Phenoxaphosphino-modified xantphos-type ligands in the rhodium-catalysed hydroformylation of internal and terminal alkenes

Article abstract of DOI:10.1002/adsc.200404066

The solubility of the modifying ligand is an important parameter for the efficiency of a rhodium-catalysed hydroformylation system. A facile synthetic procedure for the preparation of well-defined xanthene-type ligands was developed in order to study the influence of alkyl substituents at the 2-, and 7-positions of the 9,9-dimethylxanthene backbone and at the 2-, and 8-positions of the phenoxaphosphino moiety of ligands 1-16 on solubility in toluene and the influence of these substituents on the performance of the ligands in the rhodium-catalysed hydroformylation. An increase in solubility from 2.3 mmol·L^-1 to > 495 mmol·L^-1 was observed from the least soluble to the most soluble ligand. A solubility of at least 58 mmol·L^-1 was estimated to be sufficient for a large-scale application of these ligands in hydroformylation. Highly active and selective catalysts for the rhodium-catalysed hydroformylation of 1-octene and trans-2-octene to nonanal, and for the hydroformylation of 2-pentene to hexanal were obtained by employing these ligands. Average rates of > 1600 (mol aldehyde) x (mol Rh)^-1 x h^-1 {conditions: p(CO/H ₂) = 20 bar, T = 353 K, [Rh] = 1 mM, [alkene] = 637 mM} and excellent regio-selectivities of up to 99% toward the linear product were obtained when 1-octene was used as substrate. For internal olefins average rates of > 145 (mol aldehyde) x (mol Rh)^-1 x h^-1 {p(CO/ H₂) = 3.6-10 bar, T = 393 K, [Rh] = 1 mM, [alkene] = 640-928 mM} and high regio-selectivities up to 91% toward the linear product were obtained.

Full text of DOI:10.1002/adsc.200404066

FULL PAPERS
Phenoxaphosphino-Modified Xantphos-Type Ligands in the
Rhodium-Catalysed Hydroformylation of Internal and Terminal
Alkenes
Raymond P. J. Bronger,a Jochem P. Bermon,a J¸rgen Herwig,b, c Paul C. J. Kamer,a
Piet W. N. M. van Leeuwena,*
a
van×t Hoff Institute for Molecular Sciences, University of Amsterdam, Nieuwe Achtergracht 166, 1018 WV Amsterdam,
The Netherlands
Fax: þ31 205255406, e-mail: pwnm@science.uva.nl
b
Celanese Chemical Europe GmbH, Werk Ruhrchemie, Otto-Roelen-Stra˚e 3, 46147 Oberhausen, Germany
c
Present address: Degussa AG, Paul-Baumannstra˚e 1, 45764 Marl, Germany
Received: March 5, 2004; Accepted: May 25, 2004
Dedicated to Dr. Joe P. Richmond on the occasion of his 60th birthday.
Supporting Information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: The solubility of the modifying ligand is an rhodium-catalysed hydroformylation of 1-octene and
important parameter for the efficiency of a rhodium- trans-2-octene to nonanal, and for the hydroformyla-
catalysed hydroformylation system. A facile synthetic tion of 2-pentene to hexanal were obtained by em-
procedure for the preparation of well-defined xan- ploying these ligands. Average rates of >1600 (mol al-
thene-type ligands was developed in order to study dehyde)Â(mol Rh)À1 ÂhÀ1 {conditions: p(CO/H2)¼
the influence of alkyl substituents at the 2-, and 7-po- 20 bar, T¼353 K, [Rh]¼1 mM, [alkene]¼637 mM}
sitions of the 9,9-dimethylxanthene backbone and at and excellent regio-selectivities of up to 99% toward
the 2-, and 8-positions of the phenoxaphosphino moi- the linear product were obtained when 1-octene was
ety of ligands 1 16 on solubility in toluene and the in- used as substrate. For internal olefins average rates
fluence of these substituents on the performance of of>145 (mol aldehyde)Â(mol Rh)À1 ÂhÀ1 {p(CO/
the ligands in the rhodium-catalysed hydroformyla- H2)¼3.6 10 bar, T¼393 K, [Rh]¼1 mM, [al-
tion. An increase in solubility from 2.3 mmol¥LÀ1 kene]¼640 928 mM} and high regio-selectivities up
to>495 mmol¥LÀ1 was observed from the least solu- to 91% toward the linear product were obtained.
ble to the most soluble ligand. A solubility of at least
58 mmol¥LÀ1 was estimated to be sufficient for a Keywords: homogeneous catalysis; hydroformylation;
large-scale application of these ligands in hydroformy- kinetics; phosphane ligands; rhodium; substituent ef-
lation. Highly active and selective catalysts for the fects; xanthene-type ligands
Introduction
To date the rhodium-catalysed hydroformylation using processes that have come on stream during the last dec-
phosphorus-based ligands is one of the most important ades.[14,15] In that respect the ligand concentration and
industrial applications of organometallic complexes as thereby the solubility of the ligand is an important pa-
5]
homogeneous catalysts.[1 Most research focuses on rameter for a number of reasons. Firstly, decomposition
tuning electronic and steric properties of the ligands in of (phosphine) ligands is followed readily by precipita-
order to gain insight into the relation between ligand tion of metal clusters, as CO will be the only remaining
structure and catalyst performance, and to optimise stabilising ligand. Ligand-free catalysis will lead to loss
the regio- and chemoselectivity and overall catalytic ac- in catalytic activity or selectivity, but might also catalyse
12]
tivity.[6
In academia, issues concerning deactivation further ligand decomposition. Secondly, a sufficiently
and decomposition of the catalytic system are rarely ad- high concentration of catalyst is required to obtain a
dressed.[13] For commercial applications, however, these high space-time yield (weight of aldehyde produced
factors are crucial. Next to catalyst selectivity and activ- per volume of catalyst per time unit, e.g., hour). Thirdly,
ity, stability has been a key issue for the many industrial ligand deposition during continuous hydroformylation
Adv. Synth. Catal. 2004, 346, 789 799
DOI: 10.1002/adsc.200404066
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
789

Raymond P. J. Bronger et al.
FULL PAPERS
processes might lead to obstruction of tubes or filters,[16] For the kinetic investigation of internal alkenes 2-pen-
and additionally volatile phosphorus compounds in the tene was preferred over trans-2-octene to simplify prod-
off-gas can lead to corrosion.
Recently, the first diphosphine ligand that shows a
uct analysis.
