from water and carbon dioxide, which catalyze the dehydration
of 2a to 3a. Thus the product distribution as the 2a/3a ratio can
selectivity to aldol 2a. In supercritical conditions, above 8 MPa,
14,15
water is more soluble than gaseous CO
2
,
and the solvated
be varied by simply controlling the pressure of CO
2
.
water reacts with CO likely to generate carbonic acid which
2
Fig. 1 shows the CO pressure dependence of the total
2
could promote the dehydration of 2a as a Brønsted acid. Thus,
under supercritical conditions, the selectivity to 3a would be
higher than that under subcritical conditions in the lower
pressure region. We also investigated another aldol reaction of
acetaldehyde under the same conditions in the presence of
MgO, and a similar pressure dependence on the product
selectivity was observed.§
In conclusion, we found that the selectivity of the self-aldol
condensation of propionaldehyde can be tuned simply by
2
adjusting the pressure of CO in the presence of MgO catalyst
with a small amount of water. At the supercritical region the
unsaturated aldehyde 3a was obtained in 94% high selectivity
around the critical pressure of 12 MPa, while the aldol
derivative 2a was attained in a satisfactory selectivity of 85% at
the subcritical region. Furthermore, acetaldehyde can be also
converted into aldol or unsaturated aldehyde in good selectivity
conversion of 1a in MgO in the presence of a small amount of
water. One can see an interesting pressure dependence, in which
the conversion decreases, reaching a minimum at 8 MPa, and
then increases to the maximum at the near-critical pressure
around 14 MPa. Taking into account the error rate in Fig. 1, this
pressure dependence showing extrema in the conversion is
considered to be convincing. Such a pressure dependence was
2
previously reported when CO was used as a solvent or a
reactant.2 The pressure dependence of the selectivity was
further investigated. Interestingly the selectivities toward 2a
and 3a which vary in opposite manners with pressure can be
seen in Fig. 2. In the lower pressure range below about 5 MPa,
an increase in pressure accelerated only the formation of 2a and
led to the maximum selectivity to 2a of 85% around 5 MPa,
when the dehydration for the formation of 3a from 2a was
depressed. However, further increase in pressure above 5 MPa
resulted in higher selectivity to 3a, attaining the maximum value
of 94% at the critical pressure of 12 MPa. This pressure
dependence would be due to the changing property on the
,3
2
by pressure manipulation of CO .
Notes and references
surface of the MgO catalyst caused by adsorption of CO
water at each pressure. At lower pressures below 8 MPa, CO
adsorbed on the surface of MgO catalyst, and carbonic acid is
formed by the reaction of CO with added water which is also
2
and
‡
MgO (JRC-MGO-4 100A) used in this study was purchased from The
2
is
Catalysis Society of Japan, and its physical properties are as follows:
2
21
particle size: ca. 14 nm, BET surface area; 120 m g . MgO was heated at
00 °C under air for 1 h before the reaction.
2
1
adsorbed on the surface of MgO, and the Brønsted acid derived
from generated carbonic acid is considered to be neutralized
with the basic sites of MgO, resulting in the increase in
§ 3-Hydroxybutyraldehyde and but-2-enal were also obtained from the
aldol reaction of acetaldehyde. At 5 MPa, 3-hydroxybutyraldehyde was
obtained in a maximum selectivity of 73%, whereas but-2-enal was obtained
in 96% selectivity at 12 MPa. For details see ESI.†
1
2
J. Hyde, W. Leitner and M. Poliakoff, in High Pressure Chemistry, ed.
R. Eldik and F.-G. Klärner, Wiley-VCH, Weinheim, 2002, p. 369.
S. Fujita, B. M. Banage, Y. Ikushima and M. Arai, Green Chem., 2001,
3
, 87.
3
4
5
6
H. Kawanami and Y. Ikushima, Chem. Commun., 2000, 2089.
H. Kawanami and Y. Ikushima, Tetrahedron Lett., 2002, 43, 3841.
H. Kawanami and Y. Ikushima, J. Jpn. Petrol. Inst., 2002, 45, 321.
H. Kawanami, A. Sasaki, K. Matsui and Y. Ikushima, Chem. Commun.,
2
003, 896.
7
Y. Ikushima and M. Arai, in Chemical Synthesis Using Supercritical
Fluids, ed. P. G. Jessop and W. Leitner, Wiley-VCH, Weinheim,
1
999.
8
9
W. Carruthers, Some Modern Methods of Organic Synthesis – 3rd edn.,
Cambridge Univ. Press., New York, 1986.
K. Tanabe, G. Zhang and H. Hattori, Appl. Catal., 1989, 48, 63.
1
1
0 H. Tsuji, F. Yagi, H. Hattori and H. Kita, J. Catal., 1994, 148, 759.
1 E. Dumitriu, V. Hulea, I. Fechete, A. Auroux, F.-F. Lacaze and C.
Guimon, Microporous Mesoporous Mater., 2001, 43, 341.
Fig. 1 Pressure dependence of the total conversion of 1a in the presence of
MgO and water.
1
2 (a) Y. Anzai, M. Goto, A. Kodama and T. Hirose, Proc. 14th Symp. On
Chem. Eng., Kyushu-Taejon/Chungnam, Taejon Univ., Korea, Dec. 1st
2
001, p. 13; (b) M. Goto, Y. Anzai, A. Komada and M. Yoshida, Chem.
Eng. Trans., 2002, 2, 85.
1
3 The typical experiment procedure is as follows: acetaldehyde (10 mmol)
or propionaldehyde (10 mmol), and MgO (50 mg) were charged into a
2
5 cm3 stainless steel reactor at room temperature. In the case of
addition of water, water (0.1 mL) was mixed with aldehydes and MgO.
The reactor was heated at 80 °C and CO was subsequently charged into
2
the reactor using a high-pressure liquid pump and compressed to the
desired pressure. Pressure control was achieved by a back-pressure
regulator. The reactions were started by stirring the mixture, continued
for 6 h. After reaction, the reactor was cooled to 0 °C with ice and the
pressure was released slowly. The MgO was removed by filtration, and
the products and yields were determined by GC-MS (HP 6890 GC
System and 5973 Mass Selective Detector) with tridecane as an internal
reference.
Fig. 2 Pressure dependence of the selectivity of 2a and 3a in the presence
of MgO and water.
14 R. Weibe and V. L. Gaddy, J. Am. Chem. Soc., 1941, 63, 475.
15 R. Wiebe, Chem. Rev., 1941, 29, 475.
CHEM. COMMUN., 2003, 2502–2503
2503