0.4	1.06
MET 00426 Ch3 Mes.	2.46	1.84	18.62	65.35	-	0.06	10.06	0.29	0.76
Semarkona Al-rich Ch. Bulk	0.25	29.51	15.92	43.45	0.03	0.02	8.87	0.33	0.39
Semarkona Al-rich Ch. Glass	2.95	4.71	11.51	65.21	-	0.50	10.58	1.47	0.19
MET 00426 Am. Sil.	-	16.3	3.0	40.9	1.5	0.2	0.5	-	29.8
QUE 99177 Am. Sil.	0.8	15.3	5.4	43.3	2.0	-	-	-	25.7

Table 2.

Ti-L fit peak positions.^a

Atomic %
Peak	Energy (eV)	Intensity (%)
A	457.98	11.3
B	459.95	32.4
C	463.23	21.6
D	465.26	34.7

^aPositions before shifting −1.8 eV.

4.2. Oxidation state

X-ray absorption spectroscopy can measure the oxidation state of transition metals in silicates. The Fe-L₃ edge, shown in Fig. 3B contains two peaks. Fe²⁺ and Fe³⁺ can be distinguished on the basis of the splitting of the L₃ edge. When Fe²⁺ is present in silicate glass then the peak at 708 eV will dominate the peak at 710 eV. As iron oxidizes and the fraction of Fe³⁺ increases, then the second peak at 710 eV begins to dominate (van Aken and Liebscher, 2002; Stodolna et al., 2013). The spectrum shown in Fig. 3B therefore indicates that a significant fraction of the Fe is oxidized to Fe³⁺ which can occur at higher oxygen fugacities such as near the fayalite-magnetite-quartz (FMQ) buffer.

Fig. 3.

X-ray near edge spectra of Manchanito. (A) The oxygen K-edge shows a small pre-edge peak. (B) Fe-L₃ line shows clear splitting with a higher intensity line around 710 eV and a lower intensity line at 708 eV. (C) Ti-L_3,2 showing splitting into multiple lines.

The Ti-L edge at 455 eV shows at least four clear peaks (Fig. 3C) which arise from splitting of 2p orbitals and 3d orbitals in a crystal field. In order to model the oxidation state of the Ti, we applied multiplet theory (de Groot, 2005; Haverkort, 2016; Retegan, 2018). Starting with the spectrum shown in Fig. 3C, we fit the peaks using four Gaussian functions, and we fit the Ti-L_2,3 absorption edges using two arctangent functions (Stoyanov et al., 2007). In order to compare the experimental spectrum against simulated spectra generated by multiplet theory, we produced a reconstituted experimental spectrum using only the Gaussian peaks since we did not model the absorption edges.

We generated three simulated spectra with variable assumptions about oxidation and coordination environment. An octahedral crystal field splitting of 10Dq = 2 eV was chosen by observing that the splitting within the A-B and C-D line pairs (see Fig. 4) was ≈ 2 eV (see also De Groot et al. (1990), Ikeno et al. (2011)). This splitting was applied to both Ti⁴⁺ and Ti³⁺ to produce “4+ (CF)” and “3+ (CF)” in Fig. 4. In cases where the crystal field splitting was small we would expect the atomic spectrum to be close to the observed spectrum. Therefore we also included an atomic Ti³⁺ spectrum with no crystal field. We did not include an atomic Ti⁴⁺ spectrum since it shows evidence for only two peaks.

Fig. 4.

Simulations of Ti-L edge compared to the experimental (i.e. measured) spectrum (“Exp”). The experimental spectrum has been fit using four Gaussian peaks and two arctangent edges, and then reconstituted using only the Gaussian peaks (cp. Fig. 3C). An energy shift has been applied to the experimental spectrum to match simulated positions. The simulated spectra were generated using multiplet field theory, with octahedral coordination in all cases. Simulated spectra are shown for Ti⁴⁺ and Ti³⁺ with a crystal field splitting (“CF”) of 2 eV, as well as Ti³⁺ without crystal field splitting (atomic, “at”). See text.

The simulated spectra were combined in various ratios in order to fit the experimental spectrum. The optimal proportions were chosen using a Nelder-Mead optimization (Nelder and Mead, 1965) with 5 variables: 3 component spectra, an energy shift applied to the experimental spectrum, and an overall amplitude. The resulting fit had 75% Ti³⁺, and 25% Ti⁴⁺. While the fit was not perfect, since the theoretical spectrum had a small peak at 467 eV which was not seen in the experimental spectrum, and the positions of the A and C lines disagreed by 0.5 eV, the overall fit was superior to that of any of the component spectra alone.

