Meteoritics & Planetary Science 39, Nr 8, 1307–1319 (2004)
Abstract available online at http://meteoritics.org
Formation of accretionary dust mantles in the solar nebula: Evidence from
preirradiated olivines in CM chondrites
Knut METZLER
Institut für Planetologie, Universität Münster, Wilhelm-Klemm Strasse 10, 48149 Münster, Germany
E-mail: [email protected]
(Received 28 October 2002; revision accepted 12 March 2004)
Abstract–CM chondrites are regolith breccias consisting of lithic clasts embedded in a fine-grained
clastic matrix. The majority of these lithic clasts belongs to a texturally well-defined rock type
(primary rock) that can be described as an agglomerate of chondrules and other coarse-grained
components, most of which are surrounded by fine-grained rims (dust mantles). Metzler et al. (1992)
explain these textures as the result of accretionary processes in the solar nebula, while an alternative
model explains them to be the result of regolith processes on the parent body (Sears et al. 1993).
The main intention of the present study is to discern between both models by investigating the
occurrence, frequency, spatial distribution, and textural setting of preirradiated (track-rich) olivines
in CM chondrites. Track-rich olivines were studied in situ in six polished thin sections from 4
different CM chondrites (Cold Bokkeveld, Mighei, Murchison, Nogoya) by optical and scanning
electron microscopy (SEM). It was found that their occurrence is restricted to the clastic matrix of
these meteorites. The primary rock seems to have formed in an environment shielded from cosmic
radiation, since fragments of this rock are free of track-rich grains and solar noble gases. This finding
supports the solar nebula model for the formation of dust mantles around chondrules and other
coarse-grained components, and points against a regolith origin.
In Cold Bokkeveld, a small breccia-in-breccia clast was found, which has been irradiated as an
entity within the uppermost millimeters to meters of its parent body for at least about 3 Ma. This clast
seems to represent a compacted subsurface layer that was later excavated by impact and admixed to
the host breccia.
Furthermore, the results of this study may affect the interpretation of compaction ages obtained by
fission track methods, since these ages may be mixtures of different contact ages between finegrained, U-rich dust and U-poor olivines. In some cases, they may date the formation of dust mantles
in the solar nebula, while in other cases the lithification of the host breccias may be dated.
INTRODUCTION
CM chondrites are primitive meteorites originating from
parent bodies that have been affected to various extent by
impact brecciation and aqueous alteration, but seem to have
escaped considerable thermal processing after their formation
4.56 Ga ago (e.g., Sears and Dodd 1988; Brearley and Jones
1998). Hence, these rocks contain information on the physical
and chemical conditions in the protoplanetary nebula and
during planetary accretion. CM chondrites are impact
breccias composed of subangular mineral and lithic clasts set
in a clastic matrix (e.g., Fuchs et al. 1973; Dodd 1981;
Metzler et al. 1992; Metzler 1995). This texture is shown in
Fig. 1 for the CM chondrite Nogoya, where a variety of lithic
clasts, both lighter and darker than the surrounding clastic
matrix, can be observed. The boundaries between clasts and
matrix are usually sharp and can be identified
macroscopically (e.g., Heymann and Mazor 1967) or using
scanning electron microscopy (SEM) techniques (Fig. 2).
Most lithic clasts belong to a texturally well-defined rock type
(primary rock) that is internally unbrecciated, consists of
chondrules, CAIs, and other coarse-grained components,
most of which are surrounded by fine-grained rims or dust
mantles (Metzler et al. 1992; see upper part of Fig. 2). The
clastic matrix, in which the fragments of primary rock are
embedded, consists of fine-grained debris (see lower part of
Fig. 2). The mineralogical and chemical composition of
clastic matrix and primary rock in a given meteorite is very
1307
© Meteoritical Society, 2004. Printed in USA.
1308
K. Metzler
Fig. 1. Polished thin section from Nogoya. The section was etched by WN solution (see text) to reveal nuclear tracks. A variety of subangular
lithic clasts (primary rock) are visible embedded in a fine-grained clastic matrix. At the lower end the meteorite fusion crust is visible.
Transmitted light. Field of view is 2.2 cm.
Fig. 2. Sharp boundary (dashed line) between a fragment of primary rock (upper part) and the surrounding clastic matrix (lower part) in Cold
Bokkeveld. The upper part demonstrates the typical accretional texture of primary rock, i.e., the majority of coarse-grained components like
chondrules are enclosed by fine-grained dust mantles. SEM-BSE image. From Metzler et al. (1992).
similar, leading to the assumption that both lithologies formed
from coherent primary rock by comminution (Metzler et al.
1992).
Most CM chondrites are regolith breccias with enhanced
concentrations of solar wind implanted noble gases (e.g.,
Heyman and Mazor 1967; Smith et al. 1978) and contain
mineral grains that were targeted by heavy ions from solar and
galactic cosmic radiation (e.g., Pellas et al. 1969; Caffee et al.
1988). These grains can be detected by optical microscopy
after etching the samples with suitable chemical agents. These
so-called track-rich grains show high densities of nuclear
tracks (lattice defects) formed by the interaction with charged
particles (e.g., Fleischer et al. 1975). Two olivine grains of this
type from Cold Bokkeveld are shown in Fig. 3.
