Downes, Hilary and Abernethy, FAJ and Smith, CL and Ross, AJ and Verchovsky, AB and Grady ...

18 Pages·2015·0.43 MB·English

BIROn - Birkbeck Institutional Research Online
Downes, Hilary and Abernethy, F.A.J. and Smith, C.L. and Ross, A.J. and
Verchovsky, A.B. and Grady, M.M. and Jenniskens, P. and Shaddad, M.H.
(2015) Isotopic composition of carbon and nitrogen in ureilitic fragments of
the Almahata Sitta meteorite. Meteoritics & Planetary Science 50 (2), pp.
255-272. ISSN 1086-9379.
Downloaded from: bbk.ac.uk/id/eprint/11809/https://eprints.
Usage Guidelines:
Please refer to usage guidelines at bbk.ac.uk/policies.htmlhttps://eprints. or alternatively
contact [email protected].
M A P S 12413-2149 Dispatch:24.12.14 CE:Malarvizhi
Journal Code Manuscript No. No.ofpages:18 PE:Bercy
Meteoritics&PlanetaryScience1–18(2014)
1
doi:10.1111/maps.12413
2
3
4
5 Isotopic composition of carbon and nitrogen in ureilitic fragments of the Almahata
6
Sitta meteorite
7
8
9 H. DOWNES1,2,3*, F. A. J. ABERNETHY4, C. L. SMITH3, A. J. ROSS2,3,5, A. B. VERCHOVSKY4,
10 1 M. M. GRADY4, P. JENNISKENS6, and M. H. SHADDAD7
11
12 1Department ofEarthand PlanetarySciences, Birkbeck University ofLondon, Malet Street,London WC1E7HX, UK
13 2UCL/Birkbeck Centrefor PlanetarySciences, UCL, Gower Street,London WC1E6BT, UK
14 3Departmentof EarthSciences, NaturalHistory Museum, Cromwell Road,London SW75BD,UK
15 4Departmentof PhysicalSciences, TheOpen University,Walton Hall,MiltonKeynes MK76AA, UK
16 5Departmentof EarthSciences, UniversityCollege London, GowerStreet, London WC1E6BT,UK
17 6SETI Institute, CarlSagan Centre,189BernardoAvenue, Mountain View, California 94043,USA
18 7Departmentof Physics,University ofKhartoum, POBox321, Khartoum 11115,Sudan
19 *Corresponding author.E-mail:[email protected]
20
(Received 04April 2014;revision accepted13November 2014)
21
22
Abstract–This study characterizes carbon and nitrogen abundances and isotopic
23
compositions in ureilitic fragments of Almahata Sitta. Ureilites are carbon-rich (containing
24
up to 7 wt% C) and were formed early in solar system history, thus the origin of carbon in
25
ureilites has significance for the origin of solar system carbon. These samples were collected
26
soon after they fell, so they among the freshest ureilite samples available and were analyzed
27 using stepped combustion mass spectrometry. They contained 1.2–2.3 wt% carbon; most
28 showed the major carbon release at temperatures of 600–700 °C with peak values of d13C
29 from (cid:1)7.3 to +0.4&, similar to literature values for unbrecciated (“monomict”) ureilites.
30 They also contained a minor low temperature (≤500 °C) component (d13C = ca (cid:1)25&).
31 Bulk nitrogen contents (9.4–27 ppm) resemble those of unbrecciated ureilites, with major
32 releases mostly occurring at 600–750 °C. A significant lower temperature release of nitrogen
33 occurred in all samples. Main release d15N values of (cid:1)53 to (cid:1)94& fall within the range
34
reported for diamond separates and acid leached material from ureilites, and identify an
35
isotopically primordial nitrogen component. However, they differ from common polymict
36
ureilites which are more nitrogen-rich and isotopically heavier. Thus, although the parent
37
asteroid 2008TC was undoubtedly a polymict ureilite breccia, this cannot be deduced from
38 3
an isotopic study of individual ureilite fragments. The combined main release d13C and d15N
39
values do not overlap the fields for carbonaceous or enstatite chondrites, suggesting that
40
carbon in ureilites was not derived from these sources.
