Hydrothermal origin of heavy potassium isotope compositions in altered oceanic crust: Implications for tracing the elemental cycle

Abstract

Stable potassium isotopes are easily fractionated by fluid-rock interactions and provide a tool to probe the role of hydrothermal alteration of the oceanic crust in modifying the chemical composition of seawater and, through subduction, the deep Earth. An in-depth understanding of K isotope fractionation during high-temperature hydrothermal oceanic crust alteration is necessary to refine the use of K isotopes to track geochemical processes. This study investigated a suite of samples of altered upper oceanic crust, that formed at the East Pacific Rise and is exposed at the Hess Deep, to constrain K isotope fractionation by combining mineralogical, elemental, isotopic, and synchrotron-based spectroscopic approaches. Relative to the protolith, isotopically heavy K is enriched in the sheeted dike complex (δ41K of -0.48 ± 0.15 to 0.14 ± 0.13‰, ave. -0.22‰) where the rocks are generally depleted in K through high-temperature (hundreds of °C) fluid-rock reaction. In contrast, samples from the lava/dike boundary (-0.59 ± 0.14 to -0.48 ± 0.12‰, ave. -0.54‰) and lavas (-0.54 ± 0.10 to -0.42 ± 0.09‰, ave. -0.46‰), commonly exhibit net enrichment of K relative to the protolith due to low-temperature (tens of °C) fluid-rock reaction and have δ41K similar to fresh basalts. We interpret the observed isotopic variation in the sheeted dikes as being primarily due to one or both of the following: (i) diffusive K loss, with 39K diffusing faster than 41K, and (ii) neoformation of albitic plagioclase, which is identified as an important 41K host in altered sheeted dikes. Using isotope mass balance calculation, we show that isotopic fractionation occurring during high-temperature alteration may be difficult to identify in hydrothermal vent fluids because the δ41K of the K leached from the rocks is similar to that of the protolith. Hydrothermal K outputs to the ocean may not contribute significantly to elevating the δ41K of modern seawater relative to the silicate Earth. In contrast to upper oceanic crust altered at low-temperature and seafloor sediments, subducting ocean crust altered at high-temperature adds rocks with δ41K higher than the protolith to the mantle. Because the high δ41K value of the sheeted dikes is generated during the loss of K from these rocks, hence expressed most strongly in K-depleted samples, tracking this signature in arc magma and the mantle will be challenging.