The Formation of Pallasites
Total Page:16
File Type:pdf, Size:1020Kb
40th Lunar and Planetary Science Conference (2009) 1042.pdf THE FORMATION OF PALLASITES. Z.A. Lavrentjeva. Institute of Geochemistry and Analytical Chemistry, Russian Academy of Sciences, Moscow 119991 Russia; e-mail: [email protected] REEs of olivine determined by Beattie [16] are Introduction. Observations of remote stellar compared with a REE abundance pattern systems have reached a point where meaningful assumed for the pallasite olivine determined by comparisons can be made between these Minova and Ebihara [17]. Pallasite olivine astronomical observations, astrophysical models, cannot be produced by a single stage and meteorite measurements. Among the differentiation from the starting material with an parameters that can be compared are the unfractionated REE abundance pattern. The timescales of formation of protoplanetary disks simplest model [17] to explain the observed REE and subsequent condensation, aggregation, and pattern in pallasite olivine needs two steps; the differentiation of solids. light REE-enriched phase is removed at first and Discussion. Pallasites are highly the olivine crystal is formed as a cumulate later. differentiated meteorites with two major phases, Phosphoran olivine is rare in nature, but has been olivine and Fe-Ni metal (e.g., [1]). Pallasites are found in four Main Group pallasites: classified into two groups, Main Group (MG) Springwater, Brahin, Rawlinna and Zaisho . and the other (anomalous) group, mainly on the Some olivine crystals within these pallasites basis of chemical composition of metal [2]. Most contain regions of phosphoran olivine that have of the main group pallasites have oxygen between 3 and 5 wt% P2O5 [1, 18], where isotopic compositions similar to HED meteorites phosphorus substitutes for silicon in the and IIIAB irons [3]. The main-group pallasites tetrahedral site of the olivine structure. The appear closely related, and it is often suggested formation of phosphoran olivine has been that all of them, including anomalous members, interpreted as subsolidus reaction between originated on a single parent body [e.g. 4]. The olivine and phosphorus-bearing metal [1]. The origin of the pallasites, mixtures of Fe-Ni metal occurrence of phosphoran olivine in a few and olivine, and their parent bodies is not well pallasites suggests that it forms as a result of late established. The main group of pallasites which stage crystallization (or recrystallization) from are thought to come from one parent body may phosphorusrich melt pockets (perhaps as a result have formed 1) near the surface [5, 6], 2) close of Ostwald ripening),and not as a result of to the center [7, 8], 3) at the metal-olivine subsolidus reaction between phosphorus-bearing contact zones of isolated metal pods [9], or 4) at metal and olivine. Its occurrence further suggests core-mantle boundaries [10, 11]. Pallasites are that the olivine cooled relatively quickly from long thought to represent a metallic core-silicate high temperature, since slow cooling at high mantle boundary, where the IIIAB irons are temperature would crystallize any remaining linked to the crystallization history of the melt pockets into additional olivine and metallic fraction, and the HED meteorites may phosphate and equilibrate the system. The be linked to the silicate fraction [11,12,13,]. presence of phosphoran olivine is consistent with However, measurement of trace elements in more recent minor and trace element research individual metallic and silicate phases is [19, 20] which suggests fast cooling through necessary in order to fully under-stand the high temperatures, but slow cooling through petrogenetic history of pallasites, as well as any lower temperatures. has been interpreted as magmatic processes which may link pallasites to subsolidus Fe-Ni metal in pallasites exhibits both IIIAB irons and HED meteorites. Widmanstätten structure indicative of slow Partition coefficients of REEs between cooling, and cooling rates inferred from zoning olivine and melt have been measured [14, 15, profiles in metal are ~1°C /My [11]. However, 16]. Saito et al. [14] asserted that the V-shaped pallasite olivines show chemical diffusion pattern was