A novel dicationic phenoxaphosphino-modified xantphos-type ligand: A ligand for highly active and selective, biphasic, rhodium catalysed hydroformylation in ionic liquids

Article abstract of DOI:10.1039/b403261f

A highly active and regioselective catalyst obtained from a novel dicationic ligand (1) and Rh(CO)₂(acac) for hydroformylation of 1-hexene and 1-octene in ionic liquids is reported. Optimisation studies of various reaction parameters led to an unprecedentedly active (TOFs > 6200 mol mol^-1 h^-1, T = 100°C), selective (1/b ratios > 40) and stable hydroformylation procedure. No catalyst leaching (Rh-loss < 0.07% of initial rhodium intake, P-loss < 0.4% of the initial phosphorus intake) or losses in performance could be measured during 1-octene hydroformylation recycle experiments in 1-butyl-3-methylimidazolium hexafluorophosphate. At low catalyst loadings activities and regioselectivities competitive with one-phase catalysis in conventional solvents were observed. At high catalyst loadings the system is extremely stable and has a long shelf-life as a result of the formation of stable, if inactive rhodium dimers.

Full text of DOI:10.1039/b403261f

View Article Online / Journal Homepage / Table of Contents for this issue
A novel dicationic phenoxaphosphino-modified Xantphos-type
ligand: a ligand for highly active and selective, biphasic, rhodium
catalysed hydroformylation in ionic liquids
Raymond P. J. Bronger,a Silvana M. Silva,b Paul C. J. Kamera and
Piet W. N. M. van Leeuwen*a
a van’t Hoff Institute for Molecular Sciences, University of Amsterdam,
Nieuwe Achtergracht 166, 1018 WV Amsterdam, The Netherlands.
E-mail: pwnm@science.uva.nl; Fax: ϩ(31)(0)20-5255604
b Laboratory of Molecular Catalysis, Instituto de Quimica, UFRGS, Av. Bento Gonçalves,
9500 Porto Alegre, 91501-970 RS, Brazil
Received 2nd March 2004, Accepted 30th March 2004
First published as an Advance Article on the web 16th April 2004
A highly active and regioselective catalyst obtained from a novel dicationic ligand (1) and Rh(CO)2(acac) for
hydroformylation of 1-hexene and 1-octene in ionic liquids is reported. Optimisation studies of various reaction
parameters led to an unprecedentedly active (TOFs > 6200 mol molϪ1 hϪ1, T = 100 ЊC), selective (l/b ratios > 40) and
stable hydroformylation procedure. No catalyst leaching (Rh-loss < 0.07% of initial rhodium intake, P-loss < 0.4%
of the initial phosphorus intake) or losses in performance could be measured during 1-octene hydroformylation
recycle experiments in 1-butyl-3-methylimidazolium hexafluorophosphate. At low catalyst loadings activities and
regioselectivities competitive with one-phase catalysis in conventional solvents were observed. At high catalyst
loadings the system is extremely stable and has a long shelf-life as a result of the formation of stable, if inactive
rhodium dimers.
retention, we designed a novel diphosphine ligand. Our recent
studies have shown that various xanthene based diphosphines
Introduction
In recent years much effort has been devoted to the develop-
show a moderate activity, but a high selectivity in rhodium
ment of catalytic systems that allow facile separation of the
catalysed hydroformylation.25–27 This high selectivity is in part
products and catalysts.1 Especially systems that enable efficient
retained when modified for use in RTILs.24,28 In addition, the
catalyst recycling have attracted much attention.2 In this field
nature of the anions and cations of both the ionic liquid and
the ligand have a major influence on the recyclability of the
the immobilisation of tailor-made catalysts to organic and
inorganic solid supports has been successful, but in these
catalyst. Excellent retention of the catalyst was observed
systems immobilisation often results in lower catalytic activity
by adjusting the ligand structure to the ions of the solvent.19
due to mass transfer limitations.3–7 The use of soluble supports,
Previously, we had found that when phenoxaphosphino-
such as dendrimers, results in immobilised homogeneous
modified xanthene-type ligands (Fig. 1) are used in conventional
catalysts that do not suffer from mass transfer limitations,8–11
solvents the catalytic activity is enhanced considerably.29,30
but the limited availability of proper membrane material
hampers efficient recycling. In recent years room-temperature
ionic liquids (RTIL) have proven to be attractive and alternative
mediums for many homogeneously catalysed reactions. The
properties of ionic liquids can be tuned easily by adjustment of
the cation–anion pair. For example, a plethora of imidazolium
Ϫ
ionic liquids can be prepared by varying the anion (XϪ, PF6
,
BF4Ϫ, CH3COOϪ, etc...) and the alkyl groups on the aromatic
ring. The combination of high density, high stability, non-
measurable vapour pressure, and the possibilities for catalyst
immobilisation via ionic interactions allows easy separation of
the catalyst by simple phase separation or distillation and
recycling, as most products are not or only slightly soluble in
the ionic phase. Comprehensive information on this field can be
found in the reviews that have appeared on this subject.12–16
Compared to the aqueous two-phase catalysis concept,17,18 the
scope of two-phase catalysis can be extended to substrates that
are poorly soluble or insoluble in aqueous media. In addition,
the use of phosphites,19 phosphonites and phosphinites as
modifying ligands for two-phase catalysis comes within reach in
RTILs because degradation reactions, such as hydrolysis,20,21
are less likely to occur.
Fig. 1 POP-Xantphos.29,30
Here we report our recent advances in the application
of a novel phenoxaphosphino-modified Xantphos-type ligand
in RTILs (Fig. 2), which was recently reported in a com-
munication.31
Results and discussion
Synthesis
For hydroformylation reactions 1-butyl-3-methylimid-
azolium hexafluorophosphate (BMI.PF6) is often used as the
reaction medium. In this medium a good activity,19,22,23 selec-
tivity19,22,24 and complete retention of the catalyst24 can be
obtained. Unfortunately, no systems have been reported that
combine all three aspects. In order to obtain a catalytic system
that combines a high activity, high selectivity and a good
Previous studies employing phenoxaphosphino-modified
Xantphos-type ligands in the rhodium catalysed hydroformyl-
ation in toluene have shown that these systems lead to very
high hydroformylation activities and selectivities. Averaged
turn-over-frequencies (TOFs) of 1700 mol molϪ1 hϪ1 (T =
80 ЊC, p(CO–H2, 1 : 1)) = 20 bar, [Rh] = 1 mM, [1-octene] = 637
mM) have been observed.29,30,32,33
D a l t o n T r a n s . , 2 0 0 4 , 1 5 9 0 – 1 5 9 6
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
1590

View Article Online
Scheme 1 Synthesis of ligand 1 (yields in parenthesis). Reagents and conditions: (i) 2.2 eq. AlCl3/2.2 eq. 5-bromovaleric chloride (90%);
(ii) (a) InCl3/chlorodimethylsilane (86%), (b) Br2 (90%); (iii) (a) n-BuLi, Ϫ80 ЊC, 30 min, (b) 10-chloro-2,8-dimethylphenoxaphosphine (48%);
(iv) (a) 1-methylimidazole, 80 ЊC, 8 days in CH3CN–toluene (46%), (b) KPF6 in H2O (77%).
