Liquid phase silanization reaction

Table 1 shows the kinetic data available for the (TMSjsSiH, which was chosen because the majority of radical reactions using silanes in organic synthesis deal with this particular silane (see Sections III and IV). Furthermore, the monohydride terminal surface of H-Si(lll) resembles (TMSjsSiH and shows similar reactivity for the organic modification of silicon surfaces (see Section V). Rate constants for the reaction of primary, secondary, and tertiary alkyl radicals with (TMSIsSiH are very similar in the range of temperatures that are useful for chemical transformations in the liquid phase. This is due to compensation of entropic and enthalpic effects through this series of alkyl radicals. Phenyl and fluorinated alkyl radicals show rate constants two to three orders of magnitude... 

The double promoter process involves the successive application of liquid promoter solutions of vinyltrichlorosilane (VTS) and 3-chloropropyltrimethoxy-silane followed by successive cure cycles in dry N2 at 90°C after each application and before photoresist application. The double promoter process evolved because it was felt that the silane reaction with the SiOH surface groups of low temperature oxides was incomplete for a single promoter application, and because vapor silane equipment did not exist at that time. Interestingly, a double HMDS liquid promoter process failed to yield adequate adhesion as well. Later in time, the successful but somewhat complex double promoter process was replaced by the vapor phase HMDS process in the Star 1000 (or 2000) then superior resist image adhesion was obtained on all four oxide substrates with all the photoresists tested. Before the advent of the HMDS vapor priming in standalone or wafer track equipment module chambers, liquid priming solutions were widely used, especially in development areas. 

In this part, we wish to focus on the study of two types of silanes. Aminoorganosilanes are special members of the alkoxysilanes group. They carry the catalyzing amine function, required for chemical bonding with the silica surface, inside the molecule. This makes them more reactive than other organosilanes and reduces the complexity of the liquid phase reaction system to be studied. Only three components, silica, silane and solvent, are present. Furthermore there is a large interest in the reaction mechanism of silica gel with APTS, since this aminosilane is the most widely used compound of the organosilane family. 

For the modification of silica with aminosilanes, the liquid phase procedure is usually applied. Only few studies have described the vapour phase APTS modification.6,7 The modification proceeds in three steps, (i) A thermal pretreatment of the silica determines the degree of hydration and hydroxylation of the surface, (ii) In the loading step, the pretreated substrate is stirred with the silane in the appropriate solvent, (iii) Curing of the coating is accomplished in a thermal treatment. On industrial scale ethanol/water is used as a solvent, on lab-scale an organic solvent is used. The reasons for this discrepancy is the increased control on the reaction processes, possible in an organic solvent. This will be clarified by the discussion of the modification mechanism in aqueous solvent and the effect of water in the different modification steps. 

The CSC precursor build-up has been studied after modification of the silica gel surface from the gas phase. This gas phase modification involves the deposition of one molecular layer at the time. For thicker coatings, a cyclic procedure is needed. Liquid phase modification of the silica surface may also yield valuable ceramic precursors. The precursor molecular structure and layer thickness is controlled by other parameters compared to gas phase procedures. Parameters such as reaction solvent, silane concentrations and presence of water are of primal importance. Those have been discussed in detail in chapter 9. In this chapter, the application of silica modified with aminosilanes, will be discussed. The aminopropylsilica is used as a prototype compound for the production of ceramics by liquid phase chemical surface coating. 

Silicon carbides are generally synthesized by the pyrolysis of precursors, prepared by liquid phase methods. One possible way for precursor synthesis is the addition of carbon black or sucrose, to a gelling silica.8 In this method, the carbon is introduced from an external source. A more intimate contact between the carbon and silicon in the precursor is assured with the use of organometallic polymer precursors. The use of silane polymers for silicon carbide production was initiated by Yajima.9,10 Polymers having a -[Si-C]- backbone are crosslinked and pyrolysed to yield SiC." In the initial work, dimethyldichlorosilane was used as a starting monomer, which was subjected to a sodium catalyzed polymerization (reaction (C)). 

