Elsevier

Earth and Planetary Science Letters

Volume 474, 15 September 2017, Pages 345-354
Earth and Planetary Science Letters

The early differentiation of Mars inferred from Hf–W chronometry

https://doi.org/10.1016/j.epsl.2017.06.047Get rights and content

Highlights

  • Martian meteorites show widespread 182W variations that are correlated with 142Nd.

  • Main cause of 182W variations is silicate differentiation and not core formation.

  • Magma ocean differentiation on Mars occurred within ∼20–25 Ma after CAI formation.

  • This was followed by prolonged crust formation until ∼15 million years later.

Abstract

Mars probably accreted within the first 10 million years of Solar System formation and likely underwent magma ocean crystallization and crust formation soon thereafter. To assess the nature and timescales of these large-scale mantle differentiation processes we applied the short-lived 182Hf–182W and 146Sm–142Nd chronometers to a comprehensive suite of martian meteorites, including several shergottites, augite basalt NWA 8159, orthopyroxenite ALH 84001 and polymict breccia NWA 7034. Compared to previous studies the 182W data are significantly more precise and have been obtained for a more diverse suite of martian meteorites, ranging from samples from highly depleted to highly enriched mantle and crustal sources. Our results show that martian meteorites exhibit widespread 182W/184W variations that are broadly correlated with 142Nd/144Nd, implying that silicate differentiation (and not core formation) is the main cause of the observed 182W/184W differences. The combined 182W–142Nd systematics are best explained by magma ocean crystallization on Mars within ∼20–25 million years after Solar System formation, followed by crust formation ∼15 million years later. These ages are indistinguishable from the I–Pu–Xe age for the formation of Mars' atmosphere, indicating that the major differentiation of Mars into mantle, crust, and atmosphere occurred between 20 and 40 million years after Solar System formation and, hence, earlier than previously inferred based on Sm–Nd chronometry alone.

Introduction

The early evolution of Mars probably involved large-scale melting and core formation, followed by magma ocean crystallization and crust formation (e.g. Elkins-Tanton, 2005, Mezger et al., 2013). The timescales of these processes can be quantified through application of the short-lived 182Hf–182W [half-life = 8.9 million years (Ma)] and 146Sm–142Nd (half-life = 103 Ma) systems to martian meteorites, which derive from compositionally distinct sources that were established during the early differentiation of Mars. The mantle sources of martian meteorites are thought to comprise mafic cumulates and late-stage crystallization products of a magma ocean (Borg et al., 1997, Borg and Draper, 2003; Elkins-Tanton, 2008, Elkins-Tanton, 2005), as well as crust (Agee et al., 2013, Humayun et al., 2013). These distinct reservoirs have different Hf/W and Sm/Nd ratios, ultimately leading to variations in radiogenic 182W and 142Nd within the martian mantle and crust. Thus, the 182W and 142Nd compositions of martian meteorites from distinct sources reflect Hf/W and Sm/Nd fractionations during the earliest evolution of Mars and as such can be used to constrain the timescales of magma ocean processes and crust formation.

Several studies have shown that large radiogenic 182W and 142Nd variations exist within Mars (Borg et al., 2016, Borg et al., 1997; Brennecka et al., 2014, Caro et al., 2008, Debaille et al., 2007, Foley et al., 2005, Kleine et al., 2004, Lee and Halliday, 1997). For instance, nakhlites display some of the most radiogenic 142Nd and 182W compositions yet reported among martian meteorites, indicating source formation within ∼25 Ma of Solar System formation (Harper et al., 1995; Kleine et al., 2004, Foley et al., 2005, Debaille et al., 2009). Similarly, early studies on shergottites suggested that Mars primordial differentiation occurred at about 20–60 Ma after Solar System formation (Borg et al., 2003, Kleine et al., 2004, Foley et al., 2005). However, mainly driven by improvements in the analytical precision of 142Nd/144Nd measurements, subsequent studies demonstrated that shergottites define a precise 142Nd–143Nd model age of 63±6 Ma after Solar System formation (Borg et al., 2016). The significance of this age, and whether the 142Nd–143Nd systematics of shergottites record a single differentiation event is debated, however. As such, the 142Nd–143Nd data for shergottites have also been interpreted to record a prolonged interval of magma ocean crystallization on Mars, lasting between 30100 Ma after Solar System formation (Debaille et al., 2007).

