Disequilibrium Melting and Melt Migration Driven by Impacts: Implications for Rapid Planetesimal Core Formation

Disequilibrium Melting and Melt Migration Driven by Impacts: Implications for Rapid Planetesimal Core Formation

Available online at www.sciencedirect.com Geochimica et Cosmochimica Acta 100 (2013) 41–59 www.elsevier.com/locate/gca Disequilibrium melting and melt migration driven by impacts: Implications for rapid planetesimal core formation Andrew G. Tomkins ⇑, Roberto F. Weinberg, Bruce F. Schaefer 1, Andrew Langendam School of Geosciences, P.O. Box 28E, Monash University, Melbourne, Victoria 3800, Australia Received 20 January 2012; accepted in revised form 24 September 2012; available online 12 October 2012 Abstract The e182W ages of magmatic iron meteorites are largely within error of the oldest solar system particles, apparently requir- ing a mechanism for segregation of metals to the cores of planetesimals within 1.5 million years of initial condensation. Cur- rently favoured models involve equilibrium melting and gravitational segregation in a static, quiescent environment, which requires very high early heat production in small bodies via decay of short-lived radionuclides. However, the rapid accretion needed to do this implies a violent early accretionary history, raising the question of whether attainment of equilibrium is a valid assumption. Since our use of the Hf–W isotopic system is predicated on achievement of chemical equilibrium during core formation, our understanding of the timing of this key early solar system process is dependent on our knowledge of the seg- regation mechanism. Here, we investigate impact-related textures and microstructures in chondritic meteorites, and show that impact-generated deformation promoted separation of liquid FeNi into enlarged sulfide-depleted accumulations, and that this happened under conditions of thermochemical disequilibrium. These observations imply that similar enlarged metal accumu- lations developed as the earliest planetesimals grew by rapid collisional accretion. We suggest that the nonmagmatic iron mete- orites formed this way and explain why they contain chondritic fragments in a way that is consistent with their trace element characteristics. As some planetesimals grew large enough to develop partially molten silicate mantles, these enlarged metal accumulations would settle rapidly to form cores leaving sulfide and small metal particles behind, since gravitational settling rate scales with the square of metal particle size. Our model thus provides a mechanism for more rapid core formation with less radiogenic heating. In contrast to existing models of core formation, the observed rarity of sulfide-dominant meteorites is an expected consequence of our model, which promotes early and progressive separation of metal and sulfide. We suggest that the core formation models that assume attainment of equilibrium in the Hf–W system underestimate the core formation time. Ó 2012 Elsevier Ltd. All rights reserved. 1. INTRODUCTION tem particles (Bizzaro et al., 2005; Carlson and Boyet, 2009; Kleine et al., 2009; Kruijer et al., 2012). However, these ages Magmatic iron meteorites are thought to represent the assume that cores formed by equilibrium fractionation of cores of planetesimals (Wasson and Richardson, 2001; Cha- metal from silicate, which is problematic given the implied bot and Haack, 2006; Schersten et al., 2006). Their tungsten rapidity of core formation. In the more widely accepted core isotope signatures have been interpreted as indicating that formation models, immiscible globules of Fe–Ni–S melt cores formed within 1.5 million years of the oldest solar sys- either settled through a partially molten silicate mantle (cf. Stevenson, 1990; McCoy et al., 2006; Wood et al., 2006; Sahijpal et al., 2007; Bagdassarov et al., 2009), or migrated ⇑ Corresponding author. Tel.: +61 3 99051643; fax: +61 3 through percolative flow (Yoshino et al., 2003, 2004; 99054903. Roberts et al., 2007), allowing equilibrium partitioning of E-mail address: [email protected] (A.G. Tomkins). siderophile and lithophile elements. A third model suggests 1 Present address: GEMOC, Earth and Planetary Sciences, that impacts may have played a role in driving melt Macquarie University, NSW 2109, Australia. 0016-7037/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.gca.2012.09.044 42 A.G. Tomkins et al. / Geochimica et Cosmochimica Acta 100 (2013) 41–59 migration (Bruhn et al., 2000; Rushmer et al., 2005), which formed before the OC parent bodies. This principle allows could happen on a comparatively rapid timescale and may us to investigate how impacts in the earliest solar system not require a silicate magma ocean, or complete equilibra- might have influenced the process of core formation, and tion between metal and silicate. Incomplete equilibration be- generate insights into the consequences for the global-scale tween segregating metal and silicate would result in mantle geochemistry of planetesimals and ultimately, planets. excess of 182W, and thus the ages derived from isotopic anal- ysis of iron meteorites would underestimate the core forma- 2. OBSERVED FEATURES OF IMPACT-INDUCED tion time (Nimmo and Agnor, 2006). Thus understanding MELTING AND METAL/SULFIDE MELT