J. Peng, Y. Deng / Tetrahedron Letters 42 (2001) 403–405
405
hexanone oxime in different ionic liquids was carried
out (entries 1, 2 and 3). If only pure ionic liquid was
used as catalytic medium there was no detectable trans-
formation. The transformation could, however, proceed
effectively when both ionic liquid and acidic phosphor-
ated compound, e.g. phosphorous pentachloride were
used. Only a small amount of ionic liquid, i.e. 5 mmol
(ca. 1 ml) was used and effective enough in comparison
with the related previous work,6 in which the ratio of
substrate(mmol)/solvent(ml) was 70.7 mmol/45 ml,
which is much smaller than that of 10 mmol/1 ml used
in this work. Although there was no great difference in
conversion between entries 1–3, the experimental re-
sults indicated that almost 100% of conversion and 99%
of selectivity for o-caprolactam could be achieved with
BPyBF4 ionic liquid (entry 2), indicating that both the
cation and anion of the ionic liquid were playing a role
in the reaction. The main side reaction was deoxima-
tion, and therefore cyclohexanone was formed. Some
other by-products (ca. 1% in concentration), including
a-chlorocyclohexanone (excluding entry 10) and cyclo-
hexane were also detected. The amounts of the ionic
liquid added had strong impact on the catalytic perfor-
mance. It is interesting to note that both the conversion
and selectivity were obviously reduced when the
amount of ionic liquid added was increased from 5 to
15 mmol (entries 3, 4 and 5). Slightly lower conversion
and selectivity would be obtained if the ionic liquid was
mixed together with acidic phosphorated compound
before the reactant substrate was added, and the details
about such a phenomenon are not clear at this stage.
There existed a suitable reaction temperature, i.e. ca.
80°C for achieving a maximum selectivity for o-capro-
lactam (entry 7), although the conversion of cyclohex-
anone oxime and the corresponding number of the
catalyst turnover would greatly increase with increasing
temperature (entries 6, 7 and 8). The performance of
different phosphorated compounds on the Beckmann
rearrangement of cyclohexanone oxime were also inves-
tigated (entries 3, 9 and 10). Phosphorous pentachloride
showed the highest catalytic activity and selectivity for
o-caprolactam, while phosphorous pentoxide showed a
very low activity and selectivity, though it was reported
that phosphorous pentoxide exhibited excellent activity
and selectivity in P2O5–DMF system for the liquid-
phase Beckmann rearrangement.6 Finally, the experi-
mental results indicated that increasing the amounts of
phosphorous pentachloride favored both the conver-
sion and the selectivity (entries 2, 11 and 12), although
the catalyst turnover number was decreased with the
increase of the amounts of phosphorous pentachloride.
to the corresponding N-methylacetamide with 89.1%
conversion and 66.8% selectivity, and the by-product
was acetone. As for cyclopentanone oxime, entry 14,
the conversion was low in comparison with the cyclo-
hexanone or acetone oximes and corresponding lactam
was not observed. But it is worth noting that Beckmann
fragmentation of cyclopentanone oxime was observed
unexpectedly,
and
cyclopentanone
54.9%,
5-
chlorovaleronitrile 40.4% and pentanenitrile 4.7% were
obtained as the products. It is possible that the ring
strain provides the driving force for fragmentation.12,13
The experimental results showed that the Beckmann
rearrangements of cyclohexanone and acetone oximes
in the catalytic media consisting of room temperature
ionic liquid based on 1,3-dialkylimidazolium or
alkylpyridinium salts and phosphorous pentachloride
could proceed with high conversion and selectivity,
while Beckmann fragmentation of cyclopentanone ox-
ime was observed in the same catalyst system. At this
stage, the catalyst turnovers are still not high enough,
but such catalyst systems and reaction conditions used
in this work have not been optimized yet. It is possible
that the ionic liquid could be recyclable since it can
easily be separated from the water containing used
phosphorated compound with a suitable organic sol-
vent after the products were extracted as mentioned
above. To our knowledge, it is the first time that
Beckmann rearrangement of ketoximes was performed
in the catalytic media consisting of room temperature
ionic liquid and phosphorated compounds under mild
conditions and without any additional organic solvents.
References
1. Izumi, Y.; Sato, S.; Urabe, K. Chem. Lett. 1983, 1649.
2. Ko, Y.; Kim, M. H.; Kim, S. J.; Seo, G.; Kim, M.-Y.;
Uh, Y. S. Chem. Commun. 2000, 829.
3. Dai, L.-X.; Koyama, K.; Miyamoto, M.; Tatsumi, T.
Appl. Catal. A: Gen. 1999, 189, 237–242.
4. Ushikubo, T.; Wada, K. J. Catal. 1994, 148, 138.
5. Nguyen, M. T.; Raspoet, G.; Vanquickenborne, L. G. J.
Am. Chem. Soc. 1997, 119, 2552.
6. Sato, H.; Yoshioka, H.; Izumi, Y. J. Mol. Catal. A:
Chem. 1999, 149, 25.
7. Izumi, Y. Chem. Lett. 1990, 2171.
8. Welton, T. Chem. Rev. 1999, 99, 2071 and references
cited therein.
9. Blanchard, L. A.; Hanch, D.; Beckman, E. J.; Brennecke,
J. F. Nature 1999, 399, 28.
.
10. BonhOte, P.; Dias, A. P.; Papageorgiou, N.; Kalyanasun-
In order to investigate the scope and limitation of such
catalytic media consisting of ionic liquid and phosphor-
ated compound as one of the promising catalysts for
the Beckmann rearrangement, other ketoximes, i.e. ace-
tone and cyclopentanone oximes as the substrates were
also tested in the BMImBF4/PCl5 catalytic medium
(Table 2). Acetone oxime, entry 13, has been isomerizd
daram, K.; Gra¨tzel, M. Inorg. Chem. 1996, 35, 1168.
11. Suarez, P. A. Z.; Dullius, J. E. L.; Einloft, S.; de Souza,
R. F.; Dupont, J. Polyhedron 1996, 15, 1217.
12. Hunadi, R. J.; Helmkamp, G. K. J. Org. Chem. 1981, 46,
2880.
13. Frater, G.; Mueller, U.; Guenther, W. Tetrahedron Lett.
1984, 25, 1133.
.
.