From Trisilanes

SiHjSiHjSiHj - SiH + SijHs SiHj SiHj SiHj S1H4 + SiHSiH, 

It is already well established that SijH decomposes via a H shift. Similar mechanisms can deflnitely account for the products formed in the SijHg system. However, the formation of SiHj and SijHj from trisilane is also consistent with an alternate mechanism involving a silicon bridge as the transition state, as shown in equation (50). Ring and coworkers argued against this 

Sakurai and coworkers have s mthesized 1,2,3-trisilacycloheptane derivatives and have shown that their photolysis with a low-pressure Hg lamp under a Nj atmosphere gave substituted silylenes as products  

The fact that the Si atom in the center is the one to be squeezed out to breed a silylene means that a silicon-bridged transition state like the one shown in equation (50) is likely. The same authors have also synthesized 5,6,11-trisilaspiro-4,6-undecane, which undergoes a similar photochemical decomposition to give a cyclic silylene [equation (52)]. 

Similarly, Ishikawa, Kumada, and coworkers formed SiMePh from the photolysis of 2-phenylheptamethyltrisilane with 253.7-nm radiation [equation (53)]. 

---

Diradical 52 (Ar = Mes) and the corresponding 2,4,6-triisopropylphenyl substituted 52 (Ar = Tip), which is available in the same manner starting from trisilane 46 (Ar = Tip), were the first directly observed diradicals of this type. 

I have focused on methyl derivatives of nonsiloxane organosilicon backbones to achieve a useful comparison of polymer backbones. There are studies on materials with pendant groups other than methyl and backbones other than siloxane. The most useful of these studies is the direct liquid-surface-tension measurement by Feher and co-workers (96) of silane oligomers from trisilane to heptasilane, including some branched species (Table XV) (96-99). The data are useful because they answer the question of the surface activity contribution of the Si-H group. The situation with SiH-containing siloxanes... 

It was reported that a substituent such as 8-dimethylaminonaphth-l-yl group, which may be able to coordinate with the generated silylene intramolecularly, facilitates the nickel and palladium-catalyzed extraction of silylene species from trisilane 96 (Eq. 50) [87]. 

Even in the presence of such negative evidence, the insertion of silylenes into Si-Si bonds is not necessarily an absolutely impossible reaction. The breeding modes of producing silylenes from trisilanes and polysilanes as observed by Sakurai and coworkers and by Ishikawa, Kumada, and coworkers are actually the reverse of such Si-Si bond insertion reactions, and they point to the possibility of a transition state as shown in equation (50). 

In the deposition of a-Si H the dissociation of the process gas silane leads to the formation of hydrogen, disilane, trisilane, higher-order silanes, and a solid film. Silane is depleted. From the consumption of silane, one may estimate the deposition rate of the solid film. The difference in the silane partial pressure between... 

FIG. 32. The partial pressures of disilane and trisilane. The dashed line is an extrapolation of the disilane partial pressure in the o -regime. (From E. A. G. Hamers, Ph.D. Thesis, Universiteit Utrecht. Utrecht, the Netherlands, 1998.)... 

The partial pressures of disilane and trisilane are shown in Figure 32, for the same process conditions as in Figure 31. Both partial pressures increase as a function of pressure. Around the a-y transition at 30 Pa the disilane partial pressure increases faster with increasing pressure, as can be seen from the deviation from the extrapolated dashed line. The disilane partial pressure amounts to about 1 % of the total pressure, and the trisilane partial pressure is more than an order of magnitude lower. Apparently, in the y -regime the production of di- and trisilane is enhanced. 

We think two mechanisms take place at the same ime. Besides the P -elimination mechanism, which has been described above as a kind of disproportionation, an a-elimination like the Harrod mechanism seems also to take place. From the experimental results, the polymerization of trisilanes or tetrasilanes yields preferably the dimeric species, hexa- or octa-silanes respectively. In addition, oligosilanes with odd numbers of silicon atoms were formed. We do not yet understand why these hexa-and octasilanes were formed in the iso-forms. 

The addition of bis(2,6-diisopropylphenyl)silylene, as a reactive divalent species, to CgQ or Cyo yields the [2+1] cycloadduct 351 as the ring-closed 1,2-bridged isomer (Scheme 4.73) [405-407]. The silylene was prepared in situ from the trisilane 350 by photolysis with a low-pressure lamp in toluene solution. 

One novel and interesting method of generating a silacarbonyl ylide occurred through the addition of a carbonyl species with a silylene formed under photolytic conditions. Komatsu and co-workers (177) found that photolysis of trisilane (315) in solution with a bulky carbonyl species led initially to the formation of a silacarbonyl ylide followed by a dipolar cycloaddition of an olefinic or carbonyl substrate. Reaction of simple, nonbulky aldehydes led to only moderate yields of cycloadduct, the siladioxolane. One lone ketone example was given, but the cycloadduct from the reaction was prepared in very low yield (Scheme 4.89). 

The photolysis or thermolysis of certain 1,3-bis(diazomethyl)trisilanes furnished products that are derived from bicyclic pyrazolines formed by intramolecular 1,3-dipolar cycloaddition at the Si= bond of a diazosilene intermediate (343,344). 

The first successful attempt to form multilayers by self-assembly was made by Netzer and Sagiv [17], who formed a layer of OTS, modified so as to have a double bond at the extremity. They then used the reactions indicated in Figure 6.6 to produce an OH group at the terminal position. It was then possible to react a further layer of molecules bearing the trisilane group and thus, at least in principle, a multilayer could be formed. X-ray diffraction studies [360] showed, however, that, even for as little as three layers, the structure obtained was far from perfect. 

Treatment of the trisilane Me3SiSiMe2SiMe3 with TBAF in HMPA, however, does not proceed by formation of SiMe3 instead, the more stable disilanyl anion Me3SiMe2Si is formed in equilibrium, as was concluded from the products which were obtained when the reaction was performed in the presence of aldehydes16. 

---