In order to quantify the effects of alkyl substituents,
good compromise between selectivity toward the linear and in order to facilitate identification of the ligands
aldehyde and activity in the rhodium-catalysed hydro- we prepared a range of well-defined ligands. For indus-
formylation of aliphatic internal alkenes was developed trial applications, however, a mixture of ligands with dif-
(Figure 1, 7).[12,17] Especially from an economical and ferent alkyl groups are preferred, because they can be
environmental point of view the selective hydroformy- prepared from less expensive and readily available start-
lation of internal alkenes to linear aldehydes is an impor- ing materials and a mixture of alkyl-substituted ligands
tant reaction as Raffinate II (a mixture of different bu- is advantageous for solubility (vide infra).
tenes obtained during the Naphtha steam-cracking[18]
)
can be used as feedstock. An inherent problem of these
xanthene-based ligands is their low solubility in most Results and Discussion
solvents that are used for hydroformylation, like pure
substrate, product aldehyde, toluene, xylene and ani-
Synthesis
sole. Also the ligand is hardly soluble in other solvents
such as ethers, ketones, and the high-boiling condensa-
tion products from aldehydes, the side-products that The different xanthene backbones of ligands 1 16 were
may be formed during the hydroformylation reaction. synthesised using a variety of methods. 2,7-Di-n-hexyl-
A possible way to improve the ligand solubility is mod- 9,9-dimethylxanthene,
2,7-dineohexyl-9,9-dimethyl-
ification of the ligand by introduction of aliphatic sub- xanthene and 2,7-di-n-decyl-9,9-dimethylxanthene
stituents. Here we report efficient and short routes for were synthesised by Friedel Crafts acylation of 9,9-di-
the synthesis of new Xantphenoxaphos-type ligands methylxanthene[20] with the corresponding acid chloride
1 16 (Figure 1) with highly improved solubility, thus cir- followed by the Huang Minlon modification of the
cumventing loss of activity and selectivity by ligand pre- Wolff Kischner reduction (Scheme 1).[21,22] 2,7-Di-s-bu-
cipitation.[16,19] A systematic variation of the ligand tyl-9,9-dimethylxanthene (as racemic mixture of differ-
backbone and the phenoxaphosphino moiety was initi- ent diastereomers), and 2,7-di-i-propyl-9,9-dimethyl-
ated in order to study the effect of different alkyl groups xanthene were obtained via a Wittig reaction of the ap-
on the solubility in toluene. Furthermore, the effect of propriate diketone with in situ generated methylenetri-
the substituents on catalytic performance in the hydro- phenylphosphorane followed by a palladium-catalysed
formylation of 1-octene and trans-2-octene has been in- hydrogenation (Scheme 2). It was found that Friedel
vestigated. To understand the crucial reaction param- Crafts acylation followed by ketone reduction was the
eters, we studied the kinetics for the hydroformylation most efficient method for preparing the substituted xan-
of 1-octene and 2-pentene using 13 as modifying ligand. thene derivatives. Due to the deactivating nature of acyl
groups for electrophilic aromatic substitution only one
acid chloride will react with an aromatic ring. Using
two equivalents of acid chloride leads to selectiveforma-
tion of 2,7-disubstituted xanthenes. Other methods also
yielded the desired xanthenes albeit with many side-re-
actions, resulting in tedious work-up procedures and/or
in low yield. Obvious procedures that were explored
comprise Friedel Crafts alkylation, variations on the
acid-catalysed condensation reaction of p-cresol with
No.
R1
R2
No.
R1
R2
1
2
3
4
5
6
7
8
H
Me
Me
neohexyl
n-hexyl
Me
Me
H
Me
9
neohexyl
neohexyl
neohexyl
neohexyl
n-hexyl
n-hexyl
t-octyl
H
Me
Me
Me
Me
i-Pr
s-Bu
t-Bu
t-Bu
10
11
12
13
14
15
16
neohexyl
n-hexyl
Me
n-hexyl
Me
Scheme 1. Synthesis of 2,7-dialkyl-9,9-dimethylxanthenes
(yields in parenthesis) conditions: i) 2.2 equivs. AlCl3/2.2
equivs. acid chloride (90 97%); ii) 1) H2NNH2 ¥H2O/
NaOH/120 2208C (19 65%).
n-decyl
Me
Figure 1. Xantphenoxaphos-type ligands.
790
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 789 799

Rh-Catalysed Hydroformylation of Internal and Terminal Alkenes
FULL PAPERS
Table 1. Ligand solubility; effect of changing ligand back-
bone.
No.
Ligand[a]
R1
R2
Solubility[b]
[mmol/L]
1
2
3
4
5
6
7
8
9
1
2
5
6
8
10
13
15
16
H
Me
i-Pr
s-Bu
Me
Me
Me
Me
Me
Me
Me
Me
Me
2.6
4.3
4.7
17
13
58
118
2.3
127
t-Bu
Scheme 2. Synthesis of 2,7-dialkyl-9,9-dimethylxanthenes
(yields in parenthesis) conditions: i) CH3PPh3I/n-BuLi
(51 74%), ii) Pd(0), H2 (3 bar) (38 99%).
neohexyl
n-hexyl
t-octyl
n-decyl
[a]
See Figure 1.
[b]
Solubility in toluene determined at room temperature.
uene were prepared followed by separation of the dis-
solved ligand from the non-dissolved ligand by decanta-
tion and evaporation of the toluene under vacuum. The
residue was weighed to calculate the amount of dis-
solved ligand in the saturated solution. Both experimen-
Scheme 3. Synthesis of 2,8-dialkyl-10-chlorophenoxaphos-
phines (yields in parenthesis)
conditions: i) 1.5 equivs.
