Triphasic liquid systems for improved separations. Trioctylmethylammonium chloride-immobilised rhodium trichloride: A phosphine-free hydroformylation catalytic system

Article abstract of DOI:10.1002/adsc.200700231

A liquid triphasic system made of isooctane, water, and trioctylmethylammonium chloride allows one to carry out the hydroformylation of model olefins using neat RhCl₃ as catalyst precursor. By using the triphasic system, the catalyst is kept separate from the reagents and products. This allows one to simply remove the product and to recycle the catalyst numerous times. No leaching of rhodium into the organic phase is observed.

Full text of DOI:10.1002/adsc.200700231

COMMUNICATIONS
DOI: 10.1002/adsc.200700231
Triphasic Liquid Systems for Improved Separations. Trioctyl-
methylammonium Chloride-Immobilised Rhodium Trichloride:
A Phosphine-Free Hydroformylation Catalytic System
Stefano Paganelli,a Alvise Perosa,b,* and Maurizio Selvab
a
Dipartimento di Chimica dell’Università Caꢀ Foscari, Dorsoduro 2137, 30123 Venezia, Italy
Dipartimento di Scienze Ambientali dell’Università Caꢀ Foscari, Dorsoduro 2137, 30123 Venezia, Italy
b
Fax: (+39)-041-234-8584; e-mail: alvise@unive.it
Received: May 8, 2007
Abstract: A liquid triphasic system made of isooc-
tane, water, and trioctylmethylammonium chloride
allows one to carry out the hydroformylation of
model olefins using neat RhCl3 as catalyst precur-
sor. By using the triphasic system, the catalyst is
kept separate from the reagents and products. This
allows one to simply remove the product and to re-
cycle the catalyst numerous times. No leaching of
rhodium into the organic phase is observed.
supported ionic liquid catalysis (SILP) have also been
described,[7] along the lines of the supported aqueous
phase catalysis (SAPC) developed earlier by Davis.[8]
Leitner and co-workers have instead addressed the
separation issue through the design of a CO2-regulat-
ed reaction/separation systems where the substitution
at the metal catalyst makes it soluble during the hy-
droformylation reaction, and insoluble in the separa-
tion stage.[9] All the above cases are inventive and ef-
ficient, but the presence of a phosphine ligand as cat-
alyst activator is always necessary, and labourious syn-
thetic steps are often required.
Keywords: hydroformylation; ionic liquids; multi-
phase catalysis; rhodium; water
A different class of promising multiphasic systems is
that made by more than two mutually insoluble liquid
phases (e.g., organic/aqueous/ionic). These represent a
valuable tool for in situ separation of products, by-
Hydroformylation is one of the largest volume pro- products, and catalysts; and for the simple recovery
cesses in the chemical industry that uses homogene- and recycle of the latter. In addition, often the catalyst
ous metal catalysis. Efficient catalyst separation re- efficiency can be boosted by choosing the appropriate
mains, however, one of the key issues for the clean ionic liquid phase to immobilise the catalyst. A
isolation of products and for catalyst recycling. The number of such multiphasic systems has been reviewed
archetype example of the sought-after separation of recently, homogeneous,[2] as well as heterogeneous.[10]
the catalytic system is still the Ruhrchemie/Rhône–
Our experience dates back to the use of liquid
Poulenc process for the aqueous hydroformylation of onium salt-modified heterogeneous catalysts.[11] More
propene. In this case the Rh catalyst is made water- recently, we have described a triphasic system made
soluble and can be easily separated from the reaction of trioctylmethylammonium chloride/isooctane/water
products.[1]
that promoted the formation, immobilisation, and sta-
The techniques used to keep hydroformylation bilisation of recyclable palladium nanoparticles start-
products and catalysts apart often involve partitioning ing from appropriate Pd complexes. These were cata-
of the two between immiscible liquid phases.[2] Final lytically active for selective hydrodehalogenation and
separation is then obtained by decanting. For exam- Heck coupling reactions, and allowed easy product-
ple, methylated cylodextrins can be employed in the catalyst separation thanks to triphasic separation, as
hydroformylation of olefins in an aqueous-organic well as straightforward catalyst recycle.[12] A biphasic
system.[3] Other examples include catalysis using bi- system for catalytic hydroformylation was also report-
phasic fluorous solvent systems,[4] and the use of su- ed, based on the use of water-soluble protein-rhodium
percritical fluids.[5] Ionic liquids have attracted atten- complexes.[13] In this instance, the catalytic system al-
tion in this field as well, due to their combined ability lowed facile recovery of the Rh-HSA (human serum
to form biphasic systems with organic species, and to albumin) complex with high chemo- and regioselectiv-
retain catalytic metal complexes.[6]
ity in the hydroformylation reaction.