Table 1 Hydroformylation of 1-octenea
Cycle
TOFb, c
% Isom.b, d(%)
Sel.b, e(%)
l/b Ratiob
1
2
3
4
65
88
93
112
107
318
305
11.8
8.3
7.8
7.7
9.6
13.3
17.7
86.2
89.8
90.1
90.3
88.1
85.0
80.8
44
49
44
44
38
49
55
5
6f
7f
a Conditions: T = 100 ЊC, p(CO–H2, 1 : 1) = 17 bar, [Rh] = 6.4 mM, [1] =
27 mM, substrate/Rh = 988, ionic liquid = BMI.PF6. Rate of stirring =
900 rpm. In none of the experiments was hydrogenation observed.
b Linear to branched ratio, percent isomerisation to 2-octene, percent
linear aldehyde and turnover frequency were determined at ∼30%
alkene conversion. c Turnover frequency = (mol aldehyde) (mol Rh)Ϫ1
Fig. 2 Novel dicationic ligand.
The high regioselectivity is mainly determined by the
differences in the rates of β-hydrogen elimination between
the respective rhodium–alkyl species rather than by prefer-
ential formation of linear alkyl rhodium intermediates.29,33 The
use of this type of ligands for immobilisation procedures results
in a relatively high activity compared to other reported systems,
while the regioselectivity remains unequivocally high.31 Studies
on related ligands have shown that modification of the
9,9-dimethylxanthene backbone at the 2- and 7-positions with
different aliphatic groups does not have a large influence on the
catalytic reaction. This is in contrast to modification of the
phenoxaphosphino-moieties, which often leads to a change in
selectivity.32 Therefore, a synthetic procedure was developed
that allows easy modification of the 9,9-dimethylxanthene
backbone by two cationic moieties at the 2- and 7-positions.
Ligand 1 was prepared via a six-step synthetic route. The first
step involved a Friedel–Crafts acylation using 5-bromovaleryl
chloride and 9,9-dimethylxanthene. Next, the ketone function-
alities were reduced catalytically using InCl3 and chloro-
dimethylsilane.34 Compared to standard methods for reduction,
like the Wolff–Kischner or Clemmensen reduction, the catalytic
reduction provides a more efficient route. Next, bromination of
the 4,5-positions of the modified-xanthene backbone followed
by dilithiation and reaction with 10-chloro-2,8-dimethyl-
phenoxaphosphine yielded 2. Reaction of 2 with two equiv-
hϪ1 d % Isom. = (octenes other than 1-octene)/(octenes other than
.
1-octene ϩ aldehydes) × 100%. e Sel. = (linear aldehyde)/(branched
aldehyde ϩ octenes other than 1-octene) × 100%. f p(H2) = 40 bar,
p(CO) = 6 bar.
Hydroformylation of 1-octene was carried out at 100 ЊC under
17 bar of CO–H2 (1 : 1) using 3 mL of ionic liquid and 3 mL of
1-octene. At approximately 30% conversion the reaction was
stopped by cooling the autoclave in an ice-bath and venting the
gasses. Reactions were run at fairly low conversion in order to
calculate the initial activity. Additionally, loss in activity due to
catalyst leaching is easier to detect compared to experiments
run at full conversion. Recycling experiments were performed
by removal of the octene/aldehyde layer by decantation followed
by charging the autoclave with 1-octene, repressurisation
and heating to the desired temperature. The results of seven
consecutive recycling experiments are shown in Table 1.31
The applied L/Rh ratio, preparation of catalyst precursor
and catalysis conditions are similar to the optimised conditions
described by Dupont et al. who used sulfonated Xantphos
as modifying ligand.28 Already from the first experiment it is
evident that very high linear to branched (l/b) ratios are
obtained by employing this ligand, albeit with moderate
activity. The increase in catalytic activity observed for recycle
experiments (entries 2–4) indicates that probably no complete
conversion toward the active rhodium–hydride species had been
reached in the first run, an effect also observed by others.24
Starting from recycle experiment 4 reproducible activities were
obtained. The relatively low activity for this ligand system and
the red colour of the rhodium complex in the ionic liquid after
catalysis suggests the presence of a dimeric rhodium species
(Scheme 2).35 Dimer formation is a known cause for lower
activity of rhodium hydroformylation catalysts. As a result the
concentration of the active monomeric catalyst species
decreases. Since most hydroformylation catalysts exhibit a first
order dependency in the concentration of this rhodium–hydride
Ϫ
alents of 1-methylimidazole at 80 ЊC followed by a halide/PF6
exchange gives the desired ligand 1 in a moderate overall yield
(Scheme 1).
Catalysis
The immobilised catalyst precursor was formed by mixing
Rh(CO)2(acac) and four equivalents of ligand in an aceto-
nitrile–BMI.PF6 mixture. After stirring for 1 h the acetonitrile
was removed by evaporation under reduced pressure at 60 ЊC
for 3 h, which afforded a red solution of the ionic liquid cata-
lyst-precursor. Prior to hydroformylation the catalyst precursor
was heated at 80 ЊC for 1 h under 15 bar of CO–H2 (1 : 1)
to start formation of the active hydroformylation catalyst.
D a l t o n T r a n s . , 2 0 0 4 , 1 5 9 0 – 1 5 9 6
1591

View Article Online
Table 2 Hydroformylation of 1-octene at different rates of stirringa
Cycle
rpm
TOFb, c
% Isom.b, d(%)
Sel.b, e(%)
l/b ratiob
1
550
230
550
900
1285
1600
3450
2900
4850
5150
6400
7400
14
14
11
11
19
21
84
84
87
87
79
77
64
63
65
59
56
52
Scheme 2 Equilibrium between Rh-dimer and Rh–hydride species.
4–5
2–3
6–7
10–11
8–9
species,36 the reaction becomes slower. In order to shift the
equilibrium between the rhodium monomer and the rhodium
dimer towards the rhodium monomer a higher partial hydrogen
pressure was applied.35
a Conditions: T = 100 ЊC, p(CO–H2, 1 : 9) = 60 bar, [Rh] = 1.7 mM,
[ligand] = 27 mM, substrate/Rh = 3823, ionic liquid = BMI.PF6. In none
of the experiments was hydrogenation observed. b Linear to branched
ratio, percent isomerisation to 2-octene, percent linear aldehyde and
turnover frequency were determined at ∼40% alkene conversion.
Indeed, the catalytic activity could be enhanced to TOFs >
300 mol molϪ1 hϪ1 by applying higher hydrogen pressures while
keeping the CO pressure constant (Table 1, entries 6 and 7).
