Mineralogy and oxygen isotope systematics of magnetite grains and a magnetite-

dolomite assemblage in hydrated fine-grained Antarctic micrometeorites

Elena DOBRIC�A 1*, Ryan C. OGLIORE2, C�ecile ENGRAND3, Kazuhide NAGASHIMA4, andAdrian J. BREARLEY 1

1Department of Earth and Planetary Sciences MSC03-2040, 1 University of New Mexico, Albuquerque, New Mexico87131-0001, USA

2Department of Physics, Washington University in St. Louis, St. Louis, Missouri 63117, USA3Centre de Sciences Nucl�eaire et de Sciences de la Mati�ere, Universit�e Paris Sud, Universit�e Paris-Saclay, 91405 Orsay Campus,

Saint-Aubin, France4Hawai’i Institute of Geophysics and Planetology, University of Hawai’i at M�anoa, Honolulu, HI 96822, USA

*Corresponding author. E-mail: edobrica@unm.edu

(Received 29 August 2018; revision accepted 02 July 2019)

Abstract–We report the mineralogy and texture of magnetite grains, a magnetite-dolomiteassemblage, and the adjacent mineral phases in five hydrated fine-grained Antarcticmicrometeorites (H-FgMMs). Additionally, we measured the oxygen isotopic compositionof magnetite grains and a magnetite-dolomite assemblage in these samples. Ourmineralogical study shows that the secondary phases identified in H-FgMMs have similartextures and chemical compositions to those described previously in other primitive solarsystem materials, such as carbonaceous chondrites. However, the oxygen isotopiccompositions of magnetite in H-FgMMs span a range of Δ17O values from +1.3& to+4.2&, which is intermediate between magnetites measured in carbonaceous and ordinarychondrites (CCs and OCs). The d18O values of magnetites in one H-FgMM have a ~27&mass-dependent spread in a single 100 9 200 lm particle, indicating that there was alocalized control of the fluid composition, probably due to a low water-to-rock mass ratio.The Δ17O values of magnetite indicate that H-FgMMs sampled a different aqueous fluidthan ordinary and carbonaceous chondrites, implying that the source of H-FgMMs isprobably distinct from the asteroidal source of CCs and OCs. Additionally, we analyzed theoxygen isotopic composition of a magnetite-dolomite assemblage in one of the H-FgMMs(sample 03-36-46) to investigate the temperature at which these minerals coprecipitated. Wehave used the oxygen isotope fractionation between the coexisting magnetite and dolomiteto infer a precipitation temperature between 160 and 280 °C for this sample. This alterationtemperature is ~100–200 °C warmer than that determined from a calcite-magnetiteassemblage from the CR2 chondrite Al Rais, but similar to the estimated temperature ofaqueous alteration for unequilibrated OCs, CIs, and CMs. This suggests that the sample 03-36-46 could come from a parent body that was large enough to attain temperatures as highas the OCs, CIs, and CMs, which implies an asteroidal origin for this particular H-FgMM.

INTRODUCTION

Aqueous alteration is one of the most widespreadprocesses that affected primitive solar system materialssuch as chondritic meteorites (carbonaceous andordinary chondrites—CCs and OCs), micrometeorites(Antarctic micrometeorites—AMMs), and interplanetary

dust particles (IDPs) (Germani et al. 1990; Bradley andBrownlee 1991; Keller et al. 1992; Kurat et al. 1992;Rietmeijer 1996; Brearley 2006; Zolensky et al. 2008a;Noguchi et al. 2017). Previous studies have shownseveral lines of chemical and mineralogical evidence forthe alteration of AMMs. First, aqueous alteration by thewater in Antarctica (for the collection from blue ice

Meteoritics & Planetary Science 1–17 (2019)

doi: 10.1111/maps.13366

1 © The Meteoritical Society, 2019.

fields) such as depletions (within a factor of two of thechondritic abundance) of Ca, S, and Ni in Cap-

Prudhomme AMMs (collected in ice) compared to

Concordia AMMs (from snow), which show a

chondritic abundance, and the absence of Fe-sulfides

and Fe-Ni metal minerals in Cap-Prudhomme AMMs,

whereas they are abundant in Concordia AMMs

(Duprat et al. 2007). Second, the aqueous alteration

took place on the parent body, indicated by the presence

of phyllosilicates and redeposition of mobile elements

(Duprat et al. 2007; Dobric�a et al. 2009, 2012).

However, up to now, only a limited number of

mineralogical, chemical, or isotopic studies have been

conducted on the effects of aqueous alteration that took

place on the AMMs’ parent bodies (Kurat et al. 1992,

1994; Engrand and Maurette 1998; Noguchi et al. 2017).

Phyllosilicates were identified for the first time in AMMs

in the mid-1990s (Kurat et al. 1994), but aqueous

alteration processes in AMMs were only vaguely

mentioned for the first time a decade later (Noguchi

et al. 2002). At that time, the abundance of

phyllosilicate-rich AMMs was estimated to be <1%(Noguchi et al. 2002). Currently, the abundance of

AMMs that have experienced aqueous alteration is still

poorly constrained and these AMMs have not received

the same attention as anhydrous AMMs. This is

primarily due to their micrometer size that makes them

difficult to study and the assumption that these

aqueously altered AMMs originated from similar

asteroidal bodies to the CI, CM, and CR carbonaceous

chondrites (Noguchi et al. 2017), and therefore did not

merit detailed studies. Furthermore, there was generally

a salient goal to focus on the most fluffy, fragile, fine-

grained pristine particles that have experienced the least

parent body aqueous alteration or terrestrial weathering,

that is, such as the AMM particles collected in snow at

the Concordia station (Duprat et al. 2007, 2010;

Dobric�a et al. 2012). This bias and the limited number

of samples have hindered the development of a new,

separate class of micrometeorites in the nomenclature

scheme of micrometeorites (Genge et al. 2008).

Nevertheless, the hydrated fine-grained AMMs were

included in the unmelted FgMMs category of

micrometeorites (Genge et al. 2008). Even IDPs, the

smallest and most fine-grained primitive solar system

materials available for laboratory investigation, have a

designated class for particles that show clear evidence of

aqueous alteration. This class is named chondritic

smooth (CS) IDPs (Bradley 2014). Therefore, following

this precedent for all other solar system samples,

hereafter, we classified all AMMs showing evidence of

aqueous alteration in a new class called hydrated fine-

grained AMMs (H-FgMMs).

The oxygen isotopic composition of magnetite(Fe3O4) in chondrites has been used successfully toprovide constraints on the water reservoir that alteredthese meteorites. Magnetite in H-FgMMs provides theopportunity to determine if H-FgMMs come from parentbodies with the same composition water reservoir as CCs,or might provide evidence that they came from a distinctreservoir, since known asteroidal materials have specificcompositions, but are limited by our imperfect samplingof the asteroid belt. Therefore, magnetite in H-FgMMsshould show an oxygen isotopic signature from theirlikely formation by aqueous processes, in either cometaryor asteroidal parent body(ies). Asteroidal water may bedistinguishable from cometary water in its oxygenisotopic composition. The oxygen isotopic composition(Δ17O = d17O�0.52 9 d18O) of water from an asteroidalsource could be between ~+1.7& (CI, avg. valuecalculated from Rowe et al. 1994) and ~�1.8& (CO, avg.value calculated from Doyle et al. 2015) if their parentbody is CC-like (similar to CS IDPs measured so far byAl�eon et al. 2009) or Δ17O ~+6.6& if their parent body isOC-like (Choi et al. 1998). Any H-FgMMs withsignificantly different Δ17O values would be evidence thatthese particles formed from a distinct water reservoir thatwas not sampled by these meteorite groups. If Jupiter-family comets (JFCs) accreted outer solar nebular water,the oxygen isotopic composition of their secondaryphases could be similar to the cosmic symplectites (COSs)from Acfer 094 (Sakamoto et al. 2007; Seto et al. 2008).

This study aims to improve our understanding ofthe origin of aqueously altered AMMs, through detailedcharacterization of their secondary mineralogy andmeasurements of their oxygen isotopic compositions. Inparticular, we have focused on magnetite and coexistingcarbonates, since these minerals are useful recorders ofparent body conditions, such as (1) the isotopiccomposition of the water reservoir, which interactedwith H-FgMMs and (2) the temperature at whichmagnetite coprecipitated with other secondary phases.

SAMPLES AND METHODS

Five unmelted H-FgMMs (94-4B-21, 99-12-45, 03-26-59, 03-36-46, and 07-13-01) were analyzed in thisstudy (Fig. 1). The first two samples were collectedfrom blue ice fields at Cap-Prudhomme (Maurette et al.1991) and the last three from the pristine Concordiasnow during the 2002 and 2006 campaigns (Dupratet al. 2007). In this study, the H-FgMMs were analyzedas whole particles collected from ice or snow, except forsample 07-13-01, which is a fragment of a largermicrometeorite. The H-FgMMs vary in size from25 9 50 lm (07-13-01) to 100 9 200 lm (94-4B-21).

2 E. Dobric�a et al.

Fig. 1. Backscattered (a, b, d–f) and secondary electron (c) images showing the hydrated fine-grained Antarctic micrometeorites(H-FgMMs) analyzed in this study (94-4B-21, 99-12-45, 03-26-59, 03-36-46, and 07-13-01). Framboidal magnetite was identifiedin all five particles. Additionally, plaquette and anhedral magnetite are present in particles 94-4B-21 (a) and 07-13-01 (d). f)Particle 03-36-46 contains one magnetite-dolomite assemblage (details of the white square in [e]. S = sulfide, Mt = magnetite,Do = dolomite, En = enstatite, P = Ca-phosphate, Fe-ox = iron oxide containing variable amount of silica.

