One of the biggest unknowns in deep carbon science is how much carbon is in Earth. Current estimates vary by a factor of ten, leading to significant differences in whole Earth models of carbon cycling and storage. A recent paper published in PNAS adds a new facet to how we understand carbon chemistry in Earth’s mantle, and therefore how much carbon could be stored in this most voluminous layer of our own and other planets .
Under the extreme temperatures and pressures of deep Earth, the characteristics of many chemical elements change from those observed at the surface. Through extensive experimentation and theoretical modeling, it is now known that deep reservoirs of carbon exist in the mantle in the form of graphite, diamond, methane, carbonates, the relative amounts of each depending on the pressure, temperature, and the local availability of other elements, particularly oxygen.
It has also been suggested that carbon could substitute for silicon in silicate minerals and melts, which account for almost half of the volume of the mantle . In this new study, however, it is not silicon that is substituted by carbon, but oxygen.
“ Such substitution occurs in polymer-derived ceramics, a class of material generated by heat from organic precursors. We found that such materials, made by a complex route at atmospheric pressure, were actually thermodynamically stable. I then had the idea that they might form more directly under mantle conditions and might constitute a significant mode of carbon substitution in silicates,” said Prof. Alexandra Navrotsky, University of California, Davis, USA, study co-author and DCO scientist.
Navrotsky and her colleagues capitalized on previous work investigating polymer-derived ceramics. These materials display no crystallinity when analyzed by X-ray diffraction, however nuclear magnetic resonance techniques (NMR) have shown the existence of nanodomains of local organization, with regions that contain mostly carbon and regions that contain a silica material with substantial carbon substitution for oxygen. These carbon-containing silica domains persist even when the silica phase is modified by cations such as lithium, producing a less polymerized network more like a geologic melt. More recent studies (in preparation) show this is true also when the modifying element is magnesium or calcium. Thus it is possible that silicate melts at moderate pressure under reducing conditions may contain carbon to the extent of several percent, and silicate crystals may contain carbon at the trace level.
This observation has enormous implications since this mode of carbon substitution (for oxygen rather than for silicon) may represent a previously unknown reservoir of carbon in the mantle. The extent of this reservoir, as well as how it is maintained and turned over as part of the whole Earth carbon cycle, is a key question for further investigation.
Image: Adapted from Figure 4 in reference . The image depicts a “defect” created by the replacement of oxygen by carbon in the lattice of an ortho-enstatite (MgSiO3) structure. The Si, O, Mg, and C atoms are shown in yellow, red, green, and dark gray, respectively.