In addition, even at these high hydrogen pressures, no hydro-
genation takes place. Unfortunately, and in contrast to our
previous studies using sulfonated Xantphos, no conclusive
high-pressure IR and high-pressure NMR data could be
obtained for this system.37 According to ICP analysis of the
octene–aldehyde mixture after catalysis neither phosphorus
(<100 ppb; <0.4% of the initial phosphorus intake) nor
rhodium (<5 ppb; <0.07% of the initial rhodium intake)
leaching was detected. Furthermore, the catalyst proved to be
extremely stable even under atmospheric conditions without
special precautions. When the hydroformylation of 1-octene
was repeated after the ionic liquid containing the catalyst had
been stored for 14 days at room temperature under air, similar
catalysis results were obtained as in the initial hydroformylation
experiment, although a slightly higher rate of isomerisation
was observed. The increased isomerisation might be due to
formation of acidic impurities caused by some anion hydrolysis
with moisture from air.38–41
c Turnover frequency = (mol aldehyde) (mol Rh)Ϫ1 hϪ1 d % Isom. =
.
(octenes other than 1-octene)/(octenes other than 1-octene ϩ alde-
hydes) × 100%. e Sel. = (linear aldehyde)/(branched aldehyde ϩ octenes
other than 1-octene) × 100%.
Alternatively, there could exist a small positive order in
hydrogen pressure. Decomposition of the catalytic system
could be excluded since under the standard reaction conditions
the same activity and selectivity were measured, and no
rhodium or phosphorus leaching was detected (Table 3,
cycle 8).
The very high hydroformylation and isomerisation activity
prompted us to test the catalyst for the selective hydroformyl-
ation of trans-2-octene to nonanal. In order to enhance the
isomerisation activity and to suppress the hydroformylation of
internal alkenes to branched aldehydes the reaction temper-
ature was increased to 120 ЊC and the CO pressure was reduced
to 2.5 bar (p(CO) = 2.5 bar, p(H2) = 47.5 bar). Severe catalyst
deterioration was observed under these conditions during the
hydroformylation of trans-2-octene, even at high ligand concen-
trations ([1] = 27 mM). Therefore we also tested the system
under identical conditions as for the hydroformylation of
1-octene (Table 3, cycle 9 and 10). High activities but very low
l/b ratios were observed, but again the catalyst was not stable
when trans-2-octene was used as substrate. Subsequent experi-
ments using 1-octene confirmed that catalyst decomposition
had taken place (Table 3, cycles 11–13). Why the use of trans-2-
octene results in catalyst decomposition is currently unclear,
as the resting state is most likely the species (diphosphine)-
RhH(CO)2 for both 1-octene and trans-2-octene hydroformyl-
ation.42 Perhaps unidentified impurities are present in trans-2-
octene, such as dienes and enones.43 The low rate of alkene
coordination is determined by the lower reactivity of internal
alkenes and the low concentration of the alkene in the ionic
liquid. Rhodium and phosphorus leaching was detected (1%
Rh and 8% P leaching of the initial intake). The low selectivity
and increased activity of experiments 11–13 compared to
recycle experiments 1–8 are typical of non-ligated rhodium,
which could point to ligand decomposition.
The effect of catalyst concentration on hydroformylation
Despite the promising initial results with this catalytic system,
the activity is still low compared to catalysis under one-phase
conditions, which was conducted at lower temperatures.29,30,32,33
The existence of dimeric rhodium species, possibly still present
at high hydrogen pressures, is considered to be the main reason
for the relatively low activity. High catalyst concentrations may
favour the formation of rhodium dimers35 and therefore we
opted to reduce the catalyst concentration. This was expected to
shift the rhodium dimer/monomer equilibrium toward the
monomeric rhodium species, but it also reduces the effect of
potential mass transfer limitations that are commonly observed
for immobilised systems. The results of catalysis at lowered
rhodium concentration ([Rh] = 1.7 mM in ionic liquid, [1] = 27
mM) and at different stirring rates (from 230 to 1600 rpm) are
summarised in Table 2.
Lowering the rhodium concentration results in a tremendous
increase in activity per rhodium and in a slight increase in
regioselectivity. The effect of stirring rate indicates that we
are operating under mass transfer limitations as the activity
increases when the reaction mixture is stirred more vigorously.
The stirring rate also has a large impact on isomerisation activ-
ity, possibly due to differences in dissolution rates of the sub-
strates. Overall the best performance was obtained at a stirring
rate of 900 rpm as a good balance between isomerisation and
hydroformylation activity was found.
Next, the effect of reducing the ligand concentration was
studied ([Rh] = 1.7 mM in ionic liquid, [1] = 7 mM) (Table 3). At
this lowered ligand concentration slightly higher activities
(TOF = 6200 mol molϪ1 hϪ1.) were obtained, while maintaining
a high l/b ratio. Reducing the p(CO–H2, 1 : 9) = 60 bar to p(CO–
H2, 1 : 1) = 12 bar resulted in a slight drop in catalytic activity
only. This indicates that still some rhodium dimer species might
be present under lowered hydrogen pressures, but that the
equilibrium is shifted almost completely to the monomeric
rhodium species. In addition, the colour of the ionic liquid
solution turns yellow instead of remaining red during recycling
experiments, supporting this hypothesis.
Remarkably, a stable catalyst system under ambient con-
ditions is only obtained at high catalyst loadings, because at
low concentrations decomposition is observed within a few
days after exposing the catalyst to air. When stored under
a pressure of CO/H2 the catalyst retains its activity and
selectivity. Apparently, only the rhodium-dimer is stable under
ambient conditions.
Overall the activity and selectivity for the 1-octene hydro-
formylation experiments are by far superior to those reported
by others under comparable reaction conditions (Table 4).
Hydroformylation of 1-octene and 1-hexene at 80 ؇C and
hydroformylation in HMI.PF6
Hydroformylation at 80 ЊC was carried out to enable a valid
comparison with literature data on the one-phase systems
(Table 5). At this temperature TOFs of 1200 mol molϪ1 hϪ1 were
D a l t o n T r a n s . , 2 0 0 4 , 1 5 9 0 – 1 5 9 6
1592

View Article Online
Table 3 Hydroformylation of 1-octene and trans-2-octenea
Cycle
Substrate
TOFb, c
% Isom.b, d(%)
Sel.b, e(%)
l/b Ratiob
1
2
3
1-Octene
1-Octene
1-Octene
1-Octene
1-Octene
1-Octene
1-Octene
1-Octene
trans-2-Octene
trans-2-Octene
1-Octene
1-Octene
1-Octene
5250
6150
6800
7200
6650
5100
5000
6650
1200
1200
9600
10100
8700
26.5
23.6
21.0
21.3
27.0
15.8
13.4
26.6
13.6g
16.3g
ndi
72
75
77
77
71
82
85
72
37
14
Nd
54
53
45
41
43
43
42
44
41
41
4
5
6f
7f
8
9
0.7h
0.2h
1.9
2.4
2.5
10
11
12
13
23.0
26.0
a Conditions: T = 100 ЊC, p(CO–H2, 1 : 9) = 60 bar, [Rh] = 1.7 mM, [ligand] = 7 mM, substrate/Rh = 3823, ionic liquid = BMI.PF6. Rate of stirring =
900 rpm. In none of the experiments was hydrogenation observed. b Linear to branched ratio, percent isomerisation to 2-octene, percent linear
aldehyde and turnover frequency were determined at ∼40% alkene conversion. c Turnover frequency = (mol aldehyde) (mol Rh)Ϫ1 hϪ1 d % Isom. =
.
