Comparative study of morphometric properties characterizing the complexity of silicate pore networks probed by...

In this work, we compare the surface and morphometric properties of the pore networks in four silicas (code names Fr1428, Fr474, Fr1386, and MM1164) with different random porosities using the adsorption isotherms of two different probe adsorbents, nitrogen and methanol. The parent material Fr1428 was a pure silica 25 microm sample. The Fr474 sample was the same one with bonded electroneutral diol groups on its outer surface. Fr1386 was the parent material with bonded electroneutral diol groups on its outer surface and sulfonic groups on its inner surface, and the MM1164 sample was the original sample with external electroneutral diol groups and internal n-octadecyl groups. The properties examined were the specific surface area S(p) and the specific pore volume V(p), the pore connectivity c, the pore anisotropy b, the tortuosity tau, and the lacunarity lambda of the pore network as well as the percentage microporosity. These properties provide a complete characterization of complexity of the porous network. The surface areas of the solids were estimated via the traditional BET plots (S(BET)) and the I-point method (S(I)). The two sets of values S(BET) and S(I) were practically identical and they decrease as the size of the functional group increases. The values of percentage microporosity were also determined by the same I-point method using the variation of the C parameter of the BET equation. The total pore volume V(p) was found to be higher in the case of methanol adsorption, compared to nitrogen, which might be related to increase condensation. The networks of the pores were simulated using a dual site bond model (DSBM) and Monte Carlo (MC) techniques for achieving their proper arrangement into the solids. From the resulting simulating networks, the pore connectivity distributions (PCD) and their mean values c(mean) were estimated and favorably compared to the values of connectivity c(Seaton) determined according to the method of Seaton. Both values decrease with the size of the functional groups and are weakly affected by the adsorbent employed. From the simulation pore network, the mean values of tortuosity tau(mean) were also estimated and found to be lower when N2 was used as adsorbate compared to MeOH. The values of lacunarity lambda, estimated according to the method by Allain and Cloitre using the moving box technique in the DSBM/MC simulation matrix of the pore network, indicate that the distribution of the poreless mass into the matrix increases with the size of the functional group. Finally, the internal relationships observed between the pore anisotropy b and the percentage microporosity as well as between the tortuosity tau and the pore connectivity c are discussed.