E N V I R O N M E N T , E A R T H A N D P L A N E T A R Y S C I E N C E S

S C I E N T I F I C H I G H L I G H T S

1 2 6 H I G H L I G H T S 2 0 2 3 I

X-ray techniques reveal tetracarbonates in silicate melts may be at the origin of hidden carbon reservoirs in the deep Earth

Reservoirs of carbon buried deep inside our planet are considered a possible source for carbon-rich emissions registered at the surface. X-ray techniques were used to investigate the behaviour of an analogue carbonate melt at pressures near the Earth s core mantle boundary (CMB), possibly explaining why carbon has been trapped there all along.

In the last century, the fast increase of CO2 in the atmosphere has progressively focused scientists attention on the carbon cycle and its evolution at the Earth s surface. Recent estimations locate up to 90% of the Earth s carbon between the Earth s mantle and core [1]. In these variable geological settings, carbonate melts may play a crucial role in favouring or hindering carbon storage within its interior. For instance, at shallow depths (< 5 GPa) carbonate melts are characterised by low densities and viscosities [2], properties which are best explained by the electronic structure of the CO32− groups that lack unpaired orbitals for covalent bonding, and thus are unable to polymerise [3], unlike the SiO44- tetrahedron in silicate melts. Their behaviour however, remains unexplored at greater depths, despite changes

in physical and chemical properties induced from compression-driven structural transformations.

In this study, compression experiments were performed on synthetic K2Mg(CO3)2 glass, used as an analogue for carbonate melts, in diamond anvil cells up to pressures (P) near the CMB, ~115 GPa. Characterisation of the sample was performed using a combination of non- resonant inelastic X-ray scattering (IXS) at the oxygen K-edge at beamline ID20 and X-ray diffraction (XRD) via pair distribution function (PDF) at ID15A and ID15B. The experiments were supported by ab initio molecular dynamics simulations.

The results indicate a structural compaction of the glass between ambient P and ~40 GPa with local atomic rearrangement and distortions. This is evidenced by the decrease in C-O, Mg-O and K-O bond lengths. Between ~40 to 85 GPa, the C=O double bond in CO3-groups begins to break, with consequent formation of CO4 units and increased polymerisation. The rupture of the C=O double bond is observed in the IXS spectra by the intensity loss of the peak near 534 eV, which reflects the resonant C=O bond associated to the CO3 units. The C-O bond length grows from 1.185(2) Å at 33 GPa to 1.323(1) Å at 85 GPa, due to the formation of CO4 larger molecules. At P > 85 GPa, the glass is fully polymerised. At ~112 GPa the transition is complete (Figure 101).

Fig. 101: a) Oxygen K-edge IXS spectra of K2Mg(CO3)2 glass at high pressures. Simulated spectra are shown as thin black lines. b) Integrated intensity of the π* peak versus pressure. c) PDF curves at different pressures (left) and a close-up on the C-O peak (right).