Scale dependence of mineral dissolution rates within single pores and fractures

https://doi.org/10.1016/j.gca.2007.10.027Get rights and content

Abstract

The possibility that gradients in concentration may develop within single pores and fractures, potentially giving rise to scale-dependent mineral dissolution rates, was investigated with experimentally validated reactive transport modeling. Three important subsurface mineral phases that dissolve at widely different rates, calcite, plagioclase, and iron hydroxide, were considered. Two models for analyzing mineral dissolution kinetics within a single pore were developed: (1) a Poiseuille Flow model that applies laboratory-measured dissolution kinetics at the pore or fracture wall and couples this to a rigorous treatment of both advective and diffusive transport within the pore, and (2) a Well-Mixed Reactor model that assumes complete mixing within the pore, while maintaining the same reactive surface area, average flow rate, geometry, and multicomponent chemistry as the Poiseuille Flow model. For the case of a single fracture, a 1D Plug Flow Reactor model was also considered to quantify the effects of longitudinal versus transverse mixing. Excellent agreement was obtained between results from the Poiseuille Flow model and microfluidic laboratory experiments in which pH 4 and 5 solutions were flowed through a single 500 μm diameter by 4000 μm long cylindrical pore in calcite. The numerical modeling and time scale analysis indicated that rate discrepancies arise primarily where concentration gradients develop under two necessary conditions: (1) comparable rates of reaction and advective transport, and (2) incomplete mixing via molecular diffusion. For plagioclase and iron hydroxide, the scaling effects are negligible at the single pore and fracture scale because of their slow rates. In the case of calcite, where dissolution rates are rapid, scaling effects can develop at high flow rates from 0.1 to 1000 cm/s and for fracture lengths less than 1 cm. Under more normal flow conditions where flow is usually slower than 0.001 cm/s, however, mixing via molecular diffusion is effective in homogenizing the concentration field, thus eliminating any discrepancies between the Poiseuille Flow and the Well-Mixed Reactor model. The analysis suggests that concentration gradients are unlikely to develop within single pores and fractures under typical geological/hydrologic conditions, implying that the discrepancy between laboratory and field rates must be attributed to other factors.

Introduction

The dissolution of minerals plays a major role in various physical, chemical, and biological processes in nature. Mineral dissolution affects the formation of soils, influences the degradation of radioactive waste and its containers (Spycher et al., 2003) and the subsequent migration of heavy metals and radionuclides to the biosphere (Lovley, 1993, Lovley and Coates, 1997, Chorover et al., 2003), and at larger space and time scales, it regulates the atmospheric concentrations of CO2 (Berner, 1995, Berner and Berner, 1997). Laboratory-measured dissolution rates of many minerals have been consistently found to be several orders of magnitude faster than those observed in the field (White and Brantley, 2003, Maher et al., 2004), although it is not clear in all cases that this represents a true discrepancy in the rate constants as opposed to a failure to take into account the intrinsic differences in chemical and/or physical conditions between laboratory and field settings (Steefel et al., 2005). Such rate discrepancies need to be resolved, however, since they seriously hinder the application of laboratory-measured dissolution rates to natural systems.

A variety of factors have been proposed to contribute to the discrepancies between laboratory and field rates, including the differences in reactive surface area of the fresh and weathered minerals (White and Peterson, 1990, Anbeek, 1993, White, 1995), the effect of reaction affinity (White, 1995, White and Brantley, 2003, Maher et al., 2006), the precipitation rate of secondary clay minerals (Steefel and Van Cappellen, 1990, Alekseyev et al., 1997, Zhu et al., 2004, Maher et al., 2006), and the age of the reacting material (White and Brantley, 2003, Maher et al., 2004). Recent studies have also shown that physical and chemical heterogeneities in soils and aquifers where subsurface flow occurs may contribute to a scale dependence to mineral dissolution rates, and thus potentially to discrepancies between laboratory and field rates (Malmstrom et al., 2000, Malmstrom et al., 2004, Li et al., 2006, Meile and Tuncay, 2006). For example, Li and coworkers (Li et al., 2006, Li et al., 2007) found that variations in the spatial distributions of minerals with differing reactivity can result in the development of concentration gradients of chemical species involved in dissolution reactions, thus leading to erroneous predictions of reaction rates when the scale dependence is not accounted for. More generally, any chemical, physical, or microbiological heterogeneity that results in the formation of concentration gradients in the subsurface can lead to a scale dependence of the rates, and thus potentially to discrepancies between laboratory and field rates.

