Some remarks on hydrogen-assisted electrical conductivity in olivine and other minerals

Abstract

Electrical conductivity in minerals is sensitive to hydrogen content, and therefore, it is a potentially important property from which one can infer hydrogen (water) distribution in the mantle. However, there has been much confusion in the reported results on hydrogen-assisted conductivity. In this paper, I review the existing experimental observations on hydrogenenhanced electrical conductivity in olivine and other minerals to identify the causes of confusion. Hydrogen loss as well as hydrogen gain could occur during a conductivity measurement at high pressures and temperatures. Particularly important is the unrecognized hydrogen gain during an experiment that could lead to a large degree of error. Many experiments were conducted under the conditions where specimens were super-saturated with hydrogen making the validity of those results unclear. A model for hydrogen loss is developed showing a strategy by which hydrogen loss can be minimized. When one selects the experimental results in which the influence of hydrogen loss/gain are carefully examined, there is no major discrepancy among the results from different laboratories except differences between the results from low and high temperatures. Differences between low-temperature and high-temperature results are caused by the change in conduction mechanism. At low temperatures, conduction is due to the migration of interstitial ("free") proton and is nearly isotropic, whereas conduction at high temperatures is due to the migration of two protons at M-site that is highly anisotropic. There is no evidence for substantial concentration dependence of activation enthalpy. Observed exceptionally large concentration dependence reported by Poe et al. (Phys Earth Planet Inter 181:103-111, 2010) is inconsistent with all other reports and is likely caused by some experimental artifact. Experimental results in the high-temperature regime explain a majority of geophysical observations on the conductivity of the oceanic asthenosphere: partial melting is not needed in most regions and is rather inconsistent with the observations on the matured oceanic mantle. Exception is the asthenosphere near the ridge and/or near the trench where very high conductivity (~0.1 S/m) is reported at the top of the asthenosphere. Partial melting might play some role in these regions. Electrical conductivity in the continental lithosphere cannot be attributed entirely to olivine. An important role of orthopyroxene and/or other minor materials (graphite, sulfide) is needed to explain high conductivity reported in some regions such as Bushveld in South Africa. The largest remaining uncertainty is the degree to which hydrogen affects electrical conductivity in the lower mantle minerals. Determining the influence of hydrogen on electrical conductivity in lower mantle minerals is critical to make progress in understanding the global water circulation.

An Overview of the Experimental Studies on the Electrical Conductivity of Major Minerals in the Upper Mantle and Transition Zone

“…The EC of hydrated olivine single crystals along the three dominant crystallographic directions ([100], [010] and [001]) has been reported by Dai and Karato [35] under conditions of a broad temperature range of 573-1373 K at 4.0 GPa (Figure 7). In the low temperature range (573-900 K), the EC of olivine single crystal with a low ∆H (74 kJ/mol) shows a feeble anisotropy, which is in good agreement with previously obtained conductivity results [12,[93][94][95]. However, within the high temperature range (T > 1000 K), the EC of sample exhibiting a higher activation enthalpy (~130-150 kJ/mol) is obviously anisotropic, which is consistent with the results of high-temperature H-D inter-diffusion in olivine [96].…”

“…Opposed to the previous study, Karato [95,116] has proposed a theory to explain (a) the experimentally observed discrepancy of the activation enthalpies of electrical conductivity and diffusion in H-bearing olivine, and (b) the transition from isotropic behavior of olivine electrical conductivity observed at low-T to anisotropic one at high-T. According to this hybrid model, isotope diffusion coefficient is given by the harmonic mean of the different forms of hydrogen (free proton, one or two protons at M-site, etc.…”

The Effect of Water on Ionic Conductivity in Olivine

“…The above evaluation of C H2O variation relies on the extent of C H2O that can produce measurable changes in conductivity. This is unknown because both conductivity and C H2O are variables, and therefore insignificant dehydration during heating and H 2 O‐reaborption during cooling is still theoretically possible (Karato, 2019). In that case, the C H2O under higher‐temperature conditions for the ionic conduction regime would be slightly underestimated; namely, the conductivity is enhanced by less H 2 O, which further supports our conclusion.…”

Electrical Conductivity of Ti‐Bearing Hydrous Olivine Aggregates at High Temperature and High Pressure

“…The impedance spectroscopy was made using the Solartron 1260 for the frequency range of 10 3 -10 6 Hz (we choose relative high-frequency range to minimize the water loss; see Karato, 2019) and the signal voltage is 1.0 V. In the impedance spectroscopy, we determine both the in-phase and outof-phase response of a sample to the applied AC voltage. Measuring both in-phase (real part of impedance [Z′]) and out-of-phase (imaginary part of impedance [Z″]) response is important because when conduction is due to migration of ions, migrating ions can be accumulated at the In most cases, the electrical conductivity of sample was acquired during the first cycle of decreasing temperature after the peak temperature was reached.…”