With the past decades of diamond inclusion research, it is now well established that the mantle-derived diamonds are originated either from the lithospheric mantle or sublithospheric mantle. The lithospheric diamonds can be further divided into mainly the peridotitic and eclogitic suites, which can be distinguished based on their inclusion chemistry, carbon, and nitrogen isotopic compositions (1, 2). The parental lithology of sublithospheric diamonds is less well established, partly due to their much lower occurrence relative to the lithospheric diamonds. But there has been growing isotopic evidence for the involvement of subducted materials in the source region of sublithospheric diamonds, such as carbon, boron, oxygen, and iron (3–6).
Precipitation of diamonds in the Earth’s mantle has been thought to require the presence of a fluid phase. Being C-O-H, saline, carbonatitic, silicic, or metallic in composition, these fluids were released upon dehydration or partial melting of the parental lithology and migrate through the mantle until they reach diamond saturation point due to either the change in pressure-temperature, or redox conditions. Understanding the parental lithology and fluid composition of different diamonds has primarily relied on their carbon and nitrogen isotope compositions and major/trace element compositions of mineral/fluid inclusions. These tools have been shown to be powerful in many cases but each could have their own disadvantages. Nitrogen isotopes, for example, are less applicable to sublithospheric diamonds due to their low N concentration. Trace element compositions, on the other hand, can be easily manipulated by small mass fractions of low degree-melt that are enriched in incompatible elements.
Understanding the diamond-forming fluids and their parental lithology require new tools that can provide a different perspective than the ones discussed above. In this presentation, we show recent developments in adapting Fe, Mg, and K isotope systems to diamond inclusion studies for a better understanding of their formation. These so-called “non-traditional” stable isotope systems were typically developed for large rocks that are not limited by sample amount. In order to adapt them to mineral inclusions tens to hundreds of micrometers in size, we’ve developed dedicated procedures to: 1) clean the diamond surface to remove contamination before extracting individual inclusions; 2) scale down the columns used for chemical purification to minimize blanks; and 3) improving sensitivity on the mass spectrometer to analyze small samples. With a Nu Plasma II at the Carnegie Institution for Science, we have shown to be able to analyze inclusion samples containing as little as 200 ng of Fe (6). With an upgraded Nu Plasma Sapphire at UCLA that is equipped with a collision cell, we are now able to analyze samples with >25 ng Fe. The same strategy has now been expanded to Mg and K isotope systems, for which a low sample limit of 25 ng and 300 ng has been achieved. With examples of Fe and Mg isotopic compositions of ferropericlase in sublithospheric diamond and K isotopic composition of fluid inclusions in fibrous diamonds, we show how isotopic compositions of major elements of mineral/fluid inclusions in diamond bring us new perspectives on their origin. Our tests show promising results to extend existing Mg and Fe protocols to silicate minerals and potentially applying similar strategies to silicon, calcium, and barium isotopes in the future.
more »
« less
Diamonds reveal subducted slab harzburgite in the lower mantle
Characterizing compositional heterogeneity in Earth’s lower mantle is critical to understanding its dynamics. Three low-nitrogen diamonds from Koffiefontein (South Africa), containing inclusion assemblages of ferropericlase ± orthopyroxene ± magnesite, constrain diamond formation in an Mg-rich lower-mantle environment. Ferropericlase inclusions have Mg# 82.7–88.5 and orthopyroxene inclusions (retrogressed bridgmanite) have Mg# 95.0–95.1 and mantle-like δ18O of +5.6‰ ± 0.2‰. Magnesite included in one diamond implicates carbonated fluids in diamond formation. High Mg# and low Ca, Al, and Na of the assemblage indicate a melt-depleted meta-harzburgitic environment, in contrast to more fertile compositions expected for primitive lower mantle. Extremely low Ca in orthopyroxene inclusions may reflect a combination of melt depletion and low equilibration temperatures at the time of trapping. Inclusion compositions implicate subducted oceanic slab meta-harzburgite as the host for diamond growth. Mantle-like δ18O of the orthopyroxene inclusions indicates unaltered oceanic lithosphere. Similar melt-depleted characteristics in lower-mantle inclusion assemblages worldwide support that meta-harzburgite is the dominant host of lower-mantle diamonds.
