spectral reflectance properties of carbonaceous chondrites ...psf.uwinnipeg.ca/publications/icarus...

Icarus 221 (2012) 328–358

Contents lists available at SciVerse ScienceDirect

Icarus

journal homepage: www.elsevier .com/ locate/ icarus

Spectral reflectance properties of carbonaceous chondrites: 6. CV chondrites

E.A. Cloutis a,⇑, P. Hudon b,1, T. Hiroi c, M.J. Gaffey d, P. Mann a, J.F. Bell III e

a Department of Geography, University of Winnipeg, 515 Portage Avenue, Winnipeg, Manitoba, Canada R3B 2E9b Astromaterials Research and Exploration Science Office, NASA Johnson Space Center, Mail Code KR, 2101 NASA Road 1, Houston, TX 77058-3696, USAc Department of Geological Sciences, Brown University, Box 1846, Providence, RI 02912-1846, USAd Department of Space Studies, University of North Dakota, PO Box 9008, Grand Forks, ND 58202-9008, USAe School of Earth and Space Exploration, Box 871404, Arizona State University, Tempe, AZ 85287-1404, USA

a r t i c l e i n f o

Article history:Received 15 May 2012Revised 11 July 2012Accepted 12 July 2012Available online 27 July 2012

Keywords:MeteoritesAsteroidsSpectroscopy

0019-1035/$ - see front matter � 2012 Elsevier Inc. Ahttp://dx.doi.org/10.1016/j.icarus.2012.07.007

⇑ Corresponding author. Fax: +1 204 774 4134.E-mail addresses: [email protected] (E.A. Clo

(P. Hudon), [email protected] (T. Hiroi), [email protected] (J.F. Bell III).

1 Present address: Department of Mining and MUniversity, 3610 rue Université, Montreal, QC, Canada

a b s t r a c t

Multiple reflectance spectra of 11 CV chondrites have been measured to determine spectral–composi-tional relationships for this meteorite class and to aid the search for CV parent bodies. The reflectanceof CV chondrite spectra is variable, ranging from �5% to 13% at 0.56 lm, and �5% to 15% at the0.7 lm region local reflectance maximum. Overall slopes range from slightly blue to red for powders,while slab spectra are strongly blue-sloped. With increasing average grain size and/or removal of the fin-est fraction, CV spectra generally become more blue-sloped. CV spectra are characterized by ubiquitousabsorption features in the 1 and 2 lm regions. The 1 lm region is usually characterized by a band cen-tered near 1.05–1.08 lm and a band or shoulder near 1.3 lm that are characteristic of Fe-rich olivine.Band depths in the 1 lm region for powdered CVs and slabs range from �1% to 10%. The 2 lm regionis characterized by a region of broad absorption that extends beyond 2 lm and usually includes bandminima near 1.95 and 2.1 lm; these features are characteristic of Fe2+-bearing spinel. The sample suiteis not comprehensive enough to firmly establish whether spectral differences exist between CVR, CVOxA,and CVOxB subclasses, or as a function of metamorphic grade. However, we believe that the mineralogicand petrologic differences that exist between these classes, and with varying petrologic subtype (CV3.0–>3.7), may not be significant enough to result in measurable spectral differences that exceed spectralvariations within a subgroup, within an individual meteorite, or as a function of grain size. Terrestrialweathering seems to affect CV spectra most noticeably in the visible region, resulting in more red-slopedspectra for finds as compared to falls. The search for CV parent bodies should focus on the detection ofolivine and spinel absorption bands, specifically absorption features near 1.05, 1.3, 1.95, and 2.1 lm, asthese are the most commonly seen spectral features of CV chondrites.

� 2012 Elsevier Inc. All rights reserved.

1. Introduction

As part of an ongoing study of the spectral reflectance proper-ties of carbonaceous chondrites (CCs) (Cloutis et al., 2011a,b, Clou-tis et al., 2012a–c), in this paper we examine and discuss thespectral properties of CV carbonaceous chondrites. The goals of thislarger study include determining relationships between CC reflec-tance spectra and composition/petrography/mineralogy. We alsoaim to determine whether variations within a group are expressedspectrally, and whether different CC groups exhibit unique spectralproperties.

ll rights reserved.

utis), [email protected]@space.edu (M.J. Gaffey),

aterials Engineering, McGillH3A 2B2.

CV chondrites are of importance and interest to planetary scien-tists for a number of reasons. For example, the fall of the AllendeCV chondrite in 1969 provided many hundreds of kilograms of asingle body for detailed analysis. This meteorite has provided thebest estimates for the age of the Solar System and timing of forma-tion, from dating of calcium-aluminum inclusions (e.g., Chen andWasserburg, 1981). CVs also provide insights into possible earlySolar System heating events of planetesimals (e.g., Weiss et al.,2010; Elkins-Tanton et al., 2011; Humayun and Weiss, 2011).CVs have also been tentatively linked to a number of possible par-ent bodies using spectroscopic criteria (e.g., Burbine et al., 2001).More recent studies suggest that other asteroids have similaritiesto CVs, but with enhancements of some CV components (e.g.,Sunshine et al., 2008a,b).

The known CV chondrites are approximately of petrologic grade3 (McSween, 1977, 1979). CVs are dominated by olivine(>75 vol.%), with lesser amounts of enstatite, plagioclase feldspar,magnetite, sulfides, and metal (Howard et al., 2009). Olivine

http://dx.doi.org/10.1016/j.icarus.2012.07.007

mailto:[email protected]






http://www.sciencedirect.com/science/journal/00191035

http://www.elsevier.com/locate/icarus

Table 2Modal mineralogy of Allende.

Phase Wt.% Vol.%

Olivine (Fo100) 20.5 23.4Olivine (Fo80) 14.6 15.4Olivine (Fo60) 20.8 20.6Olivine (Fo50) 21.4 20.6Olivine (Fo25) 4.3 3.9Pentlandite 11.1 8.0Clinoenstatite (En90) 5.9 6.6Plagioclase (An100) 0.9 1.2

E.A. Cloutis et al. / Icarus 221 (2012) 328–358 329

composition can vary widely, but is fairly fayalitic (>Fa35)(Weisberg and Prinz, 1997). Phyllosilicates are normally absentor rare, but are abundant in at least one case (Keller et al., 1994).

Reflectance spectra of a suite of 11 CV chondrites have beenexamined in order to determine the range of spectral propertiesof this meteorite group, whether they exhibit differences with pet-rologic subgrade, subtype (CVR, CVOx, CVOxA, CVOxB), viewing geom-etry, and grain size, whether they are spectrally distinct from otherCCs, and to provide spectral–compositional guidelines that may beuseful in the search for CV parent bodies.

Magnetite 0.3 0.2Fe-metal 0.2 0.1

Source of data: Bland et al. (2004).

Table 3Modal mineralogy and selected phase abundance of CV chondrites.

Vol.%a Allende Grosnaja Leoville Mokoia VigaranoCV3OxA CV3OxB CV3R CV3OxA CV3R

Forsterite 22.4 11.0 19.3 24.6Fo90 7.6 10.8 13.3 18.2Fo80 9.4 16.9 21.6 11.6Fo60 20.6 10.8 6.1 12.4Fo50 19.7 15.0 7.3 11.9Fo 1.9 5.8 7.1 3.7

2. Compositional properties of CV chondrites

In general, CV chondrites contain �32–67 vol.% chondrules, 6–12 vol.% inclusions, 0–4 vol.% lithic/mineral fragments, 3–8 vol.%opaque minerals, and 17–51 vol.% matrix (McSween, 1977, 1979;Buseck and Hua, 1993). Phyllosilicates, when present, generally oc-cur in small amounts (Buseck and Hua, 1993). Fayalitic olivine andmagnetite abundances are positively correlated; magnetite + faya-litic olivine versus metal are inversely correlated (Howard et al.,2009). Their overall composition in comparison to other CC groupsis provided in Table 1. A detailed analysis of Allende and a numberof other CVs are provided in Tables 2 and 3 a compilation of CAIabundances in CVs is provided in Table 4.

40

Fo20 1.1 6.3 1.7 2.6Fayalite 0.0 0.0 0.0 0.0Enstatite 6.3 4.8 6.3 8.1Anorthite 1.1 1.3 1.7 1.1Magnetite 0.3 1.6 2.6 1.4Sulfide 6.6 7.9 8.1 2.4Fe–Ni metal 1.0 0.8 1.1 2.0Fe-oxide 0.0 2.7 0.0 0.0Phyllosilicate 1.9 4.2 3.7 0.0Total 100 100 100 100

Magnetiteb <0.5 wt.% 8.0 wt.% 3.8 wt.% 3.2 wt.% 4.1 wt.%Troilitec 11.4 16.9 10.8 13 23.6Silicatesc 87.7 62.4 56 67.5 51.2Fe–Nic n.d. n.d. n.d n.d. 4.9Magnetitec n.d. 11.8 16.4 15.6 14.7Paramagc 0.9 5.9 16.7 3.8 5.5

Note: only includes phases present at >1 wt.%. Analysis performed on samples freeof large inclusions, such as CAIs and dark inclusions. A number of CVs also containglassy mesostasis within chondrules, which was not included in the totals.

a Source of data: Howard et al. (2010); abundances determined by PSD-XRD.b Bland et al. (2000).c Bland et al. (2008); numbers indicate proportions of Fe in different phases as a

fraction of total Fe; paramag = paramagnetic component.

2.1. Subgroups

The CV3 chondrites have been subdivided into oxidized (CV3Ox)and reduced (CV3R) groups based on their modal metal:magnetiteratios, Ni content of the metal and sulfides (McSween, 1977) andother petrographic and mineralogic criteria (Krot et al., 1995).The oxidized group is further subdivided into two subgroups (Al-lende-like (CV3OxA) and Bali-like (CV3OxB)) on the basis of alter-ation features (Weisberg et al., 1997) which are thought to relateto the degree and temperature of aqueous alteration that theyexperienced (Krot et al., 1998) and other characteristics, summa-rized in Table 5.

The oxidized and reduced subgroups have different metal ver-sus magnetite abundances. Members of the CV3OxA group havelow abundances of Fe–Ni-metal, sulfides and magnetite (Blandet al., 2000). The CV3OxB oxidized subgroup meteorites have vari-able to high abundances of magnetite but low or absent metaland low sulfides (Bland et al., 2000). The CVR meteorites haveintermediate magnetite abundances and high contents of metaland sulfides (Bland et al., 2000). It is suggested that the reducedsubgroup may be the precursor to the oxidized subgroup (Howardet al., 2009). Bland et al. (2002) suggested that the reduced and oxi-dized subgroups may be reflecting some other mineralogic/petro-logic effect, as magnetite abundance is higher in Vigarano (areduced CV) than Allende (an oxidized CV). Magnetite abundancesseems to increase from CVOxA (<0.5 wt.% for Allende), to CVOxB

Table 1Petrographic characteristics of C-chondrite groups. Source: Brearley and Jones (1998).

Group Chondrule Matrixabundance(vol.%)

Refractoryinclusionabundance(vol.%)

Metalabundance(vol.%)

Chondrulemeandiameter(mm)

CI �1 >99 �1 0 –CM 20 70 5 0.1 0.3CR 50–60 30–50 0.5 5–8 0.7CO 48 34 13 1–5 0.15CV 45 40 10 0–5 1.0CK 15 75 4 <0.01 0.7CH �70 5 0.1 20 0.02

(2.6–3.2 wt.% for Grosnaja and Mokoia), to CVR (3.8–4.1 wt.% forLeoville and Vigarano) (Bland et al., 2000; see also Tables 3 and5). A comparison of the phase abundances for CV3R chondrites(specifically Vigarano and Efremovka) and CV3Ox chondrites (spe-cifically Allende, Mokoia, Grosnaja, Kaba) is provided in Table 5.

The CV3OxB subgroup is characterized by the presence of sec-ondary low-Ca phyllosilicates (saponite and Na-phlogopite), mag-netite, Ni-rich sulfides, fayalite (Fa>90), and Ca–Fe-rich pyroxenes(Fs10–50 Wo45–50) (Krot et al., 1998). Phyllosilicates replace primaryCa-rich minerals, and magnetite is replaced to varying degrees byfayalite and Ca–Fe-rich pyroxenes (Krot et al., 1998). In CV3OxA

chondrites, secondary fayalite and phyllosilicates are virtually ab-sent in chondrules and CAIs. These observations suggest thatCV3OxA lithologies experienced higher alteration temperaturesthan CV3OxB lithologies (Krot et al., 1998). It has been suggestedthat the sparse phyllosilicates in the oxidized subgroup, repre-sented by Mokoia, were formed from Fe-rich olivine by removaland oxidation of Fe2+ to Fe3+ planar precipitates (Tomeoka andBuseck, 1990). Subsequent hydration formed Fe oxides and

Table 4CAI abundances in CV chondrites.

Meteorite CAI abundance Source

Allende �2.2 vol.% Kornacki and Wood (1984)2.52 area% Kornacki and Wood (1984)9.4 vol.% McSween (1977)7.1 vol.% Simon and Haggerty (1979)4.47–5.94 area% Hezel et al. (2008)

Grosnaja 2.8 vol.% McSween (1977)Leoville 6.6 vol.% McSween (1977)

1.89 area% May et al. (1999)Mokoia 3.5 vol.% McSween (1977)

0.65 area% May et al. (1999)Vigarano 5.3 vol.% McSween (1977)

1.70 area% May et al. (1999)

Notes: Abundance in Allende by Kornacki and Wood (1984) determined by opticalinspection of a thin section. Abundances determined by McSween (1977) and Mayet al. (1999) using point counting of thin sections. Abundances determined by Hezelet al. (2008) from X-ray mapping of thin sections.

Table 5Characteristics of CV3 subgroups.

Property CV3R CVOx CV3OxA CV3OxB

Matrix/chondrule ratioa 0.5–0.6 0.6–0.7 0.7–1.2Metal/magnetite ratioa 2–46 0.2 Trace metalFayalitic olivine Fa contenta 32–60 32–60 10–90Hydrous phyllosilicatesa Rare Rare More commonVol.% olivineb 83–85 76.3–83.9Vol.% enstatiteb 6.5–8.1 4.8–7.8Vol.% anorthiteb 1.1–1.2 1.1–1.7Vol.% magnetiteb 1.4–1.8 0.3–6.1Vol.% sulfideb 2.4–5.1 2.9–8.1Vol.% Fe–Ni metalb 2.0–2.2 0.2–1.1Vol.% Fe-oxideb �0 0–2.7Vol.% phyllosilicatesb �0 1.9–4.2

a Weisberg et al. (1997); analysis based on: CV3R: Leoville, Vigarano; CV3OxA:Allende, ALH 84028; CV3OxB: Grosnaja, Mokoia; CV3Ox(unassigned).

b Howard et al. (2010); analysis based on: CV3R: Efremovka, Vigarano; CV3Ox:Allende, Grosnaja, Kaba, Mokoia.

Table 6Petrographic subtypes of CV chondrites included in this study.

Meteorite Subtype Petrographicsubgrade rangea

Averagepetrographicsubtypea

Petrographicsubtypeb

ALHA 81003 CV3 3.0–3.5 3.0ALH 84028 CV3OxA 3.2 3.2ALH 85006 CV3Ox

Allende CV3OxA 3.1–3.6 3.2 >3.6Grosnaja CV3OxB 3.0–3.3 3.3 �3.6Leoville CVR 3.0–3.6 3.0 3.1–3.4Mokoia CV3OxA 3.0–3.3 3.2 �3.6NWA 3118 CV3QUE 93744 CV3Vigarano CV3R 3.0–3.3 3.3 3.1–3.4Y-86751 CV3OxA–OxB

a Guimon et al. (1995).b Bonal et al. (2006).

330 E.A. Cloutis et al. / Icarus 221 (2012) 328–358

hydroxides, while the Fe-poor olivine was altered to saponite(Tomeoka and Buseck, 1990).

In general, CVR chondrites show the least evidence of eitheraqueous alteration or metasomatism; the CVOxA chondrites pri-marily show evidence of metasomatism; the CVOxB chondrites ap-pear to have been most affected by aqueous alteration (Keller et al.,1994; Krot et al., 1995; Brearley, 2006; Huss et al., 2006). Howardet al. (2010) found that mineralogic parameters, such as the abun-dances of fayalite, ferrous olivine, and magnetite overlap betweenthe CVR and CVOx groups, and show no relationship to petrographictype. These mineralogic relationships also show no correlation topetrographic type, suggesting that low temperature anhydrousmetamorphism affected all CVs and was responsible for the forma-tion of these phases; in addition, Allende is distinct from other CVs(Howard et al., 2010). The major mineralogy differences betweenthe CVR and CVOx groups seem to be the relative abundance of me-tal and absence of phyllosilicates in reduced CVs (Howard et al.,2010).

A metamorphic series for CV3 chondrites has been suggestedbased on various criteria, although the degree of metamorphismis less than is seen for CO chondrites (Guimon et al., 1995), andmultiple processes seem to have affected many CVs (Huss et al.,2006). Guimon et al. (1995) proposed a sequence based largelyon induced thermoluminescence, which was compared to variouspetrologic criteria. Increasing grade seems to correlate withincreasing olivine Fa content (from Fa<8 to Fa>14.5), decreasing oliv-ine Fa heterogeneity, decreasing C content (from >1.05 to

<0.15 wt.%), and decreasing H2O content (from >3.5 to <0.5 wt.%)(Guimon et al., 1995). The petrographic subtypes of the CVs in-cluded in this study and characterized by Guimon et al. (1995)are provided in Table 6. It should be noted that for many CV chon-drites, different criteria lead to different petrologic subtypes. Forexample, infrared spectroscopic categories of insoluble organicmatter in CVs do not correlate with degree of metamorphismdetermined by other means (Kebukawa et al., 2011).

Bonal et al. (2006) used Raman spectroscopic properties of or-ganic matter in the matrix, as well as petrographic information,to derive petrologic subgrades for a variety of CVs. They deter-mined that the induced thermoluminescence approach used byGuimon et al. (1995) underestimated petrologic grades for a num-ber of CVs common to both studies. The derived petrologic sub-grades from Bonal et al. (2006) are provided in Table 6. Frieset al. (2011) determined, from Raman spectroscopy of organicmaterial in a variety of CCs, that Vigarano is more metamorphosedthan Kainsaz (a CO3.2 chondrite).

