Table 2 Hydroformylation of 1-octene in presence of different
1-alkyl-3-methylimidazolium bromides
both additives, the rhodium leaching in the organic phase was less
than 0.5 ppm.
a
b
21
In summary, we have demonstrated that 1-octyl-3-methylimi-
dazolium bromide can be used successfully as an additive for the
aqueous-biphasic hydroformylation of higher alkenes. The reac-
tion proceeds at commercially competitive rates without greatly
impairing the catalyst retention and with rapid phase separation.
We thank Sasol and EaStCHEM for a studentship (S. D.).
Entry
Additive
TOF
0
/h
Phase separation
c
1
2
a
HexMimBr
DecMimBr
47.2
1239.3
,10 min
Stable emulsion
Reaction conditions: T 5 100 uC; P 5 20 bar (CO : H
2
5 1 : 1);
O:
cm ; alkene: 2 cm . HexMimBr 5 1-hexyl-3-methylimidazolium
3
8
h; [Rh]H2O 5 1.25 mM; P/Rh 5 50; [additive] 5 0.5 M; H
2
3
3 b
c
bromide; DecMimBr 5 1-decyl-3-methylimidazolium bromide. The
phases were fully separated by the time the autoclave was opened,
ca. 10 min after the reaction was stopped.
Notes and references
{
A stock solution was prepared as follows: weighed amounts of
Rh(acac)(CO) ] and TPPTS were dissolved in a known volume of water
containing 1-alkyl-3-methylimidazolium bromide at the desired concentra-
tion. CO–H was then bubbled through the resulting yellow solution at
0 uC for 1 h. The solution obtained was stored under CO–H
Typical procedure for reaction in an open reactor: an autoclave, fitted
[
2
dissociation necessary for the formation of the 4-coordinated
active rhodium species, [RhH(CO)(TPPTS) ], giving a drop in rate.
2
2
6
2
.
The initial insensitivity of the reaction rate toward the excess ligand
suggests that the reaction rate is not limited by simple kinetics.
The rhodium leaching into the organic phase has been
determined using ICP-MS for the reactions carried out in presence
of OctMimBr. With a 10-fold excess of ligand, the leaching of the
metal was high, 126.2 ppm, but with a 50-fold excess the leaching
dropped to 0.49 ppm. The phase separation at the end of the
reaction was clean and fast, especially when a P/Rh ratio of 50 was
used. Strangely, the linear to branched ratio was not greatly
affected by the excess ligand, 2.84 with a 10-fold excess of
phosphine and 3.19 with a 50-fold excess.
with a mechanical stirrer, thermocouple pocket, pressure transducer and
attached to a ballast vessel via a catalyst injector and mass flow controller,
2
was degassed by pressurising three times with CO–H and releasing the
3
25
pressure. The stock catalyst solution (8 cm , 1.25 6 10 mol Rh and the
appropriate amount of TPPTS in water) was transferred into the autoclave
2
and degassed by pressurising three times with CO–H and releasing the
2
pressure. The autoclave was pressurised with CO–H (1 : 1, 15 bar) and
heated to 100 uC (stirrer speed 5 1000 rpm). Meanwhile, the substrate
injector was charged with 1-octene (2 cm ). Once the reactor had reached
3
1
00 uC, the substrate was injected using an overpressure of CO–H
2
and the
pressure was brought to 20 bar. CO–H was then fed from the ballast vessel
2
so as to maintain the pressure in the autoclave at 20 bar and the pressure in
the ballast vessel was monitored electronically. At the end of the reaction,
the autoclave was cooled and depressurised and the contents analysed using
GC for the organic products. The results presented are the average of at
least 3 runs under each set of conditions.
There are two possible explanations for the increased rate when
adding OctMimBr to these aqueous-biphasic systems. The additive
could have surface active properties and form micelles in which the
substrate is solubilised. In this case, the positively charged
imidazolium moiety could attract the negatively charged ligand,
creating a high concentration of catalyst around the substrate.
Alternatively, the OctMimBr could act as a phase transfer agent,
exchanging the imidazolium with the sodium cations of the TPPTS
making the catalyst soluble in the organic phase. The reverse
exchange would bring the catalyst back into the aqueous phase.
