Elsevier

Chemical Geology

Volume 442, 28 November 2016, Pages 11-22
Chemical Geology

186Os/188Os variations in upper mantle peridotites: Constraints on the Pt/Os ratio of primitive upper mantle, and implications for late veneer accretion and mantle mixing timescales

https://doi.org/10.1016/j.chemgeo.2016.08.033Get rights and content

Abstract

186Os/188Os variations in mantle peridotites provide constraints on the long-term Pt/Os evolution of the depleted mantle and the Pt/Os ratio of the primitive upper mantle (PUM). We report new 186Os/188Os data for mantle peridotites from continental (Rio Grande Rift and Colorado Plateau) and oceanic (Lena Trough and Hawaiian Islands) settings that span a wide range in fertility (Al2O3 0.67-4.42 wt.%) and 187Os/188Os ratios (0.1138-0.1305). Although peridotite 186Os/188Os values span only a narrow range (from 0.1198345 to 0.1198384), 186Os/188Os broadly correlates with indices of melt depletion including bulk rock Al2O3, spinel Cr#, and clinopyroxene Cr#, consistent with Pt depletion in residual peridotites.

PUM 186Os/188Os is estimated to be 0.1198378 ± 23 (2 SD) based on extrapolation of 186Os/188Os-fertility trends, which is very slightly lower than H-chondrites [≈ 0.1198398 ± 16 (2 SD); Brandon et al., 2006]. This value is consistent with a PUM Pt/Os of 1.7 ± 0.2, similar to average Pt/Os ratios of fertile continental peridotites. The inferred PUM Pt/Os is slightly lower than but within error of Pt/Os values measured in several classes of chondrites [Carbonaceous ≈ 1.8 ± 0.2, Ordinary ≈ 1.9 ± 0.1, and Enstatite 1.9 ± 0.1 (Brandon et al., 2006)] indicating that PUM Pt/Os is broadly chondritic. In contrast, estimates for PUM Ru/Ir and Pd/Ir (cf. Becker et al., 2006) are suprachondritic. The addition of a chondritic late veneer alone cannot create a combination of chondritic and suprachondritic HSE ratios for the PUM. Instead, minor core segregation occurring concurrently with the addition of a late veneer may explain the observed mantle HSE abundances and ratios.

Combined 186Os/188Os-187Os/188Os isotopic and Pt/Os and Re/Os variability in peridotites suggest an average mantle homogenization timescale of ~ 1.2 Ga. In contrast, combined Hf-Nd isotopic and Lu/Hf and Sm/Nd variability in peridotites indicate much shorter homogenization timescales (< 0.4 Ga), potentially reflecting enhanced homogenization by melt-rock interaction to which the Pt-Os and Re-Os systems are relatively immune. The mechanical mixing timescale inferred from Os isotopes is consistent with timescales predicted for whole mantle convection.

Introduction

Core formation resulted in almost complete sequestration of the Highly Siderophile Elements (HSE) from the Earth’s mantle (cf. Newsom, 1990). However, abundances of HSE in the Earth’s mantle are substantially higher than expected from metal-silicate partition coefficients experimentally determined at low pressures (Capobianco et al., 1993, Newsom, 1990). Furthermore, most of the HSE occur in broadly chondritic relative proportions, although there is significant variation in measured HSE ratios of mantle peridotites (cf. Meisel et al., 2001, Becker et al., 2006, Brandon et al., 2006). These observations have been taken as evidence for the addition of a ‘late veneer’ of chondritic material to Earth’s mantle after core formation (cf. Morgan, 1986), which would have elevated mantle HSE abundances and established their chondritic relative proportions. Although core segregation under high pressure and/or high temperature (high P-T) conditions may explain the mantle abundances of many slightly and moderately siderophile elements and possibly some HSE such as Pd (cf. Righter et al., 2008, Righter, 2011), it is uncertain whether high P-T core segregation can account for the near- chondritic relative abundances of a wide range of HSE.

