Diamonds from the Continental Lithosphere and the Convecting Upper Mantle: Implications for the Mantle Carbon Cycle
Diamonds provide an opportunity to trace the origin of carbon-bearing fluids in the deep mantle because of their non-reactivity, their antiquity, their retention of C and N isotopic signatures and their ability to host co- crystallizing mantle mineral inclusions. Advances in analytical sensitivity and spatial resolution now enable the study of single diamonds and those with multiple inclusions. Ideally, such samples can reveal their age, relative depth of origin, and by geological inference, the source of their carbon. The chief challenge to this research lies in the limited availability of large enough specimens, the rarity of inclusions, the confinement of occurrences to the ancient continents and the stability of diamonds during transport only in magmas of the kimberlite-lamproite association. For lithospheric diamonds, the eclogitic paragenesis is known to have a compliment of diamonds with isotopically light carbon whereas the peridotitic paragenesis has only diamonds with mantle-like carbon. Unambiquous source characterization is complicated by the potential for substantial isotopic fractionation of C and N during diamond-forming fluid passage through lithospheric host lithologies. Re-Os isotopic data on eclogitic sulfides supports recycling of isotopically light subducted carbon because the diamond formation ages match known episodes of subduction around craton margins. If intra-mantle fractionation produces isotopically light C signatures it must be confined to eclogite host lithologies. Diamonds from transition zone or lower mantle may be recognized by their unique majoritic garnet or ferropericlase inclusions. Typically, sublithospheric diamonds are nitrogen free or very nitrogen poor. For individual occurrences, diamonds that host majoritic garnet have distinct but restricted carbon isotopic compositions. These restricted ranges may fit derivation from subducted organic matter (e.g. at Jagersfontein with light isotopic compositions) or from subducted carbonates (e.g. at Kankan with isotopically heavy diamond carbon) and in both cases are associated with negative Eu anomalies in the included majoritic garnets. These features are consistent with the origin of their carbon by the recycling of subducted slabs into the deep mantle. Diamonds are some of the most clear indicators that carbon is recycled into the base of the lithospheric mantle and into the asthenosphere.
Base of lithosphere as major CO2 reservoir
Studies of kimberlite clan rocks indicate extensive reaction of these magmas with the deep lithosphere (Francis and Patterson, Lithos 2008). Heat balance constraints indicate implies that this process implies a major geochemical deep lithospheric reservoir for CO2. Stagnant-lid (including chemical-lid) convection supplies heat to the base of continental lithosphere and old oceanic lithosphere in balance with the conductive heat flow. The thermal gradient beneath cratons is well constrained by xenolith geotherm and surface heat flow studies. It is ca. 17 mW m-2. The upwelling velocity at the base of the boundary layer is dimensionally ρ C V Δ T, where ρ C is volume specific heat (4 MJ m-3 K-1), V is vertical velocity, and Δ T is the temperature change. The temperature change scales to the temperature to change viscosity by a factor of e, ca. 60 K. The velocity is thus ca. 2 km per million years, enough to circulate a column of material equivalent to the thickness of the mantle over geological time. The column of CO2 released has an equivalent pressure of 10s of bars. One would thus expect that low fraction melts have carried much CO2 into the lowermost lithosphere and that the lowermost lithosphere was extensively metasomatized by this process. One would also expect that kimberlite clan magmas and carbonate metasomatized mantle xenoliths would be much more common that observed. It is possible that the top of the zone of partial melting acts as a rheological trap for ascending dikes. The intrusion of a moderate volume of dikes into the overlying solid lithosphere puts it into compression and retards further dike intrusion. It is also possible that the low viscosity of CO2-rich mantle aids its entrainment into downwelling convection more than its low density retards entrainment.
The History of Exosphere Carbon Storage and Consequences for Mantle-Exosphere Volatile Fluxes
The storage of volatiles in the mantle and their fluxes between the mantle and the near surface environment (exosphere) are constrained in part from the history of volatile storage in the exosphere. Evidence for the early formation of the oceans indicates extensive initial degassing of the mantle, but raises the question as to the fate of the carbon that must have been degassed with the H2O. Long-term storage of carbon in the exosphere is thought to require large continental areas, as carbon in the oceanic domain is rapidly returned to the mantle. Consequently, early degassing of the mantle may have been followed by rapid massive return of carbon to the mantle via subduction, leading to very high H/C ratios in the early exosphere. Alternatively, the C may have been lost to space by impact ablation of a Venus-like CO2-rich atmosphere. Less plausibly, the C could have remained in the exosphere stored in the oceanic domain but somehow escaping recyling to the mantle. Assuming that exosphere carbon storage was in fact limited by continental area, gradual regrowth of the carbon exosphere budget would then parallel that of growth of the continents. Interestingly, this suggests that the relatively high H/C ratio of the modern exosphere compared to the mantle (Hirschmann and Dasgupta, 2009), is a remnant of very early Earth processes which have not been erased by subsequent volatile fluxes. A key problem with this scenario, however, is that the gradual regrowth of the exosphere carbon budget cannot have occurred with parallel growth of the exosphere H2O budget. Otherwise, there would have been substantial growth of the oceans coinciding with continental growth, which violates constraints from continental freeboard. This requires either that outgassing of carbon exceeded that of H2O, or that H2O subduction has been more efficient than CO2 subduction. The former is unlikely unless typical degrees of melting are very small. On the other hand, petrologic constraints generally suggest that carbon subduction is more efficient than H2O subduction. One possible explanation is that hotter Archean subduction zones effectively stripped carbon from subducting slabs, thereby facilitating growth of the exosphere carbon budget.