AlCl3/PCl3/8 h, ii) pyridine/1 h (65 87%).
acetone to yield 2,7,9,9-tetramethylxanthene as pro- tal methods gave results that agreed within 5%. Toluene
¾
posed by Caruso et al.,[23] or palladium-catalysed cross- was used to study the solubility instead of Texanol ,
coupling reactions using 2,7-dibromoxanthene and al- which is a widely applied mimic for the heavy ends
kylmagnesium chlorides. Especially the former two syn- that are produced during hydroformylation, because
thetic procedures give a mixture of substituted xanthene toluene is easier to remove by evaporation and conse-
backbones that require elaborate identification and/or quently results in a higher accuracy. Additionally, we
purification procedures, but from an industrial point of used toluene in all our hydroformylation experiments.
view these methods are attractive for their straightfor- Only the solubility of the ligands itself was measured
ward synthesis from low-cost starting compounds, addi- and not of their metal complexes as in industry often a
tionally the mixture of alkyl substituted xanthene back- large excess of ligand (L/Rh ratio>10) is used during
bones is advantageous for solubility reasons.
catalysis. In addition, during catalysis many different
In order to study the effect of different alkyl groups on complexes are formed, which probably all have a differ-
the phenoxaphosphino moiety on the solubility and cat- ent solubility. Measurement under catalytic conditions
alytic performance, four different 10-chlorophenoxa- would also be cumbersome.
phosphines were prepared. Again, Friedel Crafts acyla-
Table 1 shows the effects of changing the substituent
tion of diphenyl ether followed by a reduction proved to R1 at the backbone while keeping the substituent at
be the most efficient way of synthesizing the required the phenoxaphosphino moiety constant (R2 ¼Me).
4,4’-dialkyldiphenyl ether, the starting material for the The first significant increase in solubility is observed
10-chloro-2,8-dialkylphenoxaphosphine synthesis. The for tert-butyl- and sec-butyl-substituted xanthene back-
10-chloro-2,8-dialkylphenoxaphosphines were pre- bones, while the sharpest increase in solubility is ob-
pared by an 8 h reflux of 4,4’-dialkyldiphenyl ether in served for substituents with six carbons in the chain. In
PCl3 in the presence of AlCl3.
that respect, the n-hexyl group (118 mmol¥LÀ1) causes
Selective dilithiation of the modified xanthenes fol- a larger increase in solubility than the bulky neohexyl
lowed by the reaction of 10-chlorophenoxaphosphines group (58 mmol¥LÀ1) (Table 1, entries 6 and 7). Rela-
at low temperature yielded the corresponding ligands tive to n-hexyl-substituted backbones a further increase
in moderate to high yields (30 80%, based on the in carbon chain length to n-decyl only causes a small
amount of xanthene backbone).
improvement in solubility from 118 mmol¥LÀ1 to
127 mmol¥LÀ1. Interestingly, the highly branched tert-
octyl- (1,1,3,3-tetramethylbutyl-)substituted ligand 15
Solubility Data
shows a solubility that is even lower (2.3 mmol¥LÀ1
)
than the solubility of the parent compound
The solubility of the ligands in toluene at room temper- (2.6 mmol¥LÀ1) (Table 1, entries 1 and 8).
ature was determined by slow addition of toluene to the To investigate the effect of alkyl substituents at the
1
ligand under continuous stirring until all ligand had dis- phenoxaphosphino moiety, four different structural var-
solved. Additionally, saturated solutions of ligand in tol- iants have been synthesised. Table 2 shows that modifi-
Adv. Synth. Catal. 2004, 346, 789 799
asc.wiley-vch.de
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
791

Raymond P. J. Bronger et al.
FULL PAPERS
Table 2. Ligand solubility; effect of changing the phenoxa-
phosphino moiety.
The neohexyl substituent enhances the ligand×s solubil-
ity to a considerably lesser extent (Table 2, compare en-
tries 1, 2 and 7, 8). This is probably the result of the close
proximity of the bulky neohexyl groups on the phenox-
aphosphino moieties of ligands 3 and 11, which might
cause hindered rotation (vide infra). Loss of degeneracy
of the different protons is observed when the constrain-
ed AA’BB’ splitting patterns of PP-ArCH2-CH2
C(CH3)3 on the phenoxaphosphino moiety are com-
pared to the splitting patterns of the neohexyl groups at-
tached to the ligand backbone. In the latter case the A
and A’ protons have (nearly) the same chemical shift,
while in the former case a significant difference in chem-
ical shift between A and A’ is observed (Figure 2).
The results indicate that the longer carbon chain and
the decreased overall aromatic character of the ligands
lead to more soluble systems. This can be rationalised
as follows: the modification causes more steric repulsion
hampering p-p stacking interactions.[24] Secondly, by in-
creasing the carbon chain length the number of possible
conformers increases and consequently the configura-
No.
Ligand[a]
R1
R2
Solubility[b]
[mmol/L]
1
2
3
2
3
4
Me
Me
Me
t-Bu
Me
4.3
20
330
4.6
neohexyl
n-hexyl
H
4
7
5
8
t-Bu
Me
13
6
7
8
9
10
11
12
9
neohexyl
neohexyl
neohexyl
neohexyl
n-hexyl
n-hexyl
n-decyl
H
Me
77
58
10
11
12
13
14
16
neohexyl
n-hexyl
Me
n-hexyl
Me
283
324
118
>495[c]
127
[a]
See Figure 1.
[b]
Solubility in toluene determined at room temperature.
[c] The solubility of this ligand is too high for an accurate mea-
surement concerning the amount of tested ligand.
cation of the phenoxaphosphino moiety, by changing R2, tional entropy of the system becomes larger. The latter
has a more pronounced effect on the solubility of the li- effect is clearly observed for the neohexyl and t-octyl
gands than placing substituents on the backbone. This modified systems. In the series n-hexyl, neohexyl to t-oc-
effect is partially caused by the number of extra alkyl tyl the rigidity of the alkyl chain increases dramatically;
groups that are attached to the ligand since modification this reduces the number of possible conformations and
of the phenoxaphosphino moiety results in four addi- gives rise to a decrease in solubility. The results obtained
tional alkyl groups compared to the two additional alkyl for the t-butyl and t-octyl modified ligands suggest that
groups obtained by modification of the ligand backbone. the increase of configurational entropy has a larger ef-
For ligands based on the 2,7,9,9-tetramethylxanthene fect on solubility than on hampering p-p stacking inter-
backbones a 76-fold increase in solubility was found actions.
for n-hexyl-substituted phenoxaphosphines compared
Comparison of the reactor content after hydroformy-
to methyl-substituted phenoxaphosphines (Table 2, en- lation reactions using 10 equivalents of ligand revealed
tries 1 and 3). Also for the other ligand backbones, the that reaction solutions containing ligands that have a
n-hexyl-substituted phenoxaphosphines showed im- solubility<13 mmol¥LÀ1 show precipitation of the li-
proved solubilities, although the effects were less pro- gand upon cooling the reaction mixture to room temper-
nounced (Table 2, compare entries 7, 9 and 10, 11). ature. It must be noted that we already started with sus-
Figure 2. Ligand 11, AA’BB’ splitting pattern for neohexyl groups on the phenoxaphosphino moiety (left) and on the ligand
backbone (right).
