Search

In the study of subduction zone processes, there has been much debate on slab dehydration against slab melting, as well as on flux melting of the mantle wedge against metasomatism and a prolonged period of storage and heating before mantle melting. H2O‐saturated solidus of rock is critical to the debate. Unfortunately, there is a significant discrepancy in H2O‐saturated solidus between different experimental studies, mainly due to the difficulty in distinguishing hydrous melt from aqueous fluid in quenched experimental products. In this study, we developed a new approach to locating H2O‐saturated solidus of rock systems by monitoring the electrical conductivity of the system in situ. A mixture of albite and >10 wt% H2O was heated stepwise in the range of 0.35–1.7 GPa in a piston cylinder apparatus. A jump in electrical conductivity (up to 1.8 log units) was observed over a small temperature interval (10–20°C). The temperature corresponding to the electrical jump decreased from 790 ± 10°C at 0.35 GPa to 640 ± 10°C at 1.7 GPa, in good agreement with previously reported H2O‐saturated solidus of albite. Examination of the experimental products quenched immediately after observation of the electrical conductivity jump confirmed the occurrence of albite melting. Our experimental results suggest that the electrical conductivity jump serves as a good indicator for rock melting under H2O‐saturated conditions, and this in situ approach can overcome the ambiguity in phase identification in quench experiments. This new approach has the potential to solve the discrepancy over H2O‐saturated solidus of a variety of rock systems and shed light on subduction zone processes. Plain Language Summary: To understand petrological processes at subduction zones, it is important to know at what temperature a rock starts to melt with plenty of water available (i.e., the H2O‐saturated solidus). However, in the products from high‐pressure quench experiments, it is often difficult to tell whether the rock system has melted or not. Not surprisingly, there are significant discrepancies between previous studies on the location of H2O‐saturated solidus, especially for peridotite. By using the albite‐H2O system as a test system, we show that melting can be recognized in situ by an electrical conductivity jump, although the jump becomes smaller toward high pressure. The solidus temperatures determined by our approach are in good agreement with widely accepted values in literature. With further improvement, this new approach can be used to solve the discrepancy over H2O‐saturated solidus of different rock systems and shed light on subduction zone processes. Key Points: A new in situ method of measuring H2O‐saturated solidus is developedMelting of albite corresponds to a jump of electrical conductivityThe amplitude of electrical conductivity jump reduces with pressure [ABSTRACT FROM AUTHOR]

Amphibolite- and granulite-facies metamorphic rocks are common in the eastern Himalayan syntaxis of southeastern Tibet. These rocks are composed mainly of gneiss, amphibolite and schist that underwent various degrees of migmatization to produce leucogranites, pegmatites and felsic veins. Zircon U–Pb dating of biotite gneiss, leucocratic vein and vein granite from the syntaxis yields consistent ages of ∼49 Ma, indicating crustal anatexis during continental collision between India and Asia. Garnets in these rocks are categorized into peritecitc and anatectic varieties based on their mode of occurrence, mineral inclusions and major- and trace-element zoning. The peritectic garnets mainly occur in the biotite gneiss (mesosome layer) and leucocratic veins. They are anhedral and contain abundant mineral inclusions such as high-Ti biotites and quartz, and show almost homogeneous major-element compositions (except Ca) and decreasing HREE contents from core to rim, indicating growth during the P - and T -increasing anatexis. Peak anatectic conditions at 760–800°C and 9–10·5 kbar are well constrained by phase equilibrium calculations, mineral assemblages, and garnet isopleths. In contrast, anatectic garnets only occur in the vein granite. They are round or subhedral, contain quartz inclusions, and exhibit increasing spessartine and trace-element contents from core to rim. The garnet–biotite geothermometry and the garnet–biotite–plagioclase–quartz geobarometry suggest that the anatectic garnets crystallized at ∼620–650°C and 4–5 kbar. Some garnet grains show two-stage zoning in major and trace elements, with the core similar to the peritectic garnet but the rim similar to the anatectic garnet. Mineralogy, whole-rock major- and trace-element compositions and zircon O isotopes indicate that the two types of leucosomes were produced by hydration (water-present) melting and dehydration (water-absent) melting, respectively. The leucocratic veins contain peritectic garnet but no K-feldspar, have lower whole-rock K2O contents and Rb/Sr ratios, higher whole-rock CaO contents and Sr/Ba ratios, and show homogeneous δ18O values that are lower than those of relict zircons, indicating that such veins were produced by the hydration melting. In contrast, the vein granite contains peritectic garnet and K-feldspar, has higher whole-rock K2O contents and Rb/Sr ratios, lower whole-rock CaO contents and Sr/Ba ratios, and shows comparable δ18O values with those of relict zircons, suggesting that this granite were generated by the dehydration melting. Accordingly, both hydration and dehydration melting mechanisms have occurred in the eastern Himalayan syntaxis. [ABSTRACT FROM AUTHOR]