more » « less- Award ID(s):
- 2025779
- PAR ID:
- 10519439
- Publisher / Repository:
- Geological Society of America
- Date Published:
- Journal Name:
- Geology
- Volume:
- 51
- Issue:
- 3
- ISSN:
- 0091-7613
- Page Range / eLocation ID:
- 238 to 241
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Composite mantle xenoliths from the Cima Volcanic Field (CA, USA) contain a variety of melt (now glassy) inclusions hosted within mantle phases. The compositions and textures of these melt inclusions have the po- tential to constrain their trapping processes, melt sources, and the rates of ascent of their parent xenoliths. Here we focus on unusual spinel-hosted melt inclusions from one composite xenolith, reporting glass and daughter mineral compositions and textures and attempting to reconstruct inclusion bulk compositions. The xenolith contains spinel-hosted melt inclusions in its harzburgite, olivine-websterite and lherzolite layers; there are none in its orthopyroxenite layer. The glass compositions and reconstructed bulk compositions of the partly-crystallized inclusions correspond to alkaline intermediate melts, mostly trachyandesites. Such melts are most likely to be generated and trapped by vapor-undersaturated phlogopite or amphibole dehydration melting to an assemblage of liquid + spinel + olivine ± pyroxenes. We modeled the near-liquidus phase relations of the inclusion bulk compositions and noted the closest approach of each inclusion to simultaneous saturation with spinel and either phlogopite or amphibole, resulting in estimated trapping pressures of ~0.5–1.5 GPa and temperatures of ~1000–1100 ◦C. The large size of the hosting spinel grains suggests a slow process associated with these breakdown reactions, probably thinning of the lithosphere and steepening of the geotherm during regional extension. A linear correlation between the vesicle area and inclusion area (as proxies for volume) suggests an in-situ exsolution process from melts of relatively uniform volatile initial contents, consistent with trapping of vapor- undersaturated melts that later exsolve vapor during cooling and daughter crystal growth. A negative correla- tion between the glass content in melt inclusions and the size of the inclusion itself suggests a control on the degree of crystallinity with the size. There appears to be a two-stage cooling history captured by the inclusions, forming first prismatic daughter crystals and large round vesicles at the wall of the inclusion, followed by quenching to form a mat of fine crystallites and small vesicles in most inclusions. We connect the final quench to rapid ascent of the xenolith in its host melt, which also triggered partial breakdown of remaining amphibole to fine glassy symplectites.more » « less
-
In the current plate tectonic regime, thermal modeling, petrology, and seismology show that subsurface portions of cold slabs carry some of their volatiles into the deep upper mantle, mantle transition zone, and uppermost lower mantle avoiding the devolatilization occurring with normal arc and wedge subduction. Slab crustal remnants at these depths can melt by intersecting their carbonated solidus whereas slab mantle remnants can devolatilize by warming and metamorphosing to ‘dryer’ mineral assemblages. Since fluid release and earthquake production (“dehydration embrittlement”) operates down to ~300 km depths in all subduction zones, we propose, that deep-focus earthquakes trace the places of fluid release at deeper levels (350 to 750 km). Fluids in faults related to earthquake generation will become diamond-forming as they react with mantle rocks along the fault walls. Diamonds thus formed will record deformation produced by mantle convection and slab buckling during mantle storage. Lithospheric diamonds, stored in static ancient continental keels, lack the connection to this type of geodynamic regime that is evident for sublithospheric diamonds. However, a comparison between the two diamond types suggests a geologic model for lithospheric diamond formation in the ancient past. Lithospheric diamonds and sublithospheric diamonds both contain evidence for the recycling of sediments or surficial rocks that have equilibrated at low temperatures with seawater. The known way to inject these materials into diamond-forming regions is slab subduction. Hence both diamond types may have formed by variants of this same process that differ in depth and style over geologic time. Lithospheric diamonds are different from sublithospheric diamonds in critical ways: higher average N content, ages extending into the Paleoarchean, inclusion assemblages indicating formation at lower pressure, and lack of ubiquitous deformation features. Nitrogen content is the key to relating lithospheric diamonds to the subducting slab. Nitrogen occurs in clays and sediments at the slab surface or uppermost crust. Regardless of whether the slab is hot or cold during subduction, nitrogen will be removed into a mantle wedge if one exists. Additionally, diamonds will not survive in the melts/fluids generated in the wedge under oxidizing conditions. For sublithospheric diamonds, their low to non-existent nitrogen content occurs because they are derived from slab fluids/melts once nitrogen has been largely removed or from rocks deeper in the slab where nitrogen is scarce. The much higher nitrogen in lithospheric diamonds suggests that they formed from fluids/melts derived near the slab surface that contained N. In the Archean, such slabs must have subducted close to the nascent mantle keel with no mantle wedge so the fluids could be directly reduced by the mantle keel. We propose a gradual temporal change from shallow, keel-adjacent, mantle-wedge-poor subduction that produced lithospheric diamonds starting in the Paleoarchean to wedge-avoiding, cold and deep subduction that produced sublithospheric diamonds in the Paleozoic. This temporal change is consistent with many geologic features: an early stagnant lid and a buoyant Archean oceanic lithosphere; the slab-imbrication, advective thickening, and diamond-richness of portions of mantle keels; and anomalously diamond-rich ancient eclogites.more » « less