2.2. Compositional properties of CV phases

2.2.1. MatrixMatrix comprises between �17 and 51 vol.% of CVs, and is com-

posed of olivine (the dominant phase), minor pyroxene (Fs10–50

Wo45–50), kamacite, taenite, magnetite (minor), pyrrhotite, pent-landite, Ca-phosphates (minor), nepheline (minor), sodalite (min-or), micas (minor, only in chondrules), saponite (minor),amorphous material, and carbonates (in chondrules only) (Buseckand Hua, 1993; Zolensky et al., 1993). Fayalitic olivine (Fa>35) com-prises up to >80 vol.% of CV3 matrix (Kojima and Tomeoka, 1996;Weisberg and Prinz, 1997; Krot et al., 1998). While most matrixolivine is Fa>35, its composition can range from Fa10 to Fa70 (Buseckand Hua, 1993; Zolensky et al., 1993). CV matrix appears to consistof at least two end member types, at least texturally (Hurt et al.,2012).

2.2.2. ChondrulesChondrules, which occupy �32–65 vol.% of CVs are generally of

two main types: I and II. Type I chondrules are olivine-rich andFa<10, but can range up to Fa45 (McSween, 1977; Brearley andJones, 1998). Type I chondrules occur as two main subtypes: opa-que mineral-poor and opaque mineral-rich, and occur in approxi-mately equal abundances: the opaques are commonly metal,magnetite and/or sulfide minerals (McSween, 1977). Type IIchondrules are rare, and contain more Fe-rich olivine, as well asfine-grained chromite (Johnson and Prinz, 1991).


2.2.3. Mafic silicatesFayalitic olivine (average Fa�32) is a major component of the

matrices and dark inclusions of CV3 chondrites (Weisberg andPrinz, 1998). It commonly occurs as rims, veins and halos in andaround chondrule silicates in CV3OxA chondrites and to a much les-ser extent in CV3R and CV3OxB chondrites (Weisberg and Prinz,1998). Fayalitic olivine makes up >80% of the matrix of Allende(Weisberg and Prinz, 1998). The range of Fa content varies, likelyas a result of equilibration (Brearley and Jones, 1998). XRD datasuggest that olivines span the entire Fe-Mg range, and all CVs con-tain a component (4–13%) of fine-grained olivine that is Fa>60

(Howard et al., 2009).

2.2.4. CAIsThe average areal abundance of CAIs in CV chondrite varies

among different studies (Table 4), ranging from 0.65 area% to9.4 vol.%. McSween (1977) determined the modal distribution oftwo types of inclusions: amoeboid olivine inclusions (AOIs) andCa, Al-rich. AOI abundances range from 1.2–8.6 vol.%; �2 vol.% inAllende, although higher abundances are seen in other CVs (Gross-man and Steele, 1976), while CAI abundances range from 0.3 to9.4 vol.%. AOIs consist predominantly of olivine (Fa1–Fa36), as wellas pyroxene, nepheline, and sodalite, with rare anorthite, spinel(2.1–13.4% FeO), and perovskite (Grossman and Steele, 1976).Alteration of CAIs in CVOxA chondrites includes entry of Fe into spi-nels (Krot et al., 1995).

Two main types of CAIs occur in CVs: type A and B. Type A inclu-sions include compact and fluffy varieties (MacPherson and Gross-man, 1984). Compact type A inclusions are composed of 80–85%melilite (<1% FeO), 15–20% spinel (0–4.5% FeO, mostly 0–1% FeO),1–2% perovskite, and rare plagioclase, hibonite, wollastonite, andgrossularite (Grossman, 1975). Clinopyroxene that may be presentis restricted to thin rims around inclusions or cavities in their inte-riors. Fluffy type A inclusions are composed of Al-rich melilite(<0.5% FeO), spinel (commonly V-rich; 0–20% FeO, mostly <2%FeO), perovskite, and hibonite (MacPherson and Grossman, 1984).

Type B inclusions, that occur exclusively in CV chondrites, arecomposed of abundant fassaite (30–60 vol.%; <1–8% FeO, <1–18%TiO2), anorthite (5–25 vol.%; An98–100), spinel (15–30 vol.%; <1–20% FeO, mostly <1% FeO), and melilite (5–20 vol.%; <1% FeO)(Grossman, 1975, 1980; Cloutis and Gaffey, 1993 and referencestherein).

CAIs are commonly rimmed by other materials. These includeWark-Lovering rims that consist of multiple layers of varying pro-portions of spinel, perovskite, hibonite, olivine, anorthite, pyrox-ene, and andradite (Wark and Lovering, 1977). Rim compositionsdiffer from the adjacent matrix and CAI cores, and also differ be-tween Type A and B CAIs, and between fluffy and compact CAIs(Wark and Lovering, 1977).

Additional rimming materials consist of dark, fine-grainedmaterials that mantle peripheries of clasts, inclusions, and chond-rules (King and King, 1981; MacPherson et al., 1985). These man-tles are multilayered and differ slightly from the matrix in termsof grain size, mineralogy, and chemistry (MacPherson et al.,1985). These ‘‘accretionary rims’’ consist largely of olivine, withlesser amounts of Fe-bearing pyroxenes, andradite, nepheline,and iron sulfides (MacPherson et al., 1985) and may be gradationalwith the surrounding matrix (King and King, 1981).

2.2.5. Dark inclusionsDark inclusions, at the few vol.% level, are present in CV chon-

drites and differ slightly, both compositionally and petrographi-cally, from the host meteorite (Fruland et al., 1978). They consistlargely of FeO-rich olivine (Fa0–45) and Ca–Fe pyroxene (Frulandet al., 1978; Brearley and Jones, 1998). Dark inclusions are also fre-quently rimmed (Fruland et al., 1978) and likely underwent a

different history of alteration than the surrounding matrix (John-son et al., 1990; Ohnishi and Tomeoka, 2002; Fries et al.,2005a,b). Different types of dark inclusions (e.g., Johnson et al.,1990) indicate that they may have undergone aqueous alterationand subsequent dehydration, some may be fragments of primitiveaccreted materials, and some experienced higher thermal meta-morphism than the host meteorite (Bischoff et al., 2006, and refer-ences therein). They may have undergone some level of processingprior to their incorporation into the CV host (Johnson et al., 1990).Raman spectroscopy of dark inclusions indicates that their carbonis more graphitic, and hence likely more thermally metamor-phosed, than carbon in adjacent matrix (Fries et al., 2005b).

2.2.6. Thermally metamorphosed clastsJogo and Krot (2010) and Jogo et al. (2011) reported the pres-

ence of thermally metamorphosed CV-like clasts in the Mokoiaand Y-86009 CV breccias. The identified clasts contain a varietyof textural and mineralogic features that suggest strong prolongedthermal metamorphism or annealing on a CV parent body.

2.2.7. CV brecciasA number of CVs contain clasts that differ from their host mete-

orites. Such breccias can include mixtures of CVR and CVOx materi-als, and CV clasts of different metamorphic grades (Bischoff et al.,2006, and references therein). Some of the CVs included in thisstudy (e.g., Mokoia, Vigarano) are regolith breccias, containing dif-ferent types of clasts (e.g., Krot et al., 1998, 2000).

2.2.8. Opaque mineralsThe opaque mineralogy differs between the oxidized and re-

duced subgroups. In CVR chondrites, metal and pyrrhotite are thedominant phases, while in CVOx chondrites metal is rare and mag-netite is the dominant phase (McSween, 1977; Brearley and Jones,1998). Opaque phases generally occur in chondrules rather thanmatrix (McSween, 1977). It also appears that average magnetitegrain size increases with increasing thermal metamorphism(Emmerton et al., 2011).

2.2.9. Carbon and carbonaceous phasesBesides metal, magnetite, and sulfides, the major opaque phase

in CVs is carbonaceous material. Carbon abundance in CVs rangesfrom 0.01 to 1.5 wt.% (Pearson et al., 2006; Alexander et al.,2007), and decreases with increasing metamorphic grade (Guimonet al., 1995). The insoluble organic matter in CVs has generally low-er H/C, N/C, and O/C ratios than other CCs (Alexander et al., 2007).CVs also contain well-crystallized graphite and carbonaceous glob-ules that are consistent with thermal metamorphism (Aoki andAkai, 2008), as are changes in various other organic compounds(Aponte et al., 2011). Carbon in CVs occurs in a variety of forms,including small (�10 lm) inclusions in chondrules (Fries et al.,2005a). Carbon in chondrules seems to have been more thermallymetamorphosed than the carbon in CV matrix (Fries and Bhartia,2010). Well-ordered graphite is more prevalent in CVs that seemto have experienced higher degrees of thermal metamorphism(Aoki and Akai, 2008). H/C ratios for insoluble organic matter aresuggestive of lower values in CVR versus CVOx (Alexander et al.,2007). Infrared spectra of CV insoluble organic matter differ fromthe material from low petrographic type CI, CM, CR and CO chon-drites, and are consistent with more condensed aromatics formedby thermal metamorphism (Kebukawa et al., 2011).

2.2.10. Aqueous alteration and secondary mineralsAqueous alteration in CVs ranges from extremely minor (e.g.,

Mokoia, Grosnaja, Vigarano) to highly altered with extensivedevelopment of phyllosilicates (e.g., Bali) (e.g., Brearley and Jones,1998; Gyollai et al., 2011), distinct aqueously altered clasts


(e.g., Tomeoka and Ohnishi, 2011), and can also be variable withina single meteorite (Noguchi et al., 2003; Komatsu et al., 2008). Thephyllosilicate mineralogy of CV3 chondrites is dominated by low-Al fine-grained saponites (Fe/(Fe + Mg) = 0–0.2), and various micas(Keller and Buseck, 1990b; Keller and McKay, 1993); serpentineand chlorite are also present in Grosnaja (Keller and McKay,1993). The saponite may be associated with fine-grained magnetite(Brearley and Jones, 1998). Ferric iron (at least in the tetrahedralsites of phyllosilicates) was inferred from stoichiometric analysisof Mokoia phyllosilicates (Cohen et al., 1983). Mössbauer analysisof Allende indicates that 10% of the total iron is present as Fe3+,presumed to be present in phyllosilicates (Fisher and Burns,1991). Keller (2011) described an aqueously altered CV (MIL090001: CVR) containing abundant metal and magnetite, as wellas highly altered matrix containing serpentine and chlorite; theytentatively classified this meteorite as a CV2, suggesting that aque-ous alteration has extensively affected some CV chondrites.

3. Experimental procedure

Many of the details of our experimental procedure have beendiscussed in previous papers in this series (Cloutis et al.,2011a,b). A total of 11 CVs, including members of the CVOxB, CVOxA,and CVR subgroups were included in this study (Table 6). As in ourprevious studies of CCs, we have applied continuum removal to thespectra to isolate absorption features of interest and use variousmetrics of spectral slope and overall reflectance to search for sys-tematic spectral–compositional trends (Cloutis et al., 2011a,b).Descriptions of the CVs are provided in Appendix A. The variousreflectance spectra included in this study are available as an on-line supplement.

Reflectance spectra used in this study were measured at the RE-LAB facility at Brown University (Pieters, 1983) at i = 30� and e = 0�and a 5 nm spectral sampling interval and spectral resolution of<5 nm), from 0.3 to 2.6 lm, and at the PSF facility at the Universityof Winnipeg at i = 30� and e = 0� and 1 nm spectral output, from0.35 to 2.6 lm. Some older spectra from Gaffey (1974) were digi-tized and are available through the RELAB archive. The RELABand PSF spectra were measured relative to halon or Spectralon�

and corrected for minor irregularities in the reflectance of thesestandards in the 2 lm region. It should be noted that in the ensuingdiscussion we have minimized comparisons of absolute reflectanceand spectral slopes between spectra measured using bidirectionalreflectance versus those measured using an integrating sphere(Gaffey, 1974), as these parameters are dependent on observa-tional conditions (Gradie et al., 1980; Gradie and Veverka, 1982,1986). We also note that sample preparation procedures for thepowders differed somewhat between Gaffey (1974) and this study.For the RELAB and PSF measurements, a powdered sample ispoured into an aluminum cup and the edge of a glass slide is drawnacross the sample to produce a flat surface. In all cases, a flat pow-dered surface is used for the spectral measurements.

We have developed a variety of easily-applicable spectral met-rics for analysis of CCs (Cloutis et al., 2011a,b), and we utilize thesesame metrics for analysis of CVs. These metrics include absolutereflectance at the visible region peak, highest absolute reflectance,various measures of overall spectral slope, band area ratios, andband depths (Cloutis et al., 1986, 2011b). Absorption features inthe 1 lm region were isolated by fitting a straight line continuumthat is tangent to the reflectance spectrum on either side of this re-gion. Absorption features in the 2 lm region were isolated by fit-ting a straight line continuum that is tangent to the reflectancespectrum at the interband peak near 1.5 lm and the long wave-length end fixed at 2.5 lm. Band centers in this region were deter-mined using a combination of chords constructed across the

feature and polynomial fitting. Band depths were calculated usingEq. 32 of Clark and Roush (1984). Band area ratios (BARs) were cal-culated using procedures first outlined in Cloutis et al. (1986):band IIA/IA and band II/band I.

4. Spectral properties of constituent phases

The spectral reflectance properties of most CV chondrite con-stituents have been discussed in previous papers in this series.The main minerals that are expected to affect CV reflectance spec-tra are olivine, Fe-bearing pyroxene, spinel, melilite, metal, magne-tite, and carbonaceous phases.

Briefly, olivine exhibits an absorption feature consisting of acentral band near 1.05 lm (due to crystal field transitions in Fe2+

located in the M2 site) and side bands near 0.85 and 1.3 lm dueto crystal field transitions in Fe2+ located in the M1 site. The posi-tion of this feature moves from �1.045 to �1.085 lm with increas-ing Fe2+ content (King and Ridley, 1987). Low-Ca pyroxene exhibitstwo strong absorption bands near 0.9 (band I) and 1.9 lm (band II)due to crystal field transitions in Fe2+ located in the pyroxene M2site; weaker bands may also be present near 0.9 and 1.2 lm whenFe2+ is located in the M1 site (Adams, 1974). With increasing Fe2+

and Ca content, the M2 bands move to longer wavelengths, rangingfrom �0.90 to 0.94 lm (band I) and �1.80 to 2.08 lm (band II) forlow-Ca pyroxenes, and �0.93 to 1.08 lm (band I) and �1.90 to2.37 lm (band II) for high-Ca pyroxenes (Adams, 1974; Cloutisand Gaffey, 1991; Klima et al., 2007). Fassaite has a Fe2+–Fe3+

charge transfer band in the 0.75 lm region, and M1 Fe2+ crystalfield transition bands near 0.95 and 1.15 lm (Cloutis and Gaffey,1991). Spinel has a strong absorption band near 2 lm due to tetra-hedrally coordinated Fe2+ crystal field transitions (Cloutis et al.,2004). Melilite has an absorption feature in the 1.6–2.0 lm regionattributable to crystal field transitions in tetrahedrally coordinatedFe2+. Meteoritic metal has a red-sloped and otherwise featurelessspectrum (Cloutis et al., 2010), although chondritic metal appearsto be flat-sloped (Gaffey, 1986). Magnetite has a broad absorptionband in the 1 lm region due to crystal field transitions in octahe-drally coordinated Fe2+ (Sherman et al., 1982). Carbonaceousphases may be flat or red-sloped depending on structure and com-position (Cloutis et al., 2011a,b). Reflectance spectra of a number ofconstituent CV phases are shown in Fig. 1.

5. Results

Our analysis of CVs includes a variety of grain sizes for a num-ber of the CVs, duplicate spectra, spectra of separate splits, andwhole rock spectra (Table 7). We also include CV spectra measuredby previous investigators in our analysis (Table 8). However, thenumber of individual CVs is limited (Table 7), therefore we maynot capture the full range of CV spectral variability, and the num-ber of CVs in the various subgroups is also limited.

5.1. Spectra of CV3, CV3R, CV3Ox, CV3OxA, and CV3OxB

Our survey and analysis of CV spectra focuses on powderedsamples because they will likely be the most representative ofthe spectral properties of CV parent bodies, which are expectedto have powdered regoliths (Dollfus, 1971; Dollfus et al., 1975).

Of the 11 available CVs, three are only described as CV3 withoutinformation on whether they belong to the oxidized or reducedsubgroups. Their reflectance spectra are shown in Fig. 2a. Withonly limited information on these samples, it is difficult to makeany firm observations. ALHA 81003 has been assigned a low petro-logic subtype (3.0) by Bonal et al. (2006) and has lower overallreflectance than the other two CVs, for which petrographic

Fig. 1. RELAB reflectance spectra of some CV constituents. (a) Olivine and pyroxene(<45 lm grain size). (b) CAI constituents: anorthite, fassaite, melilite, and spinel(<45 lm grain size). (c) Opaque analogues: Fe–Ni metal, magnetite, graphite,carbon black (grain sizes indicated for each).

Table 7CV reflectance spectra included in this study.

Meteorite Sample type/grain size RELAB or PSF file ID

ALH 81003 <125 lm c1mt16ALH 84028 <125 lm c1mb78, c2mb78

<25 lm camb7825–45 lm cbmb7845–75 lm ccmb7875–125 lm cdmb78<500 lm c1mc05

ALH 85006 <125 lm c1mt46Allende Roughened slab (6 spots) 120215a.001-006

<125 lm 120216a.022, 120216a.023CAI unsorted powder caca00<150 lm (with mask) 90814c.009, 90814a.010<150 lm 120216a.001, 120216a.009<180 lm camh57, cbmh57<150 lm mgp124<45 lm s1rs40<1000 lm cls746<125 lm c2mb63<63 lm c3mb6363–125 lm c4mb63<63 lm, 400 �C camb63<63 lm, 500 �C cbmb63<63 lm, 600 �C ccmb63<63 lm, 700 �C cdmb63<63 lm, 800 �C cemb63<63 lm, 900 �C cfmb63<63 lm, 1000 �C cgmb63<63 lm, 1100 �C chmb63<63 lm, 1200 �C cimb63500–2000 lm (JFB-01) 120216a.006, 120216a.0140–>2000 lm (JFB-02) 120216a.005, 120216a.01375–250 lm (JFB-03) 120216a.008, 120216a.016<150 lm (JFB-04) 120116a.007, 120216a.015

Grosnaja <75 lm mgp12875–150 lm mgp130150–500 lm mgp132<45 lm s1rs43

Leoville <150 lm mgp134<150 lm (with mask) 091104a.017, 091104a.018<150 lm 120216a.004, 120216a.012

Mokoia <150 lm mgp136<150 lm (with mask) 90814c.011, 90814c.012<150 lm 120216a.002, 120216a.010

NWA 3118 <125 lm c1mp129<125 lm, dark inclusion c1mp130

QUE 93744 <75 lm c1ph41Vigarano <150 lm mgp138

<100 lm c1mb59<150 lm (with mask) 90817b.002, 90817b.003<150 lm 120216a.003, 120216a.011

Y-86751 <125 lm c1mp09

Note: ‘‘mgpxxx’’ spectra were acquired with an integrating sphere (Gaffey, 1974)and have been subsequently digitized; all other spectra were acquired at i = 30� ande = 0�.


subgrades are not available; it is also the most blue-sloped. Allthree spectra show absorption bands in the 1 lm region, centerednear 1.05 lm and an inflection near 1.3 lm – these features areconsistent with Fe-rich olivine, and their presence is expected gi-ven the high olivine abundance in CV3s. They also show a weakerabsorption feature centered near 2.1 lm. This feature is most con-sistent with spinel, as discussed below.

