Supplementary Figure 1. Polished slabs of Gujba and Hammadah al Hamra 237 (HH 237). HH
237 is more fine-grained than Gujba, but contains some metal-sulfide (met-sf) nodules and
skeletal olivine (SO) chondrules similar in size to those in Gujba. Courtesy of E. Twelker.
Supplementary Figure 2. Backscattered electron images of chondrules in the Gujba meteorite.
Regions outlined in a, c, e are shown in detail in b, d, f, respectively. Chondrules have skeletal
olivine textures and lack relict grains, coarse-grained igneous rims, and fine-grained matrix-like
rims. They consist of forsteritic olivine (Fa1-4), low-Ca pyroxene (Fs2±2Wo3±2), high-Ca pyroxene
(Fs2±1Wo44±8), and alkali-poor glassy mesostasis. Both pyroxenes contain high and variable
Al2O3 contents (see Supplementary Table 1), suggesting fast cooling rates during solidification.
Abbreviations: cpx, high-Ca pyroxene; mes, mesostasis; ol, olivine; px, low-Ca pyroxene.
Supplementary Figure 3. Backscattered electron images of inter-chondrule material in the Gujba
meteorite. Region outlined in a is shown in detail in b. The interchondrule material consists of
coarse chondrule fragments (fr); fine-grained matrix material is absent.
SUPPLEMENTARY DOCUMENT
Pb-isotopic results for Gujba
Each chondrule was coarsely crushed, and several fractions of chondrule fragments
devoid of visible terrestrial weathering (unless indicated otherwise) were picked from each
chondrule. Fraction descriptions and Pb-isotopic results are presented in Table 1. Each fraction
was washed in distilled acetone, followed either by repeated ultrasonic agitation in 2M HCl and
1M HBr, or by heating in 7M HNO3 and 6M HCl. Pb separation was done using anion exchange
in HBr+HNO3, modified from Lugmair and Galer (1992). Fractions containing 50-300 pg of Pb
were analyzed using a two-step static multicollector procedure on a Triton TI mass spectrometer
at the Geological Survey of Canada. Chondrule fractions grinded and washed in hot 7M HNO3
and in hot 6M HCl ±1M HBr, followed by several cycles of dilute HCl and HBr with ultrasonic
agitation. Pb separation was done using anion exchange in HBr+HNO3, modified from Lugmair
and Galer 1992. Procedure blanks were between 1-3 pg Pb. Most loads contain between 30-300
pg Pb. Pb isotopic analysis by TIMS using 2-step static multicollector procedure, Triton TI at the
GSC, Merck silicic acid high efficiency emitter. Isochron regressions and weighted average
calculations were performed using Isoplot-Ex 3.00 (Ludwig, 2003).
0.68
Gujba
Chondrules 3,4,5 (n=11):
Age = 4562.69 ± 0.49 Ma
MSWD = 1.3
0.64
Chondrules 1,2 (n=11):
Age = 4551.9 ± 7.2 Ma
MSWD = 33
207
Pb/
206
Pb
0.66
0.62
0.60
0.000
0.002
0.004
0.006
204
0.008
Pb/206Pb
0.010
0.012
Fragments from chondrules 3, 4 and 5 produced highly radiogenic Pb-isotopic analyses
with measured 206Pb/204Pb ratios of 307-2602 (four out of 11 analyses yielded 206Pb/204Pb>1000).
The common Pb content in most of these fractions is close to analytical blank, therefore the
influence of common Pb heterogeneity on the age calculation is very small. Isochron regression
of these data yielded an age of 4562.7±0.5 Ma (MSWD=1.3), which we consider the best
estimate for the timing of formation of the Gujba silicate chondrules.