792
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 789 799

Rh-Catalysed Hydroformylation of Internal and Terminal Alkenes
FULL PAPERS
Table 3. Hydroformylation of 1-octene at 808C.[a]
Ligand
R1
R2
TOF[b, c]
l/b[b]
Isom.[b] [%]
9.8
Sel[b] [%]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
H
Me
Me
Me
neohexyl
n-hexyl
Me
Me
H
Me
H
1800
1660
1200
1800
1640
1700
1900
1500
2100
1900
1900
1350
1700
1200
1800
1400
97: 3
97: 3
92: 8
95: 5
97: 3
97: 3
99: 1
97: 3
99: 1
98: 2
96: 4
96: 4
97: 3
96: 4
96: 4
95: 5
87
87
84
84
87
87
87
87
87
87
88
87
87
88
86
85
10
8.6
11.3
10.4
10.4
12
i-propyl
s-butyl
t-Bu
9.8
neohexyl
12.3
10.7
8.3
9.3
9
8.3
10.5
10.4
Me
neohexyl
n-hexyl
Me
n-hexyl
Me
n-hexyl
t-octyl
n-decyl
Me
[a]
Conditions: p(CO/H2)(1: 1)¼20 bar, ligand/Rh¼5, substrate/Rh¼637, [Rh]¼1.00 mM in toluene, number of experiments
is 3 (data represent average numbers). In none of the experiments was hydrogenation observed.
Linear to branched ratio, percent linear aldehyde, percent isomerisation to 2-octene and turnover frequency were deter-
mined at ꢀ20% alkene conversion.
[b]
[c]
Turnover frequency¼(mol aldehyde)Â(mol Rh)À1 ÂhÀ1
.
pensions of these ligands in toluene before the catalytic there is hardly any effect of changes of the ligand back-
reactions were run (with the exception of 8 that gave a bone on the catalytic performance. This is different for
clear solution in toluene). From the reaction mixture the ligands bearing the more bulky neohexyl- and n-hex-
with ligand 6, which has a solubility of 17 mmol¥LÀ1
,
yl-substituted phenoxaphosphino moieties (Table 3, en-
crystals were formed within two days after stopping tries 3, 4, 11, 12 and 14). In the latter cases both activity
the reaction. The ligands that have a higher solubility and linear/branched (l/b) ratio are clearly affected.
did not show any precipitation or crystallisation.
Comparison of the effect of different substituents on
It has been reported that the application of 10 in a con- the phenoxaphosphino moieties shows that both the
tinuous hydroformylation process did not show any loss rate of isomerisation and the l/b ratio increase when
in activity or selectivity during at least the first 168 h of less bulky substituents are applied. As the overall selec-
application {substrate was either a mixture of cis- and tivity for the linear product is hardly influenced, the re-
trans-2-butene or Raffinate II, T¼1258C, p(CO/H2)- sults imply that the amount of branched alkyl-rhodium
(1:1)¼25 bar, [rhodium]¼1.85 mM, [ligand]¼28 mM, species formed during the reaction is about equal in all
catalysis was performed at 73% conversion of the sub- cases (Scheme 4). Small changes in ligand structure de-
strate}.[16]
termine the ratio between the rate of b-hydrogen elimi-
nation (Scheme 4, C ! F) versus the rate of CO-inser-
tion (Scheme 4, C ! E).
Hydroformylation of 1-Octene
Apparently it is more favourable to perform CO inser-
tion (steric requirement diminishes) than b-hydrogen
The effect of the different alkyl substituents on the per- elimination (steric requirement increases) in the case
formance of the ligands during the hydroformylation of of large steric hindrance, e.g., when large substituents
1-octene was investigated. The reactions were per- are used on the phenoxaphosphino moieties. The nega-
formed in toluene at 808C under 20 bar of 1:1 CO/H2 us- tive effect of bulky substituents (of either the ligand or
ing a 1.0 mM solution of rhodium diphosphine catalyst substrate) on selectivity for the linear aldehyde is com-
prepared from Rh(CO)2(acac) and 5 equivalents of li- monly encountered.[25,26]
gand. The formation of octene isomers, nonanal, and
2-methyloctanal was monitored by gas chromatography.
Turn-over frequencies were determined and averaged Hydroformylation of trans-2-Octene
over the initial ꢀ20% conversion. The results of the hy-
droformylation experiments are shown in Table 3.
Previous studies with related phenoxaphosphino-modi-
Comparisons of the ligands with small substituents on fied xanthene backbones showed that these types of li-
the phenoxaphosphino moiety (R2 ¼H or Me) show that gands are suitable for efficient hydroformylation of in-
Adv. Synth. Catal. 2004, 346, 789 799
asc.wiley-vch.de
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
793

Raymond P. J. Bronger et al.
FULL PAPERS
Scheme 4. Partitioning of the branched alkyl species C to E and F. B reverts almost directly to the linear aldehyde.
ternal alkenes to linear aldehydes.[12],[17] Therefore, the Table 4. Hydroformylation of trans-2-octene at 1208C.[a]
current series of ligands was used to investigate the ef-
Ligand R1
R2
TOF[b, c] l/b[b] Sel[b, d] [%]
fects of the different alkyl groups on the catalytic activity
and selectivity using 2-alkenes as substrate. Hydrofor-
mylation of trans-2-octene was carried out at 1208C un-
der 3.6 bar of 1:1 CO/H2 using a 1.0 mM solution of rho-
dium diphosphine catalyst prepared from Rh(PPh3)3
H(CO) and 10 equivalents of ligand. We chose to use
Rh(PPh3)3H(CO) as catalyst precursor instead of
Rh(CO)2(acac) for several reasons: 1) no incubation
time is needed and under the applied hydroformylation
conditions the catalyst formation is less efficient when
starting from Rh(CO)2(acac) and ligand, 2) the presence
of PPh3 might additionally stabilise the active species,
and 3) ligand exchange of PPh3 for these diphosphines
occurs rapidly, already at room temperature. The pro-
duction of octene isomers, nonanal, and branched C9-al-
dehydes was monitored by gas chromatography. Turn-
over frequencies were determined and averaged over
the initial 2 h of reaction time. The results of the hydro-
formylation experiments are shown in Table 4.