Formation of accretionary dust mantles in the solar nebula
1309
Fig. 3. Two preirradiated olivine grains with high densities of nuclear tracks in a polished thin section of Cold Bokkeveld. The sample was
etched by WN solution (see text) to reveal nuclear tracks. Transmitted light. Widths of field is 280 µm.
The main intention of the present study is to investigate
the occurrence, frequency, spatial distribution, and textural
setting of track-rich olivines using polished thin sections.
Although there is a considerable data base (e.g., Price et al.
1975; MacDougall and Price 1974; Goswami and Lal 1979),
most studies are based on olivines that were extracted from
their host meteorites by crushing the samples and hand
picking of suitable grains. Although generally appropriate for
basic investigations and for obtaining statistical data, this
method destroys the original textures and leaves no
information on the textural settings. The use of this standard
procedure makes it impossible to identify the host lithology of
track-rich components, and important information gets lost.
The work by Goswami and MacDougall (1983) seems to be
the only exception. They present a sketch of a thin section
from the CM chondrite Murray, which shows the locations of
in situ etched olivines in track-rich inclusions. However, at
the time, little was known about the internal texture of
brecciated CM chondrites and no attempt was made to
correlate the locations of these grains with distinct
lithological units.
To address this shortcoming, large polished thin sections
from Cold Bokkeveld, Mighei, Murchison, and Nogoya were
mapped with SEM, subsequently etched to reveal the tracks,
and investigated by optical microscopy. This method
preserves the textural relationship between track-rich grains
and their host lithologies.
The results of this study are relevant to the interpretation
of the bulk texture and the formation history of CM
chondrites, and help to answer the following questions (see
the Discussion section):
1. Is there a distinct host lithology for track-rich grains in
CM chondrites?
2. Is there a systematic correlation between the occurrence
of preirradiated grains and solar wind implanted noble
gases?
3. Did dust mantles around chondrules and other coarsegrained components form by solar nebula processes or
by regolith activity on the parent body?
SAMPLES AND ANALYTICAL TECHNIQUES
Eight polished thin sections (surface 1.5–5.3 cm2;
thickness about 40 microns) from four CM chondrites were
studied by optical microscopy (transmitted and reflected
light) and SEM (Table 1). Chemical analyses were obtained
using an energy dispersive X-ray analyzing system (Link
AN10000) attached to a JEOL 840A scanning electron
microscope.
For each thin section, a mosaic of backscattered electron
(BSE) images was prepared to map the texture and the
distribution of the main lithologies. These mosaics store
detailed information on the original state of the etched thin
sections and help identify textural features on the µm scale.
Six of the eight thin sections were etched in boiling WN
solution (Krishnaswami et al. 1971) for about 4.5 hr to reveal
the nuclear tracks in olivine. To protect the sensitive surfaces
from erosion during etching, the thin sections were stored in a
teflon container permeable to the etching solution. After
etching, the sections were cleaned in hot distilled water and,
subsequently, dried under red light.
After etching, each thin section was scanned by optical
1310
K. Metzler
Fig. 4. Texture of a fragment of primary rock from Cold Bokkeveld after etching with WN solution (see text). The typical accretional texture
with dust-mantled chondrules is now easily visible in transmitted light. Field of view is 1.4 mm.
Table 1. List of investigated samples and polished thin
sections.
Meteorite
Cold Bokkeveld
Mighei
Murchison
Nogoya
Polished thin
sections (PTS)
PL 92175
PL 92176
PL 91392
PL 92174
PL 93050
PL 94113a
PL 93048
PL 94043
Track analyses
X
X
X
X
X
X
microscopy with a magnification of 800× to identify and
locate track-rich olivines. These show either a much higher
density of nuclear tracks than the surrounding transit
irradiation background and/or a steep track gradient due to
irradiation by solar cosmic rays in the uppermost regolith
layers. Track-rich olivine grains identified by this method
were located and mapped on the BSE photomosaics.
Additionally, the size and track density of each grain was
determined to identify the track density distribution and
possible correlations between grain size and track density.
RESULTS
primary rock strongly contrast with the clastic matrix because
they appear lighter or darker than the surrounding material
(Fig. 1). The observation that fragments of primary rock in
Nogoya react differently to the etching solution can be
explained by the fact that these clasts have been affected to
various degrees by aqueous alteration on the parent body
(e.g., Zolensky and McSween 1988; Metzler 1995).
Much of the fine-grained material, e.g., clastic matrix and
fine-grained rims around chondrules, was removed during
etching. Only a thin layer of these components, directly glued
to the glass holder by the resin sheet, remains. On the other
hand, coarser components like chondrules and large mineral
grains are preserved showing their original thickness of about
40 microns. This observation implies that most olivine grains
smaller than the original thickness of the sections (about 40
microns) have been removed during etching and are lost for
track investigations. Due to the removal of fine-grained opaque
components, the characteristic agglomeration texture of
primary rock, previously visible only in backscattered electron
images, is now easily visible in transmitted light (Fig. 4).
The most important finding is that the textural
relationships between various components in all etched thin
sections are preserved, i.e., all track-rich grains can be
unambiguously assigned to their host lithologies.
Etching Effects on Polished Thin Sections
Background Track Densities (Transit Irradiation)
Due to chemical treatment, all thin sections changed their
appearance from nearly opaque to translucent. After etching,
many textural details, previously detectable only by scanning
electron microscopy, are visible by the naked eye, as shown
for Nogoya in Fig. 1. After etching, most fragments of
Background track densities in meteoritic minerals are
produced by galactic cosmic rays (GCR) during meteoroid
transit to Earth (e.g., Caffee et al. 1988; Vogt et al. 1990).
