Available online at www.sciencedirect.com ScienceDirect Geochimica et Cosmochimica Acta 200 (2017) 201–217 www.elsevier.com/locate/gca The accretion and impact history of the ordinary chondrite parent bodies Terrence Blackburn a,⇑, Conel M.O’D. Alexander b, Richard Carlson b, Linda T. Elkins-Tanton c a Earth and Planetary Sciences, University of California, Santa Cruz, Santa Cruz, CA 95064, United States b Department of Terrestrial Magnetism, Carnegie Institution for Science, Washington, DC 20015, United States c School of Earth and Space Exploration, Arizona State University, Tempe, AZ 85287, United States Received 12 February 2016; accepted in revised form 24 November 2016; Available online 9 December 2016 Abstract A working timeline for the history of ordinary chondrites includes chondrule formation as early as 0–2 Ma after our Solar System’s earliest forming solids (CAIs), followed by rapid accretion into undifferentiated planetesimals that were heated inter- nally by 26Al decay and cooled over a period of tens of millions of years. There remains conflict, however, between metallo- graphic cooling rate (Ni-metal) and radioisotopic thermochronometric data over the sizes and lifetimes of the chondrite parent bodies, as well as the timing of impact related disruptions. The importance of establishing the timing of parent body disruption is heightened by the use of meteorites as recorders of asteroid belt wide disruption events and their use to interpret Solar System dynamical models. Here we attempt to resolve these records by contributing new 207Pb–206Pb data obtained on phosphates isolated from nine previously unstudied ordinary chondrites. These new results, along with previously published Pb-phosphate, Ni-metal and thermometry data, are interpreted with a series of numerical models designed to simulate the thermal evolution for a chondrite parent body that either remains intact or is disrupted by impact prior to forming smaller unsorted ‘‘rubble piles”. Our thermal model and previously published thermometry data limit accretion time to 2.05–2.25 Ma after CAIs. Measured Pb-phosphate data place minimum estimates on parent body diameters of 260–280 km for both the L and H chondrite par- ent bodies. They also consistently show that petrologic Type 6 (highest thermal metamorphism) chondrites from both the H and L bodies have younger ages and, therefore, cooled more slowly than Type 5 (lesser metamorphism) chondrites. This is interpreted as evidence for Type 5 chondrite origination from shallower depths than Type 6 chondrites within initially con- centrically zoned bodies. This contrasts metallographic cooling rate data that are inconsistent with such a simple onion shell scenario. One model that can reconcile these two data sets takes into account subtle differences in temperature to which each system responds. This working model requires that disruption occur early enough such that the Ni-metal system can record the cooling rate associated with a rubble pile (<70 Ma), yet late enough that the Pb-phosphate system can record an onion shell structure (>30 Ma). For this 30–70 Ma timeline, reaccretion into smaller rubble piles will ensure that the originally dee- ply buried and hot Type 6 samples will always cool faster as a result of disruption, yielding nearly uniform ages that record the time of parent body disruption. This is consistent with the available Pb-phosphate data, where all but one Type 6 chondrite (H, n =3;L,n = 4) yields a cooling age within a narrow 4505 ± 5 Ma timeframe. These data collectively imply that both the H and L chondrite parent bodies were catastrophically disrupted at 60 Ma. In addition, combined Ni-metal and Pb-phosphate models confirm that a subset of Type 4 chondrites record early rapid cooling likely associated with erosional impacting of the H and L parent bodies on 5 Ma timescales. Ó 2016 Elsevier Ltd. All rights reserved. ⇑ Corresponding author. E-mail address: [email protected] (T. Blackburn). http://dx.doi.org/10.1016/j.gca.2016.11.038 0016-7037/Ó 2016 Elsevier Ltd. All rights reserved. 202 T. Blackburn et al. / Geochimica et Cosmochimica Acta 200 (2017) 201–217 Keywords: Ordinary; Chondrites; Solar system disruption; Thermochronology 1. INTRODUCTION alone yield a compelling case for the concentrically zoned ‘‘onion shell model” for the OC parent bodies. Multiple sys- Ordinary chondrites (OCs) provide a record of planetes- tems, including Hf-W (Kleine et al., 2008), phosphate imal formation in the asteroid belt through the accretion of 207Pb–206Pb (Gopel et al., 1994; Amelin et al., 2005; some of the Solar System’s earliest formed solids, including Blinova et al., 2007) and phosphate fission track (Pellas silicate chondrules, metallic Fe, Ni grains and fine-grained and Storzer, 1981; Trieloff et al., 2003), yield dates that matrix. They preserve an early stage of planetesimal assem- are inversely correlated with metamorphic Type. This has bly that can guide our understanding of early Solar System been interpreted by a number