Copper recycling and redox evolution through progressive stages of oceanic subduction: Insights from the Izu-Bonin-Mariana forearc

https://doi.org/10.1016/j.epsl.2021.117178Get rights and content

Highlights

  • Oxidized hydrous boninites show low Cu contents and no clear slab addition of Cu.

  • Oceanic slabs from initial to mature subduction add little Cu to mantle wedge.

  • Slab materials added to the mantle wedge are generally reducing.

  • Hydrous slab melt-peridotite reaction is likely vital to produce oxidized arc magma.

  • The metasomatized mantle wedge is not enriched in Cu from the oceanic slab.

Abstract

Addition of subducted materials from the slab to the mantle wedge is often thought to elevate the oxygen fugacity of arc magmas and also to fertilize the mantle wedge in metals for subduction-related Cu-Au deposits. However, it remains controversial if slab-driven metal addition is effective and whether it occurs at all stages of subduction. The Izu-Bonin-Mariana (IBM) forearc preserves a full record of arc development from the initiation of subduction of the Pacific plate to mature arc volcanism. Here, we study the ore-forming and redox-sensitive metal Cu and its isotopes (δ65Cu) in the type-locality forearc basalts (FABs), boninites and high-Mg andesites from the IBM forearc, as well as MORBs from the East Pacific rise. Overall, the FABs display variably high Cu contents and MORB-like δ65Cu, consistent with limited Cu isotopic fractionation during partial melting of typical MORB mantle sources. Beginning with the boninites, the magma products from the IBM record strong slab signals, high water contents and high fO2 (ΔFMQ > +1). However, the oxidized, hydrous boninites and subsequent high-Mg andesites display low Cu contents, and mantle-like Cu/Sc ratios and δ65Cu, with no clear indication of the addition of Cu from the slab. The boninites show noticeably lower TiO2, Yb and Cu contents than the FABs, consistent with refractory mantle sources. Combined with available data from boninites and arc basalts worldwide, our results lead to a general conclusion that subducted oceanic slabs from initial to mature subduction contribute little Cu to the mantle wedge. This is further supported by the compiled Cu contents of arc peridotites. Adding Cu-poor, water-rich slab melts to the mantle wedge at any stage of Pacific oceanic subduction causes limited release of Cu, which remains trapped predominantly by reduced sulfides in the subducting slab. This likely reflects the overall reducing nature of slab materials added to the mantle wedge. However, we propose that the subsequent reaction of such reducing, hydrous slab melts with peridotites and flux melting produce oxidized primitive arc magmas deep in the mantle wedge. The reactive process promotes sulfide dissolution and metal release from the mantle wedge itself to oxidized arc magmas with high sulfur solubility, which explains the inheritance of mantle-like δ65Cu and the sulfide-undersaturated early-stage evolution in boninites and arc basalts. This study elucidates the role of subducted oceanic slabs in metal transfer and redox evolution and implies no significant Cu enrichment in the metasomatized mantle sources of magmatic arcs.

Introduction

Most of the world's economic Cu resources are from porphyry Cu-Au deposits which are mainly sited in magmatic arcs above subduction zones (e.g., Sillitoe, 2010; Sun et al., 2015). Slab subduction has long been recognized to play a key role (e.g., Hedenquist and Lowenstern, 1994; Richards, 2011; Sun et al., 2015). Arc magmas display high oxygen fugacity (fO2) and high water contents which are often thought to reflect the addition of oxidized fluids and/or melts released from the slab (or water-rich slab melts) to the sub-arc mantle wedge (e.g., Bénard et al., 2018; Debret and Sverjensky, 2017; Evans, 2012; Kelley and Cottrell, 2009; Parkinson and Arculus, 1999). Consequently, as theoretical calculations and experimental constraints indicate, the oxidized slab fluids or melts would be expected to destabilize sulfides and release sulfur and the metals contained within them from the subducted oceanic lithosphere, resulting in noticeable enrichment of the mantle wedge (e.g., Canil and Fellows, 2017; Evans and Tomkins, 2011; Mungall, 2002; Sun et al., 2012, Sun et al., 2017). Metasomatic components in the mantle wedge are preferentially melted at later stages and so, if metal-enriched, should favor the formation of Cu-Au deposits (Griffin et al., 2013; Groves et al., 2019; McInnes et al., 1999; Richards, 2009). However, detailed studies and global compilations of natural samples indicate that the Cu concentration in primitive arc basalts (∼80 ppm) is similar to those of MORBs, implying that Cu is mainly sourced from the mantle wedge with an indiscernibly low contribution from the subducted oceanic lithosphere (e.g., Chiaradia, 2014; Jenner et al., 2010; Lee et al., 2012; Rezeau and Jagoutz, 2020; Wang et al., 2019). Also, metasomatized arc peridotites, even if oxidized, often do not display the expected enrichment of Cu and Au (e.g., McInnes et al., 1999; Parkinson and Pearce, 1998; Savov et al., 2007; Secchiari et al., 2020).

