GRANTEE: YALE UNIVERSITY

Department of Geology and Geophysics

New Haven, Connecticut 06511

GRANT: DE-FG02-90ER14153

PERSON IN CHARGE: A. C. Lasaga (203-432-3114; Fax 203-432-3134) and D. M. Rye (203-432-3174)

Objectives: The objectives are to integrate new results and to develop new techniques along four directions: (a) theoretical modeling of coupled fluid flow-chemical reactions, (b) experimental and theoretical studies of water-rock reactions, (c) the collection and interpretation of stable isotopic data from fluid flow systems at the regional scale, the vein scale, and the mineral grain scale in several geological environments (d) geochemical field based studies of transport and reaction in environments ranging from soils to sedimentary basins to high grade metamorphic rocks.

Project Description: The main thrust of the theoretical modeling of coupled fluid flow and chemical reactions, involving kinetics, has been to understand the differences between equilibrium, steady- state, and non-steady-state behavior of the chemical evolution of open fluid-rock systems. These differences have not been fully appreciated in previous models. The computer codes developed in this project numerically treat multi-component, finite-rate reactions combined with advective and dispersive transport in one, two and three dimensions and incorporate isotopic exchange as well as heat and mass transfer. The codes also model flow and reaction along both fracturess and porous media simultaneously.

Experimental work has focused on obtaining the kinetic rate laws of pertinent silicate-water reactions. Theoretical studies have used ab-initio quantum mechanical techniques to obtain the kinetics and mechanisms of silicate surface reactions.

Stable isotope studies have focused on new sampling strategies and techniques. These strategies and techniques provide data sets that can be input into transport and reaction models to determine the length and time scales relevant to fluid transport in a wide variety of geologic settings. These settings include: 1) the Wepawaug metamorphic schist, 2) the Irish base-metal sediment hosted ore system, 3) the Olympics accretionary wedge, 4) and the Dalradian metamorphic complex in Scotland.

Geochemical field based studies have been carried out in the Pijiguaos bauxite deposit in Venezuela, in the Columbia River flood basalts, and in each of the stable isotope study areas. The field data, geochemical data, isotopic data, and experimental and theoretical rate data are used as input to or constraints on the numerical models.

Results: We have embarked on a modeling effort which should revolutionize our understanding of isotope systematics in the Earth's crust. Our effort is the first attempt ever to consistently model the simultaneous effects of diffusion within grains, along with dissolution and precipitation (or recrystalization) and fluid flow (Bolton, Lasaga, Rye, and Chakraborty, 1997). We have examined both closed and open systems, and are comparing these model results to experimental and field studies. For open systems, we also solve fluid and solute transport equations with advection, diffusion, and source/sink terms at the fluid/grain boundaries, and we are currently coupling these detailed isotopic models with our other models of reactive flow and transport. This treatment represents a major step forward, as we will be able to link the kinetics of mineral exchange with that of isotopic exchange. The results with both diffusion and dissolution/precipitation acting in combination are quite different from a simple addition of their individual effects, and a host of characteristic time scales have been found which influence crustal fluid properties. The results of our comprehensive isotope model underscore the importance of such a unified approach which links the kinetics of both mineral dissolution/precipitation and the exchange of isotopes.

We are also in the midst of exploring the dynamics of flow and multimineralic reaction systems in weathering environments. The present models include dissolution and precipitation of the minerals albite, quartz, gibbsite, kaolinite, and paragonite. Our new flow and reaction code allows us to examine more complex feedbacks (Bolton, Lasaga and Rye, 1997c) as a nonisothermal extension of Soler and Lasaga (1996) (i.e. bauxites in Venezuela) to a heterogeneous porous medium. The newly developed subgrid-scale grain models can handle the full range of porosity and permeability variations and heterogeneity which were inaccessible to previous models. These new capabilities are essential for comparisons to realistic field scenarios.

