Continuous, selective hydroformylation in supercritical carbon dioxide using an immobilised homogeneous catalyst

Article abstract of DOI:10.1039/b002526g

A continuous process for the selective hydroformylation of higher olefins in supercritical carbon dioxide (scCO₂) is presented; the catalyst shows high selectivity and activity over several hours and no decrease in performance was observed over several days.

Full text of DOI:10.1039/b002526g

Continuous, selective hydroformylation in supercritical carbon dioxide using
an immobilised homogeneous catalyst
Nicola J. Meehan,a Albertus J. Sandee,b Joost N. H. Reek,b Paul C. J. Kamer,b Piet W. N. M. van Leeuwen*b
and Martyn Poliakoff*a
a School of Chemistry, University of Nottingham, Nottingham, UK NG7 2RD.
E-mail: Martyn.Poliakoff@nottingham.ac.uk
b Institute of Molecular Chemistry, University of Amsterdam, Nieuwe Achtergracht 166, 1018 WV Amsterdam, The
Netherlands. E-mail: pwnm@anorg.chem.uva.nl
Received (in Liverpool, UK) 28th March 2000, Accepted 29th June 2000
Published on the Web 19th July 2000
A continuous process for the selective hydroformylation of
higher olefins in supercritical carbon dioxide (scCO2) is
presented; the catalyst shows high selectivity and activity
over several hours and no decrease in performance was
observed over several days.
insoluble catalysts14 have demonstrated significant activity in
scCO2 systems. The use of an immobilised homogeneous
catalyst overcomes both solubility and catalyst recovery
problems. At Nottingham, continuous processing in scCO2 has
been successfully applied to a wide range of hydrogen-
ations,15,16 Friedel–Crafts alkylations17 and etherification reac-
tions18 using heterogeneous catalysts supported on polysiloxane
(Deloxan®, Degussa AG). Here, we show how this technique is
effectively applied in the hydroformylation reaction using an
immobilised rhodium–diphosphine catalyst containing a large
P–Rh–P bite angle.
Hydroformylation is one of the mildest and most efficient
methods of producing aldehydes and it is therefore widely
applied in the petrochemical industry. The cleanest, industrially
important hydroformylation process is the aqueous biphasic
process, developed by Ruhrchemie/Rhône-Poulenc, affording a
straightforward separation of the organic products from the
catalyst.1 The applicability of this system, however, is strictly
limited to substrates which are slightly water soluble, such as
propene and but-1-ene. In the industrial hydroformylation of
higher olefins, the catalyst is separated by either extraction,
catalyst destruction or distillation. One of the major challenges
in this field is the development of a continuous hydro-
formylation process combining a high catalytic activity and
selectivity with facile product separation, in which catalyst
leaching does not occur.2 To date, no such system has been
reported; often at higher CO pressures the ligand on the metal is
easily exchanged for CO molecules resulting in rhodium
leaching.3 Recently Van Leeuwen and coworkers reported a
promising approach based on rhodium–diphosphine hydro-
formylation catalysts with a large P–Rh–P bite angle in a
multiphase batch process.4,5
The catalyst used is the rhodium complex of N-(3-tri-
methoxysilyl-n-propyl)-4,5-bis(diphenylphosphino)phenox-
azine A, immobilised on silica (particle size 200–500 mm).†
Typically, 1 g of ligand-modified silica was reacted with 4 mg
of Rh(acac)(CO)2 and then loaded into a 5 mL supercritical flow
reactor.‡
The catalyst performed well in the hydroformylation of oct-
1-ene with selective production of linear nonanal. The average
linear to branched aldehyde ratio (l b) was 40 1. Oct-1-ene
conversions of up to 14% were obtained (Table 1, entries 5 and
7) and only a few percent of octene isomers and a trace amount
(ca. 1%) of alcohol were observed as byproducts.
The use of scCO2 is becoming increasingly important as a
reaction medium in metal catalysed reactions.6–8 The absence of
a gas–liquid phase boundary and the ability of scCO2 to support
high concentrations of dissolved gases combined with facile
product and catalyst separation makes scCO2 a competitive
alternative to conventional solvents. However, homogeneous
catalysts often require modification in order to increase their
solubility in scCO2,9–12 although unmodified13 and even
Table 1 Results from the hydroformylation of oct-1-ene, values shown are average numbers over a period of 3–6 ha
Linear
aldehydec
(%)
Branched
aldehydec
(%)
Alkene
isomersc
(%)
Linear
alcoholc
(%)
Linear to
branched
ratio
Oct-1-ene
conversion
(%)
Entry
TOFb
1d
2
39
87
112
117
160
44
96.1
92.9
94.4
92.6
93.5
90.7
96.0
91.3
2.4
3.0
2.4
3.8
2.8
4.4
1.9
4.1
1.5
3.8
2.5
2.5
2.9
3.7
1.1
4.3
0
40
32
40
24
33
21
50
23
3.6
9.4
10.1
10.3
14.3
4.1
0.3
0.7
1.0
0.8
1.3
0.9
0.3
3e
4f
5g
6h
7i
8j
93
96
14.3
4.6
a Ligand Rh ratio is 10 1 and the catalysis was performed at 80 °C, 120 bar CO2 at 0.65 L min21 flow rate (at 20 °C, 1 atm), 50 bar overpressure syngas
and an oct-1-ene flow rate of 0.05 mL min21 (substrate syngas = 1 10) unless otherwise stated. b Average turnover frequencies were calculated as mol
aldehyde (mol catalyst)21 h21 c Determined by means of GC analysis using decane as an internal standard. d Reaction temperature is 70 °C. e Syngas
.
overpressure is 25 bar (substrate syngas = 1 5). f 0.3 L min21 CO2 flow rate (at 20 °C, 1 atm). g Reaction temperature is 90 °C. h 180 bar CO2. i Oct-1-ene
flow rate of 0.03 mL min21 j Oct-1-ene flow rate of 0.1 mL min21
.
.
DOI: 10.1039/b002526g
Chem. Commun., 2000, 1497–1498
This journal is © The Royal Society of Chemistry 2000
1497