of studies (Trieloff et al., processes, including how planetesimal accretion times and 2003; Kleine et al., 2008; Henke et al., 2012) as evidence parent body sizes varied within the protoplanetary disk. that the H-chondrite parent body had a long-lived Numerical simulations of planetesimal growth in the terres- (>100 Ma) layered structure – with Type 4 representing trial planet region predict rapid early growth of bodies that the shallowest and most quickly cooled of the equilibrated were large enough to melt within 105–106 years (Chambers, samples, and Type 6 representing the deeper more slowly 2004). This rapid, early timeline for terrestrial region plan- cooled planetesimal centers. Prior estimates of parent body etesimal formation is supported by radioisotopic studies of size are consistently 200 km in diameter (e.g., Trieloff et al., igneous achondrite meteorites, e.g., eucrites (Kleine et al., 2003; Kleine et al., 2008; Henke et al., 2012). 2002), angrites (Amelin, 2008), and irons (Kruijer et al., Despite this seemingly concordant result, the onion shell 2014), whose documented early differentiation likely model appears to be inconsistent with metallographic cool- occurred as a result of melting induced by the decay of ing rates estimated from Ni diffusion profiles in Fe-metal short-lived radioactive nuclides (26Al, 60Fe). The unmelted (Bennett and McSween Jr, 1993; Taylor et al., 1987; OCs, however, preserve accretion textures, primitive geo- Tomiyama et al., 2003; Scott et al., 2014). The metallo- chemistries and a range of chondrule ages (e.g., 0–3 Ma graphic cooling rate data (Ni-metal) have been interpreted after CAIs; Mostefaoui et al., 2002; Gounelle and Russell, as a record of chondrite parent body disruption (Taylor 2005; Connelly et al., 2008; Kleine et al., 2008) that reflect et al., 1987) or impact driven mixing (Scott et al., 2014) a younger timeline for planetesimal accretion relative to prior to cooling below the temperature (500 °C) at which achondrites. Thermal modeling studies have attempted to Ni diffusion in Fe-metal is negligible. Parent body disrup- define an OC parent body accretion time by constraining tion would had to have been followed by rapid reaccretion the conditions under which a chondrite parent body would of material into unsorted rubble piles while the fragments remain undifferentiated (Miyamoto et al., 1982; Hevey and were still hot (Taylor et al., 1987). This last step is consis- Sanders, 2006; Sugiura and Fujiya, 2014). The main criteria tent with both the lack of correlation between Ni-metal used to constrain these models include: (1) remaining below cooling rates and petrologic Type, as well as the existence the onset of melting, defined by either the Fe–FeS eutectic of chondrite regolith breccias. Both petrologic (e.g., at 980 °C(Dodd, 1981), or a silicate solidus at 1050– Ruzicka et al., 2015), and modeling (e.g., Davison et al., 1100 °C(Johnson et al., 2016), (2) reaching the central tem- 2013) studies have stressed the inescapable importance that peratures estimated by thermometry of Type 6 chondrites impact processing has for the thermal and mechanical mix- (900 ± 50 °C) (e.g., Olsen and Bunch, 1984; Slater- ing histories of chondrite parent bodies. Reynolds and McSween, 2005), and (3) a mechanism for The importance of impacts in the evolution of the chon- generating the ranges in thermal metamorphism observed drite parent bodies is supported by dynamical simulations in chondrite suites, classified as petrologic Types 3–6 (Van of Solar System evolution that predict that the asteroid Schmus and Wood, 1967). The latter is most easily pro- belt’s first several hundred million years will have been duced by a thermal gradient across the radius of a body punctuated by episodes of planetesimal disruption. Possible resulting from internal radiogenic heating and variable drivers for increases in high velocity impact events are degrees of insulation. While strongly dependent upon the numerous and include the injection of material into the thermal modeling constants chosen, the above criteria limit inner Solar System or clearing of the asteroid belt due to: undifferentiated OC parent bodies to those accreting (1) Dissipation of the gas disk, gas giant planet formation (instantaneously) after 2.0 Ma (Miyamoto et al., 1982; and/or early migration (1–10 Ma) (Chambers and Hevey and Sanders, 2006; Sugiura and Fujiya, 2014). Wetherill, 2001; Walsh et al., 2011; Turrini et al., 2012; Reconstruction of chondrite parent body size may be Kaib and Chambers, 2016); (2) terrestrial planet formation provided by chondrite cooling rate estimates from a num- (30–100 Ma) (Bottke et al., 2006; Bottke et al., 2015); (3) or ber of temperature-sensitive systems, including major ele- through dynamical events responsible for the late heavy ment distributions within silicate minerals (Ganguly et al., bombardment (700–800 Ma) (Gomes et al., 2005; 2013), metallographic cooling rates (Ni distribution in Levison et al., 2009; Bottke et al., 2012).