The paradox about Cu contents has been argued to reflect the reduced nature of primitive arc magmas similar to MORBs (e.g., Lee et al., 2012). Magmatic differentiation is probably a key mechanism to increase the oxygen fugacity of evolving magmas in thick continental arcs (Lee and Tang, 2020; Tang et al., 2020). However, high fO2 is widely observed for metasomatized arc peridotites (Parkinson and Arculus, 1999) and particularly primitive arc magmas of oceanic islands, e.g., sulfate phases and ΔFMQ of +1 to +1.5 (e.g., Bénard et al., 2018; Zelenski et al., 2018). The high fO2 nature and yet lack of Cu enrichment in primitive arc magmas is apparently contradictory, and the specific cause is enigmatic. It largely reflects the complexity of the deep mantle wedge processes and the lack of sufficient constraints from natural rocks. Recent work suggests that the fO2 may differ between subduction stages or depths (Cannaò and Malaspina, 2018), and that slab components cannot be as oxidizing as previously thought, and possibly even reducing (Li et al., 2020; Tollan and Hermann, 2019). This further complicates the understanding of redox-sensitive sulfide stability and thus metal release and transfer in subduction zones.

Copper is one of the most important ore-forming metals; combined with the concentration, its isotopes (δ65Cu) can provide further constraints on metal transfer and enrichment in the metasomatized mantle wedge (Liu et al., 2015; Zheng et al., 2019), because oxidized Cu mainly displays elevated δ65Cu with Δ65Cu fluid-residue of 0.5–3‰ (e.g., Fernandez and Borrok, 2009; Mathur and Fantle, 2015). In addition, slab fluids or melts released from variable subduction depths and stages are rich in chlorine (Cl), one of the key transport agents for Cu (Barnes et al., 2018; Kendrick et al., 2020; Rustioni et al., 2019). If the slab fluids or hydrous melts released are rich in Cu, fluid-driven isotopic fractionation would cause Cu complexed by Cl to display heavier δ65Cu by 0.1–0.7‰ than the residual phases in the slab (Guo et al., 2020). Any Cu transferred from the slabs to the mantle wedge should thus increase both the Cu content and δ65Cu of the wedge, features that are expected in derivative primitive arc magmas as well.

Boninites belong to high-Mg andesites and are widely accepted to form by melting of highly depleted mantle wedge fluxed by hydrous slab melts during the embryonic stage of subduction (Crawford et al., 1989; Pearce and Reagan, 2019). They display water contents (2–5 wt.%) and fO2 (mainly ΔFMQ: +1 to +2) as high as mature arc lavas, reflecting rapid fluid flux and oxidation of the mantle wedge (e.g., Brounce et al., 2015, Brounce et al., 2021; Umino et al., 2015). Given the incompatible behavior of Cu, previous episodes of high-degree melt extraction would have largely depleted the mantle source of boninites in Cu. Therefore, refractory boninites are more sensitive than primitive arc basalts to show whether subducted slabs are the source of significant quantities of metals. Also, boninites are magmatic products from a large mantle source region, minimizing the effect of local heterogeneity in the mantle wedge.

Here we report the Cu contents and δ65Cu of type-locality forearc basalts (FAB, n=5), boninites (n=12) and high-Mg andesites (n=8) from the renowned Izu-Bonin-Mariana (IBM) forearc, as well as fresh MORBs (n=8) from the East Pacific rise for comparison. The IBM samples represent a series of magmatic products formed from the early stages of oceanic subduction. When compared to data from mature IBM arc basalts, the results allow us to better understand the role of subducted oceanic slabs in metal recycling through progressive stages of oceanic subduction and the related evolution of redox states. We show that the Pacific oceanic slabs add little Cu to the mantle wedge at all stages from initial to mature subduction, suggesting that the sulfides and Cu are largely retained in the subducting slabs. We also propose a mechanism to explain why primitive arc magmas display no Cu enrichment despite high fO2.

