3818
Communications to the Editor On the Role of Alkoxysilylketenes and the Absence of Carbon-Silicon Double Bonds in the Gas-Phase Decomposition of Trimethylsilyldiazoacetates' Sir: It has become recently f a ~ h i o n a b l e to ~ ? claim ~ silaethylenes as intermediates in the formation of products from trimethylsilylcarbenes. W e present here a case in which such an inference would be reasonable a t first glance, but in which a closer look reveals that more circuitous reaction paths are being followed. When methyl trimethylsilyldiazoacetate (1) was evaporated through a Pyrex tube maintained a t 360' and the products then treated with methanol, three compounds were 3 is almost certainly formed by an ino b ~ e r v e dCompound .~ sertion of carbene 2 into the proximate carbon-hydrogen bond of the methoxyl group. Although p-lactones have not been isolated previously, they have been p ~ s t u l a t e d ,and ~ larger ring lactones have been found in similar reactions.6 T h e ' H N M R spectrum of 3 (CC14, T 5.6-5.9, m, 1 H; 5.9-6.3, m, 1 H; 6.5-6.9, m, 1 H; 9.83, s, 9 H) and infrared absorptions at 1813 and 1095 cm-' serve to establish the structure.' N?
1I
(CH ),SiCCOOCH, 1
-
CH,
I (CH,),SiCHCOOCH,
(CH,),Si(%OOCH, 2
k!?
3
(CH,),SiCHCOOR + (CH,),SiCHCOOR
1. 360'
(CH3)3SiCCOOCH,
2.CD3OD
OCH, 3
OCD,
Pyrolysis of 1 a t 360° followed by addition of benzene led to 8 in 12% yield.g The structure follows from the 'H N M R spectrum (CC14, 7 6.57, s, 3 H; 8.39, s. 3 H: 9.78, s, 6 H) and the presence of infrared bands a t 2085 and 1088 cm-l.' A t 60' in alcohols 8 gave the twice-exchanged products quantitatively. Thus the source of 5 is not the silaethylene but rather ketene 8, and there is no need to invoke n-bonded silicon species, a t least in the gas phase.
-
(CH7)&3iCCOOCH1 " "
1. 360'
2. benzene
Compound 4 is the evident product of methanolysis of the ketene 6, itself formed by Wolff rearrangement of carbene 2.8
-
OCH,
I
OCH,
4
(CH,),SiCHCOOR I
OR
77
(CH,),SiL-C=o
6
/
6
CH3 I (CH,),SiC=C=O I
OCH, 8
(CH,),Si