(octenes other than 1-octene)/(octenes other than 1-octene ϩ aldehydes) × 100%. e Sel. = (linear aldehyde)/(branched aldehyde ϩ octenes other than
1-octene) × 100%. f p(CO–H2, 1 : 1) = 12 bar. g Predominantly isomerisation to cis-2-octene. h The differences in l/b ratio between cycle 9 and cycle 10
are most likely caused by some aldehydes present in the RTIL from the previous catalysis experiment (cycle 8). i nd = Not determined.
Table 4 Comparison of different systemsa
Ligand backbone
Ref.
TOFb
Rate c
l/b Ratio
Rh leachingd(%)
P leachingd(%)
Phenole, f
19
23
22
24
240
552
810
52
32
6200
1.3
1.3
2.0
ndk
0.1
5.3
13
1
16
21
13
44
2
nr
nrg
nr
nr
2-Imidazoliumh
Cobaltoceniumi
Xanthenej
<0.2
<0.07l
nr
nr
Xanthenem
Xanthenei
28
nr
This work
<0.07l
<0.4j
a Conditions: T = 100 ЊC, p(CO–H2) = 10–60 bar, substrate = 1-octene, ionic liquid = BMI.PF6. In none of the experiments was hydrogenation
observed. b Turnover frequency = (mol aldehyde) (mol Rh)Ϫ1 hϪ1 c Rate = (mol aldehyde) LϪ1 hϪ1 d Percentage of leached rhodium/phosphorus of
.
.
initial intake. e Monophosphite ligand. f Substrate = 1-hexene, T = 80 ЊC. g nr = Not reported. h Monophosphine ligand. i Diphosphine ligand.
j Phenylguanidium modified diphosphine ligand. k nd = No sufficient data to calculate the rate in (mol aldehyde) LϪ1 hϪ1 l Detection limit of ICP
.
analysis. m Sulfonated diphosphine ligand.
Table 5 Hydroformylation of 1-hexene and 1-octenea
Cycle
Substrate
T /ЊC
TOFb, c
% Isom.b, d(%)
Sel.b, e(%)
l/b Ratiob
1
2
3
1-Octene
1-Octene
1-Octene
1-Octene
1-Hexene
1-Hexene
1-Hexene
1-Octene
100
100
100
80
80
80
4800
6250
6200
1200
4000
5800
8900
950
12.3
11.1
13.0
10.3
nd
nd
nd
86
87
85
88
nd
nd
nd
85
45
42
44
50
54
58
54
31
4–6
7f, g
8f, g
9f, g
10–15
80
80
12.2
a Conditions: T = 100 ЊC, p(CO–H2, 1 : 9) = 60 bar, [Rh] = 1.7 mM, [ligand] = 7 mM, substrate/Rh = 3823, ionic liquid = BMI.PF6. Rate of stirring =
900 rpm. In none of the experiments was hydrogenation observed. b Linear to branched ratio, percent isomerisation to 2-octene, percent linear
aldehyde and turnover frequency were determined at ∼40% alkene conversion. c Turnover frequency = (mol aldehyde) (mol Rh)Ϫ1 hϪ1 d % Isom. =
.
(octenes other than 1-octene)/(octenes other than 1-octene ϩ aldehydes) × 100%. e Sel. = (linear aldehyde)/(branched aldehyde ϩ octenes other than
1-octene) × 100%. f Substrate/Rh = 4821. g Cycle 7: 89% conversion, cycle 8: 77% conversion, cycle 9: 49% conversion.
observed; these surprisingly high activities are competitive with
results obtained under typical one-phase hydroformylation
conditions (TOF = 1700 mol molϪ1 hϪ1, solvent = toluene, T =
80 ЊC, p(CO–H2, 1 : 1) = 20 bar, [Rh] = 1 mM, [1-octene] = 637
mM).29,30,32,33 To investigate the influence of substrate solubility
1-hexene was tested as substrate in BMI.PF6, because the solu-
bility of 1-hexene in BMI.PF6 is higher than that of 1-octene.
Under one-phase conditions, the hydroformylation of 1-hexene
or 1-octene gives similar results as regards both activity and
selectivity.33,44 Differences in performance between the two sub-
strates in ionic liquids might be attributed to differences in
solubility, but the activation parameters of the two processes
for the two substrates have not been determined. Comparison
of the results shows that a much higher activity for the
hydroformylation of 1-hexene is obtained (Table 5).
The use of 1-hexene leads to 8% rhodium leaching, deter-
mined by ICP analysis, after three recycle experiments. The
large increase in TOF that is observed from cycle 7–9 can there-
fore in part be attributed to catalysis not taking place in the
ionic liquid phase, but in the substrate phase. Leaching was
corroborated by consecutive hydroformylation reactions with
1-octene at T = 80 ЊC since the TOF dropped from 1200 mol
molϪ1 hϪ1 for the initial hydroformylation experiments (cycle
4–6) to 950 mol molϪ1 hϪ1 for the experiments performed after
1-hexene hydroformylation (cycle 10–15). This loss in activity
(21%) is larger than the detected loss of rhodium (8%), which
might indicate catalyst decomposition. Catalyst decomposition
would also explain the loss in regioselectivity. During none of
the 1-octene hydroformylation experiments rhodium or phos-
phorus leaching was detected.
D a l t o n T r a n s . , 2 0 0 4 , 1 5 9 0 – 1 5 9 6
1593

View Article Online
a
Table 6 Hydroformylation of 1-octene in HMI.PF6
rate, substrates and ionic liquids employed. The best results
were obtained by using a low rhodium concentration, a high
hydrogen pressure and a stirring rate of 900 rpm in BMI.PF6
as the ionic liquid and 1-octene as the substrate. Under
these conditions an unrivalled combination of activity, regio-
selectivity and recyclability has been observed in the rhodium
catalysed hydroformylation of 1-octene in ionic liquids.