Stoyanov et al. (2007) describe a method for computing the Ti³⁺/Ti⁴⁺ ratio using a similar method, which gives 16% Ti⁴⁺, in rough agreement with our multiplet calculation. We conclude that the oxidation state of the Ti is dominantly Ti³⁺, but includes a minor but non-zero Ti⁴⁺.

At fixed oxygen fugacity, Ti is more oxidized than Fe. The Ti₂O₃–TiO₂ (Ti³⁺ -Ti⁴⁺) buffer is at $\text{log}\left({\text{f}}_{{\text{o}}_2}\right)=-16$ at 1500 K compared to the QFM buffer (Fe²⁺–Fe³⁺) with $\text{log}\left({\text{f}}_{{\text{o}}_2}\right)=-8$ at 1500 K. The Ti XANES analysis indicates that Manchanito formed in a very reducing environment, but the Fe XANES indicates a more oxidizing environment. The spectra do not show significant variation throughout the sample, so Manchanito does not appear to have oxidation state zoning.

It is not obvious how to reconcile the oxidized Fe with the reduced Ti. Since the diffusion rate of Fe varies from Ti as a function of composition and temperature, it is most likely that Ti and Fe in Manchanito were both originally reduced or oxidized and subsequent alteration flipped one element. Zhang et al. (2010) give diffusion rates for both Fe and Ti in a number of compositions including andesites which have similar Si and Al content to Manchanito, though lower Ca (andesite1 and andesite2 from Zhang et al. (2010)). The diffusion rates for Ti and Fe are both D ≈ 10⁻⁵ m²/s² at 1500 K. The slope of Fe diffusion is steeper than for Ti, so as temperature drops below 1500 K, we would expect Ti to be relatively more mobile than Fe. Above 1500 K, we would expect the opposite.

We conclude that it is not possible to determine from the XANES analyses alone whether Manchanito was originally reduced and then oxidized or the reverse. However, we can tell from the XANES that it has seen both oxidizing and reducing environments in its history. Serpentization of anhydrous minerals during aqueous alteration can cause an increase in the Fe oxidation state. Such a process may have altered amorphous silicates (Guillou et al., 2015) and secondary phases of CM chondrites (Pignatelli et al., 2017). However, Manchanito is not similar in composition to amorphous silicates (Table 1) or secondary phases found in chondrites, so such an alteration environment is unlikely.

4.3. O isotopic composition

We measured the O isotopic composition of two spots on Manchanito. The results for both spots were consistent with each other when taking into account statistical and systematic uncertainty. The weighted mean of these two spots was δ¹⁷O = 7.0 ± 1.2‰, δ¹⁸O = 14.0 ± 1.1‰ (2σ uncertainties) (Fig. 5).

Fig. 5.

Oxygen isotopic composition of Manchanito (this work) compared to anhydrous and hydrous IDPs measured by Aléon et al. (2009), a low-Ca pyroxene grain measured by Nakashima et al. (2012), refractory IDPs measured by McKeegan (1987), and chondrule mesostases measured by Chaussidon et al. (2008). The terrestrial fraction line and carbonaceous chondrite anhydrous mineral (CCAM) line are denoted in the figure.

4.4. Al-Mg isotope systematics

We established an upper limit for the initial ²⁶Al/²⁷Al ratio in Manchanito of 2.2 × 10⁻⁶ (2σ). The Mg isotopic compositions and isochron are shown in Fig. 6; data are shown in Table 3. Assuming an ²⁶Al half-life of 705,000 years (Norris et al., 1983) and an initial ²⁶Al/²⁷Al ratio of 5.2 × 10⁻⁵ in CV CAIs (Jacobsen et al., 2008), Manchanito solidified from a melt, or was last significantly heated, more than 3.2 Myr after the formation of CV CAIs. Alternatively, we conclude that Manchanito may have formed in region of the Solar System that lacked ²⁶Al (see discussion in Section 5.2).

Fig. 6.

(Left) Mg isotope compositions of standards and Manchanito. (Right) Isochron plot showing excess ²⁶Mg (Eq. 1) vs. Al/Mg ratio in Manchanito. Dashed line is canonical (5 × 10⁻⁵, Jacobsen et al. (2008)), solid line is the 2σ upper bound of our measurements: 2 × 10⁻⁶. White circle is potted butt, red dots are measured on microtome slices. Isochron is tied to the origin (solid black triangle, from measured standards), uncertainty of the origin point is the standard error of measurements of the anorthite standard.

Table 3.

Manchanito Al and Mg ion probe data.