While the discrepancies between laboratory and field rates cannot be attributed entirely to the effect of physical and chemical heterogeneities, as indicated by studies of weathering rates in physically and chemically homogeneous media (Maher et al., 2006), it is clear that a comparison of lab and field rates requires careful consideration of the inherent differences between the laboratory and natural systems as suggested above. The laboratory measurement of dissolution rates usually employs well-mixed batch or flow through reactors. In these experimental systems, the aqueous phase is stirred rapidly enough that the aqueous phase becomes well-mixed, thus eliminating the effect of transport. In such cases, mineral dissolution is surface-controlled and depends only on the uniform chemistry of the aqueous solution. In natural systems, however, reactions are inevitably subject to the influence of transport via advection, molecular diffusion, and/or dispersion. As such, the mineral dissolution rates are an outcome of coupling between the reaction and transport processes.

In addition to these differences, natural porous or fractured media are also very different from laboratory settings in terms of their structure and heterogeneous nature. For example, natural porous media typically possess hierarchical structures and spatial scales that range from the pore scale, to the continuum scale, and finally to the field scale. The pore scale focuses on individual pores at the spatial scale of tens to thousands of microns, while the continuum scale, often at the scale of millimeters to centimeters, contains a sufficient number of pores that allows the definition of statistically averaged properties of porous media, including porosity and permeability (Bear, 1972). The field scale, often from meters to kilometers, is the scale at which we examine specific processes in natural field settings. Heterogeneities in the physical and chemical properties of porous media exist at all spatial scales. While at the continuum and field scales porous media are often represented by properties and processes “averaged” over a large number of pores, the pore scale is where processes such as flow, transport, and reactions actually take place. As such, pores are the fundamental building blocks of natural porous media and are an important starting point to examine the scaling issue with regard to reaction rates.

The effect of transport, especially diffusion, on mineral reaction rates in natural systems has been discussed previously in the context of transport-controlled versus surface-controlled reaction kinetics and the validity of the local equilibrium assumption (Berner, 1978, Dibble and Tiller, 1981, Casey, 1987, Murphy et al., 1989). These studies did not address the detailed flow patterns developed in single pores and fractures and focused primarily on single component systems so that analytical solutions could be derived for the time evolution of the aqueous concentrations. In this work, we examine the effects of advective and diffusive transport on mineral dissolution that involve multiple species in single pores and fractures, where flow can no longer be described completely by an average Darcy velocity—gradients in flow velocity inevitably exist within individual pores as a result of the fundamental physics. Such flow gradients, as well as the limited rates of diffusive transport, can present conditions where reactions are limited by flow and transport, which are very different from conditions in well-mixed batch or flow through reactors in laboratory settings. We aim to understand the mechanisms that contribute to the rate discrepancies between laboratory and field settings at the single pore and fracture scale and to identify the conditions under which such rate discrepancies become significant.

In this study, laboratory-measured reaction kinetics are applied at the scale of the mineral–water interface where they occur, and then coupled with a rigorous treatment of flow and transport. Although the model for a single pore is highly idealized in terms of its geometry, first-order effects in terms of the coupling of flow, transport, and reaction kinetics are captured. The numerical model used here for flow and transport is verified with an analytical solution. In addition, the coupled representation of flow, transport, and reaction by the model is validated with a microfluidic reactive flow experiment carried out at a spatial scale of hundreds to thousands of microns. We then compare rates within the single pores and fractures obtained from these fully coupled reactive transport simulations to rates obtained from a model that assumes complete mixing within the pore, while maintaining the same reactive surface area, average flow rate, geometry, and multicomponent chemistry.

