TY - JOUR
T1 - Magnesite dissolution rates at different spatial scales
T2 - The role of mineral spatial distribution and flow velocity
AU - Salehikhoo, Fatemeh
AU - Li, Li
AU - Brantley, Susan L.
N1 - Funding Information:
This work is supported by start-up funds for Li Li through the Penn State Institutes of Energy and the Environment (PSIEE), the Penn State Laboratory for Isotopes and Minerals in the Environment, and the Penn State Earth and Environmental Systems Institute. S.L.B. cites support from DOE Office of Basic Energy Sciences DE FG02 05ER15675. We acknowledge two anonymous reviewers for their careful and insightful comments that have significantly improved the paper.
PY - 2013/5/1
Y1 - 2013/5/1
N2 - We examined the role of mineral spatial distribution and flow velocity in determining magnesite dissolution rates at different spatial scales. One scale is the column scale of a few to tens of centimeters where dissolution rates are measured. Another is the " local" in situ scale defined as approximately 0.1. mm. The experiments used two columns with the same bulk concentration but different spatial distributions of magnesite. In the " Mixed" column, magnesite was evenly distributed spatially within a quartz sand matrix across the whole column, while in the " One-zone" column, magnesite was distributed in one zone in the middle of the column. The two columns were flushed with the same inlet acidic solution (pH 4.0) under flow velocities varying from 0.18 to 36. m/d. Columns of different lengths (22, 10, and 5. cm) were run to understand the role of length scales. Reactive transport modeling was used to infer local-scale and column-scale dissolution rates.Under the acidic-solution flushing conditions used in this study, local in situ dissolution rates vary by orders of magnitude over a length scale of a few to tens of centimeters. Column-scale rates under different conditions vary between 6.40×10-12 and 1.02×10-9mol/m2/s. The distribution of local-scale rates, which collectively determine the column-scale rates, depend on flow velocity, column length scale, and mineral distribution. A two orders of magnitude difference in flow velocity results in more than two orders of magnitude difference in the column-scale rates. Under the same conditions of flow velocity and mineral distribution, column-scale rates are higher in short columns and are lower in long columns. Mineral spatial distribution made a maximum difference of 14% in the medium-flow velocity regime where the reaction kinetics of the system operates under mixed-control conditions. Under such mixed-control conditions, the larger difference between the two columns in their spatial variation of pH and saturation state lead to a larger difference in the spatial distribution of local dissolution rates and therefore column-scale rates. In contrast, under slow-flow velocity conditions, the system is mostly at equilibrium without much spatial variation, i.e., the regime of local equilibrium. Under fast-flow velocity conditions, the system is kinetically controlled, the local aqueous geochemistry is everywhere similar to the inlet condition, and is also relatively uniform. Under these two conditions, there is almost no difference between the two columns. Column-scale rates were best understood in terms of the Damkohler number (DaI) that quantifies the relative dominance of advection and dissolution processes. The observations in this study lead us to surmise that rates of weathering and other natural processes may be similarly affected by chemical heterogeneity in natural systems under conditions where reaction rate and flow rate are comparable.
AB - We examined the role of mineral spatial distribution and flow velocity in determining magnesite dissolution rates at different spatial scales. One scale is the column scale of a few to tens of centimeters where dissolution rates are measured. Another is the " local" in situ scale defined as approximately 0.1. mm. The experiments used two columns with the same bulk concentration but different spatial distributions of magnesite. In the " Mixed" column, magnesite was evenly distributed spatially within a quartz sand matrix across the whole column, while in the " One-zone" column, magnesite was distributed in one zone in the middle of the column. The two columns were flushed with the same inlet acidic solution (pH 4.0) under flow velocities varying from 0.18 to 36. m/d. Columns of different lengths (22, 10, and 5. cm) were run to understand the role of length scales. Reactive transport modeling was used to infer local-scale and column-scale dissolution rates.Under the acidic-solution flushing conditions used in this study, local in situ dissolution rates vary by orders of magnitude over a length scale of a few to tens of centimeters. Column-scale rates under different conditions vary between 6.40×10-12 and 1.02×10-9mol/m2/s. The distribution of local-scale rates, which collectively determine the column-scale rates, depend on flow velocity, column length scale, and mineral distribution. A two orders of magnitude difference in flow velocity results in more than two orders of magnitude difference in the column-scale rates. Under the same conditions of flow velocity and mineral distribution, column-scale rates are higher in short columns and are lower in long columns. Mineral spatial distribution made a maximum difference of 14% in the medium-flow velocity regime where the reaction kinetics of the system operates under mixed-control conditions. Under such mixed-control conditions, the larger difference between the two columns in their spatial variation of pH and saturation state lead to a larger difference in the spatial distribution of local dissolution rates and therefore column-scale rates. In contrast, under slow-flow velocity conditions, the system is mostly at equilibrium without much spatial variation, i.e., the regime of local equilibrium. Under fast-flow velocity conditions, the system is kinetically controlled, the local aqueous geochemistry is everywhere similar to the inlet condition, and is also relatively uniform. Under these two conditions, there is almost no difference between the two columns. Column-scale rates were best understood in terms of the Damkohler number (DaI) that quantifies the relative dominance of advection and dissolution processes. The observations in this study lead us to surmise that rates of weathering and other natural processes may be similarly affected by chemical heterogeneity in natural systems under conditions where reaction rate and flow rate are comparable.
UR - http://www.scopus.com/inward/record.url?scp=84874483209&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84874483209&partnerID=8YFLogxK
U2 - 10.1016/j.gca.2013.01.010
DO - 10.1016/j.gca.2013.01.010
M3 - Article
AN - SCOPUS:84874483209
SN - 0016-7037
VL - 108
SP - 91
EP - 106
JO - Geochimica et Cosmochimica Acta
JF - Geochimica et Cosmochimica Acta
ER -