Sample	27Al/24Mg	σ(27Al/24Mg)	δ26Mg	σ(δ26Mg)	δ25Mg	σ(δ25Mg)	δ26Mg*	σ(δ26Mg*)
Slice 1	44.7	2.0	−9.5	2.4	−3.6	2.5	−2.1	7.5
Slice 2	44.5	2.5	−8.2	3.0	0.5	3.2	−11.0	9.4
Slice 3	44.3	1.9	−7.4	2.2	−2.2	2.4	−4.7	7.1
Slice 4	49.1	4.9	−23.1	5.5	−14.8	5.8	4.3	17.9
Slice 5	45.4	3.4	−16.7	3.9	−5.9	4.1	−6.6	12.4
Slice 6	40.9	3.4	−12.6	4.2	−5.9	4.5	−2.4	13.4
Slice 7	38.6	2.2	−10.6	2.8	−5.8	2.9	−0.5	8.9
Potted Butt	43.3	6.7	37.6	0.9	19.9	1.0	0.0	2.7

All errors are 2σ. ²⁷Al/²⁴Mg ratio is in atom percent.

4.5. Ne and He isotopic compositions and release profiles

Several observations (see Fig. 7) are pertinent to understanding the He and Ne distributions in Manchanito:

1. Compositions and abundances of He and Ne in Manchanito compared to those found in cometary material separated from Stardust Track 41, aerogel cell C2044.

2. Implications of Manchanito’s amorphous structure, indicated by its TEM diffraction pattern, for the very different Ne release patterns seen in the left (Track 41) and right (Manchanito) panels in Fig. 7.

3. The concentration of He and Ne in the adhering carbonaceous matter on Manchanito.

Fig. 7.

Stardust and Manchanito: Observations relevant to sources and distributions of noble gases. Left panels) Stardust Track 41 and typical samples Thera-1 and St-2 keystoned from the track wall; gas carrier grains embedded in the aerogel estimated to be ~0.26 ng in mass (Marty et al., 2008). Far left) Cumulative ²⁰Ne fraction released vs. temperature for the Ne-rich Stardust sample St-1 (not shown). Right panels) Manchanito, its diffraction pattern indicating amorphous structure, image of adhering organic matter, and Ne release profile where the dotted curve shows the release pattern from terrestrial obsidian glass (Matsuda et al., 1989). A more detailed plot of this profile, including the Matsuda et al. (1989) data points, is shown in Fig. 10 in Section 5.3.

See discussions of amorphization, Ne release vs. temperature, and gas host in Sections Sections 5.1 and 5.3.

Ne and He isotope ratios measured in Stardust Track C2044,0,41 (Marty et al., 2008) (“Track 41”) are shown in Fig. 8A,B. Manchanito data from the present study are listed in Table 4 and shown in red in Fig. 8. The two Track 41 measurements in Fig. 8A, St-1 and Thera-2, are consistent within error with the Ne-Q compositional range determined by Busemann et al. (2000). The measured Manchanito value falls in the Ne-Q data field, although uncertainties are larger, particularly for ²¹Ne/²²Ne.

Fig. 8.

Neon and helium isotope ratios and Ne concentrations in Stardust Track 41 and Manchanito. Stardust data as shown in Figs. 1–3 of (Marty et al., 2008), with Manchanito measurements (this work) superimposed in red. (A) ²⁰Ne/²²Ne vs. ²¹Ne/²²Ne compared with the Ne-Q range from (Busemann et al., 2000). (B) ³He/⁴He for Manchanito, Stardust St-1 and St-2, He-Q (Busemann et al., 2000), Jupiter (Mahaffy et al., 1998), and solar wind (Heber et al., 2009). The air ratio of 1.4 × 10⁻⁶ falls ~2 orders of magnitude below the plot boundary. (C) Neon concentrations in cm³ STP/g, with IDP and lunar regolith concentrations slightly updated from Fig. 3 in (Marty et al., 2008). See caption of that figure for the estimates of possible grain loading by adsorbed and dissolved nebular Ne. Meteoritic Ne-phase Q range is from Busemann et al. (2000), which is likely an underestimate because the unknown actual Q-carrier probably constitutes only a fraction of the Q-phase masses.

Turning to He in Fig. 8B, one sees that the ³He/⁴He ratios in Manchanito and those measured in Track 41 (Marty et al., 2008) are statistically indistinguishable, and therefore that both Ne and He compositions in these samples, deriving from very different present-day provenances, are identical within error. The two St-He ratios obtained at Minnesota were subsequently replicated in independent analyses of additional Track 41 materials at another laboratory (Palma et al., 2019). This strengthens the argument for a cometary origin for Manchanito. However, these three ³He/⁴He measurements are significantly elevated above He-Q in Fig. 8B and so are incommensurate with the Q composition, unlike the Ne. See Section 5.3 for discussion.