Three of our CVs (ALH 84028, Allende, and Mokoia) have beencategorized as subtype CV3OxA, and their spectra are shown inFig. 2b. Guimon et al. (1995) classified them all as CV3.2, althoughBonal et al. (2006) suggested that Allende is much more metamor-phosed (>CV3.6), and Mokoia is �CV3.6. Mokoia and Allende havesimilar overall reflectance, in spite of a small difference in maxi-mum grain size and the fact that Mokoia was measured using anintegrating sphere (Gaffey, 1974). Both are red-sloped, in contrast

to ALH 84028 which is neutral to blue-sloped. All three spectra ex-hibit an absorption band in the 1 lm region, near 1.05 lm for ALH84028 and Allende, while Mokoia shows two apparent absorptionfeatures near 0.87 and 1.05 lm. ALH 84028 and Allende also showa weak absorption band in the 2 lm region, which is not as appar-ent in the Mokoia spectrum.

Grosnaja is the only CVOxB in our suite, Y-86751 is categorizedas CV3OxA–OxB (Komatsu et al., 2007), and ALH 85006 is only cate-gorized as CVOx. Their spectra are neutral to blue-sloped, and onlyY-86751 has well-defined 1 and 2 lm region absorption bands(Fig. 2c). Fine-grained (<74 lm) Grosnaja powder spectrally char-acterized by Johnson and Fanale (1973) is slightly red-sloped withwell-defined 1 and 2 lm region absorption features.

Two CVs in our suite have been categorized as CV3R: Leovilleand Vigarano. They both exhibit a broad concave-shaped region

Table 8Previous spectral reflectance studies of CVs (�0.3–2.5 lm) included in this study.

Meteorite Sample type Source

ALH84028

<125 lm (0.3–2.5 lm) 1

Allende <75 lm (0.6–2.5 lm) 2<125 lm (0.3–2.5 lm) 1<150 lm (0.35–2.5 lm) 3Carbonaceous extract (0.35–2.5 lm) 4Room temp, 800 �C (IW + 2; IW � 1), <100 lm (0.2–2.5 lm)

5

White inclusions on slab (0.6–2.5 lm) 6AOIs, fluffy type A CAI, type B CAI,CAI-free matrix, bulk; all <38 lm (0.3–2.5 lm) 7<74 lm (0.35–2.5 lm) 8[150 lm (0.45, 0.55, 0.65, 0.70, 0.95 lm) 9[150 lm (0.5–>2.5 lm) 10<100 s of lm (0.633 lm) 11Solar furnace condensates (0.45–2.5 lm) 12<45 lm, 45–75 lm, 75–150 lm (0.4–0.6 lm) 13<75 lm (0.4–1.2 lm) 14, 15,

16Grosnaja <150 lm (0.35–2.5 lm) 3

147–495 lm (0.35–2.5 lm) 474–147 lm (0.35–2.5 lm) 4<74 lm (0.35–2.5 lm) 4

Leoville <150 lm (0.35–2.5 lm) 3Mokoia <150 lm (0.35–2.5 lm) 3Vigarano <150 lm (0.35–2.5 lm) 3

Sources of data: [1] Hiroi et al. (1993). [2] Burbine et al. (1992). [3] Gaffey (1974).[4] Johnson and Fanale (1973). [5] Miyamoto et al. (2000). [6] Rajan and Gaffey(1984). [7] Sunshine et al. (2008a,b). [8] Salisbury et al. (1975). [9] Beck et al.(2012). [10] Grisolle et al. (2011). [11] Kamei and Nakamura (2002). [12] King et al.(1983). [13] French and Veverka (1983). [14] Gradie et al. (1980). [15] Gradie andVeverka (1982). [16] Gradie and Veverka (1986).

Fig. 2. RELAB reflectance spectra of powders of CV chondrites (grain size indicated for eacThe Mokoia (2b), and Leoville and Vigarano (2c) spectra are from Gaffey (1974); i.e., me


of reflectance, with an overall neutral to blue slope (Fig. 2d), and incontrast to the other CV spectra, have less well-defined absorptionbands in the 1 and 2 lm regions. One possibility is that the lack ofwell-defined mafic silicate absorption bands is due to their lowerabundance of Fe-bearing silicates compared to other CVs (Blandet al., 2008; Table 3).

5.2. Metamorphic subtypes

The CV spectra can also be compared on the basis of metamor-phic subtype. As has been found for other CCs, we expect increas-ing metamorphism to lead to increasing reflectance, bluer slopes,and better-defined mafic silicate absorption bands due to aggrega-tion/loss of organics, and crystallization of amorphous/poorlycrystalline silicates. It is also possible that metal/silicate re-equili-bration could lead to incorporation of iron into the silicates. Fig. 3ashows spectra of powdered CVs along with the available subtypesas determined by Guimon et al. (1995) and Bonal et al. (2006). Be-sides the aforementioned difference in slope between the reducedand oxidized groups, there appears to be no systematic variation asa function of metamorphic subtype, even when the oxidized andreduced subgroups are separated. The 1 lm continuum-removedspectra (Fig. 3b) do not show any systematic relationships betweenband depths or positions with petrographic subtype. Most of theseCVs show band minima near the expected position for olivine(�1.05 lm), while Vigarano (CV3.0–3.4) shows a longer than ex-pected band minimum. The 2 lm region continuum-removedspectra (Fig. 3c) show much more diverse behavior. Well-resolvedspinel absorption features near 1.95 and 2.1 lm are not present inall the spectra. An absorption feature in the 2.4 lm region ofsome of the spectra may be attributable to fassaite (Rajan andGaffey, 1984).

h). (a) CV3, unspecified subtype. (b) CV3OxA. (c) CV3Ox, CV3OxA–OxB, CV3OxB. (d) CV3R.asured using an integrating sphere.

Fig. 3. Reflectance spectra for CVs for which subtypes have been determined by Guimon et al. (1995) and Bonal et al. (2006). (a) RELAB powder bidirectional spectra. (b) Sameas (a) for continuum-removed 1 lm region. (c) Same as (a) for continuum-removed 2 lm region. (d) CV spectra measured by Gaffey (1974) with an integrating sphere anddigitized by RELAB. (e) Same as (d) for continuum-removed 1 lm region. (f) Same as (d) for continuum-removed 2 lm region. (g) Gaffey (1974) samples remeasured at PSF inbidirectional (i = 30�, e = 0�) mode. (h) Same as (g) for continuum-removed 1 lm region. (i) Same as (g) for continuum-removed 2 lm region. Linear vertical offsets applied tocontinuum-removed spectra are indicated in brackets on the relevant figures.


If we limit our analysis to CV spectra of a common grain size(<150 lm) acquired with an integrating sphere (Gaffey, 1974),we find that the two most metamorphosed CVs (Bonal et al.,2006) are more red-sloped (Fig. 3d), opposite the expectation ofbluer slopes with increasing metamorphism. Bluer slopes are ex-pected as carbonaceous phases become increasingly condensedand C-rich (Cloutis, 2003). Overall reflectance also does not corre-late with metamorphic grade. The 1 lm region continuum-re-moved spectra (Fig. 3e) show no systematic variations withmetamorphic grade; absorption features near 0.87–0.90 may beartifacts, as they do not appear in RELAB and PSF CV spectra. The2 lm continuum-removed spectra (Fig. 3f) all have a broad,

shallow, and generally featureless absorption feature, and noapparent correlations of depth with metamorphic grade.

The Gaffey (1974) samples were re-measured at the PSF for thisstudy. Their reflectance spectra (Fig. 3g) vary from red- to blue-sloped, but there is no systematic relationship between absolutereflectance or slope with metamorphic grade. The continuum-re-moved 1 lm region spectra all show an olivine-associated absorp-tion band in the 1.30–1.35 lm region (similar to the Gaffey (1974)spectra. Band minima are at generally longer wavelengths than forpure olivine, suggesting either an inappropriate continuum orspectral contributions from other phases; other possible phasesinclude amorphous or poorly crystalline Fe-bearing phases,

Fig. 3 (continued)


magnetite, or fassaite. The 2 lm continuum-removed spectra allshow evidence of spinel absorption features near 1.95 and2.1 lm; band depths do not appear to be correlated with metamor-phic grade, however. The lack of systematic spectral variationswith metamorphic grade is likely attributable to limited numberof CVs, grain size variations, heterogeneities within any CV, anddisagreements on degree of metamorphism (Bonal et al., 2006).

5.3. Duplicate spectra

Duplicate spectra are available for a number of the CVs, allinvolving the same split, with some comparative measurementsavailable for bidirectional versus integrating sphere. Duplicatespectra allow us to explore spectral variability within individualCV chondrite subsamples.

Fig. 4a and b shows six �8 mm diameter spots on a slab of theAllende CV chondrite that were spectrally characterized. No effortwas made to target specific phases in the slab. The reflectancespectra are all strongly blue-sloped with peak reflectance near0.55–0.65 lm and with maximum reflectance ranging from 8.6%to 10.5%. The cause of a blue slope in solid samples is not yet wellunderstood. All the spectra exhibit absorption features in the 1 and2 lm regions that show little variation from spot to spot. The 1 lmregion continuum-removed spectra (Fig. 4c and d) are consistentwith Fe-rich olivines: an absorption band centered near 1.05–1.07 lm, an inflection near 0.85 lm, and a better-defined absorp-tion band near 1.3 lm. The well-defined nature of this latter fea-ture is characteristic of ferrous olivines (King and Ridley, 1987).The 2 lm region continuum-removed spectra (Fig. 4e and f) exhibitabsorption features near 1.95 and 2.1 lm. The 1.95 lm featuremay be attributable to hydrated phases (which are rare in Allende)

or, more likely, to Fe2+-bearing spinels in CAIs (Cloutis et al., 2004).This phase would also account for the 2.1 lm feature and overallbroadness of the 1.9–2.1 lm interval.

Duplicate spectra of <125, <150, <180, 75–250, 500–2000, and0–>2000 lm fractions of Allende are also available for comparison.The <125 lm fraction spectra (Fig. 4g) are very similar: both arered-sloped and exhibit 1 and 2 lm region absorption features.The continuum-removed spectra show a few differences. The1 lm region spectra (Fig. 4h) have the characteristic features ofolivine, but one of the spectra shows a broadened 1 lm absorptionband minimum. The 2 lm region spectra (Fig. 4i) are similar to thecontinuum-removed slab spectra, with absorption features near1.95 and 2.1 lm.

The <150 lm fraction of Allende was measured with and with-out a black mask over part of the field of view of the spectrometer(to ensure that only a flat sample surface is viewed by the spec-trometer, due to the limited sample size). The spectra show slightdifferences in slope and position of the peak near 0.7 lm, and theunmasked spectra show differences in overall reflectance, whichmay be attributable to differences in sample surface texture(Fig. 4j). The 1 lm region spectra (Fig. 4k) all show the expectedolivine absorption features near 1.07 and 1.3 lm, with some differ-ences in the appearance of these bands in terms of the width of theband minimum and intensity of the 1.3 lm band. The 2 lm regionspectra (Fig. 4l) are broadly similar to each other and to other slaband powder Allende spectra, with a broad absorption feature cen-tered near 2.1 lm, and an additional feature near 1.95 lm. Two ofthe spectra also show evidence of weak bands near 1.67 and2.35 lm, that may be attributable to the carbonaceous phases inthe sample or organic contaminants (as the history of this sampleis unknown).

Fig. 4. Reflectance spectra of Allende. (a and b) PSF spectra of random spots on roughened slab. (c and d) Same as (a and b) for continuum-removed 1 lm region. (e and f)Same as (a and b) for continuum-removed 2 lm region. (g) PSF < 125 lm powder duplicate spectra. (h) Same as (g) for continuum-removed 1 lm region. (i) Same as (g) forcontinuum-removed 2 lm region. (j) <150 lm Gaffey (1974) sample measured with and without a mask at PSF, those measured with a mask were done to minimize sampleholder contributions. (k) Same as (j) for continuum-removed 1 lm region. (l) Same as (j) for continuum-removed 2 lm region. (m) <180 lm powder duplicate RELAB spectra.(n) Same as (m) for continuum-removed 1 lm region. (o) Same as (m) for continuum-removed 2 lm region. (p and q) Duplicate spectra of four different size powdersmeasured at PSF (grain sizes indicated for each sample). (r and t) Same as (p and q) for continuum-removed 1 lm region. (s and u) Same as (p and q) for continuum-removed2 lm region. Linear vertical offsets applied to continuum-removed spectra are indicated in brackets on the relevant figures.


The two <180 lm fraction spectra (Fig. 4m) are similar in over-all shape, but differ by �1% in absolute reflectance. The contin-uum-removed spectra (Fig. 4n and o) are again consistent witholivine, although the 1.3 lm absorption feature is less apparentin these spectra. The 2 lm region shows the expected broadabsorption band centered near 2.1 lm, but the spectra are noisierthan other Allende spectra.

Coarser-grained Allende duplicate powders spectra are shownin Fig. 4p. The two sets of powders, with grain sizes of 500–

2000 lm and 0–>2000 lm exhibit well-resolved olivine and spinelabsorption bands. The sample with the finest fraction removed(JFB-01; 500–2000 lm) is more blue-sloped. The spectra have var-iable overall reflectance, likely attributable to differences in grainsize distributions at the surfaces of the samples between duplicateruns (the sample cups were emptied and repacked for the dupli-cate measurements). Two other sets of Allende powders (Fig. 4q),with sizes of 75–250 lm and <150 lm show little variability with-in each pair, with overall reflectance varying by <1% absolute

Fig. 4 (continued)


between duplicate spectra. As with the previous Allende spectra,the sample from which the finest fraction has been removed(JFB-03: 75–250 lm) is darker and bluer-sloped than the sampleswith the finest fraction retained (JFB-04: <150 lm). These spectraare also characterized by recognizable olivine and spinel absorp-tion bands. The continuum-removed 1 lm region spectra of thecoarsest fractions (Fig. 4r) suggest that the sample with the finestfraction removed (JFB-01) has a less well resolved olivine absorp-tion feature in the sense that the band minimum is broader andis shifted to longer wavelengths. Band depths for each duplicatepair vary by up to 1.5% absolute for JFB-01 and 0.4% absolute forJFB-02. The 2 lm continuum-removed spectra (Fig. 4s) also showvariability in band depths, again greater for JFB-01 (3.5% versus5.5%) than for JFB-02 (2.8% versus 4.0%), but spinel-associated fea-tures are apparent in all the spectra.

For the finer-fraction spectra (Fig. 4t), the 1 lm continuum-re-moved spectra show less variability within each pair, with banddepth varying by <0.4% absolute. The continuum-removed 2 lmregion spectra (Fig. 4u) also show less variability within pairs, withband depths varying by <0.5% absolute. Band ‘‘centers’’ show somevariability within pairs, likely due to their broad, shallow nature.

Duplicate spectra of a <150 lm fraction of Leoville were alsomeasured with and without a black mask (Fig. 5a). The four spectraare similar in terms of overall reflectance and shape. The 1 lm-re-gion continuum-removed spectra (Fig. 5b) show absorption fea-tures near 0.95, 1.1, and 1.3 lm, whose intensity variessomewhat among the different spectra. The 0.95 lm feature maybe attributable to low-calcium pyroxene, which is abundant in Leo-ville (Krot et al., 1998) or hydrated silicates, which are also presentin Leoville (Oliver, 1978; Kracher et al., 1985). The 2 lm-region

Fig. 4 (continued)


continuum-removed spectra (Fig. 5c) are all similar, with weakabsorption features in this region, but resolvable bands near 1.95and 2.1 lm.

Reflectance spectra of a <150 lm fraction of Mokoia, measuredwith and without a black mask (Fig. 5d) show similar overall reflec-tance, although the masked spectra are more red-sloped than theunmasked sample spectra. The cause of this reddening is unknown,but we do not believe it is attributable to the mask, as this effect isnot seen in other masked spectra. The 1 lm region continuum-re-moved spectra (Fig. 5e) show some variability in band centers,with the 1 lm region centers varying from �1.05 to �1.13 lm;the 1.3 lm region feature occurs near 1.33 lm. This could be dueto improper removal of a continuum or mineralogical variations;the broad and shallow nature of the 1 lm region absorption

feature also makes it sensitive to choice of continuum. The 2 lmregion continuum-removed spectra (Fig. 5f) show weak absorptionfeatures; noise in the masked spectra makes continuum construc-tion and band identification difficult, but the unmasked spectra ex-hibit the familiar spinel features near 1.95 and 2.1 lm, although acontribution by adsorbed water to the 1.95 lm feature cannot beruled out.

The Vigarano spectra (Fig. 5g) show no appreciable differencebetween duplicate or between masked and unmasked spectra.The continuum-removed 1 lm region spectra (Fig. 5h) show smalldifferences in band shapes and positions between duplicate pairs,suggesting some variability this single sample. All the spectra exhi-bit ‘‘olivine-like’’ absorption features. As with Mokoia, the masked2 lm region continuum-removed spectra (Fig. 5i) are noisy, mak-

Fig. 4 (continued)


ing continuum construction and band identification difficult. Nev-ertheless, the duplicate pairs of spectra show similar absorptionbands and depths.

