The nature of the Ni(II) phase formed on silica during the preparation by deposition−precipitation (DP) of Ni/SiO2 samples is shown to depend on the silica surface area and the time of deposition−precipitation. Ni/SiO2 samples have been characterized by XRD, IR, EXAFS, TPR, TEM, STEM-EDX, and BET. With silica of low surface area and for short DP time (≤4 h), the Ni(II) phase is mainly a turbostratic nickel hydroxide with a small amount of 1:1 nickel phyllosilicate. With silica of high surface area and for short DP time (≤100 min), the Ni(II) phase is mainly a 1:1 nickel phyllosilicate. For longer DP time (>4 h) and with both types of silica, the Ni(II) phase is an ill-crystallized 1:1 nickel phyllosilicate. However, the latter is better crystallized with silica of low surface area. Almost the same Ni(II) phases were obtained whether silica was porous or not. However, the Ni(II) phase is better crystallized and the interface with the support is larger with nonporous silica than with porous one.

A molecular mechanism explaining the chemical phenomena that occur during the preparation of silicasupported nickel catalysts by the method of deposition-precipitation is proposed. This mechanism takes into account the fact that either 1:1 nickel phyllosilicate or nickel hydroxide may be formed on silica during deposition-precipitation, depending on the silica surface area and the deposition-precipitation time, as shown in our previous paper (Burattin, P.; Che, M.; Louis, C. J. Phys. Chem. B 1997, 101, 7060). The mechanism is mainly based on the kinetic competition between two types of reactions: (i) Ni-O-Si heterocondensation/ polymerization, which leads to the growth of 1:1 nickel phyllosilicate on a Ni(II) brucitic layer bonded to silica; (ii) Ni-OH-Ni olation/polymerization, which leads to the formation and the growth of nickel hydroxide on a Ni(II) brucitic layer bonded to silica. The former type of reaction is faster than the latter but is limited by the concentration in solution of silicic acid arising from silica dissolution and by its diffusion rate. The changes in the pH curves, the yield of Ni(II) deposition-precipitation, and the nature of the supported Ni(II) phase during deposition-precipitation under different experimental conditions are also interpreted on the basis of the mechanism proposed.

The influence of the parameters of the deposition−precipitation method used to prepare Ni/SiO2 samples on the size of nickel metal particles was investigated by temperature-programmed reduction, transmission electron microscopy, and thermogravimetry. The results show that the average metal particle size, which varies between 27 and 79 Å, depends on the nature and the reducibility of the supported Ni(II) phase (nickel hydroxide or 1:1 nickel phyllosilicate), and on the extent of the interface between the supported Ni(II) phase and silica. These parameters themselves depend on the characteristics of the silica support (surface area and morphology) and on the preparation parameters (urea and silica concentrations). At the interface, a nickel phyllosilicate is detected whatever the nature of the supported Ni(II) phase. The metal particles are smaller and the size distribution is narrower when the supported phase is a 1:1 nickel phyllosilicate (d̄ ≤ 50 Å, 10−100 Å) than when it is a nickel hydroxide (d̄ ≈ 80 Å, 20−280 Å). The metal particles are smaller when the extent of the Ni(II) phase−silica interface increases. This arises from the comparison of Ni/SiO2 samples prepared from silicas of high and low surface area, and from nonporous and porous silica.