Angrite
(Basaltic/Quenched)
Found July 1979
37° 40' S., 61° 39' W.
This relatively fresh, 16.55 kg, shield-shaped, regmaglypted meteorite is by far the largest of the angrites found so far. The mass was found in Buenos Aires Province, Argentina by a farm worker who struck it with a plow. Thinking he had unearthed an Indian artifact, possibly an old mortar, he gave it to the landowner who set it by his house for the next ~20 years. Not until 1998, after reading an article on meteorites, did the owner seek to have the stone analyzed. In September of 2000, Dr. Gero Kurat of the Naturhistorisches Museum of Vienna, Austria made the determination that it was an angrite.
D'Orbigny is an unusual achondrite that shows no evidence of brecciation, shock metamorphism, or significant thermal metamorphism, and some believe that it may not have an igneous origin. It consists predominantly of Ca-bearing olivine and anorthite, which compose an intergrowth of plates and networks (Kurat et al., 2004). The other major constituent is Ti–Al-augite, which likely formed later than the olivine–anorthite intergrowths. A late oxidizing event produced strongly zoned grains of the clinopyroxene Ti–Al-hedenbergite, previously considered to be fassaite, which fill most of the intergranular spaces. Large (up to 1 cm) clear to milky, green to greenish-white, magnesian olivine megacrysts and polycrystalline olivinites represent one of the earliest phases of the host rock. Also representing very early constituents, Cr-bearing Al-spinel and Fe-bearing spinel occur within some olivine and anorthite grains. Minor kirschsteinite, ulvöspinel, and troilite (and other sulfides) are present, along with rare awaruite, calcium silico-phosphate (a late-stage crystallization phase similar to silicocarnotite which only crystallizes from Fe-, Ca-, and P-enriched basalt; Mikouchi et al, 2010), and an unidentified Fe–Al–Ti silicate. Rare, cm-sized, magnesian, Cr-rich olivine and spinel xenocrysts with granoblastic textures have been identified, similar to those found in greater abundance in other angrites (except Sah 99555). The xenocrysts are indicative of a rapid ascension of magma.
D'Orbigny has a heterogeneous composition consisting of alternating layers of a dense, coarse-grained texture, and a porous texture containing abundant round vugs or hollow shells up to 2.5 cm, along with plates and druses composed primarily of augite (diopside-hedenbergite) crystals and some anorthite crystals. Formation of these vesicles is consistent with bubble growth at CO2 concentrations of 10–20 ppm in a magma as it rises within a dike, undergoing decompression and coalescence of smaller bubbles, which then solidifies near the surface (McCoy et al., 2003). It was ascertained that the druse pyroxenes formed under oxidizing conditions near the QFM buffer, then underwent rapid cooling from ~1000°C, perhaps as a result of the dissipation of a hot vapor (Abdu et al., 2009).
Conversely, instead of formation of the vesicles through an igneous process, it was proposed by Kurat et al. (2002) that they were originally solid spheres composed of one of the earliest and most reduced phases, possibly CaS, which was covered by anorthite-olivine rims and plates. The cores were subsequently lost through an oxidizing Fe–Mn–Cr metasomatism process incorporating water, with the calcium being utilized in the formation of the augite, kirschsteinite, and hedenbergite. This druse formation scenario is consistent with a pneumatolytic formation process. Some of the vugs are now filled with glass.
Ubiquitous primary glasses present in D'Orbigny have unfractionated chondritic relative abundances of refractory lithophiles, indicating a possible origin through bulk rock melting, but excluding an origin as residues of a partial melt. Solar-like trapped noble gases present in these glasses are thought to have originated from primordial reservoirs of solar wind gases accumulated very early in Solar System history. These noble gases were subsequently implanted within the glasses by way of an ascending deep magma (Busemann et al., 2006). Vesicles that are present in several angrites attest to this rapid ascent and cooling of volatile-enriched magma. Schiller et al. (2010) argue that angrites experienced such volatile depletion associated with accretion within a short time after CAI formation, which was ~1 m.y. before volatile depletion occurred on the HED parent body. Two further episodes of volatile depletion for the angrite parent body are likely impact related. Formation of the glasses was contemporaneous with formation of the bulk of D'Orbigny.
