Influence of dehydration on the electrical conductivity of epidote and implications for high‐conductivity anomalies in subduction zones

Abstract

The anomalously high electrical conductivities (~0.1 to 1 S/m) in deep mantle wedge regions extensively detected by magnetotelluric studies are often associated with the presence of fluids released from the progressive dehydration of subducting slabs. Epidote minerals are the Ca‐Al‐rich hydrous silicates with huge stability fields exceeding those of amphibole (>70–80 km) in subducting oceanic crust, and they may therefore be transported to greater depth than amphibole and release water to the mantle wedge. In this study, the electrical conductivities of epidote were measured at 0.5–1.5 GPa and 573–1273 K by using a Solartron‐1260 Impedance/Gain‐Phase Analyzer in a YJ‐3000t multianvil pressure within the frequency range of 0.1–106 Hz. The results demonstrate that the influence of pressure on electrical conductivity of epidote is relatively small compared to that of temperature. The dehydration reaction of epidote is observed through the variation of electrical conductivity around 1073 K, and electrical conductivity reaches up to ~1 S/m at 1273 K, which can be attributed to aqueous fluid released from epidote dehydration. After sample dehydration, electrical conductivity noticeably decreases by as much as nearly a log unit compared with that before dehydration, presumably due to a combination of the presence of coexisting mineral phases and aqueous fluid derived from the residual epidote. Taking into account the petrological and geothermal structures of subduction zones, it is suggested that the aqueous fluid produced by epidote dehydration could be responsible for the anomalously high conductivities in deep mantle wedges at depths of 70–120 km, particularly in hot subduction zones.

Effect of chemical composition on the electrical conductivity of gneiss at high temperatures and pressures

“…Aside from the chemical composition, other available alternative causes for high conductivity anomalies can be considered, such as water in nominally anhydrous minerals (Wang et al, 2006;Yang, 2011;Karato, 2009, 2014a), interconnected saline (or aqueous) fluids (Hashim et al, 2013;Shimojuku et al, 2014;Sinmyo and Keppler, 2017;Guo et al, 2015;Li et al, 2018), partial melting (Wei et al, 2001;Maumus et al, 2005;Gaillard et al, 2008;Ferri et al, 2013;Laumonier et al, 2015Laumonier et al, , 2017Ghosh and Karki, 2017), interconnected secondary high conductivity phases (e.g., FeS, Fe 3 O 4 ; Jones et al, 2005;Bagdassarov et al, 2009;Padilha et al, 2015), dehydration of hydrous minerals (Wang et al, 2012(Wang et al, , 2017Manthilake et al, 2015Manthilake et al, , 2016Hu et al, 2017;Sun et al, 2017a, b;Chen et al, 2018) and graphite films on mineral grain boundaries (Freund, 2003;Pous et al, 2004;Chen et al, 2017). In consideration of the similar formation conduction and geotectonic environments, the Himalaya-Tibetan orogenic system was compared with the Dabie-Sulu UHPM belt and explained high electrical conductivity anomalies.…”

“…According to magnetotelluric (MT) and geomagnetic depthsounding results, the electrical conductivity of geological samples at high temperature and pressure can be used to extrapolate the mineralogical composition and thermodynamic state in the Earth's interior (Maumus et al, 2005;Dai et al, 2008;Hui et al, 2015;Manthilake et al, 2015;Li et al, 2016;Hu et al, 2017). High conductivity anomalies are widely distributed in the middle to lower crust and upper mantle, and there are various causes of these anomalies in different regions (Xiao et al, 2007(Xiao et al, , 2011Pape et al, 2015;Novella et al, 2017).…”

Electrical Conductivity of Clinopyroxene‐NaCl‐H2O System at High Temperatures and Pressures: Implications for High‐Conductivity Anomalies in the Deep Crust and Subduction Zone

“…As shown in Figure 5, the impedance spectra of the Cpx‐H 2 O system can be fitted by the equivalent circuit that was composed of components of CPE S , R S1 , R S2 , and L K . If we refer to the relevant equivalent circuit in previous studies (Barsoukov & Macdonald, 2005; Hu et al, 2017; Wang & Karato, 2013), we see that CPE S represented the constant‐phase element of the sample, the sum of R S1 and R S2 represented the sample resistance, and L K was the inductive reactance in the equivalent circuit. It has been proposed that the low‐frequency inductive behavior hardly affected the high‐frequency arc that was related closely to the sample conduction (Bai & Conway, 1991; Hu et al, 2017; Macdonald, 1978; Wang & Karato, 2013).…”

“…Although the chemical compositions of fluxing fluids in the Earth's interior cannot be obtained accurately by aqueous fluid inclusions, the main solutes in the aqueous fluids can be determined. Free water in the Earth's interior can derive from hydrous mineral dehydration (e.g., amphibole, chlorite, epidote, and antigorite) (Hu et al, 2017, 2018; Manthilake et al, 2016; Wang et al, 2017) and seawater in the pores of subduction plates (Holland & Ballentine, 2006; Sharp & Barnes, 2004). Solutes are provided by minerals, seawater, and mantle gasses (Franck & Bounama, 1997; Shmulovich et al, 2001).…”

Response to reviewer 1

“…According to magnetotelluric (MT) and geomagnetic depthsounding results, the electrical conductivity of geological samples at high temperature and pressure can be used to extrapolate the mineralogical composition and thermodynamic state in the Earth's interior (Maumus et al, 2005;Dai et al, 2008;Hui et al, 2015;Manthilake et al, 2015;Li et al, 2016;Hu et al, 2017). High conductivity anomalies are widely distributed in the middle to lower crust and upper mantle, and there are various causes of these anomalies in different regions (Xiao et al, 2007(Xiao et al, , 2011Pape et al, 2015;Novella et al, 2017).…”