Nickel hydrosilation

It was as early as 1946 that de Lange and Yisser (1) published their finding that the precipitate formed on adding alkali hydroxide to a suspension of diatomaceous earth in a nickel salt solution is not merely a mixture of nickel hydroxide and diatomaceous earth, but that a reaction has taken place between the hitherto supposed inert carrier and the nickel compound. They ascertained the occurrence of such a reaction from the very great increase in accessible surface and the more difficult reduction of the reaction-product as compared with that of nickel hydroxide, and, on the basis of their X-ray work, suggested the formation of nickel hydrosilicates having a layer structure. Further, they assumed that after reduction... 

In the first place, we learned more about the formation of nickel hydrosilicates under certain circumstances from an investigation of Van Eijk van Voorthuysen and Franzen (2). These investigators made a number of preparations by combining boiling dilute solutions of nickel nitrate and alkali silicate in various proportions. In order to find out to what extent co-precipitation is required for the formation of hydrosilicate structures, acid was added to a nickel hydroxide suspension in a silicate solution by which silica is precipitated, or conversely, alkali was added to a suspension of silica gel in a nickel nitrate solution. Some of the preparations were subjected to a hydrothermal treatment at 250° C. for 50 hrs. with a sufficient quantity of water for developing the best possible structure. 

Fig. 2. Thermal analysis curves of nickel hydroxide and nickel hydrosilicates.

For some systems, however, the influence of the temperature on the phase composition can be predicted based on chemical considerations. For instance, the composition of bismuth molebdatc catalysts is believed to be determined by the nature of the molybdate anion present in solution [28] which is dependent on the solution temperature. For Ni/SiCh catalysts the differences between catalysts prepared at high or low temperatures are explained by the formation of nickel hydrosilicate at high temperatures, while at low temperature the main precipitate is nickel hydroxide [29]. 

Nickel reacts even more strongly than copper with silica indeed, silica completely reacts with nickel(II) ions precipitating at 90°C to nickel hydrosilicate, provided sufficient nickel is supplied. As a result, the structure or the support completely disappears from the electron micrographs, only the hydrosilicate platelets being in evidence. 

In the preparation of Ni/Hp catalysts by the deposition-precipitation method (DP), nickel hydrosilicates are formed mainly but not exclusively in the external surface of the Hp zeolite. The strong metal-support interaction induced by the DP preparation method prevents the Ni metal particles from sintering during the activation of the catalysts (calcination and reduction) and a homogeneous distribution of small nickel particles is obtained. The catalyst prepared by DP showed better catalytic activity in the hydrogenation of naphthalene than the catalyst prepared by cationic competitive exchange. 

Evidence has recently been provided by the work of de Lange and Visser (10) on nickel catalysts deposited on kieselguhr. This has established that the normal lattice of metallic nickel does not occur in materials prepared in this way, and that the reduced active catalyst is obtained as a result of the attack of hydrogen on a nickel hydrosilicate. By interaction of the diatomite and the nickel hydroxide deposited on it, an entirely new lattice is produced which provides a greatly increased total surface of catalyst, and which leads in the finished state to a widely dispersed and very stable arrangement of nickel atoms. It will be of great interest to obtain further evidence as to the exact distance between pairs of nickel atoms in this catalyst, which is of exceptionally high activity. 

The phenomenon of neoformation of a mixed compound has been extensively studied in the case of the Ni/SiOj system, which forms nickel phyllosiUcate of 1 1 type, also referred to as nickel hydrosilicate. 1 1 nickel phyllosilicate exhibits a stacked structure, each layer consisting of a brucite-type sheet containing Ni(II) in octahedral coordination and a sheet containing linked tetrahedral Si04 units (Figure 14.3a). 

The work by Coenen and others confirmed that the reduction of nickel hydrosilicates is inhibited by water in the lattice. This had led to the early problems in reducing the catalyst industrially because water forms continuously during reduction, not only as the nickel compounds are converted to metal but also as the remaining hydroxyl layers gradually decompose. Catalysts become active only after reduction at temperatures in the range 300-400°C, and even then a significant proportion of the nickel oxide is umeduced and the lattice still contains water. 

It was concluded that the active catalyst was a nickel suboxide, probably because of the difficulty in reducing nickel hydrosilicates in the catalyst. 

Although the actual reaction mechanism of hydrosilation is not very clear, it is very well established that the important variables include the catalyst type and concentration, structure of the olefinic compound, reaction temperature and the solvent. used 1,4, J). Chloroplatinic acid (H2PtCl6 6 H20) is the most frequently used catalyst, usually in the form of a solution in isopropyl alcohol mixed with a polar solvent, such as diglyme or tetrahydrofuran S2). Other catalysts include rhodium, palladium, ruthenium, nickel and cobalt complexes as well as various organic peroxides, UV and y radiation. The efficiency of the catalyst used usually depends on many factors, including ligands on the platinum, the type and nature of the silane (or siloxane) and the olefinic compound used. For example in the chloroplatinic acid catalyzed hydrosilation of olefinic compounds, the reactivity is often observed to be proportional to the electron density on the alkene. Steric hindrance usually decreases the rate of... 