Cycle
TOFb, c
% Isom.b, d(%)
Sel.b, e(%)
l/b Ratiob
1
2
1400
4950
9000
5250
10.8
5.9
16.6
19.0
87
92
81
79
41
44
44
43
3–6
7–9f
a Conditions: T = 100 ЊC, p(CO–H2, 1 : 9) = 60 bar, [Rh] = 1.7 mM,
[ligand] = 7 mM, substrate/Rh = 3823, ionic liquid = HMI.PF6. Rate
of stirring = 900 rpm. In none of the experiments was hydrogen-
ation observed. b Linear over branched ratio, percent isomerisation to
2-octene, percent linear aldehyde and turnover frequency were deter-
mined at ∼40% alkene conversion. c Turnover frequency = (mol alde-
Experimental
General procedures
hyde) (mol Rh)Ϫ1 hϪ1 d % Isom. = (octenes other than 1-octene)/
.
All air- or water-sensitive reactions were performed using
standard Schlenk techniques under an atmosphere of purified
argon. Toluene was distilled from sodium, THF from sodium/
benzophenone, and hexanes from sodium/benzophenone/
triglyme. Isopropanol and dichloromethane were distilled
from CaH2. Chemicals were purchased from Acros Chimica,
and Aldrich Chemical Co. Silica gel 60 (230–400 mesh)
purchased from Merck was used for column chromatography.
1-n-Butyl-3-methylimidazolium hexafluorophosphate ionic
liquid (BMI.PF6)45 and 2,8-dimethyl-10-chlorophenoxaphos-
phine46 were prepared according to literature procedures. It is
important to note that the quality of the ionic liquid has a large
impact on the percentage of isomerisation (compare the results
of Tables 3 and 5). Traces of water in the ionic liquid might lead
to some anion hydrolysis, which might cause a higher propor-
tion of isomerisation. However, reactions with the same ionic
liquid give reproducible results; laboratory-prepared ionic
liquid led to a lower percentage of isomerisation than the
commercially available ionic liquid.
(octenes other than 1-octene ϩ aldehydes) × 100%. e Sel. = (linear alde-
hyde)/(branched aldehyde ϩ octenes other than 1-octene) × 100%.
f p(CO–H2, 1 : 1) = 12 bar.
The cause of this different behaviour of the catalyst system
when using different substrates is unclear, therefore 1-hexyl-3-
methylimidazolium hexafluorophosphate (HMI.PF6) was used
as the reaction medium as the solubility of 1-octene is increased
in this solvent.15,40 The results for the recycling experiments are
summarised in Table 6.
The activity for 1-octene hydroformylation increases from
TOFs of ∼6200 mol molϪ1 hϪ1 in BMI.PF6 to TOFs of ∼9000
mol molϪ1 hϪ1 in HMI.PF6, while maintaining a high l/b ratio.
The catalyst can again be recycled for at least 10 times without
showing any noticeable loss in activity or selectivity, but in
contrast to the recycle experiments in BMI.PF6 detectable
rhodium leaching occurs, albeit in trace amounts (0.07–0.08%
of the initial rhodium intake). Surprisingly, only minor differ-
ences in activity between catalysis in BMI.PF6 or HMI.PF6 are
measured when the pressure is reduced to 12 bar of CO–H2
(1 : 1). Although the differences in catalysis could be ascribed to
differences in substrate solubility, it must be noted that other
minor differences between the two ionic liquids exists that
influence catalysis, like viscosity and surface tension, as both
parameters affect the rates of mass transfer.39 Also differences
in dissolution rates of CO and H2 influence the catalysis results,
as corroborated by the effect of stirring rate on activity.
From a thermodynamic point-of-view the differences in
catalysis results between the two ionic liquids are not clear. If
the three phases of the reaction mixtures are in equilibrium
than the chemical potential of the gas-phase, substrate-phase
and ionic liquid-phase should be equal. For the experiments
with BMI.PF6 and HMI.PF6 the same CO and H2 pressures
were applied, the ionic liquids have no measurable vapour
pressure, and therefore the vapour pressure of the substrate
should be equal and independent of the ionic liquid. Con-
sequently, the activity (or fugacity) of CO, H2 and substrate
should be irrespective of the ionic liquid employed, and the
catalytic activities should be the same if the activation
parameters of the reaction were the same. Since the rates are
not the same, more research is required to elucidate the exact
thermodynamic parameters of all mixtures. The first results
seem to suggest that above 900 rpm no mass-transfer limit-
ations are involved, but this needs further scrutiny for all
substrates and conditions.
Melting points were determined on a Gallenkamp MFB-595
melting point apparatus in open capillaries and are reported
uncorrected. NMR spectra were recorded on a Varian Mercury
300 and Inova 500 spectrometer. 31P and 13C spectra were
measured in 1H decoupled mode. TMS was used as an external
1
standard for H and 13C NMR and 85% H3PO4 in H2O as an
external standard for 31P NMR. Hydroformylation reactions
were carried out in a 75 mL home-made stainless steel
autoclave. The alkene was purified by percolation over neutral
activated alumina. The reactions were terminated by cooling on
ice, stopping the mechanical stirring and by venting the gases.
Synthesis gas (CO–H2, 1 : 1, 99.9%), CO and H2 were purchased
from Air Liquide. Gas chromatographic analyses were run on
an Interscience HR GC Mega 2 apparatus (split/splitless
injector, J&W Scientific, DB-1 30 m column, film thickness
3.0 mm, carrier gas 70 kPa He, FID detector) equipped with
a Hewlett Packard Data system (Chrom-Card).
2,7-Di(5-bromopentanoyl)-9,9-dimethylxanthene. At 0 ЊC 24.9
g of AlCl3 (187 mmol) was added slowly to a stirred solution of
17 g of 9,9-dimethylxanthene (81 mmol) and 25 mL of
5-bromovaleric chloride (187 mmol; 2.3 equivalents) in 250 mL
of CH2Cl2. After overnight stirring the reaction mixture was
poured into 100 mL of ice–water and extracted with CH2Cl2
(3 × 100 mL). Subsequently, the organic layer was dried over
MgSO4. The solvents were removed in vacuo and the resulting
dark green solid was purified by flash column chromatography
over silica (eluent: CH2Cl2). Yield: 39 g of a yellow–green solid
(90%) that was used without further purification. 1H NMR
(CDCl3): δ 8.10 (d, 4J(H,H) = 1.8 Hz, 2H), 7.83 (dd, 3J(H,H) =
8.7 Hz, 4J(H,H) = 2.1 Hz, 2H), 7.12 (d, 3J(H,H) = 8.4 Hz, 2H),
3.46 (t, 3J(H,H) = 6.6 Hz, 4H), 3.00 (t, 3J(H,H) = 6.9 Hz), 1.93
(m, 8H), 1.70 (s, 6H). 13C{1H} NMR (CDCl3): δ 198.41 (s),
153.49 (s), 132.96 (s), 130.10 (s), 128.40 (s), 127.38 (s), 116.90
(s), 37.46 (s), 34.95 (s), 34.35 (s), 33.32 (s), 33.11 (s), 23.09 (s).