The cumulative ²⁰Ne release pattern vs. temperature profile for Manchanito is plotted in the right panel of Fig. 7. It is precisely congruent with that measured for terrestrial obsidian by Matsuda et al. (1989), indicating that Ne release from the obsidian and Manchanito glasses is governed by the same diffusive mechanism. It is likely that this is also the case for He, where obsidian diffusivities are several orders of magnitude larger than for Ne (Matsuda et al., 1989). Consequently, these glasses would readily lose most of their He inventories, even at low temperatures. Such losses are reflected in Manchanito’s He release pattern; in contrast to the ²⁰Ne profile, no ⁴He at all evolved from the sample prior to the heating steps at 650 °C (45% release) and 790 °C (100% release). The implication is that large He losses from Manchanito have occurred, at temperatures reached by drag heating on atmospheric entry, during the 150 °C bakeout or even at room temperature. Neon, however, is fully retained at 20 °C, although a fraction of its inventory could have been lost in bakeout.

4.6. Neon concentration

Strikingly high noble gas concentrations were measured in the Stardust Track 41 particles, comparable to those implanted into lunar fines and many IDPs by solar wind ions (Fig. 8C) and suggesting a similar gas acquisition mechanism, in this case by Q-ion irradiation, for the track carrier grains (Marty et al., 2008). A correspondingly large concentration of Q-gases is observed in Manchanito: measured ²⁰Ne abundance (9.45 ± 0.37 × 10⁻¹² ccSTP) combined with the estimated mass of the sample after milling from the potted butt (~1.45 ± 0.36 ng, determined from the shape and size of the sample) yields the Ne concentration of 6.5 ± 1.6 × 10⁻³ ccSTP/g shown in Fig. 8C. Sample mass was derived from the volume analyzed for noble gases (~5.30 × 10⁻¹⁰ cm³), calculated from the potted sample dimensions of 13.7 μm × 6.7 μm × 6.8 μm, minus estimated volume losses of 10% in FIB milling from the butt and 32 μm³ excavated in O, Al-Mg analyses. Density was assumed to be within 10% of the anorthite value of 2.73 g/cm³. Errors are propagations of conservative uncertainties in sample volume measurements, density, and ²⁰Ne abundance.

Manchanito’s Ne concentration is remarkably large. Two possibilities for creating it are diffusive ingassing from ambient Ne, or implantation of irradiating Ne ions. A demonstrated mechanism for diffusive acquisition from ambient atmospheric Ne is that described by (Matsuda et al., 1989) for obsidian glass, a reasonable structural analog of Manchanito. Their measurements of abundances and diffusion coefficients demonstrate that obsidian cooling from a melt at Earth’s surface passes through a temperature regime between 300–600 °C in which Ne diffusion coefficients are large and Ne is consequently diffusively ingassed from ambient air at full Ne partial pressure, before being frozen in as temperatures and diffusivities decline. The resulting measured Ne concentration in obsidian was 4.0 × 10⁻⁸ ccSTP/g (Matsuda et al., 1989). This diffusive mechanism could have operated in the solar nebula, but since Manchanito’s ²⁰Ne concentration of 6.5 × 10⁻³ ccSTP/g exceeds that in the obsidian glasses by more than 5 orders of magnitude, nebular conditions such as Ne partial pressure and duration of diffusive ingassing must have been very different from those governing the Matsuda et al. (1989) experiments. An alternative environment in which loading by implantation of energetic Ne ions could have occurred is discussed in Section 5.1.

5. DISCUSSION

5.1. Amorphization in the solar system?

Manchanito’s euhedral shape can be understood if it originally formed as a crystal and then was later amorphized. Manchanito’s elemental composition does not match any known mineral, so amorphization would have to be accompanied by preferential elemental loss to explain Manchanito’s composition. In this section we explore the possibility that Manchanito was amorphized by energetic particles in the Solar System, and conclude that Manchanito could have been amophized in a near-Sun environment 3–5 Myr after the formation of the first Solar System solids.

During its entire history, from formation to analysis, Manchanito was shielded from bombardment by energetic particles inside its parent body except for two time periods: (1) immediately after its formation, presumably in the solar nebula, but before its incorporation into a larger body and (2) after its ejection from its parent body but before it was accreted by Earth (during this time, it was lightly shielded by its host cluster particle). In the first case, gas drag in the solar nebula is the dominant force for determining the lifetime of a ~10 μm particle before it is lost to the Sun (Dohnanyi, 1972). For the second case, Poynting-Robertson drag (i.e. radiation-pressure drag) and collisional loss determine the particle’s characteristic lifetime. Dohnanyi (1972) estimated that the lifetime of 10 μm particles is less than 10⁵ years. Since gas drag will decrease this characteristic lifetime, we take 10⁵ as the upper limit of time Manchanito may have spent near 1 AU in either scenario where it could be exposed to significant energetic particle radiation.