One potential issue in the chronological interpretation of 142Nd–143Nd systematics is the presence of nucleosynthetic Nd isotope variations that arise through the heterogeneous distribution of presolar matter at the bulk meteorite and planetary scale (Burkhardt et al., 2016). For instance, recent studies have shown that the ∼10–20 parts-per-million 142Nd difference observed between chondrites and terrestrial samples (Boyet and Carlson, 2005) reflects nucleosynthetic Nd isotope heterogeneity between chondrites and the Earth (Burkhardt et al., 2016, Bouvier and Boyet, 2016), rather than an early Sm/Nd fractionation and subsequent radiogenic ingrowth from short-lived 146Sm. The 142Nd difference between terrestrial samples and chondrites, therefore, does not provide a record of an early differentiation of the silicate Earth. This example highlights that quantifying the extent of nucleosynthetic Nd isotope variations is essential for using the 146Sm–142Nd system to obtain meaningful ages for early differentiation processes. For Mars the extent of nucleosynthetic Nd isotope anomalies is not well known, however, and this may impact the chronology of Mars' early differentiation inferred from 146Sm–142Nd systematics. For instance, assuming an ordinary chondrite-like bulk 142Nd/144Nd for Mars provides a ∼30 Ma model age for the formation of the source of depleted shergottites (Debaille et al., 2007), whereas this age changes to ∼60 Ma if an Earth-like 142Nd144Nd is assumed for bulk Mars (Borg et al., 2016). Thus, the aforementioned uncertainties in the 146Sm–142Nd timescale for Mars' early differentiation at least partially reflect uncertainties in the 142Nd composition of bulk Mars.

The Hf–W chronometer is ideally suited to investigate the duration of magma ocean differentiation on Mars and to distinguish between an early differentiation at ∼30 Ma and a later differentiation at ∼60 Ma after Solar System formation. This is because owing to the much shorter half-life of 182Hf compared to 146Sm, significant 182W variations can only be produced within the first ∼50 Ma of the Solar System (e.g., Kleine et al., 2009). Thus, if the martian magma ocean crystallized at ∼60 Ma as suggested by a 142Nd–143Nd isochron for shergottites (Borg et al., 2016), then these meteorites should all have the same 182W composition. Conversely, if Mars' early differentiation largely occurred at ∼30 Ma, then there should be 182W variations among the shergottites. Published 182W data for shergottites do not show resolvable 182W variations (Foley et al., 2005, Kleine et al., 2004, Lee and Halliday, 1997) and, therefore, seem to be consistent with differentiation of Mars at ∼60 Ma after Solar System formation. However, the precision of the 182W measurements achievable at the time of these earlier studies was significantly lower than at present and was insufficient for resolving potential 182W variations among the shergottites formed from relatively young source regions.

To assess the extent of 182W variations in the martian mantle, and to better constrain the timescales of Mars' early differentiation, we obtained high-precision 182W data for a comprehensive suite of martian meteorites, including samples derived from some of the most enriched and depleted sources known on Mars. To help interpret the 182W data in terms of differentiation timescales, we also obtained high-precision 142Nd data for some of the same samples, coupled with data for non-radiogenic Nd isotopes to assess whether Mars shows a nucleosynthetic Nd isotope anomaly relative to Earth. Combined, these data provide new insights into the timescales of core formation, magma ocean crystallization, and crust formation on Mars.