Conclusions
A synthetic route toward a novel functionalised phenoxa-
phosphino-modified ligand has been described. While this
ligand was specifically designed for use in ionic liquids, minor
adjustments of the synthetic procedure will allow easy modifi-
cation of the ligand with other functional groups. During
recycling experiments under different reaction conditions we
have shown that the efficiency of this catalyst system is sensitive
to the catalyst concentration, partial hydrogen pressure, stirring
2,7-Di(5-bromopentyl)-9,9-dimethylxanthene. To
suspension of 1.3 g InCl3 (5.9 mmol) and 27.6 mL of chloro-
a stirred
D a l t o n T r a n s . , 2 0 0 4 , 1 5 9 0 – 1 5 9 6
1594

View Article Online
dimethylsilane (4.8 eq., 245 mmol) in 80 mL of CH2Cl2 was
added 27.2 g of 2,7-di-5-bromopentan-1-one-9,9-dimethyl-
xanthene (51 mmol) in 80 mL of CH2Cl2. The reaction was
followed by GC-MS and IR and quenched by addition of 100
mL of water after complete reduction of the ketone function-
alities (∼4 h reaction time). Next, the mixture is extracted with 3
× 80 mL of CH2Cl2. Subsequently, the organic layer was dried
over MgSO4. The solvents were removed in vacuo and the result-
ing solid was purified by flash column chromatography (eluent:
hexanes). Yield: 22 g of a slightly yellow compound (86%) that
was used without further purification. 1H NMR (CDCl3): δ 7.18
(d, 4J(H,H) = 1.8 Hz, 2H), 6.99 (dd, 3J(H,H) = 8.4 Hz, 4J(H,H)
= 2.1 Hz, 2H), 6.94 (d, 3J(H,H) = 7.8 Hz, 2H), 3.41 (t, 3J(H,H) =
6.6 Hz, 4H), 2.60 (t, 3J(H,H) = 7.8 Hz, 4H), 1.89 (m, 4H), 1.66
(m, 4H), 1.61 (s, 6H), 1.49 (m, 4H). 13C{1H} NMR (CDCl3):
δ 148.96 (s), 136.91 (s), 129.98 (s), 127.49 (s), 126.09 (s), 116.36
(s), 35.58 (s), 34.24 (s), 34.19 (s), 32.94 (s), 32.67 (s), 31.13 (s),
28.08 (s). GC-MS (m/z, rel. intensity): 508 (Mϩ, 8), 493 (100),
413 (20), 371 (9), 356 (14), 276 (14), 221 (22), 207 (27), 131 (18),
55 (22).
2,7-Di(5-(3-methylimidazolium)pentyl)-9,9-dimethyl-4,5-di-
(2,8-dimethyl-10-phenoxaphosphino)xanthene hexafluorophos-
phate (1). To a stirred solution of 490 mg of 2,7-di(5-bromo-
pentyl)-9,9-dimethyl-4,5-bis(2,8-dimethyl-10-phenoxaphos-
phino)xanthene (0.52 mmol) in 30 mL of an acetonitrile–toluene
mixture (1 : 1) was added 0.1 mL of 1-methylimidazole (2.7 eq.,
1.4 mmol). The mixture was heated to 80 ЊC and stirred for 8 days
to obtain complete conversion. Next, the solvents were removed
in vacuo and the resulting solid was purified by precipitation from
toluene–acetonitrile. Yield: 250 mg of a white solid (46%). Mp =
1
216 ЊC (decomp.). 31P{1H} NMR (CD2Cl2): δ Ϫ69.03. H NMR
(CD2Cl2): δ 10.45 (s, 2H), 7.98 (br d, 3J(P,H) = 5.4 Hz, 4 H), 7.3–
7.1 (m, 14 H), 6.51 (br s, 2H), 4.22 (t, 3J(H,H) = 7.5 Hz, 4H), 4.01
3
(s, 3H), 2.40 (m, 4H), 2.36 (s, 12H), 1.84 (q, J(H,H) = 7.5 Hz,
4H), 1.53 (s, 6H), 1.48 (m, 4H), 1.27 (m, 4H). 13C{1H} NMR
(CD2Cl2): δ 156.17 (s), 152.72 (t, 20.5 Hz), 139.17 (s), 137.52 (t,
22.0 Hz), 135.24 (s), 133.83 (s), 133.36 (s), 132.36 (s), 129.33 (s),
128.86 (vt, unresolved), 125.34 (s), 123.83 (s), 120.05 (s), 119.51
(s), 51.87 (s), 38.58 (s), 36.93 (s), 36.64 (s), 34.12 (s), 32.47 (s),
32.07 (s), 27.49 (s), 22.53 (s). To a suspension of 150 mg of 2,7-
di(5-(3-methylimidazolium)pentyl)-9,9-dimethyl-4,5-di(2,8-di-
methyl-10-phenoxaphosphino)xanthene bromide (0.14 mmol) in
H2O was added 78 mg of KPF6 (3 eq., 42 mmol). After over-
night stirring the solvent is removed in vacuo and 15 mL of
CH2Cl2 was added. After filtration of the salts (KBr) the
CH2Cl2 is removed in vacuo yielding a white powder (yield: 130
mg, 77%). Mp = 221 ЊC (decomp.). 31P{1H} NMR (CD3CN):
4,5-Dibromo-2,7-di(5-bromopentyl)-9,9-dimethylxanthene. To
an ice-cooled solution of 13.5 g of 2,7-di-5-bromopentyl-9,9-
dimethylxanthene (26.6 mmol) in 130 mL of CH2Cl2 was added
dropwise 4.9 mL of Br2 (3.6 eq., 95.0 mmol) in 4.9 mL of
hexane. The reaction mixture was warmed to room temperature
and stirred overnight. Next the excess of Br2 is quenched with
100 mL of an aqueous NaSO3 solution, and the mixture
extracted with 3 × 80 mL of CH2Cl2. Subsequently, the organic
layer was dried over MgSO4. The solvents were removed
in vacuo and the resulting solid was purified by flash column
chromatography (eluent: CH2Cl2). Yield: 15.9 g of a yellow
solid (89.6%) that was used without further purification.
1H NMR (CDCl3): δ 7.28 (s, 2H), 7.11 (s, 2H), 3.4 (t, 3J(H,H) =
1
δ 61.42 (heptet, J(P,F; PF6Ϫ) = 706 Hz), Ϫ66.02 (s). 19F{1H}
1
1
NMR (CD3CN): δ Ϫ67.80 (d, J(P,F) = 704 Hz). H NMR
(CD3CN): δ 8.38 (br s, 2H), 7.97 (dd, 3J(P,H) = 6.6 Hz, 4J(H,H)
= 1.4 Hz, 4H), 7.34 (m, 4H), 7.28 (dd, 3J(H,H) = 8.1 Hz, 4J(H,H)
= 2.1 Hz, 4H), 7.24 (d, 4J(H,H) = 2H), 7.15 (d, 3J(H,H) = 8.1 Hz,
4H), 6.45 (d, 4J(H,H) = 1.5 Hz, 2H), 4.06 (t, 3J(H,H) = 7.2 Hz,
3
4H), 3.83 (s, 6H), 2.38 (t, J(H,H) = 7.8 Hz, 4H), 2.36 (s, 6H),
1.76 (m, 4H), 1.51 (s, 6H), 1.42 (m, 4H), 1.21 (m, 4H). 13C{1H}
NMR (DMSO-d6): δ 154.15 (s), 150.45 (t, 22.6 Hz), 137.67,
137.15 (s), 135.49 (t, 21.4 Hz), 133.64 (s), 132.85 (s), 131.21 (s),
130.52 (s), 128.41 (s), 126.73 (vt, unresolved), 124.31 (s), 122.91
(s), 118.21 (s), 117.75 (s), 49.56 (s), 36.09 (s), 34.52 (s), 31.79 (s),
30.38 (s), 29.59 (s), 25.27 (s), 19.95 (s). Anal. Calc. for
C61H65F12N4O3P4: C, 58.42; H, 5.22; N, 4.47. Found: C, 58.26;
H, 5.40; N, 4.58%.