Duplicate spectra suggest that the CV chondrites included inthis study can be spectrally heterogeneous. However, spectral dif-ferences for duplicate samples are generally minor, restricted tosmall variations in absorption band shapes, as well as absolutereflectance. The differences are small enough that they are unlikelyto affect meteorite comparisons with asteroids.

5.4. Grain size effects and slabs versus powders

For a number of our CV samples, we have reflectance spectra fordifferent grain sizes, allowing us to examine how grain size affectsCV reflectance spectra. One caveat is that some of the grain sizespectra may be for different subsamples of a particular CV, thusthere may be compositional heterogeneities between differentsample splits. A number of the CVs are breccias (e.g., Mokoia,Vigarano), introducing the possibility of heterogeneity betweendifferent subsamples. In addition, crushing a sample to make dif-ferent size fractions may result in preferential depletion ofenhancement of specific phases in different size fractions, due todifferences in the strength of different phases.

5.4.1. ALH 84028For ALH 84028, six different grain size sample spectra are avail-

able (Table 7). For the ALH 84028 for which only a maximum grainsize is specified (Fig. 6a), the spectra get progressively darker andbluer with increasing average grain size. Reflectance in the0.7 lm region varies from �11% to 13.5%. The 1 lm region

continuum-removed spectra (Fig. 6b) show variations in band cen-ters and overall absorption feature shape, suggesting that compo-sitional variations may exist between different subsamples ofthis meteorite. While band position is not always within the pureolivine field (1.045–1.085 lm), the olivine-associated inflectionnear 1.3 lm is present in all of the spectra; band depths do notvary systematically with average grain size. This could be due tothe competing effects of increasing grain size leading to lowerreflectance and deeper silicate absorption bands, at least to thepoint where silicate bands become saturated, possible mineralogi-cal differences between the subsamples, and the relative impor-tance of comminuted dark versus bright components (Johnsonand Fanale, 1973). The shorter-than expected band positions couldbe associated with the presence of pyroxene (whose abundance inthis meteorite is unknown) or Fe oxyhydroxide terrestrial weather-ing products (Schwarz and MacPherson, 1985). The 2 lm regioncontinuum-removed spectra (Fig. 6c) are all generally broad andfeatureless, consistent with the presence of Fe2+-bearing spinels.Band depth in this region generally increases with increasing aver-age grain size.

The ALH 84028 spectral series with increasing average grainsize (presumably made from a single subsample of ALH 84028)show increasingly blue slopes and decreasing overall reflectance(Fig. 6d), consistent with the results of Johnson and Fanale(1973) for CVs. The increasingly blue slope also shifts the 0.7 lmlocal maximum to shorter wavelengths. The 1 lm continuum-re-moved spectra (Fig. 6e) show small non-systematic differences inband center (�1.04–1.08 lm) and the appearance of the 1.3 lmolivine band; band depths are essentially constant for the differentsize fractions. As with the continuum-removed 2 lm region

Fig. 5. Bidirectional reflectance spectra measured at PSF for Gaffey (1974) <150 lm CV3s. Spectra were measured with and without a mask, those measured with a mask weredone to minimize sample holder contributions. (a) Leoville. (b) Same as (a) for continuum-removed 1 lm region. (c) Same as (a) for continuum-removed 2 lm region. (d)Mokoia. (e) Same as (d) for continuum-removed 1 lm region. (f) Same as (d) for continuum-removed 2 lm region. (g) Vigarano. (h) Same as (g) for continuum-removed 1 lmregion. (i) Same as (g) for continuum-removed 2 lm region. Linear vertical offsets applied to continuum-removed spectra are indicated in brackets on the relevant figures.


spectra of the other fractions of ALH 84028, this second series ofspectra (Fig. 6f) shows little detail, and band depths increase withincreasing grain size.

5.4.2. AllendeAllende is the most spectrally characterized CV (Table 7), with

multiple grain size fractions and slab spectra. Spectra of Allendepowders with no constrained minimum grain size (Fig. 7a) showno systematic variation in overall reflectance, suggestive of varia-tions in absolute reflectance between different subsamples. Thefines-free spectrum (63–125 lm) has the lowest overall reflectanceand bluest spectral slope, similar in behavior to previously mea-sured CV spectra (Johnson and Fanale, 1973). Band shapes and

depths are similar for the <45, <63, <125, and <180 lm fractionsin the 1 lm region, with band centers ranging between �1.05and 1.07 lm (Fig. 7b). The 2 lm region continuum removed spec-tra (Fig. 7c) all exhibit a broad shallow absorption band, but appar-ent band centers vary.

The Allende spectra with constrained minimum grain sizes(JFB01: 500–2000 lm; JFB3: 75–250 lm) are also blue sloped;the inclusion of the finest fraction (JFB-02: 0–>2000 lm; JFB-03:<150 lm) results in the appearance of an overall red slope. Thesefour size separates of Allende exhibit variations in both overallreflectance and spectral slope (Fig. 7d).

Comparing slab spectra to a <125 lm powder made from thesame slab (Fig. 7e), the slab spectra are darker and more

Fig. 5 (continued)


blue-sloped, and all show absorption features in the 1 and 2 lm re-gion. The continuum-removed 1 lm (Fig. 7f) and 2 lm (Fig. 7g)absorption features are similar in terms of overall shape, band cen-ter, and band depth, suggesting that straight line continuum re-moval is effective for recovering these features in spite ofvariations in slope and overall reflectance, and that band depthsare, unexpectedly, relatively insensitive to grain size variations.

5.4.3. GrosnajaSpectra of four size separates of Grosnaja are shown in Fig. 8a

(the <75, 75–150, and 150–500 lm spectra are from Gaffey(1974)). The spectra show no systematic variations in absolutereflectance, although the coarser samples are darker, as expected.The absorption feature in the 1 lm region (Fig. 8b), is located inthe 1.0–1.05 lm interval; the <75 lm spectrum shows an addi-tional absorption band near 0.9 lm which may be a spectral arti-fact. The 1.3 lm olivine absorption feature is also less apparentin the coarser grained sample spectra. The 2 lm region absorptionfeature is deepest in the finest fraction (<45 lm), with well-definedbands near 1.95 and 2.27 lm (Fig. 8c).

5.4.4. VigaranoSpectra of two size fractions of Vigarano, likely from different

subsamples are shown in Fig. 9a. The <150 lm fraction was mea-sured with an integrating sphere (Gaffey, 1974), and the<100 lm fraction was measured in bidirectional mode. The twospectra have similar overall reflectance, within �1% absolute ofeach other, and have flat overall slopes and weak 1 and 2 lm re-gion absorption features. The 1 lm region continuum-removedspectra (Fig. 9b) are different in all respects, and some of this var-iation may be due to the weakness of this absorption band, and

possibly mineralogic differences between the two subsamples.Both spectra show only weak evidence for the presence of olivine,as band centers fall outside the olivine range, but both do exhibitevidence of the 1.3 lm olivine band. The variations in band centerfor such weak bands are sensitive to the choice of continuum tan-gent points. The 2 lm region continuum-removed spectra (Fig. 9c)are also different. The <150 lm spectrum has a deeper absorptionfeature, centered near 2.1 lm, consistent with spinel.

5.5. Viewing geometry effects (bidirectional versus integrating sphere)

Four of the CV samples (<150 lm) measured by Gaffey (1974)using an integrating sphere (and subsequently digitized at RELAB)were remeasured in bidirectional mode at the PSF (i = 30�, e = 0�):Allende, Leoville, Mokoia, and Vigarano. These spectra allow us toexamine the effects of changing viewing geometry. Gradie et al.(1980), and Gradie and Veverka (1982, 1986) found that viewinggeometry affects spectral slopes for Allende, and therefore compar-isons of asteroid and laboratory spectra should be approached withthis effect in mind. They also found that bidirectional measure-ments differ from integrating sphere measurements, with bidirec-tional spectra being generally redder than integrating spherespectra, and the degree of difference increasing with increasingbrightness. They determined that integrating sphere measure-ments are preferred for analysis of asteroidal spectra obtained atsmall phase angles. Differences in absolute reflectance on the orderof 10%, relative, were found between integrating sphere and bidi-rectional measurements at a phase angle of 60� at 1.2 lm.

Gradie et al. (1980) and Beck et al. (2012) found that Allendeshows spectral reddening, similar to other meteorites, withincreasing phase angle in bidirectional measurements. French

Fig. 6. Reflectance spectra of various fractions and samples of ALH 84028 measured at RELAB. (a) Fractions with no minimum grain size or unknown grain size distribution.(b) Same as (a) for continuum-removed 1 lm region. (c) Same as (a) for continuum-removed 2 lm region. (d) Grain size series for ALH 84028. (e) Same as (d) for continuum-removed 1 lm region. (f) Same as (d) for continuum-removed 2 lm region. Linear vertical offsets applied to continuum-removed spectra are indicated in brackets on therelevant figures.


and Veverka (1983) found that limb darkening (decrease in reflec-tance with increasing emission angle) was independent of wave-length over the 0.4- to 0.6-lm range. Kamei and Nakamura(2002) and Beck et al. (2012) found that reflectance of Allendepowders decreases with increasing phase angle up to �60–90�and then increases toward larger phase angles.

The integrating sphere spectrum of Allende (Fig. 10a) is morered-sloped and slightly brighter than the bidirectional spectra. Allthree spectra show 1 and 2 lm region absorption bands. The con-tinuum-removed 1 lm region spectrum of the integrating spheredata are noisier than the bidirectional data (Fig. 10b), ‘‘masking’’the olivine signature. Band depths are generally comparable (5–6%). The 2 lm region absorption feature is broad in all the spectra

(Fig. 10c), and the bidirectional data show a better resolvedabsorptions feature in the 2 lm region. Differences in band minimaare likely attributable to the broad and shallow nature of this fea-ture, which makes it sensitive to continuum slope.

The Leoville spectra (Fig. 10d) are similar to each other in termsof overall slope, shape, and absolute reflectance; the integratingsphere spectrum appears to be bluer sloped than the bidirectionalspectra, and has an absorption feature near 0.87 lm (discussed be-low). As with Allende, the integrating sphere spectra are noisier.The 1 lm region continuum-removed spectra (Fig. 10e) differ be-tween the integrating sphere and bidirectional spectra in termsof apparent absorption band positions. The bidirectional spectrahave bands near 1.1 and 1.32 lm; these are at longer than

Fig. 7. Reflectance spectra of Allende. (a) Samples with unconstrained minimum grain size and 63–125 lm measured at RELAB. (b) Same as (a) for continuum-removed 1 lmregion. (c) Same as (a) for continuum-removed 2 lm region. (d) Four different grain size spectra measured at PSF. (e) Slab versus <125 lm powder made from slab. (f) Same as(e) for continuum-removed 1 lm region. (g) Same as (e) for continuum-removed 2 lm region. Linear vertical offsets applied to continuum-removed spectra are indicated inbrackets on the relevant figures.


Fig. 8. (a) RELAB reflectance spectra of Grosnaja. (b) Same as (a) for continuum-removed 1 lm region. (c) Same as (a) for continuum-removed 2 lm region. Linearvertical offsets applied to continuum-removed spectra are indicated in brackets onthe relevant figures.

Fig. 9. (a) RELAB reflectance spectra of Vigarano. (b) Same as (a) for continuum-removed 1 lm region. (c) Same as (a) for continuum-removed 2 lm region. Linearvertical offsets applied to continuum-removed spectra are indicated in brackets onthe relevant figures.


expected wavelengths for olivine and likely are a function of theshallowness of this feature and the lack of a well-defined reflec-tance peak in the 1.5 lm region for fixing a continuum. In the2 lm region (Fig. 10f), the integrating sphere spectrum has a shal-lower absorption feature with no well-defined absorption bands,while the bidirectional spectra are deeper and display absorptionbands near 1.95 and 2.1 lm, as seen in other CV spectra, such asAllende. There are also indications of an additional feature near2.3 lm, which may be attributable to hydrous phases which arepresent in Leoville (Oliver, 1978; Kracher et al., 1985).

The integrating sphere spectrum of Mokoia is brighter and red-der in the visible region, and shows more ‘‘spectral features’’ than

the bidirectional spectra (Fig. 10g). These include the 0.87 and1.37 lm features discussed below. As with some of the other spec-tra discussed above, the positions of the ‘‘olivine bands’’ differ(Fig. 10h): �1.05 and 1.25 lm for the integrating sphere versus�1.11 and 1.32–1.35 lm for the bidirectional spectra. Once again,some of these variations are probably related to their sensitivity tosmall differences in the continuum, and the wavelength offset ofthe spectrometer used to collect the Gaffey (1974) data (McFaddenet al., 1982); however band depths are comparable for all threespectra. Similar to some of the previous CV spectra, the bidirec-tional spectra show better resolved 1.95 and 2.1 lm bands, as wellas a possible 2.3 lm feature (Fig. 10i), but the 2 lm region absorp-tion feature is shallow (<4% depth) and wide.

Fig. 10. Integrating sphere (Gaffey, 1974) versus bidirectional (i = 30�, e = 0�) PSF reflectance spectra of <150 lm CVs and continuum-removed 1 and 2 lm regions,respectively, for: (a–c) Allende; (d–f) Leoville; (g–i) Mokoia; (j–l) Vigarano. Linear vertical offsets applied to continuum-removed spectra are indicated in brackets on therelevant figures.


For Vigarano (Fig. 10j), the integrating sphere spectrum isbrighter and less blue-sloped (redder) than the bidirectional spec-tra, with peak reflectance of 8% versus 5.5%. All the spectra exhibitbroad 1 and 2 lm region absorption features. The continuum-re-moved 1 lm region (Fig. 10k) show deeper 1.1 and 1.3 lm absorp-tion features for the bidirectional spectra. As with Mokoia, thesetwo absorption bands are at longer than expected wavelengthsfor olivine; this could be attributable to a number of causes, suchas amorphous Fe-bearing material, phyllosilicates, or magnetite.Uncertainty in continuum position seems unlikely, as the Vigaranospectra have a well-defined local reflectance maximum in the1.5 lm region. The 2 lm region (Fig. 10l) shows deeper overallabsorption and better-resolved absorption bands near 1.95 and2.1 lm in the bidirectional spectra.

The 0.87 and 1.37 lm features are seen in all the integratingsphere spectra. It seems likely that these are artifacts whose causeis unknown. The 0.87 lm feature could plausibly be assigned to Feoxyhydroxides, but the expected UV–visible reflectance edge is notpresent; the 1.37 lm feature occurs at longer wavelengths than ex-pected for olivine and shorter wavelengths than expected for OH/H2O. The differences in band positions between some of the inte-grating sphere and bidirectional spectra could be due to a wave-length offset that was recognized well after the Gaffey (1974)data were acquired (McFadden et al., 1982). The Gaffey (1974) datahave not been corrected for this offset, as its magnitude is likelywavelength dependent and may have varied over time. The betterresolved nature of the absorption bands in the bidirectional data,particularly in the 2 lm region is likely attributable to the higher

Fig. 10 (continued)


signal to noise of the bidirectional data, as spectral resolution forthe Gaffey (1974) and RELAB data are comparable (5 nm).

The lack of systematic reddening or bluing between integratingsphere and bidirectional measurements is likely due to the ‘‘natu-ral’’ spectral variability that seems to characterize some CVs.Duplicate measurements of various CVs, discussed above, canshow changes in overall slope and absolute reflectance. This canlikely be attributed to differences in surface texture, and the rela-tive abundances of different minerals and grain size distributionin the uppermost surface of the sample when the samples are emp-tied and repacked for duplicate spectral measurements. These dif-ferences appear to become more apparent in the larger grain sizefractions.

The integrating sphere measurements show what are likelysome spurious absorption features. Nevertheless, it appears thatthey type of spectral measurement (integrating sphere, bidirec-tional) does not negatively impact our ability to characterize CVson the basis of their diagnostic absorption features, particularlyFe-rich olivine features in the 1 lm region and Fe2+-bearing spinelfeatures in the 2 lm region.

5.6. CAIs, AOIs

The reflectance spectrum of the unsorted powder (�1 g) of amixture of two large CAIs from Allende (made by extracting theCAIs using a rotary tool) are much brighter than a bulk Allende

Fig. 11. (a) RELAB reflectance spectra of unsorted powdered CAIs from Allendeversus bulk Allende (<45 lm). (b) Same as (a) for continuum-removed 1 lm region.(c) Same as (a) for continuum-removed 2 lm region. Linear vertical offsets appliedto continuum-removed spectra are indicated in brackets on the relevant figures.

Fig. 12. (a) RELAB reflectance spectra of <125 lm NWA 3118 host and darkinclusion. (b) Same as (a) for continuum-removed 1 lm region. (c) Same as (a) forcontinuum-removed 2 lm region. Linear vertical offsets applied to continuum-removed spectra are indicated in brackets on the relevant figures.


powder (Fig. 11a), and absorption features in the 1 and 2 lm re-gion are apparent. The continuum-removed 1 lm region spectrumof the CAI powder consists of a broad region of absorption from�0.6 to �1.4 lm, with suggestions of contributions from a numberof absorption bands; the bulk Allende spectrum, by contrast isdominated by olivine (Fig. 11b). The 1 lm region CAI spectrum ismost consistent with spectral contributions from fassaite, whichhas bands near 0.75, 0.95, and 1.15 lm (Fig. 1b), The 2 lm regionshows how much deeper absorption in this region is for a CAI con-centrate compared to bulk Allende (Fig. 11c). The CAI spectrumshows indications of absorption bands near 1.95 and 2.1 lm, butthese are better expressed in bulk CV spectra, and are most consis-tent with major spectral contributions from Fe2+-bearing spinel

(Fig. 1b). Fassaite does not seem to contribute significantly to the2 lm region of CAI spectra.

Rajan and Gaffey (1984) measured reflectance spectra of vari-ous high-albedo inclusions in a slab of Allende and found diversespectral behavior that could be related to mineralogy. Some weredominated by spinel, others showed spectral contribution fromfassaite and augite (0.9–1.1 and 2.4 lm). In a more detailed studyof Allende inclusions by Sunshine et al. (2008a,b), fluffy type Ainclusions were found to have the most intense 2 lm regionabsorption feature, AOIs were spectrally dominated by olivine,and type B CAIs, which contain fassaite, were red-sloped and hada shallower 2 lm region absorption feature.