More recently, hydroformylation techniques based
In the present paper, we turn our attention to hy-
on the heterogenisation of the catalytic species using droformylation in a triphasic organic/aqueous/ionic
1858
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 1858 – 1862

Trioctylmethylammonium Chloride-Immobilised Rhodium Trichloride
COMMUNICATIONS
Table 1. Hydroformylation of styrene catalysed by RhCl3 in
system where the catalyst is neat RhCl3 immobilised
in trioctylmethylammonium chloride
water/TOMAC/isooctane triphasic system.[a]
a
[(C8H17)3N+CH3ClÀ =TOMAC] phase. The driving
force of this study is the improved separation and re-
cycle of the catalyst, as well as simplification of the
standard operating procedures, thanks (i) to the use
of one of the most widely available Rh precursors “as
such”, that is, without any activation or synthetic step,
and (ii) thanks to the lack of any additional catalyst
ligand/activator, such as phosphines. Rhodium tri-
chloride as a hydroformylation catalyst has, in fact,
been described, mostly however in conjunction with
an activator, for example, alkyl- or arylphosphines or
nitrogen-containing ligands.[14]
Run T
t
Conv.
Aldehyde yield Branched/linear
[8C] [h] [%]
[%]
[%]
1[b] 80
2
18
-
-
-
16 100
16 100
16 100
16 100
100
100
100
100
1
98/2
99/1
98/2
98/2
100
3[c]
4[c]
5[c]
6[d]
16
1
7[c,e]
18 100
18 57
18 55
18 54
100
57
55
54
99
96/4
>99
>99
>99
98/2
8
60
9[c]
10[c]
11[f] 80
5
99
RhCl3 was reported as a hydrogenation and isomer-
isation catalyst under phase-transfer catalysis condi-
tions in a biphasic aqueous-organic system, in the
presence of Aliquat 336 (a commercial mixture
[a]
Substrate=1.0 mmol; RhCl3 =0.10 mmol; isooctane/
H2O/TOMAC=4/4/1 mL; p(CO)=p(H2)=45 atm.
TOMAC=trioctylmethylammonium chloride.
Experiment carried out without water. At the end of the
reaction the organic isooctane phase is colourless and
RhCl3 is completely dissolved in TOMAC.
Reaction carried out by using the catalytic phase recov-
ered from the previous experiment.
Reaction carried out by using a styrene to rhodium
molar ratio=100/1.
To the TOMAC with dissolved RhCl3 recovered from ex-
periment 1, was added water (4 mL), isooctane (4 mL)
and styrene (1 mmol).
Experiment catalysed by RhCl3 dissolved in water
(4 mL) with isooctane (without TOMAC): at the end of
the reaction rhodium leaching is complete.
[b]
consisting
predominantly
of
TOMAC
[(C8H17)3N+CH3ClÀ]).[15,16] In this case the suggested
soluble catalyst precursor was the ion pair
[c]
[d]
[e]
[(C8H17)3NCH3]+[RhCl4]À, based on an elemental
G
analysis of the related species [(C7H15)4N]+
[RhCl3Br]À, and on the correlation between the con-
centration of Aliquat 336 and the partition coefficient
of the metal ion between the organic and aqueous
phases [Eq. (1)].[16]
[f]
The difficulty in this case was in the incomplete re-
On the other hand, the same reaction conducted in
covery of the catalyst, that had to be re-extracted re- the presence of a third phase made by water, yielded
peatedly from the organic to the aqueous phase by quantitative conversion of the aldehydes after 16 h,
ion exchange with sodium perchlorate.