Elemental ratios are susceptible to modification by processes such as melt-infiltration and metasomatism and/or weathering and alteration. This makes reliable estimation of Primitive Upper Mantle (PUM) compositions using measured elemental ratios subject to large uncertainties reflecting inherent scatter in the data resulting from the complicated, multi-stage history of many peridotites. For example, Becker et al. (2006) estimated Re/OsPUM ≈ 0.096 ± 0.029 (1 SE) by extrapolating Re/Os ratios in variably depleted peridotites to primitive mantle compositions (Al2O3 ~ 4.45 wt.%; McDonough and Sun, 1995), but the sizable error in this estimate reflects the large inherent scatter in mantle peridotite Re/Os ratios. Long-lived radiogenic isotope ratios (e.g. 187Os/188Os) can potentially provide more precise constraints on some primitive mantle elemental ratios as they reflect time-integrated parent/daughter ratios and are largely immune to recent modification. Extrapolation of PUM parent/daughter ratios from isotope systematics in peridotites also provides an independent check on estimates based on elemental abundances. Because 187Re decays to form 187Os [t1/2 ~ 42 Ga (Selby et al., 2007)], 187Os/188Os ratios in peridotites can constrain long-term Re/Os ratios in the mantle. Previous studies (cf. Meisel et al., 2001) have inferred that the 187Os/188Os of the PUM is ≈ 0.1296 ± 9. From this estimate, Re/OsPUM is calculated to be ≈ 0.090 ± 0.001 (1 SE), which is consistent with but significantly more precise than estimates utilizing measured elemental ratios (cf. Becker et al., 2006). This estimate is statistically indistinguishable from the estimated Re/Os of ordinary and enstatite chondrites (cf. Horan et al., 2003, Brandon et al., 2006). Thus, the Re/Os of the silicate Earth appears to be broadly chondritic, consistent with the late veneer model (cf. Becker et al., 2006).

Because 190Pt undergoes alpha decay to form 186Os [t1/2 ~ 468 Ga; Begemann et al., 2001], 186Os/188Os can be similarly used to infer the Pt/Os ratio of the PUM. Previous studies have examined 186Os/188Os in a variety of upper mantle materials. This dataset includes Os-Ir alloys (Walker et al., 2005) and chromitites (cf. Brandon et al., 1998, Walker et al., 2005), most of which have sub-chondritic 187Os/188Os. Other analyzed upper mantle materials include a small set of abyssal peridotites (AP) from a single section of the Kane Fracture Zone (KFZ). These peridotites span a narrow range in isotopic composition (187Os/188Os ≈ 0.1226-0.1276) and extent of melt depletion (whole rock Al2O3 ≈ 1.2-1.8 wt.%) (Brandon et al., 2000), which makes it difficult to evaluate correlations between 186Os/188Os and peridotite composition or melt-depletion history.

In this paper, we report 186Os/188Os in peridotites from continental (Rio Grande Rift and Colorado Plateau) and oceanic (Lena Trough and Hawaiian Islands) settings that span a wide range in fertility and 187Os/188Os. Correlations between fertility and 186Os/188Os are utilized to investigate the extent to which long-term Pt/Os fractionation in the upper mantle is controlled by melt depletion and to constrain the 186Os/188Os and thus the Pt/Os ratio of the PUM. Pt/OsPUM inferred from 186Os/188Os of fertile peridotites is combined with PUM estimates for other HSE ratios to evaluate the late veneer hypothesis. Because the addition of a late veneer alone cannot explain a combination of chondritic (e.g., Pt/Os) and suprachondritic (e.g., Pd/Ir) mantle HSE ratios, we examine alternate models involving minor metal or sulfide segregation concurrent with late veneer addition. Finally, the degree of isotopic (186Os/188Os and 187Os/188Os) and parent/daughter (Pt/Os and Re/Os) variability observed in upper mantle peridotites is utilized to constrain the timescale of mantle mixing and homogenization.

Section snippets

Samples

Mantle peridotites were selected from both continental (Rio Grande Rift and Colorado Plateau) and oceanic (Hawaiian Island and Lena Trough) settings that span a wide range in fertility and 187Os/188Os. Details of each of the sampling locations are given below.

  • a)

    Lena Trough

Abyssal peridotites (AP) from the Lena Trough were recovered from deep-sea dredges during the Polarstern ARK-XV-2 (1999) and Polarstern ARK-XX-2 (2004) cruises. Samples have undergone variable degrees of seafloor weathering and

Analytical methods

Detailed description of all analytical methods including standard values can be found in the supplementary materials. Most of the samples selected for this study have been previously analyzed for whole rock and mineral major and trace element compositions and detailed procedures are reported elsewhere (cf. Bizimis et al., 2007, Byerly and Lassiter, 2012, Lassiter et al., 2014). For this study, whole rock major and trace element compositions were analyzed via XRF and ICP-MS at the Geo-Analytical

Results

Bulk rock major and trace element concentrations are reported in Table S.1. Mineral major element compositions are reported in tables S.2-S.3 and clinopyroxene trace element concentrations are reported in Table S.4. Measured peridotites span a wide range of compositions. Bulk rock Al2O3 ranges from 0.67 to 4.42 wt.%. Spinel Cr# [molar Cr/(Cr + Al)] ranges from 8.2 to 50. Clinopyroxene Cr# varies from 5 to 25. Whole rock Al2O3, spinel Cr#, clinopyroxene Cr# and clinopyroxene Yb contents are all