41
42
43
44 INTRODUCTION Nakamuta 2005). Carbon occurs in ureilites as graphite, 2
45 diamond and amorphous carbon (Vdovykin 1970;
46 Ureilites are ultramafic achondrite meteorites that Berkley et al. 1976; Mittlefehldt et al. 1998; Hezel et al.
47 are considered to represent the mantle residue of a 2008; Le Guillou et al. 2010; Ross et al. 2011a), with 3
48 single parent asteroid body (Takeda 1987; Scott et al. other rare carbon polymorphs (El Goresy et al. 2004;
49 1993; Goodrich et al. 2004; Downes et al. 2008; Herrin Ferroir et al. 2010; Kaliwoda et al. 2011) and carbide
50 et al. 2010). As such they can be considered as phases, e.g., cohenite (Goodrich and Berkley 1986;
51 analogues of the early mantle of terrestrial planets. Warren and Kallemeyn 1994; Goodrich et al. 2013).
52 Ureilites are the most carbon-rich of all meteorite types, Other compounds, such as carbonate weathering
53 typically containing 2–7 wt% C (Mittlefehldt et al. products (Grady et al. 1985) are likely to represent
54 1998; Smith et al. 2001a; Grady and Wright 2003; terrestrial contamination, whereas organic molecules
1 ©TheMeteoriticalSociety,2014.
2 H. Downes et al.
1 such as those found by Glavin et al. (2010), Sabbah graphite and diamond in ureilites, with diamond having
2 et al. (2010) and Burton et al. (2011) are thought to be values of (cid:1)100& and graphite having values of +19&.
3 extraterrestrial but unlikely to be derived from the Nitrogen in polymict ureilites differs from that in
4 ureilitic precursor. unbrecciated (monomict) ureilites but the paucity of
5 Despite occasionally occurring in veins (e.g., Corder nitrogen isotope data available for polymict ureilites
6 et al. 2014), ureilitic carbon is usually intergranular and means that only the studies of Grady and Pillinger
7 found on grain boundaries. It is considered to be a (1988), Rooke et al. (1998) and Rai et al. (2003) are
8 primary constituent (Berkley and Jones 1982; Rubin directly comparable to our investigation. In polymict
9 1988) rather than having been introduced late in the ureilites, the main nitrogen release occurs below
10 formation history of the meteorites. Carbon abundances ~600 °C and is isotopically heavy with d15N up to
11 are commonly higher in ureilites with low core olivine +530&.
12 Fo contents (Nakamuta 2005). Diamond in ureilites is The number of known ureilites has increased greatly
13 usually considered to have formed via shock in recent years because of numerous finds in both hot
14 metamorphism from graphite (Lipschutz 1964; Bischoff and cold deserts. In particular, the 2008 fall of the
15 et al. 1999; Grund and Bischoff 1999; Nakamuta and polymict ureilite Almahata Sitta (Jenniskens et al. 2009;
16 Aoki 2000; El Goresy et al. 2004; Fisenko et al. 2004; Horstmann and Bischoff 2014) yielded around 600
17 Hezel et al. 2008; Ross et al. 2011a). However, some additional individual stones. These samples are very
18 studies have suggested that ureilitic diamonds were fresh, thereby minimizing the potential problems with
19 formed by other processes such as chemical vapor terrestrial weathering reported by Ash and Pillinger
20 deposition (e.g., Fukunaga et al. 1987; Nagashima et al. (1995). The aim of the current study is to characterize
21 2012; Miyahara et al. 2013). the carbon and nitrogen isotopic values in several
22 The first systematic study of carbon isotopes in ureilitic fragments of Almahata Sitta using a stepped
23 ureilites was performed by Grady et al. (1985), using a combustion technique that permits simultaneous
24 stepped combustion technique on 14 of the 20 ureilites analysis of d13C and d15N values. Stepped combustion
25 then known, including the only two polymict ureilites mass spectrometry is a powerful tool for investigating
26 known at that time, Nilpena and North Haig. They the isotopic composition of carbonaceous components
27 showed that there is a window between 500 and 800 °C within meteorites (Swart et al. 1983; Wright et al. 1983;
28 within which graphite/diamond combustion occurs, and Pillinger 1984). Stepped heating allows identification of
29 suggested that diamond and graphite in ureilites have different carbon phases, and their isotopic
30 the same d13C values, but showed that d13C values compositions, by their characteristic combustion
31 clearly vary between different ureilites. Later work by temperatures (Smith et al. 2001a; Grady et al. 2004).