Abiotic Organic Synthesis as a Path to Biosynthesis
The disequilibrium between seawater and igneous rocks engenders conditions for organic synthesis. In some cases, the process is strictly abiotic, as inferred from the Lost City hydrothermal system near the mid-Atlantic Ridge. In other cases, microorganisms take advantage of the thermodynamic drive for organic synthesis to generate biomolecules. In either case, the reduction of inorganic carbon and nitrogen compounds to form organic compounds can be coupled to the oxidation of ferrous iron in the rock through reduction of water to hydrogen during the iron oxidation reactions. In fact, thermodynamic drives generated during hydrothermal alteration of igneous rocks can eliminate the energetic costs of some biochemical processes. Conditions at which either abiotic organic synthesis or biosynthesis becomes favorable can be detected through thermodynamic calculations. Such calculations rely on a strong foundation of experimental measurements that allow correlation algorithms for estimating thermodynamic data for aqueous organic compounds and biomolecules at elevated temperatures and pressures. Recent progress has led to data for a vastly expanded variety and quantity of aqueous organic compounds. New results indicate that hydrolytic oxidation of the fayalite component of olivine and the ferrosilite component of pyroxene can provide sufficient hydrogen to reduce nicotinamide adenine dinucleotide at attainable pH and silica activities. Costly biosynthesis processes, such as adenosine triphosphate synthesis can be coupled to other overall metabolic reactions like methanogenesis that release more than enough energy to compensate. In addition, at relatively low temperatures igneous alteration provides ample energy for overall biosynthetic pathways including lipid and amino acid synthesis. As a consequence, critical steps in biosynthesis appear to have roots in the abiotic organic synthesis favored by the alteration of igneous rocks by seawater.
Carbon Isotopes, Carbon Cycling and the Early Evolution of Carbon Reservoirs
The record of carbon isotopes from 3.5 Ga to the present shows a striking constancy in δ13C (near 0) for carbonate sediments, and more variable but typically strongly negative values for organic-rich sediments. During the last 500 Ma the greater apparent variations, may partly be due to a more detailed record, but could also be due to biological innovations and their consequences. In general the record indicates a more or less constant mass ratio between the photosynthetic carbon pool (PC) and the carbonate carbon pool (CC) of about 1:4. If this ratio was initially established about 3.5 GaBP, perhaps as an indication of the emergence of oxygenic photosynthesis (a very profound biological innovation), it could indicate a massive amount of carbon fixing in a relatively short period of time, if the dominant reservoir of carbon at that time was carbonate (consistent with early degassing of CO2). The oxygen equivalence of such massive photosynthetic carbon fixing would clearly have been enough to produce a surface environment with free oxygen, but this is not consistent with geological observations. This problem can be resolved if the major early Archean carbon pool was in the form of abiotic reduced carbon (ARC) rather than carbonate. The gradual oxidation of this ARC by subduction and magmatic production of CO2 implies a slow accumulation of PC and deposition of some of that carbon as well as carbonate. This model implies that deposition of PC and CC takes place in disequilibrium with the ARC pool, or isolation from it. This is conceivable if the location of carbon sedimentation is in an environment favoring PC and CC deposition, but where ARC is not deposited in significant amounts. It is not unreasonable to think that shallow water environments could meet this requirement. Regardless of the actual history of carbon reservoirs, there is the interesting question of why the deposition of PC and CC should occur in a more or less fixed ratio over much of Earth's history. Consideration of the equilibrations attending photosynthetic conversion of CO2 and carbonate mineral deposition suggests a plausible reason for this long-term consistency.
Carbon and hydrogen isotope effects in the open-system Fischer-Tropsch type reactions. Implications for abiogenic hydrocarbons in the Earth Crust
Fluid Monitoring and Geochemical Modeling of Carbon Storage in a Depleted Gas Field, Australia
The CO2CRC's (Cooperative Research Centre for Greenhouse Gas Technologies) Otway Project in Victoria, Australia, is a demonstration site for the storage of CO2 in a depleted natural gas field. Injection of an 80:20 CO2/CH4 supercritical fluid, is through the CRC-1 well into the Waarre C sandstone located 300 m downdip from the recompleted Naylor-1 monitoring well. Tracers SF6, Kr and perdeuterated methane (CD4) were added early in the injection stream to provide monitoring targets for establishing plume breakthrough and to uniquely identify the injection stream. Fluid monitoring is carried out through a U- tube downhole assembly at Naylor-1, enabling the collection of high quality samples of the supercritical and the aqueous phases. Extensive monitoring of formation fluids and geochemical modeling of the site indicates that initially the geochemistry is controlled by CO2-water-rock interactions and that the migration of the CO2 plume has resulted in predictable changes in the chemistry of the fluids. The Waarre C mineralogy has low reactive mineral content, however, it contains low but sufficient calcite to buffer pH and fluid composition. Geochemical modeling of the system under elevated CO2 content predicts some minor changes in the fluid chemistry including lower pH and an increase in the Ca2+ and HCO3- content. Except for a drop in pH, the predicted compositional changes have not been observed. The differences between the observed and modeled results are discussed in terms of multiphase flow and mixing of this 3- component system.