An alternative formation mechanism for the glass phase has been proposed by Varela et al. (2003). Based upon the finding that some elemental abundances such as FeO and MnO, and some elemental ratios such as CaO/TiO and FeO/MnO, are similar to those in CI chondrites, and also that the glass shares many characteristics with glass inclusions in olivine grains from carbonaceous chondrites (e.g., the presence of volatile elements such as C and N, thought to have been incorporated as refractory material, which were subsequently volatilized through oxidation reactions and the depletion of volatile lithophile elements), they proposed that the glass crystals grew from a vapor phase (conceivably in the solar nebula) upon moist surfaces within interstitial spaces during olivine formation. Later, an oxidizing metasomatic alteration event that was intrinsically chondritic affected the glass and bulk rock.
Prior studies have shown that the close textural and compositional trends present in the angrites D'Orbigny, Sahara 99555, Asuka 881371, LEW 87051, NWA 1670, and possibly NWA 1296 provide evidence for their crystallization from a common magma source. It was suggested that this D'Orbigny group of angrites underwent rapid cooling and crystallization at depths of less than 0.5 m. However, since D'Orbigny contains no solar implanted gases, it could not have been exposed to the surface environment of the parent body. Angra dos Reis, LEW 86010, and NWA 2999 show evidence of a slower cooling history than the angrite grouping above, and they are probably not co-magmatic with them. Furthermore, precise U–Pb ages obtained for these three slowly cooled angrites indicate that they crystallized at least 0.9 m.y. apart, inferring an independent source magma for some or all (Amelin, 2007).
Trace element data argues for a more complex history for D'Orbigny and most angrites, including a non-igneous formation from a refractory condensate of a chondritic nature. Late phases of D'Orbigny are enriched in moderately volatile elements compared to early phases, and the two phases were formed under very different redox conditions—the early phases grew under highly reducing conditions while the late phases grew under highly oxidizing conditions (Varela et al., 2005). Moreover, highly incompatible elements in all olivine phases are far out of equilibrium (highly enriched) with the parental melt that formed the bulk rock. Curiously, plagioclase, which formed together with the olivine, contains very different abundances of incompatible elements, suggesting that the olivine and plagioclase formed from melts of dissimilar compositions. It is thought that the solid spheres (possibly CaS), enriched in Ca and unfractionated trace elements, were subsequently decomposed under the highly oxidizing conditions. By this process the previously bound trace elements were converted into a vapor phase and became available for late phase metasomatism and augite formation.
Angrites are extremely ancient meteorites, some with absolute ages as old as ~3–4 m.y. after the time when the first nebular condensates were formed (e.g., D'Orbigny). Other angrites demonstrate that basaltic extrusion on the angrite parent body continued for ~7 m.y. longer (e.g., LEW 86010 and AdoR). The early thermal history of the angrite parent body is most consistent with a relatively large sized planetesimal of at least 100 km in diameter (Sahijpal et al., 2007). One scenario for the formation of angrites involves an igneous history.
From formation models developed by Sahijpal et al. (2007), and from Rb–Sr and Hf–W systematics ascertained by Hans et al. (2009), it can be inferred that the angrite parent body experienced an early onset of accretion and volatile loss within ~2–3 m.y. of Solar System history, which then proceeded rapidly to completion over a timeframe of <10 t.y. Crystallization of the angrites proceeded as a two-stage process, beginning with partial melting from a CV-like chondritic source, composed of olivine, orthopyroxene, and clinopyroxene, at low pressures and elevated oxygen levels. Heat generated by the decay of short-lived radiogenic nuclei produced metal–silicate melting, differentiation, and basaltic melt extrusion within ~100 t.y. of the onset of accretion. This basalt was slowly cooled to ~650°C, while some of the melt experienced rapid quenching (7–13°C/hr) during eruption onto the surface or through a severe impact event.
Severe outgassing of volatiles occurred at this time, possibly hastened by the reduced strength of the gravitational field of the fragmented planetesimal.