We calculated the fraction of atomic vacancies (a proxy for crystallinity) created in Manchanito by exposure to solar energetic particle (SEP) irradiation in the heliosphere near 1 AU. We used SEP fluxes by Mewaldt et al. (2007) from H to Fe, though most of the irradiation damage (vacancies) are from H and He. Solar wind particles are higher in flux but will not effectively amorphize a 10 μm particle like Manchanito due to their low energy–solar wind would only create vacancies in the top <1 μm layer, whereas Manchanito, with characteristic dimensions of several μm, is fully amorphized. We used the “SRIM - The Stopping and Range of Ions in Matter” program (Ziegler et al., 2010) to simulate by a Monte Carlo technique the vacancy production in an object of Manchanito’s elemental composition caused by SEP irradiation. The results of our simulation (vacancy fraction as a function of depth) are shown in Fig. 9. A calculated vacancy fraction of at least 25% is required to explain Manchanito’s amorphous structure. We conclude that SEP irradiation at present-day average flux is insufficient to amorphize Manchanito, as it would take >1 Myr of SEP exposure near 1 AU to amorphize this particle, which is ~100 × longer than an IDP’s expected lifetime in the heliosphere. We also calculated the amorphization effect that galactic cosmic rays (Simpson, 1983) would have on Manchanito, but these particles are much higher energy (~100 MeV/nucleon) and do little to amorphize the top few micrometers.

Fig. 9.

Vacancy fraction calculated by SRIM and scaled for 10⁴, 10⁵, and 10⁶ years of exposure.

A lifetime of >1 Myr near 1 AU is required to amorphize Manchanito by energetic particle irradiation. Since this is much larger than the upper-bound characteristic lifetime of dust particle like Manchanito in the Solar System (0.1 Myr), we conclude that Manchanito could not have been amorphized by present-day fluxes and energies of solar energetic particles near 1 AU. However, amorphization on reasonable time scales near 1 AU could have occurred during an active T-Tauri phase of the young Sun (e.g. Sossi et al., 2017), prior to Manchanito’s incorporation into its host parent body. Fluxes of solar energetic particles may have been five times higher than present day fluxes (Kööp et al., 2018).

Alternatively, irradiation and possibly amorphization of Manchanito could have occurred in an environment 3–5 million years (constrained by Manchanito’s Al-Mg isotope systematics) after the formation of the first Solar System solids (Pepin et al., 2015), characterized by brief particle residences in an energetic and assumedly Q-rich gas reservoir close to the young, intensely flaring Sun. Energetic flare-generated ion fluxes, as observed near young solar-type stars in the Orion Nebula, exceed present-day solar energetic particle fluxes by a factor of ~10⁵ (Feigelson et al., 2002). Grains condensing in such an environment would be exposed to high-energy proton and Q-ion radiation, prior to expulsion and transport to distant comet-forming regions of the accretion disk (Brownlee et al., 2012). Manchanito’s noble gas inventories could have been populated by Q-ion implantation during a brief residence in this early near-Sun reservoir.

This irradiation scenario is based on efforts to account for high levels of spallogenic ²¹Ne in several grains from the U2–20GCA giant cluster particle (GCP), also thought to be cometary in origin (Messenger et al., 2015; Pepin et al., 2015). A subset of the grains also carry Q-gases. These spallation abundances are too large to be attributed to contemporary solar (SCR) or galactic (GCR) cosmic ray irradiation in any contemporary environment, whether generated by SCR irradiation of small individual particles in space or by GCR if buried in a regolith; both require implausibly long exposure times, in some cases approaching or exceeding the age of the Solar System. However, during residence in an early near-Sun environment subject to intense high-energy proton fluxes, spallation ²¹Ne in the GCP grains can be produced in centuries to millennia. Manchanito’s measured Q-like ²¹Ne/²²Ne ratio suggests the absence of a spallation component, but its large uncertainty (Table 4, Fig. 8A) allows exposure times up to ~1350 years, a range occupied by most of the spallation-rich GCP grains (Pepin et al., 2015).