For single fractures, the effect of transverse mixing is evaluated by comparing the rates from the fully coupled 2D reactive transport simulations to those calculated from a 1D plug flow reactor model that assumes complete transverse mixing within the fracture. The comparisons were made under various flow, pore size, and fracture length conditions to identify and quantify the conditions under which gradients in concentration (and thus, in rates) may develop.

Section snippets

Reactions and rate laws

In this work, we focus on the dissolution of three important subsurface mineral phases with a range of reaction rates: calcite, plagioclase, and iron hydroxide. Calcite dissolution is one of the most important and rapid mineral reactions in the subsurface (Morse and Arvidson, 2002), while plagioclase dissolution is one of the slowest, with its rate about 5–6 orders of magnitude less than that of calcite (Blum and Stillings, 1995, White and Brantley, 1995). Dissimilatory iron reduction is an

Model for a single pore

To analyze the scale dependence of mineral dissolution rates at the pore scale, we developed two models: (1) a Poiseuille Flow model that applies laboratory-measured dissolution kinetics at the pore or fracture wall and couples this to a rigorous treatment of both advective and diffusive transport, and (2) a Well-Mixed Reactor model that assumes complete mixing within the pore, while maintaining the same reactive surface area, average flow rate, geometry, and multicomponent chemistry as the

Verification of transport for a cylindrical pore

The numerical model implemented in CrunchFlow is verified against an analytical solution for Taylor dispersion (Taylor, 1953). Using Eq. (12) for the flow field within a cylindrical pore, Taylor derived an analytical expression for hydrodynamic dispersion (Taylor, 1953)Dh=D+UR2D,where Dh is the dispersion coefficient, and D is the molecular diffusion coefficient. Note that the molecular diffusion coefficient appears both in the first term on the right hand side of Eq. (17), where it contributes

Development of concentration gradients at the pore scale

To illustrate the effects of coupling between flow, transport, and reaction on the scale dependence of reaction rates (and thus to discrepancies between lab and field rates that may be due to processes operating at the pore scale), we show a steady-state concentration field that develops within a calcite pore of 100 μm in length at a flow velocity of 0.1 cm/s. The pore is infiltrated by a dilute pH 5 solution with 0.01 mM dissolved NaCl, equilibrated with atmospheric CO2 gas. Under these slightly

Discussion

The results presented here demonstrate that a scale dependence to mineral dissolution rates, leading potentially to a discrepancy between laboratory and field rates, arises when large concentration gradients develop. Where such conditions arise, the situation is inherently different from the well mixed laboratory conditions where most mineral reaction rates are measured. This implies that it is only in transport regimes where homogeneous concentration fields form that laboratory-measured

Conclusions

In summary, we examined the effects of flow and transport processes on mineral dissolution kinetics in single pores and fractures. Laboratory-measured reaction kinetics were coupled with Poiseuille flow and advective and diffusive transport to unravel some of mechanisms that might contribute to a scale dependence of the dissolution rates, and thus to a discrepancy between laboratory and field rates. The simulations were intended to identify the conditions under which such a rate discrepancy

Acknowledgments

Funding was provided to the Center for Environmental Kinetics Analysis (CEKA) by the U.S. Department of Energy’s Environmental Remediation Science Program as part of a joint NSF-DOE Environmental Molecular Science Institute at Pennsylvania State University. Additional funding was provided by the Laboratory-Directed Research and Development program at Lawrence Berkeley National Laboratory. We are grateful for the review of an early version of the manuscript provided by Dr. Donald DePaolo. We

References (71)

  • K.G. Knauss et al.

    Evaluation of the impact of CO2, co-contaminant gas, aqueous fluid and reservoir rock interactions on the geologic sequestration of CO2

    Chem. Geol.

    (2005)
  • L. Li et al.

    Upscaling geochemical reaction rates using pore-scale network modeling

    Adv. Water Resour.

