Rate limitations of nano-scale weathering front advance in the slow-eroding Sri Lankan Highlands

Abstract

In the Sri Lankan Highlands denudation and chemical weathering represent the low-end member in global weathering rates. Here we report on the causes for these low rates in corestones from a highly weathered saprolite profile. By using electron microprobe and transmission electron microscopy (TEM) analyses, we investigated weathering reactions, and derived rates of pyroxene and biotite oxidation. High-resolution TEM analyses on primary minerals showed that the initial weathering products are non-crystalline and that these form the precursors of secondary minerals (kaolinite and smectite). The dissolution of primary minerals is characterised by sharp reaction fronts in the absence of chemical gradients, hence, dissolution can best be described by a dissolution – reprecipitation process. The first observable weathering reaction is the oxidation of structural Fe(II) in pyroxene and biotite. This oxidation is restricted to distinct zones within the minerals without the formation of oxidized layers. While the oxidation is not accompanied by chemical changes, the oxidation of structural Fe(II) in biotite may cause lattice distortion. Pyroxene and biotite oxidation rates were calculated at the corestone-scale by using the gradient of bulk rock Fe(II)/Fetotal ratios, assuming that oxidation is coupled to the weathering front advance rate. At the mineral-scale, oxidation rates were calculated by using gradients of in situ Fe(II)/Fe(III) ratios measured with electron energy-loss spectroscopy (EELS) within the minerals assuming a coupled oxidation-cation diffusion process. The mineral-scale oxidation rates are significantly higher, log Ominin situ = −11 molmin m−2 s−1, than corestone-scale oxidation rates, log Omin = −13 to −15 mol m−2 s−1. We explain the difference to result from the fact that corestone-scale rates average the oxidation over the entire mineral surface. Because at the corestone-scale pyroxene oxidation rates are also similar to dissolution rates, we infer that oxidation preconditions Fe(II)- bearing primary silicate minerals to weathering. However, although oxidation is initiating chemical weathering in this setting and is limited by the supply of O2, overall conversion of bedrock to saprolite is driven by the formation of porosity by fracture formation during this reaction, which allows for fluid transport and subsequent plagioclase weathering. We conclude that in compact bedrock nm- to µm-scale fracture- and porosity generation are the processes that drive weathering in tectonically quiescent regions. They are also the processes at depth that couple the weathering zone with processes at its surface. Finally, the formation of secondary surface layers at the nm-scale is essential in controlling the equilibrium in the local weathering environment and hence mineral dissolution. These layers are likely also those that control isotope fractionation in the weathering zone.