CO-IR spectra were applied to investigate the concentration of active metals over serial sulfide NiMo catalysts, which were shown in Fig. S17. It is well investigated that MoS₂ slabs possess two types of edges including the sulfur-terminated edge (S-edge) and Mo-terminated edge (Mo-edge).^[53] Because the sample are detected without H₂ treatment, the symmetrical band at about 2105 cm^-1 assigned to CO adsorption on the Mo-edge sites with a high sulfur coordination number can be observed, which is also considered as fully sulfide species.^[54] Three bands at 2065 cm^-1, 2045 cm^-1 and 2020 cm^-1 signify the existence of CO adsorbed on the S-edge with different sulfur coverage.^[7,56] The band at about 2130 cm^-1 should be assigned to partial sulfided MoOxSy phase or NiMoS phase.^[55,56] The band appears at 2156 cm^-1 is due to CO adsorbed on surface OH group.^[55] The decomposition data of CO-IR spectra were concluded in Table 5. As the addictive amounts of metals are same, a higher total concentration of Mo-edge and S-edge will indicate a better dispersion of active metals. It shows that the total concentrations of Mo-edge and S-edge increase after Al and Ti modification, which follows the sequence of NiMo/SBA-16 < NiMo/AT-10 < NiMo/AT-0 < NiMo/AT-7.5 < NiMo/AT-5 < NiMo/AT-2.5, demonstrating that the incorporation of Al and Ti species into SBA-16 material can enhance the dispersion degree of metals. The total concentration of Mo-edge and S-edge for NiMo/AT-10 catalyst is lower than NiMo/AT-0 catalyst, verifying that Ti modification can improve higher dispersion degree of active metal than that of Al modification. However, as the mass percentage of TiO₂ in support reaches 10%, the total concentration for NiMo/AT-0 catalyst is lower than other catalysts containing both Al and Ti species. It should be noteworthy that the concentration of S-edge exhibits an increasing trend with the Al contents in supports, peaking at 9.70 μmol/g for NiMo/AT-7.5 catalyst. Hence, incorporation of Al species into SBA-16 silica can promote the formation of S-edge. In order to acknowledge the influence of acidity on the formation of S-edge sites, the plots of amount of B acid sites and the concentration of S-edge are shown in Fig. S18. It shows that the S-edge concentration presents an increasing tendency with the amounts of B acid, indicating that the B acid sites can facilitate the formation of S-edge sites. However, the lower concentration of S-edge for NiMo/AT-10 catalyst than NiMo/AT-7.5 catalyst should be assigned to its lower dispersion degree f_Mo of MoS₂, which can be recognized from HRTEM results. Moreover, the concentration of fully sulfide Mo-edge sites over sulfide NiMo catalysts exhibit an increasing trend with the contents of Ti species. As the TiO₂ content reaches 10% in support, the concentration of Mo-edge site becomes lower than other catalysts containing Ti species, which may be also related to its relatively low dispersion degree of active metals. In addition, the ratios of S-edge and Mo-edge are also calculated and exhibit an increasing trend with the contents of Al species.

HRTEM measurement was applied for obtaining the dispersion information of active phase MoS₂. The representative images of different sulfide catalysts are displayed in Fig. S19. The stacking-number distributions of MoS₂ crystallines for serial sulfide catalysts are exhibited in Fig. S20. It is obvious that the proportion of MoS₂ phases with lower stacking layers, especially the 2 layers, show an increasing tendency with the Al and Ti compositions in the catalysts. It can be verified from H₂-TPR result that the interaction strength between active Mo (Oh) phase and support become stronger after Al and Ti modification, which will lead to the decrement of stacking layers of MoS₂. Fig. S21 exhibits that the frequency of MoS₂ phases with short slab length ranging from 0-2 nm and 2-4 nm increases after Al and Ti modification. Whereas, the frequency of MoS₂ with higher slab length present the reverse trend. Moreover, the average length and number of slabs and dispersion degree (f_Mo) for MoS₂ slabs are displayed in Table S1. As compared with NiMo/SBA-16 catalyst, the average length and stacking layers of various NiMo/Al-Ti-SBA-16 catalysts are relatively lower. The dispersion degree of MoS₂ phase over sulfide NiMo/Al-Ti-SBA-16 is improved after the incorporation of Al and Ti species into SBA-16 materials, which follows the order of NiMo/SBA-16 (0.14) < NiMo/AT-10 (0.16) < NiMo/AT-0 (0.18) < NiMo/AT-7.5 (0.20) < NiMo/AT-5 (0.21) < NiMo/AT-2.5 (0.23). Above all, sulfide NiMo/SBA-16 catalyst presents the highest slab numbers, longest slab length and lowest dispersion degree, which are adverse for HDS reaction. Al and Ti modification can make an improvement of properties for active metals and further enhance the HDS efficiency for the corresponding catalysts. However, the NiMo/AT-0 catalyst exhibits relatively lower dispersion degree of MoS₂, which may be caused by the fact that the pore size of AT-0 support is the lowest from the above BET result.