Section snippets

Samples and analytical methods

Samples selected for this study include several shergottites (5 enriched, 3 intermediate, 5 depleted), augite basalt Northwest Africa (NWA) 8159, orthopyroxenite Allan Hills (ALH) 84001, and polymict breccia NWA 7034. The last two samples derive from the most incompatible trace element–enriched sources, whereas Tissint and NWA 7635 derive from the most depleted sources known on Mars (Agee et al., 2013, Brennecka et al., 2014, Humayun et al., 2013; Lapen et al., 2017, Lapen et al., 2010; Nyquist

Results

The investigated samples show large and well-resolved ε182W variations from approximately +0.1 (ALH 84001) to approximately +1.8 (NWA 7635) (Fig. 2a, Table 1), indicating that the total spread in ε182W is much larger than found for shergottites by previous studies (Fig. 2b; Foley et al., 2005, Kleine et al., 2004; Lee and Halliday, 1997). In particular, in contrast to earlier work, we find that different groups of shergottites exhibit distinct ε182W values. The enriched shergottites have a

Origin of 182W variations and Hf–W age of core formation

Utilizing the Hf–W system to date core formation on Mars requires knowledge of the ε182W composition of the bulk martian mantle, that is, the martian mantle composition set solely by core formation. Previous studies have estimated this value using the co-variation of ε182W and ε142Nd (Kleine et al., 2004, Foley et al., 2005, Mezger et al., 2013). Because silicate differentiation leads to correlated 182W–142Nd variations, the ε182W of samples having the ε142Nd of bulk Mars should represent the ε

Conclusions

Large 182W variations among martian meteorites require silicate differentiation on Mars within 20–40 million years after Solar System formation and, hence, earlier than previously inferred based on 146Sm–142Nd systematics of shergottites alone. The 182W and 142Nd compositions of ALH 84001 and NWA 7034 are the least radiogenic compositions yet reported for martian rocks, indicating that these samples derive from the most strongly enriched sources known from Mars. The combined 182W–142Nd

Acknowledgments

NASA and the Meteorite Working Group are gratefully acknowledged for providing several Antarctic meteorite samples for this study, and the UNM Meteorite Museum for providing samples of NWA 7034 and NWA 8159. We thank Richard Walker, Tim Elliott, and Alex Halliday for their constructive and very helpful reviews, and Derek Vance for his editorial efforts. Celeste Brennecka is acknowledged for comments on the manuscript. This study was performed under the auspices of the US DOE by Lawrence

References (45)

  • T. Kleine et al.

    182Hf–182W isotope systematics of chondrites, eucrites, and martian meteorites: chronology of core formation and early mantle differentiation in Vesta and Mars

    Geochim. Cosmochim. Acta

    (2004)
  • T. Kleine et al.

    Hf–W chronology of the accretion and early evolution of asteroids and terrestrial planets

    Geochim. Cosmochim. Acta

    (2009)
  • T.S. Kruijer et al.

    Nucleosynthetic W isotope anomalies and the Hf–W chronometry of Ca–Al-rich inclusions

    Earth Planet. Sci. Lett.

    (2014)
  • T.S. Kruijer et al.

    Hf–W chronometry of core formation in planetesimals inferred from weakly irradiated iron meteorites

    Geochim. Cosmochim. Acta

    (2012)
  • B. Marty et al.

    Signatures of early differentiation of Mars

    Earth Planet. Sci. Lett.

    (2002)
  • R. Morishima et al.

    N-body simulations of oligarchic growth of Mars: implications for Hf–W chronology

    Earth Planet. Sci. Lett.

    (2013)
  • F. Nimmo et al.

    How rapidly did Mars accrete? Uncertainties in the Hf–W timing of core formation

    Icarus

    (2007)
  • K. Righter et al.

    Magmatic fractionation of Hf and W: constraints on the timing of core formation and differentiation in the Moon and Mars

    Geochim. Cosmochim. Acta

    (2003)
  • G.A. Snyder et al.

    A chemical model for generating the sources of mare basalts: combined equilibrium and fractional crystallisation of the lunar magmasphere

    Geochim. Cosmochim. Acta

    (1992)
  • C.B. Agee et al.

    Unique meteorite from early Amazonian Mars: water-rich basaltic breccia Northwest Africa 7034

    Science

    (2013)
  • R. Andreasen et al.

    Constraints on the isotopic composition of the shergottite mantle sources — from observation based on the expanding rock record

  • L.E. Borg et al.

    A petrogenetic model for the origin and compositional variation of the martian basaltic meteorites

    Meteorit. Planet. Sci.

    (2003)
  • Cited by (83)

    View all citing articles on Scopus
    View full text