Mineralogy and oxygen isotope systematics of micrometeorites 3

The initial size before the fragmentation of themicrometeorite 07-13-01 was 90 9 130 lm. Magnetitewas identified in polished blocks of five different H-FgMMs (94-4B-21, 99-12-45, 03-26-59, 03-36-46, and 07-13-01). These H-FgMMs have representative textures andchemical compositions, typical of unmelted AMMs, since(1) they contain phyllosilicates; (2) they are not depletedin volatile elements and mineral phases like Fe-Ni sulfidesand carbonates; and (3) they lack vesicles, which arecommon in partially melted scoriaceous AMMs (Dobric�aet al. 2008, 2009). The polished blocks were firstcharacterized by scanning electron microscopy (SEM)using backscattered electron imaging on a FEI Quanta3D Dualbeam� field emission gun (FEG) SEM/focusedion beam (FIB) operating at 30 kV. Then, electrontransparent sections of eight different regions containingmagnetite were prepared by the in situ FIB techniqueusing the FEI Quanta 3D Dualbeam� FIB instruments atthe University of New Mexico and WashingtonUniversity in St. Louis. We prepared FIB sections of fourH-FgMMs (99-12-45, 94-4B-21, 03-26-59, and 07-13-01).No FIB sections were made in the particle 03-36-46 sincethe magnetites are intimately mixed with carbonates,which will be measured for 53Mn-53Cr isotopes in a futurestudy. A platinum protective layer (2 lm in thickness and2 lm in width) was deposited on top of the region ofinterest to avoid gallium primary ion beam damageduring the FIB sample preparation. The section wastransferred to Cu transmission electron microscopy(TEM) half grids with an Omniprobe 200micromanipulator. The final ion milling of the 2 lm thicksection to electron transparency was carried out with thesample attached to the TEM grid. The final thinningstages were performed at 5 kV with a current of 10 pA.Each FIB section was studied using a variety of TEMtechniques, including bright-field TEM imaging, high-angle annular dark-field scanning TEM, selected areaelectron diffraction, and energy-dispersive X-rayspectroscopy (EDS). All imaging and analysis werecarried out at 200 kV on a JEOL 2010F FEG-TEM/Scanning TEM. In situ energy-dispersive X-ray analyseswere obtained using an Oxford Instruments AZtec EDSsystem equipped with an Oxford X-MaxN 80T SDD EDSdetector (80 mm2 sensor).

Secondary ion mass spectrometer (SIMS) analyseswere carried out using a Cameca ims-1280 ionmicroprobe in the W. M. Keck CosmochemistryLaboratory, at the University of Hawai’i at M�anoa. Weused a ~25 pA Cs+ primary beam focused to ~2 lm withtotal impact energy of 20 keV. Masses 16O�, 17O�, and18O� were measured simultaneously in multicollectionmode on a Faraday cup and two electron multipliers,respectively. Prior to the SIMS measurements, theregions of interest were marked/sputtered using the ion

beam from the FEI Quanta 3D Dualbeam� FIBinstrument. The mass-resolving power for 17O� was~5500 to minimize the contribution from 16OH�. We alsomeasured 16OH� in each measurement cycle byelectrostatically deflecting secondary ions (using DSP2-x)without changing magnetic field, to quantify anycontribution of this interference to 17O (<0.5&). Data formagnetite and dolomite were corrected for instrumentalmass fractionation using terrestrial magnetite (Kusakabeet al. 2004) and UW6250 dolomite (�Sliwi�nski et al. 2015)standards, respectively. After SIMS measurements, thesputtered craters were investigated in the SEM toascertain the clean isotopic measurement of the desiredphases without having a mixing of adjacent phases.

Quantitative bulk chemical analyses of the larger(>10 lm) phases were determined by wavelength-dispersive spectrometry (WDS) using two different JEOLJXA-8200 electron microprobes (EPMA), operating at anacceleration voltage of 15 kV and a 20 nA beam current.The measurements were performed at the University ofNew Mexico and Washington University in St. Louis.The detection limit for the EPMA WDS measurements isestimated to be <0.05 wt%. In order to avoid Na loss,analyses were obtained by broad-beam analysis using a10 lm beam diameter. A variety of mineral standardsincluding orthoclase (K), albite (Al, Na), chromite (Cr,Fe), forsterite (Mg), spessartine (Mn), diopside (Si, Ca),and rutile (Ti) were used as standards. Data reductionwas carried out using standard ZAF matrix correctionprocedures.

RESULTS

Petrography of the Hydrated Micrometeorites

The SEM images of the five H-FgMMs (polishedblocks) show coarse-grained minerals embedded in a veryhomogeneous, fine-grained material composed ofphyllosilicates (Table 1; Fig. 1). These minerals areenstatite, sulfides, Ca-phosphates, iron silicates, possiblycronstedtite, dolomite, and magnetite (Fig. 1). Table 1shows the chemical compositions of the coarse-grainedminerals, which were measured using the electronmicroprobe. The enstatite grain (En88) identified in thesample 07-13-01 is ~15 lm in length and has an anhedralmorphology. The iron silicates, possibly cronstedtite (seeTable 1), were identified using EMPA in the center ofparticle 94-4B-21 and at the edge of sample 99-12-45(Fig. 1).

Magnetite is present in a variety of differentmorphologies including framboids, plaquettes, andspherules, as well subhedral to anhedral grains embeddedin fine-grained materials (Fig. 1). The magnetite grainsvary in size from ~200 nm up to 7 lm. The framboidal

4 E. Dobric�a et al.

magnetites always occurs in irregular-shaped clusters,with sizes up to ~25 lm in length (Fig. 1). The magnetitesin irregular-shaped clusters have sizes between 5 and20 lm. Sample 03-36-46 contains multiple Ca-phosphategrains (Table 1), from a few hundreds of nanometers to~3 lm in size (see white arrow in Fig. 1e). Additionally,we identified a magnetite-dolomite assemblage in 03-36-46 (Figs. 1e and 1f). The size of the magnetite-dolomiteassemblage is about 13 9 15 lm. The dolomite aggregateis composed of multiple grains with sizes between ~1.5and ~3.8 lm (Fig. 1f).

Detailed TEM observations confirm that themagnetite has different morphologies (framboidal,plaquettes, subhedral to anhedral, and spherules) (Fig. 2).Sample 99-12-45 is the only one that contains magnetitewith all four morphologies (Fig. 2d); however, theplaquette discs are the dominant morphology in thissample. The magnetite framboids have different (1)morphologies varying from euhedral to anhedral and (2)apparent diameter varying from a few nm up to 1.7 lm.Vesicles were identified in the framboidal magnetite grains

(94-4B-21) that are in contact with a continuous, <1 lm-thick magnetite rim at the exterior of the sample causedby atmospheric entry heating (Fig. 3b). The plaquettediscs from the barrel-shaped stack shown in Fig. 2d varyin width from 200 to 300 nm and in length from 1.9 to3 lm. The adjacent plaquette discs are found in contactwith each other and only rare, elongated pore spaces(~200 nm in width) were observed in between these discs.The subhedral to anhedral magnetites are found either incontact with magnetites with different morphologies(~2 9 2 lm, Fig. 2d) or isolated, embedded in the fine-grained materials (~2.5 9 3 lm, Fig. 3a). The spherulesshow smooth, well-rounded exteriors with no radiatingtextures. The largest sphere identified in the FIB sectionsis about 4 lm in diameter (Fig. 2d). The EDS/TEManalyses of the framboidal magnetite grains in the sample07-13-01 and the magnetite grains from 94-4B-21 thatwere identified in the rim modified by the atmosphericentry heating show that they contain minor quantities ofMgO (07-13-01: up to 5.2 wt%, N = 12, avg. 2.7 wt%,SD 1.3; 94-4B-21: up to 3.3 wt%, N = 7, avg. 1.8 wt%,SD 0.9) and SiO2 (07-13-01: up to 6.7 wt%, N = 12, avg.3.5 wt%, SD 1.4; 94-4B-21: up to 3.2 wt%, N = 7, avg.0.8 wt%, SD 0.5). No other minor elements higher thanthe detection limit (<0.5 wt%) were measured in othermagnetite crystals.

Figure 3 shows one representative FIB section of eachH-FgMM analyzed in this study. We observe thatmagnetite is associated with Mg- and Fe-rich coarse-grained and fine-grained phyllosilicates (Figs. 3d–g and 4,respectively) in all the samples analyzed. The chemicalcompositions of the phyllosilicates (serpentine andsmectites were identified by EDS) are shown in Table 2(Fig. 4). Sample 03-26-59 contains a coronal, “flower”-like,microtexture consisting of coarse-grained phyllosilicates(serpentine, Figs. 3a and 3b; Table 2). The materialsaround the framboidal magnetites in sample 07-13-01 arevesicular, fine-grained, and polycrystalline as shown in thediffraction ring pattern indicating that they are thermallymodified Fe-rich phyllosilicates (15.5 wt% FeO, Fig. 3g).However, the chemical composition of thermally modifiedphyllosilicates is similar to that of the phyllosilicates and tothe composition of the materials identified around themagnetites in sample 94-4B-21 (Table 2). Additionally, theDF-STEM images show atmospheric entry modification ofthe AMM and the formation of micron-sized vesicles andthe magnetite shell rim in sample 94-4B-21 (Fig. 3).

Oxygen Isotopes of Hydrated Micrometeorites

Oxygen isotope compositions of magnetite and onedolomite were measured in situ with the UH Cameca ims-1280 ion microprobe. These data are listed in Table 3 andplotted in Fig. 5. The Δ17O values of magnetites in the H-

Table 1. Electron microprobe analyses (in atom% andwt% oxide) of the phases associated with the magnetite.

