Partitioning of hydrogen (often referred to as H2O) between periclase (pe) and perovskite (pvk) at lower mantle conditions (24-80GPa) was investigated using quantum mechanics, equilibrium reaction thermodynamics and by monitoring two H-incorporation models. One of these (MSWV) was based on replacements provided by Mg2+↔2H+ and Si4+↔4H+; while the other (MSWA) relied upon substitutions in 2Mg2+↔Al3++H+ and Si4+↔Al3++H+. H2O partitioning in these phases was considered in the light of homogeneous (Bulk Silicate Earth; pvk: 75%-pe:16% model contents) and heterogeneous (Layered Mantle; pvk:78%-pe:14% modal contents) mantle geochemical models, which were configured for lower and upper bulk water contents (BWC) at 800 and 1500ppm, respectively. The equilibrium constant, BWCK(P,T), for the reactions controlling the H-exchange between pe and pvk exhibited an almost negligible dependence on P, whereas it was remarkably sensitive to T, BWC and the hydrogen incorporation scheme. Both MSWV and MSWA lead to BWCK(P,T)≤1, which suggests a ubiquitous shift in the exchange reaction towards an H2O-hosting perovskite. This took place more markedly in the latter incorporation mechanism, indicating that H2O-partitioning is affected by the uptake mechanism. In general, the larger the BWC, the smaller the BWCK(P,T). Over the BWC reference range, MSWV led to BWCK(P,T)-grand average (〈BWCK〉) calculated along lower mantle P-T-paths of ≈0.875. With regard to the MSWA mechanism, 〈BWCK〉 was more sensitive to BWC (and LM over BSE), but its values remained within the rather narrow 0.61-0.78 range. The periclase-perovskite H2O concentration-based partition coefficient, KdH2Ope/pvk, was inferred using 〈BWCK〉, assuming both hydrous and anhydrous-dominated systems. MSWV revealed a KdH2Ope/pvk-BWC linear interpolation slope which was close to 0 and KdH2Ope/pvk values of 0.36 and 0.56 (for anhydrous and hydrous system, respectively). MSWA, in turn, yielded a KdH2Ope/pvk trend with a slightly steeper negative BWC-slope, while it may also be considered nearly invariant with KdH2Ope/pvk values of 0.31-0.47 in the 800-1500ppm interval. Combining the MSWV and MSWA results led to the supposition that KdH2Ope/pvk lies in the narrow 0.31-0.56 interval, as far as the P-T-BWC values of interest are concerned. This implies that water always prefers pvk to pe. Furthermore, it also suggests that even in lower mantle with low or very low bulk water content, periclase rarely becomes a pure anhydrous phase.

Merli, M., Bonadiman, C., Diella, V., Pavese, A. (2016). Lower mantle hydrogen partitioning between periclase and perovskite: A quantum chemical modelling. GEOCHIMICA ET COSMOCHIMICA ACTA, 173, 304-318 [10.1016/j.gca.2015.10.025].

Lower mantle hydrogen partitioning between periclase and perovskite: A quantum chemical modelling

MERLI, Marcello;
2016-01-01

Abstract

Partitioning of hydrogen (often referred to as H2O) between periclase (pe) and perovskite (pvk) at lower mantle conditions (24-80GPa) was investigated using quantum mechanics, equilibrium reaction thermodynamics and by monitoring two H-incorporation models. One of these (MSWV) was based on replacements provided by Mg2+↔2H+ and Si4+↔4H+; while the other (MSWA) relied upon substitutions in 2Mg2+↔Al3++H+ and Si4+↔Al3++H+. H2O partitioning in these phases was considered in the light of homogeneous (Bulk Silicate Earth; pvk: 75%-pe:16% model contents) and heterogeneous (Layered Mantle; pvk:78%-pe:14% modal contents) mantle geochemical models, which were configured for lower and upper bulk water contents (BWC) at 800 and 1500ppm, respectively. The equilibrium constant, BWCK(P,T), for the reactions controlling the H-exchange between pe and pvk exhibited an almost negligible dependence on P, whereas it was remarkably sensitive to T, BWC and the hydrogen incorporation scheme. Both MSWV and MSWA lead to BWCK(P,T)≤1, which suggests a ubiquitous shift in the exchange reaction towards an H2O-hosting perovskite. This took place more markedly in the latter incorporation mechanism, indicating that H2O-partitioning is affected by the uptake mechanism. In general, the larger the BWC, the smaller the BWCK(P,T). Over the BWC reference range, MSWV led to BWCK(P,T)-grand average (〈BWCK〉) calculated along lower mantle P-T-paths of ≈0.875. With regard to the MSWA mechanism, 〈BWCK〉 was more sensitive to BWC (and LM over BSE), but its values remained within the rather narrow 0.61-0.78 range. The periclase-perovskite H2O concentration-based partition coefficient, KdH2Ope/pvk, was inferred using 〈BWCK〉, assuming both hydrous and anhydrous-dominated systems. MSWV revealed a KdH2Ope/pvk-BWC linear interpolation slope which was close to 0 and KdH2Ope/pvk values of 0.36 and 0.56 (for anhydrous and hydrous system, respectively). MSWA, in turn, yielded a KdH2Ope/pvk trend with a slightly steeper negative BWC-slope, while it may also be considered nearly invariant with KdH2Ope/pvk values of 0.31-0.47 in the 800-1500ppm interval. Combining the MSWV and MSWA results led to the supposition that KdH2Ope/pvk lies in the narrow 0.31-0.56 interval, as far as the P-T-BWC values of interest are concerned. This implies that water always prefers pvk to pe. Furthermore, it also suggests that even in lower mantle with low or very low bulk water content, periclase rarely becomes a pure anhydrous phase.
2016
Merli, M., Bonadiman, C., Diella, V., Pavese, A. (2016). Lower mantle hydrogen partitioning between periclase and perovskite: A quantum chemical modelling. GEOCHIMICA ET COSMOCHIMICA ACTA, 173, 304-318 [10.1016/j.gca.2015.10.025].
File in questo prodotto:
File Dimensione Formato  
1-s2.0-S001670371500616X-main.pdf

Solo gestori archvio

Tipologia: Versione Editoriale
Dimensione 1.1 MB
Formato Adobe PDF
1.1 MB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10447/165133
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 8
  • ???jsp.display-item.citation.isi??? 8
social impact