View Article Online
provides a direct and quantitative separation of the products
from the catalyst and avoids any solubility limitations of
homogeneous catalysts. Finally, the robustness of the catalyst
and absence of Rh leaching, makes this system an interesting
candidate for sustainable processes.
We are grateful to Dr W. K. Gray for initiating this
collaboration and thank Dr S. K. Ross for his help. We thank
Thomas Swan & Co. Ltd. and the EPSRC (GR/M73644) for
funding the work at Nottingham and the Innovation Oriented
Research Program (IOP-katalyse) for financing the work in
Amsterdam. We thank J. Elgersma for performing the Rh
analyses and Dr P. A. Arnold, Dr M. Glenny, M. Guyler and K.
Stanley for their assistance.
Fig. 1 Turnover number (TON) for the hydroformylation of oct-1-ene in
scCO2 at 70 and 80 °C.
Notes and references
† Catalyst preparation: 100 mg (1.40 3 1024 mol) of N-(3-trimethoxysilyl-
n-propyl)-4,5-bis(diphenylphosphino)phenoxazine (Siloxantphos) was
added to a suspension of 1 g silica (200–500 mm) (predried at 140 °C for
several days) in 25 ml toluene and the mixture was mechanically stirred at
80 °C for 20 h. After cooling to room temperature, the liquids were removed
from the residue and the silica was washed with three portions of toluene.
The ligand-on-silica was dried in vacuo and was then suspended in a
mixture of 5 mL THF and 1 mL Et3N. The suspension was mixed for 10 min
and 4 mg (1.55 3 1025 mol) Rh(acac)(CO)2 was then added. The mixture
was mechanically stirred for 15 min, after which the THF was removed and
the catalyst was further washed with three portions of THF. The catalyst was
dried in vacuo and was either used directly or stored under argon at 220
°C.
The rate of hydroformylation is moderate (39 h21) at 70 °C
with an oct-1-ene flow rate of 0.05 mL min21 (Table 1, entry 1).
The rate increased to 87 h21 with the catalyst bed at 80 °C
(Table 1, entry 2) and improved further to 112 h21 on
decreasing the syngas pressure from 50 to 25 bar (Table 1, entry
3). This improvement in rate is consistent with the negative
order in CO pressure that is commonly observed in hydro-
formylation reactions.19 A TOF of 117 h21 was observed on
increasing the residence time of the substrate in the reactor by
decreasing the CO2 flow rate (Table 1, entry 4). An alternative
method of increasing the residence time is to increase the CO2
pressure, however this resulted in a decrease in TOF (Table 1,
entry 6). This may be explained as a higher pressure results in
a higher density of CO2 which will alter the transport properties
in the reactor. The highest TOF (160 h21) was observed at 90 °C
(Table 1, entry 5). Varying the oct-1-ene flow rate was found to
have an effect on the l b ratio but not on the TOF (Table 1,
entries 2, 7 and 8). As the CO concentration in scCO2 is
relatively high, it is remarkable that the hydroformylation rate is
over three times faster than the batch reaction in toluene (TOF
= 35 h21) and only half the rate of the homogeneous analogue
(TOF = 283 h21) under comparable reaction conditions (80 °C,
50 bar syngas).5 The high rate in scCO2 is probably caused by
enhanced mass-transport properties and the lower viscosity of
the solvent medium.20
‡ The substrate + internal standard, supercritical CO2 (Pc = 73.8 bar, Tc
=
31.1 °C) and CO/H2 are brought together in a heated mixer, passed through
the reactor containing catalyst, and then expanded to separate the fluid
product from the process-stream. The reactor is assembled from commer-