The large Manchanito Ne loading shown in Fig. 8C is consistent with ion implantation. Energetic heavy ion irradiation is a well-studied mechanism for inducing amorphization in crystalline minerals (e.g. Jäger et al., 2003). Manchanito’s Ne content divided by its average projected surface area–modeled as that of a randomly oriented, equal–area cylinder exposed to a directional ion flux yields a minimum Ne fluence, assuming no impinging ions were lost from the sample, of ~2.7 × 10¹⁴ cm⁻². Argon fluence is ~25 × higher, ~7 × 10¹⁵ cm⁻², for Q-composition radiation (Busemann et al., 2000). A ~ 10 × smaller fluence of 0.4 MeV Ar ions, 5 × 10¹⁴ cm⁻², is sufficient to amorphize crystalline enstatite to the depth of the Ar ion range (Jäger et al., 2003). However at 0.4 MeV the range is only ~0.3 μm. Penetration to ~10 × this depth, corresponding to Ar ion energies of several MeV, would be required to fully amorphize a particle of Manchanito’s size.

5.2. Manchanito’s origin: late-forming refractory chondrule glass

Manchanito’s refractory composition and amorphous structure link it to refractory objects typically found in meteorites, but which are also known to be present in cometary solids at the percent level (Joswiak et al., 2017). Refractory IDPs composed of Ca-, Al-, and Ti-rich minerals typically found in CAIs have been identified (Christoffersen and Buseck, 1986; Zolensky, 1987; Greshake et al., 1996; Joswiak et al., 2017). Refractory glasses (enriched in Ca and Al) have been identified in chondrules (e.g. Reid et al., 1974) and CAIs (e.g. Grossman et al., 1988; Kimura et al., 1993) from carbonaceous chondrites, though such objects are relatively rare. Zolensky (1987) reports refractory interplanetary dust particles that contain hibonite, gehlenite, and perovskite as well as glass of refractory composition.

We conclude that Manchanito most likely originated in an igneous object, possibly an Al-rich chondrule (Bischoff and Keil, 1983), which broke apart and liberated this grain (either initially amorphous, or amorphized after formation by a high flux of solar energetic particles) for later incorporation into the giant cluster IDP. During its formation and any subsequent alteration, Manchanito experienced both oxidizing and reducing environments (Section 3.2).

Manchanito’s O isotopic composition is more ¹⁶O-poor than other refractory and most other anhydrous IDPs and is consistent with the composition of hydrous IDPs (Fig. 5). However, based on its composition, Manchanito is not a hydrated IDP and therefore likely did not form by the same mechanism. A more-appropriate comparison is with chondrule mesostasis, which is only slightly more ¹⁶O-rich than Manchanito (Fig. 5). We conclude, based on its O isotopic composition and mineralogy, that Manchanito formed by similar mechanisms and from similar source material as chondrule mesostasis glass. Manchanito may have originally filled the space between adjacent euhedral crystals in its host CAI or chondrule, resulting in the particle’s euhedral shape.

Our measurements of the ²⁶Al-²⁶Mg constrain the time when Manchanito formed, or was last significantly heated, to be more than 3.2 Myr after the formation of CV CAIs. Under the assumption that Manchanito and the cluster IDP originated in a comet, our measurements provide only the fourth measurement of the ²⁶Al-²⁶Mg system in cometary material. The other three measurements were made on Stardust samples from comet Wild 2, so Manchanito is likely the first measurement from a different comet (though we don’t know which one). None of the cometary objects studied so far show evidence for the former presence ²⁶Al (Matzel et al., 2010; Ogliore et al., 2012; Nakashima et al., 2015). Formation times for these cometary objects constrain their ages to be at least as young as some of the youngest chondrules in carbonaceous chondrites, particularly chondrules in CR chondrites (Ogliore et al., 2012). This implies either that comets preferentially sampled material that formed late, after the bulk of the initial ²⁶Al had decayed, or comets accreted objects that formed in a region of the disk that lacked ²⁶Al, possibly in the giant-planet region (eliminating the need for large-scale nebular transport of these objects to the comet-forming region, Scott et al. (2018)).

We conclude that Manchanito formed in a relatively late (>3.2 Myr after the formation of CAIs) chondrule-forming event from source material similar to the source materials of chondrules in chondrites, either in the inner (Boss and Durisen, 2005) or outer (Scott et al., 2018) Solar System.

Manchanito was found within the giant cluster IDP L2071-Cluster 2 on the collector flag. It is most likely that Manchanito was originally associated with this cluster, however we cannot rule out the small possibility that Manchanito arrived on the cluster as an individual cosmic particle and serendipitously landed on L2071-Cluster 2. In either scenario, Manchanito must have an extraterrestrial origin because of the measurements reported here, particularly Manchanito’s ³He/⁴He ratio which is ~100 higher than the terrestrial atmosphere.

