Search

Mesoporous silicate−surfactant composites were prepared via washing of as-synthesized silicate−surfactant−expander materials in appropriate solvents. The expander can be generated in situ from the surfactant decomposition under high-temperature conditions, or supplied externally at the stage of the synthesis or postsynthesis hydrothermal treatment. The washing removed the expander as well as a minor part of the surfactant ions present and thus opened the mesopores whose walls were covered by a relatively dense layer of surfactant ions. Depending on the synthesis method for silicate−surfactant−expander composites, both ordered and disordered mesoporous silicate−surfactant composites were obtained with narrow pore size distributions and a wide range of adjustable pore sizes (4−12 nm). Despite the fact that these materials contained 25−35 wt % of surfactant, they exhibited large BET specific surface areas (360−550 m² g^-1) and total pore volumes (from 0.4 to as large as 1.8 cm³ g^-1). It is postulated that the surfactant retained in the silicate−surfactant composites is bonded to the silicate walls via electrostatic interactions, which are originally responsible for the formation of silicate−surfactant or silicate−surfactant−expander mesophases. The mesoporosity of the composites is mostly constituted by the space initially occupied by expander, which was later removed upon washing. The synthesis approach described herein is the first successful method for synthesis of ordered and disordered mesoporous silicate−surfactant composites with hydrophobic surfaces, the latter being predominantly formed by long alkyl chains of surfactant ions.

Microporosity and connectivity of ordered mesopores of SBA-15 silica were studied using nitrogen adsorption and novel methods based on selective pore blocking via organosilane modification, and on the imaging of inverse platinum replica of ordered mesoporous structure. It was found that SBA-15 exhibits a relation between the pore size, pore volume and specific surface area which is significantly different from that for cylindrical or hexagonal pores, which suggests that the SBA-15 structure is more complex than an array of hexagonally ordered channels, even if they are corrugated. Nitrogen and argon adsorption measurements provided evidence that large mesopores are accompanied by a certain amount of significantly smaller pores (of the size below about 3.4 nm) with a broad distribution primarily in the micropore/small-mesopore range. The modification of SBA-15 via chemical bonding of small trimethylsilyl ligands partially blocked the complementary pores, and the bonding of larger octyldimethylsilyl groups made them essentially fully inaccessible to nitrogen molecules, which manifested itself in dramatic changes in the relation between the pore size, pore volume, and specific surface area. After dissolution of the SBA-15 framework, platinum wires grown inside the porous structure formed bundles, as seen from transmission electron microscopy. These results provided strong and unambiguous evidence that large ordered mesopores of SBA-15 are accompanied by much smaller disordered pores and that an appreciable fraction of the latter is located in the pore walls, providing connectivity between the ordered large-pore channels. The complementary pores are suggested to form as a result of penetration of poly(ethylene oxide) chains of the triblock copolymer template within the silica framework of as-synthesized SBA-15. We also studied thermal stability of SBA-15 structure and its complementary porosity. As inferred from nitrogen adsorption data, the complementary porosity was retained to a significant extent even after calcination at 1173 K, but most likely completely disappeared at 1273 K. The heat treatment was accompanied not only by a significant decrease in the specific surface area and pore volume but also by narrowing the pore size distribution at temperatures up to 1173 K. Thus, we were able to demonstrate for the first time that the SBA-15 sample with nitrogen adsorption properties similar to those of MCM-41 can be obtained via calcination at 1273 K, although the pore volume and specific surface area of such SBA-15 material is relatively low.

A procedure of the template displacement by organosilanes was utilized to prepare pure-silica and cerium-substituted mesoporous materials with tailored surface functionality. These two self-assembled mesostructured systems exhibit hexagonal arrangement of mesopores but they show different reactivity with respect to organosilanes. Several samples with different pore size and functionality were prepared and for the first time a cerium-substituted mesostructured silicate was functionalized. Some peculiarities of organosilane interactions with the self-assembled cerium-substituted silicates reflect the difference in the acidic and coordination properties of cerium centers in the mesostructure. The study of the template displacement kinetics shows that about 4 h are needed to remove 70% of surfactant and about twice longer time is required for its complete removal. In addition, it is shown that the template displacement procedure stabilizes the ordered mesoporous structures.

Two series of alkyl-modified molecular sieves were prepared using the FSM-16 mesoporous silica. The attachment of alkyl groups to the surface, the structure of the bonded layer, and changes in the mesoporous structure of the FSM-16 material upon functionalization were studied by means of high-resolution thermogravimetry, elemental analysis, nitrogen adsorption, and ¹³C and ²⁹Si NMR. Surface coverages of attached ligands were in the range of 0.85−2.60 mmol/g. All modified samples showed a decrease in the pore size, which was in relation to the size of attached ligands. The uniformity of pores upon modification proves that the mesoporous structure of FSM-16 is retained after modification. The attachment of ligands takes place in a solvent either with a small amount of water or anhydrous indicating a direct reaction between the chlorosilane molecule and silanol groups and/or a reaction autocatalyzed by the small amount of water present on the FSM-16 surface. Nitrogen adsorption isotherms and the adsorption energy distributions calculated from these isotherms revealed the presence of highly ordered alkyl ligands and the existence of the screening effect produced by alkyl groups as a result of weak interactions of nitrogen probe molecules with attached long alkyl chains. Thermal stability of modified samples greatly depends on the conformation of attached ligands. By using appropriate solvent and/or thermal treatment of modified samples, it was possible to achieve higher ordering of attached groups and consequently higher thermal stability of the resulting bonded phases. The studies of silicon spin−lattice relaxation times T1 support the formation of ligand structures on the surface and show that the surface accessibility greatly decreases with increasing length of attached ligands and with refinement of the ordering of ligands on the surface. Mesoporous FSM-16 material appeared to be a good model for studying functionalization of the silica surface.

A new approach to pore size and surface area analysis for hydrophobic mesoporous solids is proposed. Well-defined materials with strongly hydrophobic surfaces were prepared via chemical bonding of octyldimethylsilyl (ODMS) ligands to the surface of large-pore MCM-41 samples. Nitrogen statistical film thickness curves (t-curves) were determined for hydrophobic pores. These t-curves were fitted with the nitrogen adsorption isotherm for a macroporous silica modified with ODMS groups in order to derive the reference t-curve valid in the entire range of relative pressures. An empirical expression for the reference t-curve was found. The reference t-curve and the previously derived corrected form of the Kelvin equation were used to calculate pore size distributions for hydrophobic mesoporous materials, allowing to correctly reproduce the total pore volume, average pore size, and surface area determined using independent calculation procedures. The reference nitrogen adsorption isotherm on hydrophobic materials is reported in a tabular form. Moreover, the accuracy of the BET specific surface area was discussed. Thus, this study allowed us to develop the methodology for accurate and consistent characterization of hydrophobic mesoporous materials using nitrogen adsorption.