Dehydrodimerization. On excitation with a mercury vapor lamp, mercury is converted to an excited state, Hg, which can convert a C—H bond into a carbon radical and a hydrogen atom. This process can result in dehydrodimerization, which has been known for some time, but which has not been synthetically useful because of low yields when carried out in solution. Brown and Crabtree1 have shown that this reaction can be synthetically useful when carried out in the vapor phase, in which the reaction is much faster than in a liquid phase, and in which very high selectivities are attainable. Secondary C—H bonds are cleaved more readily than primary ones, and tertiary C—H bonds are cleaved the most readily. Isobutane is dimerized exclusively to 2,2,3,3-tetramethylbutane. This dehydrodimerization is also applicable to alcohols, ethers, and silanes. Cross-dehydrodimerization is also possible, and is a useful synthetic reaction. 

The silanization reaction in liquid phase (usually in dry toluene) is technologically more convenient, although it is practically impossible to ensure the complete absence of water in a reaction mixture that leads to hydrolysis of chlorosilanes. Using deuterium exchange, Roumeliotis and Unger [55] analyzed surface silanol concentration before and after the modification with different reagents (Table 3-2). 

The intermediate metal hydride has been isolated on occasion for Co and Mn , and Eq. (b) has actually been used to prepare silicon-metal bonds (see 5.2.3.2.2.). Inspection of Table 1 reveals the ease of reaction of Co2(CO)g compared with the other carbonyls. Normally this reaction is performed simply by condensing volatile silane onto the carbonyl in the absence of solvent and then allowing rapid reaction in the liquid phase at room temperature, but for the remaining carbonyls it is necessary to use elevated temperatures and sealed, evacuated tubes. The products are volatile and readily purified by vacuum fractionation or sublimation, but are often oxygen and moisture sensitive. The route is most efficient for RjSi derivatives of Co, Mn and Re, which are not generally obtainable by the reactions of silicon halides with metal carbonyl anions (see S.8.3.3.I.). In this way lCo(SiR,)(CO -] = Et, Phj, Clj -, (OEt)j, F/, ... 

An interesting question is whether these results are characteristic of vapor-phase silanizations or whether they extend to liquid-phase reaction conditions. Silanizations are most commonly performed in solutions. We are pursuing this question currently. 

Silanisation is a heterogeneous reaction. Silanes can be in the gas or liquid phase or in solution. The reaction is carried out at elevated temperatures, depending on the volatility of the silane and solvent, in a vessel under gentle stirring or in a fluidized bed reactor. To enhance the kinetics, catalysts are added. With chlorosilanes, organic bases are added as acid scavengers acids are employed in case of alkoxysilanes as reagents. By-products must be carefully removed by extraction with solvents. 

The kinetics of the thermal decompositions of 2,2-difluoroethylmethyl-difluorosilane (in the gas phase at temperatures from 182 to 246 °C and at pressures from 20 to 186 Torr) and 2,2-difluoroethyltri-n-butoxysilane (gas phase, 240—320 °C, 7—97 Torr liquid phase, 230—266 °C, silicone fluid MS 550 as solvent ) into vinyl fluoride and, respectively, methyltrifluoro-silane and tri-n-butoxyfluorosilane have been studied. The reactions are unimolecular and presumably proceed via four-centre transition states, as postulated for thermal decompositions of other j8-halogenoalkyl derivatives of silicon ... 

Silane and ethylene are present at very high concentrations so that homogeneous nucleation dominates the process. As the gases enter into the hot part of the injector, the silane will decompose and form small Si liquid droplets or solid microcrystals, depending on the temperature. The ethylene will also take part in the reaction, forming microparticles of Si C. It has been noted [39] that even a small addition of hydrocarbons converts the Si droplets to stable particles of Si C (or non-stoichiometric SiC). The stability may, in a hand-waving circumstantial way, be intuitively understood from a solubility point of view. The solubility of carbon in silicon is very low, thus, when carbon is added to the Si droplets, the phase will be solid rather than liquid. 

The reaction of several a,/ -unsaturated ketones with allyltrimethyl silane in either [C4Ciim][BF4] or [C4Ciim][PF6] has been compared with several classical solvents, see Scheme 9.45.[152] With indium(III) chloride as catalyst, only in some cases was a higher activity observed in the ionic liquids, but the differences relative to the commonly used solvent, dichloromethane, were small and no obvious advantage in performing the reaction in the ionic media was evident. Very high catalyst loadings, ca. 20-50 mol%, were used but no attempts were made to recycle the catalyst phase. 

---