5.3. Manchanito’s formation environment: evidence from noble gases

Q-type Ne isotopic distributions are observed in Manchanito, in five Stardust aerogel samples extracted from the Wild 2 comet coma collection tray (Palma et al., 2019), and in several grains from the U2–20GCA giant cluster particle (Pepin et al., 2015). All are either direct comet samples or IDPs plausibly originating from comets, and all exhibit comparably high ²⁰Ne concentrations. These Ne ratios collectively point to an intrinsic Q-Ne component in comets. Data on He compositions are more scattered. Average ³He/⁴He in the U2–20GCA grains, corrected for large spallation components, is 1:4 ± 0:8 × 10⁻⁴, similar to Q-³He/4He but with large uncertainty (Pepin et al., 2015). In samples from the Wild 2 coma collection displaying Q-like Ne compositions, the ³He/⁴He ratios range from 2.6 × 10⁻⁴ in the Track 41 particles to higher values that indicate possible solar wind components (Palma et al., 2019). However, the identical ratios of 2.6 × 10⁻⁴ in Manchanito and Wild 2 track 41 suggest that neither is perceptibly elevated by recent solar wind additions. We think it improbable that the two samples, deriving from provenances as different as a free IDP and an ablating comet, could each have coincidentally acquired just enough solar wind He to elevate an underlying Q-He ratio to their identical measured values. The fact that the He concentration in track 41 (Marty et al., 2008) ~is 20 × higher than in Manchanito (Table 4) presents a further difficulty: a much longer solar wind exposure than experienced by Manchanito would be required to equally increase their ³He/⁴He ratios from Q. Conditions allowing a substantial implantation of solar wind into a particle on Wild 2, an actively ablating short period comet, are difficult to realize. It appears much more likely that these two ratios actually reflect primordial compositions.

This suggests that the ³He/⁴He ratio of 2.6 × 10⁻⁴ is indigenous to Manchanito, and that this IDP originates from a comet isotopically similar to Wild 2. If so, there are constraints on its space exposure history. A few hundred years of solar wind implantation with ³He/⁴He = 4.6 × 10⁻⁴ (Heber et al., 2009) would increase Manchanito’s ³He/⁴He from an initial Q ratio of ~1.33–1.45 × 10⁻⁴ (Busemann et al., 2000) to its measured value. However, if Manchanito is cometary and its He mostly indigenous, exposure to solar wind must have been limited to a decade or so. This would point to origin from a Jupiter family comet with a recently created dust stream, similar to Schwassmann-Wachmann 3 (Messenger, 2002).

Manchanito is part of a large cluster particle (Fig. 1) that must have been relatively fragile to have disaggregated when it impacted the collector at 200 m/s. Cometary meteors tend to be weak (Trigo-Rodríguez and Llorca, 2006) compared to even the most fragile meteorites like Tagish Lake (Brown et al., 2002; Flynn et al., 2018). Based on Manchanito’s noble gas signatures and its association with a fragile cluster IDP, we conclude that its most likely parent body is a comet.

Organic materials can trap noble gases at high concentrations (Busemann et al., 2000). We estimate Manchanito’s carbon-rich coating (Fig. 2B) B) to have a mass of ~5 pg. If Manchanito’s gases were sited in this mass, it would contain ~10⁻³ Ne and ~6 × 10⁻³ He atoms per C atom for a pure C carrier, levels that could plausibly be accommodated in an organic structure (Busemann et al., 2000). Moreover a larger organic mass is likely since the weak carbon Kα X-ray from possible deposits on other faces of Manchanito would have been largely absorbed in transit to the detector.

However there are strong arguments, shown in Fig. 10, that the surficial organic matter imaged in the upper panel is not the host of the Q gases in Manchanito. The most compelling evidence is displayed in the cumulative Ne release profile on the left side of the lower diagram, where Ne evolution vs. temperature from the vitreous Manchanito sample is congruent with that from terrestrial obsidian glass (Matsuda et al., 1989). The obsidian Ne, which comes from the terrestrial atmosphere, devolves from the glass itself, and this is clearly indicated for Manchanito as well. Moreover, the organic residues containing Q-gas carriers in carbonaceous chondrites (phase Q) are refractory, as seen in the lower panel; at the temperature where degassing of Manchanito and obsidian Ne is 90% complete, essentially no Ne has been released from the Orgueil and Leoville residues (Huss et al., 1996). The Ne release profile for the adhering Manchanito organics is unknown, so this material can be definitively ruled out as the Q-gas carrier only to the extent that its degassing systematics at low temperatures may resemble the release profiles of the refractory meteoritic residues, and not that of Manchanito. Nevertheless, we conclude that the noble gases are indeed carried by the bulk amorphous phase of Manchanito, and the organic coating attached to the particle was accreted later on Manchanito’s parent body (comets are known to contain organics, Capaccioni et al. (2015)).