A magnetic field with a strength ~20% that of present-day Earth was imparted to the angrite PB during its earliest phase of crystallization (as observed from D'Orbigny); this magnetic field may possibly be attributable to an orbital residence very near to the early T-Tauri phase solar field, or to an internal core-dynamo mechanism (Weiss et al., 2008).
Based on studies of how kirschsteinite-lamellae profiles relate to cooling rates, as well as results of crystallization experiments, the burial depth of the angrites as they were rapidly crystallized in a thin lava flow is inferred to have been within 1 m of the surface.
Another possible petrogenetic history involves a non-igneous formation:
Kurat et al. (2004) and Varela et al. (2005) have conducted extensive studies of D'Orbigny and other angrites in which they utilized mutiple sources of data (e.g., structural, textural, chemical, and redox evidence). They concluded that the angrites are most consistent with a non-igneous origin from refractory solar nebula condensates with chondritic abundances—basically an asteroid-sized version of a CAI—which record unusual circumstances (e.g., changing redox conditions) in the early history of the solar system.
A number of whole-rock and mineral isochrons have been calculated for the angrites. A U–Pb age of 4,555.3 (±1.7) m.y. was previously reported for D'Orbigny (Jotter et al., 2002), an age that is slightly younger than that determined for other angrites using this isotopic system (4,557.8 ±0.5 m.y. for LEW 86010 and AdoR). Other studies of matrix and druse pyroxenes from D'Orbigny have yielded a range of U–Pb ages between 4,549 (±2) m.y. and 4,563 (±1) m.y (Jagoutz et al., 2003), with a mean age of 4,563.9 (±0.6) m.y. (Zartman et al., 2006). A more precisely determined measurememnt of the Pb–Pb isotopic ages yielded an even older age for D'Orbigny of 4,564.42 (±0.12) m.y. (Amelin, 2007). These ages are consistent with the Pb–Pb age determined for the A-881371 angrite (4,562.4 [±1.6] m.y.). Recently, a more precisely determined 238U/235U ratio of 137.822 (۪.028) was brought to light by Brennecka et al. (2010) and substituted for the previously accepted ratio of 137.88; the Pb–Pb anchor age of D'Orbigny is now slightly younger at 4,563.8 (±0.5) m.y.
In addition, Glavin et al. (2004) calculated an absolute Mn–Cr isotopic age for D'Orbigny of 4,562.9 (±0.6) m.y., which is concordant with the Al–Mg age calculated by Spivak–Birndorf et al (2005, 2009) for both D'Orbigny and Sahara 99555, and is concordant with the calculated Mn–Cr ages and Hf–W ages for both angrites (Nyquist et al., 2003; Spivak–Birndorf et al, 2009). These extinct radionuclides provide ages that are slightly younger than the measured Pb–Pb ages.
The ~4,564 m.y. Pb–Pb age is ~7 m.y. older than some other angrites such as AdoR, LEW 86010 and NWA 2999 (~4,557.8, ~4,558, and ~4,557.9 m.y., respectively) and attests to very early accretion, igneous activity, differentiation, partial melting, and production of basaltic magma. Attainment of isotopic equilibrium and crystallization will have occurred very soon thereafter (Shukolyukov and Lugmair, 2007). The decay products of extinct radionuclides such as 53Mn, 146Sm, 244Pu, and 182Hf suggest that the entire sequence from nebular condensation through parent body accretion, partial melting, siderophile–lithophile element fractionation, multiple metasomatic alteration events, and final cooling to temperatures low enough to retain fission tracks and noble gases was on the order of only a few m.y.
Trace and major element compositions and textures of D'Orbigny and Sah 99555 are almost identical (Nyquist et al., 2003; Floss et al., 2003), suggesting a possible relationship. In addition, Asuka 881371 and LEW 87051 have trace element trends similar to D'Orbigny and Sah 99555, suggesting that they may all share a common origin, or at least experienced similar petrographic histories. Trace element trends for LEW 86010 and AdoR are significantly different from each other and from all the other angrites, and they represent distinct lithological sources.