Sample 03-26-59 99-12-45 94-4B-21 03-36-46

Mineral F-g mx Cronst.* Cronst.* Phosphate DolomiteFigure 1c 1b 1a 1e 1f

Si 19.68 8.28 3.83 0.36 3.19

Al 1.40 0.45 0.97 0.03 0.45Cr 0.13 0.26 0.12 0.02 0.01Fe 2.73 29.68 33.38 0.30 5.11

Mn 0.00 0.02 0.01 0.13 2.47Mg 15.08 0.38 0.06 0.41 15.67Ca 0.03 0.22 0.00 23.40 15.67

Na 0.25 0.11 0.32 0.70 0.22K 0.06 0.03 0.04 0.00 0.00S 0.00 0.00 0.00 0.00 1.00P 0.00 0.46 0.17 13.91 0.28

Ni 0.00 0.04 0.16 0.00 0.00O 60.64 59.86 59.38 60.60 54.40SiO2 56.56 17.91 8.04 0.85 1.31

Al2O3 3.41 0.82 1.73 0.07 0.08Cr2O3 0.95 1.42 0.65 0.13 0.05FeO 9.39 76.74 83.86 0.85 3.56

MnO 0.02 0.05 0.03 0.38 3.73MgO 29.08 0.56 0.09 0.66 18.02CaO 0.07 0.45 0.00 52.17 24.57

Na2O 0.07 0.45 0.00 0.86 0.06K2O 0.13 0.05 0.07 0.00 0.00SO3 0.00 0.57 4.34 4.78 0.23P2O5 0.00 1.18 0.43 39.25 0.08

NiO 0.00 0.12 0.42 0.00 0.00Total 99.98 100.32 99.66 100.00 51.69

Cronst.* = iron silicate mineral, possibly cronstedtite.

Mineralogy and oxygen isotope systematics of micrometeorites 5

FgMMs range from +1.3& to +4.2& (Fig. 5). Themagnetite d18O values span a range of ~+30& (d18O:�12.5& to 17.4&). Generally, the magnetites measuredshow a large range within individual H-FgMMs. Forexample, the magnetites measured in sample 94-4B-21span the largest range in d18O: �9.8& to +17.4&. Theoxygen isotope compositions of magnetite also vary as afunction of the morphology (see Table 4). The d18Ovalues measured in framboidal and subhedral magnetiteare different than the values measured in the magnetitespherules from the same sample 99-4B-21 (framboids andsubhedral: avg. 1.6& d18O after 1.6& and spherules: avg.�9.1& d18O; see Table 4 for more details). Additionally,we observe that the average d18O of the spherulesidentified in 99-4B-21 (avg. �9.07& d18O) is similar tothe average d18O of the plaquettes measured in sample99-12-45 (avg. �8.44& d18O). Furthermore, the oxygenisotopic compositions of different morphologies are notthe same in different samples, there is not consistency in

the isotopic compositions (99-12-45: plaquettes avg.1.74& Δ17O and 99-4B-21: framboids, subhedral, andspherical magnetite varies between 2.61& Δ17O to 3.47&Δ17O; see Table 4 for more details). We measured bothmagnetite and dolomite in the magnetite-dolomiteassemblage in 03-36-46. Their Δ17O values are similarwithin uncertainty (Δ17O ~1.6 to ~3.0&), but themagnetite (2.6& d18O) and dolomite (20.6& d18O) d18Ovalues are different by 18�3& (2r).

DISCUSSION

Despite the extensive documentation of secondaryphases in chondritic meteorites (carbonaceous andordinary chondrites), the effects of aqueous alterationhave been largely overlooked in the most abundantextraterrestrials materials that fall on Earth,micrometeorites and IDPs (Love and Brownlee 1993;Duprat et al. 2005). Therefore, this is the first study

Fig. 2. Dark-field scanning transmission electron microscopy (DF-STEM) images of the different morphologies of magnetitesidentified in the hydrated fine-grained Antarctic micrometeorites (H-FgMMs) studied (a—framboidal, b—plaquettes, c—spherules, d—subhedral). Most adjacent plaquette discs have similar sizes, with the exception of the adjacent discs shown in (d)and indicated by the black arrow. Fram = framboids, Subh = subhedral, Shp = spherules, Plaq = plaquettes.

6 E. Dobric�a et al.

Fig. 3. Dark-field scanning transmission electron microscopy (DF-STEM) images showing the mineralogy and textures of onerepresentative focused ion beam (FIB) section of each Antarctic micrometeorite (AMM) analyzed by transmission electronmicroscopy (94-4B-21, 99-12-45, 03-26-59, and 07-13- 01). a, b) Framboidal and anhedral (top, right) magnetite in the sample 94-4B-21. The DF-STEM images show the atmospheric entry modification of the AMM and the formation of micron-sized vesiclesand the magnetite shell rim. The white square from (a) outlines the area shown in (b). c) DF-STEM mosaic of one of the FIBsections made in sample 99-12-45 showing the association of magnetite plaquettes and coarse-grained phyllosilicates.d) Higher magnification DF-STEM image of the white square from (c) showing the association of magnetite (left, top corner)and coarse-grained phyllosilicates. e, f) Detailed DF-STEM image of the coarse-grained phyllosilicates similar to the onesdescribed by Blinova et al. (2014) in Tagish Lake. The composition of these coarse-grained phyllosilicates is shown in Table 2. g,h) DF-STEM mosaic of the FIB section made in sample 07-13-01 showing the presence of framboidal magnetite associated withfine-grained phyllosilicates. h) Higher magnification DF-STEM image of the white square from (g) showing the presence ofmicron-sized vesicles. The sample was heated during atmospheric entry, which is indicated by the presence of micron-sizedvesicles. However, the atmospheric-entry heating was limited since the phyllosilicates are still crystalline (see the selected areaelectron diffraction (SAED) ring pattern, [g]. C-g phyllo = coarse-grained phyllosilicates, F-g phyllo = fine-grainedphyllosilicates, Mt = magnetite, Pt = platinum.

Mineralogy and oxygen isotope systematics of micrometeorites 7

reporting a combined mineralogical and oxygen isotopicstudy of magnetites from H-FgMMs.

Formation Processes of Secondary Phases

Magnetite with different morphologies was identifiedin five H-FgMMs. Several magnetite grains identified intwo H-FgMMs (94-4B-21, 07-13-01) contain minorquantities of SiO2 and MgO. These minor elementcontents could be explained by the fact that theseparticles were heated during atmospheric entry, sincethese magnetites were either in contact with the fusioncrust (Figs. 3e and 3f, H-FgMM 94-4B-21) or in theinterior of sample 07-13-01 that was slightly heated at thetime of atmospheric entry, as indicated by the presence ofmicron-sized vesicles (Figs. 3g and 3h). However, most ofthe magnetites analyzed in this study have escapedheating during the atmospheric entry, as indicated by thepresence of phyllosilicates and sulfides around themagnetite crystals.

Previous studies have identified magnetite in allprimitive solar system materials including chondriticmeteorites (CCs and OCs), AMMs, IDPs, and even incometary samples (81P/Wild 2) returned by the Stardustmission (Kerridge et al. 1979; Schramm et al. 1989; Roweet al. 1994; Genge et al. 1997, 2008; Noguchi et al. 2002,2017; Brearley 2006; Zolensky et al. 2008a; Stodolna et al.2012; Blinova et al. 2014; Doyle et al. 2015; Jilly-Rehaket al. 2018). However, not all the aqueously formedmagnetite morphologies are present in all these differenttypes of samples. For example, in the case of OCs,magnetite forms by a very different mechanism, that is, bythe replacement of primary Fe,Ni metal (Krot et al. 1997)and framboidal and plaquette magnetite morphologies donot occur in OCs. Magnetite with different morphologieswas only identified in CI chondrites and Tagish Lake(Greshake et al. 2005; Chang et al. 2016). In the other

Fig. 4. Ternary diagram (Si-Fe-Mg, in atom% element)showing the compositions of the Mg- and Fe-rich coarse-grained and fine-grained phyllosilicates (serpentine andsmectites were identified by EDS) identified in association withthe magnetite in all the focused ion beam sections analyzed.Three of the micrometeorites contain Mg-rich phyllosilicates(07-13-01—orange circles, 03-26-59—red circles, and 94-4B-21—blue circles). Only sample 99-12-45 contains Fe-richphyllosilicates (green circles). Table 2 shows the major andminor element compositions (in atom% and wt% oxide)determined by energy-dispersive X-ray spectroscopy (EDS) ofthese coarse-grained (c-g) and fine-grained (f-g) phyllosilicates(phyllo.) associated with the magnetite. Additionally, we showthe chemical compositions of the phyllosilicates identified inthe matrix of Bells (CM chondrites—light gray, Brearley1995), Tagish Lake (C2-ungrouped—dark gray, Zolenskyet al. 2002), Semarkona (OC—black, Alexander et al. 1989),and AMMs (area and squares—line pattern, Sakamoto et al.[2010] and Noguchi et al. [2017]).

Table 2. Major and minor element compositions (inatom% and wt% oxide) determined by energy-dispersive X-ray spectroscopy of the coarse-grained (c-g)and fine-grained (f-g) phyllosilicates (phyllo.) associatedwith the magnetite.