with 98% of the branched isomer (run 2), the same
Exploratory tests on the triphasic hydroformylation efficient reaction was observed by adding water to the
catalytic system were run on the model reaction of residual mixture of run 1 and subjecting it to the
styrene with syngas [Eq. (2)] by using 10 mol% same hydroformylation conditions (run 7). In all these
cases the catalyst precursor remained dissolved exclu-
sively in the TOMAC phase – presumably in the form
[(C8H17)3NCH3]+[RhCl4]À – and the TOMAC/Rh/
A
water ensemble could be recycled 3 times with no de-
tectable loss of activity (runs 3–5). A ten-fold reduc-
tion of the amount of RhCl3 resulted in the reaction
being practically inhibited (run 6). Lower tempera-
RhCl3 as the catalyst and 1.0 mL of TOMAC , the re- tures resulted in slower reactions (run 8), albeit with
sults are shown in Table 1.
equally reproducible recycling (runs 9 and 10). In the
The initial experiment was conducted using a bi- presence of a biphasic isooctane/water system, with-
phasic TOMAC/isooctane system, on the lines of re- out TOMAC, the RhCl3-catalysed hydroformylation
ported procedures where the catalyst is kept separate of styrene was fast, but total leaching of rhodium spe-
from reagents and products by the use of an ionic cies to the organic phase was observed, making sepa-
liquid phase.[17] The RhCl3 catalyst was dissolved in ration and recycle impossible (run 11).
TOMAC, the reagent in isooctane, and the biphasic
Therefore, the triphasic system effectively promot-
mixture was reacted under 90 atm of syngas at 808C ed the reaction, not only by allowing recovery and re-
(run 1). No reaction occurred.
cycle of the catalytic system, but also by avoiding rho-
Adv. Synth. Catal. 2007, 349, 1858 – 1862
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1859

Stefano Paganelli et al.
COMMUNICATIONS
dium leaching (see later), and by a favourable effect systems. In that case the effect of water was explained
due to the presence of water. by an extraction mechanism whereby the onium salt
Water, in fact, was indispensable for the reaction to regenerates the active Pd(0) species in the organic
proceed, a fact that has been already observed in phase by a phase-transfer mechanism.[19]
other instances of catalysts that are active in the pres-
When the long-chain aliphatic olefin tetradecene
ence of an ionic environment. For example, Davisꢀs was hydroformylated [Eq. (3)] in the triphasic system
studies on supported aqueous-phase (SAP) hydrofor- under the conditions of Table 1, practically complete
mylation using a sulphonated Rh catalyst, indicated conversion was observed after 21 h, and the TOMAC/
that the water amount greatly influenced catalytic Rh/water system was recycled for 5 times with identi-
performance: a too low or too high water content was cal results (solid bars Figure 1). No selectivity was ob-
detrimental, the optimum amount being intermediate served: 1- and 2-formyl-tetradecane were observed in
(4–12 wt%).[18] At low water loadings the Rh catalyst approximately equimolar amounts (cross-hatched and
was strongly adsorbed to the surface of the support, it diagonal hatched bars, Figure 1), and up to 16% of
was very stable, but lost the mobility necessary for it products deriving from double bond isomerisation fol-
to be active. At high water content (>20 wt%) the lowed by hydroformylation were detected (hatched
SAP materials lost activity and stability, approaching bars, Figure 1).
the behaviour of unsupported catalysts. The positive
In this case the filtered clear organic phase was
effect of water was also reported by Jeffery during his checked for leaching of rhodium by adding a different
studies on the Pd-catalysed Heck reaction in the pres- olefin – styrene (1 mmol) – 100 atm of syngas and
ence of tetraalkylammonium salts in water-organic heating to 808C for 24 h. Complete lack of conversion
indicated no metal leaching into the organic phase.
Norbornene was also hydroformylated quantitative-
ly in the triphasic system under the conditions of
Table 1, with 91% selectivity towards the exo isomer
[Eq. (4)]. Recycle of the catalytic system was again
achieved simply, without any loss of activity. As a
matter of fact conversion after the second run in-
creased up to quantitative, likely due to the complete
transformation of the rhodium trichloride into the
active catalytic species (Figure 2).