To distinguish between these two mechanisms, we determined
the critical micelle concentration of OctMimBr in water from
conductivity measurements. We found that, at room temperature,
OctMimBr starts to aggregate to form micelles above 2.16 6
{ Typical procedure for a closed reactor: an autoclave, fitted with a
mechanical stirrer, thermocouple pocket, pressure transducer, gas inlet and
2
injection port, was degassed by alternate vacuum–N . The stock catalyst
3
3
25
solution (8 cm , 1.25 6 10 mol Rh) and the substrate (2 cm ) were
transferred into the autoclave. The autoclave was purged three times with
2 2
pressurised CO–H . The autoclave was pressurised with CO–H (1 : 1,
20 bar) and heated, with stirring, to 100 uC. At the end of the reaction, the
autoclave was cooled, depressurised and the contents analysed using GC
for the organic products. The results presented are the average of at least
3 runs under each set of conditions.
1
For different methods of catalyst recycling, see: (a) Catalyst Separation,
Recovery and Recycling: Chemistry and Process Design, ed. D. J. Cole-
Hamilton and R. P. Tooze, Springer, Dordrecht, 2005; (b) D. J. Cole-
Hamilton, Science, 2003, 299, 1702.
22
10
23
mol dm . This corresponds approximately to the concen-
tration of OctMimBr at which we start to observe a rate
improvement. We then conducted the hydroformylation of
2 For a review on multiphase homogeneous catalysis, see: Multiphase
Homogeneous Catalysis, ed. B. Cornils, W. A. Herrmann, I. T. Horvath,
W. Leitner, S. Mecking, H. Olivier-Bourbigou, and D. Vogt, Wiley-
VCH, Weinheim, 2005.
1-octene in the presence of imidazolium salts with alkyl chains
of different lengths (Table 2).
3
P. Purwanto and H. Delmas, Catal. Today, 1995, 24, 135.
It appears that the shorter chain alkylimidazolium bromide,
HexMimBr, has only a limited effect on the reaction. On the other
hand, the longer chain additive, DecMimBr, greatly increases the
rate of the reaction but leads to stable emulsions. Therefore, we
propose that the rate improvement observed with OctMimBr is
due to its surfactant properties. Nevertheless, the emulsions
formed under the reaction conditions are not stable and the
phases are fully separated by the time the autoclave is opened, ca.
4 (a) E. Monflier, H. Bricout, F. Hapiot, S. Tilloy, A. Aghmiz and
A. M. Masdeu-Bulto, Adv. Synth. Catal., 2004, 346, 425; (b) T. Mathivet,
C. M e´ liet, Y. Castanet, A. Mortreux, L. Caron, S. Tilloy and E. Monflier,
J. Mol. Catal. A: Chem., 2001, 176, 105; (c) E. Karakhanov, T. Buchneva,
A. Maximov and M. Zavertyaeva, J. Mol. Catal. A: Chem., 2002, 184,
11.
5
(a) H. Chen, Y. Z. Li, J. R. Chen, P. M. Cheng, Y. E. He and X. J. Li,
J. Mol. Catal. A: Chem., 1999, 149, 1; (b) L. B. Wang, H. Chen, Y. E. He,
Y. Z. Li, M. Li and X. J. Li, Appl. Catal., A, 2003, 242, 85.
6 C. Yang, X. Y. Bi and Z. S. Mao, J. Mol. Catal. A: Chem., 2002, 187, 35.
7
(a) G. Papadogianakis, in Aqueous Phase Organometallic Catalysis, ed.
B. Cornils and W. A. Herrmann, Wiley-VCH, Weinheim, 2nd edn, 2004,
pp. 158–173; (b) Z. Jin, Y. Wang and X. Zheng, in Aqueous Phase
Organometallic Catalysis, ed. B. Cornils and W. A. Herrmann, Wiley-
VCH, Weinheim, 2nd edn, 2004, pp. 301–312.
1
0 min after the reaction was stopped. If a genuine phase-transfer
mechanism were operating, significant rate improvement would
have been expected when HexMimBr was used as the additive.
Moreover, it would significantly promote rhodium leaching. With
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 1933–1935 | 1935