(242a) Modeling the Synthesis of Periodic Mesoporous Silicas
MCM-41 has a regular arrangement of pores but the atoms within the walls are arranged irregularly. Realistic models have to account for the heterogeneity of the pore wall in order to predict the properties of the material, e.g. the adsorption performance, correctly.
This work combines two different approaches that have previously been used individually [1-3] to create models of MCM-41: (1) equilibrium Monte Carlo (eMC) simulations of the formation of the phases using a coarse grained model in which the molecules are confined to the nodes of a lattice, and (2) kinetic Monte Carlo (kMC) simulations of the hydrothermal synthesis and calcination of the silica using a fully atomistic, continuum representation.
eMC simulations in the NVT ensemble are used to model the aggregation behavior of an amphiphilic solution co-assembling with silica or organosilica precursors. The model surfactant consists of hydrophobic tail segments and hydrophilic head segments, whereas the inorganic precursors are modeled according to their solubility in the solvent and their hydrophobic/hydrophilic nature. Solvent molecules are not modeled explicitly. The lattice model is able to detect the main features of the system such as the self-assembly of surfactants in ordered structures and the phase separation. In particular, hexagonal liquid-crystal phases are observed at high surfactant concentrations (around 50% by volume).
A single micelle obtained by the eMC simulation is then isolated and used as the template around which the silica network forms in the kMC simulation, replacing the simple rod-shaped micelle used previously . The kMC method follows the reaction path of the hydrothermal synthesis. Monte Carlo moves allow for the condensation reaction around the micelle. The output of the kMC simulation is the positions of the atoms in the pore walls.
This work follows the real synthesis path of mesoporous silica, giving us model materials that improve our understanding of the corresponding real materials, and which can be used to predict the adsorption performance for gas separation applications.
 A. Patti, A.D. Mackie, F.R. Siperstein, Studies in Surface Science and Catalysis,2006, 495.  F.R. Siperstein, K.E. Gubbins, Langmuir 2003, 19, 2049-2057.  C. Schumacher, J. Gonzalez, P.A. Wright, N.A. Seaton, J. Phys. Chem. B 2006, 110, 319-333.