By applying catalytic conditions that promote b-hy-
drogen elimination[27] (a high reaction temperature
and low syngas pressure), the rate of isomerisation in
these systems is substantially enhanced, thereby contin-
uously replenishing the reacting terminal alkenes. With
all ligands preferential formation of the linear aldehyde
was observed. Better selectivities can be obtained by
employing even lower syn-gas pressures [p(CO/H2)
(1:1)¼2 bar instead of p(CO/H2) (1:1)¼3.6 bar] al-
though at a slight expense of activity (Table 4, entry 7).
1
2
3
5
H
Me
Me
Me
143
144
4.3
4.3
2.9
4.3
4.9
6.6
9.5
4.8
6.6
4.9
3.4
4.6
4.3
3.2
4.2
3.6
81
81
74
81
83
87
90
83
87
83
77
82
81
76
81
78
neohexyl 155
i-propyl
s-butyl
t-Bu
Me
Me
H
H
Me
H
149
141
140
112
134
142
134
6
7
7[e]
8
9
10
neohexyl
Me
11
12
13
14
15
16
neohexyl 152
n-hexyl
Me
n-hexyl
Me
Me
144
144
179
150
162
n-hexyl
t-octyl
n-decyl
[a]
Conditions: p(CO/H2)(1: 1)¼3.6 bar, ligand/Rh¼10, sub-
strate/Rh¼637, [Rh]¼1.00 mM in toluene, number of ex-
periments is 2 (data represent average numbers). In none
of the experiments was hydrogenation observed.
Linear to branched ratio, percent linear aldehyde and
turnover frequency were determined at ꢀ50% alkene
conversion.
[b]
[c]
[d]
Turnover frequency¼(mol aldehyde)Â(mol Rh)À1 ÂhÀ1
.
Percentage of linear aldehyde of all products other than
octenes.
p(CO/H2)¼2.0 bar.
[e]
These results are in line with the influence of CO and by the preference of b-hydrogen elimination over CO
H2 on activity and selectivity as observed for the hydro- insertion from the rhodium branched alkyl species
formylation of 2-pentene (vide infra). Again, the use of with the less hindered phenoxaphosphino-moieties.
different ligands bearing the same phenoxaphosphino CO insertion leads to (branched) product formation,
moiety gives similar activity and selectivity. Ligands and therefore to an increased activity and a decreased
bearing sterically less hindered phenoxaphosphino moi- selectivity, while b-hydrogen elimination is non-produc-
eties show a higher selectivity for the linear aldehyde al- tive.
beit with lower activity. This effect can also be explained
794
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 789 799

Rh-Catalysed Hydroformylation of Internal and Terminal Alkenes
FULL PAPERS
KineticStudies
reduces the concentration of the active hydroformyla-
tion species, which results in a lower activity. At very
The rate equations of the hydroformylation of 1-octene low hydrogen pressures (pH2 ¼2.5 bar) only half the ac-
at T¼808C and 2-pentene at T¼1208C using 13 were tivity was measured compared to catalysis run at hydro-
determined by a kinetic study. The concentration de- gen pressures>5 bar. This might indicate the presence
pendency of all the reactants was investigated (see Ta- of rhodium dimers at very low hydrogen pressures since
ble 5). The initial rate of aldehyde production was deter- at higher pressures a zero order dependency in pH2 was
mined by HP-IR spectroscopy or by taking samples dur- measured. Alternatively, at very low pressures hydroge-
ing the first ꢀ10% conversion. The results for 1-octene nolysis may become rate limiting. For the CO pressure a
hydroformylation are summarised in Tables 6 and 7.
À0.9 order rate dependence and for the alkene concen-
The dimerisation reaction of (diphosphine)RhH- tration a reaction order of 0.9 were measured. The reac-
(CO)2 to [(diphosphine)Rh(CO)2)]2 is an important tion orders of all parameters indicate that similar reac-
side-reaction observed under hydroformylation condi- tion kinetics as reported for the triphenylphosphine-
tions (Scheme 5). Low hydrogen pressures, low temper- based catalyst are applicable for (13)Rh(CO)2H and
atures and high rhodium concentrations promote the agree with the Type I kinetics [Eq. (7)] as proposed by
formation of these dimeric rhodium species.[28] The first van Leeuwen et al.[29] Both ligand concentration and
order dependency of reaction rate on rhodium concen- CO pressure influence the formation of the coordina-
tration indicates that within these concentration limits tively unsaturated rhodium intermediate ((diphosphi-
the dimerisation reaction is directed toward the rhodi- ne)RhH(CO)), but since there exists a 0.9 order in al-
um monomer. A small influence of ligand concentration
on initial activity was observed. From 1 mM to 5 mM no
change in activity was observed, indicating a very strong
binding of the ligand to the rhodium. The very small ef-
fect on activity at higher ligand concentrations indicates
that association of a second ligand is very difficult, but
Scheme 5. Equilibrium between Rh-dimer and Rh-hydride
feasible. The association of a second ligand effectively species.
Table 5. Concentration ranges of all reactants used for kinetic study.
1-Octene hydroformylation
2-Pentene hydroformylation
0.5 mMꢁ[Rh]ꢁ4.0 mM
212 mMꢁ[1-octene]ꢁ1062 mM
1 mMꢁ[13]ꢁ30 mM
0.25 mMꢁ[Rh]ꢁ2.0 mM
213 mMꢁ[1-octene]ꢁ1067 mM
5 mMꢁ[13]ꢁ48 mM
2.5 barꢁpH2 ꢁ32 bar
2 barꢁpH2 ꢁ16 bar
5.5 barꢁpCOꢁ30 bar
2 barꢁpCOꢁ8 bar
Table 6. Hydroformylation of 1-octene at 808C using 13.[a]
No.
pH2 [bar]
pCO [bar]
% isom. [%)
TOF[b, c]
l/b[b]
Sel[b, d] [%]
1
2
3
4
5
6
7
8
9
9.5
10.0
10.5
10.4
2.4
5.5
10.0
14.5
29.9
9.4
16.5
9.4
5.4
3150
1550
1200
700
59
39
17
14
32
31
36
33
31
34
82
88
89
90
82
87
86
83
85
85
3.0
15.3
10.6
11.9
14.6
12.0
12.9
850
5.5
9.4
1450
1550
1550
1700
1550
10.0
14.6
20.6
31.6
10.0
10.5
9.3
10
8.7
Reaction order: pH2 ¼0; pCO¼ À0.9 (R2 ¼0.981)
[a]
Conditions: ligand/Rh¼5, substrate/Rh¼637, [Rh]¼1.00 mM in toluene, number of experiments is 2. In none of the ex-
periments was hydrogenation observed.