cially available units: scCO2 pump PM101, CO/H2 compressor CU105 and
Expansion Module PE103 (all from NWA GmbH, Lörrach, Germany), a
high pressure mixer (Medimix) and a Gilson 305 pump (for the organic
substrate). CAUTION: Flow reactors have a comparatively small volume
under pressure. Nevertheless, equipment with the appropriate pressure and
temperature rating should always be used for high pressure experiments.
1 E. G. Kuntz, Chemtech, 1987, 570.
2 W. A. Herrmann and B. Cornils, Angew. Chem., Int. Ed. Engl., 1997, 36,
1048.
3 M. Beller, B. Cornils, C. D. Frohning and C. W. Kohlpainter, J. Mol.
Catal. A, 1995, 104, 17.
4 A. J. Sandee, V. F. S. Slagt, J. N. H. Reek, P. C. J. Kamer and
P. W. N. M. van Leeuwen, Chem. Commun., 1999, 1633.
5 A. J. Sandee, L. A. van der Veen, J. N. H. Reek, P. C. J. Kamer, M. Lutz,
A. L. Spek and P. W. N. M. van Leeuwen, Angew. Chem., Int. Ed., 1999,
38, 3231.
It was also found that the expansion system in our apparatus
facilitated the separation of the aldehyde product from the
residual oct-1-ene. In our preliminary experiments, we were
able to remove ca. 90% of oct-1-ene from the product by simply
controlling the two-step depressurization of CO2.
The catalyst appears to be very robust, as its performance is
constant over at least 30 h. Fig. 1, shows a plot of the turnover
number (TON) vs. reaction time. The TON increased linearly
with time at both 70 and 80 °C. Moreover, we were able to
continue using the catalyst for six non-consecutive days with no
observable decrease in either activity or selectivity. Fur-
thermore, no rhodium leaching was detected (detection limit of
used technique (AES) is 0.2% of the total amount of rhodium of
the catalyst). This demonstrates that the rhodium–diphosphine
bond in this catalyst remains stable under hydroformylation
conditions.
In conclusion, we have presented the first example of
continuous selective hydroformylation of higher olefins in
scCO2 using an immobilised homogeneous rhodium catalyst.
The process is potentially interesting in the manufacture of fine
chemicals and our approach has several advantages compared to
conventional homogeneously catalysed reactions. Firstly,
scCO2 is a clean, environmentally benign medium which can be
easily separated from the organic phase. Secondly, the applica-
tion of an immobilised homogeneous catalyst in the flow reactor
6 R. Noyori, Chem. Rev., 1999, 99, 475.
7 J. A. Darr and M. Poliakoff, Chem. Rev., 1999, 99, 495.
8 B. M. Bhanage, Y. Ikushima and M. Shirai, Chem. Commun., 1999,
1277.
9 S. Kainz and W. Leitner, Catal. Lett., 1998, 55, 223.
10 D. Koch and W. Leitner, J. Am. Chem. Soc., 1998, 120, 13 398.
11 D. R. Palo and C. Erkey, Ind. Eng. Chem. Res., 1998, 37, 4203.
12 D. R. Palo and C. Erkey, Organometallics, 2000, 19, 81.
13 I. Bach and D. J. Cole-Hamilton, Chem. Commun., 1998, 1463.
14 M. F. Sellin and David J. Cole-Hamilton, J. Chem. Soc., Dalton Trans.,
2000, 1681.
15 M. G. Hitzler and M. Poliakoff, Chem. Commun., 1997, 1667.
16 M. G. Hitzler, F. R. Smail, S. K. Ross and M. Poliakoff, Org. Proc. Res.
Dev., 1998, 2, 137.
17 M. G. Hitzler, F. R. Smail, S. K. Ross and M. Poliakoff, Chem.
Commun., 1998, 359.
18 W. K. Gray, F. R. Smail, M. G. Hitzler, S. K. Ross and M. Poliakoff,
J. Am. Chem. Soc., 1999, 121, 10 711.
19 J. A. Moulijn, P. W. N. M. van Leeuwen and R. A. van Santen,
Catalysis: an integated approach to homogeneous, heterogeneous and
industrial catalysis, Elsevier, Amsterdam, 1993.
20 A. Baiker, Chem. Rev., 1999, 99, 453.
1498
Chem. Commun., 2000, 1497–1498