Item

Abstract

Detailed knowledge of the reaction kinetics of silica nanoparticle formation in cooling supersaturated waters is fundamental to the understanding of many natural processes including biosilicifcation, sinter formation, and silica diagenesis. Here, we quantified the formation of silica nanoparticles from solution as it would occur in geothermal waters. We used an in situ and real-time approach with silica polymerisation being induced by fast cooling of a 230 degrees C hot and supersaturated silica solution. Experiments were carried out using a novel flow-through geothermal simulator system that was designed to work on-line with either a synchrotron-based small angle X-ray scattering (SAXS) or a conventional dynamic light scattering (DLS) detector system. Our results show that the rate of silica nanoparticle formation is proportional to the silica concentration (640 vs. 960 ppm SiO2), and the first detected particles form spheres of approximately 3 nm in diameter. These initial nanoparticles grow and reach a final particle diameter of approximately 7 nm. Interestingly, neither variations in ionic strength (0.02 vs. 0.06) nor temperature (reactions at 30 to 60 degrees C, mimicking Earth surface values) seem to affect the formation kinetics or the final size of the silica nanoparticles formed. Comparing these results with our previous data from experiments where silica polymerisation and nanoparticle formation was induced by a drop in pH from 12 to near neutral (pH-induced, Tobler et al., 2009) showed that (a) the mechanisms and kinetics of silica nanoparticle nucleation and growth were unaffected by the means to induce silica polymerisation (T drop or pH drop), both following first order reactions kinetics coupled with a surface controlled reaction mechanism. However, the rates of the formation of silica nanoparticles were substantially (around 50%) slower when polymerisation was induced by fast cooling as opposed to pH change. This was evidenced by the occurrence of an induction period, the formation of larger critical nuclei, and the absence of particle aggregation in the T-induced experiments. (C) 2013 Elsevier Ltd. All rights reserved.