To complicate matters, the results of CRE age studies (Eugster et al., 2002) utilizing cosmogenic nuclide data indicate that the CRE age of D'Orbigny (12.3 ±0.9 m.y.) is significantly different from that of other angrites studied: Sah 99555 (6.6 ±0.8 m.y.), Asuka 881371 (5.4 ±0.7 m.y.), Angra dos Reis (55.5 ±1.2 m.y.), LEW 86010 (17.6 ±1.0 m.y.), and LEW 87051 (~0.2 m.y.). All or most of these angrites represent unique ejection events on the angrite parent body, indicating that this object resides in a stable orbit (planetary or asteroid belt) that has permitted continuous sampling over at least the past 55 m.y.
Spectral data from studies of these new angrites, especially D'Orbigny, have yielded two possible spectral analogs among main-belt asteroids: the A-type 289 Nenetta and the Sr-type 3819 Robinson. Both show the strong spectral reddening characteristic of the Al–Ti-hedenbergite-rich angrites. However, important differences exist—the spectra of 289 Nenetta and 3819 Robinson contain distinct olivine bands that are absent in D'Orbigny, and the spectra of 3819 Robinson matches D'Orbigny only in the visible, not in the near-infrared.
It was inferred by Nyquist and Bogard (2003) that since D'Orbigny was spectroscopically similar to these two asteroids, which are located between ~2.8 and 2.9 AU, then it was also probable that the angrite parent body formed in this same region. They argued that asteroids at this heliocentric distance accreted too slowly to permit the accumulation of enough 26Al to cause global melting and differentiation before a diameter greater than ~200 km could be attained; i.e., given a body with a diameter larger than ~200 km, there would not be enough radiogenic heat produced to melt and differentiate this size of an object. By this line of reasoning, it may be concluded that the differentiated angrite PB was either not as large as 200 km in diameter, or that it formed at a smaller heliocentric distance than ~2.8 AU.
Without regard to heliocentric distance, Sanders and Scott (2007) argued that any body which accreted to a diameter >60 km (i.e., large enough to minimize heat loss from the surface through conduction) within ~2 m.y. of CAI formation (the oldest objects dating to 4.567 b.y. ago) as the angrites did, must contain enough 26Al to produce global melting and differentiation. In contrast, Senshu and Matsui (2007) determined that accretion to a diameter of only ~14 km occurring within 2 m.y. of CAI formation was all that was required for global differentiation to occur, while a diameter of 40–160 km occurring within 1.5 m.y. was cited by Hevey and Sanders (2006) and Sanders and Taylor (2005) as the minimums.
The spectrum of asteroid 3628 Boznemcová has been studied and compared to those of the angrite meteorites (Cloutis et al., 2006). Boznemcová is thought to have experienced partial melting and fractional crystallization under oxidizing conditions, and is considered to have a surface composition akin to an angritic crust; i.e., a composition of ~55–75 wt% clinopyroxene, ~20–33 wt% plagioclase feldspar, and 0–20 wt% olivine plus kirschteinite. It has a spectral type A (Fe³+)-free clinopyroxene phase known only from angrites. However, despite its similarities in reflectance spectra, and thus mineralogy, to that of angrite meteorites, the latter typically contain more olivine than is observed on Boznemcová. On the other hand, studies of the orbits of the LL6 ordinary chondrites Bensour and Kilabo by Alexeev et al. (2009) suggest that these meteorites cross the orbit of 3628 Boznemcova located at the inner asteroid belt (~2.2 AU) associated with two efficient resonances.
The angrites LEW 86010 (6.9 g), LEW 87051 (0.6 g), Asuka 881371 (11.2 g), Sah 99555 (2,710 g), D'Orbigny (16.55 kg), Angra dos Reis (~150 g of 1.5 kg preserved), NWA 1296 (810 g), NWA 1670 (29 g), NWA 2999 (392 g plus 928 g in pairings designated NWA 3164), NWA 4590 (212.8 g), and NWA 4801 (252 g) compose our limited sampling of the angrite parent body. Another recent find from Antarctica, Y-1154, is an anomalous Al–Ti-hedenbergite-rich meteorite compositionally similar to angrites, but with a unique fine-grained, dendritic texture.
A petrographic thin section micrograph of D'Orbigny can be seen on J. Kashuba's page. The specimen of D'Orbigny pictured above is a 1.6 g partial slice.