-
more » « less
Abstract The nature of the magma plumbing system of Large Igneous Provinces is still poorly understood. Among these exceptional magmatic events from Earth's past, the end-Triassic Central Atlantic Magmatic Province (CAMP) and the end-Cretaceous Deccan Traps (Deccan) coincided in time with two of the most catastrophic biotic crises during the Phanerozoic. In order to constrain the architecture of their magma plumbing system, glomerocrysts containing abundant bubble-bearing melt inclusions from basaltic lava flows of both CAMP and Deccan were investigated via in situ geochemical and microstructural analyses. The analysed glomerocrysts, dominated by augitic clinopyroxene crystals, represent fragments of a crystal mush entrained by basaltic magmas before eruption. The analysed melt inclusions, consisting of an intermediate to felsic composition glass and CO2-bearing bubbles, represent relics of interstitial melts and fluids within a porous crystal framework forming the crystal mush. The different volume proportions between bubbles and whole inclusions reveal that melt entrapment occurred after volatile exsolution. The minimum observed bubble/inclusion fraction indicates that the CO2 concentration in CAMP and Deccan melts was at least 0.3 wt.%, consistent with a maximum entrapment pressure of about 0.5 GPa at CO2–H2O fluid-saturated conditions. The MgO-rich composition of host clinopyroxene crystals and whole rocks is in contrast with the SiO2-rich composition of (trachy-) andesitic to rhyolitic glass of melt inclusions, pointing to disequilibrium conditions. Thermodynamic and geochemical modelling shows that fractional crystallization alone cannot explain the evolved composition of glass in melt inclusions starting from their whole rock composition. On one side, the oxygen isotope composition of clinopyroxene crystals in glomerocrysts ranges from +3.9 (± 0.3) to +5.8 (± 0.3) ‰ and their sample-averaged oxygen isotope composition spans from +4.4 (N = 10) to +5.6 (N = 10) ‰, implying that glomerocrysts crystallized from mafic melts with normal (i.e., mantle-like) to slightly low δ18O values. On the other side, the oxygen isotope composition of glass in melt inclusions ranges from +5.5 (± 0.4) to +22.1 (± 0.4) ‰, implying that melt inclusions entrapped intermediate to felsic melts with normal (i.e., mantle-like) to extremely high δ18O values, typical of (meta-) sedimentary rocks. Some melt inclusions are compatible with fractionation from the same mafic melts that crystallized their host mineral phase, but most melt inclusions are compatible with variable degrees of crustal assimilation and partial mixing, potentially followed by minor post-entrapment isotope re-equilibration. In the CAMP, where sedimentary basins are abundant, (meta-) pelites and occasionally granitoids were the most likely assimilants. On the contrary, in the Deccan, where sedimentary basins are rare, granitoids and metapelites were the most likely assimilants. Oxygen isotope compositions of glass in melt inclusions, spanning from mantle-like to crust-dominated signatures, suggest that the CO2 within their coexisting bubbles likely derived partly from the mantle and partly from assimilated crustal materials. The investigated glomerocrysts and their bubble-bearing melt inclusions are relics of a multiphase (i.e., solid + liquid + gas phases) crystal mush revealing a dynamic evolution for the magma plumbing system of both CAMP and Deccan, where crystals, silicate melts and exsolved fluids coexisted and interacted through most of the transcrustal section.
-
null (Ed.)Subducting tectonic plates carry water and other surficial components into Earth’s interior. Previous studies suggest that serpentinized peridotite is a key part of deep recycling, but this geochemical pathway has not been directly traced. Here, we report Fe-Ni–rich metallic inclusions in sublithospheric diamonds from a depth of 360 to 750 km with isotopically heavy iron (δ 56 Fe = 0.79 to 0.90‰) and unradiogenic osmium ( 187 Os/ 188 Os = 0.111). These iron values lie outside the range of known mantle compositions or expected reaction products at depth. This signature represents subducted iron from magnetite and/or Fe-Ni alloys precipitated during serpentinization of oceanic peridotite, a lithology known to carry unradiogenic osmium inherited from prior convection and melt depletion. These diamond-hosted inclusions trace serpentinite subduction into the mantle transition zone. We propose that iron-rich phases from serpentinite contribute a labile heavy iron component to the heterogeneous convecting mantle eventually sampled by oceanic basalts.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1130/G50675.1