Recently we found that freshly prepared nickel powder is an efficient hydrosilation catalyst when continuously irradiated(49). 

The square-planar complex (34) NiCI2-(P-/i-Bu3)2 was a better catalyst than the tetrahedral complex NiBr2 (PPh3)2 for hydrosilation of styrene with trichlorosilane at temperatures of 150°-170°C. A nickel(0) complex, Ni[P(OPh)3]4, was as good as NiCl2(NC5H5)4, which was best among known nickel catalysts for this reaction. Addition of copper(I) chloride... 

The activity of Ziegler-type systems such as M(acac) -AlEt3 (M = Cr, Mn, Fe, Co, or Ni acac = acetylacetonate) was examined with 1-olefins and triethyl- or triethoxysilanes (55). Systems with nickel or cobalt showed low activity for hydrosilation but isomerized the olefin and were reduced to the metal. The study was extended to dienes and acetylenes (56). Isoprene gave the same products with these catalysts as are made with chloroplatinic acid. Penta-1,3-diene with Pt gave l-methylbut-2-en-ylsilanes. The Ziegler catalysts gave mainly penta-2-enylsilanes... 

With dienes and acetylenes, only M = Fe, Co or Ni showed activity. The Fe and Co systems gave many side products. Therefore, the Ni catalyst was studied in greater detail. Bis-(7r-cycloocta-l,5-diene)nickel(0) was equivalent to the Ziegler catalyst in hydrosilation of penta-1,3-diene. 

These studies were extended to hydrosilation of cyclopentadiene with trichlorosilane (52). This is most difficult with platinum catalysts. Palladium complexes favored production of 1 1 adducts as a mixture of 3- and 4-trichlorosilylcyclopentene. Nickel complexes produced substantial amounts of 1 2 adducts as trichlorosilyl-substituted 4,7-methylene-4,7,-8,9-tetrahydroindanes, with the exception of nickel tetracarbonyl, which was very active and selectively formed almost exclusively 3-trichlorosi-lylcyclopentene with no 1 2 adduct. 

On comparing reducibility and the results of differential thermal analysis, it can be seen that the order of difficulty with which the preparations are reduced is the same as the order of difficulty with which they are decomposed. One might therefore expect the decomposition of hydrosilicate into nickel oxide, silica, and water to determine the velocity of the reduction... 

Hydrosilicate formation is also in evidence in the Cu(II)-Si02 system. Via precipitation from a homogeneous solution one can obtain highly dispersed copper oxide on silica (cf. above, Fig. 9.10, where it should be noted that the Cu case is more complicated than the Mn one in that intermediate precipitation of basic salts can occur). Reaction to copper hydrosilicate is evident from temperature-programmed reduction. As shown in Fig. 9.12 the freshly dried catalyst exhibits reduction in two peaks, one due to Cu(II) (hydr)oxide and the other, at higher temperature, to Cu(II) hydrosilicate. Reoxidation of the metallic copper particles leads to Cu(II) oxide, and subsequent reduction proceeds therefore in one step. The water resulting from the reduction of the oxide does not produce significant amounts of copper hydrosilicate, in contrast to what usually happens in the case of nickel. 

Nickel and palladium compounds have also been found to exhibit an interesting reactivity pattern in the hydrosilation of diynes (vide infra). 

The main product in hydrosilation of a,P-unsaturated ketones and aldehydes catalyzed by chloro-platinic acid, platinum on alumina, or metallic nickel is the corresponding silyl enol ether. With nickel catalyst, product distribution is highly dependent on the enone structure. Hydridosilanes add to a, -unsaturated esters, producing the corresponding silyl enolate as well as carbon silylated products. The course of addition depends on substrate structure and the hydridosilane utilized. Thus, triethylsilane undergoes 1,4-addition to methyl acrylate in the presence of chloroplatinic acid, while trichlorosilane with either chloroplatinic acid or Pt/C gives the -silyl ester (Scheme 65). ... 

In the DP samples, after two hours deposition-precipitation, there is formation of a clear shoulder with DP time at about 1005 cm, which according to the literature [7], points to the presence of 1 1 nickel phyllosilicate. This is more clearly seen in the subtraction spectra. It appears that in the case of the cationic competitive exchange-prepared sample the formation of hydrosilicate species is only incipient. In fact, these... 

Silanes normally reduce aldehydes or ketones under catalytic conditions to the silyl ethers. However, with certain catalysts such as nickel sulfide,Co2(CO)8, or (Ph3P)3RhCl, carbonyl compounds react with silanes to yield an equilibrium mixture of enol silyl ethers (Scheme 17). In a similar vein, the silyl-hydroformylation reaction of cycloalkenes with CO and silanes may be a practical way to prepare enol silyl ethers. An example is the preparation of compound (49). Catalytic 1,4-hydrosilation of a, -un-saturated ketones or aldehydes gives the corresponding enol silyl ethers. The reaction is similar to the reductive silylation referred to previously, but the reaction conditions are neutral and milder. The formation of the enol silyl ether (50) is outlined below. ... 

---