32 Smith et al. (1999, 2001a) showed a somewhat greater However, Smith et al. (2001b) have shown that
33 range of d13C and carbon contents for a wider variety combustion temperatures of graphite depend to some
34 of ureilite samples. Carbon in ureilites generally has degree on grain size and crystallinity. Our new data will
35 bulk d13C values of (cid:1)11& to 0& (Grady et al. 1985; be compared with literature data for unbrecciated and
36 Mittlefehldt et al. 1998; Smith et al. 2001a; Grady and polymict ureilites in order to assess whether the carbon
37 Wright 2003; Hudon et al. 2004). However, d13C values and nitrogen components in Almahata Sitta samples are
38 derived by bulk combustion techniques could include a similar to, or different from, previously analyzed
39 contribution from material released at low ureilites. We also aim to determine whether ureilitic
40 temperatures. Consequently, we have applied a stepped carbon was derived from carbonaceous chondrites, as is
41 heating technique to the Almahata Sitta samples. commonly assumed (e.g., Rubin 1988), and to
42 Nitrogen is present in ureilites in much lower investigate the variation in d13C and d15N in the early
43 abundances than carbon (ca 10–150 ppm), but is mostly solar system.
44 associated with the carbon phases. Rai et al. (2003)
45 suggested that diamond is the major carrier of nitrogen. SAMPLES
46 Nitrogen is far more complex than carbon in terms of
47 its behavior during stepped combustion. The few The meteorites collectively known as Almahata
48 nitrogen isotope analyses that have been made of Sitta fell in the Nubian desert of northern Sudan on
49 ureilites (Grady et al. 1985; Grady and Pillinger 1988; October 7 2008 (Jenniskens et al. 2009; Jenniskens and
50 Russell et al. 1993; Yamamoto et al. 1998; Rai et al. Shaddad 2010). The rapid collection of the Almahata
51 2002, 2003; Fisenko et al. 2004) have shown that there Sitta stones means that they can be considered as very
52 are multiple nitrogen isotopic components within these fresh samples, having undergone minimal terrestrial
53 meteorites. Rai et al. (2002, 2003) showed that, in weathering through processes such as rainfall. The
54 contrast to carbon isotopes, d15N values differ between samples studied in this paper were all recovered during
Carbon and nitrogen in Almahata Sitta 3
1 the first two collecting expeditions in December 2008— et al. 2010b; Kwok 2009; Jenniskens et al. 2010a;
2 less than 3 months after asteroid 2008TC collided with Shaddad et al. 2010). Thus, it is not known to what 4
3
3 the Earth (Jenniskens et al. 2009; Shaddad et al. 2010). extent the collected material is representative of the
4 Almahata Sitta has been described as an anomalous original asteroid.
5 polymict (i.e., brecciated) ureilite (Jenniskens et al. Five AlmahataSittastones wereselected foranalysis
6 2009). Although not all Almahata Sitta samples have (AS7, AS22, AS27, AS36, and AS44). Samples were
7 yet been studied, Horstmann and Bischoff (2014) primarily chosen based on availability of material for
8 collated the information available for 110 stones, of destructive analysis, but also cover the range of
9 which 75 are ureilites and 35 are chondrites. This heterogeneities found within Almahata Sitta. Apart
10 confirms the conclusions of initial mineralogical studies from AS22, the stones studied here have already
11 that Almahata Sitta is predominantly ureilitic (Bischoff been investigated by various authors using different
12 et al. 2010; Zolensky et al. 2010); ureilites make up techniques. AS7 represents the unusual fine-grained
13 68% of the samples and 73% of the total mass of the porous lithology and is the same sample initially
14 stones studied. Of the Almahata Sitta ureilites, just over described from the Almahata Sitta fall and identified as
15 half (38 samples, ~65% of the ureilite mass (Horstmann an anomalous ureilite (Jenniskens et al. 2009). The rest
16 and Bischoff 2014)) are coarse-grained akin to the main are all coarse-grained compact main group ureilites.
17 group of unbrecciated coarse-grained olivine-pyroxene AS22 and AS427 were large stones (115.32 g and
18 ureilites that have long been dominant in ureilite 283.84 g, respectively), whereas AS7 (1.52 g) and AS44
19 collections (Goodrich 1992; Mittlefehldt et al. 1998; (2.21 g) were much smaller (Shaddad et al. 2010). Three
20 Goodrich et al. 2004). However, the first sample of of the samples (AS7, AS36, and AS44) have been
21 Almahata Sitta analyzed (AS7, also studied here) described by Zolensky et al. (2010), while samples AS7,
22 represented a new fine-grained, porous lithology AS27, AS36, and AS44 were analyzed by Sandford et al.
