Amorphous boron oxide at megabar pressures via inelastic X-ray scattering

Abstract

Structural transition in amorphous oxides, including glasses, under extreme compression above megabar pressures (>1 million atmospheric pressure, 100 GPa) results in unique densification paths that differ from those in crystals. Experimentally verifying the atomistic origins of such densifications beyond 100 GPa remains unknown. Progress in inelastic X-ray scattering (IXS) provided insights into the pressure-induced bonding changes in oxide glasses; however, IXS has a signal intensity several orders of magnitude smaller than that of elastic X-rays, posing challenges for probing glass structures above 100 GPa near the Earth's core-mantle boundary. Here, we report megabar IXS spectra for prototypical BO glasses at high pressure up to ∼120 GPa, where it is found that only four-coordinated boron (B) is prevalent. The reduction in the B-O length up to 120 GPa is minor, indicating the extended stability of-bonded B. In contrast, a substantial decrease in the average O-O distance upon compression is revealed, suggesting that the densification in BO glasses is primarily due to O-O distance reduction without the formation of B. Together with earlier results with other archetypal oxide glasses, such as SiO and GeO, the current results confirm that the transition pressure of the formation of highly coordinated framework cations systematically increases with the decreasing atomic radius of the cations. These observations highlight a new opportunity to study the structure of oxide glass above megabar pressures, yielding the atomistic origins of densification in melts at the Earth's core-mantle boundary.

Structural Transitions in MgSiO3 Glasses and Melts at the Core‐Mantle Boundary Observed via Inelastic X‐ray Scattering

“…The detailed analysis of the O K ‐edge IXS features of MgSiO 3 glasses combined with the ab initio MD simulations revealed the densification of the depolymerized silicate magmas at the lowermost mantle. Recent IXS studies on simple oxide crystals and glasses have revealed that overall shifts in the O K ‐edge features, particularly shifts in the edge onset energy serve as structural proxies of the interatomic distances and bulk densities (Lee et al, , ; Yi & Lee, , ). For the MgSiO 3 glasses, Figure a presents the systematic shifts in the edge energy at the center of gravity ( E c ) values of the MgSiO 3 crystals (Yi & Lee, , ), glasses (Lee et al, ), and liquids with increasing density.…”

“…The IXS spectra were collected by scanning the X‐ray energy relative to the analyzer energy of 9.9045 keV. Scattered photons were collected at a scattering angle of 25° with a polycapillary post sample collimator, together with a single spherical Si(555) analyzer operating in a backscattering geometry (Chow et al, ; Lee et al, ). The O K ‐edge IXS spectra for MgSiO 3 glasses with varying pressures up to 130 GPa were collected with the step size and the energy loss (incident energy − analyzed energy) range of 0.5 eV and from 530 to 565 eV, respectively.…”

“…The presence of partial melts denser than the lower mantle silicate crystals in ULVZ near the bottom of the mantle has mainly been attributed to the potential partitioning of iron in the melts rather than in crystals, accounting for the observed density contrast (Caracas et al, ; Rost et al, ). In addition to such densification through iron enrichment, a dense packing of oxygen atoms, unique to compressed silicate melts above megabar pressures, may account for the presence of melts at the CMB (Lee et al, , ). Probing of bonding environments around oxygen is, therefore, crucial to understanding the nature of densification of silicate melts at the CMB.…”

“…Synchrotron inelastic X‐ray scattering (IXS) reveals otherwise hidden information on the direct bonding environments around target low‐z elements at high pressure (e.g., Lee et al, , ; Lin et al, ; Lin et al, ; Sternemann & Wilke, , and references therein) (see Text S1 in the supporting information). Indeed, oxygen K ‐edge IXS has been applied to explore the densification of MgSiO 3 glasses up to ~40 GPa (Lee et al, ).…”

“…However, the pressure condition of the earlier study was not sufficient to fully account for the changes in melt properties at CMB (Text S1). While the inefficiency of IXS poses challenges for probing structures above 100 GPa, recent experimental advances in X‐ray optics involving the postcollimation of scattered X‐rays enabled collection of oxygen K ‐edge IXS signals for simple oxides, such as B 2 O 3 and SiO 2 above megabar pressures (Chow et al, ; Lee et al, ; Lee et al, ). These studies offer opportunities to study transitions in oxygen environments in complex glasses.…”