I.K. Ball et al. / Journal of Organometallic Chemistry 599 (2000) 185–194
193
A reaction scheme that accounts for the formation of
all products is shown in Scheme 1. Due to the strength
of the SiꢀO band, the initial process would be ho-
molytic cleavage of the skeletal CꢀO bond generating
trimethylsilyloxyl and acetyl radicals (step 1). Two reac-
tion pathways are possible for the further reaction of
the dimethyethylsilyloxyl radical:
1. abstraction of hydrogen to form trimethylsilanol
(step 2),
2. elimination of an ethyl radical with the formation of
dimethylsilanone (step 3).
processes are fast relative to this, then loss of
trimethylsilyl acetate in the neat reaction will be first-
order as observed.
As carbon monoxide is present at the start tempera-
ture and rises to a maximum, the acetyl radical formed
in step 1 must thermally decompose to give carbon
monoxide and a methyl radical (step 7). After reaching
a maximum value, the carbon monoxide abundance
then decreases with time. Oxidation, presumably at the
wall surface where there is a large presence of oxide,
provides carbon dioxide (step 8).
Ethene is not present at the start, but grows in
continuously with time later in the thermolysis. Ethyl
radicals, produced into the reaction sphere, can propa-
gate to form ethene (step 9). Methane forms later than
ethene in the reaction and is probably produced from a
methyl radical hydrogen abstraction (step 10).
A third possible reaction pathway, silyloxyl radical
attack at a silicon centre, as far as we are aware, has
not as yet been observed. Hydrogen abstraction by
trimethylsilyloxyl radicals is known to occur with a
variety of substrates [18–21]. In particular, hydrogen
abstraction from bistrimethylsilyl peroxide giving the
Me3SiOOSiMe2CH2 radical has been observed. The
product of hydrogen abstraction, trimethylsilanol, read-
ily undergoes condensation to hexamethyldisiloxane
and water [17], and an analogous process (step 6)
accounts for the formation of diethyltetramethyldisilox-
ane in the present system. The hexamethylcyclotrisilox-
ane and octamethyltetrasiloxane observed almost
certainly result from the oligomerization of dimethyl-
silanone [22,23] (steps 4 and 5), formed by ethyl radical
elimination from the dimethylethylsilyloxyl radical (step
3).
The acetyl radical formed in step 1 is known to
eliminate carbon monoxide [24–27] (step 7). Methyl
radicals are also formed in this step, leading to methane
formation by hydrogen abstraction (step 10). Ethanoic
acid may be formed by the formation of an acetoxyl
radical, followed by hydrogen abstraction as described
above. An alternative pathway for the formation of
ethanoic acid is via internal hydrogen abstraction from
dimethylethylsilyl acetate (Eq. (8)):
Acknowledgements
We thank the EPSRC for a research studentship (to
A.T.)
References
[1] P.G. Harrison, A. Torr, J. Organomet. Chem. 538 (1997) 19.
[2] V.D. Sheludyakov, V.I. Zhun, M.I. Shumilin, V.N. Bochkarev,
T.F. Slyusarenko, Zh. Obshch. Khim. 58 (1988) 1583.
[3] M.I. Kabachnik, L.S. Zakharov, G.N. Mochanova, T.D. Droz-
dova, P.V. Petrovskii, Izv. Akad. Nauk. SSSR Ser. Khim. 7
(1989) 1664.
[4] R.G. Mirskov, S.V. Basenko, M.G. Voronkov, Zh. Obshch.
Khim. 57 (1987) 1907.
[5] S. Ambasht, S.K. Chiu, P.E. Peterson, J. Queen, Synthesis 4
(1980) 318.
[6] J. Pola, Z. Papouskova, V. Chvalovsky, Coll. Czech. Chem.
Commun. 38 (1973) 3163.
[7] J. Pola, Z. Papouskova, V. Chvalovsky, Coll. Czech. Chem.
Commun. 40 (1975) 2487.
[8] P.G. Harrison, D.M. Podesta, Organometallics 13 (1994) 1569.
[9] T.N. Srivastava, M. Onyszchuk, Can. J. Chem. 41 (1963) 1244.
[10] K. Licht, H. Kriegsmann, Z. Anorg. Allg. Chem. 323 (1963) 190.
[11] W. Weltner, J. Am. Chem. Soc. 77 (1955) 3941.
[12] C. Eaborn, F.M.S. Mahmoud, R. Taylor, J. Chem. Soc. Perkin
Trans. 2 (1982) 1313.
[13] A.R. Bassindale, A.G. Brook, P.F. Jones, J.M. Lenna, Can. J.
Chem. 53 (1975) 332.
[14] D. Seyferth, F.M. Armbrecht, J. Am. Chem. Soc. 91 (1969)
2616.
[15] J.G. Martin, M.A. Ring, H.E. O’Neal, Organometallics 5 (1986)
1228.
[16] I.M.T. Davidson, S. Ijadi-Maghsoodi, Organometallics 5 (1986)
2086.
[17] A.A. Turovskii, V.V. Petrenk, N.A. Turovskii, A.K. Litkovets,
Zh. Obshch. Khim. 47 (1977) 2556.
[18] A.G. Davies, B.P. Roberts, M.-W. Tse, J. Chem. Soc. Perkin
Trans. 2 (1977) 1499.
(8)
The loss of dimethylethylsilyl acetate is first order
over the whole temperature range studied. The value of
the activation energy (110(4) kJ mol−1) is relatively
low. The fact that the activation energy is much lower
than is reasonable for the SiOꢀC(O)CH3 bond dissocia-
tion energy indicates that step 1 cannot be a totally
gas-phase process. We have previously observed that
the presence of available surfaces can facilitate dissocia-
tive processes with a decrease in activation energy
[28,29], and propose that such enhancement is also
occurring in the present case.
If Me3SiOꢀC(O)CH3 bond homolysis is the rate-de-
termining step and the rates of all other subsequent
[19] J.A. Babon, J.P. Goddard, B.P. Roberts, J. Chem. Soc. Perkin
Trans. 2 (1986) 1269.