23 (Jenniskens et al. 2009). An additional 18 samples (2010). AS27 has also been investigated by Warren and
24 (making up 25% of ureilite samples but only 9% of the Rubin (2010) and Burton et al. (2011) and AS44 was
25 ureilite mass) are similarly anomalous ureilites (Bischoff described by Goodrich et al. (2010a). Samples AS36 and
26 et al. 2010; Warren and Rubin 2010; Zolensky et al. AS44 were analyzed by Murty et al. (2010) for nitrogen
27 2010; Hutchins and Agee 2012; Horstmann and and noble gases. Among our samples, the presence of
28 Bischoff 2014). diamond has been confirmed by Raman spectroscopy in
29 Four Almahata Sitta chondrites have been AS7 (Ross et al. 2011a) and has also been identified in
30 analyzed for cosmogenic radionuclides (Bischoff et al. AS22,AS27,andAS44.
31 2010; Meier et al. 2012), three of which are within The forsterite (Fo) content of olivine cores in
32 error of values from the Almahata Sitta ureilites previously studied ureilites ranges from ~75 to 96;
33 (Welten et al. 2010), indicating that they are all part however, they are not evenly distributed, showing a
34 of the same fall. The abundance of non-ureilitic significant peak at Fo~78–79 with a secondary lesser
35 material has been interpreted to indicate that peak at Fo~90–92 (Goodrich et al. 2004; Downes et al.
36 Almahata Sitta was formed from an asteroid that was 2008). Reported Fo values for Almahata Sitta ureilites
37 dominated by ureilitic material but on which a large cover this entire range from Fo~76 to Fo~95 (Bischoff
38 proportion of non-indigenous debris had also collected et al. 2010; Warren and Rubin 2010; Kita et al. 2011;
39 (Horstmann and Bischoff 2014). Almahata Sitta Horstmann and Bischoff 2014). Even within this small
40 provides important evidence that the source of ureilite sample set, there is a major peak at ~Fo79. Oxygen
41 meteorites is a brecciated impact-disrupted asteroid isotope analyses of Almahata Sitta ureilites (Bischoff
42 (Goodrich et al. 2004; Warren and Huber 2006; et al. 2010; Rumble et al. 2010; Kita et al. 2011;
43 Downes et al. 2008; Borovi(cid:1)cka and Charv(cid:3)at 2009; Horstmann et al. 2012) also cover the range of values
44 Herrin et al. 2010). previously found in ureilites (Clayton and Mayeda 1988,
45 Estimates of the mass of the 2008TC asteroid 1996; Kita et al. 2004). These data are indicative of the
3
46 before impact vary from 8 t (Kohout et al. 2011) to 108 heterogeneity within the original 2008TC asteroid. Our
3
47 t (Jenniskens et al. 2009). The combined mass of all studied samples range in Fo values from 79 (AS44) to
48 Almahata Sitta stones collected to date is approximately 90 (AS36) and their D17O values range from (cid:1)0.4 to
49 11 kg (Horstmann and Bischoff 2014). Therefore, no (cid:1)1.5& (Rumble et al. 2010).
50 matter which calculation of pre-impact mass is used, Polished blocks were prepared for petrographic
51 more than 99% of the original asteroid was lost during study. AS22 and AS27 were split into petrographic and
52 the explosions observed as it entered the Earth’s isotopic sub-samples, with AS36 being already
53 atmosphere (Borovi(cid:1)cka and Charv(cid:3)at 2009; Jenniskens completely disaggregated and requiring no further
54 et al. 2009; Jenniskens and Shaddad 2010; Jenniskens subdivision. AS7 was provided as two separate pieces,