produced with crystallization near profiles that suggest cooling rates of a few to the liquidus temperature. Partition coefficients of tens of degrees per year, roughly a million times faster [11,13,]. Metal and olivine do not seem to 40th Lunar and Planetary Science Conference (2009) 1042.pdf be telling the same story. The Widmanstätten that olivine-metal mixing occurred <10 My after pattern in meteorite metal develops between chondrule formation [28]. ~700-800 °C and ~400-500 °C where diffusion References: [1] Buseck P. R. (1977) GCA, effectively stops. Inferred time scales for cooling 41, 711-740. [2] Scott E. R. D. (1977) GCA, 41, through this temperature range in pallasites are a 349-360 [3] Clayton R. N. and Mayeda T. K. few degrees per million years [21]. The profiles (1996) GCA, 60, 1999-2017 [4] Wasson J.T. in olivine imply cooling rates of a few degrees (1999) manuscript. prep.and personal comm. [5] per year at ~1100 °C [20, 22]. Diffusion Mittlefehldt D.W. (1980) Earth Planet. Sci. Lett. gradients in olivine imply that the olivine moved 51, 29-40. [6] Davis A. M. and Olsen E. J. from a chemical environment where (1991) Nature, 353, 637-640. [7] Wahl W. concentrations were in equilibrium to an (1965) GCA, 29, 177-181. [8] Buseck P. R. and environment where they were not. The presence Goldstein J. I. (1968) Science, 159, 300-302. [9] of gradients means that the system failed to Urey H. C. (1956).The Astrophy. J., 124, 623- completely reequilibrate, probably because of 637. [10] Scott E. R. D. (1977) Mineral. Mag., rapid cooling. Thus, the concentration gradients 41, 265-272. [11] Wasson J. T. and Choi B.G. in pallasite olivines would appear to be a major (2003) GCA, 67, 3079-3096. [12] Scott E.D.R. problem for the core-mantle boundary model of (1977) GCA, 41: 349-360. [13] Mittlefehldt pallasite formation. A variety of origins for D.W. et al. (1998) In: Planetary Materials, pallasites have been put forward, including Papike J.J. (Ed), p. 195. [14] Saito T. et al., crystallization near the surface of an externally (1998) Geochem. J. 32, 159-182. [15] heated asteroid [5]; crystallization of an impact Consolmagno G. J. (1979) Icarus 40,522-530. melt [23]; and nebular condensation [24]. [16] Colson R.O. et al. (1988) GCA, 52, 539-553 However, the leading theory is that pallasites [17] Minova H. and Ebihara M.(2002) LPS were generated at the core–mantle interface of a XXXIII, #1386 [18] Buseck P. R. and Clark J. differentiated asteroid. Two mechanisms have (1984) Mineralogical Magazine 48, 229-235. been proposed. The first is an equilibrium [19] Hsu W. (2003) Meteoritics 38, 1217-1240. process, where the intercumulate silicate melt [20] Miyamoto, M. (1997) JGR 102, 21613- between cumulate mantle olivine is replaced by 21618. [21] Buseck P. R. and Goldstein J. I. molten metal [1, 25]. The alternative scenario (1969) GSA Bull. 80, 2141-2158. [22] involves an unequilibrium, impact-induced Miyamoto M. et al. (2004) MAPS 39, A70 [23] shock wave, which resulted in violent mixing of Malvin D.J. et al. (1985) Meteoritics 20, 259- cumulate olivine crystals and molten metal. A 273. [24] Kurat G. (1988) Philos. Trans. R. Soc. combination of these two mechanisms has also Lond. A325, 459-482 [25] Wood (1978) LPSC been proposed [26]. Although pallasites are XII, 1200-1202.[26] Scott E.D.R. and Taylor certainly mixtures of core and mantle material, G.J. (1990) LPSC XXI, 1119-1120. [27] Asphaug links between specific groups of irons and E. et al. (2006) Nature, 438, 155-160. [28] pallasites are tenuous suggesting that they may Tomiyama T. et al. (2007) LPS,38 ,# 2007. come from different parent bodies. Pallasites may have formed not at core-mantle boundaries as widely inferred, but from impact-generated mixtures of core and mantle materials. Mixing of small amounts of core metal with olivine mantle may have resulted from large impacts between asteroids. However, Asphaug et al. [27] infer that irons and stony-irons come from bodies that formed during glancing impacts between Moon- to- Mars-sized protoplanets. Such collisions may have converted differentiated projectiles into chains of differentiated bodies with diverse metal-silicate ratios. In the case of MG pallasites, 53Mn-53Cr isotope systematic suggest.