3
6.6 Hz, 4H), 2.57 (t, J(H,H) = 7.5 Hz, 4H), 1.88 (m, 4H),
1.59 (s, 6H), 1.59–1.47 (m, 4H). 13C{1H} NMR (CDCl3):
δ 145.84 (s), 138.65 (s), 131.80 (s), 131.25 (s), 124.99 (s), 110.88
(s), 35.66 (s), 35.29 (s), 34.09 (s), 32.81 (s), 32.07 (s), 30.84 (s),
27.93 (s).
2,7-Di(5-bromopentyl)-9,9-dimethyl-4,5-di(2,8-dimethyl-10-
phenoxaphosphino)xanthene (2). At Ϫ78 ЊC 2.2 mL of n-butyl-
lithium (2.5 M in hexanes, 5.5 mmol) was added to a stirred
solution of 1.6 g of 4,5-dibromo-2,7-di(5-bromopentyl)-9,9-
dimethylxanthene (2.4 mmol). The resulting solution was
stirred for 30 min at Ϫ78 ЊC. Subsequently, a suspension of
1.5 g of 2,8-dimethyl-10-chlorophenoxaphosphine (5.7 mmol)
in 15 mL of toluene was added dropwise. The reaction mixture
was slowly warmed to room temperature and stirred overnight.
Next the diethyl ether was removed in vacuo and the mixture
was diluted with 40 mL of CH2Cl2 and hydrolyzed with 10 mL
of a 10% aqueous HCl solution. The water layer was removed
and the organic layer was dried over MgSO4. The solvents were
removed in vacuo and the resulting yellow–white solid was
crystallized from 2-propanol–toluene. Yield: 1.1 g of white
Hydroformylation. The catalyst precursor was prepared by
adding 3 mL of BMI.PF6 to a stirred solution of 1.3 mg of
Rh(CO)2(acac) (5 µmol) and 4 equivalents of ligand in 5 mL
of acetonitrile. After 1 h the acetonitrile was removed under
reduced pressure at 60 ЊC for 3 h affording a red ionic liquid
solution.
In a typical hydroformylation experiment a home-made
75 mL autoclave was charged with this catalyst-precursor
containing solution. After purging the autoclave three times
with CO–H2 (1 : 1), the reactor was brought to 5 bar of CO and
45 bar of H2. Next the autoclave was heated to 80 ЊC. After 1 h
at 80 ЊC the autoclave was cooled down to room temperature
and the gases were vented. The substrate (3 mL) was added
and the reactor was pressurised to 5 bar of CO and 45 bar of
H2. Next the autoclave was heated to 100 ЊC, reaching a total
pressure of 60 bar. Reactions were started by stirring at 900
rpm. After 15 min the stirring was stopped and the autoclave
was rapidly cooled on ice. After venting the gases, the top layer
was removed by decantation. New substrate was added, the
reactor was purged, pressurised and heated for the next
hydroformylation cycle.
1
crystals (48%). 31P{1H} NMR (CDCl3): δ Ϫ70.51. H NMR
(CDCl3): δ 7.94 (br d, 3J(P,H) = 5.5 Hz, 4 H), 7.19 (dd, 3J(H,H)
= 8.0 Hz, 4J(H,H) = 1.5 Hz, 4H), 7.11 (d, 3J(H,H) = 8.5 Hz, 4H),
7.06 (d, 4J(H,H) = 1.5 Hz, 2H), 6.51 (d, 4J(H,H) = 1.5 Hz, 2H),
3.47 (t, 3J(H,H) = 6.5 Hz, 0.8H; CH2Cl), 3.45 (t, 3J(H,H) = 6.5
Hz, 3.2H; CH2Br), 2.40 (t, 3J(H,H) = 7.5 Hz, 4H), 2.35 (s, 12H),
3
1.79 (q, J(H,H) = 7.5 Hz, 4H), 1.54 (s, 6H), 1.43 (m, 4H),
1.32 (m, 4H). 13C{1H} NMR (CDCl3): δ 154.38 (s), 150.88
(t, 10.3 Hz), 136.92 (s), 135.77 (vt, 21.7 Hz), 133.64 (s), 132.91
(t, 5.2 Hz), 131.76 (s), 131.69 (s), 130.05 (s), 127.11 (s),
118.30 (br s), 117.58 (s), 35.10 (s), 34.68 (s), 33.91 (s),
32.83 (s), 32.58 (s), 30.27 (s), 27.66 (s), 20.85 (s). Anal. Calc.
for C53H54Br2O3P2: C, 66.26; H, 5.67. Found: C, 66.68; H,
5.62%.
Acknowledgements
Financial support from Celanese Chemicals Europe, GMBH,
Germany is gratefully acknowledged.
D a l t o n T r a n s . , 2 0 0 4 , 1 5 9 0 – 1 5 9 6
1595

View Article Online
25 M. Kranenburg, Y. E. M. van der Burgt, P. C. J. Kamer and
P. W. N. M. van Leeuwen, Organometallics, 1995, 14, 3081.
26 L. A. van der Veen, M. D. K. Boele, F. R. Bregman, P. C. J. Kamer,
P. W. N. M. van Leeuwen, K. Goubitz, J. Fraanje, H. Schenk and
C. Bo, J. Am. Chem. Soc., 1998, 120, 11616–11626.
27 L. A. van der Veen, P. H. Keeven, G. C. Schoemaker, J. N. H. Reek,
P. C. J. Kamer, P. W. N. M. van Leeuwen, M. Lutz and A. L. Spek,
Organometallics, 2000, 19, 872–883.
References
1 R. T. Baker and W. Tumas, Science, 1999, 284, 1477–1479.
2 D. J. Cole-Hamilton, Science, 2003, 299, 1702–1706.
3 P. W. N. M. van Leeuwen, A. J. Sandee, J. N. H. Reek and P. C. J.
Kamer, J. Mol. Catal. A: Chemical, 2002, 182, 107–123.
