Real-Time Measurements of Orthclase Dissolution

Background: Despite the extensive investigation of feldspar dissolution processes in the past several decades, the dissolution mechanisms remain controversial, and a number of fundamental issues have yet to be properly resolved. For instance: Why is dissolution non-stoichiometric at acidic pH and stoichiometric at alkaline pH (near room temperature)? What is the nature of the non-stoichiometric layers? What is the relative reactivity of the difference sites exposed at a feldspar mineral surface? We have addressed these questions using in-situ X-ray reflectivity and atomic force microscopy.

AFM results reveal distinct processes involving mainly terrace roughening and pitting with no significant step reactivity at pH = 1.1, versus step motion with little reactivity on terrace areas at pH = 12.9.  This observation immediately implies that the distribution of primary reactive sites are strongly pH dependent.  Separate measurements showed that a gel-like surface coating formed at acidic pH at slow fluid flow-rates.  This coating could be removed by increasing the flow rate implying that the widely observed nonstoichiomietry of reaction is not a result of differential leaching of mineral components but instead is due to the deposition of a weakly bound film consisting of the less soluble reaction products.

Real-time X-ray reflectivity data exhibits an oscillatory behavior at both acidic and alkaline pH values, due to the removal of individual orthoclase layers.  The oscillatory reflectivity immediately implies that two distinct reactive sites (e.g., terrace and step sites) are found on the surface in each pH regime.  While dissolution at pH 12.9 is fully stoichiometric and dominated by lateral dissolution processes producing a layer-by-layer dissolution process, at pH 1.1 we observe a very different damped oscillatory pattern, indicative of a more random dissolution process in which the orthoclase surface is substantially disrupted and roughened.  Additional measurements revealed that the surface remained stoichiometric during dissolution at pH 1.1 except for the outermost unit cell (6.5 Å).

How does X-ray reflectivity probe dissolution processes in real-time?

Measurements are made at the "anti-Bragg" condition, specified by Bragg's Law:

nl = 2d sin(a)

Where l is the x-ray wavelength, d= 6.459 Å is the spacing lattice planes, a is the angle of incidence, and n = 1/2 for the anti-Bragg condition. At this condition, x-rays reflected from neighboring terraces are out of phase and destructively interfere (see figure above).  Therefore the reflectivity is maximized when the surface is very smooth, and minimized when the surface is covered with a half-occupied layer.  Therefore, each oscillation in the reflectivity as a function of time at this condition corresponds to the removal of an individual orthoclase layer. 

The reflectivity is determined by a "rocking scan" at regular intervals (these measurements are performed at pH = 12.9 at 73 °C.  Measurements at the anti-Bragg condition (n = 1/2; open circles, above) show strong oscillatory behavior corresponding to the removal of individual layers (noted by 1 ML and 2 ML).  At scattering conditions that are insensitive to the surface termination (e.g., n = 1.97; filled diamonds, above) show no changes are observed, as expected. 

Reference: Fenter et al. Geochim. Cosmochim. Acta, 2003, 67, 197-211

back to highlights...