Fig. 10.

Surficial organic matter vs. vitreous bulk Manchanito as the noble gas carrier: Evidence from Ne release patterns. Upper panel) Carbon Kα₁ map of one surface of Manchanito showing adhering organic matter. Lower panel) Cumulative Ne release profiles vs. temperature for Manchanito, terrestrial obsidian glass (Matsuda et al., 1989), and acid-resistant residues separated from the carbonaceous chondrites Orgueil and Leoville (Huss et al., 1996). Measured Ne isotope ratios are dominated by the Q-Ne composition reported by Busemann et al. (2000), pointing to the presence of a significant abundance of the Q-carrier in the meteorite residues, probably accompanied by other mineral species. The small releases at low temperatures are evidence that the organic Q-carrier phase and all other components of the residues are thermally retentive, contrasting with the low-temperature release profile displayed by Manchanito.

The Ne release patterns from Manchanito and the Track 41 carriers (Marty et al., 2008) are distinctly different (Fig. 7). Neon devolves from the former over an extended low-temperature range of ~400 to 650 °C. In contrast, a sharp release from the Track 41 carriers occurs within a narrow ~50 °C temperature interval centered around 1200 °C, implying either crystalline melting near this temperature or melting of more refractory minerals fluxed by molten aerogel SiO₂. Synchrotron SXRF assays of the Track 41 bulb wall (Marty et al., 2008) identified mostly refractory crystalline minerals, primarily metal and metal-bearing compounds (~75%) and silicates (~25%). A more detailed TEM study of part of the wall found a wide variety of crystalline silicates (Joswiak et al., 2012). The high temperature required to outgas He and Ne from their Track 41 host grains points to refractory crystalline gas carriers, while the low-temperature release pattern displayed by Manchanito is attributable to its vitreous nature as discussed above. Whatever the Track 41 gas carriers are, they are not amorphous grains like Manchanito.

Manchanito’s origin as late-forming refractory chondrule-like glass, from its composition and O and Al-Mg isotope systematics, does not naturally lead to a scenario where Manchanito’s Q-like noble gases could be acquired, e.g. by energetic particle irradiation. One possibility is that Manchanito solidified rapidly and amorphously from a liquid in the presence of a compositionally Q-like atmosphere. Helium and Ne diffuse very quickly into glass at moderate temperatures (Matsuda et al., 1989). If exposure to ambient Q-gases occurred within the 200–600 °C temperature range of the diffusivity measurements, Q-Ne would be populated throughout Manchanito’s volume in time scales of days to seconds. Following residence in this Q-rich atmosphere, Manchanito must have experienced rapid ejection to a cooler spatial environment in order to preserve its Ne content. Linear extrapolation of Ne diffusivities measured by Matsuda et al. (1989) to lower temperatures suggest that the Q-Ne inventory would be retained for hundreds of Myr or longer at temperatures at or below approximately −55 °C.

Our Al-Mg isotope measurements of Manchanito constrain this Q-gas exposure to have taken place more than 3.2 Myr after CV CAI formation, towards the end of the estimated lifetime of the gaseous solar nebula (Wang et al., 2017). After its formation and Q-gas exposure, Manchanito was likely transported to the comet-forming region in the outer Solar System and accreted, along with organics and ices, into its parent-body comet. Manchanito’s formation age, high-temperature origin, and cometary parent-body are all similar, respectively, to those of the comet Wild 2 particle Iris (Ogliore et al., 2012; Gainsforth et al., 2015). The rocky component of comets may be dominated by high-temperature objects with complicated formation histories like Iris and Manchanito, embedded in extremely volatile ices.

6. CONCLUSIONS

1. Manchanito is a refractory, amorphous interplanetary dust particle with elemental and O isotopic composition consistent with chondrule mesostasis glass, and likely formed in a similar manner.

2. Manchanito formed relatively late, >3.2 Myr after CV CAI formation based on ²⁶Al-²⁶Mg systematics, consistent with other cometary objects that have been measured.

3. After its formation in a high-temperature event, Manchanito was exposed to Q-like gases which facilitated incorporation of large amounts of He and Ne.

4. Manchanito’s He and Ne composition is identical to that measured in Stardust Track 41 from comet Wild 2, indicating the Manchanito (and its host cluster IDP) probably originated in a comet.

5. The source body of Manchanito, likely cometary, is composed of high-temperature objects in intimate contact with organics, volatile ices, and primordial noble gases that have remained unchanged since the formation of the Solar System.