Background track densities observed in the investigated
samples (Table 2) vary between 3.6 × 104 tracks/cm2
Formation of accretionary dust mantles in the solar nebula
1311
Table 2. Nuclear track data for olivine grains from the investigated samples.
Meteorite
PTSa surface
(cm2)
Number of
analyzable
olivines
Background
track density
(tracks/cm2)b
Percentage of
preirradiated
olivine grains
Percentage of
preirradiated
olivines with
track gradient
Cold Bokkeveld
Inclusion CoBo-ii-1
Mighei
Murchison
Nogoya
4.6 (3PTS)
0.08
2.6
5.3
1.8
2400
25
810
2700
310
2.6 × 105
2.2 (3.5)d
100
–
2.1 (3.2)e
2.3
17 (27)d
0
–
25 (24)e
29
5.1 × 105
7.1 × 104
3.6 × 104
Mean track
density in
preirradiated
olivines
(tracks/cm2)c
1.8 × 107
1.5 × 107
–
2.5 × 107
1.1 × 107
a Polished
thin sections.
irradiation
c Track densities >5 × 107 were taken as 5 × 107 for calculation.
d MacDougall and Phinney (1977).
e Goswami and Lal (1979).
b Transit
(Nogoya) and 5.1 × 105 tracks/cm2 (Mighei). These values,
although depending on the shielding depth within the
corresponding meteoroids, roughly correspond to the
exposure ages of the host meteorites, which vary between
0.25 and 3 Ma (Metzler 1990). These exposure ages are based
on 21Ne production rates of 0.243 × 10−8 cc STP/g Ma
(Nishiizumi et al. 1980).
Textural Setting of Track-Rich Olivines
I inspected 6220 olivine grains in 6 thin sections using
optical microscopy (Table 2) and found that 115 grains have
been irradiated before transit of the parent meteoroid to Earth.
Olivines with track densities >106 tracks/cm2 were classified
as track-rich and were found in three of the four studied
meteorites in a small but remarkably constant quantity of
about 2.1–2.3%. These values are similar to the results by
MacDougall and Phinney (1977) and Goswami and Lal
(1979) who obtained values between 1.8 and 3.5%. The
grains were mapped in the previously prepared BSE
photomosaics and the results were transformed into sketch
maps shown in Figs. 5a–5f. These sketch maps show the main
CM lithologies, i.e., clastic matrix and embedded fragments
of primary rock.
All track-rich grains are located in the clastic matrix of
these meteorites (Fig. 5). The locations are indicated by black
dots. No track-rich grain has been found within fragments of
primary rock (hatched areas). The sketch map of the Mighei
thin section (Fig. 5d) indicates that it is a large fragment of
primary rock, free of any track-rich olivine. Track-rich grains
have been found in the clastic matrix of Murchison, Cold
Bokkeveld, and Nogoya. A common feature is their distinctly
inhomogeneous spatial distribution. In Murchison and
Nogoya, large areas of clastic matrix are free of them, while
they are concentrated in others. This is most obvious in the
case of Murchison (Fig. 5e) where the left part of the section
shows a fairly homogeneous distribution, contrasted by the
inhomogeneous distribution on the right hand side. This
figure visualizes earlier findings by Price et al. (1975) and
Goswami and Lal (1979), who describe large variations in the
proportion of track-rich grains in bulk olivine separates
obtained from different locations in a given CM specimen.
In addition to isolated track-rich olivines, two thin
sections of Cold Bokkeveld, which had been sliced parallel to
each other, contain a specific lithology full of track-rich
olivines. This fragment differs from the surrounding clastic
matrix by its distinct texture and mineralogical composition.
It is named CoBo-ii-1, has a size of about 4 × 1 mm, and is
characterized by its elongated shape (Fig. 6). This fragment
can be observed in 4 parallel thin sections, two of which were
not etched for track revelation (see Table 1; Fig. 7). The 3dimensional information from Fig. 7 indicates that it
represents a platy-shaped fragment that must have been
admixed to the Cold Bokkeveld breccia before compaction.
Petrographic investigations of the four parallel thin sections
revealed that this fragment itself is a breccia (breccia-inbreccia-structure) consisting of fragments of primary rock,
embedded in fine-grained clastic matrix. CoBo-ii-1 has a
mineralogical composition that is distinctly different from its
surrounding material and is characterized by the absence of
tochilinite and the frequent occurrence of magnetite clusters.
Its texture and mineralogy resembles the more heavily
(aqueously) altered portions of other CM chondrites, e.g.,
Nogoya (e.g., Metzler 1995). Although it is beyond the scope
of this paper to describe the composition of this fragment in
detail, it seems clear that it represents a brecciated subsurface
layer of the Cold Bokkeveld parent body, which has
undergone a specific evolution (see below).