4 H. Downes et al.
1 Table 1. Summary data of C and N in Almahata Sitta samples (see Appendix for full data set). For AS22 N%
2 yield in brackets for main release is value without the release at 1000 °C. 5
3
AS7 AS22 AS27 AS36 AS44
4
Sample mass(mg) 4.736 5.129 5.466 4.41 4.948
5
Focontent in olivine 86 80 85 90 79
6
px*: px/(ol+px) % 0.31 0.28 0.22 0.35 0.15–0.23
7
TotalC(wt%) 1.57 1.18 1.20 2.30 1.88
8
Carbon
9 Averagetotald13C (&) 0.0 (cid:1)7.6 (cid:1)2.8 (cid:1)5.4 0.1
10 Averageerror(&) 0.3 0.3 0.3 0.3 0.3
11 Mainrelease range (° C) 500–700 500–700 500–700 500–700 500–700
12 Cin mainrelease (ng) 70895 58566 58409 91630 84041
13 Cin mainrelease (wt%) 1.49 1.14 1.07 2.06 1.70
14 %of totalCin mainrelease 94.7 96.6 89.3 89.6 90.2
15 Weighted avd13Cin mainrelease & 0.4 (cid:1)7.3 (cid:1)2.3 (cid:1)5.2 0.4
Error(&) 0.3 0.3 0.3 0.3 0.3
16
Nitrogen
17
TotalN (ppm) 18.11 27.11 21.19 9.35 11.78
18
Averagetotald15N (&) (cid:1)48.2 (cid:1)29.4 (cid:1)7.5 (cid:1)24.1 (cid:1)52.1
19
Averageerror(&) 0.9 4.5 4.3 2.2 2.4
20 Mainrelease range (° C) 600–700 600–700 650–700 650–700 600–700
21 N in mainrelease (ng) 54.63 36.69 34.11 12.41 37.19
22 N in mainrelease (ppm) 11.47 7.15 6.24 2.79 7.52
23 %of Nin mainrelease 63.3 26.4 (60.5) 29.5 29.9 63.8
24 Weighted avd15Nin mainrelease & (cid:1)91.0 (cid:1)69.0 (cid:1)61.3 (cid:1)95.6 (cid:1)94.7
25 Error(&) 0.3 4.8 5.3 1.0 1.2
26
27
28 one with fusion crust and one without; the former was AS7 shows a distinctive porous texture. Olivine
29 made into the polished block and the latter used for core compositions are Fo . This sample was analyzed
86
30 isotope analyses. AS44 was provided as two distinct by micro Raman techniques (Ross et al. 2011a), which
31 samples, but petrographic and mineralogical analysis of revealed the presence of diamond as well as graphite
32 the polished blocks showed them to have identical and amorphous carbon. AS22 has not previously been
33 features and so only one was sub-sampled for isotopic described. It shows a typical coarse-grained (up to
34 analysis. 2 mm) olivine-dominated ureilitic texture with cores of
35 The polished blocks of samples studied here are all Fo (Ross et al. 2011b). Sample AS27 consists of
80
36 olivine-dominated (Table 1). When compared with the olivine (Fo ) and low-Ca pyroxene. Warren and
85
37 results from infrared spectroscopy by Sandford et al. Rubin (2010) suggested that AS27 may have
38 (2010), the px* values (where px* = modal px/(modal experienced some degree of shock, resulting in
39 px + modal ol)) for AS27 are almost identical and our undulose extinction in olivine, and that its original
40 results for AS36 are very close to one of three analyses texture may have been poikilitic. Our sample of AS36
41 splits of AS36. However, our sample of AS44 is was completely disaggregated into individual crystal
42 considerably more olivine-rich and our sample of AS7 is grains with a small proportion of mixed grains, and
43 marginally more olivine-rich. All samples contain therefore it tends to lack interstitial material. It is
44 carbon, with one of the AS44 blocks appearing to composed largely of coarse-grained (up to 3 mm)
45 contain a vein-like carbon-rich area. The samples also olivine Fo crystals (Ross et al. 2011c). AS44 is a
90
46 contain metal and sulfide; grain-boundary Fe-metal is coarse-grained (>2 mm) olivine-dominated ureilite
47 the most common form, which can contain minor (Fo ) with some pigeonite. Its texture resembles those
79
48 amounts of Ni and Si and trace amounts of Co and P. of impact-smelted ureilites (Goodrich et al. 2010a; Ross
49 AS27 and AS44 both contain carbon-bearing metal et al. 2011b).
50 phases. In AS27 these are included within silicate grains
51 as “spherules”. In AS44, the C-bearing metal is present METHODOLOGY
52 in the grain boundaries alongside Fe-metal,
53 schreibersite, and sulfide (Goodrich et al. 2010a; Almahata Sitta samples were analyzed for C and
54 Mikouchi et al. 2011; Agoyagi et al. 2013). N abundance and isotopic compositions using a
Carbon and nitrogen in Almahata Sitta 5
1 stepped combustion technique on the FINESSE static RESULTS
2 vacuum mass spectrometer at the Open University
3 (UK). For detailed information regarding the Carbon
4 components and processes within the FINESSE
5 instrument, see Verchovsky et al. (1998, 2002). Care In Table 1 we report the average d13C values of the
6 was taken to ensure that no sub-samples used for carbon across all temperature ranges (equivalent to the
7 isotopic analyses had fusion crust attached. The bulk carbon composition), and the weighted average
8 meteorite chips were hand crushed to a powder in a d13C of the carbon released during the temperature
9 class 10000 clean room, using an agate pestle and steps where >~90% of carbon was liberated (referred to
10 mortar. This was transferred to a class 100 clean as the “main release carbon”). The main carbon release
11 room where approximately 5 mg of the powdered temperature range is the same in all the samples,
12 ureilites were weighed into 0.25 lm thick Pt buckets occurring between 500 and 700 °C. The total amount of
13 that had been pre-cleaned by heating to 900 °C in the carbon released in all of the samples was 1.18 to 2.3 wt
14 presence of CuO in a sealed quartz tube pumped to %. The d13C value of the main carbon release is always
15 10(cid:1)7 mbar. heavier than that of the total C (Table 1), because of
16 After being loaded in the mass spectrometer the contribution from an isotopically light, low-
17 extraction system, the samples were heated temperature component. Differences in d13C between
18 incrementally from 200 to 1400 °C in the presence of the main release component and the average
19 oxygen derived from thermal decomposition, at 930 °C, d13C across all temperatures (i.e., including the
20 of CuO present in a separately heated unit with an inlet low-temperature component) range from 0.2 to 0.5&.