Sample 03-26-59 99-12-45 94-4B-21 07-13-01

MineralC-gserpentine

C-gsmectite F-g phyllo F-g phyllo

Figure 3e,f 3c,d 3a 3g

N 10 SD 3 SD 3 SD 3 SD

Si 16.4 0.6 13.7 0.8 17.0 0.4 16.4 0.7Al 1.1 0.2 0.8 0.1 1.3 0.0 1.8 0.1Cr 0.1 0.0 0.2 0.1 0.2 0.0 0.2 0.1

Fe 2.2 0.2 10.4 1.1 3.3 0.2 4.4 0.4Mn 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0Mg 14.4 0.8 0.0 0.3 14.0 0.4 14.0 0.3Ca 0.0 0.0 0.3 0.0 0.0 0.0 0.0 0.0

Na 0.5 0.3 0.0 0.0 0.7 0.4 0.4 0.1K 0.0 0.0 0.1 0.1 0.0 0.0 0.0 0.0S 0.0 0.0 0.3 0.3 0.1 0.1 0.2 0.0

P 0.0 0.1 0.3 0.3 0.0 0.0 0.1 0.1Ni 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0O 65.0 1.0 71.9 0.7 63.1 0.1 62.2 0.9

SiO2 53.7 1.1 46.8 3.9 52.0 1.5 48.8 0.7Al2O3 3.0 0.4 2.3 0.5 3.3 0.1 4.5 0.4Cr2O3 0.7 0.4 2.1 0.7 1.4 0.1 1.4 0.6

FeO 9.4 0.6 42.5 3.5 13.0 0.8 15.5 1.0MnO 0.0 0.0 0.1 0.1 0.1 0.1 0.1 0.0MgO 31.7 1.4 2.3 0.7 29.2 0.7 27.9 1.2CaO 0.0 0.1 0.8 0.1 0.1 0.1 0.1 0.1

Na2O 0.8 0.4 0.0 0.0 1.0 0.5 0.6 0.3K2O 0.1 0.1 0.2 0.2 0.1 0.0 0.0 0.0SO3 0.1 0.2 1.4 1.3 0.3 0.3 0.7 0.1

P2O5 0.2 0.3 1.4 1.3 0.1 0.1 0.4 0.3NiO 0.0 0.0 0.0 0.0 0.1 0.2 0.1 0.1

N = number of analyses, SD = standard deviation.

8 E. Dobric�a et al.

classes of chondrites (e.g., CV, CO, CK, etc.), magnetitedoes not occur as plaquettes (Chang et al. 2016).Therefore, the question is what are the limiting conditionsfor the formation of magnetite with these particularmorphologies? Kerridge et al. (1979) first proposed thatmagnetite in CI chondrites formed on the meteoriteparent body from an aqueous gel-like phase containingcolloidal crystals. Recent studies suggest that theformation of colloidal crystals implies several restrictionson the growth conditions, such as (1) supersaturation ofthe aqueous fluid; (2) the aqueous fluid must have beenconfined in small voids, in which colloidal crystallizationtakes place; and (3) the nucleation must have taken place

at the same moment in a short period of time, with nofurther nucleation occurring during subsequent growth(Nozawa et al. 2011; Kimura et al. 2013). The lastrestriction points to a sudden change of the conditions inthe parent body, for instance, a change in temperature orpressure, which rapidly altered the supersaturation of thesolution (Nozawa et al. 2011). Therefore, the lack offramboidal and plaquette magnetite morphologies in OCscould be explained by the absence of any of theserestrictions on the OCs’ parent bodies. In AMMs,previous studies indicate that the abundance offramboidal-bearing AMMs is ~4% (Kurat et al. 1992;Genge et al. 2008). Therefore, the presence of framboidal

Table 3. Oxygen isotopic composition of magnetites (mt) and one dolomite (do) measured in five hydrated fine-grained Antarctic micrometeorites (H-FgMMs).

Samples Measurements d18O d18O 2 r d17O d17O 2 r Δ17O Δ17O 2 r

99-4B-21 99-4B-21_01 �2.68 2.72 0.49 1.52 1.89 1.3099-4B-21_02 �3.89 2.72 �0.48 1.52 1.54 1.30

99-4B-21_03 5.70 2.72 6.20 1.52 3.24 1.3099-4B-21_04 �8.94 2.72 �3.35 1.52 1.29 1.3099-4B-21_05 �0.74 2.72 2.18 1.52 2.57 1.3099-4B-21_06 �2.15 2.72 1.94 1.52 3.05 1.30

99-4B-21_07 17.39 2.72 12.26 1.52 3.22 1.3099-4B-21_08 9.99 2.42 7.78 1.53 2.59 1.4399-4B-21_09 7.04 2.42 6.10 1.53 2.44 1.43

99-4B-21_10 �1.50 2.42 2.08 1.53 2.86 1.4399-4B-21_11 4.66 2.42 5.31 1.53 2.88 1.4399-4B-21_12 �3.35 2.42 1.84 1.53 3.58 1.43

99-4B-21_13 �0.93 2.42 1.98 1.53 2.47 1.4399-4B-21_14 �8.62 2.42 �0.57 1.53 3.91 1.4399-4B-21_15 �10.72 2.42 �2.07 1.53 3.51 1.43

99-4B-21_16 �0.51 2.42 3.39 1.53 3.66 1.4399-4B-21_17 �8.04 2.42 �1.42 1.53 2.76 1.4399-4B-21_18 �8.20 2.42 �1.23 1.53 3.03 1.4399-4B-21_19 �9.78 2.42 �0.94 1.53 4.14 1.43

99-4B-21_20 6.64 2.42 6.42 1.53 2.97 1.4399-4B-21_21 8.85 2.42 7.80 1.53 3.20 1.4399-4B-21_22 �6.40 2.42 �0.34 1.53 2.99 1.43

99-4B-21_24 �2.82 2.42 2.14 1.53 3.60 1.4399-12-45 99-12-45_01 �12.53 2.72 �4.52 1.52 2.00 1.30

99-12-45_02 �11.87 2.72 �4.38 1.52 1.80 1.30

99-12-45_03 �10.70 2.72 �4.08 1.52 1.48 1.3099-12-45_04 �3.88 2.72 �0.55 1.52 1.46 1.3099-12-45_05 �4.87 2.72 �1.04 1.52 1.49 1.30

99-12-45_06 �4.69 2.72 �0.88 1.52 1.56 1.3099-12-45_07 �10.52 2.72 �3.09 1.52 2.38 1.30

03-26-59 03-26-59_02 14.59 2.42 9.08 1.53 1.50 1.4303-26-59_03 �10.70 2.42 �1.35 1.53 4.21 1.43

03-26-59_04 10.44 2.42 9.45 1.53 4.02 1.4303-26-59_5 7.43 2.42 7.25 1.53 3.39 1.43

03-36-46 03-36-46 dolomite 20.6 2.37 12.3 2.36 1.63 1.67

03-36-46 magnetite 2.6 1.71 4.4 1.18 3.01 1.6007-13-01 07-13-1_01 �12.32 2.42 �2.84 1.53 3.57 1.43

07-13-1_02 �2.79 2.42 2.34 1.53 3.79 1.43

All values reported in permil deviation from SMOW (&).

Mineralogy and oxygen isotope systematics of micrometeorites 9

magnetite was used to define AMMs that have a similarorigin as CI chondrites but not necessarily the samesource (Kurat et al. 1992; Genge et al. 2008). Our studyindicates that the magnetite grains identified in H-FgMMsshow similar textures to those described previously in CIs,indicating that similar processes and conditions tookplace on the H-FgMMs’ parent body.

Phyllosilicates are ubiquitous minerals in mostextraterrestrial samples, including those of asteroidal andcometary origins, although no evidence for phyllosilicateshas been reported in comet 81P/Wild 2 samples (Tomeokaand Buseck 1984; Sandford and Walker 1985; Mauretteet al. 1990; Thomas et al. 1990; Zolensky and Lindstrom1992; Kurat et al. 1994; Genge et al. 1997; Noguchi et al.2002; Brearley 2006; Lisse et al. 2006; Zolensky et al.

2008b; Dobric�a et al. 2009; Sakamoto et al. 2010). Theyare indicative of parent body aqueous alteration (Brearley2006). We have identified fine- and coarse-grained Mg-rich and Fe-rich phyllosilicates associated with magnetitein all FIB sections made in H-FgMMs. Previous studiesindicate that phyllosilicates in AMMs collected fromsurface snow near the Dome Fuji Station are either Mg-rich (Sakamoto et al. 2010) or Fe-rich (Noguchi et al.2017). Noguchi et al. (2017) concluded that the presenceof phyllosilicates in AMMs and their absence among thereturned samples of comet 81P/Wild 2 suggest thatAMMs may have been derived from porous icy asteroidssuch as active asteroids, as well as P- and D-typeasteroids, rather than comets. However, although noevidence for phyllosilicates has been reported in comet

Fig. 5. Comparisons between d18O values and Δ17O values of the magnetite from the five AMMs (this work 94-4B-21, 99-12-45,03-26-59, 03-36-46, and 07-13-01) and one dolomite (03-36-46*, aquamarine circle, this work), OCs (gray squares, Choi et al.1997, 1998, 2000), IDPs (black triangles, Al�eon et al. 2009), carbonaceous chondrites CO, CR, CV with gray symbols (Doyleet al. 2015; Jilly-Rehak et al. 2018), CM, and CI (Rowe et al. 1994). The oxygen isotopic compositions of magnetites in CM andCI chondrites have been determined for whole-rock and separated components compared with the other data that were measuredin situ using secondary ion mass spectrometer (SIMS). Framboidal magnetites (fram—blue circles) measured in the same particle94-4B-21 have higher average d18O (avg. +1.9&) than the spherule magnetites (sph—yellow circles with black lines, d18O = avg.�9.1&). The subhedral magnetites have similar d18O values (subh—yellow circles with red lines, d18O = avg. 1.3&) to theframboidal magnetites measured in the same particle (94-4B-21). Uncertainties are 2r.

Table 4. Oxygen isotopic composition of magnetites as a function of their morphologies.

Sample Morphologies d18O SD d17O SD Δ17O SD

99-4B-21 Fram 1.65 8.17 3.44 4.51 2.58 0.67

Subh 1.66 6.38 4.34 3.10 3.48 0.43Sph �9.07 1.15 �1.24 0.56 3.47 0.58

99-12-45 Plaq �8.44 3.78 �2.65 1.77 1.74 0.35

Fram = framboids, Subh = subhedral, Shp = spherules, Plaq = plaquettes, SD = standard deviation.

10 E. Dobric�a et al.

81P/Wild 2 samples (Zolensky et al. 2008b), this could beexplained by laboratory simulations of hypervelocitycapture, which showed that phyllosilicates <1 lm aremelted on impact (Noguchi et al. 2007). Therefore, it ishard to determine the origin of AMMs with certaintybased only on the texture and chemical composition ofphyllosilicates.