Track Densities in Preirradiated Olivines
The mean track density in preirradiated olivines is
remarkably constant, ranging from 1.1–2.5 × 107 tracks/cm2
(Table 2). The percentage of olivines with track gradients,
irradiated by solar cosmic rays directly on the parent body
surface, varies between 17 (Cold Bokkeveld) and 29
1312
K. Metzler
Fig. 5. Sketch maps indicating the spatial distribution of the main lithologies and preirradiated olivines (black dots) in thin sections of the
investigated samples. The distribution of track-rich olivines is inhomogeneous and the occurrence of these grains is restricted to the clastic
matrix. The hatched areas represent fragments of primary rock, areas without signature represent clastic matrix. Dashed lines in sketch maps
from Cold Bokkeveld indicate the preirradiated breccia-in-breccia clast CoBo-ii-1 (compare Figs. 6 and 7): a, b, and c) Cold Bokkeveld; d)
Mighei; e) Murchison; f) Nogoya.
Formation of accretionary dust mantles in the solar nebula
1313
Fig. 6. Outline and texture of the preirradiated breccia-in breccia clast CoBo-ii-1 from Cold Bokkeveld. This object has a mineralogical
composition distinctly different from the surrounding material, characterized by the occurrence of magnetite and the absence of tochilinite.
SEM-BSE image.
of Mighei, which consists entirely of a single fragment of
primary rock. Olivines in this lithology show track density
distributions typical for single stage irradiations of the
corresponding meteoroids during transit from parent body to
Earth (see above).
Inclusion CoBo-ii-1 (Breccia-in-Breccia Structure)
All olivines in this lithology that are suitable for track
counting (25 grains) are track-rich. The track densities vary
from grain to grain between 0.52 and 2.78 × 107 tracks/cm2,
with a mean track density of 1.48 × 107 tracks/cm2 (Table 2;
Fig. 8). The variation factor of 5.4 is close to the value of 4,
which is the variation in track revelation due to different
crystal orientations in thin sections (Pellas, personal
communication).
Clastic Matrix
Fig. 7. Sketch map showing the shape and location of the
preirradiated breccia-in breccia clast CoBo-ii-1 in four parallel thin
sections from Cold Bokkeveld.
(Nogoya). These values are similar to literature data obtained
from extracted olivines (MacDougall and Phinney 1977;
Goswami and Lal 1979).
Fragments of Primary Rock
Although large areas of the studied thin sections consist
of fragments of primary rock (Fig. 5), none of these fragments
contain track-rich grains. The same holds for the thin section
Data for preirradiated olivines in the clastic matrix from
Cold Bokkeveld and Murchison are shown in Fig. 9. In both
meteorites, the track densities vary between 1 × 106 tracks/
cm2 and >5 × 107 tracks/cm2. Grains with track gradients
show similar distributions. Please note that the data for Cold
Bokkeveld (Fig. 9a) include those from the breccia clast
CoBo-ii-1, leading to the peak between 1 and 2 × 107 tracks/
cm2 (compare Fig. 8). The data for several track-rich grains
included in the database of Table 2 are not shown in this
figure because they were not suitable for track counting (due
to inclusions, opacity, etc.).
To identify possible correlations between grain sizes and
track densities, the data from Murchison are used and shown
in Fig. 10. The grain sizes of track-rich grains vary between
1314
K. Metzler
a
Fig. 8. Track density distribution in the preirradiated breccia-in
breccia clast CoBo-ii-1 from Cold Bokkeveld. A distinct narrow peak
is evident, indicating that this clast was irradiated near the parent
body’s surface as an entity. Grains with track gradients are absent.
a
40 and 700 microns. Grains with track gradients were
observed exclusively in the size fraction between 110 and 700
microns, but this observation might result from the fact that
track gradients in smaller grains are difficult to detect.
DISCUSSION
Brecciation of CM Chondrites in a Planetary Environment
Track-rich olivines were found throughout the clastic
matrix of the investigated samples, while lithic clasts are free
of these grains. This is the typical finding for extraterrestrial
impact breccias, where the track-rich components were
admixed to the fine-grained portion during impact activity on
the parent body’s surface (e.g., Bunch and Rajan 1988; Caffee
et al. 1988). During this process, solar and galactic particle
irradiation penetrated the uppermost millimeters to meters of
the parent body surfaces and produced freshly irradiated
components for admixture. In the case of CM chondrites,
fragments of primary rock probably represent fresh samples
of unbrecciated bedrock excavated by impacts from the
depths that were never reached by cosmic radiation. These
fragments were admixed to the CM breccias together with
track-rich grains that previously resided at the very surface of
the parent body.
b
Fig. 9. Track densities in preirradiated olivines in the clastic matrix
from Cold Bokkeveld and Murchison. Note that the data for Cold
Bokkeveld include those from the preirradiated breccia-in breccia
clast CoBo-ii-1 (compare Fig. 8). Grains with track gradients are
homogeneously distributed over the entire range of track densities: a)
Cold Bokkeveld; b) Murchison.
Formation of accretionary dust mantles in the solar nebula
1315
Fig. 10. Grain size versus track density diagram for preirradiated olivines from Murchison.
On the other hand, it is remarkable that the amount of
preirradiated grains is relatively low compared to other types
of regolith breccias, e.g., those from the Moon (Macdougall
1982). The absence (or at least scarcity) of preirradiated lithic
clasts also discerns CM chondrites from achondritic, lunar,
and OC regolith breccias. These differences may be explained
by the very old compaction ages of these meteorites, which
give values in the order of 4.5 Ga (see below). These data
imply that CM breccias are not the result of recent impact
activity but formed during an epoch of planetary evolution
that was dominated by rapid growth and burial of preexisting
surfaces rather than by regolith reworking.