21 into the furnace, resulting in the liberation of individual The five Almahata Sitta ureilite samples have very
22 components. Carbon (in the form of CO ) and similar carbon release patterns (Fig. 1). During the steps
2
23 molecular nitrogen were cryogenically separated from at <500 °C, less than 2% of the total carbon was
24 each other before analysis. Simultaneous analysis was released and the d13C values were typically around
25 made possible by using multiple mass spectrometers (cid:1)25&. As combustion reached about 600 °C, d13C
26 connected to a common extraction line: two magnetic showed a rapid increase to “peak” (main release) values
27 sector mass spectrometers for determination of carbon ranging from (cid:1)7.3& (AS22) to +0.4& (AS44).
28 isotopes and nitrogen abundance, and a quadrupole Following the main release of carbon, the samples
29 mass spectrometer for nitrogen isotopes. Abundance behaved in slightly different ways. Samples AS7 and
30 yields are recorded in nanograms with errors of <1% in AS22 released less than ~4% of their total carbon above
31 the carbon measurement, made on a capacitance 750 °C. The carbon stepped combustion plots (Fig. 1)
32 manometer, and <5% in the nitrogen measurements, show that these samples have a very subtle second minor
33 which were determined from the previously calibrated release, at ~900 °C (AS7) and ~1000 °C (AS27). In AS7,
34 relationship between m/z = 28 intensity and the the second release is almost identical in isotopic
35 nitrogen volume. The system blank levels of the composition to the major release at 650 °C. Thus, the
36 instrument are <10 ng and <1 ng for carbon and second release probably represents combustion either of
37 nitrogen, respectively. Isotopic compositions are higher crystallinity graphite or of genetically related
38 reported in the delta (d) notation as a deviation in diamond. The latter is more likely as this sample
39 parts per thousand (&) from the Peedee Belemnite definitely contains diamond (Zolensky et al. 2010; Ross
40 (vPDB) standard for carbon and from terrestrial air for et al. 2011a). Samples AS27, AS36, and AS44 released a
41 nitrogen. Isotopic reproducibility is typically <1&. As a slightly larger amount of carbon (~10%) at temperatures
42 result of the high yields normally found in ureilites >750 °C; the isotopic composition of these releases
43 (Grady et al. 1985; Smith 2002), blank correction was ranges from +1.5& (AS44) to (cid:1)5.7& (AS36). Two of
44 based on the average yield and isotopic composition of these samples (AS27 and AS44) contain C-rich metal
45 a system blank run prior to the experiment. Total which may account for these high-temperature releases.
46 isotopic error was calculated through an error The extreme change in isotopic composition of the
47 propagation calculation, based on the contribution of highest temperature release of carbon from AS44 is
48 the blank to the total yield. Summary results for almost certainly an artifact of the method (possibly
49 carbon and nitrogen concentrations and isotopic furnace residue).
50 compositions in the Almahata Sitta samples are given
51 in Table 1, and full results for each temperature step Nitrogen
52 can be found in the Appendix. No N analyses are
53 available for the highest temperature step for AS7 and Table 1 shows the d15N results for the five
54 AS36 because of system failure. Almahata Sitta samples; step combustion profiles are
6 H. Downes et al.
1
G
2
FI
3
N
4 O
5 TI
U
6 L
O
7 S
E
8
R
9 W
10 O
L
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53 Fig. 1. Carbon release patterns (solid line) and d13C compositions (points with error bars) of five samples of Almahata Sitta 8
54 ureilitesduring stepped heatingcombustion.
Carbon and nitrogen in Almahata Sitta 7
1 shown in Fig. 2. Both the nitrogen abundance whereas for AS7, the high-temperature component had
2 combustion profiles and corresponding nitrogen isotopic a much less variable isotopic composition.