A new two-dimensional kinetic model of general applicability to a range of contact and regional metamorphic systems (Luettge, Bolton, Rye, and Lasaga, 1997) has been recently developed. Our model solves for transport and reactions at metamorphic conditions with supercritical H₂O-CO₂ mixtures in a heterogeneous permeability medium. This new model enables us to examine features inaccessible to any previous study, such as the variation of T-X stability fields with pressure, the influence of high-permeability zones, and multiple trajectories of fluid parcels in complex flow fields (each with its own T-X_CO evolution). Depending on the kinetic rates, our calculated T-X_CO paths often do not follow the equilibrium curves to the isobaric invariant points. Although mineral assemblages produced may be the same as those expected on the basis of equilibrium, equilibrium curves can be overstepped leading to a major breakdown in the usual interpretation of metamorphic isograds, and a host of other traditional geologic assumptions. Combining these new calculations with experimental observations is leading to a paradigm shift in our understanding of reaction pathways in both laboratory and field environments. This requires a view of petrology which includes the essential role of metastable reactions, which are triggered by overstepping the equilibrium-buffered reactions. In short, our current two-dimensional model greatly extends the simpler treatment of Lasaga and Rye (1993), and underscores the fact that in order to understand the petrologic processes we have to consider the nonequilibrium reaction pathway, not just the mineral assemblages.

Our recent modeling of kinetically controlled mineral dissolution and precipitation revealed features never before observed. For quartz as the representative mineral phase, these include: 1) regions of downwelling oversaturated fluid experiencing heating, and regions of upwelling undersaturated fluid experiencing cooling, neither of which would be expected from models based on equilibrium, 2) significant shifts in flow direction due to precipitation/dissolution induced permeability changes, 3) the location of the basal stalk of thermal plumes rising from the heated lower boundary is inherently unstable. This stalk migrates with time, as the core of the flow generally clogs via precipitation, whereas the edges of the stalk are dissolving, via kinetic effects, 4) Although one could expect runaway dissolution to occur in downwelling up-temperature flow, by the sequence undersaturation 6 dissolution 6 higher permeability 6 faster flow 6 greater disequilibrium 6 etc., runaway dissolution can be moderated by shifts in the locations of saturation state reversals. However, the runaway regime is still observed when the kinetics are faster or the grain sizes are smaller than the standard test case (Bolton, Lasaga and Rye, 1997b).

In forced flux injection simulations, undersaturated injection leads toward permeability homogenization along the flow direction, whereas oversaturated injection tends to increase permeability heterogeneities along the flow direction. Flow rates are significantly enhanced even between isolated high permeability zones, an effect which is even more dramatic for both closer "crack" spacing and higher permeability contrasts (Bolton, Lasaga and Rye, 1997a).

There are many factors that govern the rate of growth and dissolution of minerals in contact with fluids in the crust. Therefore, a very large effort has been made in recent years to formulate a general framework that can be used to extract the overall rate laws for the geologically important minerals. In particular, this effort has found that the structure and dynamics of the mineral surface must be fully understood to make substantive progress towards the kind of rate law needed in the modern-day hydrogeologic, hydrothermal, and weathering models. The kinetic study of mineral surfaces requires a combination of kinetic theory, macroscopic experiments, field observations and atomic scale observations. While the research in our own group has been involved in all these aspects, the atomic scale observations of mineral surface dynamics needed to make fundamental changes in our kinetic paradigms has been quite limited. In particular, the study of common minerals with "slow" rates of reactions has been very meager.

Most of our present kinetics experiments use conventional flow-through reaction cells and mineral powders. The use of mineral powders gives the advantage of reaction at a large number of surfaces of the mineral, thereby providing an average reaction rate. The factors in the general rate law can be tested without concern for the heterogeneity of mineral surface sites. However, these factors can be studied in more detail through direct observation of the mineral surface in the process of reacting.

Phase shift interferometry (PSI) is a relatively new technique that provides ~1 nm height resolution with a lateral resolution on the order of ~1 micron. Phase shifted interferograms are recorded and used to calculate the phase map of the surface, which is related to the surface topography.

The near-atomic scale hydrothermal system is the first of its kind. The equipment opens up a whole new area of information on the dynamics of features such as steps, etch pits and spirals on the surfaces of many of the important silicates. We have studied the dissolution process of anorthite as a function of pH (Luettge, MacInnis, and Lasaga, 1997). Our initial measurements were made at 25EC. We are now focussing on measurements at hydrothermal conditions (300 oC and 200 bars). The dissolution rates at specific locations on the anorthite surface were measured directly from the rate of retreat of the surface (using a reference surface positioned directly on the sample).The development of etch pits in initially flat regions of the surface is an important mechanism of the dissolution process, at least at low pH values. At pH 3, the rate of retreat normal to the anorthite surface is in the same range that can be calculated from Oelkers and Schott (1995) based on bulk rates. Lateral dissolution rates can be an order of magnitude faster than surface normal rates. These results elucidate the expected complexity of the dissolution process and fill a very important gap in our understanding of reactive surface area.