Figure 1. Hydroformylation of tetradecene (1 mmol), RhCl3 (0.1 mmol), isooctane/H2O/TOMAC=4/4/1 mL; p(CO)=
p(H2)=45 atm.
1860
asc.wiley-vch.de
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 1858 – 1862

Trioctylmethylammonium Chloride-Immobilised Rhodium Trichloride
COMMUNICATIONS
Figure 2. Hydroformylation of norbornene (1 mmol), RhCl3 (0.1 mmol), isooctane/H2O/TOMAC=4/4/1 mL; p(CO)=
p(H2)=45 atm, t=16 h.
The active Rh species was not fully characterised >90% at 808C), and regioselective: >98% branched
due to the impossibility of extracting it from the ionic aldehyde in the case of styrene, and >90% exo
liquid phase without modifying its structure. Howev- isomer in the case of norbornene.
er, the IR spectrum of the Rh-containing ionic liquid
Based on one other report, RhCl3 likely forms an
phase showed two sharp bands at 2063.7 and ion pair with TOMAC [Eq. (1)], this ion pair may
1988.5 cmÀ1, assigned to a rhodium carbonyl species represent the catalytic precursor which, under the re-
(Figure 3).[20]
action conditions, forms the catalytically active Rh-
In conclusion, the triphasic system here described is carbonyl species observed by IR. The formation of
active for the hydroformylation reaction of model ole- the ion pair might also explain the relatively high con-
fins, and it allows simple recovery and recycle of the centration of Rh (10% mol) needed in this study. The
catalytic system for repeated runs, without detectable large excess of TOMAC used here might mask the
loss of activity, and without the need for phosphine li- active catalyst when this is too diluted.
gands. In addition, the described triphasic system ef-
Noticeably, the presence of the aqueous phase is
fectively prevents leaching of rhodium to the organic necessary for the reaction, a fact that is not clearly
phase. This is of fundamental importance to avoid understood yet, but that might be connected either to
metal contamination of the products. The described increased mobility afforded by water within the
hydroformylation is generally very efficient (yields TOMAC phase to the species involved in the reac-
tion, or by a phase-transfer mechanism.
Experimental Section
Synthesis of Methyltri-n-octlyammonium Chloride
(TOMAC)[21]
A 250 mL, round-bottomed flask equipped with a magnetic
stirring bar was charged with 35.6 g (0.101 mol) of tri-n-oc-
tylamine and 40 mL of toluene. The mixture was placed
under a nitrogen atmosphere and 13.0 g (0.106 mol) of di-
methyl sulphate were added at room temperature. The mix-
ture was heated in an oil bath kept at 1408C for 18 h, then
2 mL of water were added and stirring was continued at
908C for additional 8 h. The mixture was cooled to room
temperature and concentrated HCl (30 mL) was added. Fol-
lowing stirring of the biphasic mixture for 16 h, the aqueous
phase was separated, and volatile material was removed
Figure 3. IR spectrum in the 2200–1800 cmÀ1 region of the
Rh-containing ionic liquid phase.
under vacuum to afford methyltri-n-octylammonium chlo-
1
ride; yield: 42.5 g (99%). H NMR (300 MHz, CDCl3): d=
Adv. Synth. Catal. 2007, 349, 1858 – 1862
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1861

Stefano Paganelli et al.
COMMUNICATIONS
0.81 (t, 9H, J=6.6 Hz), 1.20–1.35 (m, 30H), 1.60 (br, 6H),
3.20 (s, 3H), 3.30 (m, 6H); 13C NMR (75 MHz, CDCl3): d=
14.0, 22.3, 22.5, 26.3, 28.9, 29.0, 31.6, 52.2, 61.2.
[6] F. Favre, H. Olivier-Bourbigou, D. Commereuc, L.
Saussine, Chem. Commun. 2001, 1360–1361.
[7] C. P. Mehnert, R. A. Cook, N. C. Dispenziere, M. Afe-
worki, J. Am. Chem. Soc. 2002, 124, 12932–12933; A.