3 values show more complexity than for carbon. The release patterns of carbon and nitrogen suggest
4 Nevertheless, all of the samples show a clear main that the two elements are clearly connected, as they are
5 release of nitrogen at around 600–700 °C, over which released at similar temperatures (Figs. 1 and 2),
6 the d15N values decrease significantly. It is difficult to although the main release of nitrogen starts at a slightly
7 define the main release peaks because of the presence of higher temperature than that of carbon. This may be
8 lower temperature releases, but in general these main because the major carrier of nitrogen is diamond which
9 releases yield average d15N values from (cid:1)61& (AS27) combusts at a slightly higher temperature than graphite.
10 to (cid:1)95.6& (AS36). In sample AS22, a second large There is a negative correlation between the total
11 release peak is seen at a much higher temperature amount of carbon released and the total amount of
12 (1000 °C) which lacks these low d15N values. nitrogen released during combustion of the Almahata
13 In contrast to the carbon release profiles, at Sitta samples (Fig. 3a). Sample AS22 represents the
14 temperatures below 400°C, up to ~20% of nitrogen was sample with the highest nitrogen and lowest carbon
15 released from all of our samples with a weighted abundances (although this value includes the anomalous
16 average d15N of +8 to +19&. Between 400 °C and release of nitrogen at high temperature), whereas sample
17 600 °C, samples AS7, AS27 and AS36 released 20.8 to AS36 represents a low nitrogen, high carbon component
18 34.5% of total N, with a weighted average d15N of +29 (this sample was disaggregated and contains very little
19 to +42&. In contrast, samples AS22 and AS44 released interstitial material). A stronger correlation is seen
20 less nitrogen (8.7% and 14.0%, respectively) over this between the total amount of N and the ratio of N/C
21 temperature range, and this nitrogen was significantly released from each sample (Fig. 3b). The correlation
22 heavier than that seen in the other three samples (+64& supports the contention that two separate components
23 and +82&). Sample AS27 was investigated by Burton dominate the nitrogen isotopes. The nature of these
24 et al. (2011) who identified the presence of components is discussed below.
25 extraterrestrial amino acids. This component may be
26 responsible for the low temperature, heavy nitrogen DISCUSSION
27 release.
28 At temperatures between 600 °C and 900 °C, Comparison of Carbon and Nitrogen With Ureilite
29 samples AS7 and AS44 released most (~70%) of their Literature Data
30 nitrogen. The isotopic composition of this release was
31 the lightest of the analyzed samples in this study, with The Almahata Sitta ureilite samples show a
32 weighted average d15N values of (cid:1)84& (AS7) and ubiquitous carbon component at <500 °C which is a
33 (cid:1)91& (AS44). Samples AS27 and AS36 also released very small (<2%) proportion of the total carbon yield
34 most of their nitrogen over this temperature range, and has a d13C value of around (cid:1)25 to (cid:1)30& (Fig. 1).
35 however, it was a much smaller release in terms of total The low-temperature combustion behavior and isotopic
36 nitrogen (~43%). Its isotopic composition was slightly composition of this carbon is suggestive either of
37 heavier than that seen in AS7 and AS44, with weighted terrestrial organic carbon (Hoefs 1997) or of extra-
38 average d15N values of (cid:1)40 and (cid:1)79&. terrestrial, but non-ureilitic, organic components as
39 AS22 is unique among the samples in that the described by Glavin et al. (2010) and Burton et al.
40 600 °C to 900 °C release only made up 27.8% of total (2011). While these are very minor components in
41 nitrogen, with a d15N of (cid:1)63&. Its largest release Almahata Sitta samples, these low-temperature
42 (56.4% of total N) occurred at 1000 °C with a components could bias the carbon isotopic value
43 corresponding d15N of (cid:1)29&. An extremely similar determined from a bulk (i.e., non-stepped combustion)
44 major release of nitrogen at 1000 °C was reported from analysis, particularly if the component is terrestrial in
45 Goalpara by Grady et al. (1985). This may suggest an origin and the sample has been in the terrestrial
46 additional, as yet unidentified, component of nitrogen in environment for a long time.
47 some ureilites, such as a nitride phase similar to those Data from three older ureilite falls (Dyalpur, Novo-
48 found in E chondrites. At temperatures >900 °C the Urei, and Havero€) show a range between 3.6 and 4.7%
49 other Almahata Sitta ureilite samples released <12% of of total carbon released below 500 °C (Grady et al.