In the Wepawaug Schist, the application of the experimental data to the kinetics of isotopic exchange of water with quartz, and the model results has already shown that fluid flow was transient in cracks (now preserved as veins). Not only was the flow transient, the isotopic composition of the fluid oscillated between values that were isotopically lighter and heavier than the surrounding host rocks. The above mentioned kinetic isotopic modeling effort has direct application to the field and isotopic work completed on the Wepawaug Schist.

Oxygen isotopic zonation in garnets as well as isotopic, modal and chemical profiles for individual minerals in wall rocks next to veins show that each fluid recorded in the veins infiltrated and reacted with the wall rocks. Garnets are isotopically zoned with the cores being isotopically lighter than the rims. Staurolite and kyanite throughout the profile formed late and are isotopically heavy. Quartz, biotite, and muscovite have intermediate isotopic compositions, and plagioclase is isotopically light. These results make it clear that we can no longer consider metamorphic rocks to be "closed or semi-closed" systems. Regional metamorphic rocks are, in many cases, metasomatic in origin. We can never go back to looking at whole rocks or single minerals, and we can never go back to looking at a single locality. However, the rocks do preserve some of the fluid flow history, and we can unravel that history.

Fluid inclusion, stable isotopic and Pb isotopic data from ore deposits and from veins within basement rocks in Ireland have shown for the first time that fluid flow in faults and cracks of basement rocks supplied most of the metals in the Irish Deposits. These results represent a real paradigm shift, as the deposits are generally believed to have formed strictly by basin fluid flow processes (Everett, Wilkinson, and Rye, 1997).

In spite of the widespread use of Sr isotopes as proxies for chemical weathering rates, almost no work has been done adequately linking the release of radiogenic Sr with mineral dissolution rates. We have begun a project to establish a quantitative relationship between the kinetics of silicate dissolution and the release of radiogenic Sr into weathering solutions. This work consists of both an experimental and a field component. Column dissolution experiments have been performed on biotite and phlogopite to determine kinetic rate laws for both mineral dissolution and the release of ⁸⁷Sr. We found that under far from equilibrium conditions ⁸⁷Sr is released from biotite about twice as fast as the overall mineral dissolution rate. For phlogopite, on the other hand, ⁸⁷Sr is released 100 times as fast the overall mineral dissolution rate. These results represent the first ever rate constants for the release of Sr isotopes during silicate weathering and suggests that sheet silicates may greatly influence the ⁸⁷Sr/⁸⁶Sr of weathering solutions. This conclusion is further supported by our measurements of the ⁸⁷Sr/⁸⁶Sr ratio in our output solutions. While it has been generally assumed that weathering solutions in contact with biotite and phlogopite will initially have a high ⁸⁷Sr/⁸⁶Sr, our experiments reveal that the ⁸⁷Sr/⁸⁶Sr ratio of the output solution is initially low. Once the mineral has reached steady state, far from equilibrium conditions, however, the Sr isotope ratio of the output solution increases to several times that of the whole mineral. These high ⁸⁷Sr/⁸⁶Sr ratios agree with the rapid release rates described above for ⁸⁷Sr. Clearly the relatively rapid release of highly radiogenic Sr from biotite and phlogopite could raise the ⁸⁷Sr/⁸⁶Sr ratio of weathering solutions above what one would expect if Sr release were stoichiometric during silicate dissolution. Our data demonstrate that typical assumptions about the nature of Sr isotope release from sheet silicates are incorrect and that only through careful experimental work can one adequately quantify the relationship between silicate dissolution and Sr isotopes.