Fig. 13. (a and b) RELAB reflectance spectra of <63 lm sample of Allende heated to different temperatures (indicated on figures). (c and d) Same as (a and b) for continuum-removed 1 lm region. (e and f) Same as (a and b) for continuum-removed 2 lm region. Linear vertical offsets applied to continuum-removed spectra are indicated in bracketson the relevant figures.


5.7. Dark inclusions

Dark inclusions are common at the few% level in a number ofCVs. Buchanan et al. (2010) noted that a type B dark inclusion fromAllende, dominated by fine-grained olivine, had an absorptionband near 1.05 lm, and its spectrum was different from that ofbulk Allende, although no details were provided. Fig. 12a showsreflectance spectra of bulk NWA 3118 and a dark inclusion fromit. The two spectra (both <125 lm) have similar overall reflectance,and the dark inclusion has a redder slope and less well-defined1 lm region absorption band. The continuum-removed 1 lm re-gions (Fig. 12b) have similar band positions and shapes, and banddepth is slightly less in the dark inclusion spectrum. The 2 lm re-gion continuum-removed spectra (Fig. 12c) both exhibit an absorp-tion feature in this region, with the dark inclusion band being

shallower, and having its minimum at a longer wavelength. Theseresults suggest that comminution of a CV largely eliminates anydifferences in overall reflectance that may exist between darkinclusions and a bulk CV, but that the CAI/spinel contents of a bulkCV and dark inclusions do differ (Fruland et al., 1978).

5.8. Thermal metamorphism and melting

Reflectance spectra of Allende (<63 lm grain size) heated tovarious temperatures (Miyamoto et al., 2000) are shown inFig. 13a and b. The sample was ground to <100 lm, heated to var-ious temperatures with an oxygen fugacity of IW � 1, and then re-ground and resieved prior to the spectral measurements. Absolutereflectance in the 0.7 lm region varies non-systematically between�8% and 11% up to �900 �C. Above this temperature, absolute


reflectance increases, particularly above 1000 �C. The spectra alsobecome more ‘‘olivine-like’’ with an increasingly prominent 1 lmregion absorption band. This is seen more clearly in the contin-uum-removed spectra (Fig. 3c and d), particularly above�1000 �C where band depth starts to increase. The 2 lm regionshows similar behavior (Fig. 3e and f): band depth does not beginto increase until �1100 �C, and above this temperature, a spinel-like absorption feature becomes very prominent and deep.

Miyamoto et al. (2000) heated samples of Allende to 800 �C atoxygen fugacities of IW + 2 and IW � 1. The IW + 2 spectrum wasbrighter than the starting sample, with the same overall slope, amore gradual reflectance rise in the visible region, and still dis-played a resolvable olivine absorption band. The IW � 1 spectrumwas also brighter than the starting material but more red-slopedand with a shallower olivine absorption band. These results sug-gest that thermal metamorphism at these temperatures and asso-ciated oxygen fugacity can affect CV spectra. The change in slopeand overall reflectance is not seen in the Allende spectra heatedto 800 �C shown in Fig. 13, suggesting that spectral changes inCV are sensitive to specific experimental conditions.

King et al. (1983) used a solar furnace to melt samples of Al-lende at temperatures of �3000 �C in vacuum (King, 1982). Theresulting condensates ranged from red-sloped and featureless tosimilar to the starting material. The spectral differences wereattributed to variations in the experimental conditions with prefer-ential loss or retention of different elements (King, 1982).

6. Discussion

The development of spectral–compositional correlations for CVsis hampered by a number of factors: the small number of CVs thathave been spectrally characterized and that they are distributedacross different subgroups (CVOxA, CVOxB, CVR); the low level ofagreement concerning petrologic subtype assignments; the lackof detailed mineralogy/petrology for a number of the CVs includedin this study; and the brecciated nature of some CVs. As a result,we can make no more than some tentative observations concern-ing CV spectral properties.

Considering the oxidized versus reduced subgroups, the majordifferences that may impact their reflectance spectra are theslightly higher metal:magnetite ratios in CVR versus CVOx chon-drites, and the fact that CVOx chondrites may contain a few wt.%phyllosilicates. We might expect the former to result in redderslopes for CVOx versus CVR. However, the CVR spectra are notred-sloped compared to the CVOx spectra, probably because metaland magnetite abundances are too low to appreciably affect overallslopes. Similarly, phyllosilicate abundances in the CVR chondritesappear to be too low to result in expected phyllosilicate absorptionbands near 0.7 and 1.1 lm as are seen in CM spectra (Cloutis et al.,2011a). The spectral properties of CV carbonaceous phases are notwell known. Only a carbonaceous extract of Allende has been mea-sured (Johnson and Fanale, 1973): it is dark (<4% reflectance) andblue-sloped.

With increasing metamorphism, we might expect changes infactors such as overall reflectance and mafic silicate absorptionband depth, largely due to the loss and aggregation of carbona-ceous phases. This is clearly seen in the heated Allende spectra.When we examine the CV spectra as a function of metamorphicgrade, even accounting for differences in assignments betweenGuimon et al. (1995) and Bonal et al. (2006), we find that overallreflectance and absorption band depths are not correlated withmetamorphic grade. This is likely attributable to the small samplesize and limited range of petrologic subtypes that are available.

When we examine the absorption feature in the 1 lm region,we find that its overall shape is most consistent with olivine. For

Allende, band centers, regardless of grain size, are confined to the1.05–1.08 lm region, within the range for pure olivine. For someof the other CV spectra, band ‘‘center’’ can range from �1.0 to�1.13 lm. We believe there are two reasons for these wider rang-ing values: many of the CV spectra have shallow and broad absorp-tions in this wavelength region, and many have poorly definedlocal reflectance maxima in the 1.5 lm region. As a result, smallvariations in the slope of the continuum used to isolate these bandscan have a measurable effect on band positions.

Interestingly, a very pervasive feature in the CV spectra is anabsorption feature near 1.3 lm. This is attributable to olivine,and its appearance in so many CV spectra is attributable to the factthat much of the matrix olivine in CVs is Fe-rich, and with increas-ing Fe, this band becomes more prominent relative to the 1.05 lmcenter olivine band (King and Ridley, 1987).

The 2 lm region of CVs is characterized by a nearly ubiquitousabsorption feature which is most consistent with Fe2+-bearing spi-nels (Cloutis et al., 2004). It commonly exhibits band minima/cen-ters near 1.95 and 2.1 lm. Differences in band positions for someCVs is again due to the broad and shallow nature of this feature,making it very sensitive to choice of continuum. Nevertheless, it ap-pears that a spinel-associated absorption feature is characteristic ofCVs regardless of their petrologic subtype, although it should benoted that the low petrologic grade CVs, Leoville and Vigarano haveamong the shallowest bands in this region; we believe that it is dueto their low petrologic subtype rather than the fact that they areCVR chondrites. Band depths in this region are not consistentlyhigher in CVOxA spectra in spite of the fact that alteration of thesemeteorites leads to entry of Fe into CAI spinels (Krot et al., 1995).The heating experiments on Allende show that the 2 lm regionabsorption feature deepens when temperature exceeds �1000 �C.

Band area ratio, which has been found to be a useful criterionfor characterizing olivine + low calcium pyroxene mixtures (Clou-tis et al., 1986) was examined for its utility in characterizing CVchondrites. Band area ratios are zero for pure olivine and the IIA/IA ratio range up to �0.9–1.1 for pure pyroxenes (Cloutis et al.,1986). The II/I area ratio ranges up to �3.5 for pure low calciumpyroxenes (Cloutis and Gaffey, 1991). The IIA/IA and II/I band arearatio values for the powdered CAIs from Allende are 6.4 and 6.8,respectively, well above the range for even pure orthopyroxene.This arises from the fact that their spectra are dominated by spinelwhich has a strong 2 lm region band and no 1 lm region band.However, most of the CV spectra band area ratios fall in the orth-opyroxene to olivine range, suggesting that band area ratios arenot useful for recognizing CAIs/spinels on their own. The presenceof spinels rather than orthopyroxene is best achieved by: (1) usingthe fact that spinel band minima are located near 2.1 lm region,while orthopyroxene band minima occur at shorter wavelengths(1.85–2.1 lm) (Cloutis and Gaffey, 1991); orthopyroxene-influ-enced spectra would exhibit their band I centers below 1.05 lm(Cloutis et al., 1986), whereas spinel-influenced spectra have no ef-fect on band I position; (3) the spinel 2 lm region absorption fea-ture is broader than the orthopyroxene feature, and hence does notshow a large upturn in reflectance longward of �2.1 lm.

Grain size variations have measurable effects on CV spectra.Johnson and Fanale (1973) first demonstrated that CV and otherCCs become darker and bluer with increasing average grain size,and our measurements on slabs versus powders for Allende con-firm this, but this trend is not seen in all CV spectra. We have alsofound that increasing average grain size and removal of the finestfraction both lead to increasingly blue overall slopes. We have alsofound that straight line continuum removal in the 1 lm region cansuccessfully and repeatedly recover band centers in the 1 and 2 lmregions.

The heating experiments on Allende suggest that CVs that arethermally metamorphosed will display more prominent olivine


and spinel features. This can be attributed to the loss/aggregationof carbonaceous phases, and possibly devitrification/crystallization(as olivine) of any amorphous or poorly crystalline phases that maybe present; such material has been found in ALH 84028 (Schwarzand MacPherson, 1985). The heating experiments on Allende sug-gest that measurable spectral changes of this sort require temper-atures on the order of 1000 �C. When CVs are exposed to meltingtemperatures (King et al., 1983), the resulting condensates can bevery spectrally variable, resembling the starting material, or beingstrongly red-sloped and spectrally featureless.

CV spectra measured in bidirectional mode versus with an inte-grating sphere suggest that integrating sphere measurements canbe similar to, or different from, bidirectional measurements interms of overall reflectance and spectral slope. Some of this varia-tion is likely attributable to differences in the uppermost surfacesof the samples between sample packing. Detailed measurements(Gradie et al., 1980; Gradie and Veverka, 1982) suggest that bidi-rectional measurements on Allende are redder than integratingsphere measurements. These differences are accentuated as phaseangle increases – Allende spectra become increasingly red withincreasing phase angle (Beck et al., 2012).

The CV falls and finds show a possible difference in visible re-gion slopes, with the falls being less red-sloped than the finds. Thisis the same behavior seen in ordinary chondrites (Salisbury andHunt, 1974), and suggests that falls should be preferred when con-ducting spectroscopy-based searches for CV parent bodies, partic-ularly when the visible region is used in such a search.

Fig. 14. (a) RELAB reflectance spectra of Allende (unheated, 1100 �C, and 1200 �C),and some CK3 and CK4 chondrites. (b) Same as (a) for the 2 lm region; somespectra have been offset for clarity.

6.1. CV versus CK chondrites

Possible links between CK and CVOx chondrites have been sug-gested by a number of investigators. Greenwood et al. (2003, 2004)suggested that the mineralogic and petrologic differences that ex-ist between CV and CK chondrites reflect higher metamorphicgrades in CKs and that CKs are a metamorphic extension of CVs.Mineralogic differences that may be consistent with thermal meta-morphism of CVs to form CKs include low chondrule to matrix ra-tio, low abundance of refractory inclusions, low C content, andmagnetite compositional differences (Greenwood et al., 2003,2004). A tentative CK4 chondrite, TNZ 057, was found to containabundant dark inclusions and CAIs similar to those found in CVs(Devouard et al., 2006). NWA 1628, also initially classified as aCK4 chondrite, has matrix olivine composition similar to CV3OxA

chondrites, but its matrix is coarser (Brearley, 2009). A comparisonof CV3OxA chondrites to CK3 and CK4 chondrites but Chaumardet al. (2009) also suggests that various mineralogic and petrologicdifferences that are seen are consistent with these meteorites rep-resenting a continuous metamorphic sequence. These differencesinclude (with increasing metamorphism) higher matrix abun-dance, lower chondrule abundance, and enrichment of Fe in CAIs(Chaumard et al., 2009).

How and whether similarities and differences between thesetwo classes would be expressed spectroscopically is not fully con-strained. CVs differ from CKs in terms of the former having a loweraverage abundance of matrix (49–67% versus 66–75%), higherabundance of chondrules (22–41% versus 18–27%), and higherabundance of CAIs (9.9–11.4% versus 2.1–12.5%) (Chaumardet al., 2009). Olivine compositions are largely similar betweenCVs and CKs (Brearley and Jones, 1998), and given that olivineabsorption bands shift only marginally (by �0.040 lm) from Fa0

to Fa100 (King and Ridley, 1987), we would not expect to find mea-surable differences in 1 lm region band centers between CV andCK spectra. The apparent lower abundance of CAIs in CKs could re-sult in weaker CAI-associated absorption bands in the 2 lm region,however, with increasing metamorphism, fine-grained CAIs

become enriched in Fe, and if this Fe is ferrous and enters spinels,their absorption bands become stronger (Cloutis et al., 2004).

With increasing metamorphism we would expect carbonaceousopaques to aggregate and possibly volatilize. This would result inhigher overall reflectance for CKs and deeper olivine and spinelabsorption bands with increasing metamorphism, which could bea potentially useful spectral parameter to distinguish CVs frommetamorphosed CKs.

A cursory comparison of Allende (CV � 3.7), heated Allende, andsome CK3 and CK4 chondrites (Fig. 14a), shows that unheated Al-lende has the lowest overall reflectance, which increases uponheating above �1000 �C; its reflectance is also lower than the over-all reflectance of the CK3 and CK4 chondrites of comparable grainsize. This strongly suggests that thermal metamorphism does leadto an increase in overall reflectance. Increasing temperature ofheating of Allende also leads to an increase in 2 lm region spinelabsorption band depth. The least metamorphosed CK, MET01149(CK3; McBride et al., 2003), does show an absorption feature in thisregion, of comparable depth to unheated Allende (Fig. 14b). Themore metamorphosed CKs, ALH 85002 (CK4; Score and Lindstrom,1994) and DAV 92300 (CK4; Score and Lindstrom, 1993) have shal-lower 2 lm region absorption features than unheated Allende,heated Allende, and MET 01149. This suggests that while Fe mayenter CAIs as a result of thermal metamorphism, it may not beincorporated into spinels, and/or the lower CAI content in CKsmay counteract any increase in spinel Fe content. A detailed com-parison of CV and CK chondrites will be included in a subsequentpaper in this series.


6.2. CV versus CO chondrites

Compositionally, CO and CV chondrites are quite similar (e.g.,Brearley and Jones, 1998). Petrologic parameters that are used todistinguish members of these two groups, such as chondrule size,texture, and matrix abundance (e.g., Dodd, 1981), would probablynot be useful in analysis of telescopic spectra for possible CO andCV parent bodies, as large asteroid surfaces are generally fine-grained (e.g., Dollfus, 1971; Dollfus et al., 1975; Helfenstein et al.,1994; Veverka et al., 2000), and hence chondrules and other mete-orite components would be comminuted and homogenized. As anexample, one of the criteria that has been adopted for discriminat-ing COs from CVs is abundance of matrix (average of 42 vol.% inCVs versus average of 34 vol.% in COs (Dodd, 1981). Other spectralparameters that could lead to measurable spectral differences,such as olivine Fe2+ content or CAI abundances, are similar enough(Table 1; Brearley and Jones, 1998) to suggest that these twogroups would, on average, be spectrally indistinguishable, particu-larly as both groups include members of varying petrographic sub-grades (e.g., Scott and Jones, 1990; Bonal et al., 2006).

Spectral differences between some COs and CVs with similaroverall reflectance and slopes are shown in Fig. 15a. Superficiallythere is little to distinguish them. When the 1 lm region is exam-ined in detail (Fig. 15b), Y-86751, has a more pronounced 1.3 lmolivine absorption band, but the other CV, ALH 84028, does not.The 2 lm region of these spectra (Fig. 15c) also shows no majorspectral differences that can reliably be applied to distinguish COfrom CV spectra; all four spectra exhibit a spinel-associatedabsorption feature in this region. The small differences in overallslope exhibit overlap between CO and CV spectra. A detailed com-parison of CO and CV spectra will be the subject of a future paper.

Fig. 15. (a) RELAB reflectance spectra of some CO and CV chondrites that havesuperficial similarities. (b and c) Same as (a) for the 1 and 2 lm regions,respectively; some spectra have been offset for clarity.

6.3. CV parent bodies

CV parent bodies among main belt asteroids have been tenta-tively identified using spectroscopic data. Burbine et al. (2001)found that the 0.4–1.6 lm spectrum of the Eos family asteroid599 Luisa was well matched in the UV and presence of an absorp-tion band near 1.1 lm by Mokoia and Allende. However, as dis-cussed above, spectral parameters that are used for matchingasteroid and meteorite spectra may not provide as much specificityas would be desired. For example, overall spectral slopes andreflectance of CVs, and other CCs, can vary within a single meteor-ite, as a function of phase angle, and with changes in grain size.Slopes in the visible region also seem to differ between CV fallsand finds, with falls having less-red slopes. Thus spectral matchingshould be undertaken using falls when possible.

As discussed above, we have found that CV spectra may be lar-gely indistinguishable from CO spectra. The most useful differencesare a generally well-defined 1.3 lm olivine absorption band in CVspectra, generally stronger than the same band in CO spectra, spi-nel absorption bands near 1.95 and 2.1 lm, also generally betterdefined in CV spectra, and a 1 lm region absorption band in the1.05–1.08 lm region in most CV spectra, whereas the same bandis often shifted to >1.1 lm in CO spectra.

Our analysis suggests that the mineralogic and petrologic differ-ences that exist between the CVOx and CVR classes, and the varyingpetrologic subtype included in our sample suite (CV3.0–>3.7), arenot significant enough to result in measurable spectral differencesthat exceed spectral variations within a subgroup, within an indi-vidual meteorite, or as a function of grain size. Similarly, CO andCV chondrites are spectrally quite similar, which is not surprisinggiven their similar mineralogies. Terrestrial weathering also seemsto affect CV spectra, most noticeably in the visible region, resultingin more red-sloped spectra for finds as compared to falls, further

obscuring spectral–compositional trends using this wavelengthregion.