[b]
Linear to branched ratio, percent linear aldehyde and turnover frequency were determined during ꢀ10% alkene conver-
sion.
[c]
[d]
Turnover frequency¼(mol aldehyde)Â(mol Rh)À1 ÂhÀ1
.
Percentage of linear aldehyde of all products other than octenes.
Adv. Synth. Catal. 2004, 346, 789 799
asc.wiley-vch.de
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
795

Raymond P. J. Bronger et al.
AU/min[c]
FULL PAPERS
Table 7. Hydroformylation of 1-octene at 808C using 13.[a]
[Rh][b]
AU/min[c]
[Substrate][b]
AU/min[c]
[Ligand][b]
0.5
1.0
2.0
3.0
4.0
0.124
0.246
0.441
0.637
0.886
212
425
637
849
1062
0.087
0.174
0.246
0.309
0.361
1
5
10
22
30
0.244
0.246
0.226
0.217
0.207
Reaction order¼1 (R2 ¼0.998)
Reaction order¼0.9 (R2 ¼0.996)
Reaction order¼ À0.05 (R2 ¼0.97)
[a]
General conditions: ligand/Rh¼5, substrate/Rh¼637, [Rh]¼1.00 mM in toluene, number of experiments is 2.
[b]
[c]
[Rh], [Substrate], and [Ligand] in mM.
Arbitrary units per minute
Scheme 6. Separate reaction steps of hydroformylation for the production of linear aldehydes. For internal alkenes the internal
alkene has to isomerise toward the terminal alkene [Eq. (2b) ! Eq. (2a) via b-hydrogen elimination).
kene concentration also a step later in the reaction cycle dium species. Usually hydroformylation reactions are
is rate limiting. The two rate influencing reactions that conducted under a 1:1 atmosphere of CO/H2 and thus
both fit the observed kinetics are alkene coordination the hydrogen pressure was also reduced, this might fa-
À
followed by rapid alkene insertion into the Rh H vour the formation of rhodium dimers. The rhodium
bond or rate determining hydride migration to the coor- concentration was investigated up to concentrations of
dinated alkene preceded by fast reversible coordination 2 mM, since at high rhodium concentrations ([Rh]ꢂ
of this alkene [Scheme 6, Eqs. (2a) and (3)].[29] The rate 2.5 mM) the catalyst is not stable under the chosen reac-
of CO dissociation does not play a role as this reaction is tion conditions. The results for the hydroformylation of
10 100 times faster than the rate of hydroformylation as 2-pentene are summarised in Tables 8 and 9. Within the
was found by van der Veen et al.[12,30]
chosen limits a broken order of 0.5 in the rhodium con-
centration was observed and thus the dimerisation reac-
tion does play a role under these conditions. The reac-
tion order of À0.07 in ligand concentration is compara-
ble to the one found for 1-octene hydroformylation. The
0.6 reaction order in the concentration of 2-pentene in-
ð7Þ
The CO pressure has a very large influence on the regio- dicates that either fast, reversible alkene coordination
selectivity. An increase in CO pressure of 5.5 bar to 30 followed by rate determining hydride-migration or
bar results in a drop of l/b ratio from 59 to 14, but the se- rate determining alkene coordination are rate influenc-
lectivity for the linear product increases since the ing [Scheme 6, Eqs (2b) and (3)]. The latter is most likely
amount of isomerisation decreases to a large extent.
due to the steric requirements of internal alkenes. Clear-
The 2-pentene hydroformylation experiments were ly other steps later in the catalytic cycle are also included
conducted under slightly different reaction conditions in the rate equation. For the CO pressure a reaction de-
than the 1-octene hydroformylation experiments. The pendence of À0.5 was found, which is the result of slow
reactions were run at a lower CO pressure and higher CO insertion at low CO pressures, especially CO inser-
temperature to increase the rate of isomerisation and tion from the branched alkyl rhodium species. The low
to suppress carbonylation from the branched alkyl-rho- CO pressure also leads to an increase of the coordina-
796
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 789 799

Rh-Catalysed Hydroformylation of Internal and Terminal Alkenes
FULL PAPERS
tively unsaturated rhodium intermediate [(diphos- (4) (5)/(6) should be included in the rate equation. It
phine)RhH(CO)], which might give rise to an increase seems that for 2-pentene hydroformylation the reaction
inactivity. Therateofisomerisationis stronglyinfluenced is changing toward Type II kinetics.
by the CO pressure [Scheme 6, Eq. (2b) ! Eq. (2a)].
Both hydrogen and CO pressure have a large impact
Isomerisation is an important reaction for internal olefin on the regioselectivity. An increase in the hydrogen
kinetics; the rate of isomerisation and the concentration pressure results in a slight increase in regioselectivity,
of terminal alkenes in the reaction mixture are influ- while raising the pCO leads to a lower regioselectivity.
enced by the CO pressure, but also by the concentration The latter observation is attributed to differences in
of the internal alkenes. At hydrogen pressures<3 bar a rate of b-hydrogen elimination versus CO insertion as
large influence of hydrogen pressure was observed, but was shown in previous studies.[31] The effect of hydrogen
at hydrogen pressures>5 bar the reaction is zero order pressure is not fully understood, because a zero reaction
in hydrogen pressure. The most likely explanation for order in hydrogen pressure was found at sufficiently
the influence of hydrogen pressure at pressures<3 bar high pressures and according to the 1-octene hydrofor-
is the formation of inactive rhodium dimers. At higher mylation experiments the hydrogen pressure has no in-
hydrogen pressures the rhodium dimer/monomer equi- fluence on the rate of isomerisation. At low hydrogen
librium shifts to the rhodium monomer. For 1-octene pressures rate determining hydrogenolysis could have
the rate limiting steps are predominantly Eq. (1), an effect on the regio-selectivity since differences in
Eqs. (2a) (3) in Scheme 6, but for 2-pentene hydrofor- rate of hydrogenolysis or CO deinsertion between the
mylation these steps are only partially rate limiting, branched and linear rhodium acyl species could exist.
thus step Eq. (2b) ! Eq. (2a) and the equilibrium Eqs.