To better understand the transport of Mo and W in granitic melts and the formation mechanism of porphyry ore deposits, we have investigated the diffusivities of Mo and W in granitic melts with 0.04–5.1 wt% H2O at 1000–1600 °C and 1 GPa using a diffusion couple approach and a Mo saturation approach with Mo sheet serving as the source. The Mo and W diffusivities obtained from diffusion profiles measured by LA-ICP-MS can be described as: D Mo , anhy = 10 − 1.47 ± 0.73 exp [ − (387 ± 25) / R T ] , D W , anhy = 10 − 1.28 ± 1.05 exp [ − (396 ± 35) / R T ] , D Mo , 2.7 wt%H 2 O = 10 − 5.37 ± 0.52 exp [ − (211 ± 18) / R T ] , D Mo , 5.1 wt%H 2 O = 10 − 6.87 ± 0.69 exp [ − (133 ± 20) / R T ] , $\begin{align}& {{D}_{\text{Mo},\text{anhy}}}={{10}^{-1.47\pm 0.73}}\exp \left[ -{\left(387\pm 25 \right)}/{\text{R}T}\; \right], \\ & {{D}_{\text{W},\text{anhy}}}={{10}^{-1.28\pm 1.05}}\exp \left[ -{\left(396\pm 35 \right)}/{\text{R}T}\; \right], \\ & {{D}_{\text{Mo},2.7\text{wt }\text{%}{{\text{H}}_{2}}\text{O}}}={{10}^{-5.37\pm 0.52}}\exp \left[ -{\left(211\pm 18 \right)}/{\text{R}T}\; \right], \\ & {{D}_{\text{Mo},5.1\text{wt }\text{%}{{\text{H}}_{2}}\text{O}}}={{10}^{-6.87\pm 0.69}}\exp \left[ -{\left(133\pm 20 \right)}/{\text{R}T}\; \right], \end{align}$ where D is diffusivity in m2/s (with the subscripts denoting water contents and "anhy" representing nominally anhydrous melt), R is the gas constant, T is the temperature in K, and the activation energies in the exponential are in kJ/mol. When the influence of H2O is incorporated, Mo diffusivity in granitic melts with <5.1 wt% H2O can be modeled as: log D Mo = − (1.94 ± 1.58) − (0.87 ± 0.36) w − [ (19341 ± 2784) − (2312 ± 620) w ] / T $log {{D}_{\text{Mo}}}=-\left(1.94\pm 1.58 \right)-\left(0.87\pm 0.36 \right)w-{\left[ \left(19341\pm 2784 \right)-\left(2312\pm 620 \right)w \right]}/{T}\;$ where w is H2O content in the melt in wt%. The diffusion behavior (low diffusivities, high activation energies, and strong H2O effects) of Mo and W indicates that they exist and diffuse in the melt in the form of hexavalent cations. Their low diffusivities imply that the bulk concentrations of Mo and W in exsolved hydrothermal fluid and those in the melt are probably not in equilibrium. However, because of the large fluid-melt partition coefficients of Mo and W, they can still be enriched in the hydrothermal fluid, although to a lesser extent than equilibrium partitioning would allow. Slow Mo and W diffusion can be a significant rate-limiting step for the formation of porphyry Mo/W deposits. [ABSTRACT FROM AUTHOR]