The search for CV parent bodies should focus on the detection ofolivine and spinel absorption bands, specifically absorption fea-tures near 1.05, 1.3, 1.95, and 2.1 lm, as these are the most com-monly seen spectral features of CV chondrites. As mentioned, anastronomer searching for a CV parent body would likely not be ableto differentiate a CO from a CV parent body (as discussed earlier).We would want to match spectra to CV falls where possible, andshould be careful to take into account grain size and viewinggeometry. Particular emphasis should be placed on characterizingdiagnostic absorption features in the 1 and 2 lm region, over otherspectral characteristics such as absolute reflectance, overall spec-tral shape, or slope. Any CV parent body can reasonably be ex-pected to have absorptions at or near 1.05, 1.3, 1.95, and 2.1 lm.


While not pursued here, there are a number of lines of evidence,including paleomagnetic studies of CVs (e.g., Elkins-Tanton et al.,2011; Carporzen et al., 2011; Humayun and Weiss, 2011), and var-ious isotopic data (Clayton and Mayeda, 1996) that link CVs to par-tially differentiated bodies. Tentative links between CVs and theEagle Station pallasite grouplet have been made on the basis ofthese various lines of evidence (e.g., Humayun and Weiss, 2011).Preliminary thermal models suggest that temperatures in the inte-rior of a 50 km diameter CV parent body could have reached1000 K (Wakita et al., 2011).

7. Summary and conclusions

CV chondrites, as with most other CC groups, exhibit spectraldifferences that can be attributed to factors such as grain size, re-dox conditions, and degree and duration of thermal metamor-phism. The most ubiquitous absorption features associated withCV spectra are broad regions of absorption in the 1 and 2 lm re-gion. The 1 lm region feature is usually centered near 1.05–1.08 lm with an additional band near 1.3 lm. These two featuresare attributable to Fe-rich olivine. Absorption in the 2 lm regionusually shows two minima near 1.95 and 2.1 lm that are attribut-able to Fe2+-bearing spinel.

As with other CC groups, CV spectra are diverse, with differ-ences in overall reflectance, slope, and intensity of absorptionbands. This complicates the derivation of spectral parameters thatcan be used to recognize CV subgroups (CVR, CVOxA, CVOxB), petro-logic subtypes, or even to reliably discriminate CVs from other car-bonaceous chondrites. In common with some other CCs, it appearsthat the least metamorphosed members exhibit the most weaklyfeatured reflectance spectra. Also in common with other CCs, fi-nely-dispersed opaque components have a large influence on CCspectra by reducing overall reflectance, reducing mineral absorp-tion band depths, and imparting red or blue slopes to the spectra.

Acknowledgments

We wish to thank the invaluable and generous assistance pro-vided by many individuals which made this study possible. In par-ticular we thank the US and Japanese Antarctic meteorite programsfor recovering the majority of the samples included in this study.The RELAB facility at Brown University is a multi-user facility oper-ated with support from NASA Planetary Geology and GeophysicsGrant NNG06GJ31G, whose support is gratefully acknowledged.The PSF facility at the University of Winnipeg was established withgenerous support from the Canada Foundation for Innovation, theManitoba Research Innovations Fund, and the Canadian SpaceAgency. This study was supported by an NSERC Discovery grantto EAC. The authors also wish to thank Julie Ziffer and an anony-mous reviewer for their many helpful comments which helped im-prove the quality and readability of this paper.

Appendix A. Composition of individual CVs included in thisstudy

A.1. ALHA 81003 (CV3.0–3.1)

ALHA 81003 contains abundant irregular shaped white inclu-sions on a black surface, and the presence of metal was noted(Score and Mason, 1983). It contains numerous chondrules andirregular crystalline aggregates in a minor amount of dark brownto back semi-opaque matrix (Score and Mason, 1983). Tonuiet al. (2002) describe it as consisting of low-Ca-rich, pyroxene-richchondrules and fayalitic olivine-rich matrix (Tonui et al., 2002).Thechondrules and aggregates consist mostly of olivine with a wide

compositional range (Fa0–40, mean Fa8) (Score and Mason, 1983).Trace amounts of Fe–Ni-metal are present as minute grains, and fi-nely dispersed sulfide is present in minor amounts (Score and Ma-son, 1983). The matrix consists largely of fine-grained iron-rich(Fa40–60) olivine (Score and Mason, 1983). The olivine is composi-tionally heterogeneous (Fa0–60), while the pyroxene is uniform incomposition (Fs1) (Score and Mason, 1983). It contains 0.8 wt.% Cand 1.5 wt.% S (Gibson and Andrawes, 1980). The chondrules showevidence of metasomatism, and the high degree of homogeneity ofthe matrix olivines is evidence of thermal metamorphism (Tonuiet al., 2002). It shows mobile trace element evidence of thermalmetamorphism (Wang and Lipschutz, 1998; Lipschutz et al., 1999).

A.2. ALH 84028 (CV3.2 – oxidized A subgroup)

The interior is gray with numerous lighter inclusions and a fewoxidation haloes (Schwarz and MacPherson, 1985). In thin section,a variety of chondrules, clasts, and inclusions are distributed in apristine matrix consisting of (at least) olivine (Fa45–50), troilite,and awaruite (Schwarz and MacPherson, 1985). The matrix con-sists largely of olivine, with disseminated sulfides (Abreu andBrearley, 2003). Matrix olivine is Fa42–58, average Fa52.6, compara-ble to Allende (Abreu and Brearley, 2003). Many chondrules con-tain devitrified glass; olivine in chondrules and larger matrixgrains range from Fa0 to Fa30, with most Fa0–10; pyroxene grainsare mostly close to Fs2Wo1; refractory inclusions are also present(Schwarz and MacPherson, 1985). Chondrule edges show evidenceof secondary fayalitic olivines, and metal and metal sulfides showevidence of extensive alteration, typical of CV3 chondrites (Abreuand Brearley, 2003). Sulfides and oxides are the most commoninclusions in olivine. The olivines in ALH 84028 and Allende mayhave formed by thermal metamorphism of preexisting phyllosili-cates (Abreu and Brearley, 2003). No magnetite was detected byMössbauer analysis, consistent with <0.5 wt.% magnetite (Blandet al., 2000). It contains 0.12 wt.% C (Alexander et al., 2007).

A.3. ALH 85006 (CV3 – oxidized subgroup)

The interior is made up of chondrules and white inclusions(Martinez and Mason, 1986). In thin section, chondrules, chondrulefragments, and irregular clasts are set in a dark brown to black ma-trix (Martinez and Mason, 1986). It contains fine-grained opaquesdispersed in the matrix and rimming some of the chondrules (Mar-tinez and Mason, 1986). The matrix consists largely of fine-grainediron-rich olivine (Fa45–47). Olivine in the chondrules and mineralfragments is near Fa0, and more iron-rich olivine is also present(Martinez and Mason, 1986). Pyroxene is much less abundantand close to Fs0 (Martinez and Mason, 1986). Gooding (1986) notesthat both the olivine (Fa0.3–43) and pyroxene (Fs1–5) are composi-tionally heterogeneous in this meteorite. It contains 8 wt.% magne-tite (Bland et al., 2000) and is a member of the oxidized group(Hoffman et al., 2000).

A.4. Allende (CV3.2–3.3 – oxidized A subgroup)

Allende is the most-studied CV chondrite. It contains 38.4 vol.%matrix and 0.18 wt.% C (Alexander et al., 2007). McSween (1977)gives a composition of: 43 vol.% chondrules, 3.2 vol.% AOIs,9.4 vol.% CAIs, 2.9 vol.% lithic and mineral fragments, 38.4 vol.%matrix, 0.2 vol.% metal, 1.7 vol.% sulfides, and 1.2 vol.% magnetite,by optical point counting. Modal mineralogy for Allende includes<0.3 vol.% (<0.5 wt.%) magnetite (Bland et al., 2000, 2004). Refrac-tory inclusions are present at �5 wt.% (Grossman, 1980). Averageareal abundance of CAIs is 5.02% (Hezel et al., 2008). Submicronchromite is common along healed cracks and grain boundaries inchondrule olivine (Housley, 1981). Fayalitic-rich rims, veins and


halos appear to be common in forsteritic olivine in chondrules,CAIs, and single grains (Peck and Wood, 1987; Hua et al., 1988).Matrix olivine compositions are centered around Fa47 (Peck,1983). Mössbauer analysis indicates a prevalence of Fe2+-bearingolivine, a small amount of either goethite or phyllosilicates, andno evidence of ferromagnetic minerals, such as iron oxides, magne-tite, metallic Fe and troilite (Oliver et al., 1984). Mössbauer spec-troscopy suggests that olivine and phyllosilicates (Fe3+-bearing)are present; Fe3+/total Fe = �10% (Fisher and Burns, 1991). Naturalremnant magnetization suggests the presence of pyrrhotite, tae-nite, and magnetite (Nagata et al., 1991). Chondrule olivine compo-sition is narrow (Fa1–9) (Simon and Haggerty, 1980). Olivine andpyroxene compositions average Fa10 and Fs4.1, respectively (McS-ween, 1977). Matrix phyllosilicate Fe/(Fe + Mg) ranges from 45 to55 (Zolensky et al., 1993). Matrix and chondrule rims average Fe/(Fe + Mg) of 0.50 and 0.50, and wt.% S of 0.45 and 0.41, respectively(Zolensky et al., 1993).

Dark inclusions are common in Allende than in other CV3 chon-drites (Kojima and Tomeoka, 1996). Dark inclusions range fromones composed of chondrules and CAIs embedded in a fine-grainedmatrix resembling host CV3 meteorites, to fine-grained homoge-neous Fe-rich olivine (Fa25–33) (Kojima and Tomeoka, 1996). Darkinclusions may have suffered aqueous alteration followed by ther-mal metamorphism (Kojima and Tomeoka, 1996).

Bulk C content is 0.27 wt.% (Pearson et al., 2006), 0.29 wt.%(Jarosewich, 1990), or 0.37 wt.% (Halbout et al., 1986).Carbona-ceous matter is concentrated on olivine surfaces, particularly inthe dark halos surrounding some chondrules and aggregates(Bunch and Chang, 1980). Allende’s magnetic properties have ledto the suggestion that it may have come from a partially differen-tiated CV parent body, possibly linked to the Eagle Station pallasite,which would represent a sample of the core-mantle boundary(Weiss et al., 2010).

A.5. Grosnaja (CV3.2–3.3 – oxidized B subgroup)

It contains 36 vol.% chondrules, 3.3 vol.% AOIs, 2.8 vol.% CAIs,2.3 vol.% lithic and mineral fragments, 51.3 vol.% matrix, a traceamount of metal, 2.9 vol.% sulfides, and 1.4 vol.% magnetite, byoptical point counting (McSween, 1977). Olivine and pyroxenecompositions average Fa11 and Fs7.4, respectively (McSween,1977). Matrix olivine compositions are fairly homogenous andcluster around Fa50 (Peck, 1983; Keller and McKay, 1993). Chon-drule olivines cluster around Fa11 (Van Schmus, 1969). Microprobetraverses show that the bulk of the olivines are Fa<20, and the rarerpyroxenes are generally Fs<5 (Wood, 1967a). Mason (1963) foundthat olivine composition ranges from Fa0 to Fa65. It contains2.6 wt.% magnetite (Bland et al., 2000), and no metal (Wood,1967b). It has undergone aqueous alteration, resulting in the for-mation of phyllosilicates in the matrix and chondrules. Serpentineis the most abundant matrix phyllosilicate, mixed with a chloritegroup mineral and rare smectite (Keller and McKay, 1993). Thephyllosilicates occur with Mg-rich carbonates, fine-grained mag-netite, chromite and pentlandite and poorly crystalline FeNi-oxide/hydroxide. The serpentines are Fe-rich (Fe/(Fe + Mg) = �0.5)(Keller and McKay, 1993). Chondrule mesostasis is altered to Na-saponite (Fe/(Fe + Mg) = 0.1) (Keller and McKay, 1993). Bulk C con-tent is 0.71 wt.% (Pearson et al., 2006). The matrix contains 0.6 wt.%C, and 26% Fe (Wood, 1967b).

A.6. Leoville (CV3.0 – reduced subgroup)

McSween (1977) describes Leoville as containing 51.4 vol.%chondrules, 1.3 vol.% AOIs, 6.6 vol.% CAIs, 1.1 vol.% lithic and min-eral fragments, 35.1 vol.% matrix, 1.8 vol.% metal, 2.1 vol.% sulfides,and 0.6 vol.% magnetite, by optical point counting. May et al.

(1999) found 1.89% CAIs by area. Leoville contains 35.1 vol.% ma-trix (Alexander et al., 2007). The matrix is dominated by fine-grained (typically <1 lm) fayalitic (Fa50–60) olivine; low-Ca pyrox-ene is abundant; metal is present and phyllosilicates are absent(Krot et al., 1998). Olivine and pyroxene compositions averageFa6.4 and Fs2.8, respectively (McSween, 1977). It is less metamor-phosed than Allende; olivine composition is Fa0–70, with mostFa<10, especially in the chondrules; pyroxene is almost uniformlyFs<6 (Kracher et al., 1985). It shows a preferred textural orientation,possibly due to shock compression (Muller and Wlotzka, 1982).Numerous dark inclusions include xenolithic CM-type materials,while others are similar to the host matrix (Kracher et al., 1985).All components of Leoville have undergone hydrous alteration tovarying degrees (Kracher et al., 1985). Mössbauer analysis indi-cates the presence of hydrated and anhydrous silicates, magnetite,troilite, and Fe–Ni metal (Oliver, 1978). It contains 3.8 wt.% magne-tite according to Bland et al. (2000), although Krot et al. (1998)found no magnetite. Bulk C content is given as 1.14 wt.% (Pearsonet al., 2006) or 0.61 wt.% (Alexander et al., 2007).

A.7. Mokoia (CV3.0–3.2 – oxidized B subgroup)

It contains 46.9 vol.% chondrules, 2.7 vol.% AOIs, 3.5 vol.% CAIs,3.6 vol.% lithic and mineral fragments, 39.8 vol.% matrix, a traceamount of metal, 1.6 vol.% sulfides, and 1.9 vol.% magnetite, byoptical point counting (McSween, 1977). It contains 39.8 vol.% ma-trix according to Alexander et al. (2007). It appears to consist of dif-ferent oxidized and reduced CV3 lithologies (Krot et al., 1998). Ofthe objects >100 lm in diameter in Mokoia, 13 vol.% are varioustypes of inclusions (CAIs: 6.5 vol.%; AOIs: 6.5 vol.%) (Cohen et al.,1983). May et al. (1999) found 0.65% CAIs by area. The matrix isdominated by fayalitic olivine (Fa20–40) (Krot et al., 1998); someolivine has been replaced by a mixture of smectite (saponite) andFe oxide (Keller and Buseck, 1990a).

Olivine compositions range from almost pure forsterite to al-most pure fayalite (Hua and Buseck, 1995). The bulk of the olivinein chondrules is Fa<5 (Cohen et al., 1983). Chondrule olivines clus-ter around Fa12 (Van Schmus, 1969). Microprobe traverses showthat the bulk of the olivines are Fa<10, and the rarer pyroxenesare generally Fs<4 (Wood, 1967a). Olivine composition ranges fromFa0 to Fa65 (Mason, 1963). The presence of Fa0 olivine was identi-fied by X-ray diffraction analysis (Bass, 1971). Olivine and pyrox-ene compositions average Fa11.7 and Fs4.1, respectively, accordingto McSween (1977).

It contains sparse phyllosilicates that are largely Fe-bearing sap-onite in the matrix, and Fe-bearing saponite (2–4 wt. %FeO) andNa-rich phlogopite in the chondrules, aggregates, and inclusions(Tomeoka and Buseck, 1990). The saponite resulted from hydrationof Fe-rich olivines, with formation of magnetite, and the phlogopitefrom alteration of other phases (Tomeoka and Buseck, 1990). Bothhigh- and low-aluminum phyllosilicates were identified (Cohenand Kornacki, 1983). Kerridge (1964) identified the presence of aphyllosilicate with a serpentine layer spacing. Mössbauer analysisshows that approximately 7% of the total Fe (24.1 wt.%) in thismeteorite is present in phyllosilicates (Herr and Skerra, 1968).

Dark inclusions that are ‘‘abundant’’ in Mokoia (Ohnishi andTomeoka, 2002) are compositionally similar to the host meteoritebut are finer-grained and contain less phyllosilicates, presumablydue to dehydration by shock-induced thermal metamorphism. Itis assigned to subtype CV3.0–3.2 because it appears to be a breccia(Krot et al., 1998, 2000), containing a range of lithologies (Guimonet al., 1995). It contains 3.2 wt.% magnetite by Mössbauer analysis(Bland et al., 2000) or 4.1 wt.% by thermogravimetry (Watson et al.,1975). It contains 0.54 wt.% C (Alexander et al., 2007), or 0.8 wt.% C,and 21.5% Fe in its matrix (Wood, 1967b).


A.8. NWA 3118 (CV3)

This CV3 chondrite shows, in thin section, curvilinear bands ofolivine grains (Fa32–36) enclosing masses of fine-grained Fe-richolivine (Fa42–58), pyroxenes, and sulfides (Russell et al., 2005).Many types of CAIs are present; no metal was observed in the ma-trix although some chondrules contain fresh metal; at least onedark inclusion is also present (Russell et al., 2005).

A.9. QUE 93744 (CV3)

The interior of this meteorite is dark-brown to black fine-grained, with abundant oxidation scattered throughout, and withsome small light-gray inclusions (Satterwhite and Mason, 1995).A thin section shows numerous chondrules and irregular aggre-gates in a black matrix which contains accessory amounts ofFe–Ni-metal and troilite (Satterwhite and Mason, 1995). Mostchondrule olivine is close to Fa0, but ranges up to Fa10; pyroxenecomposition is Fs1–2 (Satterwhite and Mason, 1995). The matrixconsists largely of Fe-rich olivine, about Fa50 (Satterwhite andMason, 1995).