Inclusion CoBo-ii-1: A Preirradiated Breccia-in-Breccia
Structure in Cold Bokkeveld
The clast Cobo-ii-1 seems to represent a preirradiated
breccia-in-breccia structure, formed by impact processes on
the parent body. It is important to note that this clast type has
texturally nothing in common with fragments of primary rock
with their typical accretionary texture.
The variation factor in track densities of only 5.4 within
this clast of 4 mm in length indicates that it was irradiated as
a single entity by a uniform dose of galactic cosmic particle
irradiation, otherwise, the observed track densities should
differ much more. Since no solar flare tracks (track gradients)
can be observed, this clast was probably irradiated in a nearsurface layer of the regolith. The clast must have been
shielded from solar radiation while located within the GCRtrack production zone, which means a depth between a few
millimeters to a few meters. Assuming a location just below
the solar flare track production zone, i.e., in the uppermost
millimeters of the regolith layer, the clast must have been
irradiated for about 3 Ma by GCR. This value was obtained
using the observed mean track density and the nomogram by
Bhattacharya et al. (1973). It is a minimum value since
locations deeper within the regolith layer would increase the
irradiation times significantly.
The clast Cobo-ii-1 seems to have undergone the
following evolution: 1) formation as part of primary rock with
its typical accretionary fabric (e.g., dust mantles around
chondrules, CAIs, etc.); 2) aqueous alteration (e.g., formation
of magnetite clusters); 3) impact brecciation and admixture of
unaltered material (e.g., Fe-rich olivine without phyllosilicate
reaction zones); 4) breccia compaction; 5) irradiation by
galactic cosmic rays in close vicinity to the parent body
surface, shielded from solar cosmic radiation; 6) second stage
of impact brecciation, launch, and admixture to the present
breccia in the shape of a lithic clast; and 7) breccia
compaction. Although this scenario seems to be rather
complicated, especially concerning the multiple impact
brecciation and compaction, it is well-known from many
other types of extraterrestrial breccias.
Formation of Accretionary Dust Mantles: Parent Body
Regolith versus Solar Nebula
Models for the Formation of Accretionary Dust Mantles
The majority of lithic clasts in CM chondrites belongs to
a texturally well-defined rock type (upper part of Fig. 2) that
can be described as an agglomerate of chondrules and other
coarse-grained components, most of which are surrounded
1316
K. Metzler
by accretionary dust mantles. Due to its ubiquitous
occurrence and accretionary texture, this rock type is called
“primary (accretionary) rock” by Metzler et al. (1992).
These authors explain the formation of dust mantles by
sticking of fine-grained dust to chondrule surfaces in the
solar nebula, a potential mechanism already proposed by
Bunch and Chang (1984) and is now widely accepted (e.g.,
Hewins and Herzberg 1996; Paque and Cuzzi 1997; Morfill
et al. 1998; Cuzzi et al. 1998; Lauretta et al. 2000; Simon
and Grossman 2001; Hua et al. 2002; Zega and Buseck
2003).
An alternative model explains the formation of dust
mantles by regolith processes on the parent body (Sears et al.
1993). This model assumes that the formation of dust mantles
and admixture of track-rich grains occurred simultaneously in
a single process, namely regolith gardening due to impact
activity on the parent body. These authors assume that dust
mantles possibly formed by the “objects sloshing about in the
mud” of the “apparently very wet regolith.” In the view of
these authors, the entire texture of CM chondrites is the
product of planetary processes.
Both models predict that CM chondrites contain trackrich grains, but there is a decisive difference concerning their
textural setting. The model by Metzler et al. predicts their
occurrence exclusively in the clastic matrix, since it assumes
that, in the first step, dust mantles formed in the solar nebula,
followed by a calm accretion of mantled objects to the parent
body and suffering mild compaction and lithification. Up to
this evolutionary state, the material was shielded from cosmic
radiation. In the second step, brecciation and irradiation are
the dominant processes in a more violent regolith
environment, leading to the formation of clastic matrix with
admixture of track-rich olivines. In contrast to this
chronological sequence, the model by Sears et al. predicts the
occurrence of track-rich components throughout the entire
breccia, i.e., within clastic matrix portions, within fragments
of primary rock, and even within dust mantles.
Shielding of Primary Rock During Accretion
The existence of primary rock as a major lithology seems
to prove that the earliest rock generation from CM parent
bodies mainly formed from dust-mantled components,
followed by compaction. The observation that the
investigated fragments of primary rock do not contain any
preirradiated olivine implies that during this process, the
material was shielded from cosmic radiation. This conclusion
is confirmed by Nakamura et al. (1999a, b), who performed in
situ noble gas analyses using a laser microprobe. It was found
that fragments of primary rock from Murray, Murchison,
Nogoya, and Yamato-791198 do not contain solar gases but
are dominated by planetary noble gas components. These
observations, together with the accretionary overall
appearance of primary rock, clearly refute a regolith origin as
proposed by Sears et al. (1993).
Shielding of Accretionary Dust Mantles During Their Formation
in the Solar Nebula
The nuclear track data presented here clearly support the
solar nebula model for dust mantle formation because the
primary rock and its dust mantled components were shielded
from radiation, possibly by nebular dust.
There is a positive correlation between the occurrence of
track-rich olivines and solar noble gases. Nakamura et al.
