1
296
K. Bell et al. / Tetrahedron Letters 50 (2009) 1295–1297
Acetylketene 2 is significantly more reactive than ketenes such
Table 1
2
,3
Trapping of acetylketene (2) with various reagents
as 9.
We wondered if similar solid-phase reaction conditions
could be utilized to generate 2 from the corresponding acid chlo-
ride 11 (3-oxobutanoyl chloride) that can, in turn, be generated
Reagent
Product
Yield (%)
Purification
None (dimer)
Acetone
3
1
12
83
Crystallization
13
from the readily available ketene dimer 10. Analogously, solu-
Column chromatography
4:6 hexane:ethyl acetate
Column chromatography
tion-phase reactions have been proposed.1
4,15
We now report that
2-Propanol
5b (R = 2-Pr)
82
this is indeed possible when a toluene of 11 is reacted with solid
sodium carbonate in the presence of catalytic triethylamine at
2:1 hexane:ethyl acetate
Ethyl vinyl
ether
16
5.4
Column chromatography 1:9 hexane:
ethyl acetate, then methanol
ꢀ
78 °C (Eq. 3). Simple filtration gives a solution of acetylketene
2) that can readily be reacted with a wide variety of trapping
(
agents.
bottom. A round-bottomed flask was connected to the outlet and a
pressure of dry nitrogen could be applied to the bottom of the flask.
The top of the column could be vented or connected to nitrogen
pressure. The column was oven dried, evacuated, and filled with
nitrogen. After having been dried by heating under vacuum,
O
HCl
PhMe
O
O
K CO
O
2
3
O
O
ð3Þ
Cl
Et N
3
1
0
11
2
2 3
K CO (8.5 g, 0.084 mol) was added to the column. Then, dry tolu-
ene (15 mL, distilled from Na) was added. A sufficient pressure of
nitrogen was applied to the stopcock and the top was vented, so
that the flow of nitrogen through the apparatus prevented the tol-
uene from dripping through the frit. Next, the column was cooled
with dry ice/acetone, and then a portion of the above-mentioned
toluene solution of 3-oxobutanoyl chloride (11, 4 mL, 0.021 mol)
was added, followed by the addition of triethylamine (0.29 mL,
0.0021 mol) dropwise. The upward flow of nitrogen was continued,
mixing the heterogeneous mixture for 10 min. Then, the nitrogen
pressure differential was reversed and the solution of acetylketene
(2) was allowed to flow into the receiving flask that contained an
excess of the trapping agent. This was stirred for 4 h at ꢀ78 °C,
then allowed to warm to room temperature. The products were
isolated as described below, and the results are summarized in Ta-
ble 1.
One additional aspect of this reaction sequence is noteworthy. If
any of the acid chloride (11) or the acetylketene (2) fails to react as
desired, this will reduce the overall yield but does not necessarily
lead to impurities in the product. Both 2 and 11 react with water to
give the carboxylic acid 12 (Eq. 4). However, 12 was not observed,
presumably because it decarboxylated to form acetone, which is
conveniently removed with the toluene solvent.
2
. Experimental
-Oxobutanoyl chloride (11) was prepared by the reaction of
3
13
anhydrous HCl with diketene (10). Diketene (10, 6.4 mL, 83
mmol) was dissolved in 9.6 mL of toluene in a flask equipped with
a gas inlet below the level of the toluene and connected to a nitro-
gen line equipped with a bubbler. The flask was cooled to ꢀ30 °C.
Dry HCl was generated by the addition of concentrated sulfuric
acid (15 mL) to ammonium chloride (15 g) and bubbled into the
reaction mixture over 3 h. The 3-oxobutanoyl chloride (11) solu-
tion could be stored at ꢀ78 °C for one to two days.
Acetylketene (2) was generated by the elimination of HCl from
3
1
-oxobutanoyl chloride (11) using the apparatus shown in Figure
. Based on the design of Lectka,1
0–12
this consists of a jacketed col-
umn fitted with a fritted glass disk and a three-way stopcock at the
120
50
sample introduction
vent or N2
O
O
CO2
O
H O
2
2
or 11
ð4Þ
1
1
20
10
OH
1
2
reaction flask
250 mL
cooling jacket
3
. Results and discussion
We first sought to determine if acetylketene (2) is generated un-
dry ice/acetone
medium glass frit
der these reaction conditions. Simply collecting the solution from
the column, allowing it to warm to room temperature and analyz-
ing the products addresses this question. As expected, the dimer
3
-way
6
0
stopcock
1
,2
3
precipitated from the toluene solution, and was isolated by fil-
vacuum or nitrogen
tration (0.21 g, 12% yield). The yield was not further optimized.
Rather, this result was taken as evidence that 2 was indeed formed;
the acid chloride 11 is not expected to give 3. However, this does
not address the question as to whether 3 is formed in the cold col-
umn or whether 2 survives in solution long enough to exit the col-
umn and react in the bottom flask.
90
24/40 joint
In the next experiment, an excess of acetone was added as the
trapping reagent to the bottom. 2,2,6-Trimethyl-4H-1,3-dioxan-4-
1
,3,4
trapping reagent
one (1, 2.48 g)
was obtained in 83% yield after column chroma-
tography. This product is known to form from the reaction of 2
with acetone, but would not be expected from the reaction of the
acid chloride 11. Since acetone was only present in the bottom
Figure 1. Apparatus for the low-temperature, solid-phase generation of acetylke-
tene (2). Dimensions are expressed in mm.