CLAY MINERAL REACTIVITY ACROSS SCALES UTILIZING SOLID-STATE NUCLEAR MAGNETIC RESONANCE

Open Access

Author:

Sanders, Rebecca L.

Graduate Program:

Chemistry

Degree:

Doctor of Philosophy

Document Type:

Dissertation

Date of Defense:

February 24, 2011

Committee Members:

• Karl Todd Mueller, Dissertation Advisor
• Karl Todd Mueller, Committee Chair
• Barbara Jane Garrison, Committee Member
• John B Asbury, Committee Member
• James David Kubicki, Committee Member

Keywords:

• reactive surface area
• clay minerals
• NMR
• dissolution
• kaolinite
• montmorillonite
• Shale Hills
• TFS

Abstract:

An understanding of the surface reactivity of minerals is crucial for characterizing and scaling numerous environmental processes. For instance, the release rate of ions during dissolution is scaled to remove differences in the available surface area. Typically, BET or geometric surface area measurements are the convention to calculate surface area. However, these methods do not necessarily report on the reactive surface area of a mineral, which provides a chemically-sensitive measure of surface area. The reactive surface area of clay minerals can be particularly challenging to quantify as they have a layered structure with two distinct surfaces: edge sites and basal planes. The edge sites dissolve preferentially compared to basal planes. When clay minerals are weathered, the reactive surface area decreases as edge sites are depleted. Throughout this thesis, the objective was to develop and demonstrate that advanced experimental tools can be used to predict reactive surface area across spatial scales for clay minerals. To measure reactive surface area, the probe molecule (3,3,3-trifluoropropyl) dimethylchlorosilane (TFS) is attached to lone Q3Si hydroxyl sites and the 19F spins in the TFS-treated samples are then quantified using 19F magic angle spinning (MAS) nuclear magnetic resonance (NMR) spectroscopy. The quantification of the number of reactive hydroxyl sites per gram of sample is proportional to the reactive surface area of each mineral, particularly the edge sites for clay minerals. Batch dissolution experiments of kaolinite were conducted at 21 °C and pH 3. During the course of three months, dissolution rate decreased over time. BET specific surface area measurements did not reflect changes in dissolution rate. However, the selective nature of TFS attachment has been utilized to demonstrate the changes in reactive surface area are tied to a concomitant decrease in the rates of Si and Al release into solution. Similar studies were conducted with bentonite, a montmorillonite-rich clay that contained an amorphous silica phase. The silica phase was identified to be opal-CT and was not observed to contribute the dissolution rate of bentonite. Similar results to the kaolinite experiments were obtained in which the decreasing dissolution rate was correlated with changes in reactive surface area. The quantity of reactive sites can be used to predict the dissolution rates of kaolinite and montmorillonite. Dissolution rates in the field can be several orders of magnitude slower than rates for the same mineral when dissolved in the laboratory. Several factors including reactive surface area can account for the differences in dissolution rates. The solid-state NMR proxy for reactive surface area was used to determine how surface reactivity impacts the rates of mineral transformations at the Shale Hills catchment, a Critical Zone Observatory located in central Pennsylvania. The reactive surface area was quantified for a series of soils collected along a downslope planar transect at Shale Hills. Variables that were investigated to explain the changes in reactive surface area include soil mineralogy, particle size, pore formation, and soil location and depth. Surface reactivity to the TFS probe molecule was a function of the extent that the soils have been weathered rather than soil age. Surface reactivity increases as soils are weathered from parent mineral, as particle size decreases, and as extended pore networks are formed. Dissolution rates of the soil samples were not a function of reactive surface area determined with TFS quantification. Rather dissolution rates for the clay minerals were all the same within error, suggesting that reactive surface area does not vary substantially over long time scales in field weathered shales. This conclusion signifies that while it is still important to quantify reactive surface area, it may be not be necessary to consider how reactive surface area changes for field weathered clays when accounting for discrepancies between laboratory and field dissolution rates. Our NMR proxy for reactive surface area has laid the groundwork for future studies using chemical methods to quantify reactive surface area, which will enable a more accurate prediction of laboratory and field dissolution. These studies indicate that the use of a chemical method to quantify reactive surface area can be successfully applied to simple system. However, the quantification of reactive surface area for multimineralic systems requires the quantification of both the quantity and reactivity of the reactive sites present. Future studies into the use of chemical probes for quantifying reactive surface area should emphasize a multifaceted approach to quantify the types and distribution of reactive sites and how the distribution of reactive sites evolves over time.