Reaction Rates and Mechanisms

Laboratory experiments designed to measure the dissolution and precipitation rates of minerals have long been one of the principal tools of ``basic'' research in diagenesis. Over the last two decades, kinetic data have been collected for a number of mineral-water systems, and a growing set of rate laws is now available. The past few years have seen particular emphasis on data for aluminosilicates, as illustrated by the kinetic expression for the smectite to illite transformation described above (Huang et al., 1993; Velde and Vasseur, 1993; Elliott et al, 1992), along with new data for gibbsite and kaolinite (Nagy and Lasaga, 1992). An ongoing challenge for these studies is to take the experimental results and apply them to larger scale, heterogeneous systems typical of most sedimentary basins, as illustrated in a brief discussion by Brantley (1992).

A number of possible factors may cause discrepancies between field estimates and experimental calculations of reaction rates, including: (1) errors in laboratory measurements not made at steady state, (2) the effects of variable solution chemistry, (3) incorrect estimates of mineral surface area, (4) surface wetting characteristics of the system, and (5) variability in temperature between the lab and the field. Recent studies have attempted to address these challenges in two major ways. First, laboratory experiments have been performed under conditions that better reflect the complexities of natural systems, e.g. the role of organic complexing (see Small, 1993; Harrison and Thyne, 1992; and Crossey, 1991, among others). A second approach has been to estimate rates of reaction from the rocks themselves, and compare those to the laboratory experiments. One disadvantage of the latter approach is that few basins have age controls on samples from a range of temperatures and pressures. A basin in which such controls do exist is the North Sea, as illustrated by the above discussions of K/Ar age dating of fibrous illite, and fluid inclusion analysis of quartz cement. Estimated precipitation rates for quartz based on fluid inclusion analyses range from about 1x10 moles/cms at 80C to 5x10 moles/cms at 140C (Walderhaug, 1994b) for solutions that are presumably near saturation with respect to quartz. Comparable rates calculated from laboratory experiments (Rimstidt and Barnes, 1980) are at least two orders of magnitude faster. As our understanding of mineral reaction rates in natural systems improves, it should be possible to better quantify the reasons for these differences.

A large problem faced by researchers trying to develop better models for predicting fluid-rock interactions is the enormous chemical and physical complexity of the interfaces between fluids and minerals. Molecular-scale surface features such as defects, microcracks, and etch pits result in compositional and structural heterogeneities that complicate theoretical models of reactions. In addition, a variety of different crystallographic faces of different minerals are likely to be exposed to complex solutions, and gases containing both inorganic and organic components. Several new developments help shed light on the near-surface properties of minerals and the first few atomic layers of adjacent fluids. These include improved atomic modeling using computational molecular models, new spectroscopic techniques for observing sorption complexes at mineral-water interfaces and characterizing redox reactions, and nanometer-scale imaging of microtopography using the scanning force microscope, among others. Brief reviews of these developments can be found in the papers collected in Kharaka and Maest (1992).