A coronal, “flower”-like, microtexture consisting ofcoarse-grained serpentine was identified in sample 03-26-59 (Figs. 3a and 3b; Table 2). A similar texture wasdescribed by Blinova et al. (2014) in Tagish Lake, a C2-ungrouped meteorite. They proposed a formation of thismicrostructure by different pulses of fluids with changingcomposition and varying duration. The presence of thismicrotexture could suggest that there are some affinitiesbetween H-FgMMs and the Tagish Lake parent body orthat similar processes took place on the parent body(ies)of AMMs.

Carbonates have been commonly identified incarbonaceous chondrites, micrometeorites, and IDPs(Sandford and Walker 1985; Tomeoka and Buseck 1986;Germani et al. 1990; Rietmeijer 1990; Bradley et al. 1992;Zolensky and Lindstrom 1992; Brearley 2006; Dupratet al. 2007; Al�eon et al. 2009; Dobric�a et al. 2009).Several calcite (CaCO3) grains and one dolomite (CaMg[CO3]2) grain with diameters up to 14 lm have beenidentified in Concordia AMMs, so far (Duprat et al.2007, 2010; Dobric�a et al. 2009, 2012). However,carbonates are rare in Cap-Prudhomme AMMs, due toterrestrial weathering in the ice (Dobric�a et al. 2009).These carbonates are mostly found in fine-grained fluffyand ultracarbonaceous AMMs (Dobric�a et al. 2009;Sakamoto et al. 2010). No systematic mineralogical studyof carbonates in AMMs is currently available. However,the average Ca content of the fine-grained matrix ofConcordia AMMs (Ca = 0.54 atom%) is not depletedcompared to CI (Ca = 0.5 atom%; see Lodders 2003) andCM chondrites (Ca = 0.62 atom%). This depletion in Cacontent was observed in Cap-Prudhomme AMMs (Kuratet al. 1994; Duprat et al. 2007; Dobric�a et al. 2009).Additionally, many AMMs display chemical andmineralogical affinities with CM carbonaceouschondrites, which contain a bulk carbonate abundance ofabout 2.3 vol% (Engrand et al. 1999; Lee et al. 2014;Alexander et al. 2015). Previous studies indicate that inMM, cronstedtite does not coexist with tochilinite, whichis different from CM2 chondrites that experienced weakto moderate aqueous alteration (Noguchi et al. 2007). CSIDPs have systematic Ca and Mg depletions relative toCI abundances and contain a stoichiometric excess ofoxygen consistent with the presence of hydrous phases(Schramm et al. 1989). A similar depletion of Ca in CMmatrices has been attributed to leaching andsequestration into carbonates, and, by analogy, it has

been suggested that CS IDPs have also been processed byaqueous alteration (Schramm et al. 1989; Zolensky et al.1989). Compared with carbonaceous chondrites, IDPcarbonates are typically associated with Fe-Ni sulfiderather than with magnetite or tochilinite (Zolensky et al.2008a). The occurrence of carbonates associated withphyllosilicates provides additional evidence that asignificant number of the CS IDPs have undergonesubstantial aqueous processing (Schramm et al. 1989).Carbonates occur both as separate, euhedral, singlecrystals and as anhedral aggregates (Tomeoka andBuseck 1986; Zolensky and Lindstrom 1992; Zolenskyet al. 2008a; Al�eon et al. 2009). Although submicrometercarbonates (calcite, dolomite, and possibly breunnerite)have been reported in comet 81P/Wild 2 samples(Mikouchi et al. 2007; Wirick et al. 2007; Flynn et al.2008; Tomeoka et al. 2008), these may be contaminants,rather than being indicative of aqueous alteration.Nevertheless, our mineralogical study of H-FgMMsshows that carbonates are associated with magnetite,which indicates that the interaction of aqueous fluids,produced by melting of water ice that accreted togetherwith anhydrous silicate, was common on thesemicrometeorite parent body(ies). The formation of thesephases in the H-FgMMs is consistent with an origin byprecipitation from an aqueous fluid, similar to theirformation in other primitive solar system materials(Fujiya et al. 2013; Alexander et al. 2015; Doyle et al.2015; e.g., Tyra et al. 2012).

Aqueous Fluid Composition

One question that this study aims to answer is whatis the oxygen isotopic composition of the aqueous fluidfrom which H-FgMMs are formed? This is the firstoxygen isotopic study of the secondary phases(magnetite and a magnetite-dolomite assemblage)identified in H-FgMMs.

Bulk oxygen isotopic compositions were reported forhydrated IDPs by Al�eon et al. (2009) and Starkey andFranchi (2013). All of the measured particles had oxygenisotopic compositions that lie on the terrestrialfractionation line, with a few particles that show evidenceof mass fractionation to heavier values (d18O � +30&,d17O � +15&). However, a total of only 14 hydratedIDPs have so far been analyzed for oxygen isotopes. Inmeteorites, the oxygen isotopic composition of water onthe LL chondrite parent body is estimated to bed18O = +21.0&, d17O = +17.5&, Δ17O = +6.6&, inferredfrom SIMS measurements of magnetite (Choi et al. 1998).Recently, Doyle et al. (2015) showed that magnetites inOCs span a range of Δ17O values from +3.1& to +5.9&,with an average of +4.5&. Secondary magnetites in CCshave Δ17O values between �5.2& and +4.4& (see Fig. 5)

Mineralogy and oxygen isotope systematics of micrometeorites 11

(Rowe et al. 1994; Choi et al. 1997, 2000; Doyle et al.2015; Jilly-Rehak et al. 2018). However, the oxygenisotopic compositions of magnetites in CM and CIchondrites have been determined for whole-rock andseparated components compared with the other data thatwere measured in situ using SIMS. Water in comet 1P/Halley was measured by the Giotto spacecraft(d18O = +12 � 75&, 3r) and it is consistent withterrestrial water (Δ17O = 0&), though the uncertaintiesare comparatively large (Eberhardt et al. 1995). Theoxygen isotopic composition of an unambiguoussecondary phase in the comet 81P/Wild 2 samples has notbeen measured so far, owing to the scarcity and small sizesof these phases. There are no spectroscopic measurementsthat constrain the oxygen isotope composition of waterfrom JFCs. Additionally, Sakamoto et al. (2007) reportedthe presence of isotopically unique materials in the matrixof the primitive carbonaceous chondrite Acfer 094(ungrouped), the first evidence for an extremely 17,18O-rich water reservoir (d17,18O = +180&, Δ17O = +86&) inthe early solar system, which is consistent with theexpected composition of water ice from a molecular cloudbased on self-shielding models (Yurimoto and Kuramoto2004). These materials are composed of nanocrystallinemagnetite and iron sulfides with a symplectic texture, andare referred to as COSs (Seto et al. 2008). A survey ofCOSs’ abundance in various chondrites showed that thistype of aggregate is present only in Acfer 094 (Abe et al.2008). Recently, Starkey et al. (2014a, 2014b) described alargely anhydrous IDP enriched in heavy oxygen isotopesthat could be related to COSs in Acfer 094; however, themineralogy is quite distinct from that of COS.

It was postulated that ~90% of small debris particles(~10–300 lm) from the zodiacal cloud originate from thebreakup of JFCs, and the remaining ~10% come fromOort cloud comets and/or asteroid collisions (Nesvorn�yet al. 2010). However, in the cosmic dust collections,approximately half of the particles are dominated byhydrous, secondary phases (Zolensky et al. 2008a) that aretraditionally thought to have formed by aqueous processeson asteroids. The origin of AMMs is not very wellconstrained. Previous studies suggest that some of thecollected AMMs are of cometary origin (e.g.,ultracarbonaceous micrometeorites) and some of themshow similarities with asteroidal materials (Kurat et al.1994; Engrand and Maurette 1998; Genge et al. 2008;Dobric�a et al. 2009, 2012; Duprat et al. 2010). Toconstrain the possible source(s) of AMM, we havemeasured the oxygen isotopic compositions of magnetiteswith different morphologies and one dolomite from fiveH-FgMMs. Their oxygen isotopic compositions span arange of Δ17O from ~0 & to ~+5 & (Fig. 5), covering adifferent range from that measured in other solar systemmaterials. These data could indicate that the aqueous fluid

in the H-FgMMs parent bodies had different, intermediateoxygen isotope compositions compared with the aqueousfluids in the CC and OC parent bodies (see Fig. 5). Acaveat is discussed by Davidson et al. (2014), whichindicated that the Δ17O of magnetite might not representthe original isotopic composition of water, since it is notknown (1) whether significant O isotope exchangeoccurred between water and silicates before and/or duringmagnetite formation, and (2) whether subsequent thermalmetamorphism disturbed the O isotope composition ofmagnetite after formation. However, their study was madeon magnetite formed by the oxidation of pre-existingmetal grains identified in chondrules from the anomalousCK chondrite, Watson 002, and the metamorphosedungrouped chondrites Asuka-881595 that have completelydifferent thermal metamorphic histories than theunmetamorphosed H-FgMMs samples. Additionally,whole-rock Δ17O and nucleosynthetic isotopic variationsfor chromium, titanium, and nickel in meteorites definetwo isotopically distinct populations between CCs andOCs, which indicate that the former originated in theouter part of the solar system, probably beyond the orbitof Jupiter and the ordinary chondrites represent innersolar system material (Warren 2011a, 2011b; Scott et al.2018). Therefore, there is clearly evidence for spatial and/or temporal variations in the source of chondriticmaterials and, by inference, the water they contain orinteracted with. This implies that the source of H-FgMMsis not similar to the asteroidal sources of CCs and OCs.Recently, similar conclusion were obtained by the IR andRaman measurements of hydrated fine-grained AMMs,indicating that the AMMs sampled parent bodies differentfrom CR and CM chondrites (Battandier et al. 2018).

