PII S0016-7037(01)00827-4 Evaluation of silica-water surface chemistry using NMR spectroscopy SUSAN A. CARROLL, 2, *ROBERT S. MAXWELL, 1 WILLIAM BOURCIER, 2 SUE MARTIN, 2 and SUZY HULSEY 1 1 Chemistry and Materials Science Directorate, Lawrence Livermore National Laboratory, P.O. Box 808, Livermore, CA 94551, USA 2 Energy and Environment Directorate, Lawrence Livermore National Laboratory, P.O. Box 808, Livermore, CA 94551, USA (Received January 25, 2000; accepted in revised form October 1, 2001) Abstract—We have combined traditional batch and flow-through dissolution experiments, multinuclear nuclear magnetic resonance (NMR) spectroscopy, and surface complexation modeling to re-evaluate amor- phous silica reactivity as a function of solution pH and reaction affinity in NaCl and CsCl solutions. The NMR data suggest that changes in surface speciation are driven by solution pH and to a lesser extent alkali concentrations, and not by reaction time or saturation state. The 29 Si cross-polarization NMR results show that the concentration of silanol surface complexes decreases with increasing pH, suggesting that silanol sites polymerize to form siloxane bonds with increasing pH. Increases in silica surface charge are offset by sorption of alkali cations to ionized sites with increasing pH. It is the increase in these ionized sites that appears to control silica polymorph dissolution rates as a function of pH. The 23 Na and 133 Cs NMR results show that the alkali cations form outersphere surface complexes and that the concentration of these complexes increases with increasing pH. Changes in surface chemistry cannot explain decreases in dissolution rates as amorphous silica saturation is approached. We find no evidence for repolymerization of the silanol surface complexes to siloxane complexes at longer reaction times and constant pH. Copyright © 2002 Elsevier Science Ltd 1. INTRODUCTION Significant research has been directed toward determining the reaction kinetics for the dissolution of quartz and its crys- talline and amorphous polymorphs. These SiO 2 phases play critical roles in a variety of geochemical and environmental processes ranging from soil formation to disposal of radioactive waste by burial to thermally enhanced oil recovery. Laboratory studies have found that quartz and amorphous silica dissolution rates are sensitive to solution pH and alkali concentration. Dissolution rates increase by 2 to 3 orders of magnitude be- tween pH 4 and 10 (Schwartzentruber et al., 1987; Bennett et al., 1988; Bennett, 1991; Knauss and Wollery, 1988; Wollast and Chou, 1988; Brady and Walther, 1990; House and Orr, 1992; Carroll et al., 1994; Mazer and Walther, 1994). Addi- tionally, the pH dependence of the dissolution rate may be coupled to its dependence on dissolved alkali concentrations. Many studies have shown that dissolution rates increase with small additions of alkali cations (Bennett et al., 1988; Bennett, 1991; Dove and Crerar, 1990; Dove and Elston, 1992, Dove 1994; Berger et al., 1994). Dove and Elston (1992) and Dove (1994) relate this co-dependence of quartz dissolution on pH and sodium concentrations to increased populations of reactive deprotonated silica (SiO - ) and outersphere sodium (SiO - Na + ) surface complexes. This agrees with the two- order-of-magnitude increase in the amount of surface adsorbed sodium from pH 4 to pH 10 (Allen et al., 1971). Another primary observation about silica polymorphs is that their dis- solution and precipitation rates (Rimstidt and Barnes, 1980; Fleming, 1986; Renders et al., 1995; Carroll et al., 1998) are directly proportional to saturation, suggesting that the rates are microscopically reversible at the mineral-solution interface (Rimstidt and Barnes, 1980). However, the effect of saturation on surface speciation is unknown. It is important to determine surface complexes directly, because the fundamental chemistry that controls interfacial reactions may depend on the types and concentrations of sur- face species. Our current understanding of silica-water surface chemistry has been derived from changes in bulk, aqueous measurements of silica gel suspensions (Bolt, 1957; Dugger, 1964; Schindler and Kamber, 1968; Tadros and Lyklema, 1969; Allen et al., 1971; Yates and Healy, 1976; Brady, 1992; Casey, 1994) and the application of various theoretical surface complexation models (Hiemstra et al., 1989; Brady, 1992; Sahai and Sverjensky, 1997). Generally, the silica-water inter- face consists of silanol complexes that deprotonate to form negatively charged complexes with increasing pH and increas- ing ionic strength. Cation sorption is also favored with increas- ing pH. Although, silica gel has an acid pH zpc between 2 and 4, positively charged complexes from sorption of protons to sila- nol sites are negligible over the normal pH range. Nuclear magnetic resonance (NMR) shows extreme promise for direct investigation of silica-water surface chemistry. Ma- ciel and Sindorf (1980) employed surface selective cross-po- larization magic-angle-spinning experiments to identify silox- ane sites with four oxygen bridging atoms and silanol sites with one or two non-bridging oxygen atoms on the surface of dry silica gel. This spectroscopic technique exploits the hetero- nuclear dipole-dipole couplings between abundant, high-fre- quency 1 H and rare and lower frequency 29 Si (Pines et al., 1973; Maciel and Sindorf, 1980; Ansermet et al., 1990). As such, it can be used to identify surface species at the silica- water interface, because protons present are primarily due to chemisorption at the solid-solution interface, though some pro- tons may be present internally (Chuang et al., 1993). Bunker et al. (1988), for example, used 29 Si and 17 O static and magic- angle-spinning NMR spectroscopy to show that stable siloxane sites polymerize from silanol sites as sodium borosilicate * Author to whom correspondence should be addressed (carroll6@ llnl.gov). Pergamon Geochimica et Cosmochimica Acta, Vol. 66, No. 6, pp. 913–926, 2002 Copyright © 2002 Elsevier Science Ltd Printed in the USA. All rights reserved 0016-7037/02 $22.00 + .00 913