Pore Size Effects on the Sorption of Supercritical CO₂ in Mesoporous CPG-10 Silica

Abstract

Excess sorption isotherms of supercritical carbon dioxide in mesoporous CPG-10 silica glasses with nominal pore sizes of 7.5 and 35 nm were measured gravimetrically at 35 and 50 °C and pressures of 0À200 bar. Formation of broad maxima in the excess sorption was observed at fluid densities below the bulk critical density. Positive values of excess sorption were measured at bulk densities below 0.7 g/ cm 3 , i.e., the interfacial fluid is denser than the bulk fluid at low pressures. Zero and negative values were obtained at higher densities, i.e., the adsorbed fluid becomes equal to and eventually less dense than the corresponding bulk fluid. Pronounced confinement effects on sorption behavior have been found and further analyzed by normalizing the excess sorption to the adsorbent surface area and pore volume, yielding new insight into supercritical fluid adsorption in this range of pore sizes and P, T conditions. If normalized to the specific surface area, the excess sorption is higher for the 35 nm pore size material, but the pore volume normalized excess sorption is higher for the 7.5 nm pore size material. With increasing pore width, the excess sorption peak position shifts to higher pressure. Both CPG-10 materials exhibit regions of constant mean pore fluid density as a function of bulk CO 2 density at 35 °C but not at 50 °C. This region is located between the excess sorption peak maximum and the adsorption/depletion transition point. Applied to the situation of CO 2 sequestration in dry sandstone formations, the results of this study indicate that carbon storage capacity is enhanced by sorption effects, particularly at low temperature and in narrow pores with high surface to volume ratios.

On sorption and swelling of CO2 in clays

“…With further pressure increases, the bulk density becomes increasingly larger than the sorbed phase density, hence resulting in negative excess sorption. Similar observations have been made in earlier studies by the same authors on silica gels with defined pore geometries (Rother et al 2013b). These findings partly confirm those by Busch et al (2008) where a decreasing excess sorption trend was observed for pressures exceeding *8 MPa, but negative excess sorption (where the sorbed phase density is below bulk density) was not observed.…”

“…Natural coals show higher adsorption capacities indicating that smaller pores, not covered by N 2 BET, significantly contribute to overall sorption capacity. Data on the clays, mudrocks and siltstones from Amann et al (2011), Busch et al (2008, Jeon et al (2014); for the silica gels from Rother et al (2013b), and for the natural coals and activated carbon from Gensterblum et al (2009Gensterblum et al ( , 2010 when charged with CO 2 . Measurements were performed on smectite samples exchanged with different cations (K, Ca, Na) using X-ray diffraction (XRD) in an environmental chamber (Giesting et al 2012a, b;Ilton et al 2012;Schaef et al 2012).…”

Molecular Simulation of CO2- and CO3-Brine-Mineral Systems

“…At conditions more relevant to GCS (high P and T, presence of water), few atomistic simulation studies have been carried out. Gravimetric and neutron scattering data show that confinement in silica nanopores can promote the formation of a dense adsorbed phase even in pores as large as 35 nm (Melnichenko et al 2010;Cole et al 2010;Rother et al 2012) but this has never been tested by atomistic simulation, to our knowledge. Molecular dynamics simulations described elsewhere in this volume show that the solubility of CO 2 in water-filled silica nanopores is highly sensivity to nanopore size and the hydrophilicity of the silica surface (Chialvo et al 2013a,b, this volume).…”

N-octane diffusivity enhancement via carbon dioxide in silica slit-shaped nanopores – a molecular dynamics simulation

“…12 Because quartz (made up by SiO4 tetrahedral structure) is an abundant mineral in earth, the cristobalite crystal with fully protonated non-bridging oxygen atoms is considered a reasonable proxy for hydrophilic rock pore surfaces. 13 Approximately, 1 n-octane molecule is 12.8 Å in length and 1 carbon dioxide is 5.4 Å. We chose to simulate a slit-shaped pore of width slightly larger than these dimensions, 1.9 nm (determined by the distance between the planes determined by the oxygen atoms of the hydroxyl groups across the pore volume), so that the results will differ substantially compared to those attainable in bulk systems because of surface effects.…”