Additionally, our study shows that the d18O values ofmagnetites in AMM 94-4B-21 have a ~27& mass-dependent spread (~15% d17O) in a single 100 9 200 lmparticle. This spread is larger than that for d18O values ofmagnetites measured in the thin sections of CR chondriteRenazzo (~23&; Jilly-Rehak et al. 2018) and CVchondrites (~24&, Allende, Kaba, Mokoia; Choi et al.1997, 2000), which are orders of magnitude larger in size.Huberty et al. (2010) indicated that the instrumental bias ind18O for magnetite varies due to crystal orientation effects.However, our magnetite d18O values span a range of~30&, which is much larger than that of any possibleanalytical artifact (~3& d18O; Huberty et al. 2010).Additionally, Caplan et al. (2015) conducted similar studiesof randomly oriented terrestrial magnetite and chromitesingle crystals to assess the magnitude of such crystalorientation effects; however, no correlation betweenorientation and the degree of fractionation was observed.Jilly-Rehak et al. (2018) concluded that the variations mayhave reflected differences in crater-bottom roughness thatdeveloped during ion probe sputtering; however, they

12 E. Dobric�a et al.

suggested that it seems likely that the variations in oxygenisotopes measured in Renazzo represent true mass-dependent fractionation trends and could indicate changesin temperature and/or fluid composition during magnetiteformation (Jilly-Rehak et al. 2018). We suggest that themuch larger heterogeneity of the oxygen isotopiccomposition (d18O values) in the H-FgMMs could indicatethat there was a localized control of the fluid composition,probably due to a low water-to-rock mass ratio, and notbecause of the orientation or morphologies of themagnetite crystals or the crater-bottom roughness.However, no SIMS analyses of magnetite plaquettes weremade in this study to determine if the crystallographicorientation of these minerals affects the oxygen isotopiccomposition. Additionally, no systematic heavy oxygenisotope enrichment due to the atmospheric entry heatingwas observed in the magnetites analyzed in this study.However, Engrand and Dobric�a (2012) showed that themagnetite identified in the outer shell and measured bySIMS presents systematic heavy oxygen isotopeenrichments that is due to atmospheric entry heating.

The presence of Mg-Fe carbonate in the sample 03-36-46 constrains the upper limit of heating degree during itsatmospheric entry, which is ~600 °C for theirdecomposition (Nozaki et al. 2006). As describe by Jilly-Rehak et al. (2018) in section 4.7, we can use the relativeO-isotope fractionation between dolomite and magnetite(equation 7 from Jilly-Rehak et al. 2018) as ageothermometer for aqueous alteration (see additionaldetails in Jilly-Rehak et al. 2018). Therefore, if we assumethat the magnetite and dolomite formed in equilibrium, asindicated by petrographic evidence, the relative equilibriumoxygen isotope fractionation between dolomite andmagnetite can be used to extract the temperature at whichthese minerals coprecipitated (Zheng 2011; Jilly-Rehaket al. 2018). This assemblage identified in sample 03-36-46contains framboidal and anhedral magnetite intergrownwith dolomite (Fig. 3f; Table 1). The isotopic differencebetween magnetite and dolomite in this assemblage is~18& in d18O (Table 3). This corresponds to theprecipitation temperature that yields a range between 160and 280 °C. These temperatures are higher than thosedetermined for a calcite-magnetite assemblage from theCR2 chondrite, Al Rais (~60 °C, Jilly-Rehak et al. 2018)but overlap with the range estimated for the aqueousalteration of CM chondrites (~20–300 °C, Clayton andMayeda 1999; Nakamura et al. 2003; Guo and Eiler 2007;Verdier-Paoletti et al. 2017; Vacher et al. 2019). Theseaqueous alteration temperature conditions are similar toestimates for the CI-chondrite parent bodies (~150 °C,Clayton and Mayeda 1999; and <220 °C, Busemann et al.2007). However, a recent study indicates low-temperaturealteration for the CI-chondrite parent body (~210 °C),which is supported by the presence of mercury sulfide, 4C

monoclinic pyrrhotite, and cubanite (Berger et al. 2011).The temperature range determined for sample 03-36-46could suggest that the H-FgMM was altered at relativelyhigh temperatures, similar to the estimated temperaturesfor unequilibrated OCs (<260 °C, Alexander et al. 1989),and compatible with some values obtained CI, and CMchondrites (Busemann et al. 2007; Verdier-Paoletti et al.2017). The D17O of the magnetite in the magnetite-dolomiteassemblage (D17O = +3&) is similar to the values of themagnetite measured in the other H-FgMMs. The oxygenisotopic compositions of the secondary phases in sample03-36-46 have different compositions than the primordialwater of the solar system sampled by the secondary phasesidentified in COSs or potentially by the JFCs (Sakamotoet al. 2007; Seto et al. 2008). Therefore, the relatively hightemperatures estimated for this sample and the lack ofoxygen isotopic compositions with a primordial watersignature could suggest that this particular H-FgMMcomes from a parent body that was large enough to attaintemperatures as high as the OCs, CIs, and CMs, whichimplies an asteroidal origin; however, this origin remainsspeculative.

SUMMARY

We have conducted transmission electron microscope(TEM) observations and SIMS measurements of fivehydrated, fine-grained Antarctic micrometeorites (H-FgMMs). We characterized the petrography and oxygenisotopic of magnetite with different morphologies and onemagnetite-dolomite assemblage in H-FgMMs. This is thefirst oxygen isotopic study of the secondary phasesidentified in hydrated fine-grained AMMs. This studyprovides useful constraints on the conditions of alterationexperienced by these dust particles. Our mineralogicalstudy of H-FgMMs shows the association of magnetite,phyllosilicates, and carbonates and indicates a closeaffinity with the carbonaceous chondrites. Thesesecondary phases demonstrate that the interaction ofaqueous fluids produced by melting of water ice thataccreted together with anhydrous silicate was common onthe micrometeorite parent body(ies). Additionally, wehave shown that magnetite and dolomite formed at atemperature between 160 and 280 °C, which is similar tothe estimated temperature for alteration of unequilibratedOCs. This suggests that the sample 03-36-46 could comefrom a parent body that was large enough to attaintemperatures as high as the OCs, which implies anasteroidal origin for this particular H-FgMM. Our studyshows that magnetite in H-FgMMs formed throughsimilar aqueous alteration processes as those described inprevious primitive solar system materials (chondriticmeteorites, AMMs, IDPs, and possibly in cometarysamples); however, the oxygen isotopic compositions of

Mineralogy and oxygen isotope systematics of micrometeorites 13

magnetite in H-FgMMs indicate that the aqueous fluidfrom which these secondary minerals formed is differentthan the aqueous fluid from which secondary minerals onthe OCs, CCs, or COSs, parent body(ies) formed, whichimplies that the source of H-FgMMs is different from theasteroidal source of CCs and OCs.

Acknowledgments—We thank T. Noguchi, an anonymousreviewer, and the associate editor, M. Zolensky for theirinsightful and constructive comments that helped tosignificantly improve this manuscript. This work wasfunded by NASA grant NNH16ZDA001N to E. Dobrica(PI). Sample preparation (focused ion beam) andtransmission electron microscopy analysis were carried outin the Electron Microbeam Analysis Facility in theDepartment of Earth and Planetary Sciences and Instituteof Meteoritics, University of New Mexico and atWashington University in St. Louis. Secondary ion massspectrometer (SIMS) analyses were carried out in the W.M. Keck Cosmochemistry Laboratory, the University ofHawai’i at M�anoa. The Micrometeorite collection atDome C benefits from the logistical help from IPEV andPNRA. We thank CNRS, CNES, and PNP for fundingmicrometeorite studies.

Editorial Handling—Dr. Michael Zolensky

REFERENCES

Abe K., Sakamoto N., Krot A. N., and Yurimoto H. 2008.Abundances of new-PCP in Acfer 094 and othercarbonaceous chondrites (abstract #1509). 39th Lunar andPlanetary Science Conference. CD-ROM.

Al�eon J., Engrand C., Leshin L. A., and McKeegan K. D.2009. Oxygen isotopic composition of chondriticinterplanetary dust particles: A genetic link betweencarbonaceous chondrites and comets. Geochimica etCosmochimica Acta 73:4558–4575.

Alexander C. M. O’D., Barber D. J., and Hutchison R. 1989.The microstructure of Semarkona and Bishunpur.Geochimica et Cosmochimica Acta 53:3045–3057.

Alexander C. M. O’D., Bowden R., Fogel M. L., and HowardK. T. 2015. Carbonate abundances and isotopiccompositions in chondrites. Meteoritics & PlanetaryScience 50:810–833.

Battandier M., Bonal L., Quirico E., Beck P., Engrand C.,Duprat J., and Dartois E. 2018. Characterization of theorganic matter and hydration state of Antarcticmicrometeorites: A reservoir distinct from carbonaceouschondrites. Icarus 306:74–93.

Berger E. L., Zega T. J., Keller L. P., and Lauretta D. S.2011. Evidence for aqueous activity on comet 81P/Wild 2from sulfide mineral assemblages in Stardust samples andCI chondrites. Geochimica et Cosmochimica Acta 75:3501–3513.

Blinova A. I., Zega T. J., Herd C. D. K., and Stroud R. M.2014. Testing variations within the Tagish Lake meteoriteI: Mineralogy and petrology of pristine samples.Meteoritics & Planetary Science 49:473–502.

Bradley J. P. 2014. Early solar nebula grains—Interplanetary dustparticles. InMeteorites and cosmochemical processes, edited byDavis A. M. Volume 1 of Treatise on Geochemistry, 2nd ed.Amsterdam: Elsevier Ltd. pp. 287–308.

Bradley J. P. and Brownlee D. E. 1991. An interplanetary dustparticle linked directly to type CM meteorites and anasteroidal origin. Science 251:549–552.

Bradley J. P., Humecki H. J., and Germani M. S. 1992.Combined infrared and analytical electron microscopestudies of interplanetary dust particles. The AstronomicalJournal 394:643.

Brearley A. J. 1995. Aqueous alteration and brecciation inBells, an unusual, saponite-bearing, CM chondrite.Geochimica et Cosmochimica Acta 59:2291–2317.

