Optimization of hydrogen storage capacity in silica-supported low valent Ti systems exploiting Kubas binding of hydrogen

Abstract

Silica-based materials grafted with low valent Ti fragments for Kubas-type binding of hydrogen were optimized for hydrogen adsorption capacity by varying the surface area, pore size, loading levels, and type of organometallic precursor. All materials were characterized by XRD, nitrogen adsorption, and XPS where appropriate. The surface area of HMS silica was optimized by varying silica-to-surfactant molar ratio, and also by tuning the pore size by varying the surfactant’s carbon chain length (C 6, C 8, C 10, C 12). Then Ti fragments originating from either benzyl, allyl, or methyl Ti precursors were grafted onto the optimal HMS surface at different loading levels to arrive at Ti grafted HMS materials with H 2 storage capacities and binding properties superior to those previously reported by our group for benzyl Ti (III) species on silica. HMS prepared with dodecylamine using a silica:surfactant ratio of 3:1 and subsequently grafted with 0.2 M equiv. of TiBz 4 had the highest H 2 adsorption at 2.45 wt% at 77 k and 60 atm, which equates to an average of 3.98 H 2 molecule per Ti metal center, just one H 2 molecule short of the theoretical saturation limit of 5 H 2/Ti predicted by the 18-electron rule. The H 2 adsorption capacities of Me 3Ti-HMS and Allyl 3Ti-HMS prepared using the same optimized sample of C 12-HMS silica at a 3:1 Si:surfactant ratio possessed H 2 adsorption values corresponding to 2.4 and 2.27 H 2 per Ti center, respectively, at 60 atm and 77 K. This performance level is significantly lower than that of the benzyl Ti (III) system. The binding enthalpies of the benzyl Ti (III) material increase with H 2 coverage to 23 kJ/mol, while the enthalpies for the newly synthesized Me 3Ti-HMS and Allyl 3Ti-HMS materials increase with H 2 coverage to a maximum of 2.66 and 4.17 kJ/mol, respectively. XPS studies on these materials suggested a trend in π-back donating ability on the Ti (III) centers of methyl > allyl > benzyl, opposite that observed experimentally. The reason for the diminished performance of the allyl and methyl Ti (III) systems may thus be related to the presence of THF ligands blocking coordination sites in the allyl and methyl systems. THF is not present in the benzyl system because this solvent is not required for synthesis.