Eleven fractions from the chondrules 1 and 2 produce a scattered array, which yields an
errorchron date of 4551.9±7.2 Ma, MSWD=33 (“errorchron” is an isochron with dispersion of
analytical points greater than expected from analytical errors). One fraction from the chondrule 1
plots on the isochron defined by chondrules 3, 4 and 5, whereas all the other analyses plot below
that isochron. The
207
Pb/206Pb ratios in highly radiogenic analyses from the chondrules 1 and 2
are dispersed as much as in less radiogenic analyses. This dispersion thus cannot be explained
entirely by common Pb heterogeneity. It indicates variable ancient loss of radiogenic Pb from
chondrules 1 and 2, which can be a result of shock metamorphism experienced by Gujba
(Weisberg & Kimura, 2004). On the other hand, the consistency of radiogenic
207
Pb/206Pb
between the fragments of chondrules 3, 4 and 5, strongly suggests single-stage evolution without
noticeable influence from the secondary processes, because there are many factors (e.g., grain
size, mineral composition, fracturing) that make shock- or alteration-related Pb-loss vary, but
there is no process known to stabilize the degree of secondary Pb-loss.
Pb-isotopic results for HH 237
Three methods were used for separation of silicate fractions from HH 237. Fractions 1-5
were separated by dissolution of metal from a coarsely crushed bulk meteorite in ~2M HCl
during 48 hours, with subsequent removal of remaining metal and other magnetic material by a
hand magnet and hand-picking under a binocular microscope. Fractions 6 and 7 were separated
from more finely crushed meteorite material by a hand magnet. In addition, two large individual
chondrules (8 and 9) were extracted from a polished meteorite slab using dental tools.
Each silicate fraction was cleaned by sequential leaching in hot 7M HNO3 and 6M HCl.
The HNO3 and HCl were collected separately and analyzed for Pb isotopes along with the
residues. Chemical separations, mass spectrometry, and data reduction procedures are the same
as those used for Gujba analyses. Leaching residues contain relatively radiogenic Pb with
measured
206
Pb/204Pb ratios between 97-405. The first (HNO3) leachates contain almost entirely
“modern” common Pb (206Pb/204Pb between 16.9-22.8), whereas the second (HCl) leachates
contain a mixture of common Pb and radiogenic Pb (206Pb/204Pb between 30-85). The Pb
isochron results for the residues are shown in Table 2 and in the isochron diagram below:
0.675
207
Pb/206Pb
0.665
0.655
Hammadah al Hamra 237
- Blue - bulk silicate, frac. 1-5
- Grey - bulk silicate, frac. 6, 7
- Red - individual chondrules,
fractions 8, 9
0.645
0.635
Fractions 1-5:
Age = 4564.3 ± 2.0 Ma
MSWD = 1.19
0.625
0.615
0.000
0.002
0.006
0.004
0.008
0.010
0.012
Pb/206Pb
204
An isochron regression for five fractions extracted by metal dissolution (blue symbols)
yields the age of 4564.3±2.0 Ma (MSWD=1.19). Although the scatter of this regression line does
not exceed analytical errors, the precision of the age is compromised by extrapolation of the error
envelope from moderately radiogenic Pb-isotopic data to the y-axis.
Two bulk silicate fractions extracted by magnetic susceptibility (6 and 7, grey symbols)
yielded slightly less radiogenic Pb analyses. The analyses of individual chondrules (8 and 9, red
207
symbols) yielded more radiogenic Pb, but lower
Pb/206Pb ratios, which are inconsistent with
the isochron for the fractions extracted by metal dissolution. This pattern can be explained if the
HH 237 silicates are pervasively affected by a secondary process, possibly similar to the process
that influenced the Gujba chondrules 1 and 2 (i.e., shock metamorphism, Meibom et al., 2005).
Minerals produced or modified by shock metamorphism have lost their earliest accumulation of
radiogenic Pb, and their presence forces the
207
Pb/206Pb ratios down. It is possible that these
minerals were converted to an acid-soluble form by a prolonged (48 hours) contact with HCl
during metal dissolution, whereas a shorter (several hours) hot-acid washing alone does not
dissolve them completely.