Brearley A. J. 2006. The action of water. In Meteorites and theearly solar system II, edited by Lauretta D. S. andMcSween H. Y. Jr. Tucson, Arizona: University ofArizona Press. pp. 584–624.

Busemann H., Alexander C. M. O’D., and Nittler L. R. 2007.Characterization of insoluble organic matter in primitivemeteorites by microRaman spectroscopy. Meteoritics &Planetary Science 42:1387–1416.

Caplan C. E., Huss G. R., Hammer J. E., Ogliore R. C., andNagashima K. 2015. Crystal orientation effects foroxygen-isotope measurements of magnetite and chromite(abstract #5333). Meteoritics & Planetary Science 78.

Chang Q., Zolenskyl M. E., Martinez J., Tsuchiyama A., andMiyake A. 2016. Magnetite plaquettes are naturallyasymmetric materials in meteorites. American Mineralogist101:2041–2050.

Choi B.-G., McKeegan K. D., Leshin L. A., and Wasson J. T.1997. Origin of magnetite in oxidized CV chondrites: Insitu measurement of oxygen isotope compositions ofAllende magnetite and olivine. Earth and Planetary ScienceLetters 146:337–349.

Choi B.-G., McKeegan K. D., Krot A. N., and Wasson J. T.1998. Extreme oxygen-isotope compositions in magnetitefrom unequilibrated ordinary chondrites. Nature 392:577–579.

Choi B.-G., Krot A. N., and Wasson J. T. 2000. Oxygen isotopesin magnetite and fayalite in CV chondrites Kaba andMokoia.Meteoritics & Planetary Science 1248:1239–1248.

Clayton R. N. and Mayeda T. K. 1999. Oxygen isotope studiesof carbonaceous chondrites. Geochimica et CosmochimicaActa 63:2089–2104.

Davidson J., Krot A. N., Nagashima K., Hellebrand E., andLauretta D. S. 2014. Oxygen isotope and chemicalcompositions of magnetite and olivine in the anomalousCK3 Watson 002 and ungrouped Asuka-881595carbonaceous chondrites: Effects of parent bodymetamorphism. Meteoritics & Planetary Science 49:1456–1474.

Dobric�a E., Engrand C., Leroux H., Duprat J., and GounelleM. 2008. Classic and exotic particles in the 2006CONCORDIA Antarctic micrometeorite collection(abstract #1672). 39th Lunar and Planetary ScienceConference. CD-ROM.

Dobric�a E., Engrand C., Duprat J., Gounelle M., Leroux H.,Quirico E., and Rouzaud J. 2009. Connection betweenmicrometeorites and Wild 2 particles: From Antarcticsnow to cometary ices. Meteoritics & Planetary Science1661:1643–1661.

Dobric�a E., Engrand C., Leroux H., Rouzaud J. N., and DupratJ. 2012. Transmission electron microscopy of CONCORDIA

14 E. Dobric�a et al.

ultracarbonaceous Antarctic micrometeorites (UCAMMs):Mineralogical properties. Geochimica et Cosmochimica Acta76:68–82.

Doyle P. M., Jogo K., Nagashima K., Krot A. N., Wakita S.,Ciesla F. J., and Hutcheon I. D. 2015. Early aqueousactivity on the ordinary and carbonaceous chondrite parentbodies recorded by fayalite. Nature Communications 6:7444.

Duprat J., Engrand C., Maurette M., Gounelle M., Kurat G.,and Leroux H. 2005. Friable micrometeorites from centralAntarctic snow (abstract #1678). 36th Annual Lunar andPlanetary Science Conference. CD-ROM.

Duprat J., Engrand C., Maurette M., Kurat G., Gounelle M.,and Hammer C. 2007. Micrometeorites from centralAntarctic snow: The CONCORDIA collection. Advancesin Space Research 39:605–611.

Duprat J., Dobrica E., Engrand C., Aleon J., Marrocchi Y.,Mostefaoui S., Meibom A., Leroux H., Rouzaud J.-N.,Gounelle M., and Robert F. 2010. Extreme deuteriumexcesses in ultracarbonaceous micrometeorites from centralAntarctic snow. Science 328:742–745.

Eberhardt P., Reber M., Krankowsky D., and Hodges R. R.1995. The D/H and 18O/16O ratios in water from comet P/Halley. Astronomy and Astrophysics 302:301.

Engrand C. and Dobric�a E. 2012. Bulk oxygen isotopiccomposition of Antarctic micrometeorites: Effect ofatmospheric entry (abstract #2636). 43rd Lunar andPlanetary Science Conference. CD-ROM.

Engrand C. and Maurette M. 1998. Carbonaceous micrometeoritesfrom Antarctica.Meteoritics & Planetary Science 33:565–580.

Engrand C., McKeegan K. D., and Leshin L. A. 1999. Oxygenisotopic compositions of individual minerals in Antarcticmicrometeorites: Further links to carbonaceous chondrites.Geochimica et Cosmochimica Acta 63:2623–2636.

Flynn G. J., Leroux H., Tomeoka K., Tomioka N., OhnishiI., Mikouchi T., Wirick S., Keller L. P., Jacobsen C., andSandford S. A. 2008. Carbonate in comets: A comparisonof comets 1P/Halley, 9P/Temple 1, and 81P/Wild 2. 39thLunar and Planetary Science Conference. CD-ROM.

Fujiya W., Sugiura N., Sano Y., and Hiyagon H. 2013. Mn–Cr ages of dolomites in CI chondrites and the Tagish Lakeungrouped carbonaceous chondrite. Earth and PlanetaryScience Letters 362:130–142.

Genge M. J., Grady M. M., and Hutchison R. 1997. Thetextures and compositions of fine-grained Antarctic microm-eteorites: Implications for comparisons with meteorites.Geochimica et Cosmochimica Acta 61:5149–5162.

Genge M. J., Engrand C., Gounelle M., and Taylor S. 2008.The classification of micrometeorites. Meteoritics &Planetary Science 43:497–515.

Germani M. S., Bradley J. P., and Brownlee D. E. 1990.Automated thin film analyses of hydrated interplanetarydust particles in the analytical microscope. Earth andPlanetary Science Letters 101:162–179.

Greshake A., Krot A. N., Flynn G. J., and Keil K. 2005.Fine-grained dust rims in the Tagish Lake carbonaceouschondrite: Evidence for parent body alteration. Meteoritics& Planetary Science 40:1413–1431.

Guo W. and Eiler J. M. 2007. Temperatures of aqueousalteration and evidence for methane generation on theparent bodies of the CM chondrites. Geochimica etCosmochimica Acta 71:5565–5575.

Huberty J. M., Kita N. T., Kozdon R., Heck P. R., FournelleJ. H., Spicuzza M. J., Xu H., and Valley J. W. 2010.

Crystal orientation effects in d18O for magnetite andhematite by SIMS. Chemical Geology 276:269–283.

Jilly-Rehak C. E., Huss G. R., Nagashima K., and SchraderD. L. 2018. Low-temperature aqueous alteration on theCR chondrite parent body: Implications from in situoxygen-isotope analyses. Geochimica and CosmochimicaActa 222:230–252.

Keller L. P., Thomas K. L., and Mckay D. S. 1992. Aninterplanetary dust particle with links to CI chondrites.Geochemica et Cosmochemica Acta 56:1409–1412.

Kerridge J. F., Mackay A. L., and Boynton W. V. 1979.Magnetite in CI carbonaceous meteorites: Origin by aqueousactivity on a planetesimal surface. Science 205:395–397.

Kimura Y., Sato T., Nakamura N., Nozawa J., Nakamura T.,Tsukamoto K., and Yamamoto K. 2013. Vortex magneticstructure in framboidal magnetite reveals existence of waterdroplets in an ancient asteroid. Nature Communications 4:1–8.

Krot A. N., Zolensky M. E., Wasson J. T., Scott E. R. D.,Keil K., and Ohsumi K. 1997. Carbide-magnetiteassemblages in type-3 ordinary chondrites. Geochimica etCosmochimica Acta 61:219–237.

Kurat G., Brandstaetter F., Maurette M., and Coeberl C. 1992. CI-Like micrometeorites from Cap Prudhomme, Antarctica(abstract). 23rd Lunar and Planetary Science Conference. p. 747.

Kurat G., Koeberl C., Presper T., Brandst€atter F., andMaurette M. 1994. Petrology and geochemistry of Antarcticmicrometeorites. Geochimica et Cosmochimica Acta58:3879–3904.

Kusakabe M., Maruyama S., Nakamura T., and Yada T.2004. CO2 laser-Br F5 fluorination technique for analysisof oxygen three isotopes of rocks and minerals. Journal ofthe Mass Spectrometry Society of Japan 52:205–212.

Lee M. R., Lindgren P., and Sofe M. R. 2014. Aragonite,breunnerite, calcite and dolomite in the CM carbonaceouschondrites: High fidelity recorders of progressive parentbody aqueous alteration. Geochimica et Cosmochimica Acta144:126–156.

Lisse C. M., Vancleve J., Adams A. C., A’hearn M. F.,Fern�andez Y. R., Farnham T. L., Armus L., Grillmair C.J., Ingalls J., Belton M. J., Groussin O., McFadden L. A.,Meech K. J., Schultz P. H., Clark B. C., Feaga L. M., andSunshine J. M. 2006. Spitzer spectral observations of thedeep impact ejecta. Science 313:635–640.

Lodders K. 2003. Solar system abundances and condensationtemperatures of the elements. The Astrophysical Journal591:1220–1247.

Love S. G. and Brownlee D. E. 1993. A direct measurementof the terrestrial mass accretion rate of cosmic dust.Science 262:550.

Maurette M., Beccard B., Bonny P., Brack A., Christophe M.,and Veyssi�ere P. 1990. C-rich micrometeorites on the earlyearth and icy planetary bodies. ESA Special Publication167–172.