whether planetesimal cores formed by an equilibrium, or MIGRATION partial disequilibrium, process is critical to our understand- ing of the temporal evolution of the early solar system. The dominant majority of OC meteorites have been af- Here, we investigate processes that controlled formation fected by shock caused by hypervelocity collisions on their of enlarged metal accumulations within ordinary chondrite parent bodies. A well-accepted petrographic classification (OC) meteorites. The parent bodies for these meteorites are scheme (Stoffler et al., 1991) has been used by scientists to thought to have formed approximately 2–3 million years. divide numerous OC meteorites into six classes reflecting after CAIs (e.g., see Harrison and Grimm, 2010). Thermal increasing intensity of shock from S1 to S6. In this section, metamorphism in the early solar system is thought to have an overview of melt-bearing features attributed to impact been dominantly caused by radiogenic heating through de- processes is provided, progressing from micro- to macro- cay of 26Al and 60Fe (e.g., Moskovitz and Gaidos, 2011), scale, highlighting key aspects that will facilitate discussion and since the half lives of these isotopes are only 0.717 on drivers of melt segregation and associated disequilib- and 1.5 million years respectively, the OC meteorites es- rium in subsequent sections. caped the most intense period of radiogenic heating. Although radiogenic metamorphism generated tempera- 2.1. Isolated melt pockets tures that may have significantly exceeded 900 °C near the cores of the OC parent bodies (e.g., Harrison and Grimm, Initial signs of shock-induced melting are recognised in 2010), they are thought to have escaped widespread metal– moderately impact-affected meteorites that have character- troilite melting (Slater-Reynolds and McSween, 2005), and istics of shock Stages S3 and S4. In these meteorites, iso- this is certainly true for less metamorphosed petrologic lated melt pockets preferentially occur at interfaces types (types 3–5). The OC meteorites thus preserve the ef- between high density metal or troilite, and low density sili- fects of impacts on metal–sulfide–silicate assemblages, with- cate phases, particularly plagioclase (Dodd and Jarosewich, out having been overprinted by melting associated with 1979; Stoffler et al., 1991). At low degrees of shock, melt radiogenic metamorphism, and are consequently our best pockets are not interconnected at scales exceeding 2 mm. record of these processes. These melt pockets are recognised as irregularly shaped do- In OC meteorites, many researchers have observed sili- mains containing variable proportions of either (1) glass- cate glasses in veins and breccias, which formed through like silicate material (cf. Stoffler et al., 1991; henceforth re- localised impact heating during collisions between asteroids ferred to as silicate glass) with spherical metal-only, metal– (Stoffler et al., 1991). Encapsulated within these silicate troilite or troilite-only inclusions (Fig. 1A), or (2) metal glasses are typically numerous micro- to nano-scale spheri- and/or troilite with numerous spherical inclusions of silicate cal balls of FeNi metal and/or troilite (FeS), indicating that glass (Fig. 1B). These represent two different liquid emul- these components were also melted during impacts and then sions with distinct physical properties. To describe the com- trapped as immiscible melt globules in the silicate melt. Mi- position of these emulsions we will refer to them, for cro-scale metal ± troilite veinlets have also been observed in example, as metal-dominated emulsions, meaning immisci- these meteorites, and it is generally accepted that they repre- ble silicate melt globules contained within metal melt, or sil- sent trapped metal and/or troilite melt (Stoffler et al., 1991). icate-dominated emulsions, meaning the opposite. Rare OC meteorites contain significantly thicker (mm- to The impact heating process responsible for this incipient cm-scale) metal-dominated veins, or metal-cemented brec- melting was typically tightly spatially focused such that cias (Rubin et al., 2001; Ruzicka et al., 2005). This metal even in conjoined metal–troilite grains, the dense phase in- enrichment is thought to be the result of migration of immis- volved in melting was either (1) FeNi metal only (Fig. 1A cible metal ± sulfide melts generated by impact heating (Ru- and B), (2) troilite only (Fig. 1C), or (3) both

View Full Text

Details

  • File Type
    pdf
  • Upload Time
    -
  • Content Languages
    English
  • Upload User
    Anonymous/Not logged-in
  • File Pages
    19 Page
  • File Size
    -

Download

Channel Download Status
Express Download Enable

Copyright

We respect the copyrights and intellectual property rights of all users. All uploaded documents are either original works of the uploader or authorized works of the rightful owners.

  • Not to be reproduced or distributed without explicit permission.
  • Not used for commercial purposes outside of approved use cases.
  • Not used to infringe on the rights of the original creators.
  • If you believe any content infringes your copyright, please contact us immediately.

Support

For help with questions, suggestions, or problems, please contact us