Section snippets

Geological settings and samples

The Izu-Bonin-Mariana forearc is one of the most renowned localities recording magmatic products at all stages of subduction from initiation (at 52–51 Ma) to mature arc magmatism (Ishizuka et al., 2011; Reagan et al., 2015). Along the entire IBM forearc, volcanic stratigraphies are similar and comprise, from oldest to youngest, MORB-like forearc basalts, boninites and other high-Mg andesites (mainly 50–43 Ma), and finally arc tholeiites and calc-alkaline rocks (Ishizuka et al., 2011; Reagan et

Analytical methods

The bulk-rock Cu contents and Cu isotopes with major and trace elements of the IBM samples, and eight fresh MORBs from the East Pacific rise have been analyzed (Table 1, Supplementary Tables S1–S2). We analyzed the bulk rock δ65Cu following established sample digestion and column chemistry procedures (Wang et al., 2019; Zhu et al., 2019). Geological reference materials BHVO-2, BCR-2, BIR-1a, AGV-2 indicate long-term external uncertainty for δ65Cu measurements better than ±0.05‰ (2sd). Here,

Results

The MORBs from the East Pacific rise display Cu contents and a limited range of δ65Cu (0.06–0.17‰, Fig. 4, Fig. 5), similar to previously reported MORB values from other localities (Jenner and O'Neill, 2012; Liu et al., 2015; Savage et al., 2015). Overall, the IBM FABs display high Cu contents and MORB-like REE patterns, including those from U1439-1442 sites (Fig. 2, Fig. 4). The boninites and adakites display similar Cu contents, significantly lower than those of FABs and MORBs (Fig. 4b). Each

Effects of seawater alteration

Supergene alteration often leads to strong variations in Cu contents and δ65Cu (e.g., Fernandez and Borrok, 2009; Liu et al., 2019; Mathur and Fantle, 2015), whereas in this study the majority of FABs, boninites and adakites display similar Cu isotopes. However, each group of FABs, boninites and adakites has two outliers in Cu isotopes, heavier or lighter than the majority of samples (Fig. 5). The outliers mostly show low Cu contents (Fig. 5b) and likely result from seawater alteration rather

Conclusions

The magmatic products of the Izu-Bonin-Mariana forearc record arc development from the initiation of subduction of the Pacific plate to mature arc volcanism. Overall, the forearc basalts display variably high Cu contents, MORB-like δ65Cu and trace element patterns. Addition of water-rich slab melts from the Pacific oceanic crust induced flux melting of the earlier refractory mantle source to form oxidized, hydrous boninites and following high-Mg andesites. They display low Yb and Cu contents,

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This study was supported by the funds from National Natural Science Foundation of China (No. 41722302, 41673027, 41372068) and the China Geological Survey (DD20190358). SF is funded by Australian Research Council grant FL180100134. We thank the colleagues in cruises for sampling and Kang Chen for insightful discussion. We appreciate Massimo Chiaradia and anonymous reviewers for constructive and critical reviews and Rajdeep Dasgupta for careful editorial handling.

References (90)

  • S. Jégo et al.

    Fluid-present melting of sulfide-bearing ocean-crust: experimental constraints on the transport of sulfur from subducting slab to mantle wedge

    Geochim. Cosmochim. Acta

    (2013)
  • M.A. Kendrick et al.

    SW Pacific arc and backarc lavas and the role of slab-bend serpentinites in the global halogen cycle

    Earth Planet. Sci. Lett.

    (2020)
  • S. König et al.

    Boninites as windows into trace element mobility in subduction zones

    Geochim. Cosmochim. Acta

    (2010)
  • C.-T.A. Lee et al.

    How to make porphyry copper deposits

    Earth Planet. Sci. Lett.

    (2020)
  • H.-Y. Li et al.

    Radiogenic isotopes document the start of subduction in the Western Pacific

    Earth Planet. Sci. Lett.

    (2019)
  • J.-L. Li et al.

    Fluid-mediated metal transport in subduction zones and its link to arc-related giant ore deposits: constraints from a sulfide-bearing HP vein in lawsonite eclogite (Tianshan, China)

    Geochim. Cosmochim. Acta

    (2013)
  • Y. Li et al.

    An essential role for sulfur in sulfide-silicate melt partitioning of gold and magmatic gold transport at subduction settings

    Earth Planet. Sci. Lett.

    (2019)
  • S.-A. Liu et al.

    Copper isotopic composition of the silicate Earth

    Earth Planet. Sci. Lett.