Maurette M., Olinger C., Michel-Levy M. C., Kurat G.,Pourchet M., Brandst€atter F., and Bourot-Denise M.1991. A collection of diverse micrometeorites recoveredfrom 100 tonnes of Antarctic blue ice. Nature 351:44–47.

Mikouchi T., Tachikawa O., Hagiya K., Ohsumi K., SuzukiY., Uesugi K., Takeuchi A., and Zolensky M. E. 2007.Mineralogy and crystallography of comet 81P/Wild 2particles (abstract #1946). 38th Lunar and PlanetaryScience Conference. CD-ROM.

Nakamura T., Noguchi T., and Yoneda S. 2003. Theconditions of aqueous alteration reactions recorded in the

Mineralogy and oxygen isotope systematics of micrometeorites 15

Sayama CM2 chondrite (abstract #5080). Meteoritics &Planetary Science 66.

Nesvorn�y D., Jenniskens P., Levison H. F., Bottke W. F.,Vokrouhlick�y D., and Gounelle M. 2010. Cometary originof the zodiacal cloud and carbonaceous micrometeorites.Implications for hot debris disks. The AstrophysicalJournal 713:816–836.

Noguchi T., Nakamura T., and Nozaki W. 2002. Mineralogyof phyllosilicate-rich micrometeorites and comparison withTagish Lake and Sayama meteorites. Earth and PlanetaryScience Letters 202:229–246.

Noguchi T., Nakamura T., Okudaira K., Yano H., Sugita S.,and Burchell M. J. 2007. Thermal alteration of hydratedminerals during hypervelocity capture to silica aerogel atthe flyby speed of Stardust. Meteoritics & PlanetaryScience 42:357–372.

Noguchi T., Yabuta H., Itoh S., Sakamoto N., Mitsunari T.,Okubo A., Okazaki R., Nakamura T., Tachibana S.,Terada K., Ebihara M., Imae N., Kimura M., andNagahara H. 2017. Variation of mineralogy and organicmaterial during the early stages of aqueous activityrecorded in Antarctic micrometeorites. Geochimica etCosmochimica Acta 208:119–144.

Nozaki W., Nakamura T., and Noguchi T. 2006. Bulkmineralogical changes of hydrous micrometeorites duringheating in the upper atmosphere at temperatures below1000°C. Meteoritics & Planetary Science 41:1095–1114.

Nozawa J., Tsukamoto K., van Enckevort W., Nakamura T.,Kimura Y., Miura H., Satoh H., Nagashima K., andKonoto M. 2011. Magnetite 3D colloidal crystals formedin the early solar system 4.6 billion years ago. Journal ofthe American Chemical Society 133:8782–8785.

Rietmeijer F. J. M. 1990. Salts in two chondritic porousinterplanetary dust particles. Meteoritics 25:209.

Rietmeijer F. J. M. 1996. CM-like interplanetary dust particlesin lower stratosphere during 1989 October and 1991 June/July. Meteoritics & Planetary Science 31:278–288.

Rowe M. W., Clayton R. N., and Mayeda T. K. 1994.Oxygen isotopes in separated components of CI and CMmeteorites. Geochimica et Cosmochimica Acta 58:5341–5347.

Sakamoto N., Seto Y., Itoh S., Kuramoto K., Fujino K.,Nagashima K., Krot A. N., and Yurimoto H. 2007.Remnants of the early solar system water enriched inheavy oxygen isotopes. Science 317:231.

Sakamoto K., Nakamura T., Noguchi T., and Tsuchiyama A.2010. A new variant of saponite-rich micrometeoritesrecovered from recent Antarctic snowfall. Meteoritics &Planetary Science 45:220–237.

Sandford S. A. and Walker R. M. 1985. Laboratory infraredtransmission spectra of individual interplanetary dustparticles from 2.5 to 25 microns. The Astrophysical Journal291:838–851.

Schramm L. S., Brownlee D. E., and Wheelock M. M. 1989.Major element composition of stratospheric micromete-orites. Meteoritics 24:99.

Scott E. R. D., Krot A. N., and Sanders I. S. 2018. Isotopicdichotomy among meteorites and its bearing on theprotoplanetary disk. The Astrophysical Journal 854:164.

Seto Y., Sakamoto N., Fujino K., Kaito T., Oikawa T., andYurimoto H. 2008. Mineralogical characterization of aunique material having heavy oxygen isotope anomaly inmatrix of the primitive carbonaceous chondrite Acfer 094.Geochimica et Cosmochimica Acta 72:2723–2734.

�Sliwi�nski M. G., Kitajima K., Kozdon R., Spicuzza M. J.,Fournelle J. H., Denny A., and Valley J. W. 2015.Secondary ion mass spectrometry bias on isotope ratios indolomite–ankerite, part I: d18O matrix effects.Geostandards and Geoanalytical Research 40:157–172.

Starkey N. A. and Franchi I. A. 2013. Insight into the silicateand organic reservoirs of the comet forming region.Geochimica et Cosmochimica Acta 105:73–91.

Starkey N. A., Franchi I. A., and Lee M. R. 2014a. Heavyoxygen isotope ratios in an interplanetary dust particle(abstract #1870). 45th Lunar and Planetary ScienceConference. CD-ROM.

Starkey N. A., Franchi I. A., and Lee M. R. 2014b. Isotopicdiversity in interplanetary dust particles and preservationof extreme 16O-depletion. Geochimica et CosmochimicaActa 142:115–131.

Stodolna J., Jacob D., and Leroux H. 2012. Mineralogy andpetrology of Stardust particles encased in the bulb of track80: TEM investigation of the Wild 2 fine-grained material.Geochimica et Cosmochimica Acta 87:35–50.

Thomas K. L., Zolensky M. E., Kloeck W., and McKay D. S.1990. Mineralogical descriptions of eight hydratedinterplanetary dust particles and their relationship tochondrite matrix. 21st Lunar and Planetary ScienceConference. p. 1250.

Tomeoka K. and Buseck P. R. 1984. Transmission electronmicroscopy of the ‘‘LOW-CA’’ hydrated interplanetarydust particle. Earth and Planetary Science Letters 69:243–254.

Tomeoka K. and Buseck P. R. 1986. A carbonate-rich,hydrated, interplanetary dust particle: Possible residuefrom protostellar clouds. Science 231:1544–1546.

Tomeoka K., Tomioka N., and Ohnishi I. 2008. Silicateminerals and Si-O glass in comet Wild 2 samples:Transmission electron microscopy. Meteoritics & PlanetaryScience 43:273–284.

Tyra M. A., Farquhar J., Guan Y., and Leshin L. A. 2012. Anoxygen isotope dichotomy in CM2 chondritic carbonates: ASIMS approach. Geochimica et Cosmochimica Acta77:383–395.

Vacher L. G., Piralla M., Piani L., and Marrocchi Y. 2019.Monitoring the thermal evolution of the CM parent body(-ies) with in situ oxygen isotope analyses (abstract#1862). 50th Lunar and Planetary Science Conference.CD-ROM.

Verdier-Paoletti M. J., Marrocchi Y., Avice G., Roskosz M.,Gurenko A., and Gounelle M. 2017. Oxygen isotopeconstraints on the alteration temperatures of CMchondrites. Earth and Planetary Science Letters 458:273–281.

Warren P. H. 2011a. Stable-isotopic anomalies and theaccretionary assemblage of the Earth and Mars: Asubordinate role for carbonaceous chondrites. Earth andPlanetary Science Letters 311:93–100.

Warren P. H. 2011b. Stable isotopes and the noncarbonaceousderivation of ureilites, in common with nearly all differ-entiated planetary materials. Geochimica et CosmochimicaActa 75:6912–6926.

Wirick S., Leroux H., Tomeoka K., Zolensky M., Flynn G. J.,Tyliszczak T., Butterworth A., Tomioka N., Ohnishi I.,Nakamura-Messenger K., Sandford S., Keller L., andJacobsen C. 2007. Carbonates found in Stardust aerogeltracks (abstract #1534). 38th Lunar and Planetary ScienceConference. CD-ROM.

16 E. Dobric�a et al.

Yurimoto H. and Kuramoto K. 2004. Molecular cloud originfor the oxygen isotope heterogeneity in the solar system.Science 305:1763–1766.

Zheng Y.-F. 2011. On the theoretical calculations of oxygenisotope fractionation factors for carbonate-water systems.Geochemical Journal 45:341–354.

Zolensky M. E. and Lindstrom D. J. 1992. Mineralogy of 12large “‘chondritic’” interplanetary dust particles. 22ndLunar and Planetary Science Conference Proceedings. pp.161–169.

Zolensky M. E., Bourcier W. L., and Gooding J. L. 1989.Aqueous alteration on the hydrous asteroids: Results ofEQ3/6 computer simulations. Icarus 78:411–425.

Zolensky M. E., Nakamura K., Gounelle M., Mikouchi T.,Kasama T., Tachikawa O., and Tonui E. 2002.

Mineralogy of Tagish Lake: An ungrouped type 2carbonaceous chondrite. Meteoritics & Planetary Science37:737–761.

Zolensky M. E., Krot A. N., and Benedix G. 2008a. Recordof low-temperature alteration in asteroids. Reviews inMineralogy and Geochemistry 68:429–462.

Zolensky M., Nakamura-messenger K., Rietmeijer F., LerouxH., Mikouchi T., Ohsumi K., Simon S., Grossman L.,Stephan T., Weisberg M., Velbel M., Zega T., Stroud R.,Tomeoka K., Ohnishi I., Tomioka N., Nakamura T.,Matrajt G., Joswiak D., Brownlee D., Langenhorst F.,Krot A., Kearsley A., Ishii H., Graham G., Dai Z. R., ChiM., Bradley J., Hagiya K., Gounelle M., and Bridges J.2008b. Comparing Wild 2 particles to chondrites andIDPs. Meteoritics & Planetary Science 43:261–272.

Mineralogy and oxygen isotope systematics of micrometeorites 17