Effect of melt depletion on mantle 190Pt-186Os evolution and constraints on the Pt/Os ratio of the PUM

Extrapolation of 186Os/188Os-fertility trends can be used to constrain PUM 186Os/188Os and Pt/Os in the same way previous studies (cf. Meisel et al., 2001) have utilized 187Os/188Os–fertility trends in peridotites to constrain PUM 187Os/188Os and Re/Os. In addition, if Pt/Os in peridotites is controlled by melt depletion, fertile peridotites may extend to higher Pt/Os, and therefore higher 186Os/188Os values than the refractory peridotites analyzed to date (Brandon et al., 2000). Previous

Conclusions

186Os/188Os ratios in variably melt-depleted peridotites from continental (Rio Grande Rift and Colorado Plateau) and oceanic (Lena Trough and Hawaiian Islands) settings correlate with various indices of peridotite fertility. This suggests that melt depletion largely controls 186Os/188Os variations in the upper mantle. Extrapolation of 186Os/188Os-fertility trends to PUM compositions suggests a mean PUM 186Os/188Os of 0.1198378 ± 23 (2 SD), which is slightly lower than H-chondrites [≈ 0.1198398 ± 16

Acknowledgements

This work was supported by NSF Grant EAR-1321937 and by the Jackson School of Geosciences, University of Texas at Austin. This manuscript benefitted from constructive comments by Laurie Reisberg, an anonymous reviewer and editor Catherine Chauvel as well as comments by Graham Pearson on an earlier draft.

References (69)

  • R. Chatterjee et al.

    High precision Os isotopic measurement using N-TIMS: Quantification of various sources of error in 186Os/188Os measurements

    Chem. Geol.

    (2015)
  • D.J. DePaolo

    Trace element and isotopic effects of combined wall rock assimilation and fractional crystallization

    Earth Planet. Sci. Lett.

    (1981)
  • M. Fischer-Godde et al.

    Rhodium, gold and other highly siderophile elements in orogenic peridotites and peridotite xenoliths

    Chem. Geol.

    (2011)
  • M.E. Fleet et al.

    Partitioning of platinum group elements (Os, Ir, Ru, Pt, Pd) and gold between sulfide liquid and basalt melt

    Geochim. Cosmochim. Acta

    (1996)
  • F.A. Frey et al.

    The mineralogy, geochemistry and origin of lherzolite inclusions in Victorian basanites

    Geochim. Cosmochim. Acta

    (1974)
  • A.W. Hofmann et al.

    Mantle plumes from ancient oceanic crust

    Earth Planet. Sci. Lett.

    (1982)
  • T.J. Ireland et al.

    186Os–187Os systematics of Hawaiian picrites revisited: new insights into Os isotopic variations in ocean island basalts

    Geochim. Cosmochim. Acta

    (2011)
  • M. Jamais et al.

    PGE and Os-isotopic variations in lavas from Kohala Volcano, Hawaii: Constraints on PGE behavior and melt/crust interaction

    Chem. Geol.

    (2008)
  • L.H. Kellogg et al.

    Homogenization of the mantle by convective mixing and diffusion

    Earth Planet. Sci. Lett.

    (1987)
  • J.C. Lassiter

    Constraints on the coupled thermal evolution of the Earth’s core and mantle, the age of the inner core, and the origin of the 186Os/188Os “core signal” in plume-derived lavas

    Earth Planet. Sci. Lett.

    (2006)
  • J.C. Lassiter et al.

    Osmium–isotope variations in Hawaiian lavas: evidence for recycled oceanic lithosphere in the Hawaiian plume

    Earth Planet. Sci. Lett.

    (1998)
  • J.C. Lassiter et al.

    Isotope and trace element variations in lavas from Raivavae and Rapa, Cook-Austral islands: constraints on the nature of HIMU- and EM-mantle and the origin of mid-plate volcanism in French Polynesia.

    Chem. Geol.

    (2003)
  • J.C. Lassiter et al.

    Constraints from Os-isotope variations on the origin of Lena Trough abyssal peridotites and implications for the composition and evolution of the depleted upper mantle

    Earth Planet. Sci. Lett.

    (2014)
  • V. Le Roux et al.

    The Lherz spinel lherzolite: refertilized rather than pristine mantle

    Earth Planet. Sci. Lett.