A.10. Vigarano (CV3.2–3.3 – reduced subgroup)

It contains 48.9 vol.% chondrules, 5.4 vol.% AOIs, 5.3 vol.% CAIs,2.1 vol.% lithic and mineral fragments, 34.5 vol.% matrix, 1.1 vol.%metal, 1.6 vol.% sulfides, and 0.3 vol.% magnetite, by optical pointcounting (McSween, 1977). It contains 34.5 vol.% matrix accordingto Alexander et al. (2007). May et al. (1999) found 1.7% CAIs byarea. Many chondrules and chondrule fragments retain fine-grained rims which are composed largely of FeO-rich olivine(Abreu and Brearley, 2002). Many olivine grains contain a highdensity of nanometer-size inclusions, most of which are amor-phous, and some of which contain amorphous carbon or poorlygraphitized carbon (Abreu and Brearley, 2002). Matrix and chon-drule rims average Fe/(Fe + Mg) of 0.59 and 0.65, and wt.% S of0.16 and 0.30, respectively (Zolensky et al., 1993). It contains28.8% Fe in its matrix (Wood, 1967b). Olivine and pyroxene com-positions average Fa5.6 and Fs3.6, respectively (McSween, 1977).Matrix olivine compositions are fairly homogenous and clusteraround Fa52 (Peck, 1983). Chondrule olivines cluster around Fa6

(Van Schmus, 1969). Olivine composition ranges from Fa0 to Fa60

(Mason, 1963). Matrix phyllosilicate Fe/(Fe + Mg) ranges from 50to 70 (Zolensky et al., 1993). It contains 2.5 vol.% (4.3 wt.%) magne-tite (Bland et al., 2000, 2002). C content is variously given as0.61 wt.% (Alexander et al., 2007), 1.50 wt.% (Pearson et al.,2006), or 1.12 wt.% (Wood, 1967b). This meteorite is a breccia con-taining both oxidized and reduced CV components (Krot et al.,1998).

A.11. Y-86751 (CV3 – oxidized subgroup)

Y-86751 has been described as transitional between the CVOxA

and CVOxB subgroups on the basis of AOI mineralogy (Komatsuet al., 2007). It contains CAIs, dark inclusions and AOIs (Komatsuet al., 2007). The matrix is reddish-brown, suggestive of aqueousalteration, and chondrule boundaries show evidence of aqueousalteration (Gyollai et al., 2011). It has the same bulk compositionas Allende (Palme et al., 1993). It contains nanodiamonds, and car-bonaceous globules, and shows structural changes in its low crys-tallinity carbonaceous phases, that are indicative of low levels ofthermal metamorphism (Aoki and Akai, 2008). It shows evidenceof being exposed to both reducing and oxidizing conditions(Murakami and Ikeda, 1994). Spinels and chondrule olivines areboth more Fe-rich than Allende; CAI spinels contain 18–25% FeO(Murakami and Ikeda, 1994). It contains �60 vol.% matrix that is

dominated by ferroan olivine; fine-grained Al-spinel is also com-mon in the matrix (Murakami and Ikeda, 1994).

Appendix B. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.icarus.2012.07.007.

References

Abreu, N.M., Brearley, A.J., 2002. Mineralogical characterization of Vigarano matrix:HRTEM and EFTEM observations. Lunar Planet. Sci. 33. Abstract #1542.

Abreu, N.M., Brearley, A.J., 2003. HRTEM and EFTEM observations of matrix in theoxidized CV3 chondrite ALH 84028: Implications for the origins of matrixolivines. Lunar Planet. Sci. 34. Abstract #1397.

Adams, J.B., 1974. Visible and near-infrared diffuse reflectance spectra of pyroxenesas applied to remote sensing of solid objects in the Solar System. J. Geophys.Res. 79, 4829–4836.

Alexander, C.M.O’D., Fogel, M., Yabuta, H., Cody, G.D., 2007. The origin and evolutionof chondrites recorded in the elemental and isotopic compositions of theirmacromolecular organic matter. Geochim. Cosmochim. Acta 71, 4380–4403.

Aoki, T., Akai, J., 2008. Carbon materials in Antarctic and non Antarctic carbonaceouschondrites: High-resolution transmission electron microscopy. J. Miner. Petrol.Sci. 103, 173–182.

Aponte, J.C., Alexandre, M.R., Wang, Y., Brearley, A.J., Alexander, C.M.O’D., Huang, Y.,2011. Effects of secondary alteration of the composition of free and IOM-derived monocarboxylic acids in carbonaceous chondrites. Geochim.Cosmochim. Acta 75, 2309–2323.

Bass, M.N., 1971. Montmorillonite and serpentine in Orgueil meteorite. Geochim.Cosmochim. Acta 35, 139–147.

Beck, P., Pommerol, A., Thomas, N., Schmitt, B., Moynier, F., Barrat, J.-A., 2012.Photometry of meteorites. Icarus 218, 364–377.

Bischoff, A., Scott, E.R.D., Metzler, K., Goodrich, C.A., 2006. Nature and origin ofmeteoritic breccias. In: Lauretta, D.S., McSween, H.Y., Jr. (Eds.), Meteorites andthe Early Solar System II. Univ. Arizona Press, Tucson, AZ, pp. 679–712.

Bland, P.A., Hoffman, E., Krot, A.N., Berry, J.F., 2000. CV3 chondrites: How oxidized orreduced? Meteorit. Planet. Sci. 35, A28 (abstract).

Bland, P.A. et al., 2002. 57Fe Mössbauer spectroscopy studies of meteorites:Implications for weathering rates, meteorite flux, and early Solar Systemprocesses. Hyperfine Interact. 142, 481–494.

Bland, P.A., Cressey, G., Menzies, O.N., 2004. Modal mineralogy of carbonaceouschondrites by X-ray diffraction and Mössbauer spectroscopy. Meteorit. Planet.Sci. 39, 3–16.

Bland, P.A., Berry, F.J., Cadogan, J.M., Howard, K.T., Cressey, G., 2008. Fe-bearingmineral abundance in primitive chondrites by Mössbauer spectroscopy.Meteorit. Planet. Sci. 43. Abstract #5304.

Bonal, L., Quirico, E., Bourot-Denise, M., Montagnac, G., 2006. Determination of thepetrologic type of CV3 chondrites by Raman spectroscopy of included organicmatter. Geochim. Cosmochim. Acta 70, 1849–1863.

Brearley, A.J., 2006. The action of water. In: Lauretta, D.S., McSween, H.Y., Jr. (Eds.),Meteorites and the Early Solar System II. Univ. Arizona Press, Tucson, AZ, pp.587–624.

Brearley, A.J., 2009. Matrix olivines in the metamorphosed CK chondrite NWA 1628:Possible affinities to olivines in the matrices of oxidized CV3 chondrites anddark inclusions. Lunar Planet. Sci. 40. Abstract #1791.

Brearley, A.J., Jones, R.H., 1998. Chondritic meteorites. In: Papike, J.J. (Ed.), PlanetaryMaterials, Reviews in Mineralogy, vol. 36. Mineralogical Society of America,Washington, DC (Chapter 3).

Buchanan, P.C., Zolensky, M.E., Hiroi, T., Le, L., Galindo, Jr., C., 2010. Reflectancespectra comparisons of an Allende dark inclusion and asteroids. Meteorit.Planet. Sci. 46. Abstract #5316.

Bunch, T.E., Chang, S., 1980. Carbonaceous chondrites – II. Carbonaceous chondritephyllosilicates and light element geochemistry as indicators of parent bodyprocesses and surface conditions. Geochim. Cosmochim. Acta 44, 1543–1578.

Burbine, T.H., Gaffey, M.J., Bell, J.F., 1992. S-asteroids 387 Aquitania and 980Anacostia: Possible fragments of the breakup of a spinel-bearing parent bodywith CO3/CV3 affinities. Meteoritics 27, 424–434.

Burbine, T.H., Binzel, R.P., Bus, S.J., Clark, B.E., 2001. K asteroids and CO3/CV3chondrites. Meteorit. Planet. Sci. 36, 245–253.

Buseck, P.R., Hua, X., 1993. Matrices of carbonaceous chondrite meteorites. Ann.Rev. Earth Planet. Sci. 21, 255–305.

Carporzen, L., Weiss, B.P., Elkins-Tanton, L.T., Shuster, D.L., Ebel, D., Gattacceca, J.,2011. Magnetic evidence for a partially differentiated carbonaceous chondriteparent body. Proc. Nat. Acad. Sci. 108, 6386–6389.

Chaumard, N., Devouard, B., Zanda, B., Ferrière, L., 2009. The link between CV and CKcarbonaceous chondrites based on parent body processes. Meteorit. Planet. Sci.45. Abstract #5206.

Chen, J.H., Wasserburg, G.J., 1981. The isotopic composition of uranium and lead inAllende inclusions and meteoritic phosphates. Earth Planet. Sci. Lett. 52, 1–5.

Clark, R.N., Roush, T.L., 1984. Reflectance spectroscopy: Quantitative analysistechniques for remote sensing applications. J. Geophys. Res. 89, 6329–6340.




Clayton, R.N., Mayeda, T.K., 1996. Oxygen isotope studies of achondrites. Geochim.Cosmochim. Acta 60, 1999–2017.

Cloutis, E.A., 2003. Quantitative characterization of coal properties usingbidirectional diffuse reflectance spectroscopy. Fuel 82, 2239–2254.

Cloutis, E.A., Gaffey, M.J., 1991. Pyroxene spectroscopy revisited: Spectral–compositional correlations and relationship to geothermometry. J. Geophys.Res. 96, 22,809–22,826 (Chapter 3).

Cloutis, E.A., Gaffey, M.J., 1993. The constituent minerals in calcium–aluminuminclusions: Spectral reflectance properties and implications for COcarbonaceous chondrites and asteroids. Icarus 105, 568–579.

Cloutis, E.A., Gaffey, M.J., Jackowski, T.L., Reed, K.L., 1986. Calibrations of phaseabundance, composition, and particle size distribution for olivine–orthopyroxene mixtures from reflectance spectra. J. Geophys. Res. 91, 11,641–11,653.

Cloutis, E.A., Sunshine, J.M., Morris, R.V., 2004. Spectral reflectance-compositionalproperties of spinels and chromites: Implications for planetary remote sensingand geothermometry. Meteorit. Planet. Sci. 39, 545–565.

Cloutis, E.A., Hardersen, P.S., Bish, D.L., Bailey, D.T., Gaffey, M.J., Craig, M.A., 2010.Reflectance spectra of iron meteorites: Implications for spectral identification oftheir parent bodies. Meteorit. Planet. Sci. 45, 304–332.

Cloutis, E.A., Hiroi, T., Gaffey, M.J., Alexander, C.M.O’D., Mann, P., 2011a. Spectralreflectance properties of carbonaceous chondrites: 1. CI chondrites. Icarus 212,180–209.

Cloutis, E.A., Hudon, P., Hiroi, T., Gaffey, M.J., Mann, P., 2011b. Spectral reflectanceproperties of carbonaceous chondrites: 2. CM chondrites. Icarus 216, 309–346.

Cloutis, E.A., Hudon, P., Hiroi, T., Gaffey, M.J., 2012a. Spectral reflectance propertiesof carbonaceous chondrites: 3. CR chondrites. Icarus 217, 389–407.

Cloutis, E.A., Hudon, P., Hiroi, T., Gaffey, M.J., 2012b. Spectral reflectance propertiesof carbonaceous chondrites: 4. Aqueously altered and thermallymetamorphosed meteorites. Icarus 220, 586–617.

Cloutis, E.A., Hudon, P., Hiroi, T., Gaffey, M.J., Mann, P., 2012c. Spectral reflectanceproperties of carbonaceous chondrites: 5. CO chondrites. Icarus 220, 466–486.

Cohen, R.E., Kornacki, A.S., 1983. Phyllosilicates in refractory inclusions in theMokoia C3(V) chondrite. Lunar Planet. Sci. 12, 128–129 (abstract).

Cohen, R.E., Kornacki, A.S., Wood, J.A., 1983. Mineralogy and petrology of chondrulesand inclusions in the Mokoia CV3 chondrites. Geochim. Cosmochim. Acta 47,1739–1757.

Devouard, B., Ferrierre, L., Zanda-Hewins, B., Messaoudi, M., 2006. Mineralogy andpetrology of TNZ 057 (C4) and comparison to the CV and CK groups. Meteorit.Planet. Sci. 41. Abstract #9025.

Dodd, R.T., 1981. Meteorites: A Petrologic-Chemical Synthesis. Cambridge Univ.Press, Cambridge, UK, 368pp.

Dollfus, A., 1971. Physical studies of asteroids by polarization of the light. In:Gehrels, T. (Ed.), Physical Studies of Minor Planets, BASA SP-267, pp. 95–116.

Dollfus, A., Geake, J.E., Mandeville, J.C., Zellner, B., 1975. The nature of asteroidsurfaces from optical polarimetry. In: Delsemme, A.M. (Ed.), Comets, Asteroids,Meteorites. Univ. of Toledo Press, Toledo, Ohio, pp. 243–251.

Elkins-Tanton, L.T., Weiss, B.P., Zuber, M.T., 2011. Chondrites as samples ofdifferentiated planetesimals. Earth Planet. Sci. Lett. 305, 1–10.

Emmerton, S., Muxworthy, A.R., Hezel, D.C., Bland, P.A., 2011. Magneticcharacteristics of CV chondrules with paleointensity implications. J. Geophys.Res. 116, E12007.

Fisher, D.S., Burns, R.G., 1991. Pre-terrestrial oxidation products of iron minerals incarbonaceous meteorites identified in Mössbauer spectra. Lunar Planet. Sci. 22,389–390 (abstract).

French, L.M., Veverka, J., 1983. Limb darkening on meteorites and asteroids. Icarus54, 38–47.

Fries, M., Bhartia, R., 2010. UV fluorescence/Raman imaging of Allende. Meteorit.Planet. Sci. 45. Abstract #5441.

Fries, M., Steele, A., McCoy, T., 2005a. Carbon and mineral phase distribution inintergrain spaces and inclusions within chondrules. Meteorit. Planet. Sci. 40.Abstract #5302.

Fries, M., Steele, A., Ruzicka, A., 2005b. Carbon and mineral phase distribution on aCV3 dark inclusion boundary – A confocal Raman imaging study. Meteorit.Planet. Sci. 40. Abstract #5236.

Fries, M., Bhartia, R., Steele, A., 2011. Carbonaceous chondrite groups discernedusing Raman spectral parameters. Lunar Planet. Sci. 42. Abstract #1860.

Fruland, R.M., King, E.A., McKay, D.S., 1978. Allende dark inclusions. Proc. LunarPlanet. Sci. 9, 1305–1329.

Gaffey, M.J., 1974. A Systematic Study of the Spectral Reflectivity Characteristics ofthe Meteorite Classes with Applications to the Interpretation of AsteroidSpectra for Mineralogical and Petrological Information. PhD Dissertation.Massachusetts Institute of Technology, Cambridge, Massachusetts, 355pp.

Gaffey, M.J., 1986. The spectral and physical properties of metal in meteoriteassemblages: Implications for asteroid surface materials. Icarus 66, 468–486.

Gibson Jr., E.K., Andrawes, F.F., 1980. The Antarctic environment and its effect uponthe total carbon and sulfur abundances in recovered meteorites. Proc. LunarPlanet. Sci. 11, 1223–1234.

Gooding, J.L. (Ed.), 1986. Index of Classified Meteorites from the 1976–1985Antarctic Collections (as of September, 1986). Antarct. Meteorite Newslett. 9(4), 1–45.

Gradie, J., Veverka, J., 1982. When are spectral reflectance curves comparable?Icarus 49, 109–119.

Gradie, J., Veverka, J., 1986. The wavelength dependence of phase coefficients. Icarus66, 455–467.

Gradie, J., Veverka, J., Buratti, B., 1980. The effects of scattering geometry on thespectrophotometric properties of powdered materials. Proc. Lunar Planet. Sci.11, 799–815.

Greenwood, R.C., Kearsley, A.T., Franchi, I.A., 2003. Are CK chondrites really adistinct group or just equilibrated CVs? Meteorit. Planet. Sci. 38, A96 (abstract).

Greenwood, R.C., Franchi, I.A., Kearsley, A.T., Alard, O., 2004. The relationshipbetween CK and CV chondrites: A single parent body source? Lunar Planet. Sci.35. Abstract #1664.

Grisolle, F., Beck, P., Quirico, E., Schmitt, B., Moynier, F., 2011. Reflectance spectra ofcarbonaceous chondrites under controlled atmosphere. Meteorit. Planet. Sci. 46.Abstract #5337.

Grossman, L., 1975. Petrography and mineral chemistry of Ca-rich inclusions in theAllende meteorite. Geochim. Cosmochim. Acta 39, 433–454.

Grossman, L., 1980. Refractory inclusions in the Allende meteorite. Ann. Rev. EarthPlanet. Sci. 8, 559–608.

Grossman, L., Steele, I.M., 1976. Amoeboid olivine aggregates in the Allendemeteorite. Geochim. Cosmochim. Acta 40, 149–155.

Guimon, R.K., Symes, S.J.K., Sears, D.W.J., Benoit, P.H., 1995. Chemical and physicalstudies of type 3 chondrites XII: The metamorphic history of CV chondrites andtheir components. Meteoritics 30, 704–714.

Gyollai, I., Nagy, Sz., Bérczi, Sz., Gucsik, A., 2011. Comparison of aqueous alterationof two CV3 (Kaba and Yamato-86751) chondrites. Lunar Planet. Sci. 42. Abstract#1039.

Halbout, J., Mayeda, T.K., Clayton, R.N., 1986. Carbon isotopes and light elementabundances in carbonaceous chondrites. Earth Planet. Sci. Lett. 80, 1–18.

Helfenstein, P., Veverka, J., Thomas, P.C., Simonelli, D.P., Lee, P., Klaasen, K., Johnson,T.V., Breneman, H., Head, J.W., Murchie, S., Fanale, F., Robinson, M., Clark, B.,Granahan, J., Garbiel, H., McEwen, A.S., Kirk, R.L., Davies, M., Neukum, G.,Mottola, S., Wagner, R., Belton, M., Chapman, C., Pilcher, C., 1994. Galileophotometry of asteroid 951 Gaspra. Icarus 107, 37–60.

Herr, W., Skerra, B., 1968. Mössbauer spectroscopy applied to the classification ofstony meteorites. In: Millman, P.M. (Ed.), Meteorite Research. Reidel, Dordrecht,pp. 106–122.

Hezel, D.C., Russell, S.S., Ross, A.J., Kearsley, A.T., 2008. Modal abundances of CAIs:Implications for bulk chondrite element abundances and fractionations.Meteorit. Planet. Sci. 43, 1879–1894.

Hiroi, T., Pieters, C.M., Zolensky, M.E., Lipschutz, M.E., 1993. Evidence of thermalmetamorphism on the C, G, B, and F asteroids. Science 261, 1016–1018.