4 A. J. Sandee, J. N. H. Reek, P. C. J. Kamer and P. W. N. M. van
Leeuwen, J. Am. Chem. Soc., 2001, 123, 8468–8476.
5 N. J. Meehan, A. J. Sandee, J. N. H. Reek, P. C. J. Kamer, P. W. N. M.
van Leeuwen and M. Poliakoff, Chem. Commun., 2000, 1497–1498.
6 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, 3231–3235.
7 K. Nozaki, Y. Itoi, F. Shibahara, E. Shirakawa, T. Ohta, H. Takaya
and T. Hiyama, J. Am. Chem. Soc., 1998, 120, 4051–4052.
8 G. E. Oosterom, J. N. H. Reek, P. C. J. Kamer and P. W. N. M. van
Leeuwen, Angew. Chem., Int. Ed., 2001, 40, 1828–1849.
9 L. Ropartz, D. F. Foster, R. E. Morris, A. M. Z. Slawin and D. J.
Cole-Hamilton, J. Chem. Soc., Dalton Trans., 2002, 1997–2008.
10 D. de Groot, B. F. M. de Waal, J. N. H. Reek, A. P. H. J. Schenning,
P. C. J. Kramer, E. W. Meijer and P. W. N. M. van Leeuwen, J. Am.
Chem. Soc., 2001, 123, 8453–8458.
11 L. Ropartz, R. E. Morris, D. F. Foster and D. J. Cole-Hamilton,
Chem. Commun., 2001, 361–362.
12 J. Dupont, C. S. Consorti and J. Spencer, J. Braz. Chem. Soc., 2000,
11, 337–344.
13 C. M. Gordon, Appl. Catal. A: General, 2001, 222, 101–117.
14 R. Sheldon, Chem. Commun., 2001, 2399–2407.
28 J. Dupont, S. M. Silva and R. F. de Souza, Catal. Lett., 2001, 77,
131–133.
29 L. A. van der Veen, P. C. J. Kamer and P. W. N. M. van Leeuwen,
Organometallics, 1999, 18, 4765–4777.
30 L. A. van der Veen, P. C. J. Kamer and P. W. N. M. van Leeuwen,
Angew. Chem., Int. Ed., 1999, 38, 336–338.
31 R. P. J. Bronger, S. M. Silva, P. C. J. Kamer and P. W. N. M. van
Leeuwen, Chem. Commun., 2002, 3044–3045.
32 R. P. J. Bronger, J. P. Bermon, J. Herwig, P. C. J. Kamer, P. W. N. M.
van Leeuwen, Adv. Synth. Catal., submitted.
33 R. P. J. Bronger, P. C. J. Kamer and P. W. N. M. van Leeuwen,
Organometallics, 2003, 22, 5358–5369.
34 T. Miyai, M. Ueba and A. Baba, Synlett, 1999, 182–184.
35 A. Castellanos-Páez, S. Castillón, C. Claver, P. W. N. M. van
Leeuwen and W. G. J. Lange, Organometallics, 1998, 17, 2543.
36 P. W. N. M. van Leeuwen, C. P. Casey and G. T. Whiteker, in
Rhodium Catalyzed Hydroformylation, ed. P. W. N. M. van Leeuwen
and C. Claver, Kluwer Academic Publishers, Dordrecht, 2000; pp.
63–105.
37 S. M. Silva, R. P. J. Bronger, Z. Freixa, J. Dupont and P. W. N. M.
van Leeuwen, New J. Chem., 2003, 27, 1294–1296.
38 J. Dupont, G. S. Fonseca, A. P. Umpierre, P. F. P. Fichtner and S. R.
Teixeira, J. Am. Chem. Soc., 2002, 124, 4228–4229.
39 J. G. Huddleston, A. E. R. W. M. Visser, H. D. Willauer, G. A.
Broker and R. D. Rogers, Green Chem., 2001, 3, 156–164.
40 P. B. Webb, M. F. Sellin, T. E. Kunene, S. Williamson, A. M. Z.
Slawin and D. J. Cole-Hamilton, J. Am. Chem. Soc., 2003, 125,
15577–15588.
41 G. Smith, D. J. Cole-Hamilton, A. C. Gregory and N. G. Gooden,
Polyhedron, 1982, 1, 97–103.
42 Although we were not able to obtain conclusive spectroscopic
evidence for this system we found during hydroformylation studies
with sulfo-Xantphos that the catalyst resting state in BMI.PF6
((sulfo-Xantphos)RhH(CO)2) is the same as in toluene (ref. 28).
43 E. B. Walczuk, P. C. J. Kamer and P. W. N. M. van Leeuwen, Angew.
Chem., Int. Ed., 2003, 42, 4665–4669.
44 S. C. van der Slot, J. Duran, J. Luten, P. C. J. Kamer and P. W. N. M.
van Leeuwen, Organometallics, 2002, 21, 3873–3883.
45 P. A. Z. Suarez, C. S. Consorti, de R. F. Souza and J. Dupont, Org.
Synth., 2002, 79, 236.
15 P. Wasserscheid and W. Keim, Angew. Chem., Int. Ed., 2000, 39,
3773–3789.
16 D. B. Zhao, M. Wu, Y. Kou and E. Min, Catal. Today, 2002, 74,
157–189.
17 C. W. Kohlpaintner, R. W. Fischer and B. Cornils, Appl. Catal. A:
General, 2001, 221, 219–225.
18 B. Cornils, J. Mol. Catal. A: Chemical, 1999, 143, 1–10.
19 F. Favre, H. Olivier-Bourbigou, D. Commereuc and L. Saussine,
Chem. Commun., 2001, 1360–1361.
20 P. C. J. Kamer, J. N. H. Reek and P. W. N. M. van Leeuwen, in
Rhodium Catalyzed Hydroformylation, ed. P. W. N. M. van Leeuwen
and C. Claver, Kluwer Academic Publishers, Dordrecht, 2000, pp.
35–62.
21 P. W. N. M. van Leeuwen, in Rhodium Catalyzed Hydroformylation,
ed. P. W. N. M. van Leeuwen and C. Claver, Kluwer Academic
Publishers, Dordrecht–Boston–London, 2000, pp. 233–251.
22 C. C. Brasse, U. Englert, A. Salzer, H. Waffenschmidt and
P. Wasserscheid, Organometallics, 2000, 19, 3818–3823.
23 D. J. Brauer, K. W. Kottsieper, C. Liek, O. Stelzer, H. Waffenschmidt
and P. Wasserscheid, J. Organomet. Chem., 2001, 630, 177–184.
24 P. Wasserscheid, H. Waffenschmidt, P. Machnitzki, K. W. Kottsieper
and O. Stelzer, Chem. Commun., 2001, 451–452.
46 L. D. Freedman, G. O. Doak and J. R. Edmisten, J. Org. Chem.,
1961, 26, 284–285.
D a l t o n T r a n s . , 2 0 0 4 , 1 5 9 0 – 1 5 9 6
1596