Conclusions
Table 8. Hydroformylation of 2-pentene at 1208C using 13.[a]
In conclusion, we have presented an easy synthetic route
for the preparation of a series of novel diphosphine li-
gands that show large differences in solubility. The de-
veloped synthetic procedure can easily be converted to
modify other ligands in order to improve the solubility
or to study the effects of different aliphatic groups on
catalytic performance. The configurational entropy of
the alkyl chains plays a more important role on solubility
than hampering p-p stacking interactions. In that re-
spect, the effect of linear alkyl groups on solubility is
larger than the effect of branched alkyl groups, e.g.,
the effect of n-hexyl groups is larger than the effect of
neohexyl groups. Modifications close to the active cen-
tre (i.e., modification of the phenoxaphophino-moiety)
have a larger effect on catalysis than modification fur-
ther away from the active centre (i.e., modification of
the xanthene backbone).
No.
pH2 (bar)
pCO (bar)
TOF[b, c]
l/b[b]
1
2
3
4
5
6
7
8
2
2
2
2
4
5
8
16
2
3
5
8
2
2
2
2
180
149
116
99
361
313
358
358
7.0
5.4
4.1
3.8
7.7
8.7
9.7
9.1
Reaction order: pH2 ¼0; pCO¼-0.5 (R2 ¼0.994)
[a]
Conditions: ligand/Rh¼5, substrate/Rh¼928, [Rh]¼
1.00 mM in toluene, number of experiments is 2. In none
of the experiments was hydrogenation observed.
Linear to branched ratio, percent linear aldehyde and
turnover frequency were determined and averaged over
the first 0.5 h.
[b]
[c]
Turnover frequency¼(mol aldehyde)Â(mol Rh)À1 ÂhÀ1
.
Table 9. Hydroformylation of 2-pentene at 1208C using 13.[a]
[Rh][b]
AU/min[c]
[Substrate][b]
AU/min[c]
[Ligand][b]
TOF[d, e]
0.25
0.5
0.75
1.0
1.25
1.5
2.0
0.017
0.024
0.031
0.035
0.039
0.044
0.051
213
427
640
853
1067
0.0177
0.0266
0.035
0.04
5
10
14
20
48
180
163
160
156
153
0.0466
Reaction order¼0.5 (R2 ¼0.999)
Reaction order¼0.6 (R2 ¼0.999)
Reaction order¼ À0.07 (R2 ¼0.85)
[a]
General conditions: ligand/Rh¼5, substrate/Rh¼928, [Rh]¼1.00 mM in toluene, number of experiments is 2.
[b]
[c]
[d]
[e]
[Rh], [substrate], and [Ligand] in mM.
Arbitrary units per minute.
Turnover frequencies were determined and averaged over 0.5 h.
Turnover frequency¼(mol aldehyde)Â(mol Rh)À1 ÂhÀ1
.
Adv. Synth. Catal. 2004, 346, 789 799
asc.wiley-vch.de
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
797

Raymond P. J. Bronger et al.
FULL PAPERS
For the hydroformylation of 1-octene the simplest ki- trans-2-Octene Hydroformylation
netics (Type I) can be used as starting point. For 2-pen-
tene several broken reaction orders are found and thus a
simple model cannot be used explain the kinetic results.
In a typical experiment the autoclave was charged with a solu-
tion of Rh(PPh3)3H(CO) (9.2 mg, 10 mmol) and 10 equivalents
of ligand in toluene (8.5 mL). After purging the autoclave
three times with CO/H2 (1:1), the reactor was brought to 2
bar of CO/H2 (1:1). Next the autoclave was heated to 1208C.
After 30 minutes at 1208C the substrate and internal standard
(decane) were charged to the reaction mixture by overpressure
of CO/H2 (1:1).
Experimental Section
High-Pressure FT-IR Experiments
General Procedures
The high pressure FT-IR experiments were conducted in a sim-
ilar way as the standard hydroformylation experiments only us-
ing a 50-mLhome-made stainless steel autoclave equipped
with mechanical stirrer and ZnS windows. Infrared spectra
were recorded on a Nicolet 510 FT-IR spectrophotometer
and on a Bio-Rad FTS-60A spectrophotometer. The produc-
tion of aldehydes was determined by the following the increase
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/tri-
glyme. Isopropyl alcohol and dichloromethane were distilled
from CaH2. Chemicals were purchased from Acros Chimica,
and Aldrich Chemical Co. 9,9-Dimethylxanthene,[32] 2,7,9,9-
tetramethylxanthene,[23] 2,7-di-t-butyl-9,9-dimethylxanthene,[33]
10-chlorophenoxaphosphine,[12] 2,7-di-t-butyl-9,9-dimethyl-
4,5-bis(10-phenoxaphosphino)xanthene (7)[12] and 2,8-dimeth-
yl-10-chlorophenoxaphosphine[33] were prepared according to
literature procedures. 2,7-Di-t-octyl-9,9-dimethyl-4,5-bis(2,8-
dimethyl-10-phenoxaphosphino)xanthene (15) was kindly
provided by Celanese Chemicals Europe, GmbH, Germany.[16]
Syntheses of compounds 1 14, 16 36 are provided in the Sup-
porting Information. Silica gel 60 (230 400 mesh) purchased
from Merck was used for column chromatography. Melting
points were determined on a Gallenkamp MFB-595 melting
point apparatus in open capillaries and are reported uncorrect-
ed. NMR spectra were recorded on Varian Mercury 300 and In-
ova 500 spectrometers. 31P and 13C NMR spectra were meas-
ured in the 1H decoupled mode. TMS was used as an external
standard for 1H and 13C NMR and H3PO4 as an external stand-
ard for 31P NMR. Hydroformylation reactions were carried out
in a 200-mLhome-made stainless steel autoclave. The hydro-
formylation reactions were stirred at 800 rpm. The alkene
was filtered over neutral activated alumina to remove peroxide
impurities. The reactions were stopped by quenching the reac-
tions with tri-n-butyl phosphite, cooling on ice and venting the
gases. Synthesis gas (CO/H2, 1:1, 99.9%) was purchased from
Air Liquide. Gas chromatographic analysis were run on an In-
terscience HR GC Mega 2 apparatus (split/splitless injector,
J&W Scientific, DB-1 30-m column, film thickness 3.0 mm, car-
rier gas 70 kPa He, FID detector) equipped with a Hewlett
Packard Data system (Chrom-Card) using decane as an inter-
nal standard.
in intensity of the aldehyde peak at 1734 cmÀ1
.
Acknowledgements
Financial support from Celanese Chemicals Europe, GmbH,
Germany is gratefully acknowledged.
References
[1] P. Arnoldy, in: Rhodium Catalyzed Hydroformylation,
(Eds.: P. W. N. M. van Leeuwen, C. Claver), Kluwer Aca-
demic Publishers, Dordrecht, Boston, London, 2000,
pp. 203 231.