Hoffman, E., Bland, P.A., Seifu, D., Oliver, F.W., 2000. Ferric iron phases in Mössbauerspectra of ‘‘oxidized’’ and ‘‘reduced’’ CV3 chondrites. Meteorit. Planet. Sci. 35.Abstract #5027.

Housley, R.M., 1981. SEM, optical, and Mössbauer studies of submicrometerchromite in Allende. Proc. Lunar Planet. Sci. 12B, 1069–1077.

Howard, K.T., Benedix, G.K., Bland, P.A., Cressey, G., 2009. Modal mineralogy of CV3chondrites by PSD-XRD: Mineralogic insights into a complex evolutionaryhistory. Lunar Planet. Sci. 40. Abstract #1235.

Howard, K.T., Benedix, G.K., Bland, P.A., Cressey, G., 2010. Modal mineralogy of CV3chondrites by X-ray diffraction (PSD-XRD). Geochim. Cosmochim. Acta 74,5084–5097.

Hua, X., Buseck, P.R., 1995. Fayalite in the Kaba and Mokoia carbonaceouschondrites. Geochim. Cosmochim. Acta 59, 563–578.

Hua, X., Adam, J., Palme, H., El Goresy, A., 1988. Fayalite-rich rims, veins, and halosaround and in forsteritic olivines in CAIs and chondrules in carbonaceouschondrites: Types, compositional profiles and constraints on their formation.Geochim. Cosmochim. Acta 52, 1389–1408.

Humayun, M., Weiss, B.P., 2011. A common parent body for Eagle Station pallasitesand CV chondrites. Lunar Planet. Sci. 42. Abstract #1507.

Hurt, S.M., Rubin, A.E., Wasson, J.T., 2012. Fractionated matrix composition in CV3Vigarano and alteration processes on the CV parent body. Meteorit. Planet. Sci.47, 1035–1048.

Huss, G.R., Rubin, A.E., Grossman, J.N., 2006. Thermal metamorphism in chondrites.In: Lauretta, D.S., McSween, H.Y., Jr. (Eds.), Meteorites and the Early SolarSystem II. Univ. Arizona Press, Tucson, AZ, pp. 567–586.

Jarosewich, E., 1990. Chemical analyses of meteorites: A compilation of stony andiron meteorite analyses. Meteoritics 25, 323–337.

Jogo, K., Krot, A.N., 2010. Heavily-metamorphosed chondritic clasts in the CV3carbonaceous chondrites Mokoia and Yamato-86009. Meteorit. Planet. Sci. 45.Abstract #5073.

Jogo, K., Krot, A.N., Nagashima, K., 2011. Heavily-metamorphosed clasts in the CVcarbonaceous chondrite Mokoia: Evidence for strong thermal metamorphismon the CV parent asteroid. Lunar Planet. Sci. 42. Abstract #1613.

Johnson, T.V., Fanale, F.P., 1973. Optical properties of carbonaceous chondrites andtheir relationship to asteroids. J. Geophys. Res. 78, 8507–8518.

Johnson, C.A., Prinz, M., 1991. Chromite and olivine in type II chondrules incarbonaceous and ordinary chondrites: Implications for thermal histories andgroup differences. Geochim. Cosmochim. Acta 55, 893–904.

Johnson, C.A., Prinz, M., Weisberg, M.K., Clayton, R.N., Mayeda, T.K., 1990. Darkinclusions in Allende, Leoville, and Vigarano: Evidence for nebular oxidation ofCV3 constituents. Geochim. Cosmochim. Acta 54, 819–830.

Kamei, A., Nakamura, A.M., 2002. Laboratory study of the bidirectional reflectanceof powdered surfaces: On the asymmetry parameter of asteroid photometricdata. Icarus 156, 551–561.

Kebukawa, Y., Alexander, C.M.O’D., Cody, G.D., 2011. Compositional diversity ininsoluble organic matter in type 1, 2 and 3 chondrites as detected by infraredspectroscopy. Geochim. Cosmochim. Acta 75, 3530–3541.


Keller, L.P., 2011. Mineralogy and petrography of MIL 090001, a highly altered CVchondrite from the reduced subgroup. Lunar Planet. Sci. 42. Abstract #2409.

Keller, L.P., Buseck, P.R., 1990a. Matrix mineralogy of the Lancé CO3 carbonaceouschondrite: A transmission electron microscope study. Geochim. Cosmochim.Acta 54, 1155–1163.

Keller, L.P., Buseck, P.R., 1990b. Aqueous alteration products in CV3 and CO3carbonaceous chondrite meteorites: Phyllosilicates. Lunar Planet. Sci. 21, 619–620 (abstract).

Keller, L.P., McKay, D.S., 1993. Aqueous alteration of the Grosnaja CV3 carbonaceouschondrite. Meteoritics 28, 378 (abstract).

Keller, L.P., Thomas, K.L., Clayton, R.N., Mayeda, T.K., Dehart, J.M., McKay, D.S., 1994.Aqueous alteration of the Bali CV3 chondrite: Evidence for mineralogy, mineralchemistry, and oxygen isotope compositions. Geochim. Cosmochim. Acta 58,5589–5598.

Kerridge, J.F., 1964. Low-temperature minerals from the fine-grained matrix ofsome carbonaceous meteorites. Ann. NY Acad. Sci. 119, 41–53.

King, E.A., 1982. Refractory residues, condensates and chondrules from solarfurnace experiments. Lunar Planet. Sci. 13, 389–390 (abstract).

King, T.V.V., King, E.A., 1981. Accretionary dark rims in unequilibrated chondrites.Icarus 48, 460–472.

King, T.V.V., Ridley, I.W., 1987. Relation of the spectroscopic reflectance of olivine tomineral chemistry and some remote sensing implications. J. Geophys. Res. 92,11,457–11,469.

King, T.V.V., Gaffey, M.J., King, E.A., 1983. Spectral reflectance measurements andsurface characteristics of meteoritic condensates from solar furnaceexperiments. Lunar Planet. Sci. 14, 371–372 (abstract).

Klima, R.L., Pieters, C.M., Dyar, M.D., 2007. Spectroscopy of synthetic Mg–Fepyroxenes I: Spin-allowed and spin-forbidden crystal field bands in the visibleand near-infrared. Meteorit. Planet. Sci. 42, 235–253.

Kojima, T., Tomeoka, K., 1996. Indicators of aqueous alteration and thermalmetamorphism on the CV parent body: Microtextures of a dark inclusionfrom Allende. Geochim. Cosmochim. Acta 60, 2651–2666.

Komatsu, M., Mikouchi, T., Krot, A.N., Miyamoto, M., 2007. Amoeboid olivineaggregates in the CV carbonaceous chondrite Yamato-86751. Lunar Planet. Sci.38. Abstract #1987.

Komatsu, M., Mikouchi, T., Miyamoto, M., 2008. Hydrous and anhydrous alterationof unsubgrouped CV chondrite Y-86751. Meteorit. Planet. Sci. 43. Abstract#5214.

Kornacki, A.S., Wood, J.A., 1984. The mineral chemistry and origin of inclusionmatrix and meteorite matrix in the Allende CV3 chondrite. Geochim.Cosmochim. Acta 48, 1663–1676.

Kracher, A., Keil, K., Kallemeyn, G.W., Wasson, J.T., Clayton, R.N., Huss, G.I., 1985. TheLeoville (CV3) accretionary breccia. Proc. Lunar Planet. Sci. 16, Part 1. J.Geophys. Res. 90 (Suppl.), D123–D135.

Krot, A.N., Scott, E.R.D., Zolensky, M.E., 1995. Mineralogical and chemicalmodification of components in CV3 chondrites: Nebular or asteroidalprocessing? Meteoritics 30, 748–775.

Krot, A.N., Petaev, M.I., Scott, E.R.D., Choi, B.-G., Zolensky, M.E., Keil, K., 1998.Progressive alteration in CV3 chondrites: More evidence for asteroidalalteration. Meteorit. Planet. Sci. 33, 1065–1085.

Krot, A.N., Meibom, A., Keil, K., 2000. Volatile-poor chondrules in CH carbonaceouschondrites: Formation at high ambient nebular temperatures. Lunar Planet. Sci.31. Abstract #1481.

Lipschutz, M.E., Zolensky, M.E., Bell, M.S., 1999. New petrographic and traceelement data on thermally metamorphosed carbonaceous chondrites. Antarct.Meteorite Res. 12, 57–80.

MacPherson, G.J., Grossman, L., 1984. ‘‘Fluffy’’ type A Ca-, Al-rich inclusions in theAllende meteorite. Geochim. Cosmochim. Acta 48, 29–46.

MacPherson, G.J., Hashimoto, A., Grossman, L., 1985. Accretionary rims oninclusions in the Allende meteorite. Geochim. Cosmochim. Acta 49, 2267–2279.

Martinez, R., Mason, B., 1986. ALH 85006. Antarct. Meteorite Newslett. 9 (3), 17.Mason, B., 1963. Olivine composition in chondrites. Geochim. Cosmochim. Acta 27,

1011–1023.May, C., Russell, S.S., Grady, M.M., 1999. Analysis of chondrule and CAI size and

abundance in CO3 and CV3 chondrites: A preliminary study. Lunar Planet. Sci.30. Abstract #1688.

McBride, K., Welzenbach, L., McCoy, T., 2003. MET 01149. Antarct. MeteoriteNewslett. 26 (2), 16.

McFadden, L.A., Gaffey, M.J., Takeda, H., Jackowski, T.L., Reed, K.L., 1982. Reflectancespectroscopy of diogenite meteorite types from Antarctica and theirrelationship to asteroids. Proc. Symp. Antarct. Meteorites 7, 188–206.

McSween Jr., H.Y., 1977. Petrographic variations among carbonaceous chondrites ofthe Vigarano type. Geochim. Cosmochim. Acta 41, 1777–1790.

McSween Jr., H.Y., 1979. Are carbonaceous chondrites primitive or processed? Areview. Rev. Geophys. Space Phys. 17, 1059–1078.

Miyamoto, M., Komatsu, M., Mikouchi, T., 2000. Diffuse reflectance spectra ofseveral chondrites heated at different oxygen fugacities. Lunar Planet. Sci. 31.Abstract #1109.

Muller, W.F., Wlotzka, F., 1982. Mineralogical study of the Leoville meteorite (CV3):Macroscopic texture and transmission electron microscopic observations. LunarPlanet. Sci. 13, 558–559 (abstract).

Murakami, T., Ikeda, Y., 1994. Petrology and mineralogy of the Yamato-86751 CV3chondrite. Meteoritics 29, 397–409.

Nagata, T., Funaki, M., Kojima, H., 1991. Magnetic properties and natural remnantmagnetization of carbonaceous chondrites containing pyrrhotite. Proc. NIPRSymp. Antarct. Meteorites 4, 390–403.

Noguchi, T., Nakamura, T., Nagao, K., 2003. Mineralogy and noble gas signatures ofVigarano meteorite: Evolution of planetary surface material on CV parent body.Meteorit. Planet. Sci. 38. Abstract #5100.

Ohnishi, I., Tomeoka, K., 2002. Dark inclusions in the Mokoia CV3 chondrite:Evidence for aqueous alteration and subsequent thermal and shockmetamorphism. Meteorit. Planet. Sci. 37, 1843–1856.

Oliver, F.W., 1978. Fe57 Mossbauer spectroscopy of the Leoville meteorite. Planet.Space Sci. 26, 289–290.

Oliver, F.W., Isuk, E.E., Wynter, C., 1984. Mossbauer study of the Allende meteorite.Meteoritics 19, 25–27.

Palme, H., Spettel, B., Ikeda, Y., 1993. Origin of chondrules and matrix incarbonaceous chondrites. Meteoritics 28, 417 (abstract).

Pearson, V.K., Sephton, M.A., Franchi, I.A., Gibson, J.M., Gilmour, I., 2006. Carbon andnitrogen in carbonaceous chondrites: Elemental abundances and stable isotopecompositions. Meteorit. Planet. Sci. 41, 1899–1918.

Peck, J.A., 1983. Chemistry of CV3 matrix minerals and Allende chondrule olivine.Meteoritics 18, 373–374 (abstract).

Peck, J.A., Wood, J.A., 1987. The origin of ferrous zoning in Allende chondruleolivines. Geochim. Cosmochim. Acta 51, 1503–1510.

Pieters, C.M., 1983. Strength of mineralogical absorption features in the transmittedcomponent of reflection: First results from RELAB. J. Geophys. Res. 88, 9534–9544.

Rajan, S., Gaffey, M.J., 1984. Spectral reflectance characteristics of Allende whiteinclusions. Lunar Planet. Sci. 15, 659–660 (abstract).

Russell, S.S., Zolensky, M.E., Righter, K., Folco, L., Jones, R., Connolly Jr., H.C., Grady,M.M., Grossman, J.N., 2005. The meteoritical bulletin, no. 89, 2005 September.Meteorit. Planet. Sci. 40, A201–A263.

Salisbury, J.W., Hunt, G.R., 1974. Meteorite spectra and weathering. J. Geophys. Res.79, 4439–4441.

Salisbury, J.W., Hunt, G.R., Lenhoff, C.J., 1975. Visible and near-infrared spectra: X.Stony meteorites. Mod. Geol. 5, 115–126.

Satterwhite, C., Mason, B., 1995. QUE93744. Antarct. Meteorite Newsletter 18 (2),18–19.

Schwarz, C., MacPherson, G., 1985. ALH 84028. Antarct. Meteorite Newslett. 8 (2), 10.Score, R., Lindstrom, M., 1993. Classification and pairing changes. Antarct. Meteorite

Newslett. 16 (2), 18–22.Score, R., Lindstrom, M., 1994. List of newly classified Antarctic meteorites. Antarct.

Meteorite Newslett. 17 (1), 8–14.Score, R., Mason, B., 1983. ALHA 81003. Antarct. Meteorite Newslett. 6 (1), 2.Scott, E.R.D., Jones, R.H., 1990. Disentangling nebular and asteroidal features of CO3

carbonaceous chondrite meteorites. Geochim. Cosmochim. Acta 54, 2485–2502.Sherman, D.M., Burns, R.G., Burns, V.M., 1982. Spectral characteristics of the iron

oxides with application to the Martian bright region mineralogy. J. Geophys.Res. 87, 10169–10180.

Simon, S.B., Haggerty, S., 1979. Petrography and olivine mineral chemistry ofchondrules and inclusions in the Allende meteorite. Lunar Planet. Sci. 10, 1119–1121 (abstract).

Simon, S.B., Haggerty, S.E., 1980. Bulk composition of chondrules in the Allendemeteorite. Proc. Lunar Sci. 11, 901–927.

Sunshine, J.M., Connolly, Jr., H.C., McCoy, T.J., Bus, S.J., La Croix, L., 2008a.Identification of refractory-rich asteroids: Evidence for the earliest accretedbodies in the Solar System. Lunar Planet. Sci. 38. Abstract #1613.

Sunshine, J.M., Connolly Jr., H.C., McCoy, T.J., Bus, S.J., La Croix, L.M., 2008b. Ancientasteroids enriched in refractory inclusions. Science 320, 514–517.

Tomeoka, K., Buseck, P.R., 1990. Phyllosilicates in the Mokoia CV carbonaceouschondrite: Evidence for aqueous alteration in an oxidizing environment.Geochim. Cosmochim. Acta 54, 1745–1754.

Tomeoka, K., Ohnishi, I., 2011. A hydrated clast in the Mokoia CV3 carbonaceouschondrite: Evidence for intensive aqueous alteration in the CV parent body.Geochim. Cosmochim. Acta 75, 6064–6079.

Tonui, E.K., Zolensky, M.E., Hiroi, T., Wang, M.-S., Lipschutz, M.E., 2002.Petrographic, chemical and spectroscopic data on thermally metamorphosedcarbonaceous chondrites. Lunar Planet. Sci. 33. Abstract #1288.

Van Schmus, W.R., 1969. Mineralogy, petrology, and classification of types 3 and 4carbonaceous chondrites. In: Millman, P. (Ed.), Meteorite Research. Springer-Verlag, NY, pp. 480–491.

Veverka, J. et al., 2000. NEAR at Eros: Imaging and spectral results. Science 289,2088–2097.

Wakita, S., Jogo, K., Krot, A.N., 2011. Thermal evolution of CV-like carbonaceouschondrite asteroid. Lunar Planet. Sci. 42. Abstract #1221.

Wang, M.-S., Lipschutz, M.E., 1998. Thermally metamorphosed carbonaceouschondrites from data for thermally mobile trace elements. Meteorit. Planet.Sci. 33, 1297–1302.

Wark, D.A., Lovering, J.F., 1977. Marker events in the early evolution of the SolarSystem: Evidence from rims on Ca–Al-rich inclusions in carbonaceouschondrites. Proc. Lunar Sci. Conf. 8, 95–112.

Watson, D.E., Larson, E.E., Herndon, J.M., Rowe, W.M., 1975. Thermomagneticanalysis of meteorites, 2. C2 chondrites. Earth Planet. Sci. Lett. 27, 101–107.

Weisberg, M.K., Prinz, M., 1997. Fayalitic olivine in CV3 chondrite matrix and darkinclusions: A nebular origin. Workshop on Parent-Body and NebularModification of Chondritic Materials. Abstract #4039.

Weisberg, M.K., Prinz, M., 1998. Fayalitic olivine in CV3 chondrite matrix and darkinclusions: A nebular origin. Meteorit. Planet. Sci. 33, 1087–1099.

Weisberg, M.K., Prinz, M., Clayton, R.N., Mayeda, T.K., 1997. CV3 chondrites:Three subgroups, not two. Annual Meteoritical Society Meeting, 60. Abstract#5156.


Weiss, B.P. et al., 2010. A partially differentiated parent body for CV chondrites?Lunar Planet. Sci. 41. Abstract #1688.

Wood, J.A., 1967a. Olivine and pyroxene compositions in Type II carbonaceouschondrites. Geochim. Cosmochim. Acta 31, 2095–2108.

Wood, J.A., 1967b. Chondrites: Their metallic minerals, thermal histories, andparent planets. Icarus 6, 1–49.

Zolensky, M., Barrett, R., Browning, L., 1993. Mineralogy and composition of matrixand chondrule rims in carbonaceous chondrites. Geochim. Cosmochim. Acta 57,3123–3148.

spectral reflectance properties of carbonaceous chondrites ...psf.uwinnipeg.ca/publications/icarus...

Documents