Molecular Simulations of the Synthesis of Periodic Mesoporous Silica Phases at High Surfactant Concentrations

Abstract

Molecular dynamics simulations of a coarse-grained model are used to study the formation mechanism of periodic mesoporous silica over a wide range of cationic surfactant concentrations. This follows up on an earlier study of systems with low surfactant concentrations. We started by studying the phase diagram of the surfactant–water system and found that our model shows good qualitative agreement with experiments with respect to the surfactant concentrations where various phases appear. We then considered the impact of silicate species upon the morphologies formed. We have found that even in concentrated surfactant systemsin the concentration range where pure surfactant solutions yield a liquid crystal phasethe liquid-crystal templating mechanism is not viable because the preformed liquid crystal collapses as silica monomers are added into the solution. Upon the addition of silica dimers, a new phase-separated hexagonal array is formed. The preformed liquid crystals were found to be unstable in the presence of monomeric silicates. In addition, the silica dimer is found to be essential for mesoscale ordering at both low and high surfactant concentrations. Our results support the view that a cooperative interaction of anionic silica oligomers and cationic surfactants determines the mesostructure formation in the M41S family of materials.

The role of charge-matching in nanoporous materials formation

“…Parameters of a coarse-grained model of the system, based on the MARTINI approach [26], were then calibrated so that micelle density profiles and other structural properties matched those of the atomistic simulations. This approach has been successfully validated and applied to elucidate the self-assembly mechanism in the synthesis of PMS materials [23,[27][28][29][30] as well as of their organosilica derivatives [4,31]. Details of the model parameterisation and validation procedure for HMS precursor solutions, including a diagram of the mapping scheme ( Figure S5) and a table with the final set of parameters (Table S7), are provided in Supporting Information.…”

Modelling the self-assembly of silica-based mesoporous materials

“…The bottom-up approach, on the other hand, starts at a high level of theory (atomistic or even quantum mechanical), and builds simpler models through a progressive coarse-graining procedure. This strategy has been applied to study the formation of MCM-41 [48,52,53,55,59,60], PMOs [106,107] and bio-inspired silica materials [94,95].…”

“…Taking theoretical and experimental evidence together, it is clear that MCM-41 synthesis starting from low concentration surfactant solutions takes place through a cooperative templating mechanism via silica/surfactant charge matching, where silicates promote changes in the morphology of surfactant aggregates, while the presence of surfactant micelles creates local concentration gradients that promote silica condensation. However, it appears that even when starting from high surfactant concentration solutions, which are already able to form mesostructures, a true liquidcrystal templating mechanism is unlikely to take place [59]. This is because the addition of anionic silica to the solution causes a profound change in the nature of the molecularlevel interactions, promoting phase separation of an otherwise homogeneous solution.…”

“…Atomistic studies, on the other hand, were able to elucidate the underlying molecular level interactions, concluding that the onset of micro-scale ordering on the pore walls of some PMO materials may take place very early in the synthesis process [107]. Despite these important advances, the picture of the mechanism of PMO synthesis is not as clear as that of MCM-41, partly because of the lack of intermediate-level studies of these systems (i.e., using coarse-grained models that are still able to retain a high degree of chemical realism [59,60]). Efforts to develop such a model are already underway at our research group, and we hope to report them in the near future [111].…”

“…The CTAB surfactant was described by four hydrophobic tail beads and one positively charged head bead [57], while the models for water and bromide followed the conventional MARTINI prescription [56]. The surfactant model was validated by comparing micelle density profiles at the CG [55] and all-atom (AA) levels [58], and was shown to quantitatively reproduce the experimental average size and aggregation number of small CTAB micelles [55], as well as most features of the experimental CTAB/water phase diagram over a wide range of concentrations and temperatures [59]. The silica monomer parameters were calibrated by matching micelle density profiles against those obtained from AA simulations [48] under the same conditions.…”