[2] B. Cornils, in: New Synthesis with Carbon Monoxide,
(Ed.: J. W. Faller), Springer Verlag, Berlin, Heidelberg,
1980, p. 1.
[3] C. D. Frohning, C. W. Kohlpaintner, in: Applied Homo-
geneous Catalysis with Organometallic Compounds: a
Comprehensive Handbook in Two Volumes, Vol. 1,
(Eds.: B. Cornils, W. A. Herrmann), VCH, Weinheim,
New York, Basel, Cambridge, Tokyo, 1996, p. 27.
[4] G. W. Parshall, Homogeneous Catalysis: the Applications
and Chemistry of Catalysis by soluble Transition Metal
Complexes, Wiley, New York, 1980.
[5] C. A. Tolman, J. W. Faller, in: Homogeneous Catalysis
with Metal Phosphine Complexes, (Ed.: L. H. Pignolet),
Plenum, New York, 1983, p. 81.
[6] C. P. Casey, G. T. Whiteker, Israel J. Chem. 1990, 30,
299 304.
1-Octene Hydroformylation
[7] C. P. Casey, E. L. Paulsen, E. W. Beuttenmueller, B. R.
Proft, L. M. Petrovich, B. A. Matter, D. R. Powell, J.
Am. Chem. Soc. 1997, 119, 11817 11825.
[8] C. P. Casey, E. L. Paulsen, E. W. Beuttenmueller, B. R.
Proft, B. A. Matter, D. R. Powell, J. Am. Chem. Soc.
1999, 121, 63 70.
In a typical experiment the autoclave was charged with a solu-
tion of Rh(CO)2(acac) (2.6 mg, 10 mmol) and 5 equivalents of
ligand in toluene (8.5 mL). After purging the autoclave three
times with CO/H2 (1:1), the reactor was brought to 16 bar of
CO/H2 (1:1). Next the autoclave was heated to 808C. After 1
hour at 808C the substrate and internal standard (decane)
were charged to the reaction mixture by overpressure of CO/
H2 (1:1).
[9] H. Klein, R. Jackstell, K.-D. Wiese, C. Borgmann, M.
Beller, Angew. Chem. Int. Ed. 2001, 40, 3408 3411.
798
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 789 799

Rh-Catalysed Hydroformylation of Internal and Terminal Alkenes
FULL PAPERS
[10] D. Selent, D. Hess, K.-D. Wiese, D. Rottger, C. Kunze, A.
Bˆrner, Angew. Chem. Int. Ed. 2001, 40, 1696 1698.
[11] 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, C. Bo, J. Am. Chem. Soc. 1998,
120, 11616 11626.
[12] L. A. van der Veen, P. C. J. Kamer, P. W. N. M. van Leeu-
wen, Organometallics 1999, 18, 4765 4777.
[13] E. B. Walczuk, P. C. J. Kamer, P. W. N. M. van Leeuwen,
Angew. Chem. Int. Ed. 2003, 42, 4665 4669.
[14] P. W. N. M. van Leeuwen, in: Rhodium Catalyzed Hydro-
formylation, (Eds.: P. W. N. M. van Leeuwen, C. Claver),
Kluwer Academic Publishers, Dordrecht, Boston, Lon-
don, 2000, pp. 233 251.
[15] P. W. N. M. van Leeuwen, Appl. Catal. A: General 2001,
212, 61 81.
[23] A. J. Caruso, J. L. Lee, J. Org. Chem. 1997, 62, 1058
1063.
[24] E. H. Elandaloussi, P. Frere, P. Richomme, J. Orduna, J.
Garin, J. Roncali, J. Am. Chem. Soc. 1997, 119, 10774
10784.
[25] K. Nozaki, T. Nanno, H. Takaya, J. Organomet. Chem.
1997, 527, 103 108.
[26] H. Riihimaki, T. Kangas, P. Suomalainen, H. K. Reinius,
S. Jaaskelainen, M. Haukka, A. O. I. Krause, T. A. Pak-
kanen, J. T. Pursiainen, J. Mol. Cat. A: Chemical 2003,
200, 81 94.
[27] A. van Rooy, E. N. Orij, P. C. J. Kamer, P. W. N. M. van -
Leeuwen, Organometallics 1995, 14, 34 43.
¬
¬
[28] A. Castellanos-Paez, S. Castillon, C. Claver, P. W. N. M.
van Leeuwen, W. G. J. Lange, Organometallics 1998, 17,
2543.
[16] H. Bohnen, J. Herwig, (Celanese Chem. Europe GmbH),
WO 02068369, 2002.
[29] P. W. N. M. van Leeuwen, C. P. Casey, G. T. Whiteker, in:
Rhodium Catalyzed Hydroformylation, (Eds.: P. W. N. M.
van Leeuwen, C. Claver), Kluwer Academic Publishers,
Dordrecht, 2000, pp. 63 105.
[30] 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, A. L. Spek, Organometallics 2000, 19, 872 883.
[31] R. P. J. Bronger, P. C. J. Kamer, P. W. N. M van Leeuwen,
Organometallics 2003, 22, 5358 5369.
[32] J. S. Nowick, P. Ballester, F. Ebmeyer, J. Rebek, J. Am.
Chem. Soc. 1990, 112, 8902 8906.
[33] L. D. Freedman, G. O. Doak, J. R. Edmisten, J. Org.
Chem. 1961, 26, 284 285.
[17] L. A. van der Veen, P. C. J. Kamer, P. W. N. M. van Leeu-
wen, Angew. Chem. Int. Ed. 1999, 38, 336 338.
[18] K. Weissermel, H.-J. Arpe, Industrielle Organische
Chemie; VCH Verlagsgesellschaft mbH, Weinheim,
Basel, Cambridge, New York, 1988.
[19] J. Herwig, H. Bohnen, P. Skutta, S. Sturm, P. W. N. M.
van Leeuwen, R. P. J. Bronger, (Celanese Chem. Europe
GmbH), WO 02068434, 2002.
[20] R. P. J. Bronger, S. M. Silva, P. C. J. Kamer, P. W. N. M.
van Leeuwen, Chem. Commun. 2002, 3044 3045.
[21] H. Minlon J. Am. Chem. Soc. 1946, 68, 2487 2488.
[22] T. Miyai, M. Ueba, A. Baba Synlett 1999, 182 184.
Adv. Synth. Catal. 2004, 346, 789 799
asc.wiley-vch.de
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
799