    (2015)
  • S.-A. Liu et al.

    Cu and Zn isotope fractionation during oceanic alteration: implications for oceanic Cu and Zn cycles

    Geochim. Cosmochim. Acta

    (2019)
  • J.P. Lorand et al.

    Platinum-group element systematics and petrogenetic processing of the continental upper mantle: a review

    Lithos

    (2013)
  • R. Mathur et al.

    Cu isotopic fractionation in the supergene environment with and without bacteria

    Geochim. Cosmochim. Acta

    (2005)
  • W.F. McDonough et al.

    The composition of the Earth

    Chem. Geol.

    (1995)
  • I.J. Parkinson et al.

    The redox state of subduction zones: insights from arc-peridotites

    Chem. Geol.

    (1999)
  • M.K. Reagan et al.

    Forearc ages reveal extensive short-lived and rapid seafloor spreading following subduction initiation

    Earth Planet. Sci. Lett.

    (2019)
  • H. Rezeau et al.

    The importance of H2O in arc magmas for the formation of porphyry Cu deposits

    Ore Geol. Rev.

    (2020)
  • J.P. Richards

    Magmatic to hydrothermal metal fluxes in convergent and collided margins

    Ore Geol. Rev.

    (2011)
  • A. Secchiari et al.

    Highly siderophile and chalcophile element behaviour in abyssal-type and supra-subduction zone mantle: new insights from the New Caledonia ophiolite

    Lithos

    (2020)
  • W. Sun et al.

    Porphyry deposits and oxidized magmas

    Ore Geol. Rev.

    (2015)
  • Z. Sun et al.

    Sulfur abundance and heterogeneity in the MORB mantle estimated by copper partitioning and sulfur solubility modeling

    Earth Planet. Sci. Lett.

    (2020)
  • M.J. Valetich et al.

    Evolution of chalcophile elements in the magmas of the Bonin Islands

    Chem. Geol.

    (2019)
  • Z. Wang et al.

    Abundances of Ag and Cu in mantle peridotites and the implications for the behavior of chalcophile elements in the mantle

    Geochim. Cosmochim. Acta

    (2015)
  • Z. Wang et al.

    Evolution of copper isotopes in arc systems: insights from lavas and molten sulfur in Niuatahi volcano, Tonga rear arc

    Geochim. Cosmochim. Acta

    (2019)
  • M. Zelenski et al.

    Silicate-sulfide liquid immiscibility in modern arc basalt (Tolbachik volcano, Kamchatka): part I. Occurrence and compositions of sulfide melts

    Chem. Geol.

    (2018)
  • J.D. Barnes et al.

    The behavior of halogens during subduction-zone processes

  • A. Bénard et al.

    Oxidising agents in sub-arc mantle melts link slab devolatilisation and arc magmas

    Nat. Commun.

    (2018)
  • M. Brounce et al.

    Temporal evolution of mantle wedge oxygen fugacity during subduction initiation

    Geology

    (2015)
  • M. Brounce et al.

    Covariation of slab tracers, volatiles, and oxidation during subduction initiation

    Geochem. Geophys. Geosyst.

    (2021)
  • E. Cannaò et al.

    From oceanic to continental subduction: implications for the geochemical and redox evolution of the supra-subduction mantle

    Geosphere

    (2018)
  • M. Chiaradia

    Copper enrichment in arc magmas controlled by overriding plate thickness

    Nat. Geosci.

    (2014)
  • M. Chiaradia

    How much water in basaltic melts parental to porphyry copper deposits?

    Front. Earth Sci.

    (2020)
  • M. Chiaradia et al.

    Stochastic modeling of deep magmatic controls on porphyry copper deposit endowment

    Sci. Rep.

    (2017)
  • J.S. Cline et al.

    Can economic porphyry copper mineralization be generated by a typical calc-alkaline melt?

    J. Geophys. Res., Solid Earth

    (1991)
  • D.A. Coulthard et al.

    Magma source evolution following subduction initiation: evidence from the element concentrations, stable isotope ratios, and water contents of volcanic glasses from the Bonin forearc (IODP Expedition 352)

    Geochem. Geophys. Geosyst.

    (2021)
  • A.J. Crawford et al.

    Classification, petrogenesis and tectonic setting of boninites

  • B. Debret et al.

    Highly oxidising fluids generated during serpentinite breakdown in subduction zones

    Sci. Rep.

    (2017)
  • Cited by (16)

    View all citing articles on Scopus
    View full text