Formation of Silyl Radical Adducts

In Table 5.1 some representative rate constants for the addition of silyl radicals to alkenes are reported [13-15]. Inspection of these data reveals that 

From Reference [13]. From Reference [14]. From Reference [15]. 

AG = —RTIn K. Therefore, the observed final composition should account for the difference in the stability of the two isomers [20]. It is worth noting that care must be taken in synthetic strategies, since isomerization can occur in situ, while accomplishing other reactions [19]. 

The addition of silyl radicals to double bonds in benzene or substituted benzenes (Reaction 5.2) is the key step in the mechanism of homolytic aromatic substitution with silanes [8,9]. The intermediate cyclohexadienyl radical 2 has been detected by both EPR and optical techniques [21,22]. Similar cyclohex-adienyl-type intermediates have also been detected with heteroaromatics like furan and thiophene [23]. 

The addition of a trialkylsilyl radical to benzene is much less exothermic than the addition to a non-activated alkene the AH of these reactions has been evaluated to be ca —50kJ/mol [9,18]. However, the rate constants for the addition of EtsSi radicals to aromatic and heteroaromatic compounds are similar to those of non-activated alkenes, i.e., 10 s [24]. Furthermore, 

---

Developments in the synthesis and characterization of stable silylenes (RiSi ) open a new route for the generation of silyl radicals. For example, dialkylsilylene 2 is monomeric and stable at 0 °C, whereas N-heterocyclic silylene 3 is stable at room temperature under anaerobic conditions. The reactions of silylene 3 with a variety of free radicals have been studied by product characterization, EPR spectroscopy, and DFT calculations (Reaction 3). EPR studies have shown the formation of several radical adducts 4, which represent a new type of neutral silyl radicals stabilized by delocalization. The products obtained by addition of 2,2,6,6-tetramethyl-l-piperidinyloxy (TEMPO) to silylenes 2 and 3 has been studied in some detail. ... 

The addition of silyl radicals to thiocarbonyl derivatives is a facile process leading to a-silylthio adducts (Reaction 5.37). This elementary reaction is the initial step of the radical chain deoxygenation of alcohols or Barton McCombie reaction (see Section 4.3.3 for more details). However, rate constants for the formation of these adducts are limited to the value for the reaction of (TMS)3Si radical with the xanthate c-C6HuOC(S)SMe (Table 5.3), a reaction that is also found to be reversible [15]. Structural information on the a-silylthio adducts as well as some kinetic data for the decay reactions of these species have been obtained by EPR spectroscopy [9,72]. 

Ando demonstrated that the photolysis of partially tert -butyl-substituted disilanes 45 in the presence of Cgo results in the formation of 1,16-adducts 46 in 54-62% yield (Scheme 8)37,38. Interestingly, unusual adducts similar to 49 were obtained as by-products with disilanes having trimethylsilyl substituents (Scheme 9). The authors explain these observations by a mechanism involving the intermediacy of silyl radicals which are generated photochemically by Si—Si bond homolysis. Initial free-radical addition of silyl... 

The product (183) is formed on irradiation of acridine (184) phenothiazine (185) crystals in which the ratio of the reactants is 3 4," This reactivity is different from the solution phase (in acetonitrile) process which affords both (183) and the dihydro dimer (186). The photoinduced electron transfer reactions of some a-silyl ethers has been investigated." The sensitizing system uses DCA/ biphenyl and irradiation at A, > 345 nm in acetonitrile/methanol. The irradiation brings about the formation of the radical cation (187) of the ether which undergoes cleavage to yield the radical (188), a hydroxymethyl equivalent. When these are generated in the presence of a,P-unsaturated esters such as (189) addition takes place affording the adducts (190). Additions to dimethyl maleate were also carried out successfully." ... 

Investigations of phthalimide photochemistry in the authors laboratories have concentrated on SET-induced excited-state reactions with a-trialkylsilyl substituted ethers, thioethers, and amines. These efforts have uncovered several interesting photochemical reactions. For example, in an early effort simple, a-silyl-substituted ethers, thioethers, and amines were observed to undergo efficient photoaddition reactions with phthalimide and its N-methyl derivative (Scheme 5.4) [20]. In these processes, thermodynamically/kinetically driven SET from the a-silyl donors 13 to excited phthalimide leads to formation of ion radical pairs 16 (Scheme 5.5). Solvent (MeOH) promoted desilylation of the cation radicals 16 and protonation of the phthalimide anion radicals 15 then provides radical pairs 17, the direct precursor of adduct 14. 

Only a few examples exist for the intermolecular trapping of allyl radicals with alkenes68,69. The reaction of a-carbonyl allyl radical 28 with silyl enol ether 29 occurs exclusively at the less substituted allylic terminus to form, after oxidation with ceric ammonium nitrate (CAN) and desilylation of the adduct radical, product 30 (equation 14). Formation of terminal addition products with /ram-con figuration has been observed for reaction of 28 with other enol ethers as well. 

The addition of (TMS)3SiH to prochiral diethyl methyl fumarate (5) gave both diastereoisomers with preferential formation of the threo isomer (Reaction 5.7) [25]. This suggests that the intermediate adduct 6 adopts a preferred conformation due to the allylic strain effect, in which the silyl moiety shields one face of the prochiral radical center, favouring hydrogen transfer to the opposite face, and therefore the threo product is predominantly formed. 

A theoretical study of the reactivity of prototype ketene CH2=C=0 with several radicals including HsSi has been reported. Scheme 5.11 shows that three adducts are possible with the corresponding energy barriers for their formation calculated at B3LYP/6-31G level of theory [75]. However, all the levels of theory used in this study predicted that silyl radical prefers to add to the carbon terminus of ketene. 

The oxidation of ketones to enones via the reaction of their silyl enol ethers with 2,3-dichloro-5,6-dicyano-l,4-benzoquinone (DDQ) has been suggested originally to proceed via allylic hydride abstraction [195-198]. A recent reinvestigation, however, [199] has established the intermediate formation of a substrate-quinone adduct 96 which was presumably formed from a geminate radical ion pair after electron transfer. Decomposition of the adduct then finally afforded the observed enone product 97. Recently, the critical role of solvent polarity in the formation of 97 from the PET reaction of 93 and chloranil has been identified by time-resolved spectroscopy [200]. 

Irradiation of the cyclohexenone 32 with the tertiary a-silylamine EtjNCHjSiMe, affords two adducts 32a and 32b in a product ratio that is solvent dependent (Scheme 20). In acetonitrile solution, the formation of 32a as major product can be attributed to selective a-CH deprotonation adjacent to the silyl substituent, whereas the formation of 32b as the major product in methanol solution can be attributed to nucleophihc displacement of the silyl group followed by addition of the radical EtNCHj to either the neutral enone or its anion radical. Intramolecular analogs of this reaction have also been studied for substrates 33 and 34 and display highly selective a-CH deprotonation adjacent to the silyl substituent in acetonitrile solution to yield 33a and 34a and desdylation in methanol solution to yield 33b and 34b (Scheme 21). Intramolecular addition can also compete with trans,cis isomerization of acychc enones, whereas intermolecular addition is too slow to do so. " ... 

---