    (2007)
  • J.P. Lorand et al.

    Platinum-group element micronuggets and refertilization process in Lherz orogenic peridotites (northeastern Pyrenees, France)

    Earth Planet. Sci. Lett.

    (2010)
  • A. Luguet et al.

    A coupled study of sulfide petrology and highly siderophile element geochemistry in abyssal peridotites from the Kane Fracture Zone (MARK area, Mid-Atlantic ridge)

    Geochim. Cosmochim. Acta

    (2003)
  • W.F. McDonough et al.

    The composition of the Earth

    Chem. Geol.

    (1995)
  • T. Meisel et al.

    Osmium isotopic compositions of mantle xenoliths: A global perspective

    Geochim. Cosmochim. Acta

    (2001)
  • J.E. Mungall et al.

    Interfacial tension between magmatic sulfide and silicate liquids. Controls on kinetics of sulfide liquation and sulfide migration through silicate rocks

    Earth Planet. Sci. Lett.

    (2005)
  • M. Rehkamper et al.

    Non-chondritic platinum-group element ratios in oceanic mantle lithosphere: petrogenic Re, Os, Ir, Pt and Pd signature of melt percolation

    Earth Planet. Sci. Lett.

    (1999)
  • K. Righter

    Prediction of metal-silicate partition coefficients for siderophile elements: An update and assessment of PT conditions for metal-silicate equilibrium during accretion of the Earth

    Earth Planet. Sci. Lett.

    (2011)
  • D.C. Rubie et al.

    Heterogeneous accretion, composition and core-mantle differentiation of the Earth

    Earth Planet. Sci. Lett.

    (2011)
  • R.L. Rudnick et al.

    Interpreting ages from Re-Os isotopes in peridotites

    Lithos

    (2009)
  • P. Schiano et al.

    On the preservation of mantle information in ultramafic nodules: glass inclusions within minerals versus interstitial glasses

    Earth Planet. Sci. Lett.

    (1999)
  • Cited by (15)

    • Secular compositional changes in hydrated mantle: The record of arc-type basalts

      2022, Chemical Geology
      Citation Excerpt :

      This uniformity of composition (recognized by Condie, 2005) supports the existence of well-mixed mantle sources of basaltic magma during this 900-Myr time window, at least in terms of incompatible elements. Platinum-group elements and Hf, Nd and W isotope distributions in early Archean mantle rocks are also consistent with mantle mixing between 3.5 and 3.0 Ga (Chatterjee and Lassiter, 2016; Van De Locht et al., 2018). Nd isotopic studies of Blichert-Toft and Albarede (1994) suggest that in terms of incompatible elements, the mantle was well mixed by 2.6 Ga in agreement with our results.

    • Constraints on mantle evolution from Ce-Nd-Hf isotope systematics

      2020, Geochimica et Cosmochimica Acta
      Citation Excerpt :

      The mantle processing rate therefore determines the present extent of mantle depletion, and is a proxy for the overall rate of silicate earth differentiation (e.g., Stracke, 2018; Tackley, 2015). Hence, if the young DM ages of our model indicate that the mantle re-melts within < 1.5 Ga, which is consistent with the ∼ 1.2 Ga estimated based on Os isotope ratios in abyssal peridotites (Chatterjee and Lassiter, 2016), the processing rate of the mantle would be considerably faster than inferred from oceanic crust production rates (e.g., Stracke, 2018; Stracke et al., 2003; Tackley, 2015). If so, the resulting large extent of mantle depletion must be counteracted, on a global scale, by a relatively large return flux of subducted oceanic and continental crust.

    • Significance of high field strength and rare earth element distributions in deciphering the evolution of the inner solar system

      2019, Geochimica et Cosmochimica Acta
      Citation Excerpt :

      In contrast, basalts from hydrated mantle sources show significant element ratio variability both before and after 2.5 Ga (Fig. A1), suggesting that even before a possible onset of subduction around 3 Ga, hydrated mantle was not involved in widespread mantle mixing, perhaps because it was protected either in a non-subducting lithosphere or that it was continually produced at the top of the convecting mantle. The overabundance of platinum group elements (PGEs) in Earth’s mantle is generally explained by the addition of a thick veneer during the late stages of planetary accretion (Maier et al., 2009; Chatterjee and Lassiter, 2016). This PGE-rich veneer appears to have mixed with the mantle between 3.5 and 3.0 Ga, although recent data from ultramafic rocks in SW Greenland suggest that PGE mixing was complete at this site by 3.8 Ga (van de Locht et al., 2018).

    View all citing articles on Scopus
    View full text