MULTI-LENGTHSCALE STUDY OF GROWTH FEATURES ON QUARTZ SURFACES

Accurately predicting the porosity in clastic reservoirs requires a quantitative model for quartz precipitation kinetics. The rates used in reservoir quality models are determined by laboratory studies of quartz precipitation and dissolution. However, these studies derive rates from measurements of changes in bulk solution composition, not the evolution of the growing crystal surface. Uncertainties in the precise relationship between the two may be a source of significant mismatch between field and laboratory-derived rates.

We have employed a combination of atomic force microscopy (AFM) and vertical scanning interferometry (VSI) to study the development of surface topography of growing quartz surfaces over nanometer to millimeter length scales. We used gem-quality, single quartz crystals (Herkimer diamonds, Herkimer County, NY), which provide large, crystallographically well-defined growth faces. Crystals were exposed to varying aqueous silica supersaturations at 100 ºC in batch autoclaves. Growth is sufficiently slow such that the supersaturation does not change significantly during the experiments.

Our AFM and VSI observation of surface features such as steps, hillocks, and two-dimensional growth “pancakes” permit the quantification of their role in the overall growth process, and lay the framework for development of a mechanistic understanding of the growth kinetics. For example, at, 5-times quartz saturation our observations reveal that growth occurs above the thermodynamic surface-roughening transition. Surprisingly, we note that the “pristine” natural surfaces also exhibit such growth features indicating that Herkimer quartz surfaces are growth surfaces, not equilibrium surfaces (apparently preserved for 300 Ma). Our mineral dissolution studies have shown that it is key to observe the mineral-water interface kinetics over a range of time and length scales. We are applying a similar approach to derive a quantitative model of quartz growth kinetics.