Riisager, P. Wasserscheid, R. van Hal, R. Fehrmann, J.
Catal. 2003, 219, 452–455.
Triphasic HydroformylationCatalysed by RhCl
3
To a glass tube containing TOMAC (1.0 mL) and H2O
(4.0 mL), RhCl3 (0.1 mmol) was added and stirred until
complete dissolution in the TOMAC phase. A solution of
olefin (1.0 mmol) in isooctane (4.0 mL) was then added to
the aqueous/ionic system. The tube was placed into a
150 mL stainless steel autoclave under nitrogen, pressurised
to 90 atm with syngas (CO/H2 =1) and heated at 60–808C
for the due time (see Table and plots). The reactor was then
cooled to room temperature and the residual gases vented.
The organic phase was separated, dried on Na2SO4 and the
reaction products characterised by GC-MS.
[8] J. P. Arhancet, M. E. Davis, J. S. Merola, B. E. Hanson,
Nature 1989, 339, 454–455; M. E. Davis, Chemtech
1992, 22, 498–502.
[9] M. Solinas, J. Jiang, O. Stelzer, W. Leitner, Angew.
Chem. Int. Ed. 2005, 44, 2291–2295.
[10] P. Tundo, A. Perosa, Chem. Soc. Rev. 2007, 36, 532–
550.
[11] P. Tundo, A. Perosa, React. Funct. Polym. 2003, 54, 95–
101.
[12] A. Perosa, P. Tundo, M. Selva, P. Canton, Chem.
Commun. 2006, 4480–4482.
[13] C. Bertucci, C. Botteghi, D. Giunta, M. Marchetti, S.
Paganelli, Adv. Synth.Catal. 2002, 344, 556–562.
[14] P. W. N. M. van Leeuwen, C. Claver, (Eds.), Rhodium
catalyzed hydroformylation, Kluwer Academic Publish-
ers, Dordrecht, 2000.
Acknowledgements
The authors wish to thank the Italian Ministry of University
and Research (MUR) for funding.
[15] J. Blum, I. Amer, A. Zoran, Y. Sasson, Tetrahedron
Lett. 1983, 24, 4139–4142.
[16] Y. Sasson, A. Zoran, J. Blum, J. Mol. Catal. 1981, 11,
293–300.
References
[17] L. Magna, Catalysis in nonaqueous ionic liquid, in:
Multiphase Homogeneous Catalysis, (Eds.: B. Cornils,
W. A. Herrmann, I. T. Horvath, W. Leitner, S. Mecking,
H. Olivier-Bourbigou, D. Vogt),Wiley-VCH, Weinheim,
2005, vol. 2, p 477.
[1] E. G. Kuntz, Chemtech 1987, 17, 570–575.
[2] B. Cornils, W. A. Herrmann, I. T. Horvath, W. Leitner,
S. Mecking, H. Olivier-Bourbigou, D. Vogt, (Eds.),
Multiphase Homogeneous Catalysis, Wiley-VCH, Wein-
heim, 2005.
[3] B. Sueur, L. Leclercq, M. Sauthier, Y. Castanet, A.
Mortreux, H. Bricout, S. Tilloy, E. Monflier, Chem.
Eur. J. 2005, 11, 6228–6236.
[18] J. P. Arhancet, M. E. Davis, J. S. Merola, B. E. Hanson,
J. Catal. 1990, 121, 327–339.
[19] T. Jefferey, Tetrahedron 1996, 52, 10113–10130.
[20] A. Haynes, B. E. Mann, G. E. Morris, P. M. Maitlis, J.
Am. Chem. Soc. 1993, 115, 4093–4100.
[4] I. T. Horvath, J. Rabai, Science 1994, 266, 72–75.
[5] P. G. Jessop, T. Ikariya, R. Noyori, Chem. Rev. 1999, 99,
475–494.
[21] K. Sato, M. Aoki, M. Ogawa, T. Hashimoto, R. Noyori,
J. Org. Chem. 1995, 61, 8310–8311.
1862
asc.wiley-vch.de
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 1858 – 1862