The second aspect of this study involves measuring weathering rates and Sr release rates in the field. Our modeling has suggested that on a geologic time scale, basalt may play an important role in determining the Sr isotope chemistry of the oceans. This role has generally been overlooked in models using Sr isotopes to describe global weathering rates because the silicate portion of the continents is typically assumed to be entirely granitic. To test the hypothesis that basalts can significantly influence marine Sr isotope chemistry, we are comparing basalt and granite weathering rates in an area where all climatic variables are held constant. This work is being performed in northern Idaho where the Columbia River Basalts and the Idaho Batholith are both exposed within 15 kilometers of each other. The results from our initial field season suggest that the long term (12,000 years) average weathering rate for basalts is about 4 times as fast as that of non-basaltic rocks. The stream data from this area, however, suggests that the modern basalt weathering rates are about twice as fast as those observed in non-basaltic terrains. These results indicate that basalt weathering may indeed be an important control over the marine Sr isotope record. Further work was conducted last Summer to expand the field area, resample many of the streams and to consider the effect of various lithologies on lake chemistry. The analysis of these new samples is an ongoing part of this project. Clearly developing a quantitative relationship between Sr isotope release and silicate dissolution rates through both a field and an experimental approach will allow us to use Sr isotopes more effectively as tracers of chemical weathering rates both in the geologic past and in modern catchments. (cf. Taylor, Lasaga, Blum, and MacInnis, 1997, Taylor and Lasaga, 1997). Bolton, E.W., A.C. Lasaga and D. Rye, 1996, A model for the kinetic control of quartz dissolution and precipitation in porous media flow with spatially variable permeability: Formulation and examples of thermal convection, Journal of Geophysical Research, 101, 22157-22187. Bolton, E.W., A.C. Lasaga and D. Rye, 1997a, Dissolution and precipitation via forced flux injection in a porous medium with spatially variable permeability: Kinetic control in two dimensions, Journal of Geophysical Research, 102, 12159-12171. Bolton, E.W., K.A. Maasch and J. M. Lilly, 1995, A wavelet analysis of Plio-Pleistocene climate indicators: A new view of periodicity evolution, Geophysical Research Letters, 22, 2753-2756. Fehan, J.G., 1997, Finite strain and fluid flow in accretionary wedges, northwest Washington state, Ph.D. thesis, Yale Univ. Ganor, J., Mogollon, J.L., Lasaga, A.C., 1995, The effect of pH on kaolinite dissolution rates and on activation energy, Geochim. et Cosmochim. Acta, 59, pp. 1037-1052.Lasaga, A.C., 1995, Fundamental approaches in describing mineral dissolution and precipitation rates, Rev. Mineral., 31, 23-86.   Luettge, A., MacInnis, I.N., and Lasaga, A.C., 1997, Kinetics of anorthite dissolution - A new surface technique using interferometry for in situ measurements under hydrothermal conditions, in the Proceedings of the fifth International Symposium on Hydrothermal Reactions (ISHR '97), edited by D.A. Palmer and D.J. Wesolowski, Gatlinburg, TN, USA, 61-64. Mogollon, J.L. Ganor, J., Soler, J., Lasaga, A.C.,1996, Column experiments and the dissolution rate law of gibbsite, Am. Jour. Sci., 296, 729-765. Soler, J.M., 1997, Coupled Reaction-Transport Modeling of Bauxite Formation: Application to the Los Pijiguaos Bauxite deposit (Venezuela), Ph.D Thesis, Yale University, New Haven, Connecticut. Soler J. M. and Lasaga A. C., 1996, A mass transfer model of bauxite formation. Geochim. Cosmochim. Acta, 60, 4913-4931. Soler J. M. and Lasaga A. C., 1996, Two-dimensional modeling of bauxite formation using the GIMRT software for reactive transport. Application to the "Los Pijiguaos" bauxite deposit (Venezuela), Proceedings of the Fourth International Symposium on the Geochemistry of the Earth's Surface. Ilkley(England), 662-665. Tanaka, N., Xiao, Y., and Lasaga, A. C., 1995. Ab Initio study on carbon kinetic isotope effect (KIE) in the reaction of CH4 + Cl, Journal of Atmospheric Chemistry, 23, 37-49. Van Haren, J. L. M., Rye, D. M., and Ague, J. J., 1996, Oxygen isotope record of channelized and pervasive fluid infiltration during regional metamorphism of pelitic schist, south-central Connecticut, USA: Geochimica et Cosmochimica Acta,60, 3487-3504. Xiao, Y. and Lasaga, A.C., 1996, Ab initio quantum mechanical studies of the kinetics and mechanisms of quartz dissolution: OH- catalysis, Geochim. Cosmochim. Acta, 60, 2283-2295.