50 their total nitrogen with weighted average d15N values 1985), so the Almahata Sitta samples have almost half
51 in the range (cid:1)32 to (cid:1)3&. In AS27 and AS36, the the abundance of contaminating carbon compared with
52 nitrogen isotopic composition changed from heavier other falls. For ureilite finds, however, the low-
53 values to lighter values with increasing temperature. In temperature component ranges from 1.7% (Y790981) to
54 sample AS44, the opposite pattern was observed, 23.7% (Goalpara) of total carbon (Grady et al. 1985).
8 H. Downes et al.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
Fig. 2. Nitrogen release patterns (solid line) and nitrogen isotope compositions (points with error bars) of five samples of
54 Almahata Sittaureilites.
Carbon and nitrogen in Almahata Sitta 9
1 composition (Grady et al. 1985; Smith et al. 2001a;
2 Hudon et al. 2004) and main release d13C values of the
3 Almahata Sitta samples (+0.4 to (cid:1)7.3&) fall well within
4 these ranges. Grady et al. (1985) suggested that ureilites
5 show a bimodality of peak d13C values, with clusters
6 around (cid:1)2& and (cid:1)10&. It appeared that the
7 isotopically lighter values were associated with ureilites
8 that had more Mg-rich olivine core compositions
9 (>Fo ), whereas the isotopically heavier values
83
10 belonged to samples with more Fe-rich (<Fo ) olivine
80
11 cores (Fig. 4). However, the ureilites studied by Grady
12 et al. (1985) cluster around only two different Fo
13 compositions, although this bias results from the
14 bimodal distribution of core Fo values in all ureilites
15 (Goodrich et al. 2004; Downes et al. 2008). In a later
16 study, Hudon et al. (2004) showed a more extensive
17 data set for both Fo (74–95) and bulk d13C ((cid:1)11 to
18 +1&) values that also shows a negative correlation
19 (Fig. 4). The reason for this correlation is not known,
20 although it might relate to mixing of different nebula
21 components in the source of the original UPB. Unlike
22 the data of Grady et al. (1985) and Hudon et al. (2004),
23 our Almahata Sitta ureilite data show no correlation
24 between olivine core Fo composition and d13C values.
25 However, the small number of data points in our study
26 precludes making any conclusion from this observation
27 and, in any case, our data plot in or very near to the
28 field defined by the data of Hudon et al. (2004).
29 Nitrogen is a trace element in unbrecciated ureilites
30 (3–55 ppm according to Grady and Wright (2003) and
31 Rai et al. (2003)). Our Almahata Sitta ureilite samples
32 yield total nitrogen abundances in the range 9.4–
33 27 ppm, well within this reported range. There are fewer
34 measurements of nitrogen within polymict ureilites but
35 nitrogen yields are generally higher, up to 152 ppm
36 9 Fig. 3. a) Total amount of C versus total amount of N (Grady and Pillinger 1988; Rooke et al. 1998; Rai et al.
37 released during stepped combustion of five Almahata Sitta 2003). Previous stepped combustion studies have
38 ureilites. b) Total amount of N released versus total N/total C suggested multiple components with distinct d15N values
39 released during stepped combustion of five Almahata Sitta with some differences in nitrogen composition
ureilites. N.B. From a mathematical point of view, one of
40 and combustion behavior between unbrecciated and
thesetwocorrelations cannotbeastraightline relationship.
41 polymict ureilites (Grady et al. 1985; Grady and
42 Pillinger 1988; Rooke et al. 1998; Yamamoto et al.
43 The low-temperature component is also significant in 1998; Rai et al. 2003). Grady et al. (1985) measured the
44 two polymict ureilite samples, North Haig (37.1%) and abundance and isotopic composition of nitrogen in five
45 Nilpena (8.2%). Excluding the anomalously high ureilite specimens using stepped combustion. Their
46 Goalpara value, and the polymict samples, the low- results showed multiple nitrogen components with
47 temperature carbon released by unbrecciated ureilite different isotopic compositions, released at different
48 finds ranges from 1.7% to 6.5%. Thus, in terms of temperatures. However, the lightest nitrogen they
49 carbon, the Almahata Sitta samples are among the least measured had a d15N value of only (cid:1)62.4& (the 800 °C
50 contaminated ureilites yet studied. step in Kenna) and bulk isotopic compositions d15N
51 The main carbon components released at higher values ranged from (cid:1)24.8& (Kenna) to +26.2& in
52 temperatures are considered to be indigenous ureilitic polymict sample North Haig. For unbrecciated
53 carbon. Previously analyzed unbrecciated ureilites all (“monomict”) ureilites, Yamamoto et al. (1998) and Rai
54 have similar ranges in carbon abundance and isotopic et al. (2003) described the main nitrogen release as