Determination of precise and accurate Pb-isotopic age in the presence of multiple
common Pb components and secondary alteration effects is difficult. Since the fractions
extracted by metal dissolution appear to contain no secondary material, we can try to improve
precision by considering two-point and three-point leachate-residue isochrons for individual
fractions (Table 3). This approach is based on the assumption that one of the common Pb
components is more acid-soluble than the other. In this case, the more soluble component is
removed by the first leaching step, whereas the second leachate and the residue would contain
only the second, relatively insoluble common Pb component. The two-point isochrons for
“residue-second wash” pairs for individual fractions would then give accurate and consistent
dates, even if the isotopic composition of the second common Pb component varies between the
fractions, and regressing all residue and second wash data together may not produce a valid
isochron. If our assumption is wrong, then the two-point isochron dates would be scattered. A
summary of two-point isochron dates:
4575
Hammadah al Hamra 237
Summary of two-point isochrons
"residue-second wash"
4565
4555
4562.8 ± 0.9 Ma
MSWD = 0.44, probability = 0.78
4545
4535
- Blue - bulk silicate, frac. 1-5
- Grey - bulk silicate, frac. 6, 7
- Red - individual chondrules,
fractions 8, 9
shows that all the two-point isochrons for “residue-second wash” pairs for fractions 1-5 give
consistent results. Their weighted average of 4562.8±0.9 Ma (MSWD=0.44) is our current best
estimate for the timing of formation of the HH 237 silicates. This result is consistent with the less
precise Pb-Pb isochron dates for the fraction 1-5 residues, and with the combined isochron date
for the fraction 1-5 residues (above) and second washes (4564.1±4.2 Ma, MSWD=7.5).
References
Ludwig, K. R. Isoplot/Ex version 3.00, A Geochronological Toolkit for Microsoft Excel,
Berkeley Geochronology Center Special Publ. 4, May 30, 2003 (2003).
Lugmair, G. W. & Galer, S. J. C. Age and isotopic relationships among the angrites Lewis Cliff
86010 and Angra dos Reis. Geochim. Cosmochim. Acta 56, 1673-1694 (1992).
Meibom, A. et al. Shocked effects in the metal-rich chondrites QUE 94411, Hammadah al
Hamra 237, and Bencubbin. Meteorit. Planet. Sci. 40, in press (2005).
Tatsumoto, M., Unruh, D. M. & Desborough, G. A. U-Th-Pb and Rb-Sr systematics of Allende
and U-Th-Pb systematics of Orgueil. Geochim. Cosmochim. Acta 40, 617-634 (1976).
Weisberg, M. K. & Kimura, M. Petrology and Raman spectroscopy of shocked phases in the
Gujba CB chondrite and the shock history of the CB parent body. Lunar Planet. Sci. 35,
#1559 (2004).
Supplementary Table 1. Representative compositions of olivines, low-Ca and high-Ca pyroxenes, and
mesostases in chondrules in Gujba.
ox\min
ol
ol
px
px
px
cpx
cpx
cpx
mes
mes
SiO2
41.5
41.6
57.6
54.4
47.9
51.2
48.4
45.4
62.5
52.2
TiO2
<0.09
<0.09
0.11
0.39
0.39
0.49
1.0
0.68
0.73
0.42
Al2O3
0.06
0.07
0.61
5.2
16.5
6.5
9.0
16.0
18.2
25.6
Cr2O3
FeO
MnO
MgO
CaO
0.22
1.0
<0.1
56.1
0.25
0.53
3.0
0.16
54.1
0.16
0.74
1.7
0.25
38.0
0.42
0.87
0.82
<0.1
35.1
2.5
0.79
0.78
<0.1
31.2
1.3
<0.08
1.2
<0.1
23.8
15.0
1.9
1.2
0.39
17.4
19.5
0.45
0.3
<0.1
15.0
21.8
0.28
1.4
<0.1
4.1
12.2
0.13
1.2
<0.1
4.2
14.2
Na2O
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
0.33
K2O
<0.03
<0.03
<0.03
<0.03
<0.03
0.06
<0.03
<0.03
<0.03 0.03
total
99.3
99.6
99.4
99.4
98.8
98.2
98.8
99.8
99.5
98.4
cpx = high-Ca pyroxene; mes = mesostasis; min = mineral; ol = olivine; ox = oxide; px = low-Ca pyroxene.