(1999a, b) have shown that solar noble gases are
preferentially located in the clastic matrix, i.e., solar gases
and track-rich olivines share the same host lithology.
The formation of accretionary dust mantles apparently
results from the sticking of dust to the surfaces of chondrules in
a nebular environment, shielded from cosmic radiation.
Summary of Arguments for a Solar Nebula Origin of
Accretionary Dust Mantles
The observations on track-rich olivines complete a long
list of arguments for a nebular origin of dust mantles around
chondrules and other coarse-grained components in CM
chondrites (see Metzler et al. 1992; Metzler and Bischoff
1996). These data indicate that the formation of dust mantles,
the agglomeration of dust mantled objects, and the
lithification and brecciation/irradiation on the parent body
seems to be the most plausible chronological sequence in CM
chondrite formation. The main arguments for a solar nebula
origin of accretionary dust mantles are:
1. No track-rich olivines exist in primary rock, but are
present in the clastic matrix (this study).
2. No solar wind implanted noble gases exist in primary
rock and dust mantles, but are present in the clastic
matrix (Nakamura et al. 1999a, b).
3. High amounts of planetary noble gases exist in dust
mantles (Nakamura et al. 1999a).
4. Presolar silicon carbide grains occur in dust mantles
(Swan and Walker 1998).
5. Dust mantles exist only around primary nebula products
like chondrules, CAIs, etc., not around lithic clasts or
other components that were formed on the parent body
(Metzler et al. 1992).
Reevaluation of Compaction Ages for CM Chondrites
MacDougall and Kothari (1976) measured fission tracks
in olivines from CM chondrites to calculate the so called
compaction ages of CM chondrites. These age calculations
are based on fission track densities produced by the
spontaneous fission of Pu and U, and counted on contact faces
between “U-poor” isolated olivine grains and “U-rich” finegrained matrix. The authors conclude that their compaction
ages, ranging from 4.22 to 4.42 Ga, represent the time when
olivine grains and matrix came intimately in contact,
followed by compaction and lithification. The data from each
meteorite show a considerable spread and the compaction
Formation of accretionary dust mantles in the solar nebula
ages, calculated from the mean fission track densities in a
given sample leave a gap of 0.2–0.4 Ga between compaction
and the formation of the solar system. Woolum and
Hohenberg (1993) recalculated the data using a reestimated
primordial Pu/U ratio and found that the ages obtained by
MacDougall and Kothari (1976) now shift to values between
4.48–4.52 Ga, i.e., much closer to the 4.56 Ga of solar system
formation. This means that both solar wind implanted noble
gases and track-rich grains result from the irradiation by the
ancient sun because both components were admixed to these
breccias before compaction and lithification.
Nevertheless, the large variation of fission track
densities between single grains in a given meteorite remains
to be explained. MacDougall and Kothari (1976) discuss
several possibilities, e.g., dislocations counted as tracks and
Pu-variations in the fine-grained matrix. Another
explanation arises when the petrographic observations in
CM chondrites (Metzler et al. 1992) and the present data are
taken into account. These data indicate a multi-stage
evolution of CM chondrites, where U-rich matrix and Upoor olivines came into contact by two different processes.
At first, fine-grained dust coated the surfaces of olivines in
the solar nebula, forming thick layers of “matrix” and
producing the earliest generation of fission tracks on olivine
surfaces. After accretion of these dust mantled components,
the resulting planetesimals were reworked by impact
processes, leading to the break-up of preexisting olivinebearing components like chondrules. A population of fresh
olivine grains with pristine surfaces was produced and
admixed to the clastic matrix of the resulting breccias,
leading to a second onset of fission track production on
fresh olivine surfaces.
Although difficult to quantify, this multi-stage evolution
may contribute to the observed data spread. Since it is not
known from which lithologies the olivine grains were
separated by MacDougall and Kothari (1976), the meaning of
mean fission track values and the corresponding mean
compaction ages appears of limited value. These values may
include data from grains that came into contact with the matrix
in the solar nebula and those from a later period, when olivines
came into contact with fine-grained material in a planetary
environment due to impact activity. In my view, the highest
fission track values from single grains date the time of dust
mantle formation, while the lowest values date the real closure
of the systems, i.e., the compaction of the host breccias.
SUMMARY AND CONCLUSIONS
Eight polished thin sections from four CM chondrites
were studied by optical and scanning electron microscopy to
investigate the occurrence, frequency, spatial distribution, and
textural setting of preirradiated (track-rich) olivines in CM
chondrites. The intention of this study was to correlate the
occurrence of these grains with the main lithological units,
1317
i.e., clastic matrix and fragments of primary rock and to
discern between two models for CM chondrite formation. The
results of the present study are:
1. The background track density (Table 2) due to transit
irradiation (3.6 × 104 to 5.1 × 105 tracks/cm2) is clearly
discernable from the track density in preirradiated
olivines (>1 × 106 tracks/cm2).
2. Three of the four samples (Cold Bokkeveld, Murchison,
Nogoya) are brecciated and contain a remarkably
constant quantity of track-rich olivines (2.1–2.3%;
Table 2). The Mighei sample appears to be unbrecciated
and is free of these grains.
3. The mean track densities in track-rich grains ranges
between 1.1 and 2.5 × 107 tracks/cm (Table 2).
4. The percentage of preirradiated olivines with track
gradients, typical of regolith irradiation, varies between
17 (Cold Bokkeveld) and 29 (Nogoya; Table 2).
Extracting the data for clast CoBo-ii-1 from the Cold
Bokkeveld data narrows the range to 25–31%, which is
close to literature data.
5. There is no correlation between the track density in a
given grain and the occurrence of a track gradient (Fig. 9).
6. The clastic matrix is the host lithology for all observed
track-rich olivines. Fragments of primary rock are free of
these grains.
7. A single lithic clast (CoBo-ii-1), rich in track-rich
olivines, is found in Cold Bokkeveld, representing a
breccia-in-breccia structure, typical for extraterrestrial
impact breccias. This lithology seems to have been
irradiated for at least 3 Ma as part of a brecciated and
aqueously altered subsurface layer on its parent body.
8. Using recent data on in situ noble gas measurements
(Nakamura et al. 1999a, b) based on the same petrologic
classification scheme as used in the present work, it turns
out that there is a clear positive correlation between the
occurrence of solar type noble gases and track-rich
olivines.
These results complete a long list of arguments for a
regolith origin of clastic matrix and a nebular origin of dust
mantles around chondrules and other coarse-grained
components in CM chondrites (Metzler et al. 1992; Metzler
and Bischoff 1996).
In interpreting calculated compaction ages, one should
consider this multistage formational history of CM
chondrites. According to the model described above,
conventional compaction ages might be mixtures of different
contact ages between fine-grained U-rich dust and U-poor
olivines, each with a distinct meaning, i.e., formation of dust
mantles in the solar nebula and the lithification of the host
breccias, respectively.
Acknowledgments–I am grateful to the late Paul Pellas and to
colleagues who gave me the opportunity to get familiar with
the methods of nuclear track investigation. I thank the
1318
K. Metzler
Museum of Natural History (London) for providing
meteorite samples. I thank U. Ott and an anonymous
reviewer for their reviews and M. Caffee for his editorial
handling, which led to substantial improvements of the
manuscript. This study was supported by the Deutsche
Forschungsgemeinschaft (DFG).
Editorial Handling—Dr. Marc Caffee
REFERENCES
Bhattacharya S. K., Goswami J. N., and Lal D. 1973. Semiempirical
rates of formation of cosmic ray tracks in spherical objects
exposed in space: preatmospheric and postatmospheric depth
profiles. Journal of Geophysical Research 78:8356–8363.
Brearley A. J. and Jones R. H. 1998. Chondritic meteorites. In
Reviews in mineralogy, edited by J. J. Papike. Washington D.C.:
Mineralogical Society of America. pp. 3-1–3-398.
Bunch T. E. and Chang S. 1984. CAI rims and CM2 dust balls:
Products of gas-grain interactions, mass transport, grain
aggregation and accretion in the nebula? (abstract). 15th Lunar
and Planetary Science Conference. pp. 100–101.
Bunch T. E. and Rajan R. S. 1988. Meteorite regolith breccias. In
Meteorites and the early solar system, edited by Kerridge J. F.
and Matthews M. S. Tucson: University of Arizona Press. pp.
144–164.
Caffee M. W., Goswami J. N., Hohenberg C. M, Marti K., and Reedy
R. C. 1988. Irradiation records in meteorites. In Meteorites and
the early solar system, edited by Kerridge J. F. and Matthews M.
S. Tucson: University of Arizona Press. pp. 205–245.
Cuzzi J. N., Hogan R. C., Paque J. M., and Dobrovolskis A. R. 1998.
Chondrule rimming by sweepup of dust in the protoplanetary
nebula: Contraints on primary accretion (abstract #1439). 29th
Lunar and Planetary Science Conference. CD-ROM.
Dodd R. T. 1981. Meteorites: A petrologic-chemical synthesis.
Cambridge: Cambridge University Press. 368 p.
Fleischer R. L., Price P. B., and Walker R.M. 1975. Nuclear tracks in
solids. Berkeley: University of California Press. 605 p.
Fuchs L. H., Olsen E., and Jensen K. J. 1973. Mineralogy, mineral
chemistry, and composition of Murchison (C2) meteorite.
Smithsonian Contributions to Earth Sciences 10:1–39.
Goswami J. N. and Lal D. 1979. Formation of the parent bodies of the
carbonaceous chondrites. Icarus 40:510–521.
Goswami J. N. and MacDougall J. D. 1983. Nuclear track and
compositional studies of olivines in CI and CM chondrites.
Proceeding, 13th Lunar and Planetary Science Conference.
Journal of Geophysical Research 88:A755–A764.
Hewins R. H. and Herzberg C. T. 1996. Nebular turbulence,
chondrule formation, and the composition of the Earth. Earth and
Planetary Science Letters 144:1–7.
Heyman D. and Mazor E. 1967. Light-dark structure and rare gas
content of the carbonaceous chondrite Nogoya. Journal of
Geophysical Research 72:2704–2707.
Hua X., Wang J., and Buseck P. R. 2002. Fine-grained rims in the
Allan Hills 81002 and Lewis Cliff 90500 CM2 meteorites: Their
origin and modification. Meteoritics & Planetary Science 37:
229–244.
Krishnaswami S., Lal D., Prabhu N., and Tamhane A. S. 1971.
Olivines: Revelation of tracks of charged particles. Science 174:
287–291.
Lauretta D. S., Hua X., and Buseck P. R. 2000. Mineralogy of finegrained rims in the ALH 81002 CM chondrite. Geochimica et
Cosmochimica Acta 64:3263–3273.
MacDougall J. D. 1982. Space exposure of breccia components. In
Workshop on lunar breccias and soils and their meteoritic
analogues, edited by Taylor G. J. and Wilkening L. Technical
Report 82-02. Houston: Lunar and Planetary Institute. pp. 94–96.
MacDougall J. D. and Kothari B. K. 1976. Formation chronology for
C2 meteorites. Earth and Planetary Science Letters 33:36–44.
MacDougall J. D. and Phinney D. 1977. Olivine separates from
Murchison and Cold Bokkeveld: Particle tracks and noble gases.
Proceedings, 8th Lunar and Planetary Science Conference. pp.
293–311.
MacDougall D. and Price P. B. 1974. Low-energy particle irradiation
and possible age indicator for components of carbonaceous
chondrites. Meteoritics 9:370–371.
Metzler K. 1990. Petrographische und mikrochemische
Untersuchungen zur Akkretions- und Entwicklungsgeschichte
chondritischer Mutterkörper am Beispiel der CM-Chondrite.
Ph.D. thesis, University of Münster, Münster, Germany.
Metzler K. 1995. Aqueous alteration of primary rock on the CM
parent body (abstract). 26th Lunar and Planetary Science
Conference. pp. 961–962.
Metzler K. and Bischoff A. 1996. Constraints on chondrite
agglomeration from fine-grained chondrule rims. In Chondrules
and the protoplanetary disk, edited by Hewins R. H., Jones R. H.,
and Scott E. R. D. Cambridge: Cambridge University Press. pp.
153–161.
Metzler K., Bischoff A., and Stöffler D. 1992. Accretionary dust
mantles in CM chondrites: Evidence for solar nebula processes.
Geochimica et Cosmochimica Acta 56:2873–2897.
Morfill G. E., Durisen R. H., and Turner G. W. 1998. An accretion rim
constraint on chondrule formation theories. Icarus 134:180–184.
Nakamura T., Nagao K., and Takaoka N. 1999a. Microdistribution of
primordial noble gases in CM chondrites determined by in situ
laser microprobe analysis: Decipherment of nebular processes.
Geochimica et Cosmochimica Acta 63:241–255.
Nakamura T., Nagao K., Metzler K., and Takaoka N. 1999b.
Heterogeneous distribution of solar and cosmogenic noble
gases in CM chondrites and implications for the formation of
CM parent bodies. Geochimica et Cosmochimica Acta 63:257–
273.
Nishiizumi K, Regnier S., and Marti K. 1980. Cosmic ray exposure
ages of chondrites, preirradiation and constancy of cosmic ray
flux in the past. Earth and Planetary Science Letters 50:156–170.
Paque J. M. and Cuzzi J. N. 1997. Physical characteristics of
chondrules and rims, and aerodynamic sorting in the solar nebula
(abstract). 28th Lunar and Planetary Science Conference. pp.
1071–1072.
Pellas P., Poupeau G., Lorin J. C., Reeves H., and Audouze J. 1969.
Primitive low-energy particle irradiation of meteoritic crystals.
Nature 223:272–274.
Price P. B., Hutcheon I. D., Brady D., and MacDougall D. 1975.
Track studies bearing on solar-system regoliths. Proceedings, 6th
Lunar and Planetary Science Conference. pp. 3449–3469.
Sears D. W. G. and Dodd R. T. 1988. Overview and classification of
meteorites. In Meteorites and the early solar system, edited by
Kerridge J. F. and Matthews M. S. Tucson: University of Arizona
Press. pp. 3–31.
Sears D. W. G., Benoit P. H., and Jie L. 1993. Two chondrule groups
each with distinctive rims in Murchison recognized by
cathodoluminescence. Meteoritics 28:669–675.
Simon S. B. and Grossman L. 2001. The isolated olivine population
and accretionary rims observed in Tagish Lake (abstract).
Meteoritics & Planetary Science 36:A189–190.
Smith S. P., Huneke J. C., and Wasserburg G. J. 1978. Neon in gas-rich
samples of the carbonaceous chondrites Mokoia, Murchison, and
Cold Bokkeveld. Earth and Planetary Science Letters 39:1–13.
Formation of accretionary dust mantles in the solar nebula
Swan P. D. and Walker R. M. 1998. Distribution of in situ silicon
carbide grains in primitive meteorites and their use in identifying
nebular components (abstract). Meteoritics & Planetary Science
33:A152.
Vogt S., Herzog G. F., and Reedy R. C. 1990. Cosmogenic nuclides
in extraterrestrial materials. Reviews of Geophysics 28:253–275.
Woolum D. S. and Hohenberg C. 1993. Energetic particle
environment in the early solar system: Extremely long precompaction meteoritic ages or an enhanced early particle flux. In
1319
Protostars and planets III, edited by Levy E. and Lunine J. I. pp.
903–919.
Zega T. J. and Buseck P. R. 2003. Fine-grained-rim mineralogy of the
Cold Bokkeveld CM chondrite. Geochimica et Cosmochimica
Acta 67:1711–1721.
Zolensky M. and McSween H. Y., Jr. 1988. Aqueous alteration. In
Meteorites and the early solar system, edited by Kerridge J. F.
and Matthews M. S. Tucson: University of Arizona Press. pp.
114–143.