    (2006)
  • D.R. Lovley et al.

    Bioremediation of metal contamination

    Curr. Opin. BioTechnol.

    (1997)
  • K. Maher et al.

    Rates of silicate dissolution in deep-sea sediment: In situ measurement using U-234/U-238 of pore fluids

    Geochim. Cosmochim. Acta

    (2004)
  • K. Maher et al.

    The mineral dissolution rate conundrum: insights from reactive transport modeling of U isotopes and pore fluid chemistry in marine sediments

    Geochim. Cosmochim. Acta

    (2006)
  • C. Meile et al.

    Scale dependence of reaction rates in porous media

    Adv. Water Resour.

    (2006)
  • M.M.A. Mohamed et al.

    Monte Carlo evaluation of microbial-mediated contaminant reactions in heterogeneous aquifers

    Adv. Water Resour.

    (2006)
  • J.W. Morse et al.

    The dissolution kinetics of major sedimentary carbonate minerals

    Earth Sci. Rev.

    (2002)
  • W.M. Murphy et al.

    surface reaction versus diffusion control of mineral dissolution and growth rates in geochemical processes

    Chem. Geol.

    (1989)
  • E.H. Oelkers et al.

    The effect of aluminum, pH, and chemical affinity on the rates of aluminosilicate dissolution reactions

    Geochim. Cosmochim. Acta

    (1994)
  • E.E. Roden

    Analysis of long-term bacterial vs. chemical Fe(III) oxide reduction kinetics

    Geochim. Cosmochim. Acta

    (2004)
  • N.F. Spycher et al.

    Fluid flow and reactive transport around potential nuclear waste emplacement tunnels at Yucca Mountain, Nevada

    J. Contam. Hydrol.

    (2003)
  • C.I. Steefel et al.

    Cesium migration in Hanford sediment: a multisite cation exchange model based on laboratory transport experiments

    J. Contam. Hydrol.

    (2003)
  • C.I. Steefel et al.

    Reactive transport modeling: an essential tool and a new research approach for the Earth sciences

    Earth Planet. Sci. Lett.

    (2005)
  • C.I. Steefel et al.

    Multicomponent reactive transport in discrete fractures: I. Controls on reaction front geometry

    J. Hydrol.

    (1998)
  • C.I. Steefel et al.

    A new kinetic approach to modeling water-rock interaction: the role of nucleation, precursors, and Ostwald ripening

    Geochim. Cosmochim. Acta

    (1990)
  • A. Stefansson

    Dissolution of primary minerals of basalt in natural waters: I. Calculation of mineral solubilities from 0 °C to 350 ° C

    Chem. Geol.

    (2001)
  • A.F. White et al.

    The effect of time on the weathering of silicate minerals: why do weathering rates differ in the laboratory and field?

    Chem. Geol.

    (2003)
  • C. Zhu

    In situ feldspar dissolution rates in an aquifer

    Geochim. Cosmochim. Acta

    (2005)
  • R.C. Acharya et al.

    Transport modeling of nonlinearly adsorbing solutes in physically heterogeneous pore networks

    Water Resour. Res.

    (2005)
  • R.T. Anderson et al.

    Stimulating the in situ activity of Geobacter species to remove uranium from the groundwater of a uranium-contaminated aquifer

    Appl. Environ. Microbiol.

    (2003)
  • C.A.J. Appelo et al.

    Surface complexation of ferrous iron and carbonate on ferrihydrite and the mobilization of arsenic

    Environ. Sci. Technol.

    (2002)
  • J. Bear

    Dynamics of Fluids in Porous media

    (1972)
  • Berner R.A. (1978) Rate control of mineral dissolution under earth surface conditions. Am. J. Sci. 278,...
  • R.A. Berner

    Chemical weathering and its effect on atmospheric CO2 and climate

    In Chemical Weathering Rates of Silicate Minerals